INDEX TO VOLUME III

PAGE

AN SHER IEVENACEN OS We \iieanin bp OageomogeaGno on 239

Abutilon austro-africanum, Hochr.......... 245

INCA Cla WWill Gl wary avant cre ect ent eceersts. e+ 67, 74, 80

Acacia....225, 226, 227, 228, 229, 230, 231, 232,

234, 235, 239, 240, 242, 244, 245, 246,

250, 251, 253, 254, 255, 258

Acacia: ataxacanthas "DN Cra mess ssc lace 504

Bentham, Rochbr-..s------.--.--- 236

debinens, "Bunches ciate cele arises 505

Gemmirchi, [seni ooo poe osouEOO ue 504

iheteracanthaye urchins set se se «2 ol 236

horridals Walldeeeynwis doesn les 337, 505

Karrooy Elayme wetter vere tee 236, 237, 505

Acacia litakunensis, Burch.............- 236, 487

Acacia pennata, Willdie os .s0. 52. oo oe 236, 505

permixta, Burtt Davy var. glabra,

Bz) edeshesieine pues bead ots 504

robusta, Burch: sce cote coe 504

LCT TOS TCHS ISU Goa ca ncn Oe GoGGee. HOUAOG 237

Acaciatsenegals Walldbeeraceioeee oe oes 237

spirocarpoides, Hngl................. 236

Stolonifera,sburcheeeome eee miee see 487, 505

Maen eTie IHL soobonsood bonoRoOoOS 487

Woodii Burtt Daviyeen asec 237

SPistiseeeea ee weer oe eee ae e 387, 469

Acaena sanguisorba, Auct.............----+- 604

Acaena sarmentosa, Carm....... 592, 595, 604, 605

Acalypha glabratan Lhbsesesses esse ee ee 242

Indicaw enna sNks teats coat ee 242

HA CAUNDEEDA CH AUB ais nee aaamersee ete ee ne 254

UNCED Seeenieritieses Vite ie eee ae et clay 343, 367

ANcTOceras macrum, Stapiene see. sees 593

Adansonia dipitatewlee sees sees sea es. 246

Adenia; Horskse saa 513, 514, 516, 517, 534

A Revision of the South African

Speciestofere en cas teres. cte ns 513

Adenia angustisecta, Burtt Davy..........-. 542

angustisecta, Engl. and Harms ex

TOE Ye ac aa aera 542, 559

Buchananii, Harms.......... 542, 554, 570

Adenia cissampeloides (Planch), Harms...... 535

Adenia digitata, Burtt Davy.............. 514, 542

Adenia digitata (Hary.), Engl..515, 516, 518, 527,

528, 529, 530, 531,

532, 536, 541, 549,

550, 551, 559, 561,

562

fruticosa, Burtt Davy. .515, 528, 530, 532,

538, 545, 546

glauca Schinz.......... 514,515, 517, 518:

S21 5225 5235) 530)

532, 539, 548, 549°

elobosassbinolies: se keris che ca 515

Adenia gummifera, Burtt Davy............. 536

Adenia gummifera (Hary.), Harms..514, 515, 519,

F 523, 524, 530,

SBP, apis

Adenia hastata, Burtt Davy.......::....... 537

Adenia hastata (Harv.), Schinz. ..514, 515, 519, 520

521, 53

Keinkaad (Masts) seb olseresee setts 514

Adenia multiflora, Potts.........514, 531, 541, 542

Adenia palmata, Forsk

Pechnuelli (Engl

B) iniammiseaeer rey acrce

repanda (Burch), Engl. . .247, 514, 515, 525,

Adenia Schlechter, Harms...........-..-..- 537

senensis (K1.), Engl.....514, 515, 530, 542,

senensis, Mast..

552, 553,

sa0oot 552, 555, 596, 557, 569

Adenia spinosa, Burtt Davy... .514, 515, 517, 522,

523, 530, 532, 533,

547

Adenia stenodactyla, Harms........:..-.--+- 542

stenophylla, Harms...........542, 558, 560

Adenia Wilmsii, Harms... .515, 518, 519, 526, 530,

sphenophyllus......

Stricolorsmaccieet

SbrihlOrlshiass-cee ere.

sie aVADe Sos ooncoG06

532, 538

623, 624

npoonDedsouDE 615, 617, 642

venenatasBorslkovs2 0" aankiese 513, 514, 519

Adromischus Lemaire, A Review of the Genus 613

Adromischus Alstoni............ 615, 618, 638, 639

JARNB OLUBLIA Ss Ae pea ranisint aie ek MOONE 617, 618, 628

A. caryophyllaceus. ....614, 615, 616, 617, 628, 629

Agiclayitoliissseeet iene 614, 616, 618, 636, 637

A. Cooperi. ..614, 615, 616, 617, 618, 632, 635, 639

IRRGCHIStALUSt EE Onion nike 614, 616, 618, 635

A. festivus........ 614, 617, 618, 632, 633, 634, 637

IA fli canister rersents Missoni oe 615, 630

(AS tusiformisermet. ochs ae oe icone 614, 615, 617, 618

A. hemisphaericus.614, 615, 617, 618, 625, 627, 628

PAE UINILIS -peynces Siete Sieeoe casa 614, 618, 639

Ate} aSMInT A ORUSy. 3/500) 1.1. coat recreate 629

AeKethackiit. tne Oicnies soe ee 620

A. kleinioides.............. 614, 615, 617, 618, 631

Amleucothrixe=eerece ae 614, 618, 637, 649

A. maculatus...... 614, 615, 616, 617, 618, 620, 622

jamenrmailbistece 5 oobeouos 614, 615, 617, 618, 631

JN, WERTENM EGS Soeocnaood 614, 615, 617, 618, 631, 639

Af Marlothiincianndeercettene oe ee 630

A. montium-Klinghardtii................ 614, 620

ASM UCrONa US hee. cre een erties 613, 622

Ananustemaneaee 614, 617, 618, 640, 641, 649, 650

Atspachylophusscms aiaacce se ee oe 614, 618, 633

AMS DROCULVUS sstarr-ryster-feser tie Clases 617, 641

Ae EHOMbifO HUSH aera eee 614, 615, 624, 625

iAesrObUStbushek GPE e Cee yas me aeier Feb. Ree 613, 623

ASOLUNICULOLUIS Sen ee eer 614, 617, 618, 627, 628

A rUpIcOlussererierser ta 614, 615, 617, 618, 642, 645

A SaRmICOlUS ate oes oe too DL ee meee 647

AeSchaeterianussnae eee on fea 614

A&Schonlan daiwa Sate oes cae 637

JAmisphaerophyllustn.c- toes et ieee 620

. umbraticolus.......

614, 615, 617, 618, 643, 644

646, 650, 651

PAGE

fds VERN Ce LEWC MoooobccoasoaoDgnecooDED 637

HA ZS YELL 5 cree ory sine tes aye aqegabats rhoueqeieee 614, 615, 635

Aecidium, Burtt-Davy Doidge............. 487

Aecidium bylianum, Syd................... 488

brideliae, P. Henn. et Evans....... 488

Aecidium Dinteri, Doidge................-- 487

HACIA, ISNVOle coo uo ocbeDs00DbAC 489

Aecidium limon, Duby.............------- 508

Aecidium litakunensis, Doidge.............. 487

Wie fereati, IDYor keke oo u cn couaceoone4 488

Aecidium statices, Desm.......... ss epaeauetd eens 508

Aecidium tetragoniae, Doidge.............. 488

Aeolanthus Rehmannii, Guerke........:.... 252

Ageratumcomyzoides) leass 2 eerie 257

NO TODY MONE LOME RUIN arti crakear 286

ASTOStISpH elena Steud anh teeters 603

ih 1b Cheeta Oe reac care cen ie a ceror 592

TamMUlosar tye, ccvocuia Seite acu: 592

Sinmulans ailem seem elie eerie 602

SP sete ie Sestavere ote scenic aera, alareas a toyororornne ene 595

ANT} BANAL), MOMS inc coo copa ca anboodoe 198, 199

caryophyillala. ne eieeieeerrees atin 185

pralecox Ls 522... src Seat ae een ee 185

IZ OACH AB sass Sia aeRO REE 233

Albizzia Rogersii, Burtt-Davy.............. 236

Allbucaraltissima D nyand sere eee eee 290, 320

ihereroensiss Balkeseee eee eens 229

MMocccocvopcoso poo DUD LaCCOOCOCUC 229

Alle ppopeRinet je -.n. 5: a cen MRO RE EE 594

AUT Gi ASBier ra seus scatye, Shenae orao 5 SOC Ee 597, 606

Alistilus bechuanicus, N. B. Br.............. 240

Allium cepa L... .361, 369, 396, 413, 425, 439, 448,

450, 454

BD. tity i sispcpaphateotsia nals be aunts eee 418

p-Uotntie Seat remaeetis Hay een re cence 6 u's 4 c 343,

Alo ctatricamale mies acetic ee ree 145

ARbOrESCENSh;..<:s-§ yee l-is ecg user ete foe ohOr

IDE Albee s Aen On ao ote o wb ol 647

Pallcartisns ccc ufero. 5 a fi 5 eqferemeiereuele EMSS 142

TOLORteia.c: visi, sisson ee ECE 144, 146

feroxi var. SUpralacyisa hese: 143

PARTIC. Ones CHrting GooeG cogdb.c co.cc 142

onandidentatarenrsne tenets r nent 142

MICLOStISMAL,. «ejecta 142

Obseurar.. iat. cider center a ETE 142

pluridens:s «04/5/21 5 caystvegernery tage Oe 142

purpurascens)... sc eee eeacee EET 142

bl oyecor=lhbnieel, Amb ooo ncoucoeceoceuve 229

Salm-dyckian aan steele eee 142

speciosa, bakermnm qe ecient 142, 143, 145,

146

spicata, Linn. f., Notes on..........2.. 142

oc, Iestlesoococsodscus 142, 143, 144, 145,

146

SUCCOETINA is cps Klacid tenes stoeto oe enone 142

Tomlmsonii.. 2... nee 143, 146

TraNSVaAaleNsIS: 3.) <eierie ea eK eR 647

Alternanthera repens (L.), O. Ktze.......... 232

Sessilis) (U2) ive Sr: eee eerie 232

AMAR AN TACHA fitmeceir ck.) eee eee 232

AMUN RGA DIUDINC ISVALID, caposoopoo boo uc KooGS 230

Amphilopha insculpta Stapf................ 497

Amphilophis insculpta Stapf.............. 299, 320

Ananas comosus, Merr...............+++- 396, 454

ANACARDIACHAH: (os ss sje «ssn 00 ceehaeleramne 242

Anapalina revoluta, N. BE. Br............+++- 510

PAGE

AEH MAMTA Wie WIrcocgocondscecccccscs 5

ANCHO NYLOMCACs soos neocga0g0sodondunEoe00 497

Andropogon amplectens, Nees.............. 512

Andropogon australis, Spreng............... 297

Andropogon contortus, Linn................. 301

Andropogon eucomis, Nees................. 497

imbayhertoMe, ICONS iooacoonesne 293, 320

hunllensiss Riend| eee eee 497

InbeLMe disseny > cee 297, 320

Ischaemum, Linn.............. 298

Andropogon lepidus, Nees................ 309, 320

Andropogon multinervis................... 512

Andropogon papillosus, Hochst............ 298, 320

METLUSUS Vall eet reer 297, 320

FOOD, ISIN NG soeocasonneccce 305, 320

TG HAG UH, VINO Soc so0coo5 2 h06 297

Schoenanthus, Linn........... 299, 320

sorghum, Brot...295, 320, 373, 396, 454

Andropogon virginicus, Linn................ 304

sp... .300, 301, 306, 309, 317, 320, 512

Androstachys Johnsonii, Prain.............. 241

AININONACH ABR senigt «steed ehreerenee 235

AMONG. Tela es oe kee 341

Amtelopererassi: se «nrg ere ene eR 593

Anthephora pubescens, Nees...........-- 295, 320

Anthericum elongatum, Willd. var. holos-

tachyunuBalkise- acl ter doer aei ee 229

Antherothamnus Pearsonii, N. B. Br......... 271

Antherothamnus rigida (L. Bolus), N. E. Br.. 271

Anthistiria Spaces oeomrratn toe eee 314, 320

Anthocleista zambesiaca, Bak.............. 249

Antholyzalal bySssivi caer rene see easier 510

nervosa: Mhunijosy aos Aerene 510

Anthospermum lanceolatum Thb...........- 256

Anthracordeas ewer vee meee 302

Antirrhinum majus L........... 361, 365, 428, 455

Apium australe, Thouars............ 592, 595, 605

APOCMANIAC HAW Ne necro tye eee caeience aS 249

APRONOGERONACHARPE for oh caine 225

Aponogeton gracilis Schinx................. 225

HolubusOlive formar.asseeeeer 225

Rehman Olivansee eee 225, 229

UN 0) 9) RRO aera cco esc © ccc Sc 455

Aptosimum lineare, Marl. and Engl.......... 253

patulum, Bremekamp........... 253

Arachis hypogaea, L............ 361, 407, 428, 455

Argyrolobium transvaalense, Schinz......... 238

Aristida adscensionis, Linn...............227, 498

ATIZONI Cal, \VASCYi.-1 te sere ers eich) 303

barbicollishilandsheereeeeeeees 227

junciformis, Trin. and Rupr..227, 303, 320

MELIdION aes Weenies cee cee 227

~eAmistica: EUs byiedsemmmrrieneieree < creieise ieee 303

AGIStida UnUpluimisswerchiteneee screen eer 227

Welwitschia, Rendly.......::.2...-. 498

3) cba or Gono. co. GOD aR nee olatiS oc0 bo 498

Arrhenatherum, ibeaivenh «1 tesla 186

Artabotrys brachypetalus, Benth............ 235

"ASCLHPTADA CHAR error > +. coche deemeee 249

Asclepias Burchellii, Schltr... .. .5. 0c. 249

Asparagus exuvialis, Burch.,............... 230

JAS OPE FAs) JEANS 5665 ci ccadeesedosoaoane 348

EIS Go abide 6 v.0, moaiobine V0 UIOS-d0 230

Aspidiotus furcillae, Brain..).......1.... 337, 469

perniciosus, Comst............. 337, 469

rapax, Comstierdrsierasieiy teers 337, 469

PAGE

Asplenium obtusatum Forst. f. var.

mrascmn (O, Chitnicochodosseoeonannne cad 595, 601

FENSII OOS Cheep C070 ECESRCHO eRe CIO ince ee a 410, 455

Aster luteus (N. E. Br.) Hutch. forma....... 257

ANSIM OI OWIDIBIAD on seanccoouoeeccudacGUUon 205

Asystasia atriplicifolia, Bremekamp......... 254

Atherstonea decussata, Pappe..........+..+-- 587

FAUT ACUI CHLAUUTIU ea. ee tie ete citi, <te 2s five = 343

GGA Parc coousocetecooacoac se 335

Awe eA, IWS. Selec ce poodoocubesocsoees 362, 469

Mica) © -pecsterexstatave: ty sisseye ereeeratab ates 594

AV GHE ED Soneneoen eo eoae oc mates eles crioo 185

Avena antarctica (Forst.), Thb........ 197, 198, 199

PAGICTONCO fii bal pis ryaiscielsy ars era ictoke > 195, 196

AS Gaby Gene Wis goncan on eedeudoo oo DOue OE 185

flavescens lit. care c.cassei he a dersirisjehey terse: ° 185

lauhtige (SO meshes cerca nec seo OneA OO de 194

“ AlGra limi, SOWEACK ooo nocomcensoeooapocde 193

PANE ANCONA: StCUG arenes eiseiiom aeetsiel atlas « 191

longa; Stapiayiat scan ose ae oar 189

AtVenanplatensiseuinerst traces cree ree 186

Avena quinqueseta, Steud.................-- 188

PAV CIIAISaUliviay lems teteie ce itebe eierseke sorte 288, 290, 320

SEMPenVITCHS Pk OSUnaraccite eee classic 185

Avena symphicarpa Trin. ex Steud.......... 193

WOE, SEY Nes cobsseeadnooedave 196, 197

INVENASECUMM, JOSH, <e\cle oe leleraise sete vare, sueseee na 185, 186

Avenastrum antarcticum, Stapf............ 191, 193

Gi fjiCi0n SUEY Nee oncod5s30503000C 195

caffrum, Stapf var. ? natalensis,

SENN cocoa pide Cone ee no oTy,. 194

Doda as tarp terete sl sert toons 197

dregeanum (Steud), Stapf,..... 189, 190

Avenastrum lachnanthum, Pilger............ 195

Avenastrum longum, Stapf var. grande, Stapf.. 189

OUHTOD SHERI 5 co goed occagoDBE 189

quinquesctum, Stapf............. 188

turgidulum, Stapf............ 192, 196

IN SOCLCO NSO eee ee eee eee ats 455, 464

Bacchariswlinne ese eee eee eee eee 205

sited Wea olde, URINE Soe oc odaoosc oboe 212, 214

MECH Wrearccecconnocsap aor 205, 207

Bacterium solanacearum Tonetti.......... 409, 462

eae DAKAMIA Clea reece cce ote eee cree ee aS 396

Balanites australis, Bremekamp............. 240

IEW Ne 6.6 1 cece poeta ois Cena oa 346, 359, 463

Barleria Bremekampi, Obermeyer........... 254

elegans, S. Moore...:..-..........- 254

Galjoian, C BsOheencssceneccasdac 254

heterotrichas) bmndawe-nenis ese ee ee 254.

ODUUSAENNEESH tate seveve.e ster evee cies steehee 254

transvaalensis, Obermeyer.......... 254

BartraMias pee che seciescle ert ae lofoi.s 6 600

liyicieinel bree eadoe ooac 66 boenose BanCoumOE 80

1 BYE Breas SOc etoncis Ine ch Cre eRe ha. SacucanR ReGen 455, 464

Beam Re paryoss stitches etey a eratopeioe. > 464

LGR NAME) Bdbe aos ob ooacooecesooonouDeeuOS 353

Becium angustifolium..................... 647

Becium obovatum N.E. Br................. 252

BiGinoo gorlboeo bec coadpe oo condbis a amor eon 439

le Gien sel oeiragios elas 5 Nae 5 oo Somat UD OOS 594

Berchemia discollar (K].) Hemsl............. 243

Berkheyopsis bechuanensis Sp. Moore........ 258

SZ DEZEMDORCH gemerteras oer cere iss eee aie 84

PAGE

Bidens pilosa; Winn-eP eee eee ae er elon 20

BIGINONTAC HAIR. : rank tees rat meee eres 253

Blechnum australe L.................594, 597, 601

Blechnum penna-marina, Kuhn......594, 595, 601

tabulare, Kuhn..592, 595, 596, 597, 601,

603, 604, 605, 607, 612

Blepharanthes, Sith. 05.2... le see oe 513, 514

Blepharis Clarket, Schinz:....-..s4c.%.....: 254

diversispina (Nees), C. 5 wer Riceetessene 254

Blindia brachystegia, Dixon.............. 598, 599

gracillimay Mitt... 2... 24. set... 2598; 599

magellanica, Schimp.............598, 599

IBHisterjrusti nn eacara cro sen ete ee ee 392

Blumea caffra (D.C.), O. Hoffm............. 257

acer all) Ocscivertinttery alot cee 257

matalensis sche Bips-cmer- eee 257

ye Bob be] aangsre nao ssa 5 te eRe eee. 540

IsoeWnREH RE) Choe, Wecococdocbacoooesosasar 233

Boerhaavia fallacissima..........-.+++.---- 233

plumbagineas Cane ease eae 232

Boer haavwial Schinziiseen es eee ee eee 233

Boerhaavia verticillata, Poir................- 233

BOMBA CAC HAW ys chs: oe carne ane oes 246

IBORAGINACHAB oo .ci eee eee 251

iBosciaialbitrunca |Gsand) Biaceae. ceri ae 235

Nvehmannian asa ee ee eee rere 235

Rehmanniana Pest. forma............ 235

PB OSO aT ah ee were Wo Grou... aC enw ela 51

Bothriochloa glabra, A. Camus......

insculpta, A. Camus....

pertusa (Willd), A. Camus..

.297, 298, 320

. 299, 320, 321

.225, 297,

320, 321

IO UWMMGO NOL) Hccocce oe gogo osto UME: 297, 321

Brachiaria semen te citacete sree Ooo 460

brizanuhasys tap ieee rere rete 371

deflexa (Schum), C. E. Hubb ex

1340) 0) 2 ap amo Aaiials died Gon car 226

grossa, Stapl.-/at jue ccc esse 226

nigropedata (Munro ex Hiern.),

Sta plirisci..torerengererepacrrce 226

Pubitoliasstapieeereeeeeee ees 346, 460

Brachiaria reguiaris, stapl...2--54s45-45 +: 226

Brachychlaena, Post et Kuntze............205, 206

Brachylaena R. Br.—

‘“A Revision of the South

African species of’—........ 205

iBrachylaena dentata. DC. 216

dentata (Thb.), Less............ 209

dentata var. B salicina DC....... 216

Brachylaena discolor, DC.. .-205, 206, 219, 221

Brachyliena elliptica, DC... REE peels

Brachylaena elliptica (Thb.), “Less. . ats 205, 206, 216

glabra (L.f.), ieee A ee 205, 206, 209

Brachylaena grandifolia, DC................. 209

Brachylaena huillensis, O. Hoffm.......... 206, 212

Brachylaena Hutchinsii, Hutch.............. 214

Brachylaena ilicifolia (Lam.), Phil and

Schwc:rce secure temccieks eamienee 205, 206, 212

Brachylaena natalensis, Sch. Bip............- 220

Brachylaena neriifolia (L.), R. Br 72052206; 207

Brachylaena racemosa (Thb.), Less........... 213

Brachylaena rotundata, 8. Moore..... 206, 218, 220

transvaalensis, Phill. and

MdikaahsoossococnomoDe 206, 214, 257

uniorayelarvetmereceioe © oc 206, 211

Brachiylaensis peer tlertee leit terr tr 257

Brachy stelma Gerrardi, Harv............. 273, 274

nigrum, R. A. Dyer... .273, 274,278

Braclreenis fee cise svene a eeat ie steve ter erst ame ee 362, ee

Brassica oleracea, Iv... ..5.-6 >> 396, 399, 428, 45:

Bredell, H. C.—

** A Revision of the South African Species

Gye Lal qoeraewUtN —, Sono oobbdsoncedvouns 571

Bredell, H. C. and E. P. Phillips—

“The Genus Elyonurus, Humb and Bonpl.

ines Adicas? =. ovata saves enters 259

Breutelia tenuifolia (Mitt.), Par............. 600

Bnd eliammolliss blutchneer ener eerie 241

Brizaymaxim a. 1... oases ke eee 311, 321

TBE OTNUS'. Sia cecaie te puss Oro dbeusnsueueperans| eae eee 460

bifidus,; Dhibe.)-atae aecl deriorscactentoter 198

seemless, WittiNesocoosocancoobsuaoar 293

unioloides, H. B. K...... PAB, BIL, B37, B73}

Broom. Corne.):).2 fae <meneucres tee aoe 396, 455

Bryum flaccidifolium, Dix...............598, 600

NeOdaMeEnse..) passes eee oe Pee wo 600

HEME IDS ssh orancoccno anne otto AUO)

Bug, Australiana: acolo eter 362, 469

iBurkearainicana st ookane ee eeeeE eee 237, 648

Brorril ars sei e pales ous Slt RR 284, 285

BURSHRACHAH.-< erie eee 241

Bush Jarrah aug teeter eee ene 51

Cab ba ee se iaisic <s; -valscecthetel ae eitoereh aveeiePe ts 399, 455

Cadaba macropoda, Gilg.................-.. 235

Cada ba termitaria, INpbe bracers ols 235

@recomaidestiens. eee eon 314

OVUVOCOUID ccarrterororeseceeee Ce Sate EO 294.

SEQCLUM xis: ete ee Eure oe 288, 290, 291, 311

SULOP NY LLU iors) apetes hele rel nicieieier siete les 311

staticis, Rudolphi... 26.2 -+-.-4-- > - 598

SOP MOGPSMH: dosbacds0 sade pecxc00d 293

Urtchoph ora... qireiersciep tele errs 291

YAS Pe ROI clots Go Ge ee ae osc otc 292

Callistephus chinensis, Nees.......... 362, 410, 455

Callymenia Harveyana, J. Ag............... 597

(Caloneetniaagniiid tlle eeu aer ere 344

De arNeSsit,.,seeegvoried iemkererhesecte 344

GOCOV)... \s,ccaiennciecce gee ree Ie 344

CUNINUD.) coc ene eee 344

MASSON neuter otreccr ere 344

[PMODMOBoancoreooacsedoaaseod 344

NAMIE OES, coocoonnooaddHboune 341

Calpurnia lasiogyne, H. Mey..............-- 237

Calpurnia subdecandra (L’Herit), Schweick... 237

CAMPANUILDA CHART cere ccitenercn reer 256

Campanula medium, L.. weeeeeese+. 000, 405

Campylopus introflexus, rid i caiace eee 599

Camanyalslamd’ Pines jysits stench heletsstes seer 594

Canterbury. bell, .icicte pane eels 455

Canthium huillensesElierns.eeeee eee 256

ventosa (L.) Sp., Moore........... 256

CARPARTDACE AIS vs aaiciy-cetielintccrererriere 235

(Oa PSTIS sacle note ere locstss overs iors keira) oft Ree 240

COMEenLOSas Mays «i leueteyeesntens Cremer 235

Caralluma atrosanguinea, N. E. Br........... 250

Caralluma carnosa, Schweick non Stent...... 250

Carallitma Carmoss, St ts «oe a yeaeys «i's ejelere 250

Caralluma grandidens, Verdoorn............ 250)

Caralluma Keithii, Dyer...........4......- 250

IMMACULATE WIN inp Mest aiid sistome: )areveiee 250

Schweicherdtii, Obermeyer....... 250,

PAGE

Cardiospermum alatum, Bremekamp and

Obermeyeni a. ccysei «ci seieyia ae eee eee 243

Carex cernua, Boott var. austro-africana, Kuk. 492

ethiopicasschicihienranen eerie ta 294, 321

petitiana, An Rich eee reece eee 507

phacotawspreng eee eee 294, 321

Tiparia,, CULLIS wetleitaeereerr ee ere 294

AMaQwenegssi, CHiN paoaoananeaobocos 596, 603

SPecie env case Coe ee eee 286, 316, 321

Carica papaya L...........362, 387, 392, 417, 422,

448, 450, 456

Carmationine cc seeoon on cain ee ere 367, 456

Carrot rn csos nie oscee Meet or ie ee 438

CAR MO RERYanWA CHA ner erste eae 604

CassianyicA ase sien nine oe oa See oe Oe eae 249

arachoidesss munch storman? ere eee eee 237

delagoensis, Hlatvicn rst see ones care 237

holosericea, TexeeE IDENAF oo cocoon aces 237

iholosericeas Hiresensepmeine rice eee 237

ObovatayColladees-seeeeeee ene 237

Cassine Schlechteri (Loes), Davison.......... 243

Catophractes..... 227, 232, 234, 235, 236, 239, 240,

241, 242, 244, 245, 247, 251, 253,

254, 257, 258

ANigxeyasbet, IDO. cagons0e0con0o46 253

CELASTRACH AK teeter 243

Celosia scabra (Schinz) (Hermbstaedtia scabra

Sohinz) sv. caem cuca eer ieee merase 232

Cenchrusiciliariseleeeee eee eee 227, 304, 321, 324

Gomis, ANON Vosancocssancnocos 304

(Ops, QA Winonacoocondssca90000 362, 456

MOTA NH Weoanddoacoose0d0000 436, 451

Ceraphoraianad-eeacoe cee ennai 343

Cerastium caespitosum, Gillib............ 594, 604

Ceratotheca triloba, HE. Mey................ 254

Cerebellar y.nccrrarch tae beri ohare 319

Ceropegia cimiciodora, Obermeyer........... 250

mlbonhyss, Wl asoodesnoopdaaccas 274.

Ceroplastisis peace ere ieee errr 387, 469

Chameleon yea eee eee ereroerr 406

CHENOPRODTACHAB eer eee secrets 232

Chenopodium ambrosoides L................ 232

OMOMES ON Seococagnccaooscogessagnece0 337, 469

CHMORORED CHAR epee tciietirntlaek 597

Chloris abyssinica, Hochst..............-. 319, 321

RENEE, INMIMHN>oooansoocuscecdes 319; 321

WOMEN as gc oocosnceuadaguponceoos 641

WAWFEENIBH invooo ons s eos aa onovacpEcosS 227

Chrysanthemum leucanthemum L......... 594, 606

Chrysomphalus aurantii, Mask............ UE 469

Chrysophyllum magalismontanum, Sond..... 248

Chrysophyllum Wilmsii, Engl...........++-- 248

@intractiawncnicrae teeter eerie k 287, 302

CHI wo000 550000000005 0N8000n 316

CAPEMSIS.. +006. eee eee esse cee 316, 323

(EINE gagocaoooSUodandG DOO uDN LS 286

GORICTC OL EEE erro eri teense 294

COV UmeEllafer Citrate lereteteiai ciaiepey nel 297

UPR a oot e DORE SONA aa OOO COS 296

IIRGDWIG Wogacacdoceonodsaooosuee 298

AWiNsgoaouoacouuoadecaobcodouN 315

KGLRTM MB ebooo gonna osud loo ROOMS 316

leucoderma.........-. 316, 317, 321, 325

leucoderma var. usambarensis...... 317

Me limishey: sartoewrtteasiierte ketal 302, 323

(UENIe@ONTG ao Goaneodadsoouasosce 293

jokbI oA Sogo soc ooKK oS 315, 316, 323

G@iibyee DIA MAL CLUL CIT D rela atotavaessiche vale lei=isie sere 6: «2 307

SOTO NE-UUNGONTS yaa edavelel aie es lets ee te 295

UENO ino oaeoo UUs CO OBOE eEe 317

WOMUFNI sc goonnaccouobosoeeone 317

UREA AONE oooooce coonoenHt 317, 325

(CHEED CH Daa HAC OeS an anit ck atin scr crac mean 68

Jonicerifolius; CAG Sm-ascee ee eos s+ a. 243

GMeKctiznoalenas Whocapaos poe enon sends 243

simulans CavAnpsmithiey- eisai alee 243

unguiformifolius, C. A. Smith......... 243

SM (QVILLIS SD’) lstezatetetstogetenniclersetorsuctotsaic 317, 321

Citrullus naudinianus Gos, Hooks fencer 256

vulgaris, Schrad. . .....-206, 420; 456

(Otic Gooeth mecca eos 378, 379, 387, 456, 461, 469

guranpitoliaatOwesee cee ee since: 452

grandis, Osb.............362, 448, 451, 456

.353, 362, 387, 407,

435, 449, 451, 456

nobilis Lour. var. deliciosa, Swingle... 362

sinensis, Osb... .353, 358, 362, 365, 376, 378,

387, 390, 392, 396, 399, 409,

418, 422, 435, 449, 451, 456

limoniassOsbrseree eee

Ssocoocsuentecnouennocayonum 337, 350, 392

Cladonia pyxidata (L.) Hoffm............... 597

Cladophora, flagelliformis (Suhr.), Ktz........ 597

Clausena anisata (Willd) Hook f............. 240

Clausena inaequalis var. abyssinica, Engl..... 240

Clawicepshor sete i ei occ eee ni 348

OAS Pals eia te sorereesrexctenereyayarecests 350, 351, 460

BD eect aee es Be 468

Clemantius Klotsch. eee eee eee 513, 514

Clemanthus senensis, Klotsch.........542, 552, 553

CleomerdtandraspBurche see see enter ere oe 235

Clerodendron myricoides, R. Br............- 252

similewRearsoneeanee: ane. 252

ternatume Scbhinzees ae en. ae: 252

@hitortiaalis ey ee neren Cette een erate 68

Cluster seinestfesce che cece es Neto ees Soscene 594

Cluny tiandaphnoidesenjesemer ee terres tes oe) ie 509

pulchella var. obtusata Sond ?....... 242

Coceinia’ Rehmanni Cogn... .2. 2.25.0. 25... 256

sessilifolia (Sond.), Cogn............ 256

Coceulusvhirsutus)(L:)Dielss--8.4.5-.5-.--- 235

(OM Ritd Babin abo 666 bu cld teen en ens SHeT 443

@otteatanabicailiae.<re5 setters siccecoscreysare 390, 392, 457

TO DUS haar iow. ree veccuchene hr pcac faces 422,

COlcome rer ee ene. nai cicieeos 392, 443, 457

COMB R BIRA GHAR ao ccr oer se reaccecionertio ores 247

Womibret wm Waste sorceress ore 80

apiculatum Sond...............- 247

Baek tered helen ee 506

(CUA ORS bee a a ee 647

mossambicense (Klotzsch), Engl. 247

sen Aicraisasbaaeensodscn ec Loar 506

Slthoone peo oR aS CoO kore aee 231

COMMETEINACHAR eee. seetitn cae eee eee 229

Commelina Forskalaei, Vahl................ 229

Commicarpus fallacissimus, Heim........... 233

fallacissimus Heim forma

PILOSapELeIN ae. ater eel aa 233

plumbagineus (Cayv.), Stanley.. 232

Commiphora calciicola, Engl................ 241

cinerens Wingless etree es 24)

Marlothii, Englsics.. 2. 22 see 24)

INOMISR se aes Ee aon ere ere 241

pyzacanthoides, Engl........... 241

IWelwatschitieccc 50 elke orcas... 241

PAGE

Commiphoraisp asec enero 249

COMP OSTDPADB A Rerytacietele ates) helentsyoeeciehs 257, 606

CONVOLVAUIZA CHAIR teases aril. e 251

(Caray zi Ibe Sapper oriole d Sto cine fiero oe 7

Conyza neriifolia, UHer........---..-+---- 207

Copatteraymopane; Karkeemcr petite ce 237

Corallocarpus sphaerocarpus var. scaberrimus

Cope sce scsi cieilerues week chet «sete Neuss s Ne 256

Corallomyces aurantiicola.................. 337

(Chor gallncleiotny 0 Ge comtmnlae NOC EER oan o ae 457

Corchorus asplenifolius, Burch.............. 244.

KarkaiseNj Jn Batysien actacictucttcone 244

pongolensis, Burtt Davy and

Greenway ts Se teeter ee 244

@ordiafovalis@Rae Bras aee eerie 251)

Cornflower: cnsacieciassthec® sce aeeemoeaairs 457

(Cia iey bese geud Ge clue oor oo oman onae 380, 461

Corylus’ cape papecsi testes rctes Fo So cpus een ee uacke 343

Cottoms (2. Bipenaenpaus oso eeia sae eee vee 457, 460

Cotularaustralissibookee- er peeeriee ceca: 594, 606

Cotyledontibinn eran eect: 613

VO elias Seon aaa Ee Oeo cep cme 639

alternamsieiats <i scamcb ger eke 622

IB OMUSU ssczisajdhepsiees sco De aes Soe 628

cacallioides: is sper creer 638

CACSPLLOSAr. 5) s1eie, saseuefevere ah eieeseersia = ks 648

CAI OP MVACe. Saccocadaascecc 628, 629

Clawatolias....: asey-tiecpusisere coaches 635, 637

CoO Perlinkiers=citd: oss cqseniacsieene 632

Cooperi var. immaculata......... 633

CIS taitiaicrscoc-s 7 eeptas ee renames 613, 635, 637

Cuneifonmis s-\rr-)-veckleoer sie 626

Gecussatawe wv). ssw siete 641

IDINGeRU SS paces ecgemiten erect ons 638

HiT GaSe) sites meee eres aie es aus 630

hemisphaerica. . .613, 622, 623, 626, 627,

628, 629, 630

lari Saeed ane oo oa colgnd SaoeE 639

jasmin onan ere tem ere 616, 628, 629

leucophiylalracm elena rer reir 647

MOGVAOMANIS 5 oocoonDE DSO DOD OOS 648

maculatus:...-.- 620, ene 627, 629, 632

mammillaris........ .613, 630, 631

MWaritanaesa\c).aeeproceic cee 639

Mamlothiiit. i. sracssers oscsiectie crus eis 630

MeEneoseomuaodondb aoG 622, 627, 639, 641

UOCLINDiccogdnascooucconCoooore 641

THOMIPILO Maer eee eee: 624, 625, 627

rhombifolia var. spathulata....... 642

MOHACTMKOEcocccceonocese 626, 628, 629

Schonlanditeeaeeeeererer eee: 637

COKICATIA saan ecole eit eae 641, 643

ADIN cogocdovon onc d0 0b akan OEMs (217)

WHALE oogkboac odohotel oc 642, 644

VEDUEI COSA sitar. lerrericrternr ere 632, 638

Wiallichiicsnve ae cieytraros ee 638

TAMANS A 5 o500 con 060 00d000 613, 635, 636

Courbonia camporum, G. and B.............. 235

Courbonia glauca (K1.), G. and B............ 235

CRASSUMACHAB Ma asmepe cerineicnae ere 236, 591

Crassula argyrophylla................... 647, 648

Comipactas Schon eerie titi 273

Cornyn nvikoses Wiha oouccocsne sune oc 273

Whettyaes aver et peterersteleys claieier-teve cies 273

mesembrianthemoides...........-- 638

Obvalllatayet swe = sie heeieee cities 641, 643

sessilicymula, Mogg............:273, 277

PrAch Sanit aleterre einer aera kee 638

Creonmectriard2 plod teint rete trae 344

Crinum buphanoides, Baker................ 230

Crotalariayatbhroophyllarer..a cet eter) 238

AUTOM no doonavosoonseuscagoos 238

inhabilis, Verdoorn............... 238

wine, Wescococcoscouccgse 362, 401, 457

Hlovaversindey, IBEW tins osco20000000000 238

Solanavat, WM tieoscoaococoduooCce 237

RR, cnoncocoononbudouonDGNOS 458

CHOWN oSoosv0csd cooscesanooSoDOODOOG SHOOK 242

Sapa, WIR 6 5scos0osecaeu500s 241

Groton Gubouga, S. Moore...........-.....- 241

Croton megalabotrys, Mull. Arg............. 241

pseudopulchellus, Pax............... 242

Cryptolepis obtusa, N. BH. Br................ 249

Gucumibers cies hess hae Sees 420, 458

Cucumis africanus, L.f. var. Zeyheri, Burtt

“SD anviy ice ol thats case oe peep pee 256

iniasoMe, SOM cooccssasccoaconsce 256

WMANDoGodgc0o bg UOoCADaK CO SDO USN ON 458

aan AMoe AD, INEGI cococcnpocanaes 256

SEunOMR, pocon se case dx 358, 362, 365, 458

CUCWRIBERAC HA aaa aeeinGee eer 256

Cucurbitalmaaxamiass Ochre itle erent ae 443, 458

Pepowerarcreecetetcerreiee 362, 443, 458

pepo var. verrucosa.............. 458

Cupressus lusitanica, Mill.................428, 458

macrocarpa, Hartw............... 594

(Chandar Galsjory, SelmivA 65 o0ccc0naocaeandas 232

Cyathula globulifera, Moq. . paRoBebo coos | a

Cyathula uncinulata (Sehiract ». Se nee Sveegoue 232

Cymbopogon elegans Spreng..............303, 321

excavatus (Hochst.) Stapf 225, 298, 321

plurinodis, Stapf. ..295, 296, 303, 321

Schoenanthus, Spreng ..298, 299, 320,

321

validus, Stapf. ex. Burtt Davy. 225

Sp iecc.) aE A ee anes 300, 322

Cymbosetaria sagittifolia (A. Rich), Schweick 226

Cynodonhyrrerrkin OC ceceeere reer 460

Dactylon, Pers... .227, 289, 290, 315, 371,

374, 594, 602, 603, 605

INcOmplebuss Neeser 290,

(OWAPIDIRACIBUNID; ccocaacooboo000GKCe 228, 596, 603

Cyperustalbostriatush..-cereeereiieeetiee 493

Cyperus congestus, Vahl... .0 cat escts een. 603

Cyperus sexangularis, Nees................. 228

tenellus:lisigeacesb eee 595, 602, 603

Supossoousodecdacenqoe Uc oot anole D oY

Cys PTESSidw:ioisrere w alcnarssaicusisce Cree OEIC: 458

Dactyloctenium aegyptium (L.) Richt..... 227, 290

Dictyloctenium aegyptiacum, Willd.......... 290,

Dactyloctenium geminatum, Hack.......... 290

Wahliapinnata, Caachi 458

Vaciabilis Deshi erick oxo ee 313

Sina nnnor oop actin diin GuaoM acocc 367, 401

so Daniping off? seer. «6c cin yee eee 465

Danthonia antarctica, Spreng............... 199

Danthonia leonina, Steud..............000.. 19]

Danthonia pPaApposay |e sees eee eee es 315

Danthoniopsis Dinteri(Pilger),C. EK. Hubbard 227

1D):740 0] Ig anonO BEB EccacciaCnaneoasacomo oc 458

Datura stramonium L......-............. 401, 458

Delphi WA eeissliitsrere ste sete eeekrene teehee) 459

Gi cchoadpindoo tocto.cumoqomb cule o 452

PAGE

Deschampsia antarctica, EZ. Desv............ 318

Caespitosa,, Be Baer eel 318

CaMV CIMA, sjzislaiarn epee oe tee 318

Desmonema cafirum, Miers................- 235

Dianthera burchelliana, Klotszch............ 235

Petersiana, Klotszch.........-..- 235.

Dianthus caryophyllus, L...353, 362, 367, 418, 432,

449, 459

Dicanthium papillosum, Stapf............ 298, 320

Dichrostachys glomerata (Forsk.), Hutch and

Wall Zaye ane ayecys scr tte 237

SPrreveiajeres sremsvenctecet si clesictereceeet est tere 232

Dicliptera clinopodia, Nees................. 255

Dicranoweisia antarctica (C.M.), Par....... 598, 599

Dicranoloma Harioti (C.M.), Par.......... 598, 599

Harioti (C.M.), Par., forma

TEIN, INONosocoocgosooeo0 599

imponens (Mont.), Par....... 598, 599

Dicranum aciphyllum, (H.f. and W.)......598, 599

Dicranum fuequanum: Wustss.ss-- 225+ sees oe 599

DSU GANTA Cy voters jajsies arsi axe) deme tee atepteperekten el steacuetteee 460

“* Digitaria—Some new species”’............ 147

Digitaria apiculata, Stent................ 147, 155

debelis (Desf.), Willd........... 226, 499

decumibens= Shentaee seers 150

ekianthasisteudmenenireaceiae 147, 152, 226,

371, 499

Digitaria eriantha var. stolonifera, Stapt...... 148

Digitamia foliosa, Stemi ai) relates 154

Peniculata | Steuben see erereeerer 155

glauca Stent 4.) /-ssitd sciitasocroe 151

glauca, Stent var. Bechuanica...... 152

Grantee bade rcneit amelie)

Inomienonceulisy, Will. cossenccg onc 151, 499

Wb uray ntoracracvsyaet Cis eases CORN ee 156

WrenOneNs, Meio caocsen0a oben nod 153

littoralis, Stent var. prostata....... 153

VANIEKMTEH TE, SEO sgancqeoacaaucons 226

monodactyla, Stapf............. 304, 372

natalensisppemscrsketae eee 152, Gs?

Pentzii, Stent..... 147, 150, 372, 374, 499

Pentzii, Stent var. minor.......... 148

iRolevansimstentemereeetreeeene 149

LigidaStentey erie ae ae eee 151

Scalarunta ChiOveneeimee nner 593

Setivalvansbenibpercrce a eieciaee ce 499

SHAMIM, SWIM so oocodono0conedn 153, 499

swazilandensis, Stent........... 150, 593

Penna tals ual purr elites 293, 322

trichopodia, Stent.............. 147, 155

VAMOS, 46 oo800 dooNoedCOGabEe CS 155

Validay wobemteererteintnccns dee siancystore 148

alida; stentavans claucans lise 149

Sy SHEEHAN S dice Ob aaa 293, 322, 593

SDikkop:”?\.\ccupeeseiers-seeeorretsnsatoronele etree 174

Dimorphotheca aurantiaca, D.C.......... 401, 459

DIOSCORBACHABR EERE rire as 230

Dioscorea cotinifolia, Kunth................ 230

Dioscorea malifolia, Bak................--- 230

Dipeadi glaucum (Burch), Baker............ 229

SP ereveus elele sare osteneneL Mn tse ene een rena 229

Dipterosiphoniais pari eeeer ener 597

Discofusarium tasmantense.....2.0+.+------ 376

Disease Bakanae’. o...:.'. saeco ceoeetaae 396

Kromnekis: -. aeecte teteie olole oie 463

ING ar Walt. 9 .ce.c MRR ee ees ecors 427

ee a

PAGE

Disease Pokkah-boeng...................-- 396

Dt teOhNESEVVMUGs -.. 2c) ndeeshenete eels se 427

Ditrichum conicum), Mont. -c-4..25+6.- 2... « 599

Dixon, JH. N. 5 Mosses72 2% faci. eR No ke. 598

WMGassansia desc «gas elec Sen 284, 285

peg cabcherie yet sateteter te Aelesl teers « 592

Doidge, E. M. “‘ Some South African Fusaria”’ 331

Doidge, E. M., “South African Rust Fungi

ME eaeinnrccaysyeyarescrersine ae arcisie wate eine Sam 487

Dolichos Schlechtvri (Harms), Burtt Davy 240

rao kesrse ayeqate A otevoe ere arte aes eney he aie ee eee 459

Drepanocladus uncinatus (Hedw.), Warnst... 601

“* Drying Cabinet for the Preparation of Plant

Specimens for the Herbarium” ........ so Ussz

Dryopteris aquilina, C. Chr............... 595, 602

Dryopteris elongata (Sw.), Sim non O. Ktze... 225

Dryopteris Pentheri (Krass.), C. Christensen. 225

thelypteris (L.), A. GBS a: we 225

Duinen\aaiboschiy- +e eee onasase 21

Dunn’s finger g Museo paaoONee ooe Ghen 005 OMB SO 593

Dyer, R. A., “ The Flora of Tristan Da Cunha :

H.M.S. Carlisle Expedition, 1937”........ 589

Dyschoriste Fischeri, Lindau............... 254

TBESHINVA'C HAAR, ayes arm ate evaysh cf raerens see esa 248

Hehinochloa colona, Link....:.......::.. 226, 292

Crus-galli, Beauv........ 292, 322, 324

Holubii, Stapf...+...... 291, 292, 322

pyramidalis, Hitche and Chase.. 593

stagnina (Retz.), Beauv......... 226

SP ieileisas ee sierra 306, 322, 593

chinothamnus, Wnely.......20e-+- 2222-2. -OL3, OLE

SD a ei kere fn cucyac Slot eetars tea eee 353, 396, 469

Mhirarta SPs to layos Aas ke 316, 322

IBA Te tial, Dict. ovr vwiorcs svar cio archon eee ee ee as 67, 247

hottentotta ue... CoS ae Ls 643

rigiday(chbs))Drucessss. +. -ese. - 252, 643

iBkebergial Meyertsbresli eee sees ee 241

Elaphoglossum laurifolium, Moore......... 601, 602

succisifolium, Moore.......595, 602

TE LALEN OMY CES ol Ao eens sealokss oso PETROS oe 294

OLA CCORR NSE S HSA SRRSTERNRETS 5 =. 294

Elephantorrhiza Burkei, Benth............. 237

Eleusine indica, Gaertn......... 348, 362, 396, 460

Hlictotrichon, Bess. ex Andrz...............-- 185

semvpervirens, Bess...........-- 185

Elionurus argenteus, Nees................ 288, 322

DIBA ASoeS 6 SURO O SCO OMLO ER OO na Tee 443

Elyonurus, Humb and Bonpl. (Anatomical)... 265

Humb. and Bonpl. (Systematic)... 259

Humb and Bonpl., “The Genus

in SouthyAtiriea 7gritse cei. = +. 259

argenteus, Nees.259, 260, 261, 262, 263,

266, 267, 269, 288

Elyonurus argenteus, Nees ex. Fl. Cap. p.p..260, 261

argenteus, Nees var. thymiodora,

Stapie Sci. oct Pierre eins =< 260

Elyonurus glaber, Phillips............ 259, 266, 267

glaber, Phillips var. villosus 259, 261, 262,

Jeni paoneee road oC oon Ole 263, 269

pretoriensis, Phillips...... 259, 262, 265,

266, 267, 269

Elyonurus thimiodorus, Nees..............-- 260

BHD BEE A @BrAWN 3, ote - eepsieae els Phaisyahease: << 605

Empetrum rubrum, Vahl........ 591, 594, 595, 605

Empogona Kirkii, Hook f. var. australis,

Bch elek ap cists siecle eyo Seem OR Ioeiale os. 255

PAGE

Endostemon ocimoides, Bremekamp.......... 252

Endostemon tereticaulis (Poir), Ashby....... 252

Hind othilas pisiai<*rraeente sce ook ok toe sie 294

Enneapogon brachystachyus, Stapf.......301, 322

cenchroides (Licht), Hubbard... 227

Pretoriensis, Stent............. 227

SPir ghee Shae ah 300, 322

Entandophragma caudatum, Sprague........ 241

Hinteromorphalspsereerere ecient ee 597

Hin py lomarreserersncttee oir ricterer: 284, 285, 287, 313

aus bral Oeertacceiemens bf aeuabalessnis 313, 325

EBESSEY UN Gace ENG) salar ete ehnel 313

Bid Onbistsnsccenac cei derek 313, 320

OOS AVM Sonoeaoso500000000C 317, 321

compositarum........... ASoma.0067 286

Walhiaenyrs cose rystx onesie eeu 313

Oleandractrr ocr cece ote oer, BPM

ETY SAMUS aerate rere eee 313

ZADDIAC MF iota PS RE TE BiB}, BH7/

Enumeration of Plants Collected in the Nor-

Cherm bransya alla. ae apiersiseiee sor ieee iris 223

iWpaltesialatassSteetzar seats ceieriaets 257

Epichloé Zahlbruckneriana.......... 341, 343, 468

Epymenia obtusa (Grev), Ktz.?............ 597

Eragrostis aspera (Jacq.), Nees....... 228, 311, 322

challcanthaa Enines ere 499

chloromelas) Steude eee eae 228

cilianensishinkess pee eee 228

cunvtla Neestery pitt ociecces e 228

gummufiias Nees: ee rere. 228

happula, Nees var. divaricata, Stapf 500

Dara hey. cet-iersic Seep eee eyes 343

Superba gheynen- cele emcee 500

Erianthus saccharoides, Michx............289, 322

BRTCAC HA ror seeoevsl: sien of eeyre ee:

Hriosema cajanoides, Benth................. 240

Eriosema psoraleoides (Lam.), Don.......... 240

iriospermumilabifoliime ee eeeeaee eee 317, 322

elinysibeyjoetedasy... omtcjam v- - areeeeeieee = 311

MUUY DUS ratetor ysis s\n cheyeh eek Ieee PPO 292

OCCU appeceta suis <i sfepee aN ext Sep eeree eee 312

OVA CCO = sora 2is Mis ee eee 294

POMUCOTUIN esses pM ERSTE ROG 314

VEL AV AVENGE =) --..¥ pape ter AAO 290

Cer G OT ACUs..;.5) 5 e A EOE 288

AHS MO CREED ERG Sa clo eons o Oe ee 291

Bra hr Ocarpuss NOC. iene eee. 513

SL IUIc( oll soy rt Hee OaR Ree mione a Pools clo oc colo Ceo 51

Bucalyptus cornuta, Labill................. 594

diversicolor, F. Muell........... 594

gomphocephala DC............. 594

iehinanniMEreissseeeeeeeeicdee eric 594

Bclea, Murry. ccreregeryeyss isco eo eee 67

divanorumeshiern eee eee ere. 248

GQuerket, Fiiern:azraeas cyrus ace eieieis ov 248

lanceolata, E. Mey ex. Drege?....... 248

mul pitlonassHoemMeeren erent ere 248

Eucomis punctata, L’Herit............... 290, 322

Eupatorium africanum, O. and H........... 257

HUPRHORBEACHAB cetaercitiseiserei ice a. 241

Euphorbia aeruginosa, Schweick............ 242

DECANE oanas sagnoodaodGUad 641, 643

Cooperi, N. E. Br..242, 250, 253, 254, 255

erassipes, Marloth.350, 362, 387, 396, 459

Gurrichianass baxter ais = 242

PAGE

ID Epo Nord omer CMA 6 coco aK eoedu CT CEs Fes 647, 648

Tirucalli. Lisette cvnnahontoncuck ass 242

transvaalensiss Schiltheses seem 242

LUN Caitaiery==ysyevertis cat wowees reesictee pers = 648

Burhynchium crassicostatum Dix ined... .598, 601

Bustachys)paspaloid esmeretiastiyereeatet teers 641

[WOUEEELa s coop pp boUD UDO DS CUOOSOOS 641

Kustichia longirostris (Brid.), C. M.......... 600

Ragen, GapEMSs, INMDo5o000000000000000004 240

PAIMSAcecnoaododooo0coDOaGGdDO COD DSbUdO 294.

RTC ASO CS ERO OC CORT 294.

OLIVA CE Actes clrerdiom tone evansucesc aie Uapeemeer: 294, 321

ING FEE ces SES OR ate oe Mara sree acco Ui enG 601

IE CUS oy xsre nyeveceleus istsicoeue oa aoe eens terme owe weneat oe 341

Capensis, bb yr jlearsreCe ere ie 230,

Pretoriae, burtin aviyia-iierierrrer 230, 231

Siaansi, Weck socccecoscascs0eda 230

SoldaniellalyVing neste aenl ee reels 231

onlay, WiG|sccoaonasecKodd0cca0dn0ae 231

My Cormo wy Wesssscouscacacoccccsaccue 231

SVGOMAC BE! Ibs HOMME Poococccesaucnone 231

IL LALG DiS ama ee eee s Siro 6 sic aca eM 225

Fimbristylis complanata, Link.............. 229

Ghydmwlla, Welllnoccocndcugousdo 229

Fingerhuthia africana, Lehm......... 322, 291, 322

IDA COWAR AMV CIIVNID., Goocdbndonobecesaeos 246

loa, Ont Wasa Cle, Cine, 55 c0caceeunb0000 589

Fluggea microcarpa, Blume................. 488

Fluggea virosa (Roxb.), Baill............. 241, 488

Fockea angustifolia, IK, Oli Pookoeacocooos 251

Forbes, H. M. L., “‘ The Genus Psoralea L.” 3

IUHAEPIEE SP oonooe6poubo son spon OoGe.d6 401, 459

Iie OholOl Weer iePecRe ee c-o1' boro eos ciaiatechs GOO Ga 343

AS Hreckle?2as Sen tie ee ere ree eho 465

Hreesia metrachanelelathmeneeer erie tinee 414, 459

SPrcrieneteatcaey ac tetemenen ne metceetner etree er re 418

Fuirena chlorocarpa, Ridley.............. 228, 494

coerulescens, Steud...............- 494.

pubescens (Lam.), Kunth......... 493, 494

uumibel laitian.\..ccueeeetete ee ke 494

RMON Gileecoleayets aaieetsi Oe eee et Ee eee 597

Fusaria, Some South African............... 331

IRWR Ey ANE HI OAK so Gaddouosoudacebongouens 386

QCTUAVOTUN 18h Ae a 447

ACULNUNATUN UA eae 366

GEFUGINOSUM ae 4 er eee ae 453

albido-violaceum:......eiee seen 405

QL eT OdUS :ccrsrstans st tek tT 360

allii-sativiannateeenee eee 447

allaviale BAe vrei ic Oe ee as 447

angustum...402, 407, 408, 455, 456, 457,

460, 462, 465

pT UWOMouesbso6000000000bD0¢ 349

CARON Op COOOL OOG.O.DOO.C DOMAINS 349

ANCUGLUWMN Vi. MAJUS........-.e 8s. 349

CAKE NOM NM CODUCid FoOeuboo concur 366

RO AOMORI GIS. ST OG abies peeks 376

asclerotiwum...c/tdhe eo eee 405

ABPUNAGUR., cists <tots rnareelae rete 453

CULOMILACUTUs elena eh eee tte 428

AVEUACCUMA as os lee siete lente 548, 460, 468

avenaceum f. 1...349, 350, 456, 459, 460,

462, 468

UACCHARICICOLC ite onan 335, 338

DUPOTINE ra ycwaysns, Sy feta oP AC NES ses 348

BOSELY COLES 5 crarejercie anette eels 401

ioe)

PAGE

IRB abr D7 [O11 COL ah eee en eee 383

bulbigenum...... 403, 412, 413, 454, 459,

460, 463

DOD CRUD His Woccaccnssneccsoque 416

bulbigenum y. batatas............ 402

bulbigenum y. blasticola.......... 402

bulbigenum vy. lycopersici. .402, 416, £417,

456, 457, 459,

461, 462

bulbigenum v. niveum..... 403, 419, 456

bulbigenum v. tracheiphilum...... 402

beblatwmis Ge Ae LER NE MSs 359

bullatum v.brevius.. 1.2.52... 359

Oullatum Veminussen cee eee 359

bullatum v. roseo-bullatum......... 359

bullatund ve TOSscunise nee een 359

COLCORCUNUA seed ku Maa eee banners 428

CHNOS POWs 00 00c0oce00sancsn6ce 335

Candidulumeree eee eeeeeec ee 421

COTICISINSis meen attic aye ee 382

catalepiiinenan Corr tee oem tee 335

caudatum vw. solant......+,-.....5 368

Celosigenans eee te ear er 304

CYMsssonsgoocswoo baad en onc 425, 438

UAV TOCOM WH s.000500500060000006 360

chlamydosporum.......... 345, 346, 460

ciliatum- tyra. haw nels 343, 344, 468

ciliatum Vv. magjus.........-...--- 343

Cutirullaeiceta tae eee aero ieee 419

CORMAN a ocacapongaoocuc80G 335, 337

COCO MMT csoccascoes0s 335, 336, 469

COMMWETIMN coo mocoo0ss00oe 443, 453, 466

COMMMULMUTOn gacdcocceoco00gDa0gGe 447

COMAMAINEAS. 5 oo500000n000G00000 402

conglutinans v. betae...........-.. 402

onglutinans v. callistephi.402, 410, 27/7

455

conglutinans vy. citrinum.......... 402

conglutinans V. Majus............. 410

CONG OCWSC a ceer arene tera teutede rey Vo ct ore 371

congoense Vv. septatius...........-- 371

Cordage... Bet Ae ee Ses 356

CHOMYOPRUMOTOM: aeiereltei testicle 412

culmorum...371, 380, 387, 460, 467, 468

Culmorum spelen. Memeo crore 380

CUUMOTUTTD TY, LLCS eee eee 380

COMMOTTNTO Vi HHLPUD: 0960 000200005¢ 380

decemeellulare........ 339, 340, 457, 468

Delacnrovaiv partir. naar tsi 376

dianthis-eceereerrreer 404, 431, 432, 459

dimerum y. pusillum............. 456

Aiscolon yc vane aise ele y= si 376

AUSCOLOTIVG: TUG US eel ele eis) = 383

discolor vy. sulphwreum........++.. 380

discolor v. triseptatum..........-.- 376

CINE TOMUNMN sc oo acagucguddtads 456

OffUSUN ns) ee Hea hhh cic Me ASRS 348

CleGan'sh:, sd ANDO eee 421

ClONG CIT eee 343, 428

equiseti......356, 357, 457, 458, 462, 467

COSA S Weslo goande adceauendoy 356

equiseti v. bullatum....... 356, 359, 462

equiseli v. bullatum f.1..........- 359

equiseti v. bullatum f. 2........--- 359

CQUUSCLUCOLA:.....a. « Hetsnei heat. steteiae 368

Cniisate Wii bgeeee cs seccoecauoouus 412

——_--

INAS ORIEN ARULCTCVOSOe Soa 6 Cab HOS homed oeboe 366

CUOLY SPOTUM revere nee aaa foie) 01 < 423

LUCA TUN scareseiohe, ici) aucgscaisccacysaeasis eto ons 356

falcatum v. fuscum............... 356

VOTUUGUNOSWIM orate = juin pc¥ei Kein 2 = 366

HALA ISCRS 38 ab oor bale oS Ono 368

LLUS DOTY otek poh cy aiiak eases ankeh Booker che 5/6 343

mie WITME SB ecdea on oeeadodeenoen 383

RRM Wee SAGO Ce COE OOOO Anon 376

i FUOMYADUDG 2b.66,08 BOGOR O OOO Moda 386

fructigenum v. majus f.1.......... 391

CINMAVT en cancedes sooesesdacedod 380

GLO VOSUNU scrote Been See eR KK: eheile 360

ULM MIAN DES ea ea DOP OSO OO. aoe 443

graminearum......... 371, 382, 383, 468

Graminearum V. CATICIS.....+.-..4- 382

HIN Be oD ORO OO OS TOO OR ORE 376

GU MOTU ars tosciniolttafaias sieke:avsto seins os 383

Retdelbenqen se craantapeser yikes 380

RCTOOT UTE Ss sretors olefayey a sysqetex® pais os 348

PTD Te Wecubotecoss dootoTone 348

her Domania Aaate aes eee apes ears 348

herbarum v. avenaceum...........- 348

herbarum vy. conii-maculati......... 376

herbarum v. gibberelloides.......... 348

herbarum v. graminum...........-. 348

herbarum v. pirinum...........+++- 348

herbarum v. tubercularioides....... 348

POAT Vin CHUCO sa concen sanded 348

heterosporum y. congoense.371, 372, 460,

467

heterosporum v. congoense f.1...... 371

heterosporum f. paspali............ 348

| ITE Set Se SANG CIEE 443

PUP POCUSLONT «5 \-\«rjnsere «ep l= 366

MOTUCUN meaty Naa ere eit ck ans ines 376

UILCOMMALUNTU coerce eyes, aRAE EE eee 353

UNSUOLOSUT eats Lana ter eis eyele 2 383

ICMAT Shee oe beaalS 443, 444, 458

javanicum vy. radicicola.443, 446, 464, 466

javanicum v. theobromae..........- 443

UQUIMALTUUWIN ate SRS eee: 353

VAGUS sey erat ares clpoerettarde isiste sls 466

LEM CCOLBI UN tare ni ann Sac ouete Wain nao 366

lateritium.385, 386, 456, 457, 459, 465, 469

VRE r ab TUT Wore recor: Ferctoy Naps aay ng tov st ove) 386

lateritium v. fructigenum.........- 386

lateritium v. fructigenum f.1....... 386

lateritium v. fructigenum f.2....... 386

lateritium v. longum....... 389, 456, 457

lateritium v. longum f.1........... 389

lateritiwm v. pallens.............. 386

laterttvum Vv. tenue.....,-<..-.0-.- 386

(LETT eR ee oO tile eoiscue Goo Ine 412

UM ONUS cic oye Ie Wace OE ee 386

et Aeon ae Lit. 402

LONCHECETASersie rok re crores RPA hee 412

loncheras v. microsporon........... 412

LONGLSPOKUTTE TAA eo OS Ae 389

WUC LATS ay2131 Ss ONTO OS 348

Latilattiniy ctor pbs ee tied elles 436

Wf CODERELCH I Area aie Re ore 416

VEC. <. loots eee Pena skate Bhs 447

TUL ACCOT UAT Hee oe c cB Ra es & 6 as 432

MOTUS OTIS ere ear oho =o = 450

Mariage Virdee ys. Geet Se ese 8 450

PAGE

IGEN TOPs con oconnosdeeKospnOoDE ae oO

MUAACHTOWN sea eas aere eae ene 348

metachrowm Vv. Minus............- 348

MWONGHCMM Dos cascade oods. botoGr 383

moniliforme.353, 393, 394, 395, 454, 455,

457, 459, 460, 462, 463, 464,

465, 466, 467, 468, 469

moniliforme V. erwmpuns......... 394

HUOPANOULE Vo IIt2l00 cod 16000050 OC 394

moniliforme V. MAjUS............. 394

moniliforme v. subglutinans.333, 398, 399,

454, 455, 457,

461, 465, 466,

468

NIVULCT OT LULU Tet etntat shel tele eho sickerap-ety=ts 356

MY COP UII a roe eho nye ee che Sie(ns) == 421

TCH Wiles scaoeesoaesedacooad 368

MY OSOULAUBE ssc eles) ave) seks ispeest ofa !s1 ss 421

MOOI EAD NSHARA6 oan ganado soodao6s 335

nectria-palmicolae................ 359

TEGLECEUTIU Sarre oleh 4c eystesaveinienexayelens (aes. 380

WUC OWMANAE iso nel AERO neh oy eecs esse 424

NUVOWIMLS «ic oy sreye Myo enesaletel ais teleysie, eas 419

ODUNUCOTUINY eters epsoeexey fen yeceree 421

OnbhOCera Skene een ee 402, 405, 466

orthoceras v. albido-violaceum...... 405

OLuhocerashvs pls ssc eee er 402

orthoceras'v. api fo 1-3-5: -.-..--- 402

orthoceras v. longius............. 402

OTbhOceras va PISie setae tell 402

orthoceras v. triseptatum........... 405

COMCMM a Argatice cuce Fagot bo 4 356

osteophilianen.. «cee oe 358

oxysporum. .404, £27, 456, 457, 461, 466

Oxy SPOUUMetee ly citrereys) ey 404, 423, 466

OXVSPOLUM Ss 2s elsewhere tote 404

OLY SPOTUIMU fin Deratae ela = eee = oho 424

OxySPOLUMGl. Oven ue el 404, 425, 426

Oxysporumi ft. 7--piie ae 2-2 = 404, 425, 454

GxySPOLuMyt | Oeeeeeeeeerier 404, 427, 465

oxysporum v. asclerotium.........- 405

oxysporum vy. aurantiacum.404, 428, £429,

455, 458, 464

oxysporum y. aurantiacum f. 1.403, 428,

429

oxysporum v. aurantiacum hyalina. 353

oxysporum subsp. aurantiacum..... 353

oxysporum vy. cubense............ 404

oxysporum Vv. cucuroitacearum..... 405

oxysporum vy. gladioli..404, 430, 431, 460

oxysporum f. lycopersict........... 416

oxysporum subsp. lycopersici....... 416

oxysporum v. lycopersici.......... 416

oxysporum v. medicaginis......... 404

oxysporum y. nicotianae. . .404, £24, 463

oxysporum Vv. resupinatum. ch DSeOT 405

Pallido-TOScUninenme se tetetNale ---- 353

PUNMOSUM 4 4a) t- Ate oe Noten S 376

WOU USUUICUIT moretore) <a ecis oot oie ors 343

PUSDAU A aA See oe sie eco aoe 348

I HOO Wee Boo nocnon Oo CCOSHGADO OOO 428

PElWgen@: «Ae ck Ate k ASSeiles ows 343

CSS tN SN TMS os RR 9 ie 450

MOH Ssacccac sano coURooURGUsIeS 345

IROOLenSIS eee yer eee eri: 419

pscudoeitusmmeys cere eee = or 366

PAGE

ISECEYAUTNEN, (HOH OMOM OH Da cocdocaacasacasucanne 376

PADUCUC OU om ccna kaleseuahege ORO RER NOS 446

TEGOLENS 22 ccyofawenesep Ge acd ICE eee 403

medolensihadice terete 403, 440, 441, 462

redolens v. angustius............-- 423

Rebicwulabuteye ep ysoreseshee cess oreo 458

MRIZOCRROMOAUSLES eee ciate 412

rhizochromatistes v. microsclerotium. 412

TRGOUCOULUT emaecs niero tedepsisdeteenene teers es 383

CHD RATION A ROE RRO To OS ORS IEC 376

MOSEO“ DILL eRe eee ere 359

MOSCUIMN «page esiate olshehe ecko erence 376, 383

roseum v. calystegiae.............. 353

roseum v. cucubali-baccifert........ 383

TOSQUD Wo HOTU ocoosanccsca0ece 383

FORGO Ta HitDonocoddcnssoonboooe 380

roseum v. solani nigri............. 360

HOU MELON Da er ctsectb oe GG GO OO Dae 380

FUMGMMOSUHOs0c00000000000000000C 380

TUSSLANULT EEC econ 366

SCOOT Da snacceocceccopoune 428

sambucinum..... 371, 375, 457, 462, 469

sambucinum f.2...... 370, 378, 457, 469

KAMA LEMME My Bosovoposncoccoace 380

Ra oOnaoven 1 Oaoacaccorte 371, 380, 469

sambucinum v. medium........... 376

SLINOCNSE ahiaen aun lem eR TERR Poids os 394

SONGUUN CUT ee i eee Reta Tao) 1--N 366

SACOM FMAG Os doacossec0aeaececda 349

SOUPIMMEB Gs o0as006 585600500 b00 Cc 380

.356, 360, 367, 454, 455, 456, 457,

458, 459, 461, 462, 463, 465, 466,

468, 469

Se POON also Aco. coo bs gaCaOU eee 360

scirpi v. acuminatum.356, 366, 367, 459,

466, 468, 469

COMME. fe rat ereNeReRO Ore sae 360

compactum . .356, 364, 455, 456,

457, 458, 461, 462,

scirpi..

SCUrpt V.

scirpi v.

463, 467

scinpt Vv. compactum fi. 1... .5--.-.- 364

scirpi v. filiferum.356, 368, 369, 454, 469

SOU Vio TOUGMHOS: oon ocboabounbee 360

SCUNPU Vi. NiQrantumMm.-)....-.++-.--- 360

ROOHOD Fo (HUTS. 05000000000000000 360

SCLETODETIN AUS ltteer eeetote aterele 360

sclerodermatis v. lycoperdonis...... 360

RUARONIROMMOMODS oanGa.no 0030000000 407

SCLENOLTOLACS sejereya1~)ahel eeretehorenteee ee) < 428

SGLENOUALIN Stason os 21 EERO oc 360

SCOLECOLMES :2\2s050iaious\etatseahe RR IRO a= 343

semitectum v. majus..352, 354, 456, 457,

459, 463, 467, 469

solani. .443, 447, 448, 454, 456, 459, 460,

466, 468

SOLAN MU eteel sis wale isetneeaeIe es 447

eM Wx UMM Do coucakicsgxaueen 447

solani v. Martii f. 1...443, 450, 457, 454,

456, 457, 458, 459,

461, 464, 465, 466,

469

FObyoiaye Wikhanbineh oanaccucu0oood 464.

BOLAT Vic MCAUUT ales vietelaielnerenyete iets 447

SOLAN Ve SUffUSCWM. ©... see ness ne 447

BORG TNs cs cKerazercustatornla iste HAH E ost Ne 348

spicariae-colorantis.........+++04- 339

PAGE

SUICLOUDES Nei sie se OES 383

stilboidess.4\..)..cmeen eer 391, 456, 457

StUiLbOTdeS V;. MINUS.» 1.20 eee e ss see 391

SUNCOM MANDocascsdnocneboseatcues 376

subpallidum v. roseum............ 378

SUDULALWIN ake tye RT eee oe 348

subul tum v. brevius..+.....+-....- 348

SMMC 0 dcp occ ecosccGo00RS 380

[QHOACIUOTILIN eee eee 424

tenella: cies nace COL ee 376

LENWISSUNUTUR Ieee cee 376

LErrestrisins Gains wo creel enOe 359

theobromucsenee one ror rene 339, 443

UMNOMMerobaccdscddhéooovebscoade 421

UUNCALUTV. wave ates aiaysavonsnccopesets omer 348

vasinfectum 403, 419, 432, 433, 460, 461

Wetter tt, Inn gonsancooconco08 403

vasinfectum f. 2..403, 434, 435, 456, 457,

462

vasinfectum vy. lutulatum..403, 436, 456,

461, 465

VARMA OIE V7 JR sooagucedsooee 427

vasinfectum v. zonatum.......... 403

vasinfectum v. zonatum f. 1 403, 238, 462

vasinfectum vy. zonatum f. 2 403, 439, 454

VET SUCOLO ix striae | Seer ree ee 380

ULOLE COUN areata: 376, 453

VUNUME Rs .oer0 lcs eRe en LTS peuseeConERe ESRC 447

LURE OOOO OTD COCO OPO Oa SOO OOM 348

onawiorin Wesankoacaasdbadsooods 438

ZOWALUII Rio latencies es) cr ci nronse rete teneter eke 439

MUENCHEN GHCBIs5.0ccaccscoscnehooaucoDe 366, 367

NOSCUMM pret Moatt cay ay aray cre ise toasty crcl aeae 383

MUSISPOLUUMAVENACEUTM ae « -e)-\-)eh-)e is cl 1220-1 348

ChENOpPOUiNUM........---+-+++--: 360

CORMAN Ds0c0o0e0conr0e006 335, 337

CULIN OTA eced ee orNerseie rere 380

SLU NEON aco oo CORO ea eGOS an 343

OCA n 365000000 000000008 353

incarnatum v. tussilago-farfarae. 368

MUCHO 0s occa00ecab0G00000 368

OSSICOLG neni er rare 356

MAllido-FOSCUT state trae 353

IS chiedenmaychineen eerie 349

GOLAN. \, cole eastey vouchers, sferensis eu nears 447

GMO Po [On co coocnacacanes 447

SOLGNT (UDOT OS Ueeitat veraetrerrenetiet 447

IMEC) MAPUTO cosacccconasacenoeasacaes 368

helminthosporvumern tet eer 360

PAL dU: cSvcrp ey yawoe eee © ke 356

UU PacaocnadoscocanoooDpDORooodDe 380

Dna Mpmeo soo genes obada@oadn bal oon GUNG 68

Garcinia Livingstonei, And................. 246

Gardenia Neuberia, HE. and Z............... 255

spatulifolia, Stapfand Hutch...... 255

Gasteriaycaminataere attr en ett eae holon 623

Be Excel lob aidan) “76.4.¢ cca uuncugoanocUpoovOnOD 159

Geigeria aspera Harv. forma?.............. 258

= Gemsbokkomkommeri? 72 e)sqei-15 ee eee ee 256

GHRANEA CEA israel iit asteist> Gime naere teri 240

Geranium jag is). Giese: bays ore sdenceaiatoiaa tne eyeter 459

Gibberella acumnimartar iri rtiteysislsieneenctteterole 367

LEK UC MRA oo Soon outs coco 387

Mujiktroil.y. .;: certo ater sere) 395

Fujikuroi v. subglutinans......... 399

PAGE

Gapberellanntricans=.- 2s... + sls -urs 359

TONAL OTTIVUS sclera achat elas} soko yee eres 395

jQUNCRT; cooooacaosooocn3aoLIeOS 376

Saubinetbitecs cote eee eta 383, 468

: Saubinetvi f£. dahliae.............. 367

Galrayruibras blellerseenss ne. eee 401, 449, 459

Giesekia pharnaceoides, L.................- 233

Gladiolusts pes scfrreeiocin yeu Takai 414, 431, 449, 460

CIGESINE YOncsoeouccbodcosvess 5000s 0000 353, 469

Clie pee, Wc ae nasesace one odopee 239

Gnaphalium luteo-album, L.............. 594, 606

Gnidia microcephala, Meisn—

(in error G., macrocephala)....... 490,

GooseiGrasss- n> se eee eee neeo 400905 460

(COG TONUEN. sognomoenneescaenegs ogo u0dDoED 243

Ehinceenouin, \WeitiooooooouEsceuoLe 245

par bad ense:tctenesrcvstae Siersieietevexsre os 433

her baceum Ws se cho cect oye tie ouey exe yess 433

Gossypium transvaalense, Watt...........-.- 245

Gossy, Plum Sp leis tess els ot pte oes 396, 409, 460

GRAMINEAE........225, 287, 398, 460, 461, 602

Soule, Avaateeidesancasnc co soo 147

Crapetcuity reer 407, 448, 449, 451, 460

(Graye@® Whose es Saeswocecnthosoncosee vues 460

Grass; (AM bELOPC yar<se; ci aysgs dicks: vauise hod Scsahe (okey -bye 593

TD Ewa os) Rhetoric eaten oy 5 alee eee ee etc 458

Dral DOK rat ie gecd ake Sateen sees 459

IDjrinvarisr IMiYEIRS Ooo easeaodoseoaoosoEe 593

GOOSC ses cacraclieieke eae 348, 396, 460

Alar Simaniae paceconaueconaodes 593

Napier: Hodd ernest tds tryst 593

Pongola: River Hanger.) <j. > ote 593

Swaziland yin gener ee eters etetsae trails 593

Ha Hakeh fe) ni prCne eee aS Cue coe EG 603

MNUSSO CK A. ne ops ioisrovorotare SZoteue ood le cenay sista, 603

Wioolliy SHing ete arr fac pe heletoiaic cities 148

(GTASSES sy eres tee: sictoren Ser alae te oS ve ede a heasuaie on aie 460

GTC VAS CALC seers occ ers ye: cis oroteguiG¥=:oynysesteiiernes: > 337, 469

Greviarlava sl Ciy cols & cack ht octets <'s 244

hexaM Tha eBuLns ok a ei eyes 244

occidentalligh (bs. x: .,<Ateisis. ce ee oheraa a 244

TELINEEVISs DUETS ac. aero eee 244

Schweiekerdtit, Burrs .s2. os seeene: 244

GURETDM RAB. avijariere citer ae eee 246

Gymnogongrus polycladus (Ktz.) J. Ag...... 597

Gymnogramma cheilanthoides, Sw........... 602

1S ILGr ence ae ater eeoT ES ORO nos cece pick 594

suligna, Kenreinb 2). sts..1rstetaeitat- als isisre 594

Harpagophytum Zeyheri, Decne?........... 253

Haworthia tessel lata tyctchrceertceita cae 641, 643

TeeriamVicion: sas cee see ete or see 45

Helichrysum Kraussi Sch. Bip.............. 257

Helicotrichum Bessiz-\c ns cee eee neaee 186

Helicotrichumt Nees.'s 53 22a ane sets 3 ot 186

Helictotrichon Bess. ex Schultes.......... 185, 186

barbatum (Nees),

Schweick........... 187, 190, 200

capense, Schweick 187, 198, 194, 200

Dodii (Stapf),Schweick 188, 197, 200

Galpinii, Schweick. . ..187, 192, 200

hirtulum (Steud),

Solves, fers here's 187, 193, 200

leoninum (Steud),

Schweick...........191, 187, 200

longifolium (Nees),

Schweick. .188, 194, 195, 196, 200

PAGE

Helictotrichon longum (Stapf),

SOivencoeeuccc 187, 189, 196, 200

namaquense,

Schweick...... 187, 189, 190, 200

natalense,

Schweick...... 187, 194, 195, 200

quinquesetum (Steud)

Schweick........... 187, 188, 200

* turgidulum (Stapf),

Schweick........... 188, 196, 200

EHeliotropiumreeee ee eee see ea ee eicier 226

Gunz pO, Wks obcckoonoood 252

lineare, C. H. Wright.......... 252

Nelsoni, C. H. Wright......... 252

Hel minthosportumusp: aera cece 348

HEMIBASIDIOMYCETES......-...-..... 283

Hemizygia canescens (Guerke), Ashby....... 252

SIGS OP HALOS rye weioraeve en eT ao RTAT Ol cKenE EO AIO OG 597

Hermannia boraginiflora, Hook............. 246

ORISA DSC RIN Zar e renee rete Lets 246

Holubii, Burtt Davy............ 246

iHermbsimedinan linea nyse sees sells atrial 232

Hermbstaedtia Rogersii, Burtt Davy........ 232

Heteropogon contortus (L.)

Roem et Schultz............. 297, 301, 322, 323

Heteropogon hirtus, Pers..........-...+.- 297, 323

lal@miiey, SOisivs sence ononooooso se ddpaNO0G 186

Hexalobus glabrescens, Hutch and Dalz..... 235

1a My ROU amene bod oae dda BOO Do po nn Goro ceer 341, 461

CONN HS scagascoogddesogcnagLs 461

Choo ois, ID; Chosodocosccdacdoce 245

OHOMIOMANG, soaoocanc 0070000 000c 433, 461

intermedius, A. Rich var. aristaeval-

Wik, Gli @eoe bea soo ee SbAoDoGONS 245

IOUT, Weosconssteatoogseoce 245

physaloides, G. and P.............. 245

physaloides, G. and P. forma?..... 245

praeteritus, R. A. Dyer............ 245

Baloo, Woo sgancondesnc0ce dr 433, 460

Schinziie Guet ke mene eer tte tear 245

FinddenlScalesertite ta ettocieiecr etre ter 469

lelgq Aone, ID. ENC! foo oocoococoosocoDDUE 4

EPPO OR AGEA CE Aen epriet irene 243

Hippocratea longipetiolata, Oliv............ 243

HOleustlanatusseliwaetresteeisctstisticesters 594, 596, 603

Hordeum asec tees teks ioe Coates rer 379

Hordeum vulgare, L................ 288, 316, 323

mb bardysquasheesaet tee rierce 443, 458, 461

Huernia Loeseneriana...................-. 647

zebrina, N. E. Br. var. magniflora,

hil ip steerer errr 251

HY DROCHARITACHAH.................. 225

Hydrocotyle asiatica, L.............. 247, 594, 605

capitata, Thouars....... 594, 595, 605

Tennolene, Wilo.sso¢cocodoso06 247

Hymenophyllum aeruginosum, Carm...... 595, 601

peltatum, Desv.......... 596, 601

Hiymenula equisett. 0. ede oe =: 412

Hyparrhenia aucta (Stapf), Stent......... 307, 323

eymbaria, Stapf............ 299, 323

hirta, Stapf... . 289, 323, 372, 374

Mey IYO cacsnouw eno 305, 320, 323

PAGE

Hyparrhenia Ruprechtii,

WOU. 5 o¢0¢ 225, 297, 298, 304, 323

Tamba, Anderss305, 307, 309, 320, 323

Mosooenqdoonoes00.0 297, 306, 323, 460

Ey Peni CUMIE eee cere rer tiel 571, 574

A Revision of the South African

SHITOS Oioanooochersecnovbce 571

aethiopicum, Thb...571, 572, 573, 575,

* 577, 578, 582

Hypericum aethiopicum, Thb. var. glaucescens

SoinGlind Gapeicnnmcmoeoes O06 572, 577

Hypericum humifusum, Linn............... 573

Lalandii, Choisy.246, 571, 572, 575, 582

Hypericum Lalandii, Choisy var. lanceclatum,

Kies os octtatit cra tata trecsis verso. o 572

Lalandii, Choisy var. lanceolata,

Sond)... saemie serge ere eres 572

Lalandii, Choisy var. latifolia,

Sito mouoeowoobe ooo sgamee sn 572

Lalandii, Choisy var. macropetala,

Sond stags eeeeeeehers cise 572

Hypericum Lalandii, Choisy var. transvaalense

Bredells ssc cten stone ate S722, a7), 327

Hypericum lanceolatum, Lam................ 580

Hypericum leucoptychoides, Steud.......... 506

leucoptychodes, Steud. ..571, 573, 575,

580, 582

Hypericum natalense, Wood and Evans...571, 572,

573, 575, 579, 582

natalense, W. and E. var.

petiolatum, Bredell....573, 575, 580,

582

Hypericum Quartinianum, Rich............. 582

Hypericum Roeperianum, Schimper..571, 573, 575,

582

SAMMI. og ganboosoosssa05edc 506

Sonderi, Bredell..... D2. DUDs Dl Ss O82

Sonderi, Bredell var. trans-

vaalense, Bredell....... 572, 575, 579

Wilmsii, Rob. Keller....571, 573, 575,

579, 582

Hypericum Woodii, eller eemeve neat 971, 573, 579

Hypnum cupressiforme (Hedw.), Brid........ 601

ligenyah puLChasicnerreeit- rire rio OLE OO

Ilysanthes capensis, Benth.................. 253

ilysanthes dubia) (i:)) Bernice -0eloleretle 253

INAS MTOES (PopeeU Piel, IBEW ponochGoAceesooececK 253

Imperata arundinacea, Cyrilli............289, 323

GylindiricaamB calvary neers 501

IniChein SONG) Cp ome gooUMEOabaco sb DO ee ce. 461

Imdicoterayadenoides, Bak. f......00-- en dene 238

Indigofera Baukeana, Vatke................ 238

Indigofera circinnata, Benth................ 238

erens NIE Breer nicer 238

aenpeEvarh IBGE. ob ooges acoobony EO 238

lakopiboris Ne Be dsins Séamegoosuddes 238

tettensis, Klotzsch............... 238

THN AH GO} Dl DYN Deep aie a cls alnaiG coo Siero Olay ots 205

TRISLOMISMPIMES oye ccfeie siasene* viciepricte rarer ere 594

““Intersepaline'Glands.. 0s. 0.) .ee esc ees 274

Ipomoea adenioides, Schinz................ 251

batatas; (vam. ave seth eeererarae 422, 461

Lugardi var. parviflora, Rendle.... 251

quinquefolia var. purpurea, Hall. f.. 251

LOE USSRABACMORCOEE Dc Cod ttt) .8 econ 461

PPA ae ciliat ve ikstZenaeccrestenetecutietatis cuiereicic 597

PAGE

Ibe SGM, WragocaconnosccooDcoooK0KS 492

KINO cqnossooHucoochudooeouce 492

Sasiduli-we-hlabituac. pci terte terete tert eter 218

NCPR ISE MWA. calooooocoonDO0bAd0000 513, 514, 534

TGV, OW gacccoccceabococscuce 534

Jasminum stenolobum, Rolfe............... 248

Jatropha erythropoda, Pax et Hoffm........ 242

Schiechteri/Paxerreee eerie 242

Aeynerin Sond se OnMmapeenEm rt 242

JUNCACH ATS. insect rer eee 229, 604

JUNCUS) DUOnTUS ss aT eee een ene 310

JACM IDG MOTD 56 o6bc00a0 595, 602. 604

Capensisssduh beerer trier aetna 316, 323

lomatophyllus, Spreng........ 229, 316, 323

Juncus: tenuistline eee eee eee 604

JuUNnCUS'Spi. Vetere eee 315, 316, 323

Jussiaea angustifolia, Lam................. 247

Musee wav, Walilscoooccodosdocaaeuadcac 255

odora, Vahle ieee eee eee 255

(Ansellita)ysprenmctimciciot: csc 255

(CalloalnenaonGles)) SPsoccacconoeaseces 255

Kathir Comic hereto: One ner eeeenen 380, 461

Kalanchoe multiflora, Schinz............... 236

PAANCMANEL, IELVAVoooccocas0asce 236, 647

pyzamidalis, Schonl.::-. 2. .--:.- 236

Kiara. i. aaj ste stainlot rs erereusermeret tenemos openers 67

Lei Camere me cnicen pceae Ae era nare OGIe TO be 67

Karri: ¢ sti tanemiecioe te tie eee oe reece 594

“Kat Rivers Walt tei seatcs ctpeta lecusneus cteuteriecs 463

£8 ely tigate carae Cunt scttertiaystt tueegcrs ate niante enti 593

Kien tialisprosanc airiren ricer reer 362, 461

KCramantivicmloo kel see eee iene 513, 54

‘ Kuirlehouttare comme ccm mrcrerr ciocreic et aoe 51

Kirkia pubescens, Burtt Davy.............. 241

IXiesiaven Gorell, ID{Chosonoscaccconsseoane 631

leyaerinore, ID(Coaceccomsccnccabedooc 258

radicans (Thunb.), Haw...... 631, 641, 643

Kniphofiaspy.rc- erecta me ieee 399, 461

Kolbia, Beaavica «nadtents eck ee ne tte al 513

SAGgommayalie GEA; ooo ocoopabaadanecooon 463

“SiKrrilorass' ?4 Vii, sented toeetee ere tetera tere: 228

Kyllinga melanosperma, Nees............ 228, 495

A BLATAN i. see eh eee ee er ie eeret 252

Lac ananehivnet: CCWOKUAD: 0059 ¢aba00c00G00CKe 447

Lachnopylis montana, ©. A. Smith.......... 249

Lagarosiphon muscoides forma longifolia,

Wrager:jccccmem pemegyvenr OLraotie te cerns 225

Lagenophora Commersonit, Cass...........595, 606

Lagenophora nudicaulis, Dus............. 595, 606

Wepaokol lola), CATIA Sco oopoc oo oncuoosnUanesD 647

Kirkii Dyer var. delagoense Dew. 249

Thanmeay ediulligh a. tcscusnaerm tee uev ares we TateboNs 647

darks purine eee eee eraer ke reer 461

Lathyrus odoratus L.. .362, 365, 422, 436, 452, 461

IDIBLELOMUMON(OKYAN DG coo cdo codes e00000000 00 236, 604

Wehmanwiter pecker heir retetoe ioe error 594

hem ont nee ete 387, 390, 407, 449, 451, 461

LENTIBULARIACHAT.......:......-.05- 254

Leonotis dysophylla, Benth................. 252

AKAD YMMl cae yee ogee cooeMo UU OL 647

Lepidosaphes Gloveri....... 337, 376, 378, 387, 469

Leptodontivm interruptum (Mitt.), Broth.598, 599

Leptospermum laevigatum, F. Muell........ 594.

PAGE

Leucadendron concolor, R. Br.............. 275

elimense sehen eevee a 275

iheucas,elabratas babies. cite seis ese 252

Bexdentatas Kan. caste rrecce eof ce 252

NOVA eeranahedeysnspsssversse.eteve Guster tye a acevo e oyare ne a0 515

NUTAING othe cho: Leeatis ciara ice: eke ender elon 235, 247, 249

HETOHENS crater ante enttee ye nane tet tense svoncns 343, 597, 606

Liebenberg, L. C. C.—

“A Revision of the South African Species

OLPAC Oma eee ciseccetenens fore eieveucy og ion ieee 513

NGMIGA' © RYAN Me Serres ee tne ss oo eee eis 229

imeum) Dimtensschellsaseer cecil cite 233

WMeyeris BenzlPrrrs eer netios ser sere 233

Limonium latifolium, Kuntze............... 509

| Poeaca Dcptins Better hiop 362, 365, 461

nim osellaymaiorDielssa-rian emer cr ete 253

Ibslanq eyo) PM ccccoaosnasocogoaqoageecodne 593

Lippia asperifolia, Rich ere 252

MUSEO RNAI UGUMOU a jrercpa aeveccts <0 55 ese hye eccute ai 395

Lobelia decipiens, Sond.........:........-: 256

Locust, Red...........353, 363, 369, 380, 452, 469

GO GANTA CHA: vat on, biciicn arlrtetens < 249

olmum) tenrulentumi W..2-ese5 eee - 380, 460

WOM ATAG MODUSLA aes ioe e steaks Dae ee 592

Lonchocarpus. . . .226, 227, 235, 236, 237, 241, 242,

243, 245, 247, 256

Capassan olfeseep nen ls. ae 239

Mopholaenacoriitoliari. > e+ le lie 1-1 648

TA OURVAINEIEEIEA GRVAIB ate eit citer Gai ein big 231

Loranthus Breyeri, Bremekamp............. 231

Dreger, Hand’ Z. vats... 3662 - 231

kalachariensis, Schinz............ 231

oleaefolius var. Leendertziae,

SPLAQuUe yercleircuen eh he sterae 231

Wotononis Bainesia bakermetecteee tees © cee ce 237

Loudetia filifolia, Schweick................. 227

WAI GEER Cc arees ce Ere vote PORTA rec ae 461

ILA GNIS ede ekso coo oo pon coco da UoDUOeOe 253

WiycoperdOonLriticias clue telat ieia--r-) OU OIL

LLOP RSI IRR ew loterenens 292

LOC TSNCNING CSS ooaseoobbooanoacesonEoe 379

esculentum, Mill..358, 362, 376, 396,

409, 417, 438, 441, 461

rViCOPR.OW EAC BRYA saaeccve) acevo soeteie ie cress 602

iby EO Hoh congsooHoes Soop aeDoSeonebooe 605

diaphanunr, Swep--.)s' tere) 595, 602

Wate hitaee cree toiesstonearnreis eipheke catecucislon bien cate 20

Macrocystis pirifera (Turn.), Ag............. 597

LOHAN EAI NBO Ges Ge aCe OU OOOE 593

Macromitrium fimbriatum (P. Beauy.),

Slater oocguooh abacoe soot e ob Lae comae 600

Maerua Legatii, Burtt Davy................ 236

TASC HONTC HM Cell 0 etererslcleeyelite sl ciciaia = 236

Maesanlanceolata, Horsk sae ci weioeiecie ocr: 247

Maize. ......< 367, 384, 395, 396, 398, 399, 462, 468

MalkamilkariiStraitiere eiecers, «isterae eineinilecteis.e.> 593

seal Dy atrea cos pave a dsc cumniasas eisiecieke ys fererse wisieics S 210

INUATTAVIA CBVAUBYE covey aoieatehe cstyeParciteiers cuei= euevase are 245

Marv at@ akc yoccce.cstete edo ie cote (atseereratesstepsteve oye s25: 594

Manuleopsis Karasmontana, Dinter.......... 271

COT en Re ee Fe SIS CR OIG RO EEE tee 458

Mariseus congestus C.B. Cl...............594, 603

dregeanus, Kunth................. 228

Marloth, R.—

‘* Notes on ‘ Aloe spicata Linn. f.”...... 142

NVA O Waar eae ete ries e rvererenel 443, 458, 462

PAGE

Marsilia ephippiocarpa, Alston............225, 229

IMESSariavenrstsatcs se merce ce ee 343

iIMasseeéllaniinespeae syd. eee eae 488

Massonia latifolia Linn. f................... 511

Matthiola incana, R. Br.....362, 365, 435, 452, 462

McClean, A. P. D. and H. H. Storey—

“ A Drying Cabinet for the Preparation of

Plant Specimens for the Herbarium ” 137

Medicacolsdbivan taser risers ac 409, “452, 462

Si ssoto.du.t0 POO COED AD Oca omas 418, 461

Melanopsichinmrarme eee re er cninee 287, 302

austro-americanum....... 302, 325

Melhania Rehmannii, Szyszyl............... 246

Melia: SaaS entre ee lcaltte ee eee mesons 341

MELTACHAR tarconcce somerset 241

Melinis tenuinervis, Stapf................ 302, 323

Meholaietcnirriaacce ccc rece tater c 348, 390

MENISPERMACHAE....................- 235

Mentha ecrinci ene ita eri 469

SC] SHEeS o OER CIS CICERO CIN CaBaaTS 362

Merremia pinnata (Hochst.), Hall. f.......... 251

Mesembryanthemum saxicolum............. 643

S[dovoowscoHees 359, 435, 462

Microblepharis (W. and A.) Roem.........513, 514

Microceralanantiicola-me wee ee 335

(COW TL1ia rerarediee is 3 icy cho Ci cseer aE 343

Coccidophithord ee eee ere ie 335

COLCO DIU Ua Laat eee ee 335

MESSANULE > Paice ere Me ae 343

TET LOF HIG oa 2ooconnbacescoosds 389

DiTtse platen cee eee 335

CASMANICA. Eee eee 376

MinmisopstZey bers Sounds.) se ere eee 248

Miscanthidium junceum, Stapf.............. 501

Sorghum Richer 501

MMatheei, WUE So obebsasnsonsocodconco soley Mle

abyssimica, Hochst..........-.....- 513

Cigiiatag Harve ete ere 514, 542

GPRM SOMMNV an occclodsodassed0 de 540

gummifera (Harv.), Harv. and Sond.

DTM SENN Boos ouebeasccceob oor 537

paschanthis sHanveree eee eer 534

RAMAN, DIANE os ccdeooeoTedoeboC 534

Senensis wast. vrei ceria oats 542

Modeccin'.:.1 5) s.) 5 tetera eb 514

seohlonecha, 2 canarseten ciation coeur 131

Mollugo Cerviana (IL.), Ser-:.---.).-..--.---- 234

mudi cauliss slammer eter tte 233

Momordica Balsamina L................... 256

Monadenium Lugardae, N. E. Br............ 242

Monsoniay glauca Kunthtececme erie 240

Monterey: Cypresssanccuacieiacer meee nee 594

MORACEAR ieee etre me tena ee cetats 230

Mossi iris ti incre toe etre crater cetaceans 606

“<a MOssesizg)a INE Dixon neice sateen 598

Me iapatayverisyciciiriine teeter eee ne 216

Mphapashane ne ehereinet rerio cere oe 67, 80

Musa Sapientum L..............353, 362, 396, 463

Muscari comosum (Abitewa.)), Will, .codesodounes 290

SAViShroOMesee mrt c eee ce ee me ene 597

Mitisk:m 6] on eotctey et eit: ieee nee 420

Mussel Scale...... 337, 376, 378, 379, 380, 387, ae

Myaris Presltsaa:\ciemGiicia vesaemece tee

Myoporum insulare, R. Br.................. 504

Wivnicatlipa trate Pees oe esa ane 47

NAR OMMEDAN ONL CIBVAIB sono uodevoeobe.qcoE 236

PAGE

Myrothamuus flabellifolia (Sond.), Welw..... 236

IMIVIRISTINGAIC EAI ioei0.so:cfeceielaer rene er gr 247

MEV RPA CH AB oo. a cceteonst eters iae one renee B volte 247

Mystroxylon Schlechteri, Loes............... 243

INEM Goode st aceeammab oe ave ae tes ono 451

Napier Hod d ener sirloin tevin beteeetret vetoes 593

INGais Gusti wma ertnneievsce oe seeiscea ances nepseereredese 463

me Nea WV allt: io aya sicecess sce ona tee cieisiett es arepevens Se 427

I CCTOGUS cbse re Hrate Talon WANG alee ais Slee Eps hep 287

WECUNE CG[MOTIAO. cooactcnoceoddncsoGuueNe 337

COIRTDIOKEH Io gobo co oDURoRAGOODOOON 337

JENGA OG Oa a OU OOOO Oe 337

COCCUCUD Oi ya, sagas voile aie ays cern She ATR 337

GOGCOCC MMO Moosccccocataccsneces 337

coccidophthora v. aurantticola........ 337

COCO MMI ooacccoaccacancna9000r¢ 337

COLLCLUGLE s SPistaaxck atte ate eee eo oie TR 337

CNV Gascaavooascasocneconeos06 337

CLAW OHO PES ata OS dias oitcecac SreG cis 367

DOCONG i sb cc arse pee Shspsponebey ouesponsy Fal spe eisiehs 344

CNC), Vin COMM 555 oo Ooo Geb Odocoe 344

la etiCOlOLsccsz¥octehe darwin ene 337

MOASSUTUAG ers avs. enero Het cdslSoicasaeievs est 344

MUUSCLUONO meine tenet Me Ne 337

IPASSEHINIANUS Eee Ee 337

SUDCOCCUN ED) miccriercntt ta ae Reet 337

SOP OUPPUTHEL es cog soaccaccsaccounnee 337

WM scodbacosangccoocenvovengoD 337

Neorautanenia edulis, C. A. Sm............. 240

Nertera assurgens, Thouars.............. 602, 605

Nertera depressa, Banks and Sol...... 592, 594, 605

Nertera granadensis (L.f.), Druce....592, 594, 595,

605, 606

Neuracanthus africanus, T. Anders ex. Sp.,

Moores. 3 Sree ctacis lis teeusiee cketstscaeoe coves one ets 254

INews Zealand bil ais aemiereracreier ek een 607, 611

INtcotlananta pacuinn ly. rer eis 396, 414, 424, 463

Nidorella resedifolia, D:C..........:....---. 257

Nomadacris septemfasciata.346, 353, 363, 369, 380,

452, 469

Nomenclature, A Question of............... 271

Northern Transvaal—

An Enumeration of Plants Collected in the 223

MY CRA GINACBHAR: - 2 oe sone eee: 232

INIA MOE ooo pC gDeOO DTA DOOD ODS OUOOUONO 232

NAW S UNBAN Digooooooue pococcod0cebad 234.

Nymphaea caerulea, Sav................... 234

CApPeNnsis, muh bs. -pssiseretrertaerttens 234

Obermeyer, A. A., H. G. Schweickerdt and

I. C. Verdoorn—

‘“ An Enumeration of Plants Collected in

the) Northern Dransvaal 4... seen eae

OCHINAC WAR iiriccscre.c sy-ocieude etolteener se aieke 246

Ochnayatropurpurea, ID! Crcny erect tyeiatenelele 246

jDKAKO AGS bo 5bo0eo000000GG0C00CdOC 246

PP WOVoocavsocnod0npaGbocnancnadgod 246

Ocimum) americanum Wy.) se se ee ele 252

Odyssea paucinervis (Nees), Stapf......... 228, 232

(Os Ee che EREOICIG Oe Oy LOR R MOTO Cre OO Gen ab OOO 461

OWA CA CHAR weeds Poe oe Piet Souereccncscsucnetiaeatere 231

Olax dissititloray Oliveneneiieiietiecise raisin 231

Oldenlandia cephalotes (Hochst.), O. Ktze.... 255

decumbens (Hochst.), Hiern..... 255

Oldenlandia sphaerocephala, Schinz.......... 255

OETA CAMS a: Weavers cetetolateletitonsce atateretapelstepepenete 248

PAGE

(OG Ppa cnneiBinoreee tin common Oe ee Oa cic 74

Oleandra articulata, Swartz.............. 317, 324

Oligocanphiary Cassineri..amrsrccisetoensieien 205, 206

Oligocarpha nervifolia, Cass................. 207

OIA GRA CHUA aiey-perctciet suclevelehere teicher riser 247

Onioneasyraece es 369, 396, 425, 438, 439, 448, 463

Ophrocanlon, TOOK freeyrete )etai-r-e)-seels eis 513, 514

Ophiocolon gummifera (Hary. and Sond), Mast 536

Gummer, Masts... sa). cite aoe 536

2(M 2? gummifera), Harv.......... 536

Ormocarpum setosum, Burtt Davy........... 239

Ormocarpum trichocarpum (Taub.), Harms.. 239

Ornithogalum lacteum, Jacq.............. 294, 324

Ornithoglossum glaucum, Salisb........... 312, 324

Orthosiphon canescens, Guerke.............. 252

Orthostichopsis subimbricata (Hampe),

Bro thei each nnte siedisiestet succes oueense 598, 600

Orygia decumbens, Worsk...2..°...05.-....--. 234

Ottelia australis, Bremek..............0.05 225

ulvifolia (Planch), Walp............. 225

Okuohaueray, Bwicaeiibin, IDKCs 5 occcuccauavesouce 240

ODUNUINDYACIBIM Bis cocboagGconondeccuonnne 591, 604

Oxalis corniculata Te... 5... 2... -- es 3) 004, 604

Orange!..... 376, 379, 387, 390, 392, 396, 399, 407,

418, 422, 435, 449, 451, 452, 457, 463,

473

Obteliaicncagaarc are snracised le eae 232

Pachypodium Saundersii, N. E. Br.......... 249

Pachystigma Hochst—

‘A Species of, from the Trans-

EN ee rota Ser octd monces apenas: 183

Bowkeri, Robyns............. 183

coeruleum, Robyns........... 183

macrocalyx, Robyns........ 183, 184

pygmaeum (Schlecht.), Robyns 183

rhodesianum (S. Moore), Robyns 183

triflorum, Robyns.......... 183, 184

TADBUNMNs ooosocadocdocaccodcce 647

Pallamsyarseeevivaciietsracvereicietorsieverstoruetehortake 362, 461

REVANCUNN goauboaocdacoodooouDe oD OEAaSONOND 463

bicolor Re Bienes oie rae 306

Carthagimenses Swen sss - e+ a 314, 324

Panicum colonum, Vinny... 2... ..sees =e 292

Pamicunm Colona twine Maire rete rele ele rene ierenens 593

Panicum Crus-galli, Linn...........-. 292, 314, 324

Panicum flavidum, Retz... 0... . 2.26... 307

Panicum helopus, Trim........-..-.....-- 301, 324

Panicum laevifolium, Hack.306, 309, 311, 324, 372

longijubatum, SHAM avocoouco08 308, 324

maximum, Jacq...226, 307, 311, 324, 373

manlberaering, Mites. ogoonoanqgooauL 314, 324

Panicum proliferum, Lam. var. paludosum,

SUE Mo adwoooeacongspsoudGoOC 308, 324

Panicum repens, Linn...........-...++...- 593

sanguinale, Linn................-. 293

trichopus, Hochst.......... 318, 319, 324

S| aeeoniolo cas Bacnths 311, 316, 324, 460

Papaver nudicaule L...................- 362, 463

124) sKoLstr\iul DR ee ioe Ce CAO .eO'G.C 365, 463

Pappophorum scabrum, Kunth.........-- 300, 324

Parinarium capense........05.+0++s+++ seo" 648

Parmelia saxatilis (L.), Ach................ 597

BPse siaieetclevlotonererei sveyeneredeketeret-t sotetenenate 597

PAGE

Paschanthus Bureh.ws cst ces cece ee ce cee 513, 514

NG Pae, POC sonore GocoDEOUeoe 534

TEPONGUS: BuUnCH 1s). velo). ele = ee = 534

Paspalum dilatatum, Pair.................. 350

distichuntelis-nis se see oo ie es 350

scrobiculatum L. var. Commersonii,

SUA Veo csatoon Leon LeUoa.oDiD 225

BLE e Aga rictiGU CO.uC cogs ou, CoeIOoK 460

IPASSINOLAM Reece etek orotate tte sees 517, 518

PASSTIPERORACHARNS 2. nc-6 cece sc 247, 517, 519

Pavetta Harbor, Sp: Moore.>.....-......-. 256

Schumanniana, F. Hoffm. ex

KER Schumer eee ee il ais 256

Pavonia Burchellii (DC.), R. A. Dyer....... 245

dentata, Burtt Davy..-::.--....... 245

PAnqiiodosen odes cs bes oa seaes 387, 392, 456, 464

LCE RIee porto cae oso Doi mae aomae 427, 464, 465

GAG HB hit se Aon Ncifenas one tsiZoyniay Stared iss serous nehciaess 464

12 Gah ss eee Se OI OCIS cit CE ENe Oe CIEE: 464

Pechuelloeschea Leubnitziae, Hoffm........... 257

EB IDVAU TAC BAGH eral’ See gers wcaearesisversleveie ors © 253

Pegolettia senegalensis, Cass...............- 257

Pelargonium acugnaticum, Thouars.......... 605

Pelargonium grossularioides (L.), Ait...... 595, 605

K[Sboobdodccso sauces boobs 446, 464

Peliostomum leucorrhizum, E. Mey.......... 253

Relbigenarrrtracccr eicriekettt recites --1- 343

iRelivoya) cornern WOE gondacccennocnocHe cue one 80

eNaaccyeqrosn, SONG soocucmoncues 237

Penicillaria spicata, Willd................ 314, 324

Pennisetum cenchroides, Rich......... 315, 324, 373

Pennisetum dichotomum, Delile.......... 315, 324

faserculatums Drink oes ace « 315, 324

purpureum, Schum............. 593

vulpinum, Stapf and Hubb. . .315, 325

aprwaedb da: Verh waracirtinnn. 460, 464

IRenstemronysprecrcriercicrtere- cre strnclecie ars 452, 464

Purgularia extensa (Jacq.), N. E. Br......... 251

Perigermiuntecrrytce eee ee ais see oe eternal 392

Rernicrousiscalezaiccitacsce on aise stents 337. 469

Persea americana, Mill.............. 396, 401, 464

PEABO PERV. CRAIR Soe Non see os 597

Pharnaceum salsoloides, Burch............. 234

Pharnaceum verrucosum, EK. and Z........... 234

Phaseolus acutifolius, Gray var. latifolius

JOG e Soo nore. clon neha 452, 464

Phaseolus Schlechteri, Harms............... 240

IBhaseolus:viutlo anise. steers creer 464

SPessot sys 2525 oes 367, 429, 449, 452

Phillips, E. P.

““ A Question of Nomenclature’ 271

and H. C. Bredell

“ The Genus Elyonurus, Humb

and Bonpl. in S. Africa” 259

and H. G. Schweickerdt

‘““A Revision of the South

African Species of Brachy-

Inenaa Ree BELOW arse + coe 205

Phlox Drummondii, Hook........... 362, 396, 464

Phormiumetenaxs WOnste se esssccs tn 607, 611

Phragmites communis, Trin................ 228

Phylica arborea, Thouars.. .591, 592, 595, 596, 601,

605

Phyllanthus reticulatus, Poir.....’.......... 241

Ehysalistansulatale...ssmcee rete sects oe 401, 464

PAGE

Lbs IDNA 56 Hees Seen obese ae 313

mn, IN, s5ococcoessa00ane 313, 325

PELUIamMas inary eae 313, 325

RENYROUACCACHAB 222-3. as eee 233

Phy tolaccaidioica, Wann) --i-)\)5)- le eae 594

Rixied pple tepureyyn tpi eis eck krish nobis 396, 464

Pinnatella tamariscina (Hampe), Broth...... 600

Pinvisicanariensisy © ysntesnenonyecee seen 594

AG, Whe oocoecoocccodsadaun or 594

INSISTS HO OULE terrace ciclo nee 594

lonpitoliasshioxbreeh ane eee 429, 465

palustrisspMilliy ase ees eke Pale 429, 465

DIMEN, NI LGlosscomencosodedoneede 594

LENCO Ei foie 5 cis c ocecienabiG oro Minin Seen 429, 465

SJMhodooosacoccncoesnge dee 362, 392, 409, 464

iPronnotesklani constr ranean eee noe 382

ISCUAONECERIC EEE eee 335

SHA MMOWVT Oo cccocancodosacbooe 447

CHM ncascogcn 6d cn5.0 un 06 sOOAeE 376

OULCIS x chee seine eee ee oe 447

Piriqueta capensis (Harv.), Urb............. 247

Pisum sativum L.......362, 396, 427, 436, 452, 465

TED UACIONAGIBVN Boga ono oc coasboopous 605

Riantaconanceolatan as eee eee 594, 605

MATOR yc? seh see ees ee ee 605

MEH OE Asta Yoooscddascoucou0ss 594

MEY ORI ION ganoecdadsanacosoon 602

* Plants Collected in the Northern Transvaal,

—An Enumeration of”.................. 223

“Plant Specimens for the Herbarium, The

Preparation, Of 7-1. asnse Pee e ioe 137

Plectromiaispin..)os <a rebe ee 337

Plectroniella armata (K. Schum.), Robyns... 256

Pluchea leubnitziae (O. Hoffm.), N. E. Br.... 257

iPlukenebiajairicanas SOnd seas 242

leo) Champ Ey Wes sonbooceaecocacacoun sac chy HOD

pratensis) Li. - 3.\cehors.o eee 595, 603

Pogonarthria squarrosa (Light), Pilg...... 501, 502

svRokkah-boeng. 775 ...ae ae ee 396, 466

POLY CY StS ix. eirsje: 5 sich et See Oe oe 312

Bolygalaxvirgatal.).... aero 409, 465

ROUYG ONAC EAI ane ere eae 232, 604

Rolyconumvacres seh tene eee ene 302

Enalonlene Iesotoceussaose 594, 602, 604

ity dro pipers cine eae 295

AK ACEN EI IC ooogdeaoaoade 302

lapathifolium Linn. var. glabrum,

IspiRih DEAYooc bobs case 232, 302, 325

Serrulaionn pap eee ee eee 232

Si Usoinino ceoeeen ictal Cloicis ates See ee 286

Polystichum adiantiforme, J. Sm............ 602

Polytrichum juniperinum, Hedw.......... 594, 601

Pongola River finger prass... 0. )h0+-.25.... 593

IBoplantt-sisscexocxerecee ee Ce Een 443

ROPDYES strc astecpine atest Nee eis ES 365, 465

Porotrichum atlanticum, Dix............. 598, 600

valdiviae (C.M.), Mitt.......... 600

FOR LULACACH AR seheen ae. 234

RortulacariasatrayJacqueeeea eee ee ne. 234

oleracea... see eee 234

quadrifidaltee cca nes: 234

trianthemoides, Bremekamp.... 234

Potato. .376, 380, 406, 422, 423, 424, 446, 465, 460

Pouzolzia hypoleuca, Wedd 231

PAGE

Preparation of Plant Specimens for the Her-

PANU os cececeterar erst ocsce, exeisy svedane iene aeeenstceacncte 137

Pretraea zanguebarica, Gay................ 254

EAR TEVE OHA CIBVAUBY: jays ccssecokoioteneh regen nme Ch eae 248

PROPRHACHARY = t52hte sae ceaeseieniree ote 591

Protomiy ces! MicvOSpoTGzUsns tree 313

IPTASOUUUAS cre1nfe ots isis sicker Seacis ee 313

Pro torhtis whinge Py scseternn tenet acetate per 5

TUN US! | COLASUS ria sitors eicierc Guster vere eee 343

persica, Sieb. et Zuce......... 387, 401, 465

sallio tars p arerwr crisis ch- sicuela reuse out ekcaemaette 597

Pseudocadia zambesiaca (Bak.), Harms...... 237

Pseudodistichium atlanticum, Dix........... 599

Pseudolachnostylis maprouneaefolia, Pax.... 241

SPbocovpascoosnaeoocooude 231

Psiadia arabica, Jaub. and Spach........... 257

PSorallearlinssasteainetgois, oe Ala sree ae 7, 116

PEbsoraleasminnnselnes Genii Ssascneriieiineen rite 116

Peo aillen, a@ulleaie, Wesootsponcuboscasceoe 118, 130

Enid Ue Aaa a oeeccone oo oot aol 117, 125

Ajai, lia 5 So6005000Rbo0006 117, 118

Enfueiuen, WN soo cgeecbs 65000 117, 128

ETEN Stel PAMBEEEGENERD Os ROO On eo Iily/, 1233

billorassElanves-r ener tener 117, 128

Inoblbuel, WOSsos550505000a0ne 117, 119

IBolnsnetORDesreErinberinerren ins 17; 123

ronan, lolaitvoccéboccosascccne Wie, M27

bractealtasslninin ser err erneeert 117, 126

Cattray BvanGeAnnene ene at eee 118, 133

CGandicanss Hie andere ones eeearne 17, Wee

CHO, WiliocoplocuonaGo0bKeoE se 117, 120

Carneas Hs Meyenaacs cic nes te 118, 131

Gecumbens-Autaeeeee eet ae IZ, Wea

fasciculanicssO\ Coneenr eerie aneioe 117, 119

claucinalaevererrtyrit tlc 117, 124

Gueimnzielanveeeere eee eee: 117, 120

Ingnaneil, IBERAYS. oo poopoo uo DDOMnD 118, 135

hirta, Winn ckaaes yaa ate eee 118, 134

Keeti Schonlssaneer Peres eee 118, 130

MEACTAGEe MIG Lae VAM eerie ne 118, 132

Wihihaelinennkl, 14 Bios) Ainccanounsoe 7 lt2

OIC, We MI o6mcotoasocoouce 118, 131

obtusttolia-sD {Cee eee 117, 129

OdoOrabissimas acdsee seee nar 117, 12

oligophylla, EH. and Z............117, 121

OHO Mle), SGaMliftooosacanecosaoe 117, 123

Patersoniae, Schonl.............. 118, 133

jonneyieh. Ist, oodcnoosnee oh oeoc 117, 125

pinnata var. latifolia, Harv......... 239

polyphylla, E.and Z............ 117, 122

WOLYSticta, Venu Mncmlettsl- lly 118, 130

neon Walleceoodadsonuodeces 118, 135

DAS Lnvgaveocosodooensoo do ae 117, 126

Testloides, By. and) Zuee sce cee n eee 117, 119

Moy iaeMn ey WAV 6 Soo nue doco cs Th7; 129

SHonaiter, (MOMs ouoceopnonoenuod 118, 136

BVM bis ogasouUco boowon coud 118, 134

AtacChydisy tec iioriervee site ever 118, 134

striata; ADD id-sctvet cetera seers ate 118, 136

tenuissima, H. Mey............: 117, 124

IDWOUMTU MEL aAT Vey cctclsiny even ecinie ieee 117, 129

TOMLOULOSA sel hiss vem efeleeenetenens retare 118, 132

PEMA, MWe aponoaon oo 7808 MPa, 112433

hovebient:, DEEN! Asaoucuoodoa0es 117, 120

WONUStAn peice AscMafasund ier. Cyaieec tare 118, 134

WeniitesEy WG paseo et ane. oc 117, 121

PAGE

Psoralea Walmsit Hianmisis. 42. aeeieenes 118, 136

Ley Next SH ALv sca ayeys erate ee 1i8. 133

Ptaeroxylon obliquum (Thb.), Radlk......... 241

Pteridium aquilinum, Kuhn..............362, 465

Pterodiscus ngamicus, N. EH. Br............. 253

Pterolobium exosum (Gml.), Bak.f........... 237

Pterolobuum lacerans, i. Bre. .))t cen eae oe 237

Ptychomnion densifolium (Brid.), Jacq....... 600

Puccinia amphilophidis, Doidge...........496, 498

Bottomleyae, Doidge.............. 498

bylianum, Dippenaar............ 488, 489

canicis-cemmucae sD O1d Ge. seein nee 492

eyperi-fastigiati, Doidge........... 493

cyperi-tagetiformis (P. Henn.), Kern 492

Puccinia cyperi-tagetiformis (P. Henn.), Kern

Vee, Cijimerial, IWOWOIES. caccaoccccs 492

Puccinia digitariae, Pole Hvams............. 499

eragrostidicola, Kern, Thurst et _

Wikte tus. cis arvcuvremnnienstoe ces 499, 500

eragrostidis-chalcanthae, Doidge.... 499

CARO SICVS IRWO No oeeocosasecube 500

eragrostidis-superbae, Doidge....... 500

CrythraceOSismmermaet cient nrrn 497

eEucomis. DoldGer aren arene 497, 498

wirenaess Oke tire nar fark ea 494

fuirenella, Doidge............... 493, 494

HUST OA AVR: Sono daaconogooddd 494

MMCVAC, DONC, soacccacsosgcane 490, 491

rien reseneKe, IWC. ccocescdccoocca 501

HACwHS (IDCs), Wells osouenccobo0de 491

kyvilimeicolam, Word per pierareme ere 494

Liebenbergii, Doidge.............. 490

Bicciniailvmoniss DiC seen eee eee 508

letnonae) IbeamMAO SHI, SHCh. co sosocsooceoce ct, Ale}

miscanthidii, Doidge............500, 501

Morganaes Doidmemeacmmte a. i 493

mysorensis, Syd. et Butler......... 495

pecleniane, Ord ceresrci reliant 492, 493

pentachun akiegeymecen re cry tpe- deere 489

pogonarthriae, Hopkins..........501, 502

TaMulipes sew old Sener ase 348, 468

schoenoxyphii, Doidge............. 495

scleriae-dregeanae, Doidge....... 495, 496

scleniicolas Ambhim-rrmermeeaier tae 496

tosta, -Arthiiriadecrvacirmcnasrect our 503

Puccinia tosta var. luxuriosa, Arth.,......... 502

Puccinia tristachyae, Doidge............... 503

Versicolon.vo pace Et chon tauren ine 497, 498

Ritudeotepococacogsoooanibanuce ogc 512

Pump Kin steps a heye sts cssieaiels elise sere 443, 458, 465

Pupalia lappacea (L.), Juss................. 232

Pycnostachys densiflorus, Cooke............- 252

Pycnostachys reticulata, Benth............. 252

Pycreus lanceus (Thb.), Lurrill..-..-......-. 228

polystachyus> Beauivec soc. sc. ee 228

IP Abia noose on boosts eh sad GOo OK bo esoe babs 469

Consus cent: coum bocobaGodoscs 337

WUMIE Wesapcaconndss 362, 396, 399, 418, 465

Ramphicarpatubulosa (Linn. f.), Benth....... 253

RANUPROSNOTD viejo: ceva: c¥eis ieee See ea Ee 313

Randia rudis, Hi eMey,...jn<te cverey-:aeaierereteetene 255

RWhiouukiataw como oad sadregoucoCc oS 255

Ranunculus biternatus, Sm............... S7,02D

MEMOS: «|. -yueiehe) eihketene ey tet nes 313

Ravenelia Evansil, Syd..........0scen50- a 504

—

eee

. . Ww

PAGE

Ravenelia Halsei, Doidge................:. 504

modesta, Doidge.......... 487, 504, 505

pretoriensis, Syd.. sore une eee 505

LEN(S) 1172) 0 0 Rr a SRA cereale ea ar 505

transvaalensis, Doidge............ 505

Redi@urramt Tneed.. «be <a ee eresiet.<.! 51

Ot yh OKEr Soa yrviws caret s eee eee os 466

EO CORTESE, te rncrks = 353, 368, 369, 380, 452, 469

Se ae NRG A > RAS Siete Btleeeay ations 337, 469, 473

VU SIREA CHAR.) iy cs sashes ste es 591

Reticularia segetum.........2.....++..+..288, 290

Rhacomitrium crispulum, H.f. and W........ 599

Rhacomitrium lanuginosum (Hedw.), Brid.598, 600

Rhacomitrium membranaceum, Mitt.......... 599

REPAMINA@B AR erase? eck oe aio>, 243, 605

Rheum rhaponticum L.............. 401, 452, 466

Rhigozum obovatum, Bureh............. 253, 643

ZAMDESIACUMG DAK eee te... 2d

SER O ID) OI EUY GERAIS 6 beanie ts cistyyevavs /=-2 597

IRhorerssus:sblamchis tenis = skh sjeustiaes, efeucves eheyere « s 4

ARR iTer ca ete entra vores cay sie stots ah% gan lo: cher ptioecicas a ene 466

Rhus L.

The South African Species......... 3

ACUMInai DC. cit Sean ae es «2. 51

acuminata, i. Micrpyt nee =... 51

acutdens Bmelijac® cesses vac 37, 40, 41

Gequglag Pets 35.28) jan ieiatsiess..< 22

atricaniany Vit 3 ae nietoee risers. +s. 14,58

PRCA Ua Ckilivey., eesie\ste mie eiete << = 113

aglaeophylla, H. et Z............... 62, 64

alain hun 23h | eee ene ks. 4

albomarginata, Sond.............. 14, 59

antboensis, Schinz... <:.i0.seeuns... 72

angustifolia L........17, 98, 100-102, 107

angustifolia ? L. (Herb. Thunb.).... 65

angustifolia (LL. a.), E. Mey......... 31, 46

angustifolium, Herb. Thin sisas ey. « 104

anenania nals ay-(scrks Pe Met eS +. «:: 92

argentese Mii As seytare = oaxsioetert 2 100

ARGenied Bin Cb, Zi. yee So ees =<. lil

argyropiylla, Presl.’. niatenaek ... 4

atomaria, JACQ. .535+ ase 129, 21, 22

Baur Schouls saree ke aeas li, 324, BR?

bicolo;aicht.<-Hee. sneer 97, 98, 100

IBolUSiiySOn Gaeta eae oc 15, 85

Burchellii, Sond........... 62, 63, 64, 643

burkeana, Sond .......)...20.5:: 16, 90, 91

LH eTGO IWCmoodduc Ca ancoeoeen 20, 21

carnosula, Schonl........ Gil24 1, 42553

Gamnrallests, WC yehcic-ticecietek 0. 54

celastroides, Sond...........13, 62, 63, 64

ciliata, achh. <tt> 1-5 15, 16, 82, 83, 89

corrhiflorum,s | bumbats..42 sels +2 - 4

colensoana Enel. ites syne aeiel. cs 50

commiphoroides Engl. et Gilg...14, 71, 72

concinnum, Burch: ..).....2...2--.- 82

Cencalor Pres Dae SEA tS tye. £2. 4

COnTIGED EMG ey SAMAR cia efetobe soo: 90, 91

EN ASSUNETIISUSETES Wor star Pray scheycueno teeters 52

crenata Thunb........ 5, 8, 12, 47, 69, 71

RCN IU OLED etNrey ECT Chi pae oral 509 252 2 eaten 68

GEIS Pamela livaers sees swerziaieta erie iaisten 15, 80, 81

CUMCOL ON Bsa Eis cole) asain, +: syehevare Severe 48

cuneifolia, Thunb..... 13, 17, 58, 111-113

cuneifolid, Pr. MOY. 06. (0 «26-10 6 2 «a0» 39

)

> \

Rhus, L.

PAGE

dentata, Thunb. .3, 5, 8, 9, 12, 37, 38, 39,

42, 53, 85

OUGHT, MUO Ningscandot5sonaceupes 72, 74

denudaig abet Zeererere ora sata. 74

digrtatum non Deer eee ee 4

Ch OGKHOTOS AY ib Senene OBS oe aoe 4

Danteri in ol seer eter 12, 46, 47, 72

duscolo7Schradeeee.er Pe aaeiecr cic. 97

discolor, E. Mey..6, 16, 78, 91-95, 97, ie

CiUspar Pres) en peat. Sad see ees ee

dissectayRhunibs---e es . WZ, NO conan

cling, IE Bit Zo se oocacdoaas 12, 49, 67

dregeana, Sondas- ese ee eeetie .. 15, 83

dunensis Gand tert. eee. 5

duras schonleey sce see te ee Haile a

ebumes, Schonlienia. seis ier. 14, 67

eckloniana, Sond.......... 14, 65, 68, 106

CCKLONIA NORE LES ere eer er ieee oe 21

cklomi Schrag sp eee eee one 97

Abgaite vias AMIN. 356 saooGou8 os 97, 98

ClONngatd, JACQeni nh Actes 18; 19; 21

iii aie, Wa GUIAS Sochoodeeoueouses 74

Hneleniy Britten iar 8, 14, 18, 66, 71

Brnestipschonlrne... ee ese Zs SY Tl, 73

EROSas mu hunMbD eee wel cei 8, 15, 84, 85

excisum, Thunb...... 46, 59, 62, 63, 64, 65

fasticiata! Kiet Aseria. eae 12, 36, 45

HUM CTFT WS EMI. 55 oon sc oscedane 86

(flexwosan Wiel ser-ey nie ra-eeeeroe 29, 30

ociida, Herbs Jaca eer: 21

RAGTANMS LACHUey.etl ere terete ee 74

Fraseri Schonl........... 4, 11,26, 27, 28

fULUCSCETISMVN GO -25a7> «10s elles sore eiatetate oe 50

Galpinn sho ee et. se eee ere 62, 64

Galpinii, Schinz..........-.%2- 37, 39, 40

Gemardia Hane eee UGS 75 Was 7/7

glauca, Desf..... 13, 50, 55, 57, 61, 62, 64

glaucescens7A. Rich...e rset 9, 68

GlaUGeESCeNS> ALM) eee 69

glaucovirens Engl.........5355.-..- 54

eracillimay nels). ferrite 6, 15, 85

grandidens) anvsacas-teeretree re 40

grandifolia Engl...........--.. 91, 93, 95

Gueinzii, Sond: .15, 67, 69, 74, 79, 80, 242

[AG ISN ee REO aS cca honed Odd Cond 26

horrida, Heti Zee eee eee 13, 16, 88

Arwmitlig Seti Zidane ioe ey te et a aoe 45, 46

impermeabilis, Dint. Ms............ 46

OKeauan WAL Ss sob oacoocaeenub eee 18, 66

incanum, Mill..4, 18, 19, 22, 23, 24, 31, 67

INCIsay Lek ws eieetee ee 8, 9, 17, 102-104

intermedia, Schonl................ 11, 28

ecu Songallosodadocogsv0esccs 6, 15, 87

ENYSTHOACH, SCHANZ Asie ate =e lei eit stat-) 12 a 4

krebsiana, Licht......2......- Ii) B33}, Bvt. BR

RGloaiokh INGIBE imc gence saan code 71

Umentgatantlirnts..t rae ieee 4°17, 19; 21

laevigata, Thunb........ 4, 5, 8, 41, 42, 51

laevigata, Herb. Jacq..............- 74

nA Be NIG fe coecosdosoCencecc 66

lena PARR GR aso os a aoees Ns eh LGR ae

Lagicea GSEs cs «charac aio s1~ oxeas.ysdeselsie 61

Cae aoa ee Eee seanUSMIONA do 104, 107

Legati, Schonl...4, 5, 6, 7, 8, 9, 13, 28, 29,

51, 52

Rhus, L.

PAGE

leptodictya, Diels...........--....- 80

(haineniay, ICME sos cacoapodcadobdoDoRS 97, 98

longifolia, Sond... ----..5.4.2----- 5

longispina, E. et Z....3, 6, 8, 9, 13, 16, 82,

88, 89

lucida, L.7, 8, 9, 13, 16, 17, 54, 55, 57, 58,

59, 60, 61, 89

lucida, BY. Mey...0.2. 0.0.65 +-- 22 61

Gi, Linennpo co morcia aia moc.K00Fo6 21

MacOwani, Schonl...3, 7. 8, 9, 11, 18, 19.

24, 25, 29, 31, 36, 45

macrocarpa, Engl................- 104

magalismontana, Sond....... 5, 16, 90, 91

margaretae, Burtt-Davy........ 65, 66, 68

Marlothii, Engl............ 14, 67, 69, 71

Meyeriana, Drcsli ech vai wept 31, 32

micrantha, W. et Z...............5. 62, 63

microcarpa, Schonl................ 15, 80

THOS, IDs WIGS cobcoacicovgdcoanb es 38

HOI, WIAs sccocdocknsouccaves 6, 78

mucronata, Thunb.5, 6, 7, 8, 9, 11, 13, 17,

18, 19, 20, 21, 23, 37, 50, 54, 65

maucronata, HW. Mey............0-..- 62

mucronata, HW. et Zi.i....5-.001 5 -- 58

mucronifolia, Sond............---- 5

mysurensis, Heyme...............- 4

natalensis, Bernh..... 9, 14, 68, 70, 71, 72

nebulosa, Schonl............ 6, 11, 33, 34

MAO, Di, Gi Aooaonavisocguvadocce 62, 64

ntsubanensis, Schonl.............. 12, 44

oblanceolata, Schinz.............--. 90

obliquum, Thunb..............---- 4

alionnite, SOMCl, socaccuebcoce 102, 103, 104

OnE Ra. No Ie Wika ono coos oGoDN 72

outeniquensis, Scez..............04, 55, 56

omuraenmulud., ORNooscccooeuca900ccK00 85

jretS, VB, Eth Bonscooososcccede 2, 64, 65

Focilirekeh, We NIE) Zoca05056ceu0000000¢ 74

paniculosa, Sond..............--.. 5

join, OM Gos ccouoneydounac 37, 38

pauciflorum, Thunb..............- 4

pendulina, H. et Z.................20, 21

pendulina, SACs «eles elel= isl = 74, 75

Pentheriy Zab bree rier 12, 48

(OU XE9, ARM ooo nocos soso 8oKnoesouK 21

platypoda, ir. MGynji-me + nine = 88

jalgera ollie, oo conc ooccds.0D0b 0806 61

falicdepolia; Br Cty Ziciareletslelele\ eels). 62, 64

Plukenetiana, H. et Z........... 4, 97, 100

polyneura, Engl. et Gilg............ 36

pondoensis, Schonl............. 16, 95, 96

populifolia, EH. Mey...17, 71, 90, 107, 108,

111

puberula, EH. et Z........31, 32, 45, oe 82

pubescens, Thunb.............. 18, 22, 29

pubescens, H. et Z.............. 19, 24, 32

pubescens, Herb., Berol............ 22

pyroides, Burch.....6, 7, 8, 11, 24, 29, 30,

32, 46, 53, 67, 80

pyroides var. gracillis (Mngl.), Burtt

IDEKpyoudGo HO s dou DOO Oc ooHdou dE 243

pyroides, Herb., S.A. Mus.......... 49

RONGECTIC, NGS s rae ejielelsnelessiel-iste 2, 63, 64

Rehmanniana, Engl......... 20; oe 36, 72

MOMENI N/A aoomaconoGuaUc IPA es)

PAGE

Rhus, L.

retinorrhoea, Steud................ 15, 75

Tigida VU ene ere ert 16, 89

TUNOSAy BiG brZeeree eee nicct nets 89, 90

Rogersii, Schonly.. 5... 3... 2... 12, 42, 43

rosmarinifolia, Vahl...6, 17, 100, 104, 107

Inlays, WEL os oonccaccnesoasor 11, 36

FUOSCAHIS, 18 Gi Acsococccascascaue 91, 106

rupicola, Wood et Evans.......... 12, 50

SOOO, So nscoocasasccesnone 5

Salicind Sond qari eee Ce 5

salicinum, Herb. Jacq.............. 100

Schoenlandit, Kngl................ 65

Schlechteri, Diels........ ys 1B}, Sis}, (e/5 Gre}

SCcOpanianw heb Anertnl tiiveie 16, 54, 55, 57

seytophylla, KH. et Z........ 13, 17, 60, 113

Sericea. Wet Zivsnist +: oc urace er eustriers 31, 45

sericophylla, Schlecht.............. 29, 30

serraefolha, Burch................. 84.

Simi schon aera erent 14, 69, 71

SH, We Oi Acaoasonnsoucavoeoce 102

SnueTK TOR, INNO) s60506000000550000 ah i)

Sondenating eee eee eee 5, 37, 39

Ganon, i i Soooososoo0descean5 62

SUCH, NMA sonvocccoposccoKc 4

spinescens, Diels.................. 14, 70

Steingroevert, Engl................ 17, 107

stenophylla, H. et Z:....100, 104, 106, 107

SUMeeaMn PAG, Wino codcosooscacaa0nen 51

subferruginata, Presl..............- 49

nM OT, IKE bod00den00000u000 21

Thunbergiana, R. et S............- 61

Thunbergii, Hook. f..............+- 4

tomentosa, L........4, 6, 8, 9, 17, 98, 102

tomentosum, Mill..............---- 22

transvaalensis, Engl........... 13, 53, 248

WRUCZDS, Bi, WIEVooocn6oss8cccc000ac 89, 90

tridactyla, Sond. et auct. al......... 82

tridactyla, Hi. et Z............- 38, 65, 106

TMOACLY LCs URC He ferenaeteten netstat rela 82

MOMENI}, SOC, Gocccsousan00000000 4, 26

tridentatum, Thunb..............+.. 4, 26

tnifoliolatay Bak. fay.) eel ella: 83

UU CALA) SCHUNZ Meera creer cheater te 37, 38

APMC SIS, IDS oooccusuoonobeod Wil

tumulicola, S. Moore.............-- 5

Meson oillip seer lae retry 50

undulata, Jacq........ 8, 9, 13, 61, 62, 63

vernicata, Schlecht................ 62, 63

nllostey, EAU, TON, 66 onoo0ccoocgone 25

villosa L. f....9, 17, 18, 19, 20, 22, 23, 30,

31, 75

villosissima, Engl.....:........ 91, 92, 95

DUNMINAVIS, ACH e we) sie ee es eee ais WPA, Tes

viminalis, Vahl...5, 6, 8, 15, 74, 75, 76, 77

Welwitschit, Bngl.................- 67

Waldingit, Dehn... 2.522.022... 74.

Wilmsv, Diels..........-.-.- 6, 15, 86, 88

Zeyheri, SONG... ++i sees see 13, 50, 54

Rhynchelytrum villosum, Chioy............. 226

Rhynchosia minima, DC.................-- 240

Rhyncospora aurea, Vahl............ 316, 317, 325

IHL) owownar clon cu do ooo oon wesadon coos 337

LFA GAIA EE AGIOS OSS OROO.Ke ORO. o0 395, 396

Riocreuxia aberrans, R. A. Dyer.......... 274, 279

NO Dina). rei s/seesisid eta s ane etericisres cloveteroferorerere 343

Robinia subd«candra, L’Herit......,........ 237

Robyns, W. “A New Species of Pachystigma

Hochst. from Transvaal” .... 183

ISO SIAC HAWES oes eotleemyal sh tie clot otetee aye cyayein eis 604

EVO ARI oe yeh ate Pete) nays SPER NEA Se ete ans wdle'es 469

BPs otestatsiReis ois. siene/dn See Shien pa Sees 337

Rostrupia scleriae, Pazschke................ 496

Rottboellia compressa L............. 296, 301, 325

Gealhimie, Ibivile oc oddoGupoos0s 301, 325

INGhGINA Sebosbcnnd ocucuon ocd ooo adobmedon 248

PRAT DET ae tere ee ahaa ates See Ara to lnieie ve 6 443

TROT E3 DAN OI DVN DISS Sida eile canis ip alte pc ancien 255, 605

1 E304 0) Fst nee pontine chee ty eno pice ose Ole os cae eet nen ne meE 379

Re llappatilasy Jac nr)iaectaiscietscdclelsi< es ole 254.

Rumex Acetosella L.....................594, 604

frutescens, Thouars........... . 594, 604

Ris chia saxicolaneee tpi + os 643

Ruspolia hypocrateriformis (Vahl), Milne-Red-

head var. australis, Milne-Redhead........ 255

Rust Fungi, South African, II]............. 487

ISOM VAN OID Diagiadiad aca Rasen aioe HAR eee 240, 591

Saccharum officinarum, Linn... .289, 325, 399, 466

Salactapehmranniiee eet reer slevcier ei cite 647

ASNT od CAN OR WAND ot ae A a EE a 230

Selle \NVbOSH, SEG oo 5c bb oodaadopaueoen 230

Saliverd onan aerttereiare Wee reeds eho sein oe DO

australis, Schweick............... 248

oleoides, Decne...........:....248, 249

persica, Gareinvemasss ss... 2-248, 249

fS] SUS BRAD OS UC cerat che ao eer anne ae 236

SAUD VAT) ORVAC HAR Heese antennae oe 21. 248

Spria@hkie: Walraacli, Unc cgdendedeseeosncounss 248

KSVANT ED NT DAN 2/2 Ge es a OI tae 243

SHOUAG IIIS! Lemmas Gam Se econ Scat aco Ciena 4

SHygoane, Quills naa ogaccon quand oblood 303

SANE O WA CHAM Meee omar ee ee terse. 248

Sarcopodium avenacewm.............+...+. 348

SACKHEVA GA CBAB onccuccas eee e cake ne 236

DicalasnGLreediyierteistvs stars cracls © slo wispeus, BOTs 409

bid d ehniseyay-y. cea Aa eroeee iesieug, tienen a's 469

Mussel... . .337, 376, 378, 379, 380, 387, 469

Redan pean ee ee oo 469. 473

Waxes tr iretstritr te cen dienes «<3 O75, 400.

Schizachyrium semiberbis, Nees............. 508

Schmideliases cette ice nee: 4

Schmidtia bulbosa, Stapf formaryee. aes n et. 227

Schoenoxyphus spartus, Kuk............... 495

Schonland, S.

“The South African Species of

EUDUSS Pacem rete ne setcta sn ois 3

Schweickerdt. H. G.

“An Account of the

South African Sp. of

Tribulus Tourn ex.

1 Vibe Gah. era ER Asta 159

“A Note on the South

African Sp. of Ximenia,

Linn and their possible

Economic Uses”’..... 179

** A Revision of the S.A.

Sp. of Helictotrichon,

Bessy, exsSChiayeiscu. LSd

PAGE

Schweickerdt, H. G., A. A. Obermeyer and

I. C. Verdoorn.

“An Enumeration of

Plants Collected in the

Northern Transvaal” 223

and EK. P. Phillips.

** A Revision of the South

African Species of

iBrachylacnaaree rt. 1-1 205

Setlaskcraissieebakererrererrhecee crore 290, 325

Mesa Myla aA KCL ells yeretyyeeiete cele 230

SP uevciorrtens alarms ter she pail enter erevonepiarats 290, 325

Scirpus antarcticus, Linn. fil................ 198

COLMUMSW mater bejatecinccsus falorogecowtenaaisieaie 602

cernuus, Vahl, var. subtilis.......... 603

corymbosus, Roth....-.-05-.-..+-,.-- 507

rimpraonnbes, Ob IB Ohigosocosncobocsc 229

sulcatus lhouars-aaemerer eee oe 596, 603

Thouarsianus, Schult......... 594, 595, 604

MHouarsiaiusnvacseee ey aerecereere een 605

Thouarsianus, Schult, var. bicolor,

CIS revere. euasaterete edenennehae/er oe ee 604

VAITCHS bOCCK He ner a eee ite aD oOo O4-

Scleria Dregeana, Kunth................... 496

SCROPHUMARVACBAB eaceeeraaeere 253, 605

Sea WCE oeicie sic crete cnet mele eek ok evan rea 597, 606

Secaleycereales Winnteerc crs siete 312

Secamone GerrardinHanveyeee eee ieee 249

zambesiaca var. parvifolia, N. BE. Br. 250

Seddera suffructicosa (Schinz), Hall f. var.

lanbysnacmas IeleMll Reco oooeenpaoccacoocKe 251

Selaginellamupestrisae-ey eect itor 648

Selenosporium bufonicola................... 383

COCTULE WIM site Oo aee rele 453

CODA B ABA Omron oo Orca aae 356

YMC Mc oscacandsoccocas 366

Semonvillea fenestrata, Fenzl............... 233

Senecio bipinnatifida...................... 489

BurchellitvD Cisne ioe cere 489

COLON AS EVAN i prcecor: arto tae lose) alters 488

lngxnpen EL WIN. coun opocoscocepbe 489

Wt tore us’ /h. jas.ihicie esac rors geet ataole aye oe 489

Ey NUK Sa ognacacsiaosadclsoaooe80 ce 489

UOMO 5 odocdocunvovoneocoNGE 489

MOAI 50 oc ooonadnoonsadasouad 489

polyanthemoides, Sch. Bip.......... 258

transvaalensis, Bolus............... 258

viuloariss [ataeercrtr van erie ae 489, 595, 606

Sesamothammnusiemereiiieceieecirsroerecrels 254

IDiwregh Lobos IN 1D, IsyeAGadeen ao 254

Seinen, Wiideoongodassoccc 254

Sesanrunmcapense, SUunmnns etl cjeie se ee 254

orientale ine enmere ice siege anaes 435, 466

Sesbania aculeatas Persiy. telecine) 239

Sesuvium digynum, Welly. ex. Oliver........ 234

DS CLO) OUT COAG Dee ee ete 294, 314, 325

Slaminey fleyieey, IieEWhyes aponcoodanacnaosoo0d 305

rer bkers (Dy), leeeithyooonuoopedonooue 289, 325

nigrirostris, Dur. et Schinz. . .294, 325, 373

perenmissvelacker nrc oer 305, 325, 374

sphacelata, Stapf et Hubb. ...294, 325, 373

vérticillata (L.), Beauv...........--- 226

Sha vocbenos Gown 6 ouneaO GAS 294, 325, 460

Sidaycorditoliasy Wisc. crater cravetese, eke isc cdavatd eval 245

flexuosa, Burtt Davy...... Bridie rie ve 245

PAGE

Sida Hoepfneri, Guerke....0 2.0208 020... 245

SEMARUIBACHAH terrane eiriceerticic.: 241

Smith, C. A.

“A Review of the Genus

Adromischus Lemaire”..... 613

Snaparag over eee riciachsotetsia nemo atekrayers 466

io] Phalehnily 6s sion Ria aio GON pnLL anemic cine Morrie 379

TAC AMUN Le setae cic aeooe vaventorerieee 253

[GloawGes ING 18Q Neo oeboonadeocude 253

pancunitonme wh Weyer re irrelsea 253

. .362, 367, 376, 396, 406,

422, 423, 424, 446, 449,

452, 454, 466

tuberosum, L..

Son@irg Cerca, Wiccoesdonsaoddctodss do 606

Hii nn ecd OSD a AO OES GA oooemA oC 490

Sorghum caffrorum, Beauv....... 295, 300, 308, 325

Sorghum halepense, Pers.............296, 308, 325

VIETSICOLOT PAGERS ire teirsistet= 299, 326

vulgare, Pers..........295, 296, 308, 326

vulgare, Pers. var. caffrorum

(Thun.), Hubb. et Rehder... ..295, 300,

308, 320, 325, 326, 373, 380, 396, 467

vulgare, Pers. v. technicum

(Koern. VEUIOVI NA cereccge ie revues 396, 467

SPiiot ka eae 296, 308, 319, 326, 460, 461

SOMO VAN, bo anonoesasonnssoendas 285, 287, 303

PHGAGEV AVES soos oubob Bonn owed 318, 324

BATU Palen eee 306, 324

austro-africanum............ 305, 323

IBORTUIMNULCT I See eee 303

Clinton eee rretemerieeacyc:: 306, 323

Genclaninytyeasedic bee eosinsvens 303, 321

COMSAN PUM WM we epee iel nel 303, B20

GAVMUMNs sc6anccanoccdo sacar 306, 322

Bivierhanbiionseeteneienceniio Oss ae

filtered kena wsuphinapeessvenens 308, 326

Flanaganianum.............309, 320

harrismithense..............309, 324

Heal diinatete ic aatictudaderesernce 305, 323

ElGIStiit eee eee Oo UOMOs lao2O)

Hotsoniteevss an hae acieereanee 306, 320

UAKTOMVOMAMUND, orebraocosanaved 304.

TUN CUS. RE NSO TRICE es 310

Maranguenenso:...--.):...0.---- 309

[BAROLO WernP ee MEER SERN an9 © ox bares ae 307, 324

ROTUCI TU CC UR a ei er 314

PLELOLIA CUS aeseiet Teele 303, 321

proliferatum..............-.d07, 323

pseudomaranguense.........309, 320

INGUIN, GaocosocnnSc 307, 326, 327

Saponariae.............-..... 303

Sevariae susiecntocimeasersenoeiie 305, 325

SUMNATs Canclo sac Secon seu ae ois 308, 326

Mem buted aseeade seer 305, 323

MOMMA UID: Go oo bak koe jonceoc 310

MMA HMI coo cudnecberacoese 304

VELECUNAMUM..... sh yd mols oot

VEVSAUILIS Veale seisuicht- aero USS Rose:

Wildemannianum...........o817, 320

ZUNCELMAM UN aeieeyeren eho 304, 323

SWC uooeconEHoGoeNeauscKoCoudeH NEE 458

Spartina arundinacea, Carm,....591, 592, 594, 596,

603, 607, 611

Species, Newly Described..............4... 273

Sphacelovlecaemewserierceterer snes seit 285, 287, 294

Amphilophis..........,,...299, 321

PAGE

Sphacelotheca Andropogonis..... 298, 320, 321, 323

Anthephorae..:0.0........ 295, 320

austro-americanum........... 302

CONCeNELICan erent 296, 321

collumfellifenaeperreeeeaerien 297, 322

ervlental... dette eos 286, 296, 326

Glemisa. as ea ccyerese ysl uber ees cree 296, 325

Dinterivs. sap ceegec eos 298

Doidgeae? 4s. ieee 296, 320, 321

Piva ns iitantenvrsieecrttonceseiicio 297, 323

tagellaiiaryeyiiiemeereiteers 301, 325

EL Leith: cence eel eee 299, 326

HISCHACTA Umer yalereieiericre er 298

WhHillloAgchtiocooogdouchousue 299, 321

MOdestand ites eee 301, 322

Mogoilseiaey Qos scactet cus 295, 321

MOMMA co sccacdodnaaoes 300, 322

natalensishtenchnyeeee 298, 321

Niyassie Sani eusnuie ee 300, 320

Panici-miliacei........ 314, 322, 324

Pappophoris.:t. .0. sa: 300, 322, 324

JDUEWOMNEIS), soaocna po oancos 301, 327

Tetlianal..scs5 eestor lens oct 307

IGCWICL. osc) sreveej ose bsracisegtvers 299, 323

Riuprechititeere mcrae 297, 323

SOrShiqeeeer a ommmnnrar 286, 295, 326

Seb Sooo ulsooeno coos 301, 320

WM oyooessaenooos cell, Bll, S43}

transvaalensis............. 298, 326

Waavionmreais Soaceoue -oo0un6 296, 326

VAN aie ein tes StI cer aoc one 300, 322

NMRIHE Odeo bdagacedboagaodoucuadoaud 344

Sphaerostilbe a ranieele SSL Cipla Coe ee Oe 337

Gaacodiphihonda aac ae ere 337

MBN ooniadgeodecdd ceo dc 337

PLQMUNE Citak reece Onis ce ean ae 337

Sphagnumy nerve haere iel seep rs tetope rales OOD

amblyphyllum, Russ..........598, 599

SMOLIN: nino osahddotooasdncnebuade 339

Spirostachys africana, Sond................ 242

Splachnidium rugosum (L.), Grev............ 597

SMONUSOTUUIM ja aacd eee tae sie aia: 5a esha epee 294

SONG Ue Rad n>’ Glo SORRENTO 295

SporobolustaenOl Cestemtieieet tet tr- iether 502

capensis (Willd.), Kunth. .502, 504, 603

MUN AER, NON soo oo kno oab soo en 503

Sporobolus indica, Auct. non (L.), R. Br..... 603

Sporobolus indicus, R. Br............326, 3841, 343

Evol, INTAtisosose0enecubode 227

(ACUMOEHHUS, ISIC Gousdoaca 0bc0- 502

joymewaraioleulisy, IEWNio oa sdno nooo He 227

SHCA HAeNILS SUNN G Apres os o oie Gwe 227

BPunaartre sertteetaic ar etevscsnces Graverusee 292

SCUMI Ni ogo conc oso tooo do obedoo mbes oN 467

Stapelia clavicorona, Verdoorn.............. 250

flanIrOSUIS nearer criclers ele, oss iekeer tee 654

Gethofais Potties crs sciences 251

ersyciariy IN Bid BieSe ooo gos ood Ob Ac 251

kwebensisie Net iradotsrci: vi cianisier tele 251

MOLT ING 184 13a oa dobum oreo 000.00 251

SEI KCCOReIN ei cietrcieae einige Rralsidieeo ofc 362, 365, 467

Suet Ute Sscme 6b AbedouGGkeoooudoeE 509

Steganotaenia araliacea, Hochst............. 247

Stenotaphrum americanum, Schrank........ 289

Stenotaphrum qlabrum.....ccevcreeceeees 288, 326

PAGE

Stenotaphrum secundatum (Walt.), Kuntze.. 288,

289, 326

Stent, S. M. *‘ South African Gramineae’’.... 147

SUBIR C UM TA CHAI saerts scctmteireiod. cet .cc5s 246

Sterculia Rogersii, N. E. Br......... Sicueusiecerst 246

Solum ammewniaeretl-ve ee ieee ee ee 335

LID aayTe eer aera heise. S oxsus advances Tie cxckon 467

So Rha ou abaytsh \Wat hire: 6 Ae tas Se oe ie Pease eae ee 427

SLOG Kae eae et mn Rect elt mV 362, 365, 467

Stomatostemma Monteiroae (Oliv.), N. E. Br. 249

Storey, H. H., and A. P. D. McClean—

“A Drying Cabinet for the Preparation of

Plant Specimens for the Herbarium..... 137

Sbrawiberrypriacscietteiee es Chee eee es 459, 467

Striga gesnerioides (Willd.), Vatke.......... 253

latea;*Loursessses 353, 358, 365, 396, 467

Striga orobanchoides, Benth................. 253

Strophanthus Gerrardii, Stapf........ BeogoA 249

SIMO MNOS in seco 60.6 aoe ane CrIa ot tener 583

Strychnos, Three Species of, with one-seeded

HES UATOS Pee set ator NetNels Acnske pe ectuccchsge is silat 583

Strychnos Atherstone, Harv...............: 585, 587

Strychnos decussata (Pappe), Gilg........585, 588

Strychnos dysophylla, Benth................ 249

Strychnos Henningsii, Gilg... . . .583, 584, 587, 588

shanevoreit:)-" Dis or Go ence ene ao ate 249

micans, Sp. Moore.........586, 587, 588

JOWINGIMNG S5acocosccdounudea0oaG6 647

Sinychnos Quaqua, Gilg... 26. -.- 0: += 5s 249

Strychnos Schumanniana, Gilg.............. 249

Stylosanthes mucronata, Willd............. 239

Sudedaifructicosa, Horsk......~ cio cies «le 6 232

SHIP al CAME sreetheyete ei cieie\ = ets ore 395, 396, 398, 466, 467

* Safire AN naan? Gxoyins se obe ae Be cep eee 458

SSH EMU eye hs isle yeretelsistere tein cvayeiee« 395: 5318 458, 467

SHIR THLE UE TOMER ease o Sheds seBoameaed | alll

Swaziland finger grass. ....... 02.0. 221s siele vi 593

SWKCOL PCAER EP actos taetesetiotal-\ spspageheh sof dvs ce: 362, 365, 467

WOUEIDcoosodgnnbonupsuoonpooeeeaeve 467

SUL GAINS eperey ctv (ee caste oye foyaysseuslieermuedecacsy® 467

Sylitra contorta (N. E. Br.), Bak. f.......... 239

Syzygium cordatum, Hochst................ 247

Maatboschix.rsie scence foe ete eee 98

Talinum Arnotii, Hook. ee MES 234

caffrum (Thb.), E. and Z. ciaenesone tet 234

transvaalensis, von Poellnitz....... 234

Tavaresia grandiflora (K.Schum), Berger.... 250

Wepary beam: «aca: seyret cee eee See 464

Tephrosia capensis (Thb.), Pers............. 239

euchroa, Verdoorn.............-- 239

NOCLINOLA cise nies atk Te ees 239

purpurea, Pers. forma?........... 239

zoutpansbergensis, Bremekamp.... 239

Perminalian.. eee. wiek oe ee ee ete: 244

prunioides, Laws................ 247

Rautanenii, Schinz.............. 247

Sericeas urchin... 2a Noses 247

Tetragonum arbusculum, Fenzl............. 488

MnatchiGrassocame cis css sea aa ciel sets 603

Themeda Forskalii, Hack............296, 303, 326

Themeda triandra, Forsk............296, 303, 326

Mheobromars tir. es cc eect este eh eee. 341

Thuidium curvatum (Mitt.): 22.000. .0.00..- 601

fULfUTOSUTDAS. c.ee et ee et tate 501

PAGE

Thuidium sparsum (H. f. and W.), Jaeg...... 601

DT PAC HAUS «5 <5 o.ciarc tI, RAE ea 244

REELEEVACKAR: 2)... ie. oie 283, 284, 285, 316

Mibletialchct eas ata SE aN a ne 285, 287, 310

ATT AC's ctejo.cilc-seedeiee PAR er ee 318

PSV LOSI a ncssr eget startle esis NG eR ES 283, 311

COTLES AI. ARE eR cen 311

CORCDEIN A cwie esses erate el RAN oss ce 318

estrienss sat PRB. Paks ey ee oe 314

TOSbATIS Er er nor eC Eee 310, 326

heterosporanmacacta ie eee: 310, 324

laevis MEP ao oe 285, 310, 329

Sclhenckian aera reece 318

SCR MOUUUBs coongegn jode+dene pon 295

transvaalensisanccebcee ets SL eee

TritiCinenser eehes elas oer: 285, 286, 311, 327

Mienn obitmninciscciock come te 311, 321

Mo bacco se cee este 414, 415, 424, 463, 467

Toddaliopsis Bremekampii, Verdoorn........ 240

Holy. OSPOLIMMT ees yaer-1 rye ip cree eee 287, 310

Amihistiniaessnancmeneieen 314, 320

Genchintsncsre der ces 303

Chionidis:-aeece aera 319, 321

ver hart syst csc Wa aloes oe 304

flier. AP eee eens 308

ristachiydishyey eee ere 310, 326

Volkensiie,.. Oshets ooo wn: 319, 326

EMG, co o5560 sobs 0b 00e 314

Renicillaniaesyseeeeeeeier 314, 324

Setanlicoltme seein 314, 325

Mowtato wach avers ates 362, 417, 418, 438, 462, 467

Trachypogon plumosus, Nees.............292, 326

DAC Van Teaser eee eee cite eter ete 284, 285

Tragia Okanyua, Pax...........0....0.+.0- 242

Tragus Berteronianus, Schult............... 225

UR ORVMEMENNE Gobo aaaseo wat scanuencooddae 174

vlree: Henny osipejsts soe sn oer see 595, 601, 608, 604

Trema guineensis (Schum.), Ficalho......... 230

Trianthema erectum, Schlechter............ 234

pentandra, L...............-.. 234

Trianthema salarium, Bremekamp........... 234

Thimigviings Mew Wels eesincooccoomsasonawode 167

A nHleis HATS, IDG (Dansocacssossoccsooc 167

Glatus Drege. 2s sete rte ne ae 166

albescensSchitiae seater re 166

aquaticus, ©. Bauhwas. aeeeeors Gee 174

AGUALCUS Liss ars. Chere Ieee) seis 174

Mri ulus! cCIstoidessikis 4) ieee, eee ere 163, 170, 171

Cristatiis! tEresl a. as sitet 161, 164, 165,

167, 178

Tribulus erectus, Engl................162, 163, 164

Tribulus excrucians, Wawra............. 160, 161,

162, 163, 164, 170, 178

Tribulus hisprdus, Presi: .\-c2 see) 172, 173, 174

Abin I OHbS, Ii, Lito 5 coo edoeabasugioemad 171

DSHS COATS, Al oc ooo Cad0Ooo ae c 162, 163

Tribulus sliscxcitacttrocere ees raetrsusr eer 159, 160, 161

MLACKOPLCLUS eS OISS er erate ert iele 167

Rribulus mires’ Preslyie it 170, 172, 173, 174,

175

murex, Schlecht. ex. Dint...168, 171, 172,

174

GUESS WIG NiGan Walia BOO dOenO ORO e 162, 163

parviflorus, Schlecht.............172, 174

parvispinus, Presl.......-.- 170, 172, 173,

174

PAGE

Tribulus Pechuellii, O. Ktze..........0.8. 162, 163

Tribulus pterocarpus, Ehrenb............ 167, 168

pterophorus, Presl... 02.2.2 5: 161, 165,

166, 167, 168, 171, 175, 178

Tribulus securidocarpus, Engl............ 166, 167,

168

securidocarpus, Engl. var. subtrun-

CHES. WOVEN sodas anon do ugoo 166, 167

securidocarpus, Engl. forma vulgaris,

IDaalogeaod sD oobacaDo eas ocus 166, 167

ar bUlUsivennes ins; eons irctleriortrtrreecnenersie: 240

MWA Wpogoaooacas 159, 160, 161, 170,

171, 172, 173, 174, 175, 178

ALO HNS Wane Olivos oogodooubbousub sence 162

terrestris B desertorum, E. and Z. 172

terrestris, L. var. B hispidissimus,

SOMbnoncdasonobosoaceddaqaadoac 172

terrestris, L. var. §. desertorum, Sond. 172

terrestris, L. var. Zeyheri, Schinz 168

Tribulus Zeyheri, Sond... ..161, 163, 168, 170, 171,

175, 178; 233, 240

Tribulus Zeyheri, Sond. var. auwrantiacus,

Dinter, 225. cess oe bobae 168, 171

Zeyheri, Sond. var. hirsutissimus,

Schinz act wace eee eakihele 168, 171

Zeyheri, Sond. var. hirtus, Schinz.168, 171

Zeyheri, Sond. var. Pechuelid,

Schinz ees sesepruinseyih ase... 162,163

IDAGES, 2075 Yelscotalivert aedaieeie: s sice'se cee erenareretenskas 226

Trichoneura grandiglumis (Rendl.), Stapf et

Eb bie seledicianeten deta eats 512

Schlechteri, Ekman............ 228

Prichopteryx Dintert, Pilger............+---- 227

Trientalis europaea, Dinn.....5.-.40.4--- 5. 312

diritoliumpnepens, Weeyraeha irritate c 595, 604

S) ipoudinid boob uddnnds do bo np0o.0 418

Trimeria grandifolia (Hochst.), Warb........ 246

Tristachya hispida, K. Schum.............. 503

Rohmannim Hack seen olO 26

Socououmanoogoonesasacndanuee 310

Tristan da Cunha, The Flora of............. 589

Triticum dicoccum, Schrank.............312, 326

durum Dest... eee oe to2O

eijerhpien, Wits 5 aoocauas00cae6 312, 326

vulgare, Vill...... 291, 310, 311, 312, 326

SPs lo wiereite eet we stoke eee 381, 396, 467

Abioyorictol bore eeIWEL Wigaoononnoescoss none 449, 468

MBE NE rave pisces c.etyeinss olde ae SOI 353, 469

ANTE) asiehin oR eRe ae, sce ono Parco 594

shiiberculanialcoccophildn..et inners 335

Murbercwlin ais 3.02). % 302 aos ees 4p OPS ue 319

ANG Ulio th oa pere eee OBO OB Ou bcd 6.0 dos ules OIG

Briospermi!.- neo ae eee Oo LUe Ore

Onnithoglossy.. ee eet eee 312

Trientalis.,.i:. par accrep ke erence 312

AULIOLCL 21.) s: sis store ora eines eue ots 312

PARLE vei es cexsicatersicjeys;)0s <P oloe 438

URINE RACHA... dslexoseneteeeeranbre. 247

Manchonanthus, inn... .dacads venerele 6 e 20D

Tarchonanthus dentatus, EF, and Z. non Thb... 207

PAGE

Tarchonanthus dentatus, Thb................ 209

ellapyicis aU hon ene eee 212, 216

glaber, Wit Oa eee aoe 209

lanceolaius ih beeeeene ee cenee 207

TACEMOSUS, Mh... eters). 212, 214

Malletiaispiinsovthais wek ete ee eee 197

“2/4 bra yplhors| WNobtan, Ge IDNs oooncco pods Nace 159

** An Account of the South African Species

Ofte

Trisetum cs) Sita ieee eo eee 185, 186

antareticum (Forst.), Trin.......... 199

Trisetum antarcticum, Nees.........- 189, 191, 196,

199

Trisetum barbatum, Nees...............-.- 190, 191

barnbatuma Steud eaeeiteeeienicel 191

barbatum, Nees. var. A............ 190

barbatum, Nees B minus, Nees .189, 190,

191

Dregeanum, Steud.......... 189, 190, 191

ORM, SECs s¢ncacccbosocdoce 193

hart, NCe@Sia sin. os aceknie aise oeeaes os 193

amberbex INeestasieinc moc a 196

imberbe cornutum, Nees..........-- 197

longifolium, Nees...:.......-... 189, 195

Steudelii Neesmane eerie te 188

TUM MCAAW, IDERNio00 04026 Doan O00" 199

Teelabel og2:22 oxic als Se oe eevee er La 84

Tussock Grass scc./50 sco tenet helen sr 603

ULEMACHAE Fines Soke cee cen eee 230

Wilms 6 eh ae ee ES ha eoceeree eee 343

Wivallactucaztieats scot ee eae 597

WIMOSITILIDINMBIRAIB Ss 55 Gocouo cue caeoeny ox 247, 605

Wamnononokrese on oa mint tien career nae es 587

Uncinaria brevicaulis, Thouars var. rigida,

WRG TU Paeesitrtcre acurcchet raed siete Mi 604

Uncinia breviculmis, Carm. var. rigida, Kitk. 596

(UIRIDIDIONUBVAID. Gog onenocgupobaducounctodetey sau

Uredo brideliae (P. Henn. et Evans), Doidge. 488

CORDON Me aera sds bet Nee eee ae 288

CONDOSAWENAe rns oe tee 290

CONDOS ELOTU CUT ec eta ee entero edt 288

CONDON RONNGUITALIAGC eee ee ere 314

CAT0O5 PTUCl 5:55 Saiee cio ak eee ees 291

carbo-vulgaris quenae..........--.-..- 290

Uredo cavricis-petitianae, Doidge............. 507

Oped 0. Cartesi. tis, ok scenic seit el ee rer 310, 311

Uredo combreticola, Doidge................ 506

UOnedo destrilens--ereeeyeiers oe eee eee 314

digntariaecola, Thuemy................ 499

HCOMEMMAoaoeohoclo COR dub dau nOC OOO 311

Wrecdonbuinensessa nH enrieene eer errr alts 494

OIG 1a hROPVY OTIS 5 6000900 boo onouGnce 295

Uredo Hyperici-leucoptychoides, Doidge..... 506

Hyperici-mysorensis, Petch........... 506

Hyperici-Schimperi, P. Henn......... 506

Rey llinciaes Penne eet seeraer iret: 495

Unedo lamoniis WDiulbyrerre cree iiyetereteriensl= ee 508

Uredo longaensis, P: Henn). ..0:.5..05....-- 506

Credo MAY dis... «cSt Rae RTT ole 292

ODE CEG ea icin cise thao RIO eee 294

pilulaeformis: .. yj... ace etna ie 315

PilUlifonmes; \..1 eee rae ev 316

pogonarthriae, SY... ue wee es ler. 501

Uredo schizachyrii, Doidgg BEE prep satoietatacs 498, 508

scirpi-corymbosi, Doidge,.,...-...... 907

PAGE

FG EMOMBEG CLUTIUS sia torasce = siaisi2 (2 aoe 2% =a letays ol 288, 291

segetum, Mays Zeae................. 292

segetum Panici-miliacet.............-- 314

SEgelLUMy DTU Pace oes eee sae 291

BULO PML. ror. 's a a)s ato tea te rete oe 311

SUMICGMRILGH 6 Sa boodboec50GoDed65 06 304

UICNOPROTA. Se cee oe ete 6 LO Ns SLO

trichophora var. Penniseti..........+. 315

AUT CEOLOTUAN w. tater telchesetaten tena a erate 315

AAT SACO DAC ATOLL COTTA 292

Lees Mays. sso sei eet Few, ee 292

Urginea Langii, Bremekamp...............- 229

Urochloa helopus, Stapf......... 301, 304, 324, 327

panicoides, Beauv......-......): =: 226

Phodesiensiss Stenbazecer wn a 226

IWEOCY SUIS treitas Mic eterna eeeer ts 285, 287, 312

; AMEN ON ESky.) eicke oe acta austen eek BU, CPAs

Anemones f. kerguelensis.......... 318

Ornithoslossitee ee ete 312, 324

SOLOSPOLIOIM ES" 4 -).f) ee oe 318

MGCL ass sss me acl MRR cc 312, 326, 327

Uromyces antholyzae, Syd................. 510

Clignyi, Pat. et Har........ 498, 511, 512

eriospermi, Kalch et Cke...-...... 511

limon (DC:) Wevancsa eerie rs 508, 509

massoniae, Doidge............. 510, 511

Strauchii, Doidge:..............- 509

trichoneurae, Doidge............. 512

UIRTIICACIMA i soaedenecceesng opoSeb ceded 231

Uciibhriciiiipasisese. sppeusmdpeatens Haoer 287

PTUICUT ee re etn eee 291

(OS IMEVAC III AIDES Soooodooducs cubcoscoses 468

USTILAGINACEAH........... 283, 284, 285, 287

USTILAGINALES............. 283, 284, 285, 287

Ustilaginales of South Africa............... 283

RUstilacimordeal secon <trce staiolor--teyeisreseatess es oiveats 319

MUIStr ao OMee tt Nek Sete e eee sera einen s ser 285, 287

Phin OS ert RaaRmeoan Gara Gacoe 288, 326

OMELUCON A ete (aFel Velo ohaseisie ee 289

Andropogonis-finitimi........... 293, 320

ATURE DRONES riage fo: 20) 2) ntadsy Pekar sate 295

ATUSHUAGE nisss ore 3 2 5 005 ayia, Se ya Rarer 303

WUStTO-AMETICANUM. . 0... eee ee eee 302

AVIENAC alarsteisteetecetarsv as 286, 288, 290, 320

PAD CTEM io OLULCOLM terete erence eerie 290

VA VENDA VALNLEVUS rece sterejveyeysiresice >: 288

CEICOLD AAR ancl etre ees: 302

BrAachYy podii serie ees dass vee 293

Brachypodii-distachyi.............. 293

IBONUL=OTVENSTS (os, c1sle Sisitss Sicie) =< ene - 293

IBrOMI-MOUS ents auemiele ee Nes ss 293

PLOMIVOras sR eee se ee iets alee 293, 321

bromivora f. Brachypodii........... 293

COPENSIS oetctayatek roe ereterayer sis eya'e!aheve 316

Carbo var. columellifera............ 297

Carbo 1. collumellifera b. trichophora. 315

Garboidestruens $k 22). bosch os 314

Carbo-vulgaris avenae............-- 290

Carbo-vulgaris bromivora........... 293

Carbo-vulgaris Hordacea............- 288

Carbo-vulgaris Triticea............. 291

COMLCTCOLAM a ayet eters oys = cond tceel shay 2:a bieeie 294

CHAT Gin Sh Roba Obe GOB Mea One Ld Oe 294

(OCT Pe peton CoO OOOO Oere 293, 304

Oramnerince rare ecole eco 289, 325

CEG littenceccvockh Uae CORO BESOS 296

PAGE

Ustilago Crus-galli.)-.- 2.21022 ecco 2 292, 322

POMP enconeso Pe Onn odes Joc e 306

CYyMOGONUISS cicero siestnices rettarei es 289

CYLUMARICA NSS Cae Coe eaten 298

Wachylochaentinenasssaas ert 290

Danthoniae eyes sees ste en ee 315

destruenss:.¢ic-1.}. ts SORE as. 314

destruens var. Digitariae............ 293

DUN ERE ATA E-eo oee tee ie ets 298

DreSeanaroretel jerarete: clove sve tet oie ee 315

MN ONUTIE Re tyes kee ae = 288, 322

BUchlenaessavaste se sehen cee ee sscars 292

IFWWANSUE eBhisticigian ca dacaa eee 294, 325

ining erhuthigese- eee er 291, 322

flagellatas s \c\ pei hes ee es LEN 301

POCA S RUS Nava cies Neicpate act ateiicrs or feceer neal 310

Helen OSporas jira aes eee 310

EV ONNUNGSU a) -jcte 5 ac oe eR oe 289

FL AlOride...cbsf- Sei ies IL TS 288

Ake) Lieve mec Gochie eH asia amis ees 290, 322

Hordeiken ee eee ee Oreo ola oeS

ATUCONSP UCU ANP e el eee ee ais 304

TiSCIULCTIOUS <teysonen hep vexcus sas sv es Bees RS 298

TOMSCNU «14.5; xeuveusi server sis HORE Ss 288

Koller seres cis trve btcess ae steee GIS ola 288

LeUcod erin yan sevae Gistete she nena eo 316

leWasey.eavsths atheism thaee 286, 288, 320

DM anys sia srs) Sah she cslStsoles es = ntsreyeun ye ose 292

LU PALO Goin 6 aE HOO AMO eS CO BROUOS 292

Mab cedieet ie. adiois tis Sat sees ate 299

MOLES OP THAN NSIT A AE yes referee 301

TAURIERs anna suogae cee0de5 S0000 300

TVUICLS, oe Ey JSR acre atchetanetons 316, 323

AUST aac oho abocabodcconoacooE 300

OliViACO a zvcce aie lost eaten ate eye 294

IRONMICIMULACC ttt ae ea 314

IAGO DOIG ies 5 don oo boaRI0o OS oo DOOR 300

(DICLOTLEN SC ee ieee ear ee eee 301

iRegleraeyes rere cee reer 294, 324

IRenmisebinene eee eres 315, 321, 324, 325

Dllulifonnrisse sleet 315, 316, 323

QUOI Shason auc obodcooboe nee 289, 323

ULUCT ACCU art ctere iyo rete s -lteete 307

Rabenhorstiana.....--- esis) 293, 322

RevOna ar oc ee tee ere: 307

Reiliana folvicola....-:.......-..»- 307

Sacchariiye marae eter tere core 289

Schlechteriiy «foc e-ciss asec sts 292, 326

Schweinriarayyraccrrcayerser tec tck eee 292

S Cilla eee se ke ieee vet cies 290

Scitamtneaseres eee 289, 322, 323, 325

INCL nosa nowdone heap 287, 288, 290, 316

Segetwm Avenger aretevsirtersi=1e rele hs 290

Segetum var. Hordet..............- 288

Segetum var. Writicte 2... -.- 24-9. 291

MAG Pacoacoce sono oben osomoLaoue 293

SCLIN esheets Sele ae ee ele 291

SUOPLIUMa ors er tre ere ericiel diel: 311

Sladeniinpsyacieccnci acetic 316, 322

OV QRU ae stec elcicsols sins Siclexigeseas 295

Sphaerozenamrpr vaste rasias a 293

ISLET OUAD eterno rt oich deter kere) 288

SUD OLWVACEE earner i Moraes terete elev 294

ISYNTNETUSMAC So oo ~ 50 eo He ye oe cele « 293

eCuayhOrdelemeroryersivlaeiae sc.cretsrocreh: 288

UFC S EGR E A BO OOOO LOE CER e Ie 297

PAGE

Ustilago Trachypogonis................. 292, 326

MAT NHa54occcccouguces send 291, 322

ADP ribiel ksh. Wee es eto lyyeat oe 291, 327

Riuiicufolvicoloment een ae 291

Meilasnei: oases Ga Ieee. 296

MAO ooccosheanvoosceedoube 304

UPAnd ensigns c craesac. cl sccuse yop PA 316, 324

Vastllan titre aires. tne 290, 320, 322, 325

VGC Ouray vagy: *) scion Oe RO 291

VET SALUUS atehepoten (Asis CL ORCC 308

PER ECUMA Cae Neus oi sian re uate 304

Wielwitschiae:cc.nccnsecue ie 318, 327

VANS gs SNE Boek soe 284, 292, 327

Leae:Mays'. otha. Vebyej ater mentee sor 292

Wimrculanisvexoletaag ice teeta ete 254

WB DOS:. jay sesscedarcv es Rep alceuoysiss Sb te CTO oe 216

Mahliaicapensiss Mhibs. jennie eee 236

Vangueria cyanescens, Robyns.............. 255

Hori bund a ELo Diy Seeeee eee 255

MAAS: 2 foie pces oan le Ses 647

tomentosa, Hochst............... 255

NCCE ERE OSC OO MAT OTe Me bio cca ate hae 594

WAM AD OVAVNCIDD oo oogas oa cdndanoo aoodaoe 230

Vellozia equisetoides, Baker................ 230

MER BEINAC BAD Res. 14 ee eee 252

Verdoorn, L. C. ’

“Three Species of Strychnos

with One-Seeded Fruits”. 583

A. A. Obermeyer and H. G.

Schweickerdt

“An Enumeration of Plants

Collected in the Northern

Mramsyvaialy?2o:. fone ti. te « 223

Vernonia amygdalina, Del.................. 257

cinerascens, Sch Bips.j).6 0 itt 257

fastigiata, O. and H..............- 257

Vernonia Luederitziana O. Hoffm............ 257

Randi Spy Moore: .). 2. cae. nti ee 257

Veronica serphyllifolia, L............594, 602, 605

Viscaria viscosa, Aschers..............0+: 362, 468

Viscum combreticola, Engl..............-.- 231

VEGRUCOSUIMeN Hem vena a niente venir 231

PAGE

VEEACH AR ahs acai ooatoc akc ne Ereene 243

Wibispvaniterasi ih teae Obes ere tr. ne incts te 468

Wittarialstrictan| Canntsaaea aati 602

watbariordess( Cy Chri scene eee 602

Vulpia bromoides, Gray.................595, 603

Waltheria americana var. indica, K.Schum... 246

Wiatermelonis arene any 420, 456, 458, 468

Water WiatyBlsis', Shenae Pee) lee eee 209

\Web-oisten! Co prnon denon Goto crooe OE o oc 387, 469

Welwitschia mirabilis, Hook............. 318, 327

Wikheat.. A202 aca See BE bee 396. 398, 468

Waldi@elonyi: Setescee detetanited sects 592

Watchweed'e.2)34 33... see 358, 396, 467, 468

Woolly Winger Grass\yecet ese reeseeeieee 148

Woollygiimgerity pe tei. ees scieelee eran 149

Man thoxylont lis. orrs atari eceleteke aero: 3

Ximenia

Linn.—

“A Note on

Species of ;

the South African

and their possible

ECONOMICTUSE Sige ite nee eres 179

LINTON, UAL bb blagoodes ok 179, 180, 181

americana, L. var. microphylla,

Weeliwisen-<emstity cee ane 179, 180, 181, 231

caffra, Sond............179, 180, 181, 232

caffra, Sond. var. natalensis,

Sondescetecacyese a aaes 179, 180, 181

Ximenia Rogersvi, Burtt Davy.............. 181

pXCUIVVETI Tay OI es es wiiere sacotsinn sets ciae shveter tere eter 181, 182

XOVIR DA CHAG f:. ci: <is o's oichausaavolasenpaeiece 229

PXGyrisicapensis)w UND select eee net 229

WOKS aie NC ehundhconace ddaods oat doddioos 603

Zea Mays, Linn.. .293, 308, 327, 367, 384, 396, 399,

449, 468

Aine paler ora Win Dineen selec erie 313, 327

Zizyphus mucronata, Willd................. 243

Zundel, G. L. I.

“ The Ustilaginales of 8. Africa”......... 283

NEE OMMSINALILACIAT do ssscqnaeedads ooexs 240

G.P.-8.3217—1940—1,000.

The South African Species of Rhus L.

By S. SCHONLAND.

(ILLUSTRATED.*)

A.—Introduction with Brief Historical Notes.

The South African species of Rhus are placed by Engler (in De Candolle, Monographiae

Phanerogamarum IV, 1883) into his section Gerontogeae. The section is characterized by

him as follows :—Drupae globosae plerumque laeves, raro pilosae initio compressae, mesocarpio

crasso paullum resinoso evittato cum endocarpio cohaerente, exocarpio demum soluto. Folia

semper trifoliata, raro digitata, 5-foliolata. The section seems to be an unusually natural

one. Since the publication of the first volume of the Flora Capensis in 1859-60 no account

of these plants which form such important constituents of our Flora has been readily

available to South African students. Quite a number of new species have been described

in Engler’s work just cited and additional ones have been published from time to time by

various authors. At the same time some species have been found to be untenable as

material accumulated. With few exceptions the work done on the plants of this section

has, by previous authors, been based on Herbarium material. This has frequently been

too scrappy. Our species are normally dioecious and both sexes are required for descriptive

purposes. In addition ripe fruits are required. Coppice shoots are frequently decidedly

different from the shoots of adult plants. The branching and size of the inflorescence yield

important characters, also the petiole and last, but not least, the leaflets. The flowers are

distinguished only by minute characters which only rarely can be satisfactorily used in

distinguishing the species. Unfortunately the vegetative organs are very variable, often

on the same plant. To quote only a few examples: In Ah. dentata the dentation may vary

on the same branch or may be absent altogether. Hairy and glabrous leaflets may be

found on the same branch in this species. In a large group to which he refers 61 species,

Engler divides these into species with winged petioles and those without wings, but these

wings are often not developed, e.g. in Rh. longispina, where sometimes they reach their

maximum development. Again, the shape of the leaflets may be quite protean, e.g. in

Rh. MacOwani 1 picked from one bush leaflets which were ovate-acute, ovate-obtuse,

obovate-obtuse, obovate-emarginate. They were mostly quite entire, but there were some

with one crenation near the apex, others had 2, some 3, a few had more. Most of the leaves

were 3-foliolate, but some were 5-foliolate.

From the descriptions of the species it will be seen that such instances are very

common. This makes the construction of satisfactory keys to guide the beginner practically

impossible. Only recently Prof. Engler, with reference to his key in “ Die Pflanzenwelt

Afrikas ” mentioned to me that it must be used in conjunction with type specimens. Diels

(in Engl. Jahrb. XXIV, 568) has had recourse to anatomical characters, but, as far as I can

judge, these are only of subsidiary value. In any case they would have to be investigated

on a much bigger scale than this has been done hitherto, to be of much value in determining

species.

In view of the extreme variability of the 8.A. species of Rhus I have for a number of

years given special attention to them in the field. In the year 1924 I had the advantage

* All figures are natural size, unless otherwise stated. The tertiary veins are in most cases drawn

in a portion of one leaflet. Hairs are omitted in the majority of drawings.

of inspecting the collections of Rhus at Kew and Dahlem-Berlin. Aided by a grant from

the Union Research Grant Board, I was enabled to examine these again more carefully in

1927, and I also studied the collections in the Linnean Herbarium and at the British

Museum, London. I also had the privilege of examining Thunberg’s Herbarium*

preserved at Upsala and Jacquin’s specimens preserved at Vienna. In addition, I have

seen most of the material preserved in S. African herbaria. The herbarium of the 8S. African

Museum was particularly helpful as it contains most of Ecklon and Zeyher’s types. Many

friends have supplied me with abundant material, and a number of foresters in the Union

Department of Forestry have, by instruction of the Chief Conservator of Forests, contributed

much live material and many notes. If in spite of all this help, for which I express my

very best thanks, my work does not come up to expectations, I can only plead that the

nature of the task made a quite satisfactory solution practically impossible, and I can

assure all my friends, to whom I owe so much in this connection, that at all events all the

material has had earnest attention and consideration.

The drawings in this paper have mostly been executed by Miss Olive Armstrong, by

kind permission of Mr. J. Hewitt, Director of the Albany Museum. Much assistance was

received from Mr. R. A. Dyer, M.Sc. of the Botanical Survey, and by his permission,

Miss G. Britten typed this paper.

There have been six more or less extensive accounts of South African species of Rhus

published, in which they have been either carefully described or their descriptions have

been short, but type specimens are mostly in existence. I refer to (1) Willdenow, Species

Plantarum I (1797), which represents the 4th edition of Linnaeus’ Species Plantarum ; (2)

Thunberg, Flora Capensis ed. Schultes (1823); (3) De Candolle, Prodromus II (1825) ;

(4) Ecklon and Zeyher, Enumeratio Plantarum Afr. Austr. Extratropicae (1835-37) ; (5)

Sonder, in Harvey and Sonder, Flora Capensis I, 504 (1859, 1860); and (6) Engler, in

De Candolle, Monographiae Phanerogamarum IV (1883).

A few remarks on some of these may not be out of place. Thunberg, who described

29. species, mistook Rh. laevigata L. His mistake has been perpetuated until now, and

I have been compelled to give it a new name, Rh. Legatv. His Rh. spicatum is Schmadelia

decipiens, which frequently is even now mistaken for a species of Rhusf [it was, e.g. described

by Schinz as Rh. knysniaca in Vierteljahrschr. d. naturf. Ges. Ziirich LX (1910) 238}.

Rh. cirrhiflorum, dimidiatum, tridentatum, and Rh. digitatum are species of Rhovcissus.

Rh. pauciflorum, Rh. alatum, and possibly Rh. obliquum are Hippobromus alata HK. et Z.

Rh. sinuatum Thunb. is unknown from South Africa, but agrees exceedingly well with an

Indian species (Rh. mysurensis Heyne).—Kcklon and Zeyher enumerate 48 species of S.A.

Rhus. An attempt is made by them to arrange them in natural groups, but this must be

looked upon as a total failure. They erroneously ascribe hermaphrodite flowers to most

of them. The number of species has been unduly increased by them and older names

have frequently been erroneously used. Their No. 1110, Rh. Plukenetiana K. et Z., which

does not appear to. have been mentioned by subsequent authors, is Rh. tomentosa L. Their

reference to one of Plukenet’s figures is wrong. This figure represents Rh. incanum Mill.,

as shown by Plukenet’s original in the British Museum.

Sonder with his usual minute care and taxonomic insight did a great deal to clear up

the synonymy of the species known to him and arranged them in accordance with an artificial

key which he constructed. He recognized 53 species. Of the new species described by

him I have had to rename his Rh. tridentata and have called it Rh. Fraserr. His numbers

47 to 52, Rh. Thunbergii Hook. (=Rh. argyrophylla Presl), Rh. dispar Presl, kh. concolor

* See Schonland: The genera Riws and Orassula in Thunberg’s Herbarium, Arkiv for Botanik X XI.

A (1927), No. 16. ;

+ E. Meyer, in Drége exsice. places it also under Rhus and has given it three different names. He

also refers to our genus all other South African species of Schmidelia ; further he does the same with

Myaris inaequalis Presl, Xanthoxylon capense Harv., and Sapindus oblongifolius Sond. as a reference

to Vol. I of Harvey and Sonder. Fl. Cap.. will show.

Presl, Rh. mucronifolia Sond. (=Rh. salicifolia Presl), Rh. salicina Sond., Rh. poniculosa

Sond., were later removed by Engler to the genus Heeria Meisn (= Anaphrenium E. Mey.)

and No. 53, Rh. longifolia Sond., to the genus Protorhus Engl.

Engler accepted almost all species recognized by Sonder. It goes without saying that

this was done after due consideration and, if there was any evidence required, it shows

itself in the much more detailed descriptions drawn up by Engler as compared with Sonder.

He must have overlooked Rh. magalismontana Sond. (a type of which is, e.g. in the

Herbarium of the S.A. Museum). He doubtfully referred Rh. sinuata Thunb., which Sonder

had seen in the Thunberg Herbarium, to Rh. refracta EK. and Z., and he raised Rh. dentata

Thunb. var. puberula Sond. to specific rank as Rh. Sonder1 Engl. However, the general

agreement is a striking testimony to Sonder’s discriminating and painstaking work, but it

further proves that our species of Rhus, in spite of their enormous range of variability,

are on the whole entities recognizable by the trained botanist. Engler in his monograph

described some additional varieties of some species and also 14 new species not included

in the Flora Capensis. About 20 additional species have been published since the appearance

of Engler’s monograph. I have seen types and descriptions of all of these with the exception

of Rh. dunensis Gandoger in Bull. Soc. Bot. de France LX (1913), 119, and Rh. tumulicola

S. Moore in Journ. of Bot. 1921, LIX, 227. The latter judging from the description

appears to be Rh. Zeyher.

Diels, in his essay entitled “ Die Epharmose der Vegetationsorgane bei Rhus L.

§ Gerontogeae Engl.” (in Engl. Bot. Jahrb. XXIV, 568), made an attempt to divide this

section into natural groups in which the Villosa group formed the centre of nine others.

He made, as mentioned before, extensive anatomical investigations, but in the delimitation

and treatment of the groups in detail he makes very little use of them except, that he lays

very great stress on the absence and, if present, on the nature of the indument which is

composed of ordinary and glandular hairs in variable proportions. He also emphasizes the

various methods of reduction in leaf-surface. While his results are undoubtedly an advance

on our previous knowledge, it will be seen later that I do not agree with them altogether.

Diels stated that, with the increase of material, the delimitation of the species, which

seemed on the whole to be comparatively easy at the time when Hngler wrote his mono-

graph, has become more and more difficult, and he pointed out that some previously

described species could not be kept up.

B.—Distribution.

Most S.A. species are found on the seaward side of the great escarpment (as defined

by Rogers in Botanical Survey Memoir No. 4, 1922, p. 10) with only a few in karroid portions.

Many are found in the Transvaal. Some are found in the Basuto highlands, the Orange

Free State, and other inland portions of South Africa, but even in parts with comparatively

good rainfall which the majority inhabit, there are some which prefer stony dry hillsides.

In fact a very intimate knowledge of local conditions is required to judge the relations of

climate and structure in this genus. In the arid parts of South Africa some species, such

as Rh. lancea and Rh. viminalis are well known to occur only in such places where their

roots can reach ground water, and it is, therefore, not astonishing to find that Cannon

found the highest rate of transpiration of Karroo plants in Rh. viminalis along stream beds

at Matjesfontein. Its leaves are very little protected by a resinous secretion and have no

fixed relation to the light (Cannon, W. A., General and physiological features of the

vegetation of the more arid portions of Southern Africa, with notes on climatic environment,

Washington, 1924, p. 145). On the other hand, there are others which grow in sand-dunes

by the sea exposed to physiological drought, e.g. Rh. Schlechteri, Rh. crenata Thunb., forms

of Rh. mucronata, etc.

C.—Habit.

The majority of South African species of Rhus are much branched shrubs from 8 to

10 feet high. A fair number of these may assume an arborescent habit reaching heights

of 20 to 30 feet, but only one, Rh. Legati (Rh. laevigata Thunb. non Linn.), may become a

conspicuous tree with a well-defined trunk reaching a height of 60 to 80 feet. Some, e.g.

Rh. lancea and Rh. viminalis, assume a willow-like habit. A few, e.g. Rh. rosmarinifolia,

Rh. discolor, Rh. gracillima, Rh. Wilmsix, Rh. Keetii, are dwarf shrubs, unbranched or very

little branched above ground. The first of these has assumed a somewhat heath-like habit

with its narrow leaflets recurved at the margin. The last four occur in grassveld and in

the last two especially an approach to a grass-like habit can be discerned.

Some species are always thorny, e.g. Rh. longispina HK. et Z. A number of others

are often thorny, e.g. Rh. pyroides. One often finds also dwarfshoots forming a transition

between ordinary shoots and thorns. In some species thorns are only found in seedlings

and coppice-shoots, e.g. regularly in Rh. Legativ. The vegetative organs vary considerably

in many species. Many species form coppice shoots readily and these shoots at first are

often very different from ordinary shoots. The variations in the ordinary shoots are

usually shown especially in the leaves, which will be referred to presently.

D.—The Leaves.

The majority of species have well-developed petioles, which are often subsemiterete,

canaliculate above. They are often slightly edged, and when this is readily discernible

they are described as winged.

Usually the leaves are trifoliolate. | Leaves with five leaflets are the rule in Rh. montana

Diels (as far as the scanty material shows), and are occasionally found in Rh. MacOwani

Schonl. Less than three leaflets are also occasionally met with.

In some species, e.g. Rh. tomentosa and Rh. Legati, the leaflets have petiolules, but

usually the leaflets are sessile. Ovate and obovate outlines, with a cuneate base, pre-

dominate, but narrowly linear and lanceolate shapes are found in a few species which, as

Diels rightly remarks (l.c. 594), are the final products of certain tendencies of evolution.

Unfortunately the leaflets yield the most important, though not the only characters, on

which the delimitation of species in this genus can be based, and it must be very annoying

to anybody who wants to study the genus to find a great vagueness in the descriptions of

even their shape. The consistency of the leaflets is more constant, but cannot always be

clearly seen in dried specimens. In one case, Rh. carnosula Schonl., the leaflets are slightly

fleshy, though when dried they appear coriaceous. The margin of the leaflets is often

more or less characteristic, but teeth (not always of the same shape) may be found in species

which have usually an entire margin and, not only in different individuals but even on

the same branch, differences in shape and margin of the leaflets may be found in many

species.

. The enormous plasticity in the shape of the leaflets may be due to Automorphosis

dependent on internal causes which we cannot trace, or, in other words, to variations in

the Darwinian sense. However, in some cases they are clearly due to hybridisation,*

though in many others hybridisation may be safely ruled out of court.

The nervation of the leaflets is fairly characteristic for the different species, and has

been indicated on most of the illustrations of leaves which accompany this paper. I have

distinguished in the descriptions between (a) midrib, (6) lateral veins which are often slightly

branched, and (c) tertiary veins, the finer ultimate ramifications. The last are often not

visible with the unaided eye or with the aid of a lens. They may even be absent altogether.

The reticulation may be coarse as in Rh. mucronata and Rh. nebulosa Schonl., or it

may be fine-meshed, but there are gradations which cannot always be expressed in the

descriptions. The trichomes found on the leaflets (as well as on the petioles, branchlets,

and inflorescence) are to a certain extent of taxonomic value, but in certain cases, e.g. in

Rh. mucronata they must be ignored as a means of specific distinction, Already Diels

stated (l.c. 605) this somewhat as follows :—The covering of trichomes cannot be traced

to exogenous conditions. There is no way to judge which circumstances can effect the

quality of the indument. ‘‘ We can only observe that within closely related species the

combination of ordinary hairs and glandular hairs varies enormously quantitatively. Espe-

* For supposed hybrids see pages 9, 28, 29, 37, 41, 42, 43, 47, 54, 58, 66, 75, 108, 112.

cially is this the case in Rhus § Villosa. Sometimes we see hairs and glands approximately

equally represented. Sometimes hairs preponderate especially in the more xerophilous

forms, which inhabit the eastern interior.” (I cannot follow his reasons in the next remarks,

which seem, on the one hand, to indicate a relation of the hairy covering to the rainfall,

but, on the other hand, contradict it.) Very hairy, subglabrous and glabrous forms of

Rhus mucronata are found almost side by side on Table Mountain. The most villous form

known to me from South Africa is Rh. EHrnesti Schon]. found at Barberton. Numbers of

species have very hairy young leaves, etc., but become more or less glabrous when adult.

Diels continues :—‘‘ In contrast to the hairs the glands (of which he illustrates some on

Taf. XIV) have a firmer existence. In thousands of specimens where the former are in the

process of dying or have disappeared altogether, the glandular trichomes remain in un-

- diminished number. Even if the lamina decreases in size, they become more numerous

per unit of area and there result races with chiefly glandular covering. | Which are the

conditions that favour such a development is impossible to me to even conjecture.”

Predominance of glands is especially pronounced in the Lucida group. That in some cases

it is associated with habitat m which the plants are subjected to physiological drought, as

in sand dunes or to arid conditions in the interior, seems to be evident. The glands frequently

secrete a resinous substance, which when dry and thin layered makes the leaflets look

“ varnished ” or when thick it may become grey or powdery, but this again can only very

cautiously be used for specific discrimination. Thus, Marloth says in “ Das Kapland ”’

(1908), 324, with reference to Rh. mucronata and Rh. lucida, “ on the leaves resin is rarely

seen near Capetown, but constantly in the Little Karroo and other dry parts. On the

other hand, Rh. glauca is richly provided with resin, even in the extreme South-West. The

young leaves are covered with a soft, sticky varnish which dries in summer and forms a

white crust. Several other species of Rhus (as well as a Psoralea and Conyza waefolia)

protect themselves in the same manner.”’

E.— Inflorescences.

The inflorescences are panicles, sometimes poorly branched, in which case they are

lax as, e.g. in Rh. mucronata and Rh. pyroides Burch. (non auct. al.). In other cases they are

richly branched and bear an abundance of flowers as, e.g. in Rh. MacOwani and Rh. Legati.

They may be axillary, usually in the axils of the upper leaves, or terminal. Usually both

axillary and terminal occur in the same species. These and similar characters are often of

great assistance in distinguishing allied species, but should be employed cautiously as

sometimes male and female differ in these points. The pedicels are usually about 1 mm.

long, though somewhat longer ones occur and sometimes they are shorter, the flowers then

forming glomerules. The floral bracts are short and usually narrowly lanceolate.

F.—Flowers.

The flowers are typically unisexual, the plants dioecious. Truly bisexual flowers occur,

but only as exceptions in usually unisexual species. Ecklon and Zeyher’s statements to

the contrary are not based on facts. The flowers are usually pentamerous, but variations

in the number of floral parts occur occasionally. The calyx has in the usual pentamerous

flowers 5 segments, which in some species are not equal. There are 5 petals, usually oblong-

ovate and greenish-yellow or whitish. Then follows a disk which is usually 5-crenate.

Very often it is slightly crisped, giving the appearance of 10 crenations, but sometimes

the disk is actually 10-crenate. In male flowers there are 5 stamens and often not a

trace of a gynaecium. On the other hand, the female flowers exhibit usually 5 staminodes.

The gynaecium has a more or less globose ovary with 3 short, separate, filamentous styles.

On the whole, there is such a want of differentiation in the structure of the flowers

that it is difficult to utilize their characters for taxonomic purposes. ven the size, though

always small, varies in the same species, yet when used with discretion, size can sometimes

be used as an aid in discriminating some allied species. Generally the length of the petals

is 14 to 14 mm., as e.g. in Rh. MacOwani Schonl. ; sometimes it reaches 2 mm. as, e.g. 10

Rh. mucronata Thunb.

G.—The Frut.

The fruit is a drupe with fleshy mesocarp, which, however, in some species dries up more

or less when the drupes ripen, while in others it remains juicy for a long time. In Rh. incisa

var. obovata the old drupes split open exposing the pyrena and we thus find here a transition to

a capsular fruit. The pyrena is usually distinctly compressed even in globose fruits. The colour

of the drupes is often greenish, sometimes red (as in Rh. Legati and Rh. dentata), sometimes

brown. The shape is often globose or subglobose, in some species it becomes occasionally or in

others always asymmetrical. Usually the drupes are quite smooth and glabrous. In the

Tomentosa and Populifolia groups hairy drupes are found, in the latter also verrucose drupes.

In the Populifolia group tricuspidate drupes occur, through the retention, hardening, and

slight thickening of the styles. This is also occasionally found in other groups.

H.—Uses.

Some species as, e.g. Rh. crenata help greatly in fixing coastal sanddunes, but I am

not aware that this property has been used by planting these species. Many species favour

streambanks, e.g. Rh. mucronata, Rh. MacOwani, Rh. dentata, and in the dry parts of South

Africa, Rh. lancea, Rh. viminalis, and other members of the Lancea group. They thus

prevent useful soil being swept away. On the other hand, a great deal of damage

has been done, where wood is scarce, by the wholesale destruction of these species for

firewood and other purposes. It is very desirable that extensive experiments should be

made to see whether they can be restored.

The bark of some species is used for rough cordage, hence the name Taaibosch.

The wood of most species is tough (frequently of a reddish colour), and various species,

e.g. Rh. mucronata, Rh. lucida, Rh. pyroides, and even Rh. Engleri, are used for making

kerries and pickhandles. Whipsticks are made of Rh. mucronata. A systematic examina-

tion of these woods is not available. They are usually passed by, because their trunks

rarely reach proper timber-size, and they are often found where better timber-trees are

available. Hven Rh. Legati (Rh. laevigata Thunb. non Linn.), which grows into a large

tree is not favoured by sawyers, though its wood is often used for wagon work. According

to Pappe, in “ Sylva Capensis ” (1869), the thicker and longer branches of Rh. viminalis

(Karee wood) are used as spars in thatching houses and also for wagon tents, as they bend

easily without breaking; the younger twigs for bows.

In former years the bark of Rh. lucida (Taaibosch or Cape Sumach) and Rh. tomentosa

was used for tanning purposes. Various species are locally used as hedge-plants ; I would

especially recommend Rh. erosa, which has a very striking appearance and should clip well.

Goats and Persian sheep have been noticed by me to eat the foliage of Rh. undulata and

Rh. longispina, and others are also reported to be eaten by stock, but as pasture plants

they will never be of much consequence except as a last resort in the most arid portions

of South Africa. Various species are reported to have edible drupes, but to appreciate this

statement one must know what other astringent fruits South African children will eat and

apparently enjoy. There is no hope that any of our species will yield a fruit that will be

generally acceptable.

I.—The Phylogeny of South African Species of Rhus.

D. H. Scott Jin Progressus Rei Botanicae (1907), 139] rightly stated that the determina-

tion of the actual course of descent is the ultimate or chief object of the scientific systematist.

The course of evolution in South African species of Rhus may be here and there traced

or, more accurately speaking, surmised, but its cause and its mechanism are for the present

quite hidden from us. At first sight the cause in many cases seems to be laid bare by the

fact that forms, such as we find, e.g. in the Tomentosa group, which deviate much from the

supposed original type, have many characters which have enabled them to adapt themselves

to a drier climate or to greater extremes of climate than the typical forms. They have

become more xerophytic, and it is evident that much of the development of South African

species of Rhus, as in so many other South African plants, has taken place in the direction

from more mesophytic types to more and more xerophytic types of the vegetative organs

by reduction in the breadth of the leaflets, thickening of the cuticle, exudation of resinous

substances, recurving of the lamina in narrow leaflets, sunk stomata, etc. However, it is

quite plain that these and similar characters, which enable some species to exist in some of

the most arid parts, are not directly caused by the climate. This statement cannot be

proved accurately, though it is significant that, e.g. the densely villous Rh. mucronata var.

villosa and Rh. Ernesti occur in parts with a considerable amount of rainfall. We find one

character, especially though not exclusively, developed in the Populifolia group, which is

only found in species of dry localities. I refer to the retention of the styles and their slight

thickening in these species. This is an evolutionary tendency for which no reason can at

present be assigned. Another evolutionary tendency, a hairy covering of the fruit has

found its clearest expression in the Tomentosa group, which, however, are found in localities

where other species with glabrous drupes are found.

How difficult it is to correlate the conditions under which species with diverse characters

grow may be illustrated by the following example : Near my home, Aylesby, near Grahams-

town, there are growing wild, under apparently identical conditions, Rh. MacOwani, Rh.

undulata, Rh. lucida, Rh. longispina, Rh. incisa var. obovata. Half a mile away Rh. tomentosa

is found under apparently the same conditions.

The clearest indication that characters which are most useful in dry localities are not

necessarily caused by arid conditions is given by the thorny species, for the most formidable

thorns are found in seedlings and coppice shoots of Rh. Legati, a species which is never

found in arid parts or dry localities. To quote another instance :—Rh. longispina and Rh.

undulata are often growing side by side, the former armed, the latter unarmed. Both are

browsed upon like many other species of Rhus by goats and other herbivorous animals.

The extreme innate variability of the leaves of many South African species of Rhus

has already been referred to. In some of these cases at all events, hybridization seems to

be clearly excluded, in many others it is not likely to be the cause of variation, yet there

can be no doubt that some species, e.g. Rh. dentata hybridize freely.

Only one with an experienced eye can recognize the hybrids, or rather he can guess

that he is dealing with hybrids, because proof could only be furnished by very lengthy

cultural and cytological investigations.

Diels has assumed that his Villosa group has formed the starting point for all other

African groups of the genus. He has made out a strong case for this assumption and we

have accepted it. His Villosa group corresponds largely to the Mucronata group of the

present paper. Diels imagines that the Rhus § Gerontogeae, which are found in the whole

of Africa, parts of Makaronesia, South Arabia, Syria, British Hast Indies and territories

adjoiing in the North East, Socotra and the Comore Isl., but not in Madagascar, have

entered Africa in the North. He draws an interesting parallel of their origin and migrations

with that of the Antelopes. The Mucronata group covers almost the whole of the distribu-

tion of the section, but already the Natalensis group, which it is assumed has been derived

from it, has almost as wide a distribution, though in South Africa it does not go beyond

districts which have a subtropical climate. There are an enormous number of forms in

Tropical Africa which have been (to my mind wrongly) referred to Rhus villosa L.f., while

other distinct tropical species also belong to Diels’ Villosa group. Rh. natalensis Bernh.

(Rh. glaucescens Rich.) is also much differentiated in Tropical Africa, but the material of Rhus

hitherto collected in Tropical Africa, though very large is mostly defective and usually not

accompanied by sufficient notes. It cannot be too often pointed out that scraps taken

from a tree or shrub with nothing but the locality are practically useless for taxonomic

purposes or at all events give more trouble than they are worth, besides often creating

confusion. Starting from the Mucronata and Natalensis groups further differentiation

took place already in Tropical Africa while the genus spread southwards, but the largest

differentiation clearly took place after it reached South Africa.

I have arranged the South African species into twelve groups. Of these, m addition

to the Mucronata and Natalensis groups, the Lancea and Discolor groups extend a good deal

into Tropical Africa ; some others also are not strictly confined to Temperate South Africa,

though their origin may be looked for at all events considerably south of the equator.

The Discolor, Tomentosa, Populifolia, and Horrida groups are more removed from the

original type than the others, and may be the youngest groups. In all of them a xerophytic

tendency is clearly discernible, especially in the three last ones. The Discolor group shows

relations to the Tomentosa and Populitfolia groups, while the Horrida group may have been

derived from the Lancea group. While these general relations can be observed, I fail to

see how a more detailed evolutionary scheme can be presented from available data.

Diels (l.c. 644) says :—“ The gerontogaeous species of Rhus furnish as it were a type,

which in the generative sphere has been stabilized since immemorial times, but whose life-

energy is represented by the elasticity of its vegetative organs in response to external

imfluences. Therefore, according to him, one succeeds here in picturing much more easily

the conditions for one kind of polymorphism, the epharmonic one, than is possible in the

majority of objects, where the greater complexity of all circumstances forces itself at once

on the observer. Thus, the Rhus § Gerontogeae furnish an interesting instance, how in many

cases epharmosis, here equivalent with what is usually known as formation of species, has

been brought about.”

Now, in the first place, the epharmosis of the vegetative organs of our species of Rhus

is not the only, though the most important point in their differentation. Further, the

detailed account of the epharmosis of the vegetative organs, while often most interesting,

and here and there no doubt rightly indicating the course of evolution, is to my mind

wrongly interpreted or at least doubtful m other cases, and with the more detailed

knowledge now available, one may say that the attempt at getting at the cause of this

evolution is a failure and the mechanism of evolution has not been touched upon. In fact,

though the study of the genus attracted me from a genetic point of view especially, I must

frankly confess that I have come to the conclusion that morphological (including anatomical),

geographical, and ecological facts are not sufficient to give a clear insight into the evolution

of our species of Rhus. Perhaps cultures on a large scale and cytological investigations,

neither of which I have touched, will help to bring about the desired result. Under the-

circumstances I do not feel justified in following Diels’ example by trying to form as it

were a genealogical tree of the species. At the same time I can recommend Diels’ account

to students of the genus, who cannot fail to be stimulated by it, and who will find many

interesting facts im 1t which I have not included in this paper.

While thus, to recapitulate, the evolution of our species of Rhus for the present largely

evades our grasp, it seems evident that the majority of the South African species have

been evolved in South Africa, and this, as stated before, applies also to some of the groups.

A separation of the Mucronata and Natalensis groups evidently has, however, taken place

long before the genus reached South Africa.

To show the probable relationship of the groups represented in South Africa the following

scheme has been drawn up :—

Miucronaiaas eae Natalensis.....Lancea..... Horrida.

Dentata.

| Discolor... . .Tomentosa.

Refracta. Eckloniana. |

| Populsfolia.

Lucida.

Legatv.

Snort Descriptions oF GRouPS AND Krys oF THE SPECIES.

MUCRONATA group.

Shrubs or rarely trees, unarmed or rarely thorny, often more or less hairy, but the

hairs in many species become less or disappear altogether on the adult leaflets and other

adult organs. Petioles canaliculate above, rarely slightly winged. Leaflets membranous

or rarely subcoriaceous, generally entire, more or less obovate, rarely sublanceolate, tertiary

veins generally distinct (unless obscured by hairs) and reticulate. Panicles lax or densely

multiflowered. Drupes glabrous, subglobose or slightly compressed.

DistRIBUTION : Widely spread in South Africa, absent in the arid north-western and

western parts; also widely spread in Tropical Africa.

A.—Panicles lax, shorter or slightly longer than the leaves,

(a) Leaflets oblong-lanceolate or oblanceolate.

a Leaflets at first villous, later pilose.......... 11. Rh. Rudatisu Engl.

BmlicatlebsmClADTOUS <./eicic ae sere edie a 6s aecveesye 1. Rh. mucronata Thunb.,

B. laevigata (L.) Schonl.

(6) Leaflets broad, usually obovate.

a Adult leaflets densely villous.

1. Hairs fairly uniform all over............ 4. Rh. mucronata Thunb.,

E. villosa (L.f.) Schonl.

2. Midrib and the larger veins covered with

lighter straights hams. 22 saat ae 2. Rh. Ernest Schoul.

B Adult leaflets more or less hairy or glabrous.

1. Drupes 5-5-5 m.m. in diam............ 1. Rh. mucronata Thunb.

2. Drupes 3-34 mm. in greatest diam.

x. Mature leaflets quite glabrous, cori-

aceous.

o. Branchlets subterete........... 10. Rh. dura Schon.

oo. Branchlets angular.......... 9. Rh. krebsiana Licht.

xx. Mature leaflets more or less covered

with fine, often appressed hairs.

o. Tertiary veins very coarsely re-

ticulate. Leaflets subcoriaceous.

oo. Tertiary veins delicately reti-

culate. Leaflets membranous.

+ Petioles less than half the length of

the leaflets....... 6. Rh.

TT Petioles slender, more , than “half the

length of the terminal leaflets.... 7. Rh.

io)

Serelte

nebulosa Schonl.

pyroides Burch.

Baurvi Schonl.

B.—Panicles densely multiflowered, terminal much longer than

the leaves and greatest diam. of fruit about 3 mm.

(a) Leaflets oblong or obovate, acuminate, cuneate (the

cuneate base often at least half the length of the

leaflet), often with a sch. gross mucronulate teeth

at the apex. Petiole 1-2 cm. long............. 4. Rh, Fraseri Schonl.

(6) Leaflets ovate or obovate, obtuse, acute or acuminate,

generally entire.

GmuetiolessabouuMl-oaGM LONG acs elt pas 3. Rh. MacOwani Schon.

fo Penales Des—1h Gin, Ikons esis 6 oe ae aun ool clo 5. Rh. intermedia Schonl.

DENTATA group.

Glabrous or hairy shrubs. Petiole canaliculate above, rarely slightly winged. Leaflets

membranous or coriaceous, rarely slightly fleshy, obovate, ovate, oblong-cuneate or cuneate,

“usually in the upper portion crenato-dentate. Tertiary veins reticulate. Panicles multi-

ramose, lax or densely multi-flowered, the axillary shorter than the leaves, the terminal

longer.

DistripuTion : Chiefly on the edges of open bush and forests and in stream-bank

scrub in the coast districts from the Zitzikamma to Swaziland, in the mountainous parts of

south-eastern Cape Province and Natal, extending along the Drakensberg to the northern

Transvaal, ; also in the eastern Orange Free State.

A.—Quite glabrous. Leaflets oblong or obovate, more or less

grossly dentate in the upper part, slightly fleshy when fresh.

IDTOS Glo@wIA |) WA, WN CUMIN, ..oscgoncsacdagevcg00buat 13. Rh. carnosula Schon.

B.—Branchlets pilose. Leafflets oblong-cuneate or obcuneate,

coriaceous, upper part more or less crenato-dentate..... 14. Rh. Rogersw Schonl.

C.—Branchlets glabrous or subglabrous. Leaflets oblong-ovate,

UTA TTT ar CAME MIUIT Chee gee eyes. aaah a ae eh Yopee Ay RR pee 15. Rh. ntsubanensisSchonl.

C.—Quite glabrous or more or less hairy. Leaflets membranous,

very variable in shape, usually obovate-cuneate and more

or less dentate, if entire much smaller than in Rh. ntsub-

anensiss, |Drupe about 4 mms imnydianiiec serene eee 12. Rh. dentata Thunb.

REFRACTA group.

Richly branched, usually unarmed shrubs with comparatively small leaves. Petioles

not winged, rarely slightly exceeding 1 cm. in length. Leaflets membranous or subcoriaceous,

ovate, obovate or oblong, more or less hairy when young, subglabrous or glabrous when

adult. Terminal leaflets rarely exceeding 2-2 cm. in length. Panicles lax, rarely multi-

flowered. Drupe, where known, subglobose, fleshy, glabrous (in one species tipped with

the styles).

DistRIBUTION : Coast districts of South Africa from the Cape Peninsula to Natal and

mountainous parts of south-eastern Cape Province and Natal, slightly extending into the

Orange Free State and the Transvaal. One species in the South-West Protectorate,.

Of the species belonging here, Diels has only classified Rh. crenata. He places it (l.c. 630)

into his Crenata group with Rh. natalensis. This, to my mind, is an artificial association,

though in both glandular hairs are predominant on the leaflets, the margins of which are

crenate. The group, as here constituted, is very close to the Mucronata group. Rh. refracta

especially can hardly be kept out of it. The Refracta group might also be united with the

Eckloniana group.

A.—Petioles about 2 mm. long. Leaflets obovate-cuneate,

HUGE Me WH lol BIO oogcasasncooabadacnsncese 19. Rh. crenata Thunb.

B.—Petioles 5 or more mm. long.

(a) Adult leaflets greyish pubescent. . 18. Rh. Dinterc Engl.

(6) Adult leaflets dark green above, cinta or fulvous

tomentosesbelow rt ete aces sae etme emce 21. Rh. diwaricata Ki. et Z.

a Leaflets entire, oblong or elliptic-lanceolate, ter-

TaN PH8y Cian OVE Csscey oo ceoehssosoneos 17. Rh. fastigiata HK. et Z.

B Leaflets obovate-cuneate, usually entire, rarely

crenulate towards the apex, terminal | -2—2-2

em. long, somewhat larger in specimens

PLOWING mear the iseameer ii sat ete Tot 16. Rh. refracta HK. et Z.

y Leaflets elongated obovate-cuneate, tridentate

or emarginate at the blunt apex, terminal 2-3

em. long’ 41). Anne ahi: Seema cas eke torn 20. Rh. Penther: Zahlbr.

0 Leaflets broadly obovate or obovate-elliptical,

entire, terminal about 1-3 cm. long........ 22. Rh. rupicola Wood et

Evans.

LEGATI group.

Shrubs or trees, mostly glabrous in all mature parts. Petioles not winged. Leaflets

membranous or subcoriaceous, usually quite entire, obovate or obovate oblong, acuminate,

narrowed at the base, sometimes petiolulate, tertiary veins reticulate. Panicles large and

much ‘divided, multiflowered in one species, very lax and usually smaller than the leaves

in the two others. Drupes shining, glabrous, subglobose or globose.

DistriputTion : Rh. Legati (Rh. laevigata Thunb. non Linn.) is found in all forest

patches and often in stream-bank bush from Swellendam to Natal and the northern

Transvaal; the two others are restricted to the eastern parts of the Drakensberg and the

neighbourhood of Pretoria.

Diels (l.c. 638) refers all three species to his Laevigata group, but also puts under this

group a number of other species which I have placed elsewhere.

A.—Perfectly glabrous or rarely with slightly pilose branchlets.

(a) Leaflets green or partly reddish, shghtly undulate, the

(enumioell 71S} oun, Moye, oc opancanesmoesaeoTe de 23. Rh. Legati Schonl.

(6) Leaflets glaucous, the terminal 2-3 cm. (rarely up to

TOP CHO MLOUD) Ber srpspckt tote calories c-atestee a ace os cue teste 25. Rh. Zeyheri Sond.

B.—Branchlets at first villous, later glabrescent. Leaflets

green, the terminal 2-5-4 cm: long.................... 24. Rh. transvaalensis Engl.

LUCIDA group.

Much branched unarmed shrubs, usually with predominance of glands over hairs.

Petioles channelled above and more or less winged. Leaflets often “‘ varnished ”’, sub-

coriaceous or coriaceous, rarely membranous, obovate-cuneate, lanceolate-cuneate or

obcordate-cuneate, entire or rarely dentate, with lateral veins distinct, butt ertiary veins

rarely visible. Panicles lax, shorter or slightly longer than the leaves.

DistTRIBUTION : Common in open bush and on the edges of forest in the coast districts

from south-western Cape Province to Natal and Gazaland. One species widely spread in

the more arid portions of western and central portions of South Africa extending to the

South-West Protectorate.

Diels (l.c. 632) has placed most of the species here united into his Lucida group, to

which he also refers Rh. horrida, Rh. longispina and Rh. cuneifolia. On the other hand, he

placed Rh. celastroides Sond., which, in my opinion, is only a variety of Rh. undulata, into

his Pyroides group next to Rh. mucronata.

A.—Leaflets lanceolate-cuneate, acute or emarginate, entire,

rarely ovate, dentate, membranous in the coastal forms.. 32. Rh. undulata Jacq.

B.—Leaflets obcordate-cuneate.

(a) Greatest diam. of drupe about 3 mm............... 32. Rh. undulata Jacq. var.

(6) Greatest diam. of drupe about 5mm............... 31. Rh. glauca Desf.

C.—Leaflets more or less obovate-cuneate.

(@) pBecatletsmpracticalliyisessilesse srairi-t2 <r 1-1-2 27. Rh. Schlechteri Diels.

(6) Leaflets distinctly petioled.

a Petals 14-14 mm. long.

1. Petioles slightly winged. Leaflets sub-

coriaceous with lateral veins delicate,

NCU CHADUCBObon aby ooh ede BBO eo ce 26. Rh. lucida L.

2. Petioles broadly winged. Leaflets cori-

aceous, lateral veins usually prominent

Qua boch@cuniacestess tiie cle 30. Rh. scytophylla E. et Z.

B Petals about 2 mm. long.

1. Glabrous. Leaflets with thickened white

ROVOLKS MVM no oenodoangokdoodoogs 29. Rh, albomarginata Sond.

2. Branchlets, panicles, ete., hairy......... 28. Rh. africana Mill.

ECKLONIANA group.

Shrubs, one species dwarf, rarely small trees, usually unarmed and usually with

comparatively small leaves. Petioles subterete or slightly winged. Leaflets membranous

or subcoriaceous, glabrous or pilose, entire or slightly crenato-dentate, oblong, lanceolate

or oblong-lanceolate rarely exceeding 3 cm. in length. Panicles lax, shorter or longer than

the leaves. Drupes glabrous, subglobose, shining.

DistRiBuTION: Transvaal, one species also in the Cape Province from the Uitenhage

to the Albany District.

Diels (l.c. 627) placed Rh. Engler: Britt. (= Rh. incana Engl.) into his Pyroides group,

without, however, indicating close relationships with any other species. Rh. eckloniana

Sond. was not classified by him.

A.—Branchlets and leaflets greyish pilose. Drupe about 3 mm.

TD VAIN. ..tees paces cA Ros em ashe rate a ait Gr on ele ee a oer 34. Rh. Englert Britt.

B.—Branchlets pubescent. Leaflets deep green above, paler

below, sparsely pilose. Drupe about 3 mm. in diam.... 35. Rh. ebwrnea Schon.

C.—Branchlets and leaflets glabrous or more or less pubescent.

Leaflets deep green above, slightly paler below, often folded

along the midrib. Drupe 5-5-6 mm. in diam.......... 33. Rh. eckloniana Sond.

NATALENSIS group.

Shrubs, rarely small trees, often with more or less greyish or glaucous foliage. Petioles

subterete or broadly canaliculate above. Leaflets subcoriaceous, generally more or less

oblong, rarely ovate or obovate, almost always crenate, glabrous or more or less pubescent.

Tertiary veins rarely distinctly reticulate, often not visible. Panicles generally lax and not

exceeding the length of the leaves or slightly longer. Drupes glabrous, shining, in two

species sometimes retaining the styles (compare Populifolia and Horrida groups).

DistriBpuTIoNn : South-West Protectorate, Bechuanaland, Transvaal, coast of temperate

South Africa from Hast London to the Tropics, and strongly represented in many parts of

Tropical Africa.

Diels places Rh. natalensis into his Crenata group and Rh. Marlothii into his Damarenses

group.

A.—Leaflets ovate or obovate, greyish, softly pubescent..... 40. Rh. commuphorordes

Engl. et Gilg.

B.—Leaflets oblong-cuneate, dark green above, glabrous..... 36. Rh. natalensis Bernh.

C.—Leaflets obovate oblong or sublanceolate or oblanceolate,

often glaucous, puberulous or glabrous.

(a) Drupe strongly compressed, often asymmetrical and

sometimes with persistent styles............... 39. Rh. Marlothii Enel.

(6) Drupe subglobose, c. 3:5 mm. in diam.............. 37. Rh. Simii Schonl.

¢. 4. min. inh diana teenie accel scart: 38. Rh. spinescens Diels.

LANCEA group.

Shrubs or small trees, usually glabrous or subglabrous, mostly unarmed. Branchlets

generally slender, rarely whole plant slender and almost unbranched. Petioles canaliculate,

sometimes slightly winged. Leaflets linear or lanceolate-cuneate, rarely oblong, sometimes

slightly faleate, entire, dentate or crenate, when mature generally glabrous or subglabrous.

Panicles lax. Drupes glabrous, subglobose, sometimes slightly compressed or depressed.

DistRIBUTION : Chiefly in the Transvaal and Natal (above 3,000 feet), mountains and

hilly parts of south-eastern Cape Province, and through the arid interior parts of South

Africa to Clanwilliam and Namaqualand, one species extending to Tropical Africa.

Diels includes Rh. ciliata in his Pyroides group (l.c. 627) and Rh. gracillima in his

Tomentosa group (l.c. 613); the other members of the Lancea group, as far as they have

been classified by him, he places into his Laevigata group (l.c. 638)—RhA. viminalis is allied

to Rh. retinorrhoea Steud., a Tropical African species.

A.—Slender, almost unbranched shrublets.

(a) pBetrolesm om guitle LONG tetas soit vet Ses emia ohn 51. Rh. gracillima Enel.

(b) Petioles 2-5 em. long.

a Leaflets suddenly contracted at the apex, lateral

veins prominent on both surfaces.......... 52. Rh. Wilmsw Diels.

B Leaflets acute at the apex, lateral veins a little

prominent or more often immersed......... 53. Rh. Keeti Schonl.

B.—Much branched shrubs or trees.

(a) Margin of leaflets crisped and sometimes shghtly and

feregularhyadentatesaans =. saci hs tok setae 46. Rh. crispa Harv.

(b) Leaflets not crisped.

a Margin of the leaflets eroso-dentate.......... 49. Rh. erosa Thunb.

f Margin of the leaflets serrato-dentate, leaflets

obovate oblong or oblong................. 50. Rh. Bolusii Sond.

y Margin of leaflets with a blunt tooth at the end of

cachwlateralvieinc ny wikietvstilys Acme en: o 44, Rh. Gueinzir Sond.

6 Margin of leaflets irregularly and usually sparsely

CTEMALOsCeMbAtEl «jot scher ns xcia/ertel shore chs ecideneress 43. Rh. Gerrard, Harv.

< Margin of leaflets entire.

1. Leaflets sessile.

x. Branchlets subvillous. Leaflets ob-

long or obovate, when mature

sparsely pilose. . . 45. Rh. microcarpa Schonl.

xx. Quite glabrous or with puberulous

branchlets. Leaflets linear or lance-

olate, rarely oblong, glabrous or

puberulous, rarely exceeding 2-5 cm.

TuMblenophiee se seat Steve lance: 47. Rh. ciliata Licht.

xxx. Quite glabrous. Leaflets narrowly

linear, often falcate, the terminal

3:5-6 em. long.. ; ... 48. Rh, dregeana Sond.

2. Leaflets petiolulate, quite ‘glabrous or

sparsely hairy.

x. Terminal leaflets 4-8 cm. long..... 42. Rh. viminalis Vahl.

xx. Terminal leaflets 9-12 cm. long.... 41. Rh. lancea L.f.

HORRIDA group.

Squarrose or erect, thorny or unarmed shrubs, with young parts covered with glandular,

often red, hairs or tomentose. Petioles winged or margined. Leaflets coriaceous or sub-

coriaceous, obovate-cuneate or linear-cuneate, entire. Panicles lax, usually shorter than

the leaves. Drupes glabrous, subglobose or oblique, often crowned with the hardened

styles.

DistRiBUTION: One species in the semi-arid parts of the coast districts from

Swellendam to Kast London, penetrating inland to Graaff-Reinet, two in Namaqualand and

Bushmanland, one in the Transvaal.

Diels (l.c. 632) places Rh. longispina next to Rh. lucida in his Lucida group, while

Rh. horrida was placed by him in the same group next to Rh. rigida, which he considers as

derived from Rh. scoparia. I do not think these relationships hold good. Rh. longispina

and Rh. horrida seem to be closely related. The former again appears to be related to

Rh. ciliata Licht. (in which ordinary hairs are often absent), and thus there would he

established a connection with the Lancea group. Rh. rigida has been placed in the Horrida

group, because like the two previously mentioned species it 1s also without ordinary hairs,

though the glandular hairs are not so conspicuous. Its drupe bears the thickened and

hardened styles, as is often the case in Rh. horrida and longispina (compare also the

Populifolia group). In its vegetative organs it approaches closely some forms of Rh.

magalismontana, but I am somewhat doubtful whether this should have been included

here, because its drupe does not bear the styles and the young parts are shortly tomentose.

Diels placed it (l.c. 627) as Rh. burkeana next to Rh. ciliata in his Pyroides group.

A.—Young parts shortly tomentose, drupe not bearing the styles.

Wnarmed ‘shrubs susan cutee econ 6. anne cara aay, 57. Rh. magalismontana

Sond.

B.—Young parts glandular.

(a) Erect, rigid, unarmed shrub. Petioles narrowly

margined. Leaflets cuneate-lanceolate or lanceo-

latetiitermumalg4 pacman on Cees n ene 56. Rh. rigida Mill.

(6) Squarrose, usually thorny shrubs. Petioles usually

distinctly winged.

a Terminal leaflets 4-8 mm. long.............. 54. Rh. horrida EH. et Z.

f Terminal leaflets 1-5 cm. long (usually about

POBIDEK GIN: Vaca ea ia eel a te matic Sch yg, 8 0 55. Rh. longispina HK. et Z.

DISCOLOR group.

Small, unarmed, poorly branched shrubs, one species more or less tomentose, the other

glabrous. Leaflets subcoriaceous, linear-lanceolate, lanceolate or oblong, rarely obovate,

always more or less cuneate at the base, entire or sparingly toothed. Lateral veins

numerous and often very conspicuous. Panicles dense or lax, shorter or slightly

longer than the leaves. Drupes subglobose, sometimes puberulous when young.

DistRipuTion : Amongst grass in the mountains of south-eastern Cape Province,

Transkei, Pondoland, Natal, Drakensberg Range, Orange Free State, Transvaal.

Diels (l.c. 614) placed Rh. discolor into his Tomentosa group. The group includes some

species found in Tropical Africa.

A.—Leaflets (at least on the under side) tomentose......... 58. Rh. discolor EK. Mey.

Bi heaflets claibrous. 2-194 e sn ae ee aes eee OO elu ee DON COCn SES SCHON.

TOMENTOSA group.

Unarmed shrubs (some dwarf) or small trees. Petioles subterete. Leaflets tomentose

on the under side, glabrous or subglabrous when mature on the upper, generally subcoriaceous,

elliptical, ovate, obovate, linear or linear-lanceolate, entire or slightly dentate or pimnatifid,

sometimes petiolulate. Panicles longer than the leaves, densely multiflowered. Drupes

villous or densely greyish or fulvous tomentose (very rarely glabrous), subglobose or

somewhat oblique.

DistripuTion: Hills of the coast districts from Namaqualand to the Transkei, but

mainly south-western.

Diels (l.c. 613) places all the species here enumerated into his Tomentosa group, to

which he also refers a number of other species which I have excluded.

A.—Leaflets narrowly linear or linear-lanceolate, entire or with

ONCROGREW OMS HALE beet Meet scricc pace ayeyetoclicyseis Ate sucvalsiral chs 63. Rh. rosmarinifoliaV ahl.

B.—Leaflets lanceolate or narrowly elliptical............... 61. Rh. angustifolia L.

C.—Leaflets more or less ovate or obovate in outline, rarely

elliptical.

(a) Leaflets entire or above the middle, more or less coarsely,

serrate. Drupes greyish tomentose............. 60. Rh. tomentosa L.

(6) Leaflets more or less pinnatifid with obtuse lobes or

nearly entire with small triangular often obtuse

teeth, rarely quite entire. Drupes villous...... 62. Rh. ineisa Lf.

POPULIFOLIA group.

Unarmed shrubs. Petioles subterete, slightly canaliculate above, sometimes slightly

winged. Leaflets coriaceous or subcoriaceous, obovate, suborbiculate or subrhomboid,

often cuneate or even petiolulate at the base, glabrous or subglabrous above, tomentose or

glabrous below, crenate, dentate or even pinnatifid, rarely entire, midrib and lateral veins

prominent, especially on the lower side. Panicles lax, shorter or longer than the leaves.

Drupe oblique, crowned by the slightly enlarged and hardened styles, glabrous or puberulous,

smooth or verrucose.

DistRiBuTion: Caledon, Stellenbosch, Worcester, Malmesbury, and Clanwilliam

districts, Namaqualand, Bushmanland, South-West Protectorate.

Diels (l.c. 613) includes Rh. populvfolia (with which I unite Rh. Steingroevert Engl.) and

Rh. dissecta in his Tomentosa group, while Rh. cunerfolia is placed by him into his Lucida

group between Rh. lucida L. and Rh. scytophylla H. et Z., for which I can see no adequate

reason, as he himself has pointed out (l.c. 596) that no reliance can be placed on the

indument in dealing with the taxonomy of Rhus.

A.—Petioles longer than the terminal leaflets............... 65. Rh. dissecta Thunb.

B.—Petioles shorter than the terminal leaflets.

(a) Greatest diam. of fruit 5-6 mm. Petioles 3-1 cm. long. 64. Rh. populifolia EK. Mey.

(6) Greatest diam. of fruit 6-8 mm. Petioles }-4 mm. long, 66. Rh. cuneifolia Thunb.

MUCRONATA group.

1. Rh. mucronata Thunb. (charact. emend.).

Remarks on Rhus villosa L.f., Rh. mucronata Thunb., Rh. laevigata L. (non Thunb.),

and some other forms distinguished by various authors as separate species.

Rh. villosa L.f. [suppl. 183 (1781)] is supposed by many authors to occur from the

south-west corner of Cape Colony to Abyssinia. Diels, in his valuable evolutionary studies

on Rhus § Gerontogeae Engl. (in Engl. Bot. Jahrb. XXIV, 1898, 586), takes this species as

a starting point from which the majority of African species are supposed to have been

derived by adaptation. It is, therefore, of some importance to determine as accurately as

possible what is meant by Rh. villosa Lf.

In the Linnean Herbarium, London (No. 26), there is a villous specimen named

Rh. villosum. Unfortunately the name is in Smith’s handwriting not in that of Linn. f.

However, I am quite prepared to take this as the type of Linn. f., who first published it,

though the name was given by Thunberg, from whom Linn. f. received many of his specimens.

In Jacquin’s herbarium the same form is named Rh. villosa. In the herbarrum of Thunberg,

Upsala, there are two sheets marked Rh. villoswm. On fol. a the leaflets are not villous,

they are only slightly hairy. On fol. 6 the young branches, petioles, leaflets, etc., are villous,

but it seems to me to be the species which I have called Rk. MacOwani, with which it shares

the pyramidal many-flowered terminal inflorescence and the many-flowered lateral inflor-

escences. In Willdenow’s herbarium, Berlin, there are specimens received from Thunberg

which are a similar mixture as found in Thunberg’s herbarium. In the Sloane collection

at the British Museum there is a specimen marked “* Rhus incanum Mill. dict. ed. 8 n. 8

(1760), Ray H. S. D 58, 14 Tab. 219, fig. 8”. The last reference is to Plukenet Alm. 319

t. 219, fig. 8 (1691). It is evidently Plukenet’s own specimen, but without flowers and

fruit. However, it agrees sufficiently with the specimen in the Linnean collection that

J. Britten was induced to dig out Miller’s name, Rh. incana, and substitute it for Rh. villosa.

(At the same time he changed Rh. incana Hngl. into Rh. Engleri—see Journ. of Bot.

XXXVIII, 1900, 316.) This, however, is only a side issue. The origin of Plukenet’s

specimen must be sought for in the neighbourhood of Capetown, and there we find the same

form still.

It looks at first sight as if it was a pity that the very expressive name Rh. villosa has

to be dropped, but even near Capetown this species is not always villous, and this introduces

us to one of the difficulties encountered in dealing with this species. Already Sonder

(l.c. 510) had distinguished a var. glabrata, which he identified with Rh. pubescens Thunb.*

(Prodr. 52, Fl. Cap. ed. Schultes 265.) In this “ variety ” the leaves are glabrous or sub-

glabrous. In March, 1925, two forms were sent by Dr. Marloth (11922 a, slopes of Table

Mountain, shrub, 8 feet, and 11922 B, ib., compact low shrub, 3 feet) with the followmg

remarks :—‘‘ On a walk yesterday I found two bushes of Rhus‘close together. Though

quite different in habit I found on closer examination no other difference but their hairiness.

A would be Rh. villosa (by a slip Marloth put tomentosa), B would be Rh. mucronata, but

even this shows a few hairs on the petioles. The fresh fruits of both are the same, globular

slightly flattened on the poles. A still bears the galls of the insect which produces the

“jumping eggs’. B has slightly but distinctly winged petioles which seem to go frequently

with the glabrescent form, but even A is not so distinctly villous all over the leaflets as

the type of Rh. villosa L.f.

When one goes further into the matter one finds it quite impossible to separate Rh.

mucronata Thunb. from Rh. incana, although the former in Thunberg’st types appear to

be quite glabrous, and only on very careful examination does one find a few scattered hairs

on the petioles. A separation into two varieties as done by Sonder (l.c. 513) with the aid

of Thunberg’s originals is quite unjustified. There is still a further complication. Linnaeus

has described a Rh. laevigatum which is quite different from the well-known species

described under the same name by Thunberg. According to the author it has lanceolate

leaflets, whereas in Rh. incana and mucronata they are more or less obovate-cuneate. The

type in Herb. Linnaeus, London, has inflorescences like Rh. incana and Rh. mucronata.

The flowers do not seem to be fully developed. Now this is the plant which Jacquin in

Hort. Schoenbr. t. 345 figured as Rh. elongata and which Sonder (Lc. 513) made his Rh.

mucronata var. Jacquini, under which he also placed the specimens of Rh. mucronata Thunb.

fol. 8. As a matter of fact, even in Linnaeus’ specimens the leaflets are not strictly

* This species is not represented in Thunberg’s Herb. and his description is insufficient.

+ However, in January, 1928, he wrote: “As regards Nos. 11922 a and B. If one sees the two

shrubs without examining them closely ” (they agree in inflorescence, flower, and fruit) “ one could always

suspect two species and yet you put them both simply under var. latifolia. B has yellowish green or

deep green foilage, almost shining; A is a taller upright shrub with greyish foliage often covered with

galls (moths galls), while I have not seen any galls on B, in spite of the fact that this is glabrous and

apparently softer. Horm A has the further peculiarity that in exposed places it sheds its leaves in winter

and about a month elapses before the new leaves appear.”

{ There is a slip in Thunberg’s description (Fl. Cap. ed. Schultes 265) Rh. mucronata: for “‘ petioli ”

in the third line read “ pedicelli ”.

lanceolate, although, again, Jacquin described the leaflets of Rh. elongata as lanceolate.

Sonder is wrong in describing them as obovate. I have seen also Jacquin’s types, and

I quite agree that specifically they cannot be separated from Rh. mucronata Thunb. We

can thus form an unbroken series from Rh. incana Mill. to Rh. laevigata Linn. (non Thunb.).

Whether we are dealing here with a very variable species or whether there are several

distinct species which hybridize freely cannot be decided without growing the various

forms for years and subjecting them to cytological investigations. The species seems to

be very sensitive to changes in ecological conditions. I have never seen a native specimen

which is exactly like Rh. laevigata L., though one from Kirstenbosch (a damp locality) is

very near it; but in the older Huropean herbaria we often meet it under its right name.

However, these specimens were all derived from European gardens.

Looking upon these diverse forms as members of one polymorphous species, the

question arises what name to give it.

Rh. laevigata L.—This name has for over a hundred years been applied to quite a different

species. The form which it represents has not been found wild in South Africa. It is a

most inappropriate name as regards the villous form.

Rh. mucronata Thunb.—The occurrence of a mucro on the leaflets is. very common in

this “ collective” species, though also found in numerous other species of Rhus. If the

name was adopted it would, however, cover, what I may call, the major part of the species.

Rh. incana Mill. (Rh. villosa L.f.)—This is the oldest name. This form is only found

in the neighbourhood of Capetown. I have examined a vast amount of material from

Tropical Africa, and I am not satisfied that the large number of forms referred to this

species can be legitimately associated with it directly, and if these two names are dropped

we clear away at once the misunderstanding about its distribution which has crept into

botanical literature. I may add, however, that the time is not ripe to deal again with the

Tropical African species referred to Rh. villosa, because the material available, large as it

is, is on the whole far too incomplete and frequently unaccompanied by detailed notes.

As regards South African species Rh. MacOwani Schonl. (=Rh. pubescens K. and Z. non

Thunb.) has frequently been mixed up with Rh. villosa L.f.

Assuming that this species easily responds to edaphic influences we may safely place

with it such forms as have been described by Jacquin as Rh. atomaria in Hort. Schoenbr.

t. 343.

In these forms which have somewhat thinner leaves than is generally found in the

species, the petioles have a tendency to become longer and thinner. Mrs. Bolus seit a

fair match of Rh. atomaria from Kirstenbosch, which has a high rainfall. On the other

hand, specimens from sand-dunes and other xerophytic habitats have a tendency to form

coriaceous leaves showing the coarse reticulation of the veins very plainly.

To my mind it will be best to adopt Thunberg’s name, Rh. mucronata Thunb., and to

deal with it under the following forms :—A. typica, B. laevigata (= Rh. laevigata L.), C. atomaria

(= Rh. atomaria Jacq.), D. latifolia, E. villosa (=Rh. incana Mill., Rh. villosa L.f.).

Rh. mucronata Thunb. in Fl. Cap. ed. Schultes, 264 (species collectiva, charact. emend.,

non H. et Z. 1129). For synonyms see varieties below.

Description: A much branched, unarmed or somewhat spiny shrub, 2-8 feet high,

villous, glabrous or glabrescent, with terete branchlets. Leaflets petiolate, petioles —3 the

length of the terminal leaflets, canaliculate above with a tendency in the glabrous forms

to produce narrow wings. Leaflets generally subcoriaceous, obovate, broadly oblong or

lanceolate, always with a more or less cuneate base and frequently acuminate at the apex,

rarely acute, generally mucronulate, margin slightly revolute usually entire ; midrib

prominent on the lower surface, sometimes on both, lateral veins very slightly prominent,

much broken and forming with the tertiary veins a coarse network, conspicuous in glabrous

or subglabrous forms. Panicles very lax, axillary shorter than the leaves, terminal generally

longer. Flowers pedicellate, calyx lobes ovate obtuse, petals oblong, yellowish green.

Drupe subglobose, very slightly depressed.

Length of petioles 8-15 mm. ; terminal leaflets generally 2-5-3-5 cm. ; lateral leaflets very variable.

Breadth of terminal leaflets generally 2-3 em. ; lateral leaflets very variable.

Length of calyx lobes about $ mm.; petals 13-2 mm.

Diameter of drupe 5-5-5 mm.

DisTRIBUTION: South-west Cape Colony im the coast districts to East London,

generally found not far from the sea.*

Before dealmg with the varieties some remarks on a few specimens may be here

inserted to show further that a clear distinction between them is impossible.

EK. & Z. 741 (wrongly named Rhus pendulina Jacq.) m Herb. S.A. Mus. ‘There are

three branches, all from the dunes near Cape Recief, and apparently rightly associated

with one another, but one is quite glabrous. The leaflets are either obtuse, mucronate or

emarginate. Two have villous inflorescences (with glabrous flowers) and more or less villous

petioles ; the leaflets are mostly obtuse (with or without a mucro), some are glabrous,

others are pilose on the thick recurved margin and on the midrib of the underside.

E. & Z. 1101 (described as Rhus Burmanni DC. Prod. 2, 69) in scrub in the 2d. alt.

on the Lion’s Head, Nov., in §8.A. Mus. Herb., is rather scrappy, very sparingly hairy, but

seems to be indistinguishable from the glabrous form of H. & Z. 741.

Pappe 22 (identified in Herb. 8.A. Mus. with E. & Z. 1101, evidently by Harvey) from

the sides of Table Mt., Nov., consists of two specimens, one of which is similar to the

glabrous form of 741, but has evidently thinner leaves in which the lateral veins are

not so coarse as in 741. The other specimen has narrower leaves and both evidently

represent a shade form. ‘The flowers in all of them appear to be the same. The petioles

in all of them are broadly furrowed above.

Pappe 22 in Herb. Alb. Mus. consists of two specimens. One, in young fruit, is quite

glabrous. In the other there are a few flowers which do not seem to differ from those of

typical mucronata: the branches seem to be young. They are densely villous as well as the

petioles and even the leaflets are more or less villous, especially on the underside ; otherwise

the leaves, though smaller and younger than in the glabrous branch, do not show any

essential differences.

Forester Norris (5096), Bellville Flats, near Capetown, formerly white drift sand, Feb.

allie, @, YO tea

Leaflets glabrous, terminal about 3-5 cm. long, pale green above, still paler below

midrib and veins reddish. Apex generally acuminate or mucronate or rarely slightly

emarginate. Venation boldly reticulate. Petiole canaliculate above, sometimes faintly

marginate. Inflorescence sparingly hairy. Drupe rather dry, pale yellow, turning into

red, globular, about 5-5 mm. in diam.

Coppice shoots about a year old included in (5096) are quite different in appearance

(though some fruiting branches are intermediate).

Leaflets much longer (terminal about 4-8 cm. long), softer and much thinner than in

fruiting branches, frequently with one or two crenate teeth, petiole distinctly winged in

upper part. Venation at first sight pimnate, but reticulations are seen faintly in trans-

mitted light.

* Marloth (Das Kapland 1908, 133) states that Rh. mucronata is found in the dwarf macchia on the

granite and limestone hills of the South-West as far as Saldanha Bay. Further, on p. 326, that this

species is bare for about two months in winter, but sometimes the old leaves remain until the young ones

have appeared, and that Rh. villosa, also an inhabitant of the macchia, loses its leaves in June, but replaces

them already the next month.

“ Deeply rooted, difficult to eradicate. Average height 10 feet. Locally known as

Duinen Taaibosch. Straight stems are used by natives to make strong kerries, long ones

are used for whipsticks ”’.

As stated above, I propose to divide the species into five varieties.

A. typica. Rh. mucronata Thunb. in FI. Cap., ed Schultes, 264; Rh. Burmanm* DC.

in Prodr. II, 69; Rh. Burmanni et pendulina EK. & Z. in Enum. No. 1101 and 1102; Rh.

Eckloniana, tenuiflora et pilipes Presl, Bem. (teste Harvey); Rh. lucidum Ait. hort. Kew,

ed. 2, 166 (teste Engler).

Leaflets generally obovate, their breadth less than half their length subcoriaceous,

quite glabrous or with very few minute hairs on the petioles or other parts. Apex obtuse

or acuminate, rarely subacute.

Rh. mucronata Thunb., A. typica, a genuina. Rh. mucronata Thunb., A. typica, a genuina.

Muller (5116). Under side. MacOwan 1303. Upper side.

Rh. mucronatum fol. a and £ in Herb. Thunberg ; sand dunes, Cape Flats, Marloth 11973, Lion’s Head

and Table Mountain, E. and Z. 1101, 1102; slopes of Devil’s Peak, 1,400 feet, height c. 4 feet, very

common ; Dee. (fl.), ? (4490); Cape Flats, on light sandy loam about 2 miles from False Bay, Van Roeper

(5059, 5063) [These are a very good match of Rh. mucronatum Thunb. fol. a and very similar to Rh. foetida

Herb. Jacquin]; Caecilia plantation, Capetown, 4481; Uitvlugt, Cape Division, in sandy soil

common, average height 6 feet, Muller (5115, 5116) [a good match of Rh. mucronatum Thunb. fol. 8, which

has slightly larger leaves than a, but there is no really tangible difference].

B. laevigata. Rh. laevigata L. (non Thunb.) in Sp. Pl. ed. 2 (1762) ; Rh. elongata Jacq.

in Hort. Schoenbr. t. 345; Rh. mucronata 6 Jacquin Sond. l|.c. 513 (p. pte.).

Leaflets subcoriaceous glabrous, oblong lanceolate, cuneate in lower half.

Herb. Linnaeus, London; Herb. Willdenow, Berlin; Herb. Jacquin, Vienna (Rh. elongata Jacq.).

The nearest specimen collected in South Africa was contributed by Mrs. F. Bolus from Kirstenbosch

(4475 in Herb. Alb. Mus.).

C. atomaria. Rh. atomaria Jacq. in Hort. Schoenbr. t. 343.

Leaves with somewhat more slender petioles than in A. Leaflets submembranous,

resembling those of A in shape, subvillous on both surfaces.

Herb. Jacquin ; Kirstenbosch, Mrs. F. Bolus (4474); Table Mountain, Pappe 13 (= Z. 347 from

Zwartland).

* All reference to Burm. Afr. t. fig. 2 is perhaps in future better omitted, since it is clear that, if it

is meant to represent the same species as Rh. mucronata Thunb., it is a very bad figure, and it appears

further from the letterpress that Burmann, like other earlier authors, lumped together different species.

D. latifolia. Breadth of leaflets half or more of length. Young branchlets, petioles,

leaflets and inflorescences villous, older pilose or often glabrescent, generally blunt,.

Rh. mucronata Thunb., D. latifolia. Pillans (5012). Upper side.

Slopes of Table Mountain, Marloth 11992 A and B; Kommetijes, Cape, Pillans (5012); above Camps

Bay, Rogers 3032; Knysna Heads, Schonland 3527, 3532; Redhouse, Paterson 523a; Sandhills, Port

Alfred West, Galpin 2941; Hast London, Rattray 178.

E. villosa. Rh. villosa Lf. suppl. 183 (non auct. plur.); Thunberg in Prodr. 52, FI.

Cap., ed. Schultes, 265; Sonder lI.c. 510 (excl. var. glabrata) ; Rh. incanum Mill. Dict. ed.

8 n. 8, 1760; Rh. tomentosum Mill. (non Linn.) Pluk. Phyt. t. 219, fig. 7 (a form with

coarsely crenate leaflets).

Leaflets, etc., retaining their heavy indument. Terminal leaflets subovate acuminate

or obtuse, mucronate, narrowed from about the middle to a cuneate base. Lateral leaflets

generally oblong, obtuse or subacute, mucronate with a cuneate base.

Herb. Linnaeus, London; Herb. Thunberg, Upsala and various other old collections [e.g. Plukenet’s

in Sloane collection, British Museum, types of Rh. mcanum Mill. and Rh. tomentosum Mill. (non L.)];

Signal Hill, Capetown, 1,000 feet, Marloth 94706 ; Uitvlugt, Cape, sandy soil, very rare, grown flat about

2 feet, Muller (5117) ; Klipfontein, Zwartland E. & Z. 1099 (in S.A. Mus.) ; side of Lion’s Head and Table

Mountain E. et Z. 1098.

The following specimens belonging to EH. villosa were noted by me in the Berlin

Herbarium, Dahlem :-—

1, ex Herb. Alex Braun, Rh. aequalis Pers. ~* non DC. fraglich mit Rh. villosum vereinigt ” h.B. 58.

2. Hort. Bot. Benth. Rh. pubescens 8 caledonica (subglabrous forms ex hort. Bot. Berol. 1843 bear

the same name).

3. Herb. Gundelsheimer (beginning of eighteenth century) “ Rh. Africana trifolia tomentosa ”’.

4. “ Rh. trifolium Africanum”’ (probably also Gundelsheimer coll.).

5. Hort. Monac. ? a narrow leaved form. (One old leaf is subglabrous and shows the typical venation

of “‘ mucronata ”’.)

Various other cultivated specimens, some of which are called Rh. atomaria Jacq.

Drége 5572 (Rh. villosa L., Rh. atomaria Jaeq.), showing stages leading to typical Rh. mucronata ;

Bergius 24, Oct. 15, mostly with smaller leaves than usual.

Hysrips: It has already been stated that the question whether the forms placed here

under Rh. mucronata represent a series of hybrids must for the present be left undecided.

Apparent hybrids with other species are very rare.

Rh. mucronata X MacOwani ?, Pappe (2), Caledon, Nov. (in Herb. 8.A. Mus.). Flowers

as in Rh. mucronata. Inflorescence as in Rh, MacOwan. Young branches, inflorescences,

upper side of leaflets and veins on lower side pilose.

Rh. mucronata X puberula ?, Z. 2236, Cape Recief (in Herb. Kew.).

See also Rh. Legati Schonl.

Rh. mucronata Thunb., E. villosa. Marloth 94706. Rh. mucronata Thunb., E. villosa. Marloth 9470b.

Upper side. Under side.

2. Rhus Ernesti Schonl. nov. sp.

Eh. incana var. Galpin Burtt-Davy Ms.

Description : Frutex vix 1 m. altus ramulis teretibus villosissimis. Folia petiolata,

petiolis villosissimis subteretibus supra leviter canaliculatis ; foliola oblonga vel subovata

apice obtusa vel subacuta saepius breviter mucronata, basi cuneata ; margine subplana

integra ; utrinque sericeo-villosissima costa venisque pilis patentibus tectis ; costa venisque

utrinque leviter prominulis. Paniculae laterales foliis breviores, terminales longiores, parce

ramosa, floribus sessilibus (2 ignotis). Calycis lobi imaequales lanceolati extus tomentosi.

Petala oblonga glabra. Drupa ignota.

Length of petioles about 2 cm.; terminal leaflets up to 4-5 em.; lateral leaflets one-half to two-thirds

of terminal leaflets.

Breadth of terminal leaflets up to 2-5 cm.; lateral leaflets one-half to two-thirds of terminal leaflets.

Floral bracts slightly exceeding 1 mm.

Length of calyx lobes 1-14 mm. ; petals 13-1? mm.

DistRiBuTIOoN : Only known from the neighbourhood of Barberton.

Saddleback Mountain, near Barberton, 4,000 feet, Galpin 1016; Barberton, Thorncroft 27777 (in

Herb. Transy. Mus.).

This has at first sight the look of typical Rh. mucronata, EK. villosa, but the

inflorescences and floral structure are different. The patent hairs on the midrib and lateral

veins (more lightly coloured than the other hairs on the leaves) are also not found in Rh.

mucronata. It is a pity there are no fruits known.

Rh. Ernesti Schonl. Galpin 2777. Under side.

Perhaps some specimens collected by Rademacher (Trans. Mus. 7270) at Carolina,

Transvaal, must be placed here. The shoots collected seem to be dwarfed for some reason

and both flowers and leaves are smaller than in typical Rh. Hrnesti. The terminal leaflets

do not exceed 1 cm. in length; the petioles are about 5 mm. long.

3. Rh. MacOwani Schon]. nom. nov.

Eh. pubescens KH. et Z. (non Thunb. ?) in Hnum. 1100.

Rh. pyroides Sond. p. pte. (non Burch.) in Fl. Cap. IH, 511, Hngler, 1.c. 430, and

Diels, l.c. 582, 594, 628; Rh. villosa auct. plur. (non L.f.).

I take as types the specimens distributed by Hcklon and Zeyher under No. 1100.

Whether they were right in naming them Rh. pubescens Thunb. must remain undecided

as there is no original of Thunberg’s species preserved, and the description in his Fl. Cap.,

ed. Schultes, 264, is not detailed enough. It is quite distinct from Rh. pyroides Burch.,

though I must admit that isolated branches from these two species are sometimes difficult

to distinguish. Generally speaking, Rh. MacOwanz is not so strikingly divaricately branched

than Rh. pyroides, its petioles are relatively shorter and thicker, its panicles are larger and

denser, its flowers are slightly smaller, its pubescence is generally not so appressed and

silky, and is more apt to be more or less shed on older leaves. It has much smaller flowers,

smaller and compressed drupes, and different inflorescence as compared with Rh. mucronata,

K. villosa, which it sometimes resembles in its leaves.

e new name was given i aory of the late Dr. P. MacOwan.

Th m given in memory of the late Dr. P. MacO

Description: A shrub or small tree, sometimes reaching a height of over 30 feet ;

branchlets terete, softly villous. Leaves petiolate, petioles subterete slightly furrowed

above, at first villous, later often glabrescent. Leaflets at first pubescent or villous, later

often glabrescent, ovate or obovate-cuneate, obtuse, acute or somewhat acuminate,

sometimes minutely mucronulate; margin not or very slightly revolute, entire or

occasionally sparingly crenate or dentate; midrib and the arcuate, slightly branched,

lateral veins slightly prominent on the upper surface, more so on the lower, tertiary veins

reticulate (especially distinct on the upper surface when the hairs have more or less

disappeared). Panicles pubescent, much branched and multiflowered, axillary about as

long or longer than the leaves, terminal much longer than the leaves; bracts small,

lanceolate, acute, pubescent. Flowers shortly pedicellate ; calyx lobes ovate, pubescent,

nearly half the length of the oblong petals. Drupe somewhat juicy, subglobose, slightly

compressed, cream-coloured or with a tinge of red.

Length of petioles about 1-5 em. ; terminal leaflets 2:5-3-5 cm. (rarely up to 5 cm.) ; lateral leaflets

about two-thirds of terminal leaflets.

Breadth of terminal leaflets 1-5-2 em. (rarely up to 3 cm.) ; lateral leaflets about two-thirds of terminal

leaflets.

Length of calyx lobes } mm.; petals 1-11 mm. Greatest diameter of drupe 3 mm.

Rh. MacOwani Schonl. E. and Z. 1002 (1100). Rh. MacOwani Schonl. E. and Z. 1100. Under

Upper side. side.

Distripution: In the coast districts from near Bains Kloof to Swaziland and the

Transvaal, perhaps also in Tropical Africa (often placed under Rh. villosa, but the

specimens placed under this species in various large Huropean herbaria are a curious mixture

of species and frequently incomplete). It is found in open scrub and in stream-bank bush

from near sea-level to an altitude of at least 3,000 feet. Sometimes, especially along

streams, it develops into a much branched tree which, e.g. at Keiskama Hoek and near

Riebeek East, reach a height of over 30 feet. It flowers from December to April.

Occasionally leaves with four or five leaflets are found. Usually coppice shoots do

not exhibit any unusual features, but all found on one stump of a tree on the farm Aylesby,

near Riebeek East, had leaves with five leaflets. The petioles had a broad shallow channel

above and were slightly winged, the branchlets and petioles were pubescent, the leaflets

almost glabrous.

On the banks of the Zwartkops River and in the woods of Krakakamma, E. and Z. 1100, KE. 1002 ;

Burchell 949, 3001, 3044, 4888 in Herb. Kew; Drége 6800, also specimens by Drége distributed as Rh.

villosa var. subdentata b and Rh. villosa ain Herb. Kew ; Dal Josaphat, 600 feet, Tyson 898; Swellendam, in

woods, Pappe 15; Zuurbraak, c. 800 feet, Schlechter 2127; mountains near Prince Alfred, 1,700 feet, Schlechter

9979 (very close to the form which Engler described from the Transvaal as Rh. Rehmanniana) ; Vet Rivier,

Riversdale Division, 400 feet, Muir 3471 (terminal leaflet up to 7-5 cm. long) ; neighbourhood of Knysna

from near sea-level to about 700 feet, in some of these the villosity is retained [as e.g. also in Schlechter

2127 and Pappe 15], on the under side of the adult leaflet, Duthie 13, 571a, Cloete 5128, Williamson 36,

Schonland 4150, 3496, 3524; edge of indigenous forest, Blaauwkrantz, 5013, 5026 ; Haarlem, near railway

bridge, 2,700 feet, Fourcade 3161; Assegai Bush, Humansdorp division, Foureade 1145, Zahn 5051,

Britten 1343, Schonland 3632 ; Kleinfontein, near Hankey, J. Sim 182; Van Stadens, and Kamaehs, near

Uitenhage, Paterson 523; Port Elizabeth, Kemsley 260; Zuurberg, Holland 119; neighbourhood of

;rahamstown up to about 2,000 feet, Britten 1523, Dyer 45, 209 (many: leaflets crenate or crenate-dentate

near apex), Schonland 4414 (many leaflets as in Dyer 209, some leaves have four, others five leaflets),

MacOwan 505/1218 (leaflets more elongated than usual and relatively narrower); Blaauwkrantz drift,

Britten 1406 (leaves very much as in Muir 3471), 2736; Lushington valley, near road to Bathurst,

Schonland 5168; Debe Nek, Rogers 4461; Katberg, Staples 5018; Keiskama Hoek, near the village ;

Komgha, 2,000 feet, Flanagan 798; Umzimkulu River, near Clydesdale, 15008, Tyson 2870; Umtata,

3,500 feet, Schlechter 6359 ; Ntsubane, Lusikisiki, Fraser 73/20/A (5099), 73/40/B (5164), 5098, Infengo

Forest, Lusikisiki district, Fegen (5162), Pondoland, Bachmann 819; Friedenau, Natal, Rudatis 112,

1624; Hlatikulu, Swaziland, Stewart 2555 and 9538 (of the Herb. Transvaal Mus.).

In Herb. Alb. Mus. there is a specimen from Oudtshoorn (Britten 34), unfortunately

rather poor, which seems to belong here. It is glabrescent. The flowers appear to be

perfectly hermaphrodite, a rather unusual occurrence.

Some Transvaal specimens, which I have seen, may belong here, but fuller material

is required to make quite sure.

forma Rehmanniana.

Rh. Rehmanmana Engl. lc. 422, Diels 1. 578, 613.

Leaflets obovate or obovate truncate, the terminal one more or less narrowed at the

base, entire in the lower part, crenate or crenate-dentate in the upper.

Rh. MacOwani Schonl., forma Rehmanniana. Rehmann 5560. Upper side.

Specimens which can be placed under this form are especially found in Natal and the

Transvaal, but can be fairly closely matched with others found right to near the western

limit of the species, e.g. Schlechter 1977. The amount of hairiness on the adult leaves

varies as in typical forms. .

Houtbosch, Transvaal, 5560, 5561; Barberton, Thorncroft 4532; Berlin forest reserve, on edge of

great escarpment, c. 4,500 feet, Keet (5166); Middelkop plantation, near Tzaneen, common, usually in

moist localities, Evans (5129); Colenso, Schlechter 6892; Friedenau, Natal, c. 500 feet, Rudatis 331

in Herb. Kew.

4. Rh. Fraseri Schonl. nom. nov.

Rh. tridentata Sond. in Fl. Cap. I, 511; Engler, ic. 425; Diels, l.c. 580, 623 (non

Rh. tridentatum Thunb. in Fl. Cap. ed. Schultes 222); Rh. hirta Harv. Ms. in

Herb. Kew.

Named after Forester G. Fraser, Ntsubane, Lusikisiki, to whom I owe a great deal of

material of Rhus, and whose No. 73/23/A (5143) agrees well with Gueinzius 390 from Natal

in Herb. Kew, the type of Rh. tridentata Sond.

Description: Shrub with elongated, terete, tomentose, densely leafy branchlets.

Leaves petiolate, petioles }-1 the length of the terminal leaflets (rarely somewhat longer),

tomentose or pilose, deeply furrowed above. Leaflets pubescent, dark green and often

glabrescent above, greyish green, pubescent below, more or less oblong, the terminal one

becoming cuneate towards the base from above the middle or lower ; margin slightly revolute,

entire or with a few mucronulate gross teeth near the apex (often the terminal ones toothed,

the lateral ones entire or all entire); midrib and lateral veins slightly prominent above,

more so on the under side, tertiary veins reticulate (but indistinct even on glabrescent

surfaces). Panicles tomentose from the axils of the upper leaves and (about equalling

them in length) and terminal (longer than the leaves), much branched and multiflowered,

bracts small, subulate ; flowers pedicelled. Calyx segments pilose, especially along midrib,

ovate 4-4 of the oblong petals. Drupe subglobose, brownish, shining.

Length of petioles 1-2 cm.; terminal leaflets about 7 cm.; lateral leaflets 4-5 cm.

Breadth of terminal leaflets 2-3 cm.; lateral leaflets 1-5-2 cm.

Length of pedicels 1-1-5 mm.; calyx segments 3} mm.; petals 14-14 mm.

Diameter of drupe c. 4-5 mm., height 5-5-6 mm.

Rh. Fraseri Schonl. Fraser (5149). Under side

Rh. Fraseri Schonl. Gueinzius 390. Upper side. (woolly hairs hiding tertiary veins).

DistripuTion: Natal, Pondoland, and on the Amatola mountaims. Engler has dis-

tinguished a variety integrifolia, which, however, I must drop as the occurence of teeth 1s

never constant.

Natal, Gueinzius 390; Weza forest reserve, Harding, c. 4,000 feet, A.D.F.O. (5054, 5043) ; Pinetown

and Inanda, Wood 852; Verulam, Wood 3908 ; Dumisa, Alexandra county, 600 m., Rudatis 801; Ntsu-

bane, near Lusikisiki, Fraser 73/22/a (5143) ; Hogsback, grassy valleys, Rattray 291. Fraser (5149), from

Ntsubane, is without flowers and has much broader leaves than the type. I think it is a coppice shoot.

5. Rh. intermedia Schonl. n. sp.

Allied to Rh. Fraseri Schonl. and extreme forms resembling Rh. Legati Schonl.

(Rh. laevigata Thunb. non Linn.), but the pubescent branchlets, petioles, etc.,

distinguish it easily.

Description : Frutex 3-5 m. altus ramulis gracilibus teretibus griseo-vel fusco-villosis.

Folia petiolata petiolis gracilibus primum villosis deinde sparse pilosis subteretibus interdum

supra leviter canaliculatis. Foliola membranacea oblonga apice acuta vel acuminata

mucronulata vel plicato-mucronulata basi saepius cuneata (terminalia saepius basi valde

angustata vel petiolulata) ; margine leviter revoluta vel plana integra vel rarius apicem

versus paucidentata ; juvenilia utrinque villosa, adulta utrimque pilosa vel glabrescentia :

costis et venis lateralibus utrinque leviter prominulis, nervis reticulatis non prominulis.

Paniculae villosae multiramosae axillares foliolis subaequilongae, terminales folis longiores

bracteis subulatis floribus parvis pedicellatis. Calyxis lobi ovati acuti. Petala oblonga.

Drupa subglobosa compressa.

Length of petioles 2-3-4 em.; terminal leaflets 7-10 cm. ; lateral leaflets about 3 cm.

Breadth of terminal leaflets 5-6-5 em. ; lateral leaflets 2-2-3 c.m.

Length of calyx lobes barely $ mm.; petals barely 1} mm.

Greatest diameter of fruit about 3 mm. (similar to drupe of Rh. MacOwani Schonl.).

Rh. intermedia Schon). Tustin (5122). Upper side.

Distripution : On the edge of forests from Natal to the northern Transvaal.

Those in search of natural hybrids might look upon this ** species ” as a hybrid between

Rh. MacOwani and Rh. Legati. Against this view is the fact that it does not occur west

of Natal, where these two species are commonly found close together.

Ngomi forest reserve, P.O. Hlobane, Natal, Tusten (5122); Diepkloof, Zoutpansberg, Burtt-Davy

5174; Halic forest station, Zoutpansberg, 4,200 feet, (5024); Woodbush plantation, P.O. Haenertsburg,

common, c. 4,000 feet, Coilett (5077) ; De Hoek, Tzaneen, Pretorius (5086); ? Graskop, Evans (5109).

6. Rh. pyroides Burch. Trav. I, 340 (1822), Cat. Geogr. 1796.

Rh. sericophylla Schlecht. in Engler Pflanzenwelt Afrikas II, 2.

Rh. flecuosa Diels in Engl. Bot. Jahrb. XL, 86 (1907).

(For further synonyms see below.)

This species was found by Burchell on the Asbestos mountains. The type is preserved

at Kew. He described it as follows: “ Frutex 6-10 pedalis. Ramuli saepe spinescentes.

Folia ovato-lanceolata integerrima glabra. Racemi axillares folio breviores; terminales

longiores ’. By comparison with De Candolle’s description in Prodr. II (1825) 70, it will

be seen that DC. seems to have described a different plant (foliolis obovatis-oblongis

mucronatis, etc.), and the error has been perpetuated by subsequent authors, At all events

both Sonder (lc. 1, 511) and Engler (l.c. 430) took it to be the species which Ecklon and

Zeyher distributed as Rh. pubescens Thunb. and which I have named Rh. MacOwani.

Descrietion: The following notes were taken from Burchell No. 1796 at Kew :--

A shrub (or scrambler) with very stout thorns, which are straight or curved and are

3-3:5 em. long. Bark of older branches grey with numerous raised lenticels. Branchlets

slender, 12-15 cm. long, terete, slightly puberulous. Leaves petiolate, petioles slender,

1-7-3 cm. in length, not exceeding half the length of the terminal leaflet, shghtly furrowed

above, puberulous. Leaflets submembranous, oblong-cuneate, usually acute, often shortly

mucronulate ; margin entire or with one or two teeth, upper surface subglabrous, lower

slightly puberulous, subpilose on midrib and veins; venation reticulate, sometimes very

distinct on both sides, but often tertiary veins hardly visible on upper side and indistinct

on lower. Terminal leaflets 4-7 cm. long, 1-7-2-5 cm. broad. Lateral leaflets variable in

size, often about } the length of the terminal, often the two belonging to one leaf slightly

different in size and shape. Panicles lax, puberulous, shorter than the leaves, lateral and

terminal on the young branches. Flowers on short pedicels. Drupe subglobose, evidently

slightly compressed, about 3 mm. in diameter.

The following agree closely with the type (except that sometimes the pubescence is more prominent) :

Karreefontein, Griqualand West, March 21, Wilman; Papkuil, Wilman, February 1921; Newlands,

Griqualand West, December 20, Wilman.

From these specimens the following DESCRIPTION was drawn up :—

Frutex squarrosus saepius spinosus ramulis patentibus pubescentibus ; foliis petiolatis,

petiolis gracilibus teretibus pubescentibus foliolis brevioribus, foliolis membranaceis glaucis

utrinque adpresse pubescentibus vel adultis interdum subglabris ovato-lanceolatis basi

cuneatis vel rarius petiolulatis apice acutis vel subacutis rarius emarginatis saepius sub-

mucronatis margine integris rarius irregulariter crenatis subrevolutis, lateralibus et

terminalibus subaequalibus vel lateralibus brevioribus costa atque nervis utrinque promi-

nentibus venis reticulatis utrinque prominentibus ; paniculis pubescentibus laxis, lateralibus

et terminalibus, bracteis anguste lanceolatis acutis ; floribus pedicellatis, calycis segmentis

triangularibus, petalis pallide luteis late oblongis subduplo longioribus, disco (in floribus

masculis) 5— (vel sub 10-) crenato; drupa subglobosa compressa.

Petioles 1-2-2 cm. long (rarely shorter or longer, but rarely half the length of the terminal leaflet).

Leaflets 3 cm. long (average size; the relative length of terminal and lateral ones varies greatly).

Panicle 21-4 cm. long (rarely longer). Floral bracts about 1 mm. long. Pedicels 3-4 mm. long.

Calyx segments 3-2 mm. long. Petals 11-1} mm. long. Drupe 33-4 mm. in diameter.

All these may be united as var. A. typica, and the following specimens may also be

referred to this form :—

Rh. pyroides Burch. A. typica. Wilman 1256.

Upper side.

Rh. pyroides Burch., A. typica. Burchell 1796.

Upper side. A thorny branch of the same.

Bloemfontein, Potts 3221 (fl. Nov.), Potts 1672 (young fr., Dec.), Potts 2052 (fl. Nov.), Potts 1337

(spreading bush 20 feet high by 20 feet wide), Potts 1220 (fl. Nov., spreading bush, sometimes a tree) ;

Krom River, Karroo, Potts 1336; Trompsburg, Potts 3302, 3305, 3306 (fr. Jan.); Vereeniging, Jan.,

Burtt-Davy 15139; Kafiraria ? Cooper 538 in Herb. Kew; Cradock district, Cooper 510 in Herb. Kew.

There is a form very common, especially in the Transvaal, with larger, sometimes

flexuous inflorescences, which, moreover, when young, have the flowers arranged in

glomerules. This may be called var. C. transvaalensis (usually named Rh. villosa L.f. in

numerous herbaria or Rh. villosa Lf. var. gracilis Engl.); Rh. flecuosa Diels and Rh.

sericophylla Schlecht. may be referred to this form.

Harrismith, Sankey 246; Johannesburg, Gilfillan 6201 and 6202 in Herb. Galpin; Magaliesberg,

Burke in Herb. Kew (erroneously marked “‘ Fish River”), Z. 342; Aapies River, 4,800 feet, Schlechter

3609 ; Pretoria, Burtt-Davy, Rehmann 4740, 4742, Wilms 242, Leendertz 34 and 508, Reck 1089, Thomson

6663; Howlett 1; Naboomspruit, Galpin M 63; Warmbaths, Burtt-Davy 5093; Lydenburg district,

Wilms 238; Rustenburg, Nation 197; Crocodile River, Schlechter 3982 ; Marico district and Schweizer

Reneke, Holub in Herb. Kew; Bowker’s Park, 10 miles west of Queenstown, 3,500 feet, Hilner 324 ;

Lesseyton drift, Queenstown, Galpin 2574; Berry Reservoir, Queenstown, Hilner 344; Ugie, Surat,

4,300-4,500 feet, Britten 4691 ; Katberg, Staples 5018 ; Ntsubane, Lusikisiki ; Mont aux Sources, Tugela

Valley, 5,000 feet, Bayer and MacClean 33.

Pott 4934, from Mavieristad, was named by Burtt-Davy Rh. incana var. sericea. It

belongs to Rh. pyroides, but requires further examination.

This form is evidently widely spread along the Drakensbergen into Swaziland, but

I am afraid of quoting further numbers without re-examination. When, as frequently

happens, the terminal inflorescence is large and densely multiflowered it cannot be easily

distinguished from Rh. MacOwani except that this is more or less incano-pubescent.

whereas the pubescence in Rh. pyroides is appressed sericeous and in the later the petioles

are more slender than in Rh. MacOwani.

Rh. pyroides Burch., B. transvaalensis. Leendertz 34. Under side.

Forms which must be referred to Rh. pyroides Burch. are also common in Tropical

Africa and are usually called Rh. villosa L.f.

C. puberula—Rh. puberula E. et Z. in enum. 1104, Sonder l.c. 511, Engler l.c. 427,

Diels l.c. 587 (excluding var. fastigiata Sond. l.c. 512); Rh. sericea HK. et Z. im enum. 1105;

Rh. Meyeriana Presl Bem.

In the Berlin Herbarium specimens belonging to the true Rh. pyroides Burch. are

placed by Engler under Rh. puberula K. et Z. without having been recognized as belonging

to Burchell’s species. As I agree that these two species belong together the name Rh.

puberula must be dropped since Burchell’s name is the older. The chief difference is mainly

in size of petiole and leaflets; the former is generally 5-8 mm. long, the terminal leaflets

1-5-2 cm. long, 8-10 mm. broad, the lateral about two-thirds the size of the terminal The

frequently oblique plicate apiculus of the leaflets seems to be a fairly characteristic feature

of this form.

Distripution: Eastern Cape Province, Natal, Orange Free State, Transvaal. In

open bush, locally common, rarely exceeding 7 feet in height, flowering generally from

January to March.

Rh. pyroides Burch., C. puberula. MacOwan 862.

Under side.

Rh. pyroides Burch., C. puberula. Zoutpansberg

(5005). Under side.

Burchell 4744, 4874 ; Drége 68084 (Rh. Meyeriana Presl) ; Albany, Zeyher 888 (leaflets broader than

usual) ; Geelhoutboom River, Humansdorp division, Fourcade 2123 ; Assegaibush (Albany), E. et Z. 1105

(type of Rh. sericea E. et Z.) ; south side of mountain drive, Grahamstown, R. Schonland ; ib., MacOwan

766; Katberg, MacOwan 862, Staples (5018), (5022) ; Fort Fordyce, Hoesslin (5123) ; Hogsback ? (5022) ;

Wolfridge, near Keiskamma Hoek, Hunter (5086b); mountain sides near Silo, EH. et Z. 1104 (type of

Rh. puberula EB. et Z.) ; near Queenstown, Hilner 324, 344, Galpin 2574; Komgha, Flanagan 800, Schlechter

6164; between Umtata and the Umzimvubu, Drége (5575 distributed as Rh. angustifolia L. a), Drége

5574, 5576; Pondoland, Bachmaan 821, 823, 825; Murchison, Wood 3126; Natal, Gerrard 1879 ;

Government plantation, Harrismith (5139); Dunelm Farm, Fouriesburg, Potts 3310, 3311; Barberton,

Thorneroft in Herb. Trans. Mus. 4495; Klaserie siding, Lydenburg, Keet 5160.

7. Rh. Baurii Schonl. n. sp. (Rh. pyroides Burch. var. glabrata Sond. in FI. Cap. II,

511).

Closely allied to Rh. pyroides Burch. Sonder characterizes his variety as follows :

‘“ Branches and leaves subglabrous or glabrous, panicles pubescent’. As he mixed up

Rh. pyroides with the species called Rh. pubescens Thunb. by Ecklon and Zeyher, quite

distinct plants may be placed under this variety. In addition to the characters mentioned

the plants placed by me under Rh. Baurii have the following distinguishing characters :—

Leaflets thinner than in Rh. pyroides. Petioles very slender, more than half the length of the

terminal leaflet, sometimes almost equalling it. The terminal leaflet is very frequently

subpetiolate, as it is usually very much narrowed at the base, but this also occurs in Rh.

pyroides, though more rarely.

Description : Frutex inermis laxe ramosus ramulis gracilibus subteretibus subglabris

vel pubescentibus. Folia petiolata, petiolis gracilibus, supra leviter canaliculatis. Foliola

membranacea oblonga acuta, terminalia saepius valde attenuata, juvenilia adpresso-

pubescentia deinde subglabra (costa venisque exceptis), ad margines integra vel paucidentata,

reticulata costa utrinque prominenti, nervis lateralibus arcuatis utrinque leviter prominen-

tibus, venis supra leviter prominentibus subtus immersis. Paniculae ex axillis foliorum

superiorum laterales folis subaequales et terminales quam folia longiores, adpresso-

pubescentes laxe ramosae, bracteis parvis, floribus pedicellatis. Calyx extus adpresso-

pubescens lobis triangularibus subacutis. Petala oblonga. Drupa (Galpin 2340) sub-

globosa leviter compressa.

Length of petiole usually about two-thirds the length of the terminal leaflet.

Length of terminal leaflet usually about 6 cm., sometimes less.

Breadth of terminal leaflet usually about 2 em.

Length of lateral leaflets two-thirds to five-sixths of the terminal leaflet ; floral bracts about } mm. ;

pedicels up to 2 mm.; calyx lobes }-? mm.; petals 14-1} mm.

Greatest diameter of drupe 3 mm.

Rh, Baurii Schonl. Baur 856. Under side.

DisrriBuTion : Midlands of Cape Province at an altitude of 2,500-4,000 feet.

Graaff-Reinet, by margin of watercourses, 2,500 feet, Oct., Bolus 134; Murraysburg, along water-

courses and in open veld, 4,000 feet, Sept., Tyson 61; Sheldon, Nov., Rogers 151; Shiloh, 3,500 feet,

Dec., Baur 856; near Queenstown, Jan., Hilner 336 (fairly commom at the Berry Reservoir), Galpin

2200 (north-western mountain, waterside), Galpin 2201 (north-western mountain, rocky ridges ; leaves much

smaller than usual), Galpin 2340 (kopje Du Plessis farm, Klaas Smits R. ; some leaves are 3 cm. wide and

more or less crenato-dentate).

8. Rhus nebulosa Schonl. n. sp.

Description: Frutex laxe ramosus interdum spinosus ramulis subteretibus glabris

vel parce pilosis; foliis petiolatis, petiolis quam foliola brevioribus pilosis, supra apicem

versus canaliculatis vix alatis; foliolis subcoriaceis supra viridibus subtus pallidioribus

utrinque sparse pilosis vel glabrescentibus ovatis basi cuneatis vel angustatis, margine

integris leviter revolutis, apice obtusis vel acutis vel mucronatis, costa, nervis atque venis

reticulatis utrinque prominentibus, paniculis parce pilosis quam folia longioribus laxe

ramosis, ramis gracilibus ultimis brevibus, axillaribus et terminalibus, bracteis brevissimis

subulatis ; floribus pedicellatis pentameris vel hexameris, calycis segmentis brevissimis

ovatis margine pilosis, petalis oblongis pallide luteis, disco in floribus masculis 5— (vel 6-)

crenato; drupa parva subglobosa compressa rubescens, mesocarpo succulento.

Height of plant rarely up to 6 feet, often scrambling. Petiole 1-2-2-5 cm. long.

Terminal leaflets usually 34 cm. long (sometimes up to 5:4 cm.); average breadth about 2-2 cm.

Lateral leaflets usually about one-third shorter.

Lateral panicles up to 8-5 cm. long; terminal up to 17 cm. long. Floral bracts 4—-? mm. long.

Pedicels 3-13 mm. long. Sepals about + mm. long. Petals about 14 mm. long.

Drupe 3-33 mm. in greatest diameter and about as high.

Rh. nebulosa Schonl. Galpin 2853. Upper side.

Distripution : Sandhills and coast bush from Port St. Johns to Durban at low

altitudes, rarely up to 150 feet; also at Inanda.

Port St. Johns, Ap. (fl.), Galpin 2853, Miss Wood 48, 74, Schonland 4049, 4099; Embotye, near

Pondoland coast, Fraser (5106) ; near Durban, Sutherland 22, Schlechter 2858 (fr.), Drége in Herb. Kew

(named Rh. pyroides Burch. c.), Rudatis 1655; Inanda, Wood 905, 894; Natal, Gerrard 354, 529.

9. Rh. krebsiana Pres] Bot. Bem. 41 (name only); Engler l.c. 409; Diels l.c. 573.

Description : A perfectly glabrous shrub (or, according to Engler, young branchlets

minutely puberulous), with somewhat corrugated branches and angular branchlets more

or less tinged with red. Leaves petiolate; petioles often attaining two-thirds the length

of the terminal leaflets, somewhat angular, furrowed and often slightly winged above.

Leaflets sessile, subcoriaceous, oblong or oblong-obovate, at the apex obtuse or slightly

emarginate, often with a small mucro, at the base cuneate (the terminal one much narrowed,

often subpetiolulate), the lateral somewhat asymmetrical ; margin entire, shghtly revolute ;

midrib and the much branched lateral veins slightly sunk on upper surface, prominent on

lower surface, tertiary veins reticulate and distinct on both surfaces. Panicles lateral, lax,

barely as long as the leaves ; flowers pedicellate ; bracts very minute, subulate. Calyx lobes

oblong-triangular subacute. Petals oblong. Immature drupe subglobose.

Length of petioles 2-3 cm.; terminal leaflets 3-6 cm.; lateral leaflets 2-3-2 em.

Breadth of terminal leaflets 1-5-2-5 em.; lateral leaflets 1-5-2 cm.

Length of inflorescences 3-6 em.; pedicels 14-2 mm.

Length of calyx lobes 1 mm.

Length of petals 2 mm..... Basie Binion,

Distripution: The type is without locality, but the species has recently been

re-discovered in Griqualand Kast.

Krebs in Herb berol.; Tyger Vley in fringing forest on a hillside stream. Mt. Currie

distr., 4500, July (young fr.), O.B. Miller, D 312—Specimens with young flowers were

collected in March.

Rh. krebsiana Licht. Krebs. Upper side.

10. Rhus dura Schonl. n. sp.

Description: Frutex ad 3-4 m. altus ramis subteretibus brunneis sparse pilosis,

foliis petiolatis, petiolis foliolis brevioribus sparse tomentosis supra canaliculatis anguste

alatis; foliolis glaberrimis coriaceis supra nitidis saturate viridibus subtus pallidioribus

ovatis vel late oblongis vel obovatis basi angustatis vel cuneatis apice obtusis vel acutis

breviter mucronatis vel rarius emarginatis, margine integris vel apicem versus leviter et

irregulatiter crenatis vel dentatis vel profunde crenatis; costa, nervis et venis valde

reticulatis utrinque prominentibus ; paniculis sparse tomentosis ex axillis foliorum superi-

orum lateralibus et terminalibus foliis vix aequantibus, floribus ignotis. Drupae pedicellatae

subglobosae nitidae.

Rh. dura Schonl. Evans (5070). Upper side.

Petiole 1-1-3 em. long.

Terminal leaflet up to 6:5 cm. long; about 3 cm. broad.

Lateral leaflets generally half the length of the terminal leaflets.

Drupe 3-34 mm. in diameter.

Distripution: Only known from Graskop, near Pilgrims Rest, Transvaal, at an

altitude of 4,800 feet. ,

Forester Evans, who collected fruiting specimens in January (No. 5070 in Herb. Alb.

Mus.), states that it is common and 10-12 feet high; further, that the fruit is edible and

the wood tough. It is evidently closely allied to Rh. krebsiana and Rh. MacOwani, forma

Rehmanniana, but it has more coriaceous leaves than either of them. As in the latter,

the reticulate tertiary veins on both sides of the leaflets are very prominent. In many

respects it resembles Rh. polynewra Engler et Gilg, in “ Baum, Kunene—Zambesi

Expedition,” p. 289, but in this the flowers are arranged in dense fascicles separated by

wide internodes, whereas in our specimens the flowers seemed to have been arranged fairly

regularly in a lax raceme.

_ 11. Rhus Rudatisii Engl. in ~ Pflanzenwelt Afrikas”’ III, 2, 217.

Description : Caulis efoliatus lignosus repens ramulis erectis aggregatis basin versus

efoliatis pubescentibus sursum villosis dense foliosis subteretibus ; folia petiolata villosa

deinde pilosa, foliolis oblanceolatis vel anguste oblongis apice acuminatis vel mucronatis

membranaceis costa et nervis lateralibus paucis adscendentibus parum ramosis utrinque

parum prominentibus. Paniculae laxae pauciflorae laterales quam folia breviores et

terminales folus paullum superantes. Flores pedicellati bracteis subulatis calycis segmentis

obtusis ovatis extus pilosis, petalis flavis subtriplo longioribus oblongis.

Branches up to 30 cm. long.

The lowest leaves are smaller than those in the middle of the branches and the size decreases upwards

again.

In i middle of the branches the length of the terminal leaflet varies from 34-4-8 cm.

The petiole may reach a length of 1-4 cm.

In one flowering branch the leaves are very much reduced in size.

Petals evidently about 2 mm. long.

Rh. Rudatisii Engl. Rudatis 698. Lower side.

Only known from one locality in Natal—Friedenau, Umgaye flats, Alexandra county,

c. 600 m. Natal, Sep., grazed pastures, fl. yellow, Rudatis 698. Type in Herb. Berlin.

Judging by nervation of leaves and the inflorescence I would place this near Rh. pyroides

Burch. var. puberula, but it is very distinct. When more material is available a transfer

to another group, perhaps next to Rh. fastigiata may be desirable.

DENTATA group.

12. Rh. dentata Thunb. (Synonyms: see below.)

This is an extremely variable species and forms a curious parallel to Rh. mucronata

Thunb., not only on account of its polymorphism but also because it includes quite glabrous,

more or less pilose and villous forms. i

The amount of hairiness here cannot be used as specific or varietal distinctions. I have’

used it to distinguish different forms. The form and size of the leaves again is very variable,

but as these variations to a certain extent are geographically circumscribed (though this.

applies to a certain extent to the forms also) I have used them to form three varieties, but

sharp distinctions between them cannot be given, though extreme forms are easily

recognized.

The division of the species would then be :—

A. typica: a genuina, B pilosa; B. parmfolia: a glabrescens, B pilosissima ; ©. grandi- )

folia: a glabra, B pilosa, y pilosissima.

Then there are a number of forms which do not fit in with such a scheme of classifica-

tion. In most cases they are intermediate between Rh. dentata and other well-known

species, and there can be no reasonable doubt that they are hybrids.

Rh. dentata Thunb., A. typica a genuina. Fourcade 965. Under side.

Engler (I.c. 435) has separated Rh. Sonderi from Rh. dentata. It coincides to a large

extent with ©. grandifolia. The leaflets are, according to Engler, not albo-marginate.:

This, I find, is not always the case. This character is also variable in typical dentata. He

further states that the petals are barely over 1 mm. long. I find them 1}-1) mm. long.

In A. typica they vary in length from 14-2 mm., so that it seems better to smk again Rh.

Sonderi in Rh. dentata. I must also unite with it Rh. parvifolia Sond., Rh. truncata Schinz,

Rh. Galpinii Schinz (non Engl.), and Rh. acutidens Engl.

Disrripution : In the coast districts from the Humansdorp division to Natal, on the

eastern mountains, and in the Transvaal.

A. typica—

Description: A perfectly glabrous or more or less pilose or villous shrub, unarmed |

or subspinous with short rigid or sometimes slender, slightly furrowed subterete branchlets.

Leaves petiolate, petioles slender, 4? the length of the terminal leaflets, narrowly canalicu-

late above. Leaflets thin, membranous, deep green above, lighter below, often subplicate, ,

sessile, obovate, usually coarsely crenato-dentate in the upper part, rarely subentire or

entire, teeth mucronate, the entire base of the leaflets cuneate ; margin flat, sometimes

white-margined ; midrib slightly sunk above, somewhat prominent below, lateral veins

immersed above and slightly prominent below, rarely also above, tertiary veins reticulate,

immersed. Panicles loose, the axillary generally shorter than the leaves, the terminal

slightly longer, bracts subulate, small, pedicels longer than the flowers. Calyx lobes

triangular, petals oblong. Drupe more or less red, shining, subglobose, shghtly depressed.

Length of petioles about 1-5 em.; terminal leaflets usually 2-2-5 cm.

Breadth of terminal leaflets usually about 1-7-2 cm.

Lateral leaflets one-half to two-thirds the size of the terminal leaflets.

Length of pedicels about 2 mm.; calyx lobes about ? mm.; petals 1}-2 mm.

Diameter of drupe about 4 mm.

a genuina—Perfectly glabrous. Rh. dentata Thunb. in Fl. Cap. ed. Schultes 265 ;

Rh. cuneifolia E. Mey. in Drége exsicc.

DistrrBuTION: From the Storms River to Bothas Hill, near Grahamstown, and on

the eastern mountains to Stutterheim.

Herb. Thunberg, Upsala ; Storms River, 250 feet, Nov. (fl.), Schlechter 5984 ; Witte Els Bosch flats,

700 feet, Fourcade 2198; Oudebosch flats, Fourcade 965; Suku, Humansdorp division, Burchell 4819

(many leaves entire or subentire); Hofman’s bush, Humansdorp division, thick scrub at foot of kloof,

Britten 1175; Kleinfontein, near Hankey, common, J. Sim 159 (many leaflets acuminate and with very

small teeth) ; Gamtoos River, Schlechter 6057 ; Van Stadens Mountains HK. et Z. 1126, Ecklon 21 ; between

Grahamstown and Assegai bush, Ecklon 368 (wrongly distributed as Rh. tridactyla Burch.) ; ib. Z., 2231;

Bothas Hill, near Grahamstown, Dyer , MacOwan 281; Boschberg, near Somerset East, MacOwan

281 bis; Cambedoosberg, Drége (“ Rh. dentata Thunb. b’’), Bedford, Mansell Weale 870; Katberg,

Staples (5020) [partly with longer petioles than the type]; Quaku forest station (4480); edge of Mkubiso

forest, common, Stayner 37.

6 pilosa.—Sparigly pilose on the branchlets, panicles, petioles and young leaflets

(especially veins) and calyx, sometimes also on older leaflets.

DisTRiBuTIoN: From Van Stadens to Umtata, also on the Amatola Range, Maclear

district, Fouriesburg, Orange Free State, Barberton and Pilgrims Rest districts, Transvaal.

Amongst bushes on the Van Stadens Mountains, Z. 2231 bis; Grahamstown, 1,900 feet, Schlechter

2639 : 3 miles north-west of Grahamstown, 4-5 feet high, 2,000 feet, Dyer 210 (one branch quite glabrous) ;

Tarka River, Kabousie, Cooper 274; Hogsback, Rattray 236; Stutterheim, 3,000 feet, Hunter (5033),

ib., Woodcock (5009); Komgha, 2,000 feet, Schlechter 6168; ib., Flanagan 656; slope towards the

Umtata waterfall, scattered, not more than 4 feet high, Schonland 3835; Cala, Royffe 183; Surat,

Maclear, Britten 4564; mountain sides, Fouriesburg, Orange Free State, 6.200 feet, Potts 1955 (“‘ leaves

deciduous”); Barberton and Pilgrims Rest districts, common, Keet 1455.

B. parvifolia (Rh. parvifolia Harv. in Fl. Cap. I, 510); Rh. mollis K. Mey. in Drége exsice.

== Rh. truncata Schinz in Bull. de Vherb. Boiss. Ser. II, viii, 86.

Leaflets smaller than in A. and generally not so deeply incised or even sometimes entire,

sometimes distinctly truncate and slightly emarginate (Rh. truncata Schinz), terminal

usually 10-15 mm. long, petiole usually slightly under 10 mm. long.

DisTRIBUTION : Queenstown district to Natal and the Transvaal.

a glabrescens—Leaflets glabrous or sparingly pilose.

Rocky places on the Insiswa Mountain, Griqualand East, 2,300 m., Schlechter 6465 (type of Rh. truncata

Schinz); Colenso, 3,200 feet, Schlechter 6882; Moorddrift, Transvaal, Leendertz 2265 ; Umbombola

Range, 4,000—4,400 feet, shrub 3 feet, Galpin 2190, and the following in Herb. Kew: Natal, 3,500-4,000

feet, Sutherland; Natal, Gerrard 1400; Umgeni waterfall, Cooper 2169.

6 pilosissima.—Leaflets retaining a covering of soft longish hairs.

FHincham’s Nek, Queenstown, Galpin 2204 ; hillside near junction of Kei Rivers, district Queenstown,

2,350 feet, shrub 3 feet high, Galpin 8099 ; Tabase, Transkei, 2,500 feet, Baur 333; Mvenyani, Cedarville,

Griqualand East, Bandert 42; Ingungo, 3,000 feet, Schlechter 6311 ; Mooi River, 6,000 feet, Schlechter

3348 (Wood 1035 in Herb. Berol. and Kew, and Rehmann 4743 in Herb. Berol.).

C. grandifolia—

Description: Young axial structures and petioles softly shaggy pubescent, rarely

glabrous. Leaflets generally retaining a covering of soft hairs, especially along midrib,

1ateral veins and margins, more rarely plant subglabrous or quite glabrous. The leaves

are generally two to three times the size of the average in A., but no sharp distinction can

be drawn. With the increase in size of the leaves the petioles often become absolutely

and relatively longer. The general outline of the leaflets, although variable, resembles A

as a tule (Rh. Sonderi Engl.). There is, however, a tendency to form acute apices, and

sometimes the leaflets are almost lanceolate (Rh. Galpinii Schinz, in Bull, de Vherb, Boiss

Ser. II, VIII, 638 Galpin 646 from Barberton [non Engl.] = Rh. dentata var. acuminata

Burtt-Davy Ms.).

Rh, dentata Thunb., C. grandifolia, y pilosissima. Fraser (5039). Under side.

DistriputTion: Chiefly on the outside of forest and in swampy ground from eastern

Cape Colony to Natal, the eastern Orange Free State to the northern Transvaal.

a glabra.—Quite glabrous.

Ntsubane, 10 miles from Lusikisiki, 5 miles from coast on Table Mountain sandstone, fairly common,

1,800 feet, Fraser (5048); Ingwangwane. P.O. Riverside (4484); Weza forest reserve, 14 miles from

Harding, Sallender (5044), 5045 p. pte.; Ladybrand, Rogers 791; between Greytown and Newcastle,

Wilms 1921; Alexandra county, Rudatis 1149.

B pilosa.—Pilose, leaflets glabrous or glabrescent (largely = Rh. Sonder: Engl. var.

pilosa Engl., e.g. Schlechter 6990), and Rh. dentata B puberula Sond.

. East London, Rattray 169; Kentani district, 1,200 feet, fragrant shrub, 1-5 feet, Pegler 900 p. pte. ;

Amanzamnyanu, Mt. Frere, 3,000 feet, common on hillside, Van der Merwe (5104); Dlokolyana forest,

‘East Griqualand, 4,500 feet, average height 20 feet, common (5046, 5072); Insizwa, 6,000 feet, Schlechter

6525 ; Howick Falls, Rogers 546 ; Weza forest reserve, near Harding, c. 4,000 feet, Sallender (5045 p. pte.) ;

Van Reenen, 5,600 feet, Schlechter 6990 (the specimens in Herb. Alb. Mus. are evidently coppice shoots

with very large leaves, petioles up to 5-5 cm. long, terminal leaflets up to 8-5 em. long) ; near Lydenburg,

Burtt-Davy 7642, 5328; Graskop, northern Transvaal, Evans (5108a); ib., branch with juvenile foliage

which is very large, similar to Schlechter 6990, Graskop, Evans (5095); Fouriesburg, Dunelm farm, on

mountain side, shrub 3-5 feet, Potts 3106; scattered over the mountain, Ladybrand, Patterson (5136).

There is a sterile branch in Herb. Alb. Mus. which agrees with many other specimens in Herb. Alb. Mus.

and a fruiting branch in which the leaves do not differ much from A. typica, but the panicles are much

longer (up to 7 cm.).

y puosissima.—Leaflets retaining largely a covering of soft longish hairs.

Congeane, 14 miles north-north-east of Qumbu, Transkei, Dwyer (5113) ; common in the neighbour-

hood of Lusikisiki, Pondoland, Fraser (5037, 5039); Ingwangwane, P.O. Riverside, (4482); Ladysmith,

Natal, Rogers 694 (leaflets with a more obtuse outline than is generally the case); Pretoria, Rehmann

4743 (petioles very variable in length from over 1 cm, to 5-2 em.); Ermelo, Tennant 6938; Hlands

spruitberg, 1,730 m., Schlechter 3390; Mavieristad, Pott 4933.

The separation of var. C. grandifolia into the forms here distinguished, being quite artificial

and not adopted until my return from Europe, a few of the following in Herb. Kew may

have to be readjusted. Some may even have to be placed under A; others are almost

certainly hybrids.

B pilosa—

Zuurbergen, Drége (marked Rh. crenata Thunb. ?), cotype of Rh. dentata 8 puberula Sond. (young

axial organs and petioles pilose, also midrib and margin of leaflets, otherwise surface of the latter sparsely

pilose) ; Burchell 4171, 4607; Shiloh, 3800, Baur 894; Br. Caffraria, Cooper 275, 422 (this and the

following very much like Rh. Galpinti Schinz); Natal, Gerrard 762; Inanda, Wood 1140; Phillips,

Bethlehem 3158, 3161 ; The Downs, Pietersburg, Rogers 22039 (agrees closely with Rh. acutidens Engl.) ;

Limpopo sources, Nelson 515 (like Rh. Galpinii Schinz.).

y pilosissima—

Basutoland, Dieterlen 76 a and b, Cooper 2171; Orange Free State, Cooper 850; Hoogeveld, Page’s

Hotel, Transvaal, Rehmann 6863; Charlestown, Natal, Mogg 9655.

Hysrips: There can be no doubt that Rh. dentata hybridizes freely with Rh. Legati.

Such evident hybrids can be readily detected by the practised eye in the field, but cannot

always be detected with certainty in herbarium specimens, and no doubt some of the speci-

mens enumerated above are hybrids, e.g. those referred to Rh. Galpinia Schinz and Rh.

acutidens Engl. 1.c. 423 ; Diels 1.c. 578,623.

Rh. dentata Thunb. A hybrid of Rh.

Rh. grandidens Harv. Schlechter 3642. dentata ?

Gerrard 1399. Rh. acutidens Engl. Rehmann 5558. Under side. Woodcock 4471.

(A hybrid ?) (A hybrid ?) (A hybrid ?) Upper side.

Rh. dentata X Legati.—Common on the Katberg, Amatolas, and near Grahamstown :

terminal leaflet often petiolulate, leaflets very irregularly dentate, axillary panicles much

longer than the leaves, e.g. Dyer 746a and 6 (Katberg), Woodcock (5010), Kubusie,

Stutterheim, Dyer 409 (Howiesons Poort).

Rh. dentata X pyroides var. puberula (an fastigiata ?).—Leaflets with a few acute teeth

(usually only in the upper half), cuneate from near the apex to the base, petioles often

under 1 cm. long, terminal leaflet 1-5-2 cm. long. A rambling shrub, about 8 feet high,

under common yellowwood, fairly plentiful, Upper Kubusie, Stutterheim district, 3,300 feet,

Woodcock (4471). The hybrid nature of this curious form (the leaves of which often

resemble those of Rh. acutidens) may also be concluded from the fact that only sterile

flowers with abortive stamens and no ovaries were found. In some flowers the calyx lobes

were 1 mm. long, in others 1}-1} mm. Another specimen, which is like Rh. acutidens Engl.,

is Moss 8324, from Witpoortje Kloof, near Johannesburg. Prof. Moss noted on the label :

“ Rh. dentata X pyroides (me judice).” Here the anthers were fully developed. Another

evident hybrid is Rh. grandidens Harv. Ms. in Herb. Kew; Engl. l.c. 440; Diels lc. 587.

It was described from Gerrard No. 1339 in Herb. Kew. It has glabrous leaflets which are

narrowly lanceolate, very acute and usually [not always, as Engler stated] provided with

a few acute teeth above the middle. Flowers and fruits are unknown. Rudatis 1310,

from the Umgaye Flats, Alexandra county, in the Berlin Herb. is very similar. The

following notes were taken from it :—A much branched small shrub, 3-6 feet high, with

ascending densely leafy branches. The young branches, petioles and petals are pubescent.

The leaflets are subglabrous and have revolute margins. The midrib is slightly sunk above,

prominent below, the lateral veins often prominent above and usually prominent below,

arising: at rather an obtuse angle. The specimen has flowerbuds, which are rather large

for the genus. They are arranged in dense lateral pyramidal panicles which are shorter

than the leaves.

13. Rh. carnosula Schonl. n. sp.

Rh. laevigata L. B foliis latioribus grosse dentatis in Drége exsice. 5569; Rh. laevigata

Thunb. f dentata HE. Mey.; Sonder l.c. 514.

Differs from Rh. laevigata Thunb. (non Linn.): Flowers larger; panicles always

terminal and lateral in the axils of the upper leaves; styles frequently persisting even

when the drupes are ripe; leaflets thicker, slightly fleshy, frequently coarsely dentate,

lateral veins thicker and more prominent. Merges near Grahamstown into some forms

of Rh. dentata in the shape and dentition of the leaflets. Probably also closely allied to

Rh. eckloniana.

Description: Frutex glaberrimus 2-15’ altus ramulis gracilibus teretibus. Folia

petiolata petiolis supra canaliculatis. Foliola carnosula subcoriacea, sessilia vel terminalia

petiolulata, oblonga vel subovata basi cuneata apice acuta vel acuminata mucronata;

margine plana saepius albocincta, parte superior! ++ grosse dentata, dentibus oblique sub-

erenatis vel oblique triangularibus mucronatis vel integra; costa venisque supra Vix

prominulis infra costa valde prominenti venis prominulis, nervis utrinque vix prominulis.

Paniculae multiflorae axillares quam folia breviores, terminales longiores, bracteis parvis

anguste lanceolatis acutis, floribus pedicellatis. Calycis lobi subovati subacuti vel obtusi.

Petala oblonga. Drupa nitida subgiobosa stylis saepius persistentibus.

Length of petioles 2-3-5 cm.; terminal leaflets 5-7 cm. ; lateral leaflets three-quarters to five-sixths of

terminal leaflets.

Breadth of terminal leaflets usually 2 cm., sometimes nearly 3 cm.

Breadth of lateral leaflets slightly less than terminal leaflets.

Length of pedicels 1-2 mm.; calyx lobes about $ mm.; petals 1}-2 mm.

Diameter of drupe about 5 mm.

DistRIBUTION : From Howiesons Poort, near Grahamstown (where, however, it 1s not

quite typical), along the coast to Pondoland.

A. typica—

Galan pecs 5569 ; Hast London, Rattray 177 (= Bolus 8839 in Herb. Bol.) ; Keimouth, 1,000 feet,

Schlechter 6212; amongst coarse valley vegetation, 2-3 feet, Feb. (young flowers and fruit), Kentani,

1,000 feet, Pegler 225; Hastgate S. Johns, Schonland 3944; Ntsubane forest station, Lusikisiki, Pondo-

land, Fraser (5080) [one branch with entire leaves]; ib., Fraser (5146) [one branch with small leaves,

terminal leaflet not exceeding 4 cm. in length]; ib., Fraser (5152) [with ovate leaflets]; Notensila forest

station, Pondoland (5038); Pondoland, Bachmann 69 in Herb. Berol.

Rh. carnosula Schonl. (5038). Under side.

B. longipetiolata.—Leaflets about the average size and shape of the typical form, but

with much longer petioles (4-6 cm.).

East London, Rattray 170 (some leaves have a fourth, small leaflet); 8 miles east of Hast London,

Dyer 1985 (not bushy, 2-4 feet high) ; Komgha, 2,111 feet, Rogers 3309 ; slope towards Umtata waterfall,

shrub 15 feet high, scattered, Schonland 3836 ; Embotye, Pondoland, Fraser 73/16/B (5100a) ; Ntsubane,

Fraser 5145, 5149.

C. parvifolia.—Leaflets smaller than in the type with shghtly prominent whitish lateral

veins on the upper surface.

Ntsubane and other places a few miles from the sea in Pondoland, Fraser (5053, open veld, sandy

soil often swept by fire, height about 2 feet), (5151), (5081).

The specimens from Grahamstown referred to above, e.g. Dyer 196, have the lateral veins of the

leaflets only very slightly prominent on the surface or they are quite immersed, the flowers are smaller

(calyx lobes 3-3 mm. long, petals 1+ mm. long), fruits also smaller than in the type.

Hyprips: Perhaps the plants with very long petioles may be hybrids with Rh. Legati

Schonl. (Rh. laevigata Thunb.). Hybridization with Rh. dentata Thunb. is also suspected.

14. Rh. Rogersii Schonl. n. sp.

Descrretion: Frutex fastigiatus 3-4’ altus ramulis novellis subteretibus striatis

pilosis. Folia petiolata petiolis striatis pilosis supra late canaliculatis subalatis. Foliola

subcoriacea sessilia oblonga infra medium cuneata vel totius cuneata, juvenilia dense et

breviter pubescentia, adulta glabrescentia vel glaberrima, supra saturate viridia subtus

valde pallidiora ; ad margines leviter revoluta, basi usque ad medium vel ultra integra,

parte superiori ++ grosse crenato-dentata dentibus mucronatis ; - costa venisque utrinque

prominulis nervis reticulatis immersis. Paniculae multiramosae multiflorae leviter pilosae

vel glabrae laterales and terminales quam folia breviora vel longiora bracteis parvis

lanceolatis, floribus breviter pedicellatis. Calycis lobi maequales subovati subacuti vel

obtusi. Petala oblonga. Drupa immatura subglobosa.

Length of petiole about 2-5 cm.

Length of terminal leaflets 8-10 cm. (generally) ; lateral leaflets 4-9 cm. (generally about 7-5).

Breadth of terminal leaflets about 3 cm. ; lateral leaflets about 2 cm.

uength ot pedicels 1-14 mm.; calyx lobes about $ mm.; petals about 14 mm.

Rh. Rogersit Schonl. Rogers 18270. Upper side.

DistRIBUTION : Barberton district and Swaziland.

Barberton, 3,000 feet, Dec. (fl.), Rogers 18270; ‘‘ Hilltop”, Barberton, in scrub forest on granite

soil, Nov. (fl.), Keet 1454 (‘‘a small bush, 3-4 feet high, common at this locality, and seen also along

kranses at Berlin, in the same district, at elevations up to 5,000 feet”); Berlin forest reserve, on the

edge of the great escarpment, Barberton division, c. 4,000 feet, Apr. (young fr.), Keet (5165); Godwan

River, Swaziland, Feb. (fl.), Jenkins (Transvaal Mus. herb. 10370).

The female inflorescences are longer and looser than the male as far as is shown by

the available material.

16. Rh. ntsubanensis Schonl. n. sp.

Description: Frutex parvus ramis juvenilibus subteretibus glabris lenticellis

numerosis orbicularibus prominentibus tectis. Folia petiolata petiolis glabris vel sub-

glabris supra canaliculatis subalatis. Foliola glabra sessilia oblongo-ovata basi cuneata

apice acuta vel subacuta saepius plicato-mucronata, facie undulata, supra saturate viridia

subtus glauca, margine integra; costa venisque supra vix prominulis, subtus conspicue

prominentibus nervis paucis reticulatis immersis. Paniculae glabrae laxae valde ramosae.

Flores ignoti. Drupa subglobosa nitida primum rubriuscula carnosa deinde subatra.

Length of petioles about 2 em.; terminal leaflets 7-8 cm. ; lateral leaflets 6-7 em.

Breadth of terminal leaflets about 3 em. ; lateral leaflets about 2-7 cm.

Diameter of drupe 4-5 mm.

Rh. ntsubanensis Schonl. Fraser (5138). Under side.

DisrripuTIon : Only one specimen known—found at Ntsubane Forest station, near

Lusikisiki, Pondoland, by Forester G. H. B. Fraser (No. 73/22/B; 5138 im Herb. Alb.

Museum).

REFRACTA group.

16. Rh. refracta HK. et Z. in Enum. 1103.

Description : A much branched, often thorny shrub, reaching a height of 8-9 feet,

branchlets often nearly horizontal, subterete, grey or pale rufous villous. Leaves petiolate,

petioles subterete, flattened or slightly furrowed above, leaflets rigidly membranous, dark

green above, paler below, when young villous, later sparsely pilose on both surfaces, often

corrugated, obovate-cuneate, obtuse or a little emarginate or mucronulate at the apex,

margin slightly revolute, entire or rarely crenulate in the upper part, midrib and the few

lateral veins usually sunk in the upper surface of the leaflets, slightly prominent on the

lower, tertiary veins few and usually not visible in dried specimens. Panicles axillary and

terminal, greyish villous, laxly branched, longer than the leaves. Bracts very minute.

Sepals subovate, pilose on the outside. Petals oblong. Drupe somewhat fleshy, sub-

globose, slightly depressed above, blackish with a greyish bloom.

Length of petiole 7-10 mm.................

Length of terminal leaflets about 1-2-2-2 a

Length of lateral leaflets about 7-10 mm..... Somewhat larger in some forms

Breadth of terminal leaflets about 1] cm...... growing near the sea.

Length of male panicle about 7 cm.; pedicels about 1 mm.

Length of calyx lobes about } mm.; petals about 1} mm.

Drupe: height about 34 mm., breadth and depth about 4 mm.

Rh. refracta HK. et Z. Dyer 211. Under side.

Distripution : In dry open scrub from the Knysna division to Natal (?) from near

sea-level to an altitude of about 2,000 feet.

Blaauwkrantz forest reserve, Knysna division, 1 mile from sea, 200 feet, Zahn 5040; in the forests

of Addo and by the Zwartkops River, E. and Z. 1103, Z. 14, 87; common in the neighbourhood of

Grahamstown, Schonland 3301, 4477, 4478, Dyer 162, 211, Rogers 27671, Britten 1504; Blaauwkrantz,

Britten 2738; Lushington valley, May, Schonland 5169 (fr.); Kareiga R. forest, Britten 2498; Port

Alfred, Tyson 50, Schlechter 2735, Britten 1883, 1828 [Port Natal, Gueinzius, Sanderson—teste Sonder

Vero:

In many respects this species approaches Rh. MacOwani ; in fact, it looks like a small-

leaved form of it, but its leaflets have not got reticulate venation and inflorescences,

flowers and drupes are different.

17. Rh. fastigiata E. et Z. in Knum. 1107.

Rh. puberula EK. et Z. var. fastigiata Sonder |.c. 511; Engler l.c. 427.

2 Rh. sericea H. et Z. in Enum, 1105.

Rh. humilis HE. et Z. in Enum. 1108.

Desortetion : Much branched bush, 2-9 feet high, with slender terete branchlets,

which, like the leaflets, petioles and branchlets, are shortly adpressed pubescent. Leaves

petioled ; petiole subsemiterete, furrowed above. Leaflets dark green above, paler green

on lower side, subcoriaceous, oblong-acute or elliptic-lanceolate narrowed at the base,

sometimes obtuse or mucronulate, mucro often plicate, surface in adult leaflets often

subglabrous, margin entire narrowly revolute ; midrib slightly prominent on upper surface,

more so on lower ; lateral veins few, slightly prominent above, more so on lower surface,

slightly branched towards the margin of the leaflets. Panicles very lax, lateral in the axils

of the upper leaves (shorter than the leaves) and terminal about as long as the leaves.

Bracts very small. Calyx lobes broadly ovate, very obtuse, very pale greenish white.

Petals oblong, very pale yellowish white. Drupe subglobose, glabrous.

Length of petiole 5-8 mm.; terminal leaflets 2-3 em.

Breadth of terminal leaflets 4-8 mm. Lateral leaflets about two-thirds the size of the terminal.

Length of pedicels 1-2 mm.; calyx lobes barely $ mm.; petals about 14 mm.

Diameter of drupe 3-3} mm. ;

Gane 325, from the “ Mayor’s Seat ” near Grahamstown, has perfectly hermaphrodite flowers.

Rh. fastigiata B. et Z. MacOwan 766. Upper side.

DistRiBuTION : In the coast districts from the Uitenhage division to Natal, extending

inland to the Katberg and Amatola mountains.

No absolutely sharp line separates this species from Rh. pyroides Burch. (non Sond.

nec Engler) var. puberula, but the absence of reticulate venation (except in coppice shoots)

in Rh. fastigiata defines it sufficiently. The leaflets of Rh. fastigiata are shorter and narrower

than in Rh. pyroides, but this difference becomes less in coppice shoots of the former. The

lateral veins are generally much more distinct in Rh. fastigiata. The fastigiate habit cannot

be used as distinguishing character, but is common in the species.

Uitenhage and Albany districts, H. et Z. 1107; Addo hills, E. et Z. 1108 (Rh. humilis HK. & Z.) ;

Z. 887; Van Stadens, Paterson 1987; Zuurberg, Paterson 18; very common in the neighbourhood of

Grahamstown, especially south of it, MacOwan 265, 766, Britten 2807, 5171, 5172, 5173 (coppice shoot),

Salisbury 208 (not typical: leaflets more leathery, petioles only about 3 mm. long, lateral veins often

not distinct ; perhaps Rh. fastigiata x lucida), Z. 2239, E. 887 (from the neighbourhood of the locality of

the type of Rh. sericea H. et Z.); Wolfridge, Amatolas, c. 5,000 feet, Hunter 5086a ; wooded kloof, Fort

Cunyngham, 3,300 feet, Galpin 2480 ; Komgha, 2,000 feet, Apr., Flanagan 800, Schlechter 6164 ; Ntsubane,

near Lusikisiki, Fraser 5097 ; near Murchison, Wood 3126; Natal, Gerrard 1879; Burchell 4744, 4814 ;

Drége (named Rh. angustifolia L. a.); Herb. Thuberg (Rh. esciswm fol. a). :

18. Rh. Dinteri Engl. in “ Pflanzenwelt Afrikas ” III, 2, 211, fig. 103, H. to G.

Rh. impermeabilis Dinter Ms.

Allied to Rh. refracta.

Description : A much branched thorny shrub up to about 6 feet high, with subterete

pubescent or subtomentose branchlets. Leaves petiolate, petioles about half the length of

the terminal leaflets or slightly longer, pubescent, slightly furrowed above. Leaflets

glaucous, obovate or oblong, usually slightly plicate-mucronulate at the apex, cuneate at

the base (the terminal often much contracted) ; margin usually entire, very slightly revolute ;

pubescent on both surfaces, but hairs longer on the lower surface ; midrib and lateral veins

slightly prominent on both surfaces, more on the lower than upper surface, lateral veins

few, arcuate, slightly branched, tertiary veins distinct, reticulate. Panicles subtomentose,

axillary and terminal, slightly shorter than the leaves, or the terminal ones a little longer,

laxly few-flowered, flowers pedicellate. Calyx lobes ovate. Petals oblong. Drupe sub-

clobose.

Length of petioles 6-9 mm. ; terminal leaflets 1-2-2 cm. ; lateral leaflets «8-1-6 cm.

Breadth of terminal leaflets -8-1-5 cm.; lateral leaflets -5-1-1 cm.

Length of pedicels about 1 mm.; calyx lobes about ? mm.; petals about 14 mm,

Diameter of drupe about 4 mm.

Rh. Dinteri Engl. Dinter 4359. Upper side.

DISTRIBUTION: Restricted to the South-West Protectorate.

Schaaprivier, Mar. (fr.), Dinter 1898; Lichtenstein and Hohenwarte, Anas mountains, Dec. (fr.),

Dinter 4359 ; between Haris and the Anas mountains, on plateau in sandy ground (young fr.), Pearson

9507 (one leaf with five leaflets).

19. Rh. crenata Thunb. in Fl. Cap. ed. Schultes 266, Sonder l.c. 512; Engler l.c. 422;

Diels |.c. 577, 631.

Description : Much branched shrub, usually about 3-7 feet high, with terete, shortly

greyish or rufous villous branches. Leaves shortly petiolate, petioles relatively thick.

Leaflets without ordinary hairs, rigid, membranous, sessile, dark green or greyish green

above, often rufous on the lower side, obovate-cuneate, margin slightly revolute, 3-5 crenate

at the blunt apex, midrib and the few lateral veins slightly prominent on both surfaces,

no tertiary veins. Panicles shortly villous, axillary not confined to the very apex of the

branches, shorter or the male ones longer than the leaves, terminal longer than leaves,

multiflowered. Bracts small, acuminate, pilose. Flowers shortly pedicellate, very often

4-merous, often 5-merous, rarely 6-merous. Calyx lobes ovate acute, pilose. Petals oblong,

about three times the size of the calyx lobes. Drupe fleshy, almost black when ripe,

subglobose.

Length of petioles about 2 mm.

Length of terminal leaflets 1-2-2-5 cm., but usually only 1-2-1-5 cm. and breadth in proportion.

Length of lateral leaflets -5-1-7 mm., but usually only 1 cm., and breadth in proportion.

Breadth of terminal leaflets 5-9 mm. ; lateral leaflets 6-9 mm.

Length of pedicels about 2 mm.; calyx-lobes +-$ mm.; petals 1}-1} mm.

Diameter of drupe about 4 mm.

Rh. crenata Thunb. Daly 1065. Under side.

Distripution: On fixed sandhills along the coast from the Cape Peninsula to the

Kei River mouth, flowering and fruiting almost throughout the year. A sand-dune fixed

chiefly by Rh. crenata and Myrica cordifolia L. is figured by Marloth (Das Kapland 1908,

fig. 10). The former favours the ridges.

Type in Herb. Thunberg, Upsala; Burchell 5331 (only few leaves crenate at the apex); between

Retreat and Diep River, Dimmer 1519; Cape Flats, about 2 miles from Eerste River, Van Roeper 5058 ;

Wagenhuis Krans, Bredasdorp, Fry in herb. Galpin 4975; sandy places near the Onrustriver, 10 feet,

Schlechter 10394 ; Great Brak River, 30 feet, Schlechter 5738 ; seashore, Plettenberg Bay, Fourcade 596 ;

Groot River mouth, Zahn 5039; sandhills on the shore near Cape Recief, HE. & Z. 1123; neighbourhood

of Port Hlizabeth, Ecklon 20, Paterson 1108, Cooper 1571, J. Sim 36, Daly 1065, Kemsley 316 ; neighbour-

hood of Port Alfred, Hutton 988, Burchell 3829, 3830, Britten 1910, 2751, White 61, Salisbury 130:

(p. pte.); Hast London, Rattray 35, Galpin 5688; Keimouth, Flanagan 846.

20. Rh. Pentheri Zahlbr. in “ Plantae Pentherianae ’ (Ann. Hofmus. Wien XV, 52).

Rh. cuneata N.W. Br. in Kew Bull. 1906, 17.

Zahlbruckner compares this species to Rh. undulata Jacq., and Rh. krebsiana Presl,

with neither of which, however, it is allied. N. H. Brown says that Rh. cuneata [which

must be sunk in Rh. Pentheri| “is allied to Rh. crenata, but easily distinguished because

the leaves are without glands on the under surface”. To my mind, it is also allied to

Rh. refracta HK. et Z.

Description : A much branched shrub up to 9 feet high, branchlets subterete cinereo-

villous. Leaves petioled, petioles subterete canaliculate above, usually about half the

length of the terminal leaflets, sparsely pilose. Leaflets membranous, at first covered with

sparsely adpressed hairs, later glabrescent or glabrous, dark green above, paler below, elongate

obovate-cuneate, apex obtuse, tridentate or emarginate, margin slightly revolute, midrib

prominent on both surfaces, lateral veins very slightly branched outwards and very

slightly prominent on both surfaces, tertiary vems few, coarsely reticulate, but barely visible

or invisible in~ dried specimens. Panicles cinereo- pubescent, lateral and terminal, laxly

branched, usually shorter than the leaves. Bracts minute, ovate oblong or oblong sub-

acute, pedicels short. Calyx lobes triangular-ovate, obtuse, pubescent. Petals ovate.

Drupe globose, laterally somewhat compressed and slightly depressed, shining brown when

ripe.

Length of petioles 5-13 mm. ; terminal leaflets 2-3 em. ; lateral leaflets 1-1-5 cm

Breadth of terminal leaflets 4-12 mm.; lateral leaflets about -5 cm.

Length of inflorescences 2-5-7 em.; pedicels 1-1-5 mm.; calyx lobes about -5 mm.

Length of petals 1-2-1:4 mm

Drupe: about 3 mm. high and about 3-5 mm. greatest diameter.

Rh. Pentheri Zahlbr. Schlechter 6301. Under side.

Distrrpution : Bashee, Pondoland, eastern Orange Free State, Natal, Barberton and

Lydenburg districts of the Transvaal.

Near Colossa, Natal [Colenso ? S. Sch.], leg. Krook, Penther 2290 in Herb. Vienna; on a rocky hill

near Ladysmith, 3,000 feet, Wood 5706 in Herb. Kew and Herb. Bolus (type of Rh. cuneata N.E. Br.) ;

Bashee River, 2,500 feet, Schlechter 6301 ; Lusikisiki, Ntsubane forest station, Fraser 5163; woods near

Gromberg, Natal, Wood 891; Inanda, Wood, 1319; Illovo, Wood 3106; Natal, Gueinzius, Sutherland

and Gerrard 1394 in Herb. Kew; between Pinetown and Umbilo, Rehmann 8072; Umkomanzi, 3,000

feet, Schlechter 6690 ; Arnoldsfarm, Newcastle, Rehmann 7052 ; common in low forest in the Barberton

and Lydenburg districts, Keet 1432 (5159), 5160, 1485.

21. Rh. divaricata H. et Z. in Enum. 1106; Sonder le. 508; Engler lc. 429; Diels

Le. 582.

Description: A small shrub with spreading branches; branchlets terete and like

the petioles minutely pubescent. Leaves petiolate, petioles about half to two-thirds of

the length of the terminal leaflet, furrowed above. Leaflets subcoriaceous, sessile, ovate or

obovate, obtuse and plicate-mucronulate or emarginate, the terminal cuneate at the base,

margin entire or paucidentate, slightly revolute, above puberulous and with the exception

of the rufescent veins dark green, below very shortly whitish or fulvous glandular tomentose

eventually subglabrous; midrib and the few slightly branched lateral veins slightly

prominent on both surfaces, tertiary veins very indistinct. Panicles axillary and terminal,

sparsely flowered, pubescent, shorter than the leaves. Flowers pedicellate. Calyx densely

pilose, segments oblong, triangular. Petals oblong, sparsely pilose. Drupe (fide Sonder)

globose, smooth, tipped with the three styles.

Length of petioles 7-14 mm. ; terminal leaflets 1-5-2 cm. ; lateral leaflets 1-5 cm.

Breadth of terminal leaflets 4-6 mm. ; lateral leaflets 4-6 mm.

Length of panicles 1-3 cm. ; pedicels 1-3 mm.; calyx lobes nearly 1 mm.; petals 2 mm.

Rh. divaricata E. et Z. Galpin 2540. Upper side.

DistrrpuTion : Mountains from the neighbourhood of Queenstown to Basutoland and

Ladybrand, Orange Free State, also in the Transvaal (?).

The description and measurements are taken from what I consider to be typical

specimens.

Mountain sides on the Klipplaat River (Tambukiland), Alt. 5, Noy. (fl.), E. et Z. 1106, Ecklon 5 (in

Herb. S.A. Mus.); Andriesberg, Jan., Galpin 2540.

There are a few other specimens which differ more or less, but which seem to have to

be referred to the same species.

Aliwal North, 5-6,000 feet, Drége 6796 in Herb. Kew (leaves up to 3 em. long). This is Rh. sub-

ferruginata Presl Bem.

Leribe, Basutoland, 5-6,000 feet, Dieterlen 17 (wrongly named Rh. pyroides Burch. in Herb. 8.A.

Mus.). Terminal leaflets up to 2-3 cm. long, 1-5 em. broad. Many leaves have more or less crenate

margins in the upper part. Drupe about 4 mm. in diameter.

Rh. divaricata E. et Z., var. fulvescens Engl. Rehmann (type).

Mountain slopes, Leribe, Basutoland, shrub, fl. white, Dieterlen 17b (wrongly named Rh. Zeyheri

Sond.). This has remarkably slender petioles (up to 4 cm. long). The largest terminal leaflets are

3-5 em. long. The texture, venation and pubescence of the leaflets are the same as in the type, also the

flowers. The margin of the leaflets show sometimes a few crenate teeth. The leaflets are obovate, narrowly

cuneate at base.—Scattered over the mountains at Ladybrand, Orange Free State (Patterson 5134), there

is alarge-leaved form with fruits as in Dieterlen 17a (subglobose, brown, shining, c. 4 mm.indiameter). The

petioles are 2-2-5 em. long. The largest terminal leaflets are 5 cm. by 2:2 cm. broad, the lateral leaflets

belonging to them about 3-5 cm. long.

Then there are Dieterlen 17¢ and Phillips 753 (Leribe in Herb. S.A. Mus.). These give one the

impression of being coppice shoots ; im the latter the leaflets reach a length of 6 cm. The shape of the

leaves, especially in Dieterlen 17c, varies considerably ; some are distinctly acute.

The material of Rh. fulvescens Engl. in Herb. Berol. (Rh. divaricata var. fulvescens Engl. l.c. 429),

Diels l.c. 582, collected by Rehmann at Trigaardsfontein, Transvaal, is hardly sufficient to judge whether

or not it should be placed with Rh. divaricata. The leaflets are paucidentate.

22. Rh. rupicola Wood et Evans in Journ. of Bot. 1897, 350.

Rh. Tyson Philips in Ann. 8. Afr. TX, 119.

? Rh. colensoana Engl. in Pflanzenwelt Afrikas III, 2.

Description: Erect, much branched shrub with terete branches, at first tomentose,

later often glabrescent. Branchlets short, ascending, terete, tomentose or softly pilose.

Leaves petiolate, petioles furrowed above, pilose, usually little more than half the length

of the terminal leaflet. Leaflets broadly obovate or obovate-elliptical, at the apex obtuse,

subobtuse or slightly emarginate, usually very distinctly plicate-mucronate ; at the base

cuneate, frequently (especially the terminal ones) subpetiolulate or petiolulate,

surfaces more or less softly pilose at first, later more or less glabrous; margin entire,

slightly revolute ; midrib barely prominent above, strongly prominent below, lateral veins

few, unbranched or slightly branched near the margin, delicate above, distinctly prominent

below, tertiary veins few, reticulate and very indistinct. Panicles pubescent, axillary

(smaller than the leaves) and terminal (longer then the leaves), lax, few flowered with

pedicelled flowers and minute subulate bracts. Calyx lobes ovate, subglabrous or pubescent.

Petals oblong. Drupe subglobose a little broader than high.

Average length of petioles about 7 mm. ; terminal leaflets about 1:3 em. ; lateral leaflets about 1 em.

Average breadth of terminal leaflets about 8 mm. ; lateral leaflets about 5 mm.

Length of axillary panicles about 1 cm. ; terminal panicles 2-2-5 cm.

Length of pedicels about 1 mm.; calyx lobes about # mm.; petals 14-1} mm.

Diameter of drupe about 4mm.

Rh. rupicola Wood et Evans. Wood 3932. Under side.

Disrripution : Mountains of Griqualand Kast and Natal at an altitude of 4,500 to

5,500 feet.

In a rocky valley, Liddesdale, Maritzbure county, 4-5,000 feet, Feb. (fl. and fr.), Wood 801, 3932 ;

Malowe forest subreserve, district Umzimkulu, 4,500 feet, Jan. (fl.), Miller D/326 (a small shrub about

2 feet high in grassland, probably dwarfed by grazing and burning, only one plant seen) ; amongst rocks

at Sibiskraal, near Matatiele, Tyson 1268; Zuurbergen, Natal, 5,500 feet, Schlechter 6584. The last

two have leaves more hairy than the others on the surface and ciliate on the margin.

Wood and Evans think that this species is allied to Rh. mucronata, while Phillips thinks it is allied

to Rh. glauca. Both of these suggestions are clearly wrong.

LEGATI group.

23. Rh. Legati Schonl. nom. nov.

Rh. laevigata Thunb. (in Prodr. 52 and FI. Cap. ed. Schultes, 264) et auct. plur.,

non Linn.; E. et Z. in Enum. 1096; Pappe, Sylv. cap. 12; Sonder l.c. 514;

Engler |.c. 443; Diels l.c. 589, 639; Sim, Forest Flora 195, t. XLIV.

Rh. acuminata KH. Mey. in Herb. Drége; Rh. crassinervia Presl Bem. 42,

Description: A shrub or (usually) a tree, often reaching a height of 50 to 80 feet,

glabrous or rarely slightly pilose, with reddish, striate branchlets. Stems of young plants

and of coppice shoots thorny. Leaves petiolate, petioles slender, furrowed above, variable

in length, but often more than three-quarters the length of the terminal leaflets. Leaflets

dark green or reddish, firmly membranous, ovate or oblong, acuminate, with a cuneate

base which is often, especially in the terminal leaflets, contracted to form a petiolule ;

surface slightly undulate, margin entire or rarely paucidentate ; midrib raised slightly on

lower surface, lateral veins delicately curved, slightly raised, tertiary reticulate, immersed.

Panicles richly and delicately branched, male chiefly terminal, much exceeding the leaves

in length and much longer than the female which are mostly in the axils of the upper leaves ;

bracts minute, subulate, pedicels very delicate, longer than flowers. Calyx lobes ovate,

subacute not quite half the length of the ovate-acute pale green or yellowish petals. Drupe

red, shining, subglobose.

Length of petioles 4-7 cm.

Length of terminal leaflets 7-13 cm. Breadth of terminal leaflets 2-4 cm.

Length of lateral leaflets 5-5-10-5 cm. Breadth of lateral leaflets 2-4 cm.

Length of pedicels 2-5 mm.; calyx lobes barely 4 mm.; petals barely 14 mm.

Diameter of drupe 4-6 mm. (usually 5 mm.).

Distrisution: In the coast districts from the Grootvadersbosch, Swellendam, to

Natal, extending somewhat inland even to the northern Transvaal from near sea-level to

nearly 6,000 feet. Rarely absent from forests and forest patches, also frequent in stream-

bank bush.

On the edges of forests and in stream-bank bush usually a large bush or small tree

with smaller leaves and fruit than when growing in forest, w here it always develops into a

conspicuous tree.

The tree is known as the * Red Currant tree’. It is also known as ~ Kiriehout ” and

“ Bush Jarrah”. In western districts of Cape Colony it is called “ Essenhout”’, the

natives of the eastern districts call it ““ Bosganna ”

The wood is red, hard, used by natives for pipes and fancy sticks, sometimes used for

yokes and skeys, but not much used by sawyers. Drupe sweetish, when ripe, and eaten

by natives, monkeys, and birds. When ‘the stem is cut into or bruised it exudes blood-red

sap.

I have felt compelled to give this plant a new name after inspecting Linnaeus’ type

of Rh. laevigata and Thunberg’s plant, which he and numerous other authors have called

Rh. laevigata. The two synonyms could not be adopted either. Rh. acuminata K. Mey.

must be dropped because there is already a Rh. acuminata DC. (=succedanea Linn. from

Eastern Asia). Nor can I accept Rh. crassinervia Presl, because Presl did not characterize

his species. Moreover, the name would be most inappropriate. I have taken the hberty

to name it after Mr. C. E. Legat, Chief Conservator of Forests, Union of South Africa,

under whose direction the Forestry Department has given me much assistance by the

supply of material and information.

In view of the fact that the species, within the limits mentioned, has such a universal distribution,

I do not, as a rule, quote exact localities of the perfectly glabrous specimens I have examined. Cape

Colony : Herb. Thunberg (Rh. laevigata Thunb. non Linn.) ; Burchell 7224, 4485 ; Drege 3451, 3452, 3568

(Rh. acuminata E. Mey.) ; E. et Z. 1096; Z. 25; Cooper 1548 ; Jordaan (5106) ; Duthie 23,35; Schlechter

5912, 5912a; Fourcade 1084; Zahn (5017) ; Paterson 1985; E. et Z. 314; Paterson 19lla; Holland

Rh. Legati Schonl. Paterson 191la. Upper side.

294 (Zuurberg, drupes exceptionally large, about 6 mm. in diameter); MacOwan 294; Britten 1573 ;

Schonland 3143 (2 miles beyond Stone’s hill, Grahamstown, leaves exceptionally broad, up to 6-5 cm.

and terminal up to 12 cm. long, more leathery than usual); Hoesslin (5126); Staples (4492, 4494, 4495) :

Dyer 790; Stayner 80; Hunter (5031); Sim 1990; Tyson 2591, 3133; Hilner 144; Schlechter 6176 ;

Flanagan 766; Kolbe and Pegler 3; Van der Merwe (5103); Bennie 392; Dwyer (5114); Tyson 3134 ;

Fegen (5161); Fraser 5036. Natal: ? (5024—Ingwangwane, near Riverside station) ; Wood 789, 2525 ;

Rudatis 558, 1290; Krauss 124; Gerrard 74, 524; Transvaal (near Vryheid, Waterval Boven, Zout-

pansberg, and Pietersburg divisions) : Tusten (5056) ; ? (5006), Rogers 18070, 22034; Pole Evans 16934,

16941.

Hyprips: Occasionally one finds slightly pilose specimens in the neighbourhood of

typically glabrous. ones. Whether this is due to hybridization cannot be decided at

present. Amongst these are :—

Duthie (25901) in (Herb. S.A. Mus.), Knysna; Mund (in Herb. §$.A. Mus.), Zitzikamma forest ;

Tusten (5122), Ngomi forest reserve, P.O. Hlobane, near Vryheid.

On the other hand, where this species occurs with Rh. dentata, there are frequently

found forms intermediate between them, about the hybrid character of which there can

be no reasonable doubt. These are referred to under Rh. dentata. Dr. J. Muir (n. 3474) sent

me a specimen from Vet River, near Novo, which I look upon as a hybrid between Rh.

Legati and Rh. mucronata. It resembles the former in the nature of the inflorescence,

shape and waviness of some leaves. The petioles vary much in length (from less than

half to over half the length of the terminal leaflets). The leaflets are more

leathery than in Rh. Legati and have very distinct reticulate venation. The flowers are

larger than in RA. Legati (petals 14 mm.). Petioles and inflorescences are sparingly hairy

and the leaflets also show a few hairs. (See further Rh. carnosula Schon.)

24. Rh. transvaalensis Engl. l.c. 440; Diels l.c. 638.*

Description: A much branched shrub, generally 3 to 4 feet high, but sometimes

reaching a height of 12 to 15 feet. Young branches villous, soon becoming glabrescent.

Leaves petiolate, petioles rather slender, half to two-thirds the length of the terminal

leaflets, furrowed above, more or less pilose, strongly so on the edges of the furrows. Leaflets

subcoriaceous, green, paler on the lower surface, the younger more or less pilose, the

older sometimes quite glabrous, oblong-elliptical or oblong-lanceolate, narrowed at the

base and apex, acute, mucronulate, margin slightly thickened, often slightly undulate,

venation reticulate, not prominent, except the midrib and, to a shght extent, the lateral

veins on the lower surface. Panicles numerous, lateral, few-flowered, shorter than leaves,

pilose, bracts small subulate, flowers pedicellate. Calyx lobes oblong-obtuse. Petals

oblong. Drupe subglobose, shining, pale red or nearly white.

Rh: transvaadlensis Engl. (5005). Under side.

Length of petioles one-half to two-thirds of terminal leaflets.

Length of terminal leaflets 2-5-4 cm. Breadth of terminal leaflets 1-1-3 cm.

Length and breadth of lateral leaflets one-half to two-thirds of the terminal.

Length of fruiting pedicels 2-4 mm. ; calyx lobes nearly 1 mm.

Diameter of drupes about 4 mm.

*R. transvaalensis Engl. should be better placed next to Rh. pyroides var. puberulaand may perhaps even

have to be sunk in it. I only recognised this when it was too late to alter the arrangement of the species.

DistRIBUTION : Common along streams and in scrub-forest on the Drakensberg in

the Barberton and Lydenburg districts.

Houtbosch, Rehmann; “ Hilltop”, Barberton district, Nov., Keet 1456; Halic forest station,

Zoutpansberg, Dec., Albany Mus. Herb. 5005 (fr.); Ngomi, 15 miles from Vryheid, 2,000-4,000 feet,

Jan., Forester Tusten 5057 (fr.).

25. Rh. Zeyheri Sond. l.c. 514; Engler l.c. 433; Diels Ic. 584, 639.

Rh. glauco-virens Engl. l.c. 432; Diels lc. 584.

Diels states that Engler’s species is difficult to separate from Rh. Zeyheri. Engler, in

his key to the species of Rhus in “ Pflanzenwelt Afrikas”’ III, 2, 213, places Rh. Zeyheri

under Rh. mucronata Thunb., while keeping up Rh. glauco-virens.

Description: A V erfectls glabrous, much branched shrub with slightly araegalom, often

purplish branchlets. Leaves petiolate, petioles rather slender, half to two-thirds the length

of the terminal leaflets, slightly furrowed above. Leaflets glaucous green, slightly lighter

on the lower surface, sessile or the terminal petiolulate, obovate or obovate-oblong, cuneate

at the base, obtuse, acute or apiculate, often mucronulate; margin entire or rarely

paucidentate, flat; midrib somewhat prominent on both surfaces, lateral veins slightly

prominent on both surfaces, tertiary veins reticulate, immersed. Panicles laxly flowered,

axillary shorter than the leaves, terminal somewhat longer, bracts subulate, Gowers

pedicellate. Calyx lobes ovate, obtuse. Petals oblong, yellowish. Drupe fleshy, brown,

shining, subglobose.

Length of petioles 1-1-5 cm.

Length of terminal leaflets 2-3 em. (rarely up to 6 cm.). Breadth of terminal leaflets about 1-7 cm.

Lateral leaflets barely two-thirds the size of the terminal leaflets.

Length of calyx lobes about $ mm. (variable in size in the same flower sometimes).

Length of petals about 14-14 mm. Diameter of drupe 5-5-5 mm.

Rh. Zeyheri Sond. Rehmann 4740. Under side.

DistRIBUTION : On edge of streams and on kopjes near Pretoria, varying considerably

according to habitat. Also recorded from the Drakensberg.

Z. 345 (Herb. Kew: type of Rh. Zeyheri Sond.) ; Rehmann 4740 (type of Rh. glauco-virens Engl.)

Burtt-Davy 291, 2532, 2691; McLea in Herb. Bolus 5618; Leendertz 83, 324, Verdoorn 8, Schlecht:

3626, Sturdy 4266.

At Groenkloof, near Pretoria, a form with deep-green leaves occurs (Howlett 5 and 5a). It cannot

be separated by any other character from the type. Some of Pole Evans’s specimens from Irene were

named by Burtt-Davy Rh. puberula var. Zeyheriana.

LUCIDA group.

26. Rh. lucida L.Sp. Pl. 382; Rh. lucida B Ait. hort. Kew, ed. 2, 11, 266 ; Comm. hort. 1,

t. 93; Thunb. Fl. Cap., ed. Schultes 264; Jacq. hort. Schoenbr. t. 347; Hngler

l.c. 413; Diels l.c. 574, 632, fig. 7 A, B (p. 634); Rh. Cavanillesii DC. Prod. II, 69 ;

Rh. outeniquensis Sez.; Rh. scoparia BK. & Z.

Rh. lucida L., is, in eighteenth century collections, often mixed up with Rh. glauca.

However, this has usually obcordate thinner leaflets with lateral veins placed close

together, etc. There is a scrap (without flowers) in the Plukenet Collection, British Museum,

named Rh. lucidum Mill. Pluk. t. 219, fig. 9, which appears to be typical Rh. lucida if

Amongst typical specimens of Rh. lucida which are found in many herbaria, E. & Z. 1113,

from the Lion’s Head, Capetown, and MacOwan 1827 in Herb. Austr. Afr. from Constantia

may be mentioned. Such typical specimens can easily be distinguished from typical Rh.

scoparia KE. et Z., but, having examined a large amount of material, I am unable to separate

the latter sharply from RA. lucida. Already Diels (l.c. 575) has, contrary to Sonder, stated

that the characters of Rh. scoparia : ~ short petioles, smaller leaves, longer downy panicles ”

have no systematic value, though I have long hesitated to unite these two species, since

Rh. Schlechteri might then also have to come into Rh. lucida.

The varieties established by Sonder I.c. 517 cannot be kept up. I divide Rh. lucida

into three varieties :—

A. typica, B. outeniquensis (Rh. outeniquensis Szysz.), and C. scoparia (Rh. scoparia

H. et Z.).

The wood is used for fencing posts and spokes in the Knysna division.

Description : A much branched shrub, often over 10 feet high, with pulverulent-

puberulous or shortly downy branchlets, sometimes covered with resmous excretion. Leaves

petioled, petioles of variable length, slightly winged and canaliculate above. Leaflets often

glossy, subcoriaceous, obovate- oblong, cuneate at the base, quite blunt or subacuminate,

rarely emarginate or coarsely crenate at the apex. Midrib slightly prominent, lateral veins

delicate but distinct, tertiary veins usually not visible. Panicles lateral and terminal,

shorter or slightly longer than the leaves, puberulous or covered with resin, lax. Flowers

sometimes hermaphrodite, shortly pedicelled. Calyx lobes subtriangular- ovate, often

puberulous. Petals oblong, about three times the length of the calyx ‘ole. Drupe brown,

shining, subglobose.

Length of petioles from 1 mm. to 3-5 cm. (see varieties).

Length of terminal leaflets 14-7 cm. Breadth of leaflets 5 mm. to 2} cm.

Lateral leaflets about two-thirds the size of the terminal leaflets.

Length of pedicels 1-2 mm. ; calyx lobes about } mm.; petals 1{-1j mm.

Diameter of drupe 3-5 mm.

Rh. lucida L., A. typica E. et Z. 1113. Under side.

Distripution : On edges of scrub and forests and isolated on mountain slopes in the

coast districts from the neighbourhood of Capetown to Natal (also found in Gazaland at

an altitude of 7,000 feet: Swynnerton M 635).

A. typica.—Petioles usually 5-6 mm. long. Terminal leaflets usually about 34-4 cm.

long (except in coppice shoots, in which they may exceed 6 cm. in length). Drupe (as far

as known) ec. 34 mm. in diameter.

Common on hillsides near Capetown, e.g. E. et Z. 35, 113, Wolley Dod 2029, Schlechter 1333, Drége |

6791 (var. subdentata DC.), 6802, MacOwan in Herb. Austr. Afr. 1827, Rogers 11242, Lichtenstein 193,

Wilms 3123, Engler 52, 55 ; in dunes at the mouth of the Onrust River, Z. 2235, 2248 (var. elliptica Sond.) ;

hills near Paarl, 300 feet, Schlechter 9208; edge of bank of Knysna River, Schonland 3430; northern

slopes of Knysna Heads, 50-200 feet, Schonland 3378 ; Humansdorp division, Rogers 28266 ; ridges near

Groendal, Zwartkops River valley, common, J. Sim 227; Springfields, near Uitenhage, J. Sim 129; Port

Alfred, Rogers 28036 (leaflets more decidedly narrowed in lower part than usual, many crenato-dentate

at apex); Blaauwkrantz, Hilner 63; common on the hills south of Grahamstown, e.g. Britten 1608,

1522, 1593, 1594, 2179, Schonland 1892 (very near Rh. scoparia H. & Z.), 82 (not distinguishable from

Capetown specimens); Aylesby, near Riebeek East, Schonland; margins of lower forest, Hogsback,

Rattray 357 ; Fort Cunynghame, Sim 2185; Ntsubane, near Lusikisiki, Fraser (5137, 5144); Bushmans

River valley, Natal, Wood 10635; Friedenau, Umgaye flats, Alexandra county, c. 700 m., Rudatis 115,

1130.

B. outeniquensis (Rh. outeniquensis Sez. m Polypetalae Disciflorae Rehmannianae,

Cracoviae 1888, 52).—Branchlets more slender than usual, leaves larger than in A, drupe

larger. Petiole generally about 1 cm. long, terminal leaflets generally 5-7 cm. long, drupe

(as far as known) 34-5 mm. in diameter.

Rh. lucida L., Bs outeniquensis. Fourcade 8. Under side.

Montagu Pass, Rehmann 272 in Herb. Berol.; Piquetberg, 1,000-1,500 feet, Schlechter 5199; K1.

Kruis R., Muir 3475; Muiskraal, Garcias Pass, 1,000 feet, Galpin 3895; Gr. Brak R., 30 feet, Schlechter

5737; Rust en Vrede, Oudtshoorn, 3,000 feet, Dyer 85; Uniondale, Paterson 3019; Knysna heads,

Williamson 29; Sourflats, Knysna, Keet 447, 612, 614, 615; Gully, north-east of Royal Hotel, Knysna,

Schonland 3494; Knysna, Marloth 7917 (petioles up to 2-5 em. long); Blaauwkrantz forest reserve,

Zahn (5014, 5027, 5028 (petioles in 5027 up to 3 em. long); margin of coast forest, Ratelsbosch, 600 feet,

Foureade 8 (petioles up to 14 em. long); Van Stadens, Paterson 736.

Perhaps Schlechter 6754 collected at Van Reenen belongs here also. The petioles here reach a length

of 3-5 cm.

__C. scoparia (Rh. scoparia BK. et Z. in Enum. 1122, Sonder lc. 518, Engler l.c. 415;

Diels l.c. 575, 632).—Petioles 1-4 mm. long. Terminal leaflets generally 14-24 cm. long.

Drupe (as far as known) about 3 mm. in diameter.

Eh, lucida L. (close to var. C. scoparia). Schonland 1892. Upper side.

Olifantshoek, Alexandria division, E. & Z. 1122 ; Nieuweveldberge, Drége 6803 ; Knysna commonage,

not common, 600 feet, Keet 660; Humansdorp, under 500 feet, Rogers 2908; Lonegvale, P.O. de Kol,

Alexandria division, common, Gant 47 ; Springfields, Uitenhage, J. Sim 129; road from Port Alfred to

Three Sisters, Britten 720; hills south of Grahamstown, Schonland 615. Britten 1594, 2804.

The following comparative notes on the structure of the leaves taken from Diels l.c. 575

are of interest :—

Rh. lucida L. Rh. scoparia EK. et Z. Rh. Schlechteri Diels.

Hairs— Hairs— : Hairs—

Glandular scales on both sur- Numerous glandular scales on Glands especially active when

faces (Pl. XIV O) both surfaces. young.

Epidermis— Epidermis— Epidermis—

Lumen, 10-15 up. Lumen, 10-15 p. Lumen, 15-20 p.

Wall, 3-5 pu. Wall, 5-7 up. Wall, 6-12 p.

Ground tissue— ; Ground tissue— Ground tissue—

Only the outer layer palisade Loosely palisade like. Loosely palisade like.

like.

Stomata— Stomata— Stomata—

Only on lower surface, slightly Many on lower surface, slightly On lower surface, numerous,

raised. raised. not raised.

These notes confirm, as far as they go, the general impression one gains in handling

these species: that they form a series towards more and more decided xerophytism, and it

is a great pity that cultural data are hitherto absent to decide in how far their characters

have become fixed or not.

HysripizATion : Occasionally one finds forms which lead to the conclusion that

hybrids with Rh. glauca Thunb. and allied species occur. The var. elliptica Sond. (e.g.

Z. 2248 from downs near Onrust River) seems to be composed of such hybrids.

27. Rhus Schlechteri Diels in Engl. Bot. Jahrb. XXIV (1898), 501;] ib., 575, 634, fig. 7 E.

This species is very close to Rh. lucida var. scoparia. The practically sessile and

relatively broader leaves distinguish it from this species. Diels describes it from Schlechter’s

specimens, collected at Elim, which are without flowers and fruits.

Description: A much branched shrub up to 10 feet in height with short, densely

leafy, slightly angular branchlets, which are pubescent when young and become glabrous

later. Leaves sessile. Leaflets subcoriaceous, often with a resinous secretion which may,

on drying, form a thin greyish crust, obovate or obcordate-cuneate, at the apex obtuse,

subtruncate or slightly emarginate; margin entire; midrib slightly raised, lateral ves

immersed or slightly raised, tertiary veins rarely visible. Panicles lateral and terminal

slightly longer (or female shorter) than the leaves, laxly flowered. Flowers pedicellate.

Sepals ovate, obtuse. Petals oblong. Drupe globose, shining.

Length of terminal leaflets 8-14 mm. Breadth of terminal leaflets 5-10 mm.

Lateral leaflets about two-thirds the size of terminal ones.

Length of pedicels about 1} mm. ; calyx lobes } mm.; petals about 14 mm.

Diameter of drupe (in J. Sim 78) 4 mm.

Rh. Schlechteri Diels. Burchell 4522. Under side.

DistRiBuTION : Chiefly on fixed sand-dunes from Bredasdorp division to the neighbour-

hood of Port Alfred, often gregarious and dense.

On rocky places near Elim, Bredasdorp division, about 150 m., Schlechter 7624 in Herb. Berlin ;

Burchell 4522 (quoted by Engler both under Rh. glauca and Rh. scoparia) ; Butialo Bay, Knysna ; Duine-

veld, between the Kromme and Zitzikamma R., J. Sim 78; near Schoenmaakers Kop, Port Elizabeth,

J. Sim 27; Walmer, Paterson 826; Humewood, Daly 1051; frequent near the Kasouga and Port Alfred,

Britten 2108, 2295, 5015.

28. Rh. africana Mill. dict. n. 11; Sonder lc. 517 p. pte; Engler I.c. 414 p. pte; Diels

le. 574, 632 p. pte; Rh. mucronata HE. et Z. nm Enum. 1129 (non Thunb.).

A curious mistake has been made in connection with this species. Under Ecklon 37

two different species have been distributed, one from Witsenberg, Tulbagh, is Rh. cunerfolva

Thunb., the other (= E. et Z. 1129) is closely allied to Rh. lucida, with which it agrees in

many characters, especially in the nature of the drupe, which is not described by Sonder

in Fl. Cap. I, 517. Unfortunately, however, Plukenet’s specimens in the British Museum

leave a little doubt as to what is really meant by Rhus africana. There are two branches

bearing this name which undoubtedly belong to two different species. The lower (without

flowers and fruit) may be the plant which Sonder took to be his var. macrophylla of Rh.

africana Mill. Next to it is a printed label bearing the words “ type specimen ”’, and I have.

therefore, taken it as the type.

Rh. africana Mill. E. et Z. 1129, Upper side.

Description : A low much branched shrub with densely pubescent branchlets. Leaves

shortly petiolate, petioles at first pubescent, later often glabrous, short, canaliculate above,

winged. Leaflets coriaceous, glabrescent or glabrous, sessile, obovate-cuneate, or oblong-

cuneate, subobtuse, shortly plicato-mucronulate, margin slightly revolute, usually entire ;

midrib and lateral veins raised on both surfaces, especially on the lower, tertiary veins

not visible. Panicles axillary, puberulous, shorter or slightly longer than the leaves, pedicels

very short. Calyx lobes ovate, petals oblong. Drupe as in Rh. lucida.

Length of petioles 8 mm. to 2-1 cm.; terminal leaflets 4-6 cm. (rarely longer).

Breadth of terminal leaflets 2-3 cm.

Lateral leaflets about two-thirds the size of the terminal leaflets (or smaller).

Length of calyx lobes 1 mm.; petals 2 mm.

Diameter of drupes about 4 mm.

DistRIBUTION : From the Tulbagh division through Clanwilliam to the Van Rhynsdorp

division (also in the Hex River Mts. ?), flowering in June, July, August.

Rocky places, Vogel Vley, 1,500 feet, Schlechter 7529 ; amongst heaths (alt. V), mountain sides near

Brakfontein, Clanwilliam, E. & Z. 1129 ; summit of Heerelogement, Z. In specimens at Kew, collected at

Els Kloof, Hex River, Wolley Dod 4044, the leaflets are crenato-dentate in the upper portion.

29. Rh. albomarginata Sond. I.c. 519, Engler l.c. 413, Diels 574, 632.

This species is allied to Rh. lucida (not excisa as Sonder said). It has large flowers.

It is, as far as I know, not represented in any South African herbarium, and, although

I know the places where it was found fairly well, I have never seen it growing.

Rh. albomarginata Sond. Burke in Herb., Kew.

Description : A glabrous shrub with angular compressed branchlets. Leaves petioled,

petioles nearly half the length of the terminal leaflets, canaliculate above and margined.

Leaflets slightly undulate, coriaceous, obovate-oblong-cuneate or the terminal subrhomboid

in the upper part passing into a narrow cuneate base, lateral oblong-cuneate, blunt or

slightly acuminate, entire with thickened, white, shghtly revolute margin; midrib com-

paratively broad in lower half, slightly prominent below, lateral veins slightly prominent

below, slightly branched towards the margin. Panicles axillary, shorter than the leaves.

Calyx lobes ovate, petals oblong-ovate. Drupe ?

Length of petioles 1-5-2-5 cm.

Length of terminal leaflets 3-4 em. Breadth of terminal leaflets 1-1-5 cm.

Lateral leaflets about two-thirds the length of terminal leaflets.

Calyx lobes about 1 mm. long (teste Engler). Petals about 2 mm. long (teste Engler).

Disrripution : Only collected in the neighbourhood of Grahamstown, where, however,

it must be very rare.

Herb. Kew: Slaay Kraal, Burke; Sidbury, 1,000 feet, MacOwan 749.

30. Rh. seytophylla E. et Z. in Enum. 1130; Sonder I.c. 517, Engler I.c. 412, Diels l.c. 574,

GBI, rex, }) 1D),

Derscrirtion: A shrub reaching a height of 6 feet with glabrous or puberulous

branchlets. Leaves shortly petioled, petioles broadly margined. Leaflets coriaceous,

sessile, obovate-oblong-cuneate, blunt or emarginate, sometimes mucronulate, usually

entire, with slhghtly revolute margin, sometimes shining; midrib slightly prominent on

the upper side, more so on the lower, lateral veins usually slightly prominent on both

surfaces, sparingly branched in the upper part of the leaflets near the margin, tertiary veins

not visible. Panicles multiflowered, lateral and terminal, puberulous, longer than the

leaves, flowers pedicelled. Calyx lobes ovate, glabrous or minutely puberulous. Petals

oblong. Drupe globose (teste Sonder “as in Rh. lucida”).

Length of petioles usually 5 mm., but may reach 1-5 cm.

Length of terminal leaflets 2-5-4 cm. Breadth of terminal leaflets 1-1-5 cm.

Lateral leaflets about two-thirds the size of the terminal ones.

Length of pedicels 2-3 mm.; calyx-lobes about 4 mm.; petals about 14 mm.

Kh. scytophylla EB. et Z. Z. 2247. Under side.

DISTRIBUTION : On the mountains near Tulbagh, Stellenbosch, Caledon, and Swellendam,

flowering from April to July.

Grietjesgat, near Palmiet R., E. & Z. 1130; Hottentots Holland, Z. 2247; Witzenberg, Pappe ;

Howhoek, 1,500 feet, Schlechter 7767, and the following in Herb. Kew: Z. 2246, Drége 68076, Burchell

7835, 8662, 8033 (in the last many leaves are crenate-dentate at the apex). y

31. Rh. glauca Desf. arb. II, 326; Thunb., Fl. Cap. ed. Schultes, 265; Sonder l.c. 516;

Engler l.c. 41P; Diels l.c. 573, 635; Marloth, Das Kapland 526.

Rh. Thunbergiana Roem. et Schult. Syst. Veg. VI, 657; Rh. lucida EH. Mey.

in Drége exsicc. (non Linn.). :

Description: An unarmed glabrous shrub with subangulate branches more or less

covered with resin, which eventually dries and becomes grey. Leaves petioled, resinous

or, when the resin dries in summer, covered with a grey powder, petioles canaliculate

above, shghtly winged. Leaflets sessile, terminal usually obcordate-cuneate, lateral obovate

or obcordate, sometimes plicate-mucronulate ; midrib and lateral veins (which are often

minutely wavy) slightly raised above and below, tertiary veins not visible. Panicles

terminal and lateral much branched, two to three times longer than the leaves, flowers

pedicelled. Calyx lobes ovate. Petals oblong. Drupe globose, shining, reddish.

Length of petioles 8-12 mm.

Length of terminal leaflets 1-5-2 cm. Breadth of terminal leaflets 1-1-2 em.

Lateral leaflets about two-thirds the size of the terminal ones (sometimes less).

Length of pedicels 1-2 mm.; calyx lobes } mm.; petals 14 mm.

Diameter of drupe about 5 mm.

Sometimes the number of leaflets in some leaves is reduced to one or two, sometimes

increased to four.

Rh. glauca Desf. Herb. Thunberg.

DisTRIBUTION : On mountains and hills in south-west Cape Colony and on sand-dunes,

extending eastwards to Grahamstown, Queenstown, and Kast London,* flowering generally

in winter and fruiting about September.

Herb. Thunberg; Burchell 378; Piquetberg 400 feet, Schlechter 7895; Darling, Malmesbury

division, Bachmann 377; Cape Peninsula, E. et Z. 1120 (Rh. Thunbergiana Roem. et Schult. ex E. et Z.

Knum.), E. et Z. 1121, Wilms 3124, Diels 2, 1273, Drége 116d, EH. & Z. 34, Engler 114, Schlechter 1007 ;

Caledon, E. & Z.; Genadendal, 800 feet, Schlechter 10331, 10332 (leaflets more elongate than usual) ;

Knysna heads, Keet 1029, Schonland 3528; Belvidere, Knysna, Duthie 663; the Glebe, Knysna,

Phillips 159; Duineveld, between Slang and Kromme R., 6-8 feet, common sand-stay, J. Sim 9, Phillips

1610; coastal and dune area near Schoenmakerskop, Port Elizabeth, J. Sim 23; Humewood, Paterson

3362 ; Grahamstown, near Golf Links, Gane 15, Bowker’s Park, near Queenstown, Hilner 313. Drége 116d,

which belongs here, was marked at Kew “ Rh. lancea Dest. ex Presl Bem. 41.”

Z. 2241 (evidently = E. et Z. 1120) was named Rh. Thunbergiana R. et 8. and Rh. plicaefolia Z.,

but is not Rh. plicaefolia E. et Z. 1118.

Specimens collected by Dr. Meyer in 1869 in the Hantam mountains are marked ‘‘ forma ad excisam

spectans”’. JI doubt whether they belong here.

Specimens of Sieber, Fl. Cap. 218, in the Berlin Herbarium, marked “typus Rh. glauca (Rh. Thunbergiana

R. et Sch.) ”’, have mostly obtuse, obovate, cuneate leaflets.

* This may have to be modified as in the absence of fruits of many of the specimens examined I am

unable in some cases to say for certain whether they belong to Rh. glauca Desf. or Rh. undulata Jacq.

Those from inland localities especially are doubtful. In old collections Rh. giauca is often named Rh.

lucida L.

32. Rh. undulata Jacq. in Hort. Schoenbr. t. 346; Sonder Lec. 518 ; Hoeiene 410; Diels

lc. 573, 594, 632.

Rh. excisa Thunb., Fl. Cap. ed. Schultes 264; Drege exsice.; EH. et Z. 1125.

Rh. nervosa K. et Z. (non Poir.) in Enum. 1115.

Rh. pallens KH. et Z. im Enum. 1114.

Rh. plicaefolia E. et Z. in Enum. 1118.

Rh. spathulata EK. et Z. in Enum. 1119.

Eh, aglaeophylla KE. et Z. in Enum. 1117.

Rh. micrantha HK. et Z. in Enum. 1124 (non Thunb.).

Rh. mucronata HK. Mey. in Drége exsice. (non Thunb.).

Rh. celastrovdes Sond. |.c. 519; Engl. lc. 417; Diels l.c. 576, 628.

Rh. Burchellii Sond. in Engl. l.c. 412; Diels 574, 633, fig. 7 G.

Rh. Rangeana Engl. in Pflanzenwelt Afrikas III, II, 205, fig. 101 A—E.

Rh. vernicata Schlecht. ib. 205.

Rh. Galpin Engl. (non Schinz) ib. 208.

Description : A very variable, usually unarmed shrub, glabrous except young leaves,

which may be puberulous, often reaching a height of 15 feet in the coast districts, and then

main branches reaching a thickness of 8-9 inches. Branchlets slightly angular. Leaves

petiolate, petioles winged, particularly in the upper part. Leaflets often covered with a ~

resinous excretion, membranous or (in the arid parts of South Africa) subcoriaceous or

coriaceous, cuneate-oblong, obovate-oblong, obtuse, acute or emarginate, often plicate-

mucronulate, margin entire or rarely irregularly dentate or rarely crenate ; midrib slightly

prominent on both surfaces, lateral veins delicate but distinct, tertiary veins not visible.

Panicles lax, axillary usually shorter than the leaves, terminal somewhat longer. Calyx

segments ovate. Petals oblong. Drupe glabrous, compressed, often remaining green

when ripe.

Length of petioles 1-2-2-2 em. (in var. celastroides and Burchellii smaller).

Length of terminal leaflets 3-5-5 em. (in var celastroides and Burchellii smaller).

Breadth of terminal leaflets usually 1-1-3 cm. (in var. celastroides and Burchellii smaller).

Lateral leaflets one-half to two-thirds of terminal leaflets.

Length of calyx lobes about } mm.; petals 14-1} mm.

Greatest diameter of fruit about 3 mm.

Rh. undulata Jacq. Herb. Jacquin.

Rh. undulata Jacq. Herb. Jacquin (teeth of obovate leaflet enlarged).

. Rh. undulata Sacq., var. Burchellii Schonl. Burchell 1722.

. Rh. excisum Thunb. Herb. Thunberg (very close to Rh. undulata Jacq. var. celastroides Schonl.).

Behe

Distrisution: In the coast districts from Damaraland to Natal and in the dry

interior of the Cape Province, southern Bechuanaland, Orange Free State, and the Transvaal,

having adapted itself to a wide range of climatic conditions. It flowers mainly in the late

summer and autumn.

Rh. undulata Jacq. is represented in Herb. Jacq. at Vienna (but without flowers and

fruits). The leaves vary in size and shape. The terminal leaflet may be from 4-6} cm.

in length. The leaflets in the type specimens are obovate, oblanceolate, obtuse or

subacute. Their margins are slightly undulate or in some leaflets irregularly dentate.

Some leaves are indistinguishable from Rh. excisa Thunb., with the type of which I have

compared them (Rh. exciswn fol. B in Herb. Thunberg, Upsala). Sonder (l.c. 519) says

that it differs from the preceding by the three times smaller, not undulate or incised-

toothed leaflets. Already Engler (l.c. 411) had stated with reference to Rh. excisa: ~ Valde

affinis Rh. wndulatae et vix nisi foliis paullo crassioribus supra haud resinosis integris

Aiversa”’. Diels (l.c. 573) states that these distinctions do not hold good, and I agree.

Thus Rh. excisa Thunb. is best sunk in Rh. undulata Jacq. Both have compressed

fruits which distinguish them from Rh. glauca Thunb. ; their leaves, however, sometimes

approach the latter in shape, texture, and size. Typical Rh. undulata (incl. Rh. excisa)

occurs in the southern coast districts of South Africa. Already in these the leaflets may

retain on their surfaces a copious supply of resin. This is often more decided in the more

arid interior parts. At the same time the leaves become smaller in size. There is a

bewildering number of forms; the most decided ones are those which have been placed by

various authors under Rh. celastroides Sond. (l.c. 519) and Rh. Burchellii Sond. (in Engl.

Le. 412). Extreme forms of these are easily distinguished, but I have failed to separate

them satisfactorily when large series of specimens were examined. I, therefore, place

them as varieties under Rh. undulata. A number of other forms described without

sufficient reason as separate species will be mentioned presently.

A. genuina.—Leaflets membranous, terminal usually about 3-5 cm. long.

a forma undulata.—Leaflets not conspicuously resinous, obovate-oblong or obovate,

cuneate at the base, apex obtuse or acute rarely deeply emarginate, margin entire, undulate,

undulate-denticulate, dentate or rarely crenate.

B forma excisa.—Leaflets not conspicuously resinous, entire or emarginate at the apex,

narrowly obovate-oblong or obovate-lanceolate, often plicate-mucronulate.

y forma contracta.—Leaflets similar to B, but very resinous. Lateral branches often

_ contracted, sometimes turning into thorns (Rh. vernicata Schlecht.).

B. celastroides (Rh. celastroides Sond.).—Leaflets subcoriaceous, usually conspicuously

resinous, smaller than in var. A, generally very acute. Plants of squarrose habit, often

thorny. (In the type, Z. 233, the leaflets are mostly lanceolate, acuminate, undulate, but

some are oblanceolate, obtuse, with or without a mucro. There are no tangible floral or

fruit characters that can be used to distinguish it from var. A). Only found in arid parts.

C. Burchellii (Rh. Burchellii Sond., Rh. Rangeana Engl.).—Leaflets coriaceous,

shining, rarely slightly exceeding 1 cm. in length, often much smaller, obcordate-cuneate,

much contracted in the lower half, entire or rarely more or less crenate near the apex.

Plants sometimes thorny. Only found in arid parts.

A. genuina a undulata.

Herb. Jacquin, Vienna; Drége 5889; Piquetberg, Diels 190; Olifants River valley, west of Clan-

william, Diels 1149; ib., on sandhills, Diels 228; Van Rhynsdorp, about 1,000 feet, Diels 558; Darling,

Girke 609 ; Saron, Schlechter 7782 (in the two last, as in some others, the leaflets are obovate and. bluntly

erenate in the upper part); Calvinia, Rijsjoen mountain, Marloth 10300; Heerelogement, E. & Z. 338 ;

expedition to the Khamiesberg, Giftberg, and Olifants River, 1-2,000 feet, Phillips 7528 ; amongst shrubs

east side of Table Mountain, near Constantia, E. & Z. 39, 1124 (Rh. micrantha E. et Z. non Thunb.) ;

mountains above Worcester, Rehmann 2517 (with larger leaves than usual. probably a coppice shoot) ;

Worcester, hills opposite station, Marloth 9952 ; Robertson, base of foothills, De Hoop Road, Britten 653 ;

Phisantefontein, Kl. Karroo, Muir 2477; hills south of Matjesfontein, Pearson 2967, Rehmann 2912 ;

Prince Albert, Marloth 11277; near Laingsburg, Marloth 3972; Ceres, Karroo, Marloth 10472; slopes

of Bokkeveld, 400 m., Marloth 7796; Wupperthal, Marloth 7502; hills near Genadendal, Schlechter

10331; Groot River Hill West, Zitzikamma, Fourcade 1238; Cango Kopje (Oudtshoorn), near hotel,

frequent, Britten 1730 ; Baakens River valley, Port Elizabeth, J. L. Drége 533 ; Port Elizabeth, James 3 ;

Natal, Wood 9340. Cultivated specimens ex hort. Paris 1823, in the Berlin Herb., agree well with Jacquin’s

types as well as some of Diels’ specimens.

The following has small entire leaflets and relatively long petioles which are broadly winged (the

broadness of the wings, however, varies in Jacquin’s types). The average length of the petioles is 1-2 em.

of the terminal leaflets 1:3 cm. The leaflets are broadly oblanceolate, obtuse, rarely emarginate: Klip-

fontein, Namaqualand, c. 3,100 feet, Bolus 9504. Some specimens of Sieber, Fl. Cap. 154, agree closely

with this.

A. genuina B excisa.

Herb. Thunberg ; Nieuwe Hantam, Drége 64086 and 6809 p. pte. ; Drége 5589; Calvinia, common

on south-east slope of the Roupmyniet, 900-1,000 feet, Diels 666 ; Tulbagh, Worcester, and Clanwilliam,

E. et Z. 1125; rocky places near Ladismith, Marloth 2989; hills near Mossel Bay, Schlechter 5725 ;

Glebe, Knysna, Phillips 15 ; Knysna heads, Keet 1029, Schonland 3528 ; Belvidere, Knysna, Duthie 668 ;

Plettenberg Bay, 10 feet, Schlechter 5932 (terminal leaflets only 1-5-2 em. long); Hankey, Paterson 20 ;

karroid places between the Zwartkops and Sundays Rivers, Z. 2243, E. & Z. 1118 (Rh. plicaefolia E. et Z.,

Rh. excisa y emarginata Sond.—H. & Z. remark: “ Folia valde variant’); Uitenhage and Addo, 50-500

feet, E. & Z. 1115 (Rh. nervosa E. et Z. non Poir.—has slightly thicker leaflets than the preceding ones.

The growth is more fastigiate) ; Harneliff, Port Elizabeth, 100 feet, Galpin 6378 (this agrees well with

Rh. nervosa E. et Z.; such forms closely approach var. celastroides) ;_ amongst shrubs on the Zwartkops

River, E. & Z. 1119 (Rh. spathulata K. et Z.); Zwartkops River and Addo, E. & Z. 39, 1114 (Rh. pallens

E. et Z., Rh. excisa B pallens Sond.) ; in scrub near Uitenhage, Schlechter 2521; along the Bushmans

River, Alicedale, Schlechter 2706 ; Bothas Hill, Grahamstown, H. & Z. 39, 1117 (Rh. aglaeophylla B. et Z.) ;

very common in open bush (except in karroid places) on the Zuurberg range, Grahamstown, e.g. EH. & Z. 406,

Britten 1503, 1524, 1533, 1555, 1571, 2739, Gane 66, 325, MacOwan 504, Rogers 27711, Dyer 59, 60;

Bathurst, Nell (5157); Kowie West, Britten 5016; Bedford, Nicol 78; Katberg, about 4,000 feet,

Staples (5022a); Wolfridge, Amatolas, Hunter; Keiskama hoek, Dawson (5073); Kingwilliamstown,

Schlechter 6128; Windvogelberg, near Cathcart, 3,500 feet, Baur 1119; Queenstown, 6,000 feet, Galpin

2142 (Rh. Galpinii Engl. non Schinz, approaches Rh. glauca in the shape of the leaflets) ; Hast London,

Gane 312; Komgha, Flanagan 320, 799; Bazija, Transkei, Baur.

A. genwina y coutracta.

Hills at PUs (Western Region), Schlechter 11434; Queenstown, Rogers ; in bush near Kingwilliams-

town, 1,500 feet, Tyson 2100; near the Kabousie River, Komgha, 2,000 feet, Flanagan 701.

B. celastroides—Rh. celastroides Sond.

Namos, Bechuanaland, Z. 233; Karreebergen, Schlechter 8196; Little and Great Namaqualand,

Pearson 5728, 3946, 3418, 3341, 3964, 4960, 3761, 2978, 4220.

C. Burchelluu—Rh. Burchellii Sond., Rh. Rangeana Engl.

At the confluence of the Vaal and Orange Rivers, Burchell 1722; Basutoland, Cooper 2172; Leribe,

5-6,000 feet, Dieterlen 581 ; Bloemfontein, Rehmann 3800, 3887 ; Elandshoek, near Aliwal North, Bolus

10487 ; near Burghersdorp, Flanagan 1532; Conway farm, Galpin 5516; Honey Nest Kloof, Wilman

1522; mountain ravines near Murraysburg, Tyson 296 ; on rocks amongst hills, Leeuwfontein, 2,600 feet,

Pearson 3228; common in Namaqualand and the South-West Protectorate, e.g. Pearson 3801, 5728,

$203, 3174, Dinter 1134, 1134a, Schafer 168, 236, Engler 6699, 6763, Blank 70, Range 125, 227, 1765,

Marloth 4655.

Blank notes that the natives eat the fruit which may indicate that it is more juicy

than in var. A.

While the specimens enumerated above can, on the whole, fairly easily be sorted out

as belonging to different varieties and forms, | have seen a large number of others which

I cannot readily assign to these varieties, yet I do not think it advisable at present to

place them under special named varieties or forms. The difficulties encountered may be

illustrated by Bolus 88, collected in a kloof on a mountain side near Graaf-Reinet, alt.

3,500 feet, with flowers and young drupes.

This is quoted by Engler (l.c. 411) under RA. excisa Thunb. B pallens Sond. It is

sufficiently distinct from Rh. pallens EK. et Z. 1114 to prevent us from placing it with it.

Its leaflets vary in length from 7 mm. to 2-8 em.; they are either slightly or decidedly

“varnished”; they are usually obovate-cuneate or broadly oblong-cuneate, rarely

obovate. | The cuneate lower portion narrows, either gradually or somewhat suddenly.

The apex is entire, obtuse or acuminate or mucronate, rarely emarginate or

crenate. The petiole is either slightly margined or distinctly winged. Other

forms placed under var. pallens Sond. in Herb. Kew and Berlin show so much diversity

that again one cannot find any tangible characters to keep them separate, e.g. Burchell

2850, 2871, 2991, 4214, 4726, Drége (named Rh. mucronata Thunb.), MacOwan 504 from

Grahamstown, Marloth 10347 from Beaufort West, Bolus 5290 from Mitchell’s Pass.

The following belonging to Rh. undulata cannot, without re-examination, be placed

by me under any of the varieties here constituted: Waterval Onder, Lydenburg district,

Burtt-Davy 396.

There can, further, be no doubt that this species sometimes hybridizes with others.

Perhaps Paterson 3362, collected at Humewood, near Port Elizabeth, is such a hybrid.

Tt has puberulous branchlets.

ECKLONIANA group.

33. Bh. eckloniana Sond. l.c. 515; Rh. tridactyla KE. et Z. in Enum. 1095 (non Thunb.) ;

Rh. angustifolia? a Thunb. herb.; Rh. margaretae Burtt-Davy Ms.

Rh. Schoenlandii Engl. in Pflanzenwelt Afrikas III, 2, 215 (Daly and Gane 747).

Description : A usually glabrous dwarf shrub, rarely over 2 feet in height, with purple

angular branchlets. Leaves petiolate. Petioles subterete, furrowed above. Leaflets

sessile, often more or less folded along the midrib, oblong or lanceolate-oblong, often

mucronate, cuneate at the base, margin entire, slightly thickened and often white

when dry. Midrib sunk above, very prominent below, lateral veins subparallel,

slightly branched towards the margin, very slightly prominent below, tertiary veins

inconspicuous. Panicles lateral in the upper portions of the branches and terminal equal

in length to the leaves or slightly shorter or longer, rather lax. Bracts small subulate.

Calyx segments ovate, bluntish. Petals oval. Drupe shiny, glabrous, subglobose.

Rh. eckloniana Sond. Dyer 4601. Under side. (Leaflets often folded along midrib.)

Petioles 1-2 cm. long.

Terminal leaflets 1-5-3-8 em. long, 5-10 mm. broad.

Lateral leafflets 4 to + shorter.

Pedicels about 1-2 mm. ; calyx-segments about ?mm. ; petals 14-2 lone

Drupe 5-5-6 mm. in diameter.

DistR1IBuTion : From the Elands River, in the Uitenhage division, to Bothas Hill and

Blaauwkrantz, near Grahamstown, also Roundhill, chiefly on dry ridges, flowering chiefly

in the late summer and fruiting in winter, Orange Free State and the Transvaal.

Ecklon 310, Z. 2230, Cooper 2173, 1538, Bolus 10638, Daly and Gane 747, Rogers 3108, Galpin 2906.

Britten 2767, 5160, Baur 1092 are typical Rh. eckloniana from south-eastern Cape Province. HE. et Z.

1095 are cultivated specimens, wrongly distributed as Rh. tridactyla Burch. Drége’s specimens were

distributed as Rh. laewgata L. fol. long. integra K. Mey. b.—There is a slight resemblance to the genuine

Rh. laevigata L. (non Thunb.) in shape of leaflets, inflorescence, etc., but the venation of the leaflets is

different.

Many of the Transvaal specimens are more or less pubescent, or even in young portions subvillous

(Rh. margaretae Burtt-Davy), e.g. Schlechter 3539, Moss 2957, Gilfillan in Herb. Galpin 1495, but the

following are quite glabrous: Galpin 1495, Burtt-Davy 15138 (=15060), Schlechter 3539a, Keet 5166 in

Herb. Alb. Mus. These last range from Parys (O.F.S.), Vereeniging to Barberton. and cannot be

distinguished from typical Rh. eckloniana.

In various specimens it was found that the number of floral parts was increased to

SIX or seven.

34. Rh. Engleri Britt. in Journ. of Bot. XX XVIII, 1900, 316.

Rh. incana Engl. |.c. 420 (non Mill.) ; Diels lc. 581, 630.

Description: A much branched shrub or small tree, about 12 feet high, with

spreading sometimes spinescent branches. Branchlets short, terete, the young ones densely

greyish pilose. Leaves petiolate, petioles, subsemiterete, furrowed above, sometimes

slightly winged, greyish pilose. Leaflets membranous, sessile, oblanceolate, obtuse or

emarginate, margin often slightly revolute, entire or slightly crenato-dentate, upper surface

dull green with puberulous midrib, and sometimes the lateral veins slightly prominent and

greyish pilose, lower surface oreyish pilose, the midrib and the few lateral veins somewhat

prominent. Panicles lateral and terminal, pilose, laxly flowered, equal to or twice the

length of the leaves. Bracts minute. Calyx lobes oblong, triangular. Petals yellowish,

oblong. Drupe somewhat asymmetrical, laterally compressed and often umbonate on

one side of the apex, brown, glabrous, shining.

Length of petioles 8-10 mm.

Length of terminal leaflets 2-5-3-5 cm. Breadth of terminal leaflets 6-7 mm.

Lateral leaflets usually about two-thirds the size of the terminal, sometimes shorter.

Length of pedicels 1-2 mm.; calyx lobes about # mm. ; petals 14 mm.

Greatest diameter of drupes about 3 mm., height about 4 mm.

Rh. Engleri Britt. Galpin 2919. Under side.

DistRiBUTION : Transvaal Bushveld (Rehmann), Naboomspruit, Wakkerstroom,

Waterberg district, Potgietersrust, Rooikop.

Rehmann sine no; Rogers 19004; Galpin 8472, M 64; Leendertz 1147, 1500, Pole Evans 2930,

Baily 933.

The fruit is similar to that of RA. Marloth Engl., but its affinities generally seem to

be with the species with which it is associated here. Hngler places it next to Rh. Welwitschii,

an Angolan species, and close to Rh. divaricata H. & Z. The latter has longer petioles,

longer petals, different fruit, etc.

Mr. HK. E. Galpin gives the following notes on some of his specimens :—

Locality : Farm Roodepoort, Springbok Flats, 8 miles south-east Naboomspruit Station,

Altitude 3,750 feet.

Nature of Ground, etc—Typical Springbok Flats red heavy loam derived from

amygdaloidal basalt. On level plaims in Sweet Acacia veld. In small bush clumps

associated with Acacia Karroo, Acacia litakunensis, Acacia robusta, Euclea lanceolata, and

Ehretia hottentotica, .te. Frequent.

Average height 12 feet. Stem 4-6 inches diameter.

Wood red brown and fairly tough. Used occasionally for pick-handles, but not nearly

so good as that of Rhus cana [—this should be pyroides.—S. Sch.].

Local name Karaa, as distinguished from Karee the name for Rhus lancea.

Sesuto name Mphapashane, which is also applied to Rhus Gueinzu. Very frequently

attacked by hairy caterpillars, by which many trees are completely defoliated every year.

35. Rhus eburnea Schonl. n. sp.

Description: Frutex fastigiatus 14-13 m. altus ramulis teretibus adscendentibus

pubescentibus, foliis petiolatis, petiolis pubescentibus supra canaliculatis foliolis terminalibus

brevioribus; foliolis supra saturate viridibus infra pallidioribus utrinque sparse pilosis lanceo-

latis vel oblongis basi cuneatis apice acutis vel acuminatis margine integris costa supra

canaliculato infra valde prominenti nervis parce ramosis distinctis sed immersis ;_ paniculis

laxis axillaribus et terminalibus quam folia longioribus rarius brevioribus, bracteis floriferis

minutis, floribus pedicellatis, calycis segmentis acutis petalis oblongis subtriplo brevioribus,

drupa subglobosa eburnea exocarpo coriaceo.

Petioles about 1 cm. long.

Leaflets: terminal ones about 2+5 cm. long; lateral one-third to one-half shorter, 6-8 mm. broad.

Panicles 3-4-5 cm. long. Floral bracts about ? mm. long. Pedicels usually 2 mm. long.

Calyx segments about + mm. long. Petals about 14 mm. long.

Drupe about 3 mm. in diameter; about 34 mm. high.

The yellowish white drupes usually show a few reddish thin irregular lines and bear-

at the apex a small black remnant of the styles.

Rh. eburnea Schonl. Keet 1547. Upper side.

DistripuTioN;:, Eastern Transvaal at an altitude of 2,500 to 4,000 feet.

Keet 1547—Marieskop forest reserve, Pilgrims Rest district (formerly Lydenburg), Dec. (fl., fr.).

Mr. Keet writes: “The bush forms one of the chief constituents of the “fynbos’ on

the edge of the forests and generally along foothills and slopes from 2,500 to 4,000 feet,

being mixed usually with a species of Huclea, a sp. of Royena, and with Cliffortia linearifolia.

It is found on all aspects and on soil from both granite and sandstone formations. The

reserve is situated about 30° 40’ Hast and 24° 30’ South.”

The vegetative organs resembles those of Rh. eckloniana (Rhus margaretae Burtt-Davy),

but the fruit is different, the panicles longer in Rh. ebwrnea, etc.

NATALENIS group.

36. Rh. natalensis Bernh. in Krauss Beitr. (1844) 46; Sonder l.c. 515; Engler lc. 421;

Diels l.c. 587, 631; Rh. glaucescens A. Rich. Tent. Fl. Abyss. I, 143; Rh. crenulata

A. Rich. Tent. Fl. Abyss. I (1847) 142; Cissus natalensis Bernh. in sched.

Description: A glabrous shrub 8-10 feet high with greyish subterete branchlets.

Leaves petiolate, petioles broadly canaliculate above. Leaflets subcoriaceous, dark green

above, light green below, broadly oblong, cuneate at the base, obtuse or emarginate at

the apex; margin crenate-dentate, or subentire, rarely entire, slightly revolute; midrib

prominent on both surfaces or sunk on the lower, lateral veins slightly prominent or

immersed, tertiary veins immersed, reticulate, barely visible as a rule. Panicles axillary,

often smaller than the leaves, lax. Flowers greenish yellow, calyx segments ovate, petals

oblong. Drupe glabrous, yellowish brown, a little compressed and depressed.

Length of petioles 1-2 cm.

Length of terminal leaflets 4-5-5 cm. Breadth of terminal leaflets 1-8-2-3 em.

Lateral leaflets nearly three-quarters the size of the terminal.

Length of calyx segments } mm.; petals 1-5 mm.

Greatest diameter of drupe 5-6 mm.

Rh. natalensis Bernh. Schonland 4025. Under side.

DistrrBuTion : From about 11 miles west ot East London along the coast to the

borders of the Tropics, and widely spread in Tropical Africa to Abyssinia, and varying

there to a considerable extent, but a number of specimens placed in European herbaria

under Rh. glaucescens Rich. should be excluded; in many others the material is quite

insufficient to allow of exact determination.

Diels has associated this species with Rh. crenata Thunb. in his Crenata group (l.c. 630)

characterized (l.c. 597) by “ crenate leaflets with glands predominating”. While there is

no decisive evidence against this procedure, I do not think it advisable to follow it. He

placed the closely allied Rh. Marlothii into his Damarensis group (l.c. 627), the characteriza-

tion of which (l.c. 597), to my mind, is the same as that of the Crenata group (“ Crenate

leaflets. Indument strongly inclined towards reduction ”’).

Port Natal, Krauss 395 in Herb. Kew ; Overton in forest 11 miles west of East London, Hilner 281 ;

Port St. Johns, in bush near beach, Schonland 4025 ; Embotye, Pondoland, near sea-level, Fraser 73/15/A

(5101), 37/38/A (5154), (5088); Durban, from near sea-level to 100 feet, Wood 8729, 1713, 1136,

Gueinzius. Rudatis 1481; Moss and Rogers 1708 (labelled “above Kirstenbosch, Cape ”. which is

evidently a mistake).

37. Rh. Simii Schon]. n. sp.—Rh. glaucescens Sim (non Rich.) in ~ Forests and Forest

Flora of South Africa” (1907) 195, t. 49, fig. 1.

Description : Frutex divaricatus inermis vel spinosus vel arbor parvus ramis griseis,

ramulis pubescentibus ; foliis petiolatis petiolis gracilibus subteretibus supra canaliculatis

foliolis brevioribus ; foliolis rigide membranaceis oblongis vel lanceolatis basi angustatis

obtusis rarius emarginatis margine leviter incrassatis integris vel--crenato-serratis

supra glaucis subtus pallide rufescentibus costa utrinque prominenti nervis utrinque leviter

prominentibus marginem versus integris vel furcatis et connatis, venis paucis vel nonnullis ;

paniculis pubescentibus foliis brevioribus ex axillis foliorum superiorum et terminalibus

subdensifloris ; floribus pedicellatis calycis segmentis ovatis subacutis extus parce puberulis

petalis oblongis calyci subtriplo longioribus, disco in floribus masculis 5-crenato; drupa

subglobosa nitida.

Length of petioles 2-2-5 cm.

Length of terminal leaflets usually 5-6 em. Breadth of terminal leaflets about 1-3 em.

Lateral leaflets not quite two-thirds the length of the terminal ones, but often broader than these.

Length of calyx segments $} mm.; petals 14-14 mm.

Diameter of drupe about 3-5 mm.

Rh. Simii Schonl. Sim 2127. Under side.

DisTRIBUTION : Woods at Toise River and Komgha, flowering in autumn, and Komati

Poort (var. lydenburgensis Schonl.).

Sim remarks that this species varies considerably in leaf-form and sometimes resembles

-wide-leaved forms of Rh. lancea. Its wood is sometimes used for hoe-handles as it is light

and tough.

Sim includes in this Rh. Gueinzii Sond., which is not justified. Its affinities are with

Rh. natalensis Bernh., which differs from it in many vegetative characters and has larger

fruits. It may perhaps have to be sunk in Rh. spinescens Diels.

Toise River, 3,500 feet, Sim 2127; Komgha, Flanagan 797.

var. lydenburgensis.—This has the same fruit as the type, but differs from it in the

following characters: Leaflets green and more thickened at the margin. The thicker

branches armed with short thorns. Common in the low veld at Komati Poort, 600-700 feet ;

‘also grown in hedges, Keet 1431 (5158).

I append a full DESCRIPTION as it may have to be separated as a distinct species or

united with Rh. spinescens Diels :—

Frutex divaricatus spinosus, ramis adultis spinis crassis ad 6-2 cm. longis armatis,

ramulis brevissime pubescentibus subangulatis; folius petiolatis, petiolis gracilibus sub-

_teretibus supra leviter canaliculatis foliolis brevioribus, foliolis glaberrimis rigide mem-

branaceis oblongis basi angustatis obtusis rarius emarginatis margine incrassatis breviter

crenato-serratis supra saturate viridibus subtus pallidioribus costa utrinque prominenti

nervis supra immersis vel leviter prominentibus subtus leviter prominentibus+-ramosis ;

-paniculis laxifloris lateralibus et terminalibus foliis brevioribus ; floribus ignotis; drupa

“subglobosa nitida.

Length of petioles c. 2-5 cm.

Length of terminal leaflets up to 5°8 em., rarely much shorter.

Breadth of terminal leaflets about 1:3 em.

Length of lateral leaflets 3-3-5 em.

Diameter of drupe 2—2-25 mm.

38. Rh. spinescens Diels in Engl. Bot. Jahrb. XL (1907) 87.

Description: A shrub about 8 feet high, with pale branches, the lateral often thorny,

branchlets greyish pubescent. Leaves petioled, petioles slender, semiterete, canaliculate

above, very sparingly pilose. Leaflets papery, glabrous or on the veins very rarely a little

hairy, above deep green, below pale glaucous, oblong or oblanceolate, margin very narrowly

revolute, subundulate, midrib slightly prominent on both surfaces, secondary veins distinct,

but not raised, tertiary veins not visible. _Panicles densely pubescent with whitish hairs,

shorter than the leaves. Calyx segments broadly ovate, truncate, glabrous, petals ovate,

obtuse. Drupe (in Keet 1493) subglobose, glabrous, shining, brown, shghtly depressed,

sometimes retaining the styles.

Rh. spinescens Diels. Schlechter 11791. Under side.

Length of petioles 1-1-5 em.

Length of terminal leaflets 3-3-5 cm.’ Breadth of terminal leaflets 8-10 mm.

Lateral leaflets nearly two-thirds the length of terminal.

‘Length of calyx segments } mm.; petals 1-2-1-5 mm.

Diameter of drupe 4 mm.

: 71

DistRiBuTION: Komati Poort, common in the low veld.

This species is closely allied to Rh. Simi Schonl.

Komati Poort, Schlechter 11791 ; ib., common in low veld, Keet 1493.

39. Rh. Marlothii Engl. in Bot. Jahrb. X, 37, Pflanzenwelt Afrikas III, 2, fig. 182; Diels

he, 581, 6277. be

Description: An unarmed shrub with minutely puberulous or shortly pilose

branchlets. Leaves petioled, petioles sulcate above, puberulous or pilose. Leaflets sub-

coriaceous, puberulous or pilose, obovate-oblong or oblong, cuneate at the base ;. margin

in the upper part slightly crenate ; midrib prominent on the lower surface, lateral veins

delicate, more or less immersed, tertiary veins not visible. Panicles at the end of the

branchlets, axillary shorter than the leaves, terminal usually a little longer, puberulous,

lax. Calyx segments puberulous, ovate, petals oblong. Drupe glabrous, shining, distinctly

compressed, very often somewhat asymmetrical and sometimes crowned by the hardened

subecuspidate styles.

Length of petioles 5-12 mm.

Length of terminal leaflets 2-14-5 cm. Breadth of terminal leaflets 1-1-8 cm.

Lateral leaflets about two-thirds the size of the terminal.

Length of calyx segments } mm.; petals 1} mm.

Greatest diameter of drupe about 4 mm.

Rh. Marlothii Engl. Marloth 1394. Under side.

DistrRIBuTION: South-West Protectorate, Bechuanaland Protectorate, Transvaal. It

flowers in summer.

This species is clearly allied to Rh. natalensis and not to Rh. crenata Thunb. as Engler

thought. In its fruit it may approach Rh. populifolia. On the other hand, it also

approaches Rh. Gerrardi, and thus forms a connection with the Lancea group. It is a

somewhat variable species. Engler has indicated a var. robusta (Dinter, 1711 = Rh,

tsemubensis Dinter in Herb. S.A. Mus.) and a var. subintegra (Dinter 691, ete.), but I find

that the length of the leaflets, their more or less pronounced crenation and the amount

of their hairiness cannot be used to establish even fairly well-defined varieties. In Dinter 69,

it was noticed that the rudimentary stamens in the female flowers remain even when the

fruit is ripe.

Otyimbingue, Hereroland, c. 2,700 feet, Marloth 1394; river scrub, Bullspoort, Dinter 2712 (leaflets

smaller than usual); Wilhelmsburg, near Okohandja, Dinter 69; other specimens from South-West

Protectorate: Engler 6469, Hartman 217, Waibel 50, 50b, Dinter 347, 1711, Fritsch 24, 130, and the

following which would fall under Engler’s var. subintegra: Dinter 691, Engler 6219, 6302, Seiner 835,

Fritsch 63; Mochudi, Rogers 6628, Pietpotgietersrust, Bolus 1021 and Burtt-Davy 2205 (Rh. Englerz

var. fulvescens Burtt-Davy).

40. Rh. commiphoroides Engl. et Gilg in Warb. Kunene-Sambesi Exped. 289.

Rh. kwebensis N.E. Br. in Kew Bull. 1909, 100.

Description : A much branched shrub up to 7 feet high with greyish, densely pilose

branchlets. Leaves petiolate, petioles densely pilose, subterete, slightly channelled above:

Leaflets subcoriaceous, glaucous, paler on the under side, pubescent on both sides, ovate

or obovate or elliptic-oblong, cuneate at the base (the terminal often much contracted),

obtuse at the apex; margin flat, coarsely crenate or bicrenate, except in lower portion

of the leaflets, which is entire; midrib and lateral vems generally depressed on upper

surface, prominent on the lower, tertiary veins finely reticulate, inconspicuous. Panicles

lateral and terminal, pubescent, multi-branched with clusters of pedicelled flowers. Calyx

segments ovate, obtuse, petals oblong. Drupe (in Dinter 2874) glabrous, subglobose,

shghtly depressed and compressed, whitish when ripe.

Length of petioles 1-1-3 cm.

Length of terminal leaflets about 5 cm. Breadth of terminal leaflets about 3 cm.

Lateral leaflets about half the size of the terminal leaflets and generally barely cuneate at the base.

Length of calyx segments about 4 mm.; petals 14-14 mm.

Greatest diameter of drupe 4-5 mm.

Rh. commiphoroides Engl. et Gilg. Lugard 200. Under side.

DisTRiBUTION : South-West Protectorate, Bechuanaland Protectorate, extending into

Rhodesia.

N. E. Brown considered that it is allied to Rh. Rehmanniana Engl. I think it should

be placed near Rh. natalensis Bernh. and is perhaps also allied to Rh. Dinter, Engl.

Kwebe hills 3,300 feet, Mrs. Lugard 200; Otjitjika, Dinter 2874, 2877; Tsumeb, Dinter 1684

(labelled Rh. omahekae N.K. Br.); Waterberg, Pole Evans 19317; Mochudi, Rogers 6318.

T have not seen Rh. amboensis Schinz in Bull. de Vherb. Boiss., Sér IT, VIII, 639. (Ojavo in Ondongo,

Ovampoland, Rautenau, bl. 31,1.) According to the author it reminds one strongly of Rh. commiphoroides,

but is clearly distinguished by smaller, conspicuously thicker leaflets (the terminal ones being 25-30 mm.

long) more prominent nervature and smaller crenations. Schinz also noted that the sepals are slightly

acute.

LANCEA group.

41. Rh. lancea L.f. suppl. 184; Linn. Syst. Veg. XIV, 294; Thunberg, Prodr. 52, Fl. Cap.

ed. Schultes 263; Sonder l.c. 544; Engler lc. 444; Diels lc. 589, 640; Sim,

Forest Flora 194, t. 46; Rh. viminalis Jacq. Hort. Schoenbr. t. 344 (non Vahl) ;

Marloth, Das Kapland, 241, figs. 97, 101.

Rh. denudata Licht. in Roem. et Schult. Syst. VI, 661.

Descrietion : Shrub or small tree (in South Africa seldom exceeding 20 feet in height)

with the habit of a willow, quite glabrous, with elongated, angulate often reddish branchlets.

Leaves petiolate, petioles long and slender, narrowly margined, concave above. Leaflets

rigidly membranous or subcoriaceous, sessile, dark green and shining above, pale green

below, linear-lanceolate narrowed at apex and base, shortly mucronate, acute or rarely

obtuse, sometimes slightly falcate, entire; midrib conspicuous, raised on both surfaces,

lateral veins branched towards the margin, slightly raised (often on the upper surface only),

Panicles lax, much

tertiary veins reticulate (usually plain on the upper surface only).

branched, axillary (shorter than the leaves) and terminal (about as long as the leaves),

bracts small, linear subulate, pedicels delicate. Calyx segments unequal (two short and two

long ones and usually one intermediate between them), petals oblong. Drupe subglobose,

slightly depressed, dull greyish to shining brown, often slightly asvmmetrical.

aa so =

PEER

Rh. lancea Lf. Burchell 2728. Upper side.

Petioles 3-4 cm. long.

Terminal leaflets 9-12 cm. long; 6 mm. to 1-5 cm. broad. :

Lateral leaflets usually nearly as long as the terminal leaflets and about as broad or slightly shorter

and narrower.

Pedicels 2-3 mm. long. Calyx segments about } mm. long.

Petals almost 14 mm. long. Drupe 4-5 mm. in diameter.

Distrisution : Widely spread in South-West Africa and the interior regions of South

Africa’ wherever there is groundwater available and penetrating coastwards to the

Oudtshoorn division, the arid parts of the Albany division, Fort Beaufort and King-

williamstown divisions. (Recorded by Sim from Natal, but no specimen seen by me.)

Occurs also in Rhodesia (reaching a height of 40-60 feet near Salisbury). Flowers generally

from February to July.

Herb. Thunberg ; Jacquin (Akh. viminalis Jacq. non Vahl): Drége in Herb. Kew and 3450 in Herb.

Berol. (Rh. viminalis Vahl. a); on the Bushmans and Kat R., H. et Z. 1091 (Rh. fragrans Licht. ex Presl

Bot. Bem. 41 ; Lichtenstein 194 in Herb. Berol. (Rh. denudata Licht.) ; on the Bushmans and Kat R.,

E..et Z. 1089 (Rh. viminalis Jacq. non Vahl); Burchell 2945, 3266; Great Namaqualand, Amboland,

and Hereroland: Schultze 466, Range 411, 966; Schafer 329, Dinter 288, 932, 988, Pfeil 87, 174,

Francois 39a, Engler 6754, Trotha 16a, Seiner 174, Fritsch 64, Engler 6284, Pearson 3121, 4726, 9031 ;

Calvinia, Diels 664 and 664a ; Hantam mountains, Meyer ; (Rondebosch, Hutton—evidently cultivated) ;

hills near Matjesfontein, Schlechter 10923; in thickets along river beds near Schuurkraal, Karroo,

Pearson 5000, 5030; Maggisfonteim, Rehmann 2911; Pearson 1601, 4807 ; Oudtshoorn, Schoemanspoort,

Britten 1651; Alicedale. Cruden 306; Piggot Bridge, Dyer 915; Table farm, near Grahamstown,

MacOwan 604, 1395; Herbert division and Griqualand West, Marloth 813, Anderson 691, Wilman 1 ;

Mafeking, Shantz 232; Orange Free State: Rehmann 2880, Smith 4365, Burtt-Davy 10730; Transvaal

(widely spread, especially common near Pretoria): Leendertz 188, Burtt-Davy 534, Galpin 6991, Zahn

1684, Engler 2819, Shantz 248, Nation 297, Meune 2984, Burtt-Davy 2211, Galpin 5172, 5173, 8472, ete.

Specimens from Rhodesia which I have examined: Penther 96; Matopos, Galpin 7066; Salisbury,

Eyles 1759; Bulawayo, Rogers 51. In the last two the terminal leaflets reach a length of over 16 cm.

Burchell No. 2728 in Herb. Berol. has leaflets only 2-3 mm. broad, and was distinguished by

Engler as “forma angustissima ”’.

Specimens from a plant collected by Keet (No. 1434) in fruit at Magnet Heights, Lydenburg district

(29° 55’ E., 24° 50’ S.), alt. 4,500 feet, though close to Rh. lancea may be distinct. It forms a small tree

about 6 feet high in bushveld on “ Norita ”’ formation associated with Acacia spirocarpoides, Olea verrucosa,

etc. The texture and venation of the leaflets is the same as in Rh. lancea. They are relatively broader

(terminal leaflets about 6 em. long, 14 em. broad), often blunt, sometimes wavy on the margin, sometimes

sparsely and irregularly toothed, the teeth being small and blunt as in Rh. Gueinzii. The fruit is dull

grey, shining, slightly depressed and compressed (?). 6 mm. in greatest diameter. Perhaps a hybrid ?.

42. Rh. viminalis Vahl, Symb. III, 50 (non Jacq. nec H. et Z.); Thunb. Fl. Cap. ed. Schultes

263; Sonder le. 515; Engler l.c. 442; Diels lc. 588, 640.

Rh. laevigata Herb. Jacquin (non L. nec Thunb.) ; Rh. denudata K. et Z. in Enum.

1090 (non Licht.) ; Rh. elongata K. et Z. in Enum. 1097 (non Jacq.) ; Rh.

pendulina Jacq., Willd. Enum. 324; Rh. Waldingit Dehnh. Revist. nap. I,

3, 172; Rh. fragrans Licht. in Roem. et Schult. Syst. VI, 661; Rh. pallida

EK. Mey. in Drége exsicce.

Description : Willowlike bush or small tree (up to 30 feet in height) with elongated,

glabrous, terete or subterete, often reddish branchlets. Leaves petiolate, petioles slender,

subsemiterete, furrowed above. Leaflets sessile or subpetiolulate, membranous, subcon-

colorous (only slightly paler below than above), glabrous or finely ciliate, oblong-

lanceolate, cuneate at the base, at the apex acuminate or acute, minutely mucronate, with

entire, sometimes slightly undulate margin ; midrib slightly raised on both surfaces, lateral

veins distinct but barely raised, tertiary veins indistinct, reticulate. Panicles hirsute,

multibranched with delicate ramifications, but ultimate branches somewhat densely

flowered, axillary shorter than the leaves, terminal longer. Calyx segments ovate, acute,

often hirsute on the outside. Petals oblong. Drupe subglobose.

Petioles 1-5-4 em. long,

Length of terminal leaflets 4-8 em. Breadth of terminal leaflets -9-1-7 cm.

Lateral] leaflets usually slightly exceeding three-quarters the size of the terminal leaflets.

Calyx segments } mm. long. Petals 1} mm. long. Drupe 4 mm. in diameter.

DistRipution: On the banks of rivers and vleys in Clanwilliam, Namaqualand,

Bushmanland, on the banks of the Orange River to Griqualand West, also at Graafi-Reinet,

flowering from September to January.

Sonder (l.c. 515) has distinguished a var. pendulina (= Rh. pendulina Jacq., Willd.

Enum. 324; Rh. pallida E. Mey., in coll. Drége). The characters by which he tries to

distinguish it are, however, of not much use. I have seen the original specimens of Jacquin

in Herb. Vienna. The branches have rather lanky growth and have evidently been

pendulous, but in herbarium specimens generally it is not advisable to try to distinguish

this variety.

Rh. viminalis Vahl. Anderson 691. Under side.

Specimens cultivated at Capetown, E. & Z. 1090; near Twenty-four Rivers (Clanwilliam), E. & Z.

1097; Namaqualand, near Klipfontein, Herb. Norm. Afr. 450 (Bolus 6526); Namaqualand, Pearson

6135, 6098 ; Bushmanland, Pearson 2582 ; near Doornpoort waterhole, Orange River expedition, 1910-11,

Pearson 3136; Ramansdrift, Pearson 3112; Sendlings Drift, Pearson 1550, 6098 ; Verleptpraam, Orange

River, Drége 3035; banks of Orange River, Pearson 3102 (some leaves extraordinarily broad, up to

2-3 cm., approaching an ovate form); Mazelsfontein, Griqualand West, Wilman 691; river banks near

Graaff-Reinet, Bolus 134 (this was named Rh. villosa L.f. var. glabrata by Sonder).—(Schimper 163 from

Amba Sea, Abyssinia, which was distributed as Rh. viminalis Vahl is Rh. retinorrhoea).

716

43. Rh. Gerrardi Harv. Ms.; Diels l.c. 588, 628.

Rh. viminalis var. Gerrard: Engl. lc. 422.

Description: A shrub or small tree with at first pilose, later glabrous, subterete

branchlets. Leaves petiolate, petioles slender, subterete with a shallow furrow above, at

first pilose, later glabrous. Leaflets membranous, sessile, narrowly oblong or oblanceolate

or oblong-ovate, narrowed at the base and generally also at the apex, which usually ends

in an acute point; surfaces subconcolorous, at first pilose, later more or less glabrous ;

margin irregularly (and usually sparsely) crenato-dentate, especially towards the apex,

teeth very shortly mucronate ; midrib prominent, especially on the lower surface, lateral

and tertiary veins delicate, but slightly prominent and very distinct, especially on the

upper surface, the latter reticulate. Racemes axillary and terminal, shorter than the

leaves or the terminal slightly longer, narrow, fairly compact (not diffuse as in Rh. viminalis)

subglabrous in the type. Flowers pedicellate forming small glomerules, greenish yellow.

Calyx lobes ovate. Petals oblong. Drupe ?

Length of petioles 1-4—2-25 cm.

Length of terminal leaflets 4-6 em. Breadth of terminal leaflets about 8 mm.

Lateral leaflets generally about four-fifths the size of the terminal leaflets.

Length of pedicels 1-1} mm. ; calyx lobes }-] mm. ; petals 1{-13 mm.

Diameter of drupe ?

Rh. Gerrardi Harv., A. typica. Schlechter 3771. Upper side.

DisTRiBvuTION : Basutoland, Tembuland, Natal (above 3,000 feet), eastern Transvaal,

flowering in midsummer.

Uses: Used in Basutoland for building huts and for hedges at Lydenburg.

Four varieties may be distinguished.

A. typica.—

Natal, Gerrard 1396; Leribe, Dieterlen 691; near Emmans, Natal, Wood 3645; bank of Tugela,

4,000 feet, Wood 3632; Little Tugela, near Glockners, 4,000 feet, Wood 3632; Estcourt, 3,700 feet,

Dimmock-Brown 91 ; near Colenso, Wood 6581 ; on the banks of the Olifants River, 5,000 feet, Schlechter

3771; Lydenburg, Schlechter 3963.

B. latifolia.—Leaflets larger and broader than in the type (up to 3 cm.). Inflorescence

subvillous and young branches, petioles, and leaflets more hairy than the type.

Rh. Gerrardi Harv., B. latifolia Schonl. Evans 5096a. Upper side.

Basutoland, Dieterlen 1201; Zomershoek, Lydenburg district, Burtt-Davy 7564 (= Wilms 246, 247,

248 = Rh. viminalis y Gerrardi Engl. forma pilosa Engl. in Herb. Berol.); Ishlet River, Piet Retief

district, Forest Dept. Herb. 5974; Graskop, Lydenburg district, along river banks, Evans (5096qa) ;

Palmer’s farm, Waterval, Lydenburg district, Burtt-Davy 5327.

The following does not agree exactly with any of the foregoing as the leaflets are often dentate to

near the base: Kaap River valley, near Barberton, Keet 1436. It is a small tree, 8-10 feet high. The

drupes are red laterally compressed, generally broader than high. Their greatest diameter is 24 mm.,

their height usually 2 mm.

C. basutorum.—tLeaflets narrowly oblong or oblanceolate, acute at the apex, often

mucronate, margin above the middle remotely and grossly dentate, teeth mucronulate,

above subglabrous with the exception of the densely pilose midrib, below subtomentose,

the prominent midrib and lateral veins densely pilose.

Of this I have only seen a single branch from Khanyane, Leribe district, Basutoland (Dieterlen 691.

Herb. S.A. Museum 6065). It grows on mountain slopes and in ravines. It was named Rh. viminalis

Vahl var. Gerrardi by Phillips in his “ Flora of Basutoland.”

The stem was simple, densely leafy, subterete, densely pilose. It bore female flowers

on a terminal, lax panicle, barely equal in length to the leaves. I thought at first that it

should be placed near Rh. discolor KE. Mey., but as it is used for building huts it must be a

large shrub or even a tree. In any case the resemblance to this species seems only to be

superficial and is only due to the fact that an isolated branch was compared to it.

D. montana.—Rh. montana Diels in Engl. Bot. Jahrb. XL, 86, 639.—Branchlets and

petioles glabrous. Leaflets 3-5 (rarely 6 or 7) foliolate, nearly glabrous or towards the

margin slightly pilose, oblanceolate or obovate oblong, grossly crenato-serrate, teeth

mucronulate. Panicle slender, longer than the leaves, pilose (as are also the ovate-

lanceolate calyx segments).

Rh. Gerrardi Harv., D. montana (with 7 leaflets). Bolus 8837. Under side.

This has only been collected at Engeobo, Tembuland, by H. Bolus (No. 8837).

I append a translation of Diels’ account, adding here and there a few additions and

corrections in square brackets :—

“Shrub with brown glabrous branches. Petioles semiterete, slender, at the base

enlarged and sheathing, glabrous. Leaves 3-5 foliolate. Leaflets papery, glabrescent or

towards the paler margin a little hairy, below pale, towards the base gradually narrowed,

otherwise oblanceolate or obovate-oblong, grossly [crenato-] serrate, teeth mucronulate,

midrib prominent below, lateral veins a little prominent below, tertiary veins immersed.

Inflorescence slender, exceeding the leaf; calyx segments triangular [ovate-lanceolate],

slightly hairy, petals ovate-elliptical, twice longer [immature drupe glabrous, globose].

Petiole about 3 cm. long, terminal leaflet about 6 cm. long, 2-2-5 cm. broad ; lateral

leaflets about 3-4 cm. long, 1-2-1-5 cm. broad. The inflorescence reaches a length of

10cm. The sepals are 1-5 mm. long [barely ? mm.], the petals 2-5 mm. long [1}-14 mm.].”

44. Rh. Gueinzii Sond. I.c. 515; Engler l.c. 442, p. pte, Diels I.c. 589, 623, p. pte.

Description: A glabrous shrub or small tree 12-15 feet high, with slender, flexuous

greyish or reddish, subterete or somewhat angular branchlets. Leaves petiolate, petioles

slender, subterete, slightly canaliculate, especially towards the apex. Leaflets membranous,

dull green above, lighter below, sessile, lanceolate or oblong-lanceolate, from below the

middle cuneate, acute or obtuse at the apex, often mucronulate, margin slightly revolute,

Rh. Gueinzii Sond. Howlett 4. Under side. Rh. microcarpa Schonl. (5025). Upper side.

with a small blunt tooth at the end of each lateral vein; midrib raised on both surfaces,

lateral veins delicate, distinct, but barely raised, tertiary veins reticulate, indistinct.

Panicles axillary, nearly as long as the leaves and terminal, slightly longer than the leaves,

very lax. Calyx segments about } mm. long, petals greenish yellow oblong. Drupe in

the type (Gueinzius 1395) subglobose, verrucose, in Transvaal specimens generally brown,

smooth or almost smooth, slightly asymmetrical, compressed or depressed.

Petiole 1-5-3-5 cm. long.

Terminal leaflets 4-7 em. long; about 1-4 em, broad.

Lateral leaflets slightly smaller than the terminal.

Pedicels 1-2 mm. long. Calyx segments $ mm. long. Petals 14-14 mm. long.

Drupe (in Gueinzius 1395) about 4 m.m, in diameter and greatest diameter in Transvaal specimens

about the same.

DistRIBUTION : Common in the Transvaal. The type (Gueinzius 1395) in herb. Kew

which I have examined is supposed to have come from Natal. It flowers in February.

Mr. H. E. Galpin» supplhed the following information on his No. M 721 :—

“ Bastard Karee—Mphapashane (Sesuto name, but also applied by them to my

No. M 64).

“ Frequent in sand veld on the Springbok Flats, associated with Acacia caffra and

Peltophorum africanum, and occasional in rooibosveld (a light sandy loam of reddish

colour carrying a zuurveld vegetation typified by Combretum apiculatum, the local name

for which is Rooibos) in association with Combretum apiculatum.

“ Also frequently found in granite soil on the mountain slopes west of Naboomspruit.

“ Its timber is greatly inferior to that of the other local species of Rhus, being twisted

and very brittle and of little value. The amount of serration on the leaves varies

considerably in different trees, and on some it is much more pronounced than on the tree

from which the specimens now sent you were gathered.”

Modderfontein, Konrath 112; Rustenburg, Nation 167; Potgietersrust, Burtt-Davy 2184 ;

Magaliesberg, Engler 2770 ; Wonderboom farm, Burtt-Davy 4100 ; Groenkloof, Pretoria, Howlett 2 and 4 :

Pretoria kopjes, Leendertz 54, 174; boschveld between Elands River and Klippan, Rehmann 5144;

Naboomspruit, Galpin M 721; Naboomfontein, Schlechter 4305; Zeerust, Burtt-Davy 90; Kaap River

valley, near Louws Creek, Keet 1486; Pietersburg district ? Grenfell 14, 16.

‘There are two rather stiff branches without flowers in Herb. Alb. Mus. with somewhat

more oblong and slightly thicker leaves than in Rh. Gueinzi. They were collected at

Pretoria (L. Reck No. 13, Colonial Herb. 1090). It is suggested that these are coppice

shoots of Rh. Gueinzw. They agree remarkably well with the type of Rh. leptodictya Diels

in the Berlin Herbarium, with which they have been compared (from tree-steppe, Bulawayo,

Engler 2915).

45. Rh. microcarpa Schonl. n. sp.

Descrietion : Frutex ramulis teretibus subvillosis. Folia petiolata, petiolis gracilibus

dense pilosis supra canaliculatis. Foliola membranacea glauco-viridia subtus pallidiora,

oblonga vel obovata, basi saepius valde angustata, apice acuta mucronata, primum utrinque

dense pilosa deinde sparse pilosa, margine integra, costa venisque prominulis, venis

reticulatis indistinctis. Paniculae dense pilosae laxae, axillares quam folia breviora,

terminales longiores, floribus pedicellatis. Calycis lobi ovati. Petala oblonga. Drupa

parva applanata.

Length of petioles 1-2 cm.

Length of terminal leaflets 4-6-5 em, Breadth of terminal leaflets 1-5-2 em.

Lateral leaflets generally a little over half the size of the terminal leaflets.

Length of pedicels $-] mm. ; calyx lobes 4-} mm. ; petals about 1} mm.

Diameter of drupe about 24 mm.

Distrrpution : Natal, flowering in midsummer.

Ingwangwe forest reserve, P.O. Riverside, about 4,000 feet, Household (5008, 5025); Ngomi, near

Vryheid, Government forester (5147); Harding, 3,700 feet, D.F.O. (5054).

This distinct species is, on the one hand, close to Rh. Gerrard: var. latifolia; on the

other hand, it leads up to Rh. pyroides Burch. var. puberula and may have been derived from

Its height is 3-6 feet. It is common at Ingwangwe.

46. Rh. crispa Harv. Ms.; Rh. Gueinzw Sond. var. crispa Engl. l.c. 443.

Description: Arbor parvus ramulis gracilibus glabris teretibus. Folia petiolata,

petiolis glabris gracilibus subteretibus supra profunde canaliculatis. Foliola glabra,

membranacea saepius crispa supra saturate viridia subtus pallidiora, oblonga vel lanceolato-

oblonga basi angustata apice obtusa, margine undulata vel subplana, rarius crenata, costa

utrinque prominula, venis paucis arcuatis utrinque leviter prominulis, nervis reticulatis

indistinctis. Paniculae axillares quam folia breviorae laxae pubescentes, floribus parvis

pedicellatis. Calycis lobi oblongo-ovati. Petala late oblonga. Drupa

Length of petioles 2-3 cm.

Length of terminal leaflets 5-8-5 cm. Breadth of terminal leaflets 1-2 em.

Lateral leaflets one-half to two-thirds the length of the terminal leaflets sometimes broader than these.

Length of panicles 4-6 cm.; pedicels about 1 mm.

Length of calyx lobes and petals not fully developed in the material available.

Diameter of drupe ?

As)

Rh. crispa Hary. Pegler 1446. Under side.

DistriputTion : Natal and Kentani district (Transkei).

Natal, Gerrard 1397; Kentani district, overhanging stream, 1,200 feet, Pegler 1446 (fl. buds in

February).

47.. Rh. ciliata Licht. in Herb. Willd; Roem et Schult. Syst. Veget. VI, 661; DC. Prodr.

le. 71; Sonder lc. 519; Hngler lc. 418; Diels l.c. 576, 628; Rh. tridactyle [sic !]

Burch. trav. I, 340; Rh. tridactyla Sond. l.c. 516; Engler lc. 446; Diels lc. 590,

641.

Description: A small shrub, not more than 6 feet in height, with spreading, stiff,

unarmed or thorny, glabrous or puberulous branchlets. Leaves petioled, petioles subterete

or slightly winged, canaliculate above, glabrous or puberulous. Leaflets subcoriaceous,

concolorous on both surfaces, glabrous or puberulous on lower surface and then often ciliate

on the margin, linear or lanceolate, rarely oblong, narrowed at the base, at the apex obtuse,

or sometimes mucronulate, more rarely subacute, straight or more rarely falcate ; margin

entire; midrib slightly sunk on upper surface, shghtly prominent on lower, secondary

veins arising at a very variable angle, sunk or fairly distinct, but rarely shghtly prominent.

Panicles glabrous or puberulous, axillary and terminal of variable length, very lax. Calyx

segments ovate, petals oblong. Drupe subglobose, a little compressed.

Length of petioles about 10-13 mm.

Length of terminal leaflets 1-5-2-5 cm. (rarely shorter or longer, up to 5 cm.).

Breadth of terminal leaflets about 3 mm. (rarely up to 4 mm.).

Lateral leaflets about two-thirds the length of the terminal leaflets, often a trifle broader than these.

Length of panicles 1-5-6 em. ; pedicels 13-2 mm. ; calyx lobes about $ mm. ; petals about 1} mm.

Greatest diameter of drupe 4-5 mm.

Rh. ciliata Licht. Wilman 1255. Under side.

It is impossible to separate Rh. tridactyla and Rh. ciliata. Hairy specimens may be

found in localities with glabrous forms. Armed or unarmed specimens occur also in the

same localities. There is also some variety in the shape of the leaflets, in the conspicuousness

of the lateral veins, in the breadth of the petiole, etc. In some respects this species

approaches Rh. longispona H. et Z.

DistRIBUTION : South-West Protectorate, arid portions of northern Cape Province,

Bechuanaland, Orange Free State, and western Transvaal, flowering from December to

March.

Herb. Lichtenstein at Berlin (type of Rh. ciliata Licht.); Klaarwater, Griquatown, Burchell 1946

in Herb. Berlin (Rh. concinwum Burch. Ms.); Asbestos mountains, Burchell 1667, 2131 (types of Rh.

iridactyla = Wilman 1339, 1255); Hay division, Karreefontein, Wilman; Dummery, Wilman 2252,

2254; Griqualand West, at Baaksfontein, Klein Greef Puts, Klein Papkuil, Wilman 1255; Kimberley,

Noran 13315, Wilman 2150, 2151, Marloth 835 (distributed as Rh. puberula HK. et Z.), MacOwan 2573,

2574; near Mafeking, Bolus 6404; near Vryburg, Rogers 26791; near Christiana, Bloemhof district,

Burtt-Davy 1614, 12850; Schoenheid, Bloemhof district, Burtt-Davy 9405; Naval Hill, Bloemfontein,

Potts 75, Drége 6804a ; Orange River, Z. 339 and Burke 504; Rhenosterkop, near the Vaal River, Z. 337

and Burke 275; Olifantsfontein, Orange Free State, Rehmann 3518 ; Ottoshoop, Transvaal, Engler

2901; Wolmaransstad, Rogers 18487, 20617.

forma fastigiata Schonl.—Similar to the type, except that it has a fastigiate (not

squarrose) mode of growth and the lateral veins are more prominent on the under side of

the leaflets, but Engler 6218 and 6227 connect extreme forms with the type. Dinter

mentions that it is 1 m. high and grows in masses on grassy plains.

South West Protectorate: Ankas, Dinter 833; Otjosandjon, Semer 142; Ovampoland and North

Hereroland, Engler 6218, 6227.

48. Rh. dregeana Sond. |.c. 516; Engler l.c. 445; Diels I.c. 590, 625, 641, fig. 6 D, 625.

Description: A small, much branched glabrous shrub, generally not over 3 feet in

height, with squarrose or ascending, sometimes purpurascent slightly angulate branchlets.

Leaves petioled, petioles margined. Leaflets coriaceous, subconcolorous, shining above,

narrowly linear, sometimes slightly faleate, narrowed at the base, at the apex acuminate,

shortly mucronate; margin entire; midrib slightly prominent on both surfaces, other

veins indistinct. Panicles very lax, axillary shorter than the leaves. Calyx lobes ovate,

obtuse, petals oblong. Drupe subglobose, yellowish, shining, slightly fleshy when ripe.

Length of petioles generally about 1 cm.

Length of terminal leaflets 3-5-6 cm. Breadth of terminal leaflets 1-2-2-5 mm.

Lateral leaflets generally two-thirds the length of the terminal leaflets and about as broad as these.

Length of pedicels 2-3-5 mm. ; calyx lobes about ? mm.; petals about 1? mm.

Greatest diameter of drupe 5-5 mm.

Rh. dregeana Sond. Bolus 43. Under side.

Diels (l.c. 641) remarks: “ What has been placed together as Rh. dregeana Sond.

represents perhaps less a strict monophytic connection, but rather the artificial union of

several branches of one type.” This type is perhaps Rh. ciliata Licht., from some of the

glabrous forms of which it is not separated by good technical characters. The leaflets in

Rh. dregeana are, however, more pointed and generally much longer, the flowers and fruits

larger, etc. [Rh. trifoliolata Bak. f. in Journ. of Bot. XXXVITI (1899) 429 coll. by Rand

(No. 66) at Bulawayo is very close to Rh. dregeana. Here, however, in most of its leaflets

the lateral veins are very distinct. ]

DistRiBuTION : Mountains of north-eastern Cape Province and also in the Orange

Free State near Spionkop, between Philippolis road and Smartryk, flowering in autumn.

Sneeuwberge, Drége 2374; Mooiplaats, Stormberge, 46,000 feet, Drége 3448 ; near Graafi-Reinet,

2,500 feet, Bolus 135 ; Sneeuwberge, 4,000 feet, Bolus 43 ; Graaff-Reinet, Sr. Francis 29; Burchell 2853

(sub Rh. lancea L. £.); Burghersdorp, 5,000 feet, Flanagan 1530; in gorge below Buffels River waterfall,

Basutoland, cire. 7,100 feet, Galpin 6599; near Philippolis, Orange Free State, circ. 5,050 feet, Smith

4489 (rare).

49. Rh. erosa Thunb. in Fl. Cap., ed. Schultes 363 ; Sonder l.c. 516; Engler l.c. 439; Diels

lic. 587, 595, 624, 625, fig. 6 A, B, C, 642.

Rh. serraefola Burch. trav. II, 100.

Description: A much branched glabrous, sometimes decumbent shrub with sub-

flexuous branchlets, usually 5-6 feet high, sometimes reaching a height of 10 feet and up

to 30 feet across. Leaves very variable in size, petioled; petioles slender, margined,

canaliculate above. Leaflets coriaceous, usually ~“ varnished” (especially on the upper

side), linear, sometimes slightly falcate, narrowed towards the base, acute; margin in a

varying degree eroso-dentate or sometimes entire on the same plant, teeth acute; midrib

prominent on both surfaces, lateral veins more or less prominent on both surfaces, tertiary

veins usually indistinct or not visible. Panicles very lax with slender branches, usually

somewhat shorter than the leaves, bracts very small. Calyx segments unequal, obovate,

very obtuse. Petals obovate. Drupe yellowish, shining, subglobose.

Length of petioles 2-3 cm.

Length of terminal leaflets up to 12 cm. Breadth of terminal leaflets (without teeth) 2—5 mm.

Lateral leaflets similar to terminal leaflets, usually about three-quarters of their leneths.

Length of pedicels 2-3 mm.; calyx lobes about } mm.; petals about 1 mm. #

Diameter of drupe about 4 mm.

Known as “ Bezenbosch.” Native name Tselabelo.

Rh. evosa Thunb. Phillips 3121. Upper side.

DistRiBUTION : Stony kopjes, usually frequent, in the mountainous districts of the

Midlands and north-eastern Cape Province, Basutoland and Orange Free State (Rogers 2484

has been distributed with the erroneous locality “Table Mountain, Capetown ”’).

It is used for making rough brooms and thatching sheds. Grown in gardens it is

quite ornamental. Phillips notes that the witch doctors in Basutoland use it for “ rain-

making and as a medicine for diarrhcea in man and cattle.” It is also used as a hairbrush

by Kaffirs in the Orange Free State.

Herb. Thunberg ; Burchell 2729, 2697 (in Herb. Kew, etc.) ; Koutveld mountains, near Murraysburg,

5,000 feet, Tyson 142; Graafi-Remet, 3,300 feet, Bolus 535, Bowker 39; common near Grootfontein,

Middelburg, C.C., Schonland ; near Rosmead station, Flanagan 1529; Cradock division, Z. 346, Cooper

476, Kuntze ; Somerset Hast, Atherstone 171; Roode Rand Farm, near Klaas Smits River, 3,550 feet,

Galpin 2512; Winterfield, Drége 813; Shiloh, Drége, E. et Z. 1133, Baur 909; Cala, Pegler 1742, ;

Colesberg, Shaw 40; Majuba Nek, and Sterkspruit, Herschel division, Hepburn 250, 33; Leribe plateau,

Basutoland, 5—6,000 feet, Dieterlen 78, Phillips 842; Bethlehem, Phillips 3121 ; Potts 3309, Bloemfontein,

Rehmann 3832 (var. subintegra, Sczyscz.), 3795, Potts 56; Ladybrand, 6,000 feet, Patterson (5076) ;

near Kicksburg, Potts 3165; near Trompsburg, Orange Free State; near Fouriesburg, Potts 3276.

50. Rh. Bolusii Sond. Ms.; Engler l.c. 436; Diels |.c. 584, 626.

This species has only been collected by Bolus near Graaff-Reinet, and may have to

be dropped in view of Diels’ statement (l.c. 626) that in a plant of the Berlin Bot. Gardens

(1854) the leaflets of Rh. erosa are distinctly broadened, and thus assume entirely the

appearance of Rh. Bolusii Sond. He is, therefore, inclined to look upon the latter as a

local form of Rh. erosa. He explains that its occurrence is possibly due to a favoured

habitat which might produce a form like the one produced under cultivation. Perhaps a

slight indication that this view is correct may be the fact that Mr. R. A. Dyer found in

one flower of Bolus’ plant a sixth petal which was smaller than the others. Similar

abnormalities are often found in cultivated plants. The flowers are larger than in Rh.

erosa. Diels’ views mentioned do not tally with his statement (l.c. 585) that he thinks that

Rh. dentata Thunb. is its nearest relation. Engler placed it next to Rh. oxyacantha Cav.

found on the Cape Verde Islands and in the Mediterranean Region.

Description: A glabrous shrub with suberect, slender, flexuous branchlets. Leaves

petiolate, petioles slender, slightly margined, canaliculate above. Leaflets subcoriaceous,

somewhat shining above, obovate-oblong or oblong, narrowed towards the base and with

acute, often mucronate apex, margin from below the middle or higher up serrato-dentate,

teeth acute, subacute or blunt, midrib and lateral veins prominent, tertiary veins not

visible. Panicles very lax with slender branches, bracts small ovate-lanceolate. Calyx

segments ovate, bluntish or triangular, acute. Petals broadly oblong. Drupe :

Length of pedicels 1-5 cm.

Length of terminal leaflets 2-3 em. Breadth of terminal leaflets about 1 cm.

Lateral leaflets resembling the terminal leaflets and usually about two-thirds their length.

Length of pedicels 2-3 mm.; calyx segments about ?-]1 mm.; petals about 2 mm.

Rh. Bolusti Sond. Bolus 737. Upper side.

Rh. graciluma Engl. Rehmann 1882.

Upper side.

DistriBuTION: Cave Mountain, near Graaff Reinet, 3,900 feet, Bolus 737.

51. Rh. gracillima Engl. l.c. 445; Diels l.c. 590, 614.

Description : A very slender shrub, 2-4 feet high, unbranched, except when it reaches

3-4 feet, sparsely and patently pilose all over (except in var. glaberruma Schonl.\. Leaflets

shortly petioled, petioles canaliculate and slightly margined above. Leaflets coriaceous,

concolorous, very narrowly linear, sometimes slightly falcate, narrowed towards the base

and at the apex very acute; midrib prominent on both surfaces, other veins not visible.

Panicles very lax, and with very delicate branches, axillary ones generally slightly shorter

than the leaves, sometimes much longer, terminal longer than the leaves, bracts small,

subulate. Calyx segments ovate acute, petals oval. Drupe . -

var. glaberrima ‘Schon. [Rh. filiformis Schinz in Vierteljahrschr. d. naturf. Ges.

Ziirich LV (1910) 239].—Glabrous and with longer leaves than the type.

Length of petioles 5-7 mm. (in var. glaberruma up to 15 mm.).

Length of terminal leaflets 6—7 cm. (in var. glaberrima up to 8 cm.).

Breadth of terminal leaflets 1-5 mm. (in var. glaberrima up to about 1 mm.).

Lateral leaflets very much like the terminal ones and almost as long.

Length of bracts 1 mm.; pedicels 1-2 mm.; calyx segments 3-1 mm.; petals 14-2 mm.

Distripution : Transvaal.

Between Mienaarsfarm and Elands River, Boschveld, Rehmann 1882; Premier Mine, Rogers 22421.

var. glaberrima—hills near Wilge River, 4,600 feet, Schlechter 3746 (fl. in Nov.).

52. Rh. Wilmsii Diels in Engl. Bot. Jahrb. XXIV (1898) 501, Lc. 589, 614, 641.

Description: A much branched glabrous shrub with slender branchlets. Leaves

petioled, petioles rather long, distinctly margined, canaliculate above with a slight ridge

in the centre of the channel. Leaflets concolorous, coriaceous, linear, narrowed towards,

the base, blunt or emarginate at the apex or narrowed into a mucro, margin entire or very

slightly undulate, midrib very prominent on both surfaces, lateral veins prominent on both

surfaces, tertiary ones not visible. Inflorescence, flowers, and fruit unknown.

Length of petioles 3-3-5 cm.

Length of terminal leaflets 7-9 cm. Breadth of terminal leaflets 4-7 mm.

Lateral leaflets very much like the terminal leaflets and not much shorter.

Rh. Wilmsti Diels. Wilms 249. Upper side.

Distripution : Only found by Wilms (No. 249) at the large waterfall, Lydenburg,

Transvaal.

53. Rh. Keetii Schon]. n. sp.

Description: Frutex glaberrimus gracillimus circ. 1 m. altus saepius simplex vel

apicem versus parce ramosus cauli basi vix 5 mm. diam. superne tenuissimo vix 1-5 mm.

diam. tereti; folis longe petiolatis, petiolis gracilibus supra canaliculatis vix alatis basi

incrassatis foliolis anguste linearibus basi et apice angustatis summo acutissimis saepius

+ falcatis utrinque subconcoloribus margine integris anguste albo-cartilagineis, costa utrinque

prominenti, nervis vix prominentibus vel saepius immersis, foliolis lateralibus et terminalibus

subaequalibus vel inaequalibus; paniculis terminalibus laxissimis bracteis parvis lanceolatis

acutis, calycis segmentis ovatis basi contractis, petalis pallide flavis late oblongis, floribus

foemineis cum staminodus ; drupa subglobosa, leviter depressa.

Rh. Keetii Schonl. Keet j1435.

Under side.

Petiole 33-5 cm. long (often curved), barely 1 mm. in diameter.

Leaflets, both terminal and lateral, 11-15 em. long, about 3 mm, broad.

Calyx segments #-] mm. long. Petals about 1-7-2 mm. long.

Drupe c. 5 mm. in diameter.

Distripution : Only found near Lydenburg, Transvaal.

Keet 1345.—On the Klip River in Steelspoort Park, Lydenburg district, 30° 0’ E. and 24° 50’ S.,

4,000 feet, June (fl. and fr.).

Mr. Keet notes: ‘ Slender bushes about 3-4 feet high, mostly single-stemmed, stems

pencil thickness. Found in grass-covered and very rocky places.”

This species is closely allied to Rh. Wilmsw Diels from the same neighbourhood, of

which the flowers and fruits are unknown. Perhaps it may have to be referred to it later

as a variety, but I have thought it best to keep it separate for the present because in Rh.

Wilms the petioles and leaves are not so slender, the wings of the petioles are more

pronounced, the leaflets generally shorter and suddenly contracted at the apex, the lateral

veins are prominent on both surfaces. In both species there is a narrow ridge in the middle

of the channelled upper side of the petiole.

HORRIDA group.

54. Rh. horrida H. et Z. in Enum. 1135; Sonder l.c. 415; Engler l.c. 415; Diels l.c. 575,

635; Rh. platypoda HK. Mey. in Drége exsicc.

Description : A shrub with rigid, patent, usually thorny branchlets. Young leaves,

panicles, pedicels and calyx densely covered with grey or red glands. Leaves more or less

fasciculate, petiolate, petioles distinctly wingefl. Leaflets coriaceous, linear-cuneate,

obtuse, entire. All veins (including midrib) immersed. Panicles few domerad, axillary,

shorter than the leaves. Pedicels very short. Flowers very small. Drupe oblique,

glabrous, beaked with short styles.

Petioles 2-8 mm. long, usually $ mm. broad.

Terminal leaflet 4-8 mm. long; 1-3 mm. broad.

Lateral leaflets a little shorter than the terminal.

Petals a little over 1 mm. long.

Drupe 4 mm. long; 5 mm. broad.

Rh. horrida BE. et Z. Schlechter 11179. Under side. 3/2.

Distripution : Namaqualand and Bushmanland, flowering throughout the summer.

In sandy soil, alt. III, mountain sides of the Khamiesberg and in Namaqualand, HE. et Z. 1135,

Keklon 35; Springbokkeel, Ecklon 45, Z. 348; hills near Rietkloof, 2,500 feet, Schlechter 11179 ; Drége

in Herb. Kew (Rh. platypoda EK. Mey. b).

55. Rh. longispina EH. et Z. in Enum, 1135; Sonder l.c. 520; Engler lc. 415;

Diels l.c. 575, 635.

Description: A thorny squarrose shrub up to 10 feet high. Stem and branches

grey. Young organs (branchlets, leaflets, panicles) covered with red glands, older ones

glabrous. Branches short, 3-4 cm. long, and with thorny end or bearing short branchlets

which turn into thorns. Leaves petiolate, mostly fasciculate, petioles very variable in

length, more or less according to the size of the leaflets, usually winged, the wings varying

in width even in leaves of the same size. Leaflets coriaceous, cuneate-obovate, obtuse or

subemarginate, entire, the lateral ones often oblong and slightly asymmetrical at the base.

Midrib and lateral veins (which are not quite straight and branched towards the margin)

raised on both surfaces, tertiary veins coarsely reticulate (not visible in small leaflets).

Panicles numerous, usually clustered with the leaves, densely covered with red glandular

hairs, usually slightly shorter than the leaves, with few and short side branches. Flowers

shortly pedicellate. Calyx segments ovate, green. Petals pale yellowish, oblong, narrowed

towards the base. Drupe green, sometimes turning brown, rather juicy at first, shining,

subglobose, getting flattened when old, often asymmetrical, retainmg sometimes the

hardened styles.

Petioles 5 mm. to 3-5 em. long and in exceptional cases up to 3 mm. broad.

Terminal leaflets 1-5 em. long; 5 mm. to 2-2 em. broad.

Lateral leaflets one-third to one-half the size of the terminal leaflets.

Pedicels 1 mm. long. Calyx segments barely $ mm. long.

Petals about 1} mm. long. Drupe 5-54 m.m. in diameter.

Rh. fongispina EB. et Z. E. and Z. 1048. Upper Rh. longispina EK. et Z. Dyer 61. Upper side.

side.

Drstripution : In somewhat arid places from Swellendam to Hast London penetrating

miand to Graaff-Remet, usually flowering in late summer or autumn.

The affinities of this species seem to point to Rh. rigida though it is often placed near

Rh. lucida. In some respects it also approaches Rh. ciliata Licht. The great variability

of this species in the size of leaflets, the length and breadth of the petioles is noteworthy,

and while in other species of Rhus great variability raises a suspicion of hybridization,

such a suspicion cannot be entertained here.

Burchell 5381, 5383, 3312; Swellendam, Mund and Maire; hills near Mossel Bay, 50 m., Schlechter

5725; Keurboom RB. hill, Duthie 747 ; Armoed, Britten 1747; west bank of Gamtoos River, near Hankey

(Presl), 150 feet, Fourcade 2280 ; Uitenhage, Ecklon (labelled ‘‘ Rhus pterota Presl”); ib., (Krakakamma)

Z. 456, 2245 ; amongst bushes on the banks of the Zwartkops River, Ecklon 46, 1048 ; Uitenhage, Addo,

and near Fort Beaufort, B. et Z. 1116 ; Redhouse, Paterson 2112 ; Aloes, J. L. Drége 3165 ; Sandflats, Rogers

191; Cookhouse, Rogers 2457; Graaff-Reinet, Bolus 660 ; commonage north of Grahamstown, Dyer 58,

61, 62; Otterburn, Fish River randt, Schonland ; Nahoon River mouth, Galpim 5689.

On the farm Aylesby, near Riebeek East, two forms are found not far from one another. One

has deep green leaflets ; in the other they are grey. In the former the leaves are always a little larger

than in the latter.

56. Rh. rigida Mill. dict. No. 14; Sonder lc. 520; Engl. l.c. 416; Diels l.c. 576, 633,

634, fig. 7 L.

Rh. vvmosa EK. et Z. in Enum. 1134.

Rh. triceps KH. Mey. in Drége exsicc.

Description : An erect, rigid, glabrous shrub about 14-3 feet high, with approximate

and sometimes purplish and resinous branchlets. Leaves shortly petiolate, petioles

canaliculate above, narrowly margined. Leaflets rigid, corfaceous, somewhat glaucous,

sometimes resinous, cuneate lanceolate or linear lanceolate, apiculate, concolorous on

both surfaces, margin entire or with one or two acute teeth at the apex; mudrib slightly

raised, immersed or slightly sunk on the upper surface, slightly prominent on the lower,

lateral veins invisible or barely visible, tertiary veins invisible. Calyx segments ovate,

petals oblong. Drupe glabrous, oval or oblong, compresseed and depressed, sometimes

slightly oblique, tricuspidate.

Length of petioles 1-1-5 cm.

Length of terminal leaflets 4-5 cm. Breadth of terminal leaflets 5-8 mm.

Lateral leaflets about two-thirds the length of the terminal leaflets and about as broad as these.

Length of calyx segments about }? mm.; petals 1{-1$ mm.

Greatest diameter of drupe about 6 mm.

Rh. rigida Mill. Pillans (5132). Under side.

DistRipuTIon : In the north of the south-west Cape region from the Winterhoek to

the Giftberg. Flowers from September to November.

It is very difficult to suggest close affinity of this species with any others. Diels

places it into his Lucida group. The fruit is very much as in RA. populifolia KH. Mey. and

allies, but the vegetative organs are somewhat close to those of some forms of Rh.

magalismontana Sond.

Sonder (l.c. 520) has separated as a variety Drége 6797 from the Giftberg with panicles

more compound than in the type as long or longer than the leaves, but there seems to be

no justification in keeping this up.

Drege in Herb. Kew (Rh. triceps E. Mey.) ; between Twenty-four River and Olifants River, Ecklon

45; mountains near Heerelogement, H. & Z. 1134 (Rh. rimosa E. et Z.); summit slopes of Pikenierspass,

Pillans 5132; Packhuisberg, 2,000 feet, Schlechter 10798; Saron, 1,000 feet, Schlechter 7884; plains

west of the Giftberg, on rocky places poor in vegetation, 300 feet, Diels 409 (this has a more straggling

habit and shorter leaves than the type).

57. Rh. magalismontana Sond. l.c. 510.

Rh. burkeana Sond. |.c. 514; Engler l.c. 417; Diels l.c. 576, 639.

Rh. corvacea Engl. l.c. 418; Rh. oblanceolata Schinz in Beitr. zur Kenntn. d. aftr.

Flora XXI, Bull. de Vherb Boiss. 2 iéme série VIII, 638.

Description: A much branched shrub with young branches, leaves and panicles

shortly tomentose (tomentum often yellowish or fuscous), branchlets terete, striate. Leaves

petiolate, petioles one-third ¢o one-half the length of the terminal leaflets, broadly furrowed

above, marginate, the margins slightly thickened. Leaflets subcoriaceous, sessile, often

glabrous when old, obovate oblong or oblong with a cuneate base and obtuse, rarely acute,

mucronate apex; margin entire, not revolute; midrib prominent on both surfaces, the

lateral and tertiary reticulate veins also usually fairly prominent. Panicles axillary and

terminal, shorter than the leaves or a little longer, loosely flowered, bracts subulate, flowers

pedicelled. Calyx lobes ovate-triangular, acute. Petals oblong. Drupe shining, subglo-

pose.

Rh. magalismontana Sond. Moss 2969. Under

side.

Rh. magalismontana Sond. Leendertz 1313. Under

side.

Length of petioles 1-2 cm.

Length of terminal leaflets 34-6 cm. (sometimes much smaller).

Breadth of terminal leaflets (average about 1 cm.).

Lateral leaflets about two-thirds the size of the terminal leaflets.

Length of calyx lobes about } mm.; petals about 1} mm.

Diameter of drupe (Burtt-Davy 5332) barely 4 mm.

DistRisution : Stony kopjes in the Transvaal.

Magaliesberg, Z. 11 (in Herb. S.A. Mus.), Engler 2812 (in Herb. Berlin), Burtt-Davy 2666; rocky

places on the Aapies River, Z. 335 (type of Rh. burkeana Sond.); near Johannesburg, Galpin 1469, 1493, Moss

2969 ; Ottoshoop, Engler 2812 ; between the Drakensbergen and Pretoria, Wilms 244; near the Houdek,

4,800 feet. Schlechter 2733; near Pretoria and Klippan, Rehmann (type of Rh. coriacea Engl.) ; Warm-

baths, Leendertz 1313, Burtt-Davy 5332; on the Olifants River, 5,000 feet, Schlechter 3773 (type of

Rh. oblanceolata Schinz) ; Paardevalley, near Zeerust. Burtt-Davy 7183 ; northern Transvaal, Le Doux 4.

Diels (1.c. 630) states that in Rh. burkeana Sond. he found most typical sunk stomata

such as he found in no other species of African Rhus. (See the illustration of his, Pl. XIV, 8S.)

DISCOLOR group.

58. Rh. discolor E. Mey. in Drége exsice ;, Sonder I.c. 507 ; Engler l.c. 447; Diels l.c. 590,

614.

Rh. rufescens B. et Z. in Enum. 1093 (non Hamilton).

Rh. villosissima Engl. lc. 447.

Rh. grandifolia Engl. l.c. 434.

Description: A low poorly branched shrub, 1}-3 feet high, with densely gtey or

fulvoustomentose branchlets, petioles and inflorescences. Leaves petiolate, petioles short,

semiterete, canaliculate above. Leaflets coriaceous, above green more or less pubescent or

pilose, rarely villous or glabrous, below more or less fulvous or whitish tomentose with

adpressed hairs, rarely villous, the midrib and veins sometimes more decidedly pilose,

linear lanceolate or lanceolate or oblong, rarely obovate, more or less cuneate at the base,

apex usually acute, sometimes obtuse, usually mucronate ; margin shghtly revolute, entire

or in the broader leaflets often more or less grossly dentate or rarely eroso-dentate ; midrib

not prominent or slightly prominent above, decidedly prominent below, lateral veins not

prominent or even depressed above, not prominent or more or less prominent below,

tertiary veins reticulate, but only sometimes plainly visible above, not visible below. Panicles

dense, axillary a little shorter than the leaves and terminal, somewhat longer, branches short,

multiflowered, flowers glomerulate, lower bracts foliaceous, lanceolate or linear, upper small

linear. Calyx segments lanceolate acute or subovate, tomentose on the outside. _—_— Petals

subovate. Drupe subglobose, when young sometimes puberulous, when ripe yellowish,

subglobose, glabrous, shining, sometimes crowned with the enlarged styles.

Length of petioles 5-8 mm. (in rare cases up to 1-6 cm. ; in one case up to 3:3 em.).

Length of terminal leaflets 5-8 em. (up to 16 cm. in Rh. grandifolia, 34 cm. in Baur 912).

Breadth of terminal leaflets 1-1-5 cm. (7-5 em. in Rh. grandifolia, 3 mm. in Baur 912).

Lateral leaflets about two-thirds to four-fifths the size of the terminal ones.

Length of pedicels barely 1 mm.; calyx segments }-1 mm. ; petals 14-14 mm.

Diameter of drupe 4-5 mm.

Rh. discolor EB. Mey. Drege 1839. Rh. discolor EK. Mey. Drége 1839.

Distripution : Amongst grass and low shrubs on mountains of south-east Cape

Province, Komgha division, Transkei, Pondoland, along the Drakensberg range, Natal,

Orange Free State, Transvaal. Flowers from November to January.

Alhed to Rh. arenaria Engl. (Angola).

Like so many other “ species” of Rhus, Rh. discolor is extremely variable, which shows

itself chiefly: (1) in the hairiness of the leaflets, ranging from glabrous upper surfaces

through pilose ones to almost villous. Rh. villosissima Engl. was based on an extreme

form. Already Diels (l.c. 590) has sunk this species in Rh. discolor and Engler has later

done this also The former says: “die Behaarung wechs !t iiberall bedeutend”’ ; (2)

Rh. discolor. HE. Mey. Rogers 1693. Uyper side

the shape of the leaflets. A broadening of these is especially seen in RA. grandifolia Enel.

but already in Meyer’s type of Rh. discolor (in Herb. Kew) some leaflets are almost obovate-

cuneate ; (3) the broadening is usually accompanied by an increase in length ; and (4) it

is also usually accompanied by the appearance of irregular teeth in the upper part of the

leaflets. These, however, are also already found sparingly in Meyer’s type.

Taking extreme forms such as Rh. grandifolia Engl. (Rehmann, Inanda) and Baur 912,

for which some measurements were given above, it seems almost incredible that they should

have to be placed into the same species, but there is no other course open, since they are

connected by intermediate forms.

Engler (l.c. 448) recognized two forms, in addition to the type :—

B paucinervis.—Leaves sometimes with five leaflets, leaflets acute, sometimes bidentate,

lateral veins less numerous, ascending (not spreading).—None of the characters given, are

distinctive.

Faku, Gerrard 1403 in Herb. Kew; Transvaal, Page’s hotel, Rehmann.

y brevifolia.—Leaves shorter, oblanceolate, 4-5 longer than broad.

Drakensberg, Rehmann in Herb. Berol.

This, again, cannot stand, as in Rehmann’s type the terminal leaflet is up to 4°5 cm.

long, which is not much less than what we find in other forms.

Rh. discolor K. Mey.

‘Baur 912.

Rh. discolor B. Mey., D. grandifolia (Bngl.). Rehmann, (‘Tertiary veins reticulated.)

I propose to divide the species into the following forms :—

A. typica.—Leaflets lanceolate or oblanceolate, rarely broadened, mucronate, above

pubescent. Lateral veins not prominent (or at all events not conspicuously so) on

the lower surface. Terminal leaflets 5-7 cm. long, rarely slightly shorter or longer.

B. villosissima (Rh. villosissima Engl.).— Adult leaves more hairy on _ both

surfaces, lateral veins generally very prominent on the lower surface. Shape of leaflets as

in A, but usually slightly larger.

C. latifolia.—Leaflets ovate or obovate, entire or dentate. Lateral veins not

prominent on the lower surface. As these veins are longer than in A and B they curve

upwards. The leaflets reach a length of 11 cm.

D. grandifolia (Rh. grandifolia Engl.)—Leaflets obovate-cuneate, often grossly

irregularly eroso-dentate, but on the same shoot there may be leaflets agreeing with a in

shape. Adult leaflets glabrous above, reaching a length of 16 cm. Lateral veins prominent

on lower surface.

A. Between Sandplaats and Komgha, Drége 3449, 5584 (one leaflet approaches c); Cowie Mt., Bedford,

Bennie 281 ; Hogsback, c. 6,000 feet, Rattray 66; Katberg, MacOwan 864 (largest terminal leaflet 12 cm.

long); Winterberg and Chumie mountain, E. et Z. 1093; mountain side, Gwetwyn farm, Queenstown

division, Galpin 8301; N’Achbanya mountain, Queenstown division, Galpin 1903; hills near Shiloh,

Baur 912 (in Herb. Berol. with the smallest leaflets known); Mvanyeni, near Cedarville, Griqualand

East, Bandert 106a (leaflets almost linear, very narrow); Qumbu district, Dwyer 5111 (the collector

notes that it forms a mass of underground roots which interfere with ploughing and that the drupes are

eaten by children); Ntsubane, near Lusikisiki, Evans (5048, 5052); Ingwangwane, Household 5047 ;

hills near Ladysmith, 4,000 feet, Wood 849; near Houderivier, c. 4,500 feet, Schlechter 3737 ; Majuba,

Rogers 53; Magaliesberg, Burke 328; Johannesburg, Moss 6175; Carolina, Rogers 11524; Wonder-

fontein, Nelson 259. :

The following approach B: Barber 611; Matatiele, Hilner 26; Sterkspruit, Herschel district,

Hepburn 2, 155; Pietpotgietersrust, Rogers ; The Downs, Pietersburg division, Rogers 21879 ; Graskop

forest station, Evans (5110).

B. Houtbosch, Rehmann 5557, Bolus 10999; Carolina, Rogers 11552; rocky mountain sides near

Komgha, 2,000 feet, Flanagan in Herb. Austr. Afr. 1421; Kokstad, 4,700 feet ; Camperdown, Alexandra

county, Natal, Rudatis 1808; Klip River, near Johannesburg, Engler 2741 ; river bank near Lydenburg,

Wilms 250.

C. Leribe, Basutoland, Phillips 551, 614, 844 (in the last number some of the leaflets are petiolulate,

and the petiole reaches a length of 3:3 cm.); ib., Dieterlen 29; Drakensberge, Symons 7588 ; Mont aux

Sources, McLean and Bayer 201 ; Besters Vley, near Harrismith, Bolus 8138, Flanagan 1864 ; Harrismith,

Herb. Alb. Mus. 5142 ; hills near Ladismith, Wood 849 ; Culvers, Weenen district, Rogers ; Johannesburg,

Leendertz 1693; Irene, Leendertz 695; Ermelo, Tennant 6939; Potgietersrust, Rogers 5106

Specimens from Xalingena forest, Polela, Natal, Alb. Mus. Herb. 5011, may also be placed here.

Others from Ntsubane, near Lusikisiki, Pondoland, Fraser 5149, 5150 have villous leaflets. This is also

the case in the young leaflets of Dieterlen 29.

D. Inanda, Natal, 1,800 feet, Wood 742 (Wood says = Gerrard 1403. This number is quoted by

Engler under his var. paucinervis).

Occasionally one meets with much branched specimens, e.g. Dieterlen 962 from mountain slopes

in Basutoland, but as these do not differ otherwise from typical Rh. discolor I have refrained from giving

them a distinct name.

The following specimens were not classified at the time when they were submitted to me: Dunelm

Farm, Fouriesburg, Orange Free State, Potts 3075; Bethlehem, Orange Free State, Potts 2999.

Burtt-Davy records Rh. discolor from the following additional Transvaal localities: Spion Kop,

Ermelo district, Burtt-Davy 8096; De Emigratie, Ermelo district, Burtt-Davy 4168; Sterkhill,

Lydenburg district, Burtt-Davy 481; Rietvley, Machadodorp, Burtt-Davy 7085: near Carolina, Bolus

11756; Spitskop, Pott 4932.

59. Rh. pondoensis Schonl. n. sp.

Description: Frutex parvus caule lignoso efoliato repenti ramis erectis aggregatis

basi subteretibus sursum angulatis dense foliosis sparse setaceo-pilosis, pilis saepius curvatis.

Folia breviter petiolata, petiolis glabris alatis. Foliola coriacea glabra lineari-lanceolata

acutissima vel mucronata basi saepius cuneata, margine integra plana conspicue incrassata,

costa utrinque valde prominenti, nervis lateralibus numerosis utrinque prominentibus.

Mixa

Ssh

Wood 3002. (Flowers 3/1.)

Lx |

Rh. pondoensis Schon.

Rh. pondoensis Schonl. Wood 3002. Under side.

Paniculae laxae ex axillis foliorum superiorum quam folia paullam longiores vel breviores,

bracteis lanceolatis acutis. Calycis segmenta triangularia acuta glabra. Petala flava quam

calycis segmenta subduplo longiora. *Drupa nitida subglobosa vel ovoidea, stilis

persistentibus.

Branches 30-40 em. high. Length of petioles 14-3°mm.

Length of terminal leaflets 4-5 cm. Breadth of terminal leaflets 8-10 mm.

Lateral leaflets generally not much shorter than the terminal ones, but slightly narrower.

Length of calyx segments nearly 1 mm.; petals 13-2 mm.

Diameter of drupe about 5 mm.

DistRIBUTION : Only known from Murchison, Pondoland, where it was collected by

the late Mr. J. M. Wood (No. 3002 in Herb. S.A. Mus.). The flower examined was note-

worthy. It had a normal ovary with three styles, some stamens were rudimentary, though

slightly larger than usual in female flowers of Rhus, other stamens were evidently normal

and functioning.

There can be no doubt that this species is closely allied to Rh. discolor EK. Mey., but the

subsessile perfectly glabrous leaves seem to differentiate it sufficiently from this very

variable species.

TOMENTOSA group.

60. Rh. tomentosa L. spec. 382 (non Mill.); Thunb. Prodr. 52, Fl. Cap. ed. Schultes 266 ;

Sonder l.c. 508; Pappe Sylv. Cap. 13; Engler l.c. 407; Diels l.c. 572, 592, 594,

615, 616, fig. 3; 618, 620.

Rh. ellipticum Thunb. Fl. Cap. ed. Schultes 263; Rh. elliptica K. Mey. in Drége

exsicc.

Rh. bieolor Licht. in Herb. Willd.

Rh. discolor Schrad. hort. Goetting. (ex Sonder).

Rh. Eckloni Schrad. Hort. Goett.

Rh. lobata Poir. in Herb. Berol.

Rh. Plukenetiana HK. et Z. in Knum. 1110 (non Rh. -afric., etc., Pluk. t. 219, fig. 7).

Description : A shrub or small tree, often 12—15 feet high, branches somewhat patent,

branchlets rufous tomentose or pubescent at first, later glabrous, somewhat angular.

Leaves petiolate, petioles subterete, slightly furrowed above, usually rather slender. Leaflets

coriaceous, dark green or greyish green and glabrous when mature above, covered with

dense short greyish or fulvous hairs below, elliptical or ovate, or obovate, acute or acuminate

at both ends or obtuse at the apex, sometimes mucronulate, very often petiolulate (petiolules

and petioles often reddish) ; margin flat, entire or very often from above the middle coarsely

serrate, teeth acute or obtuse; midrib on upper surface sunk or in the lower part of the

leaflets sightly prominent, on lower surface prominent, lateral ves not prominent on

upper surface or even sunk, slightly prominent on lower, tertiary veins not visible on either

surface or reticulate and plainly visible on upper. Panicles terminal, longer than the

leaves, much branched, multiflowered, densely greyish tomentose, bracts linear lanceolate,

flowers distinctly pedicelled. Calyx segments ovate, obtuse, villous on the back. Petals

oblong, greenish, slightly pilose or pubescent on back. Drupe densely greyish tomentose,

compressed and depressed, often somewhat asymmetrical or sometimes subglobose.

Length of petioles 1-5-2 cm.

Length of petiolules up to about 8 mm. in terminal leaflets, shorter or absent in lateral leaflets.

Length of terminal leaflets 5-7 cm., occasionally much smaller.

Breadth of terminal leaflets 1-3-5 cm.

Lateral leaflets generally about two-thirds the size of the terminal leaflets and resembling them.

Length of floral bracts 2-3 mm. ; pedicels 1-2 mm. ; calyx segments about } mm. ; petals 14 to nearly

2mm. Greatest diameter of drupe 5-6 mm., smallest 3 mm., height about 4 mm.

Rh. tomentosa L. Zahn (5050). Under side. Rh. tomentosa L. Zahn (5050). Under side.

DistrisutTion : Common amongst shrubs on the hills and mountains of south-west

Cape Province (altitude about 200 feet to about 3,000 feet) from Little Namaqualand to

George and then in isolated localities to the Transkei (reaching on the south-eastern

mountains an altitude of about 5,000 feet and not descending below about 1,500 feet),

also in the Wakkerstroom district, Transvaal. It flowers from June to November. Forester

Jordaan, Knysna, notes that the flowers have a most disgustingly offensive smell. Sim,

in Forest Flora 195, says that it is seldom of timber size or used for other than firewood

purposes. Like many other species of Rhus it is known as “ Taaibosch ” from the bark

being used as rough cordage.

This species is often represented in eighteenth century herbaria, and was probably

one of the earliest introductions to European gardens of South African plants. In

Thunberg’s herbarium it is found as Rh. tomentosa and as Rh. elliptica. In Willdenow’s

herbarium it appears also as Rh. tomentosa and Rh. bicolor Licht., much reliance having

evidently been placed on the shape of the leaflets. Rh. lobata Poir. in Herb. Berol. was

based on specimens cultivated at Teneriffe. Ecklon and Zeyher in Enum. 1109, besides the

type from south-west Cape Province, distinguished three varieties: f witenhagensis

(Uitenhage), y sylvatica (Krakakamma), and 6 swellendamensis (Swellendam). Sonder

(l.c. 509) has the type and a var. petiolaris in which the leaflets are elliptic oblong, acute

or acuminate at both ends, with long petiolules without, however, separating them

geographically. Hngler (l.c. 408) besides the type has two varieties: 6 petiolaris Sond.

and y swellendamensis. In the last the leaflets are oblong elliptical or lanceolate mucronulate,

quite entire. He remarks that, through the last, Rh. tomentosa comes very close to Rh.

angustifolia. As I have stated under Rh. angustifolia, the question whether this should be

united with Rh. tomentosa is, with our present knowledge, entirely a matter of opinion.

With the large amount of material which I have examined I am unable to separate the

ND,

iis; GEA

— ~~, Hayes Mtn

NW) i

Rh. tomentosa LL. Fourcade 840.

100

species into distinct varieties or even fairly well-defined forms as the distinctive characters

relied upon by various authors may be found on the same plant and in widely separated

localities. Coppice shoots have often larger leaves than fruiting branches.

I have in the description of most species not referred to the disk which is usually

5-crenate, though with a tendency to curl so as to become almost 10-crenate. In MacOwan

296 and Schlechter 8634 I found the margin with ten regular blunt teeth, while in Rh.

angustifolia I found sometimes five acute, divided teeth. I do not know whether these are

general reliable distinguishing characters.

Herb. Thunberg; Herb. Willdenow ; Burchell 299, 608, 3460, 5491; common on hills near Leeuw-

fontein, Pearson 3221 (leaflets approaching those of typical Rh. populifolia in size and shape) ; Packhuis-

berg, Schlechter 8634 ; Riebeckskasteel, Zwartland, Ecklon 688 (with narrow leaflets) ; Sevenweekspoort,

Phillips 1424; Cape Peninsula: E. & Z. 1110 (type of Rh. Plukenetiana HE. et Z.), Worsdell 4119, Ecklon

686, Scholl 750, Bowie 193, Diels 39, 85, 1315, Engler 126, 108, 125a, Wilms 3120, 3122, Wolley Dod 13938,

Bolus 3491, Marloth 11976, Lichtenstein 196 (Rh. bicolor Licht.), Drége 353, MacOwan sine No. (narrow

leaved), Rogers 11243 (narrow leaved) ; Hottentotshollandberg, Ecklon 526; Stellenbosch, Ecklon 8 and

Potts 2136 ; common in the mountains of Stellenbosch and Worcester, E. & Z. 1109; on the edge of the

forests of the Grootvadersbosch, Swellendam, Z. 2232 and EH. and Z. 1109 (type of var. swellendamensis

KE. et Z.); Plattekloof, near Fountain, Muir 482; Albertinia, Rogers 16739 (leaves only about half the

usual size, panicles very loosely flowered) ; George, Rogers 4290; edge of the Knysna forest, Jordaan ;

Melkhoutkraal, Knysna division, Keet 557; flats, Witte Els bosch, Fourcade 840; margin of forest,

Ratelsbosch, Fourcade 20; Hofmansbosch, occasional in river bush, Britten 1310; common in sandy

soil at the Storms River Pass, about 3 miles from the coast, Zahn (5050); Humansdorp, under 500 feet,

Rogers 2901; Van Stadensberg, EH. et Z. 1109 (var. uitenhagensis E. et Z.); Rust en Vrede, Oudtshoorn,

Dyer 84; Boschberg, c. 3,500 feet, MacOwan 296; hills near Grahamstown, MacOwan 12/1276, Britten

1599, Dyer 70, Daly and Sole 469; Hogsback, about 5,000 feet, Rattray 368; Buffels mountain, near

Kingwilliamstown, Tyson 1039; Fort Cunynghame, Sim 262; fairly common 18 miles from Tsolo and

23 miles from Maclear, c. 4,700 feet, Dwyer (5085); Slangapies mountains, Wakkerstroom district,

Transvaal, Burtt-Davy 1942.

61. Rh. angustifolia L. spec. 382; Thunb. Fl. Cap. ed. Schultes 263; Sonder lc. 507 ;

Engler l.c. 405; Diels l.c. 571, 592, 617, fig. 4 B, 618; Pluk. Alm. t. 219, fig. 6.

Rh. salicinum Herb. Jacq. ; Rh. argentea Mill. dict. No. 11.

In the Linnaen herbarium, London, No. 21 is called ~ angustifoliwm.” This is the

same plant which Sonder and others call Rh. angustifolia L., but No. 27 is also named

“ angustifolium ” in the younger Linnaeus’ handwriting. This is the same plant which

H. et Z. later described as Rh. stenophylla. Under Rh. angustifolium Linnaeus quotes

with his original description Burm. afr. 251, t. 91, fig. 1. This is a bad figure of Rh.

rosmarimmfolia Vahl. Though I have not been able to clear up this matter satisfactorily,

I take the species in the sense in which it was taken by Sonder and Engler.

Rh, angustifolia LL. Schlechter 5617. Under side.

Rh. angustifolia L. Schlechter 9133. (Flowers 4/1.)

102

Description: A shrub, usually not more than 4-8 feet high, often with brown

branches and with glabrous, pubescent or villous branchlets. Leaves petiolate, petioles

subterete. Leaflets coriaceous, petiolulate, narrowly elliptic or lanceolate somewhat

tapering at the base, mucronate or rarely emarginate, smooth, shining above, densely

covered with short, grey hairs below; margin entire, flat or slightly recurved; midrib

sunk above, slightly prominent below, lateral veins distinct but not raised above, often

not visible below, tertiary veins sometimes very indistinctly seen on upper surface.

Panicles terminal and lateral, longer than the leaves (male longer and denser than female),

pubescent or more rarely glabrous. Calyx segments ovate acute. Petals oblong. Drupe

densely greyish-pilose, compressed, sometimes slightly oblique.

Length of petioles rather variable, usually less than 1 cm., rarely up to 1-5 cm.

Length of terminal leaflets about 5 cm. Breadth of terminal leaflets 6-10 mm.

Lateral leaflets about four-fifths the length of the terminal leaflets and resembling them.

Petiolules of terminal leaflets 4-5 mm. long, those of the lateral ones slightly shorter.

Length of pedicels about 1 mm.; calyx segments ?-1 mm.; petals about 14 mm.

Greatest diameter of drupe 4-5 mm.

DisTRIBUTION : Common on slopes of hills (up to 2,000 feet) from Capetown to Caledon,

Malmesbury district, Cedarberge, the Pikenierskloof, Hex River, and Swellendam, often on

river banks, flowering in spring.

The species exhibits considerable differences in hairiness of young branches, petioles,

and panicles, but it cannot be satisfactorily divided into varieties or even fairly well-defined

forms. [Engler (l.c. 406) has separated a var. conerea without, however, quoting a type.]

There are also variations in the length of the petioles and in the texture and shape of the

leaflets, which, however, rarely have a tendency to form short, blunt, toothlike projections.

Whether this species should be made a variety of Rh. tomentosa is entirely a matter of

opinion. As it is circumscribed in its geographical distribution, though overlapped by

Rh. tomentosa, it is perhaps best to keep it separate, especially as its petioles are always

shorter, and even the narrow-leaved forms of Rh. tomentosa retain a more oval outline of

the leaflets.

Marloth, in Das Kapland, 1908, shows on fig. 14 B the habit of the plant near Palmiet

formation. His fig. 24, “ Macchia near Stellenbosch ”, includes it also.

Herb. Linnaeus (No. 21), London; Herb. Willd., Berlin; Cape Peninsula, Bergius, Ecklon 692,

E. et Z. 1092 (also near Stellenbosch and Worcester divisions), Z. 4832, Engler 51, Paterson 68; French

Hoek, Phillips 1086; Stellenbosch, Marloth 2822; Bainskloof, Schlechter 9133, 9157; Breede River

valley, near Bainskloof, Bolus 2745; Wellington, Marloth 11974c; Caledon, Marloth 11974 a and 6;

River-zonder-einde, Schlechter 5617; Hex River valley, Rehmann 2817, Tyson 776; banks of the Buffel-

jagdsr., Swellendam, Z. 2229; river-bed, Naidouwkloof, Pillans 5341; overhanging stream east side of

Pikeniers Pass, Pearson 5225; Hopefield, Malmesbury, Girke 1566; in stony rivulets, Cedarberge,

between Pakhuis and Groenberg, Diels 558 (panicle flexuous). Further, the following specimens in Herb.

Kew: Drége 6810a, Ecklon 689, Cooper 2168, Burchell 248, 7280, Sieber 217.

62. Rh. incisa Lf. suppl. 183; Thunb. Prodr. 52, Fl. Cap. ed. Schultes 267 ; Sonder I.c. 509 ;

Engler-l.c. 408; Diels lc. 572, 592, 594, 619, 621 and fig. 5 F, G, H, 621.

Rh. obovata Sond. |.c. 508; Engler l.c. 408; Diels 571, 592, 615 and fig. 5, H,

F, 621.

Rh. sinuata EK. et Z. Enum. 1111 (non Thunb.).

Descrietion : A much branched shrub, 3-10 feet high, with rigid spreading branches

and small puberulous branchlets. Leaves petiolate, petioles puberulous or rarely canescent,

subterete, slightly canaliculate above. Leaflets (in outline) ovate or obovate-cuneate,

rarely obtuse, mucronulate, more or less pinnatifid with obtuse lobes or only with small

triangular often obtuse teeth in the upper half or quite entire, above dark green, minutely

puberulous or velutinous, below whitish tomentose ; margin flat or (in the pinnatifid forms

often) revolute, midrib sunk on the upper surface and more or less prominent on the lower,

lateral veins not raised or even sunk on the upper surface, more or less prominent on the

103

lower. Panicles tomentose, terminal on short lateral branches bearing clusters of subsessile

flowers, generally only slightly longer than the leaves, sometimes much larger and more

luxuriant at the end of elongated branchlets. Calyx segments oblong or ovate, tomentose

on the outside ; petals oblong-ovate, tomentose on the outside. Drupe dry, subglobose,

densely villous, eventually dehiscing (always 2).

Length of petiole 3-10 mm.

Length of terminal leaflets 1-5 to 2 cm. (rarely up to 3:5 cm.). Breadth of terminal leaflets 1-1-5 cm.

Lateral leaflets generally a little over half the size of the terminal ones and resembling these.

Length of calyx lobes about 1 mm.; petals about 2 mm.

Diameter of drupe 5-7 mm.

DistrisuTion : Namaqualand, south-west Cape Province and scattered in open scrub

in the coast districts to Komgha, favouring situations near river banks and flowering from

July to October.

Rh. wncisa and Rh. obovata have been kept separate by Hcklon and Zeyher (who

wrongly applied the name Rh. sinuata to the latter), Sonder, Engler and Diels. It is true

that extreme forms seem to be quite distinct on account of the pinnatifid leaflets of the

former as contracted with the entire or subentire leaflets of Rh. obovata. As long as little

material was available, the geographical distribution seemed to confirm this view. Hven

now forms with pinnatifid leaflets are unknown to me from Riversdale eastwards. On the

other hand, plants with more or less entire leaflets occur as far as Namaqualand, e.g. in

Drége 6793 6, Schlechter 5095, 7854, EK. & Z. 1112. Marloth, in “ Das Kapland”’ (1908)

291, stated with reference to Rh. obovata, that it is common on mountainous parts near

O’okiep and Steinkopf in Namaqualand (where, as he informed me later, stock feeds on

it). The amount of dissection of the leaflets may vary considerably on the same plant.

Under the circumstances one can only consider these two so-called species as local forms

which may be distinguished as follows (but it must be understood that isolated branches

of A may be practically indistinguishable from B) :—

A. typica.—Leaflets more or less pinnatifid or nearly entire, often with the lateral veins

very prominent on the lower surface, margin usually revolute. Drupe about 5 mm. in

diameter. South-west Cape Province (extending to the macchia of the Bokkeveld) and

Namaqualand.

—

Rh. incisa L.f., A. typica. Schlechter 4991a.

Rh. incisa Lf. Schlechter 7854. Upper side.

B. obovata.—Leaflets entire or with small teeth at the apex, lateral veins rarely

prominent on the lower surface, margin flat. Drupe 6-7 mm. in diameter. From

Riversdale to Komgha. (Engler rightly remarks that in non-flowering branchlets the

leaflets are sometimes much longer than usual, more dentate, and the petioles are longer.)

104

Rh. incisa L.f., B. obovata. Britten 1532. Upper side.

A. Herb. Thunberg; Clanwilliam, on the Olifants River and near Brakfontein, Drége; Olifants

River, 500-1,000 feet, Schlechter 4991, 8720; Clanwilliam district, Bachmann 711, Schlechter 8720,

Diels 279, E. & Z. 1112, Marloth 2618, MacOwan 3311; Namaqualand, Marloth 6715, 6767 (p. pte.),

11153; Bolus 6527, Scully 42, Pearson 6530; mountains above Worcester, Rehmann 2518; between

Paarlberg and Paardeberg, Drege 6793b; Hex River mountains, Drége 277 (wrongly issued as Rh.

tomentosa L.); Saron, Schlechter 7854; Simonstown, 1,000-2,000 feet, E. et Z. (this locality requires

confirmation).

B. Drége 6794aa, Burchell 2194, 3011, 3332, 3499, 3824, 4802 in Herb. Kew; banks of Zoutmelks

River, Riversdale, 600 feet, Muir 2473; Gauritz River, Gamtoos River, and Assegaibosch (Albany),

H. et Z. 1111; Assegaibosch (Humansdorp), Rogers 2841; between Gamtoos and Kromme Rivers,

118 miles north-west of Humansdorp, Fourcade 413; flats east of Kabeljauws River, near mouth,

Foureade 721; scrub near Gamtoos River, 200 feet, Schlechter 6039; Port Elizabeth, 100 feet, Tyson

2261; mountain sides near Graaff-Reinet, 4,000 feet, Bolus 625; Alicedale, Cruden 40, 78; hills near

Bushman and Kareiga River, Z. 2240; between Port Alfred and the Kasouga, not far from the Kasouga

mouth, Britten 2260; near Grahamstown, 1,200-—2,300 feet, Schonland 574, 3293, MacOwan 34, 164,

Britten 1532, 1597, Daly and Sole 303 ; Buffalo mountain, near Kingwilliamstown, 3,000 feet, Tyson 1038 ;

British Kafiraria, Cooper 42, 421; woods and shrubby places, Prospect farm, near Komgha, Flanagan 292

(in these specimens the leaves are a little longer and relatively narrower than usual and often shortly

mucronulate).

63. Rh. rosmarinifolia Vahl Symb. III, 30; Willd. sp. I, 1484; Thunb. Fl. Cap. ed.

Schultes 262; EH. et Z. in Enum. 1088; Sonder l.c. 506; Engler lc. 404; Diels

lic. 570, 592, 617, fig. 4 H, 618; Burm. Afr. 251, t. 91, fig. 1.

Rh. stenophylla K. et Z. in Enum. 1094; Sonder lc. 509; Engler lc. 404;

Diels l.c. 571, 592, 594, 617, fig. 4 C, 618.

Rh, lawandulaefolia Pres! Bot. Bem. 42.

Rh. angustifolium in Herb. Linnaeus (non Linn.).

Rh. macrocarye Fingl. 1.c.. 449.

Description : A shrub, 2-4 feet high, with virgate branches and glabrous or puberulous

branchlets. Leaves shortly petiolate. Leaflets straight or curved, narrowly linear or

linear lanceolate ; acute on both ends, sometimes mucronate, glabrous and shining above,

whitish-tomentose below, with the exception of the midrib ; margin revolute, entire or with

one or more sharp teeth; midrib sunk on the upper surface, prominent below, lateral

veins deeply sunk on uppor' surface, rarely visible on lower, tertiary veins not visible.

Panicles lateral and terminal, the latter generally longer than the leaves, glabrous,

sparsely pilose or pubescent, multiflowere1, bracts subulate. flowers shortly pedicellate.

Calyx segments ovate acute, subacute or obtuse, sometimes pilose, petals oblong. Drupe

large, subglobose, a little compressed, usually whitish or ferruginous tomentose, rarely

glabrous.

105

Rh. rosmarinifolia Vahl. E. and Z%. 1088 (flowers enlarged).

106

Length of petioles generally 5-8 mm.

Length of terminal leaflets 2-5 cm. long, rarely longer.

Breadth of terminal leaflets 1-2 mm., in var. B. up to 5 mm. or even more.

Lateral leaflets generally about four-fifths the size of terminal leaflets and very much like them.

Length of calyx segments about #-1 mm.; petals about 2-2} mm.

Greatest diameter of drupe 8-10 mm.

Distrisution : This species, the smallest of the south-west species of Rhus, is almost

always present on hill-heathland, and extends under similar conditions to Port Elizabeth.

(Drége, im Herb. 8.A. Mus., is labelled “inter Shiloh et Los Tafelberg Dec. 47”. This

locality is, without doubt, wrong.)

Flowers mainly in winter.

Diels (lc. 618) pomts out that there are only artificial limits drawn between RA.

rosmarinifolia Vahl and Rh. stenophylla K. et Z. The latter is only a broader leaved variety

of the former, as was already suggested by Sonder (l.c. 507). Ecklon and Zeyher in Enum.

1088, divide Rh. rosmarinifolia into the followimg four varieties. They evidently did

not realize that their No. 1094, which they described as Rh. stenophylla, is very closely allied

to it, and placed it between Rh. rufescens HK. et Z. [—=Rh. discolor], and the plant which they

wrongly called Rh. tridactyla Burch. [=Rh. eckloniana Sond.].

a capensis.—Leaflets linear, sessile, very long. Amongst shrubs, north and east side

of Table Mountain (alt. If), Hottentots Holland Mountain, flowering May, June.

B witenhagensis.—Leaflets linear, petiolate, very long. In stony places (alt. IT) mountain

sides of the Zuurberg and Van Stadens (Uitenhage), flowermg July.

y caledonica.—Leaflets petioled, linear, very short. In stony places (alt. IV) mountain

sides on the Zwarte- and Baviaansberg, near Genadendal (Caledon), flowering June.

0 swellendamensis.—Leaflets shorter, linear-lanceolate. Mountain sides along the

Rivier-zonder-einde (alt. II), Swellendam, flowering May. They add: “ perhaps a distinct

species ”.

I propose to divide the species into three varieties :—

A. typica.—Leaflets narrow linear, acute or mucronate, panicles glabrous.

B. stenophylla.—Leaflets linear-lanceolate or oblanceolate, acuminate, mucronate, up

to about 7 mm. broad (in the type, EH. et Z. 1094, some are 5 mm. broad and 4 cm. long,

others on the same branch are only 14 mm. broad).

=s>

bey

aa)

I pri

ALN

—-s

Rh. rosmarinifolia Vahl., B. stenophylla. 6/2. Paterson 2269. Upper side.

107

C. brevifolia.—tLeaflets elliptico-lanceolate, acute or paucidentate at the top, c. 15 mm.

long, 2-5-4 mm. broad.

A sharp distinction between these varieties is not possible, and it may be that some

relatively broad leaved specimens are hybrids between Rh. rosmarinifolia and Rh. angusti-

folia L.

Schlechter 7872 (Saron, 600 feet, flowermg June) was looked upon by the collector

as a hybrid between Rh. rosmarinifolia and Rh. dissecta.

A. typica.—Herb, Thunberg fol.a@; Herb. Willdenow; on the sides of Table Mountain and Hottentots

Holland Mountains (var. capensis EH. et Z.); other specimens from the Cape Peninsula: Ecklon 1,

Schlechter 769, 1227, Z. 4279, Wolley Dod 1095, 2501, Cooper 2492, Pappe, Prior, Ecklon 6099 ; Piquetberg,

Drége 6812; Dal Josaphat, Tyson 847; stony places in the third altitude on the Zwartebergen and the Baviaans

River, near Genadendal, H, et Z. 1088 (var. caledonica EH. et Z.); stony places on the Van Stadens

mountain, Z. 2227 (this shades into B as do also the following:) Port Elizabeth golf course, Cruden 417,

Kemsley 167; near Redhouse, Paterson 2269; Burchell 451, 873, 943, 6374, 8374, 8535 in Herb. Kew.

Rh. lavandulaefolia Pres] Bot. Bem. 42 (in Herb. Berol.) is a form with long narrow leaflets which

are strongly curved. Rehmann 1346, from Stinkwater, Capetown, must be referred to it. The leaflets

reach a length of 5-6 em., while they are barely 1 mm. broad.

B. stenophylla.—Herb. Thunberg fol. 8 and y; Herb. Linneaus No. 27; stony places east side of

Table Mountain, H. et Z. 1094 (type of Rh. stenophylla EK. et Z.); Lions Head, Wilms 3127; above the

blockhouse, Capetown, Wolley Dod 36; Prince Alfreds Pass, 2,400 feet, Fourcade 1281; stony places

at Van Stadens mountain, MacOwan 1030; Port Elizabeth, J. L. Drége; Burchell 4470, 4528, 6766,

Z. 331 in Herb. Kew.

In Sieber 216 (Cape Peninsula ?) the breadth of leaves is extraordinarily variable, one is 1 cm. broad.

Some specimens in Herb. §8.A. Mus. distributed by Ecklon and Zeyher under 1094 as Rh. stenophylla

0 swellendamensis have rather long petioles (up to 1 cm.), and the shape and size of the leaves varies also

considerably.

C. brevifolia.—Mountain sides near Rivier-zonder-einde, Swellendam, EK. and Z. sine No. in Herb.

S.A. Mus. (var. swellendamensis K. et Z.); Caledon, ? in Herb. S.A. Mus.

Rh. macrocarpa Hngl. l.c. 449; Diels lc. 591, 592, 517, fig. 40.

This is only known from Burchell 6756, 6758, collected on the Zoutmelksrivier, near

Riversdale. In the Kew Herbarium it was placed by Sonder under Rh. stenophylla EH. & Z.

It is only distinguished from Rh. rosmarinifolia Vahl C brevifolia by perfectly glabrous

drupes, but already Diels (l.c. 592) has poited out that the haircovering of the fruit in

Rh. rosmarinifolia is not constant. He found on a specimen collected by Mundt and Maire

(on the Cape Peninsula?) that the young carpels and consequently the fruitwall were

perfectly glabrous. The absence of a haircovering on the fruit alone can, therefore, be

scarcely looked upon as a specific difference.

POPULIFOLIA group.

64. Rh. populifolia HK. Mey. in Drége exsicc.; Sonder I.c. 508; Hnglerl|.c. 407; Diels l.c. 571,

592, 622.

Rh. Steingroeveri Engl. Bot. Jahrb. XXIV, 500; Diels J.c. 572, 592, 621, fig. 5 A.

Description : A much branched shrub, 5-10 feet high, with glabrous or puberulous

branchlets. Leaves petiolate, petioles slightly furrowed above and slightly winged.

Leaflets subcoriaceous, often very shortly petiolulate, rhombeo-suborbiculate, triangulate

or subovate, cuneate at the base, obtuse at the apex, dark green, subglabrous above,

greyish or greyish-red tomentose below; margin flat, usually crenato-dentate or entire ;

midrib and lateral veins yellowish and not prominent (or even sunk) above, prominent

below, tertiary ves few and distinct below. Panicles puberulous, lateral and terminal,

shorter than the leaves with pedicelled flowers. Calyx segments ovate subacute, petals

oblong. Drupe tricuspidate, obliquely depressed and compressed, sometimes slightly

verrucose, puberulous or glabrous when ripe.

Length of petiole about 1 cm.

Length of terminal leaflets usually 1-3-3-5 cm. Breadth of terminal leaflets 2-3 cm.

‘Lateral leaflets about two-thirds the size of the terminal leaflets.

Length of calyx segments {|-} mm.; petals about 1-5 mm.

Greatest diameter of fruit 5-6 mm.

108

Distrisution : Namaqualand, Bushmanland, South-West Protectorate.

Already Diels (1.c. 572) has stated that Rh. Stengroeveri Engl. can only be distinguished

from Rh. populifolia KH. Mey. by smaller size of the leaflets and sometimes by the shape

of the leaflets. I do not see any reason, therefore, to keep up Engler’s species. The

leaflets sometimes approach those of the form of Rh. incisa, which is known as Rh. obovatu

Sond. Sonder says that the fruit is tipped by the styles, but it is more than that. The

styles enlarge slightly and form small spines as in a few other South African species of

Rhus (the fruits distributed with Pearson 4465 and 4695 do not belong here and not even

to the genus Rhus). Flowers June, July.

Rh. populifolia BE. Mey. ; forma. | 6 ee

Schlechter 11438.

Under side.

Rh. populifolia BE. Mey. Phillips 8963.

Mouth of Orange River, Drége (in Herb. Kew) ; rocky places on the Orange River, first alt., Droge

3032; Aus, Steingroever 56, 51 (in Herb. Berol.) (type of Rh. Steingroevert Engl.) ; amongst rocks near

spring at Karibis, Warmbad district, 900 m., Marloth 4793; hills near Wolveton, Schlechter 11438 (too

109

Rh. dissecta Thunb. Mader.

Upper side.

Rh. dissecta Thunb., A. obovata. female. Tyson 635 (female).

(Flowers 4/1.)

110

Rh. dissecta Thunb., A. obovata. Tyson 635 (male, flowers enlarged).

111

scrappy); Steinkopf, Namaqualand, Marloth 6767 (p. pte.) ; Namaqualand and Bushmanland, between

Steinkopf and the Orange River, Phillips 1568; Bushmanland, 2,700 feet, Pearson 4695; sand at foot

of Gesellschaftsbank, 2,700 feet, forming thickets, Pearson 4695; Great Namaqualand, Pearson 4017,

5263, 4465, 6164; South-West Protectorate (in Herb. Berol.), in the coast desert and on hills up to 3,000

feet : Hermann 20, Range 416, 833 (in one branch leaflets narrowly ovate up to 5 cm. long and 3-2 cm.

broad), 1080, 1171, Schulze 100.

65. Rh. dissecta Thunb. Fl. Cap., ed. Schultes 267; Sonder l.c. 509; Engler l.c., Diels

lic. 572, 592, 622, fig. 5 B, C, D.

Rh. argentea HK. et Z. in Enum. 1127.

Description: A shrub, 5-6 feet high, with terete, glabrous, reddish brown, sometimes

filiform branchlets. Leaves petiolate, petioles often reddish, slender, subterete. Leaflets

coriaceous, glabrous and dark green above, greyish tomentose (except on midrib and veins)

below, more or less obovate in upper part, strongly contracted and narrowly obcuneate

in lower part ; margin slightly thickened and often slightly revolute, entire in lower part,

sharply cut in upper part with teeth generally triangular, mucronate, or leaflets pmnatifid ;

midrib and lateral veins yellowish and barely raised on the upper surface, pale reddish

and prominent on the lower, tertiary veins not visible. Panicles glabrous, loose and few-

flowered, terminal on short lateral branches, bracts subulate, acute, flowers pedicelled.

Calyx segments oblong-triangular, obtuse, petals oblong. Drupe glabrous, oblique,

depressed and compressed, shining, more or less verrucose, tricuspidate.

Length of petioles 1-2-2-4 cm.

Length of terminal leaflets 8-1:5 cm. Breadth of terminal leaflets 5-12 mm.

Lateral leaflets about two-thirds the size of the terminal leaflets.

Length of pedicels 1-24 mm.; calyx segments }-1 mm.; petals 1?-2} mm.

Greatest diameter of drupe 6-8 mm.

Disrripution: From north of the Olifants River through the Clanwilliam district,

Malmesbury district, Hex River valley. Flowers in winter and spring.

This species is nearest to Rh. populifolia HK. Mey. as shown by its drupe. The venation

of leaves in both is also very similar. Sonder (l.c. 509) distinguishes two varieties, to

which the following names may be given, though a sharp line cannot be drawn between

them :—

A. obovata.—‘ Leaflets obovate or suborbiculate, cuneate, incised dentate” (Rh.

argentea HK. et Z. 1127; Rh. dissecta EK. Mey. in Drége exsicc.).

B. pinnatifida.— Leaflets pinnatifid, lobes lanceolate, acute’ (Rh. dissecta Thunb.,

E. et Z. 1128; EH. Mey. a and f).

A. obovata.—Burchell 992, 7460, in Herb. Kew ; Giitberg, common, 2,000 feet, Phillips 7529; banks

of Olifants River, 400 feet, Schlechter 5027a, Stephens 7123; Clanwilliam, Mader, Schlechter 1637, Diels

280 ; stony places above Wupperthal, MacOwan 3273 ; swampy ground, mountain sides near Brakfontein,

Clanwilliam, EH. & Z. 1127 (“ Corynthebosses ”’) (this is the type of Rh. argentea K. et Z. and includes their

var. brevifolia, in which the leaflets are a little smaller and not so sharply cut as in the type); between

Pikenierskloof and Clanwilliam, Z. 232; Cedarberge, near Groenberg, c. 2,500 feet, Diels 860 ; Piquetberg,

Edwards 193, Bolus 13530 ; near Groenekloof, Bolus 4265; Kradouw Krantz, Pillans 5,300; Klitzkop,

near Darling, Bachmann 608, 1564; Moorreesburg, Bachmann 1562; Waterboerskraal, Malmesbury,

Bachmann 1563; hills near Malmesbury, Schlechter 167 ; near Hopefield, Bachmann 1802; Hex River

valley, at the foot of the mountains, Engler 273, Rehmann 2821, Tyson 635.

B. pinnatifida.—Herb. Thunberg ; Groenekloof, Pappe 10; Nieuwekloof, 1,500 feet, Schlechter 7504 ;

Saron, 1,000 feet, Schlechter 7861; ib., 600 feet, Schlechter 7872 (Schlechter has marked this ‘‘ Rh.

rosmarimifolia KX dissecta’’).

66. Rh. cuneifolia Thunb. Prodr. 52, Fl. Cap. ed. Schultes 267; Sonder I.c. 512; Engler

Le. 419; Diels l.c. 576, 637, fig. 8 A, B, C.

Description: An erect shrub with terete or slightly angulate, sometimes purplish,

minutely puberulous branchlets. Leaves shortly petiolate or subsessile. Leaflets coriaceous,

glabrous, obcuneate, obovate-cuneate or subrhomboid, margin thickened and slightly

112

WA at af

i 4 @ PO. sOVKS

i IRPOP

i LAA Ze

I+ 7} Artes

Rh. cuneifolia Thunb. Schlechter 4824.

Under side.

Rh. cuneifolia Thunb. Herb. MacOwan 1795, (Flowers enlarged.)

115

revolute, entire except in upper part which is grossly dentate, teeth acute or acuminate,

sometimes mucronulate ; midrib and lateral veins slightly prominent on the upper surface,

more so on the lower, tertiary veins not visible. Panicles axillary and terminal, pilose,

the female about as long as the leaves, the male much longer, rather loose, flowers pedicelled.

Calyx segments ovate, subacute, petals oblong. Drupe obliquely ovoid, depressed and

compressed, shining, tricuspidate.

Length of petioles $-3 mm.

Length of terminal leaflets 1-2-1:5 cm. Breadth of terminal leaflets 8-10 mm.

Lateral leaflets one-half to two-thirds the length of the terminal leaflets and resembling them, but often

asymmetrical.

Length of pedicels about 1 mm.; calyx segments barely 4 mm.; petals about 1} mm.

Greatest diameter of drupe about 6 mm.

Distrisution: Caledon district, Stellenbosch district, and Clanwilliam, in rocky

situations. Marloth (Das Kapland, p. 261) also quotes it from the Guarri formation,

Montagu, Oudtshoorn, Ladismith, north side of the Zwartebergen. Flowers in spring.

Diels (l.c. 577) says that there are transitions between this and Rh. scytophylla EK. & Z.,

but this applies only to the shape of the leaflets. I am inclined to place it close to Rh.

dissecta Thunb., with which it agrees in the fruit and the venation of the leaflets.

Herb. Thunberg ; Burchell 8197, 8244, 8288, Pearson 7782 in Herb. Kew; Zwarteberg, Caledon,

Ecklon 37 (Rh. africana Ecklon non Mill.) ; Hottentotshollandsbergen, near Grietjesgat and Steenbrass

River (Stellenbosch) in Langehoogde, and near Bontjeskraal (Caledon), EH. and Z. 1131; near Caledon

warmbaths, 1,000 feet, MacOwan 2795 ; Howhoek, c. 1,500 feet, Bolus 9191 ; Sir Lowry’s Pass, Schlechter

4824; near Clanwilliam, Leipoldt in Herb. MacOwan 2795.

INDEX.

(Names of species marked with an asterisk have been retained by the author.)

PAGE PAGR

Ravus L. Ruus L. (continued).

Aone, IDC ccoocagausnoncbobo0d5cnboo0 dbl +cammosulanschonilamreimrirrriier 6h) 2S 44 253

acuminata Bem Mey ntl eeieiee acl eee OL Chnyeimlllsstl, IDXCe sn noobeoboodooogoocodéulooas 54

ACuUbIdenswE NS ler re deieiyeieeiertstn ton £05) celastroides’ Sond... .. 6.0... 0c 13, 62, 63, 64

dequaliswPerse ager tied a aa.s tosis lake wales eee oe Giles, IME bo cenooceuLocKd 15, 16, 82, 83, 89

—Enancepia), Wolo soapop oo bo eoogodomouedo po 14, 58 Ghadadpakoerren Wow ooadccancoacbccvdecvuues 4

Ehovlorhiey 1 Delle qo cincdd daa agcn od noes cae 113 colensoanas Helis ston see trite 50

aglaeophylla H. et Z....................62, 64 *commiphoroides Engl. et Gilg........14, 71, 72

alatumeihun Darden eciree erie 4 CONCINn UMP BUTCHER ere terrence te 82

Falbomarginmatasondss soe rcrder iii tls 14, 59 concolormbresl wri aceite tae aeeie 4

amiboensisy Schinze-stim ceria ae acl NZ Conlaceap lin oleae eeccerererer 90, 91

yxangustifolia line eee eek 17, 98, 100-102, 107 Clrassinenyla Eres seaeret eerie,

angustifolia ? L. (Herb. Thunb.)............ Gomet crenata numb errerriiie rnc 5, 8, 12, 47, 69, 71

angustifolia (L. a.) EH. Mey.............. 31, 46 crenulatapA Riche ecuec)ice cine eee 68

anpustifoliumy Herbs, Wimney cleeieisls sills 104 zc CHISMA MLL ORVier weaker arnelataste 15, 80, 81

EM NdloomsansasoGodohooo dH oooecbCG 92 cuneata, Ne Hh Breciact Acts cine sate nnyetlet 48

Enz, Nihil noo opoabeooondncdnosgodonso9 100) *cuneifolia hunbi nasa 13, 17, 58, 111-113

Heh ID Cb Asongansacousenoaconbsooce 111 cuneifoliaghine Men Tener eieer coo

ena aiylia, IBS conansccodcbaoecnooanandS 4

Monee VKH osaceoounandonceb 06 19, 21, 22

j *dentata Thunb....3, 5, 8, 9, 12, 37, 38, 39, 42,

53, 85

“een SelaOall, nconcacasesuocacodcn il, 84, 3383 Gentuclatamlichitwermerrneeererrom erent et Coan:

DiCOlonPlLACH ta vier. cepa ceheterecrn eds 97, 98, 100 denudatawhweteZir-mrreracticrccttcslerccislereieere 74

AIR hnisil Sfoyavel ncsocbo pb benpos boMaaDdar ou 15, 85 lier INN) coc cocooecodcoodnnonnonese 4

BurchellimtSondeerciciaeleiee ane 62, 63, 64 dimuidratumieihunibseeeeeeeenitei en cieniee ee 4

loaning, StomGlsooqooondenancouegaaen 16, 90, 91 IDyioern ob, Soap ocansonaccuabe 12, 46, 47, 72

Burman ai) Cie eee «sieve tacts Mere etehelators 20m GiscolonsSchradhasaceieciiicie eee eke iekererorniee 97

114

PAGE PAGE

Ruus L. (continued). Ruvus L. (continued).

*discolor EK. Mey.......6, 16, 78, 91-95, 97, 106 Flanceaelustareemear a ate simneamaae o, 6, 8, 15, 72

dispar Presl........ Pdcaguosgooscondodansus 4 lan cea SDest reer em acres vets teeter octets 61

~OhEEG Ora, AMIN Gcoognonoeuobe 17, 107, 109-111 layvandulaetoliapbreslaeeerere ener tce 104, 107

“tobivenuccenias 10 Gti Anosoocconeueco00d 12, 49, 67 *Legati Schonl..4, 5, 6, 7, 8, 9, 13, 28, 29, 51, 52

Soham) SOV socoososcoogesoososnoa0s 15, = leptodictiyamDiel Striientieteraeeeieiciiee 80

dunensis! Gam@iriinisccrscre isis) aos Cereacioereroerele lobatapPoirs-escciiecnacttac semen ee Sie ee

* ray SChOn Ss peralesiete/s cieieiets #) slostorte reser ll, 35 longitohanSonderenncnemcn reenter aeons

poner Pina E. et Z, .3, 6, 8, 9, 13, iG, 82, '88, 89

AlmeGle, Wis. 55'S O, WB, WG) My, Bek, Bis, Br, BS,

Feburmeayschonlsssemeaer seein Cer re 14, 67 59, 60, 61, 89

Fecklonianay Souder eteileliietry 14, 65, 68, 106 lucidaglg Meyercetsaset ie ecco reece 61

ecklomanameresleae ee eneeoLencn erento. 21 lucidumipAi tt ee sete ee ee 21

Mekdonia Scbradjyeytorteresste sere fercierereyereceerorretare 97

ellipticumyy nin beret rere 97, 98

Clonee, VEE l6 a conscspsscecoono0d0s 18, 19, 21 *MacOwani Schonl...3, 7, 8, 9, 11, 18, 19, 24,

Cloraseiin, IB Gi Zocacnesccaceccsgooco0cccn0 74 25, 29, 31, 36, 45

“ADhaalleya, IH os bcos ooacKoN OC 8, 14, 18, 66, 71 MacCrocal paw hn clei eer errr tat se iia 104

RE TMestiaschonleseeninn care eetn Do Os Il, Be} *magalismontana Sond............ 5, 16, 90, 91

“ROSY, AMANO. .cacgaacooounnooecos0 8, 15, 84, 85 margaretae Burtt-Davy.............. 65, 66, 68

excisum Thunb.........- 46, 59, 62, 63, 64, 65 2 Manlothiig@hnel eee eee ele 14, 67, 69, 71

Meyerlanapbresla pete ereiiritiiecaiecrc Bils 337

norma, 13, Ch Zroccctascgsves0cceodne 62, 63

AweineaaNEN 1D, Cb Zoocososacooccxsscac 12, 36, 45 TTMNOMOGIGIE) SOM; os0sn00s0008000ec0000 15, 30

iOS Nea co GondoooboeDoooDSO DS OHBDH 86 WAS IDS WE coscabosancadoondsaococgedae 38

flexuosa "Dielsen. sere rece eee 29, 30 MON tara 1elsemer sin wee ate ee en ene ate 6, 78

TORUCE, IIH; DACs cascqososvso0ccs0nc00~ 21 ~MuUcronataeLhun bees OMS One lemme alias

memes ILICMsoscooooo poco Coe RBdONoCOOFEC 74 18, 19, 20, 21, 23, 37, 50, 54, 65

cAngcml Solo@mlls 65 gaccosdecccss 4 Ve 26) 27, 28 MaAMOKOMME) 1B, WocatococnseddaunedoduGeC 62

Tu ESCoMS IVE, cocooa goon dn ndoOCDOOOOAGUOL 50 TAROOMAE LW Gi Zoran oooedoacoodsenaseocoss 58

VOTO MUNONE, ONC! Goccccpanccosccsuccenges 5

NAPUS IIVNO35 oa odoccduveccoscooCNnTeN 4

Pelborat WINE oss cop nono o Dodo b oR DOO OOOES 62, 64

Cllorbonel SolstbavA,. 5 o cbcosoconnccscnece 37, 39, 40

*Gerrardi) Hater stele) 15: (0) Gide en atalensismpern heer ee 9, 14, 68, 70, 71, 72

Sol aucamDestemmrieeirten 13, 50, 55, 57, 61, 62, 64 *nebulosaechonlcenen erences @, til, BB St!

elbrn@esteais JN, INC! 5 550565000095 90c0008% 9, 68 MEINVOSE) 15 GW Zocossccccccda000on000000 2, 64

SL AUCESCENS MS LIM eee ett teen 69 FAMISMOPIAIISS Selaralll oon sacocanenaceuoes 12, 44

glaucovamens Hao rere liteperieryetert titel t 54

-ancoilbiene, Ides coosoobocoooseuo0c ac 6, 15, 85

enandidensiHlaty lec) erect 40 Oblanceolatam Schinzarm neritic 90

Fapacbiolhicy IBhadll.coeccoaao acne oho 91, 93, 95 Ol KG IIE WAGMO. creansedancodeaeocnneceous 4

CEMA SOW. >scosend0c 15, 67, 69, 74, 79, 80 Oly oveli, Nolo soaasaogdanancovoc 102, 103, 104

OMIT Whe I WR sosancccccgde0n9000%000 72

OQUUMAMOMAAENS Wc ongdossadouco0ccuce 54, 55, 56

intapblanvaerlocerteryieieieietiek keer eet 26 Omyacant Mam Cavite iter yer tiki iserriaers :..85

Khorridiash. eb Liisi aciceitere cin cierene re 13, 16, 88

InpeeovIbs} 1D, ih Aooococogoccqcacsc00 29960 45, 46

jeans IB, Gh Aacassoonaonugepovcanse 62, 64, 65

jonllliiols, 13, W\Voobocdsqdacbnogouccgoponddoo 74

impermeabilis Dint. Ms.................... 46 PHIONOBMISE! SOM coooduconusdovascosSness0C6 5

maven, [Bail coo oe oupadoaDoO DONO ON OOOsE 18, 66 AMO, Onl ococossoonogong0eecK000s 37, 38

incanum Mill... .4, 18, 19, 22, 23, 24, 31, 67 jPRMUC MT OVADUE ANAM), > 550g00000sG0000000cn00 4

AGE, Win onoboconDK0 DOO DO aac 8, 9, 17, 102-104 jooaclwiiing, 136 G8 Ascoocaccsovcg0casao00 20, 21

shayermeataolke, SAdoyallagoosoadonascosasbocac 11, 28 pendiulimad ACO eteremerietnteveleneetneLaterersteere 74, To

#Pentheri :Zahl bityecve-letersciertele sorte ett 12, 48

jolyoras} LENCE oops poupoddooudaenedobKoveD bn 21

“AEH SINS Als sss oogdoodcocqouDdsoD 6, 15, oh Egle, IE Wyo ocovooodnodccospo0os00as 88

oayySaMleVeR), SOMA ooo 4 obo boo. aie save alee mes eeone TORN, An ohn coodnogonusocbdonoovouseoe 61

Aalosenecl Wists 6 ooboano oD banouc 11, 33, 34, 56 jObenCrollia, 1BL GH Zoccaogonocedcsbg00000 62, 64

HWE DONBISIONG) ss aT sietetersrerche sconce tekalenerenere 7) Plukenetiania Be 6b Zeeecns © ccc ees 4, 97, 100

polyneura Engl. et Gilg... ee se 36

*pondoensis Schonl.....:............. 16, 95, 96

La@Vagaibay Wir -rciatet-t telonstclovenel orcnaveteretsy = 4, 17, 19, 21 *populifolia E. Mey....17, 71, 90, 107, 108, 111

laevigata Thunb............- 4, 5, 8, 41, 42, 51 puberulay Hi web Ze viele nterlelrer 31, 32, 45, 54, 82

lbvexatepnia) lle, WEkelehacocooogncacagadenouos 74. jvorateevols| Ma, 5 ee ocoosboeo buds Dc 18, 22, 29

laevigata, Hy Meya-yririrains ehisetriy recat: 66 pubescens) EH. et Zisnecssveweeessens 19, 24, 32

PAGE PAGE

Ruvs L. (continued). Ruavs L. (continued).

pulbescenspHerbisberolyeeeare rennin ir 22 tridachylen BUNCH pm ceerirr ner reli see rs 82

*pyroides Burch...6, 7, 8, 11, 24, 29, 30, 32, 46 Wu mienie, (oC ooooagdoconseccnodedGcds 4, 26

53, 67, 80 Mukai WMG. ac5ocosgaaonnodacooer 4, 26

TAHOE Ieloae Shas We socosossaconcseaon 49 (Hub, IPI toc cin ondsaonogseouas000000 83

CAA, Sao cconascoggsaccac0500Na0n 37 38

tsemulbensism Dinter cmeeeetiioecececieeeiic 71

Nema EIN, IMil, oon poogdeeHoansooee 62, 63, 64 bumulicolam Soares iiss

Rehmanniana Engl.............. 25, 26, 36, 72 Ry sonimehillips eperamrcer tte melee ae ors 50

iano, 195 Gi Asoooacascogucuscosd oo 5, 12, 45

HEWMIMAO MOC, HAO ccocooscsubacvgdocaue 15, 75

ri 1c veal iy Stereterska se cisieiee eicietecve etere ats 16, 89 “undiuilatard cq areca 8, 9, 13, 61, 62, 63

WMO, ID CU Acsobongoccoabdbos uo oddeod 89, 90 5

RAMOS Solaoyalll 5 sans ccaoncncduesoan 12, 42, 43

*rosmarinifolia, Vahlls.... 2... 6, 17, 100, 104, 107 Vern catamschlechteeaneeieer iain 62, 63

-ARWCEIHEN Dial, copoodo oo Ged doe Ooo Odes 11, 36 willosayauct-p plutei ter vcracracyeiee cise es 25

TESCEMS, IDS Gh Accoooovoccoosdadounac 91, 106 villosa L. £..9, 17, 18, 19, 20, 22, 23, 30, 31, 75

*rupicola Wood) et) Hivans))-)-))111)4)- 1-01 12, 50 willosissim/aywEng ley elite rel stele 91, 92, 95

VALI VEC so occotas0asn000c0b0edD AG 72, 74

*viminalis Vahl........ 5, 6, 8, 15, 74,75, 76, 77

SallicitoliaMP resi |-ststoxe chefs sakes sista cre reo sient 5

Saliciman Sondiprescie raat ice eotete so esi crees ate 5

Salicimmmplerbwacden asic acres 100 We Wmiselautl WA oo os ocoadoaccHdoocuo noe 67

Saocrinkyachi! Da, obo oocoooecscuugoonduacd 65 WaldinoniDehnlh race ttre rer cere r 74

e Schlechter Dielseeeerarre ee Dom DOMOUEOS =Walmsii@ Diels. aceasta cocks eee es 6, 15, 86, 88

Reojcnnweh Wh CH Aoccoocosacogcous 16, 54, 55, 57

*scytophylla EH. et Z...2........ 13, 17, 60, 113

SEIgKCEE) 1p GW Ab oly camco is Dome cabo Abed 31, 45 Thor SonGlageaccocoocoomnousonod 13, 50, 54

Sencopuyllamschlechtesp perimeter: 29, 30

SerractOliam UL hermmtn yamine icrieisiec rae er 84

“Huai ITM os oorcoccananepacoouoe 14, 69, 71 The following genera have been mentioned in the

Giang, 13, Gh Aooasascesucocoqsusosusaoas 102 text :—

eaeeN WN coonocooosesougbaodeObUe AY) AXONOWN, WX coo comb nbooacgosa0eogc 67, 74, 80

ouch IsWele Gob wasedboooseossuboer &, Bi, co AayAareamionnom® Ib We Wooongoocdncdobeadsauenoe 5

Sonnimleie 1 Gs “os cooccgvaccpnsg6cuocusn G2 CIR SWE, Wecooodonodadocnddsc0KD cob oueOaGDEOS 68

OCHAIIN INNS cocnsscnonassooncaonndeenDe Zh Oia. Mig ooo pnopnccanpoonuerouBoaDegonK 68

Gomes IDG, co soaodancadodcabac008 IA) Coutinsinon? Woockooshoousec bopbosooSGoCUE EE 80

Siren! INVl oageqoscsnncoccouaa Ny OP. CORRGAN IDES Hooloc bo ooo ocoopbodegoooUREOCCeS 7

stenophylla E. et Z......... TOO), LOS, 108; UO. WismMWN Inco ooccdevoco Doda obcconaos cabo node 67

Ssuccedane are lair acti errr arises cr Bil Doo, Whi So owaootausoobnoooooadouUDouas 67

Subfemmucinatam eres mmr reeriair rete eieistere AGS SE mER TAM Veisniye teeter yet cartatnenct ace) ect het oe 4,5

Ishaanomnonmog) 1, GU Ad ccobosodsooopaacnoadecc 4

IMivJARIS PETES eae) jepcrcqe eter oteter taetetarottones ekel sted oterel- 4

cenuiflorawibres lacy itvalenseaclsrtame rcberavve tesa Mil WARIO SS aglncabaoqnecasuabadaonanaougD0g00 47

(hain semeieioe, 18, Gi Sasoccoccooqgugn0c00Kc Gil OWA Tis o.oo Slate ko SOO nT Oreo CORRODE ee oC 74

huniberoinplookamt-loer vier shel ioral erate le AL PEON WOR pocoocencacensoans000Cbad 80

tomentosa lnm eso AL Gs 3, Os ily, Os, Ow 1inoonsing), Winvells Seco sovesagdeouoosdoouLeacs 5

iomemesmiad Will, A coganacoeboondoabbe Good Wp) IEROTWATINA WeagoecadsonsadooodanDoubaDbdoooOS 7

Piransyaalensismbog lies stelye suse tele) oi 18, G3) ITMOWessoS JARI, sondoouedoarpoobooos udodor +

WHOS) IDE Wie\aooboce ogmaaduCoeseousoon EY, OO SAmasiWS Wo onevapecodgéonceocn0uomodcUboduT 4

imidachylagsondemebmalctalem annie reee S82 SCHMIDRLTA Tae ycpaeiceciccucll sneredsierstone ier ietejoka 4

WNGE CIE IDC Wie oodaooudansouce 26) OF, LOG SCuNantiopaaiony Ih, oksccoccscossbodsooupucuGdD 3

116

The Genus Psoralea Linn,

By HELENA M. L. FORBES, Natal Herbarium, Durban.

The South African species of the genus Psoralea were last monographed by Harvey

in Harvey and Sonder’s “ Flora Capensis” (Vol. II) and since then much fresh material

has accumulated in the South African Herbaria. As the genus is a fairly typical repre-

sentative of the family, especially in the South-Western portion of the Cape Province, it

was suggested to the writer that the local material be brought up to date. The Curators

of the Bolus Herbarium, the National Herbarium, and the herbaria of the Albany and

South African Museums very kindly placed their respective collections at the disposal of

the writer. Archdeacon F. A. Rogers also allowed use to be made of his private collection.

The writer is also indebted to Dr. I. B. Pole Evans, C.M.G., for granting facilities to under-

take the work, and to Dr. H. P. Phillips for assistance and criticism.

The genus Psoralea comprises a large group of shrubs and herbs belonging to the

sub-order Papilionaceae of the family Leguminosae. The species, numbering about 124, are

found throughout the tropical and sub-tropical parts of the world, and of these, 49 occur

in South Africa. Most of the species occur in the south-western districts of the Cape

Province, but extend through the south-eastern districts into the Transkei and Natal.

The majority of the Transvaal species are found in the northern spurs of the Drakensbergen,

a few in the western districts. With one or two exceptions, the genus is not represented

in the Karoo proper, nor in the north-western districts of the Cape Province, and but few

species have been recorded from Namaqualand and South West Africa.

The habit of the species varies considerably ; some attain a height of 15 feet, while

others are low shrubs or diffuse herbs, while at least one species forms a dense cushion-like

growth. <A characteristic feature of the genus is the presence of resinous glands on

practically every part of the plant. With the exception of P. biovulata, all the species are

uni-ovulate. The flowers are blue, purple, or rarely yellow, and the majority of the species

have sweetly-scented flowers.

PSORALEA (Linn.).

From “ The Genera of South African Flowering Plants,” by KH. P. Phillips.

Calyx gland-dotted, sub-equally lobed, or the lowest segment the largest; the upper

segments often connate. Standard ovate or orbicular, clawed, with inflexed sides ; wings

oblong, sub-faleate ; keel incurved, obtuse, shorter than the standard. Stamens diadel-

phous; the vexillary stamen free or more or less connate with the others ; anthers uniform

or the alternate fixed higher. Ovary sessile or shortly stalked, 1-ovuled, in 1 species 2-

ovuled; style filiform or dilated at the base, incurved above; stigma terminal. Pod

ovate, indehiscent, concealed in the calyx. Seed without an appendage; funicle very

short.

Herbs, shrubs or undershrubs; leaves gland-dotted, digitately 3 to many-foliate,

pinnate or 1|-foliate; stipules embracing the branches, scarcely adnate to the petiole ;

inflorescence a head, spike or raceme, axillary or terminal, or flowers fascicled or rarely

solitary ; bracts 2-3; bracteoles 0,

i. Leaves reduced or apparently absent, or leaflets filiform or linear to linear

obovate, not much broadened from the base upwards.............0.0+eeeeeceeeeee,

Leaflets obovate, oblanceolate or elliptic, broader at the apex than at the

ESDadoblandooc GoodOOt Hoe OO00.0 LO BORO AHOn imc Scie Ont MenUE EH Me noc isin cee tena

2, Ibenyes winea, ischial, meer & wii, VOM .0000000000000000000L0UC Ob FGabeRhOOeKoouOnAHE

Leaflets long-linear, sometimes reduced to a single leaflet, but usually over

Ul Gilly IOWA San Puna MoO KON) MbOUS.CObe D DOBRO RO NRO CON MME ER RnR iAS CoUe Docc Geom estn f

Soee CAL Xe Pll OSE MWA UDLIAR tevey heey ott eh etisral choy seleatcter -uenatie/cs vl shcteracsyaticdanaveraneval a oicvere eee 1 aphylla.

Calyx glabrous within; ovary, 2-ovuled) 22. sas e ess eens. seca es 2 biovulata.

4. Stipules connate, forming a tubular sheath enclosing the petiole.........................

Stipules free or slightly connate, not forming a tubular sheath.........................

5. Stipules with twe short tooth-like points, leaflets sometimes solitary.... 3 restioides.

Stipules with two long acuminate points, often exceeding 1 cm. in length. 4 fascicularis.

GC Weaieus B, Or Ikea nechices! wo Il IeeMieits oooeboocabuopoboncgan0dbccobdUUEbOUunbOUOOUE

Heatletsnmoreythanea setae mais aoc tke tear create RIES eth], Shae rah Ot ER tee CE DI pede lay B,

To. IUCENES WLR Sob GoD Boob oo con bcd Sono ODen Soom b eee crbbc rcen ors 5 Guenzit.

Meavesttrifoliateserarely mnitoliate died cer cyeicts te eae cca ee te See SEU inal gern

OAViexal lamp Tard ante .terteteretep vette nc aise eiaietavaiesecsesielcie 5 sits 2.640 sislaiereiseveual< 6 capitata.

Verllini®” Gb A AChIEs bo tcoogescube oooh bpoDoaensoapnoones odunaoUsUDOOUmbONeoduonD

10. Leaflets uncinate, with a continuous row of glands round margin........ 7 uncinata,

Glands scattered ; if in any way appearing as a marginal band, then leaves not uncinate....

ihe. TSS CEC sss clos died Cid'd B OOOO COCCI S Hen Ames dont Tema kiana Mee eee &

Stemsrdifiuseror decumlbentiante vices sche eee ecto eats cet ek ts eecgs Rac ENON: TPE SAS E 3 Ms co

2leatletss anceolatemuoplinearlanceolatesseriery icin islsisltecicein sei ies sae icicles treo

eafletsulinear) to:lin ear=oblon gss.5cs1s<0-h=raskeiciey oheisnershel fe rater siete a oe ciate alta ei ate lets

14. Upper leaflets unifoliolate; stems laxly leafy........................ 8 oligophylla.

Leaflets all trifoliolate; stems leafy.................0.+.esecereceeee 9 verucosa.

ela dshmmense disinwleaveseny ntti teltkttrstststy-y st ieleaietr-eeleieierrctetctettetel- 10 polyphylia.

Glands oti mMerseditysre aerial) ioterctavel orci re erleteene note rotate tiara eee elerlann alee e otere

Gum lceakletsmpellucid- dotted sey maciaaillaleriehlcivee tte ttre erie iin ee 11 Mundtiana.

Meat etsymiono=punciabewre necktie tl-iel ei cttiie tt titel ieee ree te-rl ick ete eeebereiars

Wicr Chile GowOUs wihldhecosdbodoboas be bavsoobeuaeooenesoouoouDeDeess 12 candicans.

Grills jae cal ding go cea nbo.s obo OD Oe OROAUOOA Cdhin.e Ob Ohm como ToMogcricce UDomEsondueaS

18. Vexillum geniculate with two slight appendages at apex of the claw.... 13 avillaris.

Vexallumenotyceniculatemmorappendagesy-. ein iieieril iene ieineke ie iereieeiciener re

UG amnlGoa vies yO la brouser: tetoeclatckegetcttaral-pedehele ehetoh sredlaietetore sanrsr etd ee aii enerer ete ruers 14 triantha.

Leaves densely hirsute when young, never quite glabrous............. 15 Bolusii.

13: Leaves not exceeding, 5)mm-. in’ length... ..3...:05....--------+- +--+ 16 oreophila.

Meavessexceedinog Iucm win wlongthiatmia vateey je) eyeetsytavanctsrele a caspaie citpetencens sie cl obehetcierete were ile overs

20. Leaves on petioles up to 2 cm. in length............................ 17 glaucina.

Leaves on petioles up to 1 cm. in length..........................-- 18 tenwissima.

8. Leaves with 5-7 pairs of leaflets; flowers white.....................- 19 odoratissima.

Weaves with 2-5) pairs of leaflets); flowers. blue....--..-2..-.--22+-ce +++ terse ctewee

21. Pedicels up to 2-5 cm. long, with connate bracts seldom exceeding 1 mm.

ti IMAI. ob8 do oo BoD OOO bOS SOS DbOUpO EOD OCU OCOD OUNOONEOOOUHODOODO 20 pinnata.

Pedicels up to 4 cm. long, with connate bracts up to 4 mm. long.......... 21 affinis.

OP, Sreowaey Chines Ow ROL Na}. soooconsodccocduoseaduanGaoodnouoDboDDooNSUOUDODCoOGDOOD A

LEMS MCLE CLMOTMASCON GIN Camels teteterlater a dcteneer lc ahevelstte tert deesleveva(ctneloredsustotedaiens Vetere tokens telco:

ORY | Voralllina aiyachiknysegocucleooce boone pve node saomuD pepe Endo S Soul Ne 22 repens.

Venilloim mom gleimclibiecccs oooccopdbogousne cg use dou boo poo bu JDOEDaUUO DOODUSCUEDOU DO

2, ILegiletis: DAUCRORTEVR cogcodgodoosnTdoosouatNs sé souNe ene OOdcUdGEbobobAodedoobdasdn auton

Lemicig VaieO ny By TROD. ccanooboocouccdoodaddavoccHGooGUCN OO G0b OOD UGODaOGCOGONaDOS

AG, Waris Ell OW oogoonoo dodo oscoUobo ad DOObobUbdSO DOC Odo oO 00 sob ot0O0Ob OgEDouDOUDG

TERS FAUITATCO ME o co con DODO OH ODDO SAnD DOOD EHD AaDONSnOUODUDDODDONDUSCORS QOeDFOONnDOOS

28. Inflorescence terminal, spicate; calyx veined................-.+e-0+. 23 bracteata.

Inflorescence axillary, racemose, calyx-tube and lobes marked with purple

RUNES: J adc goowiag Com mald.ed.0 6. UD miOo Deco GO OOed SDR eGor ESE nuncolemoeoD 24 Bowieana.

29° Leaves migro-punctate, pubesCeMmbr). ..) eclele\iel- « «\clelae e-lelcleiel«lelsiecieleleleil els 25 decumbens.

Meaves pellucid-dotteds (canescenites. jie tyele/<l-l sel siele elelalolel el siclelsialesieleleyaiere 26 argenta.

2 peduncles thea mSESSIlemtl OWLS sy -lsyeterersicicactoisve) shor sicecicteioceiel t-ycicusictatenerkatceres 27 bifiora.

Peduncles with 3 or more shortly pedicelled flowers.................. 28 obtusifolia.

OY; TLEENGE TMOG 6 ob otonooandophotooudoK Go Lddeo comodo oOUbbopoDoOr OOOO GOGUROdGODSDO

ILERRREE WHO MEI. 5 od5 pace ooo dudobENOOC OL AD HUDd OO ODO OU AU NOOOUOD DONO duO4odOOGOGOUDOE

BRO, Wes AMBION os cooocbegoudyuNodKDS DUDDODUODEOBOODUSHOOUOCgDOaGS 29 rotundifolia.

117

KEY TO SPECIES.

WeRnVES INGTCL Joos cbbosacadcadddecdosonbugagoooddnoudoDOUOONNHODOOE 30 Thom

~ ww bo

SB Or

3) Calyxglabrous-cmo ceria. coe eee asian tee cic erat hee nee Geren erener re 32

Calyx: pubescomtbectetiecisevecspaisiers sucnteverota’s loses ierstaiciclciedsteverclet citer noe ree ICI eer ee nia 33

32. Stipules persistent; leaflets with a recurved mucro................... 31 aculeata.

Stipules not persistent ; leaflets mucronulate......................... 32 Keetii.

Souehlowersi| si cosebhen, ga xl anya teiepecieter-iele erersiey ier) tected eter een eg ren 34

Mlowerspinkterminal¥oraxillanysheadssonsspikesseee eee eee eer eee eerie eer 35

345) Weatletswithsay recurved: MUCKOs.jecl elec eae eee ee eee eee eee caer erence 36

Weatlets swith? fine) mucrowoRmmoneyatiallae erie eerie isericiitiee teenie tare 37

RG, Cava mT] UIMNOUBIE ocon noo oon cos oauouaD OOOO OO DaDADOOSNDODDGGONR 33 polysticta.

Meaflets*pelluciddotted ois ssevuss-\-\-veretteletel state tice cisterns 34 carnea.

37. Leaflets elliptic-oblong; lateral leaflets unequal sided................. 35 obliqua.

Leaflets cuneate ; with prominent wart-like glands on the lower surface. 36 macradenia.

gun Ini orescencexonl peduncles;exceedingwsmmms lono eee rer) seein eee nner 38

Imflorescence)|sessile or on peduncles) up) to Snem longa --lel-teieaiesisaeecie estes 39

osambedunclesxexceedin oi LO Lemire cineieleeicie etch fore cit nee emit cies Senate 40

Peduncles\ not exceeding. 10) iceman re eee eesti Cerentino meinen 41

40. Leaves densely canescent on the lower surface ; heads globose on peduncles

UpMbo L4MeMaisyercrvyeresrssvyereriaste Tere reticent te eee eae 37 tomentosa.

Leaves glabrescent ; inflorescence a compound spike on peduncles up to

NG Giiinaoopooos oo HOOPER DO OAC OUODOOD UDO OOOO UDO ODOC Doda OOEbOD OS 38 Zeyheri.

41k) Margin of leaflets) dentate-crentlater. eerie ieee 39 Patersoniae.

Margin: ofvleaflets entire... .-1.j.sciciet toe aistretstotte tiie of-reers kere ae ic Carre Ore 42

42, Flowers in groups of 3 on short axillary and terminal peduncles.................-...-.---- 43

Flowers in lax whorls of 3-5 on rigid woody axillary and terminal peduncles............. 44

43. Terminal leaflet on petiole up to 1 em. long; apex rounded........... 40 caffra.

Terminal leatletmsub:sessile;sODtUSe sae eee eee eee aoe ee 41 venusia.

Adv eavespandsstemispsilicyacanescente peter eit t ee ee eer ae 42 hirta.

iheavessandystemssroushlyshirsutesmet rier etch erei ee riieetr eae 43 stachydis.

39 Ovary; Clabrous seers Shoe) Mevoneget> ors ceiocorokewelousle eich uae ine ey<pats ot eed cnn OTT eRS eR ee eter erento 45

Ovary. pilose sa Sisk, Plies wrerene ee eke us yee iohel ehe eyo whaie ioe oes ie eli oa Sete arene TCI eRe ae RIES RS Crashes 46

45. Leaves densely nigro-punctate on both surfaces...................... 44 spicata.

Leaves sparsely glandular; glands near margins..................... 45 racemosa.

462) Spikes sessile jn.ctacrie else rier Wloka ieee hier ieee ci eae eer cir icine rece rere 47

Spikes) on#pedunclessup)to) 22 bcm eerie eet eer eee ne een erence cir ieeeicrsreiet: 48

47. Leaves obovate, hooked! at the apex... <0... 4.0.05 seems ees 46 hamata.

Leaves, oblancedlater Hatevicisisciciit\ au rakeytleveteye ieee fasten a tel even en ae ete 47 Wilmsit.

AK IVE RES CUCs, CHbaeenns wAhWKerRo) WeOWIACECl noo obodbadcscsnonanbodsnoe. 48 striata.

Pubeseence yellowish; leawes mucronulate.........-5.5-.0+-ss0ne-e ee 49 Royffer.

1. P. aphylla Linn.

A tall erect virgate shrub up to 30 metres with drooping branches. Stems glabrescent,

glandular, canescent when young. Leaves unifoliate, 5-17 mm. long, linear, acute; pre-

sent only on very young stems, their place on older stems being taken by ovate acute

bracts up to 5 mm. long. Strpules adnate to the base of the petiole, up to 3 mm. long,

filiform, acute. Flowers axillary, 1-3 subtended by a bract. Pedvcels bracteolate about

the middle, with scattered black hairs on upper portion. Bracts up to 5 mm. long, connate,

bi-lobed, acute, glandular, pilose. Calyx-tube 2-5-6 mm. long, 6-10 mm. in diameter, 10

nerved, glandular, glabrous to nigro-hirsute ; lobes linear-lanceolate to lanceolate-ovate,

acute, glandular, ciliate, glabrous or laxly nigro-hirsute; shortly pilose on the inner

surface; upper and lateral lobes 2-6 mm. long, 1-2 mm. broad; slightly sub-faleate ;

lowest lobe 4-12 mm. long, 1-5-8 mm. broad. Veaxillum 7-14 mm. long, 6-15 mm. broad,

obovate or sub-orbicular, geniculate, shading from white to deep blue; claw 1-4 mm.

long, with two small appendages at the apex; alae pure white, 7-12 mm. long, 2-8

mm. broad, with a linear claw 2-5 mm. long; carina with a deep blue blotch at the apex,

5-9 mm. long, 3-6 mm. broad, with a linear claw 2-7 mm. long. Ovary glabrous ;

style 6-9 mm. long, glabrous.

Flowers: June—January.

Carp.—Clanwilliam district: Klavoer, Wolley Dod 298; Piquetberg district :

Piquetberg, Edwards 248; Worcester district: Mitchell’s Pass, Bolus 2607; Tulbagh

district : Waterfall, Ecklon and Zeyher; Rogers 17395; Cape district: Table Mountain,

MacOwan 924: Bolus; Ecklon and Zeyher 1530; Galpin 3695, Ecklon and Zeyher, Forbes

1g)

140; Wynberg, Wolley Dod 21; Wynberg and Kalk Bay, Bolus 2607 ; Camps Bay, Rogers

3022 without precise locality, Wolley Dod 2257; Caledon district: Steenbras, Moss and

Rogers 1540; Rogers 10510; Zwart Berg, Galpin 3694 ; Mossel River, Potts 1586; Rivers-

dale District: Riversdale, Schlechter 1881.

2. P. biovulata Bolus.

A small slender shrub. Stems -3—-6 mm. long, slender, thinly canescent when young,

becoming glabrescent. Leaves up to 6 mm. long, resembling scales, adpressed, exstipulate,

rigid, few, linear-acuminate or subulate, glabrescent, ciliate, with imcurved margins.

Flowers terminal, 1-3 together on short pedicels or subsessile. Pedicels bracteate. Calyx-

tube up to 1-5 mm. long, 3 mm. in diameter, pilose with short silky white hairs, 10-nerved ;

lobes unequal, lanceolate, acuminate, ciliate, minutely glandular, glabrous on the inner

surface ; 2 lobes 1-1-5 mm. long, about 5 mm. broad ; 3 lobes 2-3 mm. long, 0-5-1 mm.

broad. Vezillum 4 mm. long, 5-6 mm. broad, broadly obovate, with a short claw 1 mm.

long ; alae 4-5 mm. long, 2 mm. broad, with a linear claw 1-5 mm. long; carina 3-3-5

mm. long, 2 mm. broad, with a linear claw 1-5 mm. long. Ovary pilose, 2 ovuled ; style

3 mm. long, pilose.

Flowered : October.

Carr.—Bredasdorp district: Mount Kim, Bolus 6902.

3. P. restioides EH. and Z.

A slender trailing suffruticose. Stems up to -6 m. long, glabrous, glandular, striate.

Leaves unifoliolate, rarely trifoliolate, petiolate, stipulate; petioles up to 1-5 cm. long ;

leaflets 0-5-3 em. long, scarcely 1 mm. broad, subsessile, filiform, mucronulate, glabrous,

glandular. Stipules adnate to the base of the petiole; up to 1-1 cm. long, stem-clasping,

glandular. Flowers solitary, axillary, pedicellate. Pedicels up to 1 cm. long, bracteolate

about the middle. Bracts membraneous, tri-lobed, glandular, ciliate; 1 lobe short; 2

lobes up to 6 mm. long, acute, filiform, sub-faleate. Calyx-tube 2-4 mm. long, 4-7 mm.

in diameter, glabrous, glandular; lobes unequal, lanceolate, ciliate, glandular, glabrous,

shortly nigro-hirsute on the inner surface ; upper and lateral lobes 1-4 mm. long, 0-4-1-5

mm. broad; lowest lobe 4-6 mm. long, 1-2 mm. broad, subulate. Vexillum 5-7 mm.

long, 5-8 mm. broad, obovate, with a linear claw 1-5-3 mm. long; alae 5-6-5 mm. long,

1-5-2-5 mm. broad, with a linear claw 2-3 mm. long; carina 3-4 mm. long, 1-2 mm.

broad, with a linear claw 2-5-4 mm. long. Ovary glabrous; style 3-4 mm. long, glabrous.

Flowered: August-December. Flowers: Violet-blue.

Care.—Clanwilliam district: Vogelgat, Schlechter 9529; Cape district: Muizenberg,

Bolus 4271, Wolley Dod 2566; Table Mountain, Bolus 4504, Wolley Dod 795; Caledon

district : Klein River, Ecklon and Zeyher, Hermanus, Burtt-Davy, Bolus 9841 ; Bredasdorp

district: Elim, Bolus 6901, Bodkin; Riversdale district: Riversdale, Rogers 4394, Muir

794, Langeberg, Schlechter 2153; Mossel Bay district: Mossel Bay, Potts.

4. P. fascicularis D.C.

Plants suffruticose. Stems up to 6 metres, slender, diffuse or ascending, glabrous.

Leaves trifoliolate, sometimes unifoliolate ; leaflets 1-3 cm. long, linear-lanceolate, apex

very acute; petioles up to 2-5 cm. long. Stipules adnate to the petiole, stem-clasping,

up to 3 cm. long; points of stipules 1-3 cm. long, subulate, acute. Flowers axillary,

solitary, or several together. Pedicels up to 6 mm. long, axillary, solitary or several together.

Calyz-tube 2 mm. long, 5 mm. in diameter, glandular, glabrous ; lobes lnear-lanceolate,

acute, glabrous, ciliate, nigro-hirsute on the inner surface; upper and lateral lobes

2-3 mm. long, up to 1 mm. broad ; lowest lobe 5 mm. long, | mm. broad. Vezxillwm 5-6

mm. long, 4-5 mm. broad, obovate with a linear claw 2 mm. long ; carina 3-3-5 mm. long,

1-5 mm. broad, with a claw 2 mm. long; alae 5 mm. long, 1-5-2 mm. broad, with a claw

2mm. long. Ovary glabrous; style 3-4 mm. long, glabrous.

Flowers: November—January.

Carr.—Stellenbosch district: Hottentots Holland near Somerset West, Ecklon and

Zeyher ; Cape district: Newlands, Forbes 89, Orange Kloof Farm, Wolley Dod 3166.

120

5. P. Gueinzii Harv.

Plant suffruticose. Stems 3-6 metres, diffuse, villose, becoming glabrescent. Leaves

unifohiolate, 1-2-5 cm. long, up to 4 mm. broad, lanceolate, very acute, sub-sessile. Stipules

ovate, acute, striate, persistent. Flowers axillary, pedicellate. Pedicels short, about 3 mm.

long, bi-bracteolate. Bracts about 3 mm. long, ovate, acute, ciliate. Calyx-tube 2 mm.

long 6 mm. in diameter, sparsely glandular and villous; lobes lanceolate, acute, ciliate, glands

small, pilose on the inner surface; upper and lateral lobes 3-3-5 mm. long, about 1 mm.

broad, lowest lobe 5 mm. long. Vellum obovate, 1 cm. long, 8 mm. broad, with a narrow

channelled claw 2 mm. long; alae 7 mm. long, 2 mm. broad, with a linear claw 2 mm. long ;

carina 5 mm. long, 3 mm. broad, with a linear claw 3 mm. long. Ovary glabrous; style

6 mm. long, glabrous.

CapE.—Stellenbosch district: Hottentots Holland, Hceklon and Zeyher.

6. P. eapitata Lf.

Plant shrubby. Stems slender, up to 12 metres long, sparsely leafy, striate, glandular,

glabrescent. Leaves uni or-trifoliolate ; upper leaves usually unifoliolate ; petioles variable,

up to 2-5 cm. long; leaflets 1-8 em. long, up to 4 mm. broad, linear to linear-lanceolate,

acute, nigro-punctate, glabrescent. Strpules adnate to the base of the petiole, up to 7 mm.

long, subulate, acute. Inflorescence spicate or compressed into an aggregated terminal

head subtended by one or more unifoliolate leaflets with sumple acute glandular pilose adnate

stipules up to 1 cm. long, or stipules simple or connate. lowers pedicellate. Pedicels about

2 mm. long, bracteate. Calyx-tube 1-5-2 mm. long, 3-5 mm. in diameter, glandular,

glabrous to villoso-hirsute ; lobes lanceolate, acute, ciliate, glandular, glabrous to villoso-

hirsute without, pilose to nigro-hirsute within; upper and lateral lobes 1-3 mm. long,

0-5-1 mm. broad ; lowest lobe 2-5-5 mm. long, 0-5-1 mm. broad. Veaxillum 5-7-5 mm.

long, 4-6 mm. broad, obovate, glandular, with a claw 0-5-2 mm. long; alae 4-5-7 mm.

long, 1-5-2 mm. broad, with a linear claw 1-2 mm. long, carina 4-6 mm. long, 1-5-2 cm.

broad, with a linear claw 1-3 mm. long. Ovary glabrous; style 3-7 mm. long, glabrous.

Flowers: November—April.

Caprr.—Caledon district: Twenty-four Rivers, Zeyher; Tulbagh district: Waterfall,

Zeyher ; Cape district: Kenilworth, Bolus 3285; Knysna district: Knysna, Zeyher, Plet-

tenberg Bay, Levpoldt; Uitenhage district: Witte Khp, Bolus, Zwartskops River, Zeyher

2386; Albany district: Grahamstown, Glass, Southey; Komgha district: Cape Morgan,

Flanagan 2370.

Locarities UnkNown.—Zeyher 4945, Tyson 3049, MacOwan 2068.

7. P. uncinata E. and Z.

Plant small, 3-6 metres high; branching chiefly from the root. Stems erect, cane- .

scent, becoming glabrous, striate; glands becoming conspicuous when the stems are

glabrous. Leaves trifoliolate, at equidistant nodes 3 em. apart, shortly petiolate, stipulate ;

petioles up to 5 mm. long; leaflets up to 3-2 em. long and 4 mm. broad, linear-oblong, a

few lower ones cuneate-oblong, acuminate, uncinate, with scattered hairs and glands, the

glands forming a continuous band round the margins. Stipules adnate to the base of the

petiole, striate, canescent, glabrescent, with two lanceolate acute points up to 7 mm.

Inflorescence terminal and axillary, in dense sub-globose heads. Peduncles very short, up

to 6 mm. long. Flowers sessile or sub-sessile. Calyx-tube 3-4 mm. long, 6-8 mm. in dia-

meter, canescent, distinctly 10-nerved, sparsely glandular ; lobes linear-lanceolate, acute,

glandular, canescent, ciliate, with the inner surface glabrous or shortly and thinly pilose ;

upper and lateral lobes 4-6 mm. long, 1-2-5 mm. broad ; lowest lobe 6 mm. long, 2-3 mm.

broad. Vexillum 6-8 mm. long, 4-6 broad, obovate with a linear claw 1-3 mm. long ;

alae 5-7 mm. long, 1-2 mm. broad, with a claw 3-4 mm. long; carina 3-4 mm. long, 1-5-2

mm. broad, with a claw 3-4 mm. long. Ovary pilose; style glabrous.

Flowers: November—January.

121

CapE.—Piquetberg district: Pentonville, Hdwards 249; Tulbagh district: near

Tulbagh, Zeyher 1554, Rogers 17341, Wintershoekberg, Zeyher ; Cape district : Capetown,

Guthrie; Caledon district: Vier-en-Twintig Rivier, Zeyher; Beaufort West district :

Nieuwe Kloof, Schlechter 9038.

8. P. oligophylla E. and Z.

A tall slender virgate shrub. Stems glabrous, glandular, striate, sparsely leafy. Leaves

uni- or trifoliolate, petiolate, stipulate ; petioles up to 1 cm. long; leaflets up to 4-5 cm.

and 4 mm. broad, linear to linear-lanceolate, acute, glabrous, glandular. Stipules minute,

subulate. Flowers axillary 1-3 together. Pedicels up to 3 cm. long, bibracteolate near the

apex. Bracts 2-4 mm. long, with acute glabrous ciliate glandular lobes. Calya-tube 2-4

mm. long, 6-9 mm. in diameter, glandular, glabrous, 10-nerved ; lobes 1-5 mm. long, 1-2

mm. broad, slightly unequal, lanceolate-ovate, acute, slightly sub-falcate, glabrous, glan-

dular, ciliate, shortly pilose on the inner surface. Vexillwm 6-11 mm. long, 7-12 mm. broad,

sub-orbicular, geniculate; claw 1-4 mm. long, with two slight appendages at the apex ;

alae 0-6-1 cm. long, 2-5-5 mm. broad, with a linear claw 2-5 cm. long; carina 4—7-5 mm.

long, 2-5-5-5 mm. broad, with a linear claw 3-6 mm. long. Ovary glabrous; style 3-8

mm. long, glabrous. Pod, when fully developed, about 6 mm. long and 3 mm. broad ;

seed about 4-5 mm. long, 2 mm. broad, brown.

A tall shrub 5-6 feet high, growing usually in moist places.

Flowers: Blue.

Cape.—Van Rhynsdorp district: Giftberg, Phillips 3236, 7488; Ceres district :

Onderbokkeveld, Ecklon and Zeyher, Koude Bokkeveld, Schlechter, Matroosberg, Phillips

1944; Worcester district: Hex River Valley, Tyson 1908, Rehmann 1533, 1534; Riversdale

district : Riversdale, Muir 81; George district : Outeniqua Mountains, Hops 52, George,

Schlechter 2227; Humansdorp district: near Humansdorp, Schonland 3086, Witte Els

Bosch, Fourcade 827; Port Elizabeth district: Hankey, Sim 173, Forest Dept. Herb. 3997 ;

Bathurst district: Bathurst, Rogers 3516; Albany district: Grahamstown, Featherstone’s

Kloof, Bolus 1935, Atherstone, Rogers 3316, Howison’s Poort, Galpin 141, Glass, Barber,

Galpin 336, Zeyher 2383, Ecklon and Zeyher, MacOwan 448, Asylum, Daly and Sole 137 ,

Hofman’s Bosch, Britten 1327, near Sanatorium, Zuurberg, Schonland 3185, Cold Spring,

Rogers 5635, 27312, 27514; Queenstown district: Queenstown, Galpin 8107, Cathcart

Shiloh, Baw 862; Kingwilliamstown district: Kingwilliamstown, Sim 2301; Komgha

district: Komgha, Flanagan 556, 3410; Tembuland, Xalanga, Bolus 8858.

Gt. Bushmanland, Vogelklip, Pearson and Pillans 5932, Leliefontein, Pearson 6344.

Locauities UnkNown.—Rogers 17584, Schlechter 10186, Brotten 140.

9. P. verucosa Willd.

Plant shrubby. Stems erect, glandular striate, tomentulose. Leaves trifoliolate,

petiolate, stipulate; petioles up to 1-5 cm. long; leaflets wp to 4 em. long and 4 mm. broad,

linear-lanceolate, acute, glabrous, nigro-punctate. Stipules up to 3 mm. long, subulate,

acute. Flowers axillary, 1-3 together. Pedicels up to 2-2 em. long, bracteolate. Bracts

about 1 mm. long. Calyzx-tube 3 mm. long, 5-9 mm. in diameter, 10-nerved, glandular,

glabrous; lobes ovate, acuminate, glabrous, glandular, ciliate, shortly pilose on the mner

surface. Veaillwm 0-5-1 ‘cm. long, 0-5-1-2 cm. broad, sub-orbicular, geniculate ; claw

0-5-2 mm. long, with two slight appendages at the apex; alae 5-9 mm. long, 2-5 mm.

broad, with a linear claw 3-5 mm. long; carina 3-5 mm. long, 2-4 mm. broad, with a

linear claw 1-6 mm. long. Ovary glabrous; style 2-8 mm. long, glabrous.

Capr.—Worcester district: Worcester, Ecklon and Zeyher 1520; Knysna district :

Wittesdrift, Plettenberg Bay, Zeyher; Humansdorp district : Humansdorp, Storms River,

Schlechter 5971; Uitenhage district: Zwartkops River, MacOwan 725, Ecklon and Zeyher

1521, Zeyher 2379; Port Elizabeth district: Walmer, Paterson 829, Braakens River,

Burchell 4218.

122

10. P. polyphylla E. and Z.

Plant shrubby, pubescent, glabrous when older. Stems erect, leafy, branched. Leaves

trifoliolate, sub-sessile, stipulate; leaflets 1-1-1-5 cm. long, 2-4 mm. broad, linear to

linear-oblong, mucronate, glands immersed ; lateral leaflets unequal sided. Stipules small,

ovate, acuminate, not persistent. Flowers axillary, 2-3 together. Pedicels very short.

Calyx-tube nigro-villous and veined; upper and lateral lobes subulate, the lowest ovate-

lanceolate, longer than the rest.

Flowered: July.

Carr.—Uitenhage district: Krakakamma, Ecklon and Zeyher 1536. Description

taken partly from “ Flora Capensis,” as there were no complete flowers on the material

seen,

11. P. Mundtiana E. and Z.

Plant shrubby, up to -6 m. high. Stems érect, pubescent, leafy. Leaves trifoliolate,

petiolate, stipulate; petioles about 5 mm. long, glabrescent; leaflets 1-5-2-5 cm. long,

4—6 mm. broad, broadly linear to linear-oblong, tapering at the base, recurved mucronate,

pellucid dotted. Stepules about 5 mm. long and 1-5 mm. broad, ovate, acute, pubescent,

scarious, striate. Flowers axillary, 2-3 together. Pedicels short. Calyx-tube 3 mm. long,

8 mm. in diameter, glandular, villous; lobes lanceolate, acute, villous, glandular, ciliate,

pilose on the inner surface; upper and lateral lobes 5-7 mm. long, 1-1-5 mm. broad ;

lowest lobe 8 mm. long, 2 mm. broad. Vezxillum 6 mm. long, 5 mm. broad, obovate, with

a narrow channelled claw 1-5 mm. long; alae 6 mm. long, 1-5 mm. broad, with a linear

claw 2 mm. long; carina 3 mm. long, 2 mm. broad, with a linear claw 2mm. long. Ovary

and style missing in the specimen examined.

Flowered : December.

Cape.—Swellendam district: Swellendam, Ecklon and Zeyher 1537.

12. P. candicans EH. and Z.

Plant shrubby, virgate, densely or thinly canescent, glabrescent. Stems erect, laxly

leafy. Leaves trifoliolate, shortly petiolate, stipulate ; petioles about 2 mm. long ; leaflets

1-5 cm. long, 0-65 cm. broad, linear to linear-oblong, obtuse, recurved mucronate, nigro-

punctate; a few of the lower leaflets cuneate. Stipules small, up to 3 mm. long, adnate

to the base of the petiole, subulate, acute, glandular. Inflorescence axillary, 1-3 flowers on

common peduncle. Peduncles up to 1-7 em. long; pedicels about 4 mm. long. Calyx-

tube 1-5-3 mm. long, 4-6 mm. in diameter, canescent, distinctly 10-nerved, with small

often inconspicuous glands; lobes linear-lanceolate, acute, pilose, glandular, glabrous on

the inner surface; upper and lateral lobes 2-4 mm. long, about 1 mm. broad, slightly

sub-faleate ; lowest lobe 4-6 mm. long, about 1 mm. broad. Vezillwm 3-5 mm. long,

3-5 mm. broad, obovate, with a linear claw 1-2 mm. broad; alae 3-5 mm. long, 1-2 mm.

broad, with a claw 1-5-2 mm. long; carina 2-3 mm. long, 1-2 mm. broad, with a claw

2mm. long. Ovary pilose; style 2-3 mm. long, glabrous.

Flowered: September—May.

Carr.—Worcester district: Hex River, Bolus 8005; Caledon Division: Zeyher,

Hlandskop, Schlechter 9757, Ruggens near Zuurberg, Galpin 3960, near Genadendal, Galpin

3959, Bolus 7381; Swellendam district: Swellendam, Grootvaders Bosch, Zeyher 1539 ;

Uitenhage district : Zwartkop River, Zeyher 2731; Albany district: Redhouse, Grahams-

town, Paterson 238; Somerset Kast district: Bruntzeshoogte, MacOwan.

Wirnour Locauity: Zeyher 2373.

123

13. P. axillaris L.

Plant shrubby. Stems erect, glabrous, glandular, striate. Leaves trifoliolate, stipulate,

petiolate ; leaflets 1-2-8 cm. long, 3-8 mm. broad, linear to linear-oblong, tapering at the

base, obtuse, mucronulate, nigro-punctate, glabrous. Stipules about 3 mm. long, subulate.

Flowers axillary, usually solitary. Pedicels up to 2 cm. long, bracteolate near the apex.

Calyz-tube 2-4 mm. long, 0-6-1 cm. in diameter, thickly nigro-punctate, glabrous ;

lobes 1-3 mm. long, 1-2 mm. broad, slightly unequal, ovate, obtuse, nigro-punctate, glabrous

shortly ciliate, pilose on the inner surface. Veaillwm 0-8-1 cm. long, 0-8-1-1 em. broad,

sub-orbicular, geniculate ; claw channelled, 1-5-2 mm. long, with two slight appendages

above; alae 0-7-1 cm. long, 3-5 mm. broad, with a claw 3-5 mm. long; carina 5-6 mm.

long, 4-5 mm. broad, with a linear claw 5-6 mm. long. Ovary glabrous; style 6-9 mm.

long, glabrous.

Care.—Tulbagh district : Tulbagh, Burchell 1033; Knysna district: Knysna, Duthie

767, Plettenberg Bay, Zeyher; Uitenhage district: Witteklip, MacOwan 1044; Port

Elizabeth district : Port Elizabeth, Kemsley 247, Paterson 829, Van Stadens River, Bolus.

14. P. triantha E. Mey.

A slender shrub up to 4 ft. Stems erect, tomentulose. Leaves trifoliolate, petiolate,

stipulate ; petioles about 6 mm. long ; leaflets 0-5-2-5 em. long, up to 6 mm. broad, linear,

spathulate, recurved mucronate, thick, glabrous, nigro-punctate. Stipules about 2-5 mm.

long, acuminate. Flowers axillary, crowded. Pedicels 0-5-1 cm. long. Calyx-tube 2 mm.

long, 4-6 mm. in diameter, pubescent, grandular; lobes lanceolate, pubescent, pilose on

the inner surface; upper and lateral lobes 1-2 mm. long, 0-5-1 mm. broad. Vexillum

4-8 mm. long, 4-6-5 mm. broad, obovate, with a linear claw 1-5-2 cm. long; alae 4-6 mm.

long, 1-5-2-5 mm. broad, with a linear claw 2-3 mm. long; carina 2-5-3 mm. long, 1-5-2

mm. broad, with a linear claw 2-5-3 mm. long. Ovary glabrous or thinly pilose; style

3-4 mm. long, filiform.

Flowers : November—February.

Cape.—Clanwilliam district: Clanwilliam, Leipoldt 632; Tulbagh district: Witsen-

berg, Zeyher 446; Cape district: Blaauwberg, Drege; Paarl district French Hoek,

Schlechter 10268, Paarl, Schlechter 202; Uitenhage district: Zwartkop River, Zeyher 263.

15. P. Bolusii Forbes n. sp.

Plant shrubby, small. Stems erect, leafy, glandular, striate; young stems densely

albotomentose becoming glabrescent. Leaves trifoliolate, petiolate, stipulate; petioles

about 7 mm. long; leaflets 1-2-5 cm. long, about 4 mm. broad, linear to linear-oblong,

nigro-punctate, densely .albo-hirsute becoming almost glabrous, recurved mucronate.

Stipules adnate to the base of the petiole, up to 5 mm. long, subulate, acute. Flowers

axillary, in groups of 3, sub-sessile. Calyzx-tube 3 mm. long, 5 mm. in diameter, glandular,

hirsute ; lobes unequal, hirsute, long-ciliate, glandular, pilose on the inner surface ; two

upper lobes 2 mm. long, about 0-25 mm. broad, linear, acute; lateral lobes 3-5-4 mm.

long, about 0-5 mm. broad, linear, acute ; lowest lobe 6 mm. long, 1-5 mm. broad, lanceo-

late, acute. Vewillum 5-5 mm. long, 5-5 mm. broad, obovate, with a channelled claw 3

mm. long; alae 5-5 mm. long, 2:5 mm. broad, with a linear claw 2-5 mm. long; carina

3 mm. long, 1-5 mm. broad, with a linear claw 3 mm. long; Ovary thinly pilose ; style

3 mm. long, filiferm.

Flowered : October.

Capr.—Piquetberg district: Piquetberg, Bolus.

16. P. oreophila Schltr.

A very small slender herb, branching from the root. Stems decumbent, up to 3 metres

long, filiform, subterete, with scattered leaves. Leaves trifoliolate, petiolate, stipulate ;

petioles slender; leaflets linear, acute, narrowing at the base. Stipules 4-5 mm. long,

equalling or about half the length of the petiole, connate, linear-lanceolate, acute, glabrous.

124:

Flowers axillary, solitary. Pedicels nearly equalling the leaves, bracteolate near the middle.

Bracts up to 5 mm. long. Calyx-tube 2 mm. long, 4-5-5 mm. in diameter, glabrous, glan-

dular ; lobes linear-lanceolate, acute, glabrous, glandular, very shortly ciliate, glabrous on

the mner surface ; upper and lateral lobes 3-5 mm. long, 1-1-25 mm. broad; lowest lobe

5-5-6 mm. long, 1-5 mm. broad. Veaillwm 5-5-6 mm. long, 6-8 mm. broad, sub-orbicular

with a channelled claw 1-5-2 cm. long; alae 4-5-5 mm. long, 2 mm. broad, with a linear

claw 2-2-5 mm. long; carina 3 mm. long, 2 mm. broad, with a linear claw 2-2-5 mm. long.

Ovary glabrous; style 3-5 mm. long, glabrous.

Flowered: January. Flowers: Deep blue.

- Cape.—Caledon district: Sir Lowry’s Pass, Schlechter 7233.

17. P. glaucina Harv.

Plant suffruticose, diffuse. Stems slender, glabrous. Leaves trifoliolate, petiolate,

stipulate ; petioles 0-7-2 cm. long; leaflets 0-9-2 cm. long, up to 2 mm. broad, linear-

lanceolate, acute, tapering at the base, glabrous, glandular. Flowers axillary, 1-3 together.

Pedicels 1-1-5 cm. long, bracteolate near the middle. Bracts 2-2-5 mm. long, tri-lobed,

ciliate, acute. Calyx-tube 2 mm. long, 5 mm. in diameter, glandular, glabrous, 10-nerved ;

lobes lanceolate, acute, glandular, glabrous, with almost reticulate veming, shortly ciliate,

very shortly pilose on the inner surface ; upper and lateral lobes 1-5-2 mm. long, up to 1

mm. broad; lowest lobe 3 mm. long, 1 mm. broad, Veaxillum 4 mm. long, 4 mm. broad,

obovate, with a channelled claw 1 mm. long; alae 3 mm. long, 1-5 mm. broad, with a

linear claw 2 mm. long; carma 2 mm. long, 1 mm. broad, with a linear claw 3 mm. long.

Ovary glabrous; style 3 mm. long, glabrous.

Care.—Doornhoogte, Heklon and Zeyher.

18. P. tenuissima H. Mey.

Plant slender, trailing. Stems up to 6 metres long, glabrous. Leaves trifoliolate,

petiolate, stipulate ; petioles up to 9 mm. long ; leaflets up to 2-5 cm. long, 1-5 mm. broad,

narrow, linear, acute, nigro-punctate, glabrous, sub-sessile. Stipules wp to 2 mm. long,

subulate, acute, glandular. Flowers axillary, solitary. Pedicels up to 3 cm. long, bracteo-

late, near the apex. Calyx-tube 2 mm. long, 5-6 mm. in diameter, glandular, glabrous ;

lobes acute, lanceolate, glabrous, ciliate, glandular, pilose on the mner surface ; upper and

lateral 2-3 mm. long, 1-1-3 mm. broad ; lowest lobe 3-4 mm. long, up to 1-5 mm. broad.

Vexillum 6-8 mm. long, 6-8 mm. broad, obovate, geniculate, with a channelled claw 1-2

mm. long; alae 6-8 mm. long, 2-3 mm. broad, with a linear claw 1-5-2 mm. long; carina

5 mm. long, 2 mm. broad, with a linear claw 2-3 mm. long. Ovary glabrous; style 9 mm.

long, glabrous.

CapE.—Tulbagh district: Tulbagh, Schlechter 7476; Swellendam district: Wage-

boom’s River, Schlechter 10701.

19. P. odoratissima Jacq.

Plant shrubby, at least 18 metres high. Stems striate, glandular, sparsely pilose.

Leaves imparipinnate, petiolate, stipulate: petioles 5-8 mm. long; leaflets 5—7 pairs,

0-7—-2-2 cm. long, about 1-5 mm. broad, linear-lanceolate, acute, glandular ; young leaves

laxly clothed with fairly long white hairs. Stepules about 4 mm. long and 2 mm. broad,

lanceolate-ovate, acute, ciliate. lowers axillary, 1-3 or more together. Pedicels up to

1-5 em. long. Calyx-tube 3-5 mm. long, 7-5 mm. in diameter, sparsely pilose with long

white hairs, glandular; lobes, lanceolate, acute, slightly sub-faleate, ciliate, shortly pilose

on the inner surface ; upper and lateral lobes 4 mm. long, 2 mm. broad ; lowest lobe | cm.

long, 2-5 mm. broad. Vexillwm 1 em. long, 1-1 em. broad, sub-orbicular, geniculate, with

a channelled claw 3 mm. long; alae 1 cm. long, 5 mm. broad, with a linear claw 5 mm.

long ; cara 6 mm. long, 4 mm. broad, with a linear claw 5-5 mm. long. Ovary glabrous ;

style 6 mm. long, glabrous.

Flowers: Pure white, very fragrant.

Flowered at Kirstenbosch, October-November, 1923, 1924. Original locality unknown.

125

20. P. pinnata Linn.

Plant a tall much branched woody shrub up to 30 metres high. Stems striate, virgate,

glandular, glabrous or pubescent. Leaves impar-pinnate, petiolate, stipulate ; leaflets up

to 5 em. long and 3 mm. broad, linear to linear-lanceolate, acute, glandular, sometimes

recurved at the apex. Stipules free, about 5 mm. long, lanceolate to ovate-lanceolate,

acute, pubescent. Flowers axillary, 1-3 or several together. Pedicels sub-sessile or up to

2-5 cm. long, bracteolate near the apex. Bracts 1-3 mm. long, connate, pilose, with acute

ciliate lobes. Calyx-tube 2-6 mm. long, 0-4-1-3 cm. in diameter, villous or glabrous,

glandular ; lobes lanceolate, ovate, acute, glabrous or villous, glandular, ciliate, pilose on

the inner surface ; upper and lateral lobes 2-7 mm. long, 0-5-3 mm. broad ; lowest lobe

0-4-1 cm. long, 2-4 mm. broad. Vezillum 0-4-1-4 cm. long, 0-4-1-5 cm. broad, sub-

orbicular or broadly obovate, geniculate ; claw channelled 1-4 mm. long, with or without

the two small appendages at the apex of the claw ; alae 0-4-1-2 cm. long, 1-5-6 mm. broad,

with a linear claw 1-5 mm. long; carina 2-5-8 mm. long, 1-5-7 mm. broad, with a linear

claw 2-7 mm. long. Ovary glabrous; style 0-3-1 cm. long. Seed 4 mm. long, 2 mm.

broad, reddishbrown.

Flowers: Pale to deep blue. This species has a wider distribution than any other

native species and is in bloom almost the whole year.

Capr.—Clanwilliam district: Clanwilliam, Rogers 16798, Olifants River, Lezpoldt :

Tulbagh district : Scott and Hlhot, Winterhoek, Phillips 1729; Cape district : Devils Peak,

Ecklon and Zeyher 1519, Table Mountain, Forbes 139, Newlands, Forbes 91, Wynberg Moun-

tain, Wolley Dod 22; Stellenbosch district : Stellenbosch, Heklon and Zeyher 1515; Paarl

district : French Hoek, Schlechter 9263, 10268, Phillips 1089; Caledon district: Caledon

Zeyher, Bolus; Swellendam district: Swellendam, Ecklon and Zeyher 1518; Riversdale

district: Riversdale, Mwr 79, Aasvogelberg, Muir 1850, Ladismith, Seven Weeks Poort,

Phillips 1432; Mossel Bay district: Mossel Bay, Rogers 4165, Potts ; George district

Mitchell (Bolus Herb. 16090), Rogers 4281, near Langekloof, Zeyher 1517, Robertson Pass,

Outeniqua Mountains, Hops 53, Garcia’s Pass, Phillips; Knysna district: Knysna, Main

Forest, Schonland 3600, Plettenberg Bay, Smart (Rogers 15457) ; Kingwilliamstown district :

Zitzikamma Mountains, Britten 1079; Humansdorp district: Humansdorp, Schonland

3018, Witte Els Bosch, Fourcade 892, 923, 924, Muir 1250, Assegai Bosch, Rogers 2834,

Slang River, Phillips 3441 ; Uitenhage district: Uitenhage, Ecklon and Zeyher 305, Schon-

land 3730, Britten 3027, Van Staadens, Paterson 734; Albany district: Grahamstown,

Zeyher 903, 2380, Rogers, Schonland, Signal Hill, Dale and Cherry 883, Coldspring, Gower ;

Queenstown district: Katberg, Schonland 4298; Fort Beaufort district: Hogsback,

Rattray 83, Stayner 11, Hofiman’s Bosch, Britten 1185, near Kat River, Ecklon and Zeyher

1514; Catheart district: Cathcart, Cotterrell 164, Galpin 7360; Komgha district: near

Kei Mouth, Flanagan 178; Kentani district: Kentani, Pegler 255; Umtata district :

Umtata, Schlechter 6337; Beaufort West district: Murraysburg, Bolus 220.

TRANSVAAL.—Lydenburg district: Pilgrims Rest, Rogers 23110, Lydenburg, Wilmer ;

Zoutpansberg district: Zoutpansberg, Rietfontein, Leendertz 808 ; Woodbush, Jenkins

7194; Barberton district: Barberton, Galpin 401, Elandshoek, Rogers 448.

SWAZILAND.—Stewart 3687.

Narat.—Drakensberg, Polela, Hvans 714; Karkloof, Dimmock-Brown 333 ; Maritz-

burg, Moss 3344; Richmond, Wood 84, 9852; Hstcourt, Wood 10261, Wyle (Wood 10006) ;

Field’s Hill, Wood 12830; Evans 146; Illovo, Wood 6411.

ZULULAND.—Inyiza, Edwards 315 ; Ngoya, Wood 9314; Nkandhla, Wylie (Wood 8817).

21. P. affinis E. and Z.

A tall virgate shrub up to 18 metres. Stems woody, striate, glandular, slightly pilose

when young. Leaves imparipinnate with 2-4 pairs of leaflets, stipulate, petiolate ; petioles

0-7-2-3 em. long ; leaflets up to 5 cm. long, 6 mm. broad, linear-lanceolate to ovate-lanceo-

late, obtuse, glandular, mucronulate. Stipules small, deltoid. Flowers axillary, 1-3 or

126

more together. Pedicels up to 4-5 cm. long, bracteolate near the apex. Bracts bi-lobed,

up to 5 mm. long, pilose, glandular, veined. Calyzx-tube 4 mm. long, 1 cm. in diameter,

thickly nigro-punctate, sparsely to densely nigro-hirsute, veined ; lobes lanceolate-ovate

to ovate, acute, hirsute, nigro-punctate, hirsute on the inner surface; upper and lateral

lobes 3-5-4 mm. long, 2 mm. broad; lowest lobe 5 mm. long, 3-4 mm. broad. Veaxillum

0-9-1-2 em. long, 1-2-1-7 cm. broad, obovate, geniculate, with a linear claw 3-4 mm. long ;

alae 0-8-1-1 em. long, 5 mm. broad, with a linear claw 5-6 mm. long; carina 6-7-5 mm.

long, 4-4-5 mm. broad, with a linear claw 6-8 mm. long. Style 0-8-1 cm. long.

Flowers: Deep blue or purple.

Capr.—Riversdale district : Riversdale, near Albertinia, Muir 1851 ; George district :

George, Rogers 4291 ; Port Elizabeth district: Port Elizabeth, Van Staadensberg, Zeyher

2378 ; Albany district: Grahamstown, Rogers 27423; Stutterheim district: Stutterheim,

Ecklon and Zeyher, Rooi Vaal, Flanagan 178.

TRANSVAAL.—Barberton district : Barberton, Pole Evans, Rogers 14015; Middelburg

district : Waterval Boven, Rogers 2743; Lydenburg district: Lydenburg, Rogers 14564,

23110.

SwazILAND.—Mbabane, Rogers 11457.

BasutToLanD.—MacaMac, Drakensberg, MacLea 3017.

Locauitins UNkNown.—Burchell 6038, Rogers 17382.

22. P. repens L.

Plant suffruticose. Stems procumbent or prostrate, sparsely pilose, glandular. Leaves

trifoliolate, petiolate, stipulate ; petioles up to 3 cm. long, glandular, sparsely pilose ; young

petioles more thickly pilose ; leaflets thick, 0-5-1-5 em. long, 2-6 mm. broad, obovate to

obovate-oblong, cuneate, retuse or obtuse, nigro-punctate, sparsely pilose. Stipules up to

2 mm. long, ovate, acute. lowers axillary, 1 or more together. Pedicels equalling the

leaves, glandular, pilose, bracteolate near the middle. Bracts about 2-5 mm. long, bi-lobed.

ovate, acuminate, pilose, glandular. Calyzx-tube 2-4 mm. long, 5-7 mm. in diameter,

glabrous or laxly hairy, thickly nigro-punctate ; lobes ovate, obtuse, glabrous or villous,

ciliate, nigro-punctate, shortly pilose on the inner surface ; upper and lateral lobes 1-5-3

mm. long 1-2-5 mm. broad, lowest lobe 1-3-4 mm. long, 1-2-5 mm. broad. Vexillum 6-8

mm. long, 6-5-9 mm. broad, obovate, glandular, with a narrow claw 1-2 mm. long ;

carina 5-6 mm. long, 2-3 mm. broad, with a linear claw 3-4 mm. long; alae 6-7-5 mm.

long, 1-5-3 mm. broad, with a linear claw 2-3 mm. long. Ovary glabrous, sometimes

glandular ; style 5-7 mm. long, filiform.

Flowers: October—April.

Carp.—Cape district: Near Retreat, Bolus 3076, Kalk Bay, Bolus 3076, Moss 3340,

Wolley Dod 2216, Wynberg, and Tigerberg, Zeyher 1525, Kuils River, Zeyher 2384 ;

Humansdorp district: Humansdorp, Slang River, Phillips; Port Elizabeth district :

Port Klizabeth, Kemsley 221, Potts 303, Christie 56, Redhouse, Paterson 587 ; Uitenhage

district : Zwartkops River, Zeyher 2377, Algoa Bay, Zeyher 1526, Zeyher 668, New

Brighton, Southey.

23. P. bracteata Linn.

A low suffruticose. Stems diffuse or erect, virgate, leafly, pubescent when young,

becoming glabrous. Leaves trifoliolate, shortly petiolate, stipulate ; petioles up to 3 mm.

long; leaflets up to 2-6 em. long and 1-5 cm. broad, broadly obovate or cordate, glabre-

scent, pellucid-dotted, recurved mucronate. Stipules adnate to the base of the petiole, up

to 8 mm. long, lanceolate, ovate, acute, striate. Inflorescence terminal, spicate, having the

appearance of a capitulum, sessile. Flowers subtended by bracts variable in size and shape.

Calyx-tube 2-5-5 mm. long, 4-5-8 mm. in diameter, laxly hirsute or villous, pellucid-dotted,

127

distinctly 10-nerved ; lobes pellucid-dotted, long ciliate, laxly hirsute or villous, glabrous

or sparsely pilose on the inner surface ; upper and lateral lobes 1-5-8 mm. long, 0-5-1-5

mm. broad, lanceolate, acute, slightly sub-falcate ; lowest lobe 0-4-1 cm. long, 1-5-5 mm.

broad, broadly ovate, cuspidate, reticulately veined. Vexillwm 0-4-1-1 cm. long, 4-7 mm.

broad, obovate-oblong or sub-orbicular, slightly geniculate with a channelled claw 1-5-3-5

mm. long ; alae 0-4-1 cm, long, 1-3 mm. broad, with a linear claw 2-5-4 mm. long ; carina

2-5-4-5 mm. long, 1-2 mm. broad, with a linear claw 3-5 mm. long. Ovary glabrous or

slightly pilose ; style 3-5 mm. long, glabrous. Pod turgid 5-5 mm. 3 mm. broad, glabrous ;

seed small, purplish-brown.

Flowered: June-January. Flowers: White to blue.

CapPE.—Paarl district: Wellington, Tyson 917, Du Toit’s Kloof, Bolus 5158; Cape

district: Capetown, Paterson 50, Rogers 27003, Green Point, Zeyher, Table Mountain,

Rogers 17809, Bolus 1085, Newlands, Forbes 90, Rondebosch and Wynberg, Ecklon and

Zeyher 1543, Wynberg, Wolley Dod 19, Muizenberg, Pillans 3427, Fish Hoek, Wolley Dod

712, Rogers 16087, Cape Point, Forbes 182, Hout Bay, Galpin 3967, Lakeside, Moss and

Rogers 1633; Caledon district: Houw Hoek, Schlechter 9435; George district : George,

Schlechter 2349, 5863, Georgetown, Hops; Knysna district: Knysna, Keet 247, between

Knysna and Grahamstown, Bolus 1085; Humansdorp district: Humansdorp, Witte Els

Bosch, Fourcade 937, near Zitzikamma, Forest Dept. 71; Uitenhage district: Uitenhage,

Zeyher 393, Zwartkops River, Ecklon and Zeyher 1544, 1549, Zeyher 2372; Port Elizabeth

district: Port Elizabeth, Kemsley 131, West 245, Staadens Mountains, Zeyher, Van

Staadens River, Zeyher 2376; Albany district: Grahamstown, MacOwan 314; King-

williamstown district: Kingwilliamstown, Tyson, Rattel River, Schlechter 9723; De Aar

district: Philipstown, Hceklon and Zeyher 1548.

24. P. Bowieana Harv.

Plant suffruticose. Stems slender, 6 or more metres long, diffuse, branching, puberulous,

Leaves trifoliolate, shortly petiolate, stipulate; petioles 2 mm. long, leaflets 2-2-5 cm.

long, 6-7-5 mm. broad, cuneate-obovate, recurved-mucronate, pellucid-dotted, ciliate,

laxly hispid when young, becoming glabrous. Peduncles axillary, longer than the leaves,

racemose. Pedicels 2-3 together, equalling the calyx. Bracts lanceolate. Calyx-tube pale,

sub-glabrous ; lobes narrow-lanceolate ; lowest lobe longest ; calyx segments each marked

with three purple striae.

Care.—Cape, Bowre.

Description taken from ‘“ Flora Capensis.’

No specimen seen.

25. P. decumbens Ait.

Plants suffruticose or decumbent. Stems slender, striate, decumbent, young parts

densely clothed with cano- or rufo-pubescence, becoming glabrous. Leaves trifoliolate,

stipulate, petiolate ; petioles up to 5 mm. long ; leaflets 0-8-1-6 cm. long, 4-6 mm. broad,

cuneate or cordate, recurved-mucronate, nigro-punctate, pubescent. Stipules up to 6 mm.

long, adnate to the base of the petiole, hyaline, ovate, very acute. Flowers axillary, 1-3

together, often crowded at the ends of the branches. Pedicels up to 2 mm. long. Calyz-

tube 2-3 mm. long, 3-5-5 mm. in diameter, villous or sparsely hirsute ; lobes lanceolate,

acute, long ciliate, villous or sparsely hirsute, shortly pilose on the inner surface ; upper

and lateral lobes 1-4 mm. long, 0-5-1 mm. broad ; lowest lobe 3-6-5 mm. long, 1-2 mm.

broad ; glands present on tube and lobes, especially on the lowest lobe. Vewillum 4-5-7

mm. long, 4-6 mm. broad, obovate, with a channelled claw 1-5-3 mm. long; alae 4-5-7

mm. long, 1-5-2-5 mm. broad, with a linear claw 2-3 mm. long; carina 2-5-4 mm. long,

1-1-5 mm. broad, with a linear claw 2-5-3 mm. long. Ovary glabrous; style 3-4 mm.

long, glabrous.

128

CapE.—Piquetberg district : Piquetberg, Bolus ; Malmesbury district : Moorreesburg,

Bolus 9957 ; Tulbagh district: Tulbagh, Rogers 17436, Zeyher ; Ceres district: Mitchell’s

Pass, Schlechter 9971 ; Cape district: Table Mountain, Bolus 2770, 2771, 4272, Ecklon and

Zeyher 1538, MacOwan 73, Moss 2610, Rosebank, Pillans 3474, Rondebosch, Hutton, Forbes

179, Rondebosch and Newlands, Wolley Dod 1901, Kirstenbosch, Verdoorn, Forbes 87 ;

Stellenbosch district: Somerset Strand, Rogers 17674; Paarl district: French Hoek,

Phillips 1090; Caledon district: Sir Lowry’s Pass, Galpin 369, Houw Hoek, Schlechter

5440; Riversdale district: Riversdale, Mar 5085; Swellendam district: Swellendam,

Kennedy; Robertson district: Barrydale, Galpin 3963; Humansdorp district:

Humansdorp, Bolus 2294; Uitenhage district: Zwartkop River, Ecklon and Zeyher 1033,

Zeyher 445; Port Elizabeth district : Redhouse, Paterson 206, Humewood, Paterson 2613,

West 246; Albany district: Grahamstown, MacOwan 818, 1083, Rogers 1567, Signal Hill,

Dale and Cherry 1025, Balcock, Kaftraria, Mount Coke, Sim 1507, Bethelsdorp, Zeyher.

26. P. argentea Thunb.

Plant a slender shrub. Stems diffuse or sub-erect, leafy, clothed with adpressed whitish

pubescence. Leaves trifoholate, petiolate, stipulate: petioles 2-4 mm. long; leaflets 6-8

mm. long, 4-5 mm. broad, obovate, recurved-mucronate, minutely pellucid-dotted, thinly

canescent on both surfaces. Flowers in threes, axillary or in an interrupted terminal

pedunculate raceme. Bracts minute. Calyx-tube silky canescent; lobes lanceolate, two

uppermost semi-connate.

Description from “ Flora Capensis.”” No specimen seen.

27. P. bifiora Harv.

Plant low-growing, forming a thick carpet, much branched, pubescent when young,

becoming glabrous. Leaves trifoliolate, petiolate, stipulate; petioles up to 4 mm. long ;

leaflets thickish 2—3 mm. long, 2—2-5 mm. broad, obovate, nigro-punctate, pubescent when

young, glabrous or with a few scattered hairs when older, with distinct veining on the lower

surface. Stipules minute, ovate, withering. Flowers axillary; two sessile flowers on

peduncle. Peduncle bracteolate at apex. Bracts very small. Calyx-tube 3 mm. long,

4-5 mm. in diameter, hispid, glandular, distinctly 10-nerved ; lobes oblong, obtuse, hispid,

ciliate, glandular, pilose on the inner surface; upper and lateral lobes 1-5-2 mm. long,

0-5 mm. broad ; lowest lobe broadest, 2 mm. long, 1 mm. broad. Vezxillum 3-5 mm. long,

4 mm. broad, obovate, geniculate, claw 1 mm. long, with a slight appendage at the apex ;

alae 3-5 mm. long, 2 mm. broad, with a linear claw 2-5 mm. long; carina 2 mm. long,

1-5 mm. broad, with a linear claw 2-5 mm. long. Ovary pilose; style 1-5 mm. long, gla-

brous or very sparsely pilose.

Flowers: Blue with a white vexillum.

South West Arrica.—Keetmanshoop to Windhoek, Pearson, in Percy Sladen

Memorial Hxpedition.

28. P. obtusifolia D.C.

A small diffuse plant. Stems up to 9 metres, slender, spreading, villoso-canescent,

becoming glabrous when older; very pale. Leaves trifoliolate, petiolate stipulate ; petioles

usually fairly long, 0-5-7 em. long ; leaflets 0-3-2-1 cm. long, 0-2-1-5 em. broad, obovate

or elliptic obtuse, thinly canescent, glandular, plicate, with margins slightly dentate.

Stipules up to 3 mm. long, ovate, acute, ciliate, canescent. Inflorescence axillary, spicate.

Peduncles with 3-8 flowers. Flowers sub-sessile. Calyax-tube 1-5-4 mm. long, 3-5 mm.

in diameter, densely albo-villous; lobes lanceolate-ovate, ciliate, villous, pilose on the

inner surface; upper and lateral lobes 1-2 mm. long, 0-5-1 mm. broad; lowest lobe

2-4-5 mm. long, 1-1-5 mm. broad. Vemrillwm 3-5 mm. long, 3-4 mm. broad, obovate,

with a channelled claw 1-1-5 mm. long; alae 3-4-5 mm. long, 1-2 mm. broad, with a

linear claw 1-5-4 mm. long; carina 2-3 mm. long, 1-1-5 mm. broad, with a linear claw

2-2-5 mm. broad. Ovary villous; style 1-3 mm. long, glabrous or sparsely pilose.

— —

129

CarE.—Cape district: Newlands, Wilman; Swellendam district: Gauritz River,

Zeyher 450 ; Calvinia district : Near Twee River, Burtt-Davy 17562 ; Graaff-Reinet district :

Graafi-Reinet, Bolus 127 ; Beaufort West district: Murraysburg, Bolus 35: Kimberley

district: Barkly West, W.G.B. 637, Bolus 6814, near Vaal River, Bolus 6803; Vryburg

district : Vryburg, Sharpe; Kuruman district: Kuruman, Burtt-Davy, Langbere, Lanham

4032. “ata

South Wesr Arrica.—Great Namaqualand, Ramansdrift, Schlechter 11459 ;

Holoog, Pearson 9808; Kiubis, Pearson 8012, 8026; Tsubgaus, Pearson 9204; Great

Fish River, Pearson 9210, 9805, 9274; Tsondat, Pearson 9189; Kalahari, Ireland 16134:

Krai Kluft Ravine, Pearson 8280; Walvis Bay, Marloth 1481.

NAMAQUALAND Minor.—Near Henkies, Phillips 1587.

TRANSVAAL.—Pretoria district: Pienaar’s River, Schlechter 4211; Bloemhof district :

Christiana, Burtt-Davy ; Zeerust district: Zeerust, Potts 4257, Burtt-Davy.

ORANGE FREE Srate.—Boshof District: Burtt-Davy ; Bloemfontein district : Bloem-

fontem, Burtt-Davy ; Heilbron district, Heilbron, Flanagan 1459.

Locatity Unknown.—Dinter 125.

29. P. rotundifolia Linn.

A suffruticose about 3 metres high. Stems ascending, leafy, densely hispid, becoming

glabrous when older. Leaves unifoliolate, petiolate, stipulate, 0-9-2-7 em. broad, 0-9-4

em. long, ovate to ovate-elliptic, obtuse or acute, mucronulate, sparsely glandular, with

reticulate veining, glabrous; petioles up to 5 mm. long. Stvpules adnate to the base of the

petiole, equalling or exceeding the petiole, lanceolate, subulate veined, glabrous, glandular.

Inflorescence spicate, many-flowered. Peduncles terminal and axillary, 4-6 mm. long,

hairy. Flowers on short pedicels, subtended by a small bract. Calyx-tube 3 mm. long,

7 mm. in diameter, densely hirsute with long white hairs, glands translucent ; lobes lanceo-

late-ovate, long hirsute, ciliate, thinly pilose on the inner surface ; upper and lateral lobes

6-7 mm. long, up to 2 mm. broad ; lowest lobe 1-1 cm. long, 3mm. broad. Vexillum about

1 em. long and 5 mm. broad, obovate, with a channelled claw 2 mm. long ; alae 8 mm. long,

2-5 mm. broad, with a linear claw 3-5 mm. long; carina 5-5 mm. long, 2-5 mm. broad,

with a linear claw 4 mm. long. Ovary hispid; style 4 mm. long. Pod membraneous,

about 6 mm. long and 5 mm. broad, hispid.

Flowers : July—Apmil.

Caprr.—Caledon district: Houw Hoek, Bolus 80, near Klein River; Clanwilliam

district: Vogelgat, Schlechter 9503, 10417.

30. P. Thomii Harv.

A small suffruticose. Stems slender, ascending, hairy. Leaves unifoliolate, petiolate,

stipulate, 1-4 em. long, 0-5-2-15 em. broad, elliptic-lanceolate to elliptic-ovate or oblong,

mucronulate, with reticulate veining, no glands present, densely hairy when young and

never becoming quite glabrous; petioles up to 5 mm. long. Stipules about 7 mm. long

and 2 mm. broad, ovate, acute, nerved, hispid. Inflorescence spicate. Peduncles terminal

and axillary, bracteate, hairy, exceeding the leaves. Flowers sessile or sub-sessile. Calya-

tube 3-3-5 mm. long, 7 mm. in diameter, veined, thickly hirsute ; lobes linear-lanceolate

to ovate-lanceolate, acute, hirsute, glabrous on the inner surface; upper and lateral

lobes 5-9 mm. long, 1-5-2 mm. broad; lowest lobe 0-8-1 em. long, 2-5-3 mm. broad.

Vexillum 1-15-1-4 cm. long, 8-9 mm. broad, obovate with a linear claw 1-5 mm. long ;

alae 0-9-1 em. long, 3 mm. broad, with a linear claw 4 mm. long; carina 5-6 mm. long,

2-5-3 mm. broad, with a linear claw 4-5-5 mm. long. Ovary pilose ; style 4-5 mm. long,

pilose.

Flowers: July—October.

Capr.—Caledon district: Hermanus, Bolus 6378, Pieter’s Fontein, Bodkin (Bolus

6378).

31. P. aculeata L.

An erect virgate shrub up to 12 metres. Stems woody, branched, leafy, glabrous, glan-

dular, spiny with persistent stipules. Leaves trifoliolate, petiolate, stipulate ; petioles 1-5-4

mm. long, glandular, glabrous ; leaflets thickish, 1-5-8 mm. long, up to 2 mm. broad, cuneate,

nigro-punctate on both surfaces, with a sharply curved mucro. Stipules adnate to the

base of the petiole, 2-3 mm. long, persistent, subulate. Flowers axillary, solitary or 1-3

together. Pedicels up to 1-3 cm. long, bracteolate near the apex, often crowded on the

upper parts of the stem. Bracts about 5 mm. long, connate, ovate, acuminate, nigro-

punctate. Calyx-tube 3-5-5 mm. long, 0-8-1-1 cm. in diameter, glabrous, thickly glandular,

nerved ; lobes ovate, acuminate, glabrous, glandular, pilose on the inner surface, upper and

lateral lobes 2-5-6 mm. long, 1-5-3-5 mm. broad ; lowest lobe 0-3-1 em. long, 2-5-5 mm.

broad. Vexillum 0-8-1-3 cm. long, 0-8-1-5 cm. broad, obovate, geniculate, with a chan-

nelled claw 1-3 mm. long; alae 0-7—1 cm. long, 3-5-6 mm. broad, with a linear claw 2-5-5

mm. long; carina 4-8 mm. long, 8-5 mm. broad, with a linear claw 3-5 mm. long. Ovary

1-2 mm. long, glabrous; style 4-5-8 mm. long, glabrous.

Flowers: August-November.

CapE.—Cape district: Table Mountain, Heklon and Zeyher 1545, MacOwan, Galpin

3966, Bolus 4660, Rogers 17744, Orange Kloof, Wolley Dod 20; Stellenbosch, Hottentots

Holland Mountains, near Gordon’s Bay, Bolus 9865, 9933; Clanwilliam district: near

Vogelgat, Schlechter 9516.

32. P. Keetii Schonl.

A shrub. Stems erect, glabrous, glandular. Leaves trifoliolate, petiolate, stipulate ;

petioles up to 1 cm. long ; leaflets 0-5-2-5 mm. long, 0-4-1-1 em. broad, obovate, thickly

nigro-punctate, glabrous, mucronulate ; terminal leaflet on short petiole ; lateral leaflets

sub-sessile. Strpules not exceeding 5 mm. long, ovate, lanceolate, acute. lowers axillary,

solitary. Pedicels up 3-5 cm. long, bracteolate at the apex. Bracts tri-lobed, 2-3 mm.

long, ovate, acuminate, glandular, glabrous. Calyx-iube 3 mm. long, 8 mm. in diameter,

glabrous, thickly glandular; lobes ovate, acute, glandular, glabrous, ciliate, pilose on the

inner surface ; upper and lateral lobes 2-4 mm. long, 1-5 mm. broad ; lowest lobe slightly

longer. Veaillum 8 mm. long, 1 em. broad, obovate ; claw channelled, 3 mm. long, with

two slight appendages at the apex; alae 8 mm. long, 4 mm. broad, with a linear claw 4

mm. long; carina 6 mm. long, 4 mm. broad, with a linear claw 6 mm. long. Ovary gla-

brous ; style 7 mm. long, glabrous.

toy te)

Flowered: January. Flowers: Colour, light blue.

Cape.—Knysna district : Hoogeberg, Keet 1055 (Type).

33. P. polysticta Benth.

A shrub, up to 10-5 metres. Stems woody, erect, striate, hairy, becoming almost

glabrous when older, glandular. Leaves trifoliolate, sub-sessile or very shortly petioled,

stipulate ; petioles about 3 mm. long; leaflets 0-9-3-6 em. long, 0-3-1-5 cm. broad,

oblong, cuneate, nigro-punctate, recurved-mucronate, glabrescent. Stipules small, deltoid.

Flowers axillary in clusters. Peduncles 2-4 mm. long with 3 or more short pedicels to

each peduncle. Calya-tube 2-3-5 mm. long, 5-7 mm. in diameter, veined, tomentulose,

glandular; lobes lanceolate, acute, tomentulose, glandular, pilose on the inner surface ;

upper and lateral lobes 1-3-5 mm. long, about 1 mm. broad ; lowest lobe 4-5 mm. long,

1-1-5 mm. broad. Veaxillwm 6-8 mm. long, 5-6 mm. broad, obovate, with a short claw

0-5-2 mm. long: alae 5-7 mm. long, 1-5-2 mm. broad, with a linear claw 2-5-3 mm. long ;

carina 38-4 mm. long, 1-5-2 mm. broad, with a linear claw 3-4 mm. long. Ovary glabrous ;

style 3-5 mm. long.

Flowers : August—April.

131

The following note is attached to Mrs. Dieterlen’s specimen No. 372 in the S.A.

Museum :—

“ Sesuto—Mohlonecha.—He who makes respectable. A chief who wishes to have

prestige in the sight of his subjects must bathe the whole of his body with a decoction

of this plant. The burnt root is smoked as a cure for cold in the head.”

Care.—Uitenhage district: Uitenhage, Drege: Oudtshoorn district: Oudtshoorn,

Taylor 323; Albany district: Zuurberg, Paterson 44, Griqualand East, Mount Currie,

Tyson 1319, Dornkop, Zeyher, Tyson 519.

TRANSVAAL.—Pietersburg district: Pietersburg, Moss and Rogers 1330; Messina

district: Louis Trichardt, Rogers 21148 ; Heidelberg district : Heidelberg, Burtt-Davy 15435.

SwaziLtanp.—Hlatikulu, Stewart 181.

ORANGE FREE State.—Cooper 1114.

BasuToLAND.—Leribe, Dvieterlen 372.

Narat.—Ladysmith, Wood 8010; Estcourt, Wood 9953; Weenen County: Blaau-

krans, Wood 3517; Boston, Wood 9900. :

No Locauiry.—Gerrard and McKen 440.

34. P. carnea HE. Mey.

A shrub 12-15 metres high. Stems leafy, reddish -pubescent, at length becoming

glabrous. Leaves trifoliolate, stipulate, sub-sessile ; leaflets up to 1-5 cm. long and 7 mm.

broad, oblong to cuneate-oblong, obtuse, pellucid-dotted, recurved mucronate, glabrous.

Stipules about 2 mm. long, deltoid. Flowers axillary, 2-3 together. Pedicels very short.

Calyx-tube 2-3-5 mm., 5-6 mm. in diameter, veied, hirsute, glandular; lobes linear-

lanceolate, lanceolate or ovate, acute, glandular, ciliate, hirsute, glabrous on the inner

surface ; upper and lateral lobes 2-3-5 mm. long, 0-5-1 mm. broad ; lowest lobe 3-5 mm.

long, 1-1-5 mm. broad. Veaxillum 6-7 mm. long, 5-6 mm. broad, obovate, with a chan-

nelled claw 1-2 mm. long; alae 5-5-7 mm. long, 2-2-5 mm. broad, with a liner claw

2-5=3 mm. long; carina 3-4 mm. long, 1-5-2 mm. broad, with a linear claw 3-4

mm. long. Ovary glabrous or very slightly pilose; style 2-4 mm. long, glabrous.

Flowers: August—March.

Cape.—Riversdale district: Riversdale, Muir 484, Langebergen, Schlechter 5691 ;

Oudtshoorn district : Oudtshoorn, Britten 80; Knysna district: Knysna, Keet 278, 2716,

Fourcade 1519, Plettenberg Bay, Rogers 15513, Clarkson, Galpin 3961, Schlechter 6012 ;

Humansdorp district: Humansdorp, Fourcade 2330, Assegai Bosch, Fourcade 857, Schon-

land 3618, Rogers 2833; Albany district: Howison’s Poort, MacOwan.

35. P. obliqua E. Mey.

An erect shrub up to 15 metres. Stems branched, villous when young, at length

glabrescent, leafy. Leaves trifoliolate, petiolate, stipulate ; petioles up to 6 mm. long ;

leaflets up to 3-1 cm. long and 1 ecm. broad, elliptic-oblong, obtuse, mucronulate, densely

glandular, glabrescent, shiny ; lateral leaflets unequal sided. Stipules membraneous, about

6 mm. long and 2-5 mm. broad, ovate. Flowers axillary, in threes. Pedicels shorter than

the leaves ; sometimes three pedicels on short peduncle. Calyx-tube 2 mm. long, 7-8 mm.

in diameter, nerved, glandular, villous; lobes villous, glandular, ciliate, glabrous on the

inner surface ; upper and lateral lobes 4-5 mm. long, 1-1-5 mm. broad, lanceolate, acute ;

lowest lobe up to 8 mm. long, and 3 mm. broad, ovate, very acute, almost cuspidate.

Vewillum 9 mm. long, 7-8 mm. broad, obovate, with a channelled claw 1 mm. long ; alae

up to 1 cm. long, 3-4 mm. broad, with a linear claw 2-3 mm. long; carina 3-4 mm long,

2 mm. broad, with a linear claw 2-3 mm. long. Ovary glabrous; style 2-5 mm. long,

glabrous.

Flowers: July—November.

Capr.—Stellenbosch district: Stellenbosch, Duthie: Caledon district: near Palmiet

River, Zeyher 1525, Hottentots Holland Mountains, Zeyher ; Paarl district: French Hoek,

Phillips 1088.

132

36. P. macradenia Harv.

A shrub. Stems branched, erect, leafy, striate, glandular, silky canescent when young,

becoming glabrous when older. Leaves trifoliolate, stipulate, sub-sessile ; leaflets 1-9 cm.

long, 8-5 mm. broad, obovate, cuneate, recurved mucronate, canescent when young,

becoming glabrous, nigro-punctate on the upper surface with large and prominent glands

on the lower surface. Stipules up to 2-5 mm. long, subulate, acute, canescent. Flowers

axillary, 1-3 together. Pedicels short; sometimes three sub-sessile flowers on a short

peduncle. Calyx-tube 3 mm. long, 7 mm. in diameter, veined, canescent, very sparsely

glandular; lobes lanceolate-ovate, acute, sparsely glandular and canescent; upper and

lateral lobes 3-4 mm. long, 1-1-5 mm. broad ; lowest lobe 5-5 mm. long, 1-5 mm. broad.

Vexillum 5 mm. broad, 6:5 mm. long, obovate, with a channelled claw 1-5 mm. long ; alae

6-5 em. long, 2 mm. broad, with a linear claw 3 mm. long ; carina 4 mm. long, 2 mm. broad,

with a linear claw 3 mm. long. Ovary glabrous; style 3-5 mm. long, glabrous.

Flowers: July—August.

Care.—Riversdale district: Langebergen, Mwy 1113.

37. P. tomentosa Thunb.

A shrub up to 24 metres. Stems erect, leafy, densely cano-pubescent, at length becom-

ing glabrous. Leaves trifoliolate, petiolate, stipulate ; petioles up to 2-5 cm. long ; leaflets

0-6-4-5 em. long, 0-5-1-8 em. broad, elliptic or elliptic-lanceolate, mucronulate, with the

upper surface glabrous or sparsely pilose and the lower surface densely canescent. Stzpules

up to 1-3 em. long and 3 mm. broad, lanceolate, acute or subulate, cano-pubescent or

glabrescent. Inflorescence axillary, or terminal. Peduncles up to 14 cm. long. Flowers

im dense globose heads, sub-sessile ; each flower subtended by a bract. Bracts up to 1 cm.

long, ovate or lanceolate, villous, striate with dark lines. Calya-iube 3-7 mm. long,

0-7-1-3 cm. in diameter, with reticulate veming, villoso-hirsute ; lobes lanceolate, acute,

reticulately veined, villous, pilose on the inner surface, ciliate; upper and lateral lobes

3-7 mm. long, 1-2-5 mm. broad ; lowest lobe 0-75-1 cm. long, 2-3 mm. broad. Veaillwm

7-8 mm. long, 2-8 mm. broad, obovate, geniculate, distinctly veined, with a channelled

claw 2-5 mm. long ; alae 4-5-8 mm. long, 2-3-5 mm. broad, with a linear claw 3-5-6 mm.

long; carina 3-5 mm. long, 2-5 mm. broad, with a lmear claw 5-6-5 mm. long. Ovary

glabrous; style 5-7 mm. long. Glands present but very minute. The whole plant has

a silvery appearance owing to the silky cano-pubescence.

Flowers: August—January.

Capr.—George district: Outeniqua Mountains, Hops (Bolus 11795), Schlechter 5771,

Pappe; Riversdale district: Riversdale, Muir; Knysna district: Knysna, Duthe 517,

Keet 1025, Williamson 134, Tyson 973 Keet 856, Zeyher, Plettenberg Bay, Rogers 15450,

27061; Humansdorp district: Humansdorp, Galpin 3958, West 247, Rogers 2910, Witte

Els Bosch, Fourcade 899, Ratels Bosch, Fourcade 69; Uitenhage district: Uitenhage,

Zeyher ; Albany district: Grahamstown, MacOwan 26, Featherstone’s Kloot, Bolus 1939,

Trapp’s Valley, Daly 601, Howison’s Poort, Britten 2980, Glass 17, Highlands, Paterson ;

Prince Albert district: Zwartberg Pass, Bolus 11476.

Locatities Uncertain: Kleinmonde, White 946a, Inalgaten River, Young 5516.

38. P. zeyheri Harv.

A suffruticose, about 3 metres high. Stems slender, ascending, pubescent. Leaves

trifoliolate. petiolate, stipulate ; petioles up to 1-5 em. long, striate, pubescent ; leaflets

0-7-4:2 em. long, 0-4-1-5 cm. broad, broadly obovate, mucronulate, villous, becoming

glabrescent ; a few of the upper leaflets lmear to ovate-linear. ‘Stepules adnate to the base

of the petiole, up to 1-3 cm. long, subulate or lanceolate, acute, hispid, ciliate. Inflo -

escence terminal, and axillary, spicate. Peduncles 5-5-16 cm. long, hirsute. Flowers

sub-sessile, each subtended by a bract. Bracts ovate or linear, acute. Calyx-tube 3-4

mm. long, 5-6 mm. in diameter, with dark and reticulate veins, densely albo-hirsute,

lobes albo-hirsute, glabrous on the inner surface; upper and lateral lobes 3-8 mm. long,

0-5-1 mm. broad, linear-lanceolate; lowest lobe 5-9 mm. long, 1:5-2-5 mm. broad,

lial neal

133

reticulately veined, ovate-lanceolate. Vexillwm 0- 6-1-2 cm. long, 5-7 mm. broad, obovate,

with a channelled claw 0-5-3 mm. long; alae 6-9 mm. long, 1-5-3 mm. broad, with a

linear claw 1-4-5 mm. long; carina 3-5 mm. long, 1-2 mm. broad, with a linear claw

4-5 mm. long. Ovary glabrous; style 6 mm. long, glabrous.

Flowers: November—January.

Cape.—Stellenbosch district: Palmiet River Mountains, Stokoe 17507: Caledon dis-

trict: Caledon, Zeyher 2375, Houw Hoek, Bolus 6935, near Sir Lowry’s Pass, Schlechter

9505, near Rooi Els, Stokoe 17342.

39. P. Patersoniae Schonl.

A shrub. Stems erect, striate, glandular-scabrid, tomentose, at length glabrescent.

Leaves trifoliolate, stipulate, petiolate ; petioles 3-5-5 cm. long ; leaflets 1-5-5-5 em. long,

up to 4 cm. broad, obovate, acute, cuneate at the base, dentate-creniculate, mucronulate,

nigro-punctate, glabrescent ; lateral leaflets sub-sessile : terminal leaflets on petioles 0-7-2

em. long. Stipules up to 1 cm. long, lanceolate, ovate, acute, tomentulose, glandular.

Inflorescence axillary, spicate. Peduncles up to 6 mm. long. Flowers 2-3, subtended by

a bract, shortly pedicelled. Bracts up to 1 em. long, ovate, lanceolate, acute, tomentulose,

glandular. Calyx-tube 2 mm. long, 5 mm. in diameter, tomentose, glandular ; lobes lanceo-

late, ciliate, tomentose, glandular, glabrous on the inner surface ; upper and lateral lobes

2-3 mm. long, 0-7 mm. broad ; lowest lobe 5 mm. long, 1-5 mm. broad. Vexillwm 5 mm.

long, 4 mm. broad, obovate, with a channelled claw 1-5 mm. long; alae 5 mm. long, 1-5

mm. broad, with a linear claw 3 mm. long ; carina 3 mm. long, 2 mm. broad, with a linea»

claw 3 mm. long. Ovary glabrous; style 3-5 mm. long, glabrous.

Flowers: December. :

Carr.—Port Elizabeth district: Redhouse, Paterson 393 (Type).

40. P. caffra E. and Z.

An erect shrub up to 6 metres. Stems leafy, striate, with prominent glands, tomentu-

lose, at length glabrescent. Leaves trifoliolate, petiolate, stipulate ; petioles up to 1-8 em.

long ; leaflets 1-4-5 em. long, 0-7-3 em. broad, broadly elliptic or obovate, rounded at

the apex, mucronulate, thickly nigro-punctate, tomentulose, at length glabrescent ; terminal

leaflet on a petiole up to 1-1 cm. long. Stipules 2-5 mm. long, deltoid, acuminate, pilose.

Inflorescence terminal or axillary, laxly or densely spicate. Peduncles up to 5 cm. long.

Flowers in whorls of three subtended by a bract. Bracts up to 3 mm. long, ovate,

acuminate, pilose. Pedicels about 1 mm. long. Calyz-tube2-3 mm. long, 4:5-6 mm.

in diameter, nerved, tomentose, nigro-punctate; lobes lanceolate, acute, tomentose,

nigro-punctate, pilose on the inner surface; upper and lateral lobes 1-2-5 mm. long,

0-5-1 em. broad ; lowest lobe 2-4 mm. long, 1-1-5 mm. broad, acuminate. Vexillwim 5-7

mm. long, 5-7 mm. broad, obovate, with a channelled claw 0-7—2:5 mm. long; alae

5-6-5 mm. long, 1-5-2-5 mm. broad, with a linear claw 2-5-4 mm. long; carina

3-3-5 mm. long, 1-5-2 mm. broad, with a linear claw 3-4-5 mm. long. Ovary pilose ;

style 3-4 mm. long, pilose. Pod membraneous, 5-5-7 mm. long, 3-4 mm. broad, hirsute,

enclosed in calyx; seed small, hard and brown.

Whole plant has large conspicuous wart-like glands.

Capr.—Cape district: Fort Beaufort district : Hogsback, Rattray 324 ; Queenstown

district: Katberg, Bolus (Herb. 1993), Galpin 2068, between Kat and Kei Rivers, Zeyher

1550; Cathcart district: Cathcart, Galpin 7361; Stutterheim district; Stutterheim,

Mountains above Dohne, Flanagan 2294; Komgha district: Komgha, Flanagan 2319 ;

Cala district: Kolbe and Pegler 1740; Griqualand East: Clydesdale, Tyson 1220, 1260,

2526.

Naraut.— Drakensberg, Giants Castle, Wylie (Wood 10674); Polela, Hvans 659; Mah-

waqua Mountains, Hvans 225; Lidgetton, Wood 6320; Byrne and Richmond, Wood 1847 ;

Maritzburg, Hutton 315; Umgeni, Cooper 1148; Umkomaas, Wood ; Zuurberg, Wood 338.

TRANSVAA‘L.—Lydenburg district: Hlandspruitberg, Schlechter 3889, Pegler 1740 is

probably from Kentani.

134

41. P. venusta Ii. and Z.

A. suffruticose. Stems flexuous, appressedly pubescent. Leaves trifoliolate, stipulate,

sub-sessile ; leaflets 2-5 em. long, 6-7 mm. broad, cuneate-oblong, obtuse, mucronulate,

pellucid-dotted, appressedly puberulent. Stipules subulate. Inflorescence terminal and

axillary, spicate, cylindric. Pedwncle and spike up to 4 cm. long. Bracts subtending

flowers half as long as the calyx. Calyx-tube 6 cm. long, silky canescent, lobes lanceolate,

acuminate, glandular; lowest lobe longest, nearly equalling the petals.

CapE.—Malmesbury district: Saldanha Bay, Ecklon and Zeyher 1553.

Description taken from “ Flora Capensis.” No specimen seen.

42. P. hirta Linn.

A branched shrub. Stems erect, rigid, canescent with appressed short hairs, sparsely

leafy. Leaves trifoliolate, petiolate, stipulate ; petioles not exceeding 5 mm. long ; leaflets

0-5-1-5 cm. long, 4-9 mm. broad, cuneate, obovate, recurved-mucronate, canescent, glan-

dular. Stipules adnate to the base of the petiole, small, subulate. Inflorescence axillary

or in interrupted spikes. Flowers in whorls of three, sessile, bracteate. Calyzx-tube 3-4

mm. long, 6-7 mm. in diameter, glandular, villous; lobes slightly unequal, 1-5 mm. long,

0-75-2 mm. broad, linear-lanceolate to lanceolate-ovate, acute, villous, pilose on the inner

surface. Vexillum 5-7 em. long, 4-5 mm. broad, obovate, with a claw 1-3 mm. long ;

alae 4-6-5 mm. long, 1-2-5 mm. broad, with a linear claw 3 mm. long; carina 3-4 mm.

long, 1-2 mm. broad, with a linear claw 3-4 mm. long. Ovary glabrous or hirsute; style

3-4 mm. long, glabrous or hirsute.

Flowers: October-November.

CarE.—Tulbagh district: Winterhoek, Zeyher; Cape district: Blaauwberg, Zeyher,

near Milnerton, Forbes 229; without locality, Bolus 2993; Stellenbosch district: Stellen-

bosch, Heklon and Zeyher, Somerset Strand, Rogers 17675; Paarl district: Paarlberg

Drege ; Caledon district: Sir Lowry’s Pass, Bolus 9195.

43. P. stachydis L.f.

A branched shrub. Stems erect, rigid, villous, with spreading rusty hairs, becoming

glabrous when older. Leaves trifoliolate, petiolate, stipulate ; petioles up to 3 mm. long :

leaflets 0-5-2:5 cm. long, 0:4-1-5 mm. broad, obovate, acute or obtuse, hirsute, recurved

mucronate, glandular. Stipules adnate to the base of the petiole, subulate. Inflorescence

axillary and terminal, in laxly flowered spikes. Peduncles up to 5 cm. long. Flowers in

whorls of three. Bracts small. Calyx-tube 3-4 mm. long, 6 mm. in diameter, villous,

glandular ; lobes ovate-acuminate or sub-lanceolate, villous, glandular, glabrous or sparsely

pilose on the inner surface ; upper and lateral lobes 2-4 mm. long, 0-5-1-5 mm. broad :

lowest lobe 3-5-5 mm. long, 1-5-2 mm. broad. Veaillum 7 mm. long, 5-7 mm. broad,

obovate, with a linear claw 1-3 mm. long; alae 5:5-6:5 mm. long, 2-3 mm. broad, with

a linear claw 3-4 mm. long; carina 3-4 mm. long, 1-5-2 mm. broad, with a linear claw

3-4 mm. long. Ovary thinly pilose or glabrous; style 4-5 mm. long, glabrous.

Flowered : October.

Capr.—Piquetberg district: Piquetberg, Bolus 3458, 7529; Cape district: Table

Mountain, Zeyher.

44. P. spicata L.

A shrub. Stems erect, virgate, pubescent, glabrescent, glandular, leafy. Leaves

trifoliolate, petiolate, stipulate ; petioles up to 4 mm. long; leaflets up to 3 cm. long, 9

mm. broad, obovate or obovate-oblong, obtuse, recurved mucronate, thickly nigro-punctate,

slightly hirsute, especially along the mid-rib and veins. Stipules adnate to the base of the

petiole, up to 5 mm long, subulate, acute, glandular. Inflorescence terminal, sessile, a

135

densely flowered spike. Bracts small. Calyx-tube 3-4 mm. long, 5-8 mm. in diameter,

veined, pubescent, glands few; lobes lanceolate-ovate, acute, long ciliate, more thickly

glandular than the tube, shortly pilose on the inner surface ; upper and lateral lobes 1-5-3-5

mm. long, 1-2 mm. broad ; lowest lobe 3-5 mm. long, 1— 2mm. broad. Vesillum 4-6 mm.

long, 4-5-5 mm. broad, obovate, slightly geniculate, with a channelled claw 2 mm. long ;

alae 3-5-6 mm. long, 1-5-2 mm. broad, with a linear claw 2-5-3 mm. long; carina 2-3

mm. long, 1-1-5 mm. broad, with a linear claw 3-4 mm. long. Ovary y glabrous ; style 3-4

mm. long, glabrous.

Flowered : September—February. Flowers: Colour, pale blue. Plants about 2-3

feet high.

Capr.—Cape district: Capetown, Bolus 1429, near “‘ Protea,’ Bolus (Herb. 7101),

Wynberg, Bolus (Herb. 71018), Table Mountain, Zeyher (?) (Ex. Herb. Gub.); Paarl dis-

trict: French Hoek, Phillips 1091; Swellendam district: Swellendam, Zeyher 2374;

Riversdale district: Riversdale, Mwr 83; George district: George, Paterson; Knysna

district: Forest Hill, Duthie 740, Plettenberg Bay, Ecklon and Zeyher 1546; Albany

district : Zuurberg, Schlechter 5702, Hofman’s Bosch, Britten 1028; East London district :

East London, Rattray 186; Komgha district: Komgha, Flanagan 403.

45. P. racemosa Thb.

A suffruticose. Stems flexuous, slender, pubescent, becoming more or less giabrous,

glandular, leafy. Leaves trifoliolate, shortly petiolate, stipulate ; petioles up to 2-5 mm.

long ; leaflets up to 1-8 cm. long, 8 mm. broad, cuneate-oblong, flat, obtuse, mucronulate,

coriaceous, hispidulous, glabrescent, nigro-punctate, especially near the margins. Stipules

up to 4 mm. long, subulate, veined. Inflorescence terminal, interrupted or axillary.

Flowers sessile, 2-3 together. Calyx-tube 3'mm. long, 5 mm. broad, veined, glands few,

translucent,’ hispid along veins, otherwise glabrous; lobes linear-lanceolate, acute, ciliate,

hispid, glands translucent, shortly hispid on the inner surface; upper and lateral lobes

3-5-4 mm. long, up to 1 mm. broad ; lowest lobe 5-5 mm. long, 1:5 mm. broad. Veaxillum

8 mm. long, 6-5 mm. broad, obovate, with a claw 1-5 mm. long; alae 9 mm. long, 3 mm.

broad, with a linear claw 2-5 mm. long; carina 3 mm. long, 2 mm. broad, with a linear

claw 3-5 mm. long. Ovary glabrous; style 3 mm. long, glabrous.

Flowers: October.

Cape.—Riversdale district : Garcia’s Pass, Bolus 11264; George district: Langekloof,

Ecklon and Zeyher.

46. P. hamata Harv.

Ashrub. Stems erect, striate, virgate, with a few scattered glands, canescent, becoming

glabrous. Leaves trifoliolate, sub-sessile, stipulate ; leaflets up to 2 cm. long and 1 cm.

broad, obovate, obtuse, recurved-mucronate, thickly glandular, canescent when young,

becoming glabrous. Stipules up to 2-5 mm. long, subulate, acute. Inflorescence terminal,

spicate, sessile. Calyx-tube 3 mm long, 7 mm. in diameter, glandular, canescent ; lobes

ovate, acuminate, glandular, canescent, shortly pilose on the inner surface; upper and

lateral lobes 1-5-3 mm. long, up to 1-5 mm. broad ; lowest lobe 4 mm. eas 2 mm. broad.

Vexillum 5 mm. long, 4:5 mm. broad, obovate, with a channelled claw 2:5 mm. long ; alae

4-5 mm. long, 2 mm. broad, with a linear claw 3 mm. long; carima 3 mm. long, 2 mm.

broad, with a linear claw 4 mm long. Ovary pilose ; style 4mm. long. Pod membraneous,

pubescent, with persistent calyx ; seed small, hard and brown.

Flowered: January.

Capr.—Worcester district: Hex River, near De Doorns, Bolus.

NamaquaLaNnp Mryor.—Without precise locality, Scully 1137.

136

47. P. Wilmsii Harms.

A tall shrub. Stems thinly pubescent, leaves 2-3 cm. equidistant. Leaves trifoliolate,

petiolate, stipulate ; petioles up to 4 mm. long; leaflets up to 4:5 em. long and 1:3 cm.

broad, oblanceolate to oblanceolate-ovate, obtuse, recurved mucronate, nigro-punctate,

thinly pubescent, becoming glabrous, sessile or sub-sessile. Stipules lanceolate, acuminate,

glandular, thinly pubescent. Inflorescence terminal and axillary, spicate. Peduncles up to

2-8 mm. long. Flowers in threes subtended by a common bract. Calyx-tube 2-5 mm.

long, 5:5 mm. in diameter, veined, glandular, hirsute; lobes ovate, glandular, hirsute,

ciliate, hirsute on the inner surface ; upper and lateral lobes 1-5-2 mm. long, up to 1 mm.

broad ; lowest lobe 2-5-3 mm. long, 1-1-5 mm. broad. Veaxillum 6-6-5 mm. long, 5-5-6

mm. broad, ovate or obovate, with a linear claw 1-5-2-5 mm. long; alae 6 mm. long,

2-2-5 mm. broad, with a linear claw 2-5-3 mm. long; carina 3-4 mm. long, 1-5-2 mm.

broad, with a linear claw 3-3-5 mm. long. Ovary pubescent ; style 3-5-4 mm. long, pilose.

Flowers: July—January.

TRANSVAAL.—Zoutpansberg district : The Downs, Rogers 20209 ; Lydenburg district :

Elandspruitberg, Schlechter 3889; EKrmelo district: Mavierstad, Pott 5086.

SwazILAND.—Without precise locality, Bolus 11794.

48. P. striata Thb.

A virgate shrub. Stems erect, striate, leafy, canescent. Leaves trifoliolate, petiolate,

stipulate ; petioles not exceeding 5 mm. ; leaflets 0-7-3-5 mm. long, 0-4-1-7 em. broad,

obovate to obovate-oblong, recurved mucronate, nigro-punctate, thinly canescent, becoming

almost glabrous when older. Stipules lanceolate, acute, nigro-punctate, canescent.

Infloresecence terminal and axillary, spicate. Spikes sometimes lax and interrupted.

Peduncles up to 2-5 cm. long. Calyx-tube 2-5-4 mm. long, 6-7 mm. in diameter, glandular,

canescent ; lobes ovate to lanceolate-ovate, acute, glandular, canescent, glabrous or pilose

on the inner surface ; upper and lateral lobes 1-3 mm. long, 0-75-1:5 mm. broad ; lowest

lobe 1-5-5 mm. long. Veaillum 6-7 mm. long, 4-5 mm. broad, obovate, with a claw 1-3

mm. long ; alae 4-5 mm. long, 1-5-2 mm. broad, with a linear claw 3-4 mm. long; carina

3-4 mm. long, 1-5-2 mm. broad, with a linear claw 3-4 mm. long. Style 3-4 mm. long,

pilose or glandular. Pod membraneous, 1-4 cm. long, 4 mm. broad, pubescent, with the

persistent calyx ; seed up to 1 cm. long, hard and brown.

Care.—Clanwilliam district: Olifant’s River, Zeyher 1556; Swellendam district :

Gauritz River, Zeyher 1551; Oudtshoorn district: Oudtshoorn, Britten 134.

Var. gracilis. Plant more slender and laxly leafy. Flowers fewer and smaller, 1-3

subtended by a common bract. Flowers on long terminal rhachi.

NamaquaLanp.—Between Brakdam and Rietkloof, Pillans 5667.

49. P. Royfiei n. sp.

Stems striate, younger parts clothed with yellowish pubescence. Leaves trifoliolate,

petiolate, stipulate ; petioles up to 6 mm long; leaflets up to 2-3 cm. long and 1-3 cm.

broad, elliptic-lanceolate to elliptic-obovate, mucronulate, densely nigro-punctate, pub-

escent on lower surface. Stipules up to 5 mm. long, subulate, acute, glandular, pubescent.

Inflorescence terminal, spicate. Peduncles up to 1 cm. long. Flowers sessile or very shortly

pedicelled, 2-3 together subtended by a bract. Calyx-tube 2-5 mm. long, 6 mm. in diameter,

pubescent, glandular ; lobes acute, pubescent, ciliate, glandular, pilose on the inner surface ;

upper and lateral lobes 2-5-3 cm. long ; lowest lobe 4 mm. long, 1-5 mm. broad. Veaxillum

7 mm. long, 6 mm. broad, obovate with a short claw 2 mm. long; alae 6-5 mm. long, 2

mm. broad, with a-linear claw 3 mm. long; carina 3:5 mm. long, 2 mm. broad, with a

linear claw 3-5 mm. long. Ovary pilose; style 3 mm. long, pilose.

Flowered: October. Flowers: Yellow.

CareE.—Tembuland, Cala, Royffe 161.

137

A Drying Cabinet for the Preparation of Plant

Specimens for the Herbarium.

By A. P. D. McCiean, Natal Herbarium, Durban, and H. H. Srorny, Hast African

Agricultural Research Station, Amani.

INTRODUCTION.

The Drying Cabinet, which is described in this paper, is the outcome of our attempt

to meet the extreme difficulty of drying plants under the normally highly humid conditions

of the Durban climate. By the ordinary plant-press method, the drying is a lengthy process

and even by exercising the greatest care the specimens are of poor quality, frequently

rendered valueless by the development of moulds. ;

The general principle of our drier evolved from a conversation several years ago with

Mr. G. H. Cunningham, of New Zealand. He is understood to have a machine of similar

type in operation, but no description appears to have been published. In view of the

interest which many visitors have shown in our drier, it appears advisable to make

a description available. We acknowledge our obligation to Mr. Cunningham for the basic

idea, though we, in collaboration with the Public Works Department, are responsible for

the details of this design.

DESCRIPTION OF THE DRYING CABINET.

The principle of our method of drying is as follows : —The plants to be dried are laid

between sheets of drying paper in the customary manner and subjected to a suitable degree

of pressure. Hot air is then drawn through the pile, the air passing through the interstices

between the sheets.

The cabinet, constructed of stout, well-seasoned teak, consists of a Pressing or Drying

Chamber and a cupboard divided by a shelf into two compartments, an upper oven and a

lower heating chamber. These features are illustrated in the accompanying photographs

and diagrams.*

The Pressing or Drying chamber is 10” « 1’ 8” x 1’ high and can accommodate ap-

proximately 100 plant specimens of a size suitable for the average mounting sheet. This

chamber is fitted with a lid consisting of a flat base with an attached letterpress screw

and screw block. The latter is maintained in a fixed position by means of a slotted arm

which fits over two hold-fasts screwed to the sides of the chamber. By this means the

lid can be lowered or raised at will. In addition the lid has certain features to procure

as nearly as possible an air-tight joint with the sides of the chamber. Pieces of solid rubber

beading of the desired length are held in movable steel holders attached by means of slotted

iron clips to the upper edges of the lid. When the lid is in the desired position, the rubber

bead can be made to press firmly against the walls of the chamber by means of 4 pivoted

cleats. Triangular pieces of rubber are provided for the corners, the ends of the steel holders

and the beadings being cut in such a way as to allow for the fitting of the corner pieces.

* Acknowledgment is made to Mr. L. H. Godfrey, District Draughtsman, Public Works Depart

ment, Natal, for the diagram drawings.

138

The Heating Chamber of the dimensions 10” x 1’ 8” x 8" is asbestos lined and

equipped with 3 Revo electric fire-baths, each consuming 500 watts. By means of a 3-way

control switch either one, two, or three of the units can be brought into operation. An

iron shelf divides the heating chamber from the oven. Access is provided to the cupboard

by means of a felt-lined door equipped with a fastening catch bevelled to create pressure

on closing. Air is drawn through the system by a Sturtevant + H.P., 2,750 R.P.M.

exhausting fan, attached to the side of the drying chamber, communication with which is

provided through a narrow vent at its base. Air enters the system through a small opening

in the base of the heating chamber—passes over the electric fire-baths, through an opening

in the cupboard shelf and enters the drying chamber through five holes in the roof of the

oven, at the end opposite to that carrying the fan. Hot air therefore enters the drying

chamber at this end and is drawn by the fans horizontally through the pile of drying papers

and plants, which le compressed between the base of this chamber and the screw-down

lid. Temperature is observed by means of a thermometer which passes through the wall

of the cabinet into the oven at the point where the air enters the drying chamber.

PROCEDURE.

The plant specimens are arranged in the usual way between sheet of some absorbent

material such as blotting paper ; three sheets between each has been found generally suitable,

The sheets are cut 10” wide so as to fit exactly into the drying chamber. A fairly high

degree of pressure can be maintained on the specimens without seriously retarding the

passage of air through the system. The fan is then set in motion and the heating switch

turned to high. In this position the temperature measured by the thermometer already

mentioned rises to about 150° C. after an hour. The switch should then be turned to

medium which maintains the system at a temperature between 130° C. and 150° C. It is

not desirable to keep the three heating units in action for long periods as this will tend to

cause over-heating, the specimens becoming brittle and crumbling when handled. A

drying temperature of about 130° C. to 140° C. seems to be the most satisfactory. The

switch may be turned to low after four or five hours.

The average plant specimen of a woody nature and not of an unduly high degree of

succulency requires about an 8-hour treatment. Delicate herbs can be dried in under four

hours, while excessively succulent plants of the Aloe type require up to 16 hours.

For seaweeds a low temperature is recommended and we have found it advisable to

keep the switch at low, as high temperatures tend to remove the colouring matter.

The oven was included to make provision for the drying of such botanical specimens

as fleshy fungi. An excellent result was obtained with a species of Polyporus and complete

drying was effected in eight hours at a temperature of 70° C. We have no doubt that

many uses will be found for this additional drying chamber.

DISCUSSION.

Our method not only provides for the very rapid drying and preparation of herbarium

specimens, but yields results of an excellent quality.

The finished specimens do not possess the brittleness which so frequently is associated

with certain plants prepared by the usual slow method of drying. Moreover, colours are

retained very much better and there is less vegetative distortion. We have little doubt

it will overcome many of the difficulties encountered with some natural families of which

the preparation of dried specimens gives normally much trouble

We commend this piece of apparatus to those in charge of Herbaria situated in the

Tropics and other regions of high humidity.

139

DRYING CABINET

)

LETTER PRESS screw B AvaPTED

SCREW BLOCK C AremM D SLOTTED TO FIT OVER

HOLD FASTS

PLANT SHEETS ©

FOR |PRESSING —

(| THERMOmETH R FILLET

BD WAY SWITCH’

Ve)

HOLE 3'x!I@ FoR

UPWARD CURRENT

ASBESTOS

LINING

INLET VENT SPACE ui

“SECTION:

LETTER PRESS ror B

RUBBER LINING soy

JROW RIM AND

SLOTTED CLIPS EXHAUST FAN

AHD HOPPER

HOLDFAST SLOT RON RIM AND SLOTTED CLIPS

PLATE TO HOLD RUBBER BEADING.

PLAN at TOP

to eS = et

STOS LINING

PLAN at \HEATING CHAMBER

140

Plate 1].—Vid of the Cabinet—Showing Details of Designs

i4]

EXPLANATION OF PLATES.

Plate J.—Drying Cabinet with the Lid in Position.

142

Notes on Aloe spicata Linn. fil.

By R. Marloth.

The identification of our older species of Aloe from descriptions and illustrations has

offered many difficulties, for nearly all these diagnoses were based upon plants cultivated

in European conservatories, that means, they had grown up in an environment with a

considerably moister atmosphere and a far less sunny climate. Under such conditions

these highly succulent plants generally assume quite a different habit, not only in stem

and foliage, but also in the nature of their inflorescence. Consequently when Volume VI

of the Flora Capensis appeared in 1897 comparatively few of our wild species of Aloe were

known correctly by name, although some of them had been in cultivation in Kurope for

two centuries. They differed to such an extent from the available descriptions that

recognition became possible only through a study, of‘ plants from various localities and of

their behaviour under cultivation in our gardens. A few of the species of which the natural

home was not known 25 years ago are the following : —Aloe succotrina, A. purpurascens,

A. microstigma, A. obscura, A.:glauca, A. falcata, A. grandidentata, A. pluridens, A. Salm-

dyckiana and the subject of this article, viz. A. spicata. For all these, with the exception

of the last named, the difficulties and doubts have been gradually overcome, and we are

now fairly certain of their identity and localities of occurrence in the country.

With regard to the last one in the list, viz., A. spicata, the writer considers that he

has at last solved this puzzle, for Linnaeus’ Aloe spicata is the plant now usually called

A. speciosa. AN

Before discussing the reasons for this conclusion it appears advisable to quote the

descriptions of Aloe spicata, as given in the three principal publications dealing with the

plant, viz. : 4

LinnagEvs fil. Supplementum, 205 (1781).

THUNBERG, Flora Capensis, Edit. Schultes, page 309 (1823).

J. G. Baker, Flora Capensis, Vol. VI, page 316 (1896).

Linnaeus fil. Supplementum, 205 (1781).

“Aloe spicata. Caulescens, foliis planis, ensiformibus, dentatis, floribus spicatis,

campanulatis, horizontalibus.” =

- THUNBERG in Flora Capensis, Edition Schultes, page 309 (1823).

“ Orescit in -Promontorii Bonae Spei interioribus regionibus. “Floret Augusto.”

Caulis teres apice foliosus, 3-4 pedalis, crassitie brachii. Folia subverticillata,

carnosa, basi lata sensim attenuata, canaliculata, dentibus remotis, patentia bipedalia.

Flores densissime approximati horizontaliter patentes; spica pedalis sensim florens.

Bractea sub singulo flore solitaria ovata acuta lata membranacea alba, striis 3 viridibus,

corolla paulo brevior. Corolla subhexapetala ; laciniae 3 interiores invicem non connatae,

latiores, ovatae, obtusae, albae, carina linea triplici viridi, unguiculares; tres exteriores

basi cum interioribus connatae, angustiores, ceterum similes, minus concavae. Filamenta

linearia sensim parum attenuata, basi alba, superne flavescentia, inaequalia, erecta, corolla

sesquilongiora. Antherae ovatae, incumbentes, fulvae. Stylus flexuosus erectus, longi-

tudine fere staminum. Stigma simplex obtusiusculum. Capsula ovata, obtusa, subtrigona,

inflexa, hine carinata, lateribus planis quadristriata, inde convexa sulco medio glabra.

Dissepimentum duplicatum. Corollae repletae succo melleo purpurascente.”

BakrR in Flora Capensis, Vol. VI, page 316 (1896).

143

A. spicata Linn. fil.

“Stem elongated below the leaves, simple ; leaves 12-20, laxly disposed, patent or the

lower recurved, ensiform, 18 inches long, 14 to 2 inches broad low down, 6 lines thick in

the centre, tapering gradually from the middle to a long point, green with a slight glaucous

tinge, obscurely lineate towards the base, mottled with copious irregular whitish oblong

spots ; margin with spreading, horny, deltoid-cuspidate prickles 14 to 2 lines long ; peduncle

slender, simple or forked, above a foot long ; raceme oblong, 6 inches long ; pedicels ascend-

ing, 6—9 lines long; bracts small, lanceolate ; perianth bright yellow, tinted with red when

young, 15 lines long, tube short ; stamens slightly exserted.”

A comparison of these accounts shows that they cannot refer to the same species, and

as Mr. Baker states that his description of the flower was taken from a plant cultivated

by Mr. Peacock (1879), it is obvious that that plant had been wrongly named, and that

Haworth, whom Baker followed otherwise, had also erred in applying Linnaeus’ name to

a plant with speckled leaves, for neither Linnaeus nor Thunberg mention the spots on

the leaves, which they could not have overlooked, nor does the latter call the inflorescence

forked and the flowers yellow. The erroneous views of these authors are probably the

principal cause which has prevented the recognizing of the true A. spicata up to the present.

A further proof that Haworth and Baker were mistaken in their identification is

afforded by Mr. N. KE. Brown’s study of Thunberg’s type specimen collected by him about

the year 1776 and preserved in his herbarium at Upsala. Mr. Brown’s account is published

in Bothalia, Vol. I, page 142 (1922), and runs as follows (omitting his comments) :—

A. spicata Linn. fil., ex Suppl. p. 205 (1781), and Thunb. Flora Capens., page 309.

“ Thunberg’s specimen is the type of this species and consists of portions of two leaves

and six detached flowers. The leaves are resp. 10? and 13? inches long and 14 and 16

lines wide at their basal ends, gradually tapering thence into a long and slender subulate

point which at 3 inches below the acute or bluntish tip is only 2 lines broad. . . . The

teeth on the margins are 6-8 lines apart and very small, being not more than half a line

long; the apical spine is also small. The margin between the spines is straight or very

faintly concave. The flower stem and pedicels are absent, but the flowers may have been

sessile. One flower has what appears to be a bract attached to it, which is 6 lines long

and 5 lines broad, and is broadly elliptic, obtuse, 3-nerved. The perianth is campanulate

in shape, 7 lines long and 5 lines in diameter as pressed, but is probably of nearly the same

dimensions when alive. The segments are nearly 3 lines broad, ovate-oblong, obtuse, 3-

nerved ; they all appear to be free to the base and are not recurved at the apex. According

to Thunberg they are white, with three green veins. The stamens are exserted 3-4 lines

beyond the tips of the perianth segments with stout purple filaments half a line broad.”

It will be seen that—

1. The leaves of the type are not mottled.

2. The marginal prickles are “‘ very small, being not more than hali a line long,” while

Mr. Baker’s plant had “spreading, horny, deltoid-cuspidate prickles 14 to 2 lines long.”

It is obvious that Mr Baker’s statements cannot be reconciled with Thunberg’s and

Brown’s descriptions.

Having recently had occasion to study a population of hybrids* between A. ferox

var. supralaevis and A. speciosa Baker, and comparing them with living material of both

parents from various sources, the writer noticed that Baker’s description of his A. speciosa

agreed remarkably well with Mr. Brown’s description of Thunberg’s type specimen of

A. spicata, and combining this account with Thunberg’s own diagnosis it became obvious

that Baker’s A. speciosa was A. spicata Linn. fil.

* The hybrid has been named A. Tomlinsonii and is figured and described in South African Garden-

ing, October, 1929.

144

Aloe spicata L. fil. and A. ferow. Hillside at Alicedale. August. The plant in the foreground

and the others with deflexed leaves and simple spike or spikes are A. spicata; those with stiffly erect

jeaves and branched inflorescence are A. ferow. [Photo by BR. Marloth.

Hillside near the Breede River. 15 miles south of Swellendam. Nearly all

Aloe spicata L. fil.

the older plants here are branched at the top and bear several spikes. August.

[Photo by L. L. Tomlinson.

145

Baker’s description of A. speciosa (Flora Cap., Vol. VI, 323), taken over from the

Journ. Linn. Soc. XVIII, 178), is as follows :—

“Habit of A. africana, reaching a height of 20-25 feet; leaves in a dense rosette,

15-18 inches long, 2 inches broad low down, ensiform, glaucous, with smaller and weaker

prickles than A. africana; peduncle short, stout, simple; raceme very dense, simple, a

foot long, 4 inches in diameter when expanded, all except the upper flowers deflexed, pedicels

2-3 lines long ; bracts sub-orbicular, obtuse, 6 lines long and broad ; perianth cylindrical,

15 lines long, rose-tinted in an early stage only, greyish white when mature ; tube scarcely

any; stamens and style bright red, conspicuously exserted.” Somerset Hast, MacOwan

No. 1922.

Note by the Writer—A. speciosa is quite different in habit from A. africana, and never

reaches a height of 20 feet. The only similarity would be that both are arborescent, with

a trunk 6 to 10 feet high, but so are quite a number of others. The perianth, when fresh,

is always more or less clavate and constricted at the mouth, as shown very well in Berger,

Aloineae Das Pflauzenreich, page 300.

If one remembers that Baker’s description was drawn up from a dried specimen, and

if one corrects the statements about similarity to A. africana and the height of the plant,

which are erroneous as every one who knows the fairly widely distributed A. speciosa Baker *

will confirm, this description of A. speciosa agrees very well with Thunberg’s type of A.

spicata, as redescribed by Mr. Brown.

There are, however, a few minor points in Thunberg’s description of the flower which

call for special consideration.

1. The bracts are described as “little shorter than the flowers,” while in A. speciosa

they are about one-third the length of the fully developed flowers.

2. The filaments are described by Thunberg as “ white at base and yellowish higher

up,’ while in A. speciosa they are yellowish at the included part and dark brown or purplish

brown in the projecting part.

3. The nectar in the flowers is said to be purple, while it is colourless or yellowish in

fresh flowers of A. speciosa.

The discrepancy concerning the bracts has been disposed of by Mr. Brown’s account

of Thunberg’s type specimen.

With regard to the colour of the filaments Mr. Brown finds that im the type they are

purple . . . whatever their original colour may have been.

On the third point I have made an observation which disposes of that as well, viz. :

if older fresh flowers are placed between paper, the nectar, if there is any, although origi-

nally colourless, soon dissolves the pigment of the filaments and stains the paper purple

while the filaments themselves become paler.

Taking all these observations together and considering the unusual colour of the

perianth, viz.: white with green lines, there is no point left that would be inconsistent

with the view that Aloe speciosa Baker is Aloe spicata Linn. fil.

It is noteworthy that Berger’s studies of these plants had led him to the conclusion

that ““ Haworth’s identification was probably wrong and that A. spicata has its nearest

ally in A. speciosa Baker.” (Aloineae, page 301). The foregoing discussion will show that

we have to go a step further and consider them to be synonymous.

After the foregoing note on Aloe spicata had been written Dr. Pole Evans drew my

attention to an article in the Gardener’s Chronicle of June, 1921, page 6, by Mr. N. E.

Brown of Kew, on Aloe spicata. In this article I find the followig passages quoted by

the author :

(a) From Thunberg’s dissertationes on Aloe: “the purest and best juice of aloes is

obtained from this species.”

(6) From Linnaeus fil. Supplementum, page 205 (1781): “ from the juice of this species

the best gum-resin of aloes is prepared, from the rest of the species the cheaper kinds.”

146

Obviously Linnaeus had merely repeated the information received from Thunberg, but

it should be noted that Thunberg has omitted this statement from his Flora Capensis—

perhaps it had been recognized in the meantime as erroneous. Whether the drug or some

variety of the drug was ever prepared from this species on a commercial scale or not does

not concern us here. The importance of the statement is in the fact that Thunberg’s

specimen must have come from a district where the drug aloes was being manufactured—

and that information settles the point at issue, for when Thunberg stayed at the Cape (1775)

such manufacturing was carried out only in the Swellendam district which then embraced

also the country now forming the districts of Riversdale and Mossel Bay, and in that area

occur only two species of tree aloe, viz., Aloe feror and A. spicata (A. speciosa). That

the other kind “ from which the cheaper quality of aloes was obtained” is A. ferox is now

generally admitted. In fact, at present no other species is used at all for this purpose and

it is remarkable that the farmers in those districts call A, ferox simply the “ alwij,” while

A. spicata is one of the “ wilde alwij,” because it cannot be used for the manufacturing

of the drug. Whether 150 years ago some attempts were made to prepare a different kind

of drug from A. spicata as well as from A. ferox, or whether Thunberg’s informant

confused the two species could hardly be elucidated at the present time. It should be

remembered that Thunberg did not see the plant in its wild state, for he merely states thet

it grows in the interior, which means anything beyond Stellenbosch. He must have

obtained his specimen either from a garden at Capetown or it was sent to him by a

correspondent.

Postscript—When writing the foregoing notes for publication in Bothalia I sent an

illustration of the hybrid Aloe Tomlinsonii to South African Gardening, where it appeared

in November, 1929, with the following footnote :—

‘ The plant described by Baker in 1880 under the name A. speciosa and now generally

known as such has been recognized to be identical with speciments described by Linnaeus

the younger and Thunberg as A. spicata, hence the former name becomes a synonym for

the latter. The reasons for this change of name will be given in the forthcoming number

of Bothalia (Pretoria, 1929.) ”’

The view on Aloe spicata expressed therein is contradicted by Mr. N. EK. Brown, Kew,

in a letter to the Editor of S.A. Gardening, published in the May number, 1930, of that

journal, just at the time when the proof of the foregoing article reached me. I cannot help

expressing my regret that Mr. Brown, a highly esteemed friend, should have rushed into

print instead of awaiting the appearance of my notes, of which he had notice, for he does

not put forward any new observation and merely reiterates what he published 8 to 10

_ years ago and what is quoted above.

His statement in that letter : “the two species have not the slightest resemblance to

one another ”’ is in direct conflict with Berger’s view that: “ Aloe spicata has its nearest

ally in Aloe speciosa Baker” (Aloineae, page 305),* especially if one remembers that this

author based his conclusions on plants grown and flowered by him at La Mortola year after

year and not only on herbarium specimens, as Mr. Baker and Mr. Brown had to do.—R.M.

* See Das Pflanzenreich, IV, 38, III, 2. Liliaceae-Aloineae (1908.)

147

South African Gramineae.

Some new species of Digitaria.

By Sypney M. Srenv.

TueE following species and varieties of Digitaria with the exception of D. apiculata and

D. trichopodia have been described from living plants that were collected from many

different and widely separated localities in the Union of South Africa and brought to the

Government Experiment Station Prinshof at Pretoria where they have become well

established. There is a pure stand of each grass of about 1/40 acre im size so that in

describing the species I have been able to study a very large number of plants of each.

I have found it exceedingly difficult, in many cases to describe the grasses so that

they may be recognized from the herbarium sheet, in the field the differences, especially

in the vegetative habit, are so marked that it is impossible to class them as one species,

but on the sheet naturally these characters are not so obvious,.and I am not surprised that

systematists who have only had dried material with which to work have classified most

of these grasses under D. eriantha adding forma or var. to account for any variations

observed. As it is I have only, out of the mass of material studied—some 60 separate

stands—classified 16 as distinct species of which 11 are new, although the Agriculturist

claims that no two stands are exactly alike and has given to each grass its distinguishing

vernacular name.

In most cases the distinguishing characters of the inflorescence are very shght. I find

the size of spikelets and degree of hairiness very variable, within certain limits, in each

species and one must depend on a correlation of characters rather than on any specially

outstanding character in determining the species.

A detailed account of all the Digitarias at Prinshof with their agricultural and grazing

value will appear in another publication.

The letters N.H.P. before a number signify National Herbarium, Pretoria.

D. Pentzii.

Perennial, densely tufted, much branched from the crown with intravaginal inmnova-

tion shoots, sending out numerous surface runners or stolons bearing dense fascicles of

short upright or spreading shoots from the rooting nodes, internodes glabrous or hirsute,

often reddish. Flowering culms erect or sub-geniculate 3—7 noded, the lower internodes

short and usually included, the uppermost long exserted and equalling half or more the

length of the culm which reaches a height of about 120 cm. ; leaf sheaths rather loose, the

upper glabrescent, the lower hairy with rather long fine hairs, densely and appressedly silky

hairy at the base, compressed, up to 14 cm. long ; ligule membranous, broadly rounded to

truncate, often denticulate, up to 5 mm. long; blades linear from a slightly narrowed base

tapering to a long very narrow apex, up to about 30 cm. long by 6mm. broad, minutely

and softly hairy above and below or quite glabrous or with a few scattered long white hairs,

scaberulous on both sides but more so on the under side, midrib raised in a prominent keel

below, grooved above, primary nerves about 5 on either side, not prominent. The leaves

of the stolon shoots are usually shorter and broader. Racemes about 3-12 and about 17

em. long, strictly umbelled or sub-digitate on an elongated axis up to 3 cm. long termina-

ting in 1, 2 or 3 racemes ; rhachis triquetrous, narrowly winged, -5 mm. wide, scabrid on

148

the margins and on the facial angle, glabrous ; pedicels 2-nate, rarely 3-nate, triquetrous,

scabrid, the longer about 2 mm. long, lower about -5 mm., glabrous ; spikelets imbricating, -

3-3-5 mm. long and up to 1 mm. wide, narrow lanceolate acute ; lower glume reduced to

an ovate rounded or truncate nerveless scale; wpper glume equalling about 2 the length

of the spikelet, 3-nerved with dense lines of rather long straight fine blunt hairs between

the nerves and on the margins, the marginal extending beyond the tip; lower valve the

size of the spikelet, 7-nerved, nerves smooth, the inner lateral nerves rather close with

dense lines of soft fine blunt hairs between them and on the margins, hairs often ultimately

spreading and up to 1 mm. long; pale and lodicwles up to -3 mm. long ; fertile floret slightly

shorter than: the lower valve, narrow lanceolate acuminate, thinly chartaceous.

TRANSVAAL: Zoutpansberg ; Mara, Kligoblin, Archer 18; Downs Junod 4086 ; Elim

Mill, Thomas 9, 4 and 3; Manner Mead, Daire N.H.P. 8376; Messina, Turner 25, 17;

Pole Evans N.H.P. 8496; Pietersburg McKechme, N.H.P. 7913; Pretoria District :

Pretoria Pentz N.H.P. 8044, 8051, 8560, 8550, 8554, 8521, 8051; Kaalfontein, Wilfontein

Farm, Bevan, N.H.P. 7602; Hamanskraal Tregenna 201, Pearce, N.H.P. 7958; Vastfon-

tein, Inebenberg N.H.P. 7955; Rustenburg District: Verdun, Cellier, N.H.P. 7910;

Marico District: Malmanie Oog, Ottoshoop, Liebenberg, N.H.P. 7966; Wolmaransstad ;

Boskuil, Sutton 108; Standerton District: Balfour, Visser, N.H.P. 8020.

ORANGE FREE State: Parys, Ramsbottom, N.H.P. 8380.

Cape Province: Vryburg District: Vryburg Pentz, N.H.P. 8501, 8509, 8510 (type) ;

Iffey, Mather, N.H.P. 7947; Bechuanaland: Vukwe, Pole Evans and Pentz, N.H.P. 8493 ;

Tati District: Francis Town, Pentz, N.H.P. 8521.

var. minor (=Digitaria eriantha var. stolonifera, Stapf.) rather weaker and more

straggling than the type, Racemes 4 or 5 rarely more numerous, usually strictly umbelled,

spikelets about 2:5 mm. rather plumper and less acuminate than the type, hairs of the

valve often mixed with fine but stiff bristles.

TRANSVAAL: Waterberg: Middelfontein Num-num, Donisthorpe, N.H.P. 7864 ;

Naboomspruit : Koornpunt, Kalpan, Langerman, N.H.P. 8702; Palala, Hspach, N.H.P.

7622; Pietpotgietersrust, Pentz, N.H.P. 7816, 8513; Pyramid Hstate, Galpin 8874 ;

Hangklip, Knothe N.H.P. 7960; Marico District: Malmanie oog, Ottoshoop Liebenberg,

N.H.P. 7965: Pretoria District: Hamanskraal Pentz, N.H.P. 8514; Pretoria, Paarde-

fontein, Liebenberg N.H.P. 8024; Steyn N.H.P. 8054.

This species varies considerably within its prescribed limits many of the variations

being probably due to hybridisation. This is the well-known “ Woolly Finger Grass ” of

South Africa that has been proved to be such an excellent pasture grass. Specimens of

this grass and of other stoloniferous species have been distributed under the name of

D. eriantha var. stolonifera Stapf. but only the one fairly constant variety, D. Pentz var.

minor, really corresponds to the type of Stapf’s variety.

D. valida.

Densely tufted perennial with intravaginal innovation shoots and sending out long

rather stout and often repeatedly branched runners, that root and shoot from the nodes,

internodes and young shoots hairy. Culms glabrous, smooth, subterete, geniculate from

about the second node and then erect, stout, up to 130 cm. high, 3-4 noded, simple or more

usually branched from one or more of the lower nodes, uppermost internode very long ;

sheaths of leaves up to 18 cm. long compressed and keeled, the lower loosely hairy all over,

densely silky tomentose at the base, the upper glabrescent except at the always hairy nodes,

firm, rather loose ; ligule delicately membranous, up to 5 mm. long, ovate obtuse, glabrous,

minutely denticulate ; blades linear lanceolate from a slightly narrowed base, long tapering

to a narrow acute apex, 18-50 cm. long by 5-9 mm. wide, firm, flat or lightly folded, gla-

brous, smooth below, with scabrid margins, midrib grooved above and forming a prominent

firm keel below, primary nerves 3-4 on either side the midrib. Racemes about 9-14, whorled

149

or sub-whorled on a short (about 3-5 cm. long), common angled axis, up to 13 em. long ;

pedicels 2-nate the longer rather less than 1-1-5 mm. long, seabrid, glabrous. Spikelets

narrow ovate- lanceolate, acute, about 3-5 mm. long -7 mm. wide, lower glume a rather

firm minute, membranous scale ; upper glume about two-thirds the length of the spikelet,

3-nerved with lines of fine blunt appressed hairs between the nerves and on the margins ;

lower valve the size of the spikelet, 7-nerved, nerves smooth, rather close, hairs as in the

upper glume ; pale -7-1 mm. long, lodicules minute, truncate ; fertile floret rather firmly

membranous, lanceolate acuminate, slightly shorter than the lower valve.

TRANSVAAL.—Pretoria, Pentz, N.H.P. 1970 (type) 8506, 8552, Wager, N.H.P. 8060;

Rustenburg, Doornpoort, Mogg, N.H.P. 8571 and 8572; Marico, Skuinsdrift, Liebenberg

564; Slurry, Pole Evans and Pentz, N.H.P. 8520; Malmanie’s Oog, Pentz, N.H.P. 8512;

Heidelberg, Uitgevallen, Burtt-Davy 13666; Carolina, Coed- Byclair Farm, Howel-Jones,

N.H.P. 8322; Lydenberg, Pentz, N.H.P. 8515; Potchefstroom, Liebenberg, N.H.P. 8630 ;

Vereeniging, Farm Zuurfontein, Pentz, N.H.P. 8497.

ORANGE Free State.—Parys, Ramsbottom, N.H.P. 8377; Heilbron, Maccauvlei,

Brandmuller 54.

Cape Province.—Mafeking, Pitsani, Pole Evans and Pentz, N.H.P. 8548. Bechuana-

land: Padden on Molopo River, Pole Evans and Pentz, N.H.P. 8558.

Var. glauca.

Culms and stolons more slender, leaves shorter, glaucous, internodes of runners glabrous.

TRANSVAAL.— Pretoria: Hamanskraal, Pentz, N.H.P. 8499: Pretoria: Groenkloof

Valley, Mogg, N.H.P. 7969; Klapperkop, Stent and Mogg, N.H.P. 7968.

A very useful grazing grass of the ‘‘ Woolly Finger ” type (D.Pentziz) but much taller

and coarser with broader leaves, stouter runners.

D. Polevansii.

Stoloniferous perennial from a thickened much branched rhizome bearing numerous

bulbous extravaginal innovation shoots covered with short silky tomentose cataphyls.

The long, stout surface runners or stolons root firmly at the nodes and send up fascicles

of bulbous based shoots, internodes purplish grey and up to 20 em. or longer. Culms erect

rather bare from a bulboid base, up to about 1-3 m. high, smooth, glabrous, 4-5 noded,

branched from the lower or middle nodes, internodes mostly included except the uppermost

which is long exserted ; leaf sheaths long, rather loose, finely ribbed, smooth, glabrous or

the lower sometimes finely hirsute or ciliate and early fallmg away leaving a rather bare

culm base; ligule membranous, mostly truncate or rounded, denticulate, up to 3 mm.

long and usually more or less appressed to the blade; blades glaucous, linear lanceolate,

slightly narrowed at the rounded base, tapering to a fine point, flat, up to 40 cm. long by

8-11 mm. broad, quite smooth, glabrous or minutely hirsute with a few tubercle based

hairs at the base, margins cartilagenous smooth, primary lateral nerves about 5 not very

conspicuous on either side the slender prominent midrib. Racemes up to 12 and about

9-16 cm. long strictly digitate or with an additional 1-4 terminating the shortly elongated

angled axis, rhachis triangular with narrow winged, smooth or sparsely scabrid margins,

-5--75 mm. wide, glabrous; pedicels 2-nate, scabrid with cupular sub-membranous tips,

the longer up to 2 mm. Spikelets appressed, sub-imbricating about 3 mm. long excluding

the hairs at tip, narrow-oval-acute, greyish green to leaden or brownish when mature ;

lower glume very small but distinct, thinly membranous, obtuse to sub-acute ; upper glume

narrow, 3-nerved, acute, equalling about three-quarters to rather less of the fertile floret,

with long silky hairs between the nerves and on the margins and tip; lower valve usually

purplish, ovate acute, the size of the spikelet, 7-nerved, midnerve prominent, lateral faint,

smooth, lines of the silky obtuse usually appressed hairs between the inner lateral nerves

and on the margins ; fertile floret eliptic, rather shorter than the lower valve, pale greyish

green to light golden brown, flaps broad meeting or in some cases overlapping.

150

Care Province.—Bechuanaland, Vryburg, N.W. of Armadillo Creek, Burtt-Davy

13871; Kuruman, Pole Evans 2065, 3322, Molopo River, Pole Evans and Pentz, N.H.P.

8961; Mafeking between Vuilnek and Inkruip, Pole Evans 2418, 2419.

A very distinctive species with bulbous-like bases to the culms. The stout runners

‘reach a length of 450-600 cm.

D. decumbens.

Perennial. Culms fascicled on a short rhizome, innovation shoots intravaginal,

stoloniferous. Flowering culms up to 113 em. rather bare at the base, many noded, usually

‘decumbent, much and often repeatedly branched and often rooting at the lower nodes,

‘glabrous, smooth, terete or slightly compressed; sheaths rather long and _ loose,

smooth, glabrescent, those of the young shoots often finely ‘hirsute especially at

the nodes; ligule thinly membranous, truncate, about 3 mm. long; blades linear

lanceolate from a shghtly narrowed base tapermg to an acutely acuminate tip,

flaccid, minutely scaberulous, glabrous, up to 14 em. long (those of barren shoots

sometimes to 28 cm.) by 7 mm. broad, light green. Racemes up to 13 ecm. long,

typically widely spreading the lower almost horizontal, 6-10 arranged in a single whorl

(usually of 6) at the apex of the culm or with the common axis elongated about 1-1-5 cm.

and bearing a second whorl of 2-4 at the tip, occasionally with one or two scattered below

the top whorl. Rhachis sub-flexuous, slender, flat on the back trigonous on the face with

narrow but distinct green winged margins, scabrid, glabrous, from -4--8 mm. wide ;

pedicels 2-nate, slender trigonous, scabrid, longer up to 1:5 mm. long; spzkelets rather

loosely imbricating, narrow lanceolate-acuminate, 2-7-3 mm. long by -6—-8 mm. wide, ~

scantily and usually appressedly hairy ; lower glume a minute obtuse membranous persis-

tent scale, wpper glume narrow oblong equalling half to three-quarters length of the

spikelet, 3-nerved, nerves smooth lines of fine closely appressed silky hairs between the

nerves and slightly longer (up -5 mm.) and spreading hairs on the margins ; lower valve

the size of the spikelet, 7-nerved, rather close and prominent and smooth, apparently

glabrous, but actually with narrow lines of closely appressed silky hairs in the lower half

between inner pairs of lateral nerves and on the margins right to the tip of valve, hairs

never longer than -5 mm. and not or seldom spreading, fertile floret narrow lanceolate acumi-

nate, slightly shorter than the lower valve, greyish yellow to purplish.

TRANSVAAL.—Barberton District: Nelspruit, Pentz, N.H.P. 8495.

The loosely fascicled culms the weak mostly desumbent and leafy stems, the widely

spreading racemes and the apparently glabrous or sparsely and shortly hairy spikelets,

distinguish it from D. Pentzi,

D. Swazilandensis.

Perennial more or less prostrate creeping grass sending out long slender rooting runners

with numerous short glabrous or sparsely hairy internodes. Culms prostrate erect,

very slender, glabrous smooth, terete, with many short, more or less included,

internodes below the comparatively long (up to 13 cm.) uppermost inter-

node; frequently branched and sending out leafy barren shoots. Sheaths somewhat

compressed at length slipping from culm, glabrous except at the long hairy nodes or

sparsely long and softly hairy, produced in an auricle on one or both sides of the mouth.

Ligule rounded or truncate, thinly membranous about -7 mm. long; blades flat narrow

ovate-lanceolate from a rounded and contracted base, tapering to a narrow acute apex,

up to 7 em. long by 4 mm. wide, glabrous or with a few scattered fine hairs towards the

base, smooth except at the tip and on the minutely scaberulous margins. Racemes 2-3

very rarely 4, slender, about 5-7 cm. long, rhachis, about -4 mm. wile, narrowly winged,

triquetrous and sharply angled on the face, elabrous and smooth except on the minutely

scaberulous margins ; pedicels 2-nate, terete or somewhat flattened or sub- -triquetrous,

longest about 1-5 mm. long, minutely scabrid on the margins; spikelets somewhat

imbricating, 2-2-5 mm. long, about -7 mm. wide, oval-acute, apparently glabrous. Lower

151

glume an obtuse to truncate, persistent, membranous scale ; wpper oval acute, rather broad

and thin, equalling half to two-thirds the length of spikelet, 3-nerved with scanty lines of

appressed straight fine hairs between the nerves and on the margin those towards the tip

longer and slightly denser ; lower valve thinly membranous the size of the spikelet, rather

prominently, 7-nerved with scanty lines of appressed hairs between the inner pair of lateral

nerves and on the margins, fertile floret oval, acute, purple.

Near to D. horizontalis but a perennial with shorter narrower and less scabrid leaves,

and fewer racemes.

Swazianp.—Stegi Mrs. Perkins N.H.P. 7931; on the border near Mrs. Andrews’

farm, Pentz N.H.P. 8557.

A prostrate grass that makes a mass of short dense matted foliage suitable for a lawn,

or for short crop for sheep, does well with clover.

D. rigida.

Perennial, caespitose, sending out long branched and rather rigid runners that root

and shoot from the nodes. Culms geniculate or prostrate erect about 75 cm. long, 4-6

noded, simple or branched, glabrous, smooth, sub-terete, compressable ; sheaths of the

leaves firm, loose finely and closely striate, smooth, glabrous except at the densely silky

nodes, or with a few scattered hairs ; ligule prominently exserted, membranous 5-14 mm:

long, dentate, adnate to the often rather long auricles of the sheath ; blades linear, flat or

folded, 5-20 cm. long by about 3-4 mm. wide, from an equally wide base long tapering to

an acute apex, very firm, smooth glabrous or softly and sparsely hairy on the upper side,

midrib rather stout and prominent below, whitish above, primary lateral nerves about 6

on either side, close and firm. Racemes 6—9 in whorls or sub-whorls on a short common

axis, about 7-9 cm. long ; rhachis narrow, about -5 mm. wide margined or very narrowly

winged, scabrid, glabrous; pedicels 2-nate or sometimes 3-nate towards the base, longest

about 2 mm., scabrid ; spikelets imbricating, about 3 mm. long, ovate lanceolate, acute,

lower glume a nerveless membranous scale; wpper glume equalling from half to two-thirds

of the spikelet, 3 nerved with very dense lines of fine, straight, blunt hairs between the

nerves and on the margins ; lower valve equalling the spikelet, 7-nerved, the inner pairs of

lateral nerves rather remote from the central nerve and equally prominent, all smooth and

with lines of rather fine, stiff, straight, blunt hairs between the inner lateral nerves and on

the margins, the hairs between the nerves rather scantily appressed, those on the margins

longer and rather denser. Fertile floret elliptic, lanceolate, shortly and acutely acuminate,

firmly membranous, finely punctate striate, light green or purplish, margin rather distant ;

anthers about 1-5 mm. purple.

TRANSVAAL.—Magalakwin, Pole Evans, N.H.P. 8530.

The very hard leaves and the long prominent ligule distinguish this species from other

stoloniferous species of the same type. Roots of the grass were collected by Dr. Pole Evans

in the Northern Transvaal and planted at Prinshof. This description is from a single

flowering specimen collected from those roots.

D. glauca.

Perennial, densely tufted, with intravaginal innovation shoots on a short oblique or

descending rhizome. Culms glabrous, terete, erect or sub-geniculate, rather stout, up to

135 em. high, much or scantily branched from the lower nodes, 2-5 noded, lower internodes

included, upper exserted, uppermost longer than half the culm ; lower sheaths compressed,

from 5-12 em. long, loose, finely ribbed, more or less densely hairy at the base, otherwise

glabrous or with scattered rather long tubercle based hairs but never densely hairy ; ligule

delicately membranous, oblong obtuse, 3-5 mm. long, 3-lobed or toothed, sides adnate with

the sheath auricles, glabrous; blades rather firm mostly strongly folded, finely ribbed,

about 3 mm. wide to 30 cm. long, culm leaves shorter, finely scaberulous on the nerves on

152

either surface, glabrous, glaucous. Racemes about 4-7, sessile, obliquely ascending, about

10-14 cm. long, the lowest usually 2-nate or 3-nate, the succeeding variously arranged on a

short angular scaberulous glabrous axis 3-5 em. long, the terminal raceme solitary and erect,

rhachis of racemes, triquetrous, very narrowly winged, -3--5 mm. wide, scabrid, finely

pubescent at the junction with the main axis otherwise glabrous ; pedicels angled, slender,

scabrid, with shallowly concave membranous tips, 2-nate or often on short contracted

branchlets at the base of raceme ; spikelets loosely appressed or spreading, not imbricating,

often their own length apart, narrow lanceolate-acuminate, 4-4-3 mm. long by 1 mm. wide,

conspicuously but not very long, hairy ; lower glume an ovate, nerveless, rather firmly

membrous scale usually about -5 mm. long; upper glume equalling about half to three-

quarters the length of the spikelet, narrow ovate-obtuse, 3-nerved with lines of soft fine

obtuse hairs between the nerves and on the margins, produced beyond the tip of the glume ;

lower valve the size of the spikelet, 7-nerved, nerves smooth, rather close and equidistant,

with lines of straight soft appressed or somewhat spreading hairs between the inner lateral

nerves and on the margins, hairs up to 1.5 mm. long; pale and lodicules, hyaline, minute ;

fertile floret narrow oval acuminate equalling or slightly shorter than the lower valve, pallid

sometimes faintly flushed with purple, the flaps of valve closely approximating.

TRANSVAAL.— Pretoria: near Government House, Pentz, N.H.P. 8519, Pretoria,

Pentz, N.H.P. 8670, Heidelberg, Henley-on-Klip, Stent, N.H.P. H. 21592.

ORANGE FREE Stare.—Kroonstad, Hall, N.H.P. 8669; Fauresmith, Botanical

Reserve, C. A. Smith 4098, 3960; Breda 16.

Care Province.—Gordonia, Inkruip, Pole Evans and Pentz, N.H.P. 8555.

A tufted species very glaucous with foliage seldom above 30 cm, high and numerous

tall erect flowering stems, very near to ervantha but differing from that species in its longer

more rigid leaves, taller stouter culms, longer laxer racemes, more densely and long hairy

spikelets, etc.

Var. Bechuanica.

Sheaths longer, densely hirsute, culms stouter, racemes more numerous and spikelets

imbricating and more crowded.

Care ProvincE.—Mafeking: Langeberg Camp, Pole Evans 2421; Olifantshoek,

Pole Evans and Pentz, N.H.P. 8556; Upington, on quartzite kopje, Pole Hvans 2194.

D. natalensis.

Densely tufted perennial. Culms rather stout, up to about 120 cm. high, erect and

very straight or more or less geniculate, simple or branched, glabrous, smooth or often

scaberulous just below the panicle, 4-6 noded, lower internodes usually included, upper

exserted, uppermost very long. Sheaths of the lower leaves firm sub-compressed, loose,

up to about 25 cm., glabrous except at the usually hirsute base or more or less densely

beset with rather rigid sharp pointed hairs especially on the margins, those of cauline leaves

usually quite glabrous or more or less finely pubescent at base ; ligule rather firmly scarious,

up to 15 mm. long, rounded or truncate, sometimes scantily but rather long ciliate, often

with the margins produced and adnate to the auricles of the sheath mouth ; blades linear,

up to 60 cm. long by 8 mm. wide (fresh material) flat or folded, tapering to an acute apex,

smooth, or the upper surface minutely scaberulous, midrib rather broad and stout, lateral

nerves prominent and close, blades of the wpper culm leaves sometimes much reduced.

Racemes many, digitate or sub-digitate on a short (up to 5 cm. long) common axis, 12-23

em. long; rhachis about -5--8 mm. wide, triquetrous, narrowly winged, straight or flexu-

ous, scabrid on the margins and the facial angle, glabrous except at the minutely

pubescent base ; pedicels 2-nate or 3-nate or sometimes in fascicles of 4-5, scabrid with

truncate tips; spikelets not or laxly sub-imbricating ovate lanceolate acute, about 3 mm.

long ; lower glume minute but distinct, membranous ; wpper glume narrow, obtuse, 3-nerved,

153

equalling about half the spikelet, lines of fine silky obtuse hairs between the nerves and on

the margins ; lower valve the size of the spikelet, 7-nerved, minutely scaberulous on the

nerves and often on the spaces between, hairs between the inner lateral nerves and on the

margin not very long, straight, acute, usually closely appressed or somewhat spreading

with age and often mixed with rather rigid acute bristle-like hairs; fertile valve greenish

grey, as long as or slightly shorter than the lower valve, rather firmly membranous,

shortly apiculate, margins rather distant.

NaraLt.—Umkomaas, van Rensburg, N.H.P. 8630 (type) ; Maritzburg, McClean 184 ;

Zululand-Mtunzini Mogg, N.H.P. H.20026, H.20042 ; Ixopo, Glen Daun, Bruce, N.H.P.

8072 ; Matubatuba, Harrison, N.H.P. 8323; without precise locality Storey 21 ; Kentani,

in valleys Pegler 1092.

Near to D. Smuts Stent but very different in the field with more rigid leaves and

longer ligule ; the hairs of the spikelet are rather stiff and the nerves of the lower valve

minutely scaberulous.

D. littoralis.

Densely tufted perennial on short stout descending rhizome. Culms erect or genicu-

late, rather stout, glabrous, smooth, 3—4-noded usually branched and often sending out

fascicles of shoots from the lower nodes, up to 120 cm. high, lower internodes included,

upper exserted ; leaves very green, crowded and more or less flabellate at the base ; sheaths

loose, rather long, the lower shortly and appressed hairy at the base otherwise glabrous

or with few scattered hairs, compressed, finely and closely striate ; ligule up to 9 mm. but

usually much shorter, minutely fimbriate ; blades flat, linear, from a scarcely narrowed

base tapering to long acuminate points, glabrous tip scaberulous, about 30 cm. long by 7

mm. broad, midrib prominent below, primary lateral nerves about 4 on either side, not

conspicuous. Racemes 7-10, obliquely spreading, 10-16 cm. long, more or less whorled

or scattered on a short common axis 2-3 cm. (very rarely more) long, rather dense or some-

times interupted, rhachis of racemes narrow, triquetrous, margined, -5 mm. wide, scabrid,

glabrous ; pedicels 2-nate, longer about 1-3 mm., scabrid, angled, with membranous discoid

tips. Spikelets sub-imbricating, somewhat crowded, lanceolate, oblong, 3-5-4 mm. long

and 1 or slightly over | mm. broad, conspicuously hairy, hairs fine obtuse, under | mm.

long ; lower glume a well developed firmly membranous nerveless ovate-lanceolate obtuse

to sub-acute scale, 1-1-5 mm. long; wpper equalling about three-quarters the length of

the spikelet, 3-nerved with hairs on margins and between the nerves; lower valve the size

of the spikelet, 7-nerved, nerves often minutely scaberulous, equidistant or the inner

lateral pair somewhat remote from the central nerve, hairs between nerves and on the

margins appressed, ultimately spreading ; fertile floret as long as the lower valve, anthers

2-3 mm. long.

Care Province.—Albany Division : Grahamstown, Schonland 3739 (type), Howiesons

Poort, Worracher, N.H.P. 8671, Dyer, N.H.P. 8502; Port Elizabeth, Gunn, N.H.P. 7866 ,

Galpin 6384; Port St. Johns, Eagle’s Nest, Howlett 36.

Near to D. natalensis which is however a coarser grass with very firm leaves, more

numerous racemes and coarser almost bristle-like hairs on the spikelet.

A densely tufted very green species, flowering culms, 126 cm. high and foliage about

50 cm. high. More resistant to frost than other species under cultivation.

Var. prostrata.

Densely tufted giving off numerous runners that root (rather lightly ?) at the nodes

and always terminate in a flowering culm—they are in fact prostrate culms—fascicles of

flowering shoots are given forth from most of the nodes. A very blue variety.

154

Care Province.—Albany: Grahamstown, Pentz, N.H.P. 8021 and 8504; Port

Klizabeth, Gumn, N.H.P. 8517, 7973, 8494.

Gunn 8494 is a rather smaller form with more slender flowering culms and very long

runners which do not always terminate in flowering culms.

D. geniculata.

Perennial on a stout descending rhizome. Culms fascicled, prostrate erect up to 90

em. high, much branched from the base and from the lower nodes, nodes usually many,

the basal internodes short and included ; leaf sheaths glabrous or hairy at the base and

along the margins, smooth, rather loose, finely ribbed; ligule delicately membranous,

glabrous, up to 7 mm. long but often appearing shorter from the tearing away or folding

over of the delicate obtuse tip ; blades linear from a slightly narrowed base tapering to an

acuminate setaceous tip about 30 cm. long by 4 mm. wide, quite glabrous or sparsely hairy

with scattered long white hairs, scaberulous on the upper surface and margins, midrib,

slender, somewhat prominent below, indistinct above, finely ribbed, primary lateral nerves

scarcely distinguishable ; racemes 4-9 digitate or sub-digatate on a short axis, about 9-12

em. long, obliquely erect or rather widely (but not horizontally) spreading ; rhachis narrow

triquetrous, margined, scaberulous, pedicels triquetrous, 2-nate, scaberulous; spikelets

narrow, ovate, lanceolate, acute or sub-acute 3-4 mm. long, rather lax, not imbricating ;

lower glume up to -5 mm. long, membranous, nerveless ; wpper 3-nerved equalling half to

three-quarters the length of the spikelet with rather dense lines of soft spreading hairs

between the nerves and on the margins ; lower valve the size of the spikelet 7-nerved nerves,

smooth, equidistant, with lines of soft hairs between the inner lateral nerves and on the

margins, hairs about -5 mm. long, margins hyaline ; fertile floret equalling or slightly shorter

than the lower valve, lanceolate, yellowish to brown; anthers yellow, about 2 mm. long.

Carre ProvincE.—Wodehouse: Dordrecht Pentz, N.H.P. 8049 (type), 8498;

Oudtshoorn, De Rust Pentz, N.H.P. 8518; Willowmore, near Pienaars River, Pole Evans

and Pentz N.H.P. 8508.

ORANGE Free Stare.—Senekal, Cyferfontein, Joubert, N.H.P. 7952.

D. foliosa.

Perennial on a shortly creeping rootstock with extravaginal innovation shoots covered

with lanceolate-ovate membranous, closely ribbed hairy cataphyls. Culms slender up to

45 cm. high, shortly exceeding the leaves, 5-6 noded, branched from most of the nodes or

only from the upper, internodes (except the uppermost) short included, leafy throughout

but more so towards the top. Sheaths of the leaves thin, loose, the lower soon withering,

glabrous except for occasionally a few tubercle based hairs towards the mouth ; ligule thinly

scarious, irregularly toothed, about 1-5 mm. 3-lobed the central lobe broadly rounded,

the side lobes short or narrowly elongated and adnate to the sheath-auricles ; blades rather

dark green, linear lanceolate, flat, acutely long acuminate, 6-13 cm. long by 3-5 mm. wide,

rather firm, with narrow cartilaginous margin, smooth, glabrous or with a few scattered

tubercle based hairs mostly towards the base, midrib slender, prominent below, grooved

above. Racemes slender, 2-7, digitate or sub-digitate, on a very shortly elongated axis,

4-6 em.long; rhachis slender, flat on the back, trigonous on the face, narrowly winged, about

-5 mm. wide; pedicels scaberulous, 2-nate, angled or the shorter terete, sometimes on

short appressed branchlets. Spikelets lanceolate acute, about 2-5 mm. long ; lower glume

a small membranous scale ; wpper glume ovate acute or sub-acute, equalling about three-

quarters the length of the spikelet and about as broad at the base, 3-nerved, the hairs

between the nerves and on the margins very fine blunt and under 1 mm. long ; lower valve

the size of the spikelet, thin, 7-nerved, three middle nerves, slender but rather prominent,

others somewhat obscure, hairs between the inner lateral pairs short and appressed, the

marginal longer but also usually appressed ; fertile floret narrow ovate acute, slightly shorter

than the lower valve, margins approximating ; anthers and stigmas purple,

155

Care Provincr.—Kuruman, Vuilnek, Pentz, N.H.P. 8527.

A distinct species presenting, when growing, a bushy effect with the flat rather firm

leaves gathered towards the tops of the culms leaving the bases covered only with the

withering sheaths. The comparatively short flowering culms and racemes are also rather

distinct.

D. trichopodia.

Upright perennial, innovations intravaginal (?). Culms about 45 em. high, 2-3 noded,

simple, rather stout; densely and rather softly hairy in the upper part below the panicle.

Sheaths of the cauline leaves rather tight exceeding to very little shorter than the internodes,

glabrous except at the rather long and densely bearded nodes, smooth; basal sheaths

brownish rather firm, loose, like the ovate acute cataphylls densely silky hairy at the base

and more or less so further up ; lgule rounded or truncate, membranous, glabrous, rather

short or up to 3 mm. long, rather firm ; blades broad, flat, up to about 19 cm. long by 10

mm. wide, smooth, or scaberulous in the upper surface, glabrous except for a few scattered

long hairs on either side the ligule, midrib slender not prominent, primary lateral nerves

5-6 on either side. Racemes compound towards the base 5-10, solitary on an elongated

common axis forming a more or less contracted panicle, axis up to 12 cm. long, angled and

ribbed and covered with long cream, silky rather thick walled acutely acuminate hairs,

branches short and appressed ; rhachis trigonous, flexuous, margined, about -5 mm. wide,

densely hirsute along the margins and on the face with the same long shining rather rigid

hairs ; pedicels 2-nate or solitary or more often in fascicles of 3-6, densely long, hairy, the

hairs at the discoid tips exceeding to spikelets. Sprkelets crowded almost hidden by the

copious hairs of the rhachis and pedicels, quite glabrous and smooth; light golden brown

paling to yellow at the tips, ovate acute, rather plump about 2 mm. long ; lower glume 0,

upper reduced to a thinly membranous emarginate nerveless or faintly 1-3 nerved scale

about -5 mm. long; lower valve the size of the spikelet, thinly membranous 3-5-nerved,

the outer nerves when present some distance from the margins, very faint and evanescing

above, the two inner lateral nerves anastomosing with the centre nerve just below the tip,

pale and lodicules wanting ; fertile valve golden brown rather tough the hyaline flaps some-

times overlapping at the tip or distant to the base ; pale resembling the valve but more

deeply coloured on the back and with the flaps narrow towards the tip and overlapping at

the base; anthers about 1-5 mm. long, brownish.

Care Province.—Komgha Division, grassy slopes near Kei Mouth, Flanagan 987.

The only specimen I have seen of this grass which was collected by Flanagan in 1892,

there are no barren shoots and the tips of all the leaves are broken off.

Very near to D. uniglumis but with broader leaves, longer and more copious setae on

the axis and pedicels and larger spikelets and more compound racemes, etc. Also very

near to D. Grantii Hubbard from which it differs in the densely bearded sheath-nodes, wider

leaves, longer and more densely silky racemes, etc.

D. apiculata.

Perennial, rather loosely fascicled with intravaginal and extravaginal innovation shoots

the latter covered at the base by linear lanceolate, strongly striate cataphylls. Culms

geniculate, slender, simple, 2-3 noded with lowest internode very short, smooth, glabrous

or softly pubescent for a short distance below the panicle ; leaf sheaths narrow, rather loose,

striate finely hirsute, those of the slender barren shoots about 3-5 cm. long; ligule not

exserted, firmly membranous and brownish, up to | mm. long, sheath margins not produced ;

blades narrow linear from 1-3 mm. to 3 mm. wide and up to 17 em. long, not or slightly

narrowed at the base, long tapering to a fine point, rather rigid hairy on both surfaces,

margins revolute, midrib slender but prominent with 3-5 primary lateral nerves scarcely

differentiated from the secondary, all rather prominent and the leaf therefore having a ribbed

156

effect. Racemes 4-7 and up to 8 cm. long solitary common axis very short, slender, angular,

ribbed, minutely pubescent and scabrid, rhachis about -5 mm. wide, triquetrous with

narrow scabrid margins and occasional fine hairs below the pedicels; internodes 2-5-5

mm. long; pedicels binate or ternate or in fascicles of 3-5, very uneven in length the

longest commonly up to 3 or 4 mm. long, slender, scabrid with small discoid tips. Sprkelets,

rather loosely spreading, about 2 mm. long, ovate acute, rather plump and quite glabrous ;

lower glume minute broadly clasping, thinly membranous to hyaline, truncate nerveless ;

upper glume ovate obtuse equalling less than half the spikelet, thinly membranous promi-

nently 3-nerved ; lower valve equalling or almost, the spikelet in length, very thin rather

narrow ovate acute, prominently 5-nerved, nerves smooth, equidistant ; pale and lodicules

minute ; fertile valve determining the size of the spikelet, elliptic oblong, minutely apiculate,

coriaceous, light yellowish brown when young soon deepening to dark brown except at the

base and the small apiculus which remain yellow ; pale of the same texture and colour

as the valve; lodicules broadly cuneate about -5 mm. long; anthers little over 1 mm.

ovary narrow ovate, stigmas long plumose at the tips of the styles, grain not seen.

Swazitanp.—Burtt-Davy 2820 (type); Transvaal: Barberton, Devils Kantoor,

Kaapsche Hoop, Pole Evans 1013.

Nearest to D. Ibura but a perennial with slenderer, fewer and not compound racemes

narrower leaf blades, etc. The small yellow apiculus of the dark valve was very

noticeable.

Vol. III. Price, 7s. 6d. Pare. 2

IB Othalta

J A RECORD OF

CONTRIBUTIONS

FROM THE

NATIONAL HERBARIUM

UNION OF SOUTH AFRICA

PRETORIA

wtlliy Yy YY

EDITED BY

. B. POLE EVANS, C.M.G., M.A. (Cantab.), D.Sc. (Wales),

CHIEF, DIVISION OF PLANT INDUSTRY, DEPARTMENT OF AGRICULTURE AND FORESTRY

—

PRETORIA ; AND

DIRECTOR OF THE BOTANICAL SURVEY OF THE UNION OF SOUTH AFRICA

PUBLISHED JANUARY. 1937.

AP .S @1790__19026__700

CONTENTS.

PAGE

An Account OF THE SoutH AFRICAN SPECIES OF TRIBULUS Tourn. ex Linn. .. 159

A Nove on THE SoutH AFRICAN SPECIES OF XIMENIA Linn. AND THEIR

POSSIBLE Economic UsEs .. mi ve te sy aie ee so | (LL)

| A Species or PacnystigMa Hochst. FROM THE TRANSVAAL : es Ss

A REVISION OF THE SoutTH AFRICAN SPECIES OF HELITOTRICHON Bess. ex

Schultes Se ae ae ot op i Pe ws i oo tet

A Revision oF THE SoutH ArricaNn SpEcrES OF BRACHYLAENA R. Brown .. 205

AN ENUMERATION OF PLANTS COLLECTED IN THE NORTHERN TRANSVAAL eZ

THE GENUS ELtyonurus Humb. and Bonpl.” 1x Sourn AFRica .. = Bo) at)

A QvEstion oF NOMENCLATURE tis ne ae 68 a si meee D(a

NEWLY DESCRIBED SPECIES ae ae a6 an Ue is as ore OS.

AN ACCOUNT

OF THE

SOUTH AFRICAN SPECIES OF TRIBULUS

Tourn. ex Linn.

H. G. Scnoweitckerpt, B.Sc., Ph.D., F.L.S.

In Southern Africa, the genus Tribulus L. is of considerable economic importance,

sipce several species have been found to be the cause of a troublesome disease in sheep,

known as “ geeldikkop.”

Preliminary experiments and investigations carried out during the past few years by

veterinary and agricultural officers in connection with this disease, have resulted in the

accumulation of much material in certain South African herbaria. Increasing difficulties

were encountered in naming such material forwarded for identification from various

sources to the National Herbarium, Pretoria. These difficulties were due to the fact that

the species appeared to be ill-defined and the writer was consequently encouraged to

undertake a study with a view of defining more clearly the limits of these species.

During the course of this investigation the material in the South African and that

from Southern Africa in several of the larger European herbaria was studied. [Royal

Botanic Gardens, Kew; Brit. Museum (Natural History); Berlin-Dahlem; German

Univ., Prague; Mus., Stockholm; Univ., Ziirich.]

The conclusions arrived at in this paper have thus been based almost entirely on the

study of dried material. As will be pointed out, the results have proved to be somewhat

inadec tate and unsatisfactory, i.e. the limits of several species as yet remain uncertain.

Recourse to field work and breeding experiments seems desirable, in order that a clear

concept of the limits of such species be obtained which appear to exhibit an extremely

wide range of variation in vegetative, flowering and fruiting characters.

The first record of the genus Tribulus in a work relating to South African plants

appears to be that of Thunb. Prodr. 79 (1794) where the Mediterranean species 7’. terrestris

L. is enumerated. Harvey in his Genera of S. Afr. Pl. 46 (1838) lhkewise only mentions

this species, and remarks that it is “a very common weed in cult. ground in the months

of Noy. and Dec. It is perhaps merely naturalised from the South of Europe.” Presl,

Bot. Bemerl:. 29-30 (1844) enumerated five species from the (present-day) limits of the

Cape Province. He considered these five species to be distinct from T. terrestris L. In

how far his views are justified will be dealt with under the species concerned. Harvey

apparently overlooked Presl’s paper, as only two species are enumerated in the second

edition of his Genera of S. Afr. Pl. 36 (1868), although the first volume of the Fl. Cap.

(1859-60) had by then been published. In the latter Sonder mentions four species, ac-

cepting and rejecting some of Presl’s species. During the time between the appearance

of Fl. Cap. I. and the present day, relatively few references to South African Tribuli are

to be found in literature. Apart from Hngler’s account in Veg. der Erde, no comprehensive

160

enumeration of the species exists. In Fedde Rep. XXIV (1927) Dinter enumerates the

species which had been recorded from South West Africa up to the year 1917 and Range

did very much the same thing, on the basis of his own collections, in Fedde Rep. XXXVI

(1934).

The following pages are thus intended to be of a revisional character as well as a

critical exposition of the conclusions drawn from the study of a wide range of herbarium

material from southern Africa.

In the list of localities given under each species only such specimens as were actually

seen and examined are quoted.

TRIBULUS Tourn. ex. Linn. Syst. ed. I (1735); Linn. Sp. Pl. ed. I. 386 (1753) :—

Sepals 5, deciduous or semi-persistent. Petals 5, spreading, shorter to longer than

the sepals, deciduous. Stamens 10, those opposite the petals somewhat longer than those

alternating with the petals; filaments subulate, those opposite the sepals with both an

extra and intrastaminal basal gland; anthers cordate or oblong-cordate ; intrastaminal

glands free or connate to form a shallow cup at the base of the ovary. Ovary sessile, covered

with erect bristle-lke hairs, 5-lobed, consisting of 5 intergrown carpels; each carpel with

3-5 ovules; style short or absent; stigma conspicuous, 5-angled, pyramidal or hemis-

pheric, formed by intergrowth of the 5 stigmatic lobes. Fruit 5-angled, at length breaking

up into 5 indehiscent cocei; cocci dorsally tuberculated, unarmed, spinous, winged, or

winged and spinous, 1-5 seeded. Seeds exalbuminous.

Xerophilous, mostly annual or perennial herbs, more rarely perennial shrubs ; branches

prostrate or ascending, more rarely erect. Leaves opposite, one usually larger than the

other, more rarely alternate, bistipulate, abruptly pinnate ; leaflets opposite, entire, some-

what oblique. Pedicels axillary, one-flowered. Petals usually yellow.

TAXONOMICALLY IMPORTANT CHARACTERS.

The habit of the plant is of some importance. The erect shrubby perennial nature

of T. excrucians Wawra is a constant and unique character among the African species

known up to the present day and by means of which it may readily be distinguished from

the other species. The remaining members are all very much alike in habit, 1.e. usually

prostrate or semi-prostrate annuals or perennials.

In purely vegetative characters each of these species shows a marked degree o. varia-

tion, being very plastic with regard to the size, shape, indumentum, etc., of the vegetative

paits. Such variation is frequently met with in plants inhabiting arid or semi-arid regions

and may probably be accounted for by the edaphic and climatic conditions under which

the plants develop.

The length of the pedicel appears to be of some taxonomic value. In the large-flowered

prostrate species it always exceeds the subtending leaf, and even in extreme cases still at-

tains the length of the leaf, whereas in the small-flowered species the pedicel is usually

exceeded by the leaf.

The size of the flower and hence the petals is of some value, since species such as 7.

terrestris LL. may be excluded from the “ large-flowered ” species on basis of this character.

The ratio of length of petals to sepals is of very little and in the writer’s opinion probably

of no taxonomic value. Many measurements carried out with a view to detecting whether

that ratio is a character of some constancy proved it to be of little value; the flowers on

even one and the same individual show such strong variation that allied species (distinguish-

able by other characters) were found to overlap.

The nature of the intrastaminal glands situated at the base of the ovary has been

found to be of great taxonomic value, and two forms may be recognised: (a) glands free,

161

distinct, and not connate (fig. I.), (6) glands connate to form a shallow cup at the base of

the ovary (fig. IT.). At times the region of intergrowth of the glands is not readily visible

in herbarium specimens, and in such cases soaking and careful dissection is essential.

The shape and size of the stigma appears to be correlated with the nature of the intra-

staminal glands. A hemispheric somewhat asymmetric stigma (fig. I.) is found in species

with free glands and small flowers, whereas a very much larger pyramidal almost sym-

metric stigma (fig. II.) is correlated with intergrown glands. The length of the style

appears to be variable and for this reason was discarded as a taxonomically useless character.

Taxonomically the fruits are of the greatest importance. In their absence several

closely allied species, resembling each other in vegetative and floral characters, are readily

confused ; in order to identify large-flowered species with certainty the presence of fruits

is indispensable. Whereas the fruits of species such as 7. cristatus Presl exhibit characters

which have proved to be fairly constant and thus taxonomically of the utmost importance,

other species exhibit but little constancy in this respect. J. excrucians Wawra may serve

as an example of the species in which the fruits exhibit great variation both in regard to size

and degree of spinosity of the individual cocci. Owing greatly to the limited amount of

material available for study to the older authors, too much value was attached to the

nature of the fruits. Study of a wide range of material has shown that fruit characters

of several species are to be used with some discretion. This remark brings forward the

question of natural hybrids about which Dinter states that he has never observed any

plant possessing “ Bastardeigenschaften.” As will be shown later, there appears to be

evidence in favour of the occurrence of natural hybrids among the species of Tribulus from

Southern Africa.

KEY TO THE SPECIES.

I. Intrastaminal glands connate to form a shallow cup at the base of the ovary; stigma

slender, usually pyramidal (fig. II.) :—

A. Plant an erect shrubby perennial attaining a height of 1 metre or

more ; main stem woody with corky bark; fruits (cocci) unarmed

or armed with one or two pairs of short lateral spines; spines

at times somewhat flattened and almost winglike................. 1. excrucians.

AA. Plant annual or perennial, prostrate or at length with the

extremities of the branches somewhat ascending :

B. Cocci winged; wings armed with spines or wholly devoid of

spines :

C. Wings of cocci with numerous pungent slender marginal spines

and very often a long spine arising from the centre of each

wing near the dorsal crest of the coccus..........-........-.-- 2. cristatus.

CC. Wings of cocci without or with several marginal blunt teeth

but never with a spine arising from the surface of the wing,

leathery and subrigid, or thin papery and brittle in texture

extremely variable’ in shape......-...-.-- + sss ete cerese ee eeee 3. pterophorus.

BB. Cocci never winged but armed with 2 or more pairs of spines ;

spines variable in length, arising from the sides of the dorsal

erestaor near the hase, of, {he coccus:.....52--..---.------~ - 4. Zeyheri.

II. Intrastaminal glands free, never connate to form a shallow cup;

stigma hemispheric, never slender (fig. I); petals 2-12 mm. long ;

sepals 2-6 mm. long...........--2-e reste tect nett nee eee cees 5. terrestris. .

162

ENUMERATION OF THE SPECIES.

1. T. excrucians Wawra, in Sitzb. Akad. Wien. Math.—Nat. XX XVIII. 557 (1860).

Syn.: ZT. micans Welw. Apont. 566 (1858), nomen subnudum.

T.. terrestris Oliv., in Fl. Trop. Afr. I, 283 (1868) pro parte, non Linn. ; Hiern.,

Welw., Cat. Afr. PJ., 105 (1896).

é T. Pechullai O. Kuntze in Jahrb. Bot. Gart. Berl. IV, 262 (1886); Engler,

Jahrb. X, 31 (1888) ; Schinz in Bull. Herb. Boiss. II, 187 (1894) ; Heering

and Grimme, Untersuch. Weideverh. Deutsch-Siidwest afr., 27 (1911) ;

Engler, Veget. der Erde. 9, III, 1, 738 (1915), cum fig. 343 P-R et Pflzfam.

ed. II, 19a, 176 (1931), cum fig. 84 P-R.

T. Zeyhert Sond. var. Pechuelit Schinz, in Verh. Bot. Ver. Prov. Brandenb.

XXIX, 54 (1887).

T. inermis Engl., Bot. Jahrb. X, 32 (1888), non Krahk.; Heering and

Grimme, Untersuch. Weideverh. Deutsch-Siidwestafr., 27 (1911).

T. erectus Engl., Bot. Jahrb. X, 32 (1888) ; Dinter, Deutsch-Siidwest-Afrika,

86 (1909); Engl., Veget. der Erde 9, III, i, 738 (1915); Dinter, in Fedde

Rep. XXIV, 14 (1927).

An erect or suberect perennial shrub up to 1.5 m. high, with a short stem, up to 5 cm:

thick near the base. Branches erect, robust, longitudinally striate, terete, pubescent:

glabrous in age; internodes up to 6 cm. long, but usually much shorter. Leaves unequal ;

the larger up to 7 cm. long, 6-10-jugate ; the smaller up to 3 cm. long, 3-6-jugate ; stipules

up to 7 mm. long, linear-lanceolate to obliquely lanceolate, acute, silky pubescent on the

outer surface, less so on the inner surface ; petiole silky pubescent, slightly winged towards

the apex; leaflets obliquely oblong to lanceolate, acute, up to 25 mm. long and 8 mm.

broad, but usually much smaller, densely silky pubescent on the lower surface. Pedicel

fairly short, 1.5—-2 cm. long, terete, silky pubescent, markedly striate in age. Flower-buds

ovate, markedly acuminate. Sepals linear-lanceolate, up to 11 mm. long and 3 mm. broad,

silky pubescent on the outer surface. Petals broadly cuneate, up to 25 mm. long, often

much smaller. Filaments 3.5 mm. long; anthers 2.5 mm. long. Intrastaminal glands

connate to form a very shallow cup at the base of the ovary. Stigma pyramidal, about

2.5mm. long. Mature fruit at length breaking up into several (usually less than 5) cocci ;

cocci unarmed or armed, some without any signs of spines, merely tubercled laterally and

dorsally, densely but very minutely pubescent, others again armed with 1-3 pairs of lateral

spines, which at times are much flattened and almost wing-like (fig. III).

Ancoua.—Loanda: July, Menyhart, 228. Lobito Bay: Aug., Obermeyer, in Herb.

Tvl. Mus. 32816. Benguela: Jan., Wawra, 299 (types in Herb. Mus. Nat. Hist. Vindob).

Between Mossamedes and Rio Bero: Apr., Hopfner, 5. Mossamedes: July:

Welwitsch, 1579, 1580. Oct., Briihl, 5, 6. Jessen, 329. March, Fritzsche, 2.

Sourn Wesr Arrica.—Swakopmund: Dinter 22. Belck, 63a; upright perennial in

bed of river, Bradfield, 550. 580. Swakopriver: May, Marloth, 1457. Nov. Rewsch, 63a.

Walfish Bay: Okahandja: Oct., Luderitz, 147. Husab: June, Fleck, 727. March,

Dinter, 8446. Kuwosis: Oct., Schenck, 433. Haigamchab: Jan., Galpin et Pearson,

7607, 7637; shrub along banks of Swakopriver, June, Gurich, 129, 147. Khanthal: Apr.,

Engler, 6058. Ukuib: Dec., Pogge, 9. Gamkoigchas: May, Dinter, 207. Usakos: June,

Marloth, 1457a; Nov., Schinz, 1120, Otjimbingue: May, Marloth, 1300, 1390. Oka-

handja, cult.: Nov., Dinter, 207.

Without precise locality: Luderitz, 208. Schenck, 452. Nels, 16. Francois, 38.

Pechuel Losche, 36.

163

In the description of 7. exerucians, Wawra l.c. cites the type as being his 290. The

specimen in the Vienna Herbarium and a duplicate of same in the Zurich Herbarium both

bear the number 299. The number Wawra 290 is thus probably a typographic error for

299.

The shrubby nature, together with the peculiar ashy-grey silky appearance of the

plant, are the most reliable characters by which this can be distinguished from other related

species. Apparently 7. excrucians has been overlooked by all authors cited below since

they do not refer to this species in their work relating to Tribulus.

O. Kuntze (1886) l.c. in describing T. pechuelic apparently did not know that he was

dealing with a shrubby species (the label on the type specimen gives no information with

regard to the habit of the plant). He consequently made use of the number of leaflets

per leaf, the nature of the fruit and other characters to distinguish his species from allied

plants such as 7’. Zeyheri and T. cistoides. Both the latter, however, are prostrate in habit,

the branches only rarely ascending to a few centimetres above the level of the substratum,

whereas 7. excrucians (T. Pechuelli of various collectors) is a perennial undershrub or shrub ¢

attaining a height of up to 1.5 metres. It also possesses a short main stem which may

equal a “ man’s wrist’ in thickness.

The colour of the flowers may either be pale yellow (Dinter) or yellow with a dark

claw (Marloth). The size of the petals is variable and may be anything from 15-24 mm.

long; the petals are + 2.5 times the length of the sepals.

Schinz (1887) l.c. pomted out that contrary to Kuntze’s statement regarding the un-

armed nature of the fruits, the cocci showed the presence of small spines. As a consequence

of the armed nature of the cocci, Schinz lowered the rank of this plant and considered it

to be a variety, viz. T. Zeyhert Sond. var. Pechuelii Schinz. Schinz at this time was un-

aware of the shrubby nature of the species.

Engler (1888) l.c. independently described a shrubby species (7’. erectus) from South

West Africa. He also hinted at the presence of a second erect species in the form of T.

imermis since he described the habit of this plant as ““ procumbens? ”’. Engler had over-

looked T. exerucians Wawra but had naturally consulted Kwntze’s type in drawing up his

descriptions ; being a hypercritical worker he considered T. erectus and T. inermis to be

distinct from Kwuntze’s species (the latter is a very fragmentary specimen only possessing

young fruits). His views which were naturally based on the material available for study

at the time appear to have been by no means unreasonable.

As a result of Engler’s work as well as the study of further material Schinz (1894) lc.

subsequently somewhat modified his earlier views regarding the taxonomic position of

T. Zeyheri Sond var. Pechuelii Schinz. He revived 7. Pechuelii Kuntze, and considered it

a valid species still closely allied to T. Zeyheri Sond. Engler’s species T. inermis and T.

erectus had also come to his notice and realising the great variability of the fruits he cor-

rectly considered them to be conspecific with 7. Pechuelii Kuntze. Attention may be

drawn to the fact that 7. micans Welw. from Angola had apparently been overlooked and

perhaps wilfully excluded in the absence of a description by all three authors mentioned

above. [Since Welwitsch did not draw up a valid diagnosis 7. micans Welw. is a nomen

subnudum.| The type of this plant in Herb. Mus. Brit. undoubtedly is conspecific with

T. excrucians Wawra, which is the oldest validly published name for this species. Dznter

(1927) l.c. also considers 7’. erectus, T. inermis and T. Pechuelii to be conspecific, but adopts

the name T. erectus to designate the species which, however, is not in accordance with the

International Rules of Nomenclature.

As a result of my own studies of dried material I have arrived at the conclusion that

all the species in question are conspecific. The fruits of many specimens seen showed

either the presence or absence of spines; at times the spines were so weakly developed

that they could easily have been overlooked. The fruits of even one and the same plant

164

show a fair degree of variation. Dunter (1927) lc. says of T. erectus that this species “ ist

in jeder Beziehung konstant, wenn auch hochst wahrscheinlich aus T. Zeyheri hervorge-

gangen,” furthermore that the fruits may either be spineless or weakly thorny.

Marloth 1457 and 1457b from Swakop, 8.W.A. and Pocock 975 from Benguella, Angola

belong to 7. excrucians. These sheets, however, have elliptic acute leaflets which are much

larger (25 mm. |.) and broader (up to 9 mm.) and more markedly veined than those of the

typical species. Hngler (1888) lec. places Marloth 1457 and 1457b under T. Pechuelia but

mentions that Marloth collected “ diese Art....... in zwei Formen.” The cocci of Marloth

1457 and 1457b are glabrous in age except for a few bristly hairs, whereas in Pocock 975

they are densely minutely pubescent. In all these specimens the cocci are armed with 1-2

pairs of short lateral spines and the whole appearance of these fruits is somewhat different

from those of typical 7. exerucians. But having studied a wide range of material I am of

opinion that all the sheets cited above probably only represent one variable species.

Study of this interesting species together with breeding experiments will undoubtedly

be necessary in order to prove whether in the foregoing only a single or perhaps more than

one closely related species is involved.

T. excrucians Wawra appears to have a very limited geographic distribution. It is

only known to occur in the western regions of the Mandate of South West Africa and Angola,

where it 1s mainly found growing in or near the dry sandy beds of desert rivers.

2. T. cristatus Presl, Bot. Bemerk., in Abh. Béhm. Ges. Wiss. V. 3.29 (1844) ; Sond.,

in Fl. Cap. I. 354 (1859-1860); Glover, in Ann. 8. Afr. Mus. IX. mi. 170 (1913); Engl.,

Veg. der Erde 9. III. 1. 738 (1915), cum fig. 343 V ; Dinter, in Fedde Rep. XXIV. 15 (1927) ;

Engl., Pilzfam. ed. II. 19a. 177 (1931), cum fig. 84 V; Range, in Fedde Rep. XXXVI.

249 (1934).

As Presl’s description is somewhat incomplete and that of Sonder in the Fl. Cap. was

based on the same gathering and consequently does not add much to it, it was thought

necessary to draw up an amended description based on the material now represented in

the South African and Kuropean herbaria consulted.

An annual, or possibly a biennial or perennial plant. Branches prostrate, radiating

from the much branched rootstock, 12-100 cm. long, usually again branched but not con-

spicuously so. Branches, stipules, leaves, peduncles and calyx, in fact all vegetative parts

of the plant usually hirsute with bulbous-based bristly hairs; internodes terete, striate,

0-5-6-0 em. long. Leaves unequal, the larger 1-5-6-5 cm. long with 3-8 pairs of leaflets ;

the smaller (subtending a branch or flower) with 2-4 pairs of leaflets and much shorter

than the first internode of the subtended branch, or the peduncle; stipules 2-5-8 mm.

long, obliquely lanceolate, acuminate, upper and lower surface hirsute or almost glabrous,

with marginal bulbous-based hairs; petiole hirsute or minutely pubescent, at times some-

what winged ; leaflets hirsute or glabrous except along the midrib above, hirsute and paler

beneath, obliquely oblong, obtuse or acute, with marginal bulbous-based hairs, up to 14 mm.

long and 6 mm. broad but usually much smaller. Pedicel 2-5-4 em. long, usually 2 to 3

times the length of the subtending leaf, striate, terete, set with tubercle-based hairs inter-

spersed between a finer indumentum. Jlower-buds abruptly acuminate. Sepals per-

sistent, acuminate, 8-12 mm. long. Petals broadly cuneate, up to 25 mm. long, apparently

always pale yellow in colour, twice or slightly more than twice as long as the sepals. Fila-

ments up to 4 mm. long, bearing anthers about 2 mm. long. Style one-third to two-thirds

as long as the stigma; stigma conical-oblong, slender and 2-3-5 mm. long. Divse at base

of ovary a shallow cup formed by the intergrowth of the intrastaminal glands. Mature

fruits variable in size, glabrous except for a few bristle-like hairs, 6-25 mm. long, 12-30 mm.

in diameter including the wings, finally breaking up into 5 cocci ; cocci with lateral oblong

rounded wings, dorsally ridged and set with tuberculate bristly hairs; wings rounded,

bransversely markedly striate with marginal subrigid acute spines of unequal length often

165

bearing an analogous spine dorsally from the centre of each wng near the body of the

coccus, a character which appears to be unique and by which this species may readily be

distinguished from all the other South African species. This spine may sometimes be

intergrown with the wing and in such cases is less conspicuous (Fig. IV).

Sout West Arrica.—Great Namaqualand: Elephantenfluss: Range, 1483; Eirup ;

nr. Marienthal: March, Steyn, 22533. Gouchassib R.: Aug., Range 727. Nabaos nr.

Keetmanshoop: Apr., Range, 1313. Kuibis: Sandstein plateau: Jan., Dinter, 1166 and

Tafelbergsandstein : Jan., Dinter, 1248. Seeheim, sandy bed of Fish river: Apr., Engler,

6633 and Apr., Dinter, 2967. Klein Karas: October, Dinter, 5055 and Apr., Ortendahl, 9%).

Satansplatz: March, Dinter, 2041. Wasserfall: Jan., Pearson, 3154. Between Dabai-

gabis and Griindoorn: Febr., Pearson, 3154.

Cape Proyince.—Little Namaqualand: Verleptpram, stony hills on the Orange river :

Drége, 7160 (type deposited in Herb. Univ. Germ. Prag.). Herbert: Douglas, Orpen in

Mus. Austr. Afr., 14494. Prieska: Prieska, on sand dunes north of Orange river: March,

Wilman, 3039 and 3040; in sand near Prieska, March, Bryant, 880 and J. 251.

Examination of many herbarium specimens has led me to believe that the above species

is an annual. Specimens such as Dinter 5055 and Pearson 7863, however, suggest that the.

plants may be biennial and even perennial since these sheets possess a moderately branched

rootstock bearing the remains of dry withered branches at the apex. Bryant 880 in Herb.

Kew. bears a note to the effect that the species is annual.

T. cristatus Presl appears to be consistently prostrate in habit with the branches radiat-

ing from the crown of the rootstock. The length of the internodes of the branches is a very

variable one; robust specimens’ usually have longer internodes than weaker specimens.

This character appears to be correlated with the size of leaf, leaflets, flower and fruit, i.e.

a weak specimen usually is smaller in all vegetative and reproductive characters than a

luxuriant specimen. There does not appear to be a great deal of variation in the degree

of hairiness in the vegetative parts of the species.

Bryant 880 bears a remark to the effect that this species does not “ vary in flower or

shape of fruit.” From my own observations of herbarium material I cannot agree with

the first part of this statement. The smallest flowers so far seen had sepals 8 mm. long

and petals about 1° mm. long, the largest sepals up to 12 mm. long and petals up to 25 mm.

long. The colour of the petals appears to be consistently pale yellow in colour and accord-

ing to Bryant the “ flowers open for an hour or two in the morning ”’ and are “ very fuga-

cious.” The shape and consistency of the fruits undoubtedly form the best diagnostic

characters of the species. Whereas the size of the fruits is variable, the shape and texture,

however, is fairly consistent. The largest fruits seen had wings about 25 mm. long and

up to 10 mm. broad, but on an average the fruits are much smaller. The species flowers

during the months of November and Apmil.

The plants appear to prefer a sandy substratum and the species therefore is primarily

psammophytic and according to Bryant is “ very rarely found on hills.”

The distribution is fairly restricted and the species appears to be limited to the sandy

regions of Griqualand West, Little and Great Namaqualand. It has not yet been recorded

further north than the Mandate of South West Africa and its southern limit appears to be

approximately the Orange River basin.

Bryant has observed that this species * often grows alongside 7. terrestris L. but never

hybridises apparently.” Dinter regards T. cristatus Presl as being a very good species

having nothing in common with T. pterophorus Presl, an opinion which I am inclined to

endorse.

If accompanied by young or preferably mature fruits, T. cristatus Presl may always

be readily indentified. It is one of the most clearly defined species in the whole genus

and owing to its characteristic fruits, taxonomically occupies a somewhat isolated position.

166

Among the material examined none was observed to exhibit intermediate characters,

an indication that this species does not tend towards natural hybridisation.

3. T. pterophorus Presl, Bot. Bemerk, in Abh. Bohm. Ges. Wiss. V.3. 29 (1844) ;

Sonder, in Fl. Cap. I. 353 (1859-1860); Dinter, Deutsch-Siidwest-Afrika, 86 (1909) ;

Glover, in Ann. §. Afr. Mus. IX. in. 170 (1913); Engl., Veg. der Erde, 9. III. i. 738 (1915),

cum fig. 343 U; Burtt Davy, Flow. Pl. & Ferns I. 187 (1926) ; Dinter, in Fedde Rep. XXIV.

15 (1927) et Fedde Rep. Beih. LIII. 50 (1928); Engl., Pflzfam. ed II. 19a. 176 (1931),

cum fig. 84 U; Range, in Fedde Rep. XXXVI. 249 (1934).

SYN. —

T. alatus Drege, Zwei Pflzgeogr. Docum. 227 (1843), non Del.

T. securidocarpus Engl., Veg. der Erde, 9. III. 1. 738 (1915) in obs., cum. fig. 343 8.

a, b, ¢; Dinter, in Fedde Rep. XXIV. 15 (1927); Hngl., Pflzfam. ed. II. 19a.

176 (1931), cum fig. 84. a, b,c; Range, in Fedde Rep. XXXVI. 249 (1934).

T. securidocarpus Engl. forma vulgaris Hngl., l.c., cum fig. 343, S. a, b.

T. securidocarpus Engl. var. subtruncatus Engl., l.c., cum fig. 343. 8. ¢.

T. albescens Schltr. ex Dinter, in Fedde Rep. XXIV, 14 (1927), nomen tantum ;

Engler, Pflzfam. ed. II. 19a, 176 (1931).

An annual. Branches prostrate, radiating from the much branched crown of the

rootstock, up to 100 cm. long and usually somewhat branched again, in all vegetative

parts more or less densely hirsute with a fine indumentum scattered between which are

much longer bulbous-based hairs; internodes up to 10 em. long, usually much shorter,

striate, terete. Leaves unequal; the larger up to 6 crh. long, 6—-9-jugate; the smaller

(subtending a branch or pedicel) up to 3-5 em. long, 3-6-jugate ; stipules 3-5-6 mm. long,

obliquely lanceolate, acute, upper and lower surface pubescent, with marginal tubercle-

based hairs; petiole not winged; leaflets obilquely oblong, sub-acute or oblique, up to

12 mm. long and 6 mm. broad, densely silky beneath, less so on the upper surface. Pedicel

1-3 times as long as the subtending leaf, up to 3-5 ¢.m long. Flower-buds ovate in outline,

hardly acuminate. Sepals lnear-lanceolate, up to 10 mm. long and 2-5 mm. broad, acute,

silky pubescent without. Petals broadly cuneate, 1-5 to 2 times the length of the sepals,

up to 20 mm. long and apparently bright yellow to orange in colour. Filaments up to

4 mm. long, bearing anthers up to 3-0 mm. long. Style short ; stigma broadly pyramidal,

1-25-2-5 mm. long. Intrastaminal glands united to form a shallow cup at the base of

the ovary. Carpels and young fruit minutely pubescent. Mature fruit winged, glabrous

or minutely pubescent, at length breaking up into 5 cocci; cocci winged, with several

spines on the dorsal crest, each terminating in a tubercle-based hair; wings extremely

variable in shape, size and texture, often rounded, oblong, narrowed and triangular or

subtruncate, transversely striate and with one or several teeth on the margins, more rarely

irregularly dentate, never spiny, papery, brittle or coriaceous and tough, minutely pube-

scent or glabrous, up to 18 mm. long and 10 mm. broad, but often very much smaller and

showing mere traces of wings (Fig. V).

Sout West Arrica.—Gt. Namaqualand: Marienthal: March, Dinter, 2022; March,

Steyn, 22555 and 22546. Haribes: 40 km. 8.W. of Marienthal, Apr., Engler, 6578, 6579,

6592. Garis: Oct., Hartman, 155, 155b. Between Packrien and Leberfluss: Tvrotha, 43.

Sandverhaar: sand dunes, Febr., Pearson, 4693; tiefer sand, Jan. Dinter, 1187. Kubub-

Flache: March, Range, 232. Seeheim: Apr., Dinter, 2956; sandy bed of Fisch river,

Apr., Engler, 663. Holoog: dry river bed, Pearson, 4120. Klein Karas: Dinter, 5101.

Great Karasberge: Noachabeb, 1918, Blank, sn. Keetmanshoop: Fenchel, 29. Sand-

fontem: Wailman, 2177. Satansplatz: Dinter, 2042; Ariamsvlei, farm Walserbrunn,

Ortendahl, 316. Orange River: Gaidib, Dec., Dinter, 5138, Without precise locality, Afr. :

Fleck, 26a.

167

CarE Proyvince.—Little Namaqualand: Orange River, Verleptpram, Drége. Wortel:

Dec., Pearson, 3631; without precise locality: Marloth, 7809. Kenhardt: Upington, Moss,

10730; Smith, 2369. Gordonia: without precise locality, Pole Evans, 2180. Prieska :

Prieska, March, Bryant, J.39. Barkly West: Danielskuil, Lawson, in McGregor Mus.,

2121.

TRANSVAAL PRoyincE.—Zoutpansberg: Messina, nr. the town, Young, in Hb. Moss.,

14675.

From the large range of material studied it appears that this species is an annual or

at the very most a biennial. The indumentum of the vegetative parts is more pronounced

than in other species, especially with regard to the bulbous-based hairs, but this character

is inadequate to identify the species in the absence of fruits.

The size of the flowers is extremely variable. Engler 6579 has flowers with petals

about 8 mm. long whereas Pearson 4693 has such of over 15 mm. in length; in fact they

may reach a length up to 20 mm. The colour, however, appears to be consistently a pure

though a somewhat pale yellow.

In fruiting characters the species, however, is extremely variable, especially with

regard to the shape, size and texture, and degree of indumentum of the wings of the indi-

vidual cocci. The colour of the wings appears to be brown, whereas in 7. cristatus Pres]

they are apparently always pale olive-green in colour. The latter character is remarked

upon by Dinter, who also states that he has never observed a transition from the one to the

other. In sheets such as Dinter 1187 and Bryant J. 39 the extreme variation in the size

and shape of the wings may be clearly seen. The texture of the wings is also of a very

variable nature; it may be papery and brittle showing all intermediate stages to rigidly

coriaceous and tough. This great variation undoubtedly led Engler (1915) l.c. to figure

and describe (inadequately?) T. secwridocarpus which he considered to be specifically

distinct from our species in question. Under his species he also distinguished forma vulgaris

and var. subtruncatus. Hxamination of the material of this species and its forms deposited

in Herb. Mus. Bot Bérol. has convinced me that neither the forms nor this species created

by Engler can be upheld, as they fall within the limits of variation of 7. pterophorus. A

further argument in favour of this view is the following: the type number of T. pterophorus

Presl in Herb. Mus. Brit. bears fruits which appear to be almost mature and comparison

of these with such of typical 7. securidocarpus Engl. have led me to believe that both these

species are conspecific. H. Meyer when working through Drége’s gatherings of Tribulus

labelled certain sheets Tribulus alatus Delile? thus (inadvertently ? ) drawing attention

to the similarity between the Cape plant and the true but distinct Indian and north-east

African Tribulus alatus Del. As Presl’s type of T. pterophorus cannot be traced in the

Herb. Un. Germ. Prag. we have no evidence that this species is distinct from T. securido-

carpus Engl., but the fact that the type numbers of T. pterophorus in both Herb. Kew and

Herb. Mus. Brit. agree perfectly with the latter species and also with Engler’s figure of

T. securidocarpus is sufficient evidence to regard them as being conspecific.

T. albescens Schlechter ex Dinter l.c. is only more densely hispid than typical 7. ptero-

phorus Presl. It agrees in fruiting characters with, and therefore is also conspecific with

the latter.

T. alatus Del., T. macropterus Boiss. and T. pterocarpus Ehrenb. are the nearest allies

of T. pterophorus. T. alatus Del. and T. macropterus, however, both have smaller flowers,

intrastaminal glands which are not joined to form a shallow cup at the base of the ovary

and fruits with smaller and shorter wings. T. pterocarpus Ehrenb. has extremely small

flowers with petals about 5 mm. long and the fruits, including wings, do not exceed 10 mm.

in both length and diameter, characters by which it is readily distinguishable from T.

pterophorus.

168

Dinter (1927) lc. states that 7. pterophorus and T. securidocarpus are undoubtedly

closely related species, but never show transitions between each other. I have had the

privilege of studying a wide range of material, my observations have led me to the following

conclusions. The extreme forms included under the two species can definitely be dis-

tinguished but since every degree of variation and in many cases great variation in the

shapes of the fruits on one and the same plant can be observed, the question arises whether

we are dealing with two distinct species and their numerous intermediates due to hybri-

disation. It is possible that 7’. securidocarpus may prove to be a hybrid of the parentage

T. pterophorus on the one hand and T. Zeyheri or an allied species on the other, but this

can only be proved by breeding experiments. The figure in Engler l.c. of T. pterophorus

does not agree in shape with the fruits of the type numbers I have studied; should 7.

pterophorus Engl. (non Presl.) eventually be recognised as a distinct species, it will have

to receive another name.

In order to decide with absolute certainty whether 7. pterophorus Presl is an extremely

variable species or includes more than one closely allied species with tendencies towards

hybridisation, careful breeding experiments will be necessary. As I have had neither the

opportunity of studying these plants in the field nor undertaking breeding experiments,

my deductions are based on the study of herbarium material, and such conclusions I have

reached and the views which are bemg forwarded here may later prove to be quite erroneous.

The distribution of the species is a fairly limited one. It is fairly frequent in the sandy

parts of the Mandate of South West Africa and its southernmost limit appears to be ap-

proximately the Orange River basin. It has not yet been recorded from as far north as

Angola.

A note on Blank s.n. (leg. anno 1918) in Herb. Mus. Bot. Berol. says “ Aufschlag, das

erste nach dem Sommerregen spriessende Griin, wachst iiberall und wird, besonders im

jungen Zustande, von allem Vich gern gefressen. Wenn die Samenkapseln gelb werden,

sollen sie Schuld en der ‘ Geelsiekte ’ der Schafe sem. Das Fett der an dieser Krankheit

verendeten Schafe ist durchweg gelb . . .”’, an indication that this species is suspected

of causing “ geeldikkop ”’? in sheep.

4. T. Zeyheri Sond., in Fl. Cap. I. 353 (1859-1860) ; Dinter, Deutsch-Siidwest-Afrika. 85

(1909) ; Heering & Grimme, Unters. Weideverh. Deutsch-Siidwestafr. 27 (1911) ; Glover,

in Ann. 8. Afr. Mus. IX. in. 170 (1913); Engler, Veg. der Erde 9. III. i. 736 (1915), cum

fig. 343Y ; Burtt Davy, Flow. Pl. & Ferns I. 187 (1926); Dinter, in Fedde Rep. XXIV.

15 (1927) ; Engler, Pflzfam. ed. IT. 19a. 176 (1931), cum fig. 84 Y ; Bremekamp, in Karsten

u. Walter: Vegetationsbilder, XXII. 3. 3 (1932), cum fig. 13; Range, in Fedde Rep.

XXXVI. 250 (1934).

Syn.: 7. Zeyheri Sond., var. hirtus Schinz, in Verh. Bot. Ver. Brandenb. XXIX,

54 (1887).

T. Zeyheri Sond., var hirsutissimus Schinz, 1.c.

T. terrestris L., var. Zeyherv Schinz, in Bull. Herb. Boiss. II, 187 (1894).

T. Zeyhert Sond., var. awrantiacus Dinter, in Fedde Rep. XXIV, 15 (1927).

T. murex Schlechter ex Dinter, in Fedde Rep. XXIV, 14 (1927), nomen

subnudum, non Presl, pro parte; Range, in Fedde Rep. XXXVI, 249

(1934).

A prostrate perennial. Branches prostrate or at length somewhat ascending, radiating

from the much branched crown of the rootstock, up to 1 metre long (sometimes even ex-

ceeding this length but usually very much shorter) and branched again, more or less hirsute

in all vegetative parts with a fine indumentum, with scattered bristle-lke bulbous-based

hairs; internodes very variable in length, depending on the robustness of the plant, up

to 9 cm. long, usually much shorter, striate, terete. Leaves unequal; the larger up to

9 em. long, up to 9-jugate; the smaller up to 5 cm. long, up to 4-jugate ; stipules up to

169

10 mm. long, usually much shorter, narrowly lnear-lanceolate to obliquely ovate, acute,

ciliate with tubercle-based hairs, more or less pubescent on both surfaces; petiole not

winged ; leaflets obliquely oblong, acute or ovate acute to slightly obovate abruptly acute,

very variable in size and shape, from 4~20 mm. long and 2-1] mm. broad, more or less

densely pubescent on both surfaces sometimes almost glabrous on the upper surface and

very often conspicuously ciliate with bulbous-based hairs. Pedicel 1} to 2 times as long

as the subtending leaf. Flower-buds ovate, obtuse or acuminate, up to 8 mm. long.

Sepals narrowly linear-lanceolate, acute, up to 12 mm. long and 2 mm. broad, unusually

densely pubescent without. Petals broadly cuneate, up to 25 mm. long, 1-7 to 2-5 times

the length of the sepals. Filaments up to 3-5 mm. long; anthers up to 3 mm. long.

Style usually fairly short ; stigma slender, pyramidal, about 2 mm. long, much exceeding

the style in length. Intrastaminal glands united to form a shallow cup at the base of the

ovary. Carpels and young fruit minutely pubescent and hirsute with bristle-like hairs.

Mature fruits armed or almost devoid of spines, extremely variable in size and shape, at

length breaking up into 5 cocci; cocci usually armed with 4 (—6) well developed spines,

or spines very much reduced as to give the coccus almost a warted appearance, tubercled

on the dorsal crest and very often laterally compressed (Fig. VI).

DaMARALAND.—Grootfontein: Jan., Lightfoot, 63; Febr., Sener, 674; Jan., Schoen-

felder, 488. At Gaub: Borle, 50. Okahandji: Dinter, 143, 303; Bradfield, 412; Hopfner,

53, 52. Swakopmund: Bradfield, 581; Liideritz, 148. bei Ukib: Dinter, 60; Pogge,

15. Onguati: Engler, 6191. Salem: Dznter, 110. Karibib, Hartman, 155a, 155¢ pro

parte; Micke, 7. Auasberge: Dinter, 1888. Windhuk: Rogers, 29766. Rehoboth :

Fleck, 150, 592. Kuiseb-bed: Fleck, 776.

Great NAMAQUALAND.—Gibeon : Pearson, 9212. Sandverhaar: Range, 942. Kuibis :

Range, 897. Aus: Schafer, 156; Schinz, 1121. Huibplateau: Schenck, 207, 211.

Holoog: Pearson, 4120. Aias: Pearson, 8039. Klein Karas : Ortendahl, 250. Choaberib :

Pearson, 9461.

Care Province.—Namaqualand: Rietfontein, Pearson, 3434. Kamabies: Pearson,

3780. Springbok: Godman, 689; Salter, 4577. Calvinia: Springbokkuil, Zeyher, 272

(type in Herb. Mus. Bot. Stockholm). Calvinia: Marloth, 10487. Gordonia: Upington,

Wagner, sn. Askkam: Lang, sn. Laigsburg: Matjesfontein, Foley, 192. Graaff

Reinet : Kruidfontemhoogte, Bolus, 836. Prieska: Bryant, J.21, J.39, J.19. Hopetown :

Orange River nr. Hopetown, Bolus, 1836; Rehmann, 3336. Herbert: St. Clair, Douglas,

Orpen, 124. Kimberley: Witpen, Pocock, sn. Barkly West: Benim, 607. Groot

Boetsap: Marloth, 1133. Likat: Wuilman, sn. Winters Rust: Wilman, sn. Kuru-

man: 50 miles from Kuruman, Lang, sn. Batlharos: Silk, 15.

ORANGE FREE SratTeE.—Fauresmith: Henrici, 2557.

TRANSVAAL PRovince.—Zoutpansberg: Blaauwberg: Bremekamp and Schweikerdt,

120. Mapagoni: Breyer, in Hb. Transv. Mus., 16044. Messina: Rogers, 19373, 19401,

18422; Scholtz, 1. Zoutpan: Obermeyer, Schweickerdt and Verdoorn, 263, 299. Water-

poort : Obermeyer, Schweickerdt and Verdoorn, 325. Lydenburg: Sekukuni, Barnard, 186.

Sonder in Fl. Cap. 1. |.c. based his description of the above species on one gathering

only; ie. Zeyher 272 from Springbokkuil, Little Namaqualand. Examination of the

type numbers of this species in various herbaria suggested that it is an annual, the crown

of the rootstock not being strong enough to give one the impression of a perennial species.

Sheets such as Pearson 3780, 3015, 9212 and 3434, and Marloth 1133 which are good matches

with the type, however, have rootstocks in which the crown exhibits the presence of remains

of dry shoots from an earlier season. Observations of this species in the field in the sandy

areas of the northern Transvaal (during the month of November) showed the presence of

persistent rootstocks (with the remains of withered and dried branches) giving rise to

young flowering shoots. This species is thus undoubtedly perennial but very probably

may reach the flowering stage within a year.

170

Sonder’s type is undoubtedly a specimen which grew under unfavourable circumstances.

In endeavouring to find a match with this type among more recently gathered material

it was found that stunted and dwarfed plants such as Pearson 3780 and Marloth 1133

approached the type most closely, whereas more robust gatherings such as Dinter 303

and Pole Evans 1, at first sight, did not appear to belong to this species. Furthermore,

luxuriant specimens such as Schweickerdt and Verdoorn 650, and Obermeyer, Schweickerdt

and Verdoorn 299 at first sight appeared wholly out of place in this species. Careful exami-

nation of vegetative, floral and fruit characters, however, have subsequently shown, that

all these exhibit such a wide range of variation that a sub-division based on these characters

would amount to the description of individual plants as species. It therefore became

obvious that owing to the great variation in the specimens examined, the adoption of a

broad view of the species seemed the only one possible.

From the appearance of the specimens examined, the notes made by different col-

lectors and from my own observations of plants in the field, it appears that the habit of

this species is a fairly constant one: the primary branches arising from the crown of the

rootstock are prostrate and towards their extremity may be ascending, but the species never

tends to become shrubby as in 7. excrucians Wawra. It is by this character (supported by

others less conspicuous) that these two species may be distinguished.

A study of the characters of the vegetative and reproductive organs, i.e. length of

internodes, size of leaflets, degree of indumentum, relative and actual size of calyx and

corolla, nature of the fruit, etc., showed T. Zeyheri to be an extremely variable species,

the limits of which are not at all clearly defined. In fact it is only with great difficulty

and uncertainty that it may be distinguished from 7. cistoides L. Sonder l|.c. states that

T. cistoides is ““ much, more robust and has larger leaves, flowers and fruits, besides a style

2 lines long and a short terminal stigma, by which character as already stated by Schlechtendal

Bot. Zeit. 1851, p. 844, it is known from other Tribuli.” This statement holds good for

Sonder’s type and a few modern gatherings of stunted plants which are in no way truly

representative of the species. As soon as a broader view of the species is adopted it is a

matter of difficulty to distinguish our South African species from the American T. cistoides

L. Hxamination of material of the latter from America, Cape Verde Islands and Tropical

Africa, etc., has shown that in many cases the style is extremely well-developed and the

stigma very much reduced; in other cases, however, a pyramidal well-developed stigma

and consequent reduction in the length of the style is exhibited by the American plants.

The character of relative size and length of style and stigma thus does not hold good as

a means of distinction between T. Zeyheri and T. cistoides. The shape of the leaflets,

however, appears to be more satisfactory, viz. in T. cistoides they tend to be oblong to

obovate-oblong apiculate, whereas in 7. Zeyheri they tend to be more or less oblong or

ovate-oblong. The fruits of these two species do not exhibit any constant differences.

From the foregoing observations it is assumed that these two species are closely related.

Whereas extreme forms of these species may be readily distinguished, occasions arise when

one is rather somewhat in doubt as to the identity of a plant expecially when it has come

from an area where both species overlap, i.e. some parts of Tropical Africa.

To take a restricted view of the above species would not further the position in any

way. It would merely mean that certain individuals would have to remain unclassified.

To regard T. Zeyher: and T. cistoides as being conspecific would amount to extending the

limits of the species too considerably. Oliver in Fl. Trop. Afr. I. 283 (1868) considers

T. cistoides Li. to be a variety of T. terrestris L. This is very unlikely to be true for the

following reasons. The flowers of 7. terrestris L. (from Southern Europe) possess intras-

taminal glands which are free and not joined to form a definite cup around the base of

the ovary, whereas T. cistoides L. shows the presence of this cup. This characteristic

appears to be of great taxonomic inportance as in the many sheets examined I have always

been able to distinguish 7. terrestris L. (including 7. parvispinus Presl and 7. murex Presl)

from the other South African members of this genus.

ilviol

Schinz l|.c. after having studied a large number of specimens from the Mandate of

South West Africa, arrives at the conclusion that T. Zeyheri is to be regarded as a variety

of T. terrestris L. I cannot agree to this view on the grounds that 7. Zeyheri has inter-

grown intrastaminal glands whereas in 7’. terrestris these are free. Schinz furthermore

states that he had not yet seen 7. cistoides from Africa, but regards all plants so-named

from that continent to be T. Zeyhert. In Herb. Kew., however, several sheets from Tropical

Africa undoubtedly belong to TZ. cistordes L., in fact they resemble the typical South

American form of that species.

Dinter, in Deutsch-Siidwest-Afrika |.c. states that T. Zeyheri differs from T. terrestris

only in the very large flowers and which are extremely variable in colour. This fine collector

evidently overlooked the nature of the intrastaminal glands.

Engler, Veg. der Erde |.c. considers T. cistoides, T. terrestris and T. Zeyheri to be

distinct and from a phylogenetic point of view equivalent species. He distinguishes 7.

Zeyheri from T. cistoides in the following characters: T. cistoides has smaller flowers,

broader sepals and larger fruits than our plant. These characters appear to be of little

or no taxonomic value since I have examined many sheets of 7. Zeyher: with flowers

smaller and fruits larger than those of the American species.

Dinter, in Fedde Rep. XXIV l.c. again stresses the variation in colour of the fiowers

which can be either uniformly yellow, cream with a saffron claw or more rarely, uniformly

cream, and in the region of the Aviser Pforte nr. Windhoek very frequently orange-yellow

to orange-red. On the basis of the latter colour he distinguishes var. aurantiacus Dtr. L.c.

from the typical plant. I very much doubt whether this represents a distinct variety and

for the present am inclined to consider it synonymous with 7. Zeyheri Sond.

Schinz’s varieties hirtus and hirsutissimus of T. Zeyheri are in my opinion only extremely

hairy individuals of the typical plant. I do not consider degree of indumentum to be of

any taxonomic importance.

T. murex Schlechter l.c. definitely falls within the range of T. Zeyhert. Dinter 110

which bears “typ. auct.” in Schlechter’s hand has extremely spiny fruits and it was no

doubt on the basis of this character that Schlechter had intended to separate it from 7’.

Zeyheri.

The fruits of 7. Zeyheri appear to vary markedly in their degree of spinosity. Brad-

field 581 and Dinter 303, both from South West Africa, exhibit fruits which are extremely

spiny and in this respect are not unlike immature fruits of the Australian 7. hystrix R. Br.

The fruits of Bryant J. 21, Miicke 52, Schoenfelder 488 and Steyn 22566 again present the

other extreme in which the cocci are laterally much compressed and warted on the dorsal

crest ; the spines are short and rigid. In fact the fruits of the specimens just mentioned

appear so typical, that may be on the basis of this character it would be possible to regard

them as a species distinct from 7. Zeyheri. It will be necessary, however, to grow plants

from such seed before any conclusions are arrived at. Another most interesting gathering

is that of Bryant J. 39 (see also under T. pterophorus) which has fruits partly devoid of spines

and in which the cocci are sharply tubercled as well as having two slender downward turned

spines arising from near the base of each coccus. ‘These fruits are not unlike the figure

representing 7. Zeyheri in Engl. l.c. (True 7. Zeyheri Sond. possesses fruits in which each

coccus is armed with four well-developed spines.) ;

Until an intensive study of the various variations outlined above has been made in

the field in conjunction with breeding experiments, it will be impossible to say whether

several varieties or even species are involved under the present concept of 7’. Zeyheri Sond.,

or whether this species is only one extremely variable species. The distribution is a much

wider one than that of any of the large-flowered South African species. It is found mainly

in the sandy arid and sub-arid regions of Southern Africa and occurs further north through

South Tropical Africa, finally overlapping with 7. cistoides L. in the Tropics proper.

172

5. T. terrestris Linn., Sp. Pl. 387 (1753); Thunb. Prodr. 79 (1794); Thunb. Fl. Cap. ed.

Schult. 543 (1823) ; DC. Prodr. I. 703 (1824) ; Eckl. & Zeyh. Enum. Pl. 95 (1835) ; Harvey,

Gen. 8. Afr. Pl. 46 (1838) ; Drege Zwei Pflzgeogr. -Doc. 58, 73, 131 (1843) ; Krauss in Flora

1844, p. 301; Sonder, Fl. Cap. I. 352 (1859-1860); Harvey, Gen. 8. Afr. Pl. ed. II. 36

(1868); Engl. et Gilg., in Warburg Kunene-Samb.-Exped. Baum: 269 (1903); Dinter,

Deutsch-Siidwest-Afr. 85 (1909); N. E. Brown in Kew Bull. 1909, p. 97; Heermg &

Grimme, Untersuch. Weideverh. Deutsch-Stidwestafr. 26, 74 (1911); Glover, in Ann. S.

Afr. Mus. IX, i, 170 (1913); Engler, Veget. der Erde 9. III. 1. 736 (1915), cum fig. 343,

E-L; Juel, Plant. Thunberg. 309 (1918); Burtt Davy, Flow. Pl. & Ferns I. 187 (1926) ;

Dinter, in Fedde Rep. XXIV. 15 (1927); Engler, Pflzfam. ed. 2. 19a. 176 (1931), cum

fig. 84 E-L; Range, in Fedde Rep. XXXVI. 250 (1934).

Syn.: ZT. terrestris B. desertorum Eckl. and Zeyher, Enum. Plant., 95 (1835).

T. hispidus Presl, Bot. Bemerk., 29 (1844).

T. murex Pres], l.c.

T. parvispinus Presl, l.c.

T. terrestris L., var. 6. desertorum Sond., Fl. Cap. I, 353 (1859-60).

T. terrestris L., var. B. hispidissimus Sond., Fl. Cap. I, 353 (1859-60) ; Burtt

Davy, Flow. Pl. and Ferns I, 187 (1926).

T. parviflorus Schlechter ex Engler, Pflzfam. ed. II, 19a, 176 (1931), nomen

tantum.

T. murex Schlechter ex Dinter, in Fedde Rep. XXIV, 14 (1927) pro parte,

nomen subnudum, non Pres]; Range, in Fedde Rep. XXXVI, 249 (1934).

A spreading prostrate usually decumbent annual. Branches radiating from the much-

branched crown of the rootstock, up to 1-5 metres long and usually branched again, in

all vegetative parts pubescent, villous or hispid or glabrescent, extremely variable as to

degree of hairiness; internodes up to 6 cm. long, usually much shorter, terete, striate.

Leaves unequal ; the larger up to 6 cm. long, usually somewhat smaller, with up to 8 pairs

of leaflets ; the smaller up to 3-5 cm. long, usually much smaller, with up to 6 pairs of

leaflets; leaflets obliquely oblong-lanceolate, or lanceolate-ovate, acute or subobtuse,

villous on both surfaces, often more or less glabrescent above, up to 15 mm. long and

5 mm. broad, usually much smaller ; stipules lear or linear-lanceolate, acute, up to 10 mm.

long, often much shorter. Pedicel shorter, as long as or more rarely slightly longer than

the subtending leaf. Flower buds ovate-acuminate or acute. Sepals 3-6 mm. long, linear-

lanceolate, acute, villous without. Petals broadly cuneate, clear yellow, 3-12 mm. long,

shorter than or up to 2-3 times the length of the sepals. Filaments 3 mm. long or some-

what longer. Style short, much reduced ; stigma hemispherical and almost sessile on the

ovary ; ovary hirsute with bristly bulbous-based hairs. Intrastaminal glands not inter-

grown to form a cup around the base of the ovary. Matwre fruit pubescent or almost

glabrous, at length breaking up into 5 cocci; cocci usually with 2 lateral divergent acute

spines inserted above the middle, and two shorter spines directed downwards and inserted

near the base of the coccus; dorsal crest tubercled and usually set with bristly hairs ; the

size and degree of spinosity of the cocci is extremely variable (Fig. VII).

AncoLa.—Mossamedes: Hopfner, sn. On the Kubango at Kalolo: Baum, 441.

AmBoLAND.—Ondonga : Liljeblad, 188; Rautanen, sn. Olukonda: Schinz, 1022,

1023, 1025.

DaMARALAND.—Okahandja: Dinter, 143; Seiner, 150. Windhuk—Walfishbay :

Rogers, 15179. Salem: Dinter, 102. Windhuk: Tvotha, 81a. Karabib: Hartman,

155b, 155¢ pro parte, 155d. Lichtenstein: Dinter, 4489. Rehoboth: Fleck, 4, 65.

Gr. NamagiiaLanpD.—Kubub : Range, 226; Schinz, 1024. Schakalskuppe: Pearson,

4784. Sandverhbaar: Pearson, 4675. Seeheim: Pearson, 3723. Naruda-siid: Pearson,

7862, 8218.

173

. Cape Province.—Little Namaqualand: Steinkopf, Schlechter, sn. Stinkfontein :

Pearson, 5521. Calvinia: Springbokkuil, Zeyher, 273. Clanwilliam: Bachmann, 332.

Nr. Wupperthal: Drege. Cape: Claremont, Schlechter, 553. Nr. Salt River stn.: Salter,

240/11. Nr. Lakeside stn.: Andreae, 229. Robertson: Montagu, Moss, 5593. Rivers-

dale: Gauritz River: Ecklon and Zeyher, 751B. Mossel Bay: Moss, 5593. Knysna:

Burchell, 1841; Fourcade, 1993. Uitenhage: Hcklon and Zeyher, 751. Bathurst: Fish

River, Burke. Komgha: Flanagan, 82. Willowvale: Bashee, river mouth and in gardens,

Drége. Queenstown: Shiloh, Baur, 973. Graaff Remet: Bolus, 261. Prieska: Bryant,

J.39 pro parte. Hay: Asbestos Hills, Wilman, 3041. Herbert: Belmont, Wilman,

Orpen, 125, 126. Kimberley: Wzlman, 2124. Riverton: Wialman, 3041.

Nata Provrnce.—Ladysmith : Wagon Hill, Wood, 18781. Zululand: Gerrard, 214.

ORANGE FREE STaTE PRovince.—Fauresmith : Henrici, 1873, 2461, 2462, 2532-2534,

2559-2561, 2676, 2678, 2704-2706, 2708, 2710-2712, 2718-2719; Verdoorn, 1561-1563 ;

Goossens, 654; Steyn, 22565. Bloemfontein : Bestersput, Welti, 22. Modderrivierdrift :

Rehmann, 3586. Kroonstad: Bothville, Schweickerdt, 1075, 1080. Vredefort : Obermeyer,

in Tol. Mus., 31665.

TRANSVAAL Province.—Potchefstroom: Burtt Davy, 854. Vereeniging, Leendertz,

3862. Witwatersrand: Johannesburg, Moss, 7469, 9557, 13779, 16146, 16171. Pretoria :

Wonderboompoort, Smith, 6192 ; Schweickerdt, 1049, 1050. Rooikop : Smuts and Gillett,

2029. Meintjes Kop: Mogg, 12307 : Schweickerdt, 1202. Rustenburg: Nation, 42

Watt and Brandwyk, 1813. Waterberg: Nylstroom, Mogg, 12535. Zoutpansberg :

Messina, Turner, 18.

BasutroLanp.—Leribe: Dieterlen, 127. White Hill: Jacottet, 236.

BECHUANALAND Prorectoratre.—Mafeking: Bolus, 6402. Mochudi: Rogers, 6443.

Kwebe (Ngamiland): Lwgard, 105, 117, 123.

PortuGguEseE Hast Arrica.—Ressano Garcia: Schlechter, 11896. Rikatla: Junod,

440. Lourenco Marques: Moss, 11806, 6947 ; Thoday, 176; Monteiro, 56: Inhambane :

Lawrence, 31.

For reasons of space it has been impossible to cite all the specimens seen. Consequently

gatherings for citation were selected mainly with a view to indicate the wide distribution

of this species in South Africa. As a resuit of studying a very wide range of material from

almost all parts of Southern Africa, the writer has come to the conclusion that among the

various species of Tribulus found in the area under consideration, 7’. terrestris Linn. show

the widest range of variation both in vegetative and reproductive characters.

T. murex Presl, T. parvispinus Presi and T. hispidus Presl based on scanty material

are very probably only forms of the cosmopolitan 7. terrestris L.

T. parvispinus Presl has petals and sepals 2-5 mm. long and in many cases the sepals

exceed the petals in length. The ratio of length of petals to sepals may thus be represented

by the factor 1-0 or a value less than 1-0.

In 7. murex Presl the petals vary between 5-12 mm. and sepals 4-6 mm. in length.

The ratio of the length of petals to sepals here has a value of | -25-2-3 and appears to be

always greater than 1-0. The flowers appear to be somewhat larger than those of 7. parvi-

spinus, but specimens have been examined in which the flowers approached in size those

of the species just mentioned.

In the type and type numbers of 7. hispidus Pres] the petals have been eaten by

insects (probably already at the time of gathering) but judging from the remains they

appear to exceed the sepals im length. The very hispid nature of the plant to my mind

is merely due to habitat, it does not appear to be of any taxonomic value, since it is not

correlated with other distinctive characters.

174

An attempt to dintinguish Presl’s species from each other on basis of the relative

ratios of length of petals to sepals has proved unsuccessful especially as the Linnean type

of 7. terrestris in some respects appears to be intermediate between these species.

The Linnean type of 7. terrestris is a complete plant having about 5 branches each

of which is more or less 8 cm. long. The petals are about 3 mm. long and the sepals being

hidden by the former are somewhat shorter than the petals. The leaves are 4-6 jugate,

and the pedicels are as long as or slightly shorter than the subtending leaves; the leaflets

are up to 7 mm. long and 3 mm. broad. The young fruits are minutely puberulous (apart

from the bristly hairs on the dorsal crests) and each coccus is armed with two pairs of lateral

spines of which one pair is shorter than the other.

[As a note of interest the following may be mentioned: Linnaeus wrote up his type

(sheet 4 in the cover of Tribulus) as Tribulus Y which actually means 7. aquaticus. In

the manuscript of the Sp. Pl, however, he correctly wrote Tribulus + the latter sign being

used by him to denote “earth,” i.e. terrestris. By mistake the sheet was thus actually

written up wrongly, but this has no bearing on Tribulus aquaticus C. Bauh. which is quite

a different plant, namely Tvapa natans (see Richter’s Index)].

Taking various standard works on the Floras of Central Europe, North America and

Australia into account, one finds that the limits by which these various authors define

T. terrestris LL. show wide discrepancy. Some define the petals as being “3-4 mm. long,”

some state “6 mm. long’” and others again ““ up to 10 mm. long,” the length of the sepais

also exhibits this wide range of variation. There is thus very little doubt with regard to

the extreme variability of this species.

It has been pointed out that the Linnean type has extremely small flowers. What

Presl has described under the name 7. parvispinus is to my mind typical T. terrestris L.

The fruit characters on which he distinguishes his species from the Linnean plant do not

hold good. A wide range of material has shown the size, etc. of the fruits to be extremely

variable. For analogous reasons 7. murex Presl and T. hispidus Presl are considered to

be merely large-flowered and more robust forms of 7. terrestris L.

Dinter 102 quoted in Fedde Rep. XXIV. 14 (1928) under 7. muwrex Schlechter is nothing

else but 7’. terrestris L. Furthermore T. parviflorus Schlechter ex Hngler L.c. is T. terrestris

Miss I. C. Verdoorn of the Division of Plant Industry, Pretoria, who has had occasion

to study plants in the field and especially on the lands of the Grootfontien School of Agri-

culture, Middelburg, Cape Province, recognised three closely allied forms of Tribulus,

(a) small plants with erect dried-up looking (greyish) inward-curling branches, (b) plants

with longer branches which were inclined to be prostrate, (c) luxuriant plants with closely

prostrate branches. Miss Verdoorn kindly prepared herbarium specimens of these forms

all of which have since been identified by the writer as being 7. terrestris Linn. These

forms could not be regarded as varietally different on basis of their habitat, since the latter

was not correlated with any other morphological difference. The small erect “ form ”

was suspected of having caused “ dikkop ” in sheep.

Dr. M. Henrici, who has grown plants at the Veld Reserve, ‘Dammam, O.F.S., states :

“At least four species seem io exist according to habit of growth and flowers: (1) a widely

spreading plant with a large yellow dark-centered flower, (2) a running variety with a large

evenly coloured yellow flower smaller than that of (1), (3) medium sized yellow flower with

calyx about half as long as petals, both running and upright forms ; when the species runs,

calyx may be nearly as long as the petals, (4) small yellow flower with calyx as long as or

longer than the petals, running and upright forms. The two latter species have forms

not only differing in their habit of growth but also in their content of green pigment. While

some of the forms are. excellent fattening fodder plants, those which grow upright and

contain less pigment cause dikkop at times, especially in the wilted state and on certain

soils (limestone and river bank soils). It seems that the species with large flowers and large

amounts of pigment and with long runners are never poisonous.”

175

The plants mentioned under (1) and (2) are definitely not 7. terrestris L. but very

probably 7. Zeyheri Sond. or one of the species with winged fruits. Since the nature of

the fruits is not mentioned it is impossible to say which of these species are involved. The

plants under (3) are probably luxuriant forms of T. terrestris L. (= T. murex Presl) and

those under (4) almost certainly the typical small-flowered Linnean species. It may be

added that according to a note by Blank on a sheet of T. pterophorus Presl in Herb. Mus.

Bot. Berol., this large-flowered prostrate species has been found to cause “ geelsiekte ” in

sheep (see notes under 7’. pterophorus). Thus not only the small-flowered species are to

be suspected of being poisonous to sheep.

T. terrestris L. makes its appearance soon after the first summer rains have fallen and

often forms extensive dense carpet-like growths near the habitations of man.

Owing to the ruderal and semi-ruderal nature and the wide (almost cosmopolitan)

distribution of this Linnean species it is readily conceivable that under different edaphic

and climatic conditions the plants may present a very variable appearance. Careful

breeding experiments would do doubt throw light on the question as to whether 7. terres-

tris LL. as conceived in this paper is only one extremely variable species or whether at present

several closely allied species are included under this name.

CONCLUDING REMARKS.

That distinguished systematists such as Engler, Schinz and Dinter had failed in defin-

ing the limits of the various species clearly, in spite of the fact that each of these authors

had travelled and collected extensively in Southern African regions where Tribulus abounds,

may serve to indicate how ill the members of this genus are defined.

A successful and clear delineation of the Southern African species may be possible

by taking recourse to special field studies including the growing of plants from seed and

the conduction of breeding experiments. A clear conception of the range of variation

exhibited by any one species may be obtained in this way.

It is furthermore a sine qua non that thorough gatherings be made of such mother

plants in both flowering and fruiting condition of which the seed is intended to be grown.

Such mother plants must be retained for future reference and as a standard for comparison

with the cultivated daughter plants. The mere growing of plants from seed without

having preserved specimens of the mother plants from which the seed had been derived is

not of much use, since it is absolutely essential that the nature of the original plant in its

wild state be known.

The most satisfactory and profitable method of procedure appears to be the following :

As many different forms of species such as for example 7. pterophorus, T. Zeyheri and

T. terrestris, including specimens of these species from their type localities (where such are

known) and which have been found to resemble the type, should be gathered and their

seeds should be grown under identical as well as different conditions (climatic and edaphic).

Gatherings should again be made of both flowering and fruiting material of these daughter

plants. By subsequent comparison of these individuals possibly an insight may be gained

into what at present is considered to be an individual but very variable species.

ACKNOWLEDGEMENTS.

It has been my privilege to examine the material in all the South African Herbaria

as well as the Southern African material in the following Kuropean Herbaria: Hort. bot.

Reg. Kew., Mus. Brit., Mus. bot. Berol., Mus. bot. Stockholm, Mus. bot. Univ. Germ. Prag.,

Mus. bot. Univ. Ziirich and Hofmuseum, Wien. I wish to tender my sincere thanks to the

Directors and Curators (Keepers) of these various institutions for allowing me to consult

this material. My special thanks are due to Sir Arthur Hill, Director of the Royal Botanic

Gardens, Kew, for the great facilities offered during the preparation of this paper.

178

EXPLANATION OF FIGURES.

I.—Ovary of T. terrestris L. (x 10).

I1.—Ovary of 7. Zeyhert Sond. (x 10).

III.—_Coeci of 7. excrucians Wawra (x 3).

IV.—Cocei of 7. eristatus Presl (x 2).

V.—Cocei of TF. pierophorus Presl (x 2).

VI.—Coeci of 7’. Zeyherr Sond (x 3).

VII.—Cocci of T. terrestris L. (x 3).

179

A NOTE ON THE SOUTH AFRICAN SPECIES

OF XIMENIA LINN. AND THEIR POSSIBLE

ECONOMIC USES.

By H. G. Scuwerickerpt, B.Sc., Ph.D., F.L.S.

Durtne the past seventeen years spasmodic interest has arisen in connection with the

possible use of the fruits of species of Ximenia as a source of oil. In 1917 the Imperial

Institute reported on a sample of fruits said to be those of X. americana Linn. Last year

Mr. A. G. S. du Toit, the Extension Officer at Ixopo, Natal, sent in specimens of a Ximenia

(National Herbarium No. 16694) accompanied by a large sample of fruits, with a request

for any information as to their economic value. Mr. du Toit in his letter stated that the

plant grows on very poor dry land—practically useless for any other plant of value and

that the fruits could be gathered in large quantities. In South Africa we have two species

of Ximenia which have always been confused and because of the interest taken in the plants,

it was thought desirable to clear up the confusion, as it was important to know which of

the two species were investigated and reported on by the Imperial Institute. The work

was commenced at the National Herbarium, Pretoria and completed at the Herbarium,

Royal Botanic Gardens, Kew.

Oliver in the Fl. Trop. Afr. 1, p. 346 (1868), mentions one species, X. americana L.

and one variety, X. americana var. microphylla Welw. Sonder in the Fl. Cap. 1, p. 234

(1859), likewise mentions only one species X. caffra Sond. and one variety, X. caffra var.

natalensis Sond. The fundamental difference between X. americana L. (and the variety

microphylla Welw.) and X. caffra Sond. (and the variety natalensis Sond.) may be found

in the type of inflorescence.

The following diagramatic drawings may serve to illustrate the morphology of different

types of inflorescences met with in the South African species and varieties of Ximenia L.

For the sake of clarity these diagrams have been somewhat exaggerated especially with

regard to the dimensions of the abbreviated shoots (B) in Figs. I. II. III.

FIG. I. - FIG. II. HIG. Lis

180

In X. americana the inflorescence is always a stalked axillary few- to many-flowered

bracteate cyme. In X. americana var. microphylla the inflorescence is also a few-flowered

stalked and bracteate axillary cyme; occasionally however, the flowers may be solitary,

but then the peduncle is always bracteate, thus suggesting a reduced inflorescence. In

X. caffra and X. caffra var. natalensis, however, each flower arises singly in the axil of

either a scale-like leaf or in the axil of a normally developed leaf; the pedicels are never

bracteate. A number of these flowers usually arise on one and the same much abbreviated

lateral shoot thus forming an axillary fascicle. By the foregoing characters X. caffra (and

its variety) may readily be distinguished from X. americana (and its variety).

A study of herbarium material has shown, that the inflorescence Fig. II (A), or solitary

flowers Fig. I and Fig. III arise in the axils of either (1) normally developed foliage leaves,

or (ii) in the axils of reduced scale-like leaves borne by shoots of limited growth (B). These

dwarf shoots (B) are usually much abbreviated and consequently the facicles of flowers

in X. caffra (and its variety) may readily be mistaken for sessile axillary cymes. The

abbreviated shoots in turn arise in the axils of either (1) normally developed foliage leaves,

or (i1) in the axils of the lower scale-like leaves, on lateral branches of limited growth (C) ;

the latter may either be so much abbreviated as to resemble warty outgrowths or may be

relatively well-developed thorns. The thorns invariably arise in the axils of normally

developed leaves (D). The latter may have fallen by the time the inflorescence develops,

but this is by no means always the case. Furthermore the leaves on the shoots (B in Figs.

I and IT) do not always develop and consequently may be absent. In other cases again

the bracts subtending the flowers (Fig. I) or those subtending the cymes (Fig. II) may be

replaced by normally developed leaves.

Burtt Davy in his Manual of Flowermg Plants and Ferns of the Transvaal with Swazi-

land 11, p. 453 (1932) seems to have overlooked the fact that X. americana L. and X. caffra

Sond. are readily distinguishable by their type of inflorescence. In fact this character is

of fundamental importance in the distinction of species (and varieties) belonging to this

genus.

It is therefore suggested that the following key to the Transvaal species (and varieties)

be substituted for the one on page 453 of Burtt Davy’s manual (L.c.)

A.—Flowers solitary in the axils of normal or scale-like leaves, or in few- to many-

flowered fascicles on abbreviated branches. Flower-stalks never bracteate near

Ghee sama AGE ea os cea cccccicesiouie oceh hee tee ore eens eI er ee B.

Inflorescence a few- to many-flowered stalked axillary bracteate cyme, or

flowers solitary arising in the axils of leaves, but then the flower-stalk always bracte-

ate mear the: middle. i ss: fav. S/o aa are aye crete eR eRe ae Teo ee ro ene I C.

B.—Branchlets and leaves densely tomentose when young; lamina becoming glabrate

above in age (even quite glabrous and shiny) ; petioles, peduncles, calyx and outer

surface of the corolla pubescent, the latter at times almost glabrous. .X. caffra Sond.

Branchlets and leaves glabrous to tomentose when young ; lamina in age above

and beneath glabrous, but never shiny ; petioles pubescent ; peduncles and calyx

scantily pubescent or glabrous; corolla externally glabrous.....................

Ra Meee SMM ciara OPO DIA Oh. U,G.0 Satta Cease Gee! X. caffra var. natalensis Sond.

C.—Leaves glabrous, but not conspicuously glaucous ; lamina 34-74 cm. long, 2-3} cm.

broad ; flowers 0-8-1-25 cm. long; calyx not deeply lobed ; fruits 245-3cm. long,

yellow when» ma turetncsa- cs meteeiicte ak he enn ot ae Earn. X. americana L.

Leaves glabrous, conspicuously glaucous ; lamina up to 3} c.m long and up to

2 cm. broad, but usually much smaller; flowers 5-6 mm. long; calyx persistently

deeply lobes; fruit up to 2 cm. long, yellow when mature; plants usually very

THOM AT. 2 HANS. Sofenentie eis ee eee X. americana var. microphylla Welw. ex Oliv.

181

According to Burtt Davy l.c., X. americana L. occurs in the bushveld and Barberton

areas of the Transvaal. Hxamination of specimens quoted in his manual, has proved these

to belong to X. caffra var. natalensis Sond. As it may easily be seen from the above key

that the latter plant is entirely distinct from X. americana L., Burtt Davy’s synonomy

X. americana L. = (X. caffra var. natalensis Sond.) is not justified ; it is merely the result

of erroneous identification of specimens. Furthermore in the ‘* Notes from the National

Herbarium and Museum, Series No. 4, Journ. Dept. Agr., South Africa, January, 1925”

the plant in question (Nat. Herb. No. 2840) is definitely not X. americana L. but X. caffra

Sond.

All the available material from the Transvaal I have so far had an opportunity to

examine, did not include any belonging to X. americana L. This species in its typical form

apparently does not occur in the Transvaal or even in the Union of South Africa.

With regard to X. Rogersii Burtt Davy: This species is described by Burtt Davy in

his Manual 11, p. xxxv (l.c.). A description of the inflorescence is not given. Hxamina-

tion of the type specimen (Rogers 22569 in Herb. Kew.) showed, that it agrees perfectly

with the type of X. americana var. microphylla Welw. ex Oliver (Welwitsch 1127 in Herb.

Mus. Brit.). Burtt Davy does not, however, quote this variety as a synonym of his species

X. Rogersw. Whether the plants placed in X. Rogersi and therefore also those placed in

the variety of X. americana are sufficiently distinct from X. americana to constitute a

separate species is at present difficult to say. They are undoubtedly closely allied to that

species and the only character by which they may be readily distinguished appears to be

the difference in length of the petals of the flowers. Until more complete material, better

field-notes and a much wider range of material can be studied it is perhaps more satis-

factory to retain X. americana var. microphylla Welw. ex Oliver in preference to X. Rogersit

Burtt Davy. It is also suggested that the enumeration of species and varieties of Ximenia

L. on p. 453-454 of Burtt Davy’s Manual be modified as follows :—

(1) X. caffra Sond.

(2) X. caffra var. natalensis Sond.

(3) X. americana var. microphylla Welw. ex Oliver = (X. Rogersii Burtt Davy).

The fruits sent to the Imperial Institute in 1934 were those of X. caffra var. natalensis

(National Herbarium 16694) and examination of the herbarium material now proves that

the samples of fruits examined by the Imperial Institute in 1917, were not those of X.

americana but of X. caffra (National Herbarium 2840).

The following is an extract frem the report of the Imperial Institute :—

“As previously mentioned it seems unlikely that the oil could be prepared by

pressing the kernels, and solvent extraction would be necessary. ‘The oil obtained in

the present case by extraction with light petroleum resembled the similarly-prepared

oil from the X. americana (i.e. X. caffra) kernels in being viscous and cloudy, and in

containing an appreciable quantity of a rubber-like constituent, the presence of which

would account for the high viscosity of the oil. Such oil could not be used for edible

purposes and would probably prove unattractive for the manufacture of soap in com-

petition with other readily available oils. Its comparatively low iodine value indi-

cates that it would be unsuitable for use in paint and varnish-making. The acetone-

extracted oil, on the other hand, proved to be practically free from the objectionable

tubber-like substance. It might therefore prove more suitable for soapmaking and

possibly, after refining, for edible use. The value of such oil at the present time would,

however, be only about £13 to £14 per ton in the United Kingdom. The residual

meal is rich in proteins, but feeding trials carried out in Germany on several kinds of

animals with the residual meal of X. americana kernels are stated (Der Pflanzer, 1911

7, 486) to have shown that the meal is not well suited for use as a feeding-stuff. The

present meal would probably give similar results, but feeding trials would be necessary

182

to determine this point. ‘In this connection it may be pointed out that the meal left

after extraction with acetone would contain most of the rubber-like constituents of

the kernels, and this might affect its suitability for use as a feeding-stuff. Owing to

the inferior quality of Ximenia kernels in comparison with other oil-seeds and in

view of the current over-production of vegetable oils generally and the consequent

low price of these commodities, it does not appear likely that under existing condi-

tions it would be profitable to exploit Ximenia kernels as a source of oil, except possibly

for local markets. It may be mentioned in connection with any effort which may

be made to utilise Ximenia oil in South Africa, that the kernels can be readily extracted

from the dry fruits by treatment in a Miller’s palm-nut cracking machine, and sub-

sequent separation by means of sieves and an air-blast such as are employed in

machines used in the preparation of palm kernels.”

A NEW SPECIES OF PACHYSTIGMA

HOCHST. FROM TRANSVAAL.

By Pror. Dr. W. Rosyns, Brussels (Belgium).

THE genus Pachystigma has its centre of dispersion in South Africa. Of the ten known

species, only two are to be found in Southern Rhodesia : P. rhodesianum (S. Moore) Robyns,

which is limited to that country and the very poisonous P. pygmaeum (Schlecht) Robyns,

a native of South Africa but extending into Southern Rhodesia.

Several species show that suffrutescent habit which is so commonly met with in South

Africa,* but others are dwarf shrublets or erect and more or less branched shrubs of 2—4 m.

high, as P. Bowkeri Robyns and P. macrocalyx Robyns. The second species is perhaps

the most variable of the whole group in its vegetative characters as a result of response to

local conditions. Amongst rocks, it shows the dwarf squat habit with short internodes

and small subtomentose more or less fulvous leaves (forma rupicola), whereas on the edge

of forests, it develops into a much branched shrub with elongated internodes and large

puberulous green leaves (forma silvicola). The extreme forms of this species look so dif-

ferent that one may easily be inclined to consider them as two distinct species, but the

differences seem to be only of degree, no qualitative characters being available.

The new species from Northern Transvaal described here through the courtesy of

Dr. E. P. Phillips, Principal Botanist, Pretoria, who kindly sent the herbarium specimens

for examination to Brussels, is, according to the collector’s label, a small tree of about

5 m. (15 feet) high. Its affinities are with P. macrocalyx on account of the habit and the

long calyx lobes. It may be incorporated in the key I have published in 19287 as follows :-—

Folia utrinque tomentosa vel rarius dense appresse pilosa; alabastra ut

calycis lobi pubescentia :

Folia anguste elliptica, tantum usque ad 1 cm. lata; cymae distincte

pedunculatae, peduncuo 3-5 mm. attingente :

Frutices nani; folia 3 cm. longa et 1 cm. lata, cyanescentia pilo-

saque; cymae 3-5-florae; calycis lobi lineares, vix tertiam

partem superiorem alabastrorum attingentes................. coerulewm

Arbores parvi, breviter ramosi; folia usque as 1-7 cm. longa et

0-8 cm. lata, cinereo-tomentella ; cymae 3-florae ; calycis lobi

anguste elliptici et alabastra plus minusve aequantes......... triflorum

Folia rotundata, ovato-rotundata vel elliptico-rotundata, semper

ultra 1 cm. lata; flores subfasciculati vel rarius ad cymas brevi-

pendunculatas referentes; calycis lobi lineari-obtusi et alabastra

CVU MENTO ATS HUGO RAS fs ERE SE Sea Soe pee an Capone Deore macrocalyx

* cf. J. Burtt Davy.—The suffrutescent habit as an adaptation to environment. Journ. of

Keology, X, p. 211 et sqq. (1922).

+ W. Robyns.—Tentamen Monographiae Vangueriae Generumque affinium. Bull Jard. Bot. Etat

Brux., XI, p. 119 (1928).

184

Pachystigma triflorum Robyns sp. nov., ex affinitate P. macrocalycis, sed habitu, foliis

ellipticis multo minoribus et calycis lobis ellipticis primo visu distinctum.

Arbor parva, secundum collectores + 5 m. alta, multo sed breviter divaricate ramosa ;

rami ramulique oppositi, cylindracei, cortice cinereo interdum subdeciduo obtecti, glabri ;

ramuli novelli breves, oppositi, divaricati, teretes, internodiis brevibus vel usque ad 6-8 mm.

attingentibus, dense cinereo-tomentelli. Stipulae subherbaceae, brevissime vaginato-

connatae, subito 0-5-2 m. longe filiformi-subulatae, omnino cinereo-tomentellae, vix

seariosae et demum deciduae. Folia ad nodos ramorum hornotinorum opposita, mox

decidua, breviter petiolata, petiolo 1-1-5 mm. attingente et griseo-tomentello ; laminae

rigide herbaceae, anguste ellipticae, basi breviter attenuatae, apice plus minusve late

obtusae, statu sicco marginibus saepe recurvatis, [-1-7 cm. longae et 0-5-0-8 cm. latae,

utrinque tomentellae sed ‘leviter discolores, pagina superiore leviter glaucescentes at pagina

inferiore cinerascentes, costa media pagina inferiore distincte prominente, costis secun-

dariis inconspicuis. Cymae as nodos ramorum annotinorum plerumque defoliatorum

insertae, plerumque oppositae, divaricatae, simplices, typice triflorae sed saepe abortu

biflorae vel etiam uniflorae, bracteolatae, omnino cinereo-tomentellae, in toto 1-5-1 -8 cm.

attingentes, distinecte pedunculatae, pedunculo crassiusculo et 3-5 mm. longa. Flores

mediocres, 3-5 mm. longe pedicellati ; alabastra oblonga, distincte apiculata, circa 6-7 mm.

longa et dense pubescentia ; calycis lobi sub anthesin plus minusve patententes, anguste

elliptici, apice obtusi, 6 mm. longi, utrinque tomentelli ; corollae tubus cylindricus, -- 2 +5

mm. longus, extus dense pubescens, intus medio annulo pilorum reflexorum instructus,

lobi lanceolati, appiculati, -- 4 mm. longi quorum 1-5 mm. pro apiculis, extus breviter

pubescentes et intus carnosuli; antherae subsessiles, lanceolatae, breviter apiculatae,

1-25 mm. attingentes ; stylus gracilis, e basi sensim attenuatus, + 4:°5 mm. altus; stigma

cylindricum, apice distincte 5-lobatum, irregulariter costatum, + 1 mm. longum; discus

annularis, glaber ; ovarium hemisphericum, + 1 mm. altum, tomentellum et 5-loculare.

Fructus pedicello elongato suffultus, globosus, apice calycis lobis auctis coronatus, + 6 mm.

diametro (an maturus ?), sicco nigricans, sparse puberulus, 5-pyrenus.

SourH Arrica.—North-Transvaal: Waterberg, Olifant’s Poort, at 9 miles north of

Nylstroom, in bush on rocky kopje, rare, tree of about 5 m., flowers pale greenish yellow,

December, 1934, Galpin, 13198 (typus floriferus) (Herb. Pretor. et Herb. Brux.); Zout-

pansberg Range, January, 1925, Smuts, in Nat. Herb., 19796 (typus fructiferus) (Herb.

Pretor.).

Observation.—This new species can easily be distinguished by the xerophytic habit,

the ashy-coloured leaves, the typical 3-flowered cymes and the form and length of the

calyx-lobes.

—

(ez)

A REVISION OF THE SOUTH AFRICAN

SPECIES OF HELITOTRICHON,

Bess. ex Schultes.

By H. G. Scuwetcxerprt, B.Sc., Ph.D., F.L.S.

(With 12 Figures.)

I.—INTRODUCTION.

THE most elaborate account of the South African species of this genus is that by Stapf,

in Dyer, Fl. Cap. VIL. 472-477 (1899) under the name Avenastrum Jess., where a fairly

broad view of the species was taken.

Stapf did not consult the actual type specimens of Stewdel and Nees ab Essenbeck, but

based his identifications on the study of type numbers. For purposes of study he further-

more had only a comparatively narrow range of material at his disposal. During the last

few years, however, a fair amount of material has accumulated in various South African

herbaria. Because of this and the foregoing reasons, the author of the present paper found

it desirable to undertake a revision of the genus with a view to elucidating several remarks

made by Stapf in his account of the genus.

In this revision the material of several of the larger European and the more important

South African herbaria has been included.

Il— HISTORY OF THE GENUS.

The generic name Helictotrichon Bess. first appears in Schultes Mant. Syst. Veg. 1.

Addit. I. 526 (1827) and most probably is merely a revised spelling of the earlier name

Elictotrichon Bess. ex Andrz. [Rys. Bot. p. 9 (1823)]. The latter appeared in a list of plants

as Elictotrichon sempervirens. Bess. unaccompanied by any description and being a nomen

nudum must thus be rejected.

At a date subsequent to the publication of the name Elictotrichon sempervirens, Besser

communicated to Schultes a new classification of Avena and Trisetwm in which he proposed

several new genera, including Helictotrichon. As Besser had already used the name £.

sempervirens for Avena sempervirens Host. and as the latter was also the first species to be

listed under Helictotrichon, Avena sempervirens Host. naturally is the type species of the

genus.

The name Avenastrum first appeared as a section to the genus Avena L. in Koch, Syn:

Fl. Germ. et Helv. ed. 1. 795 (1837). This section was subsequently used in the same sense

by authors such as Benth. and Hooker [Gen. Pl. i. 1160 (1883)], Hackel [Engl. Pflzfam. 11.

2. 56 (1887) et True Grasses, 123 (1896)], Aschers. and Graebner [Syn. Mitteleurop. Flor.

i. 243 (1899)], Hitchcock [Man. Grass. Un. States, 297 (1935)], and as a subgenus by Rouy

[Fl. France, XIV, 126 (1913)].

Jessen, in Deutsch. Graser, 214 (1863) put forward the name Avenastrum as a genus

and included under it species such as (1) Avena flavescens L., (2) Avena elatior L., (3) Aira

caryophyllea L. and (4) Aira praecor L. He thus not only raised Koch’s sectional name

186

to generic rank, but modified its sense appreciably by including in it the heterogenous

elements quoted above. The name and genus Avenastrum Jessen is superfluous since

Jessen had included under it the older valid names and genera Trisetum Pers. (1805) [for (1)

above] and Arrhenatherum Beauv. (1812) [for (2) above|and for this reason must be rejected

as a “nomen illegitimum.”

Beck [F. Nieder.-Ost. 1. 72, in Ann. Nat. Hofmuseum. Wien V. 560-561 (1890)] recog-

nised the genus Avenastrum Jess. but restricted its use and conceived it generically in the

sense that Koch used it sectionally, 1.e. differing only from Koch’s conception in rank. The

genus is used in the same restricted sense by Stapf [Dyer, Fl. Cap. VII, 472 (1899)] where

strictly speaking the genus 1s Avenastrum Jess. pro parte.

Apparently Beck did not consider the genus Heuffelia Schur. [Enum. Pl. Transs. 760

(1866)] under which Avena sect. Avenastrum Koch [Syn. ed. 2. 918 (1844)] is cited as a

synonym. Avena pratensis L. was included in this genus as well as in Helictotrichon Besser

and on this account Heuffelia Schur. is congeneric with the older genus i.e. Helictotrichon

Besser.

It may be pointed out that the name Helictotrichon Bess. is not an orthographic variant

of Helicotrichum Nees (1818) as the former was derived from the adjective helictos, whereas

the latter was derived from the noun helix and on this account both genera have the right

to stand independently.

Furthermore the name Helicotrichum Bess. has been taken up in the Index. Kew-

from Reichenb. Fl. Germ. Excurs. 1406. no. 352 (1832) where it is considered a synonym

of Avena planiculmis Schrad. Benth. and Hook. f. [Gen. Pl. IIL. 11. 1160 (1883)] and Hackel

[Hngl. Pflzfam. IT. 1. 56 (1887)] quote it as a synonym under Avena sect. Avenastrum Koch.

It is probably only a misprint for Helictotrichon Bess.

III.—DESCRIPTION OF THE GENUS.

Helictotrichon Bess. ex Schult. Mant. Syst. Veg. 11. Addit. I. 526 (326 errore) (1827).

Spikelets narrowly-oblong to oblong or elliptic-oblong, laterally compressed, usually

erect or suberect, of medium size (8-15 mm. long, rarely smaller or larger), in nearly always

erect often stiff panicles, rarely nodding. Rhachilla disarticulating above the glumes and

between the valves, produced into a short bristle beyond the uppermost floret or ending

with a rudimentary lemma; rhachilla-internodes glabrous, or short- or long-hairy, 1 -5—4

mm. long. Florets 2-6, ¢, or the uppermost more or less reduced. Glwmes 2, persistent,

hyaline or subhyaline, subequal or unequal, acute or acuminate, more or less distinctly

keeled ; the lower 1—3-nerved; the upper 3—7-nerved (3-nerved in the South African

species). Lemmas usually distinctly exserted from the glumes, rarely subincluded, more

or less herbaceous with scarious or hyaline tips, often rather firm, at length becoming

indurated, acute or acuminate, bifid (rarely 4-fid), with or without bristles from the lobes,

5-ll-nerved, awned; awn dorsal, from the middle or shghtly above the middle of the

lemma, kneed and twisted (at times spuriously) below the bend; callus short or elongate,

villous. Paleas shorter than the lemmas, 2-keeled, ciliate. Lodicules 2, rather large,

hyaline. Stamens 3. Ovary hairy from above the middle or at the apex only; styles

distinct, short; stigmas usually laterally exserted, plumose. Caryopsis oblong, slightly

laterally compressed, usually grooved in front, hairy at the apex, pallid, soft, embraced by

the hardened lemma and palea; hilum linear, up to half the length of the grain; embryo

small. Caespitose perennials; leaf blades linear, usually narrow, flat or convolute, some-

times setaceous ; ligules hyaline or scarious; panicle narrow, more rarely diffuse, erect or

nodding.

Species about 65; mainly natives of the temperate regions of the northern hemisphere,

but also occurring on the mountains of Java, extending through the high mountain regions

of tropical Africa to South Africa.

187

IV.—KEY TO THE SPECIES.

A.—Rhachilla-internodes from 2-5-4 mm. long, densely bearded for the

greater part of their length :

B.—Inflorescence very compact, composed of numerous spikelets :

C.—Lemma with prominently raised nerves, finely granular

between but not on the nerves; callus 1-1} mm. long ;

glumes narrowly lanceolate, prominently nerved and

Lnijy2 WAS seer spe RR NN CE OITA er RE ane

CC.—Lemma not prominently nerved, often scabrous at the in-

sertion of the awn but smooth dorsally below it, and minutely

scaberulous towards the margins, distinctly so towards

the apex of the lobes ; callus 14-2 mm. long ; glumes broadly

lanceolate, not very prominently nerved, not sub-hyaline

BB.—Inflorescence short, lax to very lax, composed only of relatively

few spikelets :

D.—Plants about 30 em. high. Lemma about 1 -0 em. long (ex-

cluding the bristles), distinctly coarsely granular on the

back in the dorsal middle third, granules also present on

Chemnervesteremes re anrackyc a ckiech. care csp tee eile. ne

DD.—Plants about 80 cm. high. Lemma slender, about 1-5 cm.

long (excluding the bristles), smooth dorsally, not promi-

nently nerved, very minutely scaberulous near the margin

in the upper half, slightly scaberulous along the nerves. .

AA.—Rachilla-internodes 1 -5-2-5 mm. long:

E.—Rachilla-internodes glabrous, smooth, acute, hardly dilated

at the apex; lobes of lemma (above insertion of the awn)

about 10 mm. long or slightly longer..................-

EE.—Rhachilla-internodes bearded and dilated near the apex ;

lobes of lemmas up to 8 mm. long, but usually much shorter :

F.—Lemmas scabrid or scaberulous :

G.—Florets scarcely exserted beyond the glumes ;

glumes broadly lanceolate ; outer surface of lemmas

minutely, scaberulous) allovera...4-2.9.-.-0:s-:-

GG.—Florets exserted well beyond the glumes; glumes lanceo-

late ; lemmas scabrid mainly at the insertion of the awn:

H.—Lemmas about 10 mm. long; rhachilla-internodes

ZB PON GEE cy veel chads Seis oi sree teee anety ccs ish Ralerat its

HH.—Lemmas about 7 mm. long; rhachilla-internodes

Aloo Wey rte, Ione e pecac acta bes oabicc Nab be Dele

FF.—Lemmas glabrous on the back below the point of insertion

of the awn:

I.—Lemmas about 7 mm. long ; column of awn with 1-3

URES a Bune irs Aaa eis aioe Pcie eae an tas Mentors a

. quinquesetum.

longum.

. namaquense.

barbatum.

leoninum.

. Galpin.

. capense.

. hirtulum.

. natalense.

188

Ii.—Lemmas usually 10 mm. long or somewhat longer ;

column of awn with more than 3 twists:

K.—Leaves many from the base of the culms, seta-

ceous and up to 40 cm. long; spikelets usually

gaping; rnachilla-internodes long-hairy and

ustallivarevidemtbse,s Ach scteratsiom » Srice-ndetc eran 10. longifolium.

KK.—Leaves not setaceous, often convolute. Rha-

chilla-internodes usually not readily evident :

L.—Glumes broadly lanceolate ; lemmas lan-

ceolate in profile ; lobes (above insertion

of awn) 3-5 mm. long; anthers 1 mm.

long, rarely 2mm. insome florets ; cleisto-

PAMOUSH ete shenisek Oe gba eae 11. turgidulum.

LL.—Glumes lanceolate, acuminate; lemmas

narrowly lanceolate, slender in profile ;

lobes (above insertion of awn) 6-8 mm.

long; anthers 1-5-2 mm. long......... 12. Dodw.

V.—ENUMERATION OF THE SPECIES.

1. H. quinquesetum (Steudel) Schweickerdt, comb. nov.

Syn.: Avena quinqueseta Steudel, in Flora, 1829, 485; Kunth., Enum. I, 305 (1833).

Trisetum Steudelii Nees, in Linnaea, VII, 308 (1832); FI. Afr. Austr., 349

(1841) ; Steudel, Syn. Pl. Glum, i, 228 (1854); Dur. and Schinz, Consp.

Fl. Afr., V, 840 (1894).

Avenastrum quinquesetum Stapf, in Dyer, Fl. Cap., VIL, 474 (1899).

Culms 50-75 em. high, glabrous, 2-4-noded, sheathed to 5-10 em. below the panicle.

Leaves 4-6, arising from near the base, 2 higher up ; lowest sheaths somewhat compressed,

firm, minutely puberulous, strongly nerved, the upper more terete and glabrous; ligule

truncate, irregularly laciniate, up to 3 mm. long; blades linear, with callous tips, flat or

conduphicate, up to 25 em. long and 4 mm. wide, the upper usually much shorter, rigid,

strongly and closely nerved, glabrous. Panicle contracted, rather dense and almost spilke-

like, narrow, 12-18 cm. long, straight or subflexuous; branches fascicled, very unequal,

branched from near the base, adpressed to the rhachis; the longest up to 5 cm. long.

Spikelets 12-18 mm. long, very loosely 2-sub—3-flowered. Glumes narrow-lanceolate,

acuminate, prominently nerved, glabrous, scarious along the margin and towards the

apex ; the lower 8-11 mm. long and about 14 mm. broad; the upper 10-13 mm. long and

13-3 mm. broad. Rhachilla-internodes 3-44 mm. long, acute, slightly broadened towards

their apex, slender, hairy almost to the base with white spreading hairs up to 4 mm. long.

Lemmas shortly exserted, linear-lanceolate, the lowest (including callus and lobes, but

excluding the awn) up to 18 mm. long, pallid, rather firm, with prominently raised nerves,

glabrous below the insertion of awn, finely granular between the nerves up to the base of

the valve ; lobes 10-12 mm. long, finely scaberulous and scarious towards the apex ; each

lobe usually terminating in 2 bristles of unequal length (termination of veins). Callus

subulate, up to 1} mm. long, densely bearded with hairs about 14 mm. long ; awn inserted

-++ 7 mm. from the base of the valve ; column 8-10 mm. long; bristle up to 18 mm. long.

Palea about 9 mm. long; keels prominently ciliolate. Anthers 24 mm. long. Ovary

pubescent for the greater part of its length.

Carn Province.—Table Mountain, near Capetown, Ecklon, 929! without precise

locality, Harvey, 295!

This species appears to be very rare ; it is represented in most herbaria only by dupli-

cates of the type gathering (Hcklon, 929). The type specimen of Avena quinqueseta Steudel

is deposited in the Fielding Herbarium, Oxford.

189

2. H. longum (Stapf) Schweickerdt comb. nov.

Syn.: Tvrisetum antarcticum Nees, Fl. Afr. Austr. 346 (1841) pro parte, exclud. syn.

pro parte; in Linnaea, XX, 254 (1847), not in Linnaea, VII, 307 (1832).

Trisetum longifolium Nees, Fl. Afr. Austr., 347 (1841), pro parte.

Avena longa Stapf, in Kew Bull., 1897, 292.

Avenastrum longum Stapf, in Dyer, Fl. Cap. VII, 473 (1899).

Avenastrum longum Stapf, var. grande Stapf, l.c.

Culms 60-110 em. high, glabrous, about 3-noded, sheathed almost up to base of in-

florescence. Leaves 3-6 from near the base, and usually 3 higher up along the culm ; sheaths

rather loose ; the lower persistent, usually glabrous, strongly striate; the upper glabrous

or more or less densely pubescent with reflexed short hairs ; ligules conspicuous, 3-5 mm.

long, truncate, usually somewhat torn; blades linear to linear-lanceolate, tapering to a

fine point, 15-30 em. long, very variable in width, 2 -5-10 mm. wide, flat or involute, flaccid,

glabrous, smooth or rough above, markedly striate. Panicle contracted or somewhat inter-

rupted, linear, linear-oblong or oblong, 20-30 cm. long, nodding and flexuous, or fairly

straight and robust; branches fascicled, very unequal; the longest up to over 5 cm. long,

branched from near the base or simple, filiform, flexuous, scaberulous. Spikelets up to

30 mm. long (including the bristles of the lemmas), usually about 20 mm. long, 4-5 flowered.

Glumes lanceolate, acuminate; the lower 6-5-9 mm. long, 1-sub—3-nerved; the upper

9-12 mm. long, 3-nerved. Rhachilla-internodes 2-5-3 mm. long, bearded throughout with

hairs up to.3 mm. long. Lemmas rather firm, pallid, sometimes purplish, up to 18 mm.

long (including the bristles but excluding the awn), not conspicuously nerved, minutely

granular on the back below the insertion of the awn, sometimes scabrid along the margins ;

lobes beyond the insertion of the awn about 10 mm. long, scabrid. Callus 1 -25-2 mm.

long, long-bearded. Awn inserted about 8 mm. distant from the base of the lemma ;

column 6-9 mm. long; bristle 15 mm. or somewhat longer. Palea about 7 mm. long,

inconspicuously ciliolate. Anthers 2-5-3 mm. long. Ovary puberulous in upper half.

The type specimen (Zeyher, 1807) is deposited in the Kew Herbarium.

Cape PENINSULA.—Cap. b. Spei, Bergius, 228! In humidis planib. Capens. atque

dunarum, Oct., Zeyher, 1807, 1807b! In arenosis ad litus maris pone van Kampsbay,

Oct., Zeyher (38?) in Herb. Bolus, 21723 et in Herb. Mus. Austro-Afr., 19430. Kenilworth

Race Course, LZ. Bolus, 15054! University grounds, Rondebosch, Nov., Levyns, 3674.

Ottery Rd., sandy flats, Adamson, sn! Klein Slangkop, about 500 ft., Wolley-Dod, 3004.

Camp Ground, Wolley-Dod, 3473. Slopes beyond Miller’s Point, Sept., Wolley-Dod, 3003!

Table Mountain, Rogers, 30434! Upper northern slopes of Lion’s Head, Wolley-Dod, 3571!

Orange Kloof below farm, Wolley-Dod, 3128! In grassy rocky places above Camps Bay,

McOwan, 1793! Stellenbosch Div., near Firgrove, Oct. C. Sandwith, 147!

Among the South African representatives, this species appears to be the most robust

with regard to inflorescence ; in width of leaf and indumentum of the vegetative parts it

appears to be somewhat variable and for this reason Stapf’s var. grande is considered merely

an extreme form which does not deserve varietal rank.

3. H. namaquense Schweicherdt nom. nov.

Syn.: Trisetum Dregeanum Steud., Syn. Pl. Geum., 227 (1854), nomen illegit. ;

Dur. and Schinz, Consp. Fl. Afr., V, 838 (1894).

Trisetum barbatum Nees B minus Nees, Fl. Afr. Austr., 345 (1841).

Avenastrum dregeanum (Steud.) Stapf, in Dyer, Fl. Cap., VII, 473 (1899).

A densely caespitose perennial. Culms about 30 cm. long, slightly bulbous at the

very base, glabrous, striate, slender, 1-noded ; node much below the middle ; sterile leafy

shoots many at the base of the culms. Leaves about 4 from near the base of the culm, one

190

or two sheathing it for the greater part of its length ; sheaths fairly tight, glabrous; the

lower persistent and sub-membranous; the upper striate, somewhat contorted; ligule

truncate, 1-1} mm. long; blades linear with callous tips; the lower up to 10 cm. long,

usually shorter and 2 mm. or less wide, flat or convolute, rigid, subglaucous, striate, hairy

above, scaberulous beneath. Panicle suberect or nodding, 6-10 cm. long, very loose,

secund; branches paired, 1—3-spiculate, very unequal, somewhat spreading, filiform,

scabrid ; the lowest up to 25 em. long. Spikelets 10-15 mm. long, very loosely 3-4-flowered.

Glumes unequal, submembranous, lanceolate, acuminate; the lower 8-9 mm. long and

2 mm. broad, l-nerved; the upper 12 mm. long, 3-3} mm. wide, 3-4-nerved, scabrid on

the main nerve. Rhachilla-internodes 24-3 mm. long, bearded in the upper two thirds with

hairs 4-5 mm. long. Lemmas long-exserted, oblong-lanceolate ; the lowest + 14 mm.

long (including the setaceous lobes but excluding the awn), glabrous, pallid, firmly coriaceous,

coarsely granular-scabrid on the back just below the insertion of the awn, smooth and

glabrous towards the base ; lobes strongly nerved, submembranous, ending in scaberulous

bristles. Callus 1{ mm. long, curved, bearded with hairs 2-5 mm. long. Awn inserted

+ 54 mm. from the base of the valve; column + 8 mm. long; bristle + 12 mm. long.

Palea about 8 mm. long, linear-lanceolate, conspicuously ciliolate. Anthers up to 3} mm.

long. Ovary hairy above the middle.

Carr Province: Namaqualand, Kamiesbergen, steinige Berggegend bei Ezelsfontein,

3-4,000 ft., Nov., Drége, 2526 !

The type specimen is deposited in the Berlin Herbarium.

Apparently a very rare species since only this gathering is known from the European

herbaria. It has apparently not been gathered again either on the Kamiesbergen or in

any other locality since Drége’s time.

Stapf [in Fl. Cap. VII, 474 (1899)], under Avenastrum dregeanum cites Drége 2625.

This number is probably an error for 2526 which is the number of the type specimen of

Trisetum barbatum Nees var. [ minus in Herb. Nees.

Steudel [Syn. Pl. Glum. 227 (1854)] created the name Tvisetum Dregeanum, with the

intention that it should replace the name Trisetwm barbatum Nees [Fl]. Afr. Austr. 345

(1841)] which he considered to be a later homonym of 7’. barbatum Steud. [Nom. ed. II. 11.

713 (1841)]. The latter, however, is a nomen tantum whereas Nees’ species bearing the same

name was validly published. Consequently 7. barbatum Nees may stand whereas Trisetum

Dregeanum Steud. is a superfluous name and as such must be rejected.

In Steudel’s description of 7. Dregeanum the culms are described as being “ pedali”’

i.e. about 30 cm. high. This obviously does not apply to 7. barbatum Nees var. a, which

is a much taller plant and of which I have seen the type in Herb. Nees. It applies more-

over to 7’. barbatum Nees var. 8 minus Nees as I have seen both the type of the latter and

the specimen of T. Dregeanum from the Steudel Herbarium (Paris) and have found them to

agree in every respect, in fact they are part of the same gathering by Drége. It should be

pointed out, however, that both these sheets represent a species distinct from 7’. barbatum

Nees var. a.

As a result of misidentification Steudel thus inadvertently applied a new name to the

wrong plant. Later authors, e.g. Dur. and Schinz, and Stapf in following Steudel have

consequently also misapplied the epithet ‘‘dregeanum.”’ It thus cannot be accepted to

designate Nees’ 7’. barbatum var. § minus. As this variety should be given specific rank,

I have named it H. namaquense.

4. H. barbatum (Nees) Schweickerdt comb. nov.

Syn.: Trisetum barbatum Nees, var. — Nees, Fl. Afr. Austr., 345 (1841), non Steudel.

Densely caespitose. Culms up to 75 cm. high, with numerous sterile leafy shoots from

the base, glabrous or very minutely scaberulous, slender, striate, about 2-noded; nodes

191

somewhat exserted. Leaves few from near the base of the stem, two to three upwards

along the culm; lowermost sheaths submembranous, pale, striate, persistent, glabrous,

finally splitting into fine fibres; upper sheaths somewhat striate, glabrous or scaberulous

towards the ligule, fairly lax and somewhat contorted ; ligule about 3 mm. long, irregularly

fimbriate ; blades narrowly linear, up to 17 cm. long, usually much shorter, 13-3 mm.

wide, striate, scaberulous. Panicle very lax, few-flowered, about 8-10 cm. long, somewhat

branched ; branches filiform, unequal; the lowest up to 24 cm. long, scaberulous, each

bearing 1-2 spikelets. Spikelets very laxly 3-4-flowered, 14-17 mm. long. Glumes lanceo-

late; lower 8-9 mm. long, submembranous, glabrous but scaberulous along the nerves,

acute ; upper 14 mm. long, 3-nerved, submembranous, acute, glabrous, somewhat scabe-

rulous along the nerves. Rhachilla-internodes 3-3-5 mm. long, bearded in the upper two

thirds with white hairs 5-6 mm. long. Lemmas 16 mm. long (including the bristly lobes,

but excluding the awn), smooth on the back below the point of insertion of the awn, minutely

but distinctly scaberulous along the nerves especially in region of the hyaline lobes beyond

the insertion of the awn; lobes about 10 mm. long, each ending in a fine bristle. Callus

about 1 mm. long, bearded with white hairs about 2mm. long. Awzn inserted about 6-7 mm.

from base of valve ; column of awn + 10 mm. long ; bristles 15-17 mm. long. Palea 8 mm.

long, not very conspicuously ciliolate. Anthers linear, 2-25 mm. long. Ovary pubescent

in upper half.

Care Province.—Namaqualand, on the Kamiesbergen, Nov., Drége, 2572b! in Herb-

Nees. et in Herb. Mus. Austr.-Afric.

The type specimen is deposited in the Berlin Herbarium.

Nees [Fl]. Afr. Austr. 345 (1841)] cites as the type of Trisetum barbatum var. a gather.

ing by Ecklon. This is probably an error as the type in Herb. Nees, viz. the specimen cited

is a gathering by Drege.

Trisetum barbatum Steud. [Nom. ed. IT. ii. 713 (1841)] is a nomen tantum and therefore

Trisetum barbatum Nees, which was validly published, can stand and takes precedence

over the former.

Steudel |Syn. Pl. Glum. 227 (1854)] renamed the above plant Trisetwm Dregeanum, but

in reality his description applies to Trisetum barbatum Nees var. § minus Nees which is

a species distinct from 7’. barbatum Nees var. a. T. Dregeanum Steud. therefore must be

sunk in synonomy under the above species. For further information the reader is referred

to the remarks in this paper under H. namaquense.

H. barbatum (Nees) Schweickerdt is apparently a rare species, since in herbaria it is

only represented by Drége’s gathering. No other collector seems to have found this species

since Drége’s time.

5. H. leoninum (Steudel) Schweickerdt comb. noy.

Syn.: Avena leonina Steud., in Flora, 1829, 484; Kunth, Rév. Gram. 11, 521, t. 175

(1831); Kunth, Enum., i, 303 (1833); Trin., in Mem. Acad. Petersb.,

Ser. VI, Sc. Nat., IV, 1, 29 (1836).

Danthonia leonina Steud. ex Kunth, Enum., i, 303 (1833), in syn.

Trisetum antarcticum Nees, in Linnaea, VII, 307 (1832), pro parte, non Trin.

Avenastrum antarcticum Stapf, in Dyer, Fl. Cap., VII, 476 (1899), pro parte.

Densely caespitose with numerous barren shoots. Culms up to 50 cm. high, usually

somewhat shorter, glabrous, 2-3-noded, internodes included or exserted. Leaves mainly

from near the base; sheaths terete, fairly tight, glabrous or with spreading fine hairs,

strongly striate ; lowermost persistent and eventually splitting into fibres; upper some-

what compressed, not slipping off the culms; ligule about 14 to 2 mm. long, irregularly

dentate ; blades linear, flat, tapering to a callous point, up to 10 cm. long, but usually much

shorter, up to 3 mm. wide, glabrous or scantily pubescent, markedly nerved. Panicle

192

contracted, linear, erect, stiff or somewhat flexuous, up to 10 em. long, often much shorter ;

lower branches in pairs, unequal in length; the longer up to 4 em. long and 2-3-spiculate,

scaberulous. Spikelets 3-5-flowered, 12-14 mm. long, greenish. Glwmes lanceolate, acumi-

nate, unequal, glabrous; margins and apex scarious; the lower 53-7 mm. long, narrow ;

the upper 74-9 mm. long. Rhachilla-internodes 14-2 mm. long, gradually tapering to an

acute point, never widened at the apex, usually glabrous but very rarely with a few scattered

hairs near the apex. Lemmas 12-14 mm. long (including callus and lobes, but excluding

the awn), coriaceous, dorsally scaberulous, decreasing in scabrosity towards the lobes above:

the insertion of the awn; lobes 9-10 mm. long, with scarious margins. Awn inserted

about 54 mm. from the base of lemma; column 7-9 mm. long; bristle + 13 mm. long.

Callus 1 mm. long, clothed with short hairs. Palea conspicuously ciliolate. Anthers 2 mm.

long. Ovary pubescent in the upper half.

Carr Province.—Cape Peninsula: Table Mountain, Pappe, pro parte! Lion’s Head.

Mountain, Heklon, 928! and Zeyher, 101! Signal Hill, nr. Lion Battery, Wolley-Dod, 2747!

Field below Prince of Wales Blockhouse, Wolley-Dod, 1474! 1477! Orange Kloof, below

Constantia Nek, Oct., Bolus, 14667! Near Maitland Stn., Oct., Wolley-Dod, 3167! Along-

side Pipe Track, Orange Kloof, Oct., F. Bolus, s.n.! Under Pine Trees, Signal Hill, Aug.,

Levyns, s.n.!

On account of the glabrous acute rhachilla-internodes this well-defined species taxo-

nomically occupies a singular position among the South African representatives of the

genus.

6. H. Galpinii Schweickerdt. spec. nov. ; affine H. twrgidulo (Stapf) Schweickerdt, sed lem-

matibus omnino scaberulis distinguitur.

Syn.: Phillips, in Ann. A. Afr. Mus., XVI, 1, 343 (1917), sub. Avenastrum turgi-

dulum Stapf.

Gramen perenne, dense caespitosum. Culmi erecti, usque ad 60 cm. alti, graciles;

2-3-nodes, paniculis et nodis superioribus exsertis, glabri laevesque, striati, ad basin ramis

foliatis. Vaginae pubescentes, infimae striatae, demum in fibras fissae, superiores striatae,

vix glabrae. Lagulafere 1-5 mm. longa. Laminae erectae, fere rigidae, lineares, 12-16 mm.

longae, nonnunquam breviores, in acumen callosum productae, planae vel leviter involutae,

2-5-3 mm. latae, subtus pilis fere dense munitae, supra sparse pubescentes et valde striatae.

Pamcula leviter contracta, 10-16 cm. longa, angusta; rhachis glabra laevisque, apicem

versus nonnunquam, scaberula ; rami fasciculati, inaequales, usque ad 3-5 cm. longi, erecti,

scaberuli, 2-4-spiculati. Spiculae 3-4-florae, floribus vix exsertis, 8-10 mm. longae, erectae.

Glumae subaequales, tota facie scaberulae (vel minute pubescentes), valde striatae, sub-

hylinae, purpureo-pictae ; inferior 9 mm. longa et 2 mm. lata, 3-nervis, late lanceolata ;

superior 10-5 mm. longa et 3 mm. lata, 3-nervis, late lanceolata. Internodia rhachillae

fere 1-25 mm. longa, apicem versus pilis 3 mm. longis valde barbata. Lemmata usque ad

8-5 mm. longa, 5-nervia ; arista 4 mm. basin lemmatis inserta, tota facie minute scaberula ;

lobi 4-4-5 mm. longi, scabri, in setas scabras producti. Callus pilis 2 mm. longis barbatus.

Avistae columna 5 mm. longa. Paleae dorso tertiis partibus superioribus scaberulae, carinis

ciliolatis. Antherae 2 mm. longae. Ovariwm 1-25 mm. longum, apicem versus dense

villosum.

Care Proyvince.—Barkly Kast distr., at an altitude of 9,700 feet (2,900 metres) on Ben

McDhui (Wittebergen), March, Galpin, 6902, pro parte (type deposited in Nat. Herb. Pretoria

and in the Kew Herbarium).

A species which appears to be well-defined by the subincluded florets, the lemmas of

which are minutely scaberulous on the outer surface.

Galpiv’s gathering 6902 is a mixture of the above species and typical H. turgidulum

(Stapf) Schweickerdt ; the latter may, however, be readily distinguished by the glabrous

Jemmas.

193

7. H. capense Schweickerdt spec. nov.

Syn.: Ave. strum antarcicum Stapf, in Dyer, Fl. Cap., VII, 476 (1899), pro parte.

affine H. hirtulo (Steud.) Schweickerdt, sed paniculis laxioribus, spiculis

majoribus, lemmate minus scabrido, rhachillae internodio longe brabato

differt.

Gramen perenne, dense caespitosum. Culmi erecti, usque ad 100 em. alti, glabri,

circiter 3-nodes, nodis exsertis et basin versus ramis foliatis. Vaginae striatae, glabrae vel

minute puberulae, inferiores demum in fibras fissae, superiores fere laxae. Ligula circiter

0-75 mm. longa. Laminae filiformis vel anguste lineares, nonnunquam involutae, usque

ad 25 em. longae, subtus glabrae vel minute scaberulae, supra pilis sparse praeditae.

Pancula erecta vel leviter flexuosa, usque ad 20 cm. longa; rhachis glabra; rami inae-

quales, fasciculati, filiformis, leviter flexuosi, usque ad 4 cm. longi. Spiculae circiter 15 mm.

longae, 4-5-florae. Gluwmae inaequales, lanceolatae, acuminatae, subhyalinae, valde

nervatae ; inferior fere 5-5 mm. longa; superior fere 7-9 mm. longa, dorso apicem versus

minute scaberula. Lemmata usque ad 12 mm. longa, lineari-lanceolata, pallida, firma,

dorso valde vel leviter scabra; lobi fere 6 mm. longi, minute scaberuli, in setis scabris

producti. Avista fere 5 mm. basin lemmatis inserta; columna 4-5-mm. longa. Inter-

nodia rhachillae 2 mm. longa, apicem versus dilatata, pilis 3 mm. longis dense barbata.

Paleae 5-5 mm. longae, carinis ciliolatis. Antherae 2 mm. longae. Ovariuwm apicem versus

villosum.

The type specimens are deposited in the Kew Herbarium and in Nat. Herb. Pretoria.

CapE PRrovince.—Cape Peninsula: Table Mountain, Pappe, pro parte!. Kalk Bay

Mountain, Bolus, 14652!. Riversdale distr.: Zoetmelksrivier, Burchell, 6694!. Hast

London distr. : East London, May, Rattray, 720!. Gonubie, Sept., Dyer, 2053!. Komgha

distr.: near Komgha, Flanagan, 935 (type)!. Kentani distr.: Among tall valley grasses,

Pegler, 2057!. King Williamstown distr.: Nahoon River, near Kei Road Station, Nov.,

Galpin, 8244!. Near Cemetery, Nov., Sim, 2803!.

Narat Province.—Durban: Clairmont, Schlechter, 3089 !.

The plants placed under this species bear a close resemblance to H. hirtulum (Steud.)

Schweickerdt and are often confused with that species. H. capense has, however, larger

spikelets and the panicle tends to be somewhat more open than that of the allied species.

8. H. hirtulum (Steud.) Schweickerdt comb. nov.

Syn.: ?Avena hirta Schrad., in Goett. Gel. Anz., 111, 2075 (1821); Schult. Mant,

pt. 11, 374 (1824).

Avena symphicarpa Trin. ex Steud., Nomencl., ed. 1, 1, 173 (1840), nomen

tantum.

Avenastrum antarcticum Stapf, in Dyer, Fl. Cap., VII, 476 (1899), pro parre

Trisetum hirtum Nees, Fl. Afr. Austr., 350 (1841), non Trin.; Linnaea, XX,

254 (1847).

Trisetum hirtulum Steud., Syn. Pl. Glum, i, 228 (1854); Dur. and Schinz,

Consp. FI. Afr., V, 838 (1894).

A weak perennial with several barren shoots between the culms. Culms up to 100 cm.

long but usually very much shorter, terete, glabrous, 2-3-noded ; upper internodes exserted.

Leaves few near the base, soon dying off, higher up somewhat distant ; sheaths persistent ;

the lower soon breaking up into fibres, glabrous; the upper pubescent with reflexed hairs

or glabrous, fairly light, terete ; ligule about 1 mm. long, irregularly dentate ; blades nar-

rowly linear, tapering to an acute point, often involute, up to 25 cm. long, subrigid or flaccid,

glabrous or hairy, finely nerved, smooth or scaberulous. Panicle contracted, linear, erect

194

or somewhat flexuous, 10-20 cm. long, branched ; lower branches usually in pairs, of un-

equal length ; the longer up to 5 or 6 cm. long. Spikelets up to 10 cm. long, 3-5-flowered.

Glumes unequal, acute ; the lower narrowly lanceolate, glabrous, almost hyaline, 1-nerved,

3-6 mm. long; the upper broadly lanceolate, narrowed at the base, glabrous, hyaline

towards the margins and apex, 3-nerved, up to 7 mm. long. Rhachilla-internodes 14-2 mm.

long, widened and flattened towards the apex, acute, clothed in the upper part with hairs

1-5-2 mm. long, glabrous in the lower part. Lemmas exserted beyond the glumes, lanceo-

late, acuminate, dorsally scabrous especially below the point of insertion of the awn; the

lower 7-8} mm. long; the upper somewhat shorter; lobes hyaline, setaceous, glabrous.

Awn inserted 34-4} mm. from the base of the valve; column 5-7 mm. long. Callus of

lemma very short, inconspicuous, bearded with hairs about 0-5 mm. long. Palae hyaline,

conspicuously and markedly ciliolate along the keels, about 6-63 mm. long. Anthers

14-2 mm. long. Ovary pubescent in upper half.

The type specimen indicated below is deposited in the Berlin Herbarium.

CarE Proyincre.—Cape Peninsula: Rondebosch, University Grounds, Levyns, 3560 !,

3669!. Caledon distr.: Zwartberg, near the Hot Springs, 1,000-2,000 ft., Hcklon and

Zeyher, 4553!. Albany distr.: Near Grahamstown, McOwan, 1302!. Kalksteinriicken auf

der Hohe des linken Buschmann—flusses Ufer, Zeyher, 143!. Grahamstown, Apr., Daly,

150, 152.! Botha’s Hill, Dyer, 1480!. Bathurst distr.: Trappes Valley, Dec., Daly, 678.!

Alexandria distr.: Urwialder bei Olifantshoek am Bosjesman-rivier, Hcklon, s.n. (type)!

Queenstown distr.: Fincham’s Nek, 4,000 ft., Galpin, 32811.

This species is allied to H. capense, but may be distinguished from the latter in being

a weaker plant with smaller spikelets and having an inflorescence which usually is more

contracted and spike-like than that of the allied species.

[Whether Avena lirta Schrad. l.c. is conspecific must remain an open question until

the type specimen has been located. It is believed to be at Leningrad but so far has not

been traced there. ]

9. H. natalense Schweickerdt stat. nov.; affine H. longifolio (Nees) Schweickerdt, sed

foliis latioribus, spiculis glumis lemmatibusque minoribus differt; affine H. hirtulo

(Steud.) Schweickerdt, sed panicula laxiora, lemmatibus glabris distinguitur.

Syn.: Avenastrum caffrum Stapf, var. ? natalensis Stapf, in Dyer, Fl. Cap., VII,

477 (1899) ; Medley Wood, Natal Plants, II, tab. 191 (1904).

Gramen perenne. Culmi erecti, usque ad 1 m. alti, 4-nodes, infra nodes minute pube-

ruli, graciles, nodis et paniculis exsertis. JFoliorwm vaginae striatae, minute scaberulae,

pilis sparsis munitae. Ligulae breves, circiter 0 -5mm.longae. Laminae planae, vixsubflaccidae,

usque ad 25 cm. longae et 4 mm. latae, perraro involutae, valde striatae, subtus scaberulae

supra scaberulae et pilis sparsis munitae. Panicula 25 em. longa, diffusa, laxa, rhachis

scaberula, filiformis ; rami fasciculati, valde inaequales, usque ad 3-5 cm. longi, simplices

vel parce ramosi, filiformes, flexuosi, scaberuli vel hispiduli. Spiculae 7-8 mm. longae,

laxe 3-sub—4-florae, floribus exsertis. Glwmae acuminatae, valde inaequales; inferior

3 mm. longa, anguste lineari-lanceolata, 1-nervis, glabra, secus carinam scaberula ; superior

6 mm. longa, lanceolate, valde 3-nervis, glabra, secus carinam scaberula. Lemmata ex-

serta, usque ad 8 mm. longa, lineari-lanceolata, pallida, rigida, valde nervata, dorso infra

aristam minute granulata ; lobi 3 mm. longi vel breviores, in setas producti, secus nervos

minute scaberuli. Arista 4-5 mm. basin lemmatis inserta, columna breve fere 3 mm. longa.

Callus brevissimus, breviter barbatus. Internodia rhachillae fere 1-5 mm. longa, apicem

versus pilis 3 mm. longis barbata. Paleae fere 4 mm. longae, carinis conspicue ciliolatis.

Antherae 1-5-2 mm. longae. Ovarium apicem versus hirsutum.

The type specimens are deposited in the Kew Herbarium and in the Natal Herbarium,

Durban.

195

Narat Province.—Umvoti distr.: Rietvlei, 4,000-5,000 ft., Buchanan, 238 (type) !.

Zululand: Melmoth, Imfulazane, 4,500 ft., Mogg, 6089 !.

TRANSVAAL Provincre.—Belfast Distr.: Dullstroom, 6,500 ft., Galpin, 130081.

Although the lemmas in Galpin 13008 are on the whole more markedly nerved and

slightly less conspicuously granulate than Buchanan 238, I have no hesitation in referring

this sheet to H. natalense Schweickerdt.

Whereas all the other South African species have markedly twisted columns to the

awn, this is not the case in the species in question, a character met with in the allied Avena-

strum lachnanthum Pilger.

10. H. longifolium (Nees) Schweickerdt comb. nov.

Syn.: Avena caffra Stapf, in Kew Bull., 1897, 293.

Trisetum longifolium Nees, Fl. Afr. Austr., 348 (1841), pro parte; Steud.,

Syn. Pl. Glum., 228 (1854); Dur. and Schinz, Consp. Fl. Afr., V, 839

(1894).

Avenastrum caffrum Stapf, in Dyer, Fl. Cap., VII, 477 (1899) ; Phillips, in

Ann. 8. Afr. Mus., XVI, 1, 343 (1917).

Densely caespitose with several sterile leafy shoots. Culms up to 90 cm. long, glabrous,

3-4-noded, sheathed to the base of the panicle. Leaves several from the base, 2—3-higher

up along the culm; lower sheaths persistent, glabrous, markedly striate, breaking up

into somewhat coarse fibres; upper not very tight, slightly contorted, glabrous, not

markedly striate or only so towards the ligule ; ligule oblong, up to 14 mm. long; blades

usually very narrow, subsetaceous, convolute; the lower up to 35 cm. long; the upper

usually shorter, glabrous, strongly and closely few-nerved, with distinctly rough margins.

Panicle usually lax and open, up to 20 cm. long; flexuous or slightly nodding and sub-

flaccid; rhachis filiform, striate; branches fascicled, very unequal; the longest up to

5 em. long, branched or simple, finely filiform, flexuous, scaberulous to finely hispidulous.

Spikelets loosely 3-4-flowered, 8-10 mm. long. Glwmes very thin, almost hyaline, strongly

nerved ; lower 5-64 mm. long, very narrowly lanceolate, glabrous, acute, 1-nerved, scaberu-

lous along the nerve ; upper 8-9 mm. long, lanceolate, acuminate, 5-nerved, scaberulous

along the main nerve and margins. Rhachilla-internodes usually prominently exposed,

about 2 mm. long, bearded with hairs 3-4 mm. long. Lemmas exserted, linear-lanceolate ;

the lowest about 10 mm. long (including the lobes, but excluding the awn), glabrous, firm

coriaceous, very minutely and evenly granular on the back below the insertion of the awn ;

lobes + 5 mm. long, scarious, produced into short fine bristles. Callus #? mm. long, bearded

with hairs 15-2 mm. long. Awn inserted 5-5} mm. from base of the lemma ; column 5-8

mm. long; bristle +10 mm. long. Palae about 6 mm. long; keels ciliolate. Anthers

2-3 mm. long. Ovary pubescent in upper half.

Care Province.—Aliwal North distr. : Witte Bergen, on rocks, 7,500 ft., Drége, 8134!

(Herb. Nees, lectotype!). Murraysburg distr.: At Snyder’s Kraal, Tyson, 278!

ORANGE FREE State.—Senekal distr.: Wonderkop, frequent in moist places on upper

and middle slopes of mountain, Goossens, 845! Doornkop, frequent on slopes of mountain,

Goossens, 701!

BasutToLanp.—Mafeteng distr.: Station Likhoele, Dieterlen, 400a! Leribe Plateau,

Dieterlen, 967!

This species is hable to infection by a smut.

The species does not appear to occur in the south-western region of the Cape Province.

Stapf’s remark [Fl]. Cap. VII, 477 (1899)] under Avenastrwm caffrum is fully justified, as the

sheets cited by Nees [Fl. Afr. Austr. 348 (1841)] under Trisetum longifolium are a mixture

196

of two distinct species. Ecklon’s specimens from the dunes near Capetown belong to

Helictotrichon longum (Stapf) Schweickerdt and in this paper are referred to that species,

whereas Drége 8134 in Herb. Nees (Berlin) has been selected as the lectotype of H. longi-

folium (Nees) Schweickerdt. The specific epithet of the latter species takes precedence

over that of Avena caffra Stapf.

11. H. turgidulum (Stapf) Schweickerdt comb. nov.

Syn.: Trisetum antarcticum Nees, Fl. Afr. Austr., 346 (1841), pro parte.

Trisetum imberbe Nees, Fl. Afr. Austr., 347 (1841); Steud., Syn. Pl. Glum.,

I, 228 (1854); Dur. and Schinz, Consp. FI. Afr., V, 838 (1894).

Avena turgidula Stapf, in Kew Bull., 1897, 293.

Avenastrum turgidulum Stapf, in Dyer, Fl. Cap., VII, 474 (1899); Medley

Wood, Natal Plants, II, tab. 190 (1904); Phillips, in Ann. 8. Afr. Mus.

XVI, 1, 343 (1917).

Densely caespitose. Culms 30-100 em. long, erect or geniculately ascending, glabrous,

2-3-noded ; upper 2-3 internodes more or less exserted ; uppermost often well exserted.

Leaves few near the base, about 3 higher up along the culm; sheaths terete, not very tight,

glabrous, puberulous or more rarely pubescent with short reflexed hairs, striate, often some-

what contorted ; ligule truncate, up to 1-5 mm. long; blades linear, tapering to an acute

point, up to 15 em. long but often much shorter, up to 4 mm. broad, flat or involute, more

or less rigid or subflaccid, subglaucous, glabrous, more rarely scantily hairy, scaberulous

above. Panicle contracted, sometimes interrupted, erect or slightly nodding, up to 30 cm.

long; branches fascicled ; the longer up to 5 cm. long, branched or simple, with spikelets

sub-erect or somewhat spreading, filiform, scabrid. Sprkelets 10-12 mm. long, greenish,

compactly 3-5-flowered. Glumes lanceolate, acuminate ; the lower 5-7 mm. long, |-nerved ;

the upper 7-9 mm. long, 3-nerved. Rhachilla-internodes 2 mm. long, bearded with hairs

+3mm.long. Lemmas exserted, oblong-lanceolate ; the lowest 7-8 mm. long, coriaceous,

glabrous and finely granular dorsally below the point of insertion of the awn ; lobes scarious,

4-6 mm. long. Callus about ? mm. long, bearded with relatively short hairs about 14 mm.

long. Awn inserted about 5 mm. from the base of the lemma; column 5-7 mm. long ;

bristle 10-12 mm. long. Palea not conspicuously ciliolate, 6 mm. long. Anthers 3-1 mm.

long, occasionally a floret with anthers up to 2 mm. long, always included and florets thus

very probably cleistogamous. Ovary pubescent from the middle, hispidulous at the apex.

Caryopsis 24 mm. long.

Care Province.—Uitenhage distr.: In somewhat moist places on the fields near the

Zwartkops River, Ecklon and Zeyher, 463! Zeyher, 4551! Bathurst distr.: Trappes Valley,

Dec., Daly, 639!. Mt. Currie distr. : Kokstad, Nov., Goossens, 323!, 339!,179!. Ingeli Moun-

tain, March, Tyson, 1270!. Umtata distr.: Bazeia, Nov., Bawr, 364!. Engcobo distr. :

Nqumakwe River, Jan., Flanagan, 2817!. Between Kngcobo and Nqumakwe River, Jan.,

Bolus, 10363!. Queenstown distr.: Katberg, Effingham, Dec., Galpin, 8398!. Queenstown,

Everett, 38!, 4!. Shiloh, Febr., Bawr, 779!. Reservoir east of Queenstown, Jan., Hilner, 311!.

Rocky banks of the Klipplaat River, 3,500 ft., Drége!. Molteno distr.: Broughton, Dec.,

Flanagan. 1673!. Molteno, June, Mogg, 2766!. Wittebergen, on Ben McDhwi, March,

Galpin, 6902 partim!. Aliwal North distr.: In a depression at Leeuwenspruit, between

Kraai River and the Wittebergen, Drége. 3918!. Without precise locality, Drége, 4250

(Herb. Nees)!.

BasutToLanp.—Leribe, 5,000-6,000 ft., Dveterlen, 400!, 753!. Febr!., Phillips, 6317!.

Mafeteng, Thaba Chicha Mountain, March, Dieterlen, 1275!.

Nata Province.—Pietermaritzburg distr.: Near Maritzburg, Dec., Medley-Wood

7228!. Klip River distr.: Umsinga, base of Biggar’s Berg, Buchanan, 100!. Umvoti

distr.: Greytown, Buchanan, 172!. Rietvlei, Buchanan, 156!. Weenen distr.: Culvers,

197

Dec., Rogers, 28309!. Estcourt distr.: Oct., Mogg, 3314!. Bergville distr.: Mount aux

Sources, Bayer and McClean, 272!. Tintwa Mountain, Strydhoek, Jan., Doidge, in Nat.

Herb., Pretoria, 20565!. Lions River distr.: Nottingham Road, Oct., Galpin, 10251!.

OraNGE Free Strate Province.—Ladybrand distr.: Pinekloof, Goossens, 1044!

Ficksburg distr. : Riverhill Farm, Jan., Potts, in Grey Un. Coll. Herb., 3689!, 3690!. Senekal

distr. : Senekal, Dec., Goossens, 815!, 821!, 949!. Bethlehem distr.: Stony veld near Beth-

lehem, Oct., Richardson, s.n!. Kroonstad distr.: Experimental Farm, Febr., Pont, 36).

Fauresmith distr.: Fauresmith, Henrici, 2310!.

TRANSVAAL PRovincE.—Potchefstroom distr. : Potchefstroom, Oct., Burt Davy, 5591!.

Theron, 6!. Wakkerstroom distr.: Vlakfontein, Burtt Davy, 4154!. Ermelo distr. : Nooit-

gedacht, Dec., Henrici, 1364! ; Burtt-Davy, 9064!. Bethal distr.: Leslie, Bell in Nat. Herb.

Pretoria, 20550!. Belfast distr. : Dullstroom, banks of Crocodile River, Dec., Galpin, 13009!.

Pretoria distr. : Wonderboompoort, Rehmann, 4493!. Division of Plant Industry Grounds,

Oct., Stent in Nat. Herb., Pretoria, 20525!. Benoni distr.: Benoni, plentiful near water,

Bradfield, T.258!. Johannesburg distr.: Wattles, near a marsh, Oct., Moss, 13586!.

Johannesburg, Oct., Rand, 920!.

The type specimen, Zeyher 463, is deposited in the Kew Herbarium.

In this species the ovaries are often infected by a species of Tilletia which may so alter

the character of the inflorescence, that specimens so affected appear at first sight to belong

to a different species. Among modern gatherings, Dieterlen 753 l|.c. represents such an

infected and malformed plant.

Trisetum imberbe cornutum Nees |.c. is furthermore such an infected plant, and since

the name of this “‘ species’ was based on a monstrosity, it 1s a “* nomen illegitimum ” and

for that reason must be rejected. It is pointed out elsewhere in this paper that Avena

antarctica Thunb. is a nomen dubium and for that reason is rejected. The only remaining

specific ephithet available to designate this species is therefore that of ‘* turgidula”’ derived

from Avena turgidula Stapf l.c. and the type of this species is naturally also Stapf’s plant,

viz. Zeyher 463!.

12. H. Dodii (Stapf) Schweickerdt comb. nov.

Syn.: Avenastrum Dodii Stapf, in Dyer, Fl. Cap. VII, 475 (1899).

Perennial. Culms erect, slender, about 100 cm. high, glabrous, smooth, 3-4—noded,

sheathed all along or nearly so, with 1-2 erect intravaginal branches from the lowest nodes.

Leaves 3 or fewer from near the base, and 3-4 higher up, distant ; sheaths not very tight ;

the upper rather loose, markedly striate, glabrous and smooth ; ligule oblong, up to 4 mm.

long ; blades linear; the lower tapering from a long attenuate base to a fine point, 30-50

cm. long and 3-5 mm. wide, flat or with involute margins, fairly rigid, more or less glaucous,

glabrous, smooth below, strongly striate and scabrid on the upper surface. Panicle con-

tracted, 20-30 em. long, narrow, dense or somewhat interrupted, slightly nodding ; rhachis

smooth ; branches fascicled, unequal, divided from the base or nearly so; longest up to

5 em., erect, scaberulous or smooth below. Spikelets about 12 mm. long, narrow, erect,

4-5-flowered. Glumes subequal, lanceolate, shortly aristulate, subhyaline; the lower

6-7 mm. long and l-nerved; the upper 7-9 mm. long and 3-nerved. Rhachilla-internodes

2mm. long, bearded upwards with hairs about 3-35 mm. long. Lemmas distinctly exserted,

lanceolate, 13-14 mm. long (including the lobes but excluding the awn), glabrous, light

green, rather firm, finely granular on the back ; lobes scarious, 7 mm. long, produced into

fine long bristles. Callus about 3 mm. long, short, bearded with hairs about 1-25 mm.

long. Awn inserted 5-6 mm. from the base of the lemma; column 5-7 mm. long; bristle

12-15 mm. long. Palea 5 mm. long, densely but not conspicuously ciliolate along the

keels. Anthers 2 mm. long. Ovary pubescent in upper half.

198

The type specimen Wolley-Dod 2775 is deposited in the Kew Herbarium.

Carg PRrovince.—Without precise locality, Lehmann (in Herb. Kunth)!. Cape distr. :

Wet slopes near Oatlands Point, Wolley-Dod, 2775!. By wet rocks, Hout Bay Fisheries.

Wolley-Dod, 3170!. Platkhp, near Capetown, along contour path, Nov., Andreae, 83,!

Rondebosch, University grounds, Noy., Levyns, 3672, 3565!.

VI—AIRA ANTARCTICA Forst. AND AVENA ANTARCTICA Tuuns.

During the investigation regarding the identity of Avena antarctica Thunb. it became

evident that a specimen of this species first described in Thunb. Prodr. Pl. Cap. 22 (1794)

is no longer deposited in Thunberg’s Herbarium, i.e. at the present time the type specimen

cannot be traced at Uppsala. All other efforts to trace the existence of a Thunberg speci-

men bearing that name in the Montin Herbarium, Bergius Herbarium (both at Stockholm),

the Fielding Herbarium at Oxford and the Banksian Herbarium in the Brit. Museum all

of which are known to contain a number of Thunberg plants, proved unsuccessful.

As a result of a request made to the authorities at Uppsala for the loan of the type of

Avena antarctica Thunb., the following sheets were received at Kew :—

(1) A sheet consisting of three culms and inflorescences of Bromus bifidus Thunb.

collected in Japan. This sheet bore the name Bromus bifidus in the lower right

hand corner and superimposed on this the name Avena antarctica. The specimen

to which these last two names referred has at some time or other been removed

from the sheet (traces of gum? can still be seen on the sheet). This may have

been Thunberg’s plant from the Cape which now cannot be traced. Apparently

Thunberg thus at some time or other tried to identify his Cape plant with the

Japanese Bromus bifidus, but later superimposed the name Avena antarctica to

replace the misidentification. This sheet furthermore bears the name Bromus

bifidus on the lower left hand corner and next to it the name Aira antarctica

Forst.

A sheet consisting of an inflorescence of Aira antarctica Forst. This specimen

agrees in every detail with Forster's co-type preserved in the Kew Herbarium

and which is a plant very different from any present day known species of

Helictotrichon from South Africa.

As a point of interest it should be mentioned that Thunberg visited England during

December, 1778—January, 1779. He met Forster who showed him the plants he collected

during Cook’s voyage round the world. Forster even gave Thunberg a fairly large number

of duplicates from his collection. These are now preserved in Thunberg’s Herbarium.

Thus sheet (2) mentioned above is probably an isotype of Aza antarctica Forst.

It may be assumed that Thunberg at some time removed the right hand specimen

from sheet (1) and remounted it on sheet (2). This would mean that the name Avena

antarctica Thunb. was based on a fragment of the isotype of Aira antarctica Forst., which

is a New Zealand plant. Accordingly this name is not applicable to a South African species.

It was cited in Thunb. Prodr. and Fl. Cap. merely as a result of an error. Since Thunberg

was acquainted with Forster, Forster's Prodr. and plants, it is highly improbable that he

would have applied the epithet “ antarctica”’ to a species from the Cape. [Although the

epithet was used by Linn. fil. to designate a Cape species, viz. Sewpus antarcticus, also

mentioned by Thunb. Prodr. Pl. Cap. (1794)]. It is thus possible that the name Avena

antarctica Thunb. found its way into South African literature by mistake and does not

199

refer to any species from the Cape but to Forster's New Zealand Aira antarctica. On the

other hand it is quite possible that Thunberg’s Avena antarctica was definitely a species

from the Cape and that both specimen and name had nothing to do with Forster’s Aira

antarctica. If this is assumed, the type of Avena antarctica Thunb. has probably been lost

since all attempts to trace its existence have failed. The absence of a type specimen is

in itself not a very serious matter if the descriptions given by Thunberg [Prodr. Pl. Cap. Lc.

or Fl. Cap. 436 (1818)] were adequate to identify a South African species by means of them.

But several Cape species of Helictotrichon are so closely allied that it is impossible to say

with certainty which of these Thunberg may have had before him at the time and to which

particular species the name could at the present time be applied.

A study of the literature regarding Aira antarctica Forst. and Avena antarctica Thunb.

has shown that these names have been the cause of some confusion.

For example Sprengel [Syst. Veg. 331 (1825)] under Danthonia antarctica cites among

others as synonyms “Avra antarctica Forst. and Avena Thunb.” Furthermore, Hooker

[Fl. New Zeal. 335 (1864)] quotes Danthonia antarctica Spreng. under T'risetum antarcticum

(Forst.) Trin. and consequently this name also includes the Cape species. Jwel [Pl. Thunb.

89 (1918)] cites “Avena antarctica (Forst.) Thunb. Prodr. 1794, 22; Fl. Cap. 1818, 436.

Siehe unter Bromus bifidus ” for the Cape species and thus assumes that Thunberg based

Avena antarctica on Forster's New Zealand species. Although this assumption may be

correct, stress must be laid on the fact that Thunberg nowhere indicated that his Avena

antarctica was actually based on Forster's plant. The combination Avena antarctica (Forst.)

Thunb. is thus not justified.

Desvauz [in G. Jay, Fl. Chilen. VI. 350 (1853)] suggests that the epithet “‘ antarcticum ”

should be retained for the New Zealand Trisetum antarcticum (Forst.) Trin. He creates

the name T'risetum Thunbergu for the species from the Cape to which Nees applied the

name Trisetum antarcticum based on Avena antarctica Thunb. Desvaua evidently realised

that the New Zealand plant had been confused with the species from South Africa. His

epithet ‘“‘ Thunbergil,” however, is superfluous, as older specific epithets are available for

the complex of species which Nees had placed under Tvisetum antarcticum (Thunb.) Nees.

Further information is to be found in studying the synonymy of the species enumerated

in this paper.

Since two names which probably refer to related but distinct plants bearing the same

specific epithet have been confused in the literature cited above and since in the absence

of the type specimen the identity of the Cape plant cannot be made out with certainty,

it is suggested that the name Avena antarctica Thunb. be regarded both as a ‘‘ nomen

ambigum” and a “nomen dubium”’ and consequently should be rejected. The name

Aira antarctica Forst., however, should be retained for the New Zealand species of which

the type and isotype specimens are extant!

VIT.— ACKNOWLEDGEMENTS.

For the kind loan of material I beg to tender my thanks to the Directors or Curators

of all the South African Herbaria and the following European Herbaria: Kew, British

Museum, Berlin-Dahlem, Oxford, Paris, Stockholm and Uppsala. My special thanks are

due to Sir Arthur Hill, Director of the Royal Botanic Gardens, Kew, for the facilities offered

during the preparation of this paper. To Mr. C. KE. Hubbard, F.L.8., I am very much

indebted for many kind suggestions and helpful advice.

STH eee ee ee we ee

200

VIII.— EXPLANATION OF FIGURES.

. quinquesetum (Steud.) Schweickerdt

. longum (Stapf) Schweickerdt

namaquense Schweickerdt

. barbatum (Nees) Schweickerdt

. leonimum (Steud.) Schweickerdt

. Galpinii Schweickerdt

. capense Schweickerdt

. hirtulum (Steud.) Schweickerdt

. natalense Schweickerdt

. longifoium (Nees) Schweickerdt

. turgidulum (Stapf) Schweickerdt

. Dodi (Stapf) Schweickerdt

The above figures show :—

1. Lower Glume.

2. Upper Glume.

3. Lemma.

4. Rhachilla-internode, anterior view (and lateral view).

202

Fig. VI.

Fig. V.

Fig. VIII.

203

Fig. XI. Fig. XII.

205

A REVISION OF THE SOUTH AFRICAN

SPECIES OF BRACHYLAENA R. Brown.

By HE. P. Puitiirs, M.A., D.Sc., and H. G. Scowerckerpr, B.Sc., Ph.D., F.L.S.

The first two species from South Africa described by Linnaeus and Lamarck respectively

were placed by these authors under the genus Baccharis Linn. In the modern conception,

this genus has no representatives in South Africa and is restricted to North and South

America. Thunberg was more correct, when in his Prodr. Pl. Cap., he described several

species under the genus Tarchonanthus Linn. Of these some were later referred by the

authors to the genus Brachylaena R. Br. and have since been found to be conspecific and

thus synonymous with the Linnaean and Lamarckian species. The species, however,

were satisfactorily grouped when R. Brown on basis of Baccharis nerifolia L. created the

genus Brachylaena, which differs in certain fundamental characters from the genus Bac-

charis L. It would be superfluous to enumerate these differences as they may readily be

seen from Benth. and Hook [Gen. Pl. II. 179 and 180 (1873]), where the former is placed

in the Inuloideae and the latter in the Asteroideae. Cassini, in Bull. Sciences. Philomat.

1817, p. 151, described the genus Oligocarpha on basis of Baccharis nertifolia L. His

account, however, appeared during September, 1817, whereas R. Brown’s description of

Brachylaena (with which it is congeneric as it was based on the same species) appeared to-

wards the middle of the same year (an exact date is not available). Brachylaena R. Br.

consequently has priority over Oligocarpha Cassin. Lessing in his Syn. Comp. 208 (1832)

correctly referred some of Thunberg’s species of Tarchonanthus to the genus Brachylaena,

and his conception has since been supported by De Candolle and Harvey as well as several

other authors. De Candolle’s account in Prodr. V. 430 (1836) is fairly comprehensive.

Investigation of the specimens cited by that author has shown that in one or two eases

the names of the species were misapplied. This is borne out more clearly by the synonomy

under the species enumerated in this paper.

The name Brachychlaena Post et Kuntze must be regarded merely as an erroneous

spelling for Brachylaena.

DISTRIBUTION.

The genus is a tropical one and the distribution of the species in South Africa, as with

species of other tropical genera, clearly indicates the common path of the migration of such

plants. In general the south western area of the Cape Province is invaded by migration

along the eastern coastal belt and from Humansdorp westward through the area lying

south of the Zwartbergen Range of mountains. In the South African flora it is not un-

common to find representatives of tropical genera as far south as Hast London or even

Bathurst where palms are found native. The two species B. elliptica and B. alicifolia show

a typical distribution in South Africa of species with tropical affinities, 1.e. they range from

Natal as far as Uitenhage. A similar distribution is shown by B. discolor but whose southern

limit is Bathurst. When the species mingle with the true Cape flora they can easily be mis-

taken for an integral part of the original flora unless the distribution of the genus is traced.

It may be significant that the two species found in the area of the Cape flora (B. nervifolia

and B. glabra) have quite glabrous leaves whereas the other South African species have

leaves which are tomentose beneath. B. neriwfolia ranges from Clanwilliam, southward

206

through Ceres and then in the southern districts as far as Humansdorp. It is also found

on the summit of the mountain ranges which carry a typical Cape flora and is a characteristic

element in the vegetation of the south-western districts of the Cape Province. B. glabra

has not become a typical “Cape” plant as B. neriifolia and though it has been recorded

from the Caledon district it still indicates its tropical affinities by having an eastern dis-

tribution as far as Natal. One species B. transvaalensis is confined to the mountainous

region of the north-eastern Transvaal and has probably entered from P.K. Africa as it is

also found at Lourenco Marques from where it has migrated southward into Zululand.

B. rotundata is confined to the Transvaal Highveld and Bushveld while another species

B. huillensis is only known from a single specimen collected in the Kruger National Park.

B. uniflora is confined to Natal.

BRACHYLAENA, #&. Br., in Trans. Linn. Soc. XII. 115 in not. (1817); DC., Prodr. V. 430

(1836); Harv., Gen. S.A. Pl. 169 (1838); Harv. et Sond., Fl. Cap. III. 115 (1865) ;

Benth. and Hook., Gen. Plant. II. 288 (1873); Engl. and Prantl, Pflanzenfam. IV. 5.

174 (1890).

Heads many to few-flowered, unisexual. Receptacle naked. Involucral scales imbri-

cate, dry, shorter than the florets; in fruit longer or shorter than the achenes. Corolla

tubular, unequally 5-lobed. Male flowers: bisexual; anthers tailed at the base, connate,

exserted; style filiform, simple or bifid; ovary hispid, abortive; pappus sparingly

developed. Female flowers: anthers abortive, separate; style bifid; achenes pubescent

or subglabrous; pappus of bristles in two rows.

Dioecious shrubs or trees. Leaves alternate, coriaceous, shortly petiolate or sub-

sessile, entire or toothed, glabrate above, often tomentose beneath. Capitula in racemes

or panicles. Flowers yellow. Geograph. distribution: Africa and Mascarene Islands.

Syn.: Oligocarpha Cass., in Bull. Soc. Philomatique, 1817, p. 151; Journ. de Physique

LXXXVITI. 26 (1818). Brachychlaena Post et Kunze, Leaxic. gen. Phan. 17

(1903).

KEY TO SPECIES.

Leaves glabrous beneath, very rarely rusty tomentose :

Leaves usually long-lanceolate, 5-8 times longer than broad........ 1. neriifolia.

Leaves usually elliptic or obovate, 2-4 times longer than broad..... 2. glabra.

Leaves white or greyish tomentose beneath :

Male heads I-3-flowered oo. ofc. cc os dancin © cranes sale eet 3. uniflora.

Male heads 7—50-flowered :

Leaves distinctly mucronate :

Involucre of bracts 3—4-seriate ; bracts densely albo-tomentose.. 4. hwillensis.

Involucre of bracts 5—8-seriate ; bracts never white tomentose.. 5. ilecfolia.

Leaves not mucronate :

Heads usually less than 1 cm. long; the imnermost bracts scarcely as

long as the fruits:

Leaves with petioles 1 or more em. long, generally over 2 cm. broad 6. transvaalensis.

Leaves sessile or subsessile, rarely up to 1-5 cm. broad......... 7. elluptiea.

Heads usually over 1 cm. long; the innermost bracts longer than the

fruits and hiding them:

Involucre of female heads 6—7-seriate ; style not swollen at the

base (a Transvaal and Rhodesian species)................:.. 8. rotundata.

Involucre of female heads 7—10-seriate ; style swollen at the base

(Cape Province and, Natal) eisai ied a ene 9. discolor.

207

1. B. neriifolia (L.), &. Brown, in Trans. Linn. Soc. XII. 115 in not. (1816) ; Steud., Nomencl.

I. 98 (1821); Less., Syn. Comp. 208 (1832); DC., Prodr. V. 430 (1836); Drége, Zwei

Pfizgeogr. Docum. 169 (1843); Krauss, in Flora. 1844. p. 671; Dvetrich, Fl. Univ. N.

Folge, t. 4 (1849) ; Harvey, Fl. Cap. III. 116 (1865); Bibl. Bot. X. No. 52. 18 (1901) ;

Bolus and Wolley-Dod, in Trans. 8. Afr. Phil. Soc. XIV. 3, 277 (1903) ; Sim, Forest Flor.

C.G.H. 246 (1907) ; Juel, Plant. Thunb. 381 (1918) ; Sim, Native Timb. 8. Afr. 44 (1921).

Syn.: Baccharis neriifolia Linn., Sp. Pl., 860 (1753) ; Willd., Sp. PL, III, ii, 1914

(1804); Pers., Syn., II, 423 (1807); Steud., Nomencl., I, 98 (1821);

Spreng., Syst. Veg., III, 462 (1826).

Tarchonanthus lanceolatus Thunb., Prodr. Pl. Cap., 145 (1794); Willd., Sp.

Pl., III, 1, 1793 (1804) ; Pers., Syn., II, 405 (1807) ; Steud., Nomencl., II,

826 (1821); Thunb., Fl. Cap. ed. Schult., 638 (1823); Spreng., Syst.

Veg., III, 456 (1826).

Oligocarpha neriifolia Cass., Dict. Sc. Nat., XXXVI, 21 (1817).

Conyza nervifolia L’Hér. ex Steud., Nomencl., I, 98 (1821).

Tarchonanthus dentatus Eckl. and Zeyh. ex DC., Prodr., V, 430 (1836), non

Thunb.

Branches grooved, glabrous or minutely puberulous. Leaves 2-11 cm. long, 0-5-1-8

em. broad (5-8 times longer than broad), usually lanceolate, more rarely lanceolate-linear

or oblanceolate, obtuse, with the midrib prominent or distinct beneath, and with close

reticulate veining, attenuated at the base into a short petiole, entire, very rarely one or two

toothed, glabrous. Inflorescence a terminal or axillary panicle, more rarely a raceme,

3-9 cm. long. Male heads: Involucre 2-5-seriate ; bracts 1-3 mm. long, 1-3 mm. broad,

ovate, more rarely ovate-elliptic or broadly elliptic, obtuse, usually with membranous

margins. Heads 7-14-flowered. Corolla-tube 2-3 mm. long, cylindric, glabrous; lobes

1-5-4 mm. long, linear, linear-lanceolate, obtuse or subobtuse, very rarely sparsely glan-

dular without. Filaments 1-2 mm. long, linear ; anthers 1-75-2-5 mm. long, linear, acute,

tailed at the base. Ovary 0-75-2 mm. long, pubescent, very rarely glabrous; style 4-7

mm. long, cylindric, sometimes bulbous at the base, 0-5-1 mm. long, linear, oblong, ovate

or ovate-lanceolate, obtuse or subacute. Pappus 3-4 mm. long. Female heads: In-

volucre 4-5-seriate. Bracts 2-4 mm. long, 1-3 mm. broad, ovate to lanceolate, obtuse,

rarely ciliate, glabrous, sometimes. with membranous margins. Heads 6-—11-flowered.

Corolla-tube 2:5-3-5 mm. long, cylindric, glabrous, rarely sparsely glandular ; lobes 1-2 mm.

long, linear, obtuse. Ovary 1-1-5 mm. long, oblong in outline, pubescent or glandular ;

style 4-5-6 mm. long, cylindric; lobes-5--75 mm. long, linear, oblong or linear-lanceolate,

acute or subobtuse. Pappus 3-5 mm. long.

Care Province.—Without locality: #.and Z.in Nat. Herb., 11934!; Schonland, 576!;

Thom, 950!; Nwen!; Wallich! Clanwilliam distr.: Wupperthal, Drege; Pakhuis Pass,

in very sandy spots under bushes and rocks, very rare, Leipoldt in Govt. Herb.,1905!; Rogers;

16836!. Ceres distr.: Mountains at Mitchell’s Pass, Febr., Schlechter, 9960!; Laaken Vlei,

at foot of Matroosberg, 3,500 ft., Phillips, 1926! and in Herb. Mus. Austro—Afric., 117261;

near Ceres, Thode, A2250!. Paarl distr.: Banks of upper Berg River, south of Roberts’

Vlei, shrub 4-10 ft., Pillans, 6754!; French Hoek Pass, on banks of River Zonder Kinde

Galpin, 12382!; river banks, Klemm Drakenstein Mountains, near farm Saleni, Galpin, 11044! ;

in respectibus Paarlberg, Drége!. Stellenbosch distr.: Stellenbosch, 2,000-4,000 ft., Ecklon,

and Zeyher!. Caledon distr.: Nieuwe Kloof, Houw Hoek Mountains, Burchell, 8080! ;

Steenbras River at Sir Lowry’s Pass, MacOwan, 184!; near Steenbras River, 900 ft., Bolus

in Natal Govt. Herb., 2559! and in Herb. Wood, 3802! and in Herb. Norm. Austro—Afric., 1841;

Hermanus, de Beer in Herb. Transv. Mus., 16529!; near Hermanus, Smuts, 1204!; Caledon,

Elbrecht in Herb. Transv. Mus., 22126!; Hottentots’ Holland Mountains, Zeyher!. Wor-

cester distr.: On mountains above Worcester, Rehmann, 2657; Hex River Valley, 2,000 ft.,

Tyson, 7571; near De Doorns, Bolus in Nat. Herb., 21079!; Du Toits Kloof, Marloth, 635! ;

208

12746!; Waterkloof, 20 miles south of Worcester, Andreae, 332!. Tulbagh distr.: Tulbagh

Waterfall, 600 ft., Febr., Schlechter, 7511!; Mund!. Wellington distr.: Baines Kloof,

Smith, 2686!. Swellendam distr.: Swellendam, 500-2,500 ft., Mund!; Kuntze!; Smith,

2733!; between Grootvaders Bosch and Zuurbraak, Burchell, 7260!; Zuurbraak, Thode,

A2332!. George distr.: George, Patterson, 1267!; in moist woods, Prior!; The Wilderness,

Febr., Moss, 5613!; Montagu Pass, Schweickerdt in Nat. Herb., 21080!. Knysna distr. :

Plettenberg Bay, Pappe!; Zeyher!; near Deepwalls, Phillips in Herb. Forest Dept., 55121;

5497!; Forest Station, Schonland, 3583!; Gouna River, Keet in Herb. Forest Dept., 2754;

Lily Vlei Forest, Keet, 712! and im Nat. Herb., 2108, 33221; Plettenberg Bay, Rogers, 26787 !.

Riversdale distr: Corente River, Muir, 214!; Glen, Muir in Nat. Herb., 3719!. Humans-

dorp distr.: Ratels Bosch, Zitzikamma, Fourcade, 576! ; north side of Kromme River, near

Wagenboom Station, Burchell, 4850!; Lottermg River, Zitzikamma, Galpin, 9489!, 4130!;

Witkliprivier, Marloth, 13064!; Storm’s River, Zahn in Herb. Forest Dept., 40491, 4095!.

Oudtshoorn distr.: Cango Valley, Marloth, 12130!. Prince Albert distr.: Zwartberg Pass,

near Kliphuis Vlei, Pocock in Nat. Herb., 21082!.

siete AR Reta 5

B. neriifolia (L.) R. Br.

{

———

209

Common in the Knysna district on banks of mountains and forest streams, up to 6 ft.

high and 1-2 in. stem diameter ; sometimes a tree 15-18-ft. high. Common name “ Water

Wit Els.”

The type specimens deposited in the Herb. Linn. Soc. Lond. and the Hort. Cliffort.

in Herb. Mus. Brit. consist of sterile material. The leaves and branches of these resemble

modern gatherings in every respect ; furthermore the characteristic rusty brown indumen-

tum on the young parts, leaves very little doubt as to the identity of the species.

2. B. glabra (L.f.) Druce, in Rep. Bot. Exch. Cl. Brit. Isles, 1916, p. 611 (1917).

Syn.: Tarchonanthus glaber Linn., f. Suppl., 360 (1781) ; Linn., f. Syst. Nat., ed. 13,

1204 (1791); Steud., Nomencl., II, 826 (1821); Thunb., Fl. Cap. ed.

Schultes, 638 (1823).

Tarchonanthus dentatus Thunb., Prodr. Pl. Cap., 145 (1794); Willd., Sp. PL.

III, ui, 1793 (1804); Pers., Syn., I, 405 (1807); Steud., Nomencl., II,

826 (1821); Thunb., Fl. Cap. ed. Schultes, 638 (1823).

Brachylaena grandifolia DC. Prodr., V, 430 (1836); Drege, Zwei Pfizgegor.

Doc., 135 (1843).

Brachylaena dentata (Thunb.) Less., Syn. Comp., 208 (1832) ; Harv., Fl. Cap.,

III, 116 (1865) ; Sim, Forest Flor. C.G.H., 246 (1907) ; Wood, Flor. Natal,

169 (1908) ; Juel, Plant. Thunb., 381 (1918); Sim, Native Timb. 8. Afr.,

44 (1921); Bews, Flor. Nat. and Zulul., 215 (1921); Henkel, Woody PI.

Nat. and Zulul., 72 (1934).

Branches grooved, minutely tomentose or glabrescent. Leaves distinctly petioled,

3-13-5 em. long, 1-2-4-5 cm. broad, elliptic-lanceolate, obovate, elliptic-oblong, rounded,

subacuminate or distinctly acuminate, obtuse, with the mid-rib and lateral veins distinct

beneath, narrowed at the base, entire or sinuate-toothed near the apex, glabrous or more

rarely rusty tomentose beneath ; petiole 1-1-5 cm. long, concave above, convex beneath.

Inflorescence a terminal or axillary panicle, 2-5-12 cm. long, rarely up to 15 cm. long. Male

heads: Involucre 3-6-seriate ; bracts 1-75-5 mm. long, 2-4 mm. broad, elliptic, ovate-

elliptic, ovate, obtuse, sometimes with membranous margins. Heads 13-24-flowered.

Corolla-tube 2-4-5 mm. long, cylindric, sometimes subcampanulate above, glabrous ; lobes

1-5-3 mm. long, linear, linear-lanceolate, obtuse or subobtuse. Filaments 1-2 mm. long,

linear; anthers 2-3 mm. long, linear, acute, tailed at the base. Ovary 1-5-2 mm. long,

oblong in outlime, pubescent, more rarely villous; style 5-7 mm. long, cylindric ; lobes

0-5-1 mm. long, linear, linear-lanceolate or ovate-lanceolate, obtuse or subobtuse. Pappus

4-4-5 mm. long. Female heads: Involucre about 6-seriate. Bracts 2-3 mm. long,

1-5-3 mm. broad with membranous edges; the outer ovate, obtuse ; the imner linear

to lanceolate. Heads 4-5-flowered. Corolla-tube 4 mm. long, cylindric ; lobes 1-25 mm.

long, linear-lanceolate, subobtuse. Ovary 3-25 mm. long, terete, linear in outline, shortly

villous; style 6 mm. long, cylindric ; lobes 1 mm. long, oblong-lnear, obtuse.

CarE Province.—Without precise locality: In Cap. b. spei circa Essenbosch rivum,

Nov., Thunberg (2 sheets in Herb. Thunberg, Uppsala)!. Caiedon distr.: Hermanus, Smuts

in Herb. Marloth, 11905!. Knysna distr.: Stinkhout Bosch, Sim and Newson in Herb.

Forest Dept., 2236!. Humansdorp distr.: Clarkson; Thode, A 860!; Storm’s River, Zahn

in Herb. Forest Dept., 4050; Kwaaibrand Forest, 700 ft., Burton im Herb. Forest Dept.,

40!; near the mouth of the Storm’s River, 200 ft., Fourcade, 537!; Stinkhout Bosch, about

2,000 ft., Ross in Herb. Forest Dept., 2061!; Sim and Hewson in Herb. Forest Dept., 22361;

Storm’s River, Keet, 543! and in Herb. Forest Dept., 3064!. Port Elizabeth distr. : Waterfall

at Van Staden’s River, MacOwan, 2076!; Van Staden’s Mountains, Zeyher, 2785!; Patterson,

879!: Longmore Forest Reserve, Long, 1051!. Alexandra distr.: Zuurbergen, 2,000—3,000

ities Drége!. Lusikisiki distr.: Egossa Forest, Sim, 2395!. Uitenhage distr. : Eastern slopes

Eland’s River Mountains, Sim in Herb Forest. Dept., 2102!.

210

Narat Provincr.— Without precise locality, Garrard, 1512!; edge of wood, Noodsberg,

Wood, 5274!, 4129! and im Natal Govt. Herb., 4731!.

Ross in Herb. Forest Dept., 2062 may be a water shoot of the species; if so then the

leaves are closely tomentose beneath.

Found in some quantity in Kwaaibrand Forest (Humansdorp distr.) which is almost

entirely on Table Mountain sandstone. Common name ‘“ Malbar.”

According to notes on various sheets this species is a tree of “ considerable height ”

(40-50 ft.) the stem of which may reach a diameter of 10-12 inches. It is a good waggon-

wood and is common in parts of the coast region, especially in open places along road-

margins in high forest (Keet).

MAP

2 OF

UNION OF SOUTH AFRICA ||.

RY OURS ‘ - % 3 f > 7 ® ie ee i

ear ;

set ie)

-GORDONIA

&. glabra (L. fil.) Druce.

211

8. B. uniflora Harv., Fl. Cap., II, 117 (1865); Wood, Flor. Natal, 170 (1908); Bews,

Flor. Nat. and Zulul., 215 (1921).

Type specimen deposited in Herb. Hort. Bot. Reg. Kew.

Branches slightly grooved, glabrous or the younger parts minutely tomentulose.

Leaves petioled, 6-12 cm. long, 1-5-4 em. broad, elliptic-lanceolate, obovate, obovate-

oblong to oblong-lanceolate, sometimes subacuminate or acuminate, rounded or obtuse at

the apex, narrowed at the base, entire or with crenate-serrated margins in the upper por-

tion, with the mid-rib and lateral veins prominent beneath; petiole 0-5-1-5 cm. long,

convex beneath, channelled above. Inflorescence a dense panicle, terminal or axillary,

5-8 em. long; the ultimate heads corymbose. Male heads: Involucre 5—6-seriate ; bracts

0. 5-2 mm. long, ovate to lanceolate or elliptic-lanceolate, obtuse, sometimes with mem-

branous margins. Heads 1—3-flowered more or less elongated, about 3 mm. long and 1 mm.

in diameter. Corolla-tube 2-3-5 mm. long, cylindric, sometimes subcampanulate above ;

lobes 1-2-5 mm. long, lanceolate, oblong-lanceolate or linear-lanceolate, obtuse or subacute.

Filaments 1 mm. long, linear: anthers 1-1:75 mm. long, linear, acute, tailed at the base.

ra 4 ” 4

rie

suareh pure

chug:

——

WAFEKING

wile oe

KEN MAROT

CARMARVON

GALWINRA

B. uniflora Harv.

212

Style 3-5-6 mm. long, cylindric ; lobes 0-5 mm. long, ovate, obtuse or subobtuse. Pappus

3-5 mm. long. Female heads: Involucre about 6-seriate ; bracts 1-25-3 mm. long, about

1 mm. broad, ovate to elliptic, obtuse, with membranous margins. Heads 4-flowered,

about 3-5 mm. long, and 1-5 mm. in diameter. Corolla-tube 2-5 mm. long, cylindric ;

lobes 0-5 mm. long, lanceolate, subacute. Ovary terete, pubescent; style 3 mm. long,

cylindric ; lobes 0-5 mm. long, ovate, subobtuse. Pappus 3 mm. long.

Nata Province.—Without precise locality: No collector, in Natal Herb., 778! ;

Gerrard, 29!; Gerrard and McKen, 1866!and in Natal Govt. Herb.,11006!. Port Shepstone

distr. : Amanzimtoti, Kotze, 432! and in Herb. Forest Dept., 6854!. Pinetown distr. : Durban,

Wood, 12670!; Sydenham, Wood, 12287!; Dumisa, Rudatis, 653!, 1052!. Inanda distr. :

Inanda, 1,800 ft., Wood, 585! and in Natal Govt. Herb., 2826!. New Hanover distr.: On

rocky hill, Great Noodsberg, Wood, 4129! and im Natal Herb., 4731!. Kranskop distr. :

Qudeni Forest, Davis, 82! and in Natal Govt. Herb., 8816!.

A common large tree in the Hlatikulu Forest, Zululand.

A species which was only imperfectly known at the time Vol. III of the Flora Capensis

was being compiled, but additional material collected since has enabled the authors of this

paper to draw up the accompanying description of both male and female plants.

According to Rudatis this species reaches a height of up to 20 metres.

4. B. huillensis O., Hoffm., in Engl. Jahrb., XXXII, 149 (1902).

A tree. Branches striate and grooved, glabrous in age; the young parts densely

minutely albo-tomentulose. Leaves petiolate; petiole 7-10 mm. long, channelled above,

convex below, densely albo-tomentulose ; lamina up to 7-5 cm. long and 2-0 cm. broad,

oblanceolate, conspicuously mucronate, acute, cuneate towards the base, and towards the

apex often somewhat undulate, glabrous, shiny and conspicuously veined above, densely

albo-tomentulose and inconspicuously vemed beneath. Inflorescence consisting of pendun-

culate glomerate capitula, about 15-20 mm. long; capitula at length sub-corymbose.

Male heads: Inwvolucre 3-4-seriate ; bracts 1-2-5 mm. long, ovate to ovate-oblong, sub-

acuminate or obtuse, densely albo-tomentose. Heads 7—-8-flowered, more or less globose,

3 mm. long and 2-5 mm. in diameter. Corolla-tube 2-0 mm. long, narrowly funnel-shaped,

glabrous ; lobes 2 mm. long, oblanceolate, acute or subobtuse. Filaments 0-6 mm. long ;

anthers 11-75 mm. long, linear, acute, tailed. Style 3-5 mm. long, cylindric; style-

branches 0-6 mm. long, subacuminate. Pappus setose ; bristles 2-5-3 mm. long.

TRANSVAAL Province.—Zoutpansberg distr.: Punda Maria, Kruger National Park,

Lang s.n., in Herb. Tvl. Mus., 32343!.

The above gathering consists of a male specimen only, and thus strictly speaking cannot

be compared with the type specimen of B. huillensis which is female. Our plant, however,

resembles the latter in characters of foliage and habit so closely, that we do not hesitate

in referring it to that species. Leemann 24 from the Blaauwberg, Pietersburg distr., which

is in leaf only is probably this species.

The description given above was drawn up solely from the Transvaal specimen cited.

This was done with a view to simplifying any questions regarding synonomy should our

plant eventually (after the female has become known from the Transvaal) prove to be a

species distinct from the Angolan B. hwillensis O. Hoftm.

5. B. ilicifolia (Lam.) Phillips et Schweickerdt, comb. nov.

Syn.: Baccharis ilicifolia Lam., Eneycl. Method. bot., 1, 345 (1783); Steud.,

Nomenel., I, 98 (1821).

Tarchonanthus racemosus Thunb., Prodr. Pl. Cap., 145 (1794); Thunb., FI.

Cap. ed. Schultes, 638 (1823); Sprengel, Syst. Veg., III, 456 (1826), sub

T. ellupticus Thunb.

213

Brachylaena racemosa (Thunb.) Less., Syn. Comp., 208 (1831); DC., Prodr.

V, 430 (1836) ; Harvey, Fl. Cap., III, 116 (1865)

; Sim, Forest Fl. C.G.H.,

246 (1907) ; Wood, FI. Natal, 170 (1908) ; Juel, Plant Thunb., 381 (1918) ;

Sim, Native Timb. 8. Afr., 44 (1921); Bews, Flor. Nat. and Zulul., 215

(1921) ; Henkel, Woody Pl. Nat. and Zulul., 72 (1934).

Brachylaena ellvptica DC., Prodr., V, 430 (1836), excl. syn., non Less.!; Drege,

Zwei Pfizgeogr.—Doc., 47 et 137 (1843).

The type specimen is deposited in Herb. Lamarck, Mus. Hist. Nat., Paris.

Branches grooved, minutely tomentulose or glabrous. Leaves subsessile or shortly

petioled, 1-4-5 em. long, 0-25-1 cm. broad, oblong, oblong-obovate, oblong-lanceolate to

linear, mucronate at the apex, rarely without a mucro, slightly narrowed to the base, with

the mid-rib prominent or distinct but lateral veins hidden, usually entire, sometimes with

a few teeth, glabrous above, albo-tomentose beneath ; petiole up to 2 mm. long.

Inflore-

scence axillary, few-headed, rarely of a solitary head, usually shorter than the leaves.

Male heads . Inwolucre about 5-seriate ; bracts ovate, ciliate. Heads 9-11-flowered, more

BE I a L

ay”

TOW elm soeno \

“

a garnet

vf BF}

wamalouacand

+ noe see etm

} \ # ipsa

u <' ay |

B. ilicifolia (Lam.) Phill. and Schw.

214

or less globose, about 3-5 mm. long, about 4 mm. in diameter above. Female heads: In-

volucre 5—8-seriate ; bracts 1-5-5 mm. long, ovate to linear-oblong, obtuse, ciliate. Heads

10-flowered, campanulate, 5-6 mm. long, 2-5 mm. in diameter above. Corolla-tube

4-5-5 mm. long, cylindric, glabrous; lobes 0-5-3 mm. long, linear or linear-lanceolate,

obtuse. Ovary 2 mm. long, somewhat angled, pubescent ; style 5 mm. long, cylindric ;

lobes 0-5 mm. long, linear, obtuse.

Care Province.—Witbout precise locality, de Mr. Sonnerat (Herbier de Lamarck) ;

in sylvula Kuka, Dec., Thunberg (2 sheets) !. British Kaffraria : Without precise locality,

Cooper, 408! and in Natal Govt. Herb., 8336!. Uitenhage distr.: Uitenhage, moist shady

places, Prior; near Uitenhage, Burchell, 4450!; woods by the Zwartskops River, Ecklon

and Zeyher, 481 !, ! 891!; Van istadent s River Momaterag, 1,000-3,000 ft., Wallich!; at Uiten-

hage, Bowve !. Port Elizabeth distr. : Fish River, Zeyher, 819! ; Redhouse, Patterson, 959! ;

Zwartkops, Marloth, 6110!. Albany distr. : Without precise locality, Prior !. Komgha

distr.: Among shrubs, near Komgha, 1,800 ft., Flanagan, 406!. Fort Beaufort distr. :

Kat River, 800-1,000 ft., Bawr, 1077!. Somerset East distr.: By the Little Fish River,

Burchell, 3267!; between Zuurberg Range and Klein Bruintjies Hoogte, 2,000-3,000 ft.,

Drege! ; between Little Fish River and Commandagga, Burchell, 3276 !. Queenstown distr. :

Between Shiloh and Table Mountain, stony and rocky places, 4,000 ft., Drége!; Junction

Farm, Galpin, 81261.

Natat Province.—Without precise locality, Gerrard, 1018!. Weenen distr.: Suther-

land !.

The intensely bitter leaves of this species are said to be a good remedy for diabetes.

Comparison of the types of Baccharis ilicifolia Lam., and Tarchonanthus racemosus

Thunb. has shown these two plants to be conspecific, and consequently Lamarck’s epithet

has priority.

6. B. transvaalensis Phillips and Schweickerdt, sp. nov.; B. Hutchins Hutch. affinis, sed

bractes capitult masculi minoribus et multiseriatis. distinguitur.

Arbor. Ramzi striati, minute tomentulosi, vel glabri. Folia petiolata, 3-5-16-5 cm.

longa, 1-2-3-7 cm. lata, lanceolata, elliptico-lanceolata vel obovato-lanceolata, nonnun-

quam breviter acuminata, rotundata vel obtusa, perraro acuta, integerrima vel margine

dentata et undulata, supra glabra, subtus albo-tomentosa ; petiolus 1-1-2 cm. longus,

supra canaliculatus, subtus convexus. Inflorescentia e paniculis terminalibus vel axillaribus

sistens, 4-5-16 em. longa; capitula deinde subcorymbosa. Capitulum masculinum : In-

volerum 4-6-seriatum ; bracteae 1-2 mm. longae, ovatae vel ovato-lanceolatae vel lanceo-

latae, obtusae, plus minusve tomentosae, marginibus membranaceae vel ciliatae. Capitula

11-16-flora, plus minusve globosa, 2 mm. longa et 2-5 mm. diametro. Corollae tubus

3-3°5 mm. longus, cylindricus, glaber; lobi 2 mm. longi, lanceolati, obtusi vel subacuti.

Filamenta 0-5-1 mm. longa, linearia ; antherae 1-5-2 mm. longae, lineares, acutae, caudatae.

Stylus 5-5-6 mm. longus, cylindricus ; lobi 0-5 mm. longi, ovati, nonnunquam subacumi-

nati, acuti et obtusi. Capitulum feomineum : Involucrum 6-seriatum ; bracteae 1-5-4 mm.

longae, ovatae vel ovato- -ellipticae, obtusae vel subacutae, glabrae vel sparse tomentosae.

Ca pitula 8-flora, ambitu plus minusve obovata, circiter 3-5 mm. longa et 3 mm. diametro.

Corollae tubus 3-5 mm. longus, angulatus ; lobi 0-5 mm. longi, ovati, sub-acuti. Ovariwm

3°5 mm. longum, pubescens, teres ; stylus 5 mm. Jongus, cylindricus, basin versus incras-

satus ; lobi 0-75 mm. longi, ovati-lanceolati, subobtusi.

TRANSVAAL PRovince.—Pietersburg distr. : Forest between Woodbush and Haenerts-

burg, 4,000-6,000 ft., Hutchins (type! deposited in Herb. Hort. Bot. Reg. Kew); O'Connor

in Herb. Forest Dept., 1471!, 3559! ; Woodbush, Hoffman, 79! ; no collector in Herb. Transv.

Mus., 9703! ; Tzaneen, Pigeon Hole Farm, McCallum, 1532! ; Woodbush, Grenfell in Col.

Herb., 1105! ; Botha in Herb. Forest Dept., 3559!, 5286! ; no collector, 1 Herb. Forest Dept.,

215

4974! ; Middelkop Plantation, Keet, 1189!. Lydenburg distr.: On farm Hebron, Ketze in

Herb. Forest Dept., 2835!. Pilgrims Rest distr. : Sabihoek, Oranje, 2! ; Lothian, Keet, 1130! ;

Graskop, Hvans in Herb. Forest Dept., 5398! ; Joubert in Herb. Forest Dept., 8724!. Barber-

ton distr.: Wooded ravines, Rimer’s Creek, 3,000—4,000 ft., Galpin, 451!; Thorneroft in

Herb. Transv. Mus., 11174!, 2773! ; Stentor, Munro, P.S. 35!.

Natat Province.—Pinetown distr.: Stella Bush, near Durban, Tyrrell in Natal

Herb., 21310!; Amanzimtoti, Gerstner in Natal Herb., 22079!. Zululand: Negomi Forest,

Tustin in Herb. Forest Dept., 3555!, 3556!; N’Kanghla Forest, Forbes, 766!; Hlatikulu

Forest, Boocock, 26! and in Herb. Forest Dept., 5323!.

P.E. Arrica.—Lourenco Marques, on beach, Munro, P.S. 166! ; Maputaland, Maputa-

land Exped., in Herb. Transv. Mus., 14354!.

Jonna

marae

f i 4 e

ae ae x oe

Ely g

Alt 3

sf Taree a ; | ay oe eri

TWA PEKING

vayeURD

mam slQUaran®

——— | ee.

wiLuisTont

CALVERT A

pbar

B. transvaadlensis Phill. and Schw.

A specimen (Hubbard in Herb. Forest Dept., 3668 and in Nat. Herb., 21083) collected

on Meintjies Kop, Pretoria, is stated by the collector to be probably a cultivated specimen

from the northern Transvaal. Burtt Davy, 1521, Sabie Hoek Forest, Lydenburg distr. ;

216

Burtt Davy, 1171, Potato Bush, Zoutpansberg distr.; Burtt Davy, 1413, near Pilgrim’s

Rest ; Burtt Davy, 2692, Madjadjes Mountains, Zoutpansberg distr., all without flowers are

probably water shoots of this species.

A large tree up to 70 ft. high and a bole up to 3 ft. in diameter but seldom sound when

so large. Wood used a great deal by natives for making assegai handles. Found through-

out the Ngomi Forest, Zululand, up to 4,000 ft. altitude. In the Transvaal found asso-

ciated with dense forest at all elevations from 3,000 ft. upwards. Mostly a crooked much-

branched tree growing along the edge of the forest but frequently inside when it makes a

fine straight bole. Timber strong elastic and durable in the ground. Common name

“Vaalbos”’; Sesuto name is “ M’pata.”

[Obermeyer, Schweickerdt and Verdoorn 349 from the southern slcpes of the Zoutpans-

berg is probably a female plant of the above species, but since no female specimens have

been seen in the Kew Herbarium, there is some uncertainty with regard to the identity of

this sheet.] The specific epithet was given by Dr. J. Hutchinson but no description

published.

7. B. elliptica (Thunb.) Less., Syn. Comp., 208 (1832) ; Harvey, Fl. Cap., III, 116 (1865) ;

Sim, Forest Flor. C.G.H., 246 t. 92 (1907); Wood, Pl. Natal, 169 (1908); Jwel, Plant.

Thunb., 381 (1918) ; Sim, Native Timb. 8. Afr., 44 (1921); Bews, Flor. Nat. and Zulul.,

215, (1921); Henkel, Woody Pl. Nat. and Zulul., 72 (1934).

Syn.: TLarchonanthus ellipticus Thunb., Prodr. Pl. Cap., 145 (1794) ; Willd., Sp. Pl.,

III, in, 1793 (1804); Pers., Syn., II, 405 (1807); Steud., Nomencl., II,

826 (1821); Thunb., Fl. Cap. ed. Schult., 638 (1823); Spreng., Syst.

Veg., III, 456 (1826), excl. syn.

Brachylaena dentata DC., Prodr., V, 430 (1836), incl. syn., non Less. ; Drége,

Zwei Pfizgeogr.—Docum., 169 (1843).

Brachylaena dentata var. B salicina DC., Prodr., V, 430 (1836).

Type specimen deposited in Herb. Thunberg, Uppsala.

Branches grooved minutely tomentulose. Leaves sessile or shortly petioled, 1-5—10

em. long; 0-5-3 em. broad, linear, linear-lanceolate, lanceolate, oblanceolate, elliptic-

lanceolate or obovate, obtuse, narrowed at the base, entire or the margins somewhat ser-

rate, toothed or irregularly lobed above, glabrous above, white tomentose beneath, with

the mid-rib and lateral veins prominent or distinct. Inflorescence a dense terminal or

axillary panicle or raceme, leafy, mostly compact but sometimes each axillary inflorescence

distinct. Male heads: Involucre 2-5-seriate; bracts 1-3 mm. long, ovate, sometimes

with membranous margins, sometimes ciliate or margins shortly lacerate. Heads 7-11-

flowered, campanulate, about. 1-5 mm. long, 1-5 mm. in diameter above. Filaments

0-75-1-5 mm. long, rarely 3 mm. long, linear; anthers 1-2-5 mm. long, linear, acute,

tailed at the base. Style 3-7 mm. long, cylindric, lobes 0:5 mm. long, ovate or lanceolate,

usually acuminate, acute or subacute. Pappus 1-75-3 mm. long, rarely 4 mm. long, usually

rough, sometimes barbellate near the tips. Female head: Involucre 3-5-seriate ; bracts

1-2 mm. long, ovate, obtuse, usually with membranous margins, usually ciliate. Heads

4-7-flowered, campanulate, about 2-5-3 mm. long, 2 mm. in diameter above. Corolla-

tube 2-35-3 mm. long, cylindric, glabrous ; lobes 0-5 mm. long, oblong or linear, subobtuse.

Ovary 2-5 mm. long, glabrous or pubescent. Achenes 4 mm. long, obscurely angled, pube-

scent. Pappus 3-3-5 mm. long.

Carpe Proyince.—Without locality, Watt and Brandwyk, 1808!. Uitenhage distr. ;

Without precise locality, MacOwan!; between Enon and Zuurberg Range, Hoffmanns

Kloof, 1,000-2,000 ft. Drege!; Zuurberg Pass, 1,800 ft., Long, 707!. Bathurst’ distr. :

Kowie West, on bush slopes, Tyson in Herb. Transv. Mus., 1725! and in Herb. Mus. Austro-

Afric., 13364! and in Govt. Herb., 12625! and in Herb. Marloth, 8876! ; Bell’s Beach, Kowie,

217

Britten, 2095!; Port Alfred, Rogers, 16601!. Albany distr.: Grahamstown, Schlechter,

2652! and in Herb. Transv. Mus., 21396! ; Broekhuizen’s Poort, South, 670!, Cooper, 1563!

and im Natal Govt. Herb., 8293! ; Zeyher, 2736! ; MacOwen, 244! and in Natal Govt. Herb.,

252!, 11005! ; Brakkloof, White, 1172!; near Riebeek Hast, between Zwartwaterpoort and

the Zwartwaterberg, Burchell, 3456! ; Zwartwaterpoort, Burchell, 3375!, 3361!, 3400! ; near

Grahamstown, Dyer, 1340!._ Bedford distr.: Kagaberg, Scott-Elliott in Herb. Galpin, 101).

Adelaide distr.: Without precise locality, Marloth, 5356!; Watt and Brandwyk, 1286! ;

summit Mungo Mountain, Galpin, 11544!. Kingwilliamstown distr.: Evelyn Valley, Perie

Mountains, Scott-Elliott, 995!. Stockenstroom distr.: Katberg, Stayner, 99!. Stutterheim

distr.: Fort Cunningham, 3,500 ft., Sim, 2107!. East London distr.: East London,

Rattray, 70!; Watt and Brandwyk in Nat. Herb., 6059! ; Dowling, 16!; Munro in Nat.

Herb., 3333!. Komgha distr.: Near Kei mouth, Flanagan, 194!, 241!. Kentani distr. :

In woods near Kentani, Pegler, 997!, 1521!. Maanduli distr.: Banks of Umtata River,

Drége. Port St. John’s distr.: Between Umtata River and St. John’s River, Drége! ;

Pungwane Forest, Boshoff in Herb. Forest Dept., 3474! ; Mtambalala Ridge, Fegen in Herb.

Forest Dept., 3340).

B. elliptica (Thunb.) Less.

218

Nata PRovrncr.—Without precise locality, Cooper, 1141!. Umzinto distr.: Near

Umzumbi, Wood, 3052! and in Natal Govt. Herb., 252! ; sunny slopes Ifafa Valley, Rudatis,

981!; Umkomaas, Wood, 4609! and im Natal Govt. Herb., 6411!; Impambantoni Valley,

King, 97!. Pietermaritzburg distr.: Without precise locality, Rehmann, 7595! ; Thorny-

bush Junction, Marriott in Natal Herb., 24335!. Pinetown distr.: Near Durban, Wood,

10977!. Inanda distr. : Umgeni, Wood, 8434! and in Natal Govt. Herb., 9133!; Verulam,

Wood, 936!. Umyoti distr. : Near Greytown, Wood in Natal Govt. Herb., 8893!. Weenen

distr.: Hulls above Weenen, Wood, 4445! and in Natal Govt. Herb., 5160!. Lion’s River

distr.: Howick, Mogg in Nat. Herb., 21078!. Ubombo distr.: Zululand, Mfongosi River,

Watt and Brandwyk, 1014!, 1209!.

A specimen, Zeyher 271 labelled as from the Kat River, near Philipstown is evidently

incorrect as no species of Brachylaena occurs so far inland.

Shrub 8 ft. high or a small tree found in the Karroid-scrub veld of the Albany district ;

eaten by stock (R. A. Dyer). Bush up to 15 ft. high in Natal. Native name “ Isiduli-

we-hlati.”

According to various authors and collectors this species is a shrub or small tree up to

12 ft. high. In leaf-shape it is one of the most variable species of the genus; the floral

characters, however, appear to be fairly constant throughout the wide range of material

seen by the authors of this paper.

8. B. rotundata 8. Moore, in Journ. Bot., 1903, p. 131; Burtt Davy and Pott, in Ann. Tvl.

Mus., III, m, 168 (1912); Sim, Native Timb. 8. Afr., 44 (1921).

The type specimen is deposited in Herb. Mus. Brit.

Branches grooved, shortly but densely albo-tomentose, rarely tomentulose. Leaves

petioled or more rarely subsessile, 2-5-10 cm. long, 1-5-2 cm. broad, mostly elliptic, more

rarely elliptic-lanceolate, obtuse, usually rounded at the base, more rarely slightly narrowed,

with entire or irregularly-toothed margins, with the mid-rib and lateral veins prominent

beneath, glabrous above, albo-tomentose beneath ; petiole 5 mm. long. Inflorescence a dense

terminal panicle with the ultimate heads sometimes cymose. Male heads: Involucre 4-10-

seriate ; bracts 1-5-5-5 mm. long; the outer ovate, obtuse, with membranous margins,

sometimes shortly ciliate and glandular on the lower half; the mner ovate to linear, with

membranous margins, sometimes acuminate, glabrous or glandular on the lower half.

Heads 13-31-flowered. Pappus 4-5 mm. long. Corolla-tube 3-4-5 mm. long, cylindric,

glabrous, sometimes sparsely glandular ; lobes 0-5-3 mm. long, linear or lanceolate, obtuse

or acute, sparsely glandular without. Filaments 1-3 mm. long, lnear; anthers 1-3 mm.

long, lnear, sometimes acuminate, acute, tailed at the base. Female heads: Involucre

6—T-seriate ; bracts 2-5 mm. long, ovate, ovate-lanceolate, lmear-lanceolate to linear,

obtuse, usually with membranous margins and often glandular on the lower half. Pappus

5-6-5 mm. long. Corrolla-twbe 4-5-5 mm. long, cylindric, sometimes sparsely glandular ;

lobes 0-5-0-75 mm. long, ovate-lanceolate to lanceolate, obtuse. Ovary 2-4 mm. long,

linear in outline, pubescent or shortly villous; style 4:5-5-5 mm. long, cylindric ; lobes

0-5-1 mm. long, elliptic to lanceolate, obtuse.

TRANSVAAL PRoyincr.—Witwatersrand distr.: Jeppe’s Ridge, Johannesburg, ul-

allan in Herb. Galpin, 6018!, 6019! ; near Johannesburg, Adlam in Natal Govt. Herb., 7181!

and in Herb. Wood, 5633! ; Johannesburg, Conrath, 359! ; Gilfillan in Herb. Galpin, 6125! ;

Moss, 2505!, 2554!, 2261! ; Rand, 738! (type). Pretoria distr. ; Pretoria, Meintjes Kop, Pole

Evans, 474! ; Pretoria Koppies, Leendertz, 235! and in Herb. Transv. Mus., 8702! and in Natal

Govt. Herb., 10507! ; Rogers, 232!; Aapies Poort, Rehmann, 4075! ; Wonderboom Poort,

Mogg in Herb. Forest Dept., 1669!; Onderstepoort, Mogg in Govt. Herb., 15688!; near

Pretoria, Goossens, 81! ; Mogg, 15165!, 15309!, 15222!, 15170!, 15088! ; Smth, 5!; Verdoorn,

456! ; Repton, 18! ; Munro in Nat. Herb., 3331!, 1747! ; Pole Evans, 150! ; Hartebeestpoort,

219

Keet, 1270!. Magaliesberg, Leemann in Nat. Herb., 21116!. Heidelberg distr. : Bonsma in

Herb. Transv. Mus., 13139!. Rustenburg distr.: Rustenburg, Collins in Herb. Transv.

Mus., 11966! ; Leendertz in Herb. Transv. Mus., 9898!; Buffels Poort, Twrner, 35!. Pot-

gietersrust distr.: Potgietersrust, Pyramid Estate, Galpin, 8818!; near Potgietersrust,

Burtt Davy, 2273!. Middelburg distr.: Near Mission Station, Botsabelo, on Little Olifants

River, Marloth, 11743!, 11755!.

Also occurs in Rhodesia.

Note.—Burtt Davy, 2273 from near Potgietersrust; Burtt Davy, 2448 and Mogg,

15048 from Pretoria; Murray, 672 from Heidelberg probably belong to this species.

B. rotundata Sp. Moore.

9. B. discolor DC., Prodr., V, 430 (1836); Drege, Zwei Pflzgeogr.—Docum., 155 and 157

(1843) ; Harv., Fl. Cap., III, 117 (1865); Wood and Evans, Natal Pl., I, t. 23 and 24

(1898) ; Sim, Forest Flor. C.G.H., 247 (1907); Wood, Flor. Natal, 169 (1908); Sm,

Forest Flor. Portug. EH. Afr., t. 73 (1909) ; Sim, Native Timb. 8. Afr., 44 (1921); Bews,

Flor. Nat. and Zulul., 215 (1921); Henkel, Woody Pl. Nat. and Zulul., 72 (1934).

220

Syn.: Brachylaena natalensis Sch. Bip., im Walp. Rep., II, 972 (1843); Krauss, m

Flora, 1844, p. 671; Harvey, in Fl. Cap., III, 117 (1865).

Branches grooved, minutely tomentulose or subglabrous. Leaves petioled, 2-5-18 cm.

long (but usually 6-10 cm. long), 1-7 cm. broad, elliptic-oblong, obovate-lanceolate or

oblong-lanceolate, obtuse, narrowed at the base into the petiole, entire or remotely toothed,

with the mid-rib and lateral veins prominent beneath, glabrous above, white tomentose

beneath ; petiole about 1 cm. long, convex beneath, scarcely channelled above. Inflores-

cence an axillary or terminal raceme or panicle, with the ultimate heads sometimes cymose.

Male heads : Involucre 5-9-seriate ; bracts 1-4-5 mm. long, ovate, obtuse, ciliate, sometimes

woolly ; the inner ovate, lanceolate to linear, usually woolly-ciliate. Heads 11—50-flowered.

Corolla-tube 3-5 mm. long, cylindric, sometimes gradually narrowing from the base up-

wards, glabrous; lobes*2 mm. long, lanceolate to lmear, usually obtuse, sometimes sub-

acuminate and subacute. Filaments 1-2-5 mm. long, linear; anthers 1-5-2-5 mm. long,

linear, acute, tailed at the base. Style 4-5-8-5 mm. long, cylindric; lobes 0-5-1 mm.

long, ovate to lanceolate, subacute. Pappus 4-5 mm. long. Female heads: Involucre

7-10-seriate. Bracts 3-5 mm. long, ovate, obtuse, usually woolly; the imner long and

lanceolate to linear. Heads 11—26-flowered. Corolla-tube 5-7-5 mm. long, cylindric,

usually widened at the base and sometimes 5-angled or with 5 veins; lobes 0-5 mm. long,

ovate to linear, obtuse, more rarely subacute. Pappus 7-8 mm. long. Ovary 2:5-4-5 mm.

long, linear in outline, grooved or sometimes angled, sparsely pubescent, style 6-5-10 mm.

long, cylindric, swollen at the base; lobes 0-5 mm. long, ovate to linear, obtuse. Young

fruits glabrescent.

Care Province.—Without precise locality, Ecklon, 269!. Somerset Kast distr. : With-

out precise locality, Bowker!. Bathurst distr.: Port Alfred, Burchell, 2823! ; Tyson, 54! ;

Marloth, 11998! ; Schonland, 3304!; Tyson in Govt. Herb., 12571! and in Herb. Marloth,

8575! ; Britten, 773!, 1837! ; at the mouth of the Great Fish River, Burchell, 3751!. Hast

London distr.: East London, Gane, 304!; Munro in Nat. Herb., 3332!. Komgha distr. :

Among shrubs near Kei Mouth, Flanagan, 860!. Ngqgeleni distr.: Bush on sand dunes,

Fegen in Herb. Forest Dept., 2060!. Port St. John’s distr.: Port St. John’s, Doran in Herb.

Forest Dept., 1980!. Bizana distr.: Between Umtentu and Umzimkulu River, under

500 ft., Drege.

Natrat Province.—Without precise locality, Gerrard and McKen, 348!, 1017! and in

Natal Govt. Herb., 251!, 802! ; Cooper, 1240! and in Natal Govt. Herb., 8311!. Port Shepstone

distr.: Port Shepstone, Shelly Beach , Letty, 222!; Burtt Davy, 2392!. Pinetown distr. :

Salisbury Island, Durban, Forbes, 230! ; Durban, Wood, 5!, 4907!, 12670! and in Herb. Transv.

Mus., 15417!; Krauss, 243!; in coastal bush, Schlechter, 2886! ; on dunes, Rudatis, 1089! ;

in dune scrub, Moss, 2502!, 1503! ; Marloth, 4174! ; Umbogintwini, Salter, 382/7! ; Illovo

Beach, Hubbard in Herb. Forest Dept., 6103!; north end of Berea Ridge, Galpin, 12114! ;

Amanzimtoti, Kotze, 450! and in Herb. Forest Dept., 6873!; Isipingo, Forbes and Ober-

meyer, 28! ; Hutchinson, Forbes and McClean, 6!. Umzinto distr. : Without precise locality,

Rudatis, 673!. Zululand: Dukuduku Forest, Kotze in Herb. Forest Dept., 6533! ; Fair in

Herb. Forest Dept., 8134! ; northern Zululand, Kotze in Herb. Forest Dept., 3516!. Mtunzini

distr. : No locality, Thode, A1532!.

P.E. Arrica.—Lourenco Marques, in coastal bush, Bolus, 1173!; Monteiro, 36! ;

Forbes! ; Borle, 6!; Inyack Island, Breyer in Herb. Transv. Mus., 20442!.

Norr.—The record given by Burtt Davy and Pott (Ann. Transv. Mus., vol. 3, p. 168)

is an error. It is most probably B. rotundata Sp. Moore.

Common in coastal bush round Durban. At the Blue Lagoon it is grown as a hedge

(Phillips).

221

ae OF SOUTH | :

; an

< |

}

We oes

Aes

i Ts

/ {s

aS Barrie tt

oe ea pe

(

Ieee Alt Wes ,

ah x 7 Ps

x 7 acanor | ‘

{ ea

piety

a *

Se Sra)

jeenmanor

Wawalauscane

Cae eae

CARNARVON

— FRABERAURG |

i ;

ee aa ep ee lpr amen

B. discolor DC.

ACKNOWLEDGEMENTS.

The authors wish to express their gratitude to the Directors and Curators of the

various herbaria for the loan of type specimens. Special thanks are due to Sir Arthur Hill,

Director of the Royal Botanic Gardens, Kew, who gave the facilities afforded for this paper

to be completed at the Herbarium, Kew.

. . Vow hf Sa

ee tse]

a ee ee

: a Petts ee FA SL, a

Pete,

Pty en

223

AN ENUMERATION OF PLANTS COLLEC-

TED IN THE NORTHERN TRANSVAAL.

By A. A. OBERMEYER, M.Sc., H. G. Scuweickerpr, Ph.D. and I. C. Verpoorn.

Tue following list is a classified account of the names of ferns and flowering plants col-

lected during two excursions to the farm “ Zoutpan”’ at the western extremity of the

Zoutpansberg range. Most of the specimens were collected around the salt pan on this

farm or in the vicinity of it, such as on the slopes of the mountain on the eastern margin

of the pan and the sandy flats stretching to the west of it. Some of the plants listed, how-

ever, were collected on the way to and from this area or on short excursions from the pan

to neighbouring places.

A paper by H. G. Schweickerdt [S.A. Journal of Science, 30, 270 (1933)] gives an

account, accompanied by a map, of the vegetation of the area surrounding this salt pan.

Reference should be made to it when studying this list. Further a paper by Dr. C. E. B.

Bremekamp “New and otherwise Noteworthy Plants from the Northern Transvaal”

{Annals of the Transvaal Museum, 15, 11 (1933)] covers more or less the same ground but

includes specimens collected during a general tour.

Our first excursion was undertaken in November, 1932, when 438 specimens were

collected and numbered “‘ Obermeyer, Schweickerdt and Verdoorn | to 438.” The route

taken on this occasion was from Pretoria to Pietersburg and thence via Kalkbank and Vivo

to the farm Zoutpan. Camp was pitched just behind the homestead under Lonchocarpus

trees and overlooking the salt pan. During 10 days spent there two short trips were made,

one westward towards the Blaauwberg to the farm Eyem and the other via Booysen’s farm

Chapudi to Waterpoort. Otherwise the time was spent surveying the area around the pan.

The route taken on the return journey was via Wyle’s Poort and Louis Trichardt.

The second excursion made in April, 1934, was undertaken by Miss Verdoorn and

Dr. Schweickerdt (Miss Obermeyer being in Europe at the time). While camping at the

homestead for five days, further collections were made in the same area and one short trip

was made to Duvenhage’s Pan near Amisfort to collect water plants. On this second

excursion 252 specimens were collected and numbered “ Schweickerdt and Verdoorn 440

to 691.” Consecutive numbers were used for the plants collected on the two excursions

to avoid confusion. The route via Louis Trichardt and Wylie’s Poort was taken both

going and returning.

The farm Zoutpan appears to be the meeting place of at least four distinct types of

vegetation, the Eastern Mountain Flora reaching to the very margin of the pan on the

east, the Kalahari-desert Flora from the west, the Bushveld from the south and the sub-

tropical plants from the north. For this reason a botanical survey of the area seemed to

hold out prospects of interesting discoveries. We consider this assumption has been

justified and hope that the following list and notes of our identified plants may help future

workers in the study of plant geography in Southern Africa.

The specimens were first examined at the National Herbarium and Transvaal Museum,

Pretoria, by the three authors and later were sent to Kew where Dr. H. G. Schweickerdt

had been temporarily transferred. There he verified or corrected the identifications and

his notes on some of the specimens in this list are initialed since they were the result of

research done in the absence of the collaborators.

224

Subtropical Plonls

Kolohori’ Desert

Feng

Nylotroome’

Map to illustrate paper.

“An enumeration of plants collected in the Northern Transvaal.”

The families and genera are arranged according to Della Torre and Harms and the

species alphabetically. Specimens have been deposited in the National Herbarium,

Pretoria, and in the Herbarium of the Transvaal Museum. Many duplicates have been

donated to Kew and several to the Botanical Museum in Berlin-Dahlem and the Hofmuseum

in Vienna.

The authors are erateful for the facilities granted by the Chief and the Principal

Botanist of the Division of Plant Industry and the Director of the Transvaal Museum,

which made this undertaking possible. We are also indebted to the Director, Royal Botanic

Gardens, Kew, for the use “at the herbar1um and library and to several members of the

staft of that Institution for assistance with the identifications ; to Mr. J. Ramsbottom and

staft of the British Museum and to Prof. Diels and staft of the Botanische Museum, Berlin—

Dahlem, for similar help.

225

FILICKS.

1. Marsilia ephippiocarpa Alston.

Along margin of Duvenhage’s Pan near Amisfort, 628.

2. Dryopteris thelypteris (L) A. Gray.

In vlei at foot of mountain behind homestead, 218.

D. Pentheri (Krass.) C. Christensen ; [D. elongata (Sw.) Sim non O. Kuntze. ]

Along margin of stream in kloof near homestead, 240.

MONOCOTYLEDONS.

8. APONOGETONACEAE.

1. Apenogeton Rehmannii Oliver.

In semi-permanent pan on farm “ Eyem,” north of the Blaauwberg, 84.

A. gracilis Schinz.

In rock-pools on first plateau of northern slopes of the mountain, 317 a. [This species

was collected only once before in the Transvaal, viz. in the eastern mountain area (Hout-

bosch). }

A. Holubii Oliv. forma.

Near Amisfort in Duvenhage’s Pan; flowers deep yellow, 625. [This form is closely

allied to, if not actually A. Holubw Oliv., in which the leaves are more acute and the in-

florescences more robust, however, the leaves are markedly cordate in both. A. Dinteri

Engl. and Krause, another close ally, the type of which I have seen, has narrower and more

acute leaves, which are less cordate and the inflorescence is laxer than in our plant

(H.G.S.).]

11. HYDROCHARITACEAE.

1. Lagarosiphon muscoides forma longifolia Wager.

Near Amisfort in Duvenhage’s Pan, 627.

2. Ottelia ulvifolia (Planch.) Walp. [O. australis Bremek., in Ann. Tvl. Mus., XV, ii

235 (1933).]

Near Amisfort in Duvenhage’s Pan, 623.

12. GRAMINEAE.

1. Hyparrhenia Ruprechti (Hack.) Fourn.

At foot of northern slopes of mountain, 590.

2. Cymbopogon excavatus (Hochst.) Stapf

In camp near vlei at foot of mountain, 224.

C. validus Stapf ex Burtt Davy

In drier parts of vlei behind homestead, 587.

. Bothriochloa pertusa (Willd.) A. Camus.

In shade of Acacia near margin of pan, 497.

4. Tragus Berteronianus Schult.

Along margin of pan, 452, 485. This species apparently is a facultative halophyte.

5. Paspalum scrobiculatum L. var. Commersonii Stapf

In marshy soil near foot of mountain, 225.

226

6. Panicum maximum Jacq.

In kloof above waterfall, 598: in shade of trees near foot of mountain, 570: near

margin of pan in shade of Acacia, 471. [No. 471 has shortly pubescent spikelets which are

smaller in size than those of plants from Rhodesia and Tanganyika. This may be due

to the young state of our material. The east African plants placed under this species in

Herb. Kew are on the whole more strongly pubescent and more robust in habit. Stapf,

in Fl. Trop. Afr., IX, iv, 657, considers hairiness a variable character and thus includes

plants such as the above under one species (H.G.S.).]

7. Urochloa rhodesiensis Stent

Growing luxuriantly in shade of Acacia and attaining a height of 120 cm., 483.

U. panicoides Beauv.

In shade of Acacia along margin of pan, 464. A rather small but very leafy plant.

8. Brachiaria deflexa (Schum.) C. E. Hubbard ex Robyns; (B. regularis Stapf).

In moist places above waterfall in kloof behind homestead, 599.

B. grossa Stapf

On rocky slopes of the mountain, 525. [The spikelets are somewhat smaller and less

turgid than those of typical sheets of this species (Angola). The inflorescence is less branched

and on the whole our specimen is much weaker than any of the material under this species

in Herb. Kew. This constitutes the first record of the species for the Transvaal (H.G.S.).]

B. nigropedata (Munro ex Hiern) Stapf

On mountain slopes behind homestead, 617.

9. Echinochloa colona Link

Near rocky pool on northern slopes of mountain, 618.

E. stagnina (Retz.) Beauv.

On grassy slopes near vlei behind homestead, 242.

10. Digitaria debilis (Desf.) Willd.

Very occasional in marshy ground, 219.

D. eriantha Steud.

On rocky ledges above kloof behind homestead, 607: on lower northern slopes of

mountain, 527.

D. milanjiana Stapf

Fairly frequent in very sandy area about 3 miles west of pan, 635.

11. Rhynchelytrum villosum Chiov.

On lower rocky slopes of mountain, 524: in sandy area about 3 miles west of pan, 633.

12. Tricholaena monachne (Trin.) Stapf et Hubbard

At foot of mountain 589.

13. Cymbosetaria sagittifolia (A. Rich.) Schweickerdt

Fairly frequent but scattered in partial shade of Lonchocarpus belt, 578. [This con-

stitutes the first record for the Transvaal. So far it has been recorded from Tropical Africa

only, including northern South West Africa. ]

14. Setaria verticillata (L.) Beauv.

Along margin of pan, growing in association with Heliotropium, 466. Our gathering

is somewhat stunted, undoubtedly due to abnormal edaphie conditions.

227

15. Cenchrus ciliaris L.

Frequent but scattered in Catophrates belt, apparently grazed by stock, 179: about

3 miles west of pan in very sandy soil, fairly frequent, 632.

16. Aristida adscensionis L.

In shade of Acacia near margin of pan, 472.

A. meridionalis Henr.

Very occasional between loose boulders on northern slopes of mountain, 604.

A. junciformis T. and R.

On grassy slopes east of pan, at foot of mountain, 572.

A. barbicollis T. and R.

On margin of pan, only one specimen found, 461.

A. uniplumis Licht.

Dominant in very sandy soil about 3 miles west of pan, 630.

17. Sporobolus panicoides Rich.

In shade of bushes at Wylie’s Poort, 672.

§. pyramidalis Beauv.

In upper drier regions of vlei behind homestead, fairly frequent, 585: in Lonchocarpus

belt, 500.

8. Smutsii Stent

A. fairly frequent markedly stoloniferous species near the margin of pan, 460: in

shade of bushes near margin of pan, 469.

18. Danthoniopsis Dinteri (Pilger) C. E. Hubbard (Trichopteryx Dinter: Pilger).

Fairly frequent on lower rocky slopes of mountain, 522. The tropical African speci-

mens of this species in Herb. Kew are generally much more robust than our gathering.

19. Loudetia filifolia Schweickerdt

On lower rocky slopes of the mountain, subdominant, 523.

20. Cynodon dactylon (L.) Pers.

In shade of species of Acacia at foot of mountain, leaves distichous, 588: frequent

near vlei below mountain, 228.

21. Chloris virgata Sw.

In shade of species of Acacia near margin of pan, 484.

22. Dactyloctenium aegyptium (L.) Beauv.

Occasional near margin of pan, 457: in shade of species of Acacia near margin of pan,

490. [No. 490 differs from typical D. aegyptiwm in being a taller plant and an annual,

but a large range of flowering and fruiting material will be necessary before it will be pos-

sible to say whether more than one species of this genus occurs in South Africa (H.G.S.).]

23. Enneapogon cenchroides (Licht.) Hubbard

Fairly frequent on rocky ledges above kloof behind homestead, 608: in shade of species

of Acacia near margin of pan, 489, 538, 478.

E. Pretoriensis Stent

Fairly frequent on northern rocky slopes of the mountain, 605.

24. Schmidtia bulbosa Stapf forma ?

Among loose stones on northern slepes, 569, 606: in very sandy area about 3 miles

west of pan, 631. [Our plants differ somewhat from typical S. bulbosa. No. 631 differs

228

particularly in the relatively short awns which only just exceed the membranous lobes of

the lemma and the palea being broader and more obovate. The glumes in all three speci-

mens are 9—-10-nerved whereas in true S. bulbosa they are 7-nerved (H.G.8.).]

This grass is locally known by the vernacular name “ krulgras ” on account of the

very curly appearance of the old withered leaves.

25. Phragmites communis Trin.

In marshy soil east of pan along margin of periodic stream, 621.

26. Trichoneura Schlechteri Ekman (Tvriodia Schlechteri Pilger, nomen !).

On rocky ledge above kloof behind homestead, 609 : among loose stones on northern

slopes of mountain, 526.

These specimens are very similar to Moisier 135 from northern Nigeria in Herb.

Kew, only differing in beimg somewhat less hairy. Furthermore they differ from typical

T. Schlechteri by being annual and having narrower leaf-blades (H.G.8.).]

27. Odyssea paucinervis (Nees) Stapt Me

Along margin of pan, 453, 463, 498, 499. This species is dominant along the margin

of the pan. It is a typical halophyte and forms a dense sward surrounding the greater

part of the pan. The spikelets are very variable with regard to the number of florets.

No. 499 has up to 20-flowered spikelets. The leaves are rigid and pungent.

28. Eragrostis cilianensis Link

In shade of species of Acacia, east of pan, 540.

E. aspera (Jacq.) Nees

In Wylie’s Poort in shade at foot of rocky ledges, 449.

E. curvula Nees

In very deep sanay soil about 3 miles west of pan, 629.

E. chloromelas Stewd.

In drier parts of vlei west of homestead, 586.

E. gummiflua Nees

On grassy slopes at foot of mountain, east of pan, 573, 574.

18. CYPERACEAE.

1. Cyperus sexangularis Nees

Eastern margin of pan under small Acacia, 287.

2. Pycreus lanceus (Thunb.) Turrill

Frequent in vlei behind homestead, 196.

P. polystachyus Béauv.

Occasional in vlei behind homestead, 211.

. Mariscus dregeanus Kunth

Along margin of stream leading from vlei behind homestead, 241.

4. Kyllinga melanosperma Nees

Frequent in vlei behind homestead, 212, 197. No. 212 appears to be a somewhat

dwarfed specimen, but otherwise agrees in essential characters with this species.

se)

5. Fuirena chlorocarpa Ridley

Occasional in vlei behind homestead, 210.

6. Scirpus muricinux C.B. Cl.

Frequent in semi-permanent pan on farm “ Eyem,” north of the Blaauwberg, growing

in association with Marsilia ephippiocarpa and Aponogeton Rehmannii, 82.

7. Fimbristylis complanata Link

Frequent in vlei behind homestead, 199.

F. diphylla Vahl.

In vlei behind homestead, 221.

20. XYRIDACEAE.

1. Xyris capensis Thunb.

Occasional in vlei behind homestead, 198.

22. COMMELINACEAE.

1. Commelina Forskalaei Vahl.

East of pan under cover of Acacia; flowers blue, 533. The capsules of this specimen

are constantly 1-seeded.

‘24, JUNCACEAE.

1. Juncus lomatophyllus Spreng.

Very frequent in vlei behind homestead, 208.

25. LILIACEAE.

1. Anthericum elongatum Willd. var. holostachyun Bak.

On southern slopes of mountain about 6 miles west of Louis Trichardt, 343.

2. Aloe rubro-lutea Schinz

Between Waterpoort and Zoutpan, 45la: farm “Stafford,” north of Blaauwberg,

very frequent in open grassland, 647.

Typical plants from §8.W. Africa have more densely bracteate inflorescences and

appear to flower in December. Our specimens do not differ sufficiently to warrant the

creation of a separate species.

3. Albuca sp.

Occasional on stony east fringe of the pan, 139. This is probably an undescribed

species.

4. Urginea Langii Bremekamp

Farm “ Kromhoek,” north of Blaauwberg; flowers white with greenish keel, gre-

garious, 80.

This species is fairly frequent in sandy areas in the vicinity of the Zoutpan. Whether

it is correctly placed in the genus Urginea is somewhat doubtful. In appearance it resembles

an Albuca more closely and is apparently allied to Albuca hereroensis Bak. (Schinz 2 in Herb.

Kew !). It differs from this in the shorter pedicels and smaller flowers. Fruiting material,

however, is necessary to decide its true affinity. (The species has recently been collected

at Rust-der-Winter, Pretoria district, by Dr. I. B. Pole Evans.)

5. Dipeadi glaucum (Burch.) Baker

Occasionally gregarious near margin of pan, 29, 259.

D. sp.

Between Zoutpan and Waterpoort, numerous plants in patches; flowers greenish,

268. This is probably an undescribed species.

230

6. Scilla megaphylla Baker

Frequent in shade above rocky ledges near waterfall on farm “ Elsteg,’’ about 6 miles

west of Louis Trichardt, 367.

7. Asparagus exuvialis Burch.

In open patches of Acacia belt, on north side of pan, 51: farm “ Kyem,” north of

Blaauwhberg, 97.

A. sp.

Creeping in species of Acacia on fringe of pan; stems pubescent ; fruits small, green,

globose, 141.

27. AMARYLLIDACEAE.

8. Crinum buphanoides Baker

Farm “ Eyem,” north of Blaauwberg, 86. This is a gregarious species commonly met

with in deep sandy soil. The perianth varies in colour from whitish to dark pink ; the tube

is 10 cm. long, very narrow and greenish in colour while the segments are only about 5 cm.

long and spread abruptly.

28. VELLOZIACEAE.

1. Vellozia equisetoides Baker.

On northern slopes of mountain, 318. This plant has branched stems reaching a

height of 1 metre. It was the first record of the species for the Transvaal Province but

has since been collected in Sekukuniland.

29. DIOSCOREACEAE.

1. Dioscorea cotinifolia Kunth (D. malifolia Bak.).

Liane in bush above waterfall on farm “ Elsteg,’”’ about 6 miles west of Louis Trichardt,

360.

DICOTYLEDONS.

36. SALICACEAE.

1. Salix Wilmsii Seem.

A tree, about 8 metres high, with slender branches; very frequent along stream in

Wylie’s Poort, 442.

38. ULMACEAE.

1. Trema guineensis (Schum.) Ficalho

On slopes of mountain of farm “ Elsteg,” about 6 miles west of Louis Trichardt, 350.

39. MORACEAE.

1. Ficus capensis Thunb.

In kloof behind homestead ; fruit borne on branches at foot of tree, 598a.

F. Pretoriae Burtt Davy

On rocky ledge at top end of kloof behind homestead, 614, 281; near foot of moun-

tain; small shrubby tree, 104.

F. Smutsii Verdoorn

On lower rocky slopes of mountain, 600.

231

F. soldanella Miq.

Northern rocky slopes behind homestead, 280: at foot of mountain, east of pan, 571.

The fruits of 571 are about 1.5 cm. in diam. and thus larger than usual.

F. Sonderi Miq.

On northern rocky slopes of pan, 230: east of pan, 560: on rocky slopes along roadside

in Wylie’s Poort, 669.

No. 230 was a very tall umbrageous tree with a canopy about 30 metres in diameter

and with conspicuously whitish to light brown stem and branches. The branches arch

downward and touch the ground, but were not found to root as in F. Pretorvae.

F. sycomorus L.

At foot of mountain; fruit with felty pubescence, 593.

F. sycomorus L. forma ?

Along stream leading from kloof to homestead, 110. This specimen resembles F.

sycomorus closely but differs in having larger and almost glabrous fruits. It may represent

an undescribed species and requires further study in the field.

40. URTICACEAE.

1. Pouzolzia hypoleuca Wedd.

Frequent along streams in Wylie’s Poort, 337.

42. LORANTHACEAE.

1. Loranthus Breyeri Bremekamp

North side of pan; parasitic on Acacia sp., flowers yellowish, 52. This species is evi-

dently of limited geographic range.

L. Dregei KE. and Z. var.

Parasitic on species of Acacia, north of pan; flowers yellow and green, 47: on species

of Ptaerorylon behind homestead, 17.

L. kalachariensis Schinz

At foot of mountain behind homestead, 517. The clusters of bright red flowers are

very striking.

L. oleaefolius var. Leendertziae Sprague

Parasitic on malvaceous shrub on northern slopes of mountain; flowers brownish-red

with the reflexed petals green within, 282.

2. Viscum combreticola Engl.

Parasitic on Pseudolachnostylis sp., on northern slopes of mountain, 311: on Com-

bretum sp. on northern slopes of mountain, 166.

V. verrucosum Harv.

Parasitic on Acacia sp., near margin of pan; fruit stippled, 492.

45. OLACACEAE.

1. Olax dissitiflora Oliv.

Very frequent on northern slopes of mountain, 153, 236.

2. Ximenia americana var. microphylla Welw. ex Oliver (X. Rogersii Burtt Davy).

Very frequent at foot of mountain behind homestead, 530, 111.

232

X. caffra Sond. E

Near margin of pan, 295: at foot of mountain in association with Dichrostachys sp., 5.

50. POLYGONACEAE.

1. Polygonum lapathifolium var. glabrum Burtt Davy

Hydrophyte, in association with Nymphaea, Ottelia, etc. in Duvenhage’s pan near

Amisfort, 626.

Burtt Davy in Flora of Transvaal, I, p. 169, states that this variety has glabrous

peduncles. The specimen collected by Holub, however, cited under this variety, shows the

presence of scattered glands on the peduncles.

P. serrulatum Lag.

In vlei behind homestead ; flowers whitish to heather pink; bracts pink, 215.

51. CHENOPODIACEAE.

1. Chenopodium ambrosoides L.

Margin of pan, occasional, 456.

2. Suaeda fructicosa Forsk.

Along margin of pan, 38, 462. This characteristic halophyte together with Odyssea

paucinervis is one of the dominant species surrounding the pan.

52. AMARANTACEAE.

1. Celosia scabra (Schinz) (Hermbstaedtia scabra Schinz).

In shade of Acacia near margin of pan; flowers pink, 481. [This species formerly

placed under Hermbstaedtia by Schinz has now been transferred to Celosia on the grounds

that the anthers are inserted on the lobes between the sinuses of the staminal-tube (see

Engl. Pflanzenfam., ed. II, 16c¢ (1934). Schinz has furthermore placed H. linearis under

this genus and following him H. Rogers Burtt Davy should also be transferred to Celosia

(H.G.S.).]

2. Cyathula crispa Schinz )

In Acacia belt north of pan, 46: occasional near Catophractes belt, 505.

C. uncinulata (Schrad.) Schinz (C. globulifera Moyq.).

Along margin of pan, 496.

3. Pupalia lappacea (L.) Juss.

Near stream in vlei behind homestead, 238.

4. Alternanthera repens (L.) O. Ktze.

In open ground east of pan, 548.

A. sessilis (L.) R. Br,

In stream leading from vlei behind homestead, 207.

53. NYCTAGINACEAE.

1. Commicarpus plumbagineus (Cav.) Standley. (Boerhaavia plumbaginea Cav.).

Growing under cover of shrubs near margin of pan; herb 3 ft. high; flowers white

to pale mauve; stamens 2, pink, long, exserted, 19, 451.

233

C. fallacissimus (Heim.) Heim. forma pilosa Heim. differt a typico C. fallacissimo:

indumento subdenso, pilis scabridulis, brevissimus patulis formato, verticillis sum-

mofere 6-floris, florum (pulchre purpureorum) pedicellis valde brevibus, usque

solum 2.5 mm. long., haud capillaribus.

Frequent in sandy soil, growing in association with Tribulus Zeyheri Sond., near Vivo ;

flowers pink; stamens long exserted ; fruits glandular, 653 (type): on northern slopes of

mountain behind homestead, 103.

The description of this new form was sent by Dr. Anton Heimerl of Vienna. In the

letter he states :—

“Tn den Abhandl. Bot. Ver. Prov. Brandenb. XXXI, 223 (1890) habe ich die

der Boerhaavia verticillata Poir nahestehende B. fallacissima (eine ebenfalls bistami-

nate Art) beschrieben, die jetzt in die durch Standley abetrennte Gattung Commi-

carpus Standley (Contrib. U.S. Nat. Herb.) XII, 373 (1908) einzureihen ist. Von

diesem Commicarpus fallacissimus Heim., den ich aus Arabien (Aden) vom Somali-

gebiete und aus dem Hererolande kenne, unterscheidet sich die (im schénsten Bliihen

aufgesammelte) Nr. 653 durch das rauhe, ziemlich dichte, ganz kurze Haarkleid, durch

die héchstens 6-bliitige quirle und durch sehr kurze, bis 2.5 mm. lange, nicht haar-

diinne Bliitenstiele. Der Arttypus ist fast kahl, die Quirle sind bis 8-bliitig und die

ganz diinnen Bliitenstiele erreichen schon zur Anthese 5-17 mm. an Linge; veil-

leicht wiirden reife Anthocarpen (sie fehlen Nr. 653) auch Unterschiede bieten. Ich

méchte daher die Transvaal-pflanze als eine bemerkenswerte Form abtrennen und sie

als f. pilosa anfiihren.”

2. Boerhaavia diffusa L.

In open veld east of pan, forming tussocks ; flowers small, purplish-pink, 551. With

regard to this gathering Heimerl remarks as follows :—

“Wenn ich die Boerhaavia einfach als B. diffusa L. bezeichne, glaube ich, keinen

argen Verstoss zu begehen; in den Formenkreis dieser weitverbreiteten und sehr

formenreichen Art gehért sie sicher, ebenso wie die von mir vor, langer zeit aufgestellte

B. Schinzi, die freilich durch einige Merkmale aus der Formenmenge vorragt; das

Fehlen von Anthocarpen macht die Sache natiirlich auch schwieriger !”

-54. PHYTOLACCACEAE.

1. Limeum Meyeri Fenzl.

Frequent in very sandy area about 3 miles west of pan; flowers white, 639, |

L. Dinteri Schell.

Between Waterpoort and Zoutpan, 264.

[In foliage and habit this gathering is a good match of Dinter 998 and Liideritz 161

in Herb. Hort. Bot. Berol. The inflorescence in our specimen, however, is more compact

and abbreviated, but this difference alone does not warrant its exclusion from the above

species (H.G.S.).]

2. Semonvillea fenestrata Fenzl.

Very sandy area about 3 miles west of pan, 636.

3. Gisekia pharnaceoides L.

Very frequent in sandy area near Vivo; petals purple-pink, 654: east of pan; petals

pink-tipped ; stamens white, 545.

55. AIZOACEAE.

1. Mollugo nudicaulis Lam.

In shade of trees south-east of pan, 584.

234

M. Cerviana (L.) Ser.

Frequent in open spaces among Acacia trees near margin of pan, 532.

2. Pharnaceum salsoloides Burch. (P. verrucosum KH. and Z.).

In sandy soil on eastern margin of pan; flowers white, 298: in shade of Acacia trees

east of pan, 534.

3. Orygia decumbens Forsk.

In open sandy soil between Zoutpan and Waterpoort, not very frequent, 269.

4. Trianthema pentandra L.

Very frequent along eastern margin of pan, 546.

T. erectum Schlechter

Along margin of pan; flowers very small in sessile clusters; perianth-segments 5,

white ; stamens 5 with pink anthers, 459.

[In Herb. Hort. Bot. Berol. four sheets of Schlechter 11790 (from Komatipoort, Tvl.)

are not unlike our plant. The specimens, however, were collected in a young and sterile

stage. In foliage and branching they agree with our plant, but appear to be of an upright

habit—in fact they closely resemble the right-hand specimen of our 459 in Nat. Herb.

Pretoria—whereas the remaining specimen on this sheet has a semi-prostrate habit

(H.G.S.).]

5. Sesuvium digynum Welv. ex Oliver (Trianthema salarium Bremekamp).

Western side of pan, Bremek. and Schweickerdt, 232; under Acacia shrubs near margin

of pan; flowers sessile purple-pink; stamens 9; styles 2, 488.

56. PORTULACACEAE.

1. Talinum Arnotii Hook. f.

Common in sandy soil between Zoutpan and Waterpoort, 260.

T. caffrum (Thunb.) E. and Z.

Fairly frequent in sandy soil on farm “ Kyem,’

T. transvaalensis von Poellnitz

In cover of shrubs in Catophractes belt, 193.

2. Portulacaria afra Jacq.

In Wylie’s Poort, decumbent on rocks or arborescent, up to 5 metres high, 332.

P. oleracea L.

Occasional along margin of pan; flowers yellow, 31.

P. quadrifida L.

Fairly frequent on eastern side of pan, 294.

P. trianthemoides Bremekamp

Frequent along margin of pan; flowers deep yellow, 458.

north of Blaauwberg, 93.

59. NYMPHAEACEAE.

1. Nymphaea caerulea Sav. :

In Nyl River about seven miles north of Nylstroom ; flowers white with yellow centre,

440.

N. capensis Thunb.

Near Amisfort in Duvenhage’s Pan; flowers blue with yellow centres, 624.

The flower-stalk of these specimens is always exserted from 15-30 em. beyond the

surface of the water, whereas in the foregoing species the flower almost floats on the surface

of the water.

239

62. MENISPERMACEAE.

1. Cocculus hirsutus (L.) Diels

Liane, common on trees in Lonchocarpus belt, 284.

2. Desmonema cafirum Miers.

Liane, frequent in Lonchocarpus belt ; berries scarlet, 136.

This plant was observed to reach the tops of fairly tall trees such as Albizzia, ete.

The stems are extremely succulent and very brittle. This is the first record of the species

for the Transvaal.

63. ANNONACEAE.

1. Hexalobus glabrescens Hutch. and Dalz.

Small spreading tree on upper slopes of mountain, 155, 315.

2. Artabotrys brachypetalus Benth.

Tree about 2 metres high, on upper slopes of mountain, 321, 160.

BED: ptfele, abup hAtrpana 20. CAPPARIDACEAE.

(ictarer. 2c 5nd lz r

1. Cleome diandra Burd! (Dianthera Petersiana Klotszch ; D. burchelliana Klotszch).

On ledge on slopes of mountain; flowers yellow, 612.

2. Capparis tomentosa Lam.

In Lonchocarpus belt near margin of pan, 13. A robust sprawler with festooning

branches inclined to cover completely smaller trees and shrubs such as Acacia and Salvadora.

3. Boscia Rehmanniana Pest. forma.

In Catophractes belt, leaves and fruit shortly pubescent, 182a.

[This tree is fairly frequent in the sandy area north of the pan where it reaches a height

of 4-5 metres. It differs from typical B. Rehmanniana in having pubescent leaves, whereas

in the latter species they are always glabrous. Lugard 27 from Kwebe Hills, Ngamiland

in Herb. Kew, resembles our gathering, but has glabrous leaves. The fruits, however, are

pubescent and the flowers fasciculate (not a pedunculate inflorescence !) in which points

it agrees with our gathering. Furthermore our specimen agrees in habit and in fruit with

Seiner 78 from Bechuanaland in Herb. Hort. Bot. Berol. but in this specimen also the

leaves are glabrous (H.G.S8.).]

B. albitrunca Gilg et Benedict.

Between Zoutpan and Waterpoort, 276.

4. Courbonia glauca (KI.) Gilg and Benedict (C. camporum Gilg and Benedict).

Between Waterpoort and Wylhe’s Poort, 328; between Zoutpan and Waterpoort, 247.

[Gilg and Benedict in Engl. Bot. Jahrb., 53, 217 (1915) distinguish between C. glauca

and C. camporum on the grounds that the former is a woody shrub, whereas the latter is

a herbaceous plant. Knowing our plant well in the field, these characters in my opinion

are of no taxonomic value. C. glauca may either be a herbaceous plant 25 cm. or so_ high,

or may reach a height of nearly 3 metres. It then is a shrub with a fairly lignified base.

I have compared our gatherings with the types of C. glauca and C. camporum and have

come to the conclusion that only one species is concerned (H.G.8.).]

5. Cadaba termitaria N. HE. Br. (C. macropoda Gilg).

Very occasional near margin of pan, 34, 286.

[The shape and size of the leaves of this species is variable, a character of many plants

growing in arid regions. Our specimens have much smaller leaves than the type of C.

termitaria, but this difference alone does not justify specific distinction. Several small-

leaved specimens in Herb. Kew were considered by Dr. J. Burtt Davy to represent a dis-

tinct species, but in my opinion are merely forms of the above species (H.G.S8.).]

236

6. Maerua maschonica Gilg

North of pan, 2.5-3 m. high, scrambling in Salvadora sp., 43. Cattle are very fond

of this plant.

M. Legatii Burtt Davy

Shrub about 1 m. high in Catophravtes belt, north of pan, 182.

76. CRASSULACEAE.

1. Kalanchoe paniculata Harv.

Kast side of pan at foot of mountain; flowers, yellew-green; leaves crenate, 561 ;

scattered in Catophractes belt, north of pan, 184. ;

[The following described species are very closely allied: K. paniculata Harv., K.

multiflora Schinz, and K. pyramidalis Schonl. In the descriptions they are distinguished

by having either entire or crenate, sessile or petiolate leaves. Whether those characters

are of taxonomic value is doubtful, since in our 561 both sessile and petiolate leaves are

found on the same plant. There are a few sheets in Herb. Kew from Ngamiland and

Rhodesia named by N. HE. Brown as K. multiflora Schinz. Those specimens have crenate

leaves, whereas Schinz describes the species as having entire leaves. It is somewhat doubt-

ful whether the three species mentioned above, differimg in leaf-shape only, a variable

character in many succulents, should occur in the same geographic region. Study of plants

in the field will probably show that the above view is correct (H.G.8.).]

77. SAXIFRAGACEAE.

1. Vahlia capensis Thunb.

In shade of shrubs above vlei1 behind homestead, 232a.

80. MYROTHAMNACEAE.

1. Myrothamnus flabellifolia (Sond.) Welw.

On northern rocky slopes of mountain, 615, 616, 171. This plant is perhaps one of

the most remarkable found in Southern Africa. It is a dioecious shrub reaching a height

of about 75 cm. and is usually gregarious. The male and female plants grow side by side.

Some branches are decumbent and root adventitiously. Fragments of the plant which are

shrivelled and dead will resurrect even after many years when soaked in water and assume

colour and habit as in the living state (probably an imbibition movement).

85. LEGUMINOSAE.

1. Albizzia Rogersii Burtt Davy

On rocky northern slopes of mountain, 170, 601; a tall tree with markedly spreading

branches, locally fairly abundant.

2. Acacia Benthamii Rochebr.

In Lonchocarpus belt close to pan, 9; small tree, only about 1.5-2 metres high, on

the very margin of pan, 470.

A. heteracantha Burch. (A. litakunensis Burch. ; A. spirocarpoides Engl.).

Between Zoutpan and Waterpoort, 274: along eastern margin of pan, 150.

A. karroo Hayne

Frequent in Lonchocarpus belt between foot of mountain and pan, 11.

A. pennata Willd.

Above stream leading from the kloof behind homestead, 108.

237

A. permixta Burtt Davy var. glabra Burtt Davy

North side of pan forming dense patches, 41, 60. Plants reaching a height of 1-2

metres and forming dense scrub adjoining the halophytic fringe flora of the pan.

A. Senegal Willd. (A. rostrata Sim).

East side of pan, 300. Only a few specimens of this low-spreading tree about 5 metres

high were found scattered along the eastern margin of the pan.

A. Woodii Burtt Davy

In Lonchocarpus belt, 12. Large round-topped trees with rough bark and clustered

branches, growing among A. karroo.

3. Dichrostachys glomerata (Forsk.) Hutch. and Dalz.

On northern slopes of mountain, 226.

4. Elephantorrhiza Burkei Benth.

In depression on northern slopes of mountain: small shrubs, occasional, fruiting

specimens only, 175.

5. Burkea africana Hook.

On upper slopes of mountain, flat-topped trees, 319.

6. Copaifera mopane Kirk

Between Waterpoort and Wylie’s Poort, 329.

7. Cassia delagoensis Harv.

Fairly frequent at foot of mountain east of pan, 563.

C. arachoides Burch. forma ?

Frequent in sandy soil, between Zoutpan and Waterpoort, 265: fairly frequent in

very sandy area about 3 miles west of pan, 642.

[Our specimens match Moss and Rogers 36 from Messina, N. Transvaal, placed under

C. holosericea Fresen. by Burtt Davy in his Flora of the Transvaal, ii, 325 (1932). The

flowers and fruits of true C. holosericea are much smaller and more hairy than those of our

265 and 642. C. obovata Collad. has fruits which differ by the conspicuous median crest

on the lateral surfaces. Our plants are most closely allied to C. arachoides Burch. of which

species they may be merely a pubescent form (H.G.8.).]

8. Pterolobium exosum (Gml.) Bak. f. (P. lacerans R. Br.).

Frequent in Wylie’s Poort, 450. A conspicuous liane on account of its showy brilliant-

red winged fruits, climbing to the tops of the highest trees in the forest-clad ravines.

9. Peltophorum africanum Sond.

Frequent on lower slopes of mountain; tree with erect racemes of bright yellow

flowers, 107.

10. Pseudocadia zambesiaca (Bak.) Harms

At foot of mountain east of pan, 596: north-western slopes of mountain, 65. Frequent

in area around pan; one of the tallest and most umbrageous trees with dark shiny foliage

and stems up to 2 metres in diameter.

11. Calpurnia subdecandra (L’Herit.) Schweickerdt comb. nov.; [Robinia subdecandra

L’Herit. in Stirp., Nov., 157, t. 75 (1784); Calpurnia lasiogyne E. Mey}.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 353.

12. Lotononis Bainesii Bak.

Frequent along furrow leading from large vlei west of homestead, 235.

13. Crotalaria Schinzii Bak. f.

Frequent in shade of trees at foot of mountain, south-east of pan, 580.

238

C. longistyla Bak. f.

Between Louis Trichardt and Pietersburg, frequent in sandy soil, 664.

C. inhabilis Verdoorn sp. nov.

Affinis ©. australi sed pubescentia appressa et ovario tomentosa differt; affinis C. athro-

ophyllae sed racemis laxioribus elongatioribusque, calyce breviore, floribus omnino flavis (nec

purpureo-striatis), et suffrutice circiter 1 m. alto differt.

Suffrutex erectus, + 1 m. altus, supra ramosissimus; ramuli petioli et pendunculi

breviter griseo-appresso-pubescentes. Folia exstipulata, longe-petiolata, trifoliolata ; petioli

6 cm. longi, internodiis et foliolis valde excedentes ; foliola limeares vel anguste-oblonga,

1-3 em. longa, 2-7 mm. lata, infra appresse pubescentes, supra glabrescentes, apice minute

mucronata vel retusa, basi leviter cuneata, breviter petiolulata. Racemi oppositifolii vel

terminales, elongati, 15-23 cm. longi, laxi, 10-15-flori. Bracteae subulatae, 2 mm. longae.

Flores flavi, -- 2 em. longi, in medio racemorum plerumque 2 cm. distantes (C. athro-

ophyllae 1 cm. distantes). Pedicelli circiter 10 mm. longi, breviter appresse griseo-pube-

scentes, infra medium 2-bracteolati. Calyx + 8 mm. longus, appresse pubescens vel

glabrescens, lobis deltoideis circiter 3 mm. longis. Veaillum dorso glabrum ; carina dorso

rotundata, 2 cm. longa, 7 mm. lata. Ovariwm longe-stipitata, argenteo-tomentosa.

Transvaal.—Zoutpansberg distr.: On sandstone outcrop about 10 miles east of

Waterpoort, Noy., Obermeyer, Schweickerdt and Verdoorn, 323.

C. inhabilis was found abundantly on a sandstone outcrop about 10 miles east of

Waterpoort on the road to Wylie’s Poort. It is an erect half-shrub up to about 4 ft. high

and is densely branched above, which gives it a top-heavy or broom-like appearance.

The clustered ultimate branchlets, which are much more slender than those from which

they arise, and the very long-petioled leaves, the petioles being about three times as long

as the leaflets, accentuate the broom-like appearance. The pure yellow flowers are borne

on the branchlets in lax elongated racemes which may be either terminal or opposite the

leaves. ;

14. Argyrolobium transvaalense Schinz

Fairly frequent on farm “ Hlsteg,” about 6 miles west of Louis Trichardt, 351.

15. Indigofera circinnata Benth.

Between Zoutpan and Waterpoort, 275. A spiny slender shrub reaching a height of

about 1 metre.

I. egens N. EK. Br.

On upper slopes of mountain, 307.

I. flavicans Bak.

Frequent in sandy area east of pan, 296, 549.

I. Holubii N.E. Br.

Growing in cover and shade of bushes ; flowers attractive, rosy pink or red, 480: east

of pan, 544.

I. tettensis Klotzsch (I. Baukeana Vatke).

In shade of bushes along margin of pan; flowers pink; calyx small, 487.

I. sp. allied to J. adenoides Bak. f.

Common on north-western slopes of mountain, 154: frequent on rocky slopes east of

pan, 566.

[Our specimens from J. adenoides having 3-5-foliolate leaves, minute stipules, fruits

set with a few short glands and stems covered with minute almost sessile glands (H.G.S.).]

239

16. Psoralea pinnata var. latifolia Harv.

Vlei behind homestead ; slender shrub attaining a height of about 5 metres, 202.

17. Sylitra contorta (N.E. Br.) Bak. f.

In sandy area north of pan near Catophractes belt, 508: between Zoutpan and Water-

poort, in open sandy soil, 267.

18. Tephrosia zoutpansbergensis Bremekamp

Very occasional, shrub 1-1-5 metres high, on northern slopes of mountain, 174.

T. capensis (Thunb.) Pers.

On lower slopes of the mountain, 531.

T. purpurea Pers. forma ?

Kast of pan in shade of Acacia, 543.

T. euchroa Verdoorn sp. nov. ab 7. noctiflorae floribus multo majoribus differt et ab specibus

omnibus Transvaale valde distincta.

Suffruter parvus plus minusve 30 cm. altus, multiramosus, erectus vel rami decum-

bentes. Rami cano-pubescentes, glabrescentes, internodiis 2-5-3 cm. longis. Folia im-

paripinnata 2—6-juga, fololis oblanceolata-oblongis vel oblongis, 1-5-3 cm. longis, 5-9 mm.

latis utrinque cano-pubescentibus, superne ultimo glabrescentibus, apice minute mucronatis

vel retusis petiolatis, petiolulis 3 mm. longis, dense cano-pubescentibus. Stipulae lanceo-

latae, acutae, 5 mm. longae, cano-pubescentes, 3-nervatae. Racemz folia longiores, laxi-

flori. Flores 1-5-2 cm. longi, rubicundi. Bracteae lanceolatae, acutae, 4 mm. longae,

cano-pubescentes. Calyx cano-pubescens, in toto 5 mm. longus, lobis triangularibus

1-3 mm. longis. Vezillum extus appresse pubescens, suborbiculare, 1-5-2 cm. longum et

latum ; carina circiter 1-6 cm. longa, 1 cm. lata, quam vexillum paullo brevior et quam

alae paullo longior. Leguwmen immaturum lineare, apresso-cano-pubescente.

Transvaal_—Zoutpansberg distr: On lower slopes of the Zoutpansberg, on farm

“ Zoutpan,” among rocks, flowers pinky-red, April, 1934, Schweickerdt and Verdoorn, 529 ;

on rocky N.W. slopes of the Zoutpansberg, November, 1932, Obermeyer, Schweickerdt and

Verdoorn, 73 (type).

This is very distinct from all the known Transvaal species of Tephrosia. The silvery-

grey colour of the bush and the pinkish-red flowers suggested the specific name. It is

somewhat like 7’. noctiflora but has much larger flowers. The flowers of each raceme develop

at different intervals and one finds several large open flowers while the buds above are still

very immature.

19. Sesbania aculeata Pers.

Herbaceous, about 60 cm. high, near margin of pan, 454.

20. Ormocarpum trichocarpum (Taub.) Harms (0. setoswm Burtt Davy).

Small tree at foot of mountain, south-east of pan, 582.

21. Stylosanthes mucronata Willd.

In dry rocky soil between Waterpoort and Wylie’s Poort, 336.

22. Lonchocarpus capassa Rolfe

Dominant, large tree in area between pan and foot of mountain, 8.

23. Abrus laevigatus E. Mey.

In vlei behind homestead, 206.

24. Glycine javanica L.

Twiner on trees along stream leading from kloof; flowers white in long erect racemes,

597.

25. Neorautanenia edulis C.A. Sm.

Sandy area about 3 miles west of pan, 645; in Catophractes belt, 183: farm “ Krom-

hoek,” north of Blaauwberg, 81: between Zoutpan and Waterpoort, 271.

26. Rhynchosia minima DC.

Twining in Capparis near margin of pan, 473.

27. Eriosema psoraleoides (Lam.) Don (2. cajyanoides Benth.).

In vlei soil behind homestead, 594.

28. Otoptera Burchellii DC.

Frequent in sandy area about 3 miles west of pan, 643: farm “ Hyem,” north of

Blaauwberg, 100.

29. Dolichos Schlechteri (Harms) Burtt Davy

Creeping in small shrubs in Catophractes belt, 185.

[This species was originally described as Phaseolus Schlechter’ Harms in Engl. Bot.

Jahrb. XXX, 91 (1901). In the type description Harms makes no mention of the shape

of the stigma, although at some time or other he apparently was somewhat in doubt as to

which genus to refer it to. Burtt Davy in his Flora of the Transvaal cites this species under

Dolichos Schlechterv Harms MS apparently not having been aware of the fact that Harms

had described it under another name. The penicillate stigma is not characteristic of

Phaseolus and Burtt Davy has correctly aligned this plant with others at present under

Dolichos (H1..G.8.).]

30. Alistilus bechuanicus N.H.Br.

On upper slopes of mountain, a creeping procumbent herb, peduncles erect, flowers

pink, 317.

86. GERANIACEAE.

1. Monsonia glauca Knuth

Very sandy area about 3 miles west of pan, 640.

90. ZYGOPHYLLACEAE.

1. Tribulus terrestris L.

Under cover of Acacia east of pan, 539.

T. zeyheri Sond.

Along roadside, Vivo, 650: between Zoutpan and Waterpoort, 263: under thorn

fence of kraal east of pan, 299.

2. Balanites australis Bremekamp

West of pan, scattered but frequent, 479: near margin of pan, 15, 33. A small, very

thorny tree.

91. RUTACEAE.

1. Toddaliopsis Bremekampii Verdoorn

Frequent on northern slopes of mountain behind homestead; a shrub or small tree

with tri-foliolate leaves; flowers small, greenish ; fruits warted, 66, 156, 567.

2. Fagara capensis Thunb.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, in shade on rocky slopes, 346,

347. Two trees growing near together, one 3 or 4 ft. high with small leaves (probably a

young tree) and the other about 8 ft. high with larger leaves (I.C.V.).

3. Clausena anisata (Willd.) Hook. f. (C. inaequalis var. abyssinica Engl.).

Farm “ Elsteg,” about 6 miles west of Louis Trichardt; small tree, bad odour, 356.

241

92. SIMARUBACEAE.

1. Kirka pubescens Burtt Davy

Large spreading tree, fairly frequent on northern slopes of mountain, 163.

93. BURSERACEAE.

1. Commiphora pyracanthoides Engl.

North of pan in Catophractes belt; shrub; flowers red and yellow, 48, 49, 512: on

slopes of mountain; small tree with greenish stems, 159.

C. Marlothii Engl.

In dry parts of kloof behind homestead; tree with papery bark, 121: frequent on

northern slopes of mountain, 165.

C. calciicola Engl.

North side of pan near Catophractes belt; flowers red; leaves 3-foliolate ; fruit red,

oblique, 180, 511: between Louis Trichardt and Pietersburg, 663.

C. cinerea Engl.

On lower slopes of mountain behind homestead, tree about 8 metres high, 152.

C. sp. near C. mollis and C. Welwitschiv but material insufficient for specific identification.

Farm “ Eyem,” north of Blaauwberg, 95.

94. MELIACEAE.

1. Ptaeroxylon obliquum (Thunb.) Radlk.

In upper drier parts of kloof behind homestead ; small slender tree, 118.

2. Entandophragma caudatum Sprague

On northern slopes of mountain; tree about 10 metres high; leaves with long drip-

points, 306.

3. Ekebergia Meyeri Presl

Above stream leading from kloof behind homestead ; spreading tree, 112.

98. EUPHORBIACEAE.

1. Pseudolachnostylis maprouneaefolia Pax

Northern slopes of mountain, 74, 75, 117.

2. Fluggea virosa Baill.

Near foot of mountain in Lonchocarpus belt, 6, 10, 521.

3. Phyllanthus reticulatus Poir

At foot of mountain, east of pan; tree about 8 metres high, 620.

4. Bridelia mollis Hutch.

In kloof behind homestead, 599b; northern slopes of mountain, 243.

5. Androstachys Johnsonii Prain

Tree, 5-8 metres high, very frequent in Wylie’s Poort, 327.

6. Croton gratissimus Burch.

Fairly frequent on northern slopes of mountain ; small tree, 157, 169.

C. megalobotrys Mill. Arg. (C. Gubouga 8. Moore).

Tree, about 5 metres high, near homestead, 4: beside stream leading from kloof, 106.

242

C. pseudopulchellus Pax

On upper slopes of mountain, locally abundant; small shrub, about 1 metre high,

303, 316.

7. Acalypha glabrata Thunb.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt ; shrub, about 2-5 metres high,

348.

A. indica L.

In shade of Acacia, east of pan, 535.

8. Tragia Okanyua Pax

Northern slopes of mountain, twining in Croton sp., 283.

9. Plukenetia africana Sond.

Climber, in Acacia sp., east of pan, 552.

10. Jatropha erythropoda Pax et Hofim.

In Catophractes belt, 194: on farm “ Zoutpan,” on road to Waterpoort, 262.

J. zeyheri Sond. forma.

Near farm “ Kyem,” north of Blaauwberg, 101.

[This specimen differs from J. Schlechter: Pax in the more strongly lobed, less pube-

scent and shorter petioled leaves and the glabrous calyx. In the latter character our

plant also differs from J. zeyhert Sond. but approaches it with regard to leaf-shape and length

of petiole. Since the last two characters appear to be of some importance in the delimita-

tion of the S. African species of this genus, I place our gathering under the above species

and regard it as a form until more material becomes available for study (H.G.8.).]

11. Spirostachys africana Sond.

In Lonchocarpus belt near homestead, tree 5-7 metres high, 77: almost on margin

of pan, 35.

12. Euphorbia aeruginosa Schweickerdt.

On northern rocky slopes of mountain, 151, 688.

E. Cooperi N.E. Br.

On northern slopes of mountain near homestead, 649.

E. Giirichiana Pax

In Catophractes belt ; small bush, branched at base, up to 50 cm. high; flowers yellow

or green, 178, 513.

E. Tirucalli L.

Plentiful in parts of Wylie’s Poort, 676.

E. transvaalensis Schltr.

On rocky slopes in Wylie’s Poort, 369.

13. Monadenium Lugardae N.E.Br.

On mountain slopes east of homestead, forming patches; stems decumbent, rooting

where they touch the ground, about 40 em. high, 648.

14. Cluytia pulchella var. obtusata Sond. ?

Shrub, near waterfall on farm “ Elsteg,” about 6 miles west of Louis Trichardt, 366.

101. ANACARDIACEAE.

1. Rhus Gueinzii Sond.

On margin of vlei, west of homestead, 231.

243

R. pyroides var. gracillis (Engl.) Burtt Davy

In marshy soil near vlei, 222.

R. transvaalensis Engl.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 365.

108. CELASTRACEAE.

1. Cassine Schlechteri (Loes.) Davison

Small tree, in Lonchocarpus belt beyond vlei, bearmg somewhat pointed ripe fruits,

577: same locality, tree, ripe fruits globose, 576: small tree, on northern slopes ; fruits

immature, 237.

Rehmann, 6459 from Houtbosch, Transvaal, referred to the above species in Herb.

Kew has oblong apiculate fruits as in our No. 577. In the latter sheet the apiculus of the

fruits is more pronounced. No. 576 has spherical fruits without an indication of an

apiculus, but does not differ from 577 in other respects. Until more material, both

flowering and fruiting, has been studied, it is impossible to decide whether the above spe-

cimens represent distinct varieties (or even species) or whether only one species variable

with regard to leaf size and shape of fruits is involved. In the type description of

Mystroxylon Schlechteri [Loes in Engl. Bot. Jahrb., 28, 159 (1901)] no mention of the

fruits is made.

104. HIPPOCRATEACEAE.

1. Hippocratea longipetiolata Oliv.

Between Zoutpan and Waterpoort ; small shrub, with long spreading branches, 254.

106. SAPINDACEAE.

1. Cardiospermum alatum Bremekamp and Obermeyer

East of pan, climber, on Gossypium, 559.

109. RHAMNACEAE.

1. Zizyphus mucronata Willd.

Between Waterpoort and Wylie’s Poort, 326.

2. Berchemia discolor (K1.) Hemsl.

Northern slopes of mountain, 279: frequent in Lonchocarpus belt, 278.

111. VITACEAE.

1. Cissus lonicerifolius C. A. Smith

Farm “ Eyem,” north of Blaauwberg; shrub, 2 metres high; leaves folded, pungent

odour, 96.

Usually a shrub, varying from 1-3 m. high, much branched; tendrils usually only

found near end of branches.

C. quadrangularis L.

Between Zoutpan and Waterpoort, climbing in small shrubs, 261.

C. simulans C. A. Smith

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, luxuriant climber, 361.

C. unguiformifolius C. A. Smith

Between Waterpoort and Wylie’s Poort, abundant, procumbent, 330.

244

112. TILIACEAE.

1. Corchorus pongolensis Burtt Davy and Greenway ?

On lower slopes of mountain behind homestead, 67. A slender shrub, up to 130 cm.

high, differing from the type by having somewhat shorter bracts, but otherwise agreeing

fairly well.

[Our plant is allied to C. Kirku N. EH. Br. but may be readily distinguished by the

indumentum (H.G.S8.).]

C. asplenifolius Burch.

Under Acacia, east of pan, 536, 541.

2. Grewia flava DC.

In open parts of Catophractes belt, 56.

G. hexamita Burr.

Shrub, about 3 metres high, near margin on east of pan, 149.

G. retinervis Burr.

Near homestead, under cover of Terminalia, 2: on slopes of mountain behind home-

stead, 244.

G. occidentalis L.

Tree, 7-8 metres high, in Wylie’s Poort, 341.

G. Schweickerdtii Burr. sp. nov.

Frutez 8-10 pedes altus. Rami cortice rubro, ramuli breviter virgati, superne pilis

stellatis multi-ramosis, sat.longis, flavidis plus minus interrupte villosi. Stipulae lanceo-

latae. Petiolus in latere longiore usque 6 mm. longus, validus, flavido-villosus ; lamina

basi valde obliqua, in altero latere rotundata, in altero auriculari formiter producta, cordata,

praeterea ambitu elliptica vel elongato-elliptica, maxima visa 5 cm. circ. longa, 3-4 cm.

lata, supra glabra, nitens, rugosa, nervis nervulisque impressis, sat dense reticulata, subtus

pallide flavido-velutina, costa atque nervis lateralibus minus pilosis i.e. colore brunneo

notalis, nervatione ulteriore vix conspicua, margine revoluto, leviter crenato, apice plerumque

rotundato, rarius subapiculato. Cymae floriferae axillares, solitariae, 1—3-florae, in modo

ramulorum pallide fusco-flavido-villosae. Pedunculus 0-5-1-0 em. cire. longus. Pedicella

bracteis oblongo-lanceolatis vel lanceolatis, intus glabris, extra pubescentibus suffulti,

pedunculo subaequilongi. Albastra oblonga. Flores speciosi. Sepala 2-4 cm. longa,

linearia, extra pilis stellatis brevissimis atque longis fusco-flavidis multiramosis interrupte

sequentibus vestita. Petala ungui subcircularia 4 mm. in diam. circ. metiente, dorso

pilosulo, antice supra aream glanduligerum squamiformiter libero, dense, late, minus late

in lateribus flavido-tomentoso ; lamina late obovata vel potius suborbicularis, apice leviter

excisa. Androgynophorum supra petalorum unguem flavide villosum conspicue productum,

dein stamina numerosa, 1-5 em. fere longa, atque gynaeceum flavido-villosum obovatum

praebeus. Stylus longitudine staminum, glaber, gynaeceo subito impositus; stigmata

dilatata, applanata, late rotundata. Ovula pro loculo 8, biseriata, pro serie 4.

Transvaal.—Zoutpansberg distr.: Farm “ Zoutpan 193,” in kloof behind homestead,

shrub, 8-9 ft. high, November, 1921, Obermeyer, Schweickerdt and Verdoorn, 120.

G. Schweickerdtii is closely allied to G. hexamita, but differs in having much longer

petioles, large flowers and much larger leaves.

G. sp. near G. flava DC. (Material too poor for description.)

Shrub, about 38 metres high, near homestead, 147.

245

118. MALVACEAE.

1. Abutilon austro-africanum Hochr.

Small shrub, near eastern margin of pan, 144: between Zoutpan and Waterpoort,

270: in shade of Acacia sp., east of pan, 558.

2. Sida cordifolia L.

Small shrub, about 60 cm. high, frequent in open patches east of pan, 556: frequent

in vlei soil and in shade of trees in Lonchocarpus belt, 234.

S. flexuosa Burtt Davy

Frequent in sandy soil between Zoutpan and Waterpoort, 253.

S. Hoepineri Guerke.

Shrub, about 1 metre high, along eastern margin of pan, 148.

3. Pavonia Burchellii (DC.) R. A. Dyer

Small shrub, about 20 cm. high, in sandy soil of Catophractes belt, 506, 189.

P. dentata Burtt Davy

On rocky ledges in drier upper parts of kloof behind homestead, 126: on rocky moun-

tain slopes, 167: in Wylie’s Poort, on rocks, 671.

4. Hibiscus micranthus L.

Shrub, about 1 metre high ; flowers small, white, turning pink or red when folding up

again; In Catophractes belt, 504, 195.

H. dongolensis DC.

On northern slopes of mountain near pan, 575a.

H. physaloides G. and P.

Near homestead, in Lonchocarpus belt; flowers yellow with red blotch at base of

petals, 25.

H. physaloides G. and P. forma ?

Frequent along eastern margin of pan, flowers red or yellow with red blotch at base

of petals, 145, 555. [In foliage and habit the above gatherings resemble H. Schinzii Guerke

very closely but have flowers twice the size. In Herb. Hort. Bot. Berol. several sheets

named H. physaloides forma matched our gatherings. I am inclined to think that our

specimens and those in Berlin Herb. represent.a distinct variety with H. Schinzii as closest

affinity, since the latter as well as our plants are prostrate in habit, whereas H. physaloides

apart from having much larger leaves is usually an erect under-shrub (H.G.S.).]

H. praeteritus R. A. Dyer

Foot of mountain, east of pan, in shade of trees; about 2 metres high, 575; between

Zoutpan and Waterpoort, 257: at northern end of Wylie’s Poort, 338. This plant had

white flowers, which, however, turned red on drying: in all other respects it agreed with

the typical form.

H. intermedius A. Rich. var. aristaevalvis Guerke

Under small Acacia, on margin of pan; flowers creamy-yellow, 32, 476. These speci-

mens agree well with Lvideritz 82 from 8.W. Africa in the Herb. Hort. Bot. Berol.

5. Gossypium africanum Watt (G. transvaalense Watt).

Frequent in scrub and bush east of pan, 564, 135.

__ [I find no specific difference between G. africanum and G. transvaalense. Both oceur

in the same geographic area (H.G.S.).]

246

114. BOMBACACEAE.

1. Adansonia digitata L.

On northern rocky slopes behind homestead, 69.

The South African material which we have so far examined differs from true A. digitata

L. from West Africa in the following characters: The flowers and fruits are shortly pedun-

culate and the petals never reflex as shown in the figure in Memorias de Sociedade Broteriana

I, 50 et t. 5 (1930). The plant figured was from Portuguese Guinea which is near the type

locality and thus may be considered to be true A. digitata L. If any value is to be attached

to the differences recorded above the South African form may deserve varietal rank.

Further investigation, however, is necessary before any decisive step can be taken.

115. STERCULIACEAE.

1, Melhania Rehmannii Szyszyl.

Near homestead, 188a.

2. Hermannia boraginiflora Hook.

Lower slopes of mountain, east of pan; flowers pale mauve, 142, 562.

H. Holubii Burtt Davy

In sandy soil near Vivo ; flowers brick red, 652: in shade of Acacia trees near margin

of pan, 486.

H. grisea Schinz

In open soil, east of pan; flowers pale pink to brownish, 553.

3. Waltheria americana var. indica K. Schum.

In open soil, east of pan, common, 554.

4. Sterculia Rogersii N. EH. Br.

In sandy area west of pan, 87. A small (stunted ?) shrub with a succulent bole.

116. OCHNACEAE.

1. Ochna atropurpurea DC.

Near summit of northern slopes of mountain, 312.

2. O. sp. nov.

Between Waterpoort and Wylie’s Poort, 335.

[This specimen matches Rogers 19398 from Messina and Baines s.n. from “S.A. Gold

Fields” in Herb. Kew. It is very probably an undescribed species. Rogers 19398 is

quoted under O. pretoriensis by Phillips in Bothalia I, ii, 95 (1922) but does not belong to

that species, differing from it in leaf character as well as the articulation of the pedicel

(H.G.S.).1

117. GUTTIFERAE.

1. Hypericum Lalandii Choisy.

Along stream leading from vlei behind homestead, 204.

2. Garcinia Livingstonei And.

Occasional trees, on lower rocky slopes of mountain ; flowers sweetly scented ; fruits

orange-coloured, oblique, fleshy, edible, 71, 62.

123. FLACOURTIACEAE.

1. Trimeria grandifolia (Hochst.) Warb.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 354.

247

124. TURNERACEAE.

1. Piriqueta capensis (Harv.) Urb.

Between Waterpoort and Wylie’s Poort, 339.

125. PASSIFLORACEAE.

1. Adenia repanda (Burch.) Engl.

Between Zoutpan and Waterpoort, in open sandy soil, climbing over small bushes,

not very frequent; tuber subterranean, 246.

137. COMBRETACEAE.

1. Combretum apiculatum Sond.

In upper drier parts of kloof behind homestead, 603.

C. mossambicense (Klotzsch) Engl.

Near eastern margin of pan, lane on Lonchocarpus ; flowers white; anthers red, 143.

This constitutes the first record of the species for the Transvaal.

2. Terminalia prunioides Laws.

Tree, + 8 metres high, frequent in Catophractes belt, 191.

T. Rautanenii Schinz

Adjoining farm “ Zoutpan,” near Vivo, 83. A shrubby tree, attaining a height of

4-5 metres with much the habit of an Hhretia. It is fairly frequent in the sandy areas

adjoining the pan. This is the first record of the species from the Transvaal.

T. sericea Burch.

In Lonchocarpus belt adjoining the homestead, 1.

1388. MYRTACEAE.

1. Syzygium cordatum Hochst.

On slopes behind homestead, 63.

140. ONAGRACEAE.

1. Jussiaea augustifolia Lam.

Growing in bed of periodic stream leading from mountain to south-eastern margin of

pan, 559a.

143. UMBELLIFERAE.

1. Hydrocotyle asiatica L.

Near vlei to west of homestead, 225.

H. verticillata Thunb.

Frequent along stream leading from kloof, 200.

2. Steganotaenia araliacea Hochst.

In Wylie’s Poort, fairly frequent; chasmophyte, 331.

146. MYRSINACEAE.

1. Maesa lanceolata Forsk.

Along water-course leading to kloof, 113.

248

147. PRIMULACEAE.

1. Samolus Valerandi L.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, abundant on banks of stream,

363.

149. SAPOTACEAE.

1. Chrysophyllum magalismentanum Sond. (C. Wilmsw Engl.).

Near summit of northern slopes of mountain, 305.

. Mimusops Zeyheri Sond.

Farm “ Elsteg,” 6 miles west of Louis Trichardt, 361a.

150. EBENACEAE.

1. Royena sp. (not matched at Kew).

Near camp; fruit small, 277.

2. Euclea divinorum Hiern

Farm “ Hyem,” north of Blaauwberg, 98.

E. Guerkei Hiern

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 358.

E. lanceolata HE. Mey. ex Drege ?

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 371.

E. multiflora Hiern

In deep sandy soil, east of pan, 622.

151. OLEACEAE.

1. Jasminum stenolobum Rolfe

Farm “ Hyem,” north of Blaauwberg ; small erect bush, about 1 m. high, 92.

152. SALVADORACEAE.

1. Salvadora australis Schweickerdt sp. nov.

Affinis §. persicae Garcin ex Linn. et §. oleoidi Decne, sed ab illa indumento persistente

calyce valde lobato antheris multo minoribus stylo distincto, ab hae antheris maioribus con-

nectivo haud producto stylo distincto indumento persistente differt.

Frutex divaricato-ramosus vel arbor usque ad 8 m. alta ; rami iuniores minute et dense

cinereo-puberuli, teretes; internodia ad 2 em. longa, 2 mm. diametro. Folia breviter

petiolata vel fere sessilia, spatulata vel oblanceolata, obtusa, basin versus sensim attenuata,

usque ad 5 cm. longa, et 1 em. lata, integra, utrinque minute sed dense puberula, nervis

infra prominentibus supra inconspicuis. Flores viridi-flavi, in paniculis terminalibus vel

axillaribus dispositi. Paniculae circiter 2 cm. longae, multiflorae. Bracteae minutae,

0-75 mm. longae, ovatae, dense cinereo-puberulae, minute ciliolatae. Calyx circiter 2 mm.

longus, extra dense et minute puberulus, dimidio lobatus, lobis reflexis obtusis dense et

minute ciliolatis. Corolla plus minusve 3 mm. longa, profunde lobata; lobi triangulares,

acuti, sub anthesi reflexi. Stamina corolla breviora ; antherae minutae, fere 0:75 mm.

longae, glandulis interstaminalibus distinctis sed minutis. Ovariwm circiter 1 mm. diametro,

globosum, minute puberulum; stylus circiter 0-25 mm. longus, distinctus, stigmate in-

conspicuo. Fructus 6 mm. diametro, semitranslucens, globosus, minute puberulus.

Portucunse Kast Arrica.—Mapae: Guija, m bush country in sandy soil, about

1 mile from the Limpopo River, July, Lea, 5 (syn-type flowers).

249

TRANSVAAL.—Zoutpansberg distr.: Farm “ Zoutpan 193,” on northern slopes of

Zoutpansbergen, tree about 8 m. high, November, Obermeyer, Schweickerdt and Verdoorn,

18 (syn-type fruits) ; Keet in Forest Dept. Herb., 6833; Messina, Pole Evans, 1453.

Narau.—Zululand : Mkuzi, on Lebombo Flats, Galpin, 13320.

This is closely related to the Indian species S. oleoides Dec. from which it may be

distinguished by having larger anthers, a connective which is never produced and columnar

style. I have not observed the latter in either S. persica or S. oleoides. The persistent

indumentum on branches, leaves and floral parts is a character by which the southern

African species may be at once recognised.

153. LOGANIACEAE.

1. Strychnos Schumanniana Gilg

On rocky ledges behind homestead, 64.

S. innocua Del. (S. Quaqua Gilg; S. dysophylla Beuth.).

Occasional on northern slopes of mountain, 161.

2. Anthocleista zambesiaca Bak.

In kloof behind homestead, 115.

This is probably one of the most conspicuous of South African forest trees, always

growing in very damp situations, usually in ravines. It is more or less 30 metres high

and is not much branched. The branches are erect and bear a crown of large simple leaves

near their apex. The flowers are gardenia white.

3. Lachnopylis montana C. A. Smith

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 344.

155. APOCYNACEAE.

1. Landolphia Kirkii Dyer var. delagoense Dew.

Upper slopes of mountain, fairly common; shrub, 1-5 metres high, 304, 172.

2. Pachypodium Saundersii N. E. Br.

Near roadside among rocks in Wylie’s Poort, 668.

3. Strophanthus Gerrardii Stapf

Frequent in bush, west of homestead, 579. This interesting liane is fairly common

in the dense bush at the foot of the mountain. It climbs to the tops of tall trees and not

much of its foliage is visible. The two-winged, twisted stems, however, attract one’s

attention immediately (two opposite wings run vertically throughout the length of the

internode with the plane of the wings at right angles to those of the internodes above and

below).

156. ASCLEPIADACEAE.

1. Cryptolepis obtusa N. E. Br.

Twiner in Cassia, in bush west of homestead, 591.

2. Stomatostemma Monteiroae (Oliv.) N. H. Br.

On upper slopes of mountain; lane on Commiphora sp., 320.

3. Asclepias Burchellii Schltr. $e

In open veld near homestead, 23.

4. Secamone Gerrardi Harv.

Farm “ Elsteg,”’ about 6 miles west of Louis Trichardt, 352.

250

S. zambesiaca var. parvifolia N. H. Br.

Twiner on Huphorbia Cooperi, east of homestead, 138.

5. Ceropegia cimiciodora Obermeyer

Farm “ Chapudi,’ between Waterpoort and Zoutpan, creeping in low bushes, 322.

A fleshy-stemmed creeper with striking flowers.

6. Tavaresia grandiflora (K. Schum.) Berger

Between Waterpoort and Zoutpan, in sandy soil under cover of Acacia scrub, 413.

7. Caralluma atrosanguinea N. EK. Br.

Farm “ Chapudi,” between Zoutpan and Waterpoort, 446.

C. Schweickerdtii Obermeyer sp. nov. (C. carnosa Schweickerdt in Flowering Plants of South

Africa, XV, plate 592, non Stent).

Caules carnosi, erecti vel adscendentes, basi ramosi, 4-angulati, 17 em. alti, 3-4-5 cm.

diam. (dentibus inclusis), glabri, virides, brunneo-maculati; anguli dentibus patentibus.

deltoideis, 1-5 cm. longis, compressis instructi. Flores ad apicem ramorum, 1-3 aggregati.

Pedicelli ad 3 mm. longi. Sepala 4 mm. longa, lanceolata, acuta, sini flagellis instructi.

Corolla 1 cm. longa, 1-2 cm. in diam., campanulata, extus glabra, viridia, intus purpureo-

rubra, paullum cremeo-maculata, papillosa, aliquando pilis bulbosis, clavatis indutis ;

tubus annulo pentagono, parvo, instructis ; lobi deltoidei, 5 mm. longi, 7 mm. lati, paullum

patentes. Corona exterior crateriformis, lobi aliquando lyrati, patentes, ad apicem paullum

concavi vel bifidi roseo-cremei; lobi interior supra antherarum incumbenti, maculati,

dorso, carnoso-gibbosi.

TRANSVAAL.—Zoutpansberg distr.: Farm “Chapudi,’ near Waterpoort, in sand,

Obermeyer, Schweickerdt and Verdoorn, 411 (National Herbariwm, Pretoria, 19597; Herb.

Transvaal Museum, 34945).

This species is closely related to C. Keithit Dyer but may be distinguished as follows :

In C. Keithii the corolla tube is thin and minutely verrucose, different to the fleshy, densely

papillate lobes. There is, however, no sharp distinction between the corolla-lobes and the

tube in C. Schweickerdtii. The very peculiar bulbous-based clavate unicellular hairs tipping

some of the long papillae are smaller and fewer, while the margin of the corolla-lobes is

not fringed with a row of minute hairs; there are, however, some scattered over the whole

inner surface. The outer corona-lobes of C. Schweickerdtii bend outwards and are more

or less lyre-shaped, while in C. Keithii they are bifurcate and somewhat incumbent over

the staminal column. The inner corona lobes are mottled with purple, while the minute

emarginate lobe, spreading slightly between the outer corona-lobes, present in C. Keithi,

is absent. The five intersepalar flagellae are present in both species, but absent from C.

carnosa Stent. From this species it may be distinguished by the different colour and smaller

size of the corolla, ete.

Between Waterpoort and Zoutpan, 411. This specimen was figured in Flowering

Plants of South Africa (plate 592) as C. carnosa Stent; Miss Obermeyer, however, redis-

covered CO. carnosa in the type locality and found that our plant differed considerably from

it. She has therefore given the above name to our specimen.

C. maculata N. E. Br.

Near Waterpoort, fairly frequent but scattered, 660. This species is rather remark-

able as it has horizontal rhizomes at a depth of about 10 cm. below the surface of the soil.

Aerial shoots arise from these at some distance from each other. This plant was figured

in Flowering Plants of South Africa as C. grandidens Verdoorn but was found by Dr.

Schweickerdt at Kew to be C. maculata N. KE. Br. (See Kew Bull., 1935.)

8. Stapelia clavicorona Verdoorn

Growing on rocky ledges in Wylhe’s Poort, 414. Apparently a very rare species since

only a few plants have so far been found and only from that locality.

251

S. Getlieffii Pott

Along margin of pan, fairly frequent, 477: between Zoutpan and Waterpoort, fairly

common, 444.

The colour of the flowers of this species is rather variable. Flowers with dark wine-

coloured markings and hairs on the corolla and others with very pale, almost yellow corollas

and but faint markings were observed on plants growing in the same patch. The inner

corona wings also vary but this variation is not constantly associated with the variations

in the colour of the corolla.

S. gigantea N. EH. Br.

On rocky ledge on slopes behind homestead, 70.

S. nobilis N. E. Br.

Growing in large patches under a large Acacia tree near Zoutpan, 445.

S. kwebensis N. E. Br.

At “ Vivo,” west of Zoutpan, 406: between Zoutpan and Waterpoort, 662.

9. Huernia zebrina N. E. Br. var magnifiora Phillips.

Near farm “Chapudi,” between Zoutpan and Waterpoort, 416, 443, 657, 658, 659.

Colour forms were noted among these specimens such as black disk and yellow lobes marked

with red stripes, and red disk and red lobes with yellow stripes. The outer corona varied

too, sometimes appearing pincher-like and on flowers in close proximity (not obviously

the same plant) with outer corona not pincher-like.

10. Pergularia extensa (Jacq.) N. EH. Br.

Climber, on shrubs and bushes, fairly frequent in sandy areas around pan; petals

green, with densely ciliate margins, 3, 90.

11. Fockea augustifolia K. Schum. ?

In Catophractes belt, north of pan, 187 (poor specimen).

157. CONVOLVULACEAE.

1. Seddera suffructicosa (Schinz) Hall. f. var. hirsutissima Hall. f.

In sandy soil of Catophractes belt, 177, 509.

2. Ipomoea adenioides Schinz.

Fairly frequent in Catophractes belt; flowers exquisite, white with wine-coloured

throat, opening at sundown, 181.

I. quinquefolia var. purpurea Hall. f.

On sandy soil, climbing over low scrub, between Louis Trichardt and Pietersburg, 665.

I. Lugardi var. parviflora Rendle

In shade of Acacia, east of pan; flowers bluish-pink with dark pink throat, 542.

3. Merremia pinnata (Hochst.) Hall. f.

On north-east side of pan, in shade of Acacia; flowers bright yellow, 568.

[The specimens from Trop. Afr. in Herb. Kew are more robust and more pubescent

than the above gathering (H.G.S.).]

159. BORAGINACEAE.

1. Cordia ovalis R. Br.

Small tree, in Catophractes belt, very occasional, 192. First record of the species for

the Transvaal.

252

2. Ehretia rigida (Thunb.) Druce

North side of pan, fairly frequent, 44, 45.

3. Heliotropium curassavicum L.

On very margin of pan, a typical halophite, 288, 465.

H. lineare C. H. Wright

Between Zoutpan and Waterpoort, in open sandy soil, 248.

H. Nelsoni C. H. Wright

Very occasional under Acacia, near margin of pan, 20, 482, 510. |

160. VERBENACEAE.

1. Lippia asperifolia Rich.

In Lonchocarpus belt and on northern slopes of mountain, 7, 129, 519, 520.

2. Clerodendron myricoides R. Br.

At upper end of kloof behind homestead, 602. The leaves of this specimen are more

coarsely serrate than in the material from Trop. Africa.

C. ternatum Schinz

Very sandy area, about 3 miles west of pan, 641.

C. simile Pearson

Undershrub, on farm “‘ Chapudi,” between Zoutpan and Waterpoort, 252.

161. LABIATAE.

1. Leonotis dysophylla Benth.

Under tree in vlei area on slopes above pan, up to 3 metres high, 229.

2. Leucas glabrata R. Br. ark

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 364.

L. sexdentata Skan

In shade and in open veld near margin of pan, 491.

3. Aeolanthus Rehmannii Guerke

On rocky slopes, rooting in fissures, Wylie’s Poort, 670.

4, Endostemon tereticaulis (Poir) Ashby (H. ocimoides Bremekamp).

Frequent among loose stones on eastern margin of pan; flowers purple, 140.

5. Pyenostachys reticulata Benth.

In shade along bank of stream, behind homestead ; flowers a beautiful blue, 518.

P. densiflorus Cooke

In sandy soil, between Louis Trichardt and Pietersburg, 447.

6. Ocimum americanum L.

Near margin of pan; flowers small, pink and white, 494.

7. Becium obovatum N. EH. Br.

Near margin of pan under cover of Acacia ; flowers dirty white with mauve markings ;

stamens long exserted, 475.

8. Hemizygia canescens (Guerke) M. Asbhy (Orthosiphon canescens Guerke).

On slopes of mountain behind homestead ; small bush with strong odour, 613.

253

162. SOLANACEAE.

1. Solanum incanum L.

Near margin of pan, growing in association with Acacia scrub, 134.

§. panduriforme E. Mey.

East fringe of pan, common, 547.

S. kwebense N. EH. Br.

In Catophractes belt; flowers white, 54: flowers purple, 57. First record of this

species for the Transvaal.

2. Lycium sp.

Five collections were made (22, 39, 42, 474, 503) which have not been specifically

identified and a revision of the genus appears very necessary.

168. SCROPHULARIACEAE.

1. Aptosimum lineare Marl. and Engl. ;

In sandy soil, between Louis Trichardt and Pietersburg, 666: between Zoutpan and

Waterpoort in open sandy soil, 266: Vivo, 655.

A. patulum Bremekamp

Under Acacia, east of pan, 565: at foot of mountain, near upper vlei, 581.

2. Peliostomum leucorrhizum EH. Mey.

Farm “‘ Kyem,” north of Blaauwberg, 94.

3. Limosella maior Diels

In stream leading from kloof and vlei, 216.

. Ilysanthes dubia (L.) Bernh. (J. capensis Benth; I. riparia Raf.).

Near vlei behind homestead; flowers white, 217.

5. Ramphicarpa tubulosa (Linn. f.) Benth.

In grassy patches on slope of mountain, 233.

6. Striga gesnerioides (Willd.) Vatke (iS. orobanchoides Benth.).

Growing under cover of and parasitic on the roots of Euphorbia Cooperi, east of home-

stead, 127.

164. BIGNONIACEAE.

1. Rhigozum obovatum Burch.

Frequent in Catophractes belt ; flowers yellow, 53.

R. zambesiacum Bak.

Between Waterpoort and Wylie’s Poort, 340.

2. Catophractes Alexandri Don.

Dominant in a belt north of pan and owing to this the term ‘‘ Catophractes belt ”’ is

spoken of in this paper, 514, 58.

165. PEDALIACEAE.

1. Pterodiscus ngamicus N. H. Br.

Along sandy western margin of pan, 427. (A poor plant of what might be another

species of Pterodiscus was observed).

2. Harpagophytum Zeyheri Decne ?

In sandy soil, west of pan, 689. The specimen is not in fruit and hence it is impossible

to name it specifically.

3. Sesamothamnus Lugardii N. E. Br.

Frequent in very sandy area, north of pan, 59. Comparison of our specimen with

fruiting material from South West Africa named S. Seineri Engl. in Herb. Hort. Bot. Berol

(see Engl. and Drude, Veg. der Erde, 9, I, 2, p. 586 and tab. 28, 2 (1910)] suggests that these

two species are the same. Flowering material of S. Sevmeri is necessary to decide this

question. In any case S. Lugardw (1906) has priority.

3. Sesamum capense Burm.

East of pan, occasional, 557.

4. Ceratotheca triloba EK. Mey.

Very frequent at foot of mountain below upper vlei, 583.

5. Pretraea zanguebarica Gay

Along south-eastern margin of pan, frequent in patches, 690.

169. LENTIBULARIACEAE.

1. Utricularia exoleta R. Br.

Plentiful, but scattered in vlei behind homestead, 209.

170. ACANTHACEAE.

1. Dyschoriste Fischeri Lindau

Shrub, up to 1 metre high, between Waterpoort and Wylie’s Poort, 333.

2. Ruellia patula Jacq.

Margin of pan, under cover of small Acacia; flowers white, 28, 467: between Zout-

pan and Waterpoort, in shade of shrubs, 250.

3. Barleria Bremekampi Obermeyer

Very spiny bush, up to | metre high, in kloof behind homestead, 124.

B. elegans 8S. Moore

Foot of mountain, near Huphorbia Cooperi, 130.

B. Galpinii C. B. Cl.

In kloof behind homestead, 123.

B. heterotricha Lindau

In kloof behind homestead, 125.

B. obtusa Nees

On northern slopes of mountain, 162.

B. transvaalensis Obermeyer

In Acacia belt, north of pan, near Sesamothamnus, 50.

4. Neuracanthus africanus T. Anders. ex Sp. Moore

Between Zoutpan and Waterpoort, fairly frequent among Acacia scrub, 251.

5. Blepharis Clarkei Schinz

South-eastern side of pan, in stony surroundings, 291.

B. diversispina (Nees) C. B. Cl.

Farm Zoutpan, 25la.

6. Asystasia atriplicifolia Bremekamp

On farm “ Hyem” near large baobab, on north-eastern boundary, 89.

255

7. Ruspolia erypocrateriformis (Vahl.) Milne-Redhead var. australis Milne-Redhead

Wylie’s Poort; shrub, with arching branches, 10 ft. long, hanging over stones;

flowers scarlet ; fruit green, turning black, 441.

8. Dicliptera clinopodia Nees

Under Acacia, near eastern margin of pan, 297.

9. Justicia flava Vahl.

Western margin of pan, in gravelly soil; flowers yellow ; stems procumbent, 21.

J. odora Vahl.

Frequent in Catophractes belt, up to 60 cm. high, 190.

J. (Calophanoides) sp. .

On ledge above kloof, behind homestead, 610.

J. (Ansellia) sp.

Above bushman cave on upper slopes of mountain, 168.

173. RUBIACEAE.

1. Oldenlandia cephalotes (Hochst.) O. Kuntze (O. sphaerocephala Schinz).

In vlei behind homestead, 201.

[Comparison of the types of O. cephalotes and O. sphaerocephala has convinced me that

they represent one species. The leaves are somewhat variable in shape and size as is very

often the case in species of Oldenlandia (H.G.S.).]

O. decumbens (Hochst.) Hiern

In shady forest patches at foot of mountain, 592.

2. Randia sp. near R. rudis EK. Mey.

Foot of hill, below a group of Euphorbia Cooperi, 128.

3. Gardenia Neuberia E. and Z.

Small tree, among rocks at roadside in Wyle’s Poort, 667.

G. spatulifolia Stapf and Hutch.

Occasional on northern slopes of mountain; tree, about 5 metres high, 158.

4. Empogona Kirkii Hook. f. var. australis Schweickerdt var. nov. <A typo foliis multo

matoribus glabrisque, pedicellis ovario et calyce multo minus pubescentibus differt.

In Lonchocarpus belt, 72 (type) ; on lower northern slopes of mountain behind home-

stead ; flowers white, densely bearded in the throat, 528.

[No 72 was collected durmg November, i.e. early summer and therefore shows the

presence of well-developed young leaves which are of a thin texture. In shape and size

those agree with leaves of 528 collected during April, which, however, are leathery in

texture and somewhat shiny on the upper surface. Indumentum of branches, pedicels,

ovary and calyx-lobes are similar in both gatherings (H.G.S.).]

5. Vanguera tomentosa Hochst.

On northern slopes of mountain, 691.

V. cyanescens Robyns

On northern slopes of mountain, 102.

[Apart from the fact that Dinter 58 (type of V. cynanescens) has turned partly black

on drying (indigo ?), there appears to be no difference between this species and V. flori-

bunda Robyns. Rogers 18214 (type of the latter) is a fairly robust specimen which is

slightly more hairy in the inflorescence than our gathering. A wider range of material

will probably prove the above-mentioned species to be identical (H.G.S.).]

6. Canthium ventosa (L.) Sp. Moore

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 359.

C. huillense Hiern

Northern slopes of mountain, 173: in kloof behind homestead, 122.

7. Pavetta Harborii Sp. Moore

Between Zoutpan and Waterpoort, locally abundant, 255. —

P. Schumanniana F. Hofim. ex K. Schum.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 357.

8. Plectroniella armata (K. Schum.) Robyns

In Lonchocarpus belt, west of homestead; tree, 5-6 metres high; flowers bearded

in throat, 26.

9. Anthospermum lanceolatum Thunb.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 562. This gathering is some-

what more pubescent than the typical plant.

176. CUCURBITACEAE.

1. Corallocarpus sphaerocarpus var. scaberrimus Cogn.

On very margin of pan, climbing in Acacia, 30.

2. Momordica Balsamina L.

Climbing in and over small Acacia in Lonchocarpus belt, 14.

3. Citrullus naudinianus (Sond.) Hook. f.

Frequent in very sandy area, about 3 miles west of pan, known locally as “ gemsbok-

komkommer,” 646.

C. vulgaris Schrad.

Margin of pan, not very frequent, 455.

4. Cucumis africanus L. f. var. Zeyheri Burtt Davy

In drier parts of vlei behind homestead, 213.

C. hirsutus Sond.

Farm “ Gaanspan,” north of Blaauwberg, 245: between Zoutpan and Waterpoort.

258: farm “ Kyem,” north of Blaauwberg, 85. Reported to be a medicinal plant:

C. myriocarpus Naud.

Along roadside, between Louis Trichardt and Pietersburg, 448.

5. Coccinia Rehmannii Cogn.

Climbing in Salvadora, Lonchocarpus belt, 40.

C. sessilifolia (Sond.) Cogn.

Farm “ Kyem,” north of Blaauwberg, 91.

177. CAMPANULACEAE.

1. Lobelia decipiens Sond.

In vlei behind homestead, fairly common, 214.

257

179. COMPOSITAE.

1. Vernonia amygdalina Del. (V. Randii Sp. Moore).

Along watercourse leading from kloof behind homestead; about 6-7 metres high,

lounging, 116.

V. cinerascens Sch. Bip. (V. Luederitziana O. Hoftm.).

Between Zoutpan and Waterpoort. shrub, 60-100 cm. high, growing in exposed sandy

soil, 249. First record of this species for the Transvaal.

[This plant has a wide distribution in Africa, occurring in Eritrea, Somaliland, Angola,

Great Namaqualand and N. Transvaal. V. Lwederitziana undoubtedly belongs to the same

species. Our gathering differs from the Abyssinian plant only in having more robust and

shorter peduncles (H.G.S.).]

V. fastigiata O. and H.

In low thorny scrub near Catophractes belt, 516.

2. Ageratum conyzoides L.

Along stream leading from large vlei, west of homestead, 227.

3. Eupatorium africanum O. and H.

Farm “ Elsteg,’ about 6 miles west of Louis Trichardt, 342a.

4. Aster luteus (N. EK. Br.) Hutch. forma.

Between Zoutpan and Waterpoort, frequent in sandy soil, 256: Vivo, in sandy soil

656.

Our specimens are very similar to the typical form of A. luteus from northern Natal,

differing from it in having blue ray-florets and slightly more pubescent achenes. The

capitula of the Natal plants appear to be generally somewhat smaller. The wide range

of material at the Transvaal Museum, however, seems to indicate that our specimens are

merely forms of a variable species.

5. Psiadia arabica Jaub. and Spach.

Between Waterpoort and Wylie’s Poort, 334.

6. Nidorella resedifolia DC.

In shade of trees near margin of pan, 495: between Zoutpan and Waterpoort, 272.

7. Brachylaena sp., probably B. transvaalensis Phill. and Schw.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 349.

8. Blumea caffra (DC.) O. Hofim. (B. natalensis Sch. Bip.).

In open soil east of pan; heads globose; florets pinkish, 550.

B. lacera DC.

Farm “ Elsteg,’ about 6 miles west of Louis Trichardt, 205.

9. Pluchea leubnitziae (O. Hofim.) N. HK. Br. (Pechuelloeschea Leubnitziae Hoftm.).

In Catophractes belt, north of pan, 55.

10. Epaltes alata Steetz.

In shade of trees near margin of pan, 501; north-east of pan, in grassland, 301.

11. Helichrysum Kraussi Sch. Bip.

Farm “ Elsteg,” about 6 miles west of Louis Trichardt, 345.

12. Pegolettia senegalensis Cass.

In open soil between stones near margin of pan, 502.

258

13. Geigeria aspera Harv. forma ?

Vivo, frequent in sandy soil, 651: in sandy soil near Catophractes belt, 507

Zoutpan and Waterpoort, 273.

14. Senecio polyanthemoides Sch. Bip.

Scattered, in vlei behind homestead, 203.

S. transvaalensis Bolus

Between Louis Trichardt and Pietersburg; flowers pink, 674.

15. Kleinia longiflora DC.

East of pan among species of Acacia, 537.

16. Berkheyopsis bechuanensis Sp. Moore

Fairly frequent in very sandy area, about 3 miles west of pan, 638.

: between

259

THE GENUS ELYONURUS Humb. and Bonpl.

IN SOUTH AFRICA.

By E. P. Puitirps, M.A., D.Sc. and H..C. Breprtzt, M.Sc.

SYSTEMATIC (By E. P. Puiturps).

THE genus Elyonurus in known by about twelve species recorded from South and North

America, America, Australia, Arabia, Persia and Africa. In the year 1841 Nees described

two species (E. argenteus and HE. thimodorus) from the eastern and north-eastern districts

of the Cape Province. Stapf in the “ Flora Capensis ” recognised only one species and

reduced the second species to a variety thymiodora.

I recently had occasion to examine some fresh specimens collected near Pretoria and

which did not conform to the description in the “* Flora Capensis.” As I suspected it to

be different from the plant named #. argenteus, an examination of all the material in the

National Herbarium was undertaken. The result of that examination is that I recognise

three species as occurring in South Africa. The species #. argenteus Nees is common in

the eastern and north-eastern areas of the Union; the plant typical of the western Trans-

vaal and previously referred to H. argenteus, has been named EF. glaber. It is characterised

by the glabrous leaf-sheaths but a form with villous leaf-sheaths is met with and has been

designated var. villosus. The third species (#. pretorvensis) I have only seen from Pretoria ;

it is characterised by having a palea present and the peduncled spikelet invariably bisexual.

Mr. C. E. Hubbard of the Kew Herbarium, to whom some specimens were referred,

is not very convinced that the specimens are specifically distinct and due consideration

has been given to his views. The distribution of the species as recognised in this paper

is distinct ; certain morphological characters are constant; the leaf-anatomy as detailed

by Mr. Bredell is distinct. For the above reasons there appears every justification for

separating the South African plants of Hlyonurus into three species.

I am indebted to the director of the Royal Botanic Gardens, Kew, for sending me on

joan, three herbarium sheets from the Kew Herbarium and to Mr. C. BE. Hubbard for com-

ments he made on specimens sent to him.

KEY TO SPECIES.

1. Pale, present in both spikelets as a hyaline scale; peduncled spikelet

TILT Diver DISCMUIO eae ee ee ese eae Mun aoe ATL ae 3. pretoriensis.

Pale, absent in both spikelets, very rarely present ; peduncled spikelets,

male 2

nicteMaMeMatelcieleielelieliel «Ne alrelielellei{siiell sire) ell=iiucteliellesjol level lvllelisife| (a) jolie) is) (0) (wflehielile el») \le))«) le) eo) oillejiai’apie)) Al

2. Sessile and peduncled spikelets of equal lengths ; lower glume of sessile

spikelet less than 1 cm. long, 2-toothed or with awns 1—2-25 mm. long,

TALC hyeslOUC Chere errs ee cia ee a EeK Sich ans cioikens’cae tone Fn wnU ba es 1. argenteus.

Sessile spikelet longer than the peduncled spikelet; lower sessile

glume of spikelet 1 cm. or more long, rarely shorter, with awns 4-6-5

Cle Oem LAL WeISHONCCE tcc ocr co eee ete a aes Shek 3. glaber.

260

1. E. argenteus Nees

Plants 25-100 cm. high, forming dense clumps. Basal leaf-sheaths persistent, some-

what villous. Blades green, 10-27 cm. long, 0-5 mm. broad, usually somewhat curled,

keeled so that leaf is almost 3-angled in cross-section, ciliate below, glabrous. Culm bearing

inflorescence 2-noded, hairy at nodes. Inflorescence 3-8 em. long. Spikelets of equal

lengths. SrssiLE SprkeLer.—Lower glume 5-5-9-5 mm. long, 1-5-2 mm. broad, usually

lanceolate, rarely ovate-lanceolate, acuminate, usually 2-toothed, more rarely produced

into short awns 1—2-25 mm. long, with a dark band round the margins, w'th the margins

narrowly inflexed and narrowly keeled, long ciliate on the keels, sometimes cilia from

tubercules, usually 9-nerved, more rarely 5—7-nerved, villous on the back. Upper glume

4-5-7 mm. long, 1-1-5 mm. broad, lanceolate, acute, deeply concave, 3-nerved, keeled on

back, shortly ciliate, pubescent on the back. Lower valve 3-5 mm. long, 0-75 mm. broad,

lanceolate, acute, flattish on the back, 2-nerved, very rarely 3-nerved, ciliate above on

hyaline marginal flaps. Upper valve 3-5 mm. long, 0-75 mm. broad, lanceolate, concave,

3-nerved, ciliate on hyaline marginal flaps. Anthers 3-5 mm. long, linear. Ovary ellipsoid ;

styles free; stigmas about twice as long as styles. Lodicules fan-shaped or triangular,

truncate. PEDUNCLED SpiIKELET.—Peduncle 2-5-4 mm. long, hollow, obtusely 3-angled,

villous, Lower glume 3-5-7 mm. long, 1-1-75 mm. broad, lanceolate, long acuminate,

sub-acuminate, 2-awned, 2-toothed, or with a small lateral tooth, with one margin narrowly

inflexed and narrowly keeled, 5—7-nerved, more rarely 8-9-nerved, ciliate on the keel, with

the cilia sometimes from tubercules, pilose or villous on the back. Upper glume 3-5-6-5

mm. long, 0-75-1-25 mm. broad, lanceolate, acute or sub-acute, deeply concave, rounded

on the back, usually 3-nerved, rarely 4-5- or 7-nerved, keeled, ciliate on hyaline marginal

flaps, usually shortly ciliate on keel, pubescent or shortly pilose on back, rarely glabrous.

Lower valve 2-5-4 mm. long, 0-75—-1-25 mm. broad, lanceolate, flattish on the back, 2-nerved,

very rarely 3-nerved, ciliate on hyaline marginal flaps. Upper valve 3-5-4 mm. long,

rarely shorter, 0-5-1 mm. broad, lanceolate, deeply concave, rounded on back, 3-nerved,

ciliate on marginal hyaline flaps. Pale, a hyaline fimbriated scale (see only in one specimen).

Lodicules fan- “shaped or triangular, truncate. H. thimiodorus Nees, Fl. Afr. Austr., 95 ;

E. argenteus var. thymiodora Stapf i in Fl. Cap, vol. 7, p. 333; EH. argenteus Nees ex Fl. Cap.

Lc. partly.

CarE Province.—Humansdorp distr. : Witte Elsbosch, 750 ft., April, Fourcade, 2542.

Albany distr. : Grahamstown, Oct., Daly and Sole, 108 ; Howison’s Poort, 2,200 ft., Dee.,

Galpin, 3094; Trapp’s Valley, Dec., Daly, 714. Kingwilliamstown distr.: Amatola

Mountain, 4,000 ft., Dec., Dyer, 260; Kei Road, 2,000 ft., Habe. Ranger, 50. Stutterheim

distr.: Blaney Junction, 1,200 ft., Jan., Galpin, 5622. Shodkenctnoain distr.: Katberg,

Dec., Sole, 405. Queenstown distr.: Effingham Peak, Katberg, 5,700 ft., Dec., Galpin,

8393; Roode Rand farm, 3,550 ft., Nov., Galpin, 2510; Hangklip Mountain, 5,400 ft.,

Febr., Galpin, 5859. Komgha distr.: Near Komgha, 2,000 ft., Sept., Flanagan, 897.

Kentani distr.: Near Kentani, 1,200 ft., Nov., Pegler, 1386. solo distr.: Idutywa,

2,500 ft., Jan., Schltr., 6273; Bazeia, Basa. 284. Barkly Hast distr.: Near Barkly East,

Febr., Greyvenstein, 12. Mount Currie distr.: Hills round Kokstad, Nov., Goossens, 222 ;

Mogg, 4826.

_ Orance Free State.—Ficksburg distr. : Riverhill Farm, high up on mountain slopes,

Jan., Potts in Grey Univ. Herb., 3706, 3721. Senekal distr. : Common on upper slopes of

mountains near Doornkop, 5,300 ft., Dec., Goossens, 709 ; lava soil on top of mountain at

Wonderkop, Dec., Goossens, 825.

Basurotanp.—Drakensbergen, Stokoe in Nat. Herb., 8342; Thabuing, Jan., Watt and

Brandwyk in Nat. Herb., 8763; Leribe, Dieterlen, 177.

Narau.—Pietermaritzburg distr. : Cedara, Dec., Phillips in Nat. Herb., 20570. Lion’s

River distr.: St. Ives, Oct., Mogg, 5661; Balgowan, Nov., Mogg, 3541; near Howick,

3,600 ft., Nov., Mogg, 3502. Impendhle distr.: Giant’s Castle, 8,000-9,000 ft., Oct.,

261

Wood, 10543. Estcourt distr.: Mont-aux-Sources, 10,000 ft., Feb., Bayer and McClean,

318; Dec., Schweickerdt in Nat. Herb., 20573 ; Mooi River, Oct., Mogg, 3063, 3277.. Berg-

ville distr.: Mt. Twinta, Jan., Doidge in Nat. Herb., 20571; Acton Homes, Jan., Doidge

in Nat. Herb., 20572.

TRANSVAAL.—Heidelberg distr.: “Heidelberg, Dec., Burtt Davy, 3147. Standerton

distr.: Near Standerton, Jan., Burtt Davy, 3083. Ermelo distr.: Near Ermelo, Jan.,

Burtt Davy, 952; Febr., Henrici, 1209. Carolina distr.: Vlei on town lands near Carolina,

Jan., Pellissier in Grey Univ. Coll. Herb., 4609; Dec., Burtt Davy in Gevt. Herb., 7364.

Middelburg distr.: Botsabelo, Dec., Fouche in Nat. Herb., 20574.

SwazILanp.—Near Bremersdorp, Burtt Davy, 3048.

A specimen (Rehmann, 5672) collected at Houtbosch, in the Pietersburg district of the

Transvaal is probably this species but the material is too poor for any examination.

2. E. glaber Phillips sp. nov. Affinis E. argenteus sed foliis basin versus glaber differt. E,.

argenteus Nees ex Fl. Cap., vol. 7, p. 332 partly.

Culmi 42-70 em. alti. Folia 15-36 cm. longa, 2-3 mm. lata, carinata, basin versus

glabra. Racemi 7-14 cm. longi, Spiculae sessiles hermaphroditae, 1-1-4 cm. longae.

Gluma inferior, lanceolata, 6—-8-nervata, profunde 2-fida, dense villosa. Spiculae pedun-

culatae 0-65-1 cm. longae. Pedunculus 2-5-4 cm. longus, villosus. Gluwma inferior

6-9-5 mm. longa, lanceolata, acuminata vel profunde 2-fida, pilosa vel villosa.

BasuTotanp.—Likhoele, March, Dieterlen, 1097.

ORANGE FREE Stare.—Draaifontein (no precise locality), Rehmann, 3658. Senekal

distr.: Near Senekal, Dec., Goossens, 956. Heilbron distr.: Heilbron, Jan., Goossens,

444. Hoopstad distr.: Wesselsbron, Jan., Goossens, 1243. Kroonstad distr.: Kroonstad,

Sept., Pont, 498.

Care Province.—Vryburg distr.: Armoed’s Vlakte, Febr., Viljoen in Nat. Herb., 77 ;

Theiler in Nat. Herb., 20675; Klipvlakte, Nov., Burtt Davy, 11131.

TRANSVAAL.—Marico distr.: Derby Station, Nov., Burtt Davy in Govt. Herb., 7168.

Bloemhof distr.: Kameelpan, near Christiana, Jan., Theron, 628; Christiana, Mch., Burtt

Davy im Govt. Herb., 14126; Cawood’s Hope, Mch., Burtt Davy, 12953. Wolmaransstad

distr.: Boskwl, May, Sutton, 114. Ventersdorp distr.: Ventersdorp, Mch., Pole Evans,

3139. Potchefstroom distr.: Welverdiend Station, Mch., Burtt Davy, 14569. Johannes-

purg distr.: Turffontein, Mch., Bryant, D.48; Johannesburg, July, Hitchcock, 24141.

Pretoria distr.: Near Pretoria, Febr., Skea, 3, 71; Lvebenberg, 3241 (typus), 3211; Mogg

in Nat. Herb., 20577; Onderstepoort, Du Toit, 28. Waterberg distr.: Springbok Flats,

Oct., Burtt Davy, 7067.

Var. villosus Phillips, Folia basin versus pilosa vel villosa.

ORANGE FREE Srate.—Bothaville distr.: Bothaville, Jan., Goossens, 1182.

Cape Provrnce.—Kuruman distr.: Near Kuruman, Dec., J. W. Mogg, 7627.

TRANSVAAL.—Bloemhof distr.: Near Christiana, Nelson, 65. Vereeniging distr. :

“ Weltevrede,” Dec., Cronje, 55. Johannesburg distr.: Johannesburg, Moss, 6852, (a

hyaline pale was found in Moss, 6852); Hlsburg, Jan., Rogers, 12135. Benoni distr. :

Benoni, Sept., Bradfield, T.187. Pretoria distr.: Premier Mine, Oct., Moss, 5451; near

Pretoria, Dec., MacDonald in Govt. Herb., 5441; Hartebeestnek, Nov., Burtt Davy, 758 ;

Wonderboom, near Pretoria, Rehmann, 4491. Z

Plants 42-70 cm. high, forming dense tufts ; new shoots from nodes on a very short

rhizome. Basal leaf-sheaths persistent, reddish, glabrous. Blades bright or dark green,

262

15-36 em. long, 2-3 mm. broad (when fresh), keeled on back, with 5-6 nerves on either side

of the mid-rib, ciliate at juncture with the sheath, otherwise glabrous. Ligule a narrow

lacerated rim, about 0-5 mm. broad. Culms simple, rarely branched, 2-5-noded, with the

nodes reddish and slightly swollen; upper internode 15-26 cm. long. Inflorescence 7-14

cm. long. Sessile sprkelet longer than the peduncled spikelet, very rarely as long. SESSILE

SPiKELET.—Lower glume 1-1-4 cm. long, rarely less than 1 cm. long, 0-15-0-175 cm. broad,

lanceolate, usually acuminate, produced into two ciliate awns 4-6-5 mm. long, rarely

shorter than 4 mm., with the margins narrowly inflexed, narrowly keeled and long ciliate

on the keels, 6—8-nerved, villous on the back. Upper glume 5-5-7-5 mm. long, 0-1—0-125

em. broad, lanceolate, usually subacuminate, acute, deeply concave, 3-nerved, keeled on

the back, pilose, more rarely pubescent on the back. Lower valve 5-6 mm. long, 0-75-1-25

mm. broad, lanceolate, flattish on the back, 2-nerved, ciliate above on hyaline marginal

flaps. Upper valve 4-4-75 mm. long, 0-75-1 mm. broad, lanceolate, concave, rounded on

the back, 3-nerved, cilate above on hyaline marginal flaps. Anthers 2-5-3-5 mm. long,

lmear. Ovary ellipsoid ; styles free ; stigmas twice as long as the styles. Lodicules fan-

shaped or triangular, truncate, somewhat fleshy. PEDUNCLED SprkELET.—Peduncle 2-5-4

mm. long, hollow, villous. Lower glume 6-9:5 mm. long, 1-1-25 mm. broad, lanceolate,

usually long acuminate, rarely 2-awned and then awns |1-5-3-5 mm. long, with one margin

narrowly inflexed and narrowly keeled on-same margin, rarely both margins narrowly

keeled, 5-nerved, rarely 6-nerved, ciliate on the keel, pilose or villous on the back. Upper

glume 5-8 mm. long, 0-75—-1 mm. broad, lanceolate, acummate, more rarely awned, concave,

rounded on the back, 3-nerved, ciliate on hyaline marginal flaps, sparsely pilose on the back

or pubescent with a few long hairs. Lower valve 5-6-5 mm. long, 0:75-1:25 mm. broad,

lanceolate or lmear-lanceolate, usually flattish on the back, more rarely shghtly concave,

2-nerved, ciliate on hyalme marginal flaps. Upper valve 3-5-5 mm. long, 0-5-1 mm.

broad, lanceolate, concave, rounded on back, 3-nerved, ciliate on upper margins. Anthers

2-75-3-5 mm. long, lmear. Lodicules triangular or fan-shaped, truncate, somewhat fleshy.

Var. villosa. Leaf sheaths pilose or villous.

3..K. pretoriensis Phillips sp. nov. Affinis #. argenteus sed spiculus pedunculatis herma-

phroditis differt.

Culmi ad 42 em. alti. Folia ad 19 cm. longa, 3-5-4 mm. lata, carinata, basin versus

cihata, villosa ; ligulae ad marginem breviter dense ciliatam redactae. Racemi 7-5-9:5 em.

longi. Spiculae sessiles hermaphroditae, 1-05-1-25 cm. longae. Glwma inferior, lanceo-

lata, 6-9-nervata, profunde 2-fida, dense villosa. Spiculae pedunculatae hermaphroditae,

0-65-1 em. longae. Gluma inferior, lanceolata, 5—9-nervata, acuminata, pilosa vel villosa.

Pedunculus 1-3-5 mm. longus, villosus.

TRANSVAAL.—Pretoria distr.: Meintjes Kop, Pretoria, Oct., Lansdell in Govt. Herb.,

16066; near Pretoria, Burtt Davy, 726; Brooklyn, near Pretoria, Mogg, 12276; Zoo

grounds, Pretoria, Mogg in Nat. Herb., 20576 (typus).

Plants up to 42 cm. high, forming dense clumps. New shoots arising from nodes on

a short rhizome with very short internodes. Basal leaf-sheaths persistent, slightly flushed

with pink, keeled, ciliate, densely villous. Blades glaucous green, up to 19 cm. long, 3-5-4

mm. broad (when fresh), distinctly keeled on the back, with about 8 lateral nerves on either

side of the mid-rib, ciliate on the basal portion, otherwise glabrous. Ligule a very narrow

membranous rim, about 0-5 mm. broad, minutely lacerated. Culm bearing inflorescence

2-noded, rarely branched ; nodes reddish, somewhat unequally swollen; upper internode

13-16 cm. long, glabrous. Inflorescence 7-5-9-5 cm. long. Peduncled and sessile spikelets

of equal lengths or one or the other slightly longer. Sessinm Sprketer.—Lower glume

1-05-1-25 em. long, 0-175-0-225 em. broad, lanceolate, more rarely ovate-lanceolate,

produced into 2 awns 3-5-6-5 mm. long, with the margins narrowly inflexed, narrowly

keeled, ciliate on the keels, usually densely villous, more rarely pilose on the back. Upper

glume 6°5-7 mm. long, 1-25-2-25 mm. broad, lanceolate, acute, deeply concave, 3-nerved,

263

keeled on back, shortly ciliate on hyaline marginal flaps, shortly pilose on the back, more

rarely pubescent. Lower valve 5-6 mm. long, 1-25 mm. broad, lanceolate, acute, flattish

on the back, 2-nerved, shortly ciliate on hyaline marginal flaps. Upper valve 4-5 mm. long,

1 mm. broad, lanceolate, concave, rounded on the back, 3—-5-nerved, minutely ciliate on

hyaline marginal flaps. Pale a hyaline fimbriated or ciliated scale 0-5-1 mm. long. Anthers

3 mm. long, linear. Ovary ellipsoid; styles free; stigmas twice as long as the styles.

Lodicules fan-shaped or quadrate, truncate above. PEDUNCLED SpikELET.—Peduncle

1-3-5 mm. long, hollow, villous. Lower glume 0-65-1 cm. long, 1-25-1-75 mm. broad,

lanceolate, long acuminate, sometimes with a minute lateral tooth, with one margin nar-

rowly inflexed and keeled, 5—9-nerved, long ciliate on keel, pilose or villous on back. Upper

glume 5-6-5 mm. long, 1-1-25 mm. broad, lanceolate, acute or sub-acute, deeply concave,

rounded on the back, 3-nerved, ciliate on hyaline marginal flaps, sparsely pubescent or

shortly pilose on the back. Lower valve 2-5-5-5 mm. long, 1 mm. broad, lanceolate, deeply

concave, flattish on the back, 2-nerved, ciliate from hyaline marginal flaps. Upper valve

2-5 mm. long, 1 mm. broad, lanceolate, acute, concave, rounded on back, 3-nerved, ciliate

on hyaline marginal flaps. Pale 0-5-1 mm. long, a hyaline ciliated or fimbriated scale.

Ovary ellipsoid ; styles free ; stigmas twice as long as the styles. Stamens 2-25-3-5 mm.

jong, linear. Lodicules fan-shaped, truncate.

AUSTEMaUKG

“ii

Distribution of #. glaber Phill. and H. argenteus Nees.

ae ‘ -

eh a IIT A Oe Pe

265

ANATOMICAL (By H. C. Breve tz).

An examination of the anatomy of certain specimens was undertaken in order to find

out whether anatomical characters would support the taxonomic characters on which the

genus was divided into three species. The specimens examined were Galpin, 3094 (E.

argenteus Nees); Mogg in National Herbarium, 20576 (E. pretoriensis Phill.); and Mogg

in National Herbarium, 20577 (£. glaber Phill.). Only the leaf and root anatomy of these

grasses were examined.

E. PRETORIENSIS Phill.

Lear Surrace (Fig. 1).

The structure and arrangement of the epidermal cells can be studied when parts of

the epidermis are torn from the underlying tissues and placed in water under the micro-

scope. The abaxial or dorsal epidermis consists of long cells with intervening short cells

opposite the parenchymatous tissues of the leaf. Opposite the primary and secondary

veins, the cells differ in being more thick-walled, narrower, and of three types: (1) short

rod-like cells: (2) medium-sized cells with the side walls constricted at the middle; and

(3) elongated cells. The walls of all the long cells are undulated. On the adaxial or

ventral surface only two types of cells can be distinguished, namely more or less elongated

cells and short cells, the latter only being present opposite the primary and secondary vas-

cular bundles. The cells on the adaxial surface are always shorter than corresponding

cells on the abaxial surface, and the cell walls are only slightly or not at all undulated.

On the abaxial surface the stomata develop in four rows, two near each leaf margin but

on the adaxial surface they are distributed in two or more rows between the veins over the

entire surface.

Lear Anatomy (Fig. II).

In cross section the following tissues of the leaf can be distinguished : (a) the abaxial

epidermis of closely packed regular cells with thickened outer tangential walls and which

are usually small and much lignified opposite the mid-rib, the primary, and secondary

veins ; (6) the adaxial epidermis of regular, relatively thin-walled cells between which many

stomata can be seen in cross section. The row of epidermal cells underlying the mid-rib

are usually bigger and more thin-walled than the rest and act as motor cells, whereas those

opposite the primary veins are smaller and thick-walled ; (c) the ground tissue made up of

thin-walled cells and in which lie (d), the vascular bundles. According to the extent of

development, three types of vascular bundles can be distinguished, viz. the primary, second-

ary and tertiary bundles. The structure of the bundles is uniform throughout. All the

primary bundles, except the mid-rib, have a sclerenchymatous tissue (stereome) developed

towards the adaxial and abaxial surface. The secondary bundles are provided with an

abaxial stereome only, whereas the tertiary bundles have no stereome associated with

them. The stereome strands are very strongly developed and give the leaf a hard texture.

The number of bundles usually averages 32, but in very broad-leaved specimens as many

as 40 were found. Two or three tertiary bundles are present between the mid-rib and the

first secondary bundles and 5-7 bundles between the mid-rib and first primary bundle.

Root Anatomy (Fig. III).

In cross section the following tissues can be distinguished : One layer of irregular more

or less dome-shaped epiblem cells some of which grow out to form root hairs. Those cells

usually disintegrate as soon as the root hairs die off only leaving remnants of some of

266

the cell walls in their place. Underlying the epiblem is the exodermis, which consists of

one layer of more or less elongated cells with a marked tertiary thickening on the outer

tangential walls. Underlying the exodermis are two or more rows of thick-walled cells

which form a definite mechanical tissue, the sclerenchyma. The sclerenchyma surrounds

and protects the imner thin-walled cortical layers (cortex). Adjoining the cortex on the

inside is a single layer of well-differentiated cells, the endodermis which have a marked

tertiary thickening on the inner tangential walls. Inside the endodermis are the vascular

bundles with big vessels in the xylem and in the centre is the pith made up of thin-walled

cells.

E. ARGENTEUS Nees

Lear Surrace (Fig. IV).

The structure and arrangement of the cells of the abaxial epidermis resemble that of

E. pretoriensis (Fig. I). On the adaxial surface the cells are irregular in outline and some

of them bulge out to form an outgrowth on the outer tangential wall. Many hairs develop

from this surface from between the relatively thin-walled cells.

Lear Anatomy (Fig. V).

The structure of the leaf is very similar to that of H. pretoriensis (Fig. Il). Two ter-

tiary bundles are present between the mid-rib and the first secondary bundle and not more

than five bundles are present between the mid-rib and the first primary bundle. The

number of motor cells varies from 5-14 and the primary bundles are provided with a well-

developed stereome which gives the leaf a hard texture.

Roor ANATOMY.

The structure of the root is the same as described under EF. pretoriensis (Fig. III) but

a sclerenchyma with slightly thick-walled cells is present in old roots only.

E. GLABER Phill.

Lear SURFACE.

The arrangement of the epidermal cells on both abaxial and adaxial surfaces is similar

to that of EH. pretoriensis. On both surfaces the stomata are arranged in two or more rows

between the veins and the cells are extremely thin-walled on the adaxial surface.

Lear Anatomy (Fig. VI).

In cross section the same tissues are observed as in #. pretoriensis and H. argenteus.

The cells of the abaxial surface differ from those of the adaxial surface in being more

thick-walled on the outer tangential walls. One or two tertiary veins are present between

the mid-rib and the first secondary vein and not more than six bundles are found between

the mid-rib and the first primary vein. Usually 3 or 4 motor cells are present opposite

the mid-rib. The cells of the sclenchymatous tissues are relatively thin-walled, with the

exception of those opposite the mid-rib, and the leaf has a soft texture.

Roor Anatomy (Fig. VII).

The cortical tissues, viz. epiblem, exodermis and cortex have thin-walled cells and a

sclerenchymatous tissue is never differentiated.

Anatomically the three species of this genus differ in many respects. The anatomical

differences which were found to be constant in the parts studied, are as follows :—

In FH. pretoriensis and H. argenteus the stomata on the abaxial surface are usually

arranged in four rows (two rows near each margin), whereas the stomata are arranged in

267

many rows between most of the veins in #. glaber. In E. pretoriensis the cells of the adaxial

epidermis are relatively thick-walled ; in EH. glaber extremely thin-walled and irregular ;

and in &. argenteus thin-walled with some bulged out on the outer tangential walls. In

E. pretoriensis and EH. argenteus the motor cells seem to be restricted to the region of the

epidermis underlying the mid-rib, but it 1s possible that all the big cells on the entire adaxial

surface of H. glaber are motor cells because the leaves of this grass have the margins inrolled

during hot days.

The space between the motor cells and the mid-rib is filled with parenchymatous tissue

which appears to be different. In H. pretorvensis a group of cells consisting of two or three

rows of cells is present, whereas only one row of cells is present in #. argenteus and EL. glaber.

In E. glaber and E. argenteus one or two tertiary bundles are found between the mid-

- rib and the first secondary bundle, whereas the presence of. two or three tertiary bundles

is a fairly common feature in #. pretoriensis.

The structure of the vascular bundles is uniform throughout. The leaf texture in

E. pretoriensis and HE. argenteus is much harder than that of #. glaber owing to a much

better developed stereome in the two former.

In the roots certain differences are apparent in the epiblem, exodermis and scleren-

chyma. The epiblem cells of #. argenteus and E. pretariensis are usually thick-walled and

more or less persistent, whereas this layer is always relatively thin-walled in LZ. glaber.

The tertiary thickening of the exodermis cells in E. argenteus and HE. pretoriensis 1s an out-

standing anatomical feature of these species. In H. pretoriensis the thickened exodermis

is always associated with an adjoining tissue of well-differentiated sclerenchymatous cells

and although this tissue may be distinguished from the adjoining tissues in #. argenteus

m very old roots, it never forms a well-developed mechanical tissue. As a result of this

lignification of the epiblem, exodermis, and underlying cortex cells, these cell layers are

persistent throughout life and form a protecting sheath round the central cylmder which

is always present in F. argenteus and EH. pretoriensis, even though the thin-walled cortical

cells may disintegrate. In H. glaber the exodermis remains relatively thin-walled through-

out life and a well-differentiated sclerenchyma is never formed with the result that the

cells of epiblem, exodermis and cortex usually break up at an early stage leaving the central

cylinder unprotected or with a few crusts of dead cells round it.

The result of the examination is that the leaf and root anatomy of the South African

species of Elygonurus may be used to determine the species. While it is probable that

shght variations will be found under different environmental conditions, I am of opinion

that the main anatomical features will remain constant for each species.

REFERENCE.

Goossens, A. P. and Theron, J. J. (1934): ““An Anatomical Study of Themeda triandra,’’

8. Afr. Journal Se., Vol. XX XT.

Key to the species based on the leaf and root anatomy.

A, Leaves with two marginal rows of stomata on abaxial surface, hard

in texture; cells of adaxial epidermis relatively thick-walled and

more or less regular; roots with a tertiary thickened exodermis :

a, Sclerenchyma well developed; some of adaxial epidermal

cellsSbulsedsout-rticrsdaceccciae : scares ene. E. argenteus.

a, Sclerenchyma not very well developed ; none of the cells of

adaxialvepidenmis) bulged vouts..1)-.% =. pesca: clea: E. pretoriensis.

A, Leaves with many rows of stomata over the abaxial surface ; soft in

texture; cells of adaxial epidermis extremely thin-walled and

irregular ; exodermis not tertiary thickened.................... E. glaber.

269

EXPLANATION OF FIGURES.

Fig. I: Surface view of abaxial epidermis of HL. pretoriensis; Fig. II: T.S. of leaf

of EL. pretoriensis; Fig. IIL: T.8. of root of H. pretoriensis; Fig. IV: Surface view of

adaxial epidermis of H. argenteus Nees; Fig. V: T.S. of leaf of H. argenteus; Fig. VI:

T.S. of leaf of H. glaber; Fig. VII: T.S. of root of EH. glaber.

Ab. e, abaxial epidermis ; Ad. e, adaxial epidermis ; 6, epidermial cell bulged out on

the outer tangential wall; c, cortex; em., cell constricted at middle; en., endodermis ;

ep., epiblem; ex., exodermis; gt., ground tissue; h., hair; lIc., long cell; Mc., motor

cells; Mr., mid-rib; p, pith; Pv., primary vein; rc., rod-like cell; s., stoma; sc., short

cell; Scl., sclerenchyma ; ster., Sterome; Sv., secondary vein; Tv., tertiary vein.

271

A QUESTION OF NOMENCLATURE.

By EH. P. Purmutes, M.A., D.Sc.

I] HAD occasion to examine a plant sent in for determination and in the National Herbarium

found specimens of a similar plant filed away under three different genera. After looking

jnto the matter I was able to settle the point by reference to literature and the specimens.

1. In 1915 Mrs. L. Bolus described a plant collected by the late Prof. Pearson as

Sutera rigida L. Bolus (Ann. 8.A. Mus., vol. 9, p. 267). She quoted Pearson, 3619 as one

of the specimens of the species.

2. In 1922 (Hook. Ic. Pl., t. 3007), N. E. Brown described a new genus Antherothamnus

and named the plants 4. Pearsonii N. KE. Br. Pearson, 3619 is one of the quoted specimens.

3. In 1928 (Sukkulentenforschung in Siidwest-Afrika, p. 29) K. Dinter mentions a

plant Manuleopsis Karasmontana Dtr. MS. He also distributed specimens (Dinter, 5088)

under that name.

Mrs. L. Bolus was good enough to lend me a sheet of Pearson 3619 and it was found

to be the same plant as Dinter 5088. The species must then be :—

Antherothamnus rigida (L. Bolus) N. E. Br. Sutera rigida L. Bolus ; Manuleopsis Karas-

montana Dinter nomen nudum.

Souta West Arrica.—Great Karasberg: Krai Kluft Ravine, Pearson, 8284. Great

Namaqualand: Warmbad, Pearson, 4377. Bushmanland: Groot Rozynbosch, Pearson,

3619. Klein Karas: Dznter, 5088.

TRANSVAAL.—Potgietersrust distr.: Farm “ Swerwerskraal,”’ about 35 miles west of

Potgietersrust, Rowland in Nat. Herb., 8792.

i Yaseig 4

(Ee BRE OE ie

273

NEWLY DESCRIBED SPECIES.

Crassula sessilicymula Mogg, sp. nov. (Crassulaceae); affinis C. corymbulosae Link. petalis

erectis, cymulis constanter sessilibus, et foliis cautinis differt.

Herba perennis, 30-75 em. alta, singulis caulibus ex base rhizomate. Caules 0-75 em:

diametro, basi ramosi, fulvi aut colores lateris infra, supra virides glanduloso-pubescentes ;

internodia 0-5-2-5 cm. longa. Folia sub-deccussata, sessilia, semi-perfoliata, horizontalia,

simplicia ; inferiora 2-4 cm. longa, 1 cm. lata, 0-3 cm. crassa, utcunque 3 cm. x 0-7 cm. x

0-1 cm. caulis media parte, et ad apice diminuendo, ovata vel lanceolata, acuminata, gla-

braque non reticulata, glanduloso-ciliata. Infloresentia cymosa. Cymulae subcapitatae,

numerosae, axillares, ad summum caulis laxae dispositae. Flores 2-6, fasciculati, sub-

sessiles. Calyx 2 mm. longus; lobi infra breve connati aliquando liberi, 1-5 mm. longi,

angusto-acuminati, pilosi, glanduloso-ciliati. Corolla alba; petala infra connata, 3 mm.

longa, ovato-oblonga, infra ventra concava, apice patentia et infra apicem dorso mucronu-

lata. Stamina petalis subaequalia, ad tubum corollae affixa; filamenta subulata, basi

lata; antherae ovatae. Carpella staminibus aequalia; ovarium oblique-ovatum ; stylus

distinctus. Squamae parvae, aurantiacae, pyriformes, emarginatae.

TRANSVAAL.— Pretoria distr. : Farm “ Klipdrift,’’ near Hammanskraal, 28 miles north

of Pretoria, in a donga inhabitated by thorn-scrub and succulents, Mogg, 12503 (type) ;

between the Saltpan and Hammanskraal, 32 miles N.W. of Pretoria, banks of a dry river-

course beneath a bush of Acacia litakunensis Burch., Mogg, 12505; farm “ Zeekoegat,”

beneath bush-groups, Mogg, 14091; Vogts in National Herbarium, 13049 ; farm “ Rooi-

kop,” Smuts and Gillett, 2511. Waterberg distr. : Naboomspruit, farm “ Mosdene,” under

tree clumps in shade, Galpin, M120A; Seringa, Galpin, 8479; farm “ Gannabosch,”

Bailey in Colonial Govt. Herb., 80. Potgietersrust distr.: Potgietersrust, farm ‘“ Riebeek

West,” Steyn, 85.

The species is also allied to C. compacta Schénl., but the leaves are not all rosulare, and

to C. Lettyae Phill. from which it differs in the well-developed leaves which subtend the

cymules and the glandular-ciliate hairs of the leaf-margins.

Brachystelma nigrum Rk. A. Dyer, sp. nov. (Asclepiadeae-Ceropegieae) ; affinis B. Ger-

rardo Harv. floribus majoribus calycis lobis superne lineari-lanceolatis corollae

indumento coronae interioris lobis brevioribus differt.

Herba perennis. Caules erecti, simplex vel non nunquam ramosi, 30-40 cm. alti,

2-5-4 mm. crassi, foliosi, subhispidi, internodiis 3-5 cm. longis apicem versus brevioribus.

Folia breviter petiolata, late ovata, cordata, obtusa vel subacuta, 2-5-4-5 em. longa,

2-4 cm. lata, juniora minus, infra prominente nervosa, utrinque pilis paucis induta, nerviis

et margine hispidis; petioli 2-5 mm. longi, subhispidi, basi stipulis 2 minimis conicis

ornati. lores axillares nodiis solitarii; pedicelli graciles, circiter 2 cm. longi, basi 1-

bracteati; bractea oblanceolato-linearis vel ovato-elliptica, circiter 1-5 cm. longa, bre-

viter petiolulata. Calycis segmenta plus minusve 7 mm. longa, basin versus breviter ovata,

submembranacea, superne lineari-lanceolata, concava, extra pubescentia, intra basi disco

coroniforme minutissime lobato ornata. Corolla nigra; tubus breviter et latissime cam-

panulatus, glaber ; lobi 7 mm. longi,3 mm. lati, infra medium constricti, subpanduriformes,

elegantissime inflexo-ciliati, supra medium elliptici, suberassi, dorso concavi, hispidul,

intra nigro-velutini, apice leviter incurvi, pilis paucis longis inflexis induti. Coronae

exterioris lobi circiter 0-5 mm. longi, bifidi, breviter lanati ; coronae interioris lobi lineares

incumbenti-conniventes.

274

Natau.—Vryheid, among rocks on summit of Lancaster Hill, locally frequent, flower

black, Jan., Galpin, 10211.

This is yet another new species to the credit of Dr. Ernest Galpin who collected the

specimens near Vryheid, Natal, in January, 1930, He recorded the colour of the fresh

flowers as black and this is the derivation of the specific epithet. Brachystelma nigrum 1s

closely allied to B. Gerrard: Harv. (Harv. Thes. Cap. 2: 61, t. 196) and is distinguished

from this mainly by the smaller size of the flowers, the shape of the upper portion of the

calyx segments and the pubescence of the corolla. The lobes are shortly, although not

densely, pubescent on the outer surface and shortly woolly pubescent on the inner surface.

The colour of the flower of B. Gerrard: is given as “ bright metallic green ” on the inner face.

An interesting character noted by Bullock when describing Ceropegia filicalyx in Hook.

Icon. Plant. under t. 3219 (1933), namely “ intersepaline glands” has an equivalent

structure in both B. Gerrardi and B. nigrum. These have a minute lobed dise or corona-

like structure attached to the base of the calyx round the base of the corolla-tube. Through

the courtesy of the Mycologist in Charge of the Natal Herbarium, Durban, I have examined

one calyx of B. Gerrardi (Wood, No. 1607) but herbarium material of this species and of

B. nigrum is insufficient for a comprehensive study of the intercalycine corona-like organ.

A duplicate of Galpin 10211 is in the herbarium of the Royal Botanic Gardens, Kew.

Riocreuxia aberrans R. A. Dyer, sp. nov. (Asclepiadeae-Ceropegieae) ; corollae tubo breviter

campanulato valde distincta.

Herba perennis caulibus numerosis. Caules erecti vel volubiles circiter 60 cm. alti

vel altiores, plus minusve ramosi, pilis uniseriatis induti, internodis 4-5 em. longis, nodiis

ciliatis. Folia patentia, petiolata, reniformi-ovata, cireiter 5 cm. longa et lata vel interdum

usque 7 cm. longa, basi profunde cordata, apice acuminata, utrinque glabrescentia, margine

ciliata ; petioli 2-4 em. longi supra pilosi. lores plures in cymis extra-axillaribus alter-

nantibus, breviter racemosae pedicellis filiformibus. Calycis segmenta lineari-lanceolata,

glabra, 4 mm. longa. Corolla glabra; tubus campanulatus, 5 mm. longus, circiter 6 mm.

diametro ; lobi lanceolato-lineares, 1-3-1-5 cm. longi, apice coherentes facile liberati. Corona

exteriora et interiora confluentes ; coronae exteriorae lobi brevissimi, bipartiti, coronae

interiorae lobi erecti, oblongo-lineares, obtusi vel emarginati, 2 mm. longi, gynostegium

multo superanti.

TRANSVAAL.—EHrmelo distr.: On farm “ The Gem,” Dec., Walker in Nat. Herb.,

14397 (type). Belfast distr.: Dullstroom, on farm “ Paardeplaats,’ 7,000 ft., local on

low, dry ridge of barren white quartzite, bush 2 ft. high, Jan., Galpin, 13302 (fruit).

Although the specimen Galpin, 13302 is in fruit and has no flowers, it resembles the

type so closely in vegetative characters, that I have little doubt that it is the same species.

The type material was collected in flower in December and the fruiting specimen in January,

which is consistent with seasonal development. Further the leaves of the fruiting specimen

are coarser and slightly larger, up to 7 cm. long and broad, the maximum figure given in

the above description. The follicles of the Galpin specimen are up to 10 cm. long and are

slightly constricted at intervals of 6-7 mm.

Riocreuxia aberrans differs markedly from all others in the genus (hence the name) by

the short campanulate tube, whereas the usual form is cylindric. At first sight there appears

to be only one corona, owing to the very small bilobed outer corona lobes fusing at the base

with the inner ones, the lobes of the outer ones having the appearance of basal expansions

of the inner ones.

279

Another question which requires further investigation is whether the corolla lobes

remain attached at the tips when the flowers open normally. Most of the open flowers

on the type have the corolla lobes free, but this may have been caused during the prepara-

tion of the specimen. If, on the other hand, it is the normal habit of the species, it con-

stitutes a second important difference in this species from the generic characters of Rio-

creuxia, the other being the comparatively short campanulate tube. Together, these

differences would be sufficient reason for placing it in a separate genus.

Leucadendron elimense Phillips (Proteaceae—Proteae); affinis L. concolor; sed foliis glabris

differt.

Suffrutex; ramuli dense pubescentes. Folia 2-5-3 cm. longa, 0-8-1-1 cm. lata,

lanceolata, apice obtusa, glabra. Inflorescentia 3 2-7 cm. lata. Receptaculum 2 cm.

longum, 5 mm. latum, oblongum. Bracteae 1 cm. longae, 1 mm. latae, lineares, apice

subacutae, carinatae, supra pubescentes. Perianthw tubus 1-1 cm. longus, cylindricus,

glaber ; segmenta 2 mm. longa, linearia, sparsim pilosa ; lobi 4 mm. longi, lineares, apice

obtusi, glabri. Antherae 3-5 mm. longae, lineares. Stylus 1-7 em. longus. Inflorescentia

? 2 cm. lata, subglobosa. Receptaculum 1:3 cm. longum, 5mm. latum, oblongum.

Bracteze 8 mm. longae, 4-5 mm. latae, apice plus minusve obtusae, infra pilosae. Perianthi

tubus 8 mm. longus, planus, pilosus ; segmenta 2 mm. longa, linearia ; lobi 1-5 mm. longi,

oblongi, apice obtusi. Ovariwm 1-5 mm. longum, ellipticum, pilosum; stylus 1-5 cm.

longus, glaber; stigma paullo expansum, semi-obliquum.

Cape PRrovince.—Caledon distr.: Koegelberg, alt. 3,500 ft., Stokoe in National Her-

barium, 21906.

The species resembles LZ. concolor R. Br. but the leaves are not densely villous along

the margins. The female heads are surrounded by ovate ciliate glabrous bracts, and both

the male and female heads are surrounded by leaves, similar to the stem leaves, but larger

and yellow and with red tips, longer than the heads.

Crassula

Alt

sessilicymula Moge

NATIONAL HERBARIUM, PRETORIA

Regio Tre /xFi4

was la Se$5ly - yma ln Magy

CAD Fe

x} Lg . ay

inno. asf 7 pebee) OP iia

No...

foxy

yer

Defermninivit

278

Brachystelma

NATIONAL HERBARIUM, PRETORIA,

Distrsct, Uy Adt it

Ui ty Des

Colt

Colivetor's No

nigrum R. A. Dyer

————

2719

CA, ath,

NASIONALI - a

NATIONAL HERBARIUM. PRETORIA,

Nat. Herb. No Rivio

A overecew5u( spor. not traleke)

AC Kus. Ottis cl the Corothe Lbe iz

Campanerlale V4 nhevtis Prinses

ums off the speeces d] fhe above gzi.-2,J)

dork Mik 1 en ba

Deter

ie placed

_ heehee

RABE 3

Riocreuxia aberrans R. A. Dyer

i Cig ue a f :

Vol. Ill. Price, 7s. 6d. Part 3,

Pothalia

A RECORD OF

2o he ee

: "

a eee Ce ae

CONTRIBUTIONS

FROM THE

NATIONAL HERBARIUM

UNION OF SOUTH AFRICA

— ee Se a ee

ae a

PRETORIA

YU yyy =

Viplityy

Vy at Lrg,

EDITED BY

I. B. POLE EVANS, C.M.G., M.A. (Cantab.), D.Sc. (Wales),

CHIEF, DIVISION OF PLANT INDUSTRY, DEPARTMENT OF AGRICULTURE AND FORESTRY

PRETORIA ; AND

DIRECTOR OF THE BOTANICAL SURVEY OF THE UNION OF SOUTH AFRICA

PustisHEepD Apri, 1938,

CONTENTS.

Tue Usritacinates or SourH AFRica ..

Some SoutH Arrican FusaRia

THE USTILAGINALES OF

SOUTH AFRICA.

GroRGE LorENzZO INGRAM ZUNDEL,

AsstIstANtT Proressor, Phant PatHotocy, EXTENSION,

Tur PENNSYLVANIA STATE COLLEGE.

The following paper is an attempt to monograph partially the smuts or Ustilaginales

of South Africa and adjacent territories. Besides species found wholly within the Union

of South Africa, there are included species from South West Africa, Portuguese East Africa,

the Tanganyika Territory, Rhodesia and Nyassaland Protectorate.

Original descriptions of South African smuts are to be found in various publications,

principally German. Paul Hennings and later the Sydows have described most of the new

species found in South Africa. Recently the author has described new species of South

African smuts.

In 1926, Len Verwoerd of the University of Stellenbosch made the first attempt at

monographing South African smuts. It is the hope of the writer that sometime in the future

he will be able to enlarge and complete the present paper as more specimens are collected

and become available. Specimens used for descriptions in the present paper are mostly

from the Mycological Herbarium, Union Department of Agriculture. Many type specimens

of species described herein were not available for examination, in fact the writer has not

been able to locate the type specimens of some of the very early South African collections.

Acknowledgment and thanks are hereby made to the Director of the Royal Botanic

Gardens, Kew, Surrey, England, for supplying type material of Tilletia Ayresii Berk. for

study ; to Miss Claribel Barnett, Librarian, and Mr. J. A. Stevenson, Mycologist, both of

the United States Department of Agriculture for furnishing original descriptions of some

of the species and for the loan of books ; to Dr. George Perkins Clinton for the loan of type

specimens from his personal herbarium. The genus descriptions have followed very closely,

in some cases verbatim, those previously published by Dr. Clinton. Dr. Robert E. Dengler,

Professor of Classical Languages, The Pennsylvania State College, has kindly written the

Latin descriptions of new species. The host identification and nomenclature have been

revised according to the latest usage by Miss L. Chippendall of the South African National

Herbarium, Pretoria.

The asterisk (*) after a host plant name or locality indicates that the record has been

secured from literature and that no specimen has been seen.

CHARACTERIZATION OF THE USTILAGINALES.

This group of fungi belongs to the sub-group Hemibasidiomycetes with the single

order Ustilaginales containing two families: (1) Ustilaginaceae, and (2) Tilletiaceae.

The Ustilaginales consist of nearly 600 species. All of them are obligate parasites

usually attacking herbaceous plants. The mycelium is internal, slender, hyaline, somewhat

septate and branched. It is either uninucleate, binucleate or occasionally multinucleate.

284

The hyphae are usually intercellular with capitate or racemoid haustoria penetrating the

host cells for food. Most of the smuts attacking cereals, however, have intercellular mycelium

without haustoria and obtain food direct from the host cells by the process of osmosis. A

third class of smut fungi, e.g. Ustilago Zeae, has mycelium that penetrates the mdividual

host cells causing death.

At maturity the hyphae enter certain organs of the host and form thick celled chlamy-

dospores which are the common smut spores or winter spores. These are produced singly,

in pairs or in balls, usually forming a dark powder at maturity. In the genera Entyloma,

Tracya, Burrvillia, and Doassansia, however, the spores are produced within the host tissues

and are light coloured. In the early stages of development, all of the spores are binucleate.

At maturity each spore is uninucleate but is produced from a binucleate hypha. Each

mature spore has a thin endospore and a thicker variously colored and sculptured exospore.

At maturity the spore masses produce various effects on the host, e.g. a shredding of the

leaves of the host ; a destruction of the inflorescence ; the replacing of the pollen by smut

spores and the utilization of the pollen distributing apparatus for the distribution of smut

spores; stimulating the development of staminoids in pistillate flowers; the formation

of boils or gall-like tumor growths; the destruction of the seed, etc.

Usually the mycelium is annual, however, in some cases it is perennial and hibernates

in the roots of the host.

The Ustilaginales are divided into two families based on the method of producing

the promycelium and basidiospores. In both families, previous to the production of a

promycelium, the nucleus divides and one of the daughter nuclei leaves the interior of the

spore and migrates into the promycelium.

The first family is the Ustilaginaceae. Here the daughter nucleus in the primary promy-

celium divides twice forming four nuclei. Following these divisions three septa, or cross

walls, are formed thus forming a four celled promycelium. Basidiospores are produced

laterally at the cross walls. As the basidiospores are formed, the nucleus in the promycelial

cell divides and one migrates into the basidiospore. This process continues as long as

basidiospores are produced. The infection tubes are produced by the basidiospores.

The second family is the Tilletiaceae. As the promycelium develops, one of the daughter

nuclei together with some of the protoplasm migrates from the interior of the germinating

spore into the base of the young aseptate promycelium. This is followed by at least four

and sometimes more nuclear divisions, thus forming at least eight nuclei which together

with the protoplasm migrate to the distal end of ‘the mature promycehum. The “lower

empty part of the promycelium is then cut off by three septa. Eight basidiospores are

then produced terminally and usually conjugate before separation from the promycelium.

Conjugation seems to be necessary in order for the smut fungus to infect its host plant,

i.e. there are apparently basidiospores and promycelium cells with two genders. The

conjugation process may take place in any of the following ways :—

(1) A basidiospore may conjugate with another basidiospore.

(2) A basidiospore may conjugate with a cell of a promycelium.

(3) A promycelium cell may conjugate with another promycelial cell.

(4) Occasionally two promycelia are produced from one smut spore and pro-

mycelial cells of one promycelium may conjugate with cells of the other pro-

mycelium.

Infection tubes are produced by the conjugated parts of the basidiospores or promycelia.

There are three principal types of infection :—

Type 1. The chlamydospores attach themselves to the seed of the host and do

not grow until the host seed germinates. In this manner there is a seed-

ling infection. The germ tube penetrates the tissues of the entire host

plant.

285

Type 2. The chlamydospores attach themselves to the stigma of the flower of the

host, where they germinate at once, sending mycelial threads down the

style into the young ovary where the mycelium becomes dormant without

deforming the seed, and resting there until favorable conditions for

seed germination arrive. As the seed germinates and grows, the dormant

hibernating mycelium becomes active and penetrates the entire host

tissue, finally replacing the inflorescence with smut spores.

Type 3. In this type the chlamydospores produce promycelia and basidiospores

in decaying vegetation. The basidiospores are then carried by the wind

to the young host plants where local infection takes place. The mycelium

penetrates the host cells only locally.

During the short time of basidiospore formation, the smuts are facultative saprophytes

forming colonies of yeast-like sporidia. Most of them can be cultivated on nutrient agar

but few if any can complete their hfe history on artificial media.

CLASSIFICATION OF THE USTILAGINALES.

The Ustilaginales are parasitic fungi that attack various parts of herbaceous plants.

Infection nearly always takes place through very young tissues, either through germinating

seed or other special parts of the host. This group can usually best be recognized by the

sooty mass of spores that are produced, singly, in pairs, or as spore balls. The black smuts

are represented by such genera as Ustilago, Sphacelotheca, Sorosporium, Urocystis and

Tilletia, while the so-called white smuts are mostly leaf inhabiting and are included in such

genera aS Burrillia, Doassansia, Entyloma and Tracya.

Two families are included in the order Ustilaginales, (1) the Ustilaginaceae, and (2) the

Tilletiaceae. They are separated by the manner in which the promycelium and sporidia

are produced as follows :—

Promycelium usually with lateral sporidia at septa... Ustilaginaceae.

Promycelium with clustered terminal sporidia......... Tilletiaceae.

In the classification of the smut fungi the viewpoint is held that morphological characters

rather than host susceptibility or the use of biometry must be the basis for determining

species. The concept of species as used in this paper is very broad. Many recently described

species are merely physiological or pathological strains that have adapted themselves to

specific hosts. Such species should be relegated to synonymy.

Within the last few years technique has been perfected whereby it 1s now possible to

hybridize the smut fungi and thus study gender and species relationship as 1s done in the

higher plants. Extensive work of this nature has been done by workers in the United States

Department of Agriculture and of the several state Agriculture Experiment Stations, also

by workers in Canada and Wales. A brief summary of a few of the numerous papers

follows :—

Reed! in 1928 reported four physiological races of Tilletia laevis and six physiological

races of Tilletia Tritici. Hach physiological race varied in its ability to attack different

varieties or even strains within a given variety of wheat.

Flor? working in Washington State reported in 1932 that Tulletia Tritici and Tulletia

laevis were heterothallic. He also reported that he hybridized Tilletia Tritici and Tilletia

laewis and had obtained evidence that hybridization occurred in nature. Examination

of over 10,000 bunted heads revealed that there were all degrees of reticulations on spore

1 Reed, George M. Physiological races of wheat bunt. Am. Jour. Bot. 15: 157-170. 1928.

2 Flor, H. H. MHeterothallism and hybridization in Tilletia Tritici and 7. laevis. Jour. Agr.

Res. 44 : 49-58. 1932.

286

of Tilletia Tritici. In some cases the reticulations were so fine that 1t was almost impossible

to see them ; on the other hand the reticulations were so coarse that they appeared almost

spiny.

In 1924, Faris? showed that there was physiological specialization of Ustilago Hordei

and that each physiologic form attacks only certain specific varieties of barley.

Holton? in 1932, reported that he had hybridized Ustilago Avenae and Ustilago levis.

He found that factors such as echinulation of spores, growth of the fungi in artificial culture,

appearance of the smut in the panicle, all followed in general a Mendelian ratio and further-

more a buff colored smut with hyaline spores was produced by crossing monosporidial lines

hybrid chlamydospores. Later he found® that the buff smut resulted from a mutation

in Ustilago levis. He further found that pathologic strains of U. Avenae and U. levis were

produced by hybridization and segregation.

In a more recent paper,® Holton reported that when U. Avenae and U. levis were hybri-

dized that (1) the factor for brown is dominant over the factor for hyaline chlamydospores ;

(2) the factor for echinulate spore walls is domimant over the factor for smooth spore walls

and that these characters generally segregate in a Mendelian ratio.

Reed and Stanton’ in 1936, reported that a distinct strain of loose smut (U. Avenae)

occurs on Red Rustproof oats. This strain of smut also attacks various strains of Avena

fatua and Avena strigosa, also the variety Canadian (a variety of Avena sativa).

Rodenhiser® working with Sphacelotheca Sorghi and Sphacelotheca cruenta found physio-

logic forms in both species. He produced an intermediate type of smut by hybridizing

S. Sorghi and S. cruenta which had a sorus different from either parent with two kinds of

sterile cells, the small sterile cells of S. Sorghi and the large spherical type of S. cruenta.

In culture, numerous mutants were observed to arise. It was possible for him to produce

new physiologic forms by hybridization and segregation.

In view of these results 1t seems that H. Sydow® in 1924, when he made a study of

Cintractia Caricis (Pers.) P. Magn. on Carex spp., and finally described ten new species

based on host species, had in reality only ten physiologic forms of Cintractia Caricis (Pers.)

P. Magn. that by natural hybridization and segregation were pathologic to specific species

of Caren,

Likewise Liro! in 1924 in his study of the smuts attacking Polygonum spp. and Ciferri!!

in 1928, in his new sub-species of Hntyloma compositarum Farlow are dealing with physio-

logical forms.

Fisher!” has recently reported the natural infection of Agropyron tenerwm in central

Washington by Tilletia Tritici and notes that the size of the sorus varies with the size of

the ovary of the host. In other words, the morphology of the fungus varies with the host.

° Faris, James A. Physiologic specialization of Ustilago Hordei. Phytopath. 14 : 5387-557.

1924.

4 Holton, C. $8. Studies in the genetics and the cytology of Ustilago Avenae and Ustilago levis.

Univ. Minn. Exp. Sta. Tech. Bull. 87: 1-34. 1932.

® Holton, C. 8. Origin and production of morphologic and pathologic strains of the oat smut

fungi by mutation and hybridization. Jour. Agr. Res. 52: 311-317. 1936.

° Holton, C. 8. Inheritance of chlamydospore characters in oat smut fungi. Jour. Agr. Res.

52 : 535-540. 1936.

* Reed, George M. and Stanton, T. R. Reaction of oat varieties to physiologic races of loose

and covered smuts of red oats. Jour. Agr. Res. 52: 1-15. 1936.

8 Rodenhiser, H. A. Studies on the possible origin of physiologic forms of Sphacelotheca Sorghi

and S. cruenta. Jour. Agr. Res. 49: 1069-1086. 1934.

® Sydow, H. Notizen tber Ustilagineon. Ann. Myc. 22: 277-291. 1924.

10 Liro, 1. J. Die Ustilagineen Finlands. [. Ann. Acad. Sci. Fenn. A. 17: XVIII—636 pp. 1924.

'I Ciferri, R. Quarta contribuzione allo studio Ustilaginales. Ann. Myc. 26: 1-68. 1928.

12 Fischer, George W. The susceptibility of certain wild grasses to Tilletia Tritict and Tilletia

laevis. Phytopath. 26: 876-886. 1936.

287

KEY TO GENERA OF USTILAGINALES REPORTED FROM SOUTH AFRICA.

I. Spores single :—

A.—Usually forming a dusty sorus at maturity—

(Ue larcemusuallyal Os oy oerwr eer cchnie tetris trot stel ane as fe Tilletia.

2. Small to medium, usually 5-18 ~—

(a) Sorus covered with a false membrane of fungoid cells... Sphacelotheca.

(6) Protecting membrane, if any, of plant tissue........... Ustilago.

B.—More or less firmly agglutinated at maturity—

1. Firmly agglutinated into irregular nodules ................ Melanopsichium.

2. Developed around a central columella (rarely becoming dusty) Cintractia.

C.—Imbedded in leaves at maturity—

ie Usuallyshyalmexorsliohticolored maa ess. acces ac. vas oes Entyloma.

II. Spores in more or less regular balls :—

A.—Forming a dusty or granular sorus at maturity—

1. Spore balls consisting only of fertile cells—

(a) Usually evanescent, olivaceous or black brown......... Sorosporium.

(6) Quite permanent, spores adhering by folds of outer coat Tolyposporium.

2. Spore balls with outer sterile cortex........................ Urocystis.

B.—Implanted in plant tissue at maturity—

1. Spore balls without definite cortex or sterile cells........... Tubureimia.

Family I—USTILAGINACEAE.

Ustilago (Pers.) Roussel!*, Fl. du Calvados ed. 2. 47. 1806.

Necrosis Paulet, Traite Champ. 1: 584. 1793.

Ustilago Pers. Syn. Fung. 224. 1801. (sub-genus.).

Ustilagyidium Heré. Zopf Beitr. Phys. Morph. Org. 5:7. 1895.

Sori on various parts of the hosts, at maturity forming dusty, usually dark colored

spore masses; spores single, produced irregularly in the fertile mycelial threads which

early entirely disappear through gelatinization; small to medium in size; germination

by means of a septate promycelium producing only infection threads or with sporidia

formed terminally and laterally near the septa; sporidia in water usually germinating

into infection threads but in nutrient solutions multiplying indefinitely, yeast fashion.

Type: Ustilago segetum Pers., on Gramineae, France.

18 Clinton, N. Am. Flora 71:3. 1906, points out that “J. Bauhin, Hist. Pl., 418, 1651, is really

the founder of Ustilago. Fries or Persoon is ordinarily cited as authority for the genus. Fries used

Ustilago as a genus in his Syst. Myc., 3: 517. 1832, with U. grandis as the first species. Persoon

used Ustilago as a subgenus under Uredo with Uredo segetum as the first species, having five varieties

of which U. Hordez is first, and this may be taken as the actual type now that U. segetum has been

broken up into several species. Roussel merely adopted Ustilago from Persoon, but raised it to full

generic rank. giving three of Persoon’s four species, of which U. segetumis one. Paulet’s name, Necrosis,

cannot be cegarded as a true generic name, but was used more as a descriptive term.”

288

* Spores smooth.

Ustilago Elionuri P. Henn. and Pole-Evans, Bot. Jahrb. (Engler) 41:270. 1908

Sori in the ovaries, 1-1.5 em. long, covered with a dark-brown membrane of host

tissue which dehisces apically revealing a semi-agglutinated brown spore mass; spores

globose—ellipsoidal, irregular, somewhat angled, dark olive-brown, 5-7 diam., smooth

under oil immersion.

Type host and locality: On Elionurus argenteus Nees, Pretoria, Transvaal, Union

of South Africa.

On Andropogoneae: Hlionurus (Hlyonurus) argenteus Nees, Transvaal.! (M.H. 102,

9316, 17268.)

Distribution: Transvaal and Argentina.

Ustilago Hordei (Pers.) Lagerh.? Mitt. bad. bot. Ver. 70. 1889.

Reticularia segetum Bull. Hist. Champ. pl. 472, flg. II. 90. 1791.

Uredo segetum Horde: Pers. Tent. Disp. Fung. 57. 1797.

Uredo carbo DC. Fl. Fr. 6:76. 1815. p.p.

Ustilago segetum Link, Ditm. im Sturm’s Deutsch. F.] HJ. 1:67. 1817. p.p.

Caeoma segetum Link, Willd. in Sp. Pl. 52:1. 1825. p.p.

Erysibe vera Horde Wallr. Fl. Crypt. ‘Germ. 2: 217. 1833. p.p.

Uredo Carbo-Horde: Phillipar, Mem. Soc. Roy. ‘Agr. Arts, Seine-et-Oise 37:195. 1837.

Ustilago Carbo vulgaris Hordeacea Tul., Ann. Sci. Nat. Bot. WOE 72 SO, IRS jo.70.

Ustilago segetum var. Horde f. tecta Jens. Om Korns. Brand. 61 : 1888.

Ustilago tecta horde: Jens. Charb. Cereals 4. 1889.

Ustilago hordei Kell. and Sw. Ann. Rep. Kans. Agr. Exp. Sta. 2: 268. 1890.

Ustilago Jensenii Rostr. Overs K. Danske Vid. Selsk. Forh. 1890:12. 1890.

Sori in the spikelets, destroying the inflorescence, forming a rather permanent purple-

black spore mass covered by a permanent membrane, about 5-8 mm. long: Spores globose-

subglobose, occasionally ellipsoidal, olivaceous-brown, 5-7 w diam. or slightly larger, smooth .

Type host and locality: On Hordeum vulgare Linn., Kurope.

On Hordeae: Hordeum vulgare Linn., Cape Colony, Transvaal, Orange Free State.

(M.H. 1181, 7080, 9823.)

Distribution: Co-extensive with the cultivation of barley.

Ustilago levis (Kell. and Sw.) P. Magn. Abh. Bot. Ver. Prov. Brandenburg 37:69. 1896

Ustilago Avenae var. levis Kell. and Sw. Ann. Rep. Kansas Agr. Exp. Sta. 2: 259. 1890.

Ustilago Kolleri Wille, Bot. Notiser 1898: 10. 1893.

Sori in spikelets, more or less destroying basal and imner glumes; spores globose-

subglobose, light olivecaous-brown, usually lighter colored on one side, 5—7 diam., sometimes

slightly larger, smooth.

Type host and locality: On Avena sativa Linn., Kansas, United States.

On Aveneae: Avena sativa Linn., Cape Colony, Trnasvaal, Rhodesia.* (M.H. 538,

940, 7094, 8352, 10971.)

Distribution : Co-extensive with the cultivation of oats.

Ustilago affinis Ellis and Ev. Bull. Torrey Club 20: 297. 1893.

Ustilago Hilariae P. Henn. Hedwigia 87: 267. 1898. Not Ustilago Hilariae Klis and

Tracy, 1890.

Ustilago Stenotaphri P. Henn. Hedwigia 87: 293. 1898. (Type from Windhoek, South

West Africa, on Stenotaphrum asus = = Stenotaphrum secundatum (Walt) Kuntze.)

Not U. Stenotaphri McAlpine, 1895.

'M.H. = Mycologic: al Herbarium, Union Dept. Agric.

2 Wakefield and Moore in T'rans. Brit. Mye. Soc. 20:97, 1936, call attention to the fact that

the authority for this smut species should be (Pers.) Lagerh. and not (Pers.) Kell. Swingle.

289

Ustilago americana Speg. Anal. Mus. Nac. Buenos Aires 6: 201. 1899.

Ustilago Henningsii Sacc. and Sydow, Sace. Syll. Fung. 16: 368. 1902.

Sori in the spikelets, usually infecting every ovary in a spikelet, at first more or less

hidden by the surrounding leaves, at first covered by a thin, delicate olive-brown membrane

which ruptures revealing an olive-brown spore mass which soon disperses leaving a naked

rachis ; spores oblong-ovate, chiefly globose-subglobose, somewhat irregular, light reddish

brown (almost hyaline), 5.5-8 p Glia, smooth.

Type host and locality : On Stenotaphrum americanum Schrank., Mandeville, Jamaica

On Paniceae: Stenotaphrum secundatum (Walt.) Kuntze, South-West Africa. (M.H

6879.)

Distribution: West Indies, South America, South Africa.

Ustilago Scitaminea H. Sydow,® Ann. Myc. 22: 281. 1924.

Sori transforming the floral stem into a long, curved, leafless, stem-like growth covered

by a thin membrane of host tissue, the lower part of the sori concealed by the sheath ;

spores globose-subglobose, reddish brown, 5-12 » diam., smooth; groups of hyaline thin

walled cells scattered through the sori.

Type locality and host: On Saccharum officinarum Linn., India, Java, Philippines.

On Andropogoneae: LHrianthus saccharoides Michx., Natal; Imperata arundinacea

Cyrilli, Natal; Saccharum officinarum Linn., Natal. (M.H. 11111.)

Ustilago puellaris Sydow, Ann. Myc. 33: 231. 1935.

Sori entirely destroying the inflorescence, concealed by the glumes, cylindrical, usually

1 mm. long, occasionally 1-4 mm. long, covered by dark brown membrane which ruptures ,

disclosing a dark brown powdery spore mass; spores globose-subglobose, rarely ellipsoidal,

regular, olivaceous brown with a narrow reddish brown epispore, 7-9 uw diam., smooth

but sometimes indistinctly verruculate under the oil immersion.

Type host and locality: On Hyparrhenia hirta Stapf, Research Station, Nelspruit,

Transvaal.

On Andropogoneae: Hyparrhenia hirta Stapf, Transvaal. (M.H. 26646.)

Distribution: Reported only from type locality.

Ustilago Crameri Korn. Jahrb. Nassau. Ver. Naturk. 27-28:11. 1875.

Sori in the ovaries, completely destroying them, ovoid, covered with a delicate membrane

which when ruptured reveals a brownish, granular spore mass ; spores globose-subglobose ,

reddish-brown, 7-10 » diam., smooth even under oil immersion.

Type host and locality: On Setaria italica (Linn.) Beauv., Zurich, Switzerland.

On Paniceae: Setaria italica (Linn.) Beauy., Transvaal, Zululand*, Orange Free State.

(M.H. 2204, 9817, 11716.)

Ustilago Cynodontis P. Henn. Bot. Jahrb. (Engler) 14: 369. 1891.

Sori destroying the inflorescence converting it into a mass of spores along the rachis,

at first covered with a whitish-black membrane which breaks away revealing a black spore

mass 2-4 em. long, sometimes slightly longer ; spores globose-subglohose, regular, reddish-

brown, 7-10 p diam., smooth or ‘almost smooth but granular.

Type host and locality: On Cynodon Dactylon Pers., Abyssinia, Africa.

° Sydow in 1924 (1.c.) pointed out that the smut on cultivated sugarcane was not Ustilago Sacchari

Rab. and therefore proposed the name Ustilago scitaminea Syd. for the smut that attacks sugarcane.

4 Verwoerd in Ann. Univ. Stellenbosch, A. 4:19. 1926, confuses Ustilago Dregeana Tul. and

U. Cynodontis P. Henn. These are two very distinct species. U. Dregeana Tul. has papillate spores,

4-5 4. diam. (See Appendix A.), wihle U. Cynodontis P. Henn. has smooth spores, 7-10 p diam.

290

On Hordeae: Cynodon Dactylon Pers., Orange Free State (M.H. 9752), Transvaal

(M.H. 947, 112, 5636, 11005, 11014), Cape Province (M.H. 12953); Cynodon incompletus

Nees, Transvaal (M.H. 1957, 8843), Orange Free State (M.H. 904), Cape Province (M.H.

25442).

Distribution: North America, Hurope, Asia, Africa.

Ustilago Vaillanti Tul. Ann. Sci. Nat. Ill. 7:90. 1847.

Ustilago scillae Ciferri, Ann. Myc. 29:24. 1931.

Sori in pistils and anthers, perianth of diseased flowers persistent, somewhat enlarg.d

and filled with spores, spores globose-ellpsoidal, sometimes angled, light reddish-brown,

7-11 p diam., smooth but contents somewhat granular.

Type host and locality: On Muscari comosum (Iann.) Mill., Kurope.

On Liliaceas: Albuca altissima Dryand., Cape Province (M.H. 15450); Hucomis

punctata L’Herit., Cape Province (M.H. 2001); Sella Krauwssii Baker,* Inanda, Natal

(M.H. 9525); Scilla sp., Cape Province,* Natal (M.H. 12956).

Distribution: United States, Europe, Africa.

Ustilago Dactyloctaenii P. Henn. Die Pflanzenwelt Ost-Afrika, Nachbar. C:48. 1895.

Sori entirely destroying the inflorescence, at first covered by a membrane which flakes

away revealing a dark purplish spore mass, at first slightly agglutinated but later powdery ;

spores globose-subglobose, regular, light olivaceous-brown, 7-14 uw diam., granular, smooth,

epispore 1-2 pw.

Type host and locality: On Dactyloctenium aegyptium (l.) Richt. (= Dactyloctenium

aegyptiacum Willd.), Zanzibar, Africa.

On Chlorideae: Doctylocteniwm aegyptium (L.) Richt., Cape Province (M.H. 9114) ;

Dactyloctenium geminatum Hack., Portuguese Kast Africa (M.H. 14175).

** Spores not smooth.

Ustilago Avenae (Pers.) Jens. Charb. Cereales 4. 1889.

Reticularia segetum Bull. Champ. pl. 472, flg. II. 90. 1791. p.p.

Uredo carbo Avenae DC., Fl. Fr. 6. 76.

Ustilago segetum Link, Ditm. in Sturms Deutsch. Fl. II. 1:67. 1817. p.p.

Caeoma segetum Link, Willd. Sp. Pl. 62:1. 1825.

Erysibe vera Avenae Walle. Fl. Crypt. Germ. 2: 217. 1833.

Uredo carbo-avenae Phillipar, Mem. Soc. Roy. Agr. Arts Seme-et-Oise 37: 194. 1837.

Ustilago carbo-vulgaris avenae Tul. Ann. Sci. Nat. Bot. II. 7:80. 1847.

Ustilago segetum Avenae Jens. Om Korns. Brand 61. 1888.

Ustilago Avenae f. foliicola Almeida, Revista Agron. (Lisbon), 1:20. 1903.

Sori in the spikelets, 5-12 mm. long, usually destroying the inflorescence rather com-

pletely and most of the spores finally bemg blown away, rarely on the leaves or culms ;

spores spherical to subspherical or sometimes more elongate, olivaceous-brown, som :times

lighter colored on one side of the spore, 5-7 » diam., minutely echinulate.

Type host and locality: On Avena sativa Linn., Europe.

On Avenae:' Avena sativa Linn., Cape Province, Transvaal. (M.H. 8928.)

Distribution : Co-extensive with cultivated oats.

Ustilago Holubii Sydow, Ann. Myc. 38: 230. 1935.

Sori entirely destroying the panicle and sometimes extending into the end of the culms

usually 2-4 em. long, covered by a dark colored membrane which flakes away revealing a

brown spore mass; spores globose-subglobose, olivaceous-brown, 5-7 » diam., apparently

smooth but minutely verruculate under oil immersion.

291

Type host and locality: On Echinochloa Holubii Stapf, Pretoria, Transvaal, Union

of South Africa.

On Panicaceae: Echinochloa (Panicum) Holubii Stapf, Cape Province,* British Bechu-

analand, Transvaal.* (M.H. 24942 (type), 17042.)

Distribution : South Africa.

Ustilago Tritici (Pers.) Rostrup, Overs. Danske. Vid. Selsk. Forh. 1890: 15. Mr. 1890.

Lycoperdon Tritic: Bjerk. Kgl. Schmedisch Akad. Wiss. Abhandl. 87: 326. 1775.

Uredo segetum Tritict Pers. Disp. Meth. Fung. 57. 1797.

Uredo carbo Tritict DC. Fl. Fr. 6:76. 1815.

Ustilago setum Ditm. Sturm’s Deuts. Flore III. 1:67. 1817.

Caeoma segetum Link, Willd. Sp. Pl. 62:1. 1825.

Erysibe vera Tritici Wallroth, Fl. Crypt. Germ. 2:217. 1833.

Uredo Carbo-Tritici Philippar, Soc. Roy. Agr. Arts Seine-et-Oise 37: 197. 1837.

Ustilago Carbo vulgaris Triticea Tul. Ann. Sci. Nat. Bot. III. 7:80. 1847.

Ustilago segetum var. Tritici Jens. Om Korns. Brand 61. 1888.

Ustilago Horde: Bref. Nach. Klub. Landw. Berl. No. 1593. 1888.

Ustilago Tritici Jens. Kelleran and Swingle in Ann. Rep. Kansas Exp. Sta. 2: 262. 1890.

Ustilago Tritici foliicola P. Henn. Zeitschr. Pflanzen. 4: 139. 1894.

Ustilagidium Tritici Herzb. Zopfi. Beitr. Phys. Morph. Neiderer Org. 5:7. 1895.

Ustilago Vavilovi Jacz. (Ann. State Inst. Exp. Agr. III. 2-4: 106-109. 1925. Transl.

Title).

Sori in the spikelets, forming a dusty olive-brown spore mass, eventually destroying

all of the inflorescence and the spores are blown away leaving only the naked rachis ; spores

spherical or subspherical, sometimes ee elongated and irregular, light olive-brown, usually

lighter colored on one side of the spore, 5-9 w diam., minutely but distinctly echinulate.

While the entire wheat spike is girly destroyed and only the naked rachis left, yet

partial destruction of the spike is not uncommon. Under rare conditions the sorus develops

on the sheaths and leaves.

Type host and locality: On Triticum vulgare Vill., Europe.

On Hordeae: Triticum vulgare, Cape Colony*, Orange Free State, Transvaal. (M.H.

1068, 9821.)

Distribution : Co-extensive with the cultivation of wheat.

Ustilago Fingerhuthiae Sydow, Ann. Myc. 33: 230. 1935.

Sori in the ovaries, 2-5 mm. long, scattered throughout the spike, covered with a

yellowish membrane that dehisces rev ealing a brownish, granular, spore mass; spores

globose-subglobose, regular, olivaceous-brown 5-10 » diam., densely but minutely verru-

culose-echinulate.

Type host and locality: On Fingerhuthia africana Lehm., Pretoria, Transvaal, Union

of South Africa.

On Festuceae: Fingerhuthia africana Lehm., Transvaal. (M.H. 1085, 7405, 8909.)

Distribution: Not reported except from type locality.

Ustilago trichophora (Link) Kunze, Flora 1: 369. 1830.

Caeoma trichophora Link, Willd. Sp. Pl. 62:3. 1825.

Uredo trichophora Kérn. Hedwigia 16:36. 1877.

Sori as small nodules in the individual ovaries, scattered in the panicle, singly or in

groups, 2-5 mm. long, covered by a yellowish tough, hispid membrane, spore mass at first

hard, agglutinated but later powdery; spores globose-subglobose, light reddish-brown

(some immature spores almost hyaline), 7-8 » diam., abundantly echinulate under oil

immersion.

292

Type host and locality: On EHehinochloa colona (=Panicum colonum Linn.), Egypt.

On Paniceae: Hehinochloa Crus-galli Beauv. (=Panicum Crus-galli Linn.), Cape

Province. (M.H. 9424.)

Distribution : Africa.

Ustilago Trachypogonis Zundel n. sp.

Sori in the ovaries, protected by the outer glumes, about 1 em. long, spore mass granular,

dark brown; spores globose-subglobose, occasionally ellipsoidal, brown with a distinctly

reddish-brown epispore, 7-9 w diam., echinulate.

Hab. in the ovaries of Trachypogon plumosus Nees., Kaalfontein, Transvaal, Union

of South Africa, Coll. A.O.D. Mogg, Feb. 22, 1918. (M.H. 11709.)

Latin description :—

Soris in ovarus, glumis externis tectis, ca. | cm. longis, globis sporarum grandularibus,

atro-brunneis ; sporis globosis v. sub-globosis, interdum ellipsoideis, brunneis, 7-9 » diam.,

echinulatis ; episporio conspicue rubro-brunneo.

Hab. in ovarus Trachypogonis plumosi Nees., Kaaltontem, Transvaal, Unione Africae

Australis, Coll. A. O. D. Mogg, Feb. 22, 1918. (M.H. 11709.)

Ustilago Schlechteri P. Henn. Hedwigia 34 : 325. 1895.

Sori in the panicles, 8 or more em. long, deforming and destroying the inflorescence,

at first covered with a membrane which ruptures revealing a dark spore mass; spores

globose-subglobose or sometimes ellipsoidal, deep clivaceous-brown with a deeper almost

reddish-brown epispore, 7-10 w diam., or sometimes shghtly larger, minutely echinulate-

verruculose under oil immersion.

Type host and locality: On Sporobolus sp., Tweedie, Natal. (M.H. 11644.)

On Agrostideae: Sporobolus sp., Natal. (M.H. 11644.)

Distribution: Union of South Africa.

Ustilago Crus-galli Tracy and Karle, Bull. Torrey Club 22:175. 1895.

Sori surrounding the stem at nodes or place of inflorescence, attacking both stem and

leaves, nodular, elongated, swollen, several cm. long, surrounded by a tough hispid mem-

brane, spore mass brown, powdery ; spores ovoid to spherical, occasionally more elongate,

olivaceous-brown, 7—LO » diam., bluntly echinulate.

Type host and locality: On Paniewm erus-galli Linn., Salt Lake City, Utah, United

States.

On Paniceae: EHchinochloa Holubii Stapf, Transvaal. (M.H. 2247.)

Distribution: United States, Hurope, Asia, Africa, Australia.

Ustilago Zeae (Beckm.) Unger. Hinfl. Bodens 211. 1836.

Lycoperdon zeae Beckm. Hannov. Mag. 6: 1330. 1768.

Uredo segetum Mays Zeae DC. Fl. Fr. 2: 596. 1805.

Uredo Zeae Mays DC., Lamarck Enc. Meth. Bot. 8: 229. 1808.

Uredo maydis DC. Fl. Fr. 6:77. 1915.

Uredo Zeae Schw. Fung. Car. No. 71. 1815.

Caeoma Zeae Link, Willd. Sp. Pl. 62:2. 1825.

Erysibe Maydis Wallroth, Fl. Crypt. Germ. 2: 215-16. 1833.

Ustilago Maydis Corda, Corda Icon. Fung. 5:3. 1842.

Ustilago Schweintzii Tul. Ann. Sci. Nat. Bot. III. 7:86. 1847.

Ustilago Zeae-Mays Wint. Rab. Krypt.-Fl. 11-97. 1881.

Ustilago Huchlaenae Arcang. Erb. Critt. Ital. 11: 1132. 1882.

Ustilago Mays zeae Magnus, Deuts. Bot. Monatschr. 18:50. 1895.

293

Sori forming on any part of the host above ground as irregular pustules measuring

a few mm. to large boils several dm. in diam., at first covered with a membrane composed

of host tissue intermixed with fungous threads which later breaks revealing an olive-brown

spore mass; spores globose to subglobose or sometimes ellipsoidal, regular, reddish-brown,

7-10 » diam., bluntly echinulate.

Type host and locality: On Zea Mays Linn., Europe.

On Andropogoneae: Zea Mays Linn., Cape Colony (M.H. 1178, 14699, 11632), Natal,*

Transvaal.*

Distribution: North America, Kurope, Asia, Africa, Philippine Islands.

Ustilago bromivora (Tul.) Fisch. de Waldh. Bull. Soc. Nat. Moscow 401: 252. 1867.

Ustilago carbo vulgaris bromivora Tul. Ann. Sci. Nat. Bot. III. 7:81. 1847.

Cintractia patigonica Cooke and Massee, Grevillea 18:34. 1899.

Ustilago Brachypodii Maire, Bull. Soc. Hist. Nat. Afr. Nord 9:46. 1918.

Ustilago Brachypodii-distachyi Maire, Bull. Soc. Hist. Nat. Afr. Nord 10:46. 1919.

Ustilago bromivora Tul. forma Brachypodi Hariot, Bull. Soc. Hist. Nat. Afr. Nord 12 : 192-

1921.

Ustilago bromi-arvensis Liro, Ann. Acad. Sci. Fenn. A. 17:93. 1924.

Ustilago Bromi-mollis Liro, Ann. Acad. Sei. Fenn. A. 17:94. 1924.

Sori in the spikelets, usually confined within the glumes, sometimes infecting base

of glumes, covered with a delicate white membrane, usually bullate, agglutinated then

powdery ; spores globose-subglobose, occasionally broadly ellipsoidal, dark reddish-brown

to olivaceous-brown, chiefly 7-11 p diam., abundantly verruculose.

Type host and locality: On Bromus secalinus Linn., Europe.

On Festuceae: Bromus unioloides H.B.K., Cape Colony (M.H. 544, 1248, 11004,

12832), Orange Free State (M.H. 2091), Transvaal (M.H. 280, 284, 285, 195, 429, 6933).

Distribution: North America, Europe, Asia, Africa, Australia, New Zealand.

Ustilago Andropogonis-finitimi Maub. Bull. Soc. Myc. (France) 22: 74-75. 1906.

Sori in the ovaries, long linear, 5-7 mm. long, covered with a membrane of host tissue ;

spore mass brown, agglutinated, surrounding a well developed columella ; spores globose-

subglobose, semi-opaque, dark reddish-brown, 9-12 yw diam., abundantly echinulate unde-

oil immersion.

Type host and locality: On Andropogon finitimus Hochst. Portuguese East Africa

On Andropogoneae: Andropogon finitimus Hochst., Portuguese Hast Africa.

Distribution: Reported only from type locality.

Ustilago Rabenhorstiana Kuhn, Hedwigia 15:4. 1876.

Caeoma Syntherismae Schw. Trans. Am. Phil. Soc. II. 4: 290. 1834.

Ustilago Setariae Rabenh. Univ. itm. Krypt. No. 1866. Year *

Ustilago destruens var. Digitariae Sacc. Nuo. Giorn. Bot. Ital. 8: 167. 1876.

Ustilago Cesati Fisch. de Waldh. Apercu 25: 1877. p.p.

Ustilago Syntherismae Auct. p.p. Cke. Grevillea 6: 138. 1878.

Sori usually destroying the entire inflorescence, linear-oblong, 3-5 cm. long, at first

concealed by the enveloping glumes but finally becoming visible as a black-brown dusty

spore mass surrounding the elongate remnants of the inflorescence ; spores globose—sub-

globose, occasionally somewhat angled, olivaceous-brown, 10-13 w diam., verruculose.

Type host and locality: On Panicum sanguinale Linn., Europe.

On Paniceae: ‘Digitaria ternata Stapf, Natal (M.H. 11703); Digitaria sp., Rhodesia

(M.H. 13999), Transvaal (M.H. 11704).

1 This species is similar to, if not identical with Ustilago sphaerogena Burrill, found in North America.

294

Ustilago Peglerae Sydow and Bubak, Ann. Myc. 12: 264. 1914.

Sori destroying the anthers, olivaceous-black ; spores ellipsoidal to oblong, seldom

globose to subglobose, frequently irregular, olivaceous-brown, 10-14 w long, verruculose.

Type host and locality : On Ornithogalum lacteum Jacq., Cape Province, Union of South

Africa.

On Liliaceae: Ornithogalum lacteum Jacq., Cape Colony (M.H. 7101).

Distribution: Cape Province.

Ustilago Evansii P. Henn. Bot. Jahrb. (Engler) 41: 270. 1908.

Sori destroying all ovaries on the spike, 2-3 mm. long, covered with an olivaceous

membrane ; spore mass granular, olive-brown ; spores globose-subglobose, regular, oliva

ceous-brown, light colored almost hyaline, spores abundant throughout the sorus, variou~

sizes, 14-21 p diam., abundantly and coarsely echinulate, bifurcate conidiophores abundant

Type host and locality: On Setaria sphacelata Stapf and Hubb. (=Setaria aurea A-

Br.), Zoutpansberg, Transvaal, Union of South Africa. <

On Paniceae: Setaria sphacelata Stapf. and Hubb., Natal (M.H. 7757), Transvaal

(M.H. 7797), Rhodesia (M.H. 14003), Zululand (M.H. 15441, 17044, 17045); Setaria niqri-

rostris Dur. and Schinz., Transvaal, Rhodesia; Setaria sp., Southern Rhodesia.

Farysia Raciborski, Sci. Cl. Sci. Math. Nat. 1. 1909: 354. 1909.

Elateromyces Bubak, Houby Céskaé Dil. II. 1912:32. 1912.

Sori in various parts of the host, at maturity forming dusty, usually dark spore masses,

intermixed with parallel, elater-like strands of host tissue and sterile hyphae ; spores single,

produced in chains as in Ustilayo but intermixed with sterile hyphae and strands of host

tissue which function as elaters.

Farysia olivacea DC. H. and P. Sydow, Ann. Myc. 17:41. 1919.

Uredo olivacea DC. Fl. Fr. 6:78. 1815.

Caeoma olivaceum Schlecht. Fl. Berol. 2: 130. 1824.

Brysibe olivacea Wallr. Fl. Crypt. Germ. 2: 215. 1833.

Ustilago olwacea Tul. Ann. Sei. Nat. III. 7:88. 1847.

Ustilago catenata Ludw. Zeits. Pflanz. 8: 139. 1893.

Cintractia caricicola P. Henn, Hedwigia 34: 325. 1895.

Ustilago caricicola Tracy and Karle, Bull. Torrey Club 26: 493. 1899.

Ustilago subolivacea P. Henn. Ann. R. Istit. Bot. Roma 6:84. 1897.

Hlateromyces olivacea Bubak, Houby Cesky Dil. II. 1912: 32. 1912.

Farysia americana Ciferri, Ann. Myc. 29:73. 1931.

Sori in occasional ovaries, often at first partly concealed by the perigynium, ovate,

2-6 mm. in diam., at first with agglutinated spores which later become powdery, with

conspicuous elaterlike threads intermixed with the spores; spores olivaceous-brown,

irregular, varying from globose to oblong or linear but sometimes more regular and then

chiefly 7-9 1 diam., the most elongate about 12 ~ long and about 4 wide, abundantly but

minutely verruculate.

Type host and locality: On Carex riparia Curtis, France.

On Cyperaceae: Carex ethiopica Schkuhr., Cape Province; Carex phacota Spreng.,

Cape Province. (M.H. 8812.)

Distribution: North America, South America, West Indies, Europe, Asia, South

Africa, New Zealand, Tasmania, Australia,

Sphacelotheca De Bary, Verg. Morph. Biol. Pilee 187. 1884.

Sporisorvum Ehrenb. Link Willd. Sp. Pl. 62:86. 1825.

Endothlaspis Sorokin, Rev. Mye. 12:4. 1890.

295

Sori usually in the inflorescence, often limited to the ovaries, provided with a definite

(more or less) temporary false membrane covering a dusty spore mass and a central columella

of plant tissue; false membrane composed largely or entirely of definite sterile fungous

cells which are hyaline or slightly tinted ; oblong to spherical, and usually more or less

firmly bound together; spores simple, usually reddish-brown, developed in a somewhat

centripital manner as in Cintractia, small to medium in size; germination as in Ustilago.

Type: Uredo Hydropiperis Schum., on Polygonum Hydropiper Linn., Europe.

Sphacelotheca Anthephorae (Syd.) Zundel n.n.

Ustilago Anthephorae Syd. Ann. Myce. 12: 197. 1914.

Sori completely destroying the inflorescence hidden by the glumes, about 1 cm. long,

surrounded by a delicate membrane which dehisces apically revealing an agglutinated

brown spore mass surrounding a well formed columella and breaking up into sterile cells ;

sterile cells subglobose—-irregular, hyaline, singly or in chains, about 7 diam.; spores

globose-subglobose, reddish-brown, 3.5-4.5 w diam., smooth.

Type host and locality: On Anthephora pubescens Nees, Grootfontein, South West

Africa.

On Zoysieae: Anthephora pubescens Nees, South West Africa,* Transvaal (M.H.

2249, 5151), Orange Free State (M.H. 26644).

Distribution : Southern Africa.

* Spores smooth.

Sphacelotheca Sorghi (Link) G. P. Clinton, Jour. Myc. 8: 140. 1902.

Sporisorium Sorghi Link, Willd. Sp. Pl. 62:86. 1825.

Tilletia Sorghi-vulgaris Tul. Ann. Sci. Nat. II..7: 116. 1847.

Ustilago sorghi Pass. Thum. Hedwigia 12: 114. 1873.

Cintractia Sorghi-vulgaris G. P. Clinton, Bull. III. Agr. Exp. Sta. 47: 404. 1897.

Sori destroying the ovaries which are elongated about twice the normal length of the

seed, covered with an evident false membrane which ruptures revealing a brown spore

mass surrounding a short, thick, well developed columella ; false emmbrane usually breaking

up into chains of small hyaline sterile cells, subglobose-ellipsoidal, 3-10 « diam. ; spores

globose-subglobose, reddish-brown, 3-8 yw diam., mostly about 5 uw diam., smooth.

Type host and locality: On Sorghum vulgare Pers., Egypt.

On Andropogoneae: Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehder

(=S. caffrorum Beauv. and =Andropogon sorghum Brot. var.), Cape Province, Natal

(M.H. 17271), Transvaal (M.H. 315, 316, 5635, 2088, 7786).

Distribution: Co-extensive with cultivated sorghums.

Sphacelotheca Moggii Zundel, Mycologia 22: 130. 1930.

Sori in the inflorescence, long linear, 5-10 mm. long, at first usually hidden by the

outer leaf sheath, later protruding, covered with a false tissue which flakes away revealing

a dark brown, agglutinated spore mass surrounding a well developed, often forked, columella ;

sterile tissue very fragile and “ tissue like,” somewhat effervescent, adhering more or less

to the sori and breaking up into groups or balls of sterile cells which are tinted brown,

vacuolated and granular, 7-12 y diam., sterile-cell balls containing 4 or more cells, globose—

subglobose, 15-22 w diam. : balls of sterile cells are scattered throughout the sori; spores

globose-subglobose, regular, tinted olivaceous-brown to almost hyaline, vacuolated, 4-6 pw

diam., under oil immersion, smooth.

Type host and locality: On Cymbopogon plurinodis Stapf, Armoedsvlakte, British

Bechuanaland.

On Andropogoneae: Cymbopogon plurinodis Stapf, British Bechuanaland. (M.H.

19859.)

Distribution: Not reported except from type locality.

296

Sphacelotheca cruenta (Kuhn) Potter, Phytopath. 2: 98. 1912.

Ustilago cruenta Kiihn, Hamburg Gart. Blumenztig. 28: 177-178. 1872.

Ustilago Tulasnei Kiihn, Rab. Fungi Eur. No. 1997. 1875.

Sphacelotheca cruenta Bubak, Archiv Pirod. Vyck. Cech. 15:27. 1912. Archiv. Naturw.

Landes. Bohmen 15: 26. 1916..

Sori destroying the ovaries, which are scarcely larger than normal, covered by an evident

membrane which flakes away into globose sterile cells revealing a brown granular spore

mass surrounding a long, curved, well developed columella ; groups of large globose sterile

cells scattered throughout the sori; sterile cells hyaline, globose-subglobose, singly or in

groups, 9-14 w diam. ; spores globose-subglobose, light reddish-brown, 5-8 w diam., smooth.

Type host and locality: Sorghum vulgare Pers., Halle, Germany.

On Andropogoneae: Sorghum halepense Pers., Tanganyika Territory ; Sorghum

vulgare Pers., Tanganyika Territory ; Sorghum sp., Tanganyika Territory.

Distribution: North America, Europe, West Indies, Africa.

Sphacelotheca Vryburgii Zundel, Mycologia 28: 298. 1931.

Sori in the inflorescence, long lnear, 5-10 mm. long, at first hidden by the glumes

but later protruding, covered by a reddish-brown, delicate false membrane which flakes

away revealing an agglutinated black spore mass surrounding a well developed much

branched columella ; sterile cells hyaline, globose, usually in groups, 9-15 w diam. ; spores

globose-subglobose, occasionally angled, very light reddish-brown, 4—8 » diam., smooth,

contents finely granular with a hyaline to light coloured wall under oil immersion.

Type host and locality: On Themeda triandra Forsk. (=Themeda Forskalii Hack.),

Vryburg, British Bechuanaland.

On Themeda triandra Forsk., British Bechuanaland (M.H. 9733.)

Distribution: Reported only from type locality.

Sphacelotheca concentrica Zundel, Mycologia 22: 138. 1930.

Sori in the inflorescence, broadly elongate, 1 cm. or less in length, at first concealed

by the glumes, covered by an evident light coloured false membrane which flakes away

revealing a partially agglutinated dark spore mass surrounding a well formed columella ;

sterile tissue breaking up into hyaline cells, globose, somewhat variable in size, ranging

from 10-21 diam. ; spores globose-subglobose, under oil immersion the spore is divided

into four concentric parts, an outer dark brown area, then a light reddish-brown area and

an inner vacuolated, ight coloured area, surrounded by a second dark brown area, 6-8 ju

diam. but sometimes 4-8 p diam., smooth.

Type host and locality: On Cymbopogon plurinodis Stapf, Pretoria, Union of South

Africa.

On Andropogoneae: Cymbopogon plurinodis Stapf, Transvaal (M.H. 10708).

Distribution: Not reported except from type locality.

Sphacelotheca densa (McAlp.) Ciferri, Ann. Myc. 26:32. 1928.

Cintractia densa McAlp., Smuts of Australia 168. 1910.

Sori destroying the inflorescence and forming a long rachis, covered with a greyish

membrane which flakes away exposing a dark spore mass; spores globose—subglobose,

light olivaceous-brown, 6—7 yx diam., smooth.

Type locality and host: On Rottboellia compressa Linn., Burnley near Melbourne,

Victoria, Australia.

On Andropogoneae: Rottboellia compressa Linn., Natal (M.H. 12957).

Distribution: Australia, South Africa.

Sphacelotheca Doidgeae Zundel, Mycologia 22: 131. 1930.

Sori in the inflorescence usually involving the entire spikelet along the rachis, long

linear, frequently irregularly branched or compound, 3-8 mm. long, covered with an evident,

297

thick, brown, false membrane, which dehiscee from the apex disclosing a brown, agglutinated

mass of spores surrounding a well developed irregular columella ; sterile tissue breaking

up into groups or chains of hyaline sterile cells, 6-10 w diam.; groups of large globose

sterile cells through the sori; spores globose-subglobose, thick walled, olivaceous brown,

6-10 « diam., under oil immersion, smooth and finely granular.

Type host and locality: On Bothriochloa glabra A. Camus, Edendale, Natal, Union

of South Africa (M.H. 1997).

On Andropogoneae: Bothriochloa sp. Transvaal (M.H. 15058); Andropogon inter-

medius R. Br., Natal (M.H. 8939) ; Bothriochloa glabra A. Camus, Natal (M.H. 1997).

Distribution : Southern Africa. ,

Sphacelotheca tenuis (H. and P. Sydow) Zundel, Mycologia 22: 137. 1930.

Ustilago tenuis H. and P. Sydow, Ann. Myc. 4: 425. 1906.

Sor) destroying the inflorescence, $-1 cm. long, covered with a more or less permanent

false membrane which flakes away revealing a semi-powdery spore mass surrounding a

well developed columella ; cells of sterile membrane are inclined to fuse and largely lose

their cellular structure, appearing as a more or less amorphous mass, however, some globose

cells retain their identity; spores globose-subglobose, somewhat irregular and angular,

thick walled, olivaceous-brown, 6-10 diam., smooth and finely granular under oil immersion.

Type host and locality: On Bothriochloa pertusa A. Camus (=Andropogon pertusus

Willd.), Hunsur, Mysore, India.

On Andropogoneae: Hyparrhenia sp., Natal (M.H. 11862).

Distribution: India, South Africa.

Sphacelotheca columellifera (Tul.) Ciferri, Ann. Myc. 26:32. 1928.

Ustilago carbo var. columellifera Tul. Ann. Sci. Nat. Bot. III. 7:81. 1847.

Cintractia columellifera (Tul.) McAlpine, Smuts of Austr., 166. 1910.

Sori destroying the inflorescence, long linear, 5-7 cm. long, at first concealed by the

sheath but later protruding, covered by an evident yellowish-white, false membrane which

flakes away revealing a dark brown agglutinated spore mass surrounding a well developed,

hollow columella ; false membrane disintegrating into groups or chains of globose, hyaline

sterile cells, 7-12 ~ diam. ; spores generally globose, regular but occasionally subglobose,

light reddish-brown, usually 7 ~ diam., but occasionally 9 4, under oil immersion, smooth

with vacuolated contents.

Type host and locality: Andropogon australis Spreng., Queensland, Australia.

On Andropogoneae: Heteropogon contortus R. and 8. (=Heteropogon hirtus Pers.),

Tanganyika Territory.

Distribution: Australia, Africa.

Sphacelotheca Ruprechtii Sydow, Ann. Myc. 88: 232. 1935. _

Sori entirely destroying the inflorescence, cylindrical, 4-8 mm. long, at first concealed

by the glumes but later protruding, covered by a leathery, olivaceous false membrane

which ruptures disclosing a dark agglutinated spore mass surrounding a well developed

simple or bifureate columella ; sterile cells globose-subglobose, hyaline, in pairs or in groups,

rarely in chains, 10-14 w diam., spores globose-subglobose, regular, olivaceous-brown

7-10 » diam., smooth but granular under the oil immersion lens.

Type host and locality: On Hyparrhenia Ruprecht. Fourn. (Andropogon Ruprechtiv

Hack.), Marikana, Rustenburg, Transvaal, Union of South Africa.

On Andropogoneae: Hyparrhenia Ruprechtii Fourn., Transvaal (M.H. 27377).

Distribution: Southern Africa.

Sphacelotheca Evansii Zundel, Mycologia 22: 133. 1930.

Sori in the inflorescence, hidden by the glumes, inconspicuous, 5-10 mm. long, covered

by an evident membrane which flakes away revealing a dark brown spore mass surrounding

298

a well developed columella ; false tissue rather permanent, breaking up into large groups

or chains of sterile cells; groups of sterile cells through the sorus; sterile cells hyaline,

uregular, globoid, 9-12 diam.; spores globose-subglobose, regular, olivaceous-brown,

8-10 » diam., under oil immersion, smooth and vacuolated.

Type host and locality: On Hyparrhenia Ruprechtw Fourn., Olifants River, Transvaal,

Union of South Africa.

On Andropogoneae: Hyparrhenia Ruprecht. Fourn., Transvaal (M.H. 14174).

Distribution: Transvaal.

Sphacelotheca Andropogonis (Opiz) Bubak, Naturw. Landes. Bohmen, 15:25. 1916.

Ustilago Ischaemi Fuckel, Enum. Fung. Nass. 22: 1861.

Ustilago cylindrica Peck, Bot. Gaz. 7:55. 1882.

Cintractia Ischaemi Syd. Oesterr. Bot. Zeitsch. 51:12. 1901.

Sphacelotheca Ischaemi Clint. Jour. Myc. 8: 140. 1902.

Sori usually involving entire inflorescence, hidden by the sheath, long lnear, 10-40

mm. long by 1-4 mm. wide, covered by a false membrane which flakes away disclosing a

brown spore mass surrounding a well developed columella; false membrane rather per-

manent, breaking up into large masses of tissue rather than individual sterile cells, sterile

cells through the sori; sterile cells globose-subglobose, flattened when in contact with

each other, hyaline or when en masse, tinted brown, 7-16 diam. ; spores globose -subglobose

medium reddish-brown, minutely granular, 8-10 » diam. smooth.

Type host and locality : On Andropogon Ischaemum Linn., Prag, Czechoslovakia.

On Andropogoneae: Bothriochloa glabra A. Camus. Natal (MH. 1080, 7759), Transvaal

(M.H. 1073, 1921) ; Cymbopogon Schoenanthus Spreng., Transvaal (M.H. 1921); Hyparr-

hena Ruprecht Fourn., Transvaal (M.H. 1156, 10096).

Distribution: North America, Europe, Asia, Africa, Philippine Islands.

Sphacelotheca Dinteri (H. and P. Sydow) Zundel, Mycologia 22: 140. 1930.

Ustilago Dinterr H. and P. Sydow, Ann. Myc. 18:37. 1915.

Sori destroying the entire enflorescence, almost entirely hidden by the terminal sheath,

long linear, 2-4 em. long, covered by an evident brown false membrane which flakes away

Trev ealing a seml- powdery, brown, spore mass surrounding a well formed columella ; sterile

cells globose-subglobose or sometimes ellipsoidal, individually or in groups, tinted brown,

7-12 w diam.; spores globose-subglobose or occasionally ellipsoidal, frequently angular,

thick walled, finely granular, olivaceous brown, 9-12 w diam., smooth.

Type host and locality: On Dicanthium papillosum Stapf (=Andropogon papillosus

Hochst.), Pijikuara-Okaharni, South West Africa.

On Andropogoneae: Dichanthium papillosum Stapf, South West Africa.

Distribution: South West Africa.

Sphacelotheca natalensis Zundel, Mycologia 22:139. 1930.

Sori in the inflorescence, long linear, 3-6 mm. long, covered by an evident brown false

membrane which flakes away revealing an agglutinated spore mass surrounding a well

developed, simple, columella ; sterile cells globose, usually hyaline, mostly in groups or

short chains, reddish-brown en masse, variable in size, 12-15 » diam. ; spores globose—

subglobose, thin walled, ight reddish-brown, 10-12 4 diam., smooth even under oil immersion.

Type host and locality: On Cymbopogon excavatus Stapf, Mooi River, Natal, Union

of South Africa.

On Andropogoneae : Cymbopogon excavatus Stapf, Natal (M.H. 11705).

Distribution: Not reported except from type locality.

Sphacelotheca transvaalensis Zundel, Mycologia 22: 139. 1930.

Sori destroying the inflorescence, broadly linear, 5-10 mm. long, covered by a thick,

dark brown, false membrane which flakes away revealing a black spore mass surrounding

299

a large, well developed, branched, root-like, central columella and numerous surrounding

smaller columellae. (Resembling a small root system of an herbaceous plant); sterile

cells globose-subglobose, hyaline, delicate, large, single or in short chains, 11-12 » diam. ;

spores globose-subglobose, regular, reddish-brown, 10-12» diam., smooth but finely granular

under oil immersion.

Type host and locality: On Sorghum versicolor Anderss., Onderstepoort, Pretoria,

Transvaal, Union of South Africa.

On Andropogoneae: Sorghum versicolor Anderss. Transvaal (M.H.. 17047).

Distribution: South Africa.

** Spores not smooth.

Sphacelotheca Milbraedii (H. and P. Sydow) Zundel, Mycologia 22: 135. 1930.

Ustilago Milbraedv. H. and P. Sydow, Wissensch. Ergebn. Deutsch Zentral. Exped. 1907-

IOS 3 sy, WOIN.

Sori in the ovaries, long linear, 3-5 em. long, covered with an evident false membrane

which flakes away disclosing a semi-powdery spore mass surrounding a well formed columella,

outer false membrane rather persistent, breaking up into large groups of sterile cells, rec-

tangular, tinted brown, groups of sterile cells through the sori, subglobose, tinted brown,

9-12 w diam. ; spores globose-subglobose, irregular, somewhat angular, thin walled, light

brown with a darker coloured centre, 3-8 » diam., under oil immersion faintly echinulate.

Type host and locality : On Cymbopogon Schoenanthus Spreng. (= Andropogon Schoen-

anthus Linn.), Mpororo, Tanganyika Territory.

On Andropogoneae : Cymbopogon Schoenanthus Spreng., Tanganyika Territory.

Distribution: Tanganyika Territory.

Sphacelotheca Amphilophis Sydow, Ann. Myc. 38: 232. 1935.

Sori destroying the entire inflorescence, 1-2 cm. long, at first covered by a false mem-

brane which flakes away revealing a dark brown agglutinated spore mass surrounding a

large, well developed, simple or bitrifurcate columella; sterile cells scattered throughout

the sori, usually in rather agglutinated groups but occasionally single or in groups of two

or three, hyaline, globose-subglobose, regular, 7-13 cham. ; spores globose-subglobose,

rarely ellipsoidal, regular, olivaceous—brown, 5-7 » diam., apparently smooth but under

the oil immersion lens sometimes indistinctly verruculose.

Type host and locality: On Bothriochloa insculpta A. Camus (= Amphilophis insculpta

Stapf), along Crocodile River at Schagen, Barberton, Transvaal.

On Andropogoneae: Bothriochloa insculpta A. Camus, Transvaal (M.H. 26023).

Distribution: Reported only from type locality.

Sphacelotheca Ritchiei Zundel, Mycologia 22: 138. 1930.

Sori in the inflorescence, long linear, 5-8 mm. long, sometimes gregarious, at first

concealed by the glumes, later protruding, covered by an evident dark brown false membrane

which flakes away apically, revealing a brown spore mass surrounding a well developed

columella ; sterile cells hyaline, in pairs, in short chains or in groups (usually in pairs)

usually larger than the spores, 9-12 » diam., sometimes up to 15 yw, thin walled and some

what fragile; spores globose-subglobose, regular, reddish-brown, 6-10 » diam., under o1

immersion minutely verruculate.

Type host and locality: On Hyparrhenia cymbaria Stapf, Morogoro, Tanganyika

Territory.

On Andropogoneae : Hyparrhenia cymbaria Stapf, Tanganyika Territory (M.H. 20650).

Distribution: Not reported except from type locality.

Sphacelotheca Holci H. 8. Jackson, Monogr. Univ. Puerto Rico, Ser. B., No. 2: 259. 1934.

Sori in the ovaries, concealed by and not exceeding the glumes, covered by a membrane

which flakes away as sterile cells (not found in sori of old specimens) revealing a granular

300

Spore mass surrounding a well formed columella ; sterile cells throughout the sorus, globose—

ellipsoidal, singly, in pairs or groups, slightly tinged very light brown, 9-17 pw diam. ; spores

globose-subglobose, olivaceous-brown with a dark reddish-brown epispore, 7-10 diam.,

finely but ev vident verruculose to echinulate.

Type host and locality.

On Andropogoneae: Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehder.

(=Sorghum caffrorum Beauv.), Tanganyika Territory.

Distribution: Venezuela and Tanganyika Territory.

Sphacelotheca Zilligii Zundel, Mycologia 22: 142. 1930.

Sori in the inflorescence, solitary, long linear, at first concealed by the sheath, 1-3 cm.

long, covered by an evident brown false membrane which flakes away revealing a dark

brown, granular spore mass surrounding a well developed, much branched, columella :

sterile cells globose-subglobose, hyaline, usually in groups or chains, angular by compression,

variable in size, 8-14 u diam. ; spores globose-subglobose, semi-regular, light reddish-brown,

7-10 » diam., medium echinulate under oil immersion.

Type host and locality : On Cymbopojon sp., Vryburg, Cape Province, Union of South

Africa.

On Andropogoneae : Cymbopogon sp., Cape Province (M.H. 20666).

Distribution: Reported only from type locality.

Sphacelotheca Pappophori (Pat.) Zundel nn.

Ustilago Pappophori Pat. Bull. Soc. Myce. (France) 22: 199. 1906.

Ustilago Pappophori Sydow, Ann. Myc. 24: 265. 1926.

Sori in the ovaries, causing complete destruction, ovoid, about 1-1.5 by 5-8 mm.,

covered by a false membrane which flakes away as sterile cells revealing a dark brown

spore mass surrounding a columella: sterile cells abundant, subglobose— ellipsoidal, often

irregular, about the size of the spores, singly or in groups, hyaline or tinged yellow ; spores

globose-subglobose, occasionally ellipsoidal, hght olivaceous-brown with a narrow, dark,

reddish-brown epispore, 7-11 w diam., finely but abundantly verruculate.

Type host and locality : On Pappophorum scabrum Kunth, Selah ad Ahaggar, Algeria.

On Festuceae: Hnneapogon sp. (M.H. 17279), South West Africa; Pappophorum

scabrum Kunth,* South Africa.

Distribution: Algeria, Tunis, South Africa.

Sphacelotheca Nyassae (H. and P. Sydow) Zundel, Mycologia 22: 133. 1930.

Ustilago Nyassae H. and P. Sydow, Ann. Myce. 18: 156. 1920.

- Sori in the ovaries which remain about normal size, inconspicous, concealed by the

elumes, 5 mm. long, covered by an evident false membrane which ruptures revealing a

brown, powdery spore mass surrounding a simple columella ; false membrane disintegrating

into hyaline, globose sterile cells, 11— 16 p diam.; spores globose-subglobose, sometimes

angular, re »ddish-brown, 9-12 be diam., under oil Tame RRSONO finely verruculose and coarsely

vacuolated.

Type host and locality: On Andropogon sp., Nyassa-Hochland, Station Kyimbila,

Nyasaland Protectorate.

On Andropogoneae: Andropogon sp., Nyasaland Protectorate.

Distribution: Not reported except from type locality.

Sphacelotheca monilifera (Ellis and Kv.) G. P. Clinton, Jour. Myce. 8: 141. 1902.

Ustilago monilifera Ellis and Kv. Bull. Torrey Club 22: 362. 1895.

Sori in the ovaries of the spikelets, 5-7 mm. long or about the length of the glumes,

at first concealed by the glumes, covered with an evident false membrane that flakes away

revealing a brownish-black spore mass with evident columella ; cells of the false membrane

301

adhering semi-permanently, interior sterile cells globose-subglobose, singly, in pairs or

in groups, 9-14 y diam., tinted light yellowish-brown ; spores globose-subglobose, usually

regular but sometimes angular, olivaceous-brown, 9-14 4 diam., under oil immersion minutely

verruculose or echinulate.

Type host and locality : On Heteropogon contortus (L.) Roem. et Schultz (= Andropojon

contortus Linn.), Tucson, Arizona, United States.

On Andropogoneae: Heteropogon contortus (Linn.) Roem. et Schultz, Tanganyika

Territory.

Distribution: South-Western United States, Mexico, Hawai, Tanganyika Territory

Sphacelotheca pretoriense (Pole-Evans) Zundel n.n.

Ustilago pretoriense Pole-Evans, Ann. Myc. 12: 263. 1914.

Sori destroying the inflorescence, 1.5—2 cm. long, tubular, covered with a dark brown

membrane which dehisces apically revealing a brown spore mass surrounding a well formed

columella; sterile cells 7-13 y diam., globose—ellipsoidal, somewhat irregular, usually

granular, hyaline ; spores globose—broadly ellipsoidal, regular light reddish-brown, 7-13 yu

diam. but chiefly 10-13 yw, echinulate to spiny.

Type host and locality: On Urochloa helopus Stapf, Pretoria, Transvaal, (M.H. 7408)

On Paniceae: Uvochloa helopus Stapf. (=Panicum. Helopus Trin.), Transvaal (M.H.

7799, 8926).

Distribution: Transvaal.

Sphacelotheca modesta (Sydow) Zundel nn.

Ustilayo modesta Sydow, Ann. Myc. 88: 231. 1935.

Sori in the ovaries, causing complete destruction, covered by a delicate yellowish

membrane which flakes away as sterile cells disclosing a dark brown spore mass surrounding

a columella; sterile cells abundant, subglobose-ellipsoidal, singly, in pairs or in chains,

hyaline, vacuolated, usually 5 x 7 pw, occasionally larger: spores slobose-subglobose,

rarely ellipsoidal, olivaceous-brown with a reddish-brown epispore, 10-14 w diam., abun-

dantly but minutely verruculose.

Type host and locality: On Enneapogon brachystachyus Stapf, Prieska, Cape Province

Union of South Africa.

On Festuceae: Enneapogon brachystachyus Stapf, Cape Province (M.H. 23506).

Distribution: Reported only from type locality.

Sphacelotheca flagellata (Sydow) Zundel n.n.

Ustilago flagellata Sydow, Ann. Myc. 9: 144. 1911.

Sori destroying the inflorescence, forming along the rachis as a long columella, 8-30 cm.

long, at first covered by a brown membrane which later flakes away as sterile cells and

revealing a dark brown spore mass ; spores globose-subglobose, dark brown, 10-14 yw diam.,

minutely echinulate; sterile cells globose-subglobose, in groups or chains, hyaline, 10-

14 » diam.

Type host and locality : On Rottboellia exaltata Linn., Province Rizal, Luzon, Philippme

Tslands.

On Andropogoneae: Rottboellia exaltata Linn., Tanganyika Territory ; Rottboellia

compressa Linn. f. Transvaal (M.H. 20331.)

Distribution: Philippine Islands and Tanganyika Territory, South Africa.

This species differs from Sphacelotheca columellifera (Tul.) Cif. mainly in having larger

darker coloured and finely echinulate spores.

Sphacelotheca Stuhlmanni (P. Henn.) Zundel, Mycologia 22:136. 1930.

Sori in the ovaries, long linear, usually 7-10 cm. long, covered with an evident brown

false membrane which flakes away disclosing a brown, somewhat agglutinated spore mass

302

sterile tissue disintegrating into large groups of sterile cells ; often in chains which sometimes

collapse, tinted brown; spores globose-subglobose, sometimes angular, thick walled,

reddish-brown, 9-14 diam., under oil immersion minutely echinulate.

Type host and locality: On Andropogon sp., Ukami, Mrigogo, Central Africa.

On Andropogoneae: Andropogon sp., Tanganyika Territory.

Distribution: Central Africa and Tanganyika Territory.

Melanopsichium G. Beck, Ann. Nat. Hofmus. Wien 9: 122. 1894.

Sori on various parts of the host, firmly agglutinated and conspicuous spore masses ;

spores simple, developed in irregular chambers or groups arising from a mixture of plant

tissue and fungous threads, thus giving a tubercular character to the sorus, enveloped by

a more or less permanent gelatinous envelope, discharging from spore mass by absorption

of water, of medium size; germination as in Ustilago.

Type: Ustilago austro-americanum Speg.', on Polygonum incarnatum auct., *Missouri,

United States. (Rabenh. Fungi Eur. No. 35901.)

Melanopsichium austro-americanum (Speg.) G. Beck, Ann. Nat. Hofmus. Wien 9: 122,

1894.

Ustilago austro-americanum Speg. Anal. Soc. Ci. Argent. 12:63. 1881.

Sphacelotheca austro-americanum Liro, Anal. Acad. Sci. Fennicae, Ser. A., 17: 124. 1924.

Sori usually in the inflorescence, occasionally on the leaves and then smaller, forming

irregular lobed masses arising from fusion of infected parts, forming a hard agglutinated

spore mass mixed with plant tissue ; spores subglobose-—ellipsoidal, often irregular with

more or less evident gelatinous env elopes, light golden brown, chiefly 7-14 uw diam. smooth.

Type host and locality : On Polygonum acre HBK., Argentina.

On Polygonaceae : Polygonum lapathifolium L. v. glabrum Burtt Davy, Natal (M.H.

20435).

Distribution: North America, South America, Europe, Asia, Africa.

The South African specimen is a rather unusual form with the sori on the leaves.

Cintractia Cornu, Ann. Sci. Nat. VI. 15: 279. 1883.

Anthracoidea Bref. Unters. Gesammt. Myk. 12: 144. 1895.

Sort on various parts of the host, often in the ovaries, forming a black, usually rather

firm, agglutinated, spore mass; spores simple, usually of medium or large size and of reddish-

black colour, formed in a centripetal manner from fertile stroma usually surrounding a

central columella of plant tissue, often freed from sorus by absorption of water ; germination

apparently of a modified Ustilago type.

Type: Ustilago axicola Berk., on Cyperus sp.3, North America.

Cintractia Melinis Zundel n. sp.

Sori destroying the ovaries, about 1 mm. long, at first agglutmated but later somewhat

powdery ; spores globose-ellipsoidal, irregular, somewhat angled, reddish-brown, 7—l1 pu

diam., apparently smooth but minutely echinulate under oil immersion lens.

Hab. in ovaries of Melinis tenuinervis Stapf, Capetown, Cape Province, Union of South

Africa, Coll. C. W. Malley, June 12, 1914. M.H. 19860.

Latin description : —

Soris ovaria destruentibus, ca. 1 mm., primum conglutinatis, deinde subpulverulentis ;

sporis globosis v. ellipsoideis, irregularibus, Sibadeulemioue! rubro-brunneis, 7-11 w diam.,

specis levibus sed minute echinulatis (sub oleo visis).

Hab. in ovariis Melinis tenwinervis Stapf, Capetown, Cape Province, in Unione Africae

australis, Coll. C. W. Malley, June 12, 1914. M.H. 19860.)

1The original type of U. austro-americanum was described form South America by Spegazzini

in Anal. Soc. Ci. Argent. 12: 63. 1881.

2 Polygonum incarnatum Auct. is now considered to be P. lapathifoliwm Linn.

3 Clinton suggests, Proc. Boston Soc. Nat. Hist. 31 : 397. 1904, that this was really Fimbristylis.

303

Sorosporium Rudolphi, Linnaea 4:116. 1829.

Sori in various parts of the host, forming dusty dark coloured spore masses ; spore

balls composed of numerous fertile cells; often rather loosely united and frequently at

maturity completely separating, of medium size; spores usually olive or reddish-brown,

of medium size; germination similar to that of Ustilago, sometimes with elongated germ

thread and no sporidia.

Type: Sorosporium Saponariae Rud., on Saponaria officinalis, Germany.

* Spores smooth.

Sorosporium pretoriaense Zundel, Mycologia 22:146. 1930.

Sori in the inflorescence, 3-8 mm. long, broad at the base, covered by a delicate false

membrane which flakes away revealing a brown granular spore mass surrounding a well

developed columella ; spore balls broadly ellipsoidal, opaque, dark reddish-brown, many

spored, temporary, usually 38-66 w, rarely 85 u diam.; spores globose -subglobose, light

olivaceous- brown with a thick yellowish wall, 5-7 p diam., smooth, contents granular to

vacuolated.

Type host and locality: On Cymbopogon plurinodis Stapf, Pretoria, Union of South

Africa.

On Andropogoneae: Cymbopogon plurinodis Stapf, Transvaal (M.H. 10045).

Distribution: Reported only from type locality.

Sorosporium Holstii P. Henn. Pflanzenw. Ost-Afrikas Nachb. C:49. 1895.

Sori in the inflorescence, long bnear, 7 mm. long, covered by an evident false membrane

which flakes away revealing a brown granular spore mass surrounding a well developed

columella; spore balls subglobose—broadly ellipsoidal, opaque, many spored, 50-114 u

diam.; spores globose-subglobose, light reddish-brown, 5-8 » diam., smooth.

Type host and locality: On Themeda triandra Forsk. (=Themeda Forskalii Hack.),

Tanganyika Territory (German Hast Africa).

On Andropogoneae : Cymbopogon elegans Spreng., Nyasaland Protectorate ; Themeda

triandra Forsk. (=Themeda Forskali Hack.), Nyasaland Protectorate, Transvaal, Tangan-

yika Territory, Transvaal (M.H. 9315).

Distribution: Kastern and Southern Africa.

Sorosporium consanguineum Ellis and Ky. Jour. Myc. 3:56. 1887.

Ustilago Aristidae Peck, Bull. Torrey Club 12:35. 1885. Not Sorosporium Aristidae Neger-

Sorosporium Bornmuller: P. Magn., Ver. Zool.-Bot. Gesell. (Wien), 50: 434. 1900.

Sori in the ovaries, almost entirely concealed by the glumes though often somewhat

visible through them; spore balls subglobose—broadly ellipsoidal, often irregular, at first

firm but with age and in old specimens separating and becoming entirely broken down,

usually 60-130 diam.; spores ovoid-subglobose but chiefly polyhedral, reddish-brown,

mostly 6-8 w diam., smooth.

Type host and locality: On ~ Aristida Rusby.” (A. arizonica Vasey).

On Agrostideae: Aristida junciformis Trin. et Rupr., Natal (M.H. 1592, 9763).

Distribution : Central and South-Western United States, Australia, South Africa.

Sorosporium Cenchri (Bref.) Zundel n.n.

Tolyposporium Cenchri Bref., Unters. Gesammt. Myk. 12: 156. 1895.

Sori in the ovaries, destroying and fillmg them with spores, concealed by the glumes,

covered by a membrane which breaks away revealing a granular spore mass; spore balls

subglobose-ellipsoidal frequently irregular, opaque, permanent, relatively few spored,

small, 25-35 w long, occasionally 60 ~; spores globose-subglobose, frequently angled due

to compression within the spore ball, olivaceous-brown, usually 7-8 diam., occasionally

10 pw, smooth.

304

Type host and locality : On Cenchrus echinatus Torrey, Rio de Janeiro, Brazil.

On Paniceae: Cenchrus ciliaris L., Transvaal (M.H. 8893).

Distribution: Brazil and South Africa.

Sorosporium inconspicuum (Pole-Evans) Zundel n.n.

Ustilago inconspicua Pole-Evans in herb.

Sori destroying the ovaries and filling them with a brown, granular mass of spores,

about 2 mm. long ; spore balls dark-brown, opaque, many spored, semi-permanent, generally

ellipsoidal or irregular, 100-120 yw long; spores globose-subglobose, often angular, light

olivaceous- brown, 7-9 yx diam., smooth.

Type host and locality: On Digitaria monodactyla Stapt.

On Paniceae: Digitaria monodactyla Stapf, Transvaal (M.H. 9416, 10716).

Distribution: Reported only from type locality.

This species is very closely related to Sorosporium setariae McAlpine, but differs in

having smaller spores.

Sorosporium Everhartii Ellis and Gall. Jour. Myc. 6:32. 1890.

Uredo Syntherismae Rav. (not Schw.) Rav. Fung. Car. II. 90. 1853.

Ustilago Cesati Fisch. v. Waldh. Apergu 25. 1877. p.p.

Tolyposporvum Everhart Diet. Nat. Pflanz. 11:14. 1897.

Sori in the ovules of the sp'kelets, long linear, 1-2 em. long, } cm. wide, at first concealed

by the glumes, covered with an evident false membrane which dehisces from the apex

rev ealing a granular dark brown spore mass; spore balls globose—ellipsoidal, opaque, dark

reddish- brown, rather permanent, many spored, usually 40-125 p diam. or occasionally

140 ; spores globose-subglobose, somewhat irregular and angled, reddish-brown (spores

on inner part of spore ball lighter coloured, sometimes almost hyaline), 7-12 p» diam., free

surface of outer spores verruculose otherwise smooth.

Type host and locality: On Andropogon virgincus Linn., Newfield, New Jersey,

Ur ited States.

On Andropogoneae : Hyparrhenia Ruprecht Fourn., Transvaal (M.H. 7770).

Distribution: Eastern United States, Congo, South Africa.

Sorosporium! verecundum (Sydow) Zundel n.n.

Ustilago verecunda Sydow, Ann. Mye. 88: 231. 1935.

Sori entirely destroying the ovaries, about 2 mm. long, almost entirely concealed by

the glumes, covered by a lemon- yellow false membrane which flakes away revealing a dark

brown agglutinated spore mass; spore balls subglobose—ellipsoidal, many spored, opaque,

disintegrating at maturity, 42-66 w long, occasionally 102 w long; spores globose—sub-

globose, somewhat angled by compression, olivaceous-brown ‘with a narrow reddish-brown

epispore, 7-12 diam., rarely 14 w smooth.

Type host and locality : On Urochloa helopus Stapf, Wonderboom, Pretoria, Transvaal.

On Paniceae: Uvrochloa helopus Stapf, Transvaal (M.H. 26609).

Distribution : Reported only from type locality.

Sorosporium Zundelianum Ciferri, Nuovo Giron. Bot. Ital. n.s. 40: 268. 1933.

Ustilago tumefaciens P. Henn. Pflanzenw. Ost-Afrikas C. 5:48. 189

Sorosporium tumefaciens” Zundel, Mycologia 22: 149. 1930. p.p.

Sori destroying the inflorescence, concealed by the glumes, 1-2 em. long, long linear

somewhat tubular, covered by an evident false membrane which flakes away disclosing

' In order to find spore balls it is necessary to secure material from the base of the sorus. The

spore balls at the tip of the sorus are usually entirely disintegrated.

* Not Sorosporium tumefaciens McAlpine, Smuts of Austr. 184, 1910, (on Stipa sp. and Stipa

pubescens RK. Br. in Queensland, Australia).

305

a dark brown granular spore mass; spore balls globose-—ellipsoidal, irregular, opaque,

dark brown, many spored, semi-permanent (may be almost entirely disintegrated in old

mature specimens), 35-95 long ; spores globose—ellipsoidal, irregular. often angular, 9-12 yu

diam., outer spores echinulate to verruculate on free surface ,inner spores mostly smooth.

Type host and locality: On Hyparrhenia rufa Stapf (=Andropogon rufus Kunth),

Kalimandscharo, Rombo, Tanganyika Territory (formerly German East Africa).

On Andropogoneae: Hyparrhenia rufa Stapf (=Andropogon rufus Kunth), Tangan-

yika Territory ; Hyparrhenia Tamba Anders., Natal (M.H. 14167).

Distribution : Southern and Eastern Africa.

Sorosporium Tembuti P. Henn. and Pole-Evans, Bot. Jahrb. (Engler) 41: 270. 1908.

Sori destroying the inflorescence, 1-3 cm. long or occasionally slightly longer, covered

by a false membrane which flakes away revealing a dark brown granular spore mass sur-

rounding a well developed columella ; spore balls ‘globose- oblong, opaque, with 60 or more

spores, dark brown, 40-90 ~ diam. ; spores globose-subglobose, sometimes angled, medium

to light reddish-brown (spores on inner part of spore ball lighter in colour), 9-11 fw diam.

rarely 14 “, outer spores echinulate on free surface, inner spores smooth.

Type host and locality : On Hyparrhenia Tamba Anders., Waterval Onder, Transvaal.

(M.H. 169).

On Andropogoneae : Hyparrhenia Tamba Anderss. Transvaal (M.H. 169, 1794, 1849).

Distribution: South Africa.

Sorosporium setariae McAlpine, Smuts of Australia 183. 1910.

Sori in the ovaries filling them with a brown mass of spores; spore balls dark-brown,

opaque, many spored, variously shaped, globose-subglobose to ellipsoidal, often angled,

usually 85-125 w long, occasionally 160 Uu long, semi-permanent ; spores globose-subglobose

often angular, light “olivaceous- brown, 10-12 ~ diam., smooth.

Type host and lopli: On Setaria glauca Beauv., near Cloncurry, Queensland,

Australia.

On Paniceae: Setaria perennis Hack., Transvaal (M.H. 17269).

Distribution: Australia, South Africa.

This species differs from Sorosporiuwm inconspicua (Pole-Kvans) Zundel principally

by the larger spores.

** Spores not smooth.

Sorosporium austro-africanum Zundel, Mycologia 22: 147. 1930.

Sori in the inflorescence, long linear, 5-8 mm. long, solitary, covered by an evident

yellowish false membrane which dehisces at the apex disclosing a granular spore mass sur-

rounding a well developed columella ; spore balls semi-opaque, broadly ellipsoidal, usually

142-190, win length but occasionally as small as 47 7, semi-permanent, many spored, reddish-

brown: spores g globose-subglobose, irregular, often somewhat angular, light reddish-brown

to almost hyaline, thick walled, 6-10 w diam., usually smooth except spores on the outer

portion of spore ball which are verruculose.

Type host and locality: On Hyparrhenia tampa Anders., Tugela River, Natal, Union

of South Africa.

On Andropogoneae: Hyparrhenia tamba Anders., Natal (M.H. 14168).

Distribution: Reported only from type locality.

Sorosporium Healdii Zundel, Mycologia 22: 147. 1930.

Sori in the inflorescence, concealed by the glumes, attacking the individual flowers

and en masse producing a witches’ broom-like growth, 2-3 cm. long, ‘covered with a yellowish

brown false membrane which dehisces from the apex revealing numerous shreds and a

dark brown granular spore mass surrounding a well formed columella ; spore balls globose-

306

broadly ellipsoidal, somewhat irregular, opaque, dark brown, permanent, 30 or more spores,

usually 40-70 uw long but occasionally 90 uw long; spores globose-subglobose or broadly

ellipsoidal, reddish-brown for the outer spores to almost hyaline for the spores on the inner

part of the spore ball, thick walled, 6-10 w diam., sparingly verruculose under oil immersion.

Type host and locality: On Hyparrhenia sp., Pretoria, South Africa.

On Andropogoneae: Hyparrhenia sp., Transvaal (M.H. 9732).

Distribution: Transvaal.

Sorosporium afrum Sydow, Ann. Myc. 33: 232. 1935.

Sori entirely destroying the panicles, covered by a brown membrane which ruptures

revealing numerous black shreds and spore balls intermixed, 2-5 c.m long, spore mass dark

brown; spore balls globose-subglobose, dense, many spored, semi-permanent; spores

globose-subglobose, somewhat angled, olivaceous-brown, 7-12 ~ diam., apparently smooth

but under oil immersion lens showing minute echinulations.

Type host and locality: On Panicum laevifolium Hack., Transvaal (M.H. 6579).

On Paniceae: Panicum laevifolium Hack., Natal (M.H. 11706), Transvaal (M.H. 608,

8932, 8929, 20335).

Distribution: Union of South Africa.

Sorosporium Hotsonii® Zundel, Mycologia 22: 152. 1930.

Sori in the inflorescence, solitary, 3-5 em. long, at first hidden by the outer sheath,

covered by an evident false membrane which flakes away revealing a brownish granular

spore mass intermixed among the shreds: spore balls globose-subglobose, semi-opaque,

semi-permanent, reddish-brown, many spored, variable in size, 50-1157 long ; spores globose-

subglobose, often somewhat angular, thick walled, light reddish-brown, 8-10 yw diam., under

oil immersion abundantly echinulate with vacuolated contents.

Type host and locality: On Andropogon sp., Hopefield, Lawley, Transvaal, Union

of South Africa.

On Andropogoneae: Andropogon sp., Transvaal (M.H. 704).

Distribution: Reported only from type locality.

Sorosporium cryptum McAlpine, Smuts of Australia 176. 1910.

Ustilago cryptum McAlpine, Proc. Linn. Soc. New South Wales 22:42. 1897.

Sori in the ovaries of spikelets, hidden by the glumes, about 3 mm. long, covered by

a thick membrane of host tissue which ruptures revealing a black spore mass surrounding

a columella of host tissue: spore balls evanescent, semi-opaque, many spored, variable

in size and shape, spheroidal to ellipsoidal, 50-80 1 diam. or larger ; spores globose—sub-

globose or sometimes ellipsoidal, regular, dark reddish-brown, 8-10 « diam., apparently

smooth but minutely echinulate under oil immersion.

Type host and locality: On Panicum bicolor R. Br., Braidwood District, New South

Wales, Australia.

On Paniceae: Hehinochloa sp., Transvaal (M.H. 18186).

Distribution: Australia, Union of South Africa.

Sorosporium Clintonii Zundel, Mycologia 22:153. 1930.

Sori in the inflorescence, large, developing in clusters as a “ witches broom,” large,

2-6 cm. long, and often 5 mm. wide, at first concealed by the glumes, covered with a dark

brown false membrane which dehisces apically revealing a granular spore mass intermixed

with shreds ; spore balls globose—oblong, irregular, often angled as so to appear rectangular,

opaque, permanent, many spored, dark reddish-brown, ranging from 47-133 jw long, but

8 In the original description the author misspelled the species name. Since the specific name is

for Dr. J. W. Hotson of the University of Washington, Seattle, the name should not be S. Hodsonii.

307

mostly 66-114 ~ long: spores globose-subglobose, irregular, often angled, thick walled,

about 1-5 4, dark reddish-brown (spores on inner part of spore ball lighter colour, mostly

tinted brown), 8-17 y diam., verruculose on free surface.

Type host and locality: On Hyparrhenia Tamba Anders., Waterkloof, Pretoria, Trans-

vaal, Union of South Africa.

On Andropogoneae: Hyparrhenia Tamba Anders., Transvaal (M.H. 9693).

Distribution : Reported only from type locality.

Sorosporium panici! MacKinnon, Jour. and Proc. Roy. Soc. N.S. Wales 46: 210. 1912.

Sori in the ovaries, at first concealed by the glumes but later protruding, 3 mm. long,

covered by a delicate membrane which flakes away revealing a granular spore mass sur-

rounding a well formed columella; spore balls globose— ellipsoidal, somewhat irregular,

60-105 «w long, many spored, semi- permanent, in old or mature specimens almost entirely

disintegrating ; spores globose-subglobose, irregular, somewhat angular, olivaceous-brown,

8-5-11 uw diam., densely echinulate.

Type host and locality: On Panicum flavidum Retz., Nyngan Experimental Farm,

New South Wales, Australia.

On Paniceae: Panicum maximum Jacq., Rhodesia (M.H. 14000), Transvaal (M.H. 20).

Distribution: Australia, South Africa.

Sorosporium proliferatum Zundel, Mycologia 22: 150. 1930.

Sori as large proliferations in the inflorescence resembling miniature ears of corn (maize),

2-8 cm. long, concealed by large outer glumes, covered by an evident false membrane

which flakes away revealing a dark brown, granular spore mass intermixed with shreds ;

spore balls globose— ellipsoidal or sometimes angular, opaque, many spored, permanent,

usually 45-60 w long, occasionally 85 yw long.

The spores in the outer part of the spore ball dense, dark reddish-brown, while the

inner spores are nearly hyaline, somewhat irregular in size and shape, ranging from globose—

subglobose, occasionally angled, most commonly 9-12 uw diam., abundantly verruculose

under oi] immersion.

Type host and locality: On Hyparrhenia aucta (Stapf) Stent, Waterval Boven, Union

of South Africa.

On Andropogoneae: Hyparrhenia aucta (Stapf) Stent, Transvaal (M.H. 11336).

Distribution: Reported only from type locality.

Sorosporium Reilianum (Kuhn) McAlpine, Smuts of Australia 181. 1910.

Ustilago reiliana Kuhn, Rab. Fungi Eur. 1998. 1875.

Ustilago Reiliana Zeae Pass., Rab. Fungi Kur. 2096. 1875.

Ustilago pulveracea Cooke, Grevillea 4: 115. 1876.

Cintractia Reihana G. P. Clinton, Bull. Ill. Agr. Exp. Sta. 57: 346. 1900.

Ustilago (Cintractia) Reiliana foliicola Kell., Ohio State Univ. Natural. 1:9. 1900.

Sphacelotheca Reiliana G. P. Clinton, Jour. Myc. 8: 141. 1902.

Sori occurring in either g or ? inflorescence, usually causing complete destruction,

covered with an evident membrane of host tissue which ruptures disclosing a brown spore

mass and numerous columellae, the sori are frequently covered by proliferations of the tassel

or ear; spore balls irregular in shape, generally opaque, dark reddish-brown, easily dis-

1Sorosporium Beelii Zundel nov. comb.

Sorosporium panici Beeli, Bull. Jard. Bot. Brux. 8:7. 1923.

Type host and locality: On Panicum sp., Bomba, Congo, Africa.

The name Sorosporium panici MacKinnon (1912) has preference over Sorosporium panici Beeli

(1923). These species differ in size of spore balls and spores. S. panici MacKinnon has spore balls

60-100 » with echinulate spores 8-5-11 p diameter. SS. panici Beeli has spore balls 15(-180 x 100 uw

with smooth spores 6-8 » diameter. The name Sorosporium Beelii nov. comb. is therefore proposed

in place of Sorosporium panici Beeli.

308

integrating at full maturity of spores ; spore balls found only in young specimens, 76-150

diam.; spores globose-subglobose, occasionally somewhat angled, thick walled, reddish-

brown, 9-14 w diam., abundantly echinulate under the oil immersion.

Type host and locality : On Sorghum vulgare Pers., Cairo, Egypt.

On Andropogoneae: Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehd.

(=Sorghum caffrorum Beauv.), Tanganyika Territory (M.H. 20645); Sorghum halepense

Pers.*, Natal; Sorghum sp., Tanganyika Territory (M.H. 20651); Zea Mays Linn., Cape

Browimoc, Natal*, Orange Free State (M.H. 10064), Transvaal (M.H. 11, 505, 1480, 2142,

6586).

Sorosporium Simii! P.Henn. and Pole-Evans, So. Afr. Jour. Sci. 12: 543. 1916.

Sori destroying the inflorescence, long linear, 5-7 cm. long, 1-3 cm. wide, covered with

a thick dark brown false membrane which flakes away disclosing a granular spore mass

intermixed with numerous long shreds; spore balls globose-subglobose, not permanent,

opaque, many spored ; sterile tissue rather permanent but breaking up chiefly into groups

or sometimes chains of sterile cells, rarely singly, tinted brown or dark brown ; sterile

cells about the size of the spores; distinctive globose groups of sterile cells consisting of

4-6 cells are scattered through the sori, 19-36 diam. ; spores globose-subglobose, olivaceous

to reddish-brown, 9-13 diam., under oil immersion, finely echinulate with granular con-

tents.

Type host and locality: On Sorghum halepense Pers., Natal, Union of South Africa.

On Andropogoneae : Sorghum halepense Pers., Natal (MH. 8978, 10031) ; % Sorghum

sp., Transvaal (M.H. 11324).

Distribution: Union of South Africa.

Sorosporium filiferum (W. Busse) Zundel.

Tolyposporium filiferum W. Busse, Arb. Biol. Abt. Landw. Forstw. Kaiserl. Gesundheit

4:383. 1904.

Sori destroying the ovaries, cylindrical elongate, 1-3 cm. long and 5-10 mm. wide,

often curved at the end, covered by a thick membrane which ruptures apically revealing

long dark brown shreds and a granular spore mass ; spore balls subglobose—-oblong, opaque,

rather permanent, many spored, dark brown, 55-115 wu long; spores globose-subglobose,

inner spores light yellowish brown, outer spores dark brown, 9-14 w diam., inner spores

smooth, outer spores papillate on free surface.

Type host and locality : On Sorghum cult., Kenya Colony.

On Andropogoneae: Sorghum vulgare Pers. v. caffrorwm (Thun.) Hubb. et Rehder.,

Union of South Africa.

Distribution: Africa.

Sorosporium versatilis (Sydow) Zundel n.n.

Ustilago versatilis Sydow, Ann. Myc. 88: 231. 1935.

Sori entirely destroying the inflorescence, oblong, 2-3 cm. long, at first covered by a

brownish membrane which dehisces apically ovis: a dark brown agglutinated spore

mass and elator-like shreds; spore lonlls permanent, many spored, opaque, usually ovoid

but sometimes irregular, usually 65-100 ju. long; spores globose-subglobose, somewhat

angled due to compression, 10— 13 lu ae dark Teddish- brown, under oil immersion abun-

dantly but minutely verruculose.

Type host and locality: On Panicum longijubatum Stapf. (=Panicum proliferum

Lam. var. paludoswm Stapt.).

On Paniceae: Panicum longijubatum Stapf., Cape Province (M.H. 9550.).

Distribution : Reported only from type locality.

1 Sorosporium Seine is probably related to but very distinct from Skxgoajnorionn reilianum hy the

possession of large groups or chains of sterile cells throughout the sori. These groups of sterile cells

are very distinctive and usually consist of from four to six cells. The sterile cells found in Sorosporrwm

reilianum are from the disintegration of the false tissue that surrounds the sori, and are not scattered

through the sorus.

309

Sorosporium Maranguenense P. Henn., Pflanzenw. Ost-Afrikas Nachb. C: 49. 1895.

Sori in the inflorescence, at first covered by the leaf sheaths but later protruding,

3-6 cm. long, covered by an evident membrane which flakes away revealing a granular

spore mass intermixed with shreds; spore balls subglobose, angular, many spored, semi-

permanent, 35-65 long ; spores subglobose, angular, irregular, thick walled, light reddish-

brown (almost a yellow), inner spores lighter coloured, the thick wall dark reddish-brown,

10-14 w diam., verruculose on free SUAS.

Type host and locality: On Hyparrhenia Tamba Anderss. (=Andropogon lepidus

Nees), Tanganyika Territory (German East Africa).

On Andropogoneae : Hyparrhenia Tamba Anderss., Tanganyika Territory.

Distribution: Reported only from type locality.

Sorosporium pseudomaranguense Zundel n.sp.

Sori in the inflorescence, 3-5 cm. long, at first concealed by the sheath, later the tips

protrude, covered by an evident membrane which flakes away revealing a dark brown spore

mass ; spore balls ovoid-ellipsoidal, dense, many spored, semi-permanent, 35-85 ~ diam. ;

spores subglobose, angular, irregular, olivaceous-brown with a thick reddish-brown epispore,

10-14 w diam., outer spores densely verruculose on the free surface, inner spores smooth.

Hab. in the inflorescence of Andropogon sp., Mooi River, Natal, Union of South Africa,

Coll. A. O. D. Mogg, March 21, 1917. (M.H. 10073.) Host det. by Agnes Chase,

Smithsonian Institute, Washington, D.C.

This species is closely related to Sorosporitum maranguensis P. Henn.

Latin description :—

Soris in inflorescentia, 3-5 cm. longis, primum spatha tectis, deinde apicibus protru-

dentibus, membrana conspicua tectis, membrana decadenti et atro-brunneum sporarum

globum revelante ; glomerulis sporarum ovoideis v. ellipsoideis, densis, multisporis, semi-

permanentibus, 35-85 w diam.; sporis subglobosis, angularibus, irregularibus, olivaceo-

brunneis, 10-14 wv diam. : episporio denso, irregulari, rubus-brunneo ; sporis externis in

superficie dense verruculosis, sporis internis levibus.

Hab. in inflorescentae Andropogonis sp., Mooi River, Natal, in Unione Africae australis.

Hospes ab Agnes Chase det.

Sorosporium Flanaganianum Zundel, Mycologia 22:155. 1930.

Sori in the inflorescence, broad, long linear, 2-4 cm. long, solitary, covered by a thick

brown false membrane which flakes away revealing a brown granular spore mass inter-

mixed with fine shreds: spore balls globose-subglobose, semi-permanent, opaque, dark

reddish-brown, usually 75-95 u long, rarely as small as 47 [, Spores subglobose, irregular,

angular, reddish- brown, 10-14 uw diam., echinulate under oil immersion.

Type host and locality : On ? Andropogon sp., Emmasdale, Heidelberg, Transvaal,

Union of South Africa.

On Andropogoneae: ¢ Andropogon sp., Cape Province (M.H. 9423), Transvaal (M.H.

713).

Distribution : South Africa.

Sorosporium harrismithense Zundel, Mycologia 22: 154. 1930.

Sori in the inflorescence, 3-4 cm. long, 5-6 mm. wide, solitary, covered by a brown

false membrane which dehisces apically revealing a granular spore mass intermixed with

numerous shreds; spore balls globose-subglobose, opaque, semi-permanent, dark reddish-

brown, 47-105 « long; spores globose—broadly ellipsoidal, angular, thick walled, reddish-

brown, 10-14 w diam., echinulate under oil immersion.

ae: Type host and locality: On Panicum laevifolium Hack., Harrismith, Union of South

rica.

On Paniceae: Panicum laevifolium Hack., Orange Free State (M.H. 1473).

Distribution: South Africa.

310

Tolyposporium Woronin, Abh. Senck. Nat. Ges. 12:577. 1882.

Sori usually in the inflorescence, more especially in the ovaries, forming a granular

spore mass at maturity; spore balls dark coloured, composed of numerous spores per-

manently united, of medium size; spores bound together by ridged folds or thickenings

of their outer walls, of small to medium size ; germination about as in Ustilago.

(Upon rupture, by pressure, of the spore balls the thickenings or ridges often show as

reticulations or as spine-like processes at the margins of the lighter coloured spores. There

is a tendency to put species of Sorosporiwm with rather permanent spore balls into this

genus.)

Type: Sorosporium Junci Schrot., on Juncus bufomius Linn., Germany.

Tolyposporium tristachydis (Sydow) Zundel n.n.

Sorosporium tristachydis Sydow, H. & P. Bot. Jahrb. (Engler) 45 : 263. 1910.

Sori in the ovaries, hidden by the outer glumes, at first covered by a delicate membrane

which ruptures revealing a granular spore mass ; spore balls permanent, held firmly together

by outer folds in the spores, many spored, globose to ellipsoidal, dark reddish-brown, usually

50-80 « diam., occasionally 109 j4; spores globose-subglobose or ellipsoidal, somewhat

angular, reddish-brown (spores on the inner part of the spore balls lighter coloured), 10-15

diam., smooth.

Type host and locality: On Tristachya sp., Leimde, Cameroon, Africa.

On Tristachya Rehmanni Hack., Transvaal (M.H. 9436).

Distribution: Africa.

A portion of the type specimens from the Clinton herbarium has been used for this

description.

Family [1.—TILLETIACEAE.

Tilletia Tulasne Ann. Sci. Nat. Bot. III. 7: 112-113. 1847.

Sori in various parts of the host, usually in the ovaries but occasionally on the leaves,

forming a dusty spore mass ; spores simple, usually formed singly in the ends of the mycelial

threads that disappear more or less completely through gelatinization, of medium to large

size; germination usually by a short non-septate promycelium which bears a terminal

cluster of elongated sporidia that usually fuse in pairs which may, in nutrient solution,

five rise to a bonciderabie mycelium bearing secondary alr conidia.

Spores are formed from hyphae, which “swell up in a gelatinous manner.

Type Uredo Caries DC. on Triticum vulgare Linn., Europe.

Tilletia foetans (B. & C.) Trel. Wisc. Acad. Sci. Trans. 6: 139. 1886.

Ustilago foetans Berk. & Curt. Rav. Fungi Carol. 100. 1860.

Ustilago foetans Berk. & Curt. Hedwigia 8:59. 1874.

Tilletia laewis Kuhn, Rabenh. Fungi Bur. 1697. 1873.

Sori in the ovaries, foetid, ovate-oblong, 5-7 mm. diam., protruding between the pro-

tecting glumes, when ruptured revealing a brownish spore mass ; spores globose-subglobose

or elliptical, often somewhat angled, light to dark olivaceous-brown, 16-21 diam., smooth.

Type host and locality : On Triticum vulgare Vill., North Carolina, United States.

On Hordeae: Triticum vulgare Vill., Cape Province, Transvaal (M.H. 1909).

Distribution: Co-extensive with cultivated wheat.

Tilletia heterospora (P. Henn.) Zundel n.n.

Ustilago heterospora P.Henn. Pflanzenw. Ost-Afrikas Nachb. C. 5:48. 1895.

311

Tilletia Ayresii Berk.1, Massee in Bull. Misc. Inf. Kew 153: 146. 1889. (Type on Panicum

maximum Nees, hill above Port Lewis, Mauritius. Ayes No. 4754. Type in Kew

Herbarium.)

Sori in the ovaries, ovoid, inflated, 3-5 mm. diam., covered by an olivaceous, leathery

membrane which ruptures revealing a dark olivaceous, semi-agglutinated spore mass ;

sterile spores globose-subglobose, hyaline, spiny, of two general sizes, 10-12 w and 19-21 yw

diam.; bifurcate conidiophores abundant; spores globose-subglobose, regular, light

olivaceous-brown, usually 13-16 yw diam., coarsely echinulate to spiny.

Type host and locality : On Panicum maximum Nees, Tanganyika Territory.

On Paniceae: Panicum laevifolium Hack., Transvaal (M.H. 7); Panicum maximum

Jacq., Mauritius, Natal (M.H. 15443, 17081), Portuguese East Africa (M.H. 8399), Rhodesia,

Tanganyika Territory ; Panicum sp., Transvaal (M.H. 11717).

Distribution: Africa, Islands of Indian Ocean.

Tilletia Tritici (Bjerk.) Wint. Rab. Krypt.-Fl. 11:110. 1881.

Lycoperdon Tritici Bjerk. Kgl. Schmed. Akad. Wiss. Abhandl. 37: 326. 1775.

Uredo caries DC. Fl. Fr. 6:78. 1815.

Caeoma segetum Nees, Syst. Pilze 1:14. 1817.

Uredo sitophila Ditm. Sturm’s Deuts. Fl. II]. 1:69. 1817.

Uredo foetida Bauer, Ann. Sci. Nat. Bot. I. 2: 167. 1884.

Caeoma sitophilum Link, Willd. Sp. Pl. 62:2. 1825.

Erysibe foetida Wallr. Fl. Crypt. Germ. 2: 213. 1833.

Tilletia caries Tul. Ann. Sci. Nat. Bot. III. 7: 113. 1847.

Ustilago sitophila Bon. Kennt. Con. Crypt. 27. 1860.

Sori in the ovaries, showing between the glumes, 5-6 mm. long, upon rupturing dis-

closing a reddish-brown spore mass; sterile cells few, subglobose, hyaline, thin walled,

14 uw diam. ; spores globose-subglobose, regular, light to dark olivaceous brown, 16-20

diam. or slightly larger occasionally, winged reticulations about 1 high and 2-3 w broad.

Type host and locality: On Triticum vulgare Vill., Sweden.

On Hordeae: Triticwm vulgare Vill., Cape Province, general through region.

Distribution: Co-extensive with cultivated wheat.

Tilletia Viennotii Syd. Ann Myc. 35; 2589 25, 1937.

Sori filling the ovaries with a dark coloured spore mass, diseased ovaries larger than

normal ones, hidden by the glumes; spores globose-subglobose, regular, with irregular

polygonal reticulations, dark reddish-brown, 18-25 w diam., reticulations projecting on the

margin of the spores, 3-3-5 wu.

Type host and locality : Briza maxima Linn., Madeira Islands.

On Festuceae: Briza maxima Linn., Capetown, (M.H. 14679).

Distribution : Madeira Islands, South Africa.

Tilletia transvaalensis Zundel, Mycologia 28: 299. 1931.

Sori in the ovaries, about 1 mm. long, at first concealed by the glumes but later the

tip protrudes slightly, infected spikelets scattered throughout the panicle; hyaline sterile

cells smaller than the spores; spores globose-subglobose, regular, yellowish to reddish-

brown, 20-26 w diam., abundantly echinulate under oil immersion.

Type host and locality: On Eragrostis aspera Nees, Mucklenburg, Zebediela, Trans-

vaal, Union of South Africa.

On Festuceae: Hragrostis aspera Nees, Transvaal (M.H. 25463).

Distribution: Reported only from type locality.

1 Miss E, M. Wakefield first called attention to the fact that Ustilago heterospora P. Henn. and

‘illetia Ayresii Berk. were identical in ‘* Notes on Uganda Fungi,” Bull. Misc. Inf. Kew 9 : 290, 1920.

Rerkeley, however, was correct in considering the fungus a Villetia, in Massee, George “‘ A Revision

the Genus Tilletia ’’ Bull. Misc. Inf. Kew 153 : 146, 1899. The Royal Botanic Gardens, Kew, kindly

supplied type material of Tilletia Ayresii Berk. for examination.

312

Tubureinia (Fries)! Woronin, emend. Abh. Sensk. Nat. Ges. 12:561. 1882. (Fries, Syst.

Myc. 3:439. 1829.) ;

Sori usually in the leaves or stems, forming dark coloured often papillate areas, rather

permanently embedded in the tissues ; spore balls composed entirely of firmly united fertile

cells ; of medium size ; spores usually dark coloured, variable, of medium size ; sometimes

preceded by conidia”, forming a conspicuous white growth on the surface of leaves, hyaline,

oblong to ovate.

Type : Tuburcinia Trientalis B. & Br., on Trientalis europaea Linn., Kurope.

To date no species of this genus have been reported from South Africa. They are

usually found in cold climates.

Uroeystis Rabenhorst, Klotsch, Herb. Viv. Mye. ed. 2. 393. 1856.

Polycystis Lev. Ann. Sei. Nat. III. 5: 269. May 1846, not Polycystis Kutz, Jan. 1846.

Sori usually in the leaves or stems, occasionally in other parts, producing dark coloured

usually dusty spore masses; spore balls permanent, composed of an enveloping cortex

of tinted sterile cells and from one to several interior fertile cells, of small to medium size :

spores usually dark coloured, variable, of medium size; the balls of spores are developed

inside coils of hyphae, which become entwined together and swell up in a gelatinous manner ;

the central spores on germination give rise to a promycelium, with terminal sporidia which

do not as a rule fuse in pairs, but grow out directly into mycelia.

Type: EHrysibe occulta Wallr., on Secale cereale Linn., Europe.

Urocystis Tritici Koern. Hedwigia 16:33. 1877.

Tuburcimia triticc Liro, Ann, Univ. Fenn. Aboensis, Ser. A. 1:17. 1922.

Sori in the leaves, culms and leaf sheaths as long striae, covered by a leaden coloured

membrane which when ruptured reveals a dark brown spore mass ; spores variable in shape,

globose-ellipsoidal, often irregular, usually with 1—2 fertile spores, sometimes 3-4, rarely

5, dark reddish-brown surrounded by lighter coloured sterile cells, chiefly 24-32 u diam. ;

spores globose-ellipsoidal, dark reddish-brown, 10-17 « diam., smooth ; sterile cells usually

completely enveloping fertile spores, globose— ellipsoidal, pale yellow, 7-12 4 diam., smooth.

Type host and locality: On Triticum vulgare Vill., New Holland, Australia.

On Hordeae: Triticwm dicoceum Schrank.* : Cape Province ; Triticum durum Dest.*,

Cape Province; Triticum turgidum Linn.*, Cape Province; Triticum vulgare Vill., Cape

Province, Transvaal (M.H. 12454, 13050, 13049).

Distribution: Australia, South Africa, Central United States, Japan, India, Hgypt,

China.

Urocystis Ornithoglossi (Sydow) Zundel n.n.

Tuburcima Ornithoglosst Sydow, Ann. Myc. 38: 233. 1935.

Sori in the leaves as inconspicuous lead coloured pustules about 2 mm. long, sometimes

becoming confluent and then longer, covered by the epidermis, spore mass dark brown,

granular; spore balls usually with one (rarely two or three) spore entirely surrounded

by numerous outer sterile cells, usually 17-24, diam., rarely 31 4 ; outer sterile cells globose—

subglobose, light olivaceous-brown, about i iG diam. : spores globose-subglobose, dark

reddish-brown, 10-14 y« diam.

Type host and locality : On Ornithoglossum ica Salisb.

On Liliaceae: Ornithoglossum glaucum Salisb., Transvaal (M.H. 1888).

Distribution: Reported only from type locality.

1 Fries in 1829 was the first to use the term T’uburcinia as a genus name. In 1882, Woronin

emended the genus so that it did not include any of Fries’ original species. In 1922, Liro (Uber de

Gattung Tuburcinia Fries, Ann. Uvin. Fenn. Aboensis, A. 1: 1-153, 1922) combined the two genera

Urocystis Rabenh. and Tuburcinia Woronin under the one genus Tuburcinia Fries. In so doing he

includes species that are cytologically and morphologically dissimilar and only adds confusion to the

taxonomy of the Ustilaginales.

3 Tuburcinia Trientalis B. & Br. is the only species known to produce conidia,

313

Entyloma De Bary, Bot. Zeit. 82: 101. 1874.

Rhamphospora D. D. Cunningham, Sci. Mem. Med. Off. Army India 3:32. 1888.

Sori usually in the leaves, generally forming discoloured but little distorted areas,

spores simple, produced terminally or intercalary on any part of the fertile mycelium which

is intercellular and never entirely disappears through gelatinization, free (sometimes adhering

irregularly through pressure), hyaline to yellowish or reddish-yellow, occasionally dark

coloured, of medium size; germination by a short promycelium bearing a terminal group

of sporidia which usually conjugate in pairs and produce secondary sporidia or infection

hyphae; conidia often present, hyaline usually elongate formed by germination of the

spores in situ; or on the mycelium produced through the stomata.

Type: Protomyces microsporus Ung., on Ranunculus repens, Germany.

Entyloma Zinniae Sydow, Ann. Myc. 33: 253. 1935.

Sori as orbicular to irregular spots in the leaves, 2-5 mm. diam., showing best on the

upper surface, at first yellowish then brown; spores globose— -subglobose, regular, with

a regular equal epispore about 2 uw, tinged olivaceous-brown, 8-10 diam., occasionally

14 nu.

Type and locality: On Zinma paueiflora Linn.

On Compositae: Zinnia pauciflora Linn., Transvaal (M.H. 14256).

Distribution: Union of South Africa.

Entyloma Bidentis P.Henn. Pflanzenw. Ost-Afrikas C. 5:49. 1895.

Sori in the leaves, forming roundish brown spots visible on both upper and lower sur-

faces, 2-3 cm. diam., often coalescing, at first yellowish then chestnut brown ; spores sub-

globose, irregular, contents sometimes with small greenish granules, olivaceous yellow,

10-14 uw diam., epispore 1-1-5 yw thick, smooth.

Type host and locality: On Bidens pilosa Linn., Tanganyika Territory.

On Compositae: Bidens pilosa Linn., Cape Province (M.H. 8853), Tanganyika Terri-

tory, Transvaal (M.H. 1508).

Distribution: Kastern and Southern Africa.

Entyloma australe Speg. Anal. Soc. Cien. Argent. 10:5. July, 1880.

Protomyces Physalidis Kalchb. Grevillea 9:22. Sept., 1880.

Entyloma Besseyi Farl. Bot. Gaz. 8: 275. 1883.

Entyloma Physalidis Wint. Hedwigia 22: 130. 1883.

Sori in the leaves and often in the inflated calyx, forming at first hght coloured and later

darker coloured, slightly raised (on lower side of leaf) spots, roundish, often angular, 0-5-7

mm. or more in diam. ; spores globose-subglobose, sometimes angled, with a thick darker

coloured epispore, light reddish-brown, 10-15 w diam., smooth; conidia forming whitish

epi- or hypophyllous growth, linear, somewhat ‘curved, usually 30-55 uw x 1-2 uw.

Type host and locality > On Physalis hirsuta Duncan, Argentina.

On Solanaceae: Physalis minima Linn., Transvaal (M.H. 1509, 5638, 11008, 14126) ;

Physalis peruniana Linn., Cape Province (M.H. 10028), Transvaal.

Distribution: North America, South America, South Africa.

Entyloma Dahliae H. & P. Sydow, Ann. Myc. 10:36. 1919.

Sori as orbicular to elliptical spots on the leaves, pale but definite and conspicuous,

1-10 mm. long; spores globose, pale olivaceous-brown, evenly coloured including the

epispore, 9-14 ~ diam., epispore 1-2-5 ; conidia unreported.

Type host and locality: On Dahlia variabilis Desf., Harden Heights, Natal, Union

of South Africa.

On Compositae: Dahlia variatilis Desf., Natal (M.H. 86C, 19761), Transvaal (M.H.

18023).

Distribution: Europe, Central America, South Africa, South Rhodesia.

314

APPENDIX A.

The following species have been reported from South Africa and adjacent territory

but no specimens have been available for examination.

Sphacelotheca Panici-miliacei (Pers.) Bubak, Naturw. Landes. Bohmen 15: 26. 1916.

Uredo segetum panici-miliacer Pers. Synop. Meth. Fung. 224. 1801.

Uredo carbo Panici miliacer DC. FI. Fr. 6:76. 1815.

Uredo destruens Duby, Bot. Gall. 2: 901. 1824.

Erysibe Panvcorum Wallr., Fl. Crypt. Germ. 2: 216. 1833.

Ustilago Carbo destruens Ley. Ann. Sci. Nat. II]. 8: 372. 1848.

Ustilago destruens Schlecht. Rab. Herb. Myc. No. 400.

Ustilago panici-miliacer Wint. Rab. Krypt.-Fl. 11:89. 1881.

Sorosporium Panici-miliacer Tak. Bot. Mag. (Tokyo) 16: 183. 1902.

Ustilago Pamei-nuiliacer (Pers.) Wint. Die Pilze p. 89, Schroet. Pilze Schles. p. 268,

Bref. Untersuch. V, pag. 97, tab. 7, fig. 9-24, Uredo (Ustilago) segetum var. Panici-miliacei

Pers. Syn. Fung. p. 224, Uredo Carbo var. d. DC. Flor. Franc VI, pag. 76, Caeoma destruens

Schlecht. Fl. Berol. II, p. 130, Uredo destruens Duby Bot. Gall. I, p. 216, n. 1670, Ustilago

Carbo var. destruens Tul. Ann. Sci. Nat. III. 7:81. 1847, Tilletia destruens Lev. Ann.

Sci. Nat. IIT. 8: 372. 1848, Ust. destruens Schlecht. in Rabenh. Herb. Myc. noy. n. 400.—

Soris atris, pulverulentis, flores destruentibus ; sporis globosis vel ellipsoideis, 9-12 8-10,

raro angulatis, episporio flavo-brunneo levi vel punctulato instructis ; promycelio filiformi

cylindraceo, plerumque 3-septulato, articulis, anastomosantibus; oblongo-ellipticis vel

ovoideis.*

Hab. in floribus paniculisque Panici-miliacer, P. carthaginiensis et. P. Crus-galli,

Gallia, Germania, Amer. bor. (Ravenel) nec non pr. Montivideo Americae australis (Archa

valeta sec. Winter). (Saccardo Syll. Fung. 7: 454-455. 1888.)

Tolyposporium Anthistiriae Cobb. in Agric. Gaz. New South Wales, 1892, p. 1006,

cum ic, Syll. XIV, p. 426, Tolyposporvum Anthistirae P.Henn. Hedwigia 1898, pag. 283.—

Sori in spices easque destruentibus, cylindraceis, 2-5 em. longis, epidermide dein lacerata

flavescente tectis; glomerulis oblongis vel subglobosis aterrimis, 70-100 w diam. : sporis

subglobosis vy. ellipsoideis, primo hyalinis, dein cinnamomeis v. atris, 9-14** 8-12,

episporio minutae verrucoso-punctulato, 1 ~ crasso.

Hab. in spices Anthistirae spec. in Africa centr. (Schweinfurth) et Australia (Cobb).

— Sec. Cobb loc. cit. massa sporarum 40-150 ~ diam.; cellulae singulae 8-14 yw. longa.

Ergo species homonyma posterior cl. Henningsii eadem videtur. (Saccardo Syll. Fung.

16: 378. 1902.)

Tolyposporium setariicolum H. et P. Sydow Ann. Myc. 10:77. 1912. Soris in spicis

evolutis easque omnino destruentibus, haud tumefacientibus atris; glomerulis rotundatis,

ellipticis v. forma irregulari, 35-70 4 diam., ex sporis numerosissimis compositis, sub-firmis :

sporis angulato- elobosis vy. angulato- -ellipsoideis, brunneis, 7-10 x 6-9, verrucis facile

deciduis obsitis.

Hab. in foliis Setariae aureae (Graminaceae), Sidderiberg Kamerun, Africae (C. Leder-

mann, no. 4803, 30.7.1909) — A Tol. pampeano Speg. quoque in Setaria omnino diversum.

(Saceardo Syll. Fung. 28: 620. 1925.)

Tolyposporium Penicillariae Bref. Unters. XII, p. 154, tab. IX, fig. 35-40. — Soris

in quaque spica solitariis, ovariicolis, bullam magnam globosam efformantibus, nigris,

glomerulis sporarum variae magnitudinis ac formae, plerumque subglobosis v. ovoideis ;

sporis solitariis dilutioribus, 10-12 4 diam., vix distinete verrucosis, flavo-brunneis ; pro-

mycelio gracili, 8-loculari ; sporidiolis capiosissimis, ad septu evolutis.

Hab. in inflorescentiis Penicillariae spicatae, Simla Indae Orient. (Barclay). (Saceardo

Syll. Fung. 14: 426. 1899.

Silby =

Uredo pilulaeformis, n.s.; sporis fusco-nigris irregularibus vel subglobosis regosiusculis

plus minus connatis conglomeratis immixtis minoribus effaetis massam compactam non

rimosam efformatibus, Zeyher, n. 89.

Destroying the germens of some species of Juncus. Uitenhage. December.

Resembling very strongly U. urceolorum, and, like that, infesting the germen; and

forming little globose or elliptic, pill-like bodies, consisting of a compact mass of deep brown,

irregular, subglobose, often angular, more or less connate, slightly rugose spores, mixed

with smaller globose, sub-elliptic, hyaline bodies, which appear to be abortive sporidia.

They become yellow when treated with iodine, and therefore are not fecula. In the centre

of the mass, towards the base, there is sometimes a pale clavate columella, extending half

way up the mass, but this is not always present. The surface of the mass is not cracked,

at least in the specimens before me.

This species resembles very much U. urceolorum, but in that the spores are distinct

and echinulate, and they are not mixed with abortive spores, or if so, in a very slight degree.

The two species are certainly very nearly allied, but are, I believe, distinct. (Berkeley,

M. J. Enumeration of fungi collected by Herr Zeyher in Uitenhage. Jour. Bot. (London)

2: 523-524. 1843.) See Cintractia piluliformis (Berk.) P. Henn.

Ustilago Danthoniae Kalchbr. Grevillea 11:18. — Sporis globosis 36 w., granulosis,

atro-fuscus.

Hab. in spices Danthoniae papposae, in summo monte Chumiberg, prope stationem

Missionis evangelicae ““ Lovedale ” dictam (Rev. T. Buchanan)—Haud procul distat Ust.?

Salveii, sed haec soris linearibus folia modo occupat, non vero spicas “ An potius Til-

letia?”’. (Saceardo Syll. Fung. 7: 373. . 1902.)

P. Hennings in Hedwigia 34 :.328. 1895, adds to the description of Ust. Danthoniae

Kalchbr. as follows: ‘‘ Die sporen werden von Kalchbrenner mit 36 u im Durchmesser

angegeben.. Dieselben sind jedoch wie bei den vorliegenden Original-Examplaren, kugelig,

schmutzigbraun, granulirt, 10-15 » im Durchmesser.” (Saccardo | Syll. Fung. 7 : 373. 1902.)

Ustilago Dregeana Tul. in Ann. Sci. Nat. 1847, p. 83, t. II, f. 13, Fisch. Apercu p. 21.

Soris atris, deformantibus ; sporis globosis vel ovideo-obtusis, 4—5 w diam., ex olivaceous

brunneo-arantiacis, papillosus.

Hab. in pedunculis inflorescentiae graminis cujusdam in C. Bonae Spei (Drege) et

Cynodontis Dactyli Somerset East Africae australis (Mac Owan) = Teste Kalchbrenner in

Grevillea 11 : 18 inflorescentiam, adhuc in vagina latentem prorsus deformantam, fungillus

pulvere atro obducit. Sed ex Kalchbrenner l.c.; sporae sunt majores, 12-15 ~ diam. leviter

tuberculatae huic ad aliam speciem forte spectantes. (Saccardo Syll. Fung. 7 : 467. 1888.)

Ustilago Penniseti Rabenh. in Hedwigia 1871, pag. 18, Fisch. Apercu. p. 14, p.p.

Uredo (Ustilago) trichophora var. Penniseti Kunze in Flora, 1830, p. 369, Ustilago

Carbo 1 columellifera b. trichophora Tul. in Ann. Sci. Nat. 1847, p. 81. —

Soris clausis, firmulus, atris, colulella regida a basi divisa instructis ; sporis magnitudine

diversissimis, 5-5-12 4 diam., valde irregularites globosis, saepe subangulosis, conglobatis

non vel vix punctatis, episporio crassissimo, |- 8-2. uw crasso, brunneis.

Hab. in ovariis Penniseti dichotomi, vulpini, cenchroidis et fasciculati in Aegypto (Schwein-

furth) et imsula Maderia (Schroeter). (Saccardo Syll. Fung. 7 : 462. 1888.)

Ustilago piluliformis (Berk.) Tul. in Ann. Sci. Nat. 1847, p. 93, t. V, f. 27-30, Uredo piluli-

formis Berk. Fungi Uitehn. p. 507, t. XXII, f. 6. —

Sori compactis, atris; sporis vel ovideo-angulatis, 16-20 « 12-16 w diam. levibus,

episporio inaequaliter incrassato, atro, partim hyalinulo.

Hab. in ovarlis Juncorum, pr. Uitenhage in Africa meridionali (Zeyher). — A Cin-

tractia Junci Trel. plane distincta. Species, ob sporae maxime glomerato—coalitar, teste

Tuslane firsan aptius inter Thecaphoras militare. (Saccardo Syll. Fung. 7 : 458-459. 1888.)

See Cintractia piluliformis (Berk.) P. Henn. for change of name.

316

Ustilago Sladenii Pole-Evans nov. spec.

Soris olivaceo-atris, pulverulentis in rachidibus floribusque, eos omnino destruentibus ;

sporis globosis vel subglobosis 5-6 « diam., glabris, dilute olivaceis.

In the flowers and branches of the inflorescence of a grass (probably EHhrharta sp.)

Garies, 1910-11.

H. H. W. Pearson, No. 6728 (Pole-Evans No. 8409).

Saccardo does not include this species in his Sylloge Fungorum. (South African College,

Ann. Bolus Herb. 1: 115. 1915.)

Ustilago nuda (Jens.) Kell. et Swingle II Rep. Agr. Kans. p. 215 et 277, t. Il, f. 7-17.

Ust segetum Auct. p.p., Ust. Horder nuda Jens. in litt.—

Massa sporarum brunneo-olivacea, laxa mox libera; sporis ovoideis, ellipsoideis vel

subglobosis, 5-7 x 5 = 6-5, subolivaceis, germinando promycelium parce ramosum, apicibus

saepe inflatum gignentibus ; sporidiolis hucusque non visis.

Hab. in ovaris Horde: vulgaris in Kuropa amer. bor. Japomis. (Saccardo Syll. Fung.

9 : 283. 1891.)

Ustilago uzandensis P. Henn. Pilz Ostafr. p. 48. 1895.

Soris atro-olivaceis, epidermide tectis, dein cintis, pulverulentis, ad apicem culmorum

in rhachibus florisbuaque, rhachidem curvam contortamque efformantibus: sporis sub-

globosis, saepe acutangularis, fusco-olivaceis, sublevibus, punctulatis, 6-8 x 5-7.

Hab. in spicis Panicum in Africa trop. (Saccardo Syll. Fung. 14: 414-415. 1899.)

Cintractia capensis (Reess) Ciferri, n. comb. Ann. Myc. 29:72. 1931.

Ustilago capensis Reess! in Sitzb. phys.-med. Soc. Erlangen 1875, pp. 70-72.

Ovariicola ; soris aureo-flavis, pulverulentis, sporis globosis, 15-16 ~ diam., episporio

reticulato, strato triplici, constituto, interiori tenui, flavido, lineolis areolisque subhyalinis.

Hab. in ovariis Junci capensis et lomatophyll, e Cap Bonae Spei proveneintium in

Bremen Germanae (Buchenau).

(Saceardo Syll. Fung. 7: 478. 1888.)

Cintractia leucoderma (Berk.) P. Henn. 84: 335. 1895.

Ustilago leucoderma Berk. Ann. Mag. Nat. Il. 9: 200. 1852.

Cintractia Krugiana Magn. Bot. Jahrb. (Engler) 17: 490. 1893.

Cintractia affinis Peck, N.Y. State Mus. Bull. 67:28. 1903.

Ustilago leucoderma Berk. Fungi 8. Domingo n. 54, Fisch. Apercu p. 16. —

Soris atris maculis magnis insidentibus, consta albida inaequali subvelatis ; sporis

rotundatis, rarius irregularibus, 13-17 diam., opacis, atris, levibus.

Hab. in vaginis foliorum Caricis et Cyperaceae (?) cujusdam in St. Domingo et in insula

Borneo; in vaginis Rhynchosporae aureae ad Ratnapoora. (Saccardo Syll. Fung. 7 : 460.

1888.)

Cintractia piluliformis (Berk.) P. Henn. Hedwigia 1898, p. 293.

Uredo piluliformis Berk. Fg. Uitenh. p. 507, t. XXII, fig. 6.

Ustilago piluliformis Tul. Ann. Sci. Nat. 1847, p. 93, t. V, fig. 27-30, Sace. Syll. VII, p. BOO es

Soris compactis, stris, globosis, v. cylindricis, duris 1-2 mm. diam.; sporis densis

conglobatis, globosis, ellipticis v. ovoideis, initio hyalinis, intus. granulatis v. punctatis,

filamentis hyalinis intermixtis, dein brunneis, postremo atris, impellucidis, 10-17** 9-14.

Hab. in ovariis Junci capensis et Junci spec. in Africa austr. et meridionale. (Saccardo

Syll. Fung. 16: 373. 1902.)

1 Reoss, Max. Ueber Ustilago ? Capensis, einem neuen Brandpilze yom Cap der guten Hoffnung. *

Sitzungsber. der physic.-medic. Soc. zu Erlangen 7 : 70-72, 1875.

317

Cintractia tangensis P. Henn. Engl. Bot. Jahrb. XX XVIII (1905), pag. 103.

Soris in axillis foliorum globosis, primo membrana hyalina vestitis, dein pulverulentis,

aterrimis, 5-6 mm. c. er; sporis subglobosis 8-11 diam. v. ellipsoidesi et c. 8-12 x 7-8, p

brunneo-olivaceis, intus granulosis, episporio levi.

Hab. in axillis foliorum Cyperi sp., Tanga, Usambara, Africa (Tanganyika Territory).

(Saccardo Syll. Fung. 21:510. 1912.)

Cintractia togoensis P. Henn. Kngl. Bot. Jahrb. XX XVIII (1905), pag. 119.

Sori in floribus globosis, duris, atris et paucis subinclusis; sporis subglobosis v. subellip-

soideis, intus granulatis, 12-14** 11-13, episporio levi, olivaceo-brunneo ; hyphis hyalinis

intermixtis.

Hab. in floribus Cyperi sp. Togo Camerum, Africae. (Saccardo Syll. Fung. 21 : 510.

il)

Cintractia usambarensis (P. Henn.) Ciferri, in Archiv f. Bot. (Stockholm) A. 28:7. 1931.

Cintractia leucoderma (Berk.) P. Henn. Hedwigia 1895, p. 335. (syn. C. Krugiana P.

Magn.) — Var. usambarensis P. Henn. Pilz. Ostafr., p. 48. 1895.

Soris cylindraceis, nigris, 14-2 em. longis, 14-16 mm. latis; sporis majoribus, dense

verrucosis, obscurioribus, 15-17 x 14-16 yp.

Hab. in Rhynchospora aurea in Africa trop. (Stuhlman). (Saccardo Syll. Fung. 14 : 420.

1899.)

Sorosporium Wildemanianum P. Henn. in Fl. du Bas.-et Moy-Congo, Ann. Mus. du

Congo V. II. fase. II (1907), page. 87. —

Soris ovaria staminaque destruentibus, atris epidermide cinerea tectis; glomerulis

ellipsoideis v. subglobosis e sporis numerosis compositis, 50-80 ** 50-60 ; sporis subglobosis

v. ellipsoideis fusco-brunneis, verrucosis, 7-10 x 6-9 wp.

Hab. in foliis Andropogonis sp. Mbele Congo (Vanderyst). (Saccardo Syll. Fung.

21:513-514. 1911. Sept.) -

Tuburcinia Eriospermi H. Sydow.

Pustulas orbiculares vel ellipticas 3-6 mm. longas formans; massa sporarum atra,

pulverulenta ; glomeruli sat regulares, globosi vel subglobosi, 16-22 w diam., fere semper

sporam unicam centralem tantum includentis; sporae globosae, fuscae, 10-13 w diam.,

leves ; cellulae peripherical numerosae, leves, pallide fuscae, 6—9 uv longae, 4—6 wv altae.

Hab. in foliis Eriospermi latifolii, Stellenbosch, 6. 1923. leg. P. A. van der Bijl (no.

1142). (Ann. Myc. 22: 237. 1924.)

Entyloma cissigenum P. Henn.! Pilz Ost-Afr. p. 49. 1895.

Maculis amphigenis, gregariis, rotundatis, circiter 2 mm. diam., saepe confluentibus,

atris ; sporis globosis, intus fuscis v. minute viridulis subgranulatis, 18-21 crasso.

Hab. in foliis vivis Cissi in Africa trop. (Volkens). (Saccardo Syll. Fung. 14: 423.

1899.)

Entyloma Oleandrae P. Henn. Hedwigia 1895, p. 326.

Soris striiformibus, amphigenis, ca. 1 cm. longis, 2 mm. latis, atris v. atro-violaceis ;

sporis globosis, raro ellipsoideis, minute granulatis, subhyalinis, 8-12 « 6-10 w; episporio

subfuscidulo.

Hab. in foliis Oleandrae articulatae, Natal (Wood). (Saccardo Syll. Fung. 14: 425.

1899.)

Urocystis Anemones (Pers.) Schrét. form kerguelensis P. Henn. Deutsche Siidpolar-Exped.

1901-1903, 8:1. 1906.

Kerguelen—Station, an der Pinguinbucht, auf lebenden Stengeln und Blattern von

Ranunculus biternatus Sm., Dr. E. Werth, leg. 26. Dezember 1902, 9. Januar 1903. Dieser

1 Tn the original description the specific name is spelled F. cissigena.

318

in ganz Huropa, Nordamerika, Sibirien usw. auf verschiedenartigen Ranunculaceen. ver-

breitete Pilz bildet auf den niederliegenden Stengeln, Blattstielen und Blattern obiger

Pflanzen aufgetriebene bis etwa 2 cm. lange, von eimer graubraunen Oberhaut bekliedete

Brandpusteln, welche bei der Reife durch einen Langsriss aufbrechen und die tiefschwarze

Sporenmasse freilegen Dieselbe besteht aus Sporenballen von ellipsoider oder rundlicheckiger

Form, die meist 20-30 u Durchmesser besitzen. In der Mitte dieser Ballen finden sich

1-3 rundlich eckige oder ellipsoide Hauptsporen, mit dunkelbrauner, undeutlich punktierter

Membran, meist 12 —20 u lang, 10-16 u breit. Diese werden meist von zahlreichen heller

getirbeten, gelblichen, halfkugeligen oder unregelmiassig eckigen, 8-12 u breiten. Neben-

sporen ungeben, mitunter fehlen diese ganz. Die ganze Pflanze wird durch den Parasiten

untérmlich verbildet.

Durch die grésseren Hauptsporen soure durch Nebensporen ist der Pilz von der typischen

Form etwas verschieden, ebense yon U. sorosporioides Korn.

It is doubtful whether Urocystis Anemones form. kerguelensis described by P. Henning

from Kerguelen Island differs from the species other than in slight variations due to host

and climate. With further search, this species should be found in South Africa. Sacecardo

does not list this form in Sylloge Fungorum.

It must be noted that the authority for the species should be Urocystis Anemones

Pers.) Winter; Rab. Krypt.-Fl. 11: 198, WSS,

Tilletia Schenckiana P. Henn. Deutche Sudpolar-Exped., 1901-1903, 8:2. 1906.

Soris ovariicolis en destruentibus paulo deformantibus, paleis laxe cireumdatis, trans-

lucentibus, ellipsoideis, firmis, membrana cinereo-fusca vestitis, ca. 1 x 0-5 mm. diam. :

sporis ellipsoideis vel subglobosis, 20-30 « 20-24 u, episporio reticulato, melleo dein fusco-

brunneo, ca. 2-3 crasso, interdum filis hyalinis, flexuosis, 2-4 ~w crassis intermixtis.

Kerguelen im Tale zurschen Station und Mittelberg, in Bluten von Deschampsia ant-

arctica BK. Desv. Dr. E. Werth, 19 Februar 1903. Diese aussert zierliche Art tritt in den

Fruchtknoten, welcher zerstért und wenig verbildet ist, in ellipsoiden, dunklen Sori von

den trockenhautigen Spelzen locker umschlossen auf und auf dem Scheitel der Sori sind

oft noch die tadigen Narben erkennbar.

Der Pilz ist oon von. Tilletia cerebrina Ellis et Ev. aus Nord-amerika in Ovarien

von Deschampsia caespitosa P. B. sowie von T. Arrae Blytit in Deschampsia calycina durch

die kleinen sori sowie die Sporen verschieden Der Pilz wurde bei der Bestimmung der Nahrp-

flanze bereits von Prof. Dr. Schenck beobachtet und als, Tvlletia erkannt.

This smut described from rocky Kerguelen Island should be found in South Africa.

It is not known where the type specimen is deposited. Saccardo does not list this species

in Sylloge Fungorum.

APPENDIX B.

The following is a list of doubtful and excluded species that have been reported from

South Africa and nearby territory.

Ustilago Welwitschiae Bres. in Sacc. Fl. mycol. Lus. p. 68. —

Soris parvis, globosis, fuligineis, pulverulentis, e squamas erumpentibus; sporis

fuscidulis, globosis, asperulis, 34-4 diam.

Hab. in squamis conorum Welwitschiae mirabilis e Mosamedes Africae allatae in hort.

bot Coimbrae (Méller). (Saccardo Syll. Fung. 14: 411. 1899.) A doubtful smut.

Sorosporium afr wcanum Sydow, Ann. Myce. 7:544, 1909.

Soris ovaria oceupantibus easque omnino destruentibus, atris, pulverulentis ; glomerulis

ésporis usque 8 compositis, 12-25 uw diam. ; sporis globosis vel angulato-globosis, subtilissime

verruculosis vel punctatis, olivaceis vel olivaceo-brunneis 5-9 fe “diam.

Hab. in ovariis Panici trichopi Portugiesich-Ostafrika. 18.4.1908. leg. C. W. Howard.

319

The portion of the type specimen examined (M.H. 681, on Panicum trichopus Hochst.,

Portuguese Hast Africa, coll. C. W. Howard, April 18, 1909) is very poor since there were

very few spores and these were not typical of the Ustilaginales, much less a Sorosporium.

A portion of the type specimen in the Clinton herbarium was examined but it contained

no spores.

The original description says that the spore balls are composed of 8 spores. This is

not typical of a Sorosporium. This fungus is apparently one of the dark spored members

of the fungi imperfecti. At any rate we will list it temporarily as an excluded smut until

more and better material is available for examination.

Tolyposporium Chloridis P. Henn. Pilz Ostafr. p. 49. 1895.

Soris atris, firmis, bullatis, undulato-plicatis, rugulosis, interdum conduentibus ;

glomerulis sporarum firmis, varia magnitudine, subglobosis, acutangulis, e 3-5 sporis com-

positis, 10-20 x 8-15, fusco-brunneis vy. atris; sporis acutangulis, sublevibus, 6-10 uw.

Hab. in fructibus Chloridis abysinicae in Afric trop. (Volkens). (Saccardo Syll. Fung.

14: 426. 1899.)

A study of specimens of Tolyposporium Chloridis P. Henn. from various parts of southern

Africa indicates that the fungus in question is not one of the Ustilagimales but rather one of

the dark coloured Hyphomycetes.

The following South African specimens from the Union Department of Agriculture,

Mycological Herbarium have been examined: No. 9056, collected by I. B. Pole-Evans,

Feb., 1915, at Groenkloof, Pretoria, Transvaal; No. 9770, collected by P. van der Bil,

May 5, 1926, at Mid Illovo, Natal: No. 1627, collected by E. M. Doidge, June 26, 1911,

at Mountain Rise near Pietermaritzburg, Natal; No. 17043, collected by A. O. D. Mogg

(no. 4290), May 1, 1919, in Zululand. In each case the host was Chloris gayana.

Tolyposporium Volkensii P. Henn. Pilz Ostafr. p. 49. 1895.

Soris ad germina pustulatis, submesenteriformibus, atris, subgelatinosis (7), rugulosis

circ. 5-8 mm. diam.; glomerulis subglobosis v. irregulariter acutangulis, firmis, 15-28 yu

diam., e 3-15 sporis compositis, plus minus fuscis; sporis subglobosis, ovoideis, oblongis,

e mutua pressione acutangulis, fusco-brunneis, minute granulatis, 5-11 wm.

Hab. in ovariis Sorghorum cultorum in Africa trop. (Volkens).

Mason! (1926) worked with type material of Tolyposporium Volkensu P.-Henn. from

Kast Africa and found that it was not one of the Ustilaginales but was Cerebella sorghi-

vulgaris Subram. This was later confirmed by Subramanian. ‘‘ Hennings’ species is a true

Cerebella having a true convoluted stroma with a surface layer of palaside-like conidiophores

each bearing at its apex a conidium with septa in three planes.”

Species of the Graphiolaceae are excluded. While they are closely related to the Ustila-

ginales, they are not a family of this order. Graphiola phoenicis Poit., the most common

species, attacks Phoenix sp. both in greenhouses and in the open.

Various species of the genera Cerebella and Ustilaginoidea are sometimes confused with

the smuts. These genera are members of the family Dematiaceae of the imperfect fungi.

Tuberculina, a member of the Tuberculariaceae of the fungi imperfecti, is another genus

that is sometimes put with the Ustilaginales.

1Mason, E. W., on two species of Tolyposporium Woronin recorded on cultivated Sorghwm.

Rrans. British Myc. Soc. 9: 284-286. 1926.

320

HOST INDEX

Albuea altissima Dryand.

Ustilago Vaillanti Tul.

Amphilophis insculpta Stapf. (see Bothriochloa insculpta. A. Camus.)

Andropogon finitimus Hochst.

Ustilago Andropogonis-finitimi Maub.

Andropogon intermedius R. Br.

Sphacelotheca Doidgeae Zundel

Andropogon lepidus Nees (see Hyparrhenia Tamba Anders.)

Andropogon papillosus Hochst. (see Dicanthium papillosum Stapf)

Andropogon pertusus Willd. (see Bothriochloa pertusa A. Camus)

Andropogon rufus Kunth. (see Hyparrhenia rufa Stapf)

Andropogon Schoenanthus Linn. (see Cymbopogon Schoenanthus Spreng.)

Andropogon sorghum Brot. (see Sorghum vulgare Pers. var. caffroruam Hubb. and Rehder.)

Andropogon sp.

Sorosporium Flanaganianum Zundel

Sorosporium Hotsonii Zundel

Sorosporium pseudomaranguense Zundel

Sorosporium Wildemannianum P. Henn.

Sphacelotheca Nyassae (H. & P. Syd.) Zundel

Sphacelotheca Stuhlmanni (P. Henn.) Zundel

Anthephora pubescens Nees

Sphacelotheca Anthephorae (Syd.) Zundel

Anthistiria sp.

Tolyposporium Anthistiriae Cobb. (Appendix A)

Aristida junciformis Trin. and Rupr.

Sorosporium consanguineum Ell. & Ev.

Avena sativa L.

Ustilago Avenae (Pers.) Jens. °

Ustilago levis (Kell. & Sw.) P. Magn.

Bidens pilosa Linn.

Entyloma Bidentis P. Henn.

Bothriochloa glabra A. Camus.

Sphacelotheca Andropogonis (Opiz.) Bubak

Sphacelotheca Doidgeae Zundel

321

Bothriochloa insculpta A. Camus

Sphacelotbeca Amphilophis Sydow.

Bothriochloa pertusa. A. Camus.

Sphacelotheca tenuis (H. & P. Syd.) Zundel

Bothriochloa sp.

Sphacelotheca Doidgeae Zundel

Briza maxima L.

Tilletia Viennotii Sydow

Bromus unioloides H. B. K.

Ustilago bromivora (Tul.) Fisch.

Carex ethiopica Schkuhr.

Farysia olivacea (D.C.) H. & P. Syd.

Carex phacota Spreng.

Farysia olivacea (D.C.) H. & P. Syd.

Carex sp.

Cintractia leucoderma (Berk.) P. Henn. (Appendix A)

Cenchrus ciliaris L.

Sorosporium cenchri (Bref.) Zundel

Ustilago Penniseti Rabh. (Appendix A)

Chloris abyssinica Hochst.

Tolyposporium Chloridis P. Henn. (Appendix B)

Chloris gayana Kunth.

Tolyposporium chloridis P. Henn. (Appendix B.)

Cissus sp. (Vitis sp.)

Entyloma cissigenum P. Henn. (Appendix A.)

Cymbopogon elegans Spreng.

Sorosporium Holstiu P. Henn.

Cymbopogon excavatus Stapf.

Sphacelotheca natalensis Zundel

Cymbopogon plurinodis Stapf

Sorosporium pretoriaense Zundel

Sphacelotheca concentrica Zundel

Sphacelotheca Moggi Zundel

Cymbopogon Schoenanthus Spreng.

Sphacelotheca Andropogonis (Opiz.) Bubak

Sphacelotheca Milbraedii (H. & P. Syd.) Zundel

322

Cymbopogon sp.

Sphacelotheca Zilligi Zundel

Digitaria ternata Stapf

Ustilago Rabenhorstiana Kuhn

Digitaria sp.

Ustilago Rabenhorstiana Kuhn

Echinochloa Crus-galli Beauv.

Sphacelotheca Panici-miliacei (Pers.) Bubak (Appendix A.

Ustilago trichophora (Link.) Kunze

Kehinochloa Holubu Stapf

Ustilago Crus-galli. Tracy and Karle.

Ustilago Holubii Syd.

Echinochloa sp.

Sorosporium cryptum McAlp.

EKhrarta sp.

Ustilago Sladenii Pole-Evans (Appendix A.)

Khonurus argenteus Nees

Ustilago Elionuri P. Henn. and Pole-Hvans

Enneapogon brachystachyus Stapf

Sphacelotheca modesta (Syd.) Zundel

Enneapogon sp.

Sphacelotheca Pappophori (Pat.) Zundel

Eragrostis aspera Nees

Tilletia transvaalensis Zundel

Erianthus Saccharoides Michx.

Ustilago scitaminea H. Sydow

Kriospermum latifolium.

Tubureinia Eriospermi H. Syd. (Appendix A.)

Kucomis punctata L’Herit.

Ustilago Vaillanti Tul.

Fingerhuthia africana Lehm.

Ustilago Fingerhuthiae Syd.

Heteropogon contortus (L.) Roem. et Schultz.

Sphacelotheca monilifera (Hll. et Hw.) Clint.

Sphacelotheca columellifera (Tue.) Ciferri

323

Heteropogon hirtus Pers. (see H. contortus R. & 8.)

Hordeum vulgare L.

Ustilago Hordei (Pers.) Lagerh.

Ustilago nuda (Jens.) Kell. et Sw. (Appendix A}

Hyparrhenia aucta (Stapf) Stent

Sorosporium proliferatum Zunde!

Hyparrhenia cymbaria Stapf

Sphacelotheca Ritchie: Zundel

Hyparrhenia hirta Stapf

Ustilago puellaris Syd.

Hyparrhenia rufa Stapf

Sorosporium Zundelianum Ciferri

Hyparrhenia Ruprechti Fourn.

Sorosporium Everharti Ell. and Gall.

Sphacelotheca Andropogonis (Opiz.) Bubak

Sphacelotheca Evansu Zundel

Sphacelotheca Ruprechtu Syd.

Hyparrhenia Tamba Anderss.

Sorosporium austro-africanum Zundel

Sorosporium Clintonii Zundel

Sorosporium Maraguenense P. Henn.

Sorosporium Tembuti P. Henn and Pole-Kvans

Hyparrhenia sp.

Sorosporium Healdu Zundel

Sphacelotheca tenuis (H. & P. Syd.) Zundel

Imperata arundinacea Cyrilli

Ustilago scitaminea H. Syd.

Juncus capensis Thunb.

Cintractia capensis (Reess) Ciferri (Appendix A.)

Cintractia piluliformis (Berk.) P. Henn. (Appendix A.)

Juncus lomatophyllus Spreng.

Cintractia capensis (Reess) Ciferri (Appendix A.)

Juncus sp.

Cintractia piluliformis (Berk.) Ciferri (Appendix A.)

Ustilago piluliformis (Appendix A.)

Melinis tenuinervis Stapf

Cintractia Melinis Zundel

324

Oleandra articulata Swartz

Entyloma Oleandrae. P. Henn. (Appendix A.)

Ornithogalum lacteum Jacq.

Ustilago Peglerae Syd. & Butler

Ornithoglossum glaucum Salisb.

Urocystis Ornithoglossi (Syd.) Zundel

Panicum Carthaginense Sw.

Sphacelotheca Panici-miliacei (Pers.) Bubak (Appendix A.)

Panicum Crus-galli Linn. (see Echinochloa Crus-galli Beauv.)

Panicum helopus Trin. (see Urochloa helopus Stapf)

Panicum laevifolium Hack.

Sorosporium afrum Syd.

Sorosporium harrismithense Zundel

Tilletia heterospora (P. Henn.) Zundel

Panicum longijubatum Stapf.

Sorosporium versatilis (Syd.) Zundel

Panicum maximum Jacq.

Sorosporium Panici McKinnon

Tilletia heterospora (P. Henn.) Zundel

Panicum miliaceum Linn.

Sphacelotheca Panici-miliacei (Pers.) Bubak (Appendix A.)

Panicum proliferum Lam. var. paludosum Stapf. (see P. longijubatum Stapf

Panicum trichopus Hochst.

Sorosporium africanum (Syd.) (Appendix B.)

Panicum sp.

Tilletia heterospora (P. Henn.) Zundel

Ustilago ugandensis P. Henn. (Appendix A.)

Pappophorum scabrum Kunth

Sphacelotheca Pappophori (Pat.) Zundel

Penicillaria spicata Willd.

Tolyposporium Penicillariae Bref. (Appendix A.)

Pennisetum cenchroides Rich. (see Cenchrus ciliaris L.)

Pennisetum dichotomum Delile

Ustilago Penniseti Rabh. (Appendix A.)

Pennisetum fasciculatum Trin.

Ustilago Penniseti Rabh. (Appendix A.)

325

Pennisetum vulpinum Stapf. & Hubb.

Ustilago Penniseti Rabh. (Appendix A.)

Physalis m nima Linn.

Entylcma australe Speg.

Physalis peruviana Linn.

Entyloma australe Speg.

Polygonum lapathifolhum Linn. var glabrum Burtt Davy.

Melanopsichium austro-africanum (Speg.) G. Beck

Ranunculus biternatus Sm.

Urocystis Anemones (Pers.) Schrot. (Appendix A.)

Rhyncospora aurea Vahl

Cintractia leucoderma (Berk.) P. Henn. (Appendix A.)

Cintractia usambarensis (P. Henn.) Ciferri (Appendix A.)

Rottboellia compressa L.

Sphacelotheca densa (McAlp.) Ciferri

Sphacelotheca flagellata (Syd.) Zundel

Rottboellia exaltata Linn.

Sphacelotheca flagellata (Syd.) Zundel

Saccharum officinarum Linn.

Ustilago scitaminea H. Syd.

Scilla Kraussii. Baker.

Ustilago Vaillanti Tul.

Scilla sp.

Ustilago Vaillanti Tul.

Setaria aurea A. Br. (see 8. sphacelata Stapf & Hubb.)

Setaria italica (L.) Beauv.

Ustilago Crameri Korn.

Setaria nigrirostris Dur. and Schinz.

Ustilago Evansii P. Henn.

Setaria perennis Hack.

Sorosporium Setariae McAlp.

Setaria sphacelata Stapf. & Hubb,

Tolyposporium setariicolum H. & P. Syd. (Appendix A.)

Setaria sp.

Ustilago Evansii P. Henn.

Sorghum caffrorum Beauv. (see S. vulgare Pers. var caffrorum)

Sorghum halepense Pers.

326

Sorosporium Reilianum (Kuhn) McAlp.

Sorosporium Simi P. Henn. and Pole-Evans.

Sphacelotheca cruenta (Kuhn) Potter

Sorghum versicolor Anders.

Sphacelotheca transvaalensis Zundel

Sorghum vulgare Pers.

Sphacelotheca cruenta (Kuhn) Potter

Sorghum vulgare Pers. var. caffrorum (Thum.) Hubb. and Rehder

Sorosporium filiferum (W. Busse) Zundel

Sorosporium Reilianum (Kuhn) McAlp.

Sphacelotheca Holci H. 8. Jackson

Sphacelotheca Sorghi (Link.) Clint.

Sorghum sp.

Sorosporium Reilianum (Kuhn) McAlp.

Sphacelotheca cruenta (Kuhn) Potter

Tolyposporium Volkensii P. Henn. (Appendix A.)

Sporobolus indicus R. Br.

Ustilago Schlechter1 P. Henn.

Stenotaphrum glabrum (see St. secundatum (Walt.) Kunze

Stenotaphrum secundatum (Walt.) Kunze

Ustilago affinis Ell. and Ew.

Themeda Forskali: Hack. (see Th. triandra Forsk.)

Themeda triandra Forsk.

Sorosporium Holst P. Henn.

Sphacelotheca Vryburg Zundel

Trachypogon plumosus Nees

Ustilago Trachypogonis Zundel

Tristachya Rehmanni Hack.

Tolyposporium trystachydis (Syd.) Zundel

Triticum dicoccum Schrank

Urocystis Tritici Koern

Triticum durum Desf.

Urocystis Tritici Koern

Triticum turgidum Linn.

Urocystis Tritici Koern

Triticum vulgare Vill.

Tilletia foetans (B. & C.) Trel.

327

Tilletia Tritici (Bjerk.) Wint.

Urocystis Tritici Koern

Ustilago Tritici (Pers.) Rostr.

Urochloa helopus Stapf

Sorosporium verecundum (Syd.) Zundel

Sphacelotheca pretoriense (Pole-Evans) Zundel

Welwitschia mirabilis Hook.

Ustilago Welwitschiae Bres. (Appendix B.)

Zea Mays Linn.

Sorosporium Reilianum (Kuhn) McAlp.

Ustilago Zeae (Beckm.) Unger

Zinnia pauciflora Linn.

Entyloma Zinniae Syd.

INDEX OF GENERA, SPECIES AND SYNONYMS.

Anthracoidea 302 Cintractia leucoderma 316, 317, 321, 325

Burrillia 284, 285 leucoderma var. usambarensis 317

Caeoma destruens 314 Melinis 302, 323

olivaceum 294 patigonica 293

segetum 288, 290, 291, 311 piluliformis 315, 316, 323

sitophyllum 311 Reiliana 307

Syntherismae 293 Sorghi-vulgaris 295

trichophora 291 tangensis 317

Zeae 292 togoensis 317

Cerebella 319 usambarensis 317, 325

Cintractia 287, 302 Doassansia 284, 285

Cintractia affinis 316 Elateromyces 294

capensis 316, 323 Elateromyces olivacea 294

caricis 286 Endothlaspis 294

carieiwola 294 Entyloma 284, 285, 287, 313

columellifera 297 Entyloma australe 313, 325

densa 296 Besseyi 313

Ischaemi 298 Bidentis 313, 320

Junci 315 cissigenum 317, 321

Krugiana 316 compositarum 286

Entyloma Dahhae 313

Oleandrae 317, 324

Physalidis 313

Zinniae 313, 327

Erysibe foetida 311

maydis 292

occulta 312

olivacea 294

Panicorum 314

vera Avenae 290

vera Hordei 288

Sorosporium Junci 310

Maranguenense 309

Paniei 307, 324

Panicer-miliacer 314

pretoriaense 303, 321

proliferatum 307, 323

pseudomaranguense 309, 320

Reilianum 307, 326, 327

Saponariae 303

Setariae 305, 325

Simii 308, 326

Tembuti 305, 323

Tristachydis 310

tumefaciens 304

vera Tritici 291

Farysia 294

Farysia americana 294 verecundum 304, 327

olivacea 294, 321 versatilis 308, 324

Lycoperdon Tritict 291, 311 Wildemannianum 317, 320

Zeae 292 Zundehanum 304, 323

Melanopsichium 287, 302 Sphacelotheca 285, 287, 294

Melanopsichium austro-americanum 302, 325 Sphacelotheca Asunplniloplans 299, 321

NV rene A Andropogonis 298, 320, 321

Necrosis 287 :

323

Anthephorae 295, 320

austro-americanum 302

concentrica 296, 321

Polycystis 312

Protomyces microsporus 313

Physalidis 313

Reticularia segetum 288, 290 collumellifera 297, 322

Ramphospora 313 cruenta 286, 296, 326

Sorosporium 285, 287, 303 densa 296, 325

Sorosporium africanum 318, 324 Dinteri 298

afrum 306, 324 Doidgeae 296, 320, 321

austro-africanum 305, 323 Evansil 297, 323

Bornmulleri 303 flagellata 301, 325

Clintoni 306, 323 Holei 299, 326

Cenchri 303, 321 Ischaemi 298

consanguineum 303, 320 Milbraedi 299, 321

cryptum 306, 322 modesta 301, 322

Everhartn 304, 323 Moggi 295, 321

filiferum 808, 326 monilifera 300, 322

Flanaganianum 309, 320 natalensis 298, 321

harrismithense 309, 324 Nyassae 300, 320

Heald 305, 323 Panici-miliacei 314, 322, 324

Holsti 303, 321, 326 Pappophori 300, 322, 324

Hotsoni 306, 320 pretoriense 501, 327

inconspicuum 304 Reiliana 307

Sphacelotheca Ritchiei 299, 323

Ruprechtii 297, 323

Sorghi 286, 295, 326

Stuhlmanni 301, 320

tenuis 297, 321, 323

transvaalensis 298, 326

Vryburg 296, 326

Zilligii 300, 322

Sporisoriwm 294

Sporisorium Sorghi 295

Tilletia 285, 287, 310

Tilletia Airae 318

Ayresii 283, 311

caries 311

cerebrina 318

destruens 314

foetans 310, 326

heterospora 310, 324

laews 285, 310, 329

Schenckiana 318

Sorghi-vulgaris 295

transvaalensis 311, 322

Tritici 285, 286, 311, 327

Viennoti: 311, 321

Tolyposporium 287, 310

Tolyposporium Anthistiriae 314, 320

Cenchri 303

Chloridis 319, 321

Everhart 304

jiliferum 308

Tristachydis 310, 326

Volkensii 319, 326

pampeano 314

Penidillariae 314, 324

setariicolum 314, 325

Tracya 284, 285

Tuburcinia 287, 312

Tuburcinia Eriospermi 317, 322

Ornithoglossi 312

Trientalis 312

Tritici 312

Tuberculina 319

329

Uredo carbo 288

carbo Avenae 290

carbo-Hordei 288

carbo Panici miliacei 314

carbo Tritici 291

carbo-vulgaris avenae 290

caries 310, 311

destruens 314

foetida 311

Hydropiperis 295

Maydis 292

olivacea 294

pilulaeformis 315

piluliformis 316

segetum 288, 291

segetum Mays Zeae 292

segetum Panici-miliacer 314

segetum Tritici 291

sittophila 311

Syntherismae 304

trichophora 291, 315

trichophora var. Penniseti 315

urceolorum 315

ZLeae 292

Zeae Mays 292

Urocystis 285, 287, 312

Urocystis Anemones 317, 325

Anemones f. kerguelensis 318

Ornithoglossi 312, 324

sorosporioides 318

Tritici 312, 326, 327

Ustilagidium 287

Ustilagidium Tritic: 291

Ustilaginoidea 319

Ustilago 285, 287

Ustilago affinis 288, 326

americana 289

Andropogonis-finitimi 293, 320

Anthephorae 295

Aristidae 303

austro-americanum 302

Avenae 286, 288, 290, 320

Avenae f. foliicola 290

Avenae var. levis 288

axicola 302

Ustilago Brachypodiu 293

Brachypodii-distachyr 293

Bromi-arvensis 293

Bromi-mollis 293

bromivora 293, 321

bromivora f. Brachypodi 293

capensis 316

Carbo var. columellifera 297

Carbo | collumellifera b. tricho-

phora 315

Carbo destruens 314

Carbo-vulgaris avenae 290

Carbo-vulgaris bromivora 293

Carbo-vulgaris Hordacea 288

Carbo-vulgaris Triticea 291

carieicola 294

catenata 294

Cesati 293, 304

Crameri 289, 325

eruenta 296

Crus-yalli 292, 322

eryptum 306

Cynodontis 289

cylindrica 298

Dactyloctaenii 290

Danthoniae 315

destruens 314

destruens var. Digitariae 293

Dinterr 298

Dregeana 315

Klionuri 288, 322

Euchlenae 292

Evansiu 294, 325

Fingerhuthiae 291, 322

flagellata 301

foetans 310

heterospora 310

Henningsu 289

Hilariae 288

Holubiu 290, 322

Hordei 286, 288, 291, 323

inconspicua 304

Ischaemi 298

Jensenit 288

Kolleri 288

leucoderma 316

levis 286, 288, 320

Maydis 292

Mays Zeae 292

Milbraediri 299

modesta 301

Ustilago monilifera 300

nuda 316, 323

Nyassae 300

olivacea 294

Panici-miliacei 314.

Pappophori 300

pretoriense 301

Peglerae 294, 324

Penniseti 315, 321, 324. 325

puuliformis 315, 316, 323

puellaris 289, 323

pulveracea 307

Rabenhorstiana 293, 322

Reiliana 307

Reiliana foliicola 307

Sacchari 289

Schlechteri 292, 326

Schweimitzii 292

Seillae 290

Scitaminea 289, 322, 323, 325

Segetum 287, 288, 290, 316

Segetum Avenae 290

Segetum var. Hordei 288

Segetum var. Tritici 291

Setariae 293

Setum 291

sitophila 311

Sladenu 316, 322

Sorghi 295

Sphaerogena 293

Stenotaphri 288

subolivacea 294.

Syntherismae 293

tecta hordei 288

tenuis 297

Trachypogonis 292, 326

trichophora 291, 322

Tritic1 291, 327

Triticr foliicola 291

Tulasner 296

tumefaciens 304

ugandensis 316, 324

Vaillanti 290, 320, 322, 325

Vavilow 291

versatilis 308

verecunda 304

Welwitschiae 318, 327

Zeae 284, 292, 327

Zeae-Mays 292

331

SOME SOUTH AFRICAN FUSARIA.

By E. M. Dorper.

THE REVISED list of plant diseases known to occur in South Africa, which was published

in 1931 (8), records a large number of Fusarium spp. found in diseased tissues, particularly

in connection with wilts and foot rots. That little is known of the role of these organisms

in plant disease in this country, is indicated by the fact that comparatively few South African

records are to be found in the recent book * Die Fusarien,” by Wollenweber and Reinking

(61); therefore as a first step in the study of wilts and foot rots and the decay of storage

organs, it was considered desirable, that as many strains as possible of the Fusaria associated

with plant disease, should be studied and classified. The present account can only be regarded

as a preliminary study of the genus Fusariwm in South Africa ; the work is far from complete,

and records are lacking of a number of forms said by Wollenweber (61) to be prevalent in

all warm countries; species of Fusarium are probably responsible for a number of wilt

diseases which have not yet been investigated. A general survey of this kind, however,

should be a useful basis for more detailed study, especially of the species causing vascular

wilts of specific plants.

A large number of strains (-+ 100) was isolated while making a study of dry root rot

of citrus trees, which is one of the most serious causes of loss in orange orchards. It was

found that a large percentage of the fungi isolated from decaying citrus roots belonged to

the genus Fusarium ; these organisms were also found in roots apparently healthy and in

the soil. Inoculation experiments have, so far, given only negative results, and it is not

known what part the fungi play in the decay of the roots.

A further 300 strains were isolated while making a survey of the fungi found in citrus

fruits decaying in storage. Apart from the citrus investigations, no systematic collections

have been made. Many strains were isolated from plants sent for examination, or were

obtained in the course of investigation of wilt diseases of tobacco, tomato, aster, cucurbits

and other plants, by officers of the Division of Plant Industry. About 850 strains were

studied in all, but a small percentage could not be brought into good sporulating condition

and was discarded unidentified ; these strains were chiefly Fusaria of the “ elegans” section.

The method employed was as follows: Small portions of the affected tissues were

planted in prune agar plates, and when sufficient growth had taken place, transfers were

made to plain agar plates from which hyphal tip isolations were made (6), or single spore

cultures were obtained by the dilution method. The culture media used for detailed ob-

servations and the methods adopted, were those recommended in * Fundamentals for

taxonomic studies of Fusarium ”’ (62) ; the synthetic medium adopted as a standard medium

by Brown (7) with the addition of starch, was also used, as it proved a useful medium for

the production of conidia ; this medium was also largely used for stock cultures. In com-

puting percentages of conidia with 0-3-5 or more septations, some 200 conidia were counted,

and a large number were measured to get extreme and average measurements. (Measure-

ments are given in microns unless otherwise stated.) Ridgways’ colour standards and nomen-

clature (40) were used for recording the colours of conidia, mycelium and stroma of the

various strains in culture on standard media. Representative conidia of each species,

variety and form were drawn to scale with the aid of the camera lucida, the magnification

being 1 : 800.

332

The general descriptions of species and varieties were adapted from those found in

the monographs of Wollenweber and Reinking (59, 61), where full synonymy and biblio-

graphy may be found, the more detailed descriptions of conidia and cultural characters

being derived frcm a study in culture of the South African strains isolated. Dried cultures

of representative strains have been deposited for reference in the Cryptogamic section of

the South African National Herbarium, these being indicated in the text by M.H. (mycolo-

gical herbarium) numbers.

I am greatly indebted to Dr. Wollenweber for his advice and co-operation. He very

kindly examined and identified some 50 strains in culture, and his annotations on the identi-

fications were of great assistance, especially in the earlier part of the work. Iam also indebted

to him for perusing and criticising the manuscript.

I also wish to acknowledge the very considerable assistance of Mr. L. J. Kresfelder,

who made a number of the original isolations, and was responsible for the majcr part of

the laborious work of conidial computations and measurements. To Dr. V. A. Wager, I

am indebted for a large number of cultures from wilting tomato and aster plants and from

a number of other hosts. His isolations and collections are distinguished by his name in

brackets after the record. I am also indebted to several other officers of the Division of

Plant Industry for cultures and material.

FUSARIUM (Link).

Link H.F. in Mag. Ges. nat. Fr. 3: 10, 1809; Spec. Plant. 2: 105, 1825. Saccardo, Michelia 2 : 35,

1880; Syll. Fung. 4 : 694, 1886. Appel and Wollenweber in Arb. Biol. Anst. f. Land.-u. Forstw. Berlin—

Dahlem 8: 60, 1910. Wollenweber and Reinking, Die Fusarien p. 9, 1935.

Syn. Fusisporium Link pr.p. in Mag. Ges. nat. Fr. 3:19, 1809; Spec. Plant. 1 : 30, 1824.

Fusidium Link pr. p. in Mag. Ges. nat. Fr. 3:10, 1809; Spec. Plant. 2:96, 1825.

Atractium Link pr. p. in Mag. Ges. nat. Fr. 3: 10, 1809.

Fusoma Corda, Icon. Fung. 1 : 7, 1837.

Selenosporium Corda Icon. Fung. 1 : 7, 1837.

Pionnotes Fries, Summa. Veg. Scand. p. 481, 1849.

Microcera Desm. pr. p. in Ann. Sci. nat. 3, sér. 10 : 359, 1848.

Discofusarium Petch in Trans. Brit. Myc. Soc. 7: 164, 1922

Pseudomicrocera Petch in Trans. Brit. Muc. Soc. 7: 164, 1922.

Discocolla Prill. et Del. in Bull. Soe. Myc. France 10 : 86, 1894.

Conidia scattered in the mycelium, in false heads forming large or small balls, in flat

or raised mucilaginous layers (pionnotes), on a smooth or wrinkled thallus or direct on

the substratum, or in masses on a tubercularia-like plectenchymatous to sclerotial stroma

of limited extent (sporodochia) ; pale or brightly coloured (orange, salmon, ochre) in mass.

Conidia often of two kinds ; microconidia which are usually 1-celled and scattered ; macro-

conidia which are usually 3- or more septate, fusiform to falcate, dorsiventral, curved in

various ways or almost straight, and often with a pedicellate base.

Conidiophores simple to compoundly sub-verticillate ; conidia produced successively

at the tips of the septate main conidiophore, or at the tips of its irregular or whorled lateral

branches, sometimes united in chains. Occasionally they are formed (yeast-like) directly

on the mother conidium or on short sterigma-like branches arising from it.

Chlamydospores usually present, globose, ovoid or pear-shaped, 1—2-celled or in chains

or clusters, terminal or intercalary, brownish in colour or becoming tinged with the colour

of the stroma.

Sclerotia spherical, solid, occurring singly, or in groups, or absent. Sclerotial stromata

occur in many groups; they are erumpent, hemispherical, smooth or rough and cauliflower-

like ; or erect, stilboid, sometimes with antler-like branching, sessile or stalked; they

serve as a stroma for the sporodochia or remain sterile.

333

Hyphae septate, sparse or abundant, branched in various ways, epi- or endo-phytic,

free or forming a mycelium which may be loosely interwoven, or form a close, coremium-like

to plectenchymatous or sclerotial mass. The mycelium is partly submerged and partly

superficial, pale or brightly coloured (red, yellow, brown, green, blue). Aerial mycelium

mould-like ; mycelium in substratum gelatinous, leathery, plectenchymatous, often with

patches of sclerotial thickening.

A number of species of Fusarium are the conidial forms of Ascomycetes of the genera

Nectria, Calonectria, Gibberella and Hypomyces.

The genus has been divided by Wollenweber (61) into sixteen sections and sub-sections.

Key to the Sections and Sub-sections.

A.—Microconidia normally present, usually 1-celled, ovoid, fusoid,

reniform or pyriform :—

B.—Microconidia more or less pyriform....................--- Sporotrichiella.

BB.—Microconidia not pyriform :

C.—Chlamydospores wanting :

D.—Microconidia in chains :

E.—Macroconidia thin-walled ; colour and form of

conidia and stromalike Lateritium............ Liseola.

EE.—Macroconidia comparatively thick-walled ;

colour, form of conidia, stroma and sclerotia

bile D Olay rere hore ore en oe cient cae Spreariordes.

DD.—Microconidia not in chains in F’. neoceras, and F.

moniliforme v. subglutinans (Liseola), and F. lateri-

tium vars. minus and uncinatum (Lateritium) :

CC.—Terminal and intercalary chlamydospores present :

D.—Conidia with thin walls and delicate, rather in-

conspicuous septa, tapering or constricted towards

the apex, pedicellate at the base, in mass brownish-

white, pink, salmon-orange or sometimes quite pale Hlegans.

DD.—Conidia with ccmparatively thick walls and distinct

septations, sub-truncate, rounded or briefly ros-

trate, somewhat curved at the apex, more or less

pedicellate at the base; in mass brownish-white,

cream, golden yellow or often taking up the blue or

PIKISIN COMOUIE Ole UNS HAKONE) cos0cccrasoousd0g 0008 Martiella.

AA.—Microconidia usually wanting, or 1—3- or more -septate, reni-

form, comma-shaped, fusoid or falcate :

B.—Macroconidia apedicellate, usually apiculate :

C.—Pionnotes typically present. Comparatively slow-

growing fungi, with thin-walled, indistinctly septate

conidia :

D.—Stroma effuse, immersed, or matted and coremium-

like, but not stilboid. Conidia subulate.......... Eupionnotes.

DD.—Stroma cone-shaped to club-shaped, hard and horny

when dry ; or sessile, flat, loose, floceose. Conidia

subulate, like those of the Rosewm section :

E.—Entomogenous fungi (on scale insects)...... Pseudomicrocera.

EE.—Mycogenous fungi (on old Sphaeriaceae)..... . Submicrocera.

CC.—Pionnotes scanty or wanting, seldom abundant. Com-

paratively quick-growing fungi...................... Arachnites.

334

BB.—Macroconidia more or less pedicellate :

C.—Terminal chlamydospores absent :

D.—Intercalary chlamydospores wanting; conidia in

sporodochia and pionnotes salmon or orange ;

stroma effuse or erumpent, stilboid :

E.—Conidia thick-walled, sub-cylindrical, curved,

abruptly constricted, curved and rostrate at

the apex (as in Martiella). Comparatively

slow-growing, entomogenous (on scale insects)

ANG! WAVCORBAOMS WOH, o's csozoosescoovsnce Macroconia.

EE.—Conidia with thin walls and delicate, rather

indistinct septa. Stroma often sclerotially

erumpent, dark blue or pale :

F.—Macroconidia subeylindrical and equila-

terally curved in the middle: abruptly

constricted, recurved and rostrate at the

apex. Mycelium white to pink. Stroma

pale, sometimes violet, olivaceous, green.

Form of conidia, colour, stroma and _ sclero-

HEY Jb JHOGOHIS. 6060565040 sagdocdhoovcdoucne LGA.

FF.—Macroconidia subeylindrical in the middle,

curvature often somewhat inequilateral,

long, subulate, faleate, tapermg to both

ends. Mycelium and stroma variable in

colour, pink, purple, yellow or pale.

Conidia mostly orange-red......:.........- Roseum.

DD.—Interealary chlamydospores present.

E.—Sporodochia. usually wanting. Free conidia

scattered in floccose mycelium, fusoid ; macro-

conidia either fusiform-lanceolate, tapering to

both ends and not pedicellate; or faleate and

pedicellate. Colour pale, varying between

Gibbostimandultos cian nee nr eee eee cree Arthrosporvella.

EE.—Sporodochia present. Sclerotia dark blue,

brownish-white or wanting.

F.—Macroconidia thin-walled, fusiform-falcate

with parabolic or hyperbolic curvature,

inequilateral ; apical cell prolonged, fili-

form to flagelliform ; base definitely pedi-

cellate ; conidia in mass pale or pink to

salmon ochre; stroma brown, seldom

Carmine) or: yellow sherri eens eels Gibbosum.

CC.—Intercalary and sometimes terminal chlamydospores

present :

FF.—Macroconidia with comparatively thick

walls and septa, fusiform-falcate, tapering

to both ends, inequilaterally curved ;

apical cell sometimes constricted, almost

rostrate, sometimes truncate or elongated;

base pedicellate ; conidia in mass ochre,

pink, salmon or orange. Stroma _ pale,

pink, carmine, purple, yellow, brown,

blue; rarely pale and concolorous.

335

Sclerotia dark blue, brown, ochre or

wanting. Mycelium white, pink or yel-

lowish, sometimes flecked with blue.... Discolor.

CCC.—Terminal chlamydospores present, intercalary wanting.

Stroma effuse, floccose to gelatinous ; the long mycelial

strands sometimes forming a coremium-like body, but

not producing tubercularia-like sporodochia. Conidia

scattered, in false heads, not forming extensive muci-

laginous layers; wedge-shaped, with thick walls and

septa, cream-colour to brownish-white............... Ventricosum.

Section MACROCONIA Wr.

Wollenweber, Fusarium-Monographie pp. 274-281, 1931. Wollenweber and Reinking, Die Fusarien

pp. 27-28, 1935.

Conidia produced in salmon-orange pionnotes and sporodochia ; comparatively large,

rather thick-walled, mostly 3-5—9- septate, sub-eylindrical, abruptly constricted or curved

and rostrate at the apex, more or less pedicellate at the base. In some species a few small,

scattered, 1—2-celled conidia occur. Chlamydospores absent or scarce. Sclerotia present

or absent. The formation of the stroma is variable and depends on climatic conditions

and on the mode of life of the fungus. It may be limited in extent or effuse, smooth or

wrinkled, or, when aerial mycelium is abundant, filamentous and loosely interwoven. Later

it may be compact and assume various forms; occasionally also it may be delicate and

evanescent or consist of hyphae penetrating the substratum, and then the conidia appear

to be borne directly on the surface of the substratum.

Entomogenous and mycogenous species parasitic on scale insects and on other fungi,

chiefly dark-coloured Sphaeriaceae. The entomogenous species, Fusarium coccophilum, is

described by Wollenweber and Reinking (61) as the conidial form of Nectria coccophila

(Tul.) Wr.

Fusarium coccophilum (Desm.) Wr. et Rkg.

Wollenweber and Reinking, Die Fusarien pp. 34-36, 1935. Wollenweber, Fus. aut. del. 344-348

351, 614, 861-868, 1124-1126.

Syn. Microcera coccophila Desm.; Tubercularia coccophila Bon. ;

Micrccera aurantiicola Petch ; M. coccidophthora Petch ;

Fusarium (Fusisporium) coccinellum (Kalch.) Thuem. ;

Atractium flammeum Berk. et Rav. ; Stilbum flammeum Tul. ;

Fusarium baccharidicola P. Henn. ; F. callosporum Pat. ;

F. cataleptum Cke. et Harkn.; F. nectriae-turraeae P. Henn. ;

Pionnotes pseudonectria Speg.; Microcera pluriseptata Cke. et Mass.

Stroma sometimes minute, evanescent, sometimes more or less plectenchymatous,

hard when dry and becoming cartilaginous, or composed of loosely interwoven hyphae ;

occasionally wanting.

Conidia either abstricted from hyphae lying on the substratum, or forming an extensive

pionnotal layer, or produced in sporodochia. The latter are composed of densely fasciculate

conidiophores arising from a plectenchymatous base, or from a stilboid body composed

of ascending hyphae united into a coremium. Sporodochia flattened-globose, conical,

clavate or cylindrical, 0:5-2-5 x 0+25-0-6 mm., peach red to scarlet, fading to flesh ochre

and rufous with age, surrounded at the base by a thin plectenchymatous sheath, which is

continuous below and terminates above in irregular poimts. Conidiophores branching

irregularly, seldom with opposite branches in pairs. (Plates "II and IV.)

Fie. 1.

Fusarium coccophilum (Desm.) Wr. et Rkg.; macroconidia from (a) red scale on Citrus (M.H. 12174,)

(6) pernicious scale on Pyrus (M.H. 21932), and (c) pionnotes of culture on oat agar, 14 days old.

Conidia thick-walled, sub-eylindrical to faleate, tapering to both ends; apical cell

somewhat abruptly bent and constricted ; base conical to subpedicellate. Conidia 5—7- or

7-9-septate, less frequently 3-4- and 10-12-septate; in mass salmon-orange to orange-

scarlet ; occasionally the cross walls are not easily perceptible.

S-SCP label: . tan ec staen sccm 35-78 xX 4-5.

D=SEPLAbE act stoiele em me 50-95 x 4-5-7.

TSOpuahe® cree mnie Sea eee: 50-112 x 4-5-8

O-ge pater. occ. steers 60-117-5 x 5-8

IJ-12-septate..-5....-...... 98-142 x 5-8.

Chlamydospores and sclerotia lacking.

337

Fusarium coccophilum has been shown by Wollenweber and Reinking (61) to be the

conidial stage of Nectria coccophila. The ascus form has not yet been observed in South

Africa, although the conidial stage is extremely common. A description of Nectria coccophila

follows :—

Nectria coccophila (Tul.) Wr. et Rkg.

Wollenweber and Reinking, Die Fusarien pp. 34-36, 1935. Wollenweber, Fus. aut. del. 679-682.

Syn. Sphaerostilbe coccophila Tul.

Nectria aurantiicola B. et Br.; Sphaerostilbe auranticola (B. et Br.) Petch.

Corallomyces aurantiicola (Berk. et Br.) Hohn.

Nectria coccidophthora Zimm.; Sphaerostilbe coccidophthora (Zimm.) Petch.

Nectria coccidophthora Zimm. v. aurantvicola (Berk. et Br.) Rehm.

N. Balansae Speg.; N. congoensis Syd.

N. laeticolor Berk. et Curt. ; Sphaerostilbe flammea Tyl.

N. Aglaothele Berk. et Curt.; NV. colletiae Rehm; N. muscivora Berk.

N. Passeriniana Cke.; Nectria coccicida Speg. (in MS.).

N. subcoccinea Sacc. et Ell.; WN. subfurfuracea P. Henn. et E. Nym.

N. Turraeae P. Henn.

Perithecia scattered or in groups, arising from a plectenchymatous or floccose stroma,

0-2-0-4 mm. diam. (average 0-28 x 0:25 mm.), orange red to dark red, darker at the

ostiole, smooth or somewhat rough, almost spherical or conical, cupulate or laterally de-

pressed. Ostiole papillate or not prominent. Asci cylindrical to club-shaped, with or without

pedicel, 8-, seldom 4-spored, 70-130 6-10, usually monostichous. Paraphyses linear,

delicate, evanescent. Spores oval to ellipsoid, at first hyaline, then yellowish pink ; later

thick-walled and finely verrucose, brown, |-septate, 9-26 « 4-11, mostly 12-18 « 6-8;

the larger spores from 4-celled asci.

The Fusarium stage has been found on the following hosts in South Africa :

Hab. Aspidiotus furcillae Brain on bark of Acacia horrida, Somerset East, in silvis,

1876, leg. MacOwan (de Thuemen Myc. Universalis 782) M.H. 21956: pr. pedem montium

Boschberg, prope Somerset Hast, Nov., 1875, Fungi MacOwaniana 1059, M.H. 20913.

Aspidiotus perniciosus Comst. (pernicious scale), on twigs of Pyrus communis, Chasedene,

Maritzbure, Natal (van der Vyver), M.H. 21932.

Aspidiotus rapax Comst. (greedy scale) on twigs of Ribes sp., Haenertsburg, N. Trans-

vaal, July 1911 (Doidge) M.H. 1684.

Chrysomphalus aurantii Mask. (red scale) on twigs and leaves of Citrus spp., Maritzburg,

Natal, May, 1919 (Kelly) M.H. 12174 ; without locality, April 1929, M.-H. 25438 ; Duivels-

kloof, N. Transvaal, August 1911 (Doidge) M.H. 1845; Elim, N. Transvaal, January 1925

(Doidge) M.H. 20344; Bredasdorp, Cape (Turner) M.H. 20602; Alkmaar, EK. Transvaal,

June 1924 (Turner) M.H. 18191 ; Richmond, Natal, March 1922, (Tedder) M.H. 15479 ;

Mt. Silinda, 8. Rhodesia, August 1931 (Lounsbury) M.H. 25973; Maritzburg, May 1932

(van der Vyver) M.H. 26322; Politsi, N. Transvaal, Sept. 1934 (Wager) M.H. 27689 ;

Elim, April 1935 (Nyenhuis) M.H. 27561; Forest Hill, Tzaneen, Aug. 1932 (Turner) M.H.

26568.

Chrysomphalus aurantii Mask. (red scale) on twigs of Rosa sp., Ravenshill, N. Transvaal

(Eland) M.H. 25932; Maritzburg, Nov. 1933 (Fouché) M.H. 27282.

Chionaspis sp., Victoria Falls, Rhodesia (Lounsbury) Aug. 1931, M.H. 25974.

Lepidosaphes Gloveri (mussel scale) on twigs of Citrus spp., Chase Valley, nr. Maritz-

burg, Feb. 1932 (van der Plank) ; Port St. Johns, Pondoland (Fraser) M.H. 26323.

Scale undet. on Plectronia sp., Cape Province, Oct. 1906 (Lounsbury) M.H. 193.

This species occurs on scale insects in tropical and subtropical regions in all parts of

the world. The South African fungus was first collected by MacOwan in 1876 on Aspidiotus

furcillae on Acacia, and was described as Fusisporium coccinellum Kalch., and then as

Fusarium coccinellum (Kalch.) Thuemen in Fungi austro-africani, Flora 1876, p. 426 (Wollen-

weber Fus. aut. del. 344 and 861). It was also collected by Medley Wood (Wood No. 157)

338

on a scale insect on an unknown tree, Port Natal, and was identified as Fusarium bacchari-

dicola P. Henn. (Wollenweber Fus. aut del. 865), which is now regarded as a synonym for

F. coccophilum.

Fusarium coccophilum occurs on a number of different scale insects in the more humid,

sub-tropical areas; it is very variable in the size and septation of the conidia. Fifteen

collections were examined in detail; 1m seven of these 7—9-septate conidia predominated,

in five (including MacOwan’s collection) 5—7-septate conidia were most frequent, and in

two of the collections most of the conidia were 3-5-septate. In culture there is an even

wider range of size and septation than in conidia developing on the natural host.

For a full discussion of the nomenclature and synonymy of Fusarium coccophilum,

see Wollenweber and Reinking (61) and Petch (33, 34).

Growth on Standard Media.

Oat agar: Growth slow, barely covering the surface of the slant in 7 days; growth

in substratum colourless at first, but after 4 weeks tinged vinaceous russet, especially near

base of slant. Pionnotes formed on older part of growth, and were well developed after

14 days; they were at first flesh colour, then salmon colour.

Hard potato agar: Growth slow ; no aerial mycelium, growth in substratum colourless.

Sporodochia formed in small groups, bitter-sweet-pink.

Standard synthetic agar plus starch: Growth slow, and mycelium almost covered with

pionnotes which were well developed after 4 weeks ; pionnotes at first salmon orange then

bitter-sweet orange. After 8 weeks a group of sporodochia had developed at the base of

the slant.

Potato agar plus 5 per cent. dextrose : Growth very slow, and conidia very freely produced.

Pionnotes bitter-sweet orange to flame scarlet; mycelium fine, white, only visible as a

ringe round the pionnotes.

Potato plug : Growth very slow, and consisting of a salmon-colour cushion-like stroma

about 10 mm. diameter, with a very little fine, white mycelium in places on the surface.

In four weeks the stroma became very much folded and wrinkled, and the colour faded

to light pinkish cinnamon.

Melilotus stem: Growth very slow, forming a small cushion 5-10 mm. in diameter,

bitter-sweet pink underneath, overlaid with a little fine, white mycelium.

Bean pod: Growth resembling that on potato, but less vigorous, and aerial mycelium

very scanty.

Rice: Growth very slow, and penetrating very little into the medium; growth on

substratum bitter-sweet pink. After 4 weeks the rice grains were covered with ‘conidia.

Measurements of Conidia.

A.—From sporodochia on red scale and other scale insects ; summary of measurements

from 15 collections recorded above. Conidia in some collections chiefly 5—7- septate, and

in others 7—9-septate ; in two cases the majority were 3—5-septate.

IMEEM. ooo scour aebvaces 100 x 6.

IOHSTEMT no oocdhecagoessae 95-107-5 X 5-6: 25.

Q-septate sy caniricr sera einen 60-117-5 x 5-6: 25.

S-sep tater: Sree me es ren 60-112-5 x 5-6-5.

T-septate ya. Maken warn 57-107-5 x 4:5-6°5.

G=septateni seen eee 60-92:5 x 4-6.

Szseptater: aaaeeh sears 50-95 x 4-5-6.

A=Sepitabe.yc. Teron Rott 52:5-82-5 x 4:0-5.

SESE PLALEn nu mece ror te ee chs ee ATO 1190) X 2-0-0) Os

339

B.—From sporodochia on pernicious scale (M.H. 21932); conidia in this collection

were mostly 5—-7-septate, more rarely 9-septate, a few 3-septate. Exact computations

were not made.

VAIS DMUSeo goes loo ono emcee 725-825 X 5=6-25

WESC PUAber cian aia Hah, Sel ka aoe 57-5-80 < 5-6-25.

DERGUAUIEE AG no 0.0 8o00 Fo O66 55-10 x 5-6.

SSC Plate ve sa ciro ce eemeueeo Ber 47-5-52-5 x 5-6.

C.—From culture derived from conidia of sporodochia on pernicious scale (M.H. 21932).

Standard synthetic agar plus starch, culture 2 weeks old, conidia from pionnotes :—

iieseptaitensenisisencrr at iepericentieee sade LOOSMO 5 =515.

1O-septate str: sl. meat Di RP BEC SS Jose, Seer 100-110 x 5-5-5.

QHSS NERS. cones done on acs 76 neg rh Mee nba REN 90-110 x 4:5-5:5

S-septate anu scien 9 ey eek IRL 90-100 x 5-5-5.

7-septate. . te. Stier, PORE Teck ees SME ee 85-100 x 5-6.

Oat agar, cultures? anaes old, conidia from pionnotes :—

i2=septateseery-sscet: 2. ZaOCTA COG! Ry mney yo 100-105 x

UilEseptateweean see. ARP ett inten en Ie os Tio UlONX< a

NOHSOINE, o6 006 occa00% 10 Be ROR IG ieee cee 95-102°5 x 5-5-3.

SHINEWE cso 0000850086 SOMES Me aD caceneaee gine 90-112-5 x 5

SHICTOUEIUS. coc dan aaneaso PH eer Aa Tot 80-105 x 5.

(-septate ep eck.) oss LOGE Bahn ante dis 77-5-105-5 x 5-9-3.

G-seplatenee eee ser Hiatt ss ANNs aPC (D8) XX. By

OCLs oo no doce dbox Bg reset b a issn acts atte = 65-87:5 x 9.

4A-septate. soen is ceee ss - IVAROSS chiens ssa tee uae: 65 x 5.

Hard potato agar, culture 2 weeks old, conidia from pionnotes :—

ESeptateeserane er cee Nae weary sree AO orL 105 x 5:3.

OSC p bate eevee ene cin ka ceomcta At iata cust ote ae 115 X 5-3.

O-Sepuaten ames ria nae 5 (DOR MCC Geo Hamels dg oe 92-5-105 x 95-3.

S-septateraas ee ciees a: 1G era Eats cence aces 82-5-102-5 x 9.

(ESepuaiie ase e ce eee 0," ec PNT anna eee an UNE 75-100 x 5.

G:septater. cease -ciauteee 2 incu rem s Peet ey sk. 75-80 x 5.

DIODES. occ ccekoodace US ieee Peeireges het vetcee nai 50-85 x 5.

AN OUAUE coo conse oos VAR Cif ctens opens By Mek ne eso 60 x 5

Section SPICARIOIDES.

Wollenweber, Sherbakoft, Reinking, Johann and Bailey, in Jour. Agric. Res. 30 : 841, 1925. Reinking

and Wollenweber, Phil. Jour. Sci. 32 : 169, 1927. Wollenweber, Fusarium-Monographie 311, 1931. Wollen-

weber and Reinking, Die Fusarien, 36, 1935.

Microconidia delicate, ovoid, developing in chains and false heads, and later scattered

in the mycelium. Macroconidia pluriseptate, thick-walled, cylindrical, moderately curved,

constricted and rostrate at the apex ; base pedicellate ; borne in sporodochia and pionnotes,

white to cream and ochre in mass. Stroma golden yellow to carmine red. Aerial mycelium

white or tinged with the colour of the stroma. Sclerotia sometimes develope ; they are

convex, rugulose or stilboid. Chlamydo-spores absent.

Fusarium decemeellulare Brick.

Brick, C. Jahresber. Ver. f. Angew. Bot. 6: 227 (1908). .Wollenweber, Fusarium-Monographie 311,

1931; Fus. aut. del. 353, 869, 870: Wollenweber and Reinking, Die Fusarien, 36-38, 1935.

Syn. Spicaria colorans van-Hall-de Jonge.

Fusarium spicariae-colorantis (van Hall-de Jonge) Sacc. et Trott.

Fusarium theobromae Lutz (nec App. et Strk.).

340

;

c) ee . 7)

a seat us {

SS Og i \

mn y <x

} ~ y ;

t X é Me}

Past ty) ‘

\) |

x 2) e eo

ry COE »

we. /~ ¥ ~

LEFT ay Os ; Eset \

¢ ; EN) ane \

NE a

2). ~ ~/ —

& ;

| |

of >

f se

‘

‘ 1

L Ki?

eax

v ANG

Fic. 2.

Fusarium decemcellulare Brick; (a) macroconida from sporodochia of culture on Melilotus stem, 4

weeks old; (b) microconidia from mycelium on plain agar, 10 days old.

Microconidia ovoid, 1—2-celled, in chains or false heads, produced on more or less

branched conidiophores in the aerial mycelium ; these form a powdery layer on the mycelium

and are easily seattered. Macroconidia formed in sporodochia and pionnotes, which are

at first white, then cream, brownish-white or ochraceous. Macroconidia large, cylindrical,

somewhat curved especially near the ends, rostrate at the apex, pedicellate at the base,

usually 5-9-, less frequently 3-4- or 10—12-septate.

341

@:septaiter ne yome anes st Dole 2-40,,mostlysn-. 22 1-9) 13-4

Icseptatemintis veces 10-28 x 2-5, mostly... oa.. 12-20 4-4-5.

S-septates aye vant: 20-67 x 3-5-6, mostly...... 25-42 x 4-5-5-5

H-septatenag. suntan ae 42-72 x 4-5-8, mostly...... 53-64 « 4-7-6-5

feseptatencas aascse is 60-95 « 4-5-8, mostly...... Doe) << Bole

Q-septate: i... 56.2.0 4. 68-114 x 4-5-8-5, mostly... 75-97 x 5-8.

Hilzseptater: ss. ieee T3V3l x7 5—9, mostly. i.) 80-111 x 5-3-7-5.

I2=M3=septate. ni ai = 90-130 x 6-9, rare.

Stroma yellow or carmine red, covered with white to pink, aerial mycelium. Plec-

tenchyma sclerotially erumpent, and forming sclerotial growths which are convex, rugulose

or stilboid. From these the ascus form developes later. Chlamydospores absent.

Hab. Epichloé Zahlbruckneriana on Sporobolus indicus, associated with Fusariwm

ciliatum on the stroma, Acton Homes, nr. Bergville, Natal, March 1931 (L. A. Doidge).

Citrus sinensis Osbeck, from discoloured centre of Valencia orange frcm Zebediela,

Transvaal, after 18 weeks in storage, 1934. :

Fusarium decemcellulare is the conidial form of Calonectria rigidiuscula ; the ascus

stage has not yet been observed occurring naturally as a saprophyte in South Africa, but

developed in cultures isolated from Epichloé, and sent to Dr. Wollenweber for identification.

It may be characterised briefly as follows :—

Calonectria rigidiuscula (Berk. et Broome) Sacc.

Saccardo, Michelia 1 : 313 (1878). Wollenweber, Fusarium-Monographie 312-314, 1931; Fus. aut.

del. 800-802. Wollenweber and Reinking, Die Fusarien, 37-38, 1934. (For complete bibliography and

synonymy see last-named publications.)

Perithecia scattered or in groups, ovoid to subconical, cream-coloured, yellow-brown

when dry, 0-27 to 0-6 x 0-18-0-4 mm. (average 0-36 x 0-28 mm.); asci 4-spored,

seldom 2- or 8-spored. Spores fusoid, slightly curved, obtrusely conical at both ends,

obliquely striate, brownish-white in mass, mostly 3-septate, seldom (up to 14 per cent.)

4-6-septate, very rarely 1—2- or 7-septate.

ISSN coc oacecs UGAMS S Coos saccon BOVIS IN) XC or

SHMUMOc.ccocacccs NSB SK Docs ancsds Mostly 19 x 5-5-7-5, the larger

spores in 2-spored asci (31 7 )

and the smaller in 8-spored asci

(15-26 x 5-5-7-5).

D-septate.... 32:24: HIBS < (Beas ssboac Average 28 x 7:6.

T-septate.......... 25-45 x 7-9.

Stroma rough or flat, pale, golden yellow or brown, sometimes evanescent. Mycelium

at first floccose, white or pink, then drying up and disappearing.

Hab.—In culture derived from mycelial mat surrounding stroma of Hpichloé Zahl-

bruckneriana on Sporobolus indicus, cult. Wollenweber, M.H. 25897 B.

This species oceurs on dry, decaying stems, fruits, ete., on various hosts (Theobroma,

Hibiscus, Melia, Anona, Ficus) in tropical and sub-tropical regions of America, Asia and

Africa.

Growth on Standard Media.

Oat agar: Mycelium at first woolly, short, white tinged rose pink, later becoming

closely matted ; it may then become spinel red and Indian lake in colour. Growth in

substratum amaranth purple. After 3-4 weeks, dense masses of buff yellow pionnotes,

or a few separate sporodochia may be produced.

Hard potato agar: In cultures studied, growth on this medium was not vigorous,

and consisted of a little white, tufted aerial mycelium on a colourless substratum. A few

sporodochia developed after 8 weeks. Reinking and Wollenweber (39) however, record

a more vigorous growth with a rather thin, matted mycelium in cultures 12 days old; this

was pomegranate purple, rose red and rose pink. Older cultures (45-90 days) had a more

342

matted mycelium with irregular, stromatic tufts, and the colour was pomegranate purple

with a Bordeaux ring at the base. Sporodochia and pionnotes produced in large masses

were warm buff and light orange yellow.

Standard synthetic agar plus starch: Aerial mycelium scanty ; growth on substratum

amaranth purple in the centre and olive ochre along the edges of the growth.

Potato agar plus 5 per cent. dextrose: Mycelium matted, pomegranate purple and olive

ochre. Growth in substratum Bordeaux.

Potato plug: In cultures 14 days old, the plug was covered with tufted mycelium,

which was white to rose colour. Spore masses were beginning to form. Groups of sporo-

dochia, developing vigorously after 4 weeks, were light ochraceous buff ; they often coalesced

into large pionnotal masses. Individual sporodochia were sometimes columnar in shape.

Melilotus stem: After 14 days, stems were covered with a short, felt-like mycelium,

which was white to tyrian pink, growth on the water at the base of the tube being amaranth

purple. After 8 weeks, several groups of ochraceous buff sporodochia had developed.

Bean pod: Growth very similar to that on Melilotus stem, but spore masses produced

less freely.

Rice: Growth Naples yellow to primuline yellow, in cultures 14 days old. In older

cultures, the mycelium is more or less powdery, and yellow ochre to ochraceous orange in

colour. Spore masses were sometimes produced.

Measurements of Conidia.

Potato plug, culture 4 weeks old, conidia from sporodochia :—

S-Sep balou erie eirn ewan tte acer ee 82-5 X 9.

(UFSANEGosccosdcss 47) OOP CBM scbocnv0cs 70-87-5 x 5-5-3

O-septatein vaecw.a CE NS eet 52-590: X 9-625

H-septate st +. lave et oil 3 Ree ee eee tem 55-12-5 x. 5-5-6

4-Sepvaes se. 7A AP LPS CS. DID Xs

O-Septabennnnee er TOO Gis eae ae. tae Gas} XC Gate Ds

Melilotus stem, culture 4 weeks old, conidia from sporodochia :—

Q=septate.....-4.5- ILS CBMs oad ooo a 06 70-85 X 5-3.

OAIS|OUENOS oo oc a4 6's Ils) Sash BP Spee 67°5-87-5 « 5-5-6.

T-septate.......... BO) ai, | asecr reer 62-5-92-5 x 5-5-6.

Gséptater.. 50) genet ey cee BT-5-82-5 x 5-5-6.

5-septate.......... PE Na | MeP eI bo 47-5-67-°5 X 5-5-6.

4-septate:......... Ding As A ae 42-5-57-5 x 4-75.

SSeptave yn. .: sce w ale aa Am aero rs isa 22-5-52-5 X 3:°0-4°5,

ORONO. coocccccs Wt Sole Tee ONT 5=8 x 35-45.

Reinking and Wollenweber (39) give the following figures for the average measurements

of conidia on several media :—

OM daodeoss TANCE. ca so0nc08 56-73 X 5:5-T7-20.

USED a0 605000 27 a MERE een recite 60-83 5:5-7-25.

Sse plate: cancers Ueane ee uname element 59-86 x 6:25-7:20.

9-septate.......... 1a ea ee aa sae. (DSO < Oo29=5° 0D

In the South African strains studied, there was a smaller percentage of conidia with

6—9 septations, and the majority were more slender.

Section SUBMICROCERA.

Wollenweber, Fusarium-Monographie, 281, 1931. Wollenweber and Reinking, Die Fusarien, 38,

1935.

Comparatively slow-growing fungi, chiefly mycogenous ; they are found on dry branches

and grasses, chiefly as parasites on other fungi (Sphaeriaceales, Hypocreales), or on the

343

thallus of lichens growin, on trees. Conidia subulate, slender, thin-walled, pointed at both

ends, never pedicillate, indistinctly septate. They occur in reddish, mucilaginous balls,

on loose, cotton-wool-like mycelial tufts, or on a stilboid stroma. The latter may be 2 mm.

high, and consists of hyphae which are loosely interwoven or coremium-like ; the base is

whitish and soft, and looks as if powdered with flour ; often it is thickened bulbously above,

and bears spherical balls of plectenchyma, on the surface of which are produced radiating

conidiophores in dense clusters. These abstrict conidia in great numbers, and they collect

in small drops or mucilaginous balls of an orange red colour. When dry, they form a con-

tinuous, adherent, resinous, brick red, crust. Chlamydospores wanting.

Only two species known, Fusarium ciliatum Lk., the type species, of which the ascus

form is Calonectria decora (Wallr.) Sace., and F. cerasi Roll. et Ferry with somewhat smaller,

paler, more curved conidia. The latter species occurs in Europe and North America on

dead branches of Prunus cerasus, Ceraphora, Alnus and Corylus.

Fusarium ciliatum Link.

Link, Spee. Plant. IL: 105 (1825); Wollenweber and Reinking, Die Fusarien 38-39, 1935; Wollen-

weber, Fus. aut. del. 54, 437, 438, 872, 1128.

Syn. Attractium ciliatum Lk. pr. p.; Microcera ciliata (Lk.) Wr.

Fusarium ciliatum Lk. v. majus Wr.

Fusarium parasiticum West. ; F..peltigera West.

F. scolecoides Sace. et Ell. ; F. elongatum Cke.

F. filisporum (Cke.) Sace. ; Fusisporium filisporum Cke.

Microcera massariae Sace.

Conidia subulate, small, slender, delicate, straight or curved, tapering to both ends,

apedicellate and more or less truncate at the base, in mass bittersweet pink to flame scarlet,

fading to orange rufous and rufous when dry. Conidia 3-7- (mostly 5-7-) septate, seldom

up to 10-septate or less than 3-septate ; 5-septate 40-90 x 1-9-2-5-8, T-septate 55-90 x

2-25-2-5. Conidia formed on floccose, loose, mycelial tufts, and adhering in mucilaginous

balls, or borne on a coremium-like columnar, stilboid stroma. The conidiophores are

simple or branched, and develope on the aerial mycelium, or in close clusters on the swollen,

confluent, spherical balls of plectenchyma, which are produced on the top of the stalked

stroma. Chlamydospores wanting.

Hab.—On the ascigerous stroma of E'pichloé Zahlbruckneriana on Sporobolus indicus and

Eragrostis plana, Thornville Junction, Natal, March 1910 (Doidge) M.H. 865; Mooi River,

Natal, March 1917 (Mogg) M.H. 10063; Cramond, Natal, April 1911 (Pole-Kvans) M.H,

1369 ; Fairy Glen, Pretoria, March 1923 (Lounsbury) M.H. 17651; Acton Homes, Natal.

February 1931 (L. A. Doidge) M.H. 25897; Hopevale, nr. Donnybrook, Natal (Morgan)

M.H. 27749.

Fusarium ciliatum is found very commonly on the stroma of Epichloé, especially in

Natal. The organism was identified by Dr. Wollenweber from a culture made from the

specimen M.H. 25897. In nature, small patches of white, cottony to arachnoid mycelium

appear on the surface of the stroma, rapidly becoming tinged with pink as the conidia

develope. These patches consist of tangled hyphae which soon form at their tips a con-

tinuous, plectenchymatous layer, which in turn gives rise to very numerous, fasciculate

conidiophores. The small patches of mycelium increase in size, coalesce, and often com-

pletely clothe the stroma of the host with a waxy or resinous, conidial layer. Occasionally

the Hpichloé stroma developes partially enveloped in the sheath of the grass leaf; in such

a case, there is a layer of white mycelium between the stroma and the sheath, and the

pionnotal layer developes on the edge of the sheath.

The fungus occurs in Europe and North America chiefly on Massaria and other fungi,

on decaying branches of Acer, Alnus, Ulmus, Fraxinus and Robima. It is also found on

lichens (Peltigera) (see Wollenweber, Fus. aut. del. 54, 437, 438, 872).

344

Fig. 3.

Fusarium ciliatum Link. ; Macroconidia from (a) the conidial mat on the stroma of Hpichloé, and (b)

sporodochia of 8 weeks old culture on oat agar.

F, ciliatum is the conidial form of Calonectria decora (Wallr.) Sacc. the ascus stage has

not yet been observed in South Africa. It may be briefly described as follows :—

Calonectria decora (Wallr.) Sacc.

Saccardo, Michelia 1: 310, 1872; Syll. Fung. IL: 543. Wollenweber and Reinking, Die Fusarien,

39, 1935. Wollenweber, Fus. aut. del. 54.

Syn. Sphaeria decora Wallr.; Nectria decora (Wallr.) Fuck.

Nectria Massariae Pass. in herb. ; Calonectria Massariae (Pass.) Sace.

Calonectria Dearnessvi Ell. et Ey. ; C. diminuta (Berk-.) Berl. et Vogl.

Nectria diploa Berk. et Curt. v. diminuta Berk.

Creonectria diploa Seay. non Berk. et Curt.

Calonectria agnina (Rob.) Sace. ; C. pyrrochlora Sace.

345

Perithecia covered with whitish or rosy-white hyphal threads, ovoid, 0:2-0-27 x

Q-15-0-24 mm. (average 0-24 x 0-2), fleshy, light orange to wax yellow, later becoming

paler, with a darker, definitely orange-red papilla, 70 broad, leaving free an ostiole formed

of radiating hyphae. Asci club-shaped, 70-96 12-18, 8-spored, sessile. Spores more

or less distichous, hyaline, narrow-ellipsoid, tapering scmewhat to both ends, straight or

slightly curved, at first smooth, then minutely verruccse, 1—3-septate, 16-35 « 5-8, (av.

23 x 5:5). Paraphyses filiform.

Conidia and hosts as described above for Fusarium ciliatum.

Growth on Standard Media.

Oat agar: Aerial mycelium fairly abundant, white, cottony, tufted. After 14 days,

there was a tinge of congo pink in the plectenchymatous layer on the substratum, and

conidial masses were beginning to appear. Conidial masses developed very slowly, elevated

on mycelial tufts, at first bittersweet pink, then grenadine ; they were up to 2 mm. diameter.

Hard potato agar : Growth rather slow, but otherwise similar to that on oat agar.

Potato agar plus 5 per cent. dextrose: Aerial mycelium flesh pink, cottony or mealy

after 14 days. Growth in substratum becoming wrinkled and felt-like and tinged buff

pink.

Potato plug: Growth advanced slowly and was not vigorous. Mycelium cottony,.

white, tinged vinaceous pink.

Melilotus stem: Growth slow, mycelium white, cottony.

Bean pod: After 7 days, about one-third of the pod was covered with white, cottony

mycelium, and pods were entirely covered in 4 weeks.

Rice: Growth was more vigorous than on the last three media mentioned ; mycelium

at first white or tinged with pink, and after 14 days pale flesh colour to flesh colour.

Measurements of conidia.

From pionnotes on stroma of Epichloé, conidia mostly 5—7-septate.

SHIM canoscoe. COME XA uews

USED sos 08 005¢ 50-15 & 2-25-2-5.

Oat agar, culture 4 weeks old, conidia from sporodochia.

Conidia most 7-septate.

T-septate.......... 45-72-5 x 2-2°5.

6-septate.......... 66-75 x 2-3. Only occasionally up to 3 pw thick.

5-septate.......... 55-61-25 x 2-3. Only occasionally up to 3 pw thick.

4-septate.......... 45-52-5 x 2.

3-septate.......... 20-40 x 2-2-5.

In the European specimens, conidia are mostly 5-septate, 50-90 « 2-25-2-75.

Section SPOROTRICHIELLA.

Wollenweber, H. W., apud Lewis in Maine Agr. Exp. Sta. Bull. 219 : 256, 1913. Sherbakoff, N.Y;

(Cornell) Agr. Exp. Sta. Memoir 6: 183, 1915. Wollenweber and Reinking, Phytopath. 15: 156, 1925.

Die Fusarien, 45-46, 1935; Reinking and Wollenweber, Phil. Jour. Sci. 32: 116, 1927.

Microconidia 1-2-celled, spherical-ovoid, lemon or pear-shaped and also fusoid-ellipsoid.

In the species Fusarium poae and F. chlamydosporum, only a few faleate macroconidia are

found scattered in the mycelium ; in the other species they are more or less abundant and

are produced in sporodochia and pionnotes. Chlamydospores usually abundant. In this

respect the section differs from the Rosewm-Fusaria, which have no chlamydospores, but

somewhat similar macroconidia. It resembles the Rosewm section in the colour of the stroma,

346

which is typically carmine to purple red or ochre yellow. From the section Arthrosporiella

it differs in the occurrence of spherical microconidia and of sporodochial and pionnotal

conidial masses.

Fusarium chlamydosporum Wr. et Rke.

Wollenweber and Reinking, Phytopath. 15: 156, 1925; Die Fusarien, 47-48, 1935. Wollenweber,

Fus. aut. del. 883. Reinking and Wollenweber, Phil. Jour. Sci. 32 : 115-116, 1927.

Conidia-bearing mycelium floccose, pale or pink ; growth on substratum plectenchyma-

tous, sometimes forming somewhat verrucose, tubercular, sclerotial bodies; of various

colours, pale, carmine to purple-red, sulphur-yellow, ochre to dark brown. The dark colour

is due to the development in the mycelium of numerous chlamydospores ; these are spherical

to pear-shaped, smooth, rough or spiny, intercalary or terminal, single, in pairs, in chains

or in clusters, 10-16 w in diameter.

QO, ca

‘

Pat

————

—

Fig. 4.

Fusarium chlamydosporum Wr. et Rkg.; (a) conidia from mycelium on synthetic agar plus starch,

culture 10 weeks old; (b) chlamydospores. from culture on plain agar, 6 weeks old.

Microconidia small, fusoid-ellipsoid, not spherical or lemon-shaped, usually 1-celled,

seldom septate; macroconidia scattered, rare, 3-septate, faleate, subpedicellate.

O-septatercias erie tec 4-1] x 2-5-4 (av. 7 X 3:25).

leseptates ins -jists amxciemnae rier 11-16 x 3-4 (av. 13 x 3:5).

S-Sep tater, jess. cians omoboeee eae 27-32 < 3:5-4 (av. 29 x 4).

Hab.—Brachiaria pubifolia Stapf, on ovaries, Nelspruit, E. Transvaal (Liebenburg).

This species has also been observed on dead locusts, Nomadacris septemfasciata, in

South West Africa (61); it was first described from banana stems, and from soil and air

in Central America (39).

347

Growth on Standard Media.

Oat agar: Aerial mycelium fairly abundant, cottony, becoming matted and felt-like,

tinged in places with sea-shell pink and a little yellow ochre ; sometimes becoming brownish

when chlamydospores are abundant. Growth in substratum carmine, with a patch of ochre

to buckthorn brown at the top of the slant, where the medium was dry.

Hard potato agar: Aerial mycelium white, cottony ; growth in substratum colourless.

Reinking and Wollenweber (39) record the development of a cameo pink and spinel red

mycelium on this medium.

Standard synthetic agar plus starch: Aerial mycelium moderate, cottony, white, be-

coming tinged cinnamon buff to clay colour with age. Growth in substratum spinel red

to dahlia carmine.

Potato agar plus 5 per cent. dextrose: Aerial mycelium fairly dense, cottony or matted

and felt-like, white or tinged with the colour of the stroma. Growth in substratum carmine

to ox-blood red, with patches of ochre.

Potato plug: Plugs covered with a cottony mycelium, which was white to bufly

brown ; there were tinges of carmine or pomegranate purple in the substratum.

Melilotus stems: Growth copious, tomentose, white to buffy brown and buckthorn

brown.

Rice: Mycelial growth dense, felt-like on the surface of the medium, at first white.

and rose pink above, and yellow ochre round the rice grains below. After some weeks,

the growth was snuff brown to bistre.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 10 weeks old, conidia from mycelium.

(ASS) OETLES a cre sete eae crc el Gre Seen Oo a neea 5-15 x 2-5-4-5.

UeSepLaten vier ete sciacinin sae atuatanede ah 10-20 x 4-5.

DESC LATCH Ar cies oe Ome es ois cutee Savas = 12-5-22-5 X 3-75-4:5

SSCP UAL Wanner ae coatronl simu omm cue acne 20-35 x 3-5-4-5

A-“SEPUALE Nan tinned) ene ee eee ee Stine pete OB) SX LEOO)

DESC PLALCLE at prem ta dee nie ae aca eis e cc cymale 50 x 4°75

Chlamydospores very abundant, single, in pairs, or in chains and clusters; at first

smooth and colourless, becoming golden brown and verrucose when mature, 9-14 p in

diameter.

Section ROSEUM.

Wollenweber, Phytopath. 3:32, 1913. Reinking and Wollenweber, Phil. Jour. Sci. 32 : 148, 1927

Wollenweber and Reinking, Die Fusarien, 49-53, 1935.

Macroconidia subulate, slender, thin-walled, only weakly refractive, curved to almost

straight, typically of even diameter for a considerable part of their length, tapering gradually

to both ends, pedicellate at the base, orange colour or lighter in mass, brick red or reddish

brown when dry. Macroconidia borne on the aerial mycelium, on the stroma, or in pionnotes

and sporodochia direct on the substratum. When the aerial mycelium is well developed,

they may also be scattered in the mycelium or in false heads. Stroma yellow, ochre, carmine

red or reddish brown. Aerial mycelium white, pink or yellowish. Blue sclerotial stromata

occur occasionally in some species and not at all in others. Chlamydospores wanting.

Ascus stage unknown.

348

Fusarium avenaceum (F'r.) Sacc.

Sace. Syll. Fung. 4: 713, 1886. Wollenweber and Reinking, Die Fusarien 53-55, 1935. Wollenweber,

Fus. aut. del. 127, 128, 130-136, 139-164, 178-194, 560-568, 572-574, 892, 894-899, 1132, 1133.

Syn. Fusisporium avenaceum Fr.; Sarcopodium avenaceum Fr.

Fusarium biforme Sherb.; F. effusum Sherb.

Fusarium herbarum (Cda.) Fr. plus f. 1 and f. 2.Wr., v. avenaceum (Fr.) We.

v. gibberelloides Wr., v. graminum (Cda.) Wr. pr. p.

v. pirinum (Fr.) Wr., v. tubercularioides (Cda.) Wr.

v. viticola (Theum.) Wr.

. heterosporum Nees f. paspali Ell. et Ev. ; Ff. lucidum Sherb.

. metachroum App. et Wr., plus v. minus Sherb.; F. paspdli P. Henn.

. sorghi P. Henn. ; F. subulatum App. et Wr., plus v. brevius Sherb.

. truncatum Sherb. ; I. zeae (West.) Sacc.

For complete synonymy, see Wollenweber and Reinking, loc. cit.

Ry ky

INS vii

a oy ij \ PO he ee ra

Fie. 5.

Fusarium avenaceum (Fr.) Sace. ; Conidia from pionnotes of 6 weeks old cultures on (a) synthetic agar

plus starch and (6 ) oat agar.

Conidia seldom scattered, usually in false heads or balls, or in sporodochia and pionnotes ;

the latter are orange or cinnabar-red to scarlet, becoming darker if drying in a resinous

mass, or becoming lighter and pink if drying in a powdery ‘condition ; conidial masses may

also become tinged ania the colour of the stroma. Stroma yellow, ochre, carmine to red

brown ;_ aerial mycelium white, or tinged with the colour of the stroma. True sclerotia,

or sclerotially erumpent, rugulose stromata rarely found. Sclerotia, when present, single

or in groups, 60-80 jx diam., “dark blue or pale. Conidia mostly 3-5-septate, long, subulate

or filiform, symmetrically arcuate to elliptically curved, or somewhat more curved near

the apex than in the middle; base more or less pedicellate :—

SHMEMosoececsoncs BOIL SK AGRO, coscdoce mostly 30-48 x 2-7-3:9.

H-septates... sas | OO= CO elt — 6 arian rar mostly 45-66 x 3-1-4-4

(SUMS conasbscsse GUR4 S< B45) edu neos mostly 64 x 4-1.

OFSep tae erie eran CHUTE 2D =A cane average 11 x 2-9.

I-septate......¢s50.. W025 K 2-424 ee average 18 x 3.

Conidiophores simple, or with irregular to fasciculate branching ; branches are irregular

or in whorls of 2-4 or rarely 5.

Hab.—Hleusine indica Gaertn. (goose grass), from stems of plant dying from attack

of Helminthosporium sp., Acton Homes, Natal, Jan. 1931 (L A. Doidge).

Puccinia ranulipes Doidge on Asparagus laricinus. Poplars near Wonderboom, Pretoria

dist., 1937 (in teleutosori).

F. Avenaceum is very widely distributed in the temperate zone, and occurs on a wide

range of host plants. It occurs on 150 different genera, including grasses, cereals, crop

plants, ete., also on other fungi, e.g., Meliola, Claviceps, Uredineae.

349

Growth on Standard Media.

Oat agar: Aerial mycelium not abundant in cultures made from conidia, copious if

vrown from a mycelial transfer, fine, floccose, white or tinged with pink. Growth on sub-

stratum at first colourless, in four weeks becoming carmine to ox-blood red. Sporodochia

apricot-buff to salmon orange, fading to ochraceous orange and cimnamon rufous with age.

Hard potato agar : Like oat agar, but growth in substratum colourless, and sporodochia

smaller and less freely produced.

Standard synthetic agar plus starch: Aerial mycelium scanty, white tinged with pink.

erowth in substratum yellowish, then carmine to pomegranate purple. Pionnotes formed

chiefly along the needle track, at first pale flesh colour, then apricot buff to apricot orange.

Potato agar plus 5 per cent. dextrose: Aerial mycelium abundant, tomentose, at first

white, then rose pink. Growth in substratum geranium pink to carmine. No spore masses

were seen.

Potato plug: Plug covered with abundant white mycelium, becoming matted and

felt-like, and tinged with pink where it touched the glass; stroma carmine. A few large

sporodochia developed ; they were apricot buff.

Melilotus stem and bean pod: Growth slow and sparse. Aerial mycelium tomentose,

not abundant, white to ochre. No spore masses were observed.

Rice: Mycelium at first white; aerial mycelium later tinged with pink; growth on

substratum at first flesh pink, then carrot red; rice grains cream to naples yellow, later

becoming wood brown.

Measurements of Conidia.

Oat agar, culture 6 weeks old, conidia from sporodochia :—

H-sepbatene reer 2 er Celt. «Men ae 30-91 < 3:5-4:5

4-septate....:..:.... 6 A GALS tes ene: 29-41 « 3-4.

OG RCPLALCIM iy lc Mil nj aul haetaccte 23-36 x 3-4.

my 0—2-seprate sn monnine: 5 a

Standard synthetic agar plus starch, culture 4 weeks old, conidia from pionnotes :—

Gseptaternan ete s. 1 o Mew .08. Se in) M2 ne:

TAYHMMOoaceaceconss ll jose Cas sgonnsc 40-52-5 <x 3°T5-4-5.

ARSED CALC aien tere ere SLOP ee eee: 32-40 x 3-5-4.

SHGVONMS.0cesccccaga AY) Le PEN ih tne EE a 25-40 « 2°75-4.

O-2-septate.......... Few

The conidia in the only strain studied were smaller than the average for the species.

See measurements given in the general description.

Fusarium avenaceum (Fr.) Sacc. f. 1. Wr. et Rkg.

Wollenweber and Reinking, Die Fusarien, 55, 1935. Wollenweber, Fus. aut. del. 174-176, 571, 890

1134.

Syn. Fusarium arcuatum Berk. et Curt.; F. arcuatum Berk. et Curt. v. majus Wr.

F. anthophilum Wr. (non A. Braun).

F’. Schiedermayeri (Thuem.)Sace. ; Fusisporium Schiedermayert Thuem.

This form differs from the species in the colour of the stroma, which is pale, white to

flesh colour or yellowish. Conidia in sporodochia and pionnotes orange, 3—5-septate, less

frequently O-1-, or 6—7-septate.

FAO DMocccaeesaoce 22-48 x 2-5:3....... Mostly 23-46 x 2-5-

RONDO. Hs edpaesiag IM Etls Xx APs —OWoue oc Mostly 40-67 x 2-8-5

CRE DURUCs Cos oes B84 de 46-91 x 3-5:5....... Average 65 x 4:2

QAI NEW. occe6se8056 65-102 x 3-5.

@=septatess-4.55-..4-. 6-14 x 1-5 -4-7....° Average 10-5) x 3.

leseptatest her sci. ssc): 12-220 2205. 4.005-9 Average li x3.

300

fs eS 2a,

¢ ait | y =

A ay SD

Ne »

~ = 7 OCA,

. ee! a Wiese ae

é PR

ante a os

bese omet A a

A ~— )

1 J f j .

= Sea é a

. ¢

a =_ ‘ 7

ae So encrtati s

mse, sat selon recon

a ae wr | aay

Ma Ww Ameren ei:

onl 2

% Sy

aS ea

> aA

os i

ra _ od

& »)

yy 0)

A 4

Fie. 6.

Fusarium avenaceum (Fr.) Sace. f.1. et Rkg.; (a) conidia from pionnotes on ovaries of Paspalum

conidia from (6) small sporodochium of 4 weeks old culture on standard synthetic agar plus starch; (c)

small sporodochium of 4 weeks old rice culture ; conidia from pionnotes of 2 weeks old culture on (d) bean

pod (e) potato plug and (f) Melilotus stem ; b-c, culture from Paspalum; d-f, culture from Citrus.

Hab. Citrus sp., from bark of tree, orange on lemon stock, shewing collar rot, Godwan

River Estates, EH. Transvaal, October 1930 (Marloth).

Euphorbia crassipes Marloth, from decaying stems, Willeston, Cape, M.H. 28379.

Paspalum dilatatum Pair, on ovaries infected with Claviceps paspali, Cedara, Natal,

January 1930 (Gull), M.H. 25473; Eshowe, Zululand, January 1930 (van der Plank) M.H.

25864.

Paspalum distichum L., on ovaries infected with Claviceps paspali, Hastern Vlei, Durban,

April 1926 (Kent) M.H. 21105; Umbilo, nr. Durban, April 1926 (Kent) M.H. 23699.

F. avenaceum f. 1. occurs on a number of hosts, usually in the temperate zone, and is

less widely distributed than the type species. It is known as a cause of rotting in apples.

Growth’ on Standard Media.

Oat agar: Aerial mycelium cottony, often tufted, scanty in cultures from conidia,

abundant from a mycelial transfer, white, becoming tilleul buff to pale pinkish buff. Growth

351

in substratum buff pink. Pionnotes and sporodochia, developing freely after 14 days,

were buff pink to salmon colour or apricot buff to apricot orange. After 2 months, sporo-

dochia up to 3 mm. in diameter were frequently present in large numbers at the base of the

slant ; they frequently coalesced to form more extensive conidial layers. In one set of

cultures, a tinge of deep delft blue developed in the stroma at the base of the slant after

8 weeks.

Hard potato agar: Aerial mycelium scant to moderate, white, cottony, covering the

medium, or developing only near the top of the slants. Sporodochia fairly well developed

after 4 weeks, apricot buff to apricot orange, remaining discrete, or coalescing to form a

pionnotal layer.

Standard synthetic agar plus starch: Aerial mycelium scant or moderate in amount,

cottony, tufted, white, sometimes tinged with barium yellow. Pionnotes well developed

after 5 weeks, at first sea-shell pink, then salmon buff. In one set of cultures, there was

a patch of blue-black plectenchyma at the base of the slant.

Potato agar plus 5 per cent. dextrose: Aerial mycelium copious, cottony to arachnoid,

white to flesh colour and primrose yellow. Growth in substratum pale to flesh colour

becoming apricot buff to olive brown with age. The pink colour fades from the aerial

mycelium after 12-14 weeks, and it is then white to citron yellow. Groups of sporodochia

sometimes developed, they were apricot orange to bittersweet orange.

Potato plug: Plug covered with a moderate to vigorous mycelial growth ; this was

white to primrose yellow and flesh pink, but the pink colour faded after 14 days, and the

mycelium was then white and citron yellow. After 4 weeks the aerial mycelium became

rather flattened and felt-like. Growth on substratum pale to flesh colour or naphthalene

yellow. Pionnotes developed freely in 14 days and were flesh ochre. In one set of

cultures, a few specks of blue-black plectenchyma developed at the back of the plug against

the glass.

Melilotus stem: Aerial mycelium scanty to moderate or copious, white, sometimes

tinged primrose yellow, cottony to sericeo-tomentose. Pionnotes developed freely after

14 days, flesh ochre. A few blue-black sclerotia developed in one strain.

Bean pod: Mycelial growth moderate to vigorous, short or sericeo-tomentose, white,

then primrose yellow in places ; pionnotes developing after 14 days, flesh pink ; sporodochia

forming occasionally, apricot buff.

Rice: Mycelial growth at first white; growth on substratum becoming pale flesh

colour to flesh colour ; grains becoming naples yellow and then barium yellow. Sporodochia

sometimes formed in groups after 14 days ; they were numerous, 0-5 to 1 mm. in diameter,

or coalescing to form spore masses up to 5 mm. in diam., carrot red to grenadine in colour.

Measurements of Conidia.

A.—Direct from the pionnotal layer on the host, Claviceps Paspali.

M.H. 25473.—

DOA 0o0 50000008 Rewer cities see 37-50 x 3-5-4.

ASeptabe.... 5. sce o. 7 VHS Oils ogo co ol 40-45 x 3-3-7.

J-SCPUAbOna ae casas acl [Bil Wii Setaecasimin airer 22-5-42-5 x 2-75-38.

ZS 600 66 6R0008 GoG) 5

ICsepbatenseeeeir ae) a A

OAIEDUUO. o00dc0000000 9-5 i

M.H. 21105.—

URED. cosccoacoee Web) [Ne CAI. ooo does 57-5 xX 3.

G-septate............°-0°5 CEEGSDE unm RS Se 50-60 x 3°7.

OAIONUAUS.cooccseavce TA AER ie enue 37:5-52°5 x 3-5.

AEROVDUMID.s oa c00e o0c DO eenne | eras cin: 32°5-42-5 x 3-4.

SH MMSsccosdnqeace (ksee Dees utitege San ee 25-45 x 2-5-8.

AROUND. 000 60 6000 Ge 1-5 a

lseptabessereeie: i: 2:5 a

M.H. 23699.—

9-l0-septate............ ESBS ban on cher Gua ovent 80-85 x 4-7-5.

S=sepbalenere ree IME Hala biPces esac osorac 10-7125 x 4-4-5

(COUN scogsesvooce Orbyjome Coiltsshandos 60-75 x 3-4-5.

OHM ccacgaeacon » We) DSA ALA ait ol ti 55-70 x 3-4.

Heseplavewe ns ras seer LL OR mM Uy ARRM Ii Eee 45-52-5 x 3-4-5.

ALSONMI@scaccocesdce lMGat acy AOR: B—0) S< Welsh.

ek icles GOGO Ae 30-47-5 < 2-5-3.

USN. osc ecccacse 0:4 53

OAESOUENWOs200600c000¢ OA ere.

In the last list of measurements, there is an unusually large percentage of conidia with

4-10 septations, and the higher septate conidia are longer than the average.

B.—Measurements of conidia frcm cultures (strain isolated from M.H. 25473).

Oat agar, culture 4 weeks old, conidia from sporodochia—

HAINES. coo gccceass GO NSP CMMisseccodc 40-55 x 3-3-75.

pe eee arin ttis 19 A Neen tte sue iene 3-5-5215 XS 28-3715

i -septate. . side old ga alcatel gp s\necr a Ree 30-52-5 x 2-8-3-75.

1-septate. . vee 0-5

Standard sy nthetic a avar plus starch, culture 4 weeks old, conidia from sporodochia—

4-septate: cee as ae tes di TOP CBMs oo oocoore 42-5-47°5 x 2:8-3-75.

GSO OEWGccacdcccveoee 96 see ics aA 35-47:°5 xX 2-5-3-75.

Rice, culture 4 weeks old, conidia from sporodochia—

SNES caccccsoces A> OOP CMs odsacas 45-50 x 2-8-3

JENN snaaosceooos «aus eis adaptor oa 45-50 x 2-5-3.

Besepbabernn emma re cc) wae) Oa win ain keene rarer 30-47-5 x 2-5-3.

meer rate 0-7 6

OA ONMccascoscose OM o

Potato agar plus 5 per cent. dextrose, culture 4 weeks old, conidia frem sporodochia—

AED. cogcanseove 1-5 per cent 2 sate 35-37-5 x 2-5-3.

S-SEPUALC ees etry OO ON tas Mir CA eRe 21-25-37-5 x 2-5-3.

AMINO. coudoesdsoc dl i

IESeptaleryaameia eae 8 7

O-septatenmere sn neene 1 my

C.—Measurements of conidia from culture derived from citrus.

Oat agar, culture 4 weeks old, conidia from pionnotes—

HSU MCsccoccosncca MNO er Cam. ooc5c008- 32-50 x 3-5-4.

A-SeVtAtOwans cockines “LOO Gs) (ions 30-45 x 3-4.

SAODWMOs coo coecooge Ghee Marae te cease 22-5-37-5 X 2:5-3-79

Bean pod, culture 14 days old, conidia from pionnotes—

OO obo 0000005 Few.. BA al fiegs Gree tn MOONE Ue

O-septatess aay ace eaper “cent. fone crit fs 37-5-55 x 3-75-4-5

as eye rs SmI ys PE Com rcs Ul, o dow 30-45 x 3-3°75.

3-septate.....5..2-..- 6 RO SAD § crea on) Geieen 22-5-37:5 x 2-5-3°75

0- 2. SCIONS Gon ogogn aoc 3

Melilotus stem, culture 14 days old, conidia from pionnotes—

DANI sno oe senads | ©) OC CMMs doc sacoec 42-5-50 x 3:25-3:79.

oe Sa irate I ee Mnatnihe NNALAiptere Cut . 0447-5 x 3°25-3-75.

S-Septaters,.se.cics cea LOU Cr mein alone. 225-45 x 2-5-3-75.

0-2-septate.... 5 5

Hard potato agar, culture 14 days old, conidia from pionnotes—

GSI] coo vcecaeae I OP CBMs coo voce oe 50-60 x 3-5-4.

OAK WERT Gwin couaacco 240 MMR ME rece ooo hag 40-50 x 3-75-4.

4-septate. 0... tes. es | 24 DEL ATS: Pees: 40-45 x 3-75-4.

SSO pPtates esc. she eye, -2Os0 | tay, ae ee 25-42°5 x 2:75-4.

O-AAROOMENEES coo coo cd06 0

303

Section ARTHROSPORIELLA.

Sherbakoff, New York (Cornell) Agric. Exp. Sta. Memoir 6: 161, 1915. Wollenweber, Ann. Myc-

15:2, 1917; Ber. deutsch. bot Ges. 35: 733, 1918; Fusarium-Monographie 324, 1931. Wollenweber

and Reinking, Phytopathology, 15: 157, 1925; Die Fusarien, 57, 1935. Reinking and Wollenweber,

ELM Jour. Sci. 32 : 118, 1927.

Aerial mycelium abundant, floccose. Stroma, more or less effuse, variable in colour >

it may be light, yellowish, pink or ochre, to light or chestnut brown. Sporodochia typically

absent, and pionnotes also usually absent. Conidia usually scattered in the aerial mycelium,

of two kinds: Small to medium in size, with 0-3 septations, fusiform, cuneate or lanceolate,

apedicellate ; or larger, 3- or more septate, fusiform-falcate, with basal cell conical, con-

stricted or papillate (rarely pedicellate). Chlamydospores intercalary, seldom terminal.

Spherical sclerotia occasionally developing, and usually pale or light brown in colour.

Fusarium semitectum Berk. et Rav. var. majus Wr.

Wollenweber, Fusarium-Monographie, 325, 1931; Fus. aut. del. 113-116, 552, 907-910. Wollenweber

and Reinking, Die Fusarien, 59, 1935.

Syn. Fusarium asparagi Briard ; Ff. incarnatum (Rob.) Sace.

Fusisporium incarnatum Rob. ; Fusarium juglandinum Peck.

Fusarium oxysporum Schl. subsp. aurantiacum Sacc. (non Corda).

EF. oxysporum Schl. v. aurantiacum f. hyalina Brun.

F. pallido-roseum (Cke.) Sace. ; FPusisporium pallido-roseum Cke.

Fusarium roseum Lk. v. calystegiae Sacc.

Aerial mycelium white to flesh-colour or isabellinous ; stroma plectenchymatous, light

brown or pink. Chlamydospores intercalary. Sporodochia wanting. Conidia powdery,

scattered in the aerial mycelium, or adherent in clusters or false heads, salmon pink in mass ;

5-septate, or less frequently 3—4-, occasionally 6-10-septate, intermingled with smaller

0-2-septate forms. Macroconidia fusiform to lanceolate, straight or slig htly curved, usually

conical at the base, sometimes papillate, exceptionally pedicellate.

: ORNs 062 og0hns ce Hab Kee ea ad dace. ES a aS seg

l-septate............ 9-24 x 2-5-4.. . Mostly 14-21 x 2-5-3-2.

3-septate............ 13-40 x 2-5- Ae Bin Mostly 19-29 x 3-4-5.

H-septate............ 29-52) x 2-5-6....... Mostly 30-48 x 3-7-4-8

oe ae a 45-70 x 3°7-6:2..... Mostly 44— el x 4-3-6.

VESNAWsasp sacs eene § DOM) XM bacon nome Average 60 x 5:3.

Hab. Citrus Limonia Osthecls from collapsed fruit from Sunday’s River, Cape, after

seven weeks in storage.

Citrus sinensis Osbeck, in decaying fruits ; from stem end and navel end rot of oranges,

after 12-18 weeks in storage (7 isolations) ; in navel oranges from White River, E. Transvaal,

Zebediela and Letaba, N. Transvaal, and from Sunday’s River, Cape.

On twigs dying back; Hankey, Cape (van der Plank) ; Ofcalaco, N. Transvaal (van der

Plank), July 1930, M.H. 28400.

Dianthus caryophyllus L., from stems of wilting plants, Durban (McClean).

Striga lutea Lour., from stems of dying plants, “Txopo, Natal (Mack).

Glossina sp., from dead Tsetse fly, Zululand, 1931 (Harris) M.H. 28446.

Musa Sapientum L., from surface of fruit, Acornhoek (Boyce).

Nomadacris septemfasciata, on red locusts dying from the attack of Beawvaria sp.,

Pretoria dist., 1933.

Eggs, from purplish-brown, discoloured patches of membrane, which at this point

adhered to the shell, albumen partially coagulated (ass. #. monaliforme), sent by Poultry

Inspector, Port Elizabeth (Bottomley).

This species is a common saprophyte on old, dry parts of plants, and on decaying

fruit, in all parts of the world.

oto

| { } 4 r —

Meo ee a) Go

| ea ——

Fr EEN Were

a VES pies

¢ i »

Samy

ce]

CoE EA PIEET A r,

TEttan |

i a. ne

en) et “y

{

TO. a 7

Se t

ono {

a ee Pee

S99}

: ae

oe = — ee Ey

ap iia Sas

Pe oe

Ie, Zo

Fusarium semitectum Berk. et Ray. v. majus Wr.; conidia from mycelium of 2,;weeks old culture on

(a) synthetic agar plus starch (b) oat agar and (c) bean pod.

Growth on Standard Media.

Oat agar: Aerial mycelium rather scant, white, cottony, becoming mealy-looking

after 14 days, owing to the production of clusters of conidia in the mycelium. Growth on

substratum light pinkish cinnamon, sometimes with a line of sepia at the base of the slant.

355

Hard potato agar: Aerial mycelium short, white, becoming mealy-looking, longer and

cottony near the top of the slant. Growth flesh pink when conidia are freely produced.

Standard synthetic agar plus starch: Aerial mycelium scant, white, cottony ; growth

in substratum white to light pinkish cinnamon.

Potato agar plus 5 per cent. dextrose: Aerial mycelium copious, fine, cottony ; at first

white, after 7 days tinged naples yellow and pale salmon colour; later it may become

pale to dark olive buff or brownish olive, the darker colour at the base of the slant. Growth

in substratum at first salmon colour to flesh ochre, later it may become citrine drab to

brownish olive. Aerial mycelium often mealy in appearance, owing to the formation of

conidia.

Potato plug: Plug covered with a copious growth of white, cottony mycelium ; later

it becomes felt-like and wrinkled, or mealy-looking if conidia are present. Growth in sub-

stratum flesh colour to flesh ochre, sometimes becoming deep olive or buff to buffy brown

with age. When conidia are produced freely in the aerial mycelium, they are pale pinkish

cinnamon in mass.

Melilotus stem: Aerial mycelium moderate in amount, white to dirty-white, or tinged

ochre, tomentose or sericeo-tomentose.

Bean pod: Aerial mycelium vigorous, tomentose to mealy, white, or tinged salmon

colour and pinkish buff owing to the presence of numerous conidia.

Rice: Growth at first white to flesh pink, and rice grains naples yellow. Growth may

remain pink or become wood brown to natal brown, with the grains also brown. Powdery

spore masses are white to pale pinkish cinnamon.

Measurements of Conidia.

Bean, culture 4 weeks old, conidia from mycelium—

S*sepualenesat rer, HOW sacs ato ys ee On A

NEse patton eer eet: HOWe cece era secn) DOD XK 3-4,

6-septate............ Me Westies meer ee cpt AO) OX An

HQ MUMO.coscocaccos Ge) JIE CEMGocosco0e 25-62°5 Xx 3-5.

APSE DtALE: cic eck cic al SOW oh ak 25-35 x 3-5-4.

3-septate...........- 30 SUL, oti Aaedrs 20-32-5 x 2-5-4.

ZSepuave ccs Leo Zieh, Mica eae 15-20 x 3-4.

lEseptatesreese eee LS i etsicus go, Bos 8-20 x 2-4.

OZseptateserarceieer 4-5 eR. eat 7-5-10 x 2-4.

Hard potato agar, culture 4 weeks old, conidia from mycelium—

UBOVMBoccocccoscea dhy MUSES doo oe 62-5-70 x 3:2.

GSE PUL merrell ce 2 Sine MEIC cogae 55-62°5 x 3-4.

D-sepbate erie 52 ae rr nc eer 35-60 x 3-4

ASSEN oo ob oo gooD ee 3 Sali! ote niolteenate 30-40 x 3-4.

3-septate............ 37:5 Beet Ss ota 20-50 x 3-3-75.

O=2-septateysstans- uci Moy

Standard synthetic agar plus strach, culture 4 weeks old, conidia from myceltum—

SAS NENG, oon oseodo0e Rares ae sch cere oe 55 x 3-4.

T-septate:.:........- Ieper Cenibayxerrsenr 65-80 x 3-75-4.

G-septate)...----.-)-- 6 Be et Phar Ee ce 57-5-72°5 x 3-3-7.

S-septate.... . 5.06.0. 20 aaa Bh ca 30-55 x 3-4-7.

4-septates.-2-5-.<os- USD hee Saved stare 3 30-45 x 2-8-3°7.

SHIM 00 0500000C 60 RS eoeee 25-30 x 3-5.

0-2-septate............ 5:5 nt

A few intercalary chlamydospores were observed on plain agar plates.

396

Section GIBBOSUM.

Wollenweber, Fusarium-Monographie 328, 1931. Wollenweber and Reinking, Die Fusarien 61-62>

1935.

Aerial mycelium white or brownish, less frequently yellow, pink or carmine. Stroma

ochre to blackish brown, sometimes golden yellow to carmine red; plectenchymatous

stroma may, or may not be rugulose and sclerotially erumpent ; spherical, brown or dark

blue selerotia present or absent. Microconidia scattered more or less freely in the young

mycelium, disappearing later. Macroconidia in sporodochia and pionnotes, also found in

false heads and clusters, or in a loose powder on the mycelium ; in mucilaginous masses,

the conidia are isabellinous to ochre and orange red; when dry and powdery, they are

light-coloured, brownish white. Typical conidia thin-walled, but distinctly 3-5—7- or more

septate, dorsiventral, slender, more or less falcate, with parabolic or hyperbolic curvature,

sometimes with rather acutely arched dorsal line and somewhat less curved ventral line,

tapering at both ends, with filiform or flagelliform apical cell, and very definitely pedicellate

base. Chlamydospores intercalary, seldom terminal, in conidia and mycelium, spherical,

single, in chains or in clusters, brown in mass. The ascus stage is Gibberella.

Key to South African Species.

A.—Curvature of macroconidia more or less parabolic or falcate.

B.—Macroconidia 3 (3-5) septate: 3-sept. 33 x 4: 5-sept.

AG SOAP Gc cia ue at eee sre nee cabasae ners ages REE aE aes F. equiseti.

BB.—Macroconidia 3-5-sept.: 3-sept. 33 x 3-75: 5-sept.

LOD Gat oie ee te ME EEA ERAN S Nae cone oe, SomIs 68 lo nce e F., equiseti v. bullatum.

AA.— Curvature of macroconidia more or less hyperbolic. Conidia

5-septate.

B.—5-septate conidia 43 « 4-4, comparatively compact, 8-9

times) as) longsasybroada.... ane een eee een ere F. scirpi v. compactum.

BB.—5-septate conidia comparatively slender, 10-12 times as

long as broad.

C.—Stroma not carmine to yellow...................... FF. seirpi.

CC Stromarcarminestoyy.cllowameener eerie eer F. scirpiv. acuminatum

BBB.—5-septate conidia very long and slender, 14—21 times as

long.:as broads... iar: cee =. we eo RIE Se: ete ee F. scirpi v. filiferum.

Fusarium equiseti (Cda.) Sacc.

Saccardo, Syll. Fung. 4: 707, 1886. Wollenweber, Fusarium-Monographie, 330, 1931; Fus. aut.

del. 202-208, 210, 211, 596, 597, 919, 920. Wollenweber and Reinking, Die Fusarien, 63-65, 1935.

Syn. Fusarium equiseti (Cda.) Sace. f. 1. Wr. ; Selenosporium equiseti Cda.

Fusarium Cordae Mass.; F. falcatum App. et. Wr.

F’. falcatum App. et Wr. v. fuscum Sherb.

FE. mucronatum Fautr. in herb. pr. p. ; Fusoma pallidum Bon.

Fusarium ossicolum (Berk. et Curt.) Sace. ; Fusisporium ossicola Sace.

Conidia sparse at first, scattered in the mycelium, which is white to yellowish, or pink ;

they are l-celled or septate, oval or oblong to fusiform-faleate, sometimes comma-shaped,

and they disappear when the typical macroconidia begin to develop. Stroma pale or brown,

not carmine red, and of varying extent. Macroconidia in tubercular sporodochia, in pion-

notes or in clusters, seldom powdery and scattered in the mycelium ; in mass they are at

first pale, almost mealy white, then ochre to salmon pink; when dry, the spore masses are

honey colour to cinnamon brown or lighter. Macroconidia typically fusiform, thick in the

centre and tapering gradually to each end, curvature parabolic, straight or bent at the apex

and tapering to a fine point, pedicellate at the base; dorsal side usually more strongly

307

curved than the ventral side; cells more or less equal, cross walls seldom more closely

crowded in the centre than at the ends; mostly 5-septate, seldom 3-4-, exceptionally up

to 12-septate.

OSCMUMICsccovacesane CMG K BeO=Gosgecuae Average 12 x 2-5.

NSCS. ooo clo edad | IDL SoD wen seo, Average 16 x 3:

SHINN. oosecocsese W2E44 Xx 23-50... .. Mostly 15-36 x 2-5-4-8.

O-Septaberae ares ese 26-14 x 2-8-5-7..... _ Mostly 29-56. x 3-5-3.

UG DUM nog pond oan | Cl)! x ZEON es mieer Mostly 52-62 x 4-2-5-3.

8—l2:septate......... Upto 83 x 5-5.

ENT

pe tS ak ee ah

eee Ce i—

j at t Se

“ ai

ee ey 1

Lone inmate ©

ee igs — Wie entiat ee oeee ae

ES) Re ca { & oe. x

Lo ee % : 4

ye coli ul oi By ~\ . : i

: ra ! fs x ui :

L> % —— i_.| . oe

oe aes, S 3 \ ~<

/ ES Eo - SS

{/ AA AERIS See NO : 4 ¥

| mma ’ \ :

/ Ne v aa |

, eens F | \ \

en ol Bl a eal Cr)

Ss {\

estas a

caked: Theat () j

Pk SF ea Dit oe ae \\ y

iy t a aR A

ANS 4 =

eat

Fic. 8.

Fusarium equiseti (Cda.) Sace.; ‘conidia from (a) mycelium of culture on plain agar, 5 days old; (b)

pionnotes of 2 weeks old culture on synthetic agar plus starch; (c) sporodochia of 8 weeks old culture

on oat agar; chlamydospores from (d) culture on plain agar, 5 days old, and (e) culture on hard potato

agar, 4 weeks old.

308

Conidiophores simple or branched; branches spreading or fasciculate, arranged in

successive whorls of 2-3 or more, and bearing at their tips groups of 1, 2 or 3 sterigma-like

pegs to each ultimate branch. Chlamydospores 6-14 yw in diameter, round, smooth or rough,

more frequently intercalary than terminal, sometimes 1-celled, but usually in chains or

clusters ; brown in mass.

Hab. Citrus sinensis Osbeck, from Valencia oranges shewing stem end rot after 12-18

weeks in storage; oranges from Zebediela, N. Transvaal, 1933.

Cucumis satus L., from stems of wilting plant, Uitenhage, Jan. 1935 (Haines).

Lycopersicum esculentum Mill., from tomato seed offered for sale, Pretoria, 1931 (Wager) ;

from petioles of dying plants, Gqaga, Transkei (Wager).

Striga lutea Lour., from stems of dying witchweed plants, Ixopo, Natal (Mack.).

Growth on Standard Media.

Oat agar: Aerial mycelium abundant, cottony, white to pale pinkish buff and pinkish

buff; growth on substratum congo pink. A few small salmon-buff sporodochia developed

in two months.

Hard potato agar : Aerial mycelium sparse to moderate in amount, cottony to arachnoid,

white ; growth in substratum colourless.

Standard synthetic agar plus starch: Aerial mycelium sparse, white, in scattered tufts.

Pionnotes began to appear after 8 days, and were well developed after 15 days ; they were

light vinaceous cinnamon.

Potato agar plus 5 per cent. dextrose: Aerial mycelium fairly abundant, cottony, white

to pale flesh colour ; aerial mycelium and growth in substratum became brown with age.

Potato plug : The plug became covered with a dense, matted, cottony mycelium,

which was at first white to seashell pink ; the pink colour soon faded. Growth on substratum

isabella colour to light brownish olive. A few salmon buff sporodochia developed after

21 days.

Melilotus stem: Aerial mycelium fairly copious, white. A few flesh ochre sporodochia

developed on some twigs, and on others a fairly extensive pionnotes.

Bean pod: Aerial mycelium abundant, cottony to arachnoid ; at first white to shell

pink, soon becoming ochraceous buff and clay colour. No conidial masses seen.

Rice: Mycelium tilleul buff to vinaceous buff, with a ring of buffy brown at the base

of the growth.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 15 days old, conidia from pionnotes—

NOEOUENG. 00000000008 OB JOE CRMs ooc 0006 82 x 5.

GHSANEM@.ac000000008 1 iolye ier aes 72°5-75 x 4-5-5.

SHIM ccscccecve0 Sa coh toe 65-15 x 4-5-5.

TSCPtate ci oaeccneen LOMB. sign) Must cess 52-5-75 x 4-5.

Gseptates. Sink... sun U2 oe iene enema 50-62°5 x 4-5.

HASEPtabes Sass. casas TOOTEE MRR Uteenetasts © 30-97°5 x 3-75-59.

a “septatenawmrr rae Tic oi) Mieheet # Seat. cee: 30-40 x 3-75-4-5.

3-septate. .. ad ins SS ale Nah 22-5-27-5 x 3-754

Potato, culture 9 weeks old, conidia from sporodochia—

TESe placement i ttniet VOIP COMly nono 00 40-52-5 x 3-75-5°8.

G-septatena- ee cecum, BREAAY vstins ete 37:5-50 x 3-75-95.

5-septatet.g- seo ae 66 ny Oaths arate 26:25-57-5 x 3-75-5°8.

A-septatews sone ec MORO ee an 2 ke 225-40 x 3:5-5°8.

Sseptate.ica sce aes 14°5 Aes Tate Sees 15-32:5 x 3:75-6.

2-septate......0....- 4-5 sau foyhartinet WAGs. 20-27-5 x 3-5-4.

I-septate: . 2h 2 5 ser 3 hates Whee LAU 10-30 x 3-5-4

309

Fusarium equiseti (Cda.) Sacc. var. bullatum (Sherb.) Wr.

Wollenweber, Fusarium-Monographie, 331, 1931; Fus. aut. del. 117, 290, 913-918. Wollenweber

and Reinking, Die Fusarien, 64-65, 1935.

| Syn. Fusariwm bullatum Sherb.

F. bullatum v. rosewm Sherb. and vy. roseo-bullatum (Sh.) Wr.

F. bullatum Shetrb. v. brevius Wr. et Rkg., and v. minus Wr. et Rkg.

F. equiseti (Cda.) Sace. v. bullatum f. 1. et f. 2. Wr.

F. nectriae-palmicolae P. Henn. ; I. terrestris Manns.

The conidia of this variety are, on the whole, somewhat less curved than those of the

type species, and of other members of the Gibbosum-section ; the foot at the base of the coni-

dium is less sharply defined. The septation is inclined to be lower than in the typical F.

equisett, and in the mycelial stage, there are often produced lanceolate forms, recalling the

conidia of the Arthrosporiella-Fusaria, or forms resembling the sub-normal conidia of the

Discolor-Fusaria. Typical conidia from sporodochia and pionnotes measure—

5-septate............ 31-47 x 44-9....... Average 42 x 4:3.

TAG OUMs ood gene do ae BURG XX GT=BSBs 566 6 Average 33 x 3-75.

They are cream to salmon colour in mass. Chlamydospores are mostly intercalary,

in chains, or in small or large clusters. Aerial mycelium is usually abundant, of average

height and density, and almost pure white in colour. The stroma is pale to brown, and the

substratum often absorbs the colour of the stroma.

Oe. b. s _

/ fi DAG Ss ~ Le

; j 7 ae ne, ay ;

/ fi f. ~N f/f

se 4 3 ¥ é in

eee a f t i

ot i E

ay { fi

if \ } ¢

=i A i e .

Fie. 9.

Fusarium equiseti (Cda.) Sacc. v. bullatum (Sherb.) Wr.; conidia from sporodochia of 4 weeks old

cultures,on (a) standard synthetic agar plus starch, (b) oat agar, and (c) Melilotus stem.

Hab. Mesembrianthemum sp., from rotting stems of succulent plants, Pretoria (Wager).

This variety occurs on decaying parts of plants belonging to a number of different

genera, in tropical and sub-tropical countries. It is known in Asia and America, and occurs

occasionally on scale insects.

The ascus stage is Gibberella intricans Wr., which is said to develop freely in pure cul-

tures. It has not been observed in South Africa, either occurring naturally or in pure

culture. It may be briefly characterised as follows :—

Gibberella intricans Wr.

Wollenweber, Fusarium-Monographie 332, 1931; Fus. aut. del. 810. Wollenweber and Reinking,

Die Fusarien, 65-66, 1935.

Perithecia solitary or in groups, ovoid, rugulose, ostiolate, 0-17-0-4 x 0-15-0-3 mm.

usually 0-3-0-35 x 0-18-0-24 mm., blue-black; asci spuriously paraphysate, clavate,

8- or 4-spored, rarely 2-spored, monostichous or obliquely distichous ; sporidia 3-septate,

rarely 1—2- or 4~7-septate, fusoid, more rarely straight than curved, slightly falcate, conical

at both ends; 3-septate sporidia 19-36 x 3-7-7, mostly 21-33 x 4-1-5-6.

The Gibberella-stage was first observed on dry leaves of banana (39).

360

Growth on Standard Media.

Oat agar : Mycelium white, tomentose, fairly abundant ; growth in substratum colour-

less. Pionnotes, fcim'ng freely aftcr 14 days, were light ochraceous salmon.

Eard potato agar wind standard synthetic agar plus starch : Growth on these media

rescmbled that on oat agar.

Potato Agar plus 5 per cent. dextrose: Aerial mycelium white, tementose ; Growth

on substratum at first white to cream colour, becoming salmon colour, and, after 4 weeks,

pale pinkish cinnamon to dark olive buff.

Potato plug: After 14 days, the plug was covered with a vigorous growth of white,

tementose mycelitm.

Melilotus stem: After 14 days, the stems were covered with a white, tomentose to

sericeo-tcmentose mycelium ; pionnotes, pinkish buff in colour, developed after 4 weeks.

Bean pod: Growth similar to that on Melilotus stems.

Rice: After 14 days, the growth was white to tilleul buff; after 4 weeks, the colour

deepened to wood brown in places.

Only one strain of this variety was studied, and in this, the colour of the stroma was

somewhat lighter than that of strains studied elsewhere (39).

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from picnnotes—

PIIONMOsccocccacces AY JOOP CSM. cocaecc 25-40 x 3-3-75.

4-septate...... 252.5. 56:5 SERN IUEY cy Teese 22-5-37-5 x 3-3:75

SHO OUD, cacccccaacg Iliols ERY? SP el Oa Wo D=D) x Bo MD,

Standard synthetic agar plus starch, culture 4 weeks old, conidia from pionnotes—

DIOOUMOoocccoesaccd XO) [NC CCM oso ccacce 31-47-5 x 3-4-5.

BAKIOUWLNUE: coocosecaae ON) Pet Wy POMEL ool 30-35 x 3-3:75.

O-BEPLAte aise Soe, 7 See oa ee Lae ewe ae 225-30 x 3-3-70.

Melhilotus stem, culture 4 weeks old, conidia from pionnotes—

DAI|DUENOs 000 nob oe0 a0 743) {NEO CMs ogo cecase 30-37°5 x 3-4.

A-Sepvateynss qc nee 50 aerate reseis aPb. 5 25-3) X 3-30-10.

3-sepuate......:.54-5 ZOOM aaias See eee 20-30 x 3-3-5.

IFS NEMEdos eeloctence dem) 4,

Fusarium scirpi Lamb. et Fautr.

Lambotte et Fautrey, Rev. Mycol., 111, 1894. Wollenweber, Fusarium-Monographie, 334-335,

1931; Fus. aut. del. 198-201, 212-218, 595, 598, 926-929, 1137. Wollenweber and Reinking, Die Fusarien,

66, 1935.

Syn. Fusarium scirpi Lamb. et Fautr. f. 7. Wr.; 2. sclerotiwm Wr.

F. scirpt Lamb. et Fautr. v. nigrantum (-nigrans) Benn.

F.. scirpi Lamb. et Fautr. v. pallens Benn., and v. comma Wr.

EF. gibbosum App. et Wr.; F. aleyrodis Petch.

Fusarium sclerodermatis Oud. v. lycoperdonis Picb.

I. chenopodinum (Thuem.) Sace.; Fusisporiwm chenopodinum Thuem.

Fusoma helminthosporii Corda.

Fusarium roseum Lk. f. solani nigri Sacc. in Myc. Ven. 367.

Sporodochia pale, yellowish pink or ochre to salmon colour; at first they are minute

bodies the size of a pin’s head, and soon become dry and powdery, or, if moist, coalesce

into an extended pionnotes. Spore masses, when dry, are pale to light brown, or occasionally

cinnamon brown. Aerial mycelium loose, cottony or filamentous, light or brownish, later

disappearing. Stroma brown, rarely with blue-black spherical sclerotia. Comnidia resemble

those of F. equiseti, but the apical cell is more prolonged and pointed, and the curvature

of the dorsal side is strongly hyperbolic ; the cross-walls are more numerous, and are more

361

closely crowded in the middle of the conidium than at the ends. Scattered microconidia

are at first comparatively abundant, 0—3-septate, oval, fusiform, reniform or comma-shaped,

and also club-shaped to lanceolate. Macroconidia produced in sporodochia and pionnotes

typically faleate, mostly 5-septate, more rarely 3-4- or 8—11-septate.

OREN cobeseecccs DVD eK 2k, ceive’ Mostly 6-10 x 2-4-3-5.

I-séptate serie soe (820) X24 ee Se Mostly 10-15 x “9. 1-30

SACOM ccocgceccus NO=55) XK AoW 7-33 0556 Mostly 17-44 x 3-3-4:5.

DaSepuabesnctt eae ser 20-73 x 3-6......... Mostly 22-66 x 4-5-3.

SRO DIME socccee sous B01 358-0 =e) Mostly 80-11) 42-55.

Q-SeMbavenmctamiibe 51-83 x 4:5-6....... Mostly 67 x 5-2

Chlamydospores intercalary, seldom terminal, mostly in chains or clusters, brown in

colour; rarely single, round and 7-14 » diam. Sclerotia, when present, spherical, brown

to dark blue, 60-80 jy diam.

Yo Er

Z . z C

Z 4 y b. \\

A f 2 3) ie

fay & VA }

— s j —~ a

j —~ / | )

A | } i fF

Sal fo /

ma \ | mo jin! (aes be 7 i

1 i

\ £ cf | | fe ,

Gl \ [ j

\ \ | eI | i ~ /

7 \ — A

? . f~/ /

\ eS \ 1] wat | f/f oF

ia a ea ]

| f

w YX \\ \

¢ KW

yy,

dees: 6

/ —— . pa reat j Re ~ 4

ny KX ( ~~? | ¢

/ 7 LS —— Cc a ") o ad t t

hie? “ nen Mesh DA

[ al gy \ ae a a Ora ibe =e = =

A a

ao <L ve if

lar ae

[Ey _ la

- A/S [\ iis is § Mate .

} Cc (

\ " fo} G RS iC i j

\ i re Gf { eo ae

a Ot y Pe on fees Cle i <

ne er Cs Ve eas CF

\ \ Be Caine: ZF 3 rset

g ae r

OS j ~ ous,

—_——T

Rie. LO:

Fusarium scirpi Lamb. et Fautr.; Conidia from (a) sporodochia of 4 weeks old culture on standard

synthetic agar plus starch, (b) ionnOtes of 4 weeks old culture on oat agar, (c) pionnotes of 2 weeks

old culture on hard potato agar, (d) minute sporodochia of 2 weeks old rice culture; (e) chlamydospores

from 4 weeks old culture on potato plug.

Hab. Allium cepa L., from rotting stems of seedlings, Pyramids, Pretoria dist., March

1932 (Mogg).

Antirrhinum majus L., from stems of wilting plants, Wepener, O.F.S., Pretoria, Trans-

vaal, and Carnarvon, Cape (Wager).

Arachis hypogaea L., on pods and seeds of peanuts attacked in the soil, Pretoria Uni-

versity Farm, 1932 (F. du Toit).

362

Callistephus chinensis Nees, from stems of wilting aster seedlings, Pretoria (associated

with Rhizoctonia and Pythium sp.) ; Hennops River, Pretoria dist. (Havenga).

Carica papaya L., from rotting pawpaw fruit, Buffelspoort, nr. Marikana, and Warm-

baths, Transvaal.

Centaurea cyanus L., from discoloured stem of cornflower, Johannesburg (Wager).

Citrus grandis Osbeck, from bark of grapefruit tree showing gummosis, Patentie, Cape,

July 1930.

Citrus Limonia Osbeck, from roots of rough lemon stock showing dry root rot, Klands-

hoek, EH. Transvaal, July 1930 (M.H. 28911) ; Bonnievale, Cape (van der Hoek) M.H. 28430.

Citrus nobilis Lour. var. deliciosa Swingle, from twig of naartje tree, associated with

Septobasidium sp., Hast London, 1929 (Turner).

Citrus sinensis Osbeck, isolated frequently from oranges showing stem end or navel

end rot, after 12-18 weeks in storage (25 strains studied) ; in fruit from Rustenburg, Zebe-

diela, Transvaal, Groot Drakenstein, Cape, and Muden, Natal.

From bark of orange trees showing root and collar rot,

Godwan River, E. Transvaal, October 1930 (van der Plank) ; Letaba Estates, N. Transvaal

(Matthew).

Crotalaria juncea L., from base of stem of wilting plants of Sunn [Elexr9, University

Farm, Pretoria, 1932 (F. ‘du Toit).

Cucumis sativus L., from cucumbers (fruits), shewing soft rot and leaking, Nelspruit,

HK. Transvaal (Wager) ; from stems of wilting plants, nr. Port Elizabeth (Haines).

Cucurbita pepo L., on young fruits of vegetable marrow rotting in garden, Irene, nr.

Pretoria (Bottomley).

Dianthus caryophyllus L., from stems of dying carnation plants, Golden Valley, Cape

(Wager) ; Hartebeestpoort, Transvaal (associated with Sclerotiwm Rolfsi) ; Duivelskloof,

N. Transvaal (Wager).

Eleusine indica Gaertn., from stems of goose yrass (associated with Helminthosporium

sp.) Acton Homes, nr. Ladysmith, Natal (ibs A. Doidge).

Euphorbia crassipes Marloth, from rotting stems of succulent Huphorbia, Williston,

Cape.

Kentia sp , on stem of dying palm (associated with Gloeosporium sp ), Uitenhage, Cape,

Sept 1932.

Lathyrus odoratus L , from stems of sweet pea seedlings, which were yellowing and dying

off (associated with Pythium sp ), Brooklyn, Pretoria.

Timonium sp., from crown of dying Statice plants, Nelspruit, E. Transvaal (Wager).

Lycopersicum esculentum Mill., from stems of wilted tomato plant, Windhoek, 8.W.A.

(Wager); petioles of dying plant, Gqaga, Transkei; from decaying stems, Karino, H.

Transvaal (Wager), M.H. 28432; from fruits, on “* blossom end rot” lesions, Pyramids,

Pretoria dist.

Matthiola incana R. Br., from discoloured stem of stock plant, Johannesburg (Wager).

Musa Sapientum L., from surface of decaying fruit, Acornhoek (Boybe).

Papaver nudicaule ., from crown of dying plants, Brooklyn, Pretoria (Wager).

Phlox Drummondii Hook., from stems of wilting plants (ass. Rhizoctonia), Brooklyn,

Pretoria. :

Pinus sp., from stems of seedlings, dying in nursery, Heidelberg, Transvaal, M.H.

28392.

Pisum sativum L., from stems of dying plants, Carnarvon, Cape (Wager).

Pteridium aquilinum Kuhn, from rhizome of diseased bracken plant (associated with

Pythium sp. and Pestalotia sp.), White River, H. Transvaal (Wager).

Pyrus malus L., from brown cores of fruit, Vereeniging, 1935-6 (Bottomley).

Solanum tuberosum L., on tubers showing “ dry rot,’ and on tubers breaking down

with “leak” due to Pythiwm sp., Pretoria.

Viscaria viscosa Aschers., from stems of dying plants, Brooklyn, Pretoria.

Icerya purchasi, on Australian bug, on Mentha sp., Grahamstown, 1932 (N. Smith).

363

Nomadacris septemfasciata, on eggs of red locust, hatching in sterilised soil, Pretoria,

1932 (Brooks).

This cosmopolitan species is extremely common in South Africa on dead or dying

parts of plants. It appears frequently to invade plant tissues which have been attacked

by other fungi or otherwise injured, and to be a secondary cause of decay.

Growth on Standard Media.

Oat agar: Aerial mycelium not abundant, fine, white, cottony; after 7 days, very

minute sporodochia were developing all over the face of the slant, these were very numerous

and in places coalesced, but for the most part remained discrete. Sporodochia pale ochraceous

salmon to light vinaceous cinnamon.

Hard potato agar: Aerial mycelium fairly abundant, fine, white, cottony to cobwebby ;

after 14 days, a copious pionnotes had developed on the substratum; this was at first

ochraceous salmon, and later salmon buff and vinaceous cinnamon.

Standard synthetic agar plus starch: Aerial mycelium not very abundant, fine, white,

cobwebby. Pionnotes copious, in one case showing a tendency to develop in concentric

rings round the point of transfer ; pionnotes light ochraceous salmon or buff pink, becoming

salmon buff to light vinaceous cinnamon.

Potato agar plus 5 per cent. dextrose: After 7 days, aerial mycelium was fairly abundant,

white, cottony ; growth on substratum congo pink. After 14 days, growth on the substratum

was dark olive buff, and the medium was stained olive brown to vandyke brown.

Potato plug: Aerial mycelium scanty to moderate in amount, cottony to cobwebby

at first white, but after 4 weeks, becoming flattened and felt-like, and olive buff in colour,

owing to the presence of numerous chlamydospores. Growth on substratum became snuff

brown to warm sepia, and the medium was stained buffy brown. Pionnotes developed

freely after 7 days, on the substratum, and were buff pink to light vinaceous cinnamon,

Melilotus stem: Aerial mycelium fairly abundant, at first white cottony, but after

4 weeks, rather flattened and felt-like, and brownish white in colour. Pionnotes formed

freely, and practically covered the stems; they were pinkish cinnamon to cinnamon.

Bean pod: Aerial mycelium scant to copious, in the latter case almost filling the tube.

Numerous small sporodochia developed, many of which coalesced in patches and formed

a continuous pionnotes; they were pinkish cinnamon to vinaceous cinnamon.

Rice: Mycelium at first white to pale flesh colour; growth on substratum snuff

brown, and grains stained the same colour. After 4 weeks, the aerial mycelium was still

white, but shading to snuff brown near the grains. Minute sporodochia were fairly numerous,

and were light vinaceous cinnamon ; where the conidia had dried to a powder at the surface

of the medium, they were vinaceous pink.

Measurements of Conidia.

Hard potato agar, culture 14 days old, conidia from pionnotes—

Q=sep babes. cia er cise He Wrrrss teeter echt er OM XO!

Szseptateue wre CHOW Meme tse to cre GO-GO) X10:

UWS. cop coocsseo APO ASP CBMs goac006 30-67-°5 X 3-75-5

G-septatempe cena: 4-5 SMe Biaeecearaate 6 52-70 x 5.

Deseptate ct oe cis) os 90 Pepi! ich atch eee 42-62:5 x 3-5, mostly 50-

57-5 x 3-4-5.

4-septate:..........- 2 42-5-52-5 x 3-75-5.

S-septate.|...5.---..--. 0: ee ate tc 30-35 x 3-5-5.

364

Bean pod, culture 14 days old, conidia from pionnotes—

eae OI) SD).

37:5-60 x 4-4-5, mostly 40-

G=sepbtate samen. sae Me Witaeqierv yo cese<

D-SEDUANOs0cosooevece BEE OEP CEM. 6 dacs ooo:

D0) <A>,

4eSepbabeememateccttie Tan nae — delaras 37-5-51 x 4-4-5.

A sp ectca Aen ORS 30-37-5 x 3-5-4

3-septate.

Melilotus, stem, culture 14 days old, conidia from pionnotes—

1-5 per cent........ 55-62°5 x 4—

GSE|DUANIOr coccovcsesos

DAG OUAWO. co0bscoccce

4-septate

90 >»

BD 5

15),

32D —O21D) <3 10=D:

30-45 x 3:75—-5.

25=al DX Be =45.

BNEW. coc coveces

On standard eynehetic agar plus starch, the conidia from pionnotes were almost all

5-septate, 40-72-5 x 3-4-5; “other septations were rare. On oat agar, the conidia were

also 99 per cent. to 100 per cent. 5-septate ; they were 40-67-5 x 3-4-5.

Fusarium scirpi Lamb. et Fautr. var. compactum Wr

Wollenweber, Fusarium-Monographie, 333, 1931; Fus. aut. del. 923-925. Wollenweber and Reinking

Die Fusarien, 66-67, 1935.

Syn. Fusarium scirpi Lamb. et Fautr. v. compactum f. 1. W:

. =. \,

“S \ = ae CN

} ie S » \ SRA \ NN

; ia) \ \ \ \\

Verh Yo \ yey FO \ »

\ : \ ha } rea ly

{ = aN j 4 | | i J po |

| ma | p -| | } Med es Han

ee , ae J

eet I~ [a ine j | j

Kf i - j } j

\ j Is | Ly } P jp | =

‘6 S f di

“J | ; AS j y } j

= } qs / 7 /

my Vif / 7 Z fr py

/ k VA g | ed

att yo dy

] or Vf

f a” —_

’ Ss

LP +

“gh > €

A 7 ar y - eo

A f~N ty Ake \

= f WN = “a * } de

bP oot A f \ L Fran e.: }

/ Bs aN \ d & ye \ Aw

fa ey te e | Veet \

| i | ioe ve L Poe

= \ le Neje

| \ { .

\ ‘ Ls

\— | 7

: od | hi awe

i / 4 es | S mh

f ] %s

j j Sey

2 AX

sity

! Hi, masa

} f \: ¢ bess Z

Encore lal

Fusarium scirpi Lamb. et Rautr. v. compactum Wr.; Conidia from (a) sporodochia of culture on bean

} }

es 2 ag

; “arnt > | r . C01

pod, (b) pionnotes of culture on standard synthetic agar plus starch, (c) pionnotes of culture on oat agar,

(d) pionnotes of culture on hard potato agar. Culture all 2 weeks old.

365

This variety differs from Fusarium scirpi in the form of its macroconidia, which are

more compact ; they are comparatively short and broad, and the apical cell is not drawn

out into a filamentous process. Conidia mostly 5-septate, more rarely 3-4-septate, seldom

0Q—2- or 6—7-septate.

3-septate. WHO) >< BD Os 55 60. Mostly 25-31 x 4-2-5-4.

5-septate. 30-55 x 3:7-6:5..... Mostly 36-47 x 4-3-6.

WESC Pate samc secre: 31-02) x 4:5-6....... Average 42 x 5.

On media rich in carbohydrates, the stroma may assume a carmine red and golden

yellow colour, which sometimes deepens to brown ochre. The red colour is sometimes

pronounced, but may be weak or absent. The form previously described as f. 1 has no red

colour in the stroma. Chlamydospores like those of the type, mostly intercalary, and often

distinctly verrucose when mature, especially after drying.

Hab. Antirrhinum majus L., from stem of wilting plant, Pretoria (Wager).

Campanula medium L., from stem of wilting Canterbury Bell, Kimberley( Wager).

Citrus sinensis Osbeck, from fruit held in storage for 12-18 weeks; from stem end rot

of Valencia oranges from Rustenburg and Zebediela, Transvaal, and from navel end rot

of navel oranges from Letaba, N. Transvaal.

From bark of branch affected with scaly bark, Mazoe Estates,

Rhodesia (Bates); from citrus trunk showing extensive bark lesions and some gummosis,

Letaba, N. Transvaal.

Cucumis sativus L., from cucumbers shewing soft rot and leaking, Nelspruit (Wager).

Lathyrus odoratus L., from stems of wilting sweet pea seedlings, Brooklyn, Pretoria

(Wager), M. H. 28416.

Limonium sp., from stems and crowns of dying Statice plants, Nelspruit (Wager).

Matthiola incana R. Br., from discoloured stem of stock plant, Johannesburg (Wager).

Papaver Rhoeas L., from stems of Shirley poppy, which was yellowing and dying (ass.

Rhizoctonia), Pretoria (Wager).

Striga lutea Lour., from stems of dying plants, [xopo, Natal (Mack).

Growth on Standard Media.

in culture this strain resembles Fusarium scirpi, except that in some strains there is

carmine or yellow colour in the stroma. In the strains on Campanula, Cu.cumis and Limo-

nium, the stroma on oat agar and standard synthetic agar plus starch was eugenia red to

pomegranate purple, and it was carmine on potato agar plus 5 per cent. dextrose; the

red colour was more definite in some sets of cultures than in others.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 14 days old, conidia from pionnotes—

US OUND cabosacatoos Rare.

6-septate............ Rare.

HHSO)OUNS 6soccocasace SOspericenitnar eran cre. 30-47-5 x 3-75-6.

A-Sep tater acm. 1s ae Sie Me later sycelits 25-37-5 x 3-75-6.

S=SEPUabOravmrvceiscmrs sae OM gy 555 sh apare 2 seuclhs 21 -5-30 xX 3-75-5.

Hard potato agar, culture 14 days old, conidia from pionnotes—

USO MAUDcccosccecsoe IL OC CBiMiio goo ao bc 46-55 x 5.

DEON oc occovoocone 1H 4 cobosoae 52:5-53 xX 9.

DHECUAUS ccoosccacooe Gill EP ie cee oe tt 30-56 x 4-6.

ALESCIUEU@andonesodhoo I Lak ay ae Ome eaatator 30-45 x 4-5

GHIIMENOsoccegnaonsa 40 PAL. Une cYsielsiGa=ehs 20-475 -x 3-75-85

366

Potato plug, culture 14 days old, conidia from pionnotes—

SE) MAND o coo00cco nc OF [OSE Clio soca0c00d 27-5-37-5 x 5-6.

A-septate sche seas 15 sinh) Sl oan! Se ae 20-35 x 4:5-5.

SHS MEMS coccodadeeec 22 Patna Mee eter ies 20-27:-5 x 4-5-5.

Plain agar plates, culture 12 days old, conidia from mycelium—

NOAIAASI NBT occosscceoce Y [AC COilllcnncocnoce 5d-T0 x 5-6.

f-S-septate Aa wanteten LOU An sheigl ate. eo Ry. 50-60 x 5-6.

Ho Osseptatel aden cictisle, / OOM wav sy etartts eee 35-50 x 4-5-5.

3-4-septate............ 10 = a es bus ACCA REE 30-35 x 5.

The figures given above are for conidia developing on one set of plain agar plates ;

on this occasion the conidia formed had a higher number of septations than on any other

medium.

Chlamydospores numerous, usually in chains or clusters. In potato cultures 4 weeks

old, they were smooth and hyaline, becoming brown and verrucose as they matured, single

spores being 10-15 ~ diam. Chlamydospores also occurred in the conidia in some of the

older cultures.

Fusarium scirpi Lamb. et Fautr. var. acuminatum (Hl. et Ev.) Wr.

Wollenweber, Fusarium-Monographie, 335, 1931; Fus. aut. del. 165-168, 170, 569, 930-933. Wollen-

weber and Reinking, Die Fusarien, 67—68, 1935.

Syn. Fusarium acuminatum Ell. et Ev. ; Fusidiwm aloes Kalch. et Cke.

Fusarium arcuosporum Sherb. ; F. erubescens App. et Ov.

F, ferruginosum Sherb.

F.. hippocastani (Cda.) Sace. ; Selenosporium hippocastani Cda.

Fusarium lanceolatum Pratt; F. pseudoeffuswm Mur.

F. russianum Manns; I. sanguinewm Sherb. (non Fries.).

= & a

x é

> ~

Q, b.

j as

: | |

i rs

j \ \

{ 5)

G, y JI

/ : (4,

— / se

} |

i a

{ =

| a. We © /

{ \ | \. 4

' » ~

\ ~\

ms | 4

\ . Seas

AN \ ~ é S .

Z al

Hie. 12:

Fusarium scirpi Lamb. et Fautr. v. acuminatum (E11. et Ev.) Wr. conidia from pionnotes of 8 weeks

old cultures on (a) hard potato agar, (b) potato agar plus 5 per cent. dextrose, (c) bean pod, and (d) standard

synthetic agar plus starch ; (e) chlamydospores from 4 weeks old culture on hard potato agar.

367

Stroma plectenchymatous, of various colours, blood red, purple, yellow, sometimes

sclerotially erumpent and dark blue, brown or pale. Aerial mycelium white or pink. Conidia

in sporodochia and pionnotes, orange-coloured, falcate, tapering at both ends, apex more

or less elongated, base pedicellate or papillate, occasionally rounded to truncate, 5-septate,

less frequently 3-4-septate, exceptionally 0-2- or 6-7-septate.

0-septate .. ; LE AVAD< Opens eect Average 8-6 x 3.

Jeseptate= cae... -s-8 LIRL2) x22" 2-473... Average Ib x 278.

d-septate............ 16-44 x 2-5-4:5..... Mostly 24-37 x 3-4-2.

5-septate.......:.... 28-6] x 3-5-3 Mostly 34-54 « 3-1-4:3.

(-Septabeprarios tah 45-84 x 3-2-5-2..... Mostly 41-61 x 3-8-5.

Chlamydospores intercalary, mostly in chains and clusters, seldom terminal, in conidia

often 1- or 2-celled, spherical, 7-20 ~ diam., 1-septate 20-30 x 10-18; brown in mass.

This variety has much the same distribution as the type, and occurs on a number of

plants in almost all parts of the world. The ascus stage has been observed in Europe and

Australia on Acer, Dahlia and maize. It may be briefly characterised as follows :—

Gibberella acuminata Wr.

Wollenweber and Reinking, Die Fusarien, 68, 1935. Wollenweber, Fus. aut. del. 1107, 1108.

Syn. Gibberella saubinetiz (Dur. et Mont.) f. dahliae Sacc.

Nectria dahliae Rich.

Perithecia olive-green to blue black, spherical to conical, rough, 0-3-0-5 mm. diam.,

single or in small groups, loosely attached to the olive-coloured stroma. Ascimostly 8-spored.

Spores fusiform, broadly conical at both ends, slightly curved, 3 (1-3) -septate.

S=se Puaberynn vie ates 17=—36 >< 4-9)... eis ele Mostly 22-26 x 5-6.

i-septatessncmassaae- ld-20>< 4-95... 28. Mostly 16-22 x 5-8.

The ascus stage has not been found in South Africa, but the conidial stage has been

isolated from several hosts.

Hab. Dianthus caryophyllus L., from stems of carnation plants affected by foot rot or

wilt (in the latter case associated with F’. dianthi), Bethlehem, O.F.S., Sterkstroom, Trans-

vaal, and Estcourt, Natal, Feb. 1936.

Phaseolus sp., from stems of bean plants which made good growth but set no seed,

Premier Cotton Estates, Mvamba.

Solanum tuberosum L., from “ seed” tubers, imported from Hamburg, Germany.

Zea Mays L., from base of stem of plant affected by foot rot (f. moniliforme also

isolated) Lourenco Marques (Fuller), M.H. 23222; Premier Mine, Transvaal (Leemann).

Fusidium aloes Kalch. et Cke. (Grevillea, 22, 1880), which is probably asynonym for

Fusarium scirpi v. acuminatum, was collected by MacOwan in South Africa in 1879, on

the leaves of Aloe arborescens (Herb. MacOwanianum No. 1170; Wr: Fus. aut. del. 167).

This number is unfortunately missing from the collection of MacOwan’s fungi in the Crypto-

gamic herbarium.

Growth on Standard Media.

Oat agar : Aerial mycelium abundant, cottony, mostly white, but with patches of yellow

ochre where it touched the tube. Growth in substratum spinel red to pomegranate purple.

Hard potato agar: Mycelium not very plentiful, tufted, cottony. Pionnotes and

small sporodochia developed in 2-4 weeks, they were pinkish cinnamon, and developed

directly on the substratum; they were partially concealed by the aerial mycelium.

Standard synthetic agar plus starch : Aerial mycelium scanty, white, cottony ; growth

in substratum isabella colour to old rose and spinel red, or cinnamon rufous in places.

After 6 weeks, pinkish cinnamon pionnotes developed in concentric rings round the point

of transfer.

Potato agar plus 5 per cent. dextrose : After 14 days, aerial mycelium was very plentiful,

white to ochraceous buff and Chatenay pink, or occasionally geranium pink ; after 4 weeks

it was white to spinel red.

368

Potato pl-g : Plug covered with a dense, matted mycelium, which was white to sea-

shell pink, with patches of chamois colour at the base, and pinkish cinnamon where it

touched the glass. Growth on substratum was eugenia red to acajou red.

Melilotus stem: Stems covered with a copious aerial mycelium which was white or

tinged with pink.

Bean pod : Pod covered with a copious aerial mycelium, which was cottony to arach-

noid and tufted ; 1t was white to vinaceous buff and vinaceous pink.

Rice: In cultures 10 days old, the mycelium was white at the surface of the medium ;

below it was mustard yellow, and, at the base, eosine pink to begonia rose. The pink colour

faded after 4 weeks. The mycelium between the grains was then mostly yellow ochre, and

the growth in the substratum mummy brown.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 8 weeks old, conidia from sporodochia

and pionnotes—

DATUM ccosooccoccc WS) JO COM, ooo a0ccc 28-75-59 x 3-6.

MKT socsocanocoe oO ie eae de Aue Ne 30-42-5 x 3-5.

3-septate . . eet Dh O Rott ASR ee 22-540 x 3-5.

O-septate .. Arce 4 og AR ee AG 13-15 x 3-3-25.

Hard potato agar, , culture 8 weeks old, conidia from pionnotes—

6- -septate. sialon) Sota eit oud BD) [SIE OSMs ocascos Al 40) << D6),

H=septatens qoman eae OO pp Pht ety tues 25-41-95 x 5-6.

AOU coccoescccae IO i Ab Ee eh es 20-40 x 4:5-5:5.

CHSC desoccoccns MO py ob occoeds 25-38 X 3:70-4:-5

SCTE « ocd ooo an 2 Saree Aes Mp) sx Ab

IRCA ous0005'00600 3 Ra SE i Fr Se 12:5-20 x 3-75.

OFSCDUAUO > ooccccc0cd 1 BE oro cnorg 15 x 3-3:25.

On plain agar plates, chlamydospores developed aiter 6 days.

interealary, in chains or irregular clusters ;

becoming olivaceous at maturity.

They were mostly

single cells were 5-10 4 diam., sparsely verrucose,

Fusarium scirpi Lamb. et Fautr. var. filiferum (Preuss) Wr.

Wollenweber, Fusarium-Monographie, 337-338, 1931 ;

and Reinking, Die Fusarien, 69, 1935.

Syn. Fusarium filiferwm (Pr.) Wr. ;

Fusarium caudatum Wr. v. solani Sherb. ;

Fus. aut. del. 219-222, 601, 936. Wollenweber

Fusoma filiferum Preuss.

F’, equiseticola All.

Fusisporium incarnatum Rob. vy. tussilaginis farfarae Sacc.

Fusarium mycophytum (W. G. Sm.) Mass. ;

Fusarium osteophilum Speg.

Fusisporium mycophytum W. G. Sm.

Stroma effuse, sometimes sclerotially erumpent, and then brown; aerial mycelium

white, floecose. Conidia in sporodochia or pionnotes, ochraceous or amber yellow to brown

ochre, with whip-like elongation of the apical cell, and base with a long foot. Conidia

5—7-septate, less frequently 3-4- or 8-10-septate, exceptionally up to 12-septate. In young

cultures, subnormal, Fusisporvwm-like conidia are found scattered in the mycelium ; they

are smaller, oval to fusiform, rounded to conical at both ends, straight or somewhat curved.

U-SEOUENE vcccacoacane

50-114 x 3-2-

O;septates: 25.45. 44 sen Ol OM xerial —dasven ye ammpANVETACC climxXaouos

I-septate.. 5. t.ho sh So lOM xed —ter nse ae Average Sexe td /0)

3-septate............ 19-40 x 2-5......... Mostly 23-34 x 2-5-4-4.

D-Sepuatene eke 22-87 x 2:5-6....... Mostly 35-76 x 3-3-4:-5.

x 3:17-5:11

x 4-5

CMU cooosvceeeue

IZ -Sepudteneee senate

58-121 x

60-132 x 3°!

3°5—

D-5°5.

Mostly 57-84

Mostly 77—90

369

Chlamydospores 6-16 y in diameter, round or oval, intercalary, smooth or verrucose,

usually in chains and clusters.

Hab. Allium Cepa L., on rotting stems of onion seedlings, Pyramids, Pretoria dist.,

March 1932 (Mogg) (seedlings dying from attack of Ff. oxysporwin f. ay

Nomadacris septemfasciata, recorded as occurring on red locusts in South West Africa

(Wollenweber and Reinking, loc. cit.).

This variety occurs on decaying parts of plants, and on other fungi, in Europe and

North America.

>» => f ae

Ifa, 118},

Fusarium scirpi Lamb. et Fautr. v. filiferwm (Wr. et Rkg.) Wr. ; Conidia from (a) pionnotes of 4 weeks

old culture on hard potato agar and (b) sporodochia of 4 weeks old culture on oat agar.

Growth on Standard Media.

Oat agar : Aerial myceluim fairly abundant, white, tufted, cottony. Growth in sub-

stratum at first colourless, with a tinge of pinkish buff at the base of the slant; after 4

weeks, the stroma at the base of the slant became sclerotially erumpent and brown. Sporo-

dochia few, scattered, cream buff to ochraceous buff.

Hard potato agar : Mycelium white, cottony, not abundant. Growth in substratum

colourless. Sporodochia resembled those on oat agar.

Standard synthetic agar plus starch: Aerial mycelium sparse ; cultures 14 days old

had a tinge of pinkish buff in the substratum at the base of the slant, which faded to ochra-

ceous tawny in 4 weeks.

Potato agar plus 5 per cent. dextrose: Aerial mycelium fairly abundant, at first white,

later flecked with brown. Growth in substratum at first pinkish butt, becoming ochraceous

tawny to snuff brown after 4 weeks ; stroma sclerotially erumpent at the base of the slant,

as on oat agar. Pionnotes along needle track, russet colour.

Potato plug: Plug covered with fine, cottony mycelium, which was white at first

but became brownish with age. There were mummy brown patches in the substratum.

Melilotus stem : Stems covered with a very vigorous, white, cottony mycelium, which

completely concealed the sporodochia which developed round the point of transfer.

Bean pod : Pods covered with a fine white mycelium. A number of small sporodochia

developed i in a group round the point of transfer, elsewhere they were scattered ; sporo-

dochia cinnamon colour.

Rice: Growth at first white to pinkish bufi, gradually becoming brown.

370

Measurements of Conidia.

Hard potato agar, culture 14 days old, conidia from sporodochia—

(BANA coccnccnecoe BD OOP CBMs ccscoov0e 56-719 x 4-5-5.

GINA socesaobonca LO Boda. pon nove ee 55-15 x 3-5.

BSCE occccccccaac al) So Pane ee 45-72°5 x 3-5.

4-septate............ ANG ay, Wins neahin Secrees 42-5-87:5 X 3-4-5.

3-septate . . Dee een 2 Mie ear ean 32:5-40 x 2:8-4-4.

Bean pod, culture 14 days old, conidia from sporodochia—

SSI NIDIDs ccovo 50000 FUaTe) seas to ae 57-5-60 x 4-7-5.

QGIUM@>ccsccccseeo — O [XE COMossocsosoe 52:-5-60 x 3-75-5

Heseptatem scene sy 5 CO Bae tabedscs Naren oie 35-59 x 2:8-4-7.

ASO MUAbe mee ete Or ise, mn eee tye 35-37°5 < 2:°8-4-7.

SECON cooassgosane 2 Fr Peed ahr stag ee A 32°5-47-5 x 3-3:75.

Section DISCOLOR.

Wollenweber, Fusarium-Monographie, 346, 1931. Wollenweber and Reinking, Die Fusarien, 69-70,

1935.

Macroconidia comparatively thick-walled, fusiform-falcate, tapering at both ends,

curved (dorsal side convex, ventral side less curved, usually concave but occasionally some-

what convex) ; apex constricted like the neck of a bottle, curved and rostrate, or conical

to truncate or rounded ; base pedicellate, when fully developed and mature. Sporodochia

and pionnotes ochre, salmon pink or orange. A few species have a Fusisporium stage,

with smaller or medium-sized conidia, which are apedicellate, 0-3- or more-septate, oval,

fusiform to cylindrical, straight or curved ; these forms may predominate, or may disappear

with the formation of sporodochia (as in F. trichothecioides). Other species have some com-

paratively slender conidia, and some more compact (fF. heterosporum). The stroma is flat,

effuse ; it is plectenchymatous, here and there sclerotially erumpent, and varied in colour ;

it may be pale, carmine to purple red, yellow, brown, or rarely blue; in a few forms it is

pale and homogeneous. Spherical sclerotia may be present or wanting; when present

they are blue, brown or colourless. Aerial mycelium well developed, white, pink, or tinged

with the colour of the stroma. Chlamydospores few, terminal or intercalary, single, in chain

or in clusters ; brown in mass. It has been established that the ascus stage of some of the

species 1s Gbberella.

Sub-sections of the Discolor-Fusaria.

A.—Apedicellate conidia of the Fusisporium-stage pre-

dominant. Mycelium floccose, Trichotheciwm-like...... Trichothecrordes.

AA.—Pedicellate conidia predominant, developing in pion-

notes and sporodochia.

B.—Triseptate conidia typically 3-4-1 mw thick....... Neesiola.

BB.—Triseptate conidia typically 4-1-7-9 w thick..... Saubinetii.

Key to the South African Species.

A.—Macroconidia 4-5 (5-5) u thick, 3-5-septate.

B.—Stroma carmine to purple red, chestnut brown,

yellow or pink.

C.—Conidia not typically heterosporus, usually in

sporodochia.

D.—Conidia comparatively compact, in sporo-

dochia and pionnotes.

—Conidia mostly 3-, seldom 4—5-sept. :

3-sept. 25 x 4:9: 5-sept. 30 x 5:3 F. sambucinum f. 2.

EE.—Conidia 3-5-sept.................. FE. sambucinum.

DD.—Conidia comparatively elongated, slender ;

GOMES S-PAGWe ooscec oocovesvcesecaen F. graminearum.

CC.—Conidia typically heterosporous, compact or

Slenderevaaae: scr posits crater ni rae ea as vase F. heterosporum v. congoense.

BB.—Stroma not becoming carmine to purple red...... F. sambucinum f. 6.

AA.—Macroconidia 5-9 w thick, 5 (3-5-7) -sept.......... F. culmorum.

Sub-section NEESIOLA.

Wollenweber, Ann. Myc. 15: 2, 1917; Fusarium-Monographie, 346, 1931. Wollenweber and Reinking,

Die Fusarien, 70-74, 1935.

Stroma floccose, effuse, often covered by a pionnotal layer, flat, more rarely sclerotially

erumpent. Sporodochia formed less frequently than pionnotes. Conidia slender, 3-4 wu

thick, more or less 3-septate, salmon colour, reddish or orange in mass, becoming brick red

when dry. Mycelium yellow or flesh colour, rarely carmine. Chlamydospores intercalary

or none.

Fusarium heterosporum Nees var. congoense Wr.

Wollenweber, Fusarium-Monographie, 350, 1931; Fus. aut. del. 306-307, 612, 1140, 1141. Wollen-

weber and Reinking, Die Fusarien, 73, 1935.

Syn. Fusarium congoense Wr.; F. congoense v. septatius Wr. (nom. nud.).

F. heterosporum Nees v. congoense f. 1. Wr.

Sporodochia orange to brick red, gelatinous, early coalescing to form a pionnotal layer.

Conidia typically fusiform to faleate, some compact, others slender, curved, tapering at

both ends, pedicellate ; apical cell in the more compact conidia constricted or rostrate ;

in the more slender forms tapering gradually and curved; the slender forms approach

the Roseuwm type. Stroma loose, floccose, with abundant aerial mycelium which is white,

or citron yellow to sulphur yellow and flesh colour; plectenchymatous layer on the sub-

stratum carmine red. Conidia scattered in false heads or in sporodochia and pionnotes,

borne on conidiophores which branch more or less freely. Conidia mostly 3-5-septate,

seldom 0-2-septate, or 6—10-septate.

3-septate, 22-40 x 2-7-6, mostly 26-39 x 3-1-5-2 (some compact, av. 29 x 4:8,

others more slender, av. 33 = 3-4).

5-septate, 29-45 x 2-7-7, mostly 32-42 x 3-1-5-:7 (some 37 x 5:2, others

43 x 3-6).

7-septate, 38-54 x 3-6, mostly 41-45 x 4-1-5-5.

9-septate, 56 x 5-5.

Chlamydospores intercalary.

Hab. Brachiaria brizantha Stapf, on ovaries, Experiment Station, Barberton, Transvaal,

April 1914 (Mogg) M.H. 7771; Salisbury, Rhodesia, March 1919 (Eyles) M.H. 11858.

Brachiaria sp., on ovaries infected with ergot, Salisbury, Rhodesia, Feb. 1915 (Walters)

M.H. 8868.

Bromus unioloides H.B.K., on ovaries infected with Ustilago bromivora, Rietpoort

Zandspruit, Wakkerstroom Dist., Transvaal, April 1907 (Gillespie) M.H. 284-285. (This

is apparently the type collection, quoted by Wollenweber in the Fuasrium-Monographie,

p. 350, as from “ Zandspruit, Wakkerstroom Dist., Congo, Central Africa, Vanderyst ie

284-285”; the fungi in the National Herbarium were at one time distinguished by the

letter F.).

Cynodon dactylon Pers., on ovaries infected with smut, Skinner’s Court, Pretoria, Feb.

1918 (Mogg) M.H. 11673; without locality (Burtt Davy) M.H. 577.

Digitaria eriantha Steud., on ovaries, Butterworth, Cape, April 1914 (Pegler) M.H.

7743.

372

Digitaria monodactyla (Nees) Stapf, on ovaries infected with ergot, Groenkloof,

Pretoria, Dec. 1919 (Pole-Kvans) M.H. 11874.

Digitaria Pentzw Stent (Woolly Finger Grass), on ovaries infected. with ergot, Durban,

Feb. 1929 (Clarkson) M.H. 23684 ; Prinshof, Pretoria, April 1930 (Liebenberg) M.H. 25369.

Hyparrhenia hirta Stapf, on ovaries, Garstfontein, Pretoria Dist., March 1915 (Pienaar)

M.H. 8905.

Panicum laevifoliuwm Hack., on ovaries infected with smut, Tzaneen, N. Transvaal,

April 1906, M.H. 6.

-2 ty 2 . sare

Hie. 14.

Fusarium heterosporum Nees v. congoense Wr.; (a-b) conidia from natural host, (a) stout form, (b)

slender form ; (c-e) conidia from cultures; (c) from mycelium of 3 weeks old culture on plain agar, (d

from sporodochia of 8 weeks old culture on rice, (e) from sporodochia of 8 weeks old culture on oat agar-

313

Panicum maximum Hack., on ovaries infected with ergot, Kentani, Cape, May 1913

(Pegler) M.H. 6649 and 6919; Butterworth, Transkei, April 1914 (Pegler) M.-H. 7738 ;

Maritzburg, Natal, April 1914 (Sim.) M.H. 7760; Groenkloof, Pretoria, Feb. 1915 (Pole

Evans) M.H. 9058; Moodie’s Estates, Barberton, Transvaal, March 1932 (Wager) M.H.

26152.

Pennisetum cenchroides Nees, on ovaries infected with ergot, Prinshof, Pretoria, March

1934 (Mogg) M.H. 26147.

Setaria mgrirostris Dur. et Sch., on ovaries, Leeuwpoort, Carolina Dist., (Burtt Davy)

M.H. 480.

Setaria perennis Hack., on ovaries, Groenkloof, Pretoria, Feb. 1919 (Phillips) M.H.

11878.

Setaria sphacelata Stapf et Hub., on ovaries infected with ergot, Garstfontein, Pretoria

Dist., March 1915 (Pienaar) M.H. 8906.

Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehd. (= Andropogon sorghum),

on ovaries infected with smut, Sphacelotheca sorghi, Clercqsvlei, Moedig, Transvaal.

Growth on Standard Media.

Oat agar: Aerial mycelium rather sparse, white, tufted, cottony; after 4 weeks,

growth on substratum barium yellow. A number of small, scattered sporodochia developed

on the lower part of the slant, and were bittersweet pink in colour.

Hard potato agar: Aerial mycelium sparse to moderate, thin, cottony, white. Very

numerous minute sporodochia developed on the lower part of the slant, and were salmon

colour.

Standard synthetic agar plus starch: Aerial mycelium fairly plentiful, fine, cottony,

white, or tinged thulite pink and naples yellow ; growth on substratum spinel red in places.

Groups of sporodochia 0-5 to 2-5 mm. in diameter, developed in 4-8 weeks, and were salmon

orange to bittersweet pink.

Potato agar plus 5 per cent. dextrose: Aerial mycelium dense, rather tufted, cottony,

at first white to safrano pink, grenadine pink and chamois colour. The colour faded some-

what after 14 days, and was then white and pale salmon colour. Growth on substratum

eugenia red to pomegranate purple and Bordeaux red. Sporodochia not numerous, bitter-

sweet pink.

Potato plug : Aerial mycelium very abundant, felt-like, white to deep rose pink; growth

on substratum pomegranate purple. No spore masses were observed.

Melilotus stem: Aerial mycelium moderate to copious in amount, white, cottony.

Sporodochia orange pink.

Bean pod: Mycelium rather abundant, cottony, white, or with patches of salmon

buff in places. No conidial masses observed.

Rice : Growth white to spinel red and pomegranate purple ; in a second set of cultures,

after the fungus had been in culture for some months, the growth was white to flesh colour.

In 4 to 8 weeks, a number of sporodochia developed ; they were bittersweet pink.

Measurements of Conidia.

A.—Direct from the host.

M.H. 284-285, on ovaries of Bromus unioloides (type collection).

Conidia mostly 5-septate, some 3-4- and some 6—7-septate.

GIBCO oo0c oo boc oud 37-40 x 5-5-5 or 40-57:°5 x 4.

DSCIOUENHS 5. 660666000000 30-45 x 4-6-25 or 32-5-42-5 x 3-4.

ARON sc dogo 09.00 00 0.0 30-37:5 « 3-7-4.

SAGE. coos auooeoc 0c 25-27:5 x 3-7-5 or 32-5 X 3.

3T4

M.H. 8905, on ovaries of Hyparrhenia hirta.

Conidia mostly 3-5-septate, some 6—9-septate.

SICHUAN ooocaccedecoas 55-57-5 x 5-5-6.

GH NAN codsocogooanns 45-52-5 x 6-25.

USED oocooascoocace 45-55 x 5-6.

G-septate.s..55. 242022 O20 —41-0) Xa —Oe 20.

d-septate.............. 30-47-5 x 4-5-6 or 37-425 X 3:7.

A-Septatenadecne reer 30-32°5 x 4-5.

SHI OUMOceocooceaoavecc 27-5-37-5 x 4-6.

M.H. 11673, on ovaries of Cynodon dactylon.

Conidia mostly 3—-5-septate.

CHSC MEWS c oosacensoodoc 37-5-40 x 5.

5-septate.............. 32°5-42°5 x 5-5-6-25 or 35-37:5 x 3-7-4.

CERNE cooscacccsouee 30 x 5 or 37-5-40 x 3-7.

S-Sepuate 4.545.505 44) 30) xX Lon 2/540 x 3S3o7.

M.H. 11878, on ovaries of Setaria perennis.

Conidia mostly 5-septate, 3-4- and 6—7-septate fairly common, some 8-10-septate.

8—l0-septate.............. 42-5-57-5 x 5-5-6-26:

(ESC Dud Lenten anna ner 37-5-50 x 5-6.

DANN aasonsocesanca ~All =50) < D6 ore Sowa) XK Bove,

GANINUENMOsocunsoasaasoa Bl BAKO) $< diets),

d-septate.............. 22°5-27°5 & 5-5-5 or 25-27-53 x 3-3-7.

B.—From culture derived from M.H. 23684, on ovaries of Digitaria Pentziv.

Maize stem, culture 6 weeks old, conidia from sporodochia.

OI NEWI5ossonccccde Ue) | Je CWMisosococe 37-45 x 3-4-4.

A-septateucka. cacths ee On Slee) aes 32°5-45 x 3-3-5

SHOWS ccoccaoscode 2S es a aa 25-35 x 3-3°5.

ISOS 6 50000400000 RSCG, eR een 25-35 < 3-3-5.

Hard potato agar, culture 6 weeks old, conidia from sporodochia.

(CSepucbe mere eres IL {OEP Cin sco oces 51-55 x 2-8-3-75.

OMB ocooocccoeca dl aCe Ie mye pest ce 52-5-57-5 x 2-8.

DAG OUMOccccacccccss WAU) Pee Naclenia Caren 32°5-55 x 2:°5-3-75.

HSH OU Mee neid Ci onal AIRY Al COO lame ety ing sugeantee 30-47°5 x 2:5-3-75.

SHICOWN@ covccocovccc 24 Se NO eeteastarac 22-5-40 x 2-3-3-79.

Oat agar, culture 8 weeks old, conidia from sporodochia.

OAINEWEcooossccosec All OOP COMWogconccces 32°5-55 x 3:75-4:-7.

A-septate............ 30 PANT geen anual 37-5-5T°5 xX 2:°5-2-8

DSC plate. cain. Geri (BOLI N in | Magee ort 37°5-51-5 x 2-5-2°8

MANN Ocsconesooace db gp

Slender forms predominated in culture; the following are measurements of conidia

of the same strain from pionnotes on the ovaries of Digitaria Pentzr

6-S:septatel:. ..0 tcl. son PROC ph er Menietr tickle 42-5-55 x 3-75-0.

DAMM occcoccesccne Bll ROP CAMs ocavec0ce 30-45 x 3:75-5:25.

4-septatersuducinwerseer a) UObenm th Usiciutauamitnes 27°5-40 x 2-8-5.

GAN ccaccvacecos 44) Ri La ae reenter 20-50 x 3:5-4:75.

AATUM Bs scoocevadces 2 SH kumG ries eete oat 22-5-37:5 « 2:8-3:-75.

I=sep tater cceey eee Ol Wale) eineeeteM en eras 30-37°5 x 2:5-2:-8.

A few intercalary chlamydospores were seen on potato agar plus 5 per cent. dextrose ;

they were 6-10 w diam. and rough walled.

375

Sub-section SAUBINETII.

Wollenweber, Fusarium-Monographie, 346-347, 1931. Wollenweter and Reinking, Die Fusarien, 70,

75, 1935. .

Differs from sub-section Neesiola in the larger conidia, 4-59-13 in diameter, and

3-5-7-12-septate, some compact, some more elongated, generally rather pale, yellow-white,

pale orange or ochraceous. Stroma carmine or yellow.

Fusarium sambucinum Fuck.

Fuckel, Symbolae myc., 167, 1869. Wollenweber, Fusarium-Monographie, 352-356, 1931; Fus.

aut. del. 311-320, 322, 323, 607, 1142-1144. Wollenweber and Reinking, Die Fusarien, 75-76, 1935.

of y

¢ A

ra 2

C a

: So

> : -

QO a

\ - ‘

/ )

bi = ~~

>»

v e ce

{

a Ne ‘

: Be

cs)

~ ¥ y Rs

f ~ iil aah)

f .)

o) <3

Fig. 15.

Fusarium sambucinum Fuck. ; Conidia from (a) sporodochia of culture on bean pod, (b) pionnotes of

culture on oat agar, (c) pionnotes of culture on Melilotus stem, (d) sporodochia of culture on hard potato

agar, (e) pionnotes and (f) mycelium of culture on standard synthetic agar plus starch, (9) mycelium on

oat agar, (a-b) from cultures 2 weeks old, and (c-g) grom cultures 4 weeks old.

376

Syn. Fusarium aridum Pratt; F. Delacroixii Sace.

F. fraxini All.; F. discolor App. et Wr.

F. discolor App. et Wr. v. triseptatum Sherb.

F. granulare Kalch. ; F. herbarum (Cda.) Fr. v. conii-maculati Roum. pr. p.

F. hordei (W. G. Sm.) Sace. ; F. maydis Kalch.

F. pannosum Mass. ; F. pulvinatum (Berk. et Br.) Sace.

F. ricini (Bér.) Bizz. ; F. roseum Link. pr. p.

FE. sambucinum v. medium Wr.; F. subcarneum Crouan.

F. tenellum Sacc. et Briard; F. tenwissimum (Peck) Sacc.

Microcera tasmanica McAlp. ; Discofusarium tasmaniense Petch.

Pionnotes vagans Speg.; Fusarium violaceum Crouan (non Fuck.).

Conidia fusiform-falcate, curved, somewhat abruptly bent inwards at both ends, con-

stricted or conical at the apex, pedicellate at the base, thick-walled. Macroconidia borne

on the aerial mycelium are sometimes mixed with 0-septate, subnormal conidia of different

form. Aerial mycelium at first white, then golden yellow or pink. Macroconidia in sporo-

dochia and pionnotes pink to salmon and orange red in mass, sometimes carmine red to

chestnut brown or ochre by absorption of the colour of the plectenchymatous or sclerotially

erumpent stroma. Conidia 3-5-, seldom 6—7-septate.

SH NUEWaaseeecocess NOAM) Xx GO occ sccs5 Milosidhy AGG) Xx Bolom,

DAVIE cneseoossss AAO) XK PWD...2.-. Milosdhy AI XK Sse,

URE OBWAs.gcekooeces CO09 KX GVO o.c0-. ovemnge 42) < Hor,

The sclerotial stroma often breaks out in rough, cauliflower-like, stilboid bodies, which

are up to 1 c.m high, and mostly dark brown in colour. Chlamydospores comparatively

rare, intercalary, spherical, single, in chains or in clusters.

The ascus stage of Fusarium sambucinum, which has not been observed in South Africa,

may be briefly characterised as follows :—

Gibberella pulicaris (Fr.) Sacc.

Saceardo, Michelia 1:43, 1877. Wollenweber, Fusarium-Monographie. 353-356, 1931; Fus. aut.

del. 27-29. Wollenweber and Reinking, Die Fusarien, 76, 1935.

(Fer complete bibliography and synonymy, see Wollenweber, loc. cit.)

Perithecial spherical, 0-18-3 « 0-15-0-25 mm. (av. 0-26 x 0-24 mm.) diam., scattered

or in groups, with bluntly conical apex ; later collapsing, umbilicate, verrucose, blue-black

or yellow brown; borne on a raised, round or elongated stroma of several millimetres

extent. Asci club-shaped, 8- or 4-spored ; spores monostichous or more or less distichous,

elongated-fusiform, straight or slightly curved, broadly rounded at both ends; mostly

3-septate, less frequently 1—2- or 4-7-septate ; 3-septate spores 17-40 « 4-9, mostly 22-31 x

5-2-7.

The conidial form has been found in South Africa on several hosts :—

Hab. Citrus sinensis Osbeck, on fruit shewing stem end rot after 12 weeks in storage ; oranges

from Sunday’s River and Groot Drakenstein, Cape, and from White River, Zebediela and

Rustenburg, Transvaal; also from the air in the citrus packhouse, Zebediela, M.A. 28439.

Lycopersicum esculent:m Mall., from rotting petioles of wilted plant, Gqaga, Transkei,

M.A. 28421.

Lepidosaphes Gloveri, on mussel scale on citrus (associated with Tetacriwm rectisporum)

Chase Valley, Maritzburg, Natal (van der Plank) M.H. 28438.

Also isolated by du Plessis (13) from rotting potato tubers (Solanum Tuberosum) from

Ceres, Cape.

This cosmopolitan species occurs as a saprophyte on decaying parts of plants. It may

act as a weak parasite and cause fruit rot in stone fruits and cucurbits ; it also occurs on

scale and other insects. ‘The ascus stage has been found in Europe, America and Australia.

OTT

Growth on Standard Media.

Oat agar : Aerial mycelium scanty or moderate in amount, short, cobwebby. Growth

in substratum venetian pink to deep rose pink, deepening to amaranth purple ; in the dryer

parts of the medium, it may be amber yellow to mustard yellow. Pionnotes developed

after 14 days, and were pale ochraceous salmon.

Hard potato agar : Aerial mycelium very sparse, white, cottony. Pionnotes developed

in 7 to 14 days, and were light ochraceous salmon to ochraceous slamon.

Standard synthetic agar plus starch ;: Aerial mycelium sparse, short, cobwebby, white ;

or absent. Growth in substratum deep rose pink to old rose and honey yellow in places ;

the yellow colour disappeared after 4 weeks. Pionnotes light pinkish cinnamon to light

ochraceous salmon, or taking up the colour of the stroma and becoming light coral red. In

one tube, branched, erect, Clavaria-like sclerotial bodies developed at the base of the tube ;

these were pale, and after some weeks, sporodochia developed on the tips of some of the

branches.

Potato agar plus 5 per cent. dextrose : A moderate amount of aerial mycelium developed ;

it was tomentose or cobwebby, white to deep pink or Indian lake and yellow ochre—the

yellow colour chiefly in hyphae in contact with the glass. Growth in substratum amaranth

purple, pomegranate purple and Bordeaux.

Potato plug: Aerial mycelium fairly vigorous, tomentose, white to deep rose pink.

Growth on substratum pomegranate purple to Bordeaux. After some weeks, the growth

sometimes became rather felt-like and wrinkled, and numerous flesh colour sporodochia

developed.

Melilotus stem: Aerial mycelium vigorous, cobwebby to sericeo-tomentose, white,

or tinged rose pink or naples yellow. Conidia were produced freely on the mycelium after

14 days, in mass light to pale ochraceous buff ; a few sporodochia developed after 8 weeks.

In one set of cultures, there were groups of rugulose sclerotia after 8 weeks; these were

pale at first, becoming brown with age.

Bean pod: Pods covered with a moderate growth of mycelium, which is tomentose

to sericeo-tomentose, white, or tinged coral pink to hight coral red. Conidia forming freely

in the mycelium were light ochraceous buff in mass.

Rice: Growth at first white and flesh pink, becoming olive ochre to honey yellow.

After 8 weeks, the growth may still be yellow, or it may be deep vinaceous to wood brown,

and the grains vandyke brown.

Measurements of Conidia.

Oat agar, culture 2 weeks old, conidia from pionnotes.

Grseptatemmerersan sty) LAECL cc: as etic s tees. otD X 47).

DANAUS ccoccococcco W) OOP CMGsccoccos 32°5-45 x 4-4-5.

ASIEN + aocc00unede 6 Sv eerste 27:5-35 x 3:75-5.

SHINS oocsecnosane wis ideas Gr st ray oct 20-40 x 3-4-4.

FACINEED. coosocnodes WS) =

l-septate.. 0-5

Mehilotus stem, culture 2 weeks old, conidia from mycelium.

Griesepbaterniy. cansots ATC h jersey oes 50 x 5.

DHA VGSccoscoocccon 440) Ke Colin eocac. 40-50 x 3-75-5.

A-Septatenad. naan) a ein Ales Hennes 22-5-45 x 3-75-5.

SEG) OUTS cotia 6 clot pid ROW) eabce) ib oieiobaaor 25-37-5 x 3-75-95.

Hard potato agar, ‘culture 4 weeks old, conidia from spore g cna.

DAMM cooccenosce & 198 CeMioucoordoos 32-95-40 x 4-7-5.

ASSO DUAUCEEP EEE mye aS See Seen 37-5325 x 4-7-5.

S-septaters-..-. viens. 16 SET TERS CHG EROTIC 20-35 x 3:75-5

DESCWtabOnr spi ace) LZ i

IESepiabererieri-s cc Com ws,

378

Potato plug, culture 8 weeks old, conidia from sporodochia.

LENE. oo06000o0 ne Me wysten stirs a eae 27-5 x 5.

SFI MUMNOonsnocccacac (ld) [XP ClMltpcagucv0c oc 16-40 x 3-75-5.

UAC) NUENO ccogooccccad 7 ae

IEGINADoccssoonccoe 8 oe

Bean pod, culture 2 weeks old, conidia from pionnotes.

G-septaterseerta nce Mewiacanc sabes 47-5-54 x 5.

D-SCPtAte). ). alo. yet a): ON PCTACEN tera erat 32-5-52°5 x 3-75-5.

#- SCD UATE. sjotgercfeiorm ogee css ern ened: 27-5-35 x 3-3-75.

J-SEP tables. fisriokso sjer BOE stb ltsnb hon stenheree saree 20-21-25 x 3-3-75.

UHM scosdsoecaes 2

Fusarium sambucinum Fuck. f. 2. Wr.

Wollenweber, Fusarium-Monographie, 357, 1931; Fus. aut. del. 611, 942, 1145. Wollenweber and

Reinking, Die Fusarien, 77, 1935.

Syn. Fusarium subpallidum v. roseum Sherb.

G: ‘

Fie. 16.

Fusarium sambucinum Fuck. f.2 Wr.; Conidia from sporodochia of 4 weeks old cultures on (a) oat

agar, (b) bean pod, and (c) Melilotus stem, (d@) chlamydospores from 4 weeks old culture on hard potato

agar.

This variety is comparatively pale; aerial mycelium pale, yellowish or pink ; stroma

not carmine, pale or pinkish, does not become blue. Conidia in sporodochia or pionnotest

pink to light orange-red or ochre in mass, and mostly 3-septate, 25 x 4-9, less frequently

4—5-septate ; 5-septate conidia about 30 x 3.

Hab. Citrus sinensis Osbeck, from fruit shewing stem end rot after 18 weeks in storage ;

fruit from Groot Drakenstein, Cape, M.H. 28350 and 28357; Rustenburg, Transvaal,

M.H. 28355; White River, Transvaal.

Lepidosaphes Gloveri, on mussel scale on Citrus (associated with Tetacrium rectisporum),

Chase Valley, Maritzburg, Natal (van der Plank) M.H. 28415.

379

This variety has been found in Europe and North America, on diseased parts of plants

of the genera Citrus, Hordeum, Lycopersicum, Rubus and Solanum; it is also found on

mussel scale and in the soil.

Growth on Standard Media.

Oat agar : Aerial mycelium sparse or moderate in amount, fine, white, cottony. Growth

in substratum colourless, or becoming tinged with congo pink near the base of the slant

after 4 weeks. Small, scattered sclerotial bodies were tilleul buff, and became brownish.

Sporodochia developed in 2 to 4 weeks; they were light ochraceous salmon.

Hard potato agar; Aerial mycelium white; it may be short and sparse, or vigorous,

cottony. Growth on substratum colourless. Pionnotes and sporodochia developed in 2

to 4 weeks; they were light ochraceous salmon.

Standard synthetic agar plus starch: Aerial mycelium scant, fine, white, cottony, or

absent. Growth in substratum colourless, or faintly pink; sometimes there was a tinge

of brown near the base of the slant. The agar was often stained coral pink. Pionnotes

and sporodochia, developing after 14 days, were light ochraceous salmon.

Potato agar plus 5 per cent. dextrose: Aerial mycelium copious or sparse, fine, white,

cottony to tomentose. Growth in substratum avellanous, pale flesh colour, congo pink,

or brownish vinaceous to deep brownish vinaceous; it sometimes became wrinkled and

felt-like, and sometimes there were masses of brown plectenchyma at the base of the slant.

A few scattered sporodochia sometimes developed.

Potato plug: Plug covered with a dense growth of fine, white, cottony mycelium ;

light brown, raised masses of plectenchyma (up to 4 mm. diameter) sometimes developed

from the stroma. Sporodochia often numerous, crowded, light ochraceous buff to light

ochraceous salmon.

Melilotus stem: Mycelium white, or tinged ochre, cottony to sericeo-tomentose,

vigorous or sparse. Numerous sporodochia developed ; they were 0-5 to 3 mm. in diameter,

single or in groups, light ochraceous buff to light ochraceous salmon.

Bean pod: Aerial mycelium white, cottony to tomentose, vigorous or sparse. Small

masses of brown plectenchyma sometimes developed between the medium and the glass.

Large groups of sporodochia developed after 14 days; they were light ochraceous buff to

ochraceous salmon.

Rice : Growth white to flesh colour; grains naples yellow. The pink colour faded

with age, and the grains often became brown.

Measurements of Conidia.

A.—Strain from oranges.

Oat agar, culture 2 weeks old, conidia from sporodochia :—

H-septatel....09... 1... HEWN aceciscpeeemin cOOMeNTO:

ZEROING > condo soangom lly @dilin coop oboe 30-32-°5 x 3-75-4-4.

SSOP tate ni caries wis 1S Fe a pane ee ne > 20-37-°5 x 2-8-4-7.

ZESCDLALe ease re ENT er OUN 55

Melilotus stem, culture 2 weeks old, conidia from sporodochia.

S-=SCPUAUC per yacpa cra WN) (OCP CBs oo6 aooD0d 20-35 x 3-75.

AOU g6o0cccb e006 NS es

Bean pod, culture 2 weeks old, conidia from sporodochia.

4-septate............ &) (iS? Cilio cons acoee 32:5-42°5 x 3-75.

SRG HMIDEYS oath ohio ota! BUTEA, Maudicings pencr 22-5-45 x 2-8-4-4.

ESepuabenaa smc ssi 1 i

B.—Strain from mussel scale.

Oat agar, culture 4 weeks old, conidia from sporodochia :—

DeSeD babes cilia. SOmper CoM rarer: lkor- 25-42°5 x 4-7-5.

A-septate............ iL bed tN Ee 22,-5-32°5 x 3-7-5.

B-SCPEAUC eteriats- «cvesels 3 Sree Wayeud onceeers 17-5-40 x 4-7-5.

380

Hard potato agar, culture 4 weeks old, conidia from sporodochia.

DI NAUD oooboaceoece MD [IE CWiliooodscocce 25-32:-5 x 3-75-5.

A-Sepuave eee eee CA ae shone yaetton 225-30 x 3°75-4-7.

SHI WUD sacsolnocos a6 Sip Mes sans caesar ele ree 18-75-30 x 3-75-4-7.

In cultures of the strain from mussel scale, 5-septate conidia were more frequent, and

the conidia, on the whole, stouter than in cultures of this variety from oranges.

Fusarium sambucinum Fuck. f. 6 Wr.

Wollenweber, Fusarium-Monographie, 358, 1931 ; Fus. aut. del. 327-329. Wollenweber and Reinking,

Die Fusarien, 78, 1935.

Syn. Fusarium sulphureum Schlecht.

F.. discolor App. et Wr. v. sulphureum (Schl.) App. et Wr.

F. genevense Daszewska.

This variety is distinguished by the sulphur yellow colour of the plectenchymatous

part of the stroma and the aerial mycelium, and the absence of the carmine colouring

found in the species and the other varieties. Spherical, dark blue sclerotia may be present

or wanting. Conidia in sporodochia and pionnotes, light orange in mass. Sclerotial plec-

Gemelnyaane Tight brown to sepia. Chlamydospores intercalary, conidia 3—5-septate ; 3-septate

28 Xx 4-5; "5- -septate 38 x 5-1.

This form was not observed in the Union, but is recorded by Wollenweber (loc. cit.)

as occurring in South West Africa on the red locust, Nomadacris septemfasciata.

Form 6 is cosmopolitan, and occurs on a large number of plants, and also on mush-

rooms, in soil, ete. It causes a tuber rot of potatoes.

Fusarium culmorum (W. G. Sm.) Sace.

Saccardo, Syll. Fung. 11, 651, 1895. Wollenweber, Fusarium-Monographie, 360, 1931; Fus. aut.

del. 330-337, 613, 943-945, 1147-1149. Wollenweber and Reinking, Die Fusarien, 79-81, 1935.

Syn. Fusisporium culmorum W.G.Sm.; Fusarium culmorum (W. G. Sm.) Sace. f. 1. Wr.

Fusarium culmorum (W. G. Sm.) Sace. v. letevus (-lethaewm) Sherb.

. culmorum (W. G. Sm.) Sace. vy. majus Wr. (nom. nud.).

. heidelbergense Sacc. ; F. mucronatum Fautr. in herb. pr. p.

. neglectum Jacz.; F. roseum Lk. v. rhei Karst.

. rubiginosum App. et Wr.; Ff. sambucinum Fuck. f. 3. Wr.

. Schribauxwt Del. ; Fusoma tenue Grove.

Fusarium versicolor Sace.

Sel ele ot

Conidia at first scattered in the aerial mycelium, free or in false heads, later sometime

forming a pionnotal layer, or covering the tubercularia-like sporodochia. Conidia in mas

varied in colour, yellow, pink, later ochre to coffee brown, often becoming more or less

tinged with the colour of the stroma. Stroma purple-red and golden yellow to ochre

brown. Conidia fusiform-falcate, gradually or abruptly attenuate at both ends ; apical cell

sometimes rostrate, constricted like the neck of a bottle; base pedicellate ; wall thick,

highly refractive, often brownish ; septations distinct. Conidia 5-septate, less frequently

3-4- or 6—-8-septate ; exceptionally less than 3-septate.

S-septate...i..-ateaee sn LORLOY xX 4eieGe oar mostly 24-32 x 4-8-7.

O-Se plaice nem circ r rer: 23-74 x 4-8°8, ...... mostly 30-60 x 4:8-7:5.

T-septate..c5-4-55-545. GO—1OIXG 4-9) yee mostly 42 —b4y <aio 8

Chlamydospores more frequently intercalary than terminal, spherical or oval, occurring

in conidia as well as in the mycelium, single, 2-celled, or in chains and clusters, brown in

mass, l-celled 9-14 uw diam., 2-celled 13-27 x 7-19 uw.

Hab. Lolium temulentum L., from stems of plants affected with foot rot (ass. Helmin-

thosporium Sp.), Waaikraal, Pretoria dist. (Wager).

Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehder, forming a pink incrus-

tation on heads of kaffir corn infected with smut, (Sphacelotheca sorghi), Ixopo, Natal, May

1923 (Storey) M. H. 17272.

381

Triticum sp., from stems of plants affected with foot rot, (ass. Heliminthosporium sp.),

Waaikraal, Pretoria dist., (Wager).

This species, which is widely distributed, is injurious to cereals, and may cause a rot

of stored fruits. It occurs on numerous genera of plants, in the soil, in the air, and on

other fungi.

Fie. 17.

Fusarium culmorum (W.G.Sm.) Sace.; conidia from (a) pink incrustation on Sorghum (M.H. 17272),

(b) pionnotes of 4 weeks old culture on bean pod ; from mycelium of 2 weeks old cultures of (c) standard

synthetic agar plus starch and (d) on potato agar.

Growth on standard media.

Oat agar: Aerial mycelium copious, cottony, white to ochraceous buff and honey

yellow, or tinged pink. Growth on substratum carmine to ox-blood red. A few large

sporodochia developed after 4 weeks; they were light ochraceous salmon.

382

Hard potato agar: Aerial mycelium scant, white, tufted. Growth in substratum

colourless ; after 14 days, the slant was covered with a thin pionnotal layer, which was

light ochraceous salmon to vinaceous cinnamon. A few small sporodochia were ochracous

salmon to orange cinnamon.

Standard synthetic agar plus starch: Aerial mycelium scant to moderate in amount

white. Growth on substratum tyrian rose to pomegranate purple, carmine and ox-blood

red. Numerous small sporodochia and pionnotes were light ochraceous salmon and salmon

buff to vinaceous cinnamon.

Potato agar plus 5 per cent. dextrose: Aerial mycelium at first white, then white to

chamois colour at the top of the slant, below stained begonia rose. Growth on substratum

carmine to ox-blood red. Spore masses were tinged with the red colour of the stroma.

Potato plug : Aerial mycelium copious, rather coarse, cobwebby, white to naples yelow

and rose pink. Growth on substratum eugenia red to ox-blood red.

Melilotus stem: Stems covered with a vigorous growth, white at the top where the

mycelium filled the tube; below clothing the stems with a growth chatenay pink to spinal

red in colour. Sporodochia not numerous, light ochraceous salmon.

Bean pod : Growth extremely vigorous, the whole tube being filled with mycelium

which was white to geranium pink. Extensive plonnotes developed, which were at first

ochraceous salmon and later vinaceous cinnamon.

Rice: Aerial mycelium copious, at first white to amber yellow, later becoming white

to ochre red. Growth on substratum alizarine pink to acajou red, becoming pompeian red

to madder brown.

Measurements of conidia.

Standard synthetic agar plus starch, culture 14 days old, conidia from sporodochia.

UBGON@oonocoscoccs 4 JOOP CMMocoessoe 42-5-45 x 5-6:3.

OGM MBccccovscsoes A MPRA CS atte 5 40-42-5 x 6-6-3

eae note Hetclotoree O10) is otc lees 30-47 x 5-7.

AI UENI® oo o0sn60005¢ GED’. sh. AN ee errer 28-35 x 5-T.

contd: SRA Eas 1:5 Pe rer st 27-36 x 4-5

Oat agar, culture a weeks old, conidia from sporodochia.

6-septate .. Soro NG) [Oe COVMPpiae ta obi fe 45-50 x 5-6.

Seseptate sees 59 sfsiliin dbadiche Ae 37:5-53 x 5-6

4-septate .. ST ies ite eR Asa. '5'5'5.0,6°8 30-52°5 x 5-6

3-septate .. BS Ra ae Li sig aah sey RR 30=37:°5 & 5.

Mehilotus stem, culture 16 days old, conidia from sporodochia.

7-septate.. 56000. © QS) JOP CBllth. ooo cue 45 x 6.

OAIOIBIO s occcoccooct 3°5 Pat WR hee... 40-45 x 5-:5-6:3.

AION cacscorccc, (Ht silane ueuetnes 30-45 x 5-6:3.

EATON occcccooccc. LG PRE ce kG as 27-5-387:5 x 5-6:3

SAEED. cocooosacca LG UE aC SRG 20-35 x 5-6-3.

From pionnotes occurring in nature on ovaries of Sorghum, M. H. 17272.

D-Sep tate -Byancs os) s on LV ALO Ee arenes 5D X 6:3.

ieseptater. mam sera malls mperiCentmanee cer 40-45 x 6.

OrSep tales... iach eletena mee Ol) Latin en wale De teers 40-47°5 x 6-6:3

JAI NUBGcoucscccoooe 09) Perm aise 35-45 x 5-6-3.

Fe Sep table rise cy suds ehh POO Me Ue Meee 27:5-35 x 5-6.

BI UENO vo gab oarc0ac 6-5 pay tetra Mee 97-5-32°5 x 4-6.

Fusarium graminearum Schwabe.

Schwabe, Fl. Anhaltina, 2 : 285, 1838. Wollenweber, Fusarium-Monographie, 363, 1931; Fus. aut .

del. 338, 339, 354-357, 948. Wollenweber and Reinking, Die Fusarien, 82- ERD 1935.

Syn. Lusarium graminearum Schw. v. caricis (Oud.) Wr.

EF’. caricis Oud. ; Pionnotes flavicans Sacc. et D. Sace.

383

?Selenosporium bufonicola Speg.; Fusarium bufonicola (Speg.) Sacc. et Trott.

Fusarium discolor App. et Wr. v. majus Wr. apud Lewis (nom. nud.).

F. fimicolum Tassi; F. gynerii Cke. et Hark.

F. Mollerianum Thuem.; F. insidiosum (Berk.) Sacc.

2F. rhoicolum Fautr.; F. roseum Lk. pr. p.; Fusidium roseum Lk. pr. p.

F. roseum Lk. v. maydis Sacc.; ?F. roseum Lk. v. cucubali-bacciferi Sace.

F. rostratum App. et Wr. (non Speg.) Ff’. stictoides Dur. et Mont.

Pes saa ee a 240) ; { ae iy oa

E cs SQ PA Zs

ial ea. y > = = 2 yy :

Pe = aa ids a: Z

Fie. 18.

Fusarium gramineaerum Schwakte; conidia from mycelium of culture on (a) oat agar, 4 weeks old, (6

and d) potato plug, 7 weeks old, (c) standard synthetic agar plus starch, 5 weeks old, (e-f) Gibberella

Saubinetii (Mont.) Sacc., (e) asci and (f) sporidia.

384

Aerial mycelium floccose, white or tinged with pink or yellow. Stroma varied in

colour, white to pink, golden yellow, ochre, or carmine to purple red; it is partly plec-

tenchymatous, effuse, more or less covered with floccose mycelium, and partly sclerotially

erumpent and clothed with conidial masses. These are less frequently sporodochial than

pionnotal, and are ochre to light orange red. Conidia sometimes compact, asin F. culmorum,

sometimes more elongated, fusiform-faleate, curved, tapering at both ends; apex conical

or constricted ; base pedicellate. Conidia 3-5-septate, less frequently 1—2- or 6—9-septate.

BAe] OU SBMA mortals O56 25-66 x 3-6 ........ mostly 30-47 x 3-3-5.

DSO beaten meaner cys 28-72 x 3-2-6 ......mostly 41-60 x 4-3-5-5.

MESOPtAbe se brraaes os canesaue DO=88"x 4a ea ssa mostly 61-82 x 4-5-6-5.

DSC pbateny ees Veer OHANOS KX 4E8 sccoccs mostly 61-96 x 4-5-7

Chlamydospores wanting or scarce, intercalary.

Hab. Zea Mays L., on grain and cob, (grain germinating on the cob), L’Orange, Louis

Trichardt, N. Transvaal, Oct. 1932 (Leemann) M.H. 28442 and 26582; on grain and cob

showing moulding and pink discolouration, Rustenburg, Aug. 1929 (Watts) M.H. 24866 ;

from maize meal, Bethal, O.F.S.; from grain (which frequently showed no sign of disease),

Kenya, 1930 (Macdonald).

The conidial stage was also isolated from maize stalks on which (ibberella fructifi-

cations had developed.

Fusarium graminearum is cosmopolitan and it occurs chiefly on cereals, to which it

is injurious, causing foot rot and seedling blight ; it also causes cob mould of maize. It

is the conidial form of :—

Gibberella Saubinetii (Mont.) Sacc. pr. p.

(For synonymy and bibliography, see Wollenweber loc. cit.)

Perithecia blue-black, solitary or in groups, verrucose or smooth, ovoid or spherical,

coriaceo-membranaceous, frequently crowned at the apex with a long-celled outgrowth of

the peridium, 0-20 x 0-17 (0-15-0-3 x 0-1-0-25) mm. (Plate II b.) Ascus 8-spored,

37-84 x 8-15, club-shaped. Spores monostichous or imperfectly distichous, fusiform,

slightly curved or almost straight, broadly conical to acute at both ends, 3-septate, 16-33

<x 3-6, mostly 18-27 x 3-4-5, less frequently l-septate, 14-24 x 2-5-5, exceptionally

4-septate.

Hab. Zea Mays, on stalks, Kenya, March 1930 (MacDonald) M.H. 25348; Hopevale, nr.

Donnybrook, Natal, Jan. 1935 (Doidge) M.H. 27723.

Growth on Standard Media.

Oat agar; Aerial mycelium fairly abundant, tufted, cottony, white to yellow ochre

and rose colour. Growth on substratum pomegranate purple to Bordeaux. After 4 weeks,

the ochre colour disappeared. No spore masses were observed.

Hard potato agar : Aerial mycelium fairly well developed, or scant, cottony, white to

rose pink. Growth in substratum colourless, or with a tinge of Bordeaux.

Standard synthetic agar plus starch - Aerial mycelium scanty, white to yellow ochre.

Growth in substratum eugenia red to carmine.

Potato agar plus 5 per cent. dextrose: Slant covered with a fairly vigorous mycelial

growth, which was floccose, white to rose colour or yellow ochre. Growth in substratum

pomegranate purple to Bordeaux. The ochre colour disappeared with age.

Potato plug: Plug covered with a cottony, tufted mycelium, which is often very

vigorous. It is white to rose pink and ochre. Growth in substratum carmine, pome-

granate purple, Bordeaux or ox-blood red.

Melilotus stem: Stems covered with a vigorous mycelial growth, which was cottony

at first, and white to rose pink or ochre; later the colour in some tubes deepened to

carmine, and the yellow colour faded.

385

Bean pod : Aerial mycelium vigorous, covering pods, at first white to rose colour and

yellow ochre. Later the growth was white to Bordeaux, and the yellow colour had faded.

Rice: Aerial mycelium white to naples yellow and yellow ochre; growth on grains

honey yellow, or carmine to ox-blood red. In 4-6 weeks the colour faded, and growth

was cinnamon buff to snuff brown.

Measurements of conidia.

Hard potato agar, culture 4 weeks old, conidia from mycelium.

6-T-septate............ Hewett: dose 52-5-62-5 x 5.

DAC] ssocecc0ccce 59 So DED Hs Bicone a 37:°5-65 x - “4-5.

AEROBIE oocacgaccdec 7) JIE CIN poaoaceas 30-57-5 x 3-5.

ORE] OUEWS s cogcondcoer 19 eT a eR eR 22-5-47-5 x 3-5.

Oat agar, culture 4 weeks old, conidia from mycelium.

8-septate............ 7) (XOP CBM. on 0000000 67:5-89 x 4-4-5

USE OU Op Coe Pee oe | eS 70-89 x 3-75-5.

G-septatenna err ar NO ees sed Bhusdeutelosennae OT-5-712-5 x 4-4-7

D-II sccasssoo0ce G0 Sanh Wythe eee ee 40-65 «x 4-4°-7.

ASOUM@ ccaconccc008 3 Bah, Fae ee seen 3 -5=50) <X B=4-45

SHIT OD aconosccaces 20) Fagh | ay eee acRE eee 22-5-42-5 x 3-3-7).

ASTON scooebcn sone 2 a4

l-septate... 10

Standard synthetic agar plus starch, culture 3 months old, conidia from mycelium.

7-8-septate............ RUA CMs Neysesn-prcrsese euch: 50-65 x 4-7-5-5.

G-SNBISscoscccaccos Il NSP CSW, cs ccoevce 45-65 « 4-5-5.

Dssep tater vemos Ole” fe ite eeepc Siete 30-54 x 4-4-5.

AULD. coga0co0 eds 36 Aas ate cio! abe 30-42°5 x 3:-7-4:7.

DATA) DUS o og goaoccdo De eeE A yfee ies ae teh: 20-45 <x 3-75-5

Section LATERITIUM.

Wollenweber, Ann. myc. 15:2 and 54, 1917; Fusarium-Monographie, 368-370, 1931. Wollen-

weber and Reinking, Die Fusarien, 86-88, 1935.

Mycelium white, pink, yellow, orange, violet to blue-black. Stroma pale or carmine

to ochre, green olive, brown or blue-black. Spherical sclerotia dark blue or pale. Micro-

conidia 0-l- or more septate, rare, usually small, ellipsoid or comma-shaped, or large,

thick-walled, ovoid to pyriform, disappearing with the formation of sporodochia and pion-

notes. Macroconidia long, cylindrical, fusiform to lanceolate, almost straight to falcate,

constricted at the apex, and more curved near the apex than in the middle, base pedicellate.

Macroconidia pink, and orange to brick red in mass, sometimes becoming darker through

absorption of the colour of the stroma, or becoming lighter if dry and powdery. Terminal

chlamydospores wanting; intercalary chlamydospores occur more or less frequently in

conidia and mycelium.

Key to the South African Species.

A.—Conidia in sporodochia and pionnotes 3—5-septate........ F. lateritiwm.

AA.—Conidia in sporodochia and pionnotes 5-septate :

B.—Stroma not carmine to ochre.................... F. lateritium v. longum.

BiB stromarcarmines to) OCHnC sy -e) cusses iti F, stilboides.

Fusarium lateritium Nees.

Nees, System d. Pilze u. Schwamme, 31, 1817. Wollenweber, Fusarium-Monographie, 370-375

1931; Fus. aut. del. 226, 228-276, 281-285, 570, 577-581, 583-587, 592, 955-957, 959-961, 1154.

Wollenweber and Reinking, Die Fusarien, 88-91, 1935.

386

Syn. Fusarium lateritium Nees f. 1. Wr., and v. pallens Wr.

F. lateritium Nees v. fructigenwm Wr. and f. 1 and f. 2 Wr.

F. lateritiwm Nees v. tenue Wr.; F. acaciae Cke. et Harkn.

F. fructigenum Fr.; F. limonis (Briosi) Penz.

(For complete bibliography and very extensive synonymy, see Wollenweber, loc. cit.).

a =

y 5 \ aN y Yh &

k Fe Nee Wo Gi a ~

Es fi j Vs y: d hs b iN 5 X s C. \ \ . ff a et ‘i \ Seg \\

i Ee | vA Y os \ ci eke 3] \ {

aa Ru AA i amy aoa |

ar lee | Pit ha) beet { | H fj ve Fe ; ite | :

Li HE Ei a | rae ew ee ee ie

rey aa fig). tale el te) ah day Dy fe } ial

ae a A el r Vee oe | |

| a a VA ey i ee Ne ae |

Ss YN ] 4 P Ue fy a 7 /

\ 14 “) | he jh j VA

\ | > ss Z (4 / oe / J, @o

| ce i

SS \ y,

\ ? o “ th

\ f \ i!

(Z \ } \ N mt fa

: i | i » i } AN

Fie. 19.

Fusarium lateritium Nees ; (a-d) strain from failing citrus buds, (e-g) from citrus twig; conidia from

pionnotes of (a) 5 weeks old culture on Melilotus stem, (6) 10 weeks old culture on potato plug, (c) 2 weeks

old culture on oat agar; conidia from (d) pionnotes of 8 weeks old culture on synthetic agar plus starch,

(e) mycelium on bean pod, culture 4 weeks old ; pionnotes of 4 weeks old culture on (f) standard synthetic

agar plus starch and (g) oat agar.

Stroma fleshy, erumpent, smooth, convex ; or cartilaginous, sclerotial, plectenchymatous

rough (cauliflower-like) ; or erect and branched; pale pink, yellow, orange, or chestnut

brown to dark blue; sometimes with spherical sclerotia, which are dark blue to pale.

Aerial mycelium pale, pink or yellow, or tinged with the colour of the plectenchymatous

stroma. Conidia at first scattered in the mycelium; later sporodochia develop, singly or

in groups; they often coalesce to form a continuous pionnotal layer. Conidia in mass

brick red to orange, golden yellow, pink, or salmon colour. Conidia 3—5-septate, rarely

with fewer or more septations, thin-walled, long, fusiform-falcate, almost cylindrical, or

slightly dorsiventral in the middle, definitely curved and often abruptly bent near the

apex; apex constricted, sometimes rostrate; base typically pedicellate.

387

O-Sepbabeayee srr T-l1 X 2°5=3-5:..... Average 7°8 x 28.

ISS NEU Sohadeoedon 1339) X< Ase oooadoae Mostly 13-30 x 2-4-4-5.

S=Seplatenenio sa: 13-52 K 2552.50.53... Mostly 2 1-42 x 2-5-4-4.

HeSeptatensencc< ss. = 2A BA TAO—O% tees): Mostly30-59 x 3-2-4-7.

USUI. ooo ooe 000% 32-84" 3=5........- Mostly 49-72 x 3-3-4-5.

Chlamydospores rare, interealary, in conidia and mycelium. Sclerotial plectenchyma

up to 5 mm. thick, blue, brownish or colourless.

The ascus stage which has not been observed in South Africa is :—

Gibberella baccata (Wallr.) Sacc.

(For extensive bibliography and synonymy, see Wollenweber loc. cit.).

Perithecia often interspersed with the sporodochia of the conidial stage, or in groups,

blue-black, obovate to spherical, rugulose, papillate at the apex, with an inconspicuous

ostiole and a delicate plectenchymatous wall, 0-2-0-3 x 0:15-0:22 mm. Asci 8-spored,

seldom 4-spored, club-shaped, delicate, pedicellate, paraphysate. Spores hyaline, smooth,

oblong-ovoid or fusiform, broadly conical at both ends, sometimes sub-dorsiventral, 3

(1-3) -septate; 3-septate spores 12-30 x 4-10, mostly 13-25 x 4-7-8 up.

The ascus stage has been found on a number of hosts in Hurope, America, Asia and

Australia. The conidial stage is cosmopolitan, occurring chiefly in the temperate zone on

a large number of hosts; it is a cause of bud rot, fruit rot and die back of twigs. It has

been found on Citrus and several other hosts in South Africa.

Hab. Carica papaya L., on decaying pawpaw fruit (a secondary form of decay associated

with Gloeosporium sp.) Bokfontein, Pretoria Dist., M.H. 28429.

Citrus limonia Osbeck, from stem end rot of lemon, developing after 18 weeks in storage ;

fruit from Rustenburg, Transvaal.

Citrus sinensis Osbeck, from fruit (75 strains studied), common in fruit shewing stem

end rot (78 per cent.) and navel end rot or lateral lesions after 6-7, or 12-18 weeks in storage,

1933-34; in navel oranges from Muden, Natal, from Rustenburg, White River, Letaba

and Zebediela, Transvaal, and from Sunday’s River, Cape; from tough, dry form of rot

on side of navel orange from Zebediela, 1931, M.H. 28395.

From twigs showing die-back, Hankey, Cape, May 1930 (van

der Plank) M.H. 28423 ; Ofcalaco, N. Transvaal, July 1930 (van der Plank).

From bark, scaling off orange trees after prolonged drought,

De Wildt, Pretoria Dist., March 1934 (Doidge) ; on bark cracking and gumming, probably

as a result of root injury, Elandshoek, EH. Transvaal (Simmonds) ; on bark of tree affected

by scaly bark, Mazoe Estates, 8. Rhodesia (Bates).

On buds in nursery stock, failing under wet conditions, White

River, E. Transvaal, Nov. 1929 (Hsselen).

Euphorbia crassipes Marloth, on rotting stem of succulent Euphorbia, Willeston, Cape,

M.H. 28378 and 28391.

Prunus persica Sieb. et Zuce., on decaying fruit, Orchard Siding, Cape, Feb. 1913

(Dicey) M.H. 5637.

Lepidosaphes Gloveri (associated with Tetacrium rectisporiwm) on mussel scale on Citrus

twigs, Case Valley, Maritzburg, Natal (van der Plank) M.H. 28391.

Ceroplastis sp., from large waxy scale, on twigs of Acacia sp., Grahamstown, Cape

(Smith) M.H. 28443.

Growth on Standard Media.

Oat agar: Aerial mycelium very sparse, fine, white, cottony. Growth in substratum

at first colourless to barium yellow ; later it often became olive ochre to brown, or, especially

near the base of the slant, dark delft blue and sclerotially erumpent. In some cultures there

were a few small, dark blue, spherical sclerotia. Pionnotes developed freely ; they were

pale flesh colour to light ochraceous salmon, later becoming flesh ochre to rufous.

388

Hard potato agar: Aerial mycelium sparse, short, white, sometimes becoming mealy-

looking, when conidia are formed in minute masses. Growth in substratum colourless,

or with a faint touch of pink. Numerous minute sporodochia rapidly coalesced to form

a continuous pionnotal layer, which was at first pale flesh colour to light ochraceous salmon,

then bittersweet orange; the last named colour soon faded to flesh ochre.

Standard synthetic agar plus starch: Aerial mycelium sparse to moderate in amount,

white, cobwebby. Growth in substratum at first colourless, remaining pale and becoming

raised and gelatinous, or becoming deep delft blue and sclerotially erumpent near the base

of the slant. Pionnotes developed freely and were light ochraceous salmon to bittersweet

pink.

Potato agar plus 5 per cent. dextrose: Aerial mycelium wanting, sparse, or moderate

in amount, cottony or tomentose, white tinged naphthalene yellow and buff pink, sometimes

becoming deep olive buff. Growth on substratum at first pale to flesh pink and pinkish

cinnamon, raised and somewhat gelatinous in places; the stroma remained pale, or became

light brownish olive to snuff brown, bister, or slate colour to blue-black. The medium some-

times became stained brown or black. Pionnotes developed freely, and were ochraceous

salmon, bittersweet pimk to grenadine pink and flesh ochre.

Potato plug: Plug usually covered with a moderate amount of aerial mycelium ; this

was fine, cottony to fe Ifelilke onseniccoe -tomentose, sometimes becoming mealy where conidia

developed, white to cream buff and olive ochre. Pionnotes usually developed, and were

ochraceous salmon to flesh ochre. Small sclerotia developed in some tubes.

Melilotus stem: Mycelium scanty or fairly, abundant, cottony to sericeo-tomentose,

white to naphthalene yellow and ochre. Sporodochia developed; they were flesh colour

to flesh ochre. In some strains, groups of small blue-black sclerotia developed.

Bean pod: Pods covered with a moderate growth of white mycelium, which was

cottony to sericeo-tomentose, or with a tough, leathery, wrinkled growth, tilleul buff in

colour. Pionnotes and sporodochia usually developed; they were flesh colour, salmon,

bittersweet pink and salmon buff.

Rice: Growth white to flesh colour and grenadine pink to carrot red; it may be

naples yellow to mustard yellow in places. The colour may fade with age. Spore masses

developed in some tubes.

Measurements of Conidia.

Bean pod, strain from Citrus twig, culture 4 weeks old, conidia from pionnotes.

AON, cococccceco Il OS CBMisascocos SA) S< Be’).

ORG NMMSsccsscgvcccs AW) SMES: On! che fe Bi °5-10 <x s-de716.

DACNBWOsoo0cdocccge 18 ya) Fi eats tape 5067-5 « 3-3-78.

AERINEMOscccascoouc. Ail SSK Alchy As sacle 50-65 x oP 1

is IOUS. casoccco00e 50-5 PORN raha Gita 271-5—50' x 2-5-3-5.

ISH. coccescacce Mls 55

Bean, strain from pawpaw fruit, culture 2 weeks old, conidia from pionnotes.

OASENWENE sno oodossce 0-5 per cent.

5-Sep Lae sere maeraecee 15 sie ah Wee oetene 30-47:5 < 3-7T-4:-5.

4-septate: cs. as gsnen 8 Pla Fete Rees oes 32°5-40 x 3-7-4.

BAOWMOscogaccccoce ld) ed RRs). Siaees 22/-5=-35 « 3—4:

O=lEsepratenen arene 1 5}5) 5

Bean, strain from Huphorbia stem, culture 2 weeks old, conidia from pionnote;

SHECOMBINOs oocasacav0d NOW ee nee OO AO IOE

6-septateancemae eee N@ Soyer ailing 5b boo joc 42-5-65 x 3-7—-4-7.

BANS|OUBNIS boa 000 aa 000 67 Ny We Ach A 42-5-65 x 3:7-4:-7.

4-septate.....acesess 12 Ae Piao cc e 42-550 x 3-7-4.

SHOUEMDs o45 28 000006 8 ss dad Nira Ai ays NE eran 225-40 x 3-3.75.

U-SSOUENGonbo ob b60 so 3

389

Hard potato agar, strain from Citrus twig, culture 4 weeks old, conidia from pionnotes.

eSepbabeyeniae sie): I OOP CMM. ooo D066 62:5-67°5 « 3-75.

G=septate. 92... .4...-- 3°5 AR TI ihe AS 60-67°5 x 3-3-75

BESO DUALC Stier. ae)eke't 22 AN aaa ais 52-5-72°5 x 2:5-3:°75.

4-septate...........- 36 hn eee Re 50-62°5 x 3-3:°5

SHIMAUOconccoosaoe 25 FR AS i a 30-45 x 3-3:5

2-septate...........- 1-5 a

[eseptater nna...) 9 ee che Pela ae 17-5-22°5 x 2-3.

OHSCIUENDs 66 Gb000000 4 3

Fusarium lateritium Nees var. longum Wr.

Wollenweber, Fusarium-Monographie, 385, 1931; Fus. aut. del. 964, 965. Wollenweber and

Reinking, Die Fusarien, 93, 1935.

Syn. Fusarium lateritium Nees v. longum f. 1 Wr.

Microcera mytilaspidis McAlp.

? Fusarium longisporum Cke. et Mass.

Fic. 20.

Fusarium lateritium Nees. v. longum Wr.; conidia from (a) sporodochia of 5 weeks old culture on

standard synthetic agar plus starch, and (6) sporodochia of 2 weeks old culture on bean pod.

The conidia are long, cylindrical, tapering at both ends, constricted at the apex, pedi-

cellate at the base; in sporodochial and pionnotal masses they are orange red. In the

aerial mycelium, there are a few, small, scattered, 0-1-septate forms, but conidia are mostly

5-septate, less frequently 3-4- or 6—T7-septate, exceptionally 9-septate.

Ozsepbabenase= 144) CLO 2ooyd nee ti: Average.9 x 2-5.

USTED. Sno an ooee ae uy DAVOS HOSS eae. © saeevete NS) <3),

SH OUMNOsaaneoeaneus | MOLE eB Lp oped Average 42 x 3-7.

D-Sepbabewem ss. 1 tO —O0) XM SOD ens om Mostly 52-69 x 3-9-4-9

(Se Plate mata (e crit. DEW <4 Os ea ee cues Mostly 67-77 x 4-2-5-2

G-Septate ser oa. <tr: 80-94 x 4:5-6....... Average 84 x 4-5.

390

Stroma pale or flecked with blue; small sclerotia occur more or less frequently, and

are spherical, dark blue or pale.

Hab. Citrus limonia Osbeck, on lemons kept 7 weeks in storage ; fruit from Sunday’s River,

Cape.

Citrus sinensis Osbeck, from stem end rot in navel oranges, after 18 weeks in storage

(5 isolations) ; fruit from Sunday’s River, Cape.

From bark of orange tree, cracking and gumming above union,

on trees of which roots were water-logged, Letaba, N. Transvaal, 1931, M.H. 28417.

Coffea arabica L., on berries, from Lemana, N. Transvaal, Jan. 1930 (Watson).

This variety has been found on trees, often associated with scale insects, or with other

fungi such as Nectria coccophila and Melola spp., in tropical and sub-tropical regions of

America, Asia and Australia, rarely in Europe.

Growth on Standard Media.

Oat agar : Aerial mycelium moderate to sparse, short, tomentose, white or tinged salmon

buff. Growth in substratum colourless. Groups of sporodochia and pionnotes developed

freely after 2 weeks ; they were pale ochraceous salmon to flesh colour, fading after 8 weeks

to light vinaceous cinnamon.

Hard potato agar: Slant covered with a moderate amount of mycelium, which was

usually short, cottony to tomentose, white or tinged salmon colour. Growth in substratum

colourless. Pionnotes and sporodochia formed after 14 days; they were salmon colour.

Standard synthetic agar plus starch; Aerial mycelium sparse, white, or none. Growth

in substratum colourless, or with a tinge of deep brownish vinaceous at the top of the slant

Pionnotes and sporodochia developed freely after 14 days; they were pale ochraceous

salmon to flesh colour, fading with age to light vinaceous salmon. In one strain, a few small,

blue-black sclerotia were present after 3 months.

Potato agar plus 5 per cent. dextrose: Aerial mycelium wanting, or short, white tomen-

tose. Growth in substratum colourless, or white to grenadine pink, fading to salmon colour,

and, after 30 days, to vandyke brown. In one strain, a few minute, deep delft blue sclerotia

developed near the base of the slant.

Potato plug: Growth rather slow; mycelium fine, short, compact, white or tinged

salmon buff. In one strain, after 4 weeks, there were patches of bluish green in the sub-

stratum, Spore masses formed between the medium and the glass; im cultures 8 weeks

old, they were light vinaceous cinnamon.

Melilotus stem: Growth slow; mycelium sparse, tomentose, white or tinged ochre

in places. Small sporodochia developed ; they were at first pale ochraceous salmon, fading

after 8 weeks to light pinkish cinnamon.

Bean pod: Aerial mycelium developed more rapidly than on melilotus stems ; 1t was

cottony to tomentose, and white or tinged salmon buff in places. Sporodochia and pionnotes

developed after 14 days; they were pale ochraceous salmon to salmon colour.

Rice: Growth slow; mycelium at first white, seashell pink and naples yellow. In

some cases the growth remained pale, in others it became flesh colour to carrot red, and

ater wood brown. In some tubes fairly numerous, small, flesh colour sporodochia developed.

Measurements of Conidia.

Oat agar, culture 2 weeks old, conidia from sporodochia.

UAROUENWOnconconcocoe Os Sy per Cenivey rar 0) 2 6,

G-septabe.. en citrate AO ie ERE is ci 75-95 x 4-5-6, mostly 5 uw wide.

O-Septaberae es menaeae 94. SPER eA 60-87:5 x 4-5-6, mostly 5 u wide.

A-S@pvabves ns. e liore J 5

S-septate:: coccinea BOWany ottewkten et 45-50 x 4-4-5.

391

Standard synthetic agar plus starch, culture 2 weeks old, conidia from pionnotes.

S-septaternas- toe OZOpericeniteraar-r 67-85 x 5-5.

UE|DUMNI@s oo06a000000 9 Pao: 50-80 x 5-5-5.

G-SepUALC mene 8 ert Serer 67-82°5 x 5-5-5.

DEON. oocasc00008 81 Ned Wee Btorotne 50-80 « 4-5:5.

AL-SIUEMIE coo socnDee oe 1 fel pers bob, coset 40-45 x 4-4-5.

D-SePuabe nie acres Oa Te is SR Be, anti 30-42-5 x 4-4-5.

Standard synthetic agar plus starch, Galeure 5 weeks old, conidia from pionnotes.

S=sepiaten gaa saa ele Ie eperyCeNbey ty. te 97-5-105 x 5-5-5.

USQNEHIEo coo ccaoucae 21 BA a Aen es 90-112-5 «x 4-5-5.

G-septates eis vase. 30 Ra poeeed cits 87:5-107:5 x 4-5-5.

FAIS OUND. oon oo one H oe 46 Sl Ln WetSaniate 62:5-95 x 45.

a-septate...........- ICES patee ys y8 bia reece 35-65 x 4-4-5.

Bean pod, culture 2 weeks old, conidia ‘from sporodochia.

8-septate............ OPS eperTiCentinny ser: 85-105 x 5.

URCIDUMG 00005000006 2 Sn Soy ve 75-89 x 5.

G-septaterss.1:aoc ac 4 ae Meanie 62-5-70 x 4-5-5:5

H=sepbabenaemek aie 88 serene aetoshocts 60-70 x 4-5.

A-septate............ 1 AIRS HOS 42-5-50 x 4-5-5.

SHE DEWAN bo Oo SHE One L505) Barrer gate oP 32°5-52-°5 xX 3:75-4:5

leseptate aii: Aes: He Wey orioee ee OMX, 45:

Orsepiatervig syseeinrs) HO Wert kthe el kort. 10-12 x 3-75-5.

Fusarium stilboides Wr.

Wollenweber, Fusarium-Monographie, 385, 1931; us. aut. del. 966-968. Wollenweber and

Reinking, Die Fusarien, 94-95, 1935.

Syn. Fusarium stilboides v. minus Wr.

F.. fructigenum Fr. v. majus f. 1 Wr. et Rkg.

Fig. 21.

Fusarium stilboides Wr. ; Conidia from (a) pionnotes of 2 weeks old culture on oat agar, (b) sporodochia

of 4 weeks old culture on Melilotus stem and (c) pionnotes of 4 weeks old culture on hard potato agar.

Aerial mycelium floccose, abundant or sparse, at first white, then becoming pink or

yellow through diffusion of colour from the stroma. Plectenchymatous stroma honey

yellow to carmine red. The mycelium may also be flecked with blue in the neighbourhood

of small dark blue, sclerotial masses, which later become covered with a conidial layer.

Conidia at first scattered or in false heads, later produced in sporodochia and pionnotes.

Conidia are often produced so freely, that sporodochia become columnar and up to several

392

millimetres long. Smaller 0—-l-septate conidia occur, but are scarce and scattered; they

are oval, fusiform of pyriform; 2-3-septate conidia are also scarce. Macroconidia in

sporodochia and pionnotes are large, cylindrical, more curved at the ends than in the middle,

constricted at the apex, definitely pedicellate at the base, mostly 5-septate, less frequently

3-7T- exceptionally 8-16-septate.

Orseptates mmc: B= Il XK 225-3. Average 7 x 2-7.

ISSO ccceccencos Wi NW) x Aocodo.c. Awermmge 15 < Boo,

S-SeMLALE saeeys eins. = AS) XX MEI Doc od oe Mostly 24-33 x 3-1-4.

DAIOOUENo doe eco ae oc WD OY XK B°SGs0 0006 Mostly 48-73 x 3-5-5:1.

MeSetaterncteyone crete. 56-105 x 3-5-6...... Mostly 66-90 x 4-5-1.

PAENMANIOs 60000006000 69-110 x 3:8-6...... Mostly 70-92 x 4-3-5-2.

NOANG=j NLS. concccccccn WW Ys K HO.

Dark blue, spherical sclerotia, 0-35—0-6 mm. in diameter, sometimes occur on the stroma.

Chlamydospores wanting.

Hab. Carica papaya L., from rotting pawpaw fruit (fruit covered with rose coloured mycelium)

Bathurst, Cape, Nov. 1931.

Citrus sinensis Osbeck, from more or less extensive, stem end rot of navel and Valencia

oranges, after 6-18 weeks in storage ; fruit from White River, E. Transvaal.

Coffea arabica L., on coffee berries, Lemana, N. Transvaal, Jan. 1931 (Watson) M.H.

28408. In this collection, F. stilboides was associated with F. lateritiwm v. longum and a

Capnodium sp. ; the presence of the latter fungus suggests that the berries had been attacked

by scale insects.

This species has often been found associated with Nectria coccophila on scale insects,

on living leaves and branches of Citrus spp., and on blister rust, (Peridermium) on Pinus

chiefly in sub-tropical regions, but also in the temperate zone. It is known in America,

Asia and Australia.

Growth on Standard Media.

Oat agar: Aerial mycelium fairly short and sparse, white, cobwebby. Growth in

substratum eugenia red to acajou red and ochraceous buff to primuline yellow. In old

cultures, there sometimes developed a few large, erect, branched sclerotial outgrowths,

which were dirty white to greenish blue. Sporodochia began to develope after 7 days -

they were numerous, minute (up to 1 mm. diam.), and frequently coalesced to form a

pionnotal layer; they were light pinkish cinnamon or were stained with the colour of the

stroma.

Hard potato agar: Aerial mycelium sparse, white, chiefly at the top of the slant;

Growth in substratum colourless. Pionnotes formed after 7 days ; they were light vinaceous

cinnamon.

Standard synthetic agar plus starch: Aerial mycelium sparse to none. Growth in

substratum eugenia red to acajou red and honey yellow to ochre ; after 8 weeks this colour

had faded. Sporodochia and pionnotes as on oat agar.

Potato agar plus 5 per cent. dextrose: Aerial mycelium none, or rose colour to honey

yellow. Growth on substratum pale to indian lake, or carmine to ox-blood red. Pionnotes,

when present, copious, pinkish buff to light pinkish cinnamon.

Potato plug : Growth slow, white, wrinkled, felt-like ; or aerial mycelium rather coarse,

rose pink to deep rose pink and yellow. Growth in substratum indian lake to pomegranate

purple. Sporodochia rather large, forming erect columns to a height of 3-4 millimetres, light

pinkish cinnamon.

Melilotus stem: Aerial mycelium scanty, or covering the stems; in the latter case

it is sericeo-tomentose, and white to deep rose pink and rose colour. Numerous minute

sporodochia developed in groups and formed slender columns 1-2 mm. long; they were

light pinkish cinnamon.

Bean pod: Aerial mycelium sparse, white ; or more vigorous, tomentose, and tinged

rose pink and mustard yellow. Pionnotes and small sporodochia appeared after 7 days

and were light vinaceous cinnamon.

393

Rice: Aerial mycelium white to naples yellow and mustard yellow ; growth on sub-

stratum honey yellow and amaranth purple. The red and yellow colour faded after 4 weeks,

and the rice was then wood brown, Some pinkish cinnamon sporodochia developed on the

grains.

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from sporodochia.

U-SQDIBNDs 000 00 bo oc0e 7) OEE CWMho coocposoc 60-65 x 4-7-5.

GND. ooo cengccne 11 Pepa eo eneY nares 52-5-65 x 5.

DS OMUGo0 0000000006 DORM east curtains ieee 40-17°5 x 3-75-5

#-septate............ 20 Le aes eaabon ee Gncothe 35-67°5 x 3:75-5

GHCWAUG. qoccccscoee 3 Pera pate Are An eee is 25-55 X 3°75-4-7

5-septate conidia sometimes over 90 per cent., 50-70 « 4-5.

Hard potato agar, culture 4 weeks old, conidia from pionnotes.

TeS-septates hs. = ste RAT ON ON pices cea wits: 73-85 X 5-6.

G-septate............ Il JOSIP CBM. oa aceac 65-75 x 5.

AIOE. oo60c000004 57 aetna ened 46-68 x 4-7-5

A-SCDLALCME Mie oe 17 sane Geresucereratas 40-60 x 4-7-5.

SEC OUMIOsbnasobonsos ZED Rewer. MONT Neary alee: 29-59 < 3:71-4:77.

A) OWENS on 66000006 0:5 oA

Potato plug, culture 4 weeks old, conidia from sporodochia.

GAISOWAMIEs ccocccavcdac Ooty Tele COME, oo co0c 52:5-65 x 5.

SONUENSs00540008K00 80:5 as Sn a 52:5-719 x 5.

AL-RINIEUB. coocoooosoc MOB ee | saeen ences 52:5-63 x 4-7-5

eu SARE ee 6°5 Pe | i ey ee 30-55 =< 3°75=5.

il[septatemeri rir 1

The size of the conidia in the strains studied was somewhat below the average for the

species ; they were at first diagnosed as v. minus this variety now being merged in the species.

Section LISEOLA.

Fungi belonging to this section have two conidia forms, micro- and macroconidia.

Microconidia minute, 0—l-septate, fusiform to ovoid, seldom pyriform, in some forms pro-

duced in long chains or false heads, later scattered freely and forming a light powder over

the mycelium. Macroconidia delicate, slender, subulate, almost cylindrical, almost straight

or curved, somewhat dorsiventral, fusiform to falcate, tapering at both ends, sometimes

bent at rather a sharp angle, abruptly constricted at the apex, more or less pedicellate at

the base. The conidia vary in form between those of the Lateritiwm and Rosewm sections,

and also somewhat resemble those of the Elegans section. Macroconidia scattered, or in

sporodochia and pionnotes; in mass they are brownish white, or isabellinous to salmon

orange, when dry becoming brick red, cinnamon brown or pale. Chlamydospores wanting.

Stroma effuse, plectenchymatous, pale, brownish white, pink or violet, smooth, wrinkled

or sclerotially erumpent, and sometimes bearing spherical, dark blue sclerotia. Some

representatives of this group are the conidial forms of Gibberella.

Only two forms have been found in South Africa :—

Microconidiasm chains. eeietre ere es F. moniliforme.

Microconidiammotmchamsi seers rner coe F. monliforme v. subglutinans.

Fusarium moniliforme Sheldon.

Sheldon, A corn mould (Fusarium moniliforme n. sp.), Nebraska Agric. Exp. Sta. Rept.17: 23-32.

1904, Wollenweber, Fusarium-Monographie, 391-395, 1931; Fus. aut. del. 197, 366, 970-973, 976.

1157-1161. Wollenweber and Reinking, Die Fusarien, 98-100, 1935.

394

Syn. Fusarium moniliforme Sheld. v. erumpens Wr. et Rkg.

F. moniliforme Sheld. v. majus Wr. et Rkg.

F. moniliforme Sheld. v. fici Caldus.

F.. celosiae Abe; F. samoense Gehrm. pr. p.

Fusarium moniliforme Sheld.; conidia from (a) sporodochia of 12 weeks old culture on standard

synthetic agar plus starch, (6) sporodochia of 11 weeks old culture on oat agar, (c) longer forms often pre-

dominant in pionnotes of 2 weeks old cultures on standard synthetic agar plus starch, (d) conidia from

sporodochia of 12 weeks old culture on standard synthetic agar plus syarch , (a-c) strain from tomato seed,

(d) from maize.

Microconidia in chains or in false heads, 1—2-celled, fusiform-ovoid, usually very

numerous, and later, when scattered, forming an inconspicuous light powder over the

aerial mycelium ; the powder is isabellinous or pinkish. Macroconidia delicate, subulate,

slightly curved or almost straight, tapering at both ends, often constricted and sometimes

rather abruptly bent at the apex, pedicellate or sub-pedicellate at the base. Macroconidia

scattered, or produced in sporodochia and pionnotes, in mass pale, isabellinous or sealiearn

orange, becoming brick red to cinnamon red or a when dry.

OAGMADsscccccoseoe 4-18 x 1:5-4........ Mostly 5-12 x 2-3.

I-septate na a.cjuieae 9-30 x 2-5.......... Mostly 12-22 « 2-2-3-5.

SUEND: hoo ev oose 20-60 x 2-4-5....... Mostly 32-50 x 2-7-3-5.

D=SCPUALCs tn iit ie 37-70 x 2-4:5....... Mostly 41-63 « 2-7-4.

T-Septate cence 58-90 x 2-5-4-5..... Mostly 61-82 x 2-7-4-2

Chlamydospores wanting. Dark blue sclerotia, 0-08 x 0-1 mm, diam., may be present

or absent. Stroma more or less plectenchymatous, yellowish, brownish or violet.

This species is very variable in the size and septation of its conidia. It occurs in tropical

and sub-tropical regions of Asia, America, Africa, Australia and Melanesia, on a number

of different hosts ; it is chiefly known as a parasite of cereals and other grasses.

395

The ascus stage of F. moniliforme is Gibberella Fujikuroi (Saw.) Wr., which was first

described on rice in 1917. It has not been found in South Africa, but is known elsewhere

on rice, sugar cane and maize, and possibly on other host plants.

Gibberella Fujikuroi (Saw.) Wr.

Syn. Lisea Fujikuroi Saw.

Wollenweber and Reinking, Die Fusarien, 99-100, 1935. Wollenweber, Fus. aut. del. 819, 820.

Gibberella moniliformis (Sheld.) Wineland.

Fig. 22z.

Fusarium moniliforme Sheld.; from plain agar plates, conidiophores bearing microconidia.

Perithecia dark blue, spherical to ovoid, verrucose, 0-19-0-39 « 0-16-0-42 in diameter.

Asci paraphysate. Paraphyses septate, clavate, 84-150 x 9-18 w. Asci cylindrical to

clavate, flattened at the apex, 66-129 x 7-14, mostly 46-spored, seldom 8-spored. Spores

monostichous or imperfectly distichous, l-septate, 10-24 x 4-9, mostly 14-18 x 4-4-7.

Spores occasionally 2—4-septate before germination.

396

This fungus is the cause of the so-called “ Bakanae’’-disease of rice seedlings, the

‘ Pokkah-boeng”’ disease of sugar cane, and of similar diseases of maize.

The conidial stage has been found on a number of hosts in South Africa.

Hab. Alliwm Cepa L., from rotting bulb of onion, Pretoria, 1929 (ass. F. vasinfectum v.

zonatum f. 2) (Wager).

Ananas comosus Merr., from brown, decaying spots in pineapples from Bathurst Dist.,

Cape, offered for sale in Pretoria (brown spots round flowers more extensive and lighter in

colour than those caused by Penicillium sp.).

Brassica oleracea L., from stems of wilting plants, Buffelspoort, Marikana, Rustenburg

Dist. (ass. Rhizoctonia and Pythiwm sp.) (Turner).

Citrus sinensis Osbeck, from fruit showing stem end rot after 12-18 weeks in storage ;

navel oranges from Sunday’s River, Cape, and Rustenburg, Transvaal, and Valencia

oranges from Sunday’s River.

Eleusine indica Gaertn., from stems of goose grass (ass. Helminthosporiwm sp.) Acton

Homes, Natal, 1931 (L. A. Doidge).

Euphorbia crassipes Marloth, on rotting stems, Willeston, Cape (ass. F. lateritiwm).

Gossypium sp., from stems of wilting seedlings, probably following Pythiwm sp., Rusten-

burg, Transvaal (Moore).

Lycopersicum esculentum Miull., on seed offered for sale, Pretoria, (several isolations

(Wager).

Musa Sapientum L., from fruit affected by ‘tip rot,’ Acornhoek, (Boyce).

Nicotiana Tabacum L., from stems of wilting seedlings, probably following Pythiwm.

sp., Rustenburg (Moore).

Persia americana Mill., from roots of trees shewing die-back (also from soil), Malelane,

EK. Transvaal (Wager).

Phlox Drummondu Hk., from stems of wilting plant (ass. Rhizoctonia), Pretoria.

Pisum sativum L., from wilting stem, Carnarvon, Cape (Wager).

Pyrus malus L., from brown cores of fruit, Vereeniging, 1935-1936 (Bottomley).

Solanum tuberosum L., from tubers affected by dry rot, Umhlanga Beach, nr. Mt.

Edgecombe, Natal (van der Plank).

Sorghum vulgare Pers. v. Caffrorum Beauv. (= Andropogon sorghum), from heads of

kaffir corn moulding in the sheath, Pretoria University farm (F. du Toit).

Sorghum vulgare Pers. v. technicum (Koern.) Jab., from rotting stem of broom corn,

Pretoria University farm (F. du Toit).

Striga lutea Lour., from stems of dying witchweed plant, Pretoria (F. du Toit).

Triticum sp., from stems of wheat plants with blind ears, Losperfontein, Transvaal

(Leeman).

Zea Mays L., from mouldy grain, cobs and maize meal; Bethal, O.F.S. (meal said

to be unfit for human consumption) M.H. 28382; Settlers, Springbok Flats (grain showed

a low percentage of germination) (H. du Toit); Louis Trichardt, N. Transvaal (maize

germinating on cobs) ; Pretoria (young green mealie cob on Pretoria market, grains turning

light brown and decaying in patches with some pink discoloration).

From stems of plants which were stunted or were affected by foot rot

(numerous isolations), Pretoria, Immerpan and Warmbaths, Transvaal.

Eggs, from purplish brown, discoloured patches on membrane, which at this point

adhered to the shell, albumen partly coagulated (ass. ’. semitectum v. majus) sent by Poultry

Inspector, Port Elizabeth (Bottomley).

Growth on Standard Media.

Oat agar ; Aerial mycelium usually fairly plentiful, matted, arachnoid, white to hy-

drangea pink and pale brownish vinaceous. Growth on substratum colourless, or deep

purplish vinaceous to perilla purple ; in older cultures it was sometimes blue-black in places.

Pionnotes, when present, pale pinkish cinnamon.

397

Hard potato agar: Mycelium scant to moderate in amount, white to seashell pink’

rather coarse, cottony, or mealy i in appearance owing to the presence of numerous conidia

Pionnotes, when present, light ochraceous salmon to vinaceous.

Standard synthetic agar plus starch; Aerial mycelium none or scanty, white or tinged

pinkish buff. Growth in substratum vinaceous lavender to deep purplish vinaceous ; the

medium sometimes had a brownish tinge. Pionnotes formed over the face of the slant ;

they were pinkish buff to pinkish cinnamon.

Potato agar plus 5 per cent. dextrose ; Aerial mycelium sparse to moderate in amount,

cottony or tomentose, white to flesh pink or light vinaceous lilac. Growth in substratum

vinaceous purple to delft blue and blue-black. Sometimes the agar under the slant was

stained acajou red.

Potato plug : Plug covered with a dense, matted mycelium, which was white to pale

flesh colour or pale vinaceous lilac. A few minute, deep delft blue sclerotia developed in

some tubes, and a few small sporodochia.

Melilotus stem : Growth scant to moderate in amount, white to seashell pink, cottony

or mealy owing to the presence of numerous conidia.

Bean pods : Pods covered with a mycelium which was dense, downy or cobwebby,

or rather sparse, coarse and tomentose; it was white to ochraceous salmon or pinkish

cinnamon. Pionnotes developed in some tubes.

Rice: Growth at first white to alizarine pink and old rose; after 6 weeks, it was

alizarine pink to eugenia red and dark vinaceous in places, spinel red next to the grains ;

at the base of the tube, there was a tinge of dusky auricula purple. The grains were light

ochraceous buff to mustard yellow.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 3 months old, conidia from mycelium.

GI HMSosconoosscce tl jxeeeaodcoseooac 5d— Been S78 18:

DIONE ooocoo0ec00e G0 gy eet eens 40-60 x 3-3-75.

4-septatel..as rer 1 Mah pae A sks okays ta: 40-45 x 3-38-75.

SHOU os oo0c000n06 32 ee kg Sistema aetc 22°5-57-5 x 3-3-75

2-septate..........-. iL ag Saparsieuens. Seemsit 20-25 x 2-5-3.

lESeptatereceri ist (ey. GA Serene 12-5-20 x 2-3

O-septate .. 5 ipa OSA OZ)» =U Rea Nn hig cacea ee ae 5-10 x 2-5-3°75.

Standard Laine agar ales starch, culture 14 days old, conidia from pionnotes.

NEU sepiaber nm: erie cs petccn dies ice ss « 117-5-147-5 x 4-5.

IOS MAOa yon 6 ASR ONE He ce Renee HO ee aes 100-120 x 3-75-4-4

ORCI eo Race Obed CAIRO ea ns 6 Oke acini 85-117-5 x 2-8-4-4

(ESC DLALC HEE orRo: Gitotcvra croton es aciiacias co 75-112-5 x 3-7-4-5

CHET NINE a cogootue eons soe oo op pao gone uaUOS 62-5-85 x 2-8-5.

HGR oc adevcie sd ogoooesade bo cehepoode ot 40-82:5 x 2:8-3-7).

LER) URIC: 2 Gro Ao oiene ob oid A Cie ROoeaS Oo Oi ee roa 56 x 2:5.

SzSCPUALC ep tee a seok- Reh Tol aie) sts Sst sven ener oie =. he 25-52:5 x 3-3-75.

Hard potato agar, culture 14 days old, conidia from pionnotes.

-Sepvaternesce aes ING iomh oGad neo nee 65-82-5 x 3-75.

G-Septabemeaeneee ter OZ Oepericentmr irr 65-717-5 x 3-3-75.

D-Sepua berries ciel 3 de Ree 47-5-72-5 x 3-4:°5

AEROIOUENKS 5 06000000006 3:5 SEs. EEE RCA hs 45-64 x 3-75.

S-Sepbabe meets i 14 set NERY, 29 ciel 32°5-710 x 2:-8-3:75

Q-I-septate............ 79 re

The above measurements were from different cultures of the same strain, and serve

to illustrate the variability in the size and septation of the conidia of F. moniliforme.

398

Fusarium moniliforme Shel. var. subglutinans Wr. et Rkg.

Wollenweber and Reinking, Phytopathology 15: 163, 1915; Die Fusarien, 100-101, 1935;

Wollenweber, Fusarium-Monographie, 397, 1935; Kus. aut. del. 974, 1121, 1122.

Fie. 234.

Fusarium moniliforme Sheld. vy. subglutinans Wr. et. Rkg.; conidia from (a) pionnotes of culture on

hard potato agar, (b) pionnotes of culture on oat agar, (c) pionnotes of culture on potato plug, (d) pionnotes

of culture on oat agar, (e) sporodochia of rice culture; cultures all 4 weeks old; (a-c) strain 632 and (d-e)

strain 631, both from maize.

This form differs from the type chiefly in the microconidia, which are not produced

in chains. The macroconidia are 3—5-septate, seldom 7-septate.

OME boooedoboos Gall) X¢ Ioftes caso ns Mostly 7-12 x 2-3-5.

l-septaitesner-ee eee 10-28 x 2-2-4-5..... Mostly 14-20 x 2-5-3-5.

SHIMANO 5406560000 e 18-60 x 2-5-4:5..... Mostly 25-48 x 3-4.

SGM cook ocoesc A061 Xe 3-40). ce Mostly 43-53 x 3-2-4.

T= RENAME oo aconnsccus 48-57 = 3:25-4-5.

Chlamydospores wanting. Dark blue, spherical sclerotia and irregular, erumpent

sclerotial stromata may be present or absent.

This variety occurs on wheat, maize, sugar-cane and other Gramineae, and on a number

of other hosts in tropical and sub-tropical regions of America, Asia, Africa and Australia ;

it occurs less frequently in Europe. The ascus stage has been found in Australia by Hdwards

(15) and may be described as follows :—

399

Gibberella Fujikuroi (Saw.) Wr. var. subglutinans Edwards.

Edwards, Dept. of Agric. New South Wales, Sci. Bull. 49, 1935. Wollenweber and Reinking,

Die Fusarien, 101, 1935. Wollenweber, Fus. aut. del. 1121, 1122.

The perithecia are dark blue, somewhat rough, spherical to ovoid, and are similar in

size to those of the type. Asci paraphysate, 4-8-spored, clavate. Paraphyses ascending

from the base of the perithecium, about 6-celled, 70-100 x 6-15. Spores 1-septate,

11-22 x 3-8; exceptionally 2—3-septate, 18-23 x 4-6.

On maize, causing a disease of seedlings and older plants. Only the conidial stage

has been found in South Africa.

Hab. Brassica oleracea L., from stems of dying cabbage plants (ass. Rhizoctonia and Pythium

sp.), Buffelspoort, Marikana, E. Transvaal (Turner).

Citrus sinensis Osbeck, from stem end rot of Valencia oranges after 18 weeks in storage,

fruit from Rustenburg, 1933-34.

Kniphofia sp., on capsules, Loskop, Natal (Galpin) M.H. 28385.

Pyrus malus L., from brown cores of fruit, Vereeniging, 1935-36 (Bottomley).

Saccharum officinarum L., from dying leaf of sugar cane, Durban, Oct. 1931 (McLean).

Zea Mays L., from mouldy grain, cob and maize meal, Bethal, O.F.S., M.H. 28379 and

28380; Zoological Gardens, Pretoria; Klip River, Natal (Watts) M.H. 28413; Kenya

(McDonald) M.H. 28422; from maize germinating on cob, Driehoek, Piet Retief (Leemann).

From stems (upper nodes) and collapsed leaf bases, Kinross, Transvaal,

M.H. 28406.

Fic. 238.

Fusarium moniliforme Sheld. v. subglutinans Wr. et. Rkg.; conidiophores bearing micro- and macro-

conidia.

400

Growth on Standard Media.

Oat agar: Aerial mycelium very sparse, or more frequently wanting. Growth in

substratum pale to deep purplish vinaceous, anthracene purple and perilla purple. When

the stroma was purple, the agar became stained the same colour. Pionnotes rather thin,

flesh colour to salmon buff.

Hard potato agar ; Aerial mycelium short, sparse, tomentose ; growth in substratum

colourless, or with patches of blue-black near the base of the slant. A few salmon buff

sporodochia developed.

Standard synthetic agar plus starch: Aerial mycelium sparse to none; growth on

substratum pale or acajou red to deep purplish vinaceous; when purplish, the colour of

the stroma became diffused into the agar. Pionnotes light vinaceous cinnamon, or becoming

reddish through absorption of the colour of the stroma.

Potato agar plus 5 per cent. dextrose: Aerial mycelium short, sparse, rather coarse,

tomentose, white to pale vinaceous lilac. Growth on substratum perilla purple to vandyke

red and Hay’s maroon. Pionnotes at first salmon buff; both pionnotes and agar became

stained with the colour of the stroma.

Potato plug: Plug covered with copious mycelium, which was cottony, white to flesh

pink. Growth in substratum was deep delft blue in patches, and a few minute, blue-black

sclerotia were sometimes present. Pionnotes developed freely, they were light ochraceous

salmon to salmon.

Melilotus stem: Stems covered with a short, tomentose mycelium, which was white

to congo pink. Conidia were produced in a dense, vinaceous cinnamon pionnotes, or light

ochraceous salmon sporodochia developed ; these coalesced when numerous, and formed

a continuous pionnotal layer.

Bean pod: Aerial mycelium short, sparse, white; patches of blue-black sometimes

developed on the substratum. Pionnotes very copious, salmon colour to cinnamon.

Rice : Growth Chatenay pink and pale vinaceous lilac to perilla purple. Pionnotes

or sporodochia often developed on the surface of the grains; they were salmon colour to

cinnamon.

Measurements of Conidia.

Hard potato agar, culture 2 weeks old, conidia from pionnotes.

DSO D td besarte 3} [UGE COs occaoccee 52°5-85 x 2:8-3-75.

4-septate............ 8 BY. SUN eee EE 47-5-TO x 2:8-3:5

Bis (c] OME eee era aI wy. ty Golnerae go. 22-5-99 xX 2-8-3-5

me en 3 Pm das aie 18-28 x 2:4-4-5,

FE ae ean 23) ns) A eines Sen 9-18 x 2:4-3-6

O-septate. .. eee) Se aga Ee 4-11 x 2-5-4.

Bean pod, culture 2 w reeks old, conidia from plonnotes.

6-septate............ ROW: okt ieaseenenteieae 50-57-5 x 3-3:70.

DANN. cos ce ccc ks 7) WEE CBM. 06 ooag00% 42-5-57:5 x 2-8-3.

4 Sep tabecis2ys-uieewecy a) Clee Pas, mame Rio eee 42-5-52-5 x 2-8-3.

SFM. coacccaceso A at atk ea At Renae 20-50: x 2-5-3.

ie iat eee 3 Be ae eee eRe 22-27 x 3-3:5.

lEseptatennn me aecne mn LS ae ip (ee eee 10-19 x 3-3°5.

OASCOUMEC 00 ceccacne SI eae se Ey ee pen eee eeee) c5 5-112 x 2-5-3°5

Potato plug, culture 5 weeks old, conidia from sporodochia.

GI NeSeccsouoccc] Jl © [eee OMe cocaanc 43-5-45 x 3-65-4:-5.

5-septate...........-. 25 st diva toast on acter: 35-47-5 x 3:75-4:95.

A-SCPLALC yoy -tP aleve aie 18 BOTY iene 4 Bie 35-47°5 x 3-4°-5

3S paves. seers 39 Fe wih iy se aC 22°5-40 x 2-8-5.

PSM 00000000086 EOE bua cele Meer of ero 20 x 2-8.

L-septate. acs ce ceue- tae 8 st all ssiepeieacversnor 18-24 x 2-8.

O=septate.-...4-5....- Wiles) ie Ge eee 4-5-2 x 2-5-3:5.

401

Oat agar, culture 5 weeks old, conidia from pionnotes.

OSD. cog oocsaads GTO scorers aeassaeenetrsit te 42-5 x 4-4.

SHSM coo coon oe Ge @) OOP CMS 6 sccococe 37-35-50 x 44:5.

A-septate.........4-- 1 yap eee ene eet 37-5-52 x 3:5-4:-5.

d-septate.....+-.. See AL 6 als Pde Side sas tees 19-42-5 x 2-8-4-5.

Z-SCP tate). jeri ss.0e « 1 SMEG Gch eee eve 18-25 «x 3-3-75.

IESephaiteuerewsrs otis OF RE Upaveiren sete 11-20 x 2-8-3.

O-septatesenraeencrin: 39 Pre ete iat 6-12 x 2-5-3.

Section ELEGANS.

Wollenweber, Phytopathology 3: 28, 1913; Fusarium-Monographie, 400-406, 1931. | Wollen-

weber and Reinking, Die Fusarien, 104-109, 1935.

Fungi with two conidial forms, microconidia and macroconidia. Microconidia ovoid-

ellipsoid, straight or reniform, 5-12 x 2-2-3-5, single on free conidiophores, or loosely

agglutinated in false heads. Macroconidia in tubercularia-like sporodochia or in an extended

pionnotal layer, on closely crowded, freely branched conidiophores. The conidial masses

are formed on ap erumpent or flat, plectenchymatous or sclerotial stroma; they form a

convex layer, or appear in small masses, like grains ot sand, which readily become coalescent ;

when dry, these form a hard, resinous crust, or a powdery layer. In some species, the macro-

conidia are elongated, fusiform to subulate, tapering at both ends or slightly constricted ;

in others they are more compact, fusiform-falcate, usually constricted and abruptly curved

at the apex, and pedicellate or papillate at the base. Macroconidia are dorsiventral to almost

cylindrical, thin-walled, usually with 3, or up to 5, delicate cross walls ; their measurements

vary, but they are of medium size, 3-septate 27-46 x 3-5, 5-septate 50-60 x 3-5; in

mass they are pale, isabellmous, brownish-white, flesh colour to salmon orange. Mycelium

white, or stained with the colour of the stroma. Stroma pale or pink, orange colour or purple

red, plectenchymatous, effuse or raised, more or less erumpent and sclerotial, with smooth

or wrinkied surface, and sometimes with elongated or stalk-like outgrowths which are light,

or dark green to blue-black. Chlamydospores plentiful, terminal and intercalary, in my-

celium and conidia. Sclerotia, which may be rough, and brown, blue or pale, present or

wanting.

This group includes a number of organisms causing vascular wilt diseases, which are

more or less specific, on certain hosts, and also organisms causing rots of bulbs, tubers,

roots and fruit.

A number of species and varieties have been recorded on various hosts in South Africa ;

it is possible that other cosmopolitan wilt-organisms are present, and have not yet been

identified. In addition to the forms recorded on the following pages, undetermined strains

of Fusaria belonging to the Hlegans section have been isolated from the following hosts :

Crotalaria juncea, Dahlia sp., Datura stramonium, Dimorphotheca aurantvaca, Fragaria sp.,

Gilia rubra, Persea americana, Physalis angulata, Prunus persica and Rheum rhaponticum.

A full key is given to the species of this section, which comprises so large a number of specific

plant parasites.

Key to the Species.

, HAN prey as rad basse oie Sub-section Orthocera.

a.—Pionnotes typically wanting. Conidia 1-celled, or

sparingly septate.

b.—Conidiophores with bostrycoid branching. .... F. bostrycoides.

bb.—Conidiophores simple, or with branches in

whorls.

c.—Stroma pale, brownish white to flesh

colour.

402

d.—Plectenchyma sometimes erumpent.

Macroconidia not numerous.

e.—Conidia when 3-sept., 34 x 3:5;

Besopes 43 x 3:6; 7T-sept., 59 x

dd.—Plectenchyma not erumpent. Macro-

conidia not numerous, 3-sept. 30 x

3:4.

e.—Pathogenic to Brassica oleracea

ee.—Pathogenic to Beta vulgaris.....

eee.—Pathogenic to Apium graveolens

eeee.—Pathogenic to Piswm satiwum.. .

cce.—Stroma red, violet, red-brown or rust-red.

d.—Stroma chestnut brown, rust-red,

pink ; pea-wilt organism...........

dd.—Stroma purple violet ; conidia when

3-sept. 33 x 3:5; 5-sept. 43 x 3-9

ddd.—Stroma red-violet; pathogenic to

celery (Apium graveolens)...........

aa.—Pionnotes sparse or wanting.

b.—Stroma pale, brownish white, then yellow, not

red.

¢.—Conidia, when 3-sept., 28 x 3°5.........

bb.—Stroma pink to purple.

c.—Conidia, when 3-sept., 39 x 4; 5-sept.,

BC GA A i ic Gace ames Caer ese tee are ee ene ces

cc.—Conidia comparatively long, tapering

gradually to the apex; 3-sept. 36 x 3-5;

Hsept./60) 9 fe Aaosstyesseeael ie ool ewe

aaa.—Fungi sometimes with sporodochia.

b.—Macroconidia 3-sept. 30 x 38:2; or in sporo-

dochia 3-sept. 35 x 4; pathogenic to flax

(CHOOT) CRUTING Wiliicocsc os caccacsceccoeccs

aaaa.—Fungi typically with sporodochia.

b.—Macroconidia comparatively slender, 3-3-7 u

thi ches iiaick fos sc bitin Ueto tens rants Seyteaecaroeeret

e.—Conidia more or less 3-sept.

d.—Conidia very slender; 3-sept. 33 x

ay a a tae Rote nestor a Cem ee Ses

dd.—Conidia somewhat thicker; 3-sept.

BO KOBE DM i HME at LV Shea tari

cc.—Conidia more or less 3—-5-sept.

d.—Stroma more or less sclerotial, pale

not blue.

e.—Conidia 3-sept. 37 x 3-4; 5-sept.

AO! SC. c319 Dixcatarsetona inn sevocuatevseoaeke

dd.—Stroma, when sclerotially erumpent,

blue to pale.

e.—Nclerotial bodies small, numerous,

0-1-3 mm.

f—Conidia 3-sept. 36 x 3:3;

5-sept. 48 x 3-3..

ee.—Sclerotial bodies comparatively

large, scattered, 5-6 mm. diam.

Rey Ry Ry Ry

10,

. conglutinans v. callistephi

. conglutinans.

. conglutinans v. betae.

. orthoceras v. apu f. 1.

. orthoceras v. isi.

. orthoceras Vv. prsi.

. orthoceras.

. orthoceras V. api.

. conglutinans v. citrinum.

. orthoceras v. longius.

. angustum.

lina.

Sub-section Constrictum.

F. bulbigenum v. tracheiphilum.

bulbigenum vy. blasticola.

F. bulbigenum v. lycopersict.

F. bulbigenum v. batatas.

403

f.—Conidia 3-sept. 38 x 3-4;

DOO 0) XX BDacccaacacoc

ff—Conidia 3-sept. 34 x 3-6;

OSepin 41 Xa Oe cect en

bb.—Macroconidia comparatively stouter, 3-7-5 w

UBT CL Nee pe ato ns Sic. vey aspeeieice? stage eeacit 2st.

c.—Sclerotia wanting; sclerotial plecten-

chyma not erumpent.

d.—Conidia not broader in the upper

third than in the middle, nor abruptly

bent at the apex. Stroma red-violet.

Not aromatic. Conidia 3-sept. 38 x

AboRye (yet) On 4 SX “4h@sac von aodouoe

dd.—Conidia often broader in the upper

third than in the middle, usually

abruptly bent near the apex ; stroma

lilac.

e.—Conidia in mass pale, cream to

flesh colour, 3-sept. 36 « 4:7;

5-sept. 44 x 4-7.

f—Aromatic odour developing

Inprice;culiburessa.-scys > et

ff—No aromatic odour.........

cc.—Sclerotia wanting, but plectenchyma more

or less sclerotially erumpent.

d.—Stroma effuse, purple.

e.—Conidia 3-sept. 37 x 3-9; 5-sept.

42 x 4-1. Strongly aromatic. .

ee.—Conidia 3-sept. 38:5 x 3:7;

5-sept. 42-1 x 4-1. Not aroma-

LIKOS no Sato linenr's einereteath eases

dd.—Stroma effuse, salmon-orange, not or

faintly aromatic.

e.—Conidia 3-sept. 37-1 x 3:8;

DH 4) X do ddossogsccc00n 0c

ccc.—Sclerotia present. Sclerotial plectenchyma

and sclerotia dark blue to green, some-

times paler.

d.—Sclerotial bodies comparatively small,

0-1-2 mm. or smaller.

e.—Conidia 3-sept. 33 x 3-7; 5-sept.

40 x 3:8.

f.—Pathogens causing cotton

wilt.

g.—Fungus aromatic.......

gg.—Fungus not aromatic...

ff.—Not causing cotton wilt....

ee.—Conidia 3-sept. 34 « 3-8; 5-sept.

MO) s@ ELS EWROSTRBCo cn eo0c00ceae

dil.—Sclerotial bodies comparatively large,

0-5-3-6-12 mm.

e.—Stroma effuse, usually white to

flesh colour.

In.

1%,

bulbigenum.

bulbigenum v. niveum.

Sub-section Oxysporum.

Ki:

oxysporum v. aurantiacum f. 1

. redolens.

. redolens f. 1.

vasinfectum v. zonatum.

vasinfectum vy. zonatum f. 2.

. vasinfectum v. zonatum f. 1.

. vasinfectum.

vasinfectum f. 1.

vasinfectum f. 2.

. vasinfectum v. lutulatum.

404

f—Conidia 3-sept. 35 x 4,

5-sept. 42 ~« 4-2. Not

aromatic, pathogenic to aster

(COMMSEGDOWS))s0000000000808

ee.—Stroma effuse, pink, violet to red.

f.—Conidia 3-sept. 34 x 4; 5-

sept. 42 x 4-2. Faintly

aromatic. Pathogenic to

onion (Allium), but not to

POCATOR ier sia cers eenscmeistoks

ff—Conidia 3-sept. 35:1 x 4;

5-sept. 42:4 x 4-2. Aroma-

tic or not aromatic. Patho-

genic to peas (Piswm)......

fff—Conidia 3-sept. 35 x 4; 5-

sept. 41 x 4:2. Usually

aromatic.

g.—Cause of potato (Sola-

F., oxysporum f. 6.

F. oxysporum f. 7.

F. oxysporum f. 8.

TU) WAGs soecencvcesn diy CY SAORUND ja Jl.

‘.—Cause of rotting in roots,

tubers, fruit, ete.; not

a specific potato organ-

ism. Aromatic or not

NOUNS dooo0c000000

gge.—Fungus with copious

plonnotes. Cause of

wilt of sweet potato

(Ipomoea). Aromatic. .

eee.—Stroma effuse, red-violet. Coni-

dia 3-sept. 35 x 4; 5-sept.

45 x 4-2.

f.—Sclerotia comparatively num-

erous. Not aromatic. Cause

of wilt of tobacco (Nicotiana)

ff.—Sclerotia comparatively few,

seldom in groups. Cause of

ceee.—Stroma effuse, dark purple to

red-violet.

f.—Conidia 3-sept. 35 x 4:3,

ff.—Conidia 3-sept. 33 x 4:3;

5-sept. 44 « 4:5; T-sept.

42 x 4-8. Not aromatic.

Cause of Gladiolus wilt....

fff—Conidia 3-sept. 40 x 4:4;

5-sept. 47 x 4-9. Not aro-

matic. Cause of wilt of

lucerne (Medicago).........

ffff—Conidia 3-sept. 34 x 4;

5-sept. 44 x 4-3. Not aro-

matic. Cause of wilt of car-

nation (Dianthus).........-

F. oxysporum.

F. oxysporum f. 2.

F. oxysporum v. mcotianae.

F.. oxysporum v. cubense.

F. oxysporum v. aurantiacum.

F. oxysporum v. gladioli.

F. oxysporum v. medicaginis.

F. dianthi.

405

Sub-section ORTHOCERA.

Without sporodochia ; pionnotal layers of limited extent sometimes occur. Macro-

conidia almost straight, fusiform, slightly constricted at both ends, papillate or sub-pedi

cellate at the base, 3-5-septate, slender, delicate. Three-septate conidia usually 8-10, (up

to 12) times as long as broad, and 5-septate 11-13, (up to 17) times as long as broad ;

3-septate conidia 27-46 « 3-4; 5-septate conidia, which may or may not be present,

33-50 (up to 60) « 3-5-4. Chlamydospores spherical to pyriform, smooth or verrucose ;

sclerotia and sclerotial stromata pale, or green to blue-black.

Fusarium Orthoceras. App. et Wr.

Appel and Wollenweber, Arb. biol. Reichanst. Land. u. Forstw. 8: 141-156, 1910. Wollen-

weber, Fusarium-Monographie, 408, 1931; Fus. aut. del. 359-362, 620, 621, 985-989. Wollenweber

and Reinking, Die Fusarien, 111-112, 1935.

Syn. Fusarium albido-violaceum Dasz.

F.. orthoceras App. et Wr. vy. albido-violaceum (Dasz.) Wr.

F. orthoceras App. et Wr. v. triseptatum Wr.

F. oxysporum Schl. v. cucurbitacearum Rabh.

F. oxysporum Schl. v. resupinatum Sherb.

F. oxysporum Schl. v. asclerotium Sherb.

F.. asclerotium (Sherb.) Wr.

Fie. 24.

Fusarium orthoceras App. et Wr.; conidia from mycelium of (a) 8 weeks old culture on Melilotus

stem, (b) 4 weeks old culture on Melilotus stem, and 4 weeks olf cultures on (c) oat agar, (d) bean pods,

(e) hard potato agar, and (f) standard synthetic agar plus starch, (g) chlamydospores from a 6 weeks old

culture on oat agar.

406

Stroma caespitose, plectenchymatous, seldom sclerotially erumpent, flesh colour, flecked

with green, or purple red to violet. Aerial mycelium usually abundant, floccose, white to

pink, readily collapsing and then becoming tough, gelatinous. Conidia, forming freely on

the mycelium, are 1-celled or sparsely septate; very few macroconidia are produced.

Conidia abstricted successively from the tips of free conidiophores; they soon fall off, or

become agglutinated into false heads ; occasionally the conidiophores branch more freely,

and form a thin, fugaceous, flesh-coloured pionnotes, or a few sporodochia. Microconidia

ovoid-cylindrical, straight or curved. Macroconidia almost straight, fusiform-falcate, slender,

delicate, indistinctly septate ; papillate at the base, or with a tendency towards the pedi-

cellate form.

OHM occancnocces Fl X A bsocoscoc0e Wllosdhy GAIA. SK Dob whqR,

IESepicive eer eet: NO XK BeBe) soe 6 Mostly 13-27 x 2-5-3-8.

SHEN do toosancsce HSI xX ModethoB) 55505 Mostly 23-36 x 3-4-3.

DAHON cacdscocc0ss AOS) % GbeB..c5000 Mostly 33-48 x 3-2-4-5.

Chlamydospores terminal and intercalary, spherical to pyriform, smooth or verrucose ;

l-celled 6-14 x 5-13; less frequently 2-celled, 10-21 x 6-13.

Hab. Solanum tuberosum L., from tubers showing various forms of storage rot ; from tubers

imported from England and Germany ; tubers from Mokeetsi, Pretoria (Wager) and Leslie,

Transvaal ; Cedara, (Gill) and Mt. Edgecombe (van der Plank) Natal. Also reported by

du Plessis (13), in rotting tubers from Stellenbosch, Paarl and George in the winter rain-

fall area.

From base of stem of diseased potato plant, Molteno, Cape,

March 1930 (Wager).

This fungus appears to be the most common cause of “ dry rot’ of potato tubers in

South Africa. It is a cosmopolitan organism, and occurs more or less commonly in all

parts of the world, on decaying subterranean parts of plants and in humus; it occurs

less frequently on dead animals, e.g. chameleon.

Growth on Standard Media.

Oat agar : Aerial mycelium short, dense, matted, or sparse, white ; Growth on sub-

stratum pale to vinaceous lilac, or vinaceous purple to slate violet. In cultures of one

strain, a thin, light pinkish cinnamon pionnotes developed on the lower half of the slant.

Hard potato agar: Aerial mycelium moderate to sparse, short, white, cottony to

tomentose. Growth on substratum colourless. A thin pionnotes occasionally developed.

Standard synthetic agar plus starch: Aerial mycelium sparse, white, mostly at the

top and the bottom of the slant. Growth on substratum tinged light vinaceous purple

to slate purple. In culture of one strain, a light pinkish cinnamon pionnotes developed,

and also one or two minute sporodochia.

Potato agar plus 5 per cent. dextrose: Aerial mycelium sparse to none, or moderate,

and then short, tomentose, white to tourmaline pink and hyssop violet. Growth on sub-

stratum flat, or raised and cushion-like, cream, pale purple drab, or light vinaceous purple

to dark slate violet.

Potato plug : Plug covered with a dense growth of cottony mycelium, which was at

first white to shell pink, or flesh pink and buff pink where it touched the glass. Later there

were flecks of dark delft blue in the mycelium at the base of the plug, or patches of dusky

green blue to slate violet against the glass. In some cases, after 4 weeks, the mycelium

was collapsed, tough and wet-looking.

Melilotus stems : Stems clothed with a fairly copious mycelium, which was tomentose

to sericeo-tomentose, or felt-like, white, tinged olive buff to ochre in the dryer parts. Oc-

casionally a few minute, light pinkish cinnamon sporodochia developed (in one strain only).

Bean pod: Pod covered with a moderate growth of mycelium, which was white to

olive bufi, tomentose, or very coarsely sericeo-tomentose. After 4 weeks, the mycelium

was collapsed and wet-looking.

407

Rice : Growth white to alizarine pink and old rose or bishop’s purple, fading to dul

purple. After 4 weeks the mycelium became collapsed and wet-looking. Cultures had a

slight or fairly pronounced odour, resembling that of over-ripe apples. The odour is described

by Reinking and Wollenweber (39) as “ benzolic.”

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from mycelium. Conidia nearly a O-septate

a few 1-2-septate, 3-5-septate conidia rare.

Ors pbabe mere rai.cis cisrusicnclsiemee aa cristae omelet acs 5-14 x 2:7-3:-5.

ISCTOMENED 08a do 6.0 CHE SHER SIMRCIE © cheer ee Cote 14-18 x 3-4.

PUA SLONEERUS Go HES ORICA Cues oy art ce ERG SRE PE 15-22 x 3-5-4

DESC PUA CC epee peue ots vccuee ie a aatel deonatea must tuclere eters oul 30-52°5 x 34:5.

LASS) ONE no oe 20.6 Oo GEER Cee CORO neta 6 42-5-55 x 3-7-4:5

DzSEPLALC rman ates, Se phtiec lence eters elena @ ox 43-45 x 3-7-4-5.

Bean pod, culture 4 weeks old, conidia from pionnotes.

ORES TENG onosoabcoo00 GORpericentar sci 5-15 x 2-3.

ISAOEUOcoocoasondoo © AOS VAS Sines 10-225 x 2-5-3:-5.

SEMAN s c0ce0ccsa0s 26 fate ella ere nets tee 25-47-5 x 3-3-75.

A-SOptater dentate IU Seip ee ARs. Tee 45-50 x 3:7

Oat agar, culture 4 weeks old, conidia from pionnotes.

DRONE sccooonaddoe AM Spericentmann sia: 40-45 x 3-7

HERON) 56 Gao gba0du USO ATi tam a NEES 82. 7 315-50) X 375-4.

3-SePuaber mene ema esac MOOS ieee selene 30-47°5 x 2-5-4.

lEsepiatenn. a-rase: IW) As

O-septate .. ate 5:5

In pionnotes there were occasionally 80-90 per cent. of 3-septate conidia; conidia

from mycelium were usually about 99 per cent. microconidia.

Chlamydospores terminal and intercalary, common in the mycelium, and occasionally

seen in the macroconidia. Form and dimensions agreed with the particulars given in the

general description.

Fusarium angustum Sherb.

Sherbakoff, New York (Cornell) Agric. Exp. Sta. Memoir 6: 203, 1915. Wollenweber, Fusarium-

Monographie, 410-411, 1931; Fus. aut. del. 365, 991-993. | Wollenweber and Reinking, Die Fusarien.

113, 1935.

Syn. Fusarium sclerostromaton Sideris.

Stroma plectenchymatous, effuse, pink to purple. Aerial mycelium more or less abun -

dant, white or tinged with the colour of the colour of the stroma, sometimes flecked with

delft blue or with green-blue patches. Conidia borne on the mycelium or in a thin

pionnotes, pinkish cinnamon in mass, or stained with the colour of the stroma. Conidia

elongated, almost cylindrical, straight or slightly curved, tapering at both ends, sometimes

curved in more than one direction.

ORC MBO scccoacccces DIB )@ 253° Ores... Average ll ~ 2-6:

ILSCNWENE coocosaceecc VW2-24 x 2-5-4.2..5.. Average 21 x 3.

GAME scocsooaacee AV08) XX Aew—e 7 Gonos Average 45-6 x 3:5

D-SCDLALC Lee ery scr 43-81 x 3:5-4-7..... Average 60 x 4:2.

6—8-septate .. sdooo SOHIOA X¢ 4thoill . Average 78 x 4

Chlamydospores L-celled, 6-13 uw diam., or 2- celled, 13-18 x 6-10.

Hab. Arachis hypogaea L., from pods and seeds decaying while still in the soil, University

Farm, Pretoria, 1932 (F. du Toit).

Citrus limonia Osbeck, from lemons decaying after 6 weeks in storage; fruit from

Sunday’s River, Cape.

From roots of rough lemon stock on which orange or grape

fruit had been budded roots, showing “ dry root rot ” Zebediela, N. Transvaal, M.H.

408

98432 and 28441; LElizabethville, Belgian Congo; Steenbokfontein, Rustenburg dist.,

Transvaal, M.H. 28427; Marikana, Rustenburg dist., M.H. 28426; Acornhoek, E.

Transvaal ; Bonnievale, Cape.

From soil in citrus orchards, Kosterfontein, Marico dist., M.H.

28427; Boskoppies, nr. Rustenburg.

hy t S = 4

— ar ee : yh 7 ye

y. Bn. / / eid y

: al

> > oN

CE 2 a Se

te A } “

Shak ss >

oS, lO

Sah fs A ‘ 1

2 , See as

we +

¢ > SE

Se ae * RB me

nee oy

ee : _ a ; 7

SE: SS Fe >

We a e

SENN =o ey : *

3 ~~ =

S >

fas

f \

, =

RS mr

= pion

ie )

> Ga a

t SET 7 >

~ Xr. =

> # ee

BLO

Hie. 25.

Fusarium angustum Sherb.; conidia from thin pionnotes of 4 weeks old cultures on (a) hard potato

agar, (b) oat ayar, and (c) standard synthetic agar plus starch, (d) chlamydospores from culture on hard]

potato agar.

409

Citrus sinensis Osbeck, from fruit showing stem end rot, after 6-18 weeks in sti rage ;

from

Valencia oranges from White River,

Zebediela and

Rustenburg, Thamronll ¢

from navel oranges from Zebediela, White River and Letaba, Transvaal, from Groot Draken-

stein, Cape, and from Muden, Natal.

FOSsypLuMm Sp.,

(Moore).

Lycopersicum esculentum Mull.,

bucinum) ;

from stem of wilting plant, (ass: yeiated with P

K. Transvaal (Wager).

Medicago sativa L.,

Transvaal.

Pinus sp., from wood with intense yellow discolouration, Hogg

(23 isolations from oranges in all).

From roots of old seedling orange tree, Villiersdorp, Cape.

Phoma sp.), Rustenburg

from rotting petioles, Gqaga, Transkei (ass. F. sam-

from stems of plants wilting from attack of Bacterium solanacearum, Tonetti,

from discoloured tissue of crown of dying plant, Pietersburg,

’s Back, Cape (Lurie).

Polygala virgata Thun., from stem wilting plant, Durban (McClean).

Oat agar :

of the slant.

lower half of the slant ;

through absorption of the colour of the stroma.

Growth on Standard Media.

Aerial mycelium sparse, consisting of a fringe of white hyphae at the top

Growth on substratum pale to dull Indian purple.

they were light pinkish cinnamon, or were tinged vinaceous purple,

In some of the strains from Citrus, pion-

Pionnotes formed on the

notes were produced more freely over the surface of the slant on this medium and on other

media.

FLiard potato agar : }

Growth on substratum colourless.

they were light pinkish cinnamon.

Standard synthetic agar plus starch :

lavender.

slant ;

to felt-like, white to vinaceous lilac.

aceous lilac.

Potato agar plus 5 per cent. dextrose :

Growth on substratum dull Indian purple.

Plug covered with a fairly plentiful fine mycelium, which was cottony

white to vinaceous

Potato plug :

to felt-like, and white, tinged seashell pink or deep vinaceous lilac.

lilac.

Aerial mycelium sparse, cottony, white or faintly tinged vinaceous

Pionnotes formed on the lower half of the

Aerial mycelium present in patches, tomentose

Growth on substratum dull Indian purple to vin-

Pionnotes light pinkish cinnamon, or tinged vinaceous lilac.

Aerial mycelium in patches, tomentose to matted,

In some tubes the

mycelium was flecked with deep delft blue, or there were patches of dark glaucous grey

to green blue slate, especially between the medium and the glass.

Melilotus stem :

white, or tinged chamois to yellow ochre in places, and cottony to sericeo-tomentose.

Stems covered with a moderate growth of mycelium, which was

Pin-

kish cinnamon pionnotes sometimes formed on the moister parts of the medium.

Bean pods :

with patches of chamois or yellow ochre.

4 weeks.

Rice :

Growth white to alizarime pink and acajou red, or sometimes carmine.

8 weeks, the colour had faded to nigrosin violet.

Measurements of Conidia.

Aerial mycelium moderate in amount, tomentose to matted, white, or

Mycelium became collapsed and wet-looking after

After

Hard potato agar, culture 12 days old, conidia from pionnotes.

SHAG DUNE cs oop op 500 a6

7-septate

DHSGMAWE so s05c 90 acc

ALISTOMEN® oc o00e docacc

S-Sepvabene mee see 2

MOMENI so 0c00 000 cet

l-septate

O-septate

UAT OMe tesca te ayia: 72-90 x 44:7.

Ss bs BO eee | {SP COMiswecescons (USD) << Ceo

Oe Pare Wares eer ae 50-825 4-4-5,

2 ee ne ee ah 3 50-74 3-4-5.

SSP elie, veamennee i aio Al oy (Wa) S< B}=tbols)

IRMAL IY) ec muntnpes eae ees 35-40) = 3-3" 75.

aia B OI 3 Foowos coed AO x Dob Re7d,

SiMe lah 2 cette ie Nace 10:5=-17-°5 x 2°3-75

410

Standard synthetic agar plus starch, culture 4 weeks old, conidia from pionnotes.

G=l-seplaterermer acum wie alert --ne enter iene 67-5-717-5 x 3-7-4-7.

DR NMIEsssrcoecco se ILob) Joe CEM sc ooes oe 52:5-80 x 3-7-4-4.

4-Seplatel seein erase 0-0, PEP MUL 55 (eae aT? 2 50-77°5 x 3-4-4.

SHMAMIO cscccccocces Ii PA NAr ctor conc 3 320-0259) X Sonor

seamen Cane ye ere 4 ae ag ca ya MSE 2-5-4) > 25 = Sie

UASCOMEWS soocccceasse Al) Pala sisi aoe oc 12-5-20 x 2-8-3.

O-septate... isa: GhOM egan etc Ane ae 5-15 x 2-3.

Chlamydospores Rorarnedl in mycelium ama conidia. Mycelial chlamydospores mostly

erminal, at the ends of long slender hyphae, single or in pairs, rugulose ; single chlamy-

ospores from 4 weeks old ‘culture on hard potato agar, 7:5-10 ~ diam.

Fusarium conglutinans Wr. var. callistephi Beach.

Beach, The Fusarium wilt of china aster, Mich. Acad. Sci. Rept. 20: 281-308, 1918. Wollen-

weber, Fusarium-Monographie, 407-408, 1931; us. aut. del. 619. 980, 981. Wollenweber and

Reinking, Die Fusarien, 110-111, 1935.

Syn. Fusarium conglutinans v. majus Wr.

Stroma pale, white, then yellowish, brownish, or pinkish-white, exceptionally with

traces of grey lilac. Microconidia scattered, or occasionally covering the substratum with

a thin pionnotes, mostly 1-celled, seldom 1-septate, interspersed more or less freely with

larger 3-5—7-septate conidia. Macroconidia cylindrical-fusiform, or somewhat curved.

Oseptate............ 5-13 x 2-3:5........ Mostly 6-10 x 2-2-3,

l-septate............ 11-20 x 2:2-4....... Mostly 13-19 x 2-5-3-5

3-septate........5.-. 23-55 x 3-4:°5....... Mostly 28-46 > 2 Be 4-4.

5-septate . . ..... 932-60 x 3-4:5....... Mostly 40-54 x 28 5-4.

7-septate . . UKTI SS, SHAD oo00 500 Average 59 xX 3-7.

Chlamy dospores 1 numerous, ‘terminal and intercalary, spherical to pyriform, 1—2-celled,

also in short chains and small clusters, smooth or rugulose. Sporodochia and sclerotia

wanting.

This variety is the cause of wilt in asters, (Callistephus chinensis), and occurs in all

countries where asters are cultivated, except in those with comparatively low summer

temperatures.

Hab. Callistephus chinensis Nees, from discoloured stems of wilting plants, Pretoria (Doidge,

Wager, van der Merwe) M.H. 28437, and Durban (McClean and Anderson). Ten isolations

were studied. The organism is probably widely distributed in the Union.

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, white, cottony. Growth on substratum colour-

less, or becoming tinged light congo pink. The pink colour faded after 4 weeks, and was

replaced by a tilleul buff or brownish tinge.

Hard potato agar: Aerial mycelium sparse, white, cottony, mostly at the top and

bottom of the slant. Growth on substratum colourless. A thin pionnotes formed on the

surface of the slant.

Standard synthetic agar plus starch: Aerial mycelium like that on hard potato agar.

Growth on substratum colourless, or with a faint tinge of flesh pik; occasionally there

was a tinge of dark vinaceous grey in the agar under the lower part of the slant. A thin

pionnotes developed along the needle track.

Potato agar plus 5 per cent. dextrose; A moderate amount of aerial mycelium devel-

oped, especially on the lower half of the slant. Growth in the substratum was white,

shining, or tinged light brown vinaceous and dark bluish grey green.

Potato plug: Plug covered with a vigorous, white, cottony mycelium.

Melilotus stem: Stems covered with a copious, white, cottony mycelium, or with a

less vigorous growth which was tomentose to sericeo-tomentose and brownish white.

411

Bean pod: — Pods covered with a fairly vigorous growth of white, cottony, aerial

mycelium. It occasionally became tinged with yellow or with light pinkish cinnamon.

Rice : Growth white, or very faintly tinged sea-shell pink. Grains cream colour to

cream buff.

(dl, }

Fig. 26.

Fusarium conglutinans Wr. v. callistephi Beach ; Conidia from thin pionnotes of (a) 8 weeks old cul-

ture on standard synthetic agar plus starch, (b) 2 weeks old culture on oat agar, (c) 4 weeks old culture

on hard potato agar; chlamydospores from (d) 5 days old culture on plain agar and (e) 4 weeks old cul-

tur eon hard potato agar.

412

Measurements of Conidia.

Oat agar, culture 2 weeks old, conidia from thin pionnotes.

DICED occcoc bono ee A JX CHM s o's 0000046 32-5-52-5 x 3-5-4-5.

4-septate............ Dip rose) Saale 32°5-40 x 2-8-3-75.

Sseptatenca.acscecee P28) ssa americans 225-45 x 2:8-3-75.

cuentas Pee ARE en ier ar Osteo ca o.ostg 20-32 X 2-5-3.

IESeptavem erence 6 sch hs ber era ee uateyeeae ae 27-5 xX ae -5-3.

O=Septate, s.rcrllarcock DO) 14 ores Bi ceisagre eee ay 5-15 x 2-2°8.

Standard synthetic agar plus starch, culture 2 weeks old, “cosidlis how thin pionnotes.

D-Sepuabenannciee ack Aro) JOE CBMs onc o¢oc 55-80 x 3:7-4-5.

ASSeptaten ile ma OO SSE ERD eRe 55-70 x 3-5-4.

THIO|OUENMO covcccooecoe : yA Se OR 40-67:5 x 3-4.

AION Go 0000000006 1

UTD sacccoceooas | O°

O-septate..... 89

Most of the conidia from the mycelium were non-septate, those which were 3- or more

septate being only about 1 per cent. of the whole number. A few conidia with more

numerous septations were observed amongst these, especially on hard potato agar and

standard synthetic agar plus starch. The measurements of these were as follows.

G=sepbartetn mia cers renner innuareecn, acta Weer en: KO) K Bot,

(S32) UNE oii Bo Mefolts O18 cum Hig nin eidlo Mele ola obo 6 dc HO =) << aio (—4bo8),

SET septate. fc De asctarscen etait Peau eae eee 55-75 = 3-7-4-5.

Chlamydospores numerous, thick-walled, rough; intercalary single and in pairs,7—9

wu diam. ; terminal usually single, 7-5-12-5 w diam.

Sub-section CONSTRICTUM.

Sporodochia and pionnotes present. Macroconidia elongated, slender, 3-3-7 x diameter,

rather more curved at the ends than at the middle, apex constricted, base pedicellate,

3-1 or 3-5-septate; the 3-septate 10-13 times, and the 5-septate 13-15 times as long as

broad. Chlamydospores, sclerotia and sclerotial stromata as in sub-section Orthocera.

Fusarium Bulbigenum Cke. et Mass.

Cooke and Massee, Grevillea 16: 49,1887. Wollenweber, Fusarium-Monographie, 411-412, 1931 ;

Fus. aut. del. 367-370, 372, 374, 994, 995, 997, 999. | Wollenweber and Reinking, Die Fusarien, 113-114,

1935.

Syn. Fusarium cromyophthoron Sid. ;

F. loncheceras Sid.; F. loncheceras v. microsporon Sid.

F. rhizochromatistes Sid.; F. rhizochromatistes vy. microsclerotium Sid.

F. laxum Peck; ? F. equisetorum (Lib.) Desm.; Hymenula equiseti Lib.

Stroma sometimes effuse, plectenchymatous, pale, or pink to violet red, covered with

pinkish white or lilac, aerial mycelium; sometimes rugulose, sclerotially erumpent, and

developing hard sclerotial bodies 0-5-5 mm. in diameter, which are from light brownish

white or green to dark blue in colour. Conidia in sporodochia, on a flat or raised stroma,

or formed directly on the substratum, or in a pionnotal layer; ochre to salmon colour in

mass. Chlamydospores terminal or intercalary, single, 2-celled or in chains, 5-12 » diam.

Microconidia 1-celled, or with 1 or 2 septations; macrocondia 3—5-septate, elongated,

subulate, straight or sub-faleate, tapering at both ends; somewhat constricted at the

apex, and abruptly bent, or symmetrical and acute ; base more or less pedicellate.

O-septate.........4.) 5-12 x 2-3-5........ Mostly 7-9 x 2-3.

leSepbatenmecn scene 11-33 x 2-3-7....... Mostly 13-20 x 2-3-3:-2.

3-septate. 05 2..0 6. 20404 Ki 23-4.) ik. Mostly 34-44 x 2-7-3-9.

5-septate............ 3466 x 3-4-5....... Mostly 45-56 x 3:2-3-9.

413

Hab. Allium cepa L., from bulb of wilting plant, Hikenhof, nr. J ohannesburg, Oct. 1932,

M.H. 28362.

Ries 277

Fusarium bulbigenum Cke. et Mass. ; (a-d) strain from onion, (e-f) strain from tobacco ; conidia from

sporodochia of (a) 10 weeks old culture on standard synthetic agar plus starch, (6) 4 weeks old culture on

oat agar, (c) 8 weeks old culture on Melilotus stem, (d) chlamydospores from culture on hard potato agar,

8 weeks old; conidia on mycelium of culture (e) 8 weeks old on potato agar plus 5 per cent. dextrose, (f)

12 weeks old on standard synthetic agar plus starch, (g) 4 weeks old on Melilotus stem, and (h) 4 weeks

old on oat agar.

414

Freesia refracta Klatt., from corms showing a form of dry rot, said to develop in storage,

Pretoria, 1929.

Gladiolus sp., from corms of indigenous species of Gladiolus showing dry, brown form

of rot when dug up in veld, Palmaryville, nr. Louis Trichardt, N. Transvaal (Koker).

Nicotiana tabacum L., from discoloured vascular tissues of stems and petioles of wilting

tobacco plants (7 isolations), Rustenburg, Transvaal (Moore).

From soil; isolated from a soil sample by Dr. Kammerman, Division of Chemistry.

This strain was extremely tolerant of copper sulphate ; it grew in concentrations of 1/750

and 1/1000; growth was inhibited by 1/500.

Fusarvum bulbigenum occurs in Europe, and less frequently in America on decaying

bulbs, tubers, rhizomes, roots, stems, fruit, ete., chiefly on plants belonging to the Liliiflorae,

but also on other hosts. It is also present in humus.

Growth on Standard Media.

Oat agar : Aerial mycelium sparse or plentiful; in the latter case it is dense, matted,

white to vinaceous pink. Growth on substratum purplish vinaceous, or vinaceous purple

to dull Indian purple. Very numerous, blue-black, rough, irregular sclerotial outgrowths

appeared in some strains after 4 weeks, pushing through the aerial mycelium, and becoming

more or less erect and stilboid. Groups of sporodochia developed on the stroma at the base

of the tube; they were 2-3 mm. diam., and pale pinkish cinnamon.

Hard potato agar: Mycelium sparse to moderate in amount, white, tomentose or

matted. Growth on substratum colourless. A few small sporodochia developed ; they were

light ochraceous salmon. A number of minute, blue-black sclerotia developed at the base

of the slant.

Standard synthetic agar plus starch : Aerial mycelium. short, sparse, tomentose. Growth

on substratum coral pink to vinaceous pink, with groups of blue-black, erumpent, sclerotial

bodies near the base of the slant; or growth in substratum anthracene purple to taupe

brown, and the agar stained the same colour. Groups of pale pinkish cinnamon sporodochia

developed after 5 weeks.

Potato agar plus 5 per cent. dextrose: Aerial mycelium rather dense, coarse, tomentose

or matted, white to slate purple. Growth on substratum dark perilla purple and dark

naphthalene violet, or almost black, and the agar stained the same colour.

Potato plug : Plugs covered with a dense mycelial growth, which was cottony to tomen-

tose, and white to sea-shell pink. Very numerous blue-black sclerotial bodies developed

after 5 weeks; these were scattered, or crowded and coalescent; in the latter case, they

formed larger sclerotial masses up to 3:5 mm. diam. Sporodochia developed on the sclerotial

masses.

Melilotus stem and bean pod: Medium covered with rather sparse mycelium, which

was short and felt-like, or sericeo-tomentose, white to pinkish buff. Colourless sclerotial

bodies developed in places, and on these the pale pinkish cinnamon sporodochia formed.

Rice: Mycelium fairly dense, white to flesh colour, venetian pink and old rose; my-

celium on grains was often eugenia red to acajou red.

Measurements of Conidia.

A.—Strain from Allium.

Oat agar, culture 8 weeks old, conidia from sporodochia.

DCU. ogg oboccane IG) JOST? OSIM, 5 oo o50056 37:°5-55 x 3-75-4:5

SO ococaunadco GY NO SER Ta PRES st 30-47°5 x 3-5-4.

BSG OUMCs cco cede eas 49 Peay Mee et yes are pers 25-45 x 3-38-75.

l-septatenaa ph cmce 1 a Coane eee ee 10-18 x 2-5-3:5

O-septate............ Il Sg bic ner ora: 7-9 x 2-3.

415

Standard synthetic agar plus starch, culture 8 weeks old, conidia from sporodochia,

BFSETOENE: 05550070056 iS) USP CMs o soo oor

AREOUENES 606 e090 006 14-5 ws WENN ee tai:

SeSCP LALO warp miacaeny SOW ys hc s akisbeneieids

Melilotus stem, culture 8 weeks old, conidia from sporodochia.

4-septate... a Apericenbe aaa seer fe 27 -5-42°1

S=SePUALem Msc s se HO) bossy Myararenen ai. 26-42 -5

Orseptabereci asses t

Chlamydospores formed freely on plain agar plates :

smooth or verrucose, 5—10 yx. diam.

B.—Strain from Nicotiana.

Hard potato agar, culture 8 weeks old, conidia from pionnotes.

37-5-50 x 3:7-4-5.

38-52 x 2-5-4:-5.

30-49 x 2°5-3-75.

they were terminal or intercalary

7- septate. Sales arcadia Rare.. 42-5'x 4.

CHR NEUS. cee noececec 2 per cent Saale s: 42-5-47-5 x 3-7-4:25

D-Sepbaueeee er els. (Gre atmo trey | US sean ones 36-57:°5 x 34:25.

APSCPLALCr corr rca GUE I PR aakee 37°0-47-5 x 3-5-4.

BOUL 000 ooaueoee A Fa a gases ah et eee 20-42:5 x 3-4.

Z-Septates caer yas -. OO PRR ua Sak Seok 22-25) K 2 5—3:

U=SEPOUENID.s ooo oo aaceac Cee | ee bee teah ae ee 14-30 x 2:5-3-25.

CEDIEN@scrcoscscoee DOD! oe She Bese es he UCB ely 9 78}

Oat agar, culture 6 weeks old,

possibly 0-1 per cent. to 0-5 per cent.

6-T-septate............ 75-80 x 4-4-5.

5-septate. . : 30-12°5 x 3-4-25

A-septate............ 30-70 « 3-5-4.

3-septate............ 20-55 « 3-4 (about 50 per cent. of the macroconidia).

2-septate............ 20-30 x 2-5—-2°8.

USGIDEUEs oon gboo Go ox 15-27-5 x 2-5-2°8.

Standard synthetic agar plus starch, culture 10 weeks old, conidia from sporodochia.

S-septate- erin ar ee sper) Centra san. 76-80 x toes 5.

(ESepuate mare acwtl On ums ee OC mcyrwre srs 72-5-15 x 4-4-5.

Orseptabereecwe rset S| Gog shoei ter AT-5—-15: x 3-1-4-25.

DaSeDbabemee rca ti ysis. OGMORN Y Gaia ho secrete Gee 40-75 x 3-5-4-25.

MRC .co6pead0e0" USGS Thee SissGs alec 20-57-5 x 3-5—-4-25.

SseMlAtCwer aac WOR Cos: 19-48 x 3-4.

UAC UES she ag sence. Ors Say ia ae ION xX 2/5:

UAGINIENO. .0coaecae 2°5

Melilotus stem, culture 6 weeks old, Sonia from mycelium.

Very few macroconidia, about 0-1 per cent. :

SPRUE Dic oad o other hee mcr no ON omrowncs ae eee oe 45-17 -5

LSND v'e 9 oi8'd5. peed DIG CIN Lore roi en 45-65 x

SACOM. ocerdbie 6 eeu hole Pee Cela onrqeingre como 28-55 xX

ESC DUDE Metre yo he ever isuseetehers uetcich mires eters 25-32

Sep tabemmarwetray ccna crop ae stems y-seey ele useae ce 15-22-5

: about 50 per cent. of these were 3-septate.

2-8-4.

The conidia of the strain from tobacco seem to be longer on the average than those

of typical Fusarium bulbigenum, and in sporodochia and pionnotes

over 50 per cent. of 5-septate conidia.

in pairs, 5-12 » diam.

there were sometimes

Chlamydospores mostly terminal, single or occasionally

*Should further study show that this fungus is a specific vascular parasite of tobacco,

it may then be regarded as a distinct variety, but for the present it must be classified as

F. bulbigenum.

416

Fusarium bulbigenum Cke. et Mass. var. lycopersici (Brushi) Wr. et Rke.

Wollenweber and Reinking, Die Fusarien, 114-115, 1935. Wollenweber, Fusarium-Monographie,

412, 1931; Fus. aut. del. 393, 996, 998.

Syn. Fusarium lycopersic: Brushi.

F. oxysporum Schl. f. lycopersici Roum.

F. oxysporum Schl. vy. lycopersici Lindau.

I. oxysporum Schl. subsp. lycopersici Sacc.

BP. bulbigenum Cke. et Mass. f. 1 Wr.

Seles oN 2

EES

(EEE SS eal

Y oe

Pe OM

rf ‘

s :

Inet, PAs},

Fusarium bulbigenum Oke. et. Mass. v. lycogersici (Brushi) Wr. et. Rkg.; conidia from (a) pionnotes

on 12 weeks old culture on standard synthetic agar plus starch, () pionnotes of 6 weeks old culture on hard

potato agar, (c) sporodochia of oat agar culture, 6 weeks old.

417

This variety differs from the type in the absence of pigment in the sclerotially eruampent

stroma, which never becomes blue. Aerial mycelium floccose, white or pinkish white.

Plectenchymatous stroma violet red or pale; sclerotial stroma colourless, smooth, flat,

later disappearing. Sporodochia minute, raised, often coalescing more or less to form a

pionnotes ; conidia in mass flesh pink to light orange. Microconidia 1-celled, or with 1—2-

septations, produced freely on the aerial mycelium. Macroconidia 3-5-septate, seldom

6—7-septate.

O-sepbatesnaemear ae: 8 x 2-5.

ISN. cacoeceedec IS} 3 Bee.

S-Septabenese. va. aa 25-66 x 2-3-4:3..... Mostly 31-50 x 2-8-3-8.

DSA ENSsconaboaeeee 32-68 x 2-8-4:-5..... Mostly 43-56 x 3-4

US ONID. coo ge ooneue 50-70 x 3-4-5....... Average 56 x 3-9.

\ 1

& ‘ vi Ej

X ; [~

j f

} sae

} SON )

ee { 2

ss \ \ gs =

~ \ i

~~ } ia) ?

a mele ~

oS <N / aA Se

= ; | )

= PLS as :

d veoh =a f

> ) A ae

Cz eet f SEM j ve

ee A ee, > Y H

mice NS N. ae

ey / if { j }

/ $ i |

Ae ; <

h =

€ i ‘ A 7

~~ i] ~

o % \

oe See

¥ 4 JZ

=, 7

ae Sate) ‘

fi i A

f) vA

— , \

VE vA '

J i oe wo )

Va ) FA alt

a V rN ‘

{ 1} f ‘, ee

} Va >, \ ;

Fig. 29.

Fusarium bulbigenum Cke. et Mass. v. lycopersici (Brushi) Wr. et Rkg.; (a) part of branched conidio-

phore from sporodochium of culture on potato plug, (6) chlamydospores and (c) microconidia from 8 weeks

old culture on standard synthetic agar plus starch.

Hab. Lycopersicum esculentum Mill., very common as a cause of wilt of tomato plants,

especially in the Kastern Transvaal; numerous isolations studied from stems of wilting

plants ; Nelspruit, Karino and Boulders, EK. Transvaal (Wager) M.H. 28428; from petioles

of wilting plants, Gqaga, Transkei (Wager) ; from fruits taken from wilted plant, Duivels-

kloof, N. Transvaal (Wager); from seed offered for sale by local seedsmen (found in 5

samples; most of the seed used by commercial growers is imported from America).

Carica papaya L., from base of stem of plant affected by foot rot, Buffelspoort, Rusten-

burg Dist.

418

Citrus sinensis Osbeck, from fruit externally sound, but showing centre rot, after 17

weeks in storage ; navel orange from Letaba, N. Transvaal (only one isolation).

Dianthus caryophyllus ., from discoloured stems of plants collapsing as a result of

foot rot, Pretoria (Doidge) M.H. 28391.

Fusarium bulbigenum v. lycopersici 1s a cause of tomato wilt, and occurs chiefly in

North America. It occurs on other plants under humid conditions, and has been proved,

by artificial inoculation, to be injurious to Allium, Freesia, Medicago, Pyrus malus and

Trifolium.

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, white tinged vinaceous lilac, or rather copious

matted, white to hydrangea pink. Growth on substratum vinaceous lilac and orange

vinaceous to vinaceous purple and deep purplish vinaceous. Pionnotes developed on the

surface of the slant, or sporodochia were formed on small, pale, sclerotial outgrowths.

Conidia in mass ochraceous buff to ochraceous salmon or apricot buff.

Hard potato agar: Aerial mycelium sparse or copious, white, cottony. Growth on

substratum colourless. Pionnotes developed after 14 days, mostly along the needle track ;

they were pinkish cinnamon.

Standard synthetic agar plus starch: Aerial mycelium very scant, white to vinaceous

lilac, cottony. Growth on substratum colourless to congo pink or vinaceous purple. In

some cultures a rather sparse pionnotes developed, and in others light congo pink sporo-

dochia.

Potato agar plus 5 per cent. dextrose: Aerial mycelium fairly abundant, white to vina-

ceous lilac; growth on substratum dull Indian purple. In some cultures the agar under

the slant was stained purplish vinaceous to dark vinaceous.

Potato plug: Aerial mycelium abundant, white to sea-shell pink and pale salmon

cottony to arachnoid, or dense, matted and becoming wrinkled and felt-like. Growth on

substratum. tinged purplish lilac and hyssop violet. In some cultures, sporodochia developed

on small, colourless sclerotial outgrowths.

Melilotus stem : Mycelium moderate in amount, white to dirty white, or tinged ochre

in the dryer part of the medium, tomentose to sericeo-tomentose. A few light ochraceous

salmon sporodochia sometimes developed.

Bean pod: Aerial mycelium moderate to copious, sericeo-tomentose, or tending to

become wrinkled and felt-like, white or dirty white to light pinkish cinnamon. A few colour-

less sclerotial bodies developed.

Rice: Aerial mycelium white to chatenay pink, becoming flesh pink and coral pink.

Growth on grains vinaceous to deep vinaceous, then eugenia red to acajou red. After 4

weeks, there was sometimes a tinge of purplish lilac near the bottom of the tube, and there

were colourless masses of plectenchyma between the medium and the glass.

In several strains, including at least one which was proved to cause wilt in tomatoes,

a few minute, dark delft blue sclerotia, or patches of dark delft blue plectenchyma appeared

in cultures when the organism was newly isolated. These did not reappear in subsequent

sets of cultures.

Measurements of Conidia.

Standard synthetic agar plus starch, culture 8 weeks old, conidia from sporodochia.

O-Septate............ Met) THR AMG so ss555 AWB XS Bete

ALIEN, odo ab cacnos | W)Ob) ee ee oan Aner 40-525 x 3-5-4.

BHU s 900 6050000 c Ts) pee Ming tM eae eas 25-45 x 2+5-3-75.

PION 35050000000 Bo ad ea eR 22°5-32°5 < 2:5-3°5.

I-septater. +...) ae LD ed ad amin 15-225 x 2-5-3-5.

OES OUE RM oworee acloric | OCG)! Heh ot a dhve Gace 15-15 x 29-3575:

419

Oat agar, culture 6 weeks old, conidia from sporodochia.

Se Pua Lenn eatery ae A OMDEE COMM prs. sr. 40-55 x 3-5-4.

LERUENKEn oon aber 6dr 9 PL Ranh are 35-50 x 3-5-4.

SH NA occ acasecce | Oy SENS eed alates 20-52°5 x 3-4.

2-SE PUdibe maaan eer ae 2-5 WO IRSA egies 3 15-22 x 3-3-5.

URE NUMICsccocccscaco & Sl Sas OORT 10-20 x 3-3-5.

OHIGOUBNIE, coo ooeoens Ue ee erie ees 5-10 x 2°5=-3-5.

Hard potato agar, culture 8 weeks old, “conidia from pionnotes.

6- septate. Peeks ee, bake (O)cts) yore CEM oc udoec 45-62-5 x 4-4-5

DACOUAUC. ncoskooaees 2 Re Male ius 50-57:5 x 3-3-75

ALSOIDUBNIDS dos coacuses 8 Be a uenpeoate 45-52°5 x 3-75.

BACON saccscdasec D2 OMemmea Way, Meares 30-52°5 x 3-3-75.

FCN cog gon oeone 33015) Sh iaataeneouee 25-30 x 3-3:°75.

Se piace sera 5 cle Rese Musa ops: 15-25 x Boe

0-septate. . ZO SEMIN. ote rere nt ete a: 7-5-17-5 x 2-3.

Hard potato agar, “culture 2 weeks old, “conidia from plonnotes.

9—10- -septate Ue cena fa Rages, sae see none 10-15 x 4-4-5.

O-S“sepiate wnat css: ATO. e gaeeh sites pate: 45-07:5 x 4-4-5

5-Septavevmer see ceoe lols) Oe CBMs o accoec 45-60 x 3-5-4-5

ARETE ooo ceocsece BOD) ie, aise 45-60 x 3:5-4-5

-septate Beate ton Foe an cy mcheue a ee 25=50) x 3-5=4-5

PERE ULE On a cle Scher ans cea c 2 Sh See 20-25 x 3-5-4.

USNEWS. coocccoce os 3 A Bee eee 125-20 x 3-3:7.

O-septaternse eee sa ee Ly hla Yah. |, Meecceeerdtes 6-10 x 2-5-3)

A few conidia with 7-10-septations were also observed in pionnotes on plain agar, and

on standard synthetic agar plus starch, 10-septate up to 90 p long, 7-9-septate 60-85

long. Chlamydospores intercalary or terminal, single or in pairs, round or pyriform, 5-10 u

diameter.

Fusarium bulbigenum Cke. et Mass. var. niveum (HE. F. Sm.) Wr.

Wollenweber, Fusarium-Monographie, 414-415, 1931; Fus. aut. del. 387, 1002, 1169. Wol-

lenweber and Reinking, Die Fusarien, 117, 1935.

Syn. Fusarium niveum Erw. F. Sm.; F. citrulli Taub.

F. Poolensis Taub.; F. vasinfectum Ferr. (non Atk.)

= = ~

OL ;. x a. oN eS) d b. ; < (Ge

» \ ie \ / ¢ ~ 5 ky

uN » Wa Ty i> \ v\ \ Li Sok

ee { L yy \\“

fe wn } | i }

| | | La ;

aad t b — it | )

EY ia Bat =

} ’ f | \ 5

' _ r

} i) S | i }

' f fs | — Ue Ton

| ! it AS

! ; ke j™

: ‘

“ A

Hie. 30.

Fusarium bulbigenum Cke. et Mass. v. nivewm (E. F. Sm.) Wr.; conidia from mycelium of 4 weeks

old culture on (a) standard synthetic agar plus starch, and (b) oat agar.

Conidia somewhat broader than those of the type, and the colour of the effuse stroma

is a deeper purple. Mycelium white, flesh colour, pink or purple. Stroma sometimes

sclerotially erumpent, dark blue. Sclerotial bodies comparatively large, up to 3-6 mm.

420

diam., occurring infrequently, and disappearing or becoming colourless when the organism

has been growing for some time in culture. Microconidia 1-celled, or with 1-2-septa,

straight or curved, formed freely in the aerial mycelium. Macroconidia in sporodochia

and pionnotes, light red orange in mass, 3-5-septate, elongated, almost cylindrical to fusi-

form-falcate, tapering at both ends; apex somewhat constricted, abruptly bent or conical ;

base truncated, conical or pedicellate.

OZseptatesnem ene eee DPA Pee ao Sci Mostly 6-7-1] x 2-2-3-3.

lFseptatesa... 00-49 LO0-2£ x20 Ona Mostly 12-18 x 2-7-3.

BHSCOUNO. ood 0occacde 24-50 x 2-4-7....... Mostly 29-40 x 3-1-4.

Hseptatencs- i> aoc MtO TOO Groen Olen r ee Mostly 43-56 x 3:-4-4-3.

Chlamydospores typical, terminal and intercalary, spherical or oval, smooth; ‘in

conidia 5-10 w diam., or if 2-celled 12-15 x 7; in mycelium larger, 7-21 x 6-17, 2-celled

15-30 x 9-20.

Hab. Citrullus vulgaris Schrad., from stems of wilting plants, Witpoort, P.O. Halfway

House, Pretoria dist., 1931-1937; Biesjesvlei, Lichtenburg, Dec. 1935; Uitenhage, Cape

(Haines).

This organism is known as the cause of a vascular wilt of watermelons, and possibly

also of musk melons and cucumbers in the United States and less frequently in Kurope.

Watermelon wilt and its causal organism cause serious losses in fields where watermelon

iS a commercial crop in parts of South Africa. The large number of strains isolated varied

considerably in culture characters, and also in the degree of pathogenicity to the host.

Growth on Standard Media.

Oat agar: Aerial mycelium moderate in amount, cottony to tomentose, white or

faintly tinged with the colour of the stroma. Growth on substratum colourless, or vin-

aceous lilac to anthracene purple; sometimes with a touch of eugenia red at the top of

the slant.

Hard potato agar: Mycelium rather sparse to moderate, white, or tinged pale mauve

to manganese violet. Growth on substratum colourless.

Standard synthetic agar plus starch: Aerial mycelium sparse, cottony to tomentose,

white or tinged with the colour of the stroma. Growth on substratum colourless to Hay’s

lilac or dark perilla purple: im the latter case, the agar was stained dusky dull violet ;

occasionally there were patches of deep delft blue in the substratum.

Potato agar plus 5 per cent. dextrose: Mycelium moderate in amount, tomentose,

white to brownish vinaceous, vinaceous lilac and deep purplish vinaceous. Growth on

substratum colourless to slate purple and anthracene purple, sometimes with a line of slate

violet at the base of the slant. Agar sometimes stained dusky dull violet.

Potato plug: Aerial mycelium copious, tomentose, or sparse, sericeo-tomentose, white

to congo pink. Growth on substratum pale to flesh colour, with patches of slate violet

between the medium and the glass. There was a line of deep delft blue at the base of the

plug.

Melilotus stem and bean pod: Mycelium sparse to moderate, tomentose to sericeo-

tomentose, white or tinged ochre.

Rice: Growth white to eugenia red and dark vinaceous or vinaceous lilac. Not

aromatic.

Measurements of Conidia.

Six strains of this fungus were studied, but no conidial masses were observed in any

of the cultures ; macroconidia were produced in limited numbers on the mycelium, and

their measurements fell within the limits indicated in the general description of the variety.

42]

Sub-section OXYSPORUM.

Sporodochia and pionnotes present. Macroconidia comparatively stout, 3-7-4-7 p

thick, fusiform-faleate, curved, tapering gradually or abruptly at both ends, with rostrate,

elongated or constricted apex, and more or less pedicellate base, 3- or 3-5-septate. The

3-septate conidia 7 times, and the 5-septate 9-10 times as long as broad. Chlamydospores

1-celled, 5-15 w diam., 2-celled 10-14 x 4-8 w. Sclerotia and sclerotial stromata pale, or

green to blue-black.

Fusarium oxysporum Schlecht.

Schlechtendahl, Flora berol. 2: 139, 1824. Wollenweber, Fusarium-Monographie, 416-418. 1931 ;

Fus. aut. del. 378, 379, 1004-1008, 1170-1174. Wollenweber and Reinking, Die Fusarien, 117-118

1935.

Syn. Fusarium candidulum Sacc; F. elegans App. et Wr. (nom. nud.)

F. mycophilum Sace.: F. myosotidis Cke.

F. opuntiarum Speg.; F. trifolii Jacz.

Fie. 31.

Fusarium oxysporum Schlecht. ; conidia from sporodochia of culture on (a) Melilotus stem, 8 weeks

old, (6) standard synthetic agar plus starch, 8 weeks old, (c) oat agar, 10. weeks old, (d) bean pod, 12 weeks

old, (e) chlamydospores from 8 weeks old culture on Melilotus stem.

Stroma brownish-white to violet, plectenchymatous, smooth, effuse; or sclerotially

erumpent and forming hard bodies, which are pale, or wood green to blue black, more or

less rugulose, 0-5-3 or 3-6 mm. diam. When the fungus grows under more humid con-

ditions, the stroma is usually covered with a filamentous aerial mycelium of medium height.

Later sporodochia develop, or, less frequently, pionnotes. Conidia 3— (4-5)-septate, fusi-

form-faleate, curved or almost straight, definitely or weakly pedicellate. Muicroconidia

1-2-celled, oval to reniform, numerous, scattered in the mycelium, but lacking in typical

sporodochia and pionnotes, which consist almost entirely of macroconidia.

Orseptatenesc. 3... el OT ay ee as re Mostly 7-10 x 2-1-3.

Sep tates mets cies sl. 10-26 x 2-4-5....... Mostly 12-17 x 2-5-3-4.

o-septate....-..-.-.- 19-45 x 2°5—5. 3... 3 - Mostly 23-37 x 3-4-6.

DACA, daaceceseeue 30-60' x 3°5-5....... Mostly 38-49 x 3-8-4:-5.

422

Chlamydospores terminal and intercalary, in hyphae and conidia, spherical, smooth or

rugulose, 1-celled, seldom 2-celled, 5-15 » diam., sometimes larger in the mycelium (10-15).

Hab. Carica papaya L., from stems of seedlings which were dying off, E. Transvaal (Wager)

M.H. 28363.

Citrus sinensis Osbeck, from stem end rot and centre of fruit kept 12-18 weeks in

storage; navel oranges from Sunday’s River, and Groot Drakenstein, Cape, M.H. 28353,

28354 and 28351; and from White River, Transvaal; also from Valencia oranges, from

Sunday's River and White River.

Coffea robusta L., from base of stem of plants which were not thriving, Hartebeest-

poort, Transvaal (Koch) M.H. 28363.

Ipomoea batatas Lam., from tubers showing superficial, sunken, dry, discoloured areas,

Humansdorp, Cape (Wager).

Lathyrus odoratus 1., from decaying stems of seedlings, Brooklyn, Pretoria (Doidge).

Solanum tuberosum L., from tubers showing dry rot and wrinkling of stem end, Stamp-

rietfontein, Windhoek, SW. A.; Pretoria and Klerksdorp, Transvaal; Port Elizabeth,

Cape; also in a consignment of potatoes from Hamburg, Germany; from tubers showing

a soft type of rot, Mokeetsi, N. Transvaal, May 1931 (Wager) ; also reported by du Plessis

(13) to have been isolated from potatoes from Paarl and Stellenbosch, Western Cape.

As defined by Wollenweber and Reinking (loc. cit.), this is a ubiquitous species, not

a specific potato parasite, but occurring on an extensive range of hosts. It is known as

a cause of rot of fruits, bulbs and tubers, but further investigations are necessary to deter-

mine its economic importance. It varies in the presence or absence of blue colouring in

the sclerotial plectenchyma, and may, or may not produce on rice media a weak or strong

aromatic odour, reminiscent of lac. Morphologically the species is fairly constant with

regard to the size, form and septation of conidia.

Growth on Standard Media.

Oat agar: Aerial mycelium very sparse. Growth on substratum tinged pale lilac to

vinaceous purple. A few blue-black sclerotial masses, 2-5 mm. diam., formed near the

base of the slant im some strains. Groups of sporodochia developed slowly. There was

a pionnotal layer along the needle track after 14 days, but groups of sporodochia were not

fully developed until after 4 weeks growth; they were pale to light vinaceous cinnamon,

or shell pink to vinaceous pink.

Hard potato agar: A little sparse, white, arachnoid mycelium covered the slant.

Growth on substratum colourless. Sporodochia small, cream colour to pale pinkish cinnamon.

In one set of cultures, a few small sclerotia developed at the base of the slant.

Standard synthetic agar plus starch: Aerial mycelium very sparse, short and patchy,

or wanting. Growth on substratum vinaceous pink to vinaceous lavender, later tinged

slate purple. A few groups of small sporodochia developed, and somtimes coalesced along

the needle track, to form a continuous pionnotal layer. Conidial masses were sea-shell

pimk to salmon buff or vinaceous pink. In some cultures one or two sclerotia formed at

the base of the slant.

Potato agar plus 5 per cent. dextrose: Slant covered with a moderate growth of aerial

mycelium, which was cottony or matted, felt-lke, and becoming wrinkled; it was white

to pale lilac and vinaceous lavender. Growth on substratum purplish vinaceous or dull

Indian purple, sometimes becoming blue-black. The agar was often stained clove brown

to almost black.

Potato plug: Plugs covered with a vigorous growth of cottony aerial mycelium, which

sometimes became feltlike and wrinkled; mycelium white, tinged in places with light

perilla purple. Sclerotial bodies very numerous, rough, minute to 5 mm. diam., at first

pale, usually becoming blue-black ; some strains produced no sclerotia on potato. Sporo-

dochia vinaceous pink to light ochraceous salmon, developing in large groups; individual

sporodochia 1-2 mm. diam.

423

Melilotus stem: Growth rather slow; after 14 days, stems were covered with a

short, close, white mycelium. Sclerotia mostly small, or up to 2-5 mm. diam., few or

numerous, scattered, remaining pale or becoming blue-black. Sporodochia began to develop

after 14 days; they were 0-5-2 mm. in diameter, light pinkish cinnamon to light och-

raceous salmon.

Bean pod : The pods were covered with a rather thin white mycelium, with scattered

opaque spots, or with a heavier growth of white cottony mycelium. Groups of sporodochia

were developing after 14 days; they were pinkish buff to ight pinkish cinnamon and light

ochraceous cinnamon. In some cultures there were numerous, minute, scattered, blue-

black sclerotia.

Rice: Growth at first white to flesh pik, or laelia pink to tourmaline pink: after

14 days, it was white to old rose, and in 4 weeks patches of slate purple sometimes developed.

There were colourless masses of plectenchyma between the medium and the glass. Cul-

tures were faintly or strongly aromatic.

Measurements of Conidia.

Oat agar, culture 2 weeks old, conidia from sporodochia.

DIMA. nocacconoce 1015) JOE CRMs co cconen 50-52:5 x 3-8-5.

ALOU nooo odcecceue LBW Ne ices seend 42-5-52-5 x 4-4-5.

S-septate: sj y.n sas: COE ee a von ae 20-47-5 x 3-7-4-5.

DESO DUAL aaerertste tpnis USO raw Ag ae. 20-22-5 x 3-7.

NSEINWERISo 05 bo odo oe ORD aun G sey Ase. 12-18 x 2-5-3-5.

Ozsep late mn ees cts Me eT Og bal Nar cisco B15) S¢ Zea),

Hard potato agar, culture 4 weeks old, conidia from pionnotes.

SHEEN oon con oon0s QE) OP COs oo ascdnes PTS 19) YK CALEY

PONE Os ooo nvacanee ea

Bean pod, culture 4 weeks old, conidia from sporodochia.

SHINN: boc oe bo cc be We} OEP COMin caccccone 22-5-37-5 x 3-9.

2-septate.....,. 22+...

ILENE ooo bos doed

Bean pod, culture 2 weeks old, conidia from sporodochia.

DCM ccaedegecos IOS) jaee Cee csdnobe 42-5-47-5 x 4-4-5.

A-septate............ 5:5 Pi lism a ekertn apres 31-°5-47-°5 x 3-T-4-5

SEINEUs cocoon coe SOLO Ay aw wees 30-45 x 3-7T-4-5.

PEEVE 650 46 AOC 0-5

1-septate ae gt

OHS ENGL bon adan oon 52 ‘5

Melilotus stem, culture 2 weeks old, conidia from sporodochia.

AR OURIES. candi oaks eos) Wd) jae Comlnbesoease 33-45 <x 4-4-5.

SANG) coc soos owe, OOP fy odusoe oc 22-5-45 x 3-4-5

MESO DUNO A bon 6 abe Ow 0-5

OMEN cadsoccccns | OO)

Fusarium oxysporum Schl. f. 1 Wr.

Wollenweber, Fusarium-Monographie, 418, 1931: Fus. aut. del. 379, 391. Wollenweber and

Reinking, Die Fusarien, 119, 1935.

Syn. Fusarium euoxysporum Wr.; F. oxysporum aut. pr. p.

? F. redolens Wr. v. angustius Lindfors.

This is a form of F. oxysporum which is a specfic parasite of potato (Solanum tuber-

osum), causing wilt.

Conidia O-septate.............. DID XK 2 Fk he rus wim Mostly 7-10 x 2-1-3.

fesepiaten ere ne lO 20R e242 be ee Mostly: 2 Nexo 20324

424

Macroconidia in sporodochia mostly 3-septate, seldom 4—5-septate. Microconidia produce

freely in the aerial mycelium. Stroma effuse, smooth, or sclerotially erumpent, pale to

green or blue-black. Cultures on rice usually aromatic. Chlamydospores typical.

Hab. Solanum tuberosum L., from stems of wilting plants, which showed more or less typical

blackening of vascular tissues and of the vascular rig in the tubers; Northern Transvaal,

April 1932: Schietfontein, De Wildt, Transvaal (Wager); Louis Trichardt, N. Transvaal

(Wager): Mbabane, Swaziland, 1931 (Wager); Moorddrift and Planknek, Potgietersrust,

Transvaal, March 1932.

Also reported by du Plessis (13) from stem of wilting plant, Darling, Cape.

Fusarium oxysporum f. 1 is a cause of potato wilt in North America, Asia and Africa,

comparatively rarely in Kurope. Cultural characters and measurements of conidia closely

resemble those of F. oxysporum.

Fusarium oxysporum Schl. var. nicotianae Fohns.

Johnson, Jour. Agric. Res. 20: 515-535, 1921. | Wollenweber and Reinking, Die Fusarien, 120,

1925. Wollenweber, Fus. aut. del. 625.

Syn. Fusarium nicotianae Oud.; F. tabaciworum Del.

F. oxysporum Schl. f. 5 Wr.

/ H f ‘ x /

tare | ft kl é fe > SS

| Fy a \ \ }

Bye | | i . * 4 ‘ee ef LD

teal 1 es Vi 4 | i ; i

Mal iat . iy i iS "

bel Ye yh | rit : f

\. | Li : \ / f a - - < S

\ 1 et \ : ik y)

\ a IS

Hie. 32:

Fusarium oxysporum Schl. v. nicotianae Johns, ; (a) conidia from sporodochia of 8 weeks old culture

on Melilotus stem, (b) chlamydospores from the same culture.

This form has rather longer conidia than F. oxysporum f. 1 and f. 2, and is a specific

parasite of the tobacco plant. Microconidia numerous, 1-celled or occasionally 1—2-septate.

Macroconidia in sporodochia and sometimes in pionnotes, 3-septate, less frequently 4—5-

septate ; 3-septate 35 x 4:2; 5-septate 44-3 x 4. Chlamydospores 6—10-2 (av. 8-2).

Sclerotia blue-black, comparatively numerous. Fungus not aromatic.

Hab. Nicotiana tabacum L., from plants affected at the collar, and showing discoloration

of the vascular tissues for some distance up the stem, Buflelspoort, Rustenburg Dist.

(Moore).

This variety 1s a cause of tobacco wilt in North America and probably also in Asia,

Africa and Kurope.

Growth on Standard Media.

Oat aga: Aerial mycelium moderate, cottony to tomentose, white to pale lilac.

Growth in substratum pale lilac to vivid purple, colour fading with age. Sclerotial bodies

developed later, especially near the base of the slant. After 4 weeks, small sporodochia

appeared ; they were light ochraceous salmon.

Hard potato agar : Aerial mycelium moderate, cottony ; growth on substratum colour-

less.

425

Standard synthetic agar plis starch: Aerial mycelium moderate, cottony. Growth

on substratum colourless at first; im older cultures stroma and agar tinged dark purple

drab. Sclerotial bodies present.

Potato agar plus 5 per cent. dextrose: Aerial mycelium fairly abundant, tomentose,

white to pale lilac. Growth on substratum at first pale lilac, with a line of naphthalene

violet at the base of the slant ; later naphthalene violet, and the agar tinged with the colour

of the stroma. A few light ochraceous salmon conidial masses developed on tufts of mycelium.

Potato plig: Plug covered with a dense mycelial growth, which was cottony, with

a tendency to become wrinkled and felt-like, white tinged with pale llac in places. There

were patches of blue-black on the substratum, and in older cultures, numerous, large,

erumpent sclerotial masses developed ; these were at first pale, then blue-black.

Melilotus stem : Mycelium thin, white or tinged purplish lilac. After some weeks, a

few irregular sclerotial outgrowths developed from the stroma, and there were a few small

sporodochia, which were light ochraceous salmon to light pimkish cinnamon.

Bean pod : Pod became covered with a scant to moderate mycelial growth, which was

white, cottony to sericeo-tomentose, and sometimes mealy-looking, owing to the presence

of numerous conidia.

Rice : Growth white to purplish blac at first. In older cultures the superficial mycelium

was white and growth on substratum dull bluish violet to dark hyssop violet and vinaceous

lilac. The culture was not aromatic.

Measurements of Conidia.

Melilotus stem, culture 8 weeks old, conidia from sporodochia.

DSC patent are ens ck, EWN rac castle beau 35-45 x 4-5.

LASTED cose oo000e 7 WII CBliltis c oo so 006 32:°5-42-5 x 3-7-5.

SHECOMNO 04 oon 0 d6n006 91 ae WNL se 22/5425 x 3-4-5.

IS NEUE caccooosecce Dio wey

OSSD co cscococce 3)

Potato agar plus 5 per cent. dextrose, culture 8 weeks old, conidia from sporodochia.

HS OHMO coccccococcg AS) USE CUMs oscar 40-48 x 45.

Aeceptatos! oh) ee: 30 5h L -coto ty pee 37-5-46-5 x 4-5.

SHIQMNOocccooosscos (ll ee ae ee 30-425 x 3-7-5.

Ieseptateime tse, ll 45

O-Sepba tenets eer DOs Fp

Fusarium oxysporum Schl. f. 7 Wr.

Wollenweber and Reinking, Die Fusarien, 120-121, 1935. Wollenweber, Fus. aut. del. 1176.

Syn. Fusarium cepae Hanz.

F. cepae Hanz. emend. Link et Bailey.

This form is a parasite of onion (Allium) ; it does not attack potato, nor is /. 1 parasitic

on onion; 3-septate conidia are 33-36 x 3-8-4-5, and 5-septate 44 x 3-9; 6-septate

conidia rare, 52 x 3-3.

Hab. Allium cepa L., from young plants dying off in seed beds, Pyramids, Pretoria

Dist., March 1932 (Mogg) M.H. 28393; from bulbs and leaf bases, Nelspruit (Wager)

M.H. 28433.

Also reported by du Plessis (12), as occurring in the winter rainfall area, and causing

pink root and bulb of onions, Caledon, Ceres, the Peninsula, Franschhoek, Riversdale,

Stellenbosch and Tulbagh.

Growth on Standard Media.

In culture, this form does not differ materially from F. oxysporum. Sporodochia

salmon-buff to salmon colour. Sclerotial masses on potato small, numerous, dark bluish

426

grey green to dark delft blue. Chlamydospores numerous, in mycelium and conidia, mostly

1-2-celled, terminal and intercalary ; 1-celled 5-6-5 w diameter. Rice cultures faintly

aromatic.

Iie! Bi},

Fusarium oaysporum Schl. f.7 Wr.; conidia from sporodochia

(b) Melilotus stem, (c) potato plug, (d) chlamydospores from culture

of 4 weeks old culture on (a) oat agar,

on oat agar, 4 weeks old.

427

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from sporodochia.

PNM ssccopsocone WH) ee CBMGse oo ba cc 37-5-42:5 x 3-7-5

LEANED ce couc dood oor | OO) | pes Maple ome 31-50 « 3-5.

SHAUN s cooecogoonc AD pe) Verctatetaes 22-5-46-5 x 3-4.

IESEINENWE sco0cc00b00 SHOE, Bes eas nee Ras 10-18 x 2-5-3-25.

OBGDIEND conocsoocese OULHO se Caras as 16-105 x 1-8-3.

Potato, culture 4 weeks old, conidia from sporodochia

DAC DENS oocconsandoc O)ets) [OG COilijo cco cese 37:°5-55 x 3-4-7.

LESETDUDU® soccooooonee 4 CaM orc ee 32°5-45 x 3-4-4.

ee ee SHILA iia cet cua outrage sie 20-42:5 «x 3-4-5.

IESep.raiene sel iel aee Dy ME epg “ct mopar net 1O=15 x 2-5-4.

0-septate . . GOOEY hey <sitormieeee 8-10 x 2-3.

Melilotus stem, culture 4 weeks old, conidia from sporodochia.

5-septate .. ob) joe CMMWcsenoaes 40-44 x 3-7-5.

d-septate..........-. Dae Mike Bek cat. mets 30-47°5 xX BTA

DASE PLAC eas Re Ateeey BOLTON has, qe steis cue 22-5-45 x 3-4-7.

Fusarium oxysporum Schl. f. 8 Snyder.

Snyder and Walker, Fusarium near-wilt of pea, Zentralbl. f. Bakt. 11 Abt. 91:

Wollenweber and Reinking. Die Fusarien, 121. 1935.

355-378, 1935.

Syn. Fusarium vasinfectum Atk. v. pisi van Hall.

de S

\ \ / ¢ =

7) ‘ < ~~

; 2 PCN ‘SS

f \ \

\ ie by | a ad [a Wah oN ign

i link i i | | fines \

| a el

ol ae

=o ot et Po)

; 71 — ij f

i | |

\ \ i _ =

Ls Sy ff

we \ i [ass j

fu /

| 5

. A eel

Y > 3 ed

Fie. 34.

Fusarium oxysporum Schl. f.8 Snyder ; Conidia from pionnotes of 2 weeks old culture on (a) bean pod,

b) hard potato agar, and (c) oat agar.

Conidia in sporodochia and pionnotes, 3-septate, less frequently 4—5-septate, ex-

ceptionally 6—7-septate ; smaller 1—2-septate comidia are also found, and numerous micro-

conidia scattered in the mycelium.

BRO NENE. o60 7645505 A aaie) x cte—Donca pec Mostly 32-38 x 3:8-4:3.

5-septate............ 35-71 x 3-4-5....... Mostly 35-50 x 44:3

Sclerotia and sclerotial bodies occur, they are 0-5-2-5 mm. diameter, blue, green

or pale. Chlamydospores 4-14 u diameter.

Hab. Pisum sativum h.,? from stems of wilting pea plants, Tygerpoort, Pretoria Dist.,

(Fourie).

F. oxysporum f. 8 is parasitic on peas (Pisum), causing the disease known as “St.

John’s wilt” in Hurope, and “ near wilt” in the United States. An organism apparently

identical with this form was obtained in pure culture from a number of ‘wilting pea plants

from Tygerpoort, but the identity of the South African fungus with F'. oxysporum f. 8

needs confirmation by inoculation.

428

Growth on Standard Media.

In culture /. 8 resembles F. oxysporum, and is just as variable in the pigmentation

of the stroma, and in the presence or absence of dark blue or pale sclerotia. The presence

of an aromatic odour in rice cultures is also variable.

Measurements of Conidia.

Oat agar, culture 2 weeks old, conidia from pionnotes.

DONO ncooccenosce 9A jJOOPCAMercsceoe 42-5-50 x 3-7-4-7.

ArSeptateiyiuswessas + Al Pee ieais ra’ 5 37-5-47-5 x 4-4.

3-septate............ 34 Sy Scams aa 27-5-50 « 3-7T-4:-7.

IASSOHAUDoccaccccccse WW 5

O-septatennesneerccic BAG) a

Hard pee agar, culture 2 weeks old, conidia from pionnotes.

9-septate .. Raa ae Onlyaonerseenine tie 72°5 x 3°75.

A-septate............ IJ2°5 Rote Duet eae Ms 40-52-5 x 3:7-4-4.

SHIVINAD ocaccvsaccde 66°5 RES eRe te R EAC 25=60) X38 5—£-5.

PASSOUEW® ooocovedooee 3 per cent.

USSPWAWO so ocossco5oe 6 Pr

OASCMAUD cocconccocce 4 59

Bean pod, culture 2 weeks old, conidia from pionnotes.

BSC OEM ca0c00 600006 Rare wicgiahe eral stele Mae 52°35 x 4:4

AL Sep baten it erare ee A JSP CBMs codaocsse 36-50 x 4-4°-7.

BHIG|OUENWS 5 occccccnces BO Uy oaseeecnegeenrcens 27-59-45 x 3-5-4-5

OHSE[MAWE ooccooecoocs Ve a5

Fusarium oxysporum Schl. var. aurantiacum (Lk.) Wr.

Wollenweber, Fusarium-Monographie, 420-422, 1931; Fus. aut. del. 381-386, 627, 1013-1016,

1185-1187. Wollenweber and Reinking, Die Fusarien, 121-122, 1935.

Syn. Fusarium aurantiacum (Lk.) Sace; F. calcareuwm (Thuem.) Sace.

F. elongatum Pratt; F. Peckiw Sace. pr. p.

F. Saccardoanum Syd.; F. sclerodermatis Peck.

F.. sclerotioides Sherb.

This variety has somewhat larger 3-5-septate conidia than the type, and a larger pro-

portion of 4-5-septate conidia. Sclerotial bodies are sometimes 1-3 mm. diam., and some-

times more extensive, 46-12 mm. On rice the stroma is a deeper purple violet than in

cultures of F. oxysporum, and is sometimes almost chestnut brown. Rice cultures are not

aromatic.

O=septatenaeeeriasei ke D:-D-950) X 22252. Averagei-2 x 226:

IESEVOUENIS Ss cocce es ones IZM SX DPO=BeS 4.50 - Average ee x 8. i.

= “Se plateraeromrieaic UB189) € SWB) o 5006 Mostly 30-42 x 3:5-4:-8.

DANN ooaecas000nc B8=10 XX SaD°Doocss 55 Mostly 38-57 x 3-8-4:-7.

7-septate <ehioneoe ate 36-75 x 3:°3-4-5..... Mostly 41-65 x 3-7-4-6.

Chlamydospores more or less common, spherical to oval, 8-5 « 8 (5-12 diam.), 2-celled

11-14 x 7-9 (average 13-5 x 8).

Hab. Antirrhinwn majus L., from stems of wilting plants, Johannesburg (Wager).

Arachis hypogaea L., from pods showing a pink discolouration of the shell, University

Farm, Pretoria (F. du Toit).

Brassica oleracea L., from stems of wilting seedlings, Witpoortjie, Krugersdorp Dist. ;

also on half grown plants showing symptoms ‘similar to those of ‘ ‘yellows.’

Cupressus lusitanica Mill., from stems of dying seedlings, Xumeni Forest, Donnybrook,

Natal, M.H. 28388.

429

Phaseolus sp. from stems of wilted plants, Nelspruit, E. Transvaal (Wager), and {from

Swaziland.

Pinus longifolia Roxb., from dying seedlings, N. Transvaal (Bottomley).

Pinus palustris Mill. and P. taeda, from stems of dying seedlings, Dukduku plantation,

St. Lucia Bay, Zululand.

This fungus occurs in Europe, Asia and America.

It is a saprophyte on decaying

parts of plants, and is parasitic on conifer and cyclamen

seedlings.

Jae, 8%

Fusarium oxysporum Schl. v. aurantiacum (Lk.) Wr.; conidia from (a) sporodochia of culture on

Melilotus stem, (b) pionnotes on hard potato agar, (c) pionnotes on standard synthetic agar plus starch,

(d) sporocochia of culture on oat agar; all cultures 2 weeks old; (e) chlamydospores from a 4 weeks old

culture on hard potato agar.

Growth on Standard Media.

Oat agar : Aerial mycelium sparse or moderate in amount, cottony or matted, white

to pale flesh colour, or tinged pinkish vinaceous near the substratum. Growth on sub-

stratum purplish vinaceous. Large, irregular, wart-like sclerotial masses developed,es-

pecially near the base of the slant ; they were up to 5 mm. diam., and were at first colourless,

then tinged with green, and finally blue-black. Sporodochia developed in groups, and were

often coalescent, forming a continuous pionnotal layer; conidia in mass were light pinkish

cinnamon to ochraceous salmon.

Hard potato agar: Aerial mycelium sparse, white, cottony ; growth on substratum

colourless. Numerous minute sporodochia developed, which soon ran together and formed

a continuous pionnotes. A few minute sclerotia appeared at the base of the slant.

Standard synthetic agar plus starch : Aerial mycelium very sparse. Growth on substratum

deep vinaceous lavender, and after 8 weeks, the agar was stained light russet vinaceous.

Pionnotes developed freely, buff pink.

430

Potato agar plus 5 per cent. dextrose : Aerial mycelium fairly plentiful, cottony or matted,

white to pale vinaceous lilac. Growth on substratum dull Indian purple to dark slate

purple. After 4 weeks, the agar was stained brown to almost black. Numerous deep delft

blue sclerotia were present after 4 weeks.

Potato plug: Plug covered with a dense mycelial growth, which was white to pale

lilac, cottony or becoming felt-like and wrinkled. Sclerotial masses at first small and pale,

becoming bluish-green and finally blue-black and developing into irregular, raised, rough

masses, up to 5 mm. or occasionally 10 mm. in diameter. A few sporodochia developed ;

they were pale pinkish cinnamon. :

Melilotus stem: Aerial mycelium rather thin, white to shell pink. Stems became

covered with numerous sclerotial masses, which were 1-3 mm. diam., at first pale, then

greenish blue, and finally blue-black. A few sporodochia developed.

Bean pod : Mycelium rather thin, or moderate in amount. Pinkish cinnamon sporo-

dochia developed in 14 days; they were small, scattered or in groups. A few minute, blue-

black sclerotia were present.

Rice; Growth white to alizarine pink, venetian pink and old rose, becoming acajou

red and pompeian red in places. Masses of plectenchyma developed between the medium

and the glass. Not aromatic.

Measurements of Conidia.

Hard potato agar, culture 14 days old, conidia from pionnotes,

SUC scossccsceso A JOS COMGossccace 30-00 X 4-5.

DHSAMEWScooogdenaccs | Pin oer Ase 36-55 x 3-7-4:-7.

A-septate............ 23 B NST craps ee 35-55 K 3-4-5.

S-sembatenaciceeeiae:. « OONON 2 has ameter tana 37:°5-45 x 3-7-5.

Q=Seplabericaricudiasccas, (OUD, pee saaenr 22:9-40 x 3-7-5.

O-septate .. Roel

Bean pod, culture 14 days old, conidia from sporodochia.

H-SO|NUAWS osoccoccenoc BU) DOR Cilio s oe o0ce 30-904 3:5-5.

MAINED coocascoeenos Ie Sak We Bsaetescarte 325-50 <x 3-5-5

BPAIINAWO onconooacsens BY) BAe Wate ti ginlo.6 27-45 x 3-3-5.

PASO > ocoeodoasas Il Sa eae 20-30 x 3-7-4.

ASCTOUAE scoop oo0ee 115) A

OASC[DUEWO oooon 06006 1

Oat oe culture 2 weeks old, conidia from sporodochia.

DAI NEMOsoooteoscace It WEP COM. ocoocccnn 42-5-60 x 3-7-4.

Acevsyaiteiie B Ment cle cia ae Thien sen NS ANE weatiat Soe 37:°5-57-5 « 3-3-5.

SHICPOUBUS ooobaccaccoc 82 il CEE AS Fe 27 -5=50! & 3+38=4-5-

Standard synthetic agar plus starch, culture 2 weeks old, conidia from sporodochia

DAL|HEWO.ooccooocacsec IQ JOOP CMs o oco00% 49-01-90 X 3-7-9.

A-Sepuaber eer tae O pi te Sst Goer 425-55 x 3-5-5.

GFEC|MEUSscogoocadcca Oy sRay VieceaeRoe reas 30-50 x 3:5-4:5.

ESeptabe nasa cen: 0-95

OSZMAUC cccnvc0ccce0 Ore

Fusarium oxysporum Schl. var. gladioli Massey.

Massey, Fusarium rot of Gladiolus corms, Phytopathology 16; 509-523, 1926. Wollenweber and

Reinking, Die Fusarien, 122-123, 1935. Wollenweber, Fus. aut. del. 1183, 1184.

The conidia of this variety are broader than those of the type, and in this respect

approach in form those of var. aurantiacum. Conidia, measure :

Ozseptatenn.c.. gece Olen.

Seu coscaccooc , Wa >< Ber,

O-septate............ 20-41 « 3-0-4938"... Average 334 x" 423"

D-II OUENE oa acowbooudc 42-46 x 4-3-4°6..... Average 44 x 4:5.

(eseptaterinn ici. screen 40-44 x 4°8......... Average 42 x 4-8.

431

The macroconidia are produced in salmon-orange sporodochia, which are up to 2 mm.

diameter; they are 3-4-septate, less frequently 5-septate, or exceptionally up to 7-septate.

Chlamydospores spherical, smooth, terminal or intercalary, mostly 1-celled ; chlamydospores

in or on the conidia are smaller (6-14 » 5-10) than those arising in the mycelium (7-17 »

7-10). The aerial mycelium is floccose, well developed, white, and up to 5 mm. high. Dark

blue sclerotia are present and numerous. ;

Hab. Gladiolus sp., from corms and leaf bases, Princess Park, Pretoria.

The younger leaves turned brown, and plants failed to flower; corms were still firm

but were discoloured brown, especially near the base. The identity of the organism with

F. oxysporum v. gladioli needs confirmation by inoculation into healthy plants. Variety

gladioli is the cause of decay of Gladiolus corms in North America and Australia.

— \

Fic. 36.

Fusarium oxysporum Schl. y. gladioli Mass.; conidia from pionnotes of 2 weeks old cultures on (a)

oat agar, (b) hard potato agar.

Growth on Standard Media.

In culture, the Gladiolus organism did not differ materially from F'. oxysporum.

Measurements of Conidia.

Melilotus stem, culture 4 week old, conidia from sporodochia.

HAIN AMSocasoscooocs Oll JOSE CBMGocoscnse 20-37-5 x 3-7-4-7.

lESepuateriy. ra tatwoe) 1a SHO! | sgh gi omierusc s. L25-18, xX 3=3"5:

O-sepbatere sat. ue... 33 Ae ens ear ee 4-1] x 2-5-4.

Standard synthetic agar plus starch, culture 12 weeks old, conidia from sporodochia

OHH BUS ccooscacoos WEB joer Calc ooo occas AZ x 4:7.

DESCPLALEMer teats aes AeA) Capes bly acaed ae 40-45 x 4-4-7.

ALS No bacon oo coe 17 PaaS Ecce ue 40-42 x 4-4-7.

S-septate......0..... 48 Sn Baa es ee i 28-42 x 3-5-4-7

MAINED ooceceocsos | 4s RR opicio ce BENG 15-25 x 3-3-4.

IESepbateree ce iacke re ey-0 ye PU oho 15-20 x 3-3-5.

Ozseptater wiki caniel.% DA emer Wi, Age ERS: 6-10 x 2-5-3

Fusarium dianthi Prill. et Del.

Delacroix, La maladie des oeillets d’Antibes, Ann. Inst. Agron. Nancy, 16: 1901. Wollenweber

and Reinking, Die Fusarien, 123-124, 1935. Wollenweber, Fus. aut. del. 1188-1189.

The conidia occur in light orange, sporodochial and pionnotal masses; they are

fusiform-falcate, pedicellate, often abruptly bent at the constricted apex, and a little thicker

432

im the upper third than in the middle, definitely dorsiventral, mostly 3- or 3—-5-septate,

exceptionally 1—2- or 6-8-septate. Micorconidia numerous, |-celled, or with 1—2-septations,

scattered in the floccose aerial mycelium, which is white to pink.

O-septate (2.4.0. 255 0-1 X 10-4-5.- 3525, Mostly 7-11 x 2-376:

IEseptateceaa. see O30) Xe 2>4e een Mostlyalla— 2 eo sti.

SNM Oagassedemace IMO iRyo< Ao —Aen cone Mostly 23-52 x 3-1-4-5.

QUAM s uses cou « 30=80) < 3-045). eee Mostly 37-69 x 3-6-5.

1-9-septate... 70-100 x 3-8-4.

The more compact 2 septate conidia average 31 x 4- i the more slender 37 x 3-7.

Chlamydospores round, smooth or rough, 6-12 (av. Sal) 2 -celled ellipsoid-oval, 13-16 x

5-13. The fungus is not aromatic.

Hab. Dianthus caryophyllus L., from stems of wilting plants, Bethlehem, O.F.S. ;

Klim and Politsi, N. Transvaal; Durban, Natal; Golden Valley, Cape.

a , b.

Hie.) 37.

Fusarium dianthi Prill. et Del. ; conidia from sporodochia of 8 weeks old cultures on (a) potato agar

plus 5 per cent. dextrose, (b) oat agar.

Growth on Standard Media.

In culture, the strains studied closely resembled F. oxysporum var. aurantiacum.

Measurements of Conidia.

Oat oe culture 8 weeks old, conidia from sporodochia

O-Sen ua. bene erteeertes 0-5 per cent x 4-5-5.

DHAIS|OUEN. cow os ob oce 2 ALC i oreerenana) See 40-60 x 4-5.

4-septate............ 4 Pe aN ate 3 35—92) x 38-425

CHAS DUE coo acceane 89 sont Ua ates ieee 23-45 x 3-4-5

septate 0-5 St aebstete 2.6 16-20 x 2-5-3.

O-septates: sis. set. 4 be ites, MN ie 8 6-12 x 2-3-4.

Potato agar plus 5 per cent. dextrose, culture 8 weeks old, conidia from sporodochia

D=Se Pt abaya Meionus beni Gye etalk cere eel aee ee tee 25-40 x 3-4:5

A-Septabe an cl jenn ley. ew eee sie eee 20-30 x 3-5-4.

L-septatenis gs Ree, «ou. ashen nec reeee 15-24 x 3-83-75.

Fusarium vasinfectum Atk.

Atkinson, Some diseases of cotton, Agric. Exp. Sta. Alabama, Bull. 41; 19,1892. Wollenweber,

Fusarium-Monographie, 423, 1931; Fus. aut. del. 376. Wollenweber and Reinking, Die Fusarien,

124, 1935.

Syn. Fusarium malvacearum Taub.

Differs from F'. oxysporum in the somewhat narrower conidia, the free development

of pionnotes, the comparatively small, green to blue sclerotial plectenchyma (0-1-2 mm.

diam.), and the purple-red plectenchymatous stromata. Microconidia 1-celled, or with one

433

or two septations, scattered. Macroconidia in sporodochia and pionnotes, isabellinous to

- light salmon orange in mass; fusiform-falcate, somewhat constricted, tapering or rostrate

at both ends, base pedicellate or papillate.

OFSepua resem tre: AND x ZEB ties e sine Mostly 6-7-9-5 x 2-3-2-8

lESeptabeser sslsner ck PAD SC AE 50s UO in Mostly 13-20 x 2-8-3-3.

3-Septate............ 23-48 x 3-4-5....... Mostly 27-40 x 3-3-4.

*H-septate.| jis 5 -- 30-50 x 3-5 rare).... Mostly 32-48 x 3-5-4.

Chlamydospores terminal and intercalary, 1-celled, 7-13 (av. 8-8), or 2-celled 12-6 x

7. The fungus has a strong, lilac-lke odour on rice media.

Hab. Hibiscus sabdariffa L., from stems of wilting plant, Schagen, E. Transvaal (Wager).

This fungus is the cause of a vascular wilt of cotton, Gossypium herbaceum and G.

barbadense, and probably also of Hibiscus esculentus. It occurs most frequently in North

America. A vascular wilt of cotton caused by a Fusarium sp., has been observed in South

Africa, and is probably due to this fungus, but it was not found during the time this work

was in progress, and has not been studied in culture, nor have any tests of its pathogenicity

been made.

Q \ ia) : rat

Fie. 38.

Fusarium vasinfectum Atk.; conidia from (a) pionnotes of culture on plain agar, 7 days old; from

sporodochia of 4 weeks old cultures on (6) oat agar, (c) standard synthetic agar plus starch, and (d) pionnotes

of 8 weeks old culture on hard potato agar. (e) chlamy dospores from my celium on hard potato agar. culture

4 weeks old.

Growth on Standard Media.

Oat agar : Aerial mycelium rather sparse, fine, white, cottony. Growth on substratum

vinaceous pink to light vinaceous pues numerous minute sclerotia developed, especially

near the base of the slant, up to 2 mm., diam., at first pale, becoming blue-black. Groups

of sporodochia were light buff to light pinkish cmnamon. In some tubes there was a patch

of dusky dull green plectenchyma ‘at the base of the slant.

Hard potato agar: Mycelium short, white, sparse. Growth on substratum colourless.

A few small sclerotia formed along the edge of the medium, especially where it was drying.

Pionnotes developed along the needle track

434

Standard synthetic agar plus starch : There was a very little aerial mycelium at the base

of the slant, elsewhere none. Growth on substratum vinaceous pink to deep vinaceous

lavender. Pionnotes and groups of sporodochia light buff to light pinkish cinnamon.

Potato agar plus 5 per cent. dextrose: Aerial mycelium moderate in amount, white to

flesh pink and deep purplish vinaceous, sometimes becoming acajou red. Growth on sub-

stratum dull Indian purple to Hay’s maroon, sometimes with patches of deep delft blue

or Russian green. The medium was, in some tubes, stained sorghum brown to vinaceous

brown.

Potato plug: Plugs covered with a moderate amount of white, cottony mycelium,

which sometimes became wrinkled and felt-like. Sclerotia numerous, minute, 1 mm. diam.,

or up to 2-5 mm., at first colourless, becoming greenish blue to blue-black. In some tubes

there were patches of dark bluish grey green on the substratum. A few small sporodochia

developed.

Melilotus stem: Aerial mycelium scant to moderate, white, cottony or tufted. Sclerotia

numerous, small, deep delft blue. Sporodochia fairly numerous, light ochraceous salmon, 2-3

mm. diam.

Bean pod : Pods covered with a white mycelium, which was thin and cottony or felt-

like and wrinkled. Fairly numerous small sclerotia and pale pinkish cinnamon sporodochia

developed.

Rice : Growth at first flesh pink to alizarine pink, or purplish hlac to pruplish vina-

ceous ; later it was jasper pink to old rose. Groups of colourless plectenchyma developed

between the medium and the glass. Culture aromatic, with an odour resembling lilac.

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from sporodochia.

ASOD. oAooaescocn oll WSOP COMsccsoo0- 30-40 x 4.

SEONG es ome sae SkOUOO nw py eedowolne 30-42°5 x 3-7-4-5

2-septate...........- 2 5OD URN Tt Edema a oko 25-35 x 3-4-5

Usseptates cis. bits cote LO eee es ae 20-24 x 3-3-7

OFseptatese ei aimee MO! OM tl, Mememat ner etme 7-10 x 2-3.

Hard potato agar, culture 8 weeks old, conidia from pionnotes.

4-H-septater.c: 2 sceayo. OW. 2ae nes eae 4047-5 x 3-7-5.

SNUG. chdssscouns 9» 2A TOE CHM cccscce06 27-5-40 x 3-4:9.

ZASEPLALC sancti Gre ai ee Cates) eee mB) SK Bh

Lseptatercneiw atch Cini lo aieeaay 2 ligt wae a eNeees: 15=25 x 3-1-4.

OBC MAMBecucocccoce. GY)

Chlamydospores developed in many conidia.

Melilotus stem, culture 8 weeks old, conidia from sporodochia.

SHAS|OUENO. 000 v0 go0000¢ 45 per cent.......... 18-7-40 x 3-4-5.

2sseptate. (iadelsacer Lm Ween eter ee 17-5-30 x 2-75-3-75.

Iesepbate naar iran. coger tern hor 12-5-20 x 2-5-3-75.

OASEDUANE. occ voconcoe DOia! Asai Te meee. 5-9-5 « 2-3.

Standard synthetic agar plus starch, culture 4 weeks old, conidia from sporodochia

BOI MMCSsoccoesccnac 0) OIF CBM cos asoeec 30-42:°5 x 3:7-4:-5.

Aseptate; Mia. s). Cee: 1 LOL Ey ein eran 22-5-35 x 3-7-4-5.

L=septatens av Ny owe) gid ane eel ke ase 17-5-30 x 3-7-4:5

OCEANS ncaccacceee |

Fusarium vasinfectum Atk. f. 2 Wr. et Rkg.

Wollenweber and Reinking, Die Fusarien, 125, 1935. Wollenweber, Fus. aut. del. 1191.

This form was established by Wollenweber and Reinking to include provisionally a

number of fungus strains, which are morphologically similar to F. vasinfectum, but which

have not proved capable of producing cotton wilt. The typical 3-septate conidia correspond

435

with those of the type, and are 24-50 x 3-5-4-5.. On rice media the stroma is purple red,

and, cultures on rice may or may not be aromatic. Chlamydospores terminal and inter-

calary.

4 gE \

S ve

¢ ae i \

ww)

(

Ime, Bk);

Fusarium vasinfectum Atk. f.2 Wr.; conidia from (a) sporodochia of 4 weeks old culture on Melilotus

stem ; pionnotes on sclerotia of 12 weeks old culture on () rice and (c) potato agar plus 5 per cent. dex-

trose, (d) chlamydospores from 12 weeks old rice culture ; conidia from pionnotes of 9 weeks old cultures

on (e) hard potato agar, and (f) oat agar.

Hab. Citrus sinensis Osbeck, from fruit: from brown lesions at the stem end, varying

from small discolourations under the button to extensive rotted areas, after fruit had been

18 weeks in storage; in Valencia oranges from Sunday’s River, Cape (2 isolations), and

White River, Transvaal (5 isolations).

On trunk ; on strips of bark, apparently killed by

lightning, Grahamstown, Cape, Nov. 1931 (Jolly) M.-H. 28420; on bark above union,

(roots of trees water-logged), Letaba Estates, N. Transvaal.

On roots of old seedling orange tree, Villiersdorp, Cape.

Citrus limonia Osbeck, from roots showing dry root rot lesions, Bonnievale, Cape

(van der Hoek) M.H. 28402; Prudential Estates, E. Transvaal, 1930, .MH. 28440.

Matthiola incana R. Br., from stems of dying plants, Uitenhage, Cape (Wilson).

Messembrianthemum sp., from stems of succulent plants which were rotting, Pretoria

(Wager).

Sesamum orientale L., from stems of wilting plants, Pretoria University Farm, April

1932 (F. du Toit).

436

Fusarium vasinfectum Atk. var. lutulatum (Sherb.) Wr.

Wollenweber, Fusarium-Monographie, 424, 1931; Fus. aut. del. 380, 1019, 1192. Wollenweber

and Reinking, Die Fusarien, 125, 1935.

Syn. Fusarium lutulatum Sherb.

a, b c {

(aU), Wey BON :

A ee

if af — \ - \

ee | Fl |

B a

jee ey (Be ae

| i j ]

eA ; /

CEN a

+ os

‘ \

Fic. 40.

Fusarium vasinfectum Atk. v. lutulatum (Sherb.) Wr.; conidia from sporodochia and pionnotes of 2

weeks old cultures on (a) oat agar, (b) hard potato agar, (c) standard synthetic agar plus starch, (d) Melilotus

stem ; chlamydospores from (e) 4 weeks old culture on hard potato agar and (f) plain agar plates, 7 days

old.

This variety has somewhat longer conidia than the type, and small, blue-black sclerotia]

bodies (up to 0-5 mm. diam.) may be numerous or absent. The conidia are mostly 3-septate,

seldom 4- or 5-septate. Numerous 1—2-celled microconidia occur in the aerial mycelium.

SSeptaber lens. «isk Meee cad ameerea tutes ks brekueieters Mostly 28-42 x 3-2-4-5.

H=sepbabes.. a. cis.ccm « Slee ae ole wane SR nese Mostly 37-47 x 3-5-4:5.

6-Tseptates.7...o cn meee Eee oO eee Exceptional, 50-66 « 3-5-5.

Chlamydospores terminal and intercalary, 1-celled, 6-8 x 5-7; 2-celled, 8-12 x 4-7.

The fungus is aromatic on rice media.

Hab. Centaurea moschata L., from stems of wilted plants, Pretoria (Wager) M.H. 28405.

Lathyrus odoratus L., from stems of seedlings which were wilting, and also from stems

of plants dying when reaching the flowering stage (sometimes associated with Pythiwm sp.),

Brooklyn, Pretoria (Doidge).

Pisum satwum L., from wilting seedlings and older plants, H. Transvaal (Wager).

437

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, white, cottony. Growth on substratum pale

vinaceous pink to pale flesh colour. A number of minute, scattered sclerotia, 0- 1-0-5 mm.

diam., which became deep delft blue, developed in some tubes. Numerous small sporodochia

developed on the lower half of the slant; these were light vinaceous cinnamon to light

ochraceous salmon, and they remained discrete, or coalesced to form a continuous pionnotal

layer.

Hard potato agar: A little, short, white, cottony mycelium developed over the face

of the slant. Small sporodochia and pionnotes formed freely ; they were pale cinnamon

pink to light vinaceous cinnamon.

Standard synthetic agar plus starch: Aerial mycelium scant. Growth on substratum

colourless, or with a purplish tinge in places. Conidial masses vinaceous cinnamon to flesh

colour, developing as on oat agar. Agar sometimes tinged pink. A few minute sclerotia

occasionally developed.

Potato agar plus 5 per cent. dextrose: Aerial mycelium scant to moderate in amount,

white to shell pink. Growth on substratum became wrinkled white to flesh colour, with

patches of dark delft blue, which, after 14 days spread all over the slant, and in 8 weeks

became almost black. The agar was stained dull Indian purple.

Potato plug: Plug covered with a mycelium which was cottony, white tinged lilac ;

or felt-like and wrinkled, pale salmon colour to seashell pink. The colour faded with age,

and patches of deep delft blue appeared on the substratum ; occasionally a few minute

sclerotia developed. Sporodochia very numerous, pale salmon colour to light pinkish

cinnamon, well developed after 4 weeks.

Melilotus stem: Stems covered with a short, white, felt-like mycelium. Sporodochia

developed after 14 days: they were very numerous, scattered or in groups, minute, buff

pink to light pmkish cinnamon.

Bean pod: Pods covered with a fair amount of white mycelium, which was cottony,

or tomentose to sericeo-tomentose. Sporodochia developed in groups after 14 days, and

were small, pale ochraceous buff.

Rice: Growth at first white to venetian pink and alizarine pink, or pale vinaceous

later white to vinaceous and old rose. A few sclerotial bodies developed against the glass

Cultures were aromatic, with an odour resembling lilac.

Measurements of Conidia.

Hard potato agar, culture 4 weeks old, conidia from pionnotes.

DREN. oscogccce be I) JOS? CSiliiic s oe no oguc 30-52°5 x 3-75-50.

AE SC MULCH er iyrsgacs HON) 6 soa Uns. musesye omg = 35-52:5 x 4-4-5.

SeSepuaverer eer OO! ly ela veae eklok 25-47-5 x 3-7-4-5.

lEsep tates ers Mes 935

(O-SETNIEN. coon ucoscae 4

Oat agar, culture 4 weeks old, conidia from sporodochia.

SH NAS. cooceccoaoe , U [Mee CAMiccooosoccc 37-5-47-5 x 3:7-4:5.

A-SEP CALC ere ele io i er aye es Bea oeiia dic 35-42-5 x 3-7-4:9.

SUMS, cho oaceseog, UO) li |e iice.upacscdon 30-47-5 x 3-7-4:-5.

O-Sep baler tretolrsern 5 se

Melilotus stem, culture 4 weeks old, conidia from sporodochia.

USO DIEM G46 cudpao bee OF by perkcenttan cerita 30-40 x 3-7-4:-5.

S-SePlatewy-ee re) Oe ele eet enys 20-45 x 3:5-4:5.

Z=SEPUALC Naas te A <y- 0-5

lEsepbaternsh a scuasst sy oy L

O-sepuatesyeree see 6 3

438

Bean pod, culture 4 weeks old, conidia from sporodochia.

AERGOUMIO. co5o oo ond o 7 ACIS Se aaa ee Vict 30-38 x 3:7-4:-5.

SHED. ooo agave see 73 Kecmaniakatonehve sae 225-40 x 3-2-4.

PSSNUMN a 504 s000 006 Ole Shaheen wseeole isle 15-17-5 x 3-2-3°5

WESC c 00 eu do co. See oe a emetane te 15-17-5 x 2-7-3.

C"septaitessa. a: 70 radiata Miami EN a 3°2-10 x 2-5-3.

Fusarium vasinfectum Atk. var. zonatum (Sherb.) f. 1. (Lk. et Bail.) Wr.

Wollenweker, Fusarium-Monographie 425, 1931; Fus. aut. del. 629, 1021. Wollenweber and Reinking,

Die Fusarien, 126, 1935.

Syn. Pusariwm cepe Walker et Tims.

F. zonatim (Sherb.) f.1 Lk. et Bail.

ay Pe a ¢

Fie. 41.

Fusarium vasinfectum Atk. v. zonatum (Sherb.) f.1 (Lk. et Bail) Wr. conidia from (a) sporodochia of

- weeks old culture on Melilotus, and (b) thin pionnotes of 8 weeks old culture on oat agar, (c) chlamydo-

spores from 12 weeks old culture on hard potato agar.

Differs from F. vasinfectum and its other varieties in the colour of the stroma and the

conidial masses; growth on some media in concentric zones. Stroma pale, cream-coloured

to salmon ochre, seldom purple red. Dark blue sclerotia and sclerotial stroma absent,

but occasionally in cultures there occur erumpent, blister-like, raised, dark brown knots

of plectenchyma from 0:5 mm. diam. Chlamydospores abundant ; microconidia scattered

in the aerial mycelium ; macroconidia in salmon buff to ochraceous salmon pionnotes and

sporodochia.

3-septate............ 27:2-46-5 x 2-8-4-6 Average 37 x 3-8.

4-septate............ 34:4-50 x 3-2-4:6.:. Average 41:5 x 4.

DAGOUMOsccoccecccss SiH) K Boal... 06 Average 43-4 x 4.

The fungus is aromatic.

Hab. Lycopersicum esculentum Mill., from stems of wilting plants, Matatiele, E. Griqualand

(Wager) M.H. 28387.

This fungus is known as a cause of bulb rot of onions, and also occurs on carrot, tomato

and tulip in North America and Europe.

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, white, cottony, mostly on the lower part of the

slant. Growth on substratum colourless: Pionnotes developed more or less freely; they

were salmon buff to ochraceous salmon.

Hard potato agar: Aerial mycelium sparse, short, white, rather coarse. Growth on

substratum colourless. Pionnotes developed slowly.

Standard synthetic agar plus starch: No aerial mycelium. Growth on substratum

colourless to pale ochraceous buff. Pionnotes developed slowly.

439

Potato agar plus 5 per cent. dextrose : Slant covered with a very small quantity of white,

cottony mycelium. Growth on substratum tinged pale to light vinaceous purple, shading

after 14 days to russet vinaceous. After 8 weeks, the agar was stained purplish brown.

Potato plug: Plugs covered with a moderate amount of aerial mycelium. No conidial

masses developed in the cultures studied.

Melilotus stem : Stems covered with a short, white, felt-lke mycelium. Sporodochia

developed in groups after 4 weeks: they were salmon buff to light ochraceous salmon.

Bean pod: Pods covered with a short, white, felt-like mycelium, with patches of

longer, cottony hyphae at the top. No conidial masses developed in the cultures studied.

Rice: Growth white to flesh pink or salmon buff. The culture was aromatic.

Measurements of Conidia.

Melilotus stem, culture 8 weeks old, conidia from sporodochia.

DAU ocoaescoses Wad) jose CaM, 2o5050- 33-45 x 3-4-5

ARNE coo do oneous 2 amarante 23-45 x 3-5-4:-5.

SFE NENIO. odo boaeucas DG OVE ae esis caarete 225-49 x 3-3-4.

PSE DIEM 6 bn oo oon Oc 0-5

ORES: conocooocus 0-5

Oat agar, culture 4 weeks old, conidia from plonnotes.

DANA ONUG. 66605 cb 01 Mota) Joe CAMs aioe o sc 40-48 x 3-5-4-5.

4-septate............ 7 se Sscocartienttors 40-46 x 3-7-4.

SHAG OIENSabde oeabeeds 63 ers her as 25-425 x 3-4-5

2-septate........-..- SOl= ase entets Seaer 13-18 x 2-5-3-5

l-Sepiatemsan sora: 2 MED, atten Ses secs 12-15 x 2-5-3.

O2sepbatesaee anor 22 ah sttonzaleahadl age 5-112 x 2-3.

Fusarium vasinfectum Atk. var. zonatum (Sherb.) f. 2 (Lk. et Bail.) Wr.

Wollenweber, Fusarium-Monographie, 425, 1931; Fus. aut. del. 1021. Wollenweber and Rein- }

king, Die Fusar en, 126, 1935.

Syn. Fusarium zonatum (Sherb.) Wr. f. 2 Lk. et Bail.

Fie. 42.

Fusarium vasinfectum Atk. v. zonatum (Sherb.) f.2 (Lk. et Bail.) Wr.; conidia from sporodochia of 4

weeks old cultures on (a) oat agar, (b) standard synthetic agar plus starch, (c) potato plug and (d) Melilotus

stem.

This fungus is only slightly aromatic, has no sclerotia, and differs from f. 1 in the red,

almost purple, stroma, a lilac-tinted aerial mycelium and freely produced pionnotes. Conidia

3-6-septate, predominantly 3-septate ; 3-sept. 38-5 x 3:7; 5-sept. 42-1 x 4-1. Chlamy-

dospores abundant in mycelium and conidia.

Hab. Allium cepa L., from bulbs (scales showing light brown discolouration, moderately

firm to soft), Nelspruit (Wager) and Pretoria (Wager) M.H. 28407: from leaf bases of

wilting plants, Nelspruit (Wager), and Hikenhof, near. Johannesburg, October 1932.

This form occurs in North America, where it is a cause of bulb rot in onions. It also

occurs on beet.

440

Growth on Standard Media.

Oat agar : Aerial mycelium short, sparse, white or tinged lilac. Growth on substratum

deep vinaceous lavender to dull Indian purple. Numerous small sporodochia developed ;

they were light ochraceous salmon and 1-2 mm. in diameter.

Hard potato agar: Aerial mycelium sparse to moderate in amount, mostly short,

longer at the top and bottom of the slant. Growth on substratum colourless. A thin

pionnotes developed over the surface of the slant.

Standard synthetic agar plus starch: Aerial mycelium scanty. Growth on substratum

tinged vinaceous lavender. A few, light ochraceous salmon sporodochia developed near the

base of the slant.

Potato agar plus 5 per cent. dextrose : Aerial mycelium short, felt-like, white to lavender.

Growth on substratum vinaceous lavender to dull Indian purple. After some time the

growth became wrinkled.

Potato plug : Cylinder covered with a fairly copious growth of white, cottony mycelium,

which tended to become wrinkled and felt-like on the face of the plug. Very numerous,

small, ochraceous salmon sporodochia developed. In many cases, these seemed to arise

from small, brown to blackish masses of plectenchyma.

Melilotus stem: Mycelium scant, white to dirty white, very short and felt-like, or

sericeo-tomentose in coarse tangled strands. Numerous small sporodochia developed, and

also pionnotes ; these were pinkish cinnamon to ochraceous salmon.

Bean pod : Aerial mycelium moderate in amount, coarse, white, tomentose. Numerous

small sporodochia developed ; they were scattered or in groups, and pinkish cimnamon to

ochraceous salmon in colour.

Rice: Aerial mycelium at first white to vinaceous lilac, becoming rhodomite pink.

Growth on grains was vinaceous lilac to deep purplish vinaceous, becoming neutral red,

and finally alizarine pink to acajou red. Numerous small plectenchymatous masses formed

against the glass; they became brownish, then sepia to almost black. The culture was

slightly aromatic.

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from sporodochia.

D-sepuateuscae eee 5) [Oe CSM, sodasonse 42-5-52:°5 x 3-7-4.

4-septate. . adresse wala] Apia HE) ape Bd—025) xX om

aseptate ei. qtee ee LOM Me sil occ teteaice O58 27-5-52-5 x 3-25-4, mostly

35-40 Igie:

OASETNIANE soc aecn00e 3

Potato plug, culture 4 weeks old, conidia from sporodochia,

5-septate. Phids aeaencleyts «> DETNCE MIG ves sernr 31°0-49 x 3-T5—4

AEOUAUE. ooo oo dccoKs BIA Fae VRS Ste epatiete 30-40 x 3-4.

BOWEN). cooper coos as TQ) ol Deg an Walshe eee 30-425 x 3-4.

Teseptate sa. es 208 NS Mn seston Sete toed (tO Quire)

OA NAUGs.ccosccccand Zi se es ibe FRA cece 5-10 x 2-3-25.

mee stem, culture 8 weeks old, conidia from sporodochia.

DISQUS 5 2000 0000 0% I of5) [OS GSM 6 cocone 31-D-45 x 4-4-5.

A-Septauesy eee oc santa ct 30-45 X 3:50-4:50,

BONO; coc cc0ce0 sa QI eck Da ihat Peypae 25-425 x 3-5-4.

O-septate............ 2

Chlamydospores formed in many of the conidia.

Fusarium redolens f.1. Wr.

Wollenweber, Fusarium-Monographie, 426, 1931; Fus. aut. del. 1022. Wollenweber and Rein-

king, Die Fusarien, 127, 1935. .

Microconidia 1-celled, 9 x 38, or 1-septate, 16 « 4:5. Macroconidia 3-septate, less

frequently 4-septate, exceptionally 5-septate, fusiform-faleate, curved, sometimes recalling

44]

those of F. solani, but, in the more compact conidial forms, somewhat thicker in the upper

third than in the middle; gradually tapering towards the base, which is pedicellate or

papillate. Conidia in sporodochia or pionnotes, brownish-white, cream-colour, or light flesh

colour im mass.

3-Septate............ Wb X 3-6°5......-. Mostly 29-43 x 3-7-5:5.

D-Seplaueneneericn ear 31-61 x 3°5-6-5..... Mostly 37-47 x 4-6.

Chlamydospores terminal and intercalary, 1-celied 3-12, (mostly 6-9) 2-celled 11-24 x 5-14,

(average 14 x 8-2), smooth or rough, in conidia or mycelium. Blue sclerotia wanting.

Plectenchymatous stroma effuse, pale, pinkish white or lilac colour. The fungus is not

aromatic.

Hab. Lycopersicum esculentum Mill., from seed offered for sale (Wager).

{ i/ “

oe \

ae ey,

>>.

3 : SZ Pa:

Fig. 43.

Fusarium redolens f.1 Wr.; conidia from (a) sporodochia of 2 weeks old culture on Melilotus stem,

and (6) pionnotes of 6 days old plate culture on plain agar, (c) microconidia and chlamydospores from plain

agar plates, 6 days old.

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, white to shell pink. Growth on substratum buff

pink to onion skin pink. Pionnotes well developed, pale pinkish cinnamon or pinkish

cinnomon to orange cinnamon.

Hard potato agar: Aerial mycelium scanty to none. Growth on substratum colour-

less. A few small sporodochia formed near the base of the slant.

Standard synthetic agar plus starch : Noaerialmycelium. Growth on substratum colour-

less or tinged cinnamon. Sporodochia at first tilleul buff, becoming pale ochraceous buff

when well developed.

Potato agar plus 5 per cent. dextrose : Aerial mycelium scant, white, in patches. Growth

on substratum colourless. Pionnotes pale cinnamon pink.

442

Potato plug: Plug covered with a matted mycelium, which was tilleul buff to pale

vinaceous fawn, and became wrinkled when cultures were 3 weeks old. Conidial masses.

developed between the medium and the glass.

Melilotus stem: Aerial mycelium scanty, white. Sporodochia tilleul buff to light

pinkish cinnamon, not very large, and scattered or in groups.

Bean pod: Pod covered with a copious aerial mycelium, which was white to pallid

vinaceous drab to pale vinaceous fawn. Growth became wrinkled and felt-like.

Rice: Growth white to pale flesh colour. Small, white plectenchymatous bodies

formed between the medium and the glass. The culture was not aromatic.

Measurements of Conidia.

Oat agar, culture 19 days old, conidia from pionnotes.

SEI OEMoganonacanoo 474 0" COMsxoondcasn 29-49 x 6-6-25.

ASCP lava nace see Dir AR vistas Hosen 28-32:5 x 3-75-5.

EUAN cov aeoccoce MW eade meant aan: 22-5-29 x 3-7-6-25.

OZseptatesaqsee eee 32 Ae Fae Gere eee 6 4-5-10-5 x 3-7-4-5.

Melilotus stem, culture 15 days old, conidia from sporodochia.

ESN ogabsoonoso. 7 jOee CdMigdvacoac 40-44 x 5-6-25.

SHESNEN@scocccocasco ww) REE RRS Sisk essih 25-425 x 3-75-6.

Oe ae ee 0:5 ic wo hes ant et 21-22 x 45.

IL-SETOUANE. 00006008000 0:5 PAM he ces ainora ain 15-20 x 3-7-5

OSHMBooccnccoceas OES “te eee aa 7-5 X 3°75.

Standard synthetic agar plus starch, culture 8 weeks old, conidia from sporodochia-

A-Septateyncme ieee OR oRperacenteeemnane 40-45 x 4-6-5.

3-septate... sees: 99 a Meee Cine 28-38 x 4-8-6:3.

WAL) NEWS 60000000800 OD ee ibe a arr ee 30-31 x 4-8-5.

The shorter conidia often stouter than the longer ones.

Plain agar plate, culture 6 days old.

Dseptatestiat ss ca cig a eRe on aoc ace 45-55 x 4-6.

A=sepbaten.a sv is wrantiee oot hoo ile ee USA 42-5-52°5 x 4-5-5.

DS Sepbaber |. Aehak Ke Meise sence, Wn ces ee A241) xX 4-5.

Section MARTIELLA.

Wollenweber, Phytopathology, 3: 30, 1913. Wollenweber and Reinking, Die Fusarien, 127,

1935.

Fungi of this group are chiefly found in the soil and in subterranean parts of plants.

Macroconidia dorsiventral, fusiform to faleate, thick-walled, curvature slight in the central

part of the conidium, more decided near the apex; apex rounded or tapering; base sub-

pedicellate or mammillate. The medial diameter of the macroconidia is of diagnostic value

in this section. Microconidia mostly 1-celled, small, oval to oblong. Conidial masses pale,

white, yellowish or brownish, or in older cultures darker, honey colour to amber, or be-

coming tinged with the colour of the stroma. Stroma yellow brown to dark blue, the

brightest colours occurring on carbohydrate media. Sclerotial bodies erumpent on certain

substrata, brown, green, violet or blue black. Chlamydospores usually produced freely,

terminal or intercalary, developing in chains or clusters, smooth or rough.

It has been established that members of the genus Hypomyces represent the ascus

stage of a few of the Martiella-Fusaria.

Key to the South African Species.

A.—Dorsiventrality of conidia distinct only at the apical end.

a.—Conidia almost cylindrical to fusiform-faleate, ob-

liquely conical or rounded at the apex; obtuse,

mammillate or with an obliane nanilla at the base.

445

b.—Conidia 3- (0-5) septate; 3-sept. 34 x 4:8;

DESC DUM 2 HOw hire cin nice eislcmis cine Foie es F. coeruleum.

AA.—Dorsiventrality of fusiform-faleate conidia distinct at

both ends. Base mammillate to sub-pedicellate.

a.—Medial diameter of 3-5-sept. conidia 4-5 :

b.—Conidia in masses mostly 3-septate............ FE. javanmicum v. radieicola.

bb.—Conidia in masses mostly 3-5-septate.......... F. javanicum.

aa.—Medial diam. of 3-5-sept. conidia 5-6 :

b.—Conidia in masses mostly 3-septate :

e.—Conidia 3-sept. 36 x 5:5; 5-sept.49 x 5:3 F. solani.

cc.—Conidia 3-sept. 39 x 5; 5-sept. 49 x 5:3 F. solani v. Martw f. 1.

Fusarium javanicum Koord.

Koord, Verh. Koninkl. Akad. Wetensch. Amsterdam, 11, 13: 247, 1907. Wollenweber, Fus.

aut. del. 424, 426-428, 1025-1027. Wollenweber and Reinking, Die Fusarien, 131, 1935.

Syn. Fusarium theobromae App. et Strk. F. javanicum Koord., v. theobromae (App. et. shk.) Wv.

Fusarium heveae P. Henn. in herb. (non Vincens).

Fusoma glandarium Corda.

Conidia in mass brownish white to ight brown; when older, coffee brown, or tinged

with the colour of the stroma. Stroma leathery to gelatinous, seldom sclerotial, usually

olive green to olive brown. Microconidia 1-celled or septate, usually scattered freely in

the aerial mycelium. Macroconidia in sporodochia and pionnotes, falcate, slightly curved

often rather more decidedly curved at the apex, constricted at both ends, more or less

pedicellate at the base, 3-5-septate, exceptionally 6—8-septate.

O-septate....... 8 x 3.

Teseptatesen ue. Ike} 3< BW,

d-Septateric......... 22-04 % 3:5-6..02.... Mostly 28-47 x 4-5.

DAVOS o5oo0seaccce CD00) X 4c occoua0s Mostly 40-56 x 4-5-5-3.

UFSEINNUD 0a0006 000000 60 x 5.

Chlamydospores 1—2-celled, 5-8 diameter.

Hab. Cucurbita Pepo L., and C. maxima Duchesne ; from stems of wilting plants of pumpkin,

marrow and Hubbard squash, Hennops River and Daspoort, Pretoria Dist. (Kresfelder)

M.H. 28414; Uitenhage, Cape, Nov. 1935 (Haines.)

This fungus causes extensive damage in commercial plantings of pumpkin, marrow

and squash. In other warm countries, fF. javanicum is known as a rot-producing organism

in coffee, cocoa, rubber, etc. It also occurs in the temperate zone on poplar and elm.

Growth on Standard Media.

Oat agar: Aerial mycelium scant to none. Very numerous, minute sporodochia

developed, which coalesced more or less completely to form a continuous pionnotes ; the

conidial masses were cream buff, and developed in irregular patches. Growth on sub-

stratum colourless to wood brown.

Hard potato agar: Mycelium not abundant, rather coarse, short, white to ivory

yellow. Sporodochia numerous, minute, crowded, coalescing to form pionnotes, cream buff

to chamois, often developing in concentric zones round the point of transfer. Pionnotes

on this, and on other media, rather dry, and inclined to crumble when touched with a needle ;

less frequently of the consistency of cream cheese.

Standard synthetic agar plus starch : Aerial mycelium scant to none. Conidial masses

developed freely all over the surface of the slant, often in concentric zones round the point

of transfer, cream buff to light pmkish cinnamon. Growth on substratum colourless, except

at the base of the slant, where it was wood brown.

Potato agar plus 5 per cent. dextrose: Growth dense, felt-like, zoned, wood brown,

avellaneous and cartridge buff, or citrine drab and yellowish olive. Pionnotes developed

freely ; they were wood brown or tinged lincoln green.

444

d LE i cd

4: ff } y ‘sy foo

f / / y, Vi YS ~

i é: S i \

~/ / 74 if \

of (ts BN

aor ——: ~<

} ij i fae / f od 4 eee

[| a 2 a : ES

| Ye f= | j | —Z »

pees 1 Net a

| | Vat S/N

} | | it ee rei | i ~ = a\ of

| = | —

a) is | eos

{ fx ~

| eT, { a)

eet = { -

| \ - act S =e

ley | f f BO

| e = : -

4 ——

\ \ KA ;

_ NA ; i

Fia. 44.

Fusarvum javanicum WKoord.; conidia from (a) mycelium of 19 days old culfure on plain agar, (b)

pionnotes of 4 weeks old culture on hard potato agar, (c) pionnotes of 5 weeks old culture on oat agar.

(d) sporodochia of 2 weeks old culture on bean pod, (e) chalmydospores from 4 weeks old culture on oat

agar, and (f) from culture 8 weeks old on Melilotus stem.

Potato plug: Aerial mycelium very sparse, white, cottony ; sometimes a few white

tufts of mycelium appeared near the top of the plug. The rest of the cylinder became covered

with conidial masses; these consisted of very numerous, minute, crowded sporodochia,

which coalesced to form a dense pionnotes. Conidial masses at first cartridge buff, becoming

pinkish buff to sage green, and in places dark bluish glaucous to Russian green.

Melilotus stem : Mycelium sparse, white to cartridge buff. Conidial masses developed

freely ; they were chamois to cinnamon buff. Numerous minute, black sclerotia formed

under the mycelium.

445

Bean pod : Mycelium white, felt-like, wrinkled. Conidial masses formed a thick crust,

which was at first chamois, and later chamois to cinnamon buff.

Rice: Aerial mycelium tilleul buff. Growth on grains vinaceous buff to avellaneous.

Grains were army brown to Natal brown.

Measurements of Conidia.

Hard potato agar, culture 2 weeks old, conidia from pionnotes.

GINS scconocacecs A. [OP GWM cocedcc 67-5-80 x 5

DENG cscosdocsoce 445) A Soe sena Bie 57-5-75 x 4:5-5-5.

A-septates....52....- 8 Fc = eae A7T-5-67-°5 x 4-5-5.

SeSePuabe wer wc. loy. 12 Sel oy eee: 25-50 x 3-7-4:5.

ASO UAUE socccocsccse ONO ce eh Gey cckerin 20-27-5 x 3-7-5.

IESepuaivee reer re: 17 gp oe at were 12-5-17:5 x 3-5-5.

O=septatels. eens Ns Deane eaemage, ee 2Daeet 10-5-12-5 x 3-5-4.

Oat agar, culture 5 weeks old, conidia from pionnotes.

DAM ceccccocess mS) JXSE CMW conoce 60-75 x 4-7-5

AESep bale aaaas sass see OD 5 sanecoce WaHaS) <¢ Soto.

GHANA oo caoocsunne 3 Sa a Re ae 45-62°5 x 3:-7-4:-7

2=SCP LALO eae ei else Leb iewla,,

USNC ccosccasese 4-5 =

OASA MND ocscacoscacs 4 =

Potato plug, culture 4 weeks old, conidia from pionnotes.

DANO scosoovsoccs NAc) jer COME ctcccos 47-5-62-5 x 4-4-5.

AND coccasceoand 56 Ah tt GOP 45-52°5 x 4-4-5.

ACM scncoccsocga ly) a al asa ene B 30-52°5 x 3:7-4-5

2-SCPtatOaaeses > 5. 1-5 ot a eeateerehoeteae 27:5-45 x 44:5.

ILAGMAUO occosccesooc GOMES est tart). Santen & 17-5-22-5 x 3-4.

OAS NENUO sobscosoese AD gue Emre Aisles 7-5-15 x 3-7-4.

Melilotus stem, culture 4 weeks old, conidia from pionnotes.

DRGMEUWE .saoccccoccs A) DOE CBMlancogsoc 55-07-5 x 4-5.

A- SOV baUC a.) paises: 30 55 rae eae 45-52-5 x 4-5.

ECINNOsooceccoagnc 13-5 NAnla ese: 42-5-47-5 x 3-T-4-5

2-septate 2-5 Ae, Betcha tas 20-32°5 x 3-3-75.

l1-septate 19 Fi) Colo to cio ee 12)-5-21 x 3-4-5:

O-Septauenerer reer 32:5 yi) Uphebiodae 6:25-12:5 x 3-4.

Standard synthetic agar plus starch, culture 4 weeks old, conidia from pionnotes.

BRC MIO docesaceuoee U [OOP CBMWocdcaeccec 42.-5-67-5 x O.

A-Se Dudte ee eee rt: ADS sere aay, thee eoh 37:5-10 x 3-7-5

_ BOM ccoodscccase 2) ain ty Cae ate Beene 35-65 x 3-7-5.

ee he 1 Sore Sol aeenteroe tanks 33°75 x 4-4.

WAGONS ccoocasencvs 1]

QI NEUS coocsaoce soe 15

Chlamydospores common in meena and conidia. In oat agar plates 4 weeks old>

they were commonly in simple or irregular chains of 2 to 9 elements; these were mostly

terminal. Intercalary chlamydospores were often solitary. Single spores 7-5-12-5 wu

diameter. They were sometimes in loose groups, but were never seen in closely united packets

as in F. solani. In hard potato agar plates, 4 weeks old, chlamydospores were forming in

a large proportion of the conidia. These were terminal (often in the basal cell or cells)

or intercalary ; they were single, in pairs, or rarely in chains of 3 to 4 elements; thick

walled, rough when mature, 5-6-11-25 diam.

446

Fusarium javanicum Koord. var. radicicola Wr.

Wollenweber, Fusarium-Monographie, 286, 1931; us. aut. del. 423, 632, 1023, 1024. Wollen-

weber and Reinking, Die Fusarien, 129-130, 1935.

Syn. Fusarium radicicola Wr.

Fie. 45.

Fusarium javanicum Koord. v. radiciola Wr.; conidia from pionnotes of culture on (a) potato plug,

(6) bean pod; both cultures 4 weeks old.

Microconidia numerous, |-celled or septate, scattered in the mycelium, or cohering

in false heads. Macroconidia in sporodochia, or less frequently in pionnotes, brownish

white in mass, becoming darker with age, or taking up colour from the olive green or coffee

brown stroma. Macroconidia 3-septate, less frequently 4-, and exceptionally 5-septate,

elongated, slightly curved, somewhat more definitely curved, and constricted at the apex,

sub-pedicellate at the base.

OFseptateria. - set ast ON ale 4.) Open MOStyaO\.3 1 OMara= a:

leseptaten a. aac ass5. .Oreill xo—Dbe Jin ae osthy all Sexe oie tad:

SHIM Icocaspaacnan A080) XX Bods}, 005 Mostly 30-42 x 3-8-4:8.

DS|DNUENO oococe dooce 38-59 x 4-5-3....... Mostly 43-46 x 4-5:1-

Chlamydospores common, terminal and intercalary, 1—2-celled, in chains or clusters ;

1-celled 9-10 x 8-5-9; 2-celled 16-22 x 5-12; smooth or rough.

Hab. Pelargonium sp., from discoloured rhizome, Pretoria (Wager).

Solanum tuberosum L., from tubers showing a black form of dry rot, Umhlanga

Beach, near Mt. Edgecombe, Natal, Jan. 1931 (van der Plank).

This fungus is known in the United States as a cause of potato rot; it is also found in

other root crops, and in ornamental plants.

Growth on Standard Media.

Oat agar. Aerial mycelium short, sparse, white, tomentose. Growth on substratum

colourless. Conidial masses did not develope freely on this medium. Reinking and Wollen-

weber (39) record the development on oat agar of “ olive buff and pea green sporodochia

in large heaps, gradually forming a pionnotes.”

Hard potato agar : Mycelium not abundant, rather coarse, tomentose, short. Sporo-

dochia, when present, in groups, cream buff to olive buff and lichen green ; often forming

in concentric rings. In some old cultures, the agar was stained Natal brown.

Standard synthetic agar plus starch : Slant covered with short, white mycelium, which

was mealy-looking owing to the presence of numerous conidia. Growth on substratum

colourless to wood brown.

Potato agar plus 5 per cent. dextrose: Aerial mycelium scant, short, mealy-looking.

stowth on substratum buff pink to vandyke brown ; agar stained Japan rose.

447

Potato plug: Aerial mycelium rather coarse, short, tomentose or felt-like, white to

cream buff and buff pink, or, when older, tilleul buff to buff pink and vinaceous brown

brown in places after 12 weeks. Sporodochia in groups, at first vinaceous buff, then olive

buff and light terre verte.

Melilotus stem : Stems covered with a rather short, white, coarse, tomentose mycelium.

Sporodochia and pionnotes developed in longitudinal lines ; they were pinkish buff to dark

olive buff.

Bean pod: Pods covered with a rather sparse, coarse mycelium, which was white to

pale cinnamon pink. Pinkish buff sporodochia and pionnotes developed in patches.

Rice: Aerial mycelium short, white, mealy ; growth on substratum purplish vinaceous

to dark livid brown and wood brown.

Measurements of Conidia.

Melilotus stem, culture 4 weeks old, conidia from sporodochia.

H-septatemasonay) oss. 9 10) pel Cent... 05.14. 0 40-59 x 4-5-5-5.

A septabercnoacerieg sylOl | a 38 0 9 tte sechnes 30-48 x 3-7-5.

CHAINS ccecn eee Ad | <a nie aan ec nae 22-5-42 x 3-7-8.

PHSANEUO soateocanoas A: cites Pee As Gas Kats 2ae 20-27-5 x 3-7-4.

NESW concncascsco & eer ERR ec 15-20 x 3-2-4.

O2septatetea.<. 4... 39 ni) Ale sc ereereeone ars 5-12-5 x 3-3:75.

Potato plug, culture 4 weeks old, conidia from mycelium.

DAL NMOS coos cocoon dGob) jie Coie saououe 30-52°5 xX 4:5-5°-5

A-SODUMUC Merrie Mota BNO he Nec stirs 32°5-40 x 4:5-4:3

3-septate . SITES: Be ee a nes 20-32-5 x 4-5.

Z=septaters.ivudenck..) OPO Ee eh ee EY 16-18 x 3-7-5.

lesepbaterccsa. «4-5. Teg Be ea 12-5-17-5 x 3-9.

0-septate.. : GOMER: Feria tes hie OT: 3:75-10 x 3-2-5-5.

Bean pod, culture 2 weeks old, conidia from pionnotes.

Beseptate. <n... sae - Renee ee Gen est 40-57-5 x 5.

ANN coscoasbooas | OG) (OO? CBMs oho o000 39-52°5 X 5.

SHIA NCWSasconboaceao Hl 50 (le 5 tie re came 22-5-45 x 3-7-5.

MASI ONEWD 4 06 90.6 610:6.0 0% 16 a

lSseptatermacmonteri cn Le cs

O-septatemen nner aera 42-5 o

Non-septate conidia oval, pyriform or spherical; 1-septate usually comparatively

slender, but occasionally resembling the 1-celled conidia in form.

Fusarium solani (Mart.) App. et Wr.

Appel and Wollenweber, Arb. K. Biol. Anst. Land.- u. Forstw. 8: 65-78, 1910. | Wollenweber,

Fus. aut. del. 396-400, 404, 405, 418-421, 1029, 1031-1033, 1194. Wollenweber and Reinking, Die

Fusarien, 135, 1935.

Syn. Fusisporum solani Martius pro parte.

Fusisporum solani Mart. v. flavum Hart.

Fusisporum solani-tuberosi Desm.; Pionnotes solani-tuberosi (Desm.) Sacc.

Fusarium commutatum Sace.

Lachnidium acridiorum (Trab.) Giard.; Fusarium acridiorum (Trab.) Brougn. et Del.

Fusarium allii-sativi All. ; F. alluviale Wr. et Rkg.; F. Malli Taub.

F. solani (Mart.) v. cyanum Sherb.; F. solani (Mart.) f. 1 Wr.

F. solani (Mart.) v. medium Wr.; F. solani (Mart.) v. suffuscum Sherb.

F. viride (Lechm.) Wr.; Pionnotes viridis Lechm.

Conidia scattered, in false heads, in sporodochia or in pionnotes, in mass brownish

white to clay yellow, or tinged with blue, or flecked with green from the stroma. Stroma

leathery, or sclerotial, green to dark blue. Macroconidia almost cylindrical-fusiform,

448

slightly curved, rounded at both ends, or tapering and bluntly conical; base with a scarcely-

perceptible papilla, which is oblique to the longitudinal axis, seldom sub-pedicellate, 3

or 3-5-septate.

OBEN 50000050080 WIS Bee.

ISCAS oc occco05508 20 x 4:3.

S_SePoMbC eet cel 4a IS=B0) << BeO=focccnoc Mostly 28-42 x 4-1-6-2.

ASCIDUENIO oooco6aac 42 xX 5:6.

H-SepbALes.,. sic aus siege poe OO! ea alee ears Mostly 42-51 x 5-6:3.

Chlamydospores terminal and intercalary, brownish, single, spherical to pear-shaped ;

l-celled 8-5 x 8; 2-celled 9-16 = 6-10; seldom in chains and clusters, smooth, or

sometimes minutely verrucose, and rough when dry.

OQ £~f7 £i 4 -: ‘

(ff li ga a % | fo

4 [| r | es ee A ) H \\ fa

| | | L—| 4 ea al &

| ey Ed ey | ea |

rl ba elt el 1 Prati & ae |

Ya Vel eae he a tq | i | ey VA

A | eat - NaN 3 | ta bea \ Pal bt | \

SUNS BTU as 1. Wee &

\ ¥e f =) x f \= i \

WA YZ ' . SS ; \ i U (\ X \ s

d, a 7

» © EF

fk re | 1) |

CEES) HI | |

Ef j '

vAS ol fa

'- >

A (e- ¢ ? ee \ \ | '

\ < a wey Ne)

Fic. 46.

Fusarium solani (Mart. pr. p.) App. et Wr.; (a-d) strain from wheat, (e-f) from collar rot of orange

tree ; conidia from (a) sporodochia of 4 weeks old culture on Melilotus stem (band e) pionnotes of 3 weeks

old cultures on hard potato agar, (c and f) pionnotes of 5 weeks old cultures on oat agar, (d) chlamydospores

from 8 weeks old culture on hard potato agar.

Hab. Allium cepa L., from discoloured bulbs, and stems of dying plants, Nelspruit (Wager)

M.A. 28383. Also recorded by du Plessis (11) as causing a dry rot of onions in storage,

Stellenbosch, Cape.

Carica papaya L., from stem of plant affected by foot rot, Maritzburg, Natal (Wager) ;

Malelane, E. Transvaal (Wager) M.H. 28375 and 28374.

Citrus grandis Osbeck, from bark of grapefruit tree showing gummosis, Patentie, Cape

(van der Plank) M.H. 28401; Coegapoort, Gamtoos Valley, Cape (van der Plank) 1930

(from collar of 2-year old tree).

449

Citrus limonia Osbeck, from roots of lemon stocks on which orange or grapefruit had

been budded, roots showing “dry root rot” lesions; Louis Trichardt, N. Transvaal ;

Bathurst dist., Cape (van der Plank); Thorndale, Hankey, Cape (van der Plank) M.H.

28371; Acornhoek, E. Transvaal, M.H. 28396; Tzaneen, N. Transvaal (Wager); Elands-

hoek, E. Transvaal, M.H. 28372 ; Coegapoort, Gamtoos Valley, Cape, 1930 (van der Plank).

Also from soil in Citrus orchard Kosterfontein, Marico dist., Transvaal (Hsselen).

Citrus sinensis Osbeck, from bark above union, which was cracking and gumming

(roots waterlogged) Letaba Estates, N. Transvaal; from bark of tree affected with scaly

bark (psorosis) Mazoe Estates, 8. Rhodesia (Bates); from roots of old seedling orange,

Viliersdorp, Cape, M.H. 28360.

Dianthus caryophyllus L., from stem of plant affected with crown rot (ass. F. bul-

bigenum v. lycopersici) Pretoria.

Gilia rubra Heller, from stems of wilting plants, Acton Homes, Natal

Gladiolus sp., from corms showing a dry brown rot (ass. F. oxysporum vy. gladioli)

Princess Park, Pretoria.

Phaseolus sp., from stems of dying plant, Swaziland (Wager).

Solanum tuberosum L., from tubers showing dry rot and soft rot in a consignment

imported from Germany. Also isolated by du Plessis (13) from rotting tubers, Ceres,

George, Paarl and Stellenbosch, Cape.

Tropaeolum majus L., from stem of dying plant, Pretoria.

Zea mays L., from base of stem of plant showing foot rot (ass. F. moniliforme) Water-

berg, Transvaal (Sellschop).

Fusarium solani is a rotting organism and is seldom a primary cause of injury. It

occurrs on a wide range of plants, chiefly in the temperate zone.

Growth on Standard Media.

Oat agar: Aerial mycelium sparse, rather coarse, tomentose, white or dirty white.

Conidial masses developed freely; pionnotes and sporodochia at first cartridge buff, be-

coming pinkish buff, and in older cultures stained bluish grey green to dark russian green.

Sporodochia were produced in large groups.

Hard potato agar : Some rather coarse, scant, white, tomentose mycelium developed over

the face of the slant, sometimes becoming mealy owing to the formation of conidia, and some-

times showing concentric zoning. Pionnotes developed freely and groups of sporodochia

near the base of the slant; spore masses were at first cartridge buff, becoming pinkish

buff and dark bluish glaucous.

Standard synthetic agar plus starch : Aerial mycelium short, scant white. Pionnotes

developed freely, especially along the needle track; they were cartridge buff. A few,

small, blackish-brown sclerotia appeared near the base of the slant.

Potato agar plus 5 per cent. dextrose: Mycelium moderate in amount, at first white

to cartridge buff. Growth on substratum cream to sage green, becoming dark olive to

clove brown. In older cultures the ‘mycelium became tinged with the colour of the stroma.

A number of blackish-brown sclerotia developed at the base of the slant.

Potato plug: Plug covered with a moderate amount of rather coarse mycelium, wnich

was white to dirty white; there were a few patches and flecks of dark terre verte at the

base and back of the plug. Large groups of sporodochia were usually produced, and

coalesced to form a pionnotal mass; spore masses were at first cartridge buff to pinkish

buff, and later tinged bluish grey green to glaucous or deep lichen green. In some tubes

a few small, dark brown sclerotial masses developed.

Melilotus stem : Mycelium sparse, loose, coarse, white to dirty white. Pionnotes and

groups of sporodochia were pinkish buff to deep bluish glacious. A few dark brown sclerotial

masses were present.

450

Bean pod: Mycelium rather plentiful, or sparse, coarse, tomentose or becoming

wrinkled and felt-like, white to cartridge buff. Large sporodochia developed in groups ;

they were cartridge or pinkish buff to deep bluish grey green.

Rice: Growth white to light purple drab; grains white to naples yellow.

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from pionnotes.

OEGIONMccgcccnvoocs th je GMI yoo c006 38-50 x 5-5-5.

4-septate 25 oie o-as See anal 37:5-50 « 5-6:25

SECU 5 oon co 0008 (3°00 (ae emer Geen 30-42 x 5-5-5.

l-septate .. ere Me Olt ‘5

Bean pod, culture 4 weeks old, conidia from sporodochia.

5-septate....... 20006 2 [NEE COMBs ooodoecne 37-5-50 x 5-5-6

4-septaten.s.na.0500 7 40 Psoeh ete cee Se 32°5-45 x 5.

S-Septaterns...% ets) OL Bee Dnt oak ake ina Oe 27-5-50 x 5-5-5, mostly

30-40 x 5.

O-septate rn ereemeecer 4

Hard potato agar, culture 3 weeks. old, conidia from pionnotes.

ORY MUMGMBoooosccvosac Mol) [oer CMs > so0006 36-245 x 5.

A-septate ocoe JUG eke eae Rae ae 25-42:-5 x.5.

SH UEMIGobanboclib hee lb. | ay le oagegse 27-5-42-5 x 4-7-5.

1-septate.. Batgirl) 1 dl. ob) e

Oat agar, culture 2 weeks old, conidia from sporodochia.

H-SepbAtenanr toe rl: Rare .. eae aaserehiests AT Dex 5.

“Eno ecesooasee | O JX8 COME soocoooc 30-45 x 4-5

SNEWS cocossovocns OY sj eee 35-40 x 4-5.

DAHOMMOcccoscocoses De neta peas 225-30 x 3-7-4.

Teseptateine = sav. «eevee PRP oe Ode REC Eee eens 15-20 x 3-7-4.

OLS Eo eeonomoee ss O20) oy ° cbeoooen 1:0-15 X 2:5-3-T9.

Fusarium solani (Mart.) v. Martii (App. et Wr.) f. 1. Wr.

Wollenweber, Fusarium-Monographie, 290, 1931; us. aut. del. 415-417, 631. Wollenweber

and Reinking, Die Fusarien, 137, 1935.

Syn. Fusarium Martii App. et Wr. v. minus Sherb.

Fusarium Martii App. et Wr. v. viride Sherb.

? Fusarium pestis Sor.

Conidia more slender than those of Fusarium solani, with hardly any curvature in the

medial portion, more definitely curved or bent near the apex, base papillate or sometimes

pedicellate. Macroconidia 3-septate, more rarely 4-septate, exceptionally 5-septate ; micro-

conidia 0-2-septate.

OREMEWD 05059008700 8 x 3:8

UEISOUENWO oobd000000¢ 17 x 4:5.

3-septate...:...:.... 22-52 x 3:5-6....... Mostly 30-44 x 4:5-5-5

4-septate............, 46 X 5-2.

D-septate.....,.:..-.. (60-02. 4:2-6)...... Mostly 43-07 x 4-3—b<5.

Conidia in mass brownish-white, or ivory yellow to light brown; often absorbing

colour from the greenish blue or brown stroma, and then with a mixture of wood brown,

grey andicoites brown to blackish tones. Chlamydosporesin conidia 8-5 x 6; in mycelium

l-celled, 10 x 8, 2-celled 8-22 x 6-12.

Hab. A ua cepa L., from rotting bulb, Pretoria, 1932 (Bottomley),

Carica papaya rare from stems of plants showing foot rot, Maritzburg, Natal (Wager) ;

Buffelspoort, Marikana, W. Transvaal (Wager) M.H. 28373 ; Nelspruit, EK. Transvaal

(Wager).

451

Centaurea moschata L., from stems of wilted plants, Pretoria (Wager).

Citrus grandis Osbeck, from bark of tree showing gummosis, Patentie, Cape (van der

Plank).

Citrus limoma Osbeck, from roots of lemon stocks on’ which orange grapefruit or

naartje had been buddea, roots showing “dry root rot” lesions; Grahamstown, Cape

(van der Plank) M.H. 28376; Godwan River, E. Transvaal, Oct. 1930 (Marloth) ; Swane-

poelsrust, near Nylstroom, M.H. 28366; Plaston, EK. Transvaal (Ksselen); White River,

EK. Transvaal; Buffelspoort, Marikana (Turner); Elizabethville, Belgian Congo, M.H.

28367; Avonmore, Rustenburg dist., and Kosterfontein, Marico dist., W. Transvaal

(Esselen) ; Addo, Cape, M.H. 28368; Magaliesburg, Transvaal, Noy. 1929 (van der Plank) ;

Citrusdal, Cape (Turner) M.H. 28369; Airlie, EK. Transvaal (van der Plank) ; White River

(Esselen) M.-H. 28370; Kakamas, Cape (Gutsche): Hankey, Cape (van der Plank) ;

Kruis River, Cape, May 1930 (van der Plank) M.H. 28397; Louisvale, Cape (Turner) _

M.H. 28399 ; Amanzi, Cape, May 1930 (van der Plank) M.H. 28399; Ofcalaco, N. Trans-

vaal July 1930 (van der Plank); Gamtoos Valley, Cape, Aug. 1930 (van der Plank) ;

Elandshoek, E. Transvaal, July 1930.

Also from roots not visibly affected by root rot, and from

the soil, Boskopjes, Rustenburg dist.

Fie. 47.

Fusarium solani (Mart. pr. p.) App. et. Wr. v. Marti (App. et. Wr.) f.1 Wr. ; conidia from (a) sporo-

dochia of 5 weeks old culture on Melilotus stem, (b) sporodochia of 10 weeks old culture on potato plug,

(e) pionnotes of 4 weeks old culture on hard potato agar, (f) chlamydospores from 4 weeks old culture on

hard potato agar.

Citrus sinensis Osbeck, from bark cracking above union (roots waterlogged) Letaba

Estates, N. Transvaal; from bark and wood, Boschrand, EH. Transvaal and Hlandsdrift,

Rustenburg dist. (Turner); from crown, bark and roots of seedling orange, near Klaver,

452

van RKhynsdorp dist., Cape (Putterill) ; from roots of old seedling orange with dry root

rot lesions, Villersdorp, Cape, M.H. 28358 and 28361; from bark of navel orange showing

gummosis, Frantzina’s Rust, E. Transvaal, August 1930 (van der Plank).

On fruit, after 12 to 18 weeks in storage, mostly from stem

end occasionally from navel end and lateral lesions; on navel oranges from Sunday’s

River, Cape, and from White River, Rustenburg and Letaba, Transvaal; also on Valencia

oranges from Sunday’s River, White River, Zebediela and Rustenburg (23 isolations).

Delphinium sp., from crown of wilting plant, Pretoria (Bottomley) and Qumbu, E.

Griqualand.

Lathyrus odoratus 1... from stems of yellowing seedlings (Ass. Pythium sp.) Brooklyn,

Pretoria.

Mathiola incana R. Br., from stems of dying plants, Uitenhage, Sept. 1932 (Wilson) ;

Durban (McClean) M.H. 28365.

Medicago sativa L., from stem of plant with rotting crown, Pietersburg, N. Transvaal

(Wager).

Penstemon sp., from stems of wilting plants (ass. Rhizoctonia) Pretoria.

Phaseolus acutifolius Gray, from stems of Tepary bean plant, which was also infected

with Colletotrichum sp., Immerpan, Springbok Flats, Transvaal (Sellschop).

Phaseolus sp., from stems of wilting plants in variety trials, Premier Cotton Estates,

Mvamba.

Pisum satiwwum L., from stems of wilting plants (probably secondary to bacterial

infection).

Rheum rhaponticum L., from crown of dying plant, Balfour, Transvaal, March 1930

(Wager) M.H. 28404.

Solanum tuberosum L., from tubers showing dry, sunken, discoloured patches, Belgian

Congo; from stems of wilting plants (ass. F. oxysporum f. 1) Moorddrift, Transvaal; from

stems of etiolated plants in greenhouse, Pretoria.

Nomadacris septemfasciata, on eggs of red locust hatching in sterilised soil, Pretoria‘

1932 (Brookes) M.H. 28364.

This form is widely distributed in humus, and on decaying parts of plants, in warm

and temperate regions. It is recorded by Reinking and Wollenweber (39 p. 220) on rotted

roots of Citrus aurantifolia Sw. in Central America.

Growth on Standard Media.

Oat agar ; Aerial mycelium sparse, rather coarse, white to cartridge buff, tomentose ;

or aerial mycelium may be lacking. Pionnotes and groups of sporodochia developed freely,

and were at first cartridge buff, becoming pinkish, and, after 8 weeks, sage green to deep

grayish blue grecn.

Hard potato agar : Aerial mycelium not abundant, coarse, short, tomentose. Pionnotes

developed along the needle track, and groups of sporodochia at the base of the slant ; conidial

masses were pinkish buff, becoming tinged deep bluish gray green to deep olive buff; in

some tubes they were deep glaucous green.

Standard synthetic agar plus starch: No aerial mycelium. Pionnotes cartridge buff,

or tinged grayish olive.

Potato agar plus 5 per cent. dextrose: Aerial mycelium short, coarse, tomentose, white

to cartridge buff. Growth on substratum onion skin pink to vinaceous tawny or clay colour.

Sporodochia pinkish buff, becoming deep olive buff or deep grayish blue green.

Potato plug : Mycelium covering the plug, short, rather coarse, loose, white to cartridge

buff, sometimes with patches of dusky dull bluish green and deep delft blue. The plugs

early became covered with a mass of closely crowded, pinkish buff sporodochia, which

coalesced to form a dense pionnotes. The conidial masses became tinged greenish glaucous

blue to dark bluish glaucous, and later Russian green to dusky dull green.

Melilotus stems: Mycelium rather scant, white, tomentose. Sporodochia pinkish

buff, becoming tinged sage green.

453

Bean pod: Mycelium sparse, white, tomentose. Sporodochia and pionnotes pinkish

buff, becoming pea green.

Rice: Mycelium white to cartridge buff, mealy, becoming tilleul buff to pinkish buff.

Rice grains naples yellow to ochreaeous buff and cinnamon brown. Small sporodochia

sometimes developed.

Measurements of Conidia.

Oat agar, culture 4 weeks old, conidia from pionnotes.

HES UALCH er ODOmper Center iit. 5D X 5:5.

APSEPtate. ons as ce. 1-5 sa chs Sea aoaks 51-55 x 5-5-5.

S-septate..........:. 72 pat Cae A 32°5-52°5 x 4-4-5, mostly

37-5-45 x 5.

Z-Sepbateaseae sage LS Site hae seh sy 32:5-35 x 5

IESe pia besarte eri 8 50

0-septate... 15

Hard potato agar, “culture 3 weeks old, “conidia from plonnotes.

SHEOUAUIC. son oooobe do SiO ypericentirrc rice 30-47-°5 x 4-5.

2-septate............ 5 Ae Bare eee ett en 25-30 « 4-5.

l-septate............ 6 aN eee, hee 15-25 x 3-2-5.

O-Septatesnee eer (Ub Palen een oestrone t's 5-12-5 x 2-5-5.

Potato plug, culture 4 weeks old, conidia from sporodochia

4-septate............ 2D per Gente a. ae): 40-45 x 4-7-5

SHENAE. coco on0H0e 64 Bee Seite ct eS toes 30-45 x 3-7-6.

PINES oocaboooceae WW) SL uetaie 27-5-32°5 x 3-75.

NASW: ndocesccenc 2 ae

O-Septaleus eine ere 28

Bean pod, culture 2 weeks old, conidia from sporodochia.

DS DUALC MEER ace: I Depericenbir cnr 40-47-5 x 5-6.

4-septate......-..--- 50 Ge i auras ct 35-45 x 4:5-5:5

3-Sepuate....4....-.. 30 Boe vince sttas 35-40 x 5-5-5.

PSION, cob coo boone 0-5

lESeptatea-memesereee I -

ORENAD.co0 00060000 il .

Fusarium coeruleum (Lib.) Sacc.

Saccardo, Syll. Fung. 4: 705, 1886. Wollenweber, Fus. aut. del. 407-410. Wollenweber and

Reinking, Die Fusarien, 134, 1935.

Syn. Selenosporiwm coerulewm Libert in herb.

Fusarium violaceum Fuckel ; F.aeruginosum Del.

Fia. 48.

Fusarium coeruleum (Lib.) Sace.; conidia from (a) myclium of 8 weeks old culture on Melilotus stem,

and (6) pionnotes of 6 weeks old culture on oat agar.

454

Conidia in sporodochia, in extended pionnotal layers, or scattered in the mycelium-

Macroconidia almost straight or sub-faleate, with obliquely conical, ellipsoid or rounded

apex, and base obtusely oval to mammillate, or with a papilla oblique to the longitudinal

axis. Conidia in mass isabellinous-ochraceous to brownish white, sometimes taking a blue-

violet to blue-black or wood green tinge from the stroma. Chlamydospores terminal or

intercalary, 1-celled, spherical (9 4) to pear-shaped (9 x 8) or 2-celled (14 x 9). Stroma

effuse or verrucose, sclerotial, light or violet to blue-black. Conidia mostly 3-septate, less.

frequently 4-5-septate, exceptionally O-2- or 6—T7-septate.

O-septate:...is5.s. 451 16 « 4:7.

leseptaterea ss cea 2 aa:

3-septate:........... 21-47 x 3-5-6........ Mostly 32-40 x 4-5-5-5.

H-septate... va-.2:46. - 42 x 5-2.

Hab. Solanum tuberosum L., from tubers showing storage rot in a consignment from Ham-

burg, Germany, Dec. 1929 (Wager) ; also isolated by du Plessis (13) from rotting tubers,

George, Paarl, Stellenbosch and Ceres, in the winter rainfall area.

No detailed notes were made of the cultural characters and conidial measurements

of the strain studied.

ANNOTATED HOST INDEX.

A.—Fusaria on Flowering Plants.

Allium cepa L.

Foot rot, root rot and bulb rot.

Fusarium oxysporum f. 7.

F. vasinfectum v. zonatum f. 2.

F. moniliforme.

F. bulbigenum.

Decaying stem tissues.

F. scrpi and F. scirpi v. filiferum.

Bulb rot in storage.

F., moniliforme.

F. solani.

F. solam v. Martu f. 1.

F. oxysporum f. 7 has been found in the Transvaal and in the winter rainfall area (12).

The other species, with the exception of F. solani, which has been isolated from rotting

onions in the Cape Province (11) have, up to the present, only been recorded from the

Transvaal; they are, however, cosmopolitan species (25, 61) and are probably more widely

distributed than this would indicate.

Ananas comosus Merr.

Fruit rot.

Fusarium moniliforme.

This fungus was isolated from water soaked patches in pineapples; the lesions were

more extensive and lighter brown than those caused by Penicillium sp. Fusarium moniliforme

and its variety subglutinans have been found in decaying tissues of pineapples in Central

America (59, 61).

Andropogon sorghum, see Sorghum vulgare.

455

Antirrhinum majus L.

Decaying stem tissues.

Fusarium oxysporum v. aurantiacum.

F. scirpr.

F. scirpi v. compactum.

According to Mes (27), the wilt of snapdragons in South Africa is caused by Phytophthora

cactorum ; Fusarium spp. isolated from badly decayed tissues were not found to be a cause

of wilt.

Apple, see Pyrus.

Arachis hypogaea L.

Pods and seeds.

Fusarvum angustum.

F.. oxysporum v. aurantiacum.

F. scirpi.

These fungi were isolated from pods and seeds attacked in the soil; the pods from

which F. oxysporum v. aurantiacum was isolated showed a pink discolouration of the shell.

Aster, see Callistephus.

Avocado, see Persea.

Bean, see Phaseolus.

Brachiaria, see Gramineae.

Bracken, see Pteridium.

Brassica oleracea L.

Decaying stems.

F. oxysporum vy. aurantiacum.

F. moniliforme.

F. monliforme v.. subglutinans.

These fungi were isolated from stems of plants which had been attacked by Rhizoctonia,

Pythium and aphides ; no true cases of *‘ cabbage yellows” have been observed.

Bromus, see Gramineae.

Broom corn, see Sorghum.

Cabbage, see Brassica.

Callistephus chinensis Nees.

Wilted plants.

Fusarium conglutinans vy. callistephi.

Decaying stems.

F. scirpi.

Aster wilt due to F. conglutinans v. callistephi is extremely prevalent in South Africa,

and was probably introduced with seed imported from overseas. Only wilt-resistant varieties

can be grown profitably (53, 55).

Campanula medium L.

Decaying stems.

| Fusarium seirpi v. compactum.

Canterbury bell, see Campanula.

456

Carica papaya L.

Foot rot.

Fusarium bulbigenum v. lycopersict.

F. solani.

F. solam v. Martu f. 1.

Fruit rot.

F. lateritium.

F. scirpn.

F. stilboides.

Seedlings damping off.

F. oxysporum.

According to Wager (52), foot rot is caused by Pythiwm spp., and Fusarium or Rhizoc-

tonia occur as a secondary cause of rot in decaying tissues; inoculations with Fusariwm

spp. did not give rise to foot rot. In Trinidad (2) a Fusarium sp. was found to cause a foot

rot under moist conditions. Ff. diversisporum and F. dimerum v. pusillum act as wound

parasites of papaw fruit in the Philippines (61).

Carnation, see Dianthus.

Centaurea cyanus L.

Discoloured stem tissues.

Fusarium vasinfectum v. lutulatum.

F. solam v. Martu f. 1.

Citrullus vulgaris Schrad.

Wilting plants.

Fusarium bulbigenum v. niveum.

Extensive wilting is reported in commercial plantings of watermelons. Varieties

selected in America for wilt resistance (23) have been tested; Iowa Belle Round, Iowa

Belle Long, Pride of Muscatine and Iowa King all showed considerable resistance under

South African conditions ; the highest degree of resistance was shown by lowa Belle Round.

Further variety tests are in progress.

Citrus spp. (C. grandis Osb., C. limonia Osb., C. sinensis Osb.).

Dry root rot.

Fusarvum angustum.

F. solam.

F. solam v. Mart f. 1.

F. vasinfectum f. 2.

F. scirpi.

Bark on branches and twigs.

F, avenaceum f. 1.

F. lateritium.

F. lateritium v. longum.

F. scurpr.

F. scirpi v. compactum.

F. semitectum v. majus.

F. solam.

. solam v. Martu f. 1.

Buds in nursery stock.

F, lateritium.

457

Decaying fruit.

. angustum.

. bulbigenum v. lycopersicr.

. decemcellulare.

. equisety.

. lateritium.

. lateritium v. longum.

. moniliforme.

. moniliforme v. subglutinans.

oxysporum.

sambucinum.

sambucinum f. 2.

scr pv.

scirpi Vv. compactum.

. semitectum Vv. majus.

. solani v. Martu f. 1.

. stilboides.

. vasinfectum f. 2.

Pay Bay Fay Bay Bay ay By By Bay Bg Bg Fy By by

Fusarium solani and F. solani v. Martu f. 1 were almost always found associated with

dry root rot, frequently in conjunction with F. angustuwm or F. vasinfectum f. 2, but are

also found on apparently sound roots and in soil in citrus orchards. None of the Fusarium

spp. isolated from decaying roots was found to be capable of causing dry root rot. Of the

species isolated from decaying fruit, F. lateritiwm, F. moniliforme plus v. subglutinans,

F. oxysporum, F. scirpi plus v. compactum, F. solani v. Martu f. 1 and F. vasinfectum

/. 2, on inoculation into oranges, produced extensive brown rots readily, if somewhat

slowly. F. lateritium, F. oxysporum and F. solani v. Marti f. 1 were the most active rot-

producing organisms (Plate I, Ila). F. angustum, F. bulbigenum v. lycopersicr, F. lateritium

v. longum, F. sambucinum plus f. 2, F. semitectum v. majus and F. stilboides rarely produced

more than small, dry lesions around the point of inoculation. Only negative results were

obtained by inoculating F. decemcellulare and F. equiseti into oranges.

Coffea arabica L.

Stem tissues of unthrifty plants.

Fusarium oxysporum.

Berries.

F. lateritium v. longum.

F. stilboides.

It is interesting to note that F. lateritiwm v. longum has been found to cause a bark

disease of coffee in Kast Africa (44, 61). It is not known whether this organism occurs on

bark in the Northern Transvaal; coffee is not now grown in that area on a commercial

scale and few observations have been made.

Coffee, see Coffea.

Coral plant, see Pentstemon.

Cornflower, see Centaurea cyanus.

Cotton, see Gossypium.

Crotalaria juncea L.

Stems of wilted plants.

Fusarium sp. (elegans section).

F. scirpi.

458

A Fusarium sp. of the elegans section (probably belonging to the I’. vasinfectwm series)

was isolated from stems of wilting plants of Sunn hemp and has been described. It occurs

in Trinidad and India, and on C. striata in Uganda (4, 28, 46, 61).

Cucumber, see Cucumis.

Cucumis sativus L.

Rotting fruits

Fusarium scirpr.

F. scirpi v. compactum.

Stems of wilting plants.

F. equiseti.

F. scirpi.

Several species of Fusarium are recorded as causing rot of cucumber in temperate

climates ; they are wound parasites, namely F’. solani, F. orthoceras, F. reticulatum and

F. culmorum (61). The above-mentioned fungi were found in cucumbers affected with

soft rot and leaking, in the sub-tropical conditions of the Hastern Transvaal.

Cucurbita pepo L, and C. maxima Duch.

Stems of wilting plants.

Fusarvum javanioum.

Decaying stem tissues.

EF. solani v. Martw f. 1.

Fusarium javanicum causes a foot rot of cucurbits, and is often responsible for serious

losses in commercial plantings (9). It has been found occurring on pumpkin, marrow and

Hubbard squash in the field. Inoculation experiments resulted in 70 per cent. to 100 per

cent. infection of marrow, pumpkin, watermelon, spanspek (Cucumis melo) and cucumber

plants. In the case of watermelons, a high percentage of plants of varieties resistant to the

vascular wilt (7. bulbigenum v. nivewm) succumbed to the attacks of this organism. The

“Sugar Through” gourd, and a plant known locally as the “ Maraka ” (Cucurbita pepo

var. verrucosa) proved to be resistant.

Cupressus lusitanica Mill.

Dying seedlings.

Fusarvum oxysporum v. aurantiacum.

This fungus is known in Hurope and North America as a cause of damping off in coni-

ferous seedlings (61).

Cynodon, see Gramineae.

Cypress, see Cupressus.

Dahlia pinnata Cav.

Dying seedlings.

Fusarium sp. (elegans section).

This fungus was associated with Rhizoctoma sp. and Pythium sp.

Darnel, see Gramineae.

Datura stramonium L.

Stems of wilting plants.

Fusarium sp. (elegans section).

459

Delphinium Ajacis L.

Decaying stem tissues.

Fusarium solani v. Martia f. 1.

Dianthus caryophyllus L.

Stems of wilting plants.

Fusarium dianthi.

Foot rot and crown rot.

F. bulbigenum v. lycopersici..

FF. scurpi.

F. scirpi v. acuminatum.

F. semitectum v. majus.

F. solani.

Severe losses from carnation wilt are recorded from Natal and the Northern Transvaal

(47). The organism found in wilting plants from these areas agrees morphologically with

F.. dianthi, and has been shown to cause wilt in carnation seedlings artificially inoculated.

Fungi found in tissues of plants affected with crown rot usually attack carnations growing

under unsuitable climatic or cultural conditions.

Digitaria, see Gramineae.

Dimorphotheca aurantiaca D.C.

Plants affected by foot rot.

Fusarium sp. undet.

Drabok, see Gramineae.

Euphorbia crassipes Marloth.

Rotting stems.

Fusarium avenaceum f. 1.

F. lateritium.

F.. monaliforme.

F. scirpr.

Fusarium lateritium was also isolated from the fleshy stem of Hwphorbia obesa.

Fragaria sp.

Foot rot.

Fusarium sp. undet.

In North America and in England, root rot of strawberries is caused by F.

orthoceras (61) and other species. The African disease needs further study.

Freesia refracta Klatt. _

Corms showing dry rot.

Fusarium bulbigenum.

The internal tissues of the corms were light brown, and there was a white, powdery

deposit on the exterior of the corms when dry ; a similar rot occurs in America (45).

Geranium, see Pelargonium.

Gilia rubra Heller.

Foot rot.

Fusarium solani.

Fusarium sp. undet. (elegans section).

A Pythium sp. was present in all affected plants, and it is likely that the Fusarium

spp. were a secondary form of decay.

460

Gladiolus spp.

Rotting corms.

Fusarium oxysporum v. gladioli.

F.. bulbigenum.

F. solani.

Fusarium oxysporum v. gladioli was isolated from corms and leaf bases of cultivated

varieties. The first sign of disease was the browning of the younger leaves ; affected plants

failed to flower. The corms were firm, but showed a brown discolouration, especially near

the base. The organism agrees morphologically with F. oxysporum v. gladioli (28, 61) but

its identity needs confirmation by inoculation experiments. Fusariwm solani was present

in the same corms and was probably a secondary cause of decay. I’. bulbigenum was isolated

from corms of an indigenous species growing in the veld in the northern Transvaal.

Goose grass, see Gramineae.

Gossypium sp.

Foot rot.

Fusarvum angustum.

F. momliforme.

Fusarium moniliforme is recorded as a cause of foot rot of cotton in the United States

(61). The presence of F'. vasinfectum in wilted cotton plants in South Africa is unproven,

although records exist of the occurrence of a Fusariwm in the vascular bundles of wilted

plants. No opportunity has occurred, during the present investigation, of identifying this

organism. Boll rots associated with Fusarium spp. have also been observed, but no suitable

material has been obtained for investigation during the past few years.

Gramineae.

Foot rot of Hlewsine indica Gaertn. (goose grass).

Fusarvum avenaceum.

F.. moniliforme.

Foot rot of Loliwm temulentum L. (darnel, drabok).

F.. culmorum.

Ovaries of grasses infected with smut or ergot.

F. avenaceum f. 1.

F. heterosporum v. congoense.

Ovaries of Brachiaria pubsfolia.

F. chlamydosporum.

Fusarium heterosporum v. congoense occurs very commonly, forming a pink incrustation

on the ovaries of various grasses, especially when they are infected with smut or ergot ;

the fungus has been found on the ovaries of Brachiaria, Bromus, Cynodon, Digitaria,

Hyparrhenia, Panicum, Pennisetum, Setaria and Sorghum. F. avenaceum f. 1 occurs

frequently on ovaries of Paspalum spp. which are infected with Claviceps paspali.

Grapefruit, see Crtrus.

Grape vine, see Vitis.

Grasses, see Gramineae.

Hibiscus sabdariffa L.

Stems of wilting plants.

Fusarvum vasinfectum.

461

Fusarium vasinfectum is reported as a cause of wilt of Hibiscus cannabinus in Tan-

ganyika Territory (22) and the same fungus probably causes a wilt of okra, Hibiscus eseu-

lentus L. (61).

Hubbard squash, see Cucurbita.

Hyparrhema, see Gramineae.

Indian sorrel, see Hibiscus.

Ipomoea batatas Lam.

Surface rot of tubers.

Fusarium oxysporum.

This fungus was isolated from small, dry, discoloured, somewhat sunken patches on

the tubers (18).

i pomopsis, see Gilia.

Kaffir corn, see Sorghum.

Kentia sp.

Stem of dying palm.

Fusarium sev pr.

Kniphofia sp.

Capsules.

Fusarium moniliforme v. subglutinans.

Lathyrus odoratus L.

Foot rot of seedlings and mature plants.

Fusarium vasinfectum v. lutulatum.

Fusarium oxysporum.

F. scirpi.

F. scirpi v. compactum.

F. solani v. Martii f. 1.

Fusarium vasinfectum v. lutulatum was always found in the yellowing and drying

stems of affected plants. The other organisms were apparently secondary causes of decay.

Larkspur, see Delphinium.

Lemon, see Citrus.

Limonium sp.

Foot rot.

Fusarium scir pi.

F. scirpi v. compactum.

Lucerne, see Medicago.

Lycopersicum esculentum Mill.

Stems of wilting plants.

Fusarium bulbigenum v. lycopersict.

462

Decaying stem tissues.

Fusarvum angustum.

F. equisete.

F. sambucinum.

F. scurpv.

F. vasinfectum v. zonatum f. 1.

Discoloured vascular tissue in fruit.

F. bulbigenum v. lycopersici.

Rotting fruit.

F. scwrpi.

Seed.

Fusarium bulbigenum v. lycopersici.

F, equiseti.

FE. momliforme.

F. redolens f. 1,

The wilt caused by Fusarium bulbigenum v. lycopersici is very prevalent in the tomato

growing areas of the eastern Transvaal. A number of varieties selected in America for wilt

resistance have been tested and the varieties Stone and Marvel were found the most suitable

for Transvaal conditions ; further selections are being made from these varieties (49, 50,

51, 54). The organisms found in decaying stem tissues were usually associated with F.

bulbigenum v. lycopersicr, with Rhizoctonia sp., Pythium sp., or with Bacterium solanacearum.

The rotting organisms entered the fruit through wounds, “ blossom end rot,” or through

cracks at the stem end. F. scirpi frequently causes a browning of the core of apparently

sound fruit. Several-species were isolated from seed offered for sale by local seedsmen ;

most of this seed is imported from America, and several of the fungi found on the seed are

causes of rot in tomato fruit.

Maize, see Zea.

Marrow, see Cucurbita.

Matthiola incana RK. Br.

Stems of yellowing and wilting plants.

Fusarium sevrpr.

F.. scirpt v. compactum.

F. solani v. Marti f. 1.

F. vasinfectum f. 2.

Medicago sativa L.

Decaying stem tissues.

Fusarium angustum.

F. solani v. Martii f. 1.

These fungi were isolated from decaying tissues of plants affected by crown rot; they

were usually plants growing in heavy soil and indiscreetly irrigated. Neocosmospora vasin-

fecta was also obtained from the same source.

Mesembrianthemum sp.

Rotting stems of succulent species.

Fusarium avenaceum f. 1.

F. equaseti v. bullatum.

F. vasinfectum f. 2.

463

Musa Sapientum L.

Fruit.

Fusarium moniliforme.

F. semitectum v. majus.

F. scirpi.

Fusarium monilforme was isolated from the internal tissues of fruit shewing “ finger tip

rot.” Hansford (17a) records finger tip rot of several varieties of banana in Uganda caused

by this organism. The other two species mentioned were growing on the surface of the

rotting fruit.

Nasturtium, see Tropaeolum.

Nicotiana tabacum L.

Dying seedlings.

Fusarium moniliforme.

Stems of wilting plants.

F. oxysporum v. nicotianae.

F. bulbigenum.

In several publications, tobacco wilt occurring in the western Transvaal has been

attributed to Fusarium oxysporum v. nicotianae, on account of the presence of a Fusarium

mycelium in the vascular system and the similarity of the symptoms to those of the American

tobacco wilt (19, 20).. During the season 1925-1926. wilt was pronounced in the western

Transvaal, probably as a result of spells of hot, dry weather. When wilt is severe, all leaves

droop, turn yellow and die within a few days. Often only one lateral root is affected and

leaves on that side of the plant alone are affected, the others remaining normal. Wilting is

accompanied by a darkening of the wood from the roots upwards. Fusarium bulbigenum

was isolated as a pure culture from all parts of the discoloured wood—from root to petiole.

Its pathogenicity has not yet been proved by inoculation, so that it is not known whether

this strain of F. bulbigenum is a specific vascular parasite of tobacco. In later publications

this fungus is referred to as Fusariwm sp. (29, 30, 32). More recently a Fusarium sp.

morphologically identical with #. oxysporum v. mcotianae was isolated from tobacco plants

from the Rustenburg district ; this fungus caused wilting in tobacco seedlings after inocu-

lation.. The role of Fusarium spp. m causing tobacco wilt in South Africa is in need of

investigation. The “ Kromnek”’ disease, previously known as the ““ Kat River wilt.” on

investigation proves to be a virus disease very similar to spotted wilt of tomatoes, etc.

The “ wilt’ in Turkish tobacco, serious in 1926 in the western Cape Province, was also

probably “ kromnek” (31). :

Onion, see Allivm.

Orange, see Citrus.

Panicum, see Gramineae.

Papaver nudicaule L. and P. Rhoeas L.

Foot rot.

Fusarium seirpi.

F. scirpi v. compactum.

Iceland poppies and Shirley poppies are grown in the Transvaal for winter and early

spring flowering. When the temperature rises in the late spring, the leaves often yellow,

and the stems rot. The stem tissues are found to be invaded by a Pythiwm sp. and Rhizoc-

tonia sp. associated with the Fusaria mentioned above.

464

Pawpaw, see Carica.

Pea, see Pisum.

Peach, see Prunus.

Peanut, see Arachis.

Pelargonium sp.

Rhizome, showing firm, brown type of rot.

Fusarium javanicum v. radicicola.

Pennisetum, see Gramineae.

Penstemon sp.

Decaying stems.

Fusarium solani v. Marti f. 1.

Fusarium sp. undet.

From stems of plants showing foot rot, associated with Rhizoctonia sp.

Persea americana Mill.

Roots.

Fusarvum moniliforme.

Fusarium sp. undet. (elegans section).

These fungi were found in roots of an avocado tree which was dying back from the tips

of the branches, and also from the soil of the orchard ; a Phytophthora sp. was also isolated.

Phaseolus vulgaris LL. and Ph. acutifolius Gray v. latifolius Freem.

Foot rot.

Fusarvum oxysporum v. aurantiacum.

F. scurpi v. acuminatum.

F. solani.

F. solani v. Marti f. 1.

The plants from which these fungi were isolated showed yellowing and wilting of the

leaves and stems, but the specific “ dry root rot” organism (Ff. solani v. Martu f. 3) was

not isolated. F. solani was found in stems of Tepary bean, the other species were isolated

from French hean plants.

Phlox Drummondii Hook.

Foot rot.

Fusarium momliforme.

F. scwpir.

These fungi were associated with Rhizoctonia sp. in the decaying stem tissues.

Physalis angulata.

Wilting stems.

Fusarium sp. undet.

Pineapple, see Ananas.

Pinus spp.

Dying seedlings.

Fusarium oxysporum Vv. aurantiacum.

F. scurpt.

Discoloured wood.

F. angustum.

From dying seedlings of Pinus palustris, Pinus taeda and P. longifolia from Zululand

and the northern Transvaal. F’. oxysporum v. aurantiacum is known in Europe as a cause

of “damping off” in seedlings of Conifers.

Pisum sativum L.

Stems of wilting plants.

Fusarium oxysporum f. 8.

F. vasinfectum v. lutulatum.

Fusarvum sp. undet. (elegans section).

Fusarium moniliforme

Fusarium moniliforme v. subglutinans.

F. scwrpr.

F. solani v. Martu f. 1.

The last four fungi are probably organisms which are saprophytic on decaying stem

tissues. Only one set of isolations studied agreed morphologically with F. oxysporum f. 8

(48, 61) but a number of strains of Fusaria of the elegans section were not obtained in good

sporulating condition and could not be identified. The Fusariwm wilt of peas in this country

needs investigation. Fusarium vasinfectum v. lutulatum was isolated from wilting seedlings,

and from plants which did not wilt but failed to set seed. This fungus has been mentioned

as a cause of wilt in America (24). For a discussion of the causes of pea wilt in Europe

and America, see Wollenweber and Reinking (61).

Polygala virgata.

Stem of wilting plant.

Fusarium angustum.

Poppy, see Papaver.

Potato, see Solanum.

Prunus persica Sieb. et Zucc.

Rotting fruit.

Fusarium lateritium.

Fusarium sp. undet. (elegans section).

The undetermined organism invaded fruit which had been severely attacked by

* freckle ” (Cladosporium carpophilum).

Pteridium aquilinum.

Dying stems.

Fusarium sew pr.

associated with Pestalotia sp. and Pythium sp.

Pumpkin, see Cucurbita.

Pyrus malus L.

Core rot of fruit.

Fusarium moniliforme.

FE. moniliforme v. subglutinans.

F. scirpi.

466

In Europe and America, core rot is attributed to F. avenaceum and F. lateritium

and less frequently to F. oxysporum v. awrantiacum and F. lactis (61). The chief cause of

core rot in South Africa appears to be Penicillium expansum.

Red hot poker, see Kniphofia.

Rheum rhaponticum L.

Decaying stems.

Fusarwm solam v. Marti f. 1.

Fusarium undet. (elegans section).

These fungi were found in rhubarb stems which had succumbed to the attack of Phy -

tophthora parasitica v. rhev.

Rhubarb, see Rhewm.

Saccharum officinarum L.

Dying leaf.

Fusarium moniliforme v. subglutinans.

For a discussion of the diseases caused by F.moniliforme and its variety subglutinans,

see Wollenweber and Reinking (61). The Pokkah-boeng disease of sugar cane has not been

observed in Natal.

Sesamum orientale L.

Stems of wilting plants.

Fusarium vasinfectum f. 2.

A wilt of Sesamum has been reported from Turkestan, India and Japan, and 1s attributed

to a Fusarium sp. morphologically similar to F. vasinfectum (61). No infection experiments

were carried out with the strain isolated from South African plants.

Setaria, see Gramineae.

Snapdragon, see Antirrhinum.

Solanum tuberosum L.

Stems of wilted plants and discoloured vascular ring in tubers.

Fusarwm oxysporum f. 1.

Black rot of tubers.

FF. javaneum vy. radicicola.

Storage rot of tubers.

. coeruleum.

. moniliforme.

. orthoceras.

. OLYSPOrUM.

. ScUrpr.

. Scirpt V. acuminatum.

. solani.

. solani v. Martu f. 1.

Rey Ry By By By yy

Fusarium orthoceras appears to be the organism most commonly causing storage of

potato tubers in South Africa. It was found in firm tissues with ight brown discolouration,

in superficial depressed areas and occasionally in tissues affected with a soft form of rot.

For a discussion of Fusarium spp. causing wilt and various forms of tuber rot, see Wollen-

weber and Reinking (61), where an extensive bibliography will also be found.

467

Sorghum vulgare Pers. v. caffrorum (Thun.) Hubb. et Rehder.

Heads moulding in sheath before unfolding.

Fusarium moniliforme.

Pink incrustation on smutted heads.

F. culmorum.

F. heterosporum v. congoense.

Sorghum vulgare Pers. v. technicum (Koern.) Job.

Stems showing foot rot.

Fusarium moniliforme.

Squash, see Cucurbita.

Statice, see Limonium.

Stinkblaar, see Datura.

Stock, see Matthiola.

Strawberry, see Fragaria.

Striga lutea Lour.

Stems and roots of dying plants.

Fusarium equiseti.

F.. moniliforme.

F. scirpi v. compactum.

F. semitectum v. majus.

Fusarium sp. undet. (elegans section).

The plants from which these fungi were isolated had been treated with a so-called

“ witchweed eradicator.” It was claimed that maize fields treated with this eradicator

were cleared of witchweed, which was attacked and killed by a parasitic fungus. Witchweed

plants were treated with this preparation, and a percentage succumbed under very humid

conditions. No specific organism was found in the affected plants, but the Fusaria named

above were isolated, and also species of Pythiwm, Rhizoctonia and Pestalotia. These fungi

are apparently saprophytes, or weak parasites which are only able to attack the plants

under very humid conditions.

Sugar cane, see Saccharum.

Sunn hemp, see Crotalaria.

Sweet pea, see Lathyrus.

Sweet potato, see Ipomoea.

Sweet sultan, see Centaurea.

Tobacco, see Nicotiana.

Tomato, see Lycopersicum.

Triticum sp.

Foot rot.

Fusarvum culmorum.

Stems of plants with blind ears.

F. moniliforme.

Glumes of stunted plants with deformed ears.

F. semitectum v. majus.

468

For a discussion of Fusaria in connection with wheat diseases, and for a bibliography,

see Wollenweber and Reinking (61).

Tropaeolum majus L.

Stem of wilting plant.

Fusarium solani.

associated with Pythiwm sp.

Viscaria viscosa Aschers.

Decaying stem tissues.

Fusarium sew pr.

Vitis vinifera L.

Mycelial growth on bark.

Fusarium sewpi v. acuminatum.

Watermelon, see Crtrullus.

Wheat, see Triticum.

Witchweed, see Striga.

Zea mays L.

Foot rot, root rot, and cob mould.

Pusarvum moniliforme.

F. moniliforme v. subglutinans.

F. graminearum (Grbberella saubineti).

Decaying stems and roots.

F. solani.

F. scirpi v. acuminatumn.

Foot rot and cob mould due to Fusarium spp. are very common and widespread diseases

of maize in South Africa.

B.—Fusaria on Other Fungi.

Hypocreales.

Epichloé Zahlbruckneriana.

Fusarium ciliatum.

F. decemcellulare.

Claviceps spp.

F. avenaceum f. 1.

F. heterosporum v. congoense.

Basidiomycetes.

Uredineae.

Puccinia ranulipes.

Fusarwmn AVENACEUM.,

Ustilagineae.

F, heterosporum v. congoense.

FI, culmorum.

469

C.—Fusaria on Insects.

Aspidiotus furcillae (hidden sclae) on Acacia.

Fusarvum coccophilum.

Aspidiotus perniciosus (pernicious scale) on Pyrus.

Fusarium coccophilum.

Aspidiotus rapax (greedy scale) on Ribes.

Fusarium coccophilum.

Ceroplastis sp. (waxy scale) on Acacia.

Fusarium laterituwm.

Chionaspis sp. on indigenous tree.

Fusarium coccophilum.

Chrysomphalus aurantii (red scale) on Crtrus and Rosa.

Fusarium coccophilum.

Glossina sp. (Tsetse fly).

Fusarium semitectum v. majus.

Icerya purchasi (Australian bug) on Mentha.

Fusarvum sew pr.

Lepidosaphes Gloveri (mussel scale) on Citrus.

Fusarium coccophilum.

F.. sambucinum.

F. sambucinum f. 2.

F. lateritium.

Nomadacris septemfasciata (red locust).

F. sambucinum f. 6.

. SCUrpr.

. Scirpr V. acuminatum.

. scrpr V. filiferum.

. semitectum V. majus.

. solani v. Marti f. 1.

Sy yy

Fusarium coccophilum occurs very commonly on scale insects in the more humid areas

near the south-east coast and in the northern and eastern Transvaal; it is an important

factor in reducing scale infestation. It does not occur where humidity is low, and experience

in other countries has shown that it is useless to try to introduce the fungus into areas where

conditions are unsuitable (35, 60).

D.—Fusaria on Animal Products.

Eggs.

Fusarium moniliforme.

F’. semitectum v. majus.

b= 9

WB,

470

BIBLIOGRAPHY.

Appel O. und H. W. Wollenweber, Grundlagen einer Monographie der Gattung Fusa-

rium (Link), Arbeiten aus der Kaiserlichen biologischen Anstalt fur Land- und Forst-

wirtschaft 8: 1-207, 1910.

Baker, R. E. D., Papaw root and collar rot, Trop. Agric. 10: 328-329, 1933.

Bisby, G. W., Minnesota Agr. Exp. Sta. Bull. 187: 1-47, 1919.

Briant, A. K. and F. B, Martyn, Diseases of cover crops, Trop. Agric. 6: 258-260,

1929.

Briosi, G. Intorno al mal di gomma degli agrumi (Fusisporium limoni Br.), Atti Acead.

Lincei, Rend. Cl. Sci. Fis. Mat. e Nat. Ser. 3a, 2: 485-496, 1878.

. Brown W., Two mycological methods, Annals of Botany, 38: 401-404, 1924.

CF go cthe ee e Studies in the genus Fusarium II. An analysis of factors which

determine the growth forms of certain strains, Annals of Botany, 39: 373-408, 1925.

Doidge, E. M. and A. M. Bottomley, A revised list of plant diseases occurring in

South Africa, Botanical Survey of South Africa, Memoir 11, 1931.

. Doidge, EK. M. and L. J. Kresfelder, A wilt disease of cucurbits, Farming in South

Africa 7: 299-300, 1932.

). Doidge E. M. and J. E. van der Plank, The fungi which cause rots of stored citrus

fruit in South Africa, Union of 8. Afr., Dept. of Agric. and Forestry, Sci. Bull. 162,

1936.

: du Plessis, 8. J., Paratisme, Morphologie en Physiologie van Fusariwm solani (Mart.)

Sace. op Uie, Ann. Univ. Stell. X: Reeks A; Afi, 2, 1932.

ca ea nea ewe een er ae Rooskleurwortel en Bolverroting van Ulie, veroorsaak deur

Fusarium cepae (Hanz.) emend. Link et Bailey. Stellenbosch- Elsenburg landb. Kol-

lege, Wetensk. Bull. 16, 1933.

snmimmmmnnuu2,n«nu««€, nawe2 Die morphologiese Kienskappe en die Parasitisme van verskil-

lende Fusaria op Aartappels. Ann. Univ. Stell. XI: Reeks A, Afl. 3, 1933.

.. Edwards, HE. T., A new Fusarium disease of maize, Agric. Gaz. of New 8. Wales,

44: 895-897, 1933.

1 Studies on Gibberella Fujikuroi var. subglutinans, the hitherto

undescribed ‘stage. ‘of Fusarium moniliforme var. subglutinans and its pathogenicity

on maize in New South Wales, Dept. of Agric., New 8. Wales, Sci. Bull. 49, 1935.

3. Fawcett, Howard 8., Gummosis of citrus, Jour. Agric. Res. 24, 191-232, 1923.

. Fawcett, Howard S. and H. Atherton Lee, Citrus diseases and their control, New

York, 1936.

Hansford, C. G., Ann. Rept. ASric. Dept. Uganda, 1930, Part II, (1938) pp. 58-65.

. Harter, L. L. and J. L. Weimer, The surface rot of sweet potatoes, Phytopathology

9: 465-469, 1919.

Johnson, J., Fusarium wilt of tobacco, Jour. Agric. Res. 20: 515-535, 1921.

Cains niente eee lea aa Tobacco diseases and their control, U.S. Dept. Agric. Bull.

1256, 1924.

21.

22.

471

Kidd, M. N. and R. G. Tompkins, An analytical study of the mortality of orange

fruits at various constant temperatures, Dept. Sci. and Indust. Res., Rept. Food

Investigation Board for 1927, 35-36, 1928.

Kirby, A. H., control gt plant pests and diseases, Rept. Dept. Agric. Tanganyika

territory for 1925, 20-22, 1925.

Layton D. V. and J. J. Wilson, Three new wilt-resistant strains of watermelons, Abst.

in Phytopath. 21: 114, 1931.

Linford, M. B., A wilt disease of peas in Wisconsin, Wisc. Agric. Exp. Sta., Res. Bull.

85: 1-44, 1928.

Link, G. K. K. and Alice A. Bailey, Fusaria causing bulb rots of onions, Jour. Agric.

Res. 33: 929-952, 1926.

Massey, L. M., Fusarium rot of Gladiolus corms, Phytopath. 16: 509-523, 1926.

Mes, Margaretha G., A wilt of snapdragon (Antirrhinum majus) in South Africa, S

Afr. Jour. Sci. 31: 281-287, 1934.

Mitra, M., Wilt disease of Crotalaria juncea L. (Sunn Hemp), Indian Jour. Agr. Sci.

4: 701-714, 1934. é

29. Moore, KE. 8., Diseases of Virginian tobacco in South Africa, Jour. Dept. Agric. 12:

428-455 (Reprint 64) 1926.

Ree eS, eee ee ee The causes of tobacco wilt, Farming in South Africa, 1: 65

1926.

cae renee) aes I Wilt in Turkish tobacco, Farming in South Africa 1: 380”

1927.

. Moore, E. 8. and A. J. Smith, Pests and diseases in tobacco seed beds, Farming in

South Africa 8: 305-306, 1933 (Reprint 46).

. Petch, T., Studies in entomogenous fungi. I. The Nectriae parasitic on scale insects

Trans. Brit. Myc. Soc. 7: 89-167, 1922.

itnsmminmmnmmmnmnimenen ETeSidential address. Fungi parasitic on scale insects, Trans.

Brit. Mye. Soe. 7: 18- 40, 1922.

Sea eee ia De ne Ae Entomogenous fungi and their use in controlling insect pests

Ceylon Dept. Agric. Bull. 71, 1925.

Saale hk ei er mee te Mek Fusarvum pallens (Nees) Link., Trans. Brit. Myce. Soc. 10

282-287, 1926.

. Plakidas, A. G., Fusarium rot of the peach, Phytopath. 15: 92-98, 1925.

. Porter, D. R. and I. E. Melhus, The pathogenicity of Fusarium niwveum and the

development of wilt-resistant strains on Citrullus vulgaris (Sebrad: ), lowa Agric. Exp.

Sta. Res. Bull. 149: 123-184, 1932.

. Reinking, O. A. and H. W. Wollenweber, Tropical Fusaria, Phil. Jour. Sci. 32: 103—

253, 1927.

. Ridgway, Robert, Color standards and nomenclature, Washington, 1912.

. Sherbakoff, C. D., Fusaria of potatoes, New York (Cornell) Agric. Exp. Sta. Memoir

6: 89-270, 1915.

. Sirag-el-din, A., Citrus gummosis in Egypt, Min. Agric. Egypt, Tech. and Sci. Service

(Mycol. Sect.) Bull. 131, 44 pp., 1934.

472

3. Snyder, W. C. and J. C. Walker, Fusarium near-wilt of pea, Zentralblatt f. Bakt. II,

Abt.: 91, 355-378, 1935.

. Storey, H. H., A bark disease of coffee in East Africa, Ann. App. Biol. 19: 173-184,

1932; also 2nd Ann. Rept. Hast Af. Agric. Res. Sta. Amani, for 1929-1930, 1930.

45: Taubenhaus, J. J. and W. N. Ezekiel, Fusarium wilt and corm rot of Freesias, Bot.

raz. 95: 128-142, 1933.

46. Thorold, C. A., Fusarium wilt disease of Sunn hemp II, Trop. Agric. 8: 176-177, 1931.

47. Van der Byl, P. A., Wilt or crown rot diseases of carnations caused by Fusarium sp.,

Ann. App. Biol. 2: 267-291, 1916.

48. Van Hall, C. J. J., Die Sankt-Johanniskrankheit der Erbsen verursacht van Fusariwm

vasinfectum Atk., Ber. Deutsches Bot. Ges. 21: 2-5, 1903.

49. Wager, V. A., Tomato diseases, 6: Fusarium wilt or sleepy disease, Farming in

South Africa 3, 1929.

50. Fusarium wilt in tomatoes: Selection of resistant varieties’

Farming i in South Africa 4, 1930.

51 . Fusarium wilt in tomatoes: Selection experiment No. 2,

Farming i in South Africa 4, 1930.

2 is eran re erence eee as Foot rot disease of papaws, Farming in South Africa 6, 435—

437, 1932.

Las ee ac ot aera See one Ae Aster wilt in South Africa, South Afr. Jour. Sci. 29: 301-312

1932.

DA plement _ Fusarium wilt of tomatoes in South Africa, South Afr. Jour.

Sei. 30: 240-246, 1933.

55), . The wilt disease of asters, Farming in South Africa 8: 1933.

56. Wollenweber, H. W., Studies on the Fusarium problem, Phytopath. 3: 24-50, 1913.

DT. smirmnimiuaninmuarvnmne OCONSPECtus analyticus Fusariorum, Ber. Deut. Bot. Gesell. 35:

732-745, 1918.

DO keene . Fusaria autographice delinata, Nrs. 1-509, ed. 1, 1916, Ann.

Myce. 15, “ilk 56, “1917.

Supplementum; Nrs. 510-659, ed. 1, 1924.

Nrs. 660-1100, ed. 1, 1930 mit index.

Nrs. 1-659, ed. 2, 1926.

Nrs. 1101-1200, Berlin, 1935.

a) Pan ee reer ehet i . Fusarium-Monographie, Fungi parasitici et saprophytici, Para-

sitenkunde, 3: 269- 516, 1951.

60. Wollenweber, H. W. and O. A. Reinking, Aliquot Fusaria tropicalia nova vel revisa,

Phytopath. 15: 153-169, 1925.

Gils (jaeeeeee ee ee eden eeeireeen Die Fusarien, ihre Beschreibing, Schadwirkung und Bekamp-

fung, Berlin, 1935.

Gil ae) pate ee oe te ee Die Verbreitung der Fusarien in der Natur, Berlin (R. Fried-

lander und Sohn), 1935.

473

61b. Wollenweber, H. W. and Hochapfel, H., Beitrage zur Kenntnis parasitarer und sapro

phytischer Pilze. III]. Fusarium und Cylindrocarpon und ihrer Beziehung zur Frucht-

faule,

Z. f. Pflanzenkrankheiten u. Pflanzenschutz, 46: Heft XI: 534-544, 1936-

62. Wollenweber, H. W., C. D. Sherbakoff, 0. A. Reinking, H. Johann and A. A. Bailey,

Fundamentals for taxonomic studies of Fusarium, Jour. Agr. Res. 3C: 833-843, 1925e

63. Young, W. J. and F. M. Read, The preservation of citrus fruit, Progress Rept. of th.

citrus preserv. Committee, Jour. Aust. Council Sci. and Indust. Res. 3: 69-76, 1930

Plate I.

Plate II.

Plate III.

Plate IV.

Fxplanation of Plates.

Fusarium-rot of oranges, caused by artificial inoculation with (a) Fusariwm

lateritium and (b) Fusarium solani var. Marti f. 1; photographed in each case

about 4 weeks after inoculation.

(a) Fusarium-rot of orange, caused by artificial inoculation with Fusariwm

oxysporum; photographed 4 weeks after inoculation.

(b) Section through perithecia of Gibberella Saubinetii, (x 75).

Branch of orange tree infested with red scale, which has been attacked by

Fusarium coccophilum. (Natural size).

(a) Detail from the branch shown in Plate III, showing F. coccophilum growing

out of the margin of the scale. (x 10).

(6) Section through sporodochium of F. coccophilum, taken from specimen

shown in Pl. 3 (~ 150).

INDEX TO SPECIES AND SYNONYMS.

Italicised numbers indicate pages with illustrations.

ciliatum 343

flammeum 335

Calonectria agnina 344

Dearnessvi 3.44

decora 344.

diminuta 344.

Massariae 344.

pyrrochlora 344

rigidiuscula 341

Corallomyces aurantiicola 337

Creonectria diploa 344

Discofusarium fasmaniense 376

Fusarium acaciae 386

acridiorum 447

acuminatum 366

aerugunosum 453

albido-violaceum 405

aleyrodis 360

allia-sativi 447

alluviale 447

angustum 402, 407, 408, 455, 456,

457, 460, 462, 465

anthophilum 349

arcuatum 349

arcuatum Vv. majus 349

arcuosporum 366

aridum 376

asclerotium 405

asparagy 453

aurantiacum 428

avenaceum 348, 460, 468

avenaceum f. 1 349, 350, 456, 459,

460,€462, 468

baccharidicola 335, 338

biforme 348

bostrycoides 401

bufonicola 383

bulbigenum 403,'412, 473, 454, 459,

460, 463

bulbigenum f. 1 416

bulbigenum y. batatas 402

bulbigenum vy. blasticola 402

Fusarium bulbigenum v. lycopersici 402, 416,

417, 456, 457, 459, 461, 462

bulbigenum v. niveum 403,419,456

bulbigenum v. tracheiphilum 402

bullatum 359

bullatum v. brevius 359

bullatum v. minus 359

bullatum vy. roseo-bullatum 359

bullatum v. roseum 359

caleareum 428

callosporum 335

candidulum 42)

caricis 382

cataleptum 335

caudatum v. solani 368

celosiae 394.

cepae 425, 438

chenopodinum 369

chlamydosporum 345, 346, 460

ciliatum 343, 344, 468

ciliatum v. majyus 343

eitrulli 419

cocemellum 335, 337

cocecophilum 335, 336, 469

coeruleum 443, 453, 466

commutatum 447

conglutinans 402

conglutinans v. betae 402

conglutinans v. callistephi 402,

410, 411, 455

conzlutinans v. citrinum 402

conglutinans v. majyus 410

congoense 3T1

congoense v. septatius 371

Cordae 356

cromyophthoron 412

culmorum 371, 380, 38/, 460, 467,.

468

culmorum f. 1 380

culmorum vy. leteius 38)

culmorum v. majus 380

decemecellulare 339, 340, 457, 468

Delacroixii 376

475

Fusarium dianthi 404, 431, 432, 459

dimerum v. pusillum 456

discolor 376

discolor v. majus 383

discolor v. sulphureum 380

discolor v. triseptatum 376

diversisporum 456

effusum 348

elegans 421

elongatum 343, 428

equiseti 356, 397, 457, 458, 462, 467

equiseti f. 1 356

equiseti v. bullatum 356, 359, 462

equiseti v. bullatum f. 1 359

equiseti v. bullatum f. 2 359

equiseticola 368

equisetorum 412

erubescens 366

euoxysporum 423

faleatum 356

falcatum v. fuscum 356

ferruginosum 366

filiferum 368

filisporum 343

fimicolum 383

fraxini 376

fructigenum 386

fructigenum v. majus f. 1 391

genevense 380

gibbosum 360

glandarium 443

sraminearum 371, 382, 383, 468

graminearum v. caricis 382

granulare 316

gynerit 383

heidelbergense 380

herbarum 348

herbarum f. 1 348

herbarum f. 2 348

herbarum v. avenaceum 348

herbarum v. conti-maculatc 376

Fusarium herbarum vy. gibberelloides 348

herbarum v. graminum 348

herbarum v. pirinum 348

herbarum v. tubercularioides 348

herbarum vy. viticola 548

heterosporum v. congoense 371,

372, 460, 467

heterosporum v. congoense f. 1 371

heterosporum {. paspali 348

heveae 443

hippocastani 366

horder 376

incarnatum 353

insidiosum 383

javanicum 443, 444, 458

javanicum v. radicicola 443, 446,

464, 466

javanicum v. theobromae 443

juglandinum 353

lactis 466

lanceolatum 366

lateritium 385, 386, 456, 457, 459,

465, 469

lateritium f. 1 386

lateritium v. fructigenum 386

lateritium v. fructigenum f. 1 386

lateritium v. fructigenum f. 2 386

lateritium v. longum 389, 456, 457

lateritium v. longum f. 1 389

lateritium v. pallens 386

lateritium v. tenue 386

larum 412

limonis 386

lini 402

loncheceras 412

loncheceras v. microsporon 412

longisporum — 389

lucidum 348

lutulatum 436

lycopersici 416

Malli 447

malvacearum 432

Marti v. minus 450

+76

Fusarium Marta v. viride £50

maydis 376

metachroum 348

metachroum Vv. minus 348

Mollerianum 383

moniliforme 353, 393, 394, 395, 454,

455, 457, 459, 460, 462, 463,

464, 465, 466, 467, 468, 469

moniliforme v. erumpens 394

monaliforme v. ficr 394

moniliforme v. majus 394

moniliforme v. subglutinans 333,

398, 399, 454, 455, 457,

461, 465, 466, 468

mucronatum 356

mycophilum 421

mycophytum 368

myosotidis 421

nectriae-turreae 330

nectria-palmicolae 359

neglectum 380

nicotianae 424

niveum 419

opuntiarum 421

orthoceras 402, 405, 466

orthoceras v. albido-violaceum 405

orthoceras v. apil 402

orthoceras v. apu f. 1 402

orthoceras v. longius 402

orthoceras v. pisi 402

orthoceras v. triseptatum 405

ossicolum 356

osteophilum 368

oxysporum 404, 427, 456, 457, 461,

466

oxysporum f. 1 404, 423, 466

oxysporum f. 2 404

oxysporum f. 5 424

oxysporum f. 6 404, 425, 426

oxysporum f. 7 404, 425, 454

oxysporum f. 8 404, 427, 465

oxysporum v. asclerotiwm 405

oxysporum v. aurantiacum 404,

428, 429, 455, 458, 464

oxysporum v. aurantiacum f. |

403, 428, 429

oxysporum Vv. aurantiacum

hyalina 353

oxysporum subsp. aurantiacum 353

Fusarium oxysporum v. cubense 404

oxysporum v. cucurbitacearum 405

oxysporum v. gladiol 404, 430

431, 460

oxysporum {. lycopersics 416

oxysporum subsp. lycopersici 416

oxysporum v. lycopersici 416

oxysporum v. medicaginis 404

oxysporum v. nicotianae 404, £24,

463

oxysporum Vv. resupinatum 405

pallido-roseum 353

pannosum 3716

parasiticum 343

paspali 348

Peckhu 428

peltigera 348

pestis 450

poae 345

Poolensis 419

pseudoeffusum 366

pulvinatum 376

radicicola 446

redolens 403

redolens f. 1 403, 440, 447, 462

redolens v. angustius 423

reticulatum 458

rhizochromatistes 412

rhizochromatistes v. microsclerotium

412

rhoicolum 383

ricint 376

roseo-bullatum 359

roseum 316, 383

roseum v. calystegiae 353

roseum v. cucubali-baccifer’ 383

roseum v. maydis 383

roseum v. rhev 380

roseum Vv. solani nigri 360

rostratum 380

rubsginosum 380

russianum 366

Saccardoanum 428

sambucinum 3871, 375, 457, 462, 469

sambucinum f. 2 370,378,457, 469

sambucinum f. 3 380

sambuecinum f. 6 371, 380, 469

Fusarium sambucinum v. medium 376

samoense 394

sanguineum 366

Schiedermayert 349

Schribauzi 380

scirpl 356, 360, 361, 454, 455, 456,

457, 458, 459, 461, 462, 463,

465, 466, 468, 469

scurpy f. 1 360

scirpl Vv. acuminatum 356, 366,

367, 459, 466, 468, 469

scirpt V. comma 360

scirpi v. compactum 356, 364, 455,

456, 457, 458, 461, 462, 463, 467

scirpt Vv. compacium f. 1 364

scirpi v. filiferum 356, 368, 369,

454, 469

scurpt V. nigrans 360

scurpr Vv. nigrantum 360

serpi v. pallens 360

sclerodermatis360

sclerodermatis v. lycoperdonis 360

sclerostromaton 407

sclerotioides 428

sclerotium 360

scolecoides 343

semitectum v. majus 352, 354, 456,

457, 459, 463, 467, 469

solani 443, 447, 448, 454, 456, 459,

460, 466 468

solani f. 1 447

solani v. cyanum 447

solani v. Martu f. 1 443,450, 457,

454, 456, 457, 458, 459,

461, 464, 465, 466, 469

solani v. Marti f. 3 464

solani v. medium 447

solani v. suffuscum 447

sorghi 348

spicariae-colorantis 339

stictoides 383

stilboides 391, 456, 457

stilboides v. minus 391

subcarneum 376

477

Fusarium subpallidum v. roseum 378

Fusidium aloes

subulatum 348

subulatum v. brevius 348

sulphureum 380

tabaciwworum 424

tenellum 376

tenuissimum 376

terrestris 359

theobromae 339, 443

trifolii 421

truncatum 348

403, 419, 432, 433,

460, 461

vasinfectum f. 1 403

403, 434, 435,

456, 457, 462

vasintectum v. lutulatum 403. 436,

456, 461, 465

vasinfectum v. pisi 427

vasinfectum

vasintectum f. 2

vasinfectum vy. zonatum 403

vasinfectum v. zonatum f.1 403,

38, 462

vasinfectum v. zonatum f. 2 403,

439, 454

versicolor 380

violaceum 376, 453

viride 447

zeae 348

zonatum f. 1 438

zonatum f. 2 439

366, 367

roseum 383

Fusisporium avenaceum 348

chenopodinum 360

339, 337

culmorum 380

filisporum 343

ancarnatum 353

coccinellum

mearnatum Vv. tussilago-farfarae

368

mycophytum 368

ossicola 356

pallido-roseum 353

Schiedermayeri 349

478

Fusisporium solani 447 Nectria coccidophthora v. aurantiicola 337

solani v. flasum 447 coccophila 337

solani-tuberost 447 colletiae 337

Fusoma filiferum 368 congoensis 337

helminthosporii 360 dahliae 367

pallidum 396 decora 344

tenue 380 diploa v. diminuta 344

Gibberella acuminata 367 laeticolor 337

baccata 387 massariae 344

Fujikuroi 395 musciyoro 337

Fujikuroi v. subglutinans 399 Passerimana 337

intricans 359 subcoceinea 337

moniliformis 395 subfurfuracea 337

pulicaris 376 turraeae 337

Saubinetii 383, 468 Pionnotes flavicans 382

Saubinet {. dahliae 367 pseudonectria 335

Hymenula equiseta 412 solani-luberosi 447

Lachnidium acridiorum 447 vagans 3716

Lisea Fujikurot 395 viridis 447

Microcera auranticola 335 Sarcopodium avenaceum 348

cihata 343 Selenosporium bufonicola 383

coccidophthora 330 coeruleum 453

coccophila 335 equiseli 356

massarvae 343 hippocastani 366

mytilaspidis 389 Sphaeria decora 344

pluriseptata 335 Sphaerostilbe aurantucola 337

tasmanica 316 coccidophthora 337

Nectria aglaothele 337 coccophila 337

aurantucola 337 flammea 337

Balansae 337 Spicaria colorans 339

coccicida 337 Stilbum flammenm 335

coccocidophthora 337 Tubercularia coccophila 335

479

Puate I.

w= ae 4

Price, 7s. 6d. . Part 4,

Bothalia

A RECORD OF

CON PRIBUTIONS

FROM THE

NATIONAL HERBARIUM

UNION OF SOUTH AFRICA

PRETORIA

ee Wy de

a Fay Bi CZy y

EDITED BY

I B. POLE EVANS, C.M.G., M.A. (Cantab.), D.Sc. (Wales),

_ CHIEF, DIVISION OF PLANT INDUSTRY, DEPARTMENT OF AGRICULTURE AND FORESTRY

‘ PRETORIA; AND

DIRECTOR OF THE BOTANICAL SURVEY OF THE UNION OF SOUTH AFRICA

PRINTED IN THE UNION oF SouTH AFRICA BY THE Govicmerar PRINTER, PRETORIA,

PousiisHED APRiL, 1939.

fe a G.-P.-S.11470—1939—1,000

SOUTH AFRICAN RUST FUNGI, _ IIL

E. M. ate

[ The first paper of this series, entitled “A Preliminary

Study of the South African Rust Fungi’ constituted Part Ia

of Vol. 2 of Bothalia (1927). South African Rust Fungi II

appeared in Vol. 2 Part 2 (1928) J.

Aecidium Burtt-Davyi nov. spec.

Aecidiis foliicolis et petiolicolis, matricem incrassatis, plerumque dense confertis,

cylindraceis, usque 2 mm. longis, 300-400 y latis, pallide flavis, primo clausis dein apertis

margine erecto vix lacerato ; cellulis peridie arctissime conjunctis, irregularibus, 25-42-5 x

10-20 , pariete exteriore striato 8-15 crasso, interiore striato-verrucoso 4-5 y crasso ;

sporis subglobosis, ovatis, ellipsoideis v. oblongis, saepe irregularibus et angulatis, sub-

tilissime verruculosis, 21-43 x 16-22-5 yw, episporio ubique 2-2-5 yw crasso, poris germina-

tionis obscuris.

Hab. in foliis petiolisque Acaciae stoloniferae Burch., Christiana, leg. Burtt Davy,

1973 et 5581.

This may possibly be the aecidial stage of Ravenelia modesta (q.v.) which occurs on the

same host, but the aecidium and the teleuto-form were collected in widely separated

localities and no connection between the two forms has yet been established.

Aecidium Dinteri nov. spec.

Pyenidiis minutis, superficialibus, inter aecidia distributis, applanato-hemisphericis,

melleo-brunneis, 50-75 y diam.

Aecidiis ramicolis et petiolicolis, ramulos incrassatos et uncinatos plus minus contortos

dense aequaliterque distributis, longe exsertis, tubulosis, 2-5-4 yw altis, 250-300 y latis,

pallide flavo-brunneis, diu clausis, tandem apertis, margine pallidiore, leniter recurvato,

denticulato ; cellulis peridie arctissime conjunctis, quoad formam valde variabilis, saepe

rectangularibus v. irregulariter rhomboideis, 25-43 x 15-20 yw, pariete exteriore striato,

12-15 u crasso, interiore striato-verrucosa 3-5-5 » crasso. Sporis irregularibus plerumque

angulato-globosis oblongis v. ellipsoideis, 25-32-5 x 17-5-22-5 w, dense minuteque ver-

ruculosis, subhyalinis; episporio 2-5-3 m crasso, poris germinationis sparsis (usque 8)

praedito.

Hab. in ramis petiolisque Acaciae uncinatze Engl., Otjihavera-Okapuka, S.W. Africa,

leg. Dinter 3488 (ex. Herb. Marloth) 26686.

Aecidium litakunensis nov. spec.

Pyenidiis inter aecidiis sparsis, melleis, applanato-hemisphericis, 60-100 diam.

Aecidiis foliicolis et ramicolis ; in foliis petiolisque gallas subglobosas usque oblongas,

plus minus flexuosas et distortas, magnitudine variabiles, 2-4 cm. longas, 5-15 mm. crassas

formantibus, dense confertis, cylindraceis, usque 1-5 mm. longis, 350-500 y latis, flavidis

v. flavo-brunneis, diu clausis ; cellulis peridie irregularibus, 20-37-5 x 7:5-17-5 yw, pariete

exteriore striato 12-15 crasso, interiore striato-verrucoso 3-5 yp crasso. Sporis globosis,

ovatis, ellipsoideis v. oblongis, saepe irregularibus et angulatis, verruculosis, pallide flavidis

v. subhyalinis, 22-30 x 17-22 yp, episporio ubique 1-5-2 y crasso, poris germinationis

obscuris.

Hab. in foliis Acaciae litakunensis Burch., inter Rustenburg et Northam, leg. Scott,

27294; Bechuanaland leg. Burtt Davy, 2443.

488

Aecidium Moggii nov. spec.

Pycnidiis amphigenis sed plerumque epiphyllis, copiose evolutis, inter aecidia distributis

vel per magnam folii partem aequaliter dispositis, ex flavidis tandem nigrescentibus, 90-130

diam.

Aecidiis hypophyllis, singulis subinde epiphyllis, plerumque per totam folii super-

ficiem dense aequaliterque distributis, cupulatis, 200-250 1 diam., margine laciniato; cellulis

eridie fere quadraticis usque rhomboideis, 20-35 x 10-20 mw, pariete exteriore striato

5-6 ps. crasso, interiore verrucoso 3-5 yw crasso. Sporis ar.gulato-globosis, oblongis v. late

ellipsoideis, minutissime verruculosis, subhyalinis, 17—22-5 x 15-20 yp, episporio 1-5-2 p

crasso.

Hab. in foliis Senecionis coronati Harv., Pretoria, leg. Mogg, 23636.

Aecidium tetragoniae nov. spec.

Pyenidiis amphigenis, inter aecidia copiose sparsis, melleis dein atro-brunneis, 120-180 px

diam.

Aecidiis amphigenis, totam folii superficiem vel magnam ejus partem dense aequaliterque

obtegentibus, diu hemispherico-clausis, tandem apertis cupulatis, 300-400 4 diam., margine

albido laciniato mox evanescente ; cellulis peridie laxe conjunctis, irregularibus, 27-42 x

15-25 w, pariete exteriore striato 5-8 ys crasso, interlore verrucoso 2-3 p crasso. Sporis

sub-globosis, oblongis, ellipsoideis vel ovatis, plerumque angulatis, 20-32-5 x 15-20 pw;

episporio 2-5-5 4 crasso, dense minuteque verruculoso.

Hab. in folus Tetragoniae arbusculae Fenzl., Fauresmith, leg. Henrici, 25892.

Masseeélla flueggeae Syd.

in Ann. Myc. 26 (1928) 424.

Syn. Uredo brideliae (P. Henn. et Evans) Doidge in Bothalhia 2 (1927) 193.

Aecidium brideliac P. Henn. et Evans in Engl. Bot. Jahrb. 41 (1908) 272;

Syd. Monogr. Ured. 4 (1924) 186.

Uredo-sori hypophyllous, on small leaf spots which may be indistinct, or yellow to

yellow-brown, irregularly scattered or in groups of 2-8, long covered by the discoloured,

yellow-brown epidermis, very minute, 200-350 4 diam. Uredospores subglobose, ovate or

ellipsoid, borne singly, 17-26 « 14-20 4; epispore hyaline, 1-5-2 » thick, minutely ver-

ruculose-echinulate, germ pores obscure.

Teleuto-sori epiphyllous—very rarely a single one occurs on the under side of the

leaf—deeply immersed, in small or larger groups, or even solitary, minute; the spores

protrude from the sori in filiform columns 1-2-5 mm. long and 50-90 y broad, these are

usually curved and become dark brown when dry. Teleutospores ovate or ellipsoid, often

slightly angular, 1-celled, smooth, deep yellow or golden yellow, embedded in a mucous

mass, 22-30 x 15-21 4; epispore about 3 y thick with a distinct apical germ pore.

Hab. on leaves of Fluggea virosa (Roxb.) Baill. (= Fluggea microcarpa Blume) near

Nelspruit, Burtt Davy, 77; Nelspruit, Liebenberg, 25968; Schagen, Liebenberg, 26359.

In Bothalia (loc. cit.) it was poimted out that number 77 was a uredo-form; more

recently the teleuto-stage has been collected by Liebenberg, and it agrees in every detail

with the fungus described by Sydow on Fluggea virosa from the Philippines.

Puccinia bylianum Dippenaar.

in the South African Journ. Sci. 28 (1931) 288.

Syn. Aecidium bylianum Syd. in Ann. Myc, 22 (1924) 236; Bothalia 2 (1927) 172.

Aecidia hypophyllous or caulicolous, developing on thickened and sometimes distorted

parts of the host plant, sub-seriate or in densely crowded, elongated groups 1—2 cm. long

(fide Sydow, amphigenous, chiefly hypophyllous, on yellow leaf spots, closely crowded in

489

groups 3-8 mm. diam.) immersed, cupulate, 250-300 4 diam., margin erect, white, lacerate.

Cells of the peridium firmly compacted, rhomboid, 20-30 x 10-16 4; outer wall striate,

6-10 » thick, inner verrucose 3-5 y thick. Spores angular-globose, subhyaline, 10-18

diam., epispore very delicately verruculose, about 1 y thick.

Fig. 1.—Puccima bylianwm, teleutospores.

Teleuto-sori caulicolous, rarely on the leaves, scattered or in groups, often developing

between the aecidia, long covered by the raised epidermis, which finally splits longitudinally,

oblong, about 4 mm. long, sometimes becoming confluent and up to 4 mm. long, black.

Teleuto-spores ablone, ellipsoid or subclavate, 35-52-5 x 12-5-22-5 p, light brown, darker

at the apex; apex rounded, truncate or obtusely acuminate, often oblique, attenuate at

the base, constricted at the septum; epispore smooth, 1-5-2 » thick, thickened at the

apex up to 8; pedicel persistent, up to 42 ~ long, subhyaline or light brown, 5-6 y thick.

Mesospores fairly numerous, elipsoid to subclavate, 25-33 x 15-17-5 w; epispore similar to

that of the teleutospores. An occasional 3-septate spore was seen.

Hab. on leaves and stems of Senecio Burchellii DC., Bloemfontein, Potts, 24875 (Grey.

Univ. Coll. No. 8017) ; I. Pretoria, Fuller, 15018.

In Bothalia 2 (1927) 171, the aecidial stage (No. 15018) was recorded as Aecidium

incertum Syd. on Senecio laevigatus Thun. On comparison, the host of this number proves

to be the closely similar species Senecio Burchellii, and the aecidium is identical with that

of the plant on which teleuto-sori were found.

The aecidium closely resembles Aecidiwm incertum Syd., which was first described on

Senecio napifolius ; no teleutosori have yet been found on the latter host, so that it is

impossible to say whether Aecidium incertum Syd. is a synonym for Puccinia bylianum.

Although the teleutospores of the Orange Free State specimen are somewhat smaller,

there can be no doubt that this is the rust found in the winter rainfall area, and described

by Dippenaar as occurring on Senecio spp., Senecio littoreus, S. pinnulatus, S. vulgaris and

S. bipinnatifida from Ceres, Hopefield, Knysna and Stellenbosch.

This species differs widely from Puccinia pentactina described as occurring on Senecio

pentactinus [Bothalia 2 (1928) 473].

bebe di

Fig. 2.—Puccinia pentactina, teleutospores.

490

Puccinia Liebenbergii nov. spec.

Aecidiis hypophyllis, maculis brunneolis 1-2 mm. diam. insidentibus, 3-7 in centro

macularum aggregatis, cupulatis, 200-250 4 diam., margine albido reflexo laciniato ; cellulis

peridie laxiuscule conjunctis, rhomboideis, 20-25 x 15-18 yw, pariete exteriore striato,

2-4 crasso, interiore verrucoso, 2-5 4 crasso. Aecidiosporis angulato-globosis v. ellipsoideis,

dense et subtilissime verruculosis, subhyalinis, 12-20 x 11-15 yp, episporio 1 y crasso.

Soris uredosporiferis amphigenis, plerumque epiphyllis, per totam foli1 superficiem

plus minusve dense sparsis, interdum circa aecidia annula efficientibus, minutis, rotundatis,

usque 4 mm. diam., epidermide bullata diu tectis, dein poro rotundo centrali apertis,

cinnamomeo-brunneis. Uredosporis subglobosis v. ovatis, flavo-brunneolis, ubique aequaliter

et sat valide aculeatis, 20-25 x 20-21 y, episporio 1-5-2 u crasso, poris germinationis

obscuris.

eciyes

dds

Fig. 3.—Puccima Liebenbergu, teleutospores.

Soris teleutosporiferis conformibus, atro-brunneis, vel teleutosporis in soris uredospori-

feris immixtis. Teleutosporis ellipsoideis, sub-globosis v. irregularibus, levibus, brunneis,

27-540 « 20-27-5 pw, apice plerumque rotundatis rarius applanatis, medio leniter con-

strictis, basi plerumque rotundatis interdum attenuatis ; episporio 1-5 1 crasso, ad apicem

haud incrassato; pedicello hyalino, deciduo, teleutosporam subaequante, verticale v.

oblique, nonnunquam transverse inserto.

Hab. in foliis Sonchi sp., Schagen, leg. Liebenberg, 26176.

Puccinia gnidiae nov. spec.

Uredosporis in soris teleutosporiferis immixtis, late ellipsoideis, ovatis v. subglobosis,

brunneis, subtiliter denseque verruculosis, 25-30 x 22-5-25 4; episporio 2-5-3 y erasso,

poris germ. ca 4-7 sparsis praeditis.

Soris teleutosporiferis caulicolis vy. amphigenis, sparsis v. aggregatis, rotundatis v.

ellipticis, interdum confluentibus, $-3 mm. longis, pulvinatis, epidermide lacerata cinctis.

Teleutosporis sub-globosis, late ellipsoideis v. ovatis, levibus, castaneo-brunneis, 30-40 x

25-30 p, apice late rotundatis interdum compressis, basi rotundatis v. subattenuatis, medio

vix constrictis ; episporio 2-5-4 m crasso, apice leniter (usque 8 yz) incrassato ; pedicello

persistente, valido, interdum oblique inserto, hyalino v. subflavescente, 7-5-10 pu crasso,

lumine haud raro fere nullo.

Hab. in caulis foliisque Gnidiae macrocephalae Meisn., Kaalfontein, leg. Pole Evans,

10083.

AQ]

The stems of the host plant showed indications of the occurrence of an aecidium earlier

in the season.

Vig. 4.—Puccinia gnidiae, teleutospores and uredospores.

Mesospores were occasionally seen ; they were ovate, 30-33 25-27-5 py, with charac-

ters similar to those of the teleutospores. The upper loculus of the teleutospore is often

compressed, directly (making the apex almost flat) or obliquely.

Puccinia iridis (DC.) Wallr.

in Rabh. Krypt. Fl. ed. I (1844) 23; Syd. in Monogr. Ured. 1 (1904) 598.

Uredo-sori amphigenous, scattered or subgregarious, round, oblong or elongated, long

covered by the epidermis, which at length splits and remains partially veiling the pulverulent

sorus, up to 2 mm. long, rusty brown. Uredospores globose, subglobose, ellipsoid or ovate,

ochraceous, 20-35 x 16-26 ; epispore rather coarsely echinulate, 1-5-2 mm. thick,

slightly thicker (2-5 4) at the base, germ pores 2-3 equatorial, rather conspicuous, papillate,

Fig. 5.—Puccinia iridis, teleutospores and one uredospore.

Teleuto-sori hypophyllous, scattered, sometimes confluent, sometimes developing

between the uredo-sori, linear or striiform, early becoming naked, black. Teleutospores

clavate or oblong, rounded, acuminate or truncate at the apex, slightly constricted at the

septum, usually attenuate at the base, ight golden brown, darker at the apex, 30-52 x

492

14-22 «; epispore smooth, delicate, about 1 thick in the lower loculus, up to 2 y thick

in the upper, very much thickened, up to 14 1, at the apex; pedicel tinted brown, persistent,

as long as the spore or shorter. The loculi separate readily at the septum.

Hab. on leaves of Iris germanica L., Johannesburg, Hollis, 25922, and Iris sp. (regilio-

cyclus) Johannesburg Hingst 30142.

This rust, which is widely distributed in Europe and America, has only recently made

its appearance in gardens in Johannesburg.

Puccinia caricis-cernuae nov. spec.

Soris uredosporiferis amphigenis, plerumque hypophyllis, sparsis v. gregariis, oblongis

v. linearibus, usque 3 mm. longis, epidermide diu tectis, tandem longitudinaliter apertis,

cinnamomeis ; Uredosporis ovatis, ellipsoideis v. subglobosis, dilute brunneis, echinulatis,

20-25 x 16-19 p. episporio 2-2-5 » erasso, poris germ. obscuris.

Fig. 6.—Puccinia caricis-cernuae, teleutospores.

Soris teleutosporiferis amphigenis, plerumque hypophyllis, maculis nullis v. indeter-

minatis flavis insidentibus, sparsis v. gregariis, oblongis v. linearibus, saepe in strias longas

(usque 6 mm.) confluentibus, diu epidermide plumbea tectis, pulvinatis. Teleutosporis

clavatis v. fusoideo-clavatis, apice rotundatis v. conico-angustatis, saepe oblique productis,

nonnunquam subtruncatis, medio leniter constrictis, basi attenuatis, levibus, pallide

brunneolis, sursum saturatioribus, 37-5-62-5 x 12-5-9 #; episporio ca. 1 crasso, apice

leniter (2:5-9 1) incrassato ; pedicello brunneolo, persistente, usque 48 » longo, 4-5 pn

crasso.

Hab. in foliis Caricis cernuae Boott. var. austro-africanae Kuk., Pelindaba, leg. Doidge

et Bottomley, 29873; Skinner’s Court, leg. Doidge et Bottomley, 23444.

Puccinia pegleriana nov. nom.

Syn. Puccinia cyperi-tagetiformis (P. Henn.) Kern var. africana Doidge in Bothalia 2 (1927)

116.

The South African fungus differs from Puccinia cyperi-tagetiformis (P. Henn.) Kern

(Mycologia XI, 1919, p. 138) in several particulars ; the uredospores are somewhat larger

and have a much thicker wall, and the teleutospores are considerably longer. P. eyperi-

tagetiformis is not represented in the Cryptogamic Herbarium at Pretoria and has not been

compared with the South African material, but in view of the distinct differences in the

descriptions of both uredo- and teleutospores, it is considered advisable to regard this as

a distinct species and not as a variety.

493

Puccinia Morganae nov. spec.

Soris teleutosporiferis hypophyllis, maculis flavidis v. brunneolis effusis saepe insidenti-

bus, minutis, rotundatis v. ellipticis, usque 0-4 mm. diam., sparsis v. irregulariter aggregatis,

interdum confertis confluentibusque, pulvinatis, diutius tectis, demum epidermide rupta

cinctis v. semivelatis, ochraceo-brunneis. Teleutosporis plerumque clavatis v. oblongo-

clavatis, interdum fusiformis, rarissime 2-septatis, apice rotundatis, recte v. oblique conico-

attenuatis, rarius truncatis, subhyalinis v. dilutissime flavo-brunneolis, 37-5-55 « 12-5—

15 p, rarius usque 17 latis ; episporio leve, tenue, usque | crasso, apice leniter (usque 3 2)

incrassato ; pedicello subhyalino persistenti, usque 35 p longo.

Me OAR

Fig. 7.—Puccinia Morganae, teleutospores.

g p

Hab. in foliis Cyperi albostriati Schrad. in silvis Xumeni, prope Donnybrook, leg.

Morgan et Doidge, 29985.

This species differs widely from Puccinia cyperi-fastigiati Doidge and P. pegleriana

(Bothalia 2, 1928, p. 473; and 2, 1927, p. 117). Teleutospores of the former species are

illustrated for comparison ; for form of the teleutospores of P. pegleriana see Bothalia,

loc. cit.

Fig. 8.—Puccinia cyperi-fastigiati, teleutospores and uredospores.

Puccinia fuirenella nov. spec.

Soris uredosporiferis sine maculis, plerumque hypophyllis, rarius epiphyllis, linearibus,

minutis, usque 1 mm. longis, primo epidermide tectis, brunneis. Uredosporis late ellip-

soideis, globosis v. ovatis, brunneis, 25-32-5 x 20-25 m; episporio 2-2-5 mw crasso,

remotiuscule echinulato, poris germinationis 2 conspicuis papillatis equatorialibus praeditis.

Soris teleutosporiferis conformibus. Teleutosporis longe clavatis v. lanceolatis, levibus,

diluto aureo-brunneis, apice plerumque pallidioribus, rotundatis truncatis v. conicis,

nonnunquam obliquis, rarissime dentibus 2 coronatis, medio leniter constrictis, deorsum in

pedicellum attenuatis. 42-72-5 « 17-5-22-5; episporio 1-2-5 yp crasso, ad apicem valde

incrassato usque 13 » ; pedicello subhyalino, usque 40 1 longo, apice 8-10 y crasso, deorsum

attenuatis.

Hab. in foliis vaginisque Fuirenae pubescentis (Lam.) Kunth., Donkerpoort, leg. Doidge

et Bottomley, 29977.

494.

The uredo-form also occurs on :—

Fwirena coerulescens Steud., Maritaburg, Pole Evans, 1444.

Fuirena pubescens (Lam.) Kunth, Nelspruit, Lvebenberg, 26074.

Fuirena chlorocarpa Ridley, Xumeni Forest, Donnybrook, Morgan and Doidye,

29823 ; Donkerpoort, Doidge and Bottoml<y, 29983.

Fig. 9.—Puccinia fuirenella, teleutospores and uredospores.

The uredo on F. coerulescens was recorded under the name Uredo Fuirenae P. Henn.

in Bothaha 2 (1927), p. 198. Uredo Fucrenae P. Henn. was originally described on Fuirena

umbellata from Brazil, and has been shown by Kern, Ciferri and Thurston in Ann. Myc.

31 (1933), p. 13, to be the uredo-form of Puccinia fuirenicola Arth. A portion of the type

collection (Mycoflora Domingensis exsiccata 117) of this rust has been examined; the

uredospores resemble those of Puccinia fuirenella, but the teleutospores differ in several

particulars from those of the South African rust. The original description reads “ teliospores

cylindric or ellipsoid, constricted at septum, rounded or somewhat pointed above and usually

narrowed below ; wall cinnamon brown, 1-5-2 #; upper cell usually darker and thickened

at the apex, 4-7 we; pedicel slightly tinted, ne length of the spore or less ”’.

Puccinia fuirenella also differs from P. Fuirenae Cke. [Grevillea 6, p. 137 ; Syd. Monogr.

Ured. 1 (1904) 687]; the teleutospores of the latter species are more slender, 11-15 p broad,

and the uredo-spores have 3-4 germ pores.

Puccinia kyllingicola nov. spec.

Soris uredosporiferis amphigenis v. calamicolis, plerumque autem hypopbyllis, maculis

conspicuis rufo-brunneis insidentibus, sparsis v. paucis ag geregatis, nonnumquam circulariter

circa sorum centralem dispositis, ellipticis v. oblongis, } to 1 mm. longis, epidermide inflata

diu tectis. Uredosporis ovatis, ellipsoideis, spalbpalelvonis, subinde angulatis, diu hyalinis v.

subhyalinis, tandem flavis v. flavo-brunneolis, 22-32:5 x 15-20 4; episporio 1-5-2 p

crasso, breviter laxiusculeque echinulato, poris germ. 2, rarius 3-4, equatorialibus praedito.

NOD,

I! 10.—Puccinia kyllingicola, teleutospores,

495

Soris teleutosporiferis minutis, compactis, epidermide tectis. Teleutosporis oblongis

v. clavatis, rectis v. curvatis, levibus, aureo-brunneis, apice rotundatis, truncatis v. attenu-

atis, interdum obliquis, medio leniter constrictis, basi saepius attenuatis, 40-77 x 15-25 wv

plerumque 15-20 y latis; episporio ca. 1-5 crasso apice incrassato usque 10 1 ; pedicello

subpersistente, brunneolo, usque 25 longo.

Hab. in foliis calamisque Kyllingae melanospermae Nees, Donkerpoort, leg. Doidge et

Bottomley, 29971; II. Karino, leg. Wager, 23414.

This rust differs from Puccinia mysorensis Syd. et Butler [Ann. Myc. 4 (1906), p. 434],

in the size of the teleutospores, which are larger and thicker at the apex. The uredo

resembles Uredo Kyllingiae P. Henn., but so far as can be ascertained the teleuto-stage of

this American species has not been described.

Puccinia schoenoxyphii nov. spec.

Soris teleutosporiferis sparsis, maculis minutis, brunneis, conspicuis, plerumque

ellipticis, usque 2 mm. longis, singulariter insidentibus ; soris minutis, ellipticis, $ to ? mm

longis, fusco-brunneis, diu epidermide bullata tectis. Teleutosporis plerumque clavatis,

rarius fusoideis v. oblongis, rectis v. leniter curvatis, pallide luteo-auranteis, ad apicem

obscurioribus, apice rotundatis rarius recte v. oblique obtuse conicis, medio plus minus

contrictis, basi plerumque attenuatis, nonnunquam subrotundatis, 40-58 x 14-20 p;

episporio leve, tenue, 1-1-5 y crasso, ad apicem valde incrassato, 7-5-12-5 yw, imterdum

usque 15:5 4: pedicello persistente, leniter brunneolo, 5 4 crasso et usque 50 « longo.

Fig. 11.—Puccinia schoenoxyphu, teleutospores.

Hab. in foliis Schoenoxyphii spartii Kuk., in silvis Xumeni, prope Donnybrook, leg.

Morgan et Doidge, 30106.

Puccinia scleriae-dregeanae nov. spec.

Soris uredosporiferis amphigenis, plerumque hypophyllis, maculis minutis, conspicuis,

elongatis, usque 1 mm. longis, sparsis insidentibus, singulis v. paucis in quoque macula,

ellipticis v. linearibus, cinnamomeis, usque 400 y longis, diu epidermide bullata tectis.

Uredosporis ellipsoideis, ovatis v. subglobosis, 20-30 x 15-20 4: episporio fere hyalino,

1-1-5 » crasso, remote subtiliterque echinulato, poris germ. obscuris, ut videtur 3, equatoriali-

bus.

Soris teleutosporiferis inter soros uredosporiferos sparsis, hypophyllis, minutis, 150-200 yu

longis, ellipticis, atro-brunneis, epidermide tectis, compactis. Teleutosporis plerumque

clavatis, rarius cylindraceis y. fusoideis, 30-45 x 10-22-5 yw, cellula superiore pallide

auranteo-brunnea usque cinnamomea, inferiore pallidiore, apice rotundatis, truncatis v.

plerumque oblique acuminatis v. melius in rostrum productis, medio non vel vix con-

strictis, basi plerumque pedicellum versus attenuatis, interdum subrotundatis; episporio

leve, tenue, 1-1-5 w crasso, ad apicem leniter incrassato 3-5-8 4; pedicello persistente,

pallide flavo-brunneolo, 5-6 1 crasso et usque 23 » longo.

496

Hab. in foliis Scleriae Dregeanae Kunth, Donkerpoort, leg. Doidge et Bottomley,

29982.

Fig. 12.—Puccimia scleriae-dregeanae, teleutospores.

The loculi separate readily at the septum and fall apart. This rust differs widely from

Rostrupia scleriae Pazschke in the size of the teleutospores and sori, and in the absence of

2-septate spores. It approaches more nearly to Puccinia sclervicola Arthur, but there has

been no opportunity of examining this species. The description differs in several details

(Mycologia 7, 1915, p. 232) particularly in the almost hyaline uredospores.

Puccinia amphilophidis nov. spec.

Soris uredosporiferis hypophyllis, maculis brunneolis indeterminatis insidentibus,

minutis, usque ? mm. longis, oblongis, sparsis v. gregariis, mox nudis, epidermide lacerata

cinctis, cinnamomeis. Uredosporis plerumque ovatis, rarius ellipsoideis v. subglobosis,

echinulatis, brunneis, 25-35 « 17-5-20 1; episporio 1-5-2-5 p crasso, ad apicem nonnun-

quam leniter incrassato usque 4 pw, poris germinationis 4-6 equatorialibus praedito. Para-

physibus numerosis, capitatis v. clavatis, flavo-brunneolis, rarius hyalinis, rectis v. leniter

curvatis, 45-85 » longis, ad apicem 12-520 y latis, membrana ca 2-2-5 crassa, ad apicem

usque 7-5 # incrassata, lumine haud raro fere nullo.

Soris teleutosporiferis consimilis, atro-brunneis. Teleutosporis subglobosis v. oblongo-

ellipsoideis, levibus, castaneo-brunneis, 30-40 x 22-25 , apice late rotundatis, basi rotun-

datis v. subattenuatis, medio vix constrictis ; episporio 2-5-3 y crasso, apice leniter incrassato,

usque 6:5 4; pedicello persistente, crassiusculo, apice 5-7-5 yw crasso, usque 115 «4 longo,

ubique brunneolo, interdum oblique, rarius lateraliter inserto.

Big. 138.—Puccinia amphilophidis, teleutospores, paraphyses and uredospores.

497

Hab. in foliis Amphilophidis insculptae Stapf., Schagen, leg. Liebenberg, 26024; II.

Godwan River. leg. Liebenberg, 26056 ; Nelspruit leg. Liebenberg, 26029 ; Warmbaths, leg.

Curson, 26394.

This rust resembles Puccinia versicolor and P. erythraeénsis, which also occur on

Andropogoneae, in the form of its teleutospores, but differs in having brown pedicels. The

uredospores differ from those of either of the two species mentioned in having 4-6, fairly

conspicuous, sub-papillate, equatorial germ pores.

Puccinia eucomis nov. spec.

Soris uredosporiferis amphigenis, dense seriatim dispositis, oblongis, }-1 mm. longis

saepe autem confluendo longioribus, aurantiacis, mox nudis, pulverulentis, epidermide

fissa cinctis. Uredosporis plerumque ovatis, rarius subglobosis v. oblongis, 25-35 pp x

18-22-52; episporio subhyalino, subtiliter verruculoso, 2-2-5 crasso, ad apicem incrassato

6-7-5 w, rarius usque 9 4; poris germinationis obscuris ; paraphysibus nullis.

big. 14.--Pucecinia eucomis, teleutospores and uredospores.

Soris teleutosporiferis conformibus, atro-brunneis, subpulvinatis. Teleutosporis

castaneis, levibus, plerumque late ellipsoideis, rarius subglobosis v. ovatis, apice late rotun-

datis, medio haud constrictis, basi rotundatis v. interdum sub-attenuatis, 35-47-5 x

22-5-30 4; episporio 2-5-3 4 crasso, apice incrassato usque 9 y, plerumque ca 7-5 p

pedicello crasso, persistente, pallide brunneolo, saepe oblique v. transverse inserto, 7-5-8 2

crasso et usque 100 y longo.

Hab. in foliis vaginisque Andropogonis eucomis Nees, Donkerpoort, leg. Doidge et

Bottomley, 30129.

The uredo has also been collected on Andropogon huillensis Rendl. Donkerpoort, Doidge

and Bottomlcy, 30132.

One 3-celled teleutospore was seen. The contents of the uredospores are bright orange,

and this gives the orange colour to the uredo-sori.

The teleutospores of the known species of Puccinia on South African Andropogoneae

resemble one another very closely ; the uredoform is more distinctive, and the species of

rust on these hosts may readily be distinguished when uredo-spores are present. The

distinguishing characters are indicated in the following key :—

A. Paraphyses present, germ pores conspicuous.

a. Uredo-spore uniformly thin walled, brown, with

5-8, rather conspicuous, scattered germ pores.... Puccinia erythraeénsis.

498

0. Wall of uredo-spore thickened at the apex.

x. Apex slightly thickened, up to 4 , spore

20-35 X 17-5-20 pw, 4-6, equatorial germ

DOLCSHR Fy Picts eke Phat: Ae te atehen et eee Puccima amphilophidis.

xx. Apex more definitely thickened, up to 8 1,

spore 30-42-5 <x 22-5-30 p, 2-3 equatorial

SELM POLES... < sho, ee SER a Uredo schizachyrit.

B. Paraphyses wanting.

a. Germ pores obscure.

x. Wall of uredo-spore thickened irregularly,

CAN, HEINE Sie ovo awessuooe docnbesoeo Puccima versicolor.

xx. Wall thickened at the apex, upto9y.... Puccinia eucomis.

b. Germ pores evident, 4-7, scattered.............. Uromyces Clignyt.

Puccinia Bottomleyae nov. spec.

Soris uredosporiferis amphigenis, plerumque epiphyllis, oblongis, cinnamomeis,

3-2 mm. longis, mox nudis, epidermide fissa cinctis, pulverulentis. Uredosporis saepe

etiam teleutosporis immixtis, sub-globosis, ovatis v. ellipsoideis, brunneis, 22-30 x 20-25 p,

laxe minuteque aculeatis; episporio 1-5-2 ™ crasso, poris germinationis 4-7, sparsis

instructis.

Vig. 15.—Puccima Bottomleyae, teleutospores, uredospores and paraphyses.

Soris teleutosporiferis conformibus, atro-brunneis. 'Teleutosporis subglobosis, oblongis

y. ovatis, castaneo-brunneis, levibus, apice rotundatis, medio haud constrictis, basi rotun-

datis v. subattenuatis, 27-5-40 ~ 20-25 p; episporio 2-2-5 1 crasso, apice leniter incrassato,

usque 6 #; pedicello persistente, apice brunneolo, 6-6-5 p crasso, 85-112°5 yw longo,

nonnumquam oblique inserto. Mesosporis paucis, subglobosis v. ovatis, 25-27-5 x 22-5 y.

Paraphysibus numerosis, capitatis, brunneolis, 50-100 y longis; capite 15-20 yp latis,

membrana leve, 1-1-5 y crassa, apice valde incrassata, usque 10 yp.

Hab. in foliis Aristidae Welwitschiae Rendl. et A. barbicollis Trin. et Rupr., Derdepoort,

leg. Doidge et Bottomley, 29793 et 29795. ~

Aristida adscensionis Linn., Buffelspoort, Marikana, leg. Doidge, 29991 ; Derdepoort,

leg. Doidge et Bottomley, 29790. ms Be

Aristida sp., Derdepoort, leg. Doidge et Bottomley, 29789.

499

Puccinia digitariae Pole Hvans.

in Ann. Bolus Herb. 2 (1917), 111; Doidge in Bothalia 2 (1927), 124.

Syn. Uredo digitariaecola Thuem. in Myc. Univ. No. 2041 (1882); Syd. Monogr. Ured. 4

(1924), 604.

The illustration of this rust in Bothalia (loc. cit.) is misleading. The teleutospores are

closely crowded in the minute, compact sori and are consequently much more irregular in

form than the drawing indicates, especially at the apex. One 3-celled spore was seen, and

in several teleutospores, germ pores at the apex and just below the septum were observed.

The teleutospores are delicate and very easily crushed.

Noir

Fig. 16.—Puccinia digitariae, teleutospores.

In addition to the hosts previously recorded, i.e. Digitaria debilis, D. eriantha and

D. Smutsii, this rust has been found on :—

Digitaria horizontalis Willd., Nelspruit, Inebenberg, 26649.

Digitaria Pentzii Stent, Donkerpoort, Doidge and Bottomley, 29750, Ashbury, Doidge

and Bottomley, 29783.

Digitaria setivalva Stent, Salisbury, Hopkins, Rhod. Myce. Herb., 2602 and 2071.

The Rhodesian collection [Hopkins in Trans. Rhod. Sc. Ass. 35 (1938), 106], is designated

“ Puceinia digitariae affinis”. From an examination of a portion of this material kindly

supplied by Dr. Hopkins, it appears that this rust is typical Puccinia digitariae.

Puccinia eragrostidis-chalcanthae nov. spec.

Soris teleutosporiferis epiphyllis, oblongis, 0-5-2-5 mm. longis, sparsis v. aggregatis

et confluendo longioribus, mox nudis, brunneis, pulverulentis. Teleutosporis subglobosis

v. oblongis, aureo-brunneis, 26-27-5 x 17-5-25 p, apice late rotundatis v. obtuse conicis,

medio haud constrictis, basi rotundatis; episporio leve, 3-4 » crasso, apice plerumque

incrassato, usque 8 4; pedicello tenue, hyalino, persistente, usque 75 » longo, ad apicem

5 # erasso, deorsum attenuato.

BOF OP

Fig. 17.—Puccinia eragrostidis-chalcanthae, teleutospores.

Hab. in foliis Hragrostidis chalcanthae Trin., Donkerpoort, leg, Doidge et Bottomley,

29760.

Through the courtesy of Dr. Kern and of Dr. Stevenson of the Division of Mycology

and Plant Disease Survey, Washington, a portion of the type collection of Puccinia

eragrostidicola Kern, Thurston and Whetzel (Mycologia 25, 1933, p. 469) was available for

study. The South African rust described above is near this species, but the teleutospores

500

are much more variable in form and are thicker walled ; the pedicel is hyaline and inserted

directly, whereas in P. eragrostidicola it 1s tinted brown and is often oblique. Puccinia

eragrostidis Petch (Ann. Roy. Bot. Gard. Peradeniya 6, 1917, pp. 209, 216) has narrower,

ovoid or oblong-ovoid teleutospores, 20-32 x 15-17 yw, with hyaline, oblique pedicel.

Puccinia eragrostidis-superbae nov. spec.

Soris pu ceospantcus amphigenis, plerumque hypophyllis, ochraceis, ellipticis v. oblongis,

minutis, usque } mm. longis, epidermide fissa cinctis. Uredosporis ovatis, fusco- brunneis,

ad apicem obscurioribus, "27-32 x 17-25 ; episporio 1-5-2 ft crasso, ad apicem saepe

leniter incrassato, usque 4 “4, sparse valideque echinulato, poris germinationis 4-6, con-

spicuis, equatorialibus praedito. Paraphysibus satis numerosis, clavatis, rectis v. leniter

curvatis, 25-65 x 8-17-5 4; membrana leve, 1-5-4 yu crasso, ad apicem incrassato usque

12-5 uw.

Fig. 18.—Puccinia eragrostidis-superbac, teleutospores, paraphyses and uredospores.

Soris teleutosporiferis amphigenis, plerumque hypophyllis, atro-brunneis, oblongis,

3-1 mm. longis, saepe confluendo longioribus, epidermide fissa cinctis. Teleutosporis sub-

globosis, ovatis v. late ellipsoideis, utrinque rotundatis, medio non vel vix constrictis,

35-45 x 25-30 4; episporio leve, castaneo-brunneo, cellula inferiore 3-4 y crasso, cellula

superiore 4-5 4 crasso, ad apicem incrassato 8-10 he pedicello persistente crasso, ad apicem

8-10 ps crasso, beuuncolos usque 112 » longo, membrana crassa.

Hab. in folius Hragrostidis swperbae Peyr., Derdepoort, leg, Doidge et Bottomley,

29811; (II) in foliis Hragrostidis happulae Nees var. divaricatae Stapf. Derdepoort, leg.

Doidge et Bottomley, 29813.

Puccinia miscanthidii nov. spec.

Soris uredosporiferis hypophyllis, minutis, sparsis, ca. } mm. lengis, pulverulentis,

luteo-aurantaceis v. cinnamomeis. Uredosporis ovatis, oblongis v. subglobosis, pallide

aurantiacis deinde brunneis, 24-30 x 17-5-25 1; episporio 1-1- 5 J Crasso, ‘subtiliter verru-

culoso, poris germinationis obscuris, ut videtur minutis sparsis.

Soris teleutosporiferis amphigenis, plerumque hypophyllis, sparsis, aggregatis v.

seriatim dispositis, {-2 mm. longis, saepe confluendo longioribus, pulvinatis, atro-brunneis,

epidermide fissa cinctis. Teleutosporis castaneo-brunneis, apice saepe obscurioribus,

oblongis, clavatis, subglobosis v. irregularibus, apice rotundatis v. conicis, rarius truncatis,

medio non vel vix constrictis, basi rotundatis v. subattenuatis, 30-50 x 20-30 4; episporio

leve. 2-2-5 pw crasso, apice incrassato usque 8 4; pedicello crassiusculo, persistente, ubique

brunneolo, ca 10 » crasso, usque 87-5 longo, saepe oblique, interdum lateraliter inserto.

501

Fig. 19.—Puccinia miscanthidii, teleutospores.

Hab. in folus Miscanthidvi sorghi Rich., Lundie’s Hill, Umkomaas Valley, leg, Doidge

30104; in foliis Miscanthidii juncei Stapf., Trigaartspoort, leg. Doidge et Bottomley,

30105.

This species is closely related to Puccinia imperatae Doidge on Imperata cylindrica

Beauv. (Bothalia 2, 1928, p. 474) from which it differs in the colour and form of the teleutos-

pores ; they are darker brown, more irregular in form and thinner walled ; the pedicel is

tinted brown throughout, and is frequently inserted obliquely, occasionally transversely.

epee

Fig. 20.—Puccinia imperatae, teleutospores.

Puccinia pogonarthriae Hopkins.

in Trans. Rhod. Sc. Soc. 35 (1938), 106.

Syn. Uredo pogonarthriae Syd. in Ann. Mye. 10 (1912), 35; in Bothalia 2 (1927), 198.

Uredo-sori amphigenous, but mostly hypophyllous, scattered, oblong, $-2 mm. long,

surrounded by the torn epidermis, pulverulent, rusty brown. Uredospores globose or

subglobose, light brown, 22-26 « 18-22 1; epispore light golden brown, 1-5—-2-5 y» thick,

delicately echinulate and with about six scattered germ pores.

Teleutosori epiphyllous, scattered, oblong, }-2 mm. long, dark brown, surrounded by

the torn epidermis, soon. becoming pulverulent. Teleutospores subglobose or oblong, broadly

rounded or obtusely conical at the apex, not constricted at the septum or very shehtly So,

rounded at the base, golden brown, 25-40 x 17-25 4; epispore smooth, 2-5-4-5 yw thick,

often slightly thickened, up to 8 », at the apex ; germ pore apical or slightly oblique in the

upper cell, one-half to one-third of the distance from septum to base in the lower cell ;

pedicel sub-persistent, hyaline, 5 » thick at the apex, and up 50 yu long.

Hab. on leaves of Pogonarthria squarrosa (Light.) Pilg., Donkerpoort, Dodge and

Bottomley, 29757.

Fig. 21.—Puccinia pogonarthriae, teleutospores.

A portion of the type collection was made available for comparison through the courtesy

of Dr. Hopkins. The type was collected on the same host at Marandellas (Rhod. Govt.

Myc. Herb., 2163) ; it is evidently the same species as the rust collected at Donkerpoort

but the teleutospores of the latter are more variable in form. The description is amended

accordingly. The uredo described by Sydow (loc. cit.) on Pogonarthria squarrosa, was found

on the same leaves.

Puccinia luxuriosa Syd.

in Monogr. Ured. 1 (1904), 812.

Syn. Puccinia tosta var. luauriosa Arth. in Bull. Torr. Bot. Club 29 (1902), 229.

Teleuto-sori amphigenous, scattered or in groups, oblong or linear, often confluent and

forming striae up to 6 mm. long, rather compact, black. Uredospores mixed with the

teleutospores, globose or subglobose, verrucoso-echinulate, yellow brown, 22-30 diam. ;

epispore about 2 » thick, with six or more scattered germ pores.

Tas

Fig. 22.—Puccinia luxwriosa, teleutospores.

Teleutospores ellipsoid or oblong, rounded at the apex, usually slightly constricted at

the septum, usually rounded at the base, brown, 36-56 x 20-28 4; epispore smooth,

3-5 pw thick, thickened at the apex, up to 10 »; pedicel stout, persistent, brown at the

apex and 7-8 » thick, up to 110 » long.

Hab. on leaves of Sporobolus pectinatus Hack., Donkerpoort, Doidge and Bottomley,

29726.

On Sporobolus capensis (Willd.) Kunth, Donkerpoort, Doidge and Bottomley, 29758.

This rust was originally described on Sporobolus aeroides in Oregon and Montana, North

America ; the South African rust agrees well with the description of the American fungus.

Fig. 23.—Puccinia tosta, teleutospores.

Puccinia luxuriosa differs from P. tosta Arthur previously recorded on Sporobolus

fimbriatus Nees (Bothalia 2, 1928, p. 474) in the larger and much more conspicuous sori,

the echinulate uredospores, and the definitely larger teleutospores. The teleutospores of

both species are illustrated for comparison.

Puccinia tristachyae Doidge

in Bothalia 2 (1927), 132.

The type specimen of this rust was collected at Kaalfontein on Tristachya Rehmanni

HHack., and the teleutospores were described as follows: “ Teleutospores clavate, ellipsoid

or oblong, upper cell usually shorter than the lower, rounded or truncate at the apex, rarely

subacute, more or less constricted at the septum, attenuate or rounded at the base, golden

brown, 40-60 x 16-23 4; epispore smooth, 1-5-2 y thick in the lower cell, 2-5-3 » thick

in the upper, rarely very slightly thickened, up to 4 4, at the apex; pedicel persistent,

rather stout, ight brown, up to 45 # long and 6-5 w thick.” The first four spores in the

figure are typical of those found in the type collection. (See also Bothalia loc. cit.)

Rust pustules on a recent collection of Tristachya hispida K. Schum. are apparently

those of the same fungus, but the teleutospores are much more variable than those of the

type specimen. A large proportion are typical in form, but comparatively few are 45-55 yu

long ; the majority are smaller. A large number are irregular in shape, as if through com-

pression ; the pedicel is often inserted obliquely and rarely transversely. Mesospores are

fairly numerous.

Fig. 24.—Puccinia tristachyae, teleutospores.

There is no doubt that the rust on T'ristachya hispida is a more variable form of Puccinia

tristachyae, and the description must be amended to include the smaller and more irregular

teleutospores found in this collection. It will then read: Teleutospores clavate, ellipsoid,

oblong or irregular ; upper cell shorter than the lower, or cells sub-equal ; apex rounded,

truncate or bluntly conical, in the latter case often oblique ; more or less constricted at the

septum, attenuate or rounded at the base, 30-60 « 15-30 4; a large proportion of the spores

do not exceed 40 » in length; pedicel as previously described, but occasionally oblique

and rarely transverse. Mesospores fairly numerous, clavate or cylindrical, 30-53 x

13-17-5 p.

Hab. on leaves Tristachya hispida K. Schum., Donkerpoort, Doidge and Bottomley,

30133.

504

Ravenelia Evansii Syd.

in Ann. Mye. 10 (1912), 440; Monogr. Ured. 3 (1915), 234; Doidge in Bothalia 2

. (1927), 144.

Uredo-sori amphigenous, mostly hypophyllous, also on the petioles, subepidermal, minute,

round, scattered or in small groups (usually 2-3), 0-1-0-25 mm. diam., surrounded by the

torn epidermis, sub-pulverulent, cinnamon brown, on the same leaflets as the teleuto-sori.

(Teleuto-sori are sometimes found on the under side of the leaf, they are not exclusively

epiphyllous as the original description might seem to indicate.) Uredospores ellipsoid or

ovate, pale yellowish brown, 20-30 « 12-5-15 yw, the length being much more variable

than the breadth ; epispore 1-5-2 » thick, occasionally thickened at the apex, up to 5 p,

rather closely verruculose-echinulate and with 4 equatorial germ pores. Paraphyses

numerous, capitate, yellowish or light brown at the apex, hyaline below, 37-5-50 y long,

12-5-17-5 yw broad.

Hab. on leaves of Acacia robusta Burch., Umzinto, Natal, McClean, 30124.

On Acacia Gerrard: Benth., Verulam, Natal, Halse, 30127.

The type specimen described by Sydow (loc. cit.) showed only the teleuto-sori ; the

aecidium was subsequently described in Bothalia. A recent collection from the Natal

coast has numerous uredo-sori and a few teleuto-sorl on the leaves. Uredo- and teleuto-

sorl were also found on Acacia Gerrardi, which is a new host for this rust.

Ravenelia Halsei nov. spec.

Soris teleutosporiferis amphigenis et petiolicolis, plerumque hypophyllis, sparsis,

minutis, rotundatis v. ellipticis, 120-300 » long, cuticula rupta cinctis, obscure brunneis,

Capitulis teleutosporarum ‘leniter convexis, ambitu orbicularibus v. suborbicularibus,

subtus leniter concavis, castaneo-brunneis, levis, 80-112 » diam., ex 9-11 sporis in omni

Fig. 25.—Ravenelia Halsei, teleutospore heads.

directione compositis; sporis singulis continuis, recte v. oblique cuneatis vel oblongis,

25-30 longis, 10-15 y latis, ad apicem incrassatis (5-6 4); cystidiis eodem numero quo

sporis marginalibus, capitulis dense adpressis, in aqua intumescentibus et tandem ruptis ;

pedicello brevi deciduo ex hyphis paucis composito, hyalino.

Hab. in foliis petiolisque Acaciae ataxacanihae DC., Ndwedwe, Natal, leg. Halse, 30117.

Ravenelia modesta nov. spec.

Soris teleutosporiferis subepidermicis, amphigenis v. petiolicolis, plerumque hypo-

phyllis, sparsis, minutis, brunneis, 250-300 » diam., Capitulis teleutosporarum convexis,

subtus leniter concavis, ambitu orbicularibus, suborbicularibus v. irregularibus, 75-105

diam., aureo-brunneis, ex sporis 5-6 in omini directione compositis ; sporis omnibus papillis

4-7 (plerumque 5) subhyalinis v. brunneolis, 3-5 y longis obsitis ; sporis singulis continuis,

cuneatis, 27-5-45 longis, 13-18 y latis, ad apicem incrassatis ; cystidiis eodem numero

quo sporis marginalibus, in aqua intumescentibus ; pedicello composito brevi, hyalino.

505

Hab. in foliis Acaciae stoloniferae Burch., Pienaar’s River, leg. Doidge et Bottomley,

30110.

Fig. 26.—Ravenelia modesta, teleutospore heads.

This species is near Ravenelia pretoriensis Syd., but it differs in several particulars.

The heads are more definitely convex; the papillae are longer and more conspicuous,

acuminate rather than verruciform, straight, oblique or somewhat curved. The sori are

very minute ; there is frequently only one and rarely more than 2 or 3 on a single leaflet.

Ravenelia pretoriensis Syd.

in Ann. Mye. 10 (1912), 441 ; Doidge in Bothalia 2 (1927), 146.

The host of this rust has been wrongly identified ; it is definitely not Acacia horrida

Willd. (= Acacia harroo). The branch and leaf characters agree well with those of Acacia

pennata Willd. and it may be this species ; an exact identification is not possible without

the pods, which are absent from the material. Further collections will therefore be necessary

before the identity of the host can be determined satisfactorily.

In Bothalia (loc. cit.) the drawing of Ravenalia pretoriensis, p. 147, is erroneously labelled

R. Pienaarii, and under that of R. Pienaarvi (p. 146) the legend reads R. pretoriensis.

Ravenelia transvaalensis nov. spec.

Soris teleutosporiferis amphigenis et petiolicolis; epiphyllis conspicuis, sparsis v.

paucis ageregatis, atro-brunneis, rotundatis v. irregularibus, usque 2 mm. diam., epidermide

lacerata cinctis; hypophyllis petiolisque inconspicuis, effusis, indeterminatis. Capitulis

teleutosporarum convexis, subtus leniter concavis, ambitu orbicularibus, suborbicularibus

y. irregularibus, castaneo-brunneis, levibus, 75-100 » diam., ex sporis 5-6 in omni directione

composito ;_ sporis singulis 30-35 » longis, 15-17-5 y latis, ad apicem incrassatis (ca 6 1) ;

cystidiis numerosis, ut videtur eodem numero quo sporis singulis, ovatis, dependentibus,

in aqua intumescentibus ; pedicello non viso.

Hab. in foliis Acaciae detinentis Burch., prope Pienaar’s River, leg. Mogg, 27382.

So?

Fig. 27.—Ravenelia transvaalensis, teleutospore heads.

506

Uredo Hyperici-leucoptychoides nov. spec.

Soris hypophyllis, maculis flavis insidentibus, sparsis v. aggregatis, minutissimis,

usque 0-2 mm. diam., aurantiacis, mox nudis, ab cellulis epidermidis peridiiformiter cinctis.

Paraphysibus numerosissimis, periphicis, clavatis v. clavato-capitatis, saepe irregularibus,

50-100 w longis, uncinatis, geniculato-incurvatis v. rectis, hyalinis, ad apicem 10-15 pu

erassis, membrana 2-4-5 jz crassa, superne plus minusve incrassata (usque 6-5 jw) lumine

Fig. 28.—Uredo hyperici-leucoptychoides, spores and paraphyses.

haud raro fere nullo. Sporis subglobosis, ovatis v. ellipsoideis, pallide flavo-brunneolis,

17-25 x 15-20 p, subtiliter laxeque echinulatis, episporio ca 1 yz crasso, poris germinationis

haud conspicuis (? 4-7 sparsis).

Hab. in folus Hyperict leucoptychodis Steud., Woodbush, leg Doidge, 28467.

This rust closely resembles Uredo Hyperici-mysorensis Petch, which occurs in Ceylon,

but the paraphyses are longer and stouter. It differs widely from Uredo Hyperici-Schim-

pert P. Henn. described on Hypericum Schimperi from Central Africa.

Uredo combreticola nov. spec.

Soris amphigenis, sine maculis, per totam folii superficiem plus minus dense sparsis,

rotundatis, ellipticis vy. irregularibus, minutis, ca } mm. diam., pallide cinnamomeis, epi-

dermide diu tectis deinde poro centrale plus minus late apertis. ‘Sporis solitarie in pedicellis

ortis, ovatis, subglobosis v. ellipsoideis, interdum polygonalibus, hyalinis v. pallide flavo-

brunneolis, remotiuscule subtiliterque verruculosis, 17-23 x 10-15 4; episporio 1-1-5

erasso, poris germinationis haud corspicuis.

Hab. in foliis Combreti Zeyhert Sond., Nelspruit, leg. Liebenberg, 26038.

This rust differs from Uvredo longaensis P. Henn. occurring on Combretum Baunii on

the Longa River, in the character and distribution of the sori, which are more numerous

on the upper than on the lower side of the leaf, and are not on leaf spots. The spores are

not in chains as in U. longaensis.

507

Uredo caricis-petitianae nov. spec.

Soris hypophyllis, maculis brunneolis indeterminatis insidentibus, plus minus aggregatis

v. seriatim dispositis, rotundatis v. ellipticis, usque $ mm. longis, cimnamomeis, epidermide

fissa cinctis. Sporis ellipsoideis, ovatis v. subglobosis, pallide flavobrunneis, laxe ver-

ruculosis, 25-30 x 15-22-5 m; episporio 2-2-5 y crasso, poris germinationis 2-3 equa-

torialibus praedito.

Fig. 29.—Uredo caricis-petitianae, spores.

Hab. in foliis Cariezs petitianae A. Rich, in silvis Xumeni, Donnybrook, leg. Morgan et

Doidge, 29830, 30107.

The sori are found almost exclusively near the tips of the long leaves; they are on

irregular, brown blotches, which are small at first, but increase in size, and finally involve

the whole upper portion of the leaf, which becomes brown and dead. Collections were made

in winter and in early spring, but no teleuto-sori were found.

Uredo scirpi-corymbosi nov. spec.

Soris calamicolis, maculis sparsis ellipticis 1-5-7 mm. longis, 1-4 mm. latis, ferrugineo-

brunneis, tandem plus minus confluentibus insidentibus, solitariis v. paucis parallele aggre-

gatis, linearibus, 2-5 yp longis, diutissime epidermide elevata tectis. Sporis ovatis, ellip-

soideis, subglobosis v. pyriformibus, pallide flavo-brunneis, subtiliter sparseque echinulatis,

27-36 x 20-25 #4; episporio tenue ca | y crasso, poris germinationis obscuris (? 3-4 equa-

torialibus).

Fig. 30.—Uredo scripi-corymbosi, spores.

Hab. in calamis Scirpi corymbosi Roth., Donkerpoort, leg. Doidge et Bottomely, 29986 ;

Skinner’s Court, leg. Doidge, 23457 ; Onderstepoort, leg, Pole Evans, 1314; Debbe’s Ravine,

leg. Bottomley, 25332.

508

Uredo schizachyrii nov. spec.

Soris amphigenis, minutis, oblongis, ca } mm. longis, epidermide fissa cinctis. Sporis

ovatis v. ellipsoideis, apice rotundatis v. obtuse conicis, castaneis, 30-42-5 x 22-5-30 yp,

crasse sparseque echinulatis ; episporio 1-5-2 y crasso, apice incrassato usque 8 4, poris

germinationis 2-3 equatorialibus praedito. Paraphysibus numerosis, plerumque capitatis

interdum clavatis, 35-62-5 yw longis, ad apicem 12-5-17-5 yw crassis; capite subgloboso

brunneolo, membrana leve, valde incrassata (7-5-15 ) et lamellata.

Fig. 31.—Uredo schizachyru, spores and paraphyses.

Hab. in foliis Schizachyrti senuberbis Nees, Donkerpoort, leg. Doidge et Bottomley,

29766.

Uromyces limonii (DC.) Lev.

in, Dict. d’Hist. Nat. Art. Uredinees (1840) 19; Syd. in Monogr. Ured. 1 (1904) 41.

Syn. Puccima limon DC. FI france. 2 (1805) 595 et Syn., p. 45.

Aecidium statices Desm. Plant. crypt. de France no. 132.

Aecidium limon Duby Bot. Gall. 2 (1830) 904.

Caeoma staticis Rudolphi in Linnaea 4 (1829) 510.

Uredo limon Duby Bot. Gall. 2 (1830) 1897.

(Aecidia amphigenous, often on brown or reddish leaf spots, in round groups, or

elongated to 5 mm. long along the veins, usually briefly cylindrical, white, with lacerated

margin. Aecidiospores angular-globose or ellipsoid, closely and minutely verruculose

yellowish, 21-32 x 18-26 yp.)

Uredo-sori amphigenous, scattered, usually round, oblong on, the stems, long covered

by the epidermis, at length naked, pulverulent, cinnamon brown. Uredospores globose,

subglobose, ovate, ellipsoid or oblong, densely verruculose, yellow brown, 22-32 x 20-28 p;

epispore 1-5-2-5 » thick, germ pores 2-3, scattered.

Teleuto-sori on the leaves and stems ; on the leaves amphigenous, but usually developing

more freely on the under surface, scattered or in concentric rings, round or oblong, remaining

for some time covered by the blistered epidermis, which finally ruptures and surrounds the

509

sorus ; sori pulvinate, dark brown or almost black. Teleutospores golden brown, to chestnut

brown, subglobose, oblong or clavate, apex broadly rounded or obtusely conical, direct or

oblique, base rounded or attenuate, 24-50 x 14-25 ; epispore smooth, 1-5-2 y thick,

thickened at the apex up to 10 #7; pedicel stout, persistent, tinted brown, 5-6 y thick at

the apex and up to 88 » long.

Fig. 32.—Uromyces limonii, teleutospores.

f=) >)

Hab. on leaves of Limonium latifolium Kuntze (= Statice latifoa Sm.), Pretoria,

Taylor, 23819 ; Hartebeestpoort, Bottomley.

This rust has not previously been recorded as occurring in South Africa, and has only

recently made its appearance in gardens where flowers are grown on a commercial scale.

The leaves examined showed a gross infection. Teleutosori were abundant, a few uredo-

spores being found mixed with the teleutospores. No aecidia were seen, the description

quoted being that given by Sydow (loc. cit.).

Uromyces Strauchii Doidge

in Bothalia 2 (1928) 473.

on leaves of Cluytia daphnoides, Alexandria, Dodge, 22370.

9000

Fig. 33.—Uromyces Strauchii, teleutospores.

No illustration of this species was published with the original description. Drawings

of the teleutospores are reproduced in the accompanying text figure, chiefly to indicate the

nature of the sculpturing of the epidermis.

510

Uromyces antholyzae Syd.

in Ann. Myce. 2 (1904) 27; Monogr. Ured. 2 (1910) 252.

Uredo-sori amphigenous, not on leaf spots, or on vaguely discoloured areas of the leaf,

round or transversely oblong, scattered or in transverse groups between the veins; or

caulicolous, scattered or in groups with the longer diameter perpendicular, in groups roughly

oval in outline and up to 5 mm. long; sori small, about } mm. diam., long covered by the

epidermis. Uredospores globose or subglobose, rarely ellipsoid, almost hyaline, 17-5-23

diam., minutely verruculose ; epispore 1-1-5 thick, germ pores obscure.

OVO

Fig. 34.—Uromyces antholyzae, teleutospores.

Teleuto-sori amphigenous, similar to the uredo-sori but dark brown. Teleutospores

subglobose, ovate or oblong, often irregular by compression, brown, 20-25 x 15-20 m,

apex rounded, flattened or obtusely conical ; epispore smooth, 1-5-2 b thick, thickened at

the apex, 4-7 4; pedicel persistent, rather stout, hght brown, 5-6-5 mu thick and up to

32 yw long.

Hab, on leaves and stems of Antholyza nervosa Thun. (= Anapalina revoluta N. BH.

Br.) Ruytersbosch, Mossel Bay, Gemmel (O.F.S. Herb. no. 8053) 30085.

The original description, gives wider limits to the size of the teleutospores, otherwise

the South African fungus agrees in every detail with the rust collected by Schimper on

Antholyza abyssinica in, Abyssinia.

This rust has been recorded on the same host by Verwoerd [in Union Department of

Agriculture, Bull. 88 (1929) p. 5] from Stellenbosch, Newlands and Knysna.

Uromyces massoniae nov. spec.

Aecidiis amphigenis, maculis indeterminatis insidentibus, im greges rotundatos, v.

elongatos usque | cm. diam. plus minus dense dispositis, cupulatis, flavidis, margine revoluto

inciso, 300-450 » diam.; cellulis peridie in series regulares dispositis, firme conjunctis,

subrhomboideis vel penta—v. hexagonis, 25-37-5 ~« 15-25 pw, pariete exteriore striato

7-10 2 crasso, interiore verrucoso 4-5 y, crasso. Sporis subglobosis vel ellipsoideis, plerumque

angulatis, 2227-5 x 15-22 p, dense minuteque verruculosis ; episporio 1-2 y crasso.

Soris uredosporiferis amphigenis, plerumque hypophyllis, sparsis, rotunddatis v. irregu-

lariter rotundatis, epidermide fissa cinctis, pulvervulentis, cinnamomeo-brunneis, usque

1 mm. diam. Uredosporis subglobosis, ovatis v. ellipsoideis, 20-27-5 x 20-25 yw, flavis,

echinulatis ; episporio 2-2-5 ym crasso, poris germinationis pluribus (74-6) sparsis parum

manifestis praeditis.

Fig. 35.—Uromyces massoniae, teleutospores.

Soris teleutosporiferis amphigenis circa soros uredosporiferos annulum efficientibus

vel sparsis, atris, minutis, diu epidermide tectis. Teleutosporis e mutua pressione quoad

formam variabilis, subglobosis, ellipsoideis, pyriformibus, saepe angulatis, castaneo-

brunneis, levibus, apice rotundatis, 25-32-5 x 22-5-27-5 #; episporio 3-4-5 yp crasso ;

pedicello subpersistente, 15-25 longo, apice leniter colorato, 5-7-5 y crasso.

Hab. in foliis Massoniae latifoliae Linn. f., Fauresmith, leg. van der Plank, 25447.

Uromyces ertospermi Kalch. et Cke. affinis.

Uromyces Clignyi Pat. et Har.

in Journ. de Bot. 14 (1900) 237; Syd. Monogr. Ured. 3 (1910) 320.

Uredo-sori hypophyllous, scattered, often in series, but rarely confluent, minute, oblong’

up to 1 mm. in length, long covered by the epidermis, yellow brown. Uredospores globose

or subglobose, golden brown, echinulate, 22-30 p diam. ; epispore 2-2-5 thick with 4-7

small, scattered germ pores.

Fig. 36.—Uromyces Clignyi, teleutospores.

512

Teleuto-sori similar in form, dark brown to black. Teleutospores globose or subglobose,

dark brown, becoming almost opaque, 22-32 » diam.; epispore smooth, 4:5-6 y thick,

not thickened at the apex; pedicel hyaline, equalling the spore or longer.

Hab. on leaves of Andropogon amplectens Nees, Donkerpoort, Doidge and Bottomley,

29751 ; Derdepoort, Doidge and Bottomley, 29788.

This species was originally described from tropical Africa on, Andropogon sp., and on

A. multinervis in the region of the river Niger and in Abyssinia. It also occurs in America.

Uromyces Cliqnyz has not previously been recorded on Andropogon spp. South Africa.

Uromyces trichoneurae nov. spec.

Soris uredosporiferis amphigenis, minutis, sparsis. Uredosporis subglobosis, late

ellipsoideis v. ovoideis, saepe subangulatis, flavo-brunneis, 27-5-37-5 x 22-5-27-5 pw;

episporio 1-1-5 « crasso, brunneolo, subtiliter echinulato, poris germinationis 3 equatorialibus

praeditis.

90006

Fig. 37.—Uromyces trichonewrae, teleutospores.

Soris teleutosporiferis plerumque epiphyllis, minutis, usque + mm. longis, interdum

confluendo hinc inde majoribus, usque 3 mm. longis. Teleutosporis quoad formam

variabilis, subglobosis, ovatis, oblongis v. piriformibus, saepe angulatis, 20-35 « 16-19 y.

apice rotundatis, truncatis v. conicis, basi rotundatis v. attenuatis ; episporio leve, 1-1-5 uw

crasso, apice incrassato usque 5 4; pedicello subpersistente, crassiusculo, apice brunneolo,

circ. 5 erasso et usque 35 y longo.

Hab. in foltis Trichoneurae grandiglumis (Rendl.) Stapf et Hubb., Donkerpoort, leg.

Doidge et Bottomley, 29762 ; Derdepoort, leg. Doidge et Bottomley, 29792.

I am indebted to the botanists in the National Herbarium for the revision of the host

plants, and in particular to Miss L. C. Chippendall, who kindly identified a large number

of grasses.

Text figures were drawn to scile, but the illustrations of spores of Puccunia spp. have

been reduced by half as compared with those of species of other genera.

513

A REVISION OF THE SOUTH AFRICAN SPECIES

OF ADENIA.

L. C. C. Liebenberg

INTRODUCTION.

The genus Adenia was founded by Forskal on a plant collected in Arabia and which he

named A. venenata. In the original description, the genus is stated as having 6-merous

flowers. Ascherson, (1)* as pointed out by Engler, first drew attention to the fact that

Adema Forsk. (1775) and Modecca Lam. (1797) were synonymous. Engler (5) states that

tis absolutely certain that Modecca abyssinica Hochst. and Adenia venenata Forsk. are

identical although the former is 5-merous, explaining that it is quite possible that Forskal

had a specimen with 6-merous flowers in front of him as “in the related genus Keramanthus,

6- and 5-merous flowers occur”. Engler further explains (l.c.) that he attempted to separate

the two genera but was unsuccessful. The majority of workers had apparently, as suggested

by Ascherson, overlooked or not recognised this fact.

Smith (17) in 1821 substituted Blepharanthes Sm. for Modecca Lam. Wight and

Arnold (20) in 1834 divided the genus Modecca into two subgenera Microblepharis Wight &

Arn. and Blepharanthes (Sm.) Wight & Arn. Roemer (14) in 1846 divided Modecca into

the genera Microblepharis (W. & A.) Roem. and Erythrocarpus Roem. while Bentham and

Hooker (3), 1867, distinguished 2 genera Ophiocaulon Hook. f. and Modecca Lam. recognising

Clemanthus Klotsch., Paschanthus Burch. and Kolbia Beauv. as synonyms of Modecca.

Baillon (2), m 1888, divided Modecca into three sections, EHumodecca Baill., Ophiocaulon

(Hook. f.) Baill. and Keramanthus (Hook. f.) Baill. thus further enlarging the conception of

the genus by incorporation of the latter two genera. Engler (l.c.) in 1892, added two more

sections to Wight and Arnold’s division of the genus Adenia (in which Modecca was included),

viz. Huadenia Engl. and Hildebrandtiothamnus Engl.

In 1893 Harms (6) wrote: “Von hohem Interesse ist die Anatomie der Gattungen

Adenia, Ophiocaulon, Echinothamnus und Keramanthus. Es scheint mir bei diesen der

anatomische Bau ganz deutlich auf eine enge Zusammengehorigkeit hinzuweisen.”” He (7)

in 1895 subscribed to Engler’s division of the genus Adenia in which he not only included

(like Bentham and Hooker) the genera Clemanthus Klotsch. and Kolbia Beauv., but also

Keramanthus Hook. f. each representing only one species. In 1897 he (8) further enlarged

the genus incorporating 3 genera, viz. Paschanthus Burch. (= Jdggia, Schinz.), Echinotham-

nus Engl. and Ophiocaulon Hook. f. The first 2 monotypic genera were combined, con-

stituting his sect. 1 Paschanthus, while the last, comprising some dozen closely related

species in tropical and southern Africa, constituted his section 2, Ophiocaulon previously

raised to this position by Baillon in 1888, as noted above. In addition to this the section

Hildebrandtiothamnus (originally his sect. 4) was sunk and the genus Keramanthus given

sectional rank as was also previously done by Baillon. De Dalla Torre and Harms (1907)

have upheld this division of Adeniain the 5 sections, Paschanthus, Ophiocrlon, Blepharanthes,

Keramarthus, Microblepharis and Euadenia. In 1921, however, the section Microblepharis

W. & A. was not included by Harms (10) but in 1925 (11) he resuscitated it.

* Figures in parenthesis refer to “Literature Cited” on page 544.

514

None of these sections, as defined by Harms, will allow for the inclusion of A. glauca

and A. spinosa, two apparently very closely related species, differing mainly in the presence

of spines and simple leaves in the one species as opposed to compound leaves in the other

species—both without glands (“ Coronaschuppen ”’, etc.). Harms (11) includes 4. glauca

in Bleparanthes presumably based on Schinz’s description of this species in which he

(Schinz) described the petals as “im Grunde des Receptaculums inseriert’”’ which the

present writer did not find to be the case. (See discussion under A. glauca on p. 523.) The

following is a summary of the synonymy of the genus Adenia Forsk. accepting the con-

ception of Harms [presumably based, to some extent at least, on his extensive studies of

anatomic characters (referred to elsewhere)| and certain other workers mentioned above :—

1775. Adenia Forsk. Fl. Aeg. Arab., p. 77.

1797. Modecca Lam. Encycl. meth. bot. lv., 208.

1807. Kolbia P. Beauv. Fl. d’Oware et Ben., 11.91.

1821. Blepharanthes Smith Gramm. of Bot., 188.

1822. Paschanthus Burch., in Burch. Travels 1, 543.

1846. Microblepharis M. Roem., Synops. 11, 133, 200.

1846. Erythrocarpus M. Roem., Synops. Mon. 11, 204.

1863. Clemanthus Klotsch., Peters Reise Mossamb. Bot., 143.

1867. Ophiocaulon Hook. f., Gen. Plant, 1, 111, 813.

1876. Keramanthus Hook. f., Bot. Mag. T., 6271.

1889. Javgia Schinz., Verh. Bot. Ver. Brand, XXX, 253.

1891. Echinothamnus Enel., Bot. Jahrb. 14, 383.

The genus has attracted attention in South Africa mainly as a result of the poisoning

of human beings from A. digitata. In 1922 poisoning (and one death) of adults was reported

as a result of the chewing of the tubers of A. digitata which were mistaken for that of a

cucurbitaceous plant. Previously Burtt Davy reported death and poisoning of children

as well as suspected poisoning from this species. In 1928 death of a child and poisoning

of others were also ascribed to this species. A study of its poisonous principles at the

Veterinary Research Laboratory, at Onderstepoort,* revealed two toxic principles, hydro-

eyanic acid and a toxalbumin, “ Modeccin”. Steyn (18) reports that hydrocyanic acid

has been found in the fresh leaves of this species as well as that of A. glauca, but in the

latter case not in the “ root ” which he states is edible. Dr. E. E. Galpin reported in Decem-

ber, 1931, that he had observed children eating the fruits of A. glauca, which they said were

very nice. According to Watt and Brandwyk (19), A. senensis, A. gunmifera and A. Kirkii

are used medicinally.

Harvey states that the fruits of A. hastata are edible; Bryant states that A. repanda

is greedily eaten by stock which is also reported for A. hastata; Potts (13) states that

natives, when thirsty, suck the sap of the “tuber” of A. multiflora, but this information

must be incorrect as will be explained further on, and the confusion is very probably attri-

butable to the fact that A. glauca occurs in the same locality and is nontoxic and edible.

Forskal described A. venenata as having poisonous tubers while A. palmata is also said to

be poisonous.

The South African species which represent only a very small percentage (approximately

10 per cent.) of the world’s species have been recorded mainly from the Transvaal. The

genus is largely restricted to the tropics of the old world and from Africa approximately

50 species have been described. It is of more than usual interest as it exhibits some most

interesting plant forms (Plates 1, 2, 3 and 4).

* Green, H. H., and Andrews, W. H., 1923: The toxicity of A. digitata Burtt Davy (Modecea digitata

Harv.) 9th and 10th Rpt. Dir. Vet. Educ. & Res., pp. 381-91. Green, H. H., and Kamerman, P., 1924 :

The protein phytotoxin with special refce. to the new “‘ modeccin”’ Journ, S.A. Chem. Inst. 7, pp. 3-5.

515

There are in South Africa (and South West Africa) the desert-arid region forms,

A. repanda (Burcb.) Engl. and A. Pechuelli (Engl.) Harms, the latter being a plant of very

strange habit. At the other extreme there is the interesting, widely distributed, liana,

A. gummifera (Harv.) Harms, which often grows to enormous size with very long woody

thick stems, up to 2} and 3 inches diameter. Intermediate forms are represented by several

species in South Africa, two types being distinguished, viz. one with a fleshy swollen axis

partly or largely above the soil (Plates 1-4), the other with a tuberlike underground portion

from which the annual branches arise. In the former case the green or above-ground swollen

main axis assumes more or less characteristic shapes in the different species. Of this type

there are only 4 described species, one, A. globosa Engl., occurring in East Africa, the other

3 having been recorded from South Africa and that, with the exception of two specimens

only, from the Transvaal. Judging from the figures of the spinescent A. globosa Engl.,

it would seem that this species is not closely related to any of the 3 South African species

of similar habit, amongst which there is also a spinescent species.

Chu fh & f g

Fig. 1—Variation in petal, gland and coronal processes Fig. 2.—Two petals from two

in two flowers from Obermeyer in Trans. Mus. 29287, 9, different flowers from Liebenberg

belonging to A. digitata (Harv.) Engl. 3366, 9, belonging to A. digitata

(Harv.) Engl.

The Root-stem Relations.

In the absence of anatomic proof the writer has adopted the view that the tuberlike

structure when below ground only (which is normally the case in A. repanda, A. senensis,

A. Wilms, A. digitata and A. hastata) is entirely a root structure except for the attenuated

perennial portion at the top thereof, from which the annual branches arise and which is

formed from accumulated annual growths or from elongation or enlargement of the bud-

producing zone. When the swollen perennial main axis is largely or partly above the

ground (as in the case of A. fruticosa, A. spinosa and A. glauca) then the underground portion

is regarded as root structure and the aboveground part (green portion) as stem structure,

the soil level beng regarded the line of demarcation between the two kinds of structures.

The annual axes or stems are referred to as branches. This characterisation is resorted

to only as a matter of convenience for the description of the species.

516

MorpHotocicaL Nores AND STUDIES.

The Flower.

The classification of certain species of Adenia is not only made more difficult by leaf

heterophylly but also by polymorphism and the presence of male and female flowers on

different individuals (very rarely on the same individual) and to some extent by marked

variations in floral structures, within the species, although on the same individual there

usually appears to be little variation (figs. 1, 2 and 3). It may be pointed out that in the

majority of the species polymorphism is very apparent, the difference in structures of the

flower being marked in the two sexes apart from the obvious differences (abortion of androe-

cium and gynaecium respectively) as a result of the unisexual nature of the flower. This

is particularly characteristic of the petals.

In the literature there are two interpretations of the morphology of the flower parts

and the writer has adopted the interpretation which follows hereunder, in which is included

a description of the main features of the flower.

A ite

——_

aa S

p TSG. is P i p J fe)

Fig. 3.—Variation in petal, gland, coronal processes, ovary and staminode in two flowers A and B

from Liebenberg 3056, °, belonging to A. digitata (Hary.) Engl. A is from the original specimen and

B from a specimen grown in the gardens of the D.P.I. 0, overy; s, staminode; g, ‘uland ; fp, coronal

processes ; p, petal.

Receptacle.

That part of the flower more or less below the “ glands ” (q.v.) and above the articula-

tion with the pedicel. The articulation is never absent.

Calyx tube.

Part of the “ zeceptacle ” of Harms, Engler, etc.

Calyx lobes.

(5). The “sepals” of Harms, Engler, etc.

Corolla.

Petals 5, alternating with the calyx lobes, inserted at the sinuses of the latter or at

varying depths on the calyx tube, depending upon the species.

517

Corona.

The filiform processes which arise from the calyx tube near its base and are arranged

in a sinuate ring or in groups in a circle. “ Korona” (in part) or “ Effigurationen” (in

part) of Harms. This is homologous to the prominent structure in certain other genera

of the family, e.g. Passiflora.

Androecvum.*

In the male flower 5 stamens and in the female flower 5 staminodes either free or other-

wise connate for part of their length at the base. The staminal or staminodal column is

adnate to the receptacle (? and/or calyx tube), at 5 points (always opposite the petals),

so forming 5 depressions or pockets from the base of which arise the “‘ glands ”

Gynuecium.*

In the female a normal stipitate ovary with numerous ovules on 3 placentae (parietal)

and a 3-branched style and pufflike stigma. In the male an abortive structure semiterete

linear-cylindric in shape somewhat tapering upwards, surrounded by the staminal column.

The fruit is a capsule, normally 3-valvate, dehiscent or indehiscent, with anatropous arillate

seeds.

Glands.

(5). The 5 structures which arise from the base of the pockets or depressions and which

“ernate with the petals. “‘ Receptaculumeffigurationen,” ““ Zungenférmige Schuppen,”

yvrona Schuppen,” “ Schuppenformige Effigurationen,”’ ‘“ Outer stamimodes,” “ glands

a aise eens term (gland) is used for convenience, as these structures may actually

ter staminodes ” or even the vestiges of an aborted coronal structure. Absent in

sence

The slender processes (“ corona”’) vary a great deal in the various species, being

rudimentary in certain individuals or certain of the South African species. Their absence,

vogether with the absence of glands in any particular species may perhaps be looked upon

as a sufficiently strong reason to exclude such species from the genus Adenia. These pro-

cesses when numerous or in a continuous circle around the calyx tube very often appear

to arise from the edge, or to be the lacerated margin or upper part, of a membranous tissue

which lines the base of the calyx tube (? or uppermost part of the receptacle) and it is

apparently also this same tissue which joins the filamental column (collar or tube) formed

by the connate filaments or staminodes, to the base of the calyx tube at 5 points, so forming

the 5 pockets or depressions. This membranous tissue is often clearly seen in certain flowers

(when dissected) and such a wider conception of the corona seems also to be justified by the

occurrence of these processes on the tissue joining stamens to calyx (and/or receptacle 7)

n such species as A. glauca or A. spinosa.

THE [NFLORESCENCE.

The inflorescence in Adenia presents interesting features and in order to understand

these and the variations better, particular attention was paid to it. The main features

n the various species are discussed elsewhere. At this stage the general features will be

dealt with.

Harms (9) has made a special study of the morpholog xy of the inflorescence and tendrils

in the Passifloraceae. Speaking about the “ Inflorescenztriger oder Pedunculus”’, he

states that for Adenia: “ Dieser tragt seltener nur einen, meist zwei Seitenaste, wahrend

er in der Mitte in eine Ranke auslaiift. Die bliitentragenden Seitenzweige zeigen cymose

Verzweigung in mannigfacher Art und verschiedenem Grade”. The typical Adenia

inflorescence is a longer or shorter ““ peduncle” terminating in a ‘tendril and having two

* As stated, in the S. African spp. dioecism is practically the rule.

518

opposite or alternate side branches or “ cymes’ which may be repeated one or more times.

Common variations—with reference to the South African species—are (1) the absence

(non-development) of one or both cymes, when the peduncle may terminate in a flower-

bearing pedicel ; (2) the peduncle terminating in a pedicel.

Harms further states: “Im allgemeinen wiederholt sich die dichasiale Verzweigung

mit Ausbildung einer Mittelbliite einige wenige Male oder es gehen durch Fehlschlagen

eines der beiden Seitenadstchen die Dichasien schon bald oder erst by Achsen hoherer

Ordnung in Monochasien itber”. Presumably under dichasial branching Harms includes

those cases where the side-branches (“ Seitenastchen ”’) are not only opposite but also

alternate and this is essentially the nature of the inflorescence in the South African species,

in which sidebranches are always alternate except perhaps those (of the first order) arising

on the peduncle. In the South African species the standard groundplan for the inflorescence

is one in which each relative main axis (always with terminal flower) has usually two lateral

branches with a tendency either towards the monochasial (only 1 bract, with or without its

side-branch present) or towards the trichasial or pleiochasial* (where more than 2 alternate

bracts, with or without their side-branches are present).

The switching over to a mono-, tri-, or pleiochasium, particularly the latter two, is

apparently of unusual occurrence in the South African species because the youngest

relative main axes practically always bear two alternate bracts with buds. On the older

axes, however, this may sometimes occur.

In accordance with the law of development in the cymose inflorescence, the upper-

most side branch of each relative main axis is the oldest (and most developed) and naturally

flowers on this are relatively older than those of the branch below it. The branches of

any relative main axis are always alternate, and it very rarely happens that such branches

fail to develop in due course, 1.e. is only represented by a bract, without a bud, unless of

course they are the youngest bracts on the inflorescence. That portion of each relative

main axis which bears the flower (that is the pedicel) is often pushed aside and is less

developed than the side-branches.

All flowers (by definition) are borne terminally on each relative main axis (which may

or may not be repeated one or more times) and are therefore pedicillate, besides being

articulate. Not only does the distance between the side-branches (or their bracts) vary a

great deal in different species but also on the same inflorescence or on the same

or different individuals. The pedicel length is subject to similar variation due to the

position of the oldest (or nearest) relative bract very rarely bemg so short as to appear to

be lacking or almost so, as, for example, in 4. Wilmsii.

If then the side-branches remain undeveloped and their respective bracts are dis-

placed to their uppermost limit (i.e. bordering the articulation) the flower will appear

sessile and bracteolate. This is observed in A. Walmsw (fig. 13). Further modification

gives us the inflorescences of many species of Passiflora as pointed out by Harms and others.

In these the “ peduncle” has undergone maximum, reduction so that tendril and side-

branches (or only 1 side-branch) arise side by side in the axil of the leaf having their bracts

forced (displaced) on to their respective axes thus giving on each 3 bracts (2 bracteoles and

1 bract)} which in the various species of Passiflora assume various positions or modifications

from the three scattered narrow bracts to involucra of various shapes and sizes.

Although, to some extent, the characters of the inflorescence differ somewhat in the

different species and are fairly constant for certain species, there is a marked variation in

one or two of the species (e.g. A. digitata) and they would therefore not appear to be suitable

for the characterisation of species.

* A 4th side branch, represented by a bract, has only been observed once (in Fig. 15).

+ This is not uncommon in the case of A. glauca.

519

Harms refers to inflorescence dimorphism in the two sexes, e.g. in A. venenata but such

has not been established for the South African species. He also records the occurrence

of “ traubenihnliche Bliitenstinde ” in certain species, in which category the reproductive

branchlets of A. gummifera and A. Wilmsii may presumably be placed.

The interesting behaviour in A. gummifera (Harv.) Harms, to which Engler and Harms

have drawn attention, should here be mentioned. In this species (and presumably in

related species, according to these workers) there are usually 2, sometimes 3, buds in the

axils of the leaves, the lowermost of which develops into a tendril or strong inflorescence,

the other into a branchlet of varying size. This branchlet may at times give the impression

(with leaf abscission or ? non-development of leaves) of bemg a compound inflorescence.

As noted elsewhere these “secondary” inflorescences of the branchlets, apparently (in

the South African specimens), always (?) end in pedicels, not in tendrils. Presumably

Engler and Harms have only observed this branchlet (“ Seitenzweig Spross ’) above (i.e.

in the axil of) a tendril and not above an inflorescence. In the South African material,

this branchlet has been observed above a strongly developed inflorescence in which the

main axes (peduncles) terminate in tendrils.

VARIATION AND ABNORMALITIES IN THE SPECIES.

In his researches on the utilisation of the anatomic structure for the limitation and

division of the Passifloraceae, Harms in 1893 drew attention to the differences as well as

variations, in the anatomic structures exhibited by the various related genera and species

of the Passifloreae (including Modecceae) for the latter group of which he particularly

suggests the possible value of characters like “ Bau des Holzes, die Excrete des Blattes

(Krystalzellen, * Driisen ’, Gerbstoffbehalter) und die Haarbildung ”’, the latter term being

used in a sense to include the curvature of the cells of the epidermis. At the same time he

points out that a number of these characters vary a great deal and should only be used

with care for the limitation of species, although the above-mentioned characters as well

as others such as strength of outer epidermis walls, structure of mesophyll, etc., could be

utilised for such purpose.

In the revision of the South African species, the present writer has experienced great

difficulty in classifying certain specimens belonging to a group which he has referred to as

the “ digitata-senensis-complex ’. He has observed great variation in leafshape, in glands

and in the number of “* ruby-coloured ” or giant cells (presumably the “ Gerbstofibehalter ”

of Harms) on the upper and lower surfaces of the leaves of specimens of the same species

and even on the same specimen, as well as variation in the waxy covering in the same

species. Variation in the “ ruby-coloured cells” (viz. the ‘‘ Gerbstofibehalter ’’) is only a

reflection of the anatomic variation of the species. In view of this fact and the variation

observed by Harms for several of his anatomic characters (in the same species) the writer

does not feel that the classification of a difficult group like the “* digitata-senensis-complex ”’

would be much facilitated by the study of anatomic differences. Differences in anatomic

characters would no doubt be more marked in the case of those plants that are not so closely

related or those that are easily separated on morphological characters. Whether, however,

anatomic characters will have taxonomic value in cases where the comparatively numerous

morphological characters have failed, that is, where the latter vary so much that classification

is impossible or very difficult, seems doubtful.

A. hastata (Hary.) Schinz

In this species fairly wide variations are evident in the characters of the inflorescence ;

in size and stoutness of the inflorescence, including tendril and in stoutness of tendril in

relation to the peduncle. The “cymes” are usually opposite and few of the peduncles

terminate in pedicels, tendrils being usually present. In the number of flowers to a “ cyme ”

there is a wide difference.. The majority of specimens have 1-flowered or 1- to 2-flowered

520

cymes but several specimens are up to 5- or 6-flowered (or more ?). On the same specimen

the number of flowers to a cymé is quite constant. It may be noted that of three sheets of

Rogers 12606 one sheet showed typical 1-flowered “ cymes”, while the other two were

both many-flowered.

It 1s interesting to note that the apical glands of the leaf (paired glands at the apex

of the leaves) are sometimes absent on some specimens.

In figs. 4 and 5 two drawings of inflorescences of this species are shown.

Fig. 4,—Typical inflorescences of Adenia hastata (Harv.) Schinz.. enlarged several times.

From Rogers in Trans. Mus. 13273, ¢.

Very great variations were observed in size and shape of flower and the parts thereof.

This was particularly noticeable in the case of the petals which varied from entire or almost

30, In some specimens, to the characteristic fimbriate ones (the processes of the petals being

of very variable length) of the species, in the males.

The following abnormalities were observed :—

Rogers 13273: A sixth petal arising much below the others and differing some-

what from them.

Thorneroft 2034: Calyx lobes and petals 4-merous (stamens 5), Apparently only

1 flower of this nature.

Breyer 17956: Small leaves in axils of some inflorescences.

521

Fig. 5.—Typical inflorescences of Adenia Hastata (Harv.) Schinz., enlarged several times.

From Thorncroft 2034, 3.

Fig. 6.—Inflorescences of A. glauca Schinz, enlarged

many times. From Galpin 11605, 2.

522

fia 9

Ferrio €

Figs. 7 and 8.—Inflorescences of A. glauca Schinz, enlarged many times. 7 and 8 are from

Galpin 13197, &.

Fig. 9.—A typical inflorescence of A. spinosa Burtt Davy, enlarged many times. From Bremekamp

and Schweickerdt in Trans. Mus. 29882, 3.

523

A. glauca Schinz

Except for fairly marked variations in the inflorescence, this species is fairly uniform

in its flower parts.

What is unique is that occasionally the inflorescences are clustered at the bases of

flowering shoots or of branches or of branchlets but also often occur at the bases of tendrils

and in these cases the “cymes”’ are very contracted or apparently only represented by

single pedicels (with flowers) on very reduced peduncles. In the latter cases there are no

tendrils, though peduncles usually end in tendrils. Usually, however, inflorescences are

axillary (axils of leaves) fairly short (with no visible peduncles) with “cymes” 1- to 2-

flowered but occasionally many-flowered, ending in tendrils though not infrequently in long

(flower-bearing) pedicels. Like the stipules the bracts are reddish brown and those of the

undeveloped side-branches are often found to be displaced along their respective axes.*

In figs. 6, 7 and 8 two typical inflorescences are shown. No abnormalities were

observed.

It should be pointed out here that the writer has not been able to find any glands in

this species although marks were observed (grouping of “veins”’) coinciding with the

position of glands in the ordinary Adenia flower. Schinz in describing this species (l.c.)

states that “Die unanschaulichen Receptaculumeffigurationen, die den Staminodien

opponiert sind, haben zungenformige Gestalt”. He further states that the petals “ finden

wir im Grunde des Receptaculums inseriert’’. The present writer has always observed the

petals to be inserted at the sinuses of the calyx lobes (“ Kelchblatter ’”’ of Schinz), that is

at the upper margin of the “ Receptaculum ”’.

A. spinosa Burtt Davy

In this species the inflorescences are usually axiilary (axils of spines or those of leaves),

the “cymes” being many-flowered although occasionally 1-few-flowered. Inflorescences

are usually much reduced (compact). Such reduced inflorescences not only occur in the

axils of spines but are, not infrequently, also clustered near the base of the latter. The

thorn may be considered as the modified main axis of the inflorescence for not only does it

often function as a tendril, but it usually has buds near the base corresponding to the

position of the main branch of the inflorescence and, in fact, occasionally possesses normal

cymes (Bremekamp and Schweickerdt 469 in Transvaal Museum). This type of inflorescence

is shown in fig. 9. It also is of interest to record the presence of leaves in, what appears

to be, normal, axillary, reduced inflorescences. There are usually 2 to 3 flowers present

and such inflorescences are perhaps to be regarded as very reduced branchlets, arising,

it would appear, in the axils of spines, not in those of leaves, as axillary inflorescences do.

Presumably such branchlets should be interpreted as truly axillary, being the development

of a second bud in the leaf axil.

No abnormalities or marked variations were observed.

A. gummifera (Harv.) Harms

The occurrence of both primary and secondary inflorescences is interesting. The

former occur in axils of leaves while the latter occur on almost leafless branches which

arise in the axils of primary inflorescences. Secondary inflorescences are not visibly axillary

as leaves have often not developed or have dropped early. The 1-many-flowered “ cymes ”’

are usually alternate and usually the peduncles end in tendrils (though sometimes in

pedicels), but “cymes” are occasionally also opposite. One or both of the cymes may be

undeveloped (rudimentary) when they are represented by their bracts only or very little

besides. Not infrequently the inflorescence consists only of 2 bracts and a long terminal

flower-bearing pedicel. Peduncles are usually well-developed, particularly in the primary

inflorescences where they are often very long (up to 10 cm.).

* The bases of tendrils and the pedicels of the primary branches are also red-coloured.

524

anda:

fia. 11.

Figs. 10 and 11.—Inflorescences of A gwnmifera (Harv.) Harms, enlarged several times. 10 is from

Wylie in Natal Herb. 23313, 9. 11 is fromWatt and Brandwyk 1497.

525

In Figs. 10 and 11 two inflorescences of this species are shown.No abnormalities wera

observed.

With regard to floral structures, it can be recorded that this species does not show any

marked variations and neither are the variations in other morphological characters very

pronounced. Although differences in leaf shape, leaf size, number of large “ Gerbstoff-

zellen” (black dots), etc., are quite noticeable.

A. repanda (Burch.) Engler

The inflorescences are usually small, isignificant, the bracts characteristically long.

In some cases, particularly at the base of branches, they are so reduced that the flowers

appear to be solitary, the “cymes” being rudimentary and the shortened peduncle ter-

minating in a flower-bearing pedicel, not a tendril. Inflorescences are axillary. The

“cymes” are very rarely opposite and one or both of them may develop, usually only the

one develops; it may be the lower or the upper. The “cymes” are characteristically

1- to few-flowered. The peduncles end in pedicels (this usually is the case at bases of

branches) or tendrils (usually this occurs towards apices of branches). Occasionally both

cymes are undeveloped.

: Vi

ee C.

Fig. 12.—Inflorescences of A. repanda (Burch.) Engl. several times enlarged. A is from Pearson 8166

(Natal Herb.), B is from Dinter 4516, ¢, and C is from Marloth 1092, 3.

In Fig. 12 are shown some typical inflorescences.

On the whole this species is very uniform, showing a very narrow range of variation

except perhaps for the insertion of petals. No abnormalities were observed other than 4

small sessile flower buds (side by side) on one of the tendrils in Marloth 1092 (in National

Herbarium).

526

A. Wilms Harms

Of this species only a few specimens are available and the variation is not great except

for leafsize. The species appears to produce different ty pes of shoots, that is vegetative,

reproductive and normal and these do not, apparently, arise at the same time on the same

plant. The normal shoots have 1- to 3- flowered axillary inflorescences with the main axis

(peduncle) always terminating in a flower and not a tendril. This terminal flower of the

main axis is normal but one or both of the sidebranches (of the main axis) may be

undeveloped. The reproductive shoots are reduced main stems with several alternate

axillary inflorescences in which the leaves are considerably reduced, almost bractlike and

the inflorescences more elaborate than in the normal shoot. The bracts on the main axis

of the inflorescence or on any of its sidebranches may occupy any position on their respective

axes and may sometimes be so close to the articulation as to give the flower the appearance

Fig. 13.—Inflorescences of A. Wilmsii Harms,

several times enlarged. From specimens colld. by

Miss van Wyk (Nation. Herb.).

527

of an axillary origin. In Fig. 13 inflorescences are depicted. The glands or nectaries at

the apex of the pedicel are sometimes present in 2 pairs and not 1, which fact along with

unusually broad petiole and peduncle, as well as the extra pair of lobes (7 altogether) rather

suggests a degree of fasciation as being hereditary for this species which character may

possibly have arisen as a mutation in the original species from which it was evolved.

A. digitata (Harv.) Engl.

In this species very marked variations were observed. The variation in leaf characters,

particularly in shape and number of leaf lobes is a very striking feature as can be gauged

from the photographs (Plates 16-36). As pointed out further on some of these forms had

been mistaken for new species by other workers. In characters of the inflorescence and of

flowers the variations are almost equally striking. Thus, the peduncle which is generally

about 2 cm. long may often be absent or up to 6-5 cm. long. Or again, the flowers, which

are usually 4-12 per inflorescence may often be only 2 and sometimes as high as 40 or 80. The

distribution of giant cells (Gerbstoffbehalter) on the underside of the leaf has also been

observed to be extremely erratic even on the same specimen and on the same leaf. Some

typical inflorescences are shown in Figs. 14-17.

A few abnormalities were noted as follows: Osborne in Nat. Herb. (2639), 3 glands.

Rogers 24185 bract adnate to the receptacle, 6 calyx lobes, 6 stamens and in a second

flower only 4 stamens and 3 minute peglike outgrowths from the base of the anther. Galpin

13196, 2 stamens have their anthers and free filaments connate. Breyer in Trans. Mus.

24215, 1 flower with 6 calyx lobes and 6 stamens.

ia, 14

Fig. 14.—Inflorescence of A. digitata (Harv.) Engl., trom Schweickerdt in

Trans. Mus. 30164, g. Enlarged.

528

A. fruticosa Burtt Davy

Inflorescences are 1- to 3-flowered, axillary ; peduncles lacking or up to 3 or 4mm.

long, ending in normal flowers not tendrils, the latter bemg bracteacte or non-bracteate,

strong or weak. Tendrils occur in axils of leaves and branches in axils of tendrils. Leaf

size Is very variable.

EYES 3243 IN HERBBoLus.

a a ae ‘

Fig 15. ff STEM

Fig. 15.—Inflorescence of A. digitata (Hary.) Engl., from Eyles 3243 (Bolus Herb.). Enlarged several

times.

529

Fig. 16.—Inflorescence of A. digitata (Harv.) Engl., from Mogg 8182, 9. Enlarged several times.

530

fie. 17. rg

Fig. 17.—Inflorescence of A. digitata (Harv.) Engl., several times enlarged.

From Osborne in Nat. Herb. 2639, 3.

NaruRAL RELATIONSHIPS AND CLASSIFICATION OF THE SouTH AFRICAN SPECIES.

Of the eight species which the writer distinguishes, six, to our present knowledge,

are practically limited to South Africa. Under such conditions one would perhaps expect

a certain degree of relationship among them and such appears to be the case. As already

pointed out glauca and spinosa are closely related while fruticosa may be regarded as a

near relative ; digitata, and Wilmsii on the other hand, are even more closely related to each

other. The writer regards the latter as having been evolved from the variable digitata.

Gummifera and repanda have (like glauca and spinosa) no glands and in addition they

lack coronae and have simple leaves; gummifera, a widely distributed species in Africa,

and hastata and repanda are not clearly related to each other or to any of the other species.

The classification of the majority of species presented little difficulty but as a result

of leaf heterophylly in one species, however, the work was much delayed, the complexity

of the problem being increased by dioecism and floral polymorphism. The classification of

the senensis—digitata material has resulted in more than 2 years delay in the completion of

this study. Early in this work it became evident that leaf variability (heterophylly) within

the senensis—digitata group has been responsible for the founding of a few additional species.

To obtain the necessary proof for this view, has involved a good deal of time. This view is

being illustrated by means of Plates 16-36 in conjunction with Plates 8-15 showing photos

of type specimens of the species which were sunk. Moreover, further delay occurred when

the South African material of senensis—digitata was forwarded to Kew for comparison,

although, unfortunately, this procedure has been of little or no assistance.

Attention is here directed to the fact that amongst the material which the writer

classifies under A. digitata a number of specimens have rather large fruits and may therefore

represent a distinct group which appears to grow only in a certain area, approximately

531

defined by the Pretoria-Rustenburg districts. It is not unlikely that in future evidence

may become available that would justify the separation of such large-fruited plants as a

distinct variety (or perhaps species). However, such classification should be based on at

least another character not associated with the fruit as otherwise male plants (and non-

fruiting female plants) from the same area, cannot be satisfactorily classified or will have

to be classified with the species. Even if at this stage one were to favour separation of

large-fruited specimens as a distinct group (variety or species) such a procedure would be

impossible on account of this very difficulty, because there is as yet no means of separating

male or non-fruiting female plants. As a result of dioecism and floral polymorphism field

studies would be necessary. Herbarium material is usually wanting in either fruit or

flowers.

It should be mentioned that giant cells (“ Gerbstoffbehalter ”) have not been observed

on the Pretoria—Rustenburg specimens but, as pointed out, the distribution of this character

is so variable and erratic that too much significance should perhaps not be attached thereto.

Another character which would perhaps assist the elucidation of this problem is fruit colour.

A Note on A. multiflora Portrs

This species was described from the growth made from an “ enormous tuber” (61 cm.

diameter and 30 cm. high) collected at Baviaanspoort, near Pretoria, by Dr. J. M. Fehrson

in September, 1913. The description was based on material obtained from growth made

“in a very sunny spot on the windowsill in the herbarium” (Transvaal Museum). The

“tuber ” was presumably not in soil and flowered first in February, 1914. ‘It never got

a drop of water and again in October of the same year it made long shoots and flowered

profusely as can be seen by the photograph taken at that time.” Somewhat abnormal

growth might be expected under such conditions and the present writer suggests that the

numerous and small flowers as well as the unusually narrow lobed leaves (and the absence

of tendrils too) that the “type” specimen showed was probably due to this or perhaps

to the wrong season in which the plant first flowered. The multiflowered character is not

unusual in A. digitata.

The “ tuber ” is stated to grow partly above ground and to have had a “ grey leathery

sk . . . greenbeneath”. This character in itself was at first considered by the

writer to be sufficiently distinct to justify a distinct species from A. digitata but on further

study of photograph and type material in comparison with A. digitata, he observed a strong

resemblance between the two and thereupon made another attempt to obtain material from

the type locality, several previous visits having been unsuccessful. He was fortunate to

locate two Adenia plants under a tree of Acacia caffra. They were growing side by side and

their branches and leaves were dry but in the latter case both were of the A. digitata type.

The two tubers were dug up and the one was completely underground and resembled a

typical A. digitata tuber. The other one, partly above ground, was found to be growing

immediately above a large root of the Acacia tree and had produced two separate “ tap-

roots ” on either side of the root on which it was, so to speak, sitting astride, Plate 7. This

obstruction possibly accounts for its appearance above the surface of the ground where the

exposed surface was grayish but green just beneath, as described for A. multiflora.

The accompanying photograph shows the two “tubers”. This, the writer contends

proves that the appearance of the “ tuber” above the surface of the ground is not due to

a hereditary factor, but to environment. Whether subterranean obstruction is the only

cause of this behaviour is doubtful for in the poisonous plant garden at the Onderstepoort

Veterinary Research Laboratory, two plants of A. digitata, the tubers of which were originally

“planted beneath the surface ”, have appeared above the surface and are said to be rising

steadily.

532

ACKNOWLEDGMENTS.

The writer wishes to thank the Chief, Division of Plant Industry, Department of

Acriculture and Forestry, for the facilities which enabled him to carry out this study.

To Dr. E. P. Phillips, Principal Botanist, he is greatly indebted for interest and guidance.

Dr. H. G. Schweickerdt, Botanist for the Division of Plant Industry at Kew, has

rendered valuable assistance in connection with the literature, the examination of type

specimens both at Kew and Berlin, the examination and comparison with Kew material

of some 100 sheets of the two closely-related species, digitata and senensis. He spared

neither trouble nor time to obtain the desired information and I therefore make grateful

acknowledgment to him.

My thanks are also due to the Curator of the Bolus Herbarium, the directors of the

Capetown Museum, Albany Museum and Transvaal Museum, as well as to the Professor of

Botany, Witwatersrand University and the Mycologist in Charge of the Natal Herbarium

for the loan of specimens. Mr. W. G. Barnard, Stock Inspector im Sekukuniland has, on

request, forwarded a range of very useful material for which I wish to thank him very

much.

KEY TO THE IDENTIFICATION OF THE SPECIES.

1. Plants with well-developed thorns; leaves\simple............................... A. spinosa.

Plants without thorns (rarely with weak thorns in A. fruticosa but then leaves lobed) 2

2. Leaves simple or not digitately compound......................5..5...200-e0e 3

IDGERVES CheAKEEHICIh7 COM YONG, coooaceccccos000Ks 5500000000900 900003 300000000000 5

3; Leavessat least 3) timesias)longyas| broad) -jctriiis serials ela ieee eer i A, repanda,

Leaves not more than 1-5 times as long as broad or leaves broader than long....... 4

4 Weavestandstlowersspunctulalten. teeter n teeter tien nee neni A. gummifera.

ILenmas Gin! TO VOAS WOW POWAOCHENIDs co ccavenoc..0n0 ps 00D0ban nD DS DOUCUDDOODSODNE A. hastata.

5. Leaves palmately 7-lobed (very rarely 5-lobed) tendrils absent.................... A. Wilmsii.

Leaves palmately 3- to 5-lobed, tendrils present................................. 6

6. Leaf-lobes more or less orbicular or somewhat broader than long; fruits variegated A. fruticosa.

Leaf-lobes at least twice as long as broad; fruits not variegated.................. if

7. Leaf-lobes broadly elliptic or elliptic oblong very glaucous, entire, conduplicate ;

petals arising in sinuses of the lobes. ‘“ Tuber” always partly above ground... A. glauca.

Leaf-lobes lanceolate or lanceolate-ovate, not strikingly glaucous, not conduplicate ;

petals arising near the base of the calyx tube. “ Tubers” very rarely above Be:

PALO, sooo ooo COOD OHNO DOOD ODDD NOD DODOOOOOOD nadoo boo EOBOpoODOOONQ0OOH A. digitata.

533

DESCRIPTION OF THE SPECIES.

Where possible the sex of the specimen has been indicated as well as the herbarium

in which the specimens may be found.

A.—Albany Museum Herbarium, Grahamstown.

B.—Bolus Herbarium. Capetown.

C.—Natal Herbarium, Durban.

N.—National Herbarium, Pretoria.

S.—South African Museum Herbarium, Capetown.

T.—Transvaal Museum Herbarium, Pretoria.

W.—Witwatersrand University Herbarium, Johannesburg.

All specimens quoted were seen by the writer.

1. A. spinosa Burtt Davy. Man. Fl. Pl. and Ferns. Pt. 1, 36, 221, 222.

Main (swollen) axis, irregular, tuberlike, variously shaped, fleshy, partly or mostly

above ground, the latter green always growing in breadth, up to over 2 metres diameter

(Bremekamp 3). Branches, numerous, “ divaricate”’, arising from attennuations of the

main stem; branchlets virgate, striate, glabrous, armed with spines; spines terete

spreading, fairly stout or slender, 1-3 cm. apart, 1-2-4 cm. long, glaucous, resembling the

branches or brownish, or the cortex thickened towards the base or in patches or from near

the base upwards, with a straight or curved point or ending in a short tendril, with or

without “ toothlike ” prominences or inflorescences near or at the base. Leaves simple,

entire, sessile or shortly petioled, 1-5-3-2 em. long, 1-2-3 cm. broad; oblong-ovate,

retuse or emarginate or rounded at the apex, cordate or rounded at the base; glands

absent on surface, two glands at the apex of the petiole and another smaller one at the

apex of midrib below ; petiole subsessile to 7 mm. long ; stipules minute, toothlike from a

broad base, reddish brown. Inflorescence short or very reduced, in axils of the spines or

clustered at the base of spines; “cymes” 1-few-flowered; peduncles usually very

abbreviated, terminating in pedicels or spines. Male flowers about 1-8 cm. long, yellowish.

Receptacle scarcely 3 mm. long. Calyx-tube obconical, one-third the length of the lobes ;

lobes spreading, linear-oblong, entire, obtuse. Petals about two-thirds as long as the calyx-

lobes, arising from their sinuses but appearing to be also partly inserted on the tissue joining

calyx-tube to the staminal-tube, membranous, transparent, l-nerved, linear, acuminate,

narrowing below, subserrate or subundulate near the apices. Corona of slender processes,

0-5-1:0 mm. long, arising around the bases of the petals, partly or largely from the tissues

joining staminal-tube to calyx-tube. Stamens much overtopped by the petals, extending

to the middel of the calyx-lobes ; filaments connate for half their length at the base, adnate

to the calyx-tube at 5 points, forming 5 shallow narrow pockets ; anthers linear or oblong :

connective not produced into a point. Glands O. Ovary rudimentary, small. Female

flowers not seen. Fruit yellowish, about 2 cm. long, ovoid; pericarp leathery-papery, ?

usually indehiscent.

TRANSVAAL.—Zoutpansberg district: Messina, September 1918, Rogers in T. 24000,

Q. Messina, Rogers 19299,? (N). Messina 2,000 ft., September 1918, Rogers 21664, 3

and 9 (N). Messina, Rogers 19341, 9 (N). Near Messina, base of kopjie above the river

Limpopo near the gorge, shrub w. elephantsfootlike base 27.5.27, Young in Herb. Moss.

14672, 3, and T. 26933. On farm “ Zoutpan No. 193”, very characteristic of northern

slopes of Zoutpansberg, main stem tuberous, about 1 ft. high and 23 ft. broad at the base,

Novem. 1932, Obermeyer, Schweickerdt and Verdoorn 137 (N & T), 2 sheets of each.

North of Fogwells, 21.12.35, Smuts and Gillett 3114 (N). Pietersburg district: At

Naauwpoort, 29.1.31, Bremekamp and Schweickerdt 469, 3 (N & T).

504

2. A. repanda (Burch.)* Engl. Bot. Jahrb. XIV 375.

Main axis a tuberlike topshaped—napiform root with a stem formed from accumulated

remains of annual growths, subterranean ? or partly exposed. Stem woody, up to about

20 cm. high and 8 mm. thick, greyish brown. Branches straggling, ? or climbing, striate,

grayish or purplish gray, glaucous, about 4 mm. thick at base, up to 120 cm. long, arising

from the main stem or directly from the rootstock ; branchlets absent. Leaves subsessile

or shortly petioled, semi-conduplicate, 6-13 cm. long, 0-5-2 cm. broad, simple, subentire,

remotely repand or short-lobed, with thinly cartilagenous green or reddish margins ; linear-

lanceolate or elongate-elliptic, obtuse, reticulated ; glands on underside of leaf, below each

marginal inequality (or lobe), below the apex at the end of the midrib and 2 at base of leaf

on each side of apex of the petiole; petiole up to about 5 mm. long; stipules brownish,

acicular, 1-1-5 mm. long. Inflorescence usually small, axillary, characteristically 2- to

few-flowered ; peduncle usually with alternate sidebranches, generally 1 sidebranch develops,

occasionally both rudimentary; peduncle ending in tendril or pedicel; flowers usually

dioecious, | “ yellowish”, “ dirty yellow ” or “ ochraceous’, or “ greenish’. Male flowers

about 2 cm. long. Receptacle usually about 3-4 mm. long, narrowly cylindric, slightly

widening upwards. Calyx obconically tubular, the limb half or slightly less the length

of the tube; lobes spreading, ovate to ovate-oblong to subovate to elliptic oblong, obtuse

or subobtuse, entire. Petals inserted a little below the sinuses of the calyx-lobes or

oceasionally towards the middle of the calyx-tube, membranous, transparent, 1-nerved or

with branched palmate veins, narrowly oblong or elliptic-oblong, acute to subobtuse,

entire. Corona 0. Stamens free, inserted usually at about the middle of the calyx-tube,

rarely lower, about equalling the petals in length ; filaments, subulate or linear-subulate,

from half to as long as the length of the anthers ; anthers broadly linear ; connective not

produced into a point. Glands 0. Ovary rudimentary, about 1 mm. in length. Female

flowers “ greenish’, just over 1 cm. long. Receptacle 1-2 mm. long, shortly funnel-shaped

or sub-cylindric. Calyx reddish or greenish, subcampanulate; the limb slightly shorter

than the tube; lobes oblong-ovate, obtuse, entire. Petals inserted more than halfway

up the tube, extending to the sinuses of the calyx-lobes, transparent, membranous,

l-nerved, short, oblong or elliptic-oblong, acute or subacute, apices curved outwards, entire.

Coronz 0. Staminodes free or connate at the extreme base, inserted at the base of the calyx,

subulate or subulate-linear, with apices curved, hooked or tipped with abortive anthers.

Glands 0. Ovary stipitate, ovoid or ellipsoid-ovoid, smooth ; style fairly long, 3-branched,

with the branches widening, terminating in a fleshy-papillate surface forming the stigma ;

ovules few, arising either in lower half or in upper half of the ovary. Fruit bright or coral

red, roundish to ovoid, roundly and shallowly 3-lobed, about 1-5-3 cm. long, leathery,

dehiscent, splitting into 3 (or 4?) valves; seeds roundish-heartshaped, regularly pitted.

Paschanthus repandus Burch. Trav. 1, 543. Modecca paschanthus Harv. Flora Cap. 11,

p- 500. Jdggia repanda Schinz.f Verh. bot. Ver. Prov. Brand. 1888, 254. Fig. in Engl.

Pflanzenwelt Afrikas TX. Bd. 111, Heft 2, 601. Paschanthus Jdaggu Schinz.§

* The present writer is inclined to endorse the view of Schinz (l.c.) who thought that this species

does not fit in well into Adenia as in several respects it differs from it, viz. absence of corona, herma-

phroditism, non-connate stamens absence of glands and insertion of the stamens. However, no critical

attitude is justified as the writer has only studied a a limited number of the species in the genus.

+ Burchell described this species as having polygamous flowers. Schinz described Jdggia, which

Harms places as synonymous with Adenia, as hermaphroditic, Harms gives for his sect. 1 Paschanthus

(Burch.) Harms (Jdggia, Schinz): Flowers hermaphrodite, polygamous or dioecious. The present writer

has not found polygamy or hermaphroditism of common occurrence.

1H. Schinz in Bull. Travaux de la Soc. Bot. Geneve XI (67) 1891 states, P. repandus Burch.

(Jdggia repanda Schinz).

§ The publication in which this epithet appeared is not known. Modecca repanda Druce. Rep. Bot.

Exch. Club, Brit. Isles 1916, 636, is given as a syn. in Kew Index, but the author has not seen this

publication.

530

TRANSVAAL.—Zoutpansberg district: Vivo, 20.1.31, Bremekamp and Schweickerdt

206, (N & T). Near farm Chapudi between Zoutpan and Waterpoort, branches flexuous

not climbing, leaves markedly glaucous, scattered specimens seen 26.11.32, Obermeyer,

Schweickerdt and Verdoorn 246, 2 (N & T). Botanical Reserve, Messina, 18.12.28, Pole

Ewans 2, 2 (N), (2 sheets).

Cape Province.—Barkly West district: At Motito, Febr. 1842, Burchell 2486/2, 3,

type (N). Windsorton 1150 M., Jan. 1910. Marloth 5840, 3 & 3 (N). Prieska district :

Without precise locality rocky hillsides, greedily eaten by stock, 26.11.28 and 3.12.35

Bryant 345, 2 (N).

BECHUANALAND.—Near Kuruman, Ga Mbani Mts. 1350 M, Febr. 1886, Marloth 1092,

3 (N), 2 specimens.

Sourn West Arrica.—Windhoek, Nov. 1924, Rogers 29798, 3 & 2 (N &$). Ditto,

Rogers 29723 and 29784 (T & 8). Great Karasberg, Narudas Siid, middle slopes, straggling

among rocks fairly common, Dec. 1912/Jan. 1913, Pearson 8166, 3 & @ (N, C, & 8).

Okahandja, niedrige Acacienbuschsteppe, 27.1.07, Dinter 362, 2 (S). Windhoek, Glimmer-

schieferberge, Dec. 1912, Dinter 4516, 2. Waterberg: Quickborn, under thorn trees, Apr.

1929, Bradfield 75, 9 (N). Near Karibib, 24.12.29, Moss 17893, 2 (W). hills S.W.

Griindoorn, erect 2-3 ft., in partial shade, Pearson 4276 (N).

3. A. gummifera (Harv.) Harms. Natiirl. Pflanzenfam. Nachtrag 1, 255.*

Main stem woody, cylindric, greyish, up to 7-5 cm. im diam., the nodes enlarging

with age, climbing to tops of large forest trees (‘liana ’’), profusely branched. Branches

green, striate, climbing in the canopies of trees, very glaucous; branchlets often numerous.

Leaves petioled, 4-3-10 em. wide and 4-8-8 cm. long, varying from kidney-shaped to deeply

3-lobed; lobes usually shallow, rounded, entire, the median one oblong, triangular or

broadly oblong-ovate, bases variously cordate, subtruncate or variously rounded ; sinuses

wide; undersurface paler, visibly net-vemed, punctulate (gland dotted) on one or both

surfaces; dots dense or scattered, same colour as leaf surface or black; glands, ? rarely

present on under surface of leaf, solitary at the apex of the petiole; petiole 3-5-9 cm.

long: stipules minute, scale-like or a scaly ridge, usually shrivellmg away with age.

Inflorescence usually fairly open (not reduced), 2-many-flowered ; sidebranches of peduncle

usually alternate, sometimes one or both undeveloped ; peduncles of primary inflorescences

2-14 cm. long, usually terminating in a tendril, sometimes in a long pedicel; flowers “ green”

“cream”, “yellowish”. Male flowers about 1-5 cm. long. Receptacle 3-5 mm. long,

narrowly subcylindric, abruptly widening at upper end or subfunnel-shaped or obconical.

Calyx-tube under 2-5 mm. long, saucer-shaped or ring-shaped, one-third to one-fifth the

length of the limb; lobes spreading, linear or narrowly oblong-ovate or oblong-elliptic or

linear-ovate or oblong or subspathulate, obtuse to subacute, entire or slightly uneven at

apices, faintly striped-splashed and remotely black-dotted (punctulate). Petals inserted at

the sinsuses of the calyx-tubes, resembling these and equalling them in length, or shorter or

slightly longer, linear-oblanceolate, subspathulate, oblanceolate or linear or elliptic-oblong,

more transparent and less punctulate than the calyx-lobes, crenate-dentate, uneven near the

apices, acute to obtuse. Corona 0. Stamens arising from the centre of the receptacle,

extending to beyond the middle of the calyx-lobes or nearly to their apices ; free portions of

filaments shortly subulate or linear-subulate, connate for almost half (or slightly more)

* There is apparently a good deal of confusion as regards the relation between this species and 4.

cissampeloides (Planch) Harms, some regarding them as identical. Masters divides them, it would appear,

as an Mastern and Western species and describes both as having the sepals inserted at the base of the

calyx which definitely is not the case in the S.A. plant which invariably has the petals arising in the sinuses

of the calyx-lobes and correctly described by Harvey in Fl. Cap. vol. 2, p. 500. Dr. H. G. Schweickerdt

writing from Kew (Jan. 6, 1937) states: “A. gummifera (Harv.) Harms and A. cissampeloides (Planch)

Harms are two good distinct species. I first believed them to be conspecific, but now am quite ecrvinced

that they are distinct ; the venation differs,”

536

their length at the base or almost free ; anthers, linear, or lmear-oblong, densely and minutely

reddish-brown spotted; connective ? usually “dotted” (punctulate). Glands 0. Ovary

rudimentary, minute or overtopping the staminal tube. #emale flowers about 5-6 mm.

long or slightly longer. Receptacle minute, 1 mm.and under. Calyx-tube absent or practically

so; sepals spreading, linear-oblong, ovate or ovate-oblong, faintly-striped-splashed and

remotely black-dotted (punctulate), obtuse to subacute, entire. Petals inserted at the

sinuses of the sepals, about half the length of these or shorter, linear, under } mm. wide,

apices curved, transparent, with black “dots” few or absent. Corona 0. Staminodes

short flat out-growths or toothlike, arising just at the base of the ovary stalk. Glands 0.

Ovary sessile, or shortly stipitate, ovoid, smooth ; style short or wanting ; stigma of short

reflexed lobes, arising abruptly from the narrowed apex of the ovary or as 3 expanded

branches from a short style; ovules, several. Fruit “ brownish-orange”’, up to 4 or 5 cm.

long, ellipsoid, leathery, dehiscent, seed 4 mm. long somewhat flattened, subovate, regularly

pitted. Modecca gummifera (Harv.) Harv. & Sond. Fl. Cap. I, 500. Ophiocolon (M. ?

gummifera) Harv Gen. 8. Afr. Pl. Ed. 2, 121. Ophiocolon gummifer Mast. Oliv. Fl. Trop.

Afr., I, 518. Ophiocolon gummifera (Harv. & Sond.) Mast. Nat. Pflanzenfam. Ed. I, III,

ta., 83. Adenia gummifera Burtt Davy FI. Pl. and Ferns, Tvl., & Swaz. 1, 222, and in Ann.

Transvaal Museum III, 121.

TRANSVAAL.—Nelspruit district: At Kaapmuiden, Febr. 1922, Thorneroft 1199, in

T.M. No. 23140, 3 (T). Just outside Nelspruit, 2,700 ft. 26.10.30, Liebenberg 2636, 3 (N).

Barberton district : Highland Creek, climbing over trees 10-20 ft. high, 4,000 ft, 29.1.90,

Galpin 782. Zoutpansberg district: Farm “ Elsteg”, 5 miles west of Lows Trichardt,

Nov. 1932, Obermeyer, Schweickerdt and Verdoorn 355, 9 (N & T). At Elim, Dec. 1930,

Obermeyer 819, 2 (T). Pisangkop, Febr. 1878, Nelson in T. 11159, 2. Pietersburg district :

Modjadjies, Rogers 18110 (W). Politsi, Dec. 1932, Schweickerdt 1039, 3 (N). Tshakoma,

Nov. 1931, Obermeyer 1063 (T). Magoebaskloof, Jan. 1933, Murray 761 (N). Lydenburg

district : Mariepskop, Nov. 1925, Fitzsimons and van Dam in T. 30631, 3. Sekukuni

Location, Kloof, Western Spur on farm Magnets Heights 4,500 ft.. 24.10.34, Barnard 128,

3 (N).

NaTAL AND ZULULAND.—Durban, Doonside, Dec. 1933, Wyllie in C. 23313, g. Berea,

150 ft., Dec. 1894, Wood 5502, 3 (N). Ditto, 17.1.98, Wood 6662, 3 & 2 (N). Farm Frie-

denau, Station Dumisa, 6.12.08, Rudatis 523, 3. Without precise locality, Gerrard and

McKen in C. 688, 3. Pietermaritzburg: Umlalaas, 1.1.33, Gerstner im C. 22614.

CarpE PRovince.—In woods near Keimouth, 100 ft., Jan. 1892, Flanagan 1156, 3 &

Q(N). Port St. John, climbing over shrubs, 15 ft., Dec. 1896, Galpin 3461, 2 (N). Umtata

to Port St. Johns, Dec. 1927, Blenkiron in W. 16053. Kentani district: Valley, immense

climbing plant, Jan. 1903, alt. O, Pegler 869, 3 (N).

4. A. hastata (Harv.) Schinz. Bot. Jahrb. XV. Beibl. 33, 3.

Root not known ? like in A. digitata. Main stem ? absent. Main branches herbaceous,

annual straggling, procumbent or climbing, striate, greyish or dark, sometimes glaucous,

up to 4mm. thick at the base and “ 250 cm. long or more ”’, ? arising from an underground ?

tuberlike rootstock ; branchlets usually absent. Leaves petioled, 3-8 cm. long and broad,

simple, entire, variously cordate-ovate to hastate* (but the lateral lobes obtuse) with the

median lobe lanceolate, acute or subobtuse ; glands 2 (paired) at the apex of the petiole,

variable, often 2 (paired) at the leaf apex, variable ; petiole, 0-8-5-0 cm. long; stipules

subulate and toothlike, 1:5-2-0 mm. long. Inflorescence axillary, varying much in size;

peduncles usually terminating in tendrils, with sidebranches usually opposite and few-

flowered. Male flowers 1-3-3-1 cm. long. Receptacle 3-8 mm. long. subcylindric, widening

towards the apex, or obconical to subfunnel-shaped. Calyx tubular, widening upwards or

* Peltate-ovate leaves were typical of Gerstner 2345 but were not observed in any other specimens.

537

subcampanulate, with the limb one-third to one-half the length of the tube; lobes ovate

to orbicular or oblong, obtuse or subacute, with the 5 interiorly overlapping margins laciniate-

lacerate and the remaining 5 entire. Petals inserted below (just above the corona) at, or

above the middle of the calyx-tube, transparent to subtransparent, l-nerved or 3-veined,

linear-lanceolate, narrowed at the base, entire to remotely serrate (or distantly narrow-

lobed) or subentire or fimbriate-laciniate (filiform processes) for two-thirds or more of their

length, with the processes varying in length and density. Corona a sinuate circle

of filiform processes, arising from a little above the base of the calyx-tube in

groups or in a continuous circle; processes 0-75-3 mm. long, scattered to very dense.

Stamens arising from the centre of the receptacle extending to the sinuses of the lobes or

well below; free portions of the filaments subulate-linear ; connate for one-fourth to two-

fifths their length at the base, forming a shallow cup adnate to the calyx-tube (at 5 points)

forming 5 narrow pockets or depressions; anthers linear-oblong to linear, with the con-

nective not produced into a point, equalling or up to 1-5 times the length of the free fila-

ments. Glands 5, arising from the base of the pockets (or depressions), approximately 1-5

mm. long, hidden in or protruding from the pockets, flattened, capitate or subcapitate,

shortly linear to shortly oblong or spathulate. Ovary subterete, protruding slightly above

the staminal cup or extending beyond the apices of the filaments. Female flowers, 1-2-1-8

em. long, “ white” ?. Receptacle 1-5-2-5 mm. long, stout, shortly funnel-shaped or sub-

funnel-shaped. Calyx campanulate or nearly so, narrowed at the base; tube twice as long

as the limb; lobes ovate to orbicular-ovate to oblong-ovate, obtuse to subacute with the

5 interiorly overlapping margins laciniate to lacerate, the remaining entire. Petals inserted

at about the middle of (or just below) the calyx-tube, extending to about the sinuses of the

calyx-lobes, narrowly linear-acute or linear-acuminate, slightly curved at the apices or

straight, entire or remotely dentate, narrowly lobed in the upper half. Corona a circle of

filiform processes or fimbriately lobed processes ; processes 0:75-1:75 mm. long, arising

from near the base of the calyx-tube. Staminodes arising from the centre of the receptacle,

bases connate, forming a collar around the stalk or ovary, adnate to the extreme base of

the calyx-tube (? or receptacle) forming 5 pockets or depressions, linear-subulate, curved

or bent near the apices. Glands arising from the side of the depressions very small or up

to 1 mm. long, flattened, variously capitate, shortly oblong or subpathulate. Ovary stipitate,

ovoid or spherical, smooth ; style fairly long, 3-branched; branches palmately widened,

terminating in a fleshy-papillate surface (stigma): ovules numerous. Frwt “as large as

an egg” (Harvey), and smaller, “green and white”, roundish “pulpy”, leathery, ?

dehiscent; seed flattened, ovate-orbicular, pitted. Modecca hastata Harv. Thes., Cap. 11,

43, pl. 167. Adenia Schlechteri Harms Engl. Bot. Jahrb. 33, 150. Adenia hastata Burtt

Davy Ann. Transv. Museum III, 121.

TRANSVAAL.—Nelspruit district : Komatipoort, Rogers (? 12606) in T. 13273. Ditto,

Nov. 1931, 1,000-2,000 ft., Rogers 12606, 3 (8, B & N). Komatipoort, 14.12.97, Schlechter

11747 (N). Nelspruit, Dec. 1917, Breyer, T. 17956, g. Karino, 28.1.29, Hutt in N. 7870, 2

Near Nelspruit, 24.1.06, Crome in N. 1489, 2. Barberton district: Barberton, Nov. 1915,

Rogers 18369, 3 (S). Barberton, Nov. 1931, Smith 7006, 3 (N). Barberton, Sept./Oct.

1889, 2,300-3,000 ft. procumbent, 3-4 ft. long, In stony ground on hillsides among rocks,

Galpin 563, $ & 9(B. & N). Hills near Barberton, climber, 3,000 ft., Aug. 1923, Thorneroft

2034, g¢ (N). Barberton, Oct. 1922, Wager in T. 23675, 3. Barberton! “Nov. 1931. Smith

7069 (N). Ditto, Thorncroft in C. 5980, 3 & 2. Ditto, Nov. 1909, Williams in T. 7648, 9.

Ditto, Oct. 1907, Thorncroft in T. 3923, 2. Kruger National Park: Skukuza, in shade of

Acacia, among rocks on N. bank of Sabi R., Letty 43, 9 (N). ? district : Witsteen, growing

in Div. Pl. Ind. Garden, Hutchinson, 3 (N).

Narat.—Without precise locality (? near Greytown), 29.10.31, Pole Evans 3836, 9

(N). Nongoma district: Mahlabatini, between diabase, 18.11.37, Gerstner 2345, 3 (N).

Middle Umkuzi, on Mr. Nagels farm, 10.1.36, Gerstner 2895 (N).

538

5. Adenia Wilmsii Harms. Engl. Bot. Jahrb. 26: 238.

Root tuberlike, napiform or variously shaped, ? not rising above ground, up to 7 lbs. ?

and over. Main stem ? always underground, an attenuated outgrowth from root or ?

formed from accumulated remains of annual growths, woody. Branches herbaceous, annual.

one or more from same rootstock or main stem, semi-erect or semi-procumbent up to 50

em. and over. Leaves digitately compound, petioled ; lobes 7, occasionally 5, the median

lobe often entire, but usually pinnati-lobed (1-2 pairs) at about the middle or below, 5-12

em. long (? often less) ; the other lobes entire and simple, unequal in length ; glands absent

on under-surface and usually also in the sinuses of the lobes; petiole stout, usually long,

4-7 cm., with 2 large (occasionally 4) fleshy circular glands at the junction with the lobes,

above ; stipules about 1-2 mm. long, toothlike. Inflorescence axillary ; peduncle stout,

4-5 cm. long, terminating in a normal flower; main branches opposite or alternate, one or

both sometimes rudimentary with the “ terminal ” flower (at the termination of the peduncle)

? always developed; peduncle branches of normal stems 1-flowered, 2- to 3-flowered on

reproductive shoots; flowers yellowish. Male flowers about 2-5cm. long. Receptacle 5-7

mm. long, linear, subeylindrie widening upwards. Calyx salver-shaped or cylindric-

campanulate ; tube cylindric or narrowly obconical, sometimes abruptly narrowed above

the base, one and one-half to twice the length of lobes; lobes ovate or oblong-ovate

or subelliptic, obtuse or subobtuse, interior margins entire or subentire. Petals inserted

a little below the middle of the calyx-tube, extending a little or well beyond the sinuses

of the lobes, membranous, transparent, palmately 3- or 5-veined, with sideveins in the latter

usually few-branched, oblanceolate or elliptic-oblanceolate, subserrate in upper } or 4,

subobtuse. Corona of slender processes, 0:5-0-75 mm. long, arranged in a circle or smuate

ring just below the insertion of the petals. Stamens well overtopped by the petals, extending

to just below the sinuses of the calyx-lobes, connate for half their length at the base; fila-

mental column adnate to the calyx-tube for 4 or its full length producing 5 pockets ; anthers

equal in length or longer than the free filaments, apiculate. Glands arising from the base

of the pockets, flattened, capitate; stems short or almost absent. Ovary rudimentary,

extending half way up the filamental column or its entire length. Female flowers not seen.

TraNsvAAL.—Lydenburg district : On High School grounds in Lydenburg, Oct./Nov.

1935, Van Wyk, 3 & 9 (N). Without precise date and locality (? at Lydenburg), van Wyk

(N). liydenburg, 1935, Pons, 3 (N). Farm Rooidraai,1/8” long stems, red loam soil,

7.12.35, Liebenberg 3488 (N). Ditto, Liebenberg 3496 (N).

6. A. fruticosa Burtt Davy Man. Fl. P. & Ferns Tvl. & Swaz. 1, 36.

Main (swollen) aavis tuberlike, flask-shaped, trunklike, fleshy, smooth, partly or mostly

above ground, the latter green, up to over 2 metres, branched or unbranched at the base,

ending in whiplike branches (Bremekamp). Branches climbing, striate, greyish-green,

glaucous : branchlets present in axils of tendrils. Leaves compound, digitately 3-

lobed, rarely 5-lobed, petioled; lobes petioluled to subsessile, the median about 1-7—

6-5 em. long and about the same width, the lateral and basal smaller, simple, entire, rotund,

orbicular or orbicular-obovate, subtruncate, retuse or rounded at the apex; glands absent

on all parts of the leaf including the sinuses of the lobes (? always) with 1 larze subreniform

subpeltate gland at the apex of the petiole ; petiole 1-5-0 cm. long ; stipules, minute, tooth-

like, 4 mm. or less long; tendrils sometimes strong or weak, Bearers breaking off giving

the appearance of thorns. Inflorescence usually on axillary branchlets, in axils a leaves or

tendrils, 1- to 3-flowered, usually alternate, sometimes one or both rudimentary ; peduncle

wanting to 4 mm. long, terminating in a flower not a tendril; tendrils without developed

cymes, in axils ofleaves, with or without bracts. Male flowers not seen, according to Burtt:

Davy: Sepals imbricate (in bud about 8 mm. long). Petals free, membranaceous (about

1 cm. long and 2:5 mm. wide). Stamens 5-6, free; filaments 2 mm. long; anthers about

039

2-5 mm. long, laterally dehiscent. Ovary rudimentary, small. Female flowers “‘ greenish ”’,

yellowish or “ yellowish-green ”, about 8 mm. long. Receptacle subcylindric, 1 mm. long.

Calyx subcampanulate ; lobes semi-erect, 3-4 times as long as the tube, elliptic-oblong, obtuse,

entire. with broad green longitudinal veins. Petals about } the length of the calyx lobes or

slightly longer, arising from the sinuses of the latter but appearing also to be partly inserted

on the tissue joining calyx-tube to staminal column, membranous, transparent, with | broad

green vein, ovate-lanceolate, tapering towards the base in lower third, serrate-uneven in upper

half. Corona a circular lacerated fringe with slender-branched filiform processes about

0-5 mm. long arising from the edge of the calyx-tube and from the tissue joining staminal

tube to calyx-tube. Staminodes connate for half their length at the base, forming a column

around the base of the ovary stalk ; staminal column joined to the calyx-tube at 5 points

by means of fleshy membranous tissue, forming 5 pockets with the free ends subulate and

apices curved. Glands broad, flattened, arising at the base of the “ pockets”. Ovary

stipitate, spherical, smooth ; style short, 3-branched ; branches long, terminating in shield-

like fleshy-papillate structures forming the stigma; ovules few. Fruit yellowish, longi-

tudinally unevenly banded with green, approximately 2 cm. long, roundish, leathery,

dehiscent, splitting into 3 valves ; seeds flat, suborbicular, regularly pitted. Bremekamp in

Vegetationsbilder 1932; 23, 3, pl. 18.

TRANSVAAL.—Pietersburg district: On slopes 2 miles beyond Chuniespoort Hotel,

pale green succulent flask-shaped, stems up to 4-5 ft. high, lounging against Peltophorum,

several branches ascending and climbing in the tree, May 1935, Obermeyer and Verdoorn

10 (N & T). M’Phatlele’s Location, climbing plant with swollen stem, 9.10.19, Pole Evans

in N. 19885, 2. Zoutpansberg district: Dongola Reserve, Messina, 15.9.34, Pole Evans

3747, 2 (N). Lydenburg district: Sekukuni, farm Driekop, dry sandy loam, 3,500 ft.,

17.12.36, Barnard 4548 (N). Ditto, between crevices on “ koppies”’, bole attains size of

a 56-gal. barrel, 4 ft. high, 13.1.36, Barnard 454 (N).

7. A. glauca Schinz Bot. Jahrb. XV. Beibl. 33 Heft 1, 1-3.

Main (swollen) axis tuberlike, ** urn-shaped”’ or irregular-shaped, fleshy, partly or,

mostly above ground, the latter greyish, green beneath the skin, of various shapes and*

sizes, up to 2 ft. (or more ?) high. Branches ? divaricate, striate, glaucous, up to 5 mm.

thick and about 150 cm. long, arising from attenuations of the main stem or directly from

a flat surface; branchlets ? usually few or absent, rarely well developed, resembling the

branches but greener, glaucous. Leaves digitately compound, petioled; lobes 5, occasionally

sub-petioluled, conduplicate, 1-5-6 cm. long, entire, with the margins thinly cartilagenous,

elliptic, sub-orbicular, rotund or obovate, eradually or abruptly narrowed at the base, obtuse ;

glands absent on all parts of the leaf including sinuses of lobes, with 2-paired flaplike

glands} at the base of the leaf; petiole 0-6-5 cm. long; stipules minute, toothlike, dark

reddish-brown. Inflorescence usually axillary, peduncles opposite or alternate with 1- to

2-flowered sidebranches, occasionally many-flowered; peduncles usually terminating in

tendrils, not infrequently in shorter or longer flower-bearing pedicels. Male flowers yellowish,

about 3 cm. long. Receptacle 1-1-3-5 cm. long, usually about 1 em. long, linear-sub-

cylindric, gradually widening towards the apex. Calyx-tube obconical or subobconical,

2-1 the length of the lobes; lobes semi-spreading, linear-oblong, sometimes broadening

towards the apex to linear, or elliptic-oblong, obtuse or subobtuse, entire. Petals more

than half the length of the calyx-lobes, arising from the sinuses of the calyx-lobes but

appearing also to be partly inserted on the tissue joining staminal cup (tube) to calyx-tube,

membranous, transparent, l-nerved or palmately 3-veined, the median vein unbranched

or remotely branched or branches absent with occasional stray veins, narrowly oblong-

elliptic or linear-lanceolate tapering towards the base or ovate-lanceolate or lanceolate at

both ends, acute to obtuse with curved or straight apices and margins remotely or unevenly

* Sometimes with “ neckline ’’ protuberances from which the branches arise. (Smith 6271.)

bracts enveloping a bud.

540

serrate or dentate in the upper third or two-thirds. Corona of a few filiform processes, about

0-5 mm. long (very rarely 1-5 mm. long), arising from around the base of the petals partly

or entirely from the edge of the tissue joining staminal cup to calyx-tube. Stamens much

overtopped by the petals, extending about half way up the calyx-lobes, free portions of

filaments subulate or linear-subulate, of varying length, connective produced into a point,

connate for half to two-thirds their length at the base forming an obconical cup adnate to

the calyx-tube at 5 points producing 5 narrow pockets; anthers short, broad, or linear

oblong, equalling the free filaments in length or occasionally up to 4 or 5 times their length.

Glands 0. Ovary rudimentary, 2 mm. long. Female flowers yellowish, about 1-4 cm. long.

Receptacle approximately 2-4 mm. long, shortly funnel-shaped or narrowly subcylindric.

Calyx-tube subcylindric to cup-shaped, one-third to one-fifth the length of the lobes ;

lobes semi-spreading or subspathulate or obovate or oblong or linear, elliptic-oblong, obtuse

to sub-obtuse, entire. Petals about half the length of the calyx-lobes, arising from the

sinuses of the latter but appearing also to be partly inserted on the tissue joining calyx-tuhe

to staminal tube, membranous, transparent, l-vemed, with scmetimes remote branches or

an occasional stray vein, oblanceolate to linear-elliptic to lanceolate in upper one-third to

one-half, gradually tapering towards the base, acute to truncate-toothed, with curved or

straight apices and margins subentire or remotely serrate-uneven in upper parts. Corona

of a few filiform processes about 5 mm. long, arising from around the base of the petals,

partly or entirely from the edge of the tissue joining staminal tube (cup) to calyx tube.

Staminodes connate for half or more of their length at the base to form a wide tube or collar

around the ovary stalk ; adnate to the calyx-tube at 5 points forming 5 narrow pockets with

the free filaments subulate-tapering, sharply curved at the apices or with rudimentary

anthers. Glands 0. Ovary stipitate, ovoid to orbicular, or rarely somewhat 4-sided, smooth

or with transverse raised bands ; style short, 3-branched, * the branches widening upwards

terminating in a papillate-fleshy stigma; ovules ? few. Fruit ? orange to yellow-coloured,

roundish to ovoid, roundly and shallowly 3 (—4) lobed, leathery, ? usually dehiscent, splitting

into 3 (4) valves. Seeds flat, roundish-heartshaped, regularly pitted. Modecca glauca,

Schinz. Bot. Jahrb. XV. Beiblatt 33, I.

TRANSVAAL.—Waterberg district : Warmbaths, 8.12.04, Burtt Davy 2622, 2 (N & B).

Near Pienaars River, 52 miles west of Warmbaths on hills, Sept. 1932, Smuts 355, Q (N).

Vierentwintig Riviere, Jan. 1920, Rogers in T. 20816. Vygeboompoort, Sept. 1913, van

Dam in T. 13191. Ditto. Oct. 1913, in T. 13715, 9. Farm Roodepoort No. 15, Palala Rd.,

rocky ridge, large epigeal tuber with 18” stems, 6.12.31, Galpin 11606, 2 (N & B). Farm

Doornfontein No. 1807, amongst felsite rocks, fleshy stem 13 ft. high and 6” diameter,

19.2.24, Galpin 9164 (N). Olifants Poort, 9 miles N.E. of Nylstroom, climbing up stem,

of trees, on hillside, 6.12.34, Galpin 13195 (N).+ Potgietersrust district: Farm Nooit-

gedacht near Naboomspruit, amongst ia he mountain top, stems trailing from large

epigeal tuber, 18.16.31, Galpin 11605, ® (N). Potgietersrust, June 1916, Rogers 18827,

3 (N & B). Kwarriehoek School, ne particularly between rocks and extended

rock outcrops, Steyn 37, 5 (N). On Temby Downs, epigeal portion of tuber conical 15”

high, malachite green, stems slender climbing 10 ft. up tree stems, 2,900 ft, 18.11.34, Galpin

13197, 9 (N). Pretoria district : On summit of Daspoort range near Fairy Glen, 4,700 ft.,

§.10.33, Mogg 14130, 3 (N). Wonderboom, Mar. 1924, van Dam in T. 25042. Premier

Mine, Aug. 1924, Verdoorn (N). Ditto, Dec. 1919, Rogers 25027 (T). Ditto, 4,000 ft.,

Menzies. Flats beyond Silverton, 12.10.19, Phillips 3021, 9 (N). Derdepoort, 7.10.28,

Mogg 15386, ) & 9 (N). Magaliesberg, May 1920, Marloth 9508, g (N). Foot of Magalies-

berg on farm ‘ Grafheim ” 4 miles west of Wonderboom Poort, growing on rocky but loose

* In Smuts 355 there were 4 style branches and 4 placentae on the flowers examined.

+ On farm “‘ Grafenheim ”’ along lower N. slopes of the Magaliesberg growing under Khretia rigida

bush in shady places and widely climbing among the branches of its support, rootstock a large tuber of

globose to obovoid shape and partly buried in the loose black sandy soil . . . the exposed part often

acquiring a highly polished surface, being thus very shiny and green . . . “* vern, name “Bobbejaan’”’,

Oct. 1933, Smith 6841 ¢ (N).

541

sandy soil, stems up to 4 ft. with 2 or 3 necklike protuberances from which the stems arise,

up to 50 ib. in weight, stems sprawling over other plants such as Ochna pulchra and Burkea

‘ufricana, 31.7.32 C., 4,200 ft., Smith 6271, 3 (N).* Without precise locality, growing in

Stellenbosch University Garden, Oct. 1928, Marloth in N. 16416,2. Ditto, Div. Pl. Ind.

Gardens, Verdoorn (N). Ditto, from Onderstepoort poisonous plant garden, Nov. 1934,

Inebenberg 3222, 2 (N). Ditto, 1932, Steyn AS; 9.11.32.

8. Adenia digitata (Harv.) Engl. Bot. Jahrb. XIV. 375.

Root tuberlike, subnapiform or variously shaped, appearing above ground (green)

with subterranean obstruction, up to about 30 Ib.? and over. Main stem usually under-

ground, from accumulated remains of annual growths, usually under 20 cm. long, and

15-0 mm. thick, woody. Branches herbaceous, annual, striate, climbing, up to 180 cms.

long and ? more, and about 6-0 mm. thick, usually arising singly from a tuber. Leaves

digitately compound, petioled ; lobes usually 5, occasionally 3, (sub) or pinnatilobed or

pinnatisect sometimes petioluled or simple, narrowly linear or linear lanceolate or ovate or

elliptic-lanceolate, entire, unequal in length ; central lobes 2-0 (? 1-5f)-16 cm. long ; glands

on lower surface circular or slightly oblong, usually present at all sinuses of primary lobes

and of the lowermost secondary lobes of the primary central lobe, sometimes present at all

sinuses of lower secondary lobes, occasionally absent in some of the sinuses of primary

lobes or on the lower surfaces of simple lobes ; petiole 0-5-7-5 cm. long with 2 paired glands

above, at the apex; stipules minute toothlike, rarely up to 3 mm. Inflorescence axillary,

two primary branches opposite or alternate, often 1-few-flowered, usually 4-8 flowered,

rarely many flowered (18-20); peduncles absent or almost so, to 6-5 cms. long, always

ending in tendrils; flowers whitish, bright-greenish, pink tinted or creamy or yellowish.

Male flowers 1-5-3-5 em. long. Receptacle 3-9-5 mm. long, linear- subcylindric, widening

upwards or funnelshaped. Calyx subcampanulate or campanulate; tube subcylindric

widening upwards or obconical sometimes with a slight constriction just above the base,

equal in length to the lobes or up to 3 times their length ; lobes elliptic-ovate or oblong-

ovate or broadly ovate (or-bicular-ovate) or oblong ovate; obtuse to subacute, interior

margins lacerate-laciniate, very rarely entire or subentire. Petals inserted near the base

of the calyx tube, rarely near the middle thereof, extending to the sinuses of the calyx

lobes or a little below or well above ; membranous, transparent, palmately 3-veined with

the sideveins entire or few-branched; oblanceolate- acuminate, oblanceolate or linear-

oblanceolate or broadly oblanceolate or elliptic-ovate with narrowed bases or lanceolate

in upper half, cuneate-tapering in lower half; acute to obtuse, upper 3 or 4 serrate or serru-

late or denticulate (dentate)—laciniate or serrate-denticulate, rarely entire or subentire.

Corona sometimes absent or nearly so, of slender processes 0-50-75 mm. long, rarely over

1 mm. ; arising in a continuous sinuate ring rarely in groups from between the bases of the

petals, sometimes subpapillate at the upper ends. Stamens usually overtopped by the calyx

tube and petals, extending below or beyond the sinuses of the calyx lobes, connate for half

their length, rarely up to 2 their length ; the filamental column adnate to the calyx tube

in its lower half very rarely only at the base or for its full length, producing 5 pockets ;

anthers shorter or longer than the filaments, apiculate, loosely adhering at these points.

Glands usually about 0-50-75 mm. long, arising from the base of the “ pockets ”, flattened,

capitate, variously shaped. Female flowers 1:5-2-5 cm. long. Receptacle 2-6-5 mm.

long, subcylindric widening upwards or funnelshaped, sometimes also widened at the base.

Calyx campanulate ; tube obconical or subcylindric widening upwards, as long as the lobes

or up to double their length ; lobes ovate or broadly oblong-ovate, or narrowly oblong or

* Without precise locality, Waterberg dist. vine w. large thick tuberous rootstock, 9.11.28. Repton

96 2 (N).

+ Some flowering specimens have central lobes 1-5 cm. long, though it is not known whether they

are full grown. In the Fehrson specimens (type of A. multiflora Pott.) the leaves are much shorter.

tIn Byles 3243 in Herb. Bolus (from Rhodesia), Fig. 15, the inflorescence is approximately

40-flowered.

542

broadly lanceolate, acute to obtuse with interior margins entire, very rarely subentire.

Petals inserted near the middle or near the base of the calyx tube, extending to the sinuses

of the lobes or below or above ; membranous, transparent, 1 veined, very rarely 3-veined,

oblanceolate or lanceolate to ovate-lanceolate in upper 4 with lower 4 gradually tapering

towards the base, linear-lanceolate, narrowed at the base, or linear-oblanceolate or narrowly

elliptic, entire, sometimes remotely toothed towards the apices, very rarely laciniate in

upper 3 or 4; acute to subobtuse or acuminate. Corona rarely absent or nearly so, of slender

appendages about 0:5-0-75 mm. or less long, arising in groups in a circle or in a sinuate

ring from near the base of the calyx tube or higher, sometimes subpapillate at their upper

ends. Stamodes partly connate, the free portions linear-subulate or subulate, or linear-

sub-spathulate, as long or longer than the rest, curving outwards and inwards or vice versa,

with apices incurved, shorter, as long as or longer than the ovary stalk ; the staminal collar

adnate for a short distance or more of its length to the calyx tube forming 5 pockets or

depressions. Glands flattened, capitate, variously shaped ; stems erect, rarely recurved,

short and broad or long, heads usually large, concave or level above, rarely bilobed. Ovary

stipitate, ovate to obicular or oblong or ellipsoid, smooth or prominently veined, rarely

furrowed or ridged or uneven; style 3-branched, very rarely 2-branched, the unbranched

portion up to % the length of the ovary or almost absent s stigma pufflike, woolly fleshy ;

ovules numerous. Fruit “orange” “ brilliant orange’, “yellow’’? ‘crimson’, “ red-

purplish ” ovoid-oblong or ellipsoid, 3-valved, dehiscent ; seeds flattened, roundish-sub-

pearshaped, regularly pitted. Adenia senensis (KI. ) Engl. Bot. Jahrb. 14 (1892), 375. Adenia

digitata Burtt Davy Ann. Transv. Mus. 111. 121.—Modecca digitata Harv. Thes. Cap. p.8.

A. multiflora Potts Ann. Transv. Mus. V. 235. Clemanthus senensis Klotsch, Peters Reise

Mosamb. Bot. 143. Modecea senensis Mast. Olw. Fl. Trop. Afr. 11,517. Adenia angustisecta*

Burtt Davy. Kew. Bull 1921. 280. Adenia stenophylla Harms. Eng. Bot. Jahrb. XX VI,

238. Adenia Buchananti Harms. ex Engler in Engl. Pflanzenw. Afr. 111, 2. (Engl. and

Drude Veg. d. Erde IX) 605, (1921) in obs.

TRANSVAAL.—Pretoria district: Muiddelkop farm near Pienaars River, C. 3680 fts

Jan. 1926, Smith 2120, 9 (N). Hartebeestpoort, Jackson, 2 (N). Rooikop, 5.1.36, Smut.

and Gillett 3450, 29 (N). Pienaars River Station, Oct. 1932, Osborne in N. 2639, ¢ (2 sheets).

Without precise locality, 23.1.23, Osborne in N. Rust- der-Winter, Jan. 1936, Pole Evans

3886, 9 (N). Strubenskop, 18.1.36, Munroin N, 2. Farm Zeekoegat, Swingbridge, 27.1. 34,

Schweickerdt 1090, 9 (N). Bon Accord, W. Pyramid Hill, 12 M. north of Pretoria, 4,200 ft.,

6.4.32, Mogg 12388, 9 (N). From Onderstepoort Poisonous Plant garden Noy. 1934, Lieben-

berg 3224, 9, 3223, (n). Pretoria, Febr. 1912 Rogers in T. 12041. Without precise locality

Magaliesberg Zeyher (S.). Brooklyn, Mar. 1914, Pott. 4826 (T). Hammanskraal, hard

deep compact clay, 17.10.34, de Lange 76, § (N). Baviaanspoort, Febr. 1914, Fehrson in

T. 13768, 3 (T.), 3 sheets. Hammanskraal, red gravel, 17.10.34, von Malititz, 60 (N). Pot-

gietersrust district ; ? at Potgietersrust 3.11.08, Leendertz 6007, 3 (A & T). 4% Potgietersrust,

21.12.28, Govt. Analyst in N. 7817, 9. * Mosdene ” near Naboomspruit, loam formation

climbing over bushes, 10.11.19, Galpin 477 M, 3 (N). Ditto, 25.2.19, Galpin 142 M, 2

(N & 8). Ditto, in Acacia veld stems 1-2 M. arenate, climbing up bushes, fls. creamy tuber

very large, 3,800 ft. 21.11.34, Galpen 13196, 2 (N.) Rustenburg district ; Brits, 29.12.27,

Watt and Brandwyk 2045, 2 (N). Ditto, 17.11.27, Watt and Brandwyk 2038, 2 (N). Farm

Welgevonden, 3,200 ft. 8.12.34, Mogg 14609, 3 (N). Brits, 20.12.27, De Ridder in N. 7504,

9 (2 sheets). Middelburg district ; Niebo, Oct. 1921, Rogers 24853, § (T). Potchefstroom

district ; on experimental farm, 5.1. 31, Theron 5, 2 (N). Bechuanaland ; Mochudi, May

1914, Rogers (W). Saberones, 10.12.36, Watt and Brandwyk 1683, 9 (N). Mochudi, Jan.

1915, Harbor in T. 17027, 9. Lydenburg district ; Farm Schoonoord, Sukukuni, poisonous,

black clay soil between norite boulders, 4,000 ft. 8.3.37, Barnard 229 A. Foothills, Camp

* A. augustisecta Hngl. & Harms ex Kngler, Pflanzenwelt Afr. III, 2 (1921), p. 605, in obs. was the

original name for A. stenodactyla Harms which now stands as Burtt Davy published his description first,

necessitating Harms to change his epithet,

543

HB. Schoonoord 4,500 ft., 3.12.33, Barnard 229, 9. Farm Korenvelden near Sukukuni,

tuber topshaped, poisonous, 3,500 ft., 3.12.34, Barnard 153 (6 sheets). Ibid, small bulb,

poisonous, used by natives, 3,000 ft., 8.11.34, Barnard 155. Ibid, 3,000 ft., 13.3.35, Barnard

305, 9, allin N. Ohrigstad valley, 23.10.08, Mundy, in Bolus Herb. 4,700, 9. Barberton

District ; without precise locality (? at Barberton), Nov. 1931, Smith 7019, 2 (N). On

road to Carolina (? from Barberton) + 4,000 ft., 5.10.30, Bremekamp in T. 28575, gd. Glen-

thorpe farm, 1.11.11, Scheuble in T. 10908, g. Without precise locality (? at Barberton)

Jan. 1908, De Beer in T. 4940, 3. Dry hillsides at Barberton, twining on shrubs, 2,000—

2,800 ft., 1889 A.p., Galpin 677, 3 & 2 (N. 2 sheets), T &8.). ? at Barberton, 2,900 ft.,

21.11.88, Thorncroft 15, 3 (B). Ditto, Nov. 1909 Willams in T. 7645, 9. Hyslops Creek,

trailing, 2,500 ft., Apr. 1926, Thorncroft 2141, ¢ (N). Nelspruit district ; Northern slopes

of Amajuba mt., Schagen, climbing on trees, etc. stems several ft. long, 3,000 ft., Dec. 1934,

Inebenberg 3362, 9 (N). Ditto grown at Div. Pl. Ind. gardens, Pretoria, 19.1.37. Same

locality, long stems, climbing, bulb 10-12 cm. diam. turbinate, 3,000 ft., 28.12.33 Lieben-

berg 3056, 9 (N). At Schagen just off main road, 16 M. from Nelspruit, tuberous topshaped

rootstock 15 cm. diam., climbing, stems several ft. long, 2,500 ft., Dec. 1934, Laebenberg

3301, ¢ & 2, (N). Ditto, 2 grown at D.P.I. gardens Pretoria, 19.1.37. From same locality

but grown at D.P.I. gardens, Lxebenberg 3055, 9, (N). On farm Suidwalliskraal, on main

road Nelspruit-Machadodorp, stems several ft. long, climbing, 2,800 ft., Dec. 1934, Ineben-

berg 3366, 2 (N). On Research Station, Nelspruit, long stems climbing up trees, 2,375 ft.,

1.11.30, Liebenberg 2544, 2 (N). Plaston, climbing herb, 3,000 ft., Oct. 1931, Holt 75, $ (N).

White River, Oct. 1919, Rogers 23288, 2 (U & N). Doornkraal, 28.12.23 Stubbs 2 (N).

Mayfern, (grown at D.P.I. gardens, Pretoria), April 1929, Mogg 8182, 2 (N), (2 sheets).

Pietersburg district ; Tzaneen, 8.12.36, Hattingh PS. 227, 2 (N). Rooikoppies, Politsi,

climber open parts of forest, Dec. 1932, Schweickerdt 1037, $ & 9 (N. & T.), 4 sheets. Wood-

bush, de Hoek, Dec. 1931, Schweickerdt in T. 30164, J, 2 sheets. Haenertsburg, Noy. 1913,

Pott in T. 13368, 3. Westfalia Estate, 17.11.36, Pole Evans 3983, (N.). Haenertsburg,

Nov. 1917, Moss and Rogers 884, 2 (U.), 2 sheets. Zoutpansberg district ; Elim, Dec.

1930, Obermeyer, in T. 29287 and 29288, 9. Louis Trichardt, Dec. 1922, Breyer in T. 24215,

3g. Tshakoma, Nov. 1931, Obermeyer in T. 30349, 9. Pigeon Hole, 28.10.18, McCallum

70, 2 (N). The Downs, Nov. 1918, Rogers 21937, 9. (A, B. & T.). Middelburg district,

Tautesberg, 9.11.33, Young A. 247, 3 & 2 (T), 3 sheets. Natal: Vryheid ditricts; Oct.

1905, Sim 2922, 2 (B), 2 sheets. At Roman Catholic Miss. Stat. on Inkawana, on the

Vryheid side of Besters Spruit, between diabase, strictly dioecious, 3 ft. high, 3 flowers

yellow, 2 flowers green, poisonous, abundant. 4,000 ft., 20.10.37, Gerstner 2330 and 2331.

36 & Q(N.). Nongoma district ; at Nongoma, native name Umbulele (= poisonous plant),

20.1.38, Gerstner 2896 (N). Kruger National Park ; Baiandbai, 24.11.32, Zang in T. 32154,

6. Ditto, 25.11.32, Lang in T. 32153, 9.

1], BAS CHERSONE Gero cei eer

2S VB AMON way eae neers

3. BrentHam & HOOKER..........

4, BREMEKAMP, C.H. B...........

He IDAMAG dy IBURMEG coaccegc00000

GAVENGEERSEACR. Oren coe ooniae

ELAR MS DEltts arntiaects rem aratosorereete

8. —

9. ae

10. —

iil. —

12. —

135~ BODTS eR astae oieies cates eee ee

14°, WROBMBR;, [Mitscecrc oie cree amore

NSE SCHINZae EL pyeeeineoin rotors

16. —

Ft SMUT coisa ee eetsrruclereteette crete

185 (SEBYNS DAG) ssa ere eer

19. Watt J. M. & BreyER-BRAND-

Wyk, M. G.

20. Wicut & ARNomD...........--

544

LITERATURE CITED.

Baillons’ Dictionaire, p. 47 (1876).

Natural History of Plants (1888) VIII 476.

Gen. Plant. 1, III, 1867, p. 813.

Merkwitrdige Sukkulententypen aus dem Nordlichen Transvaal.

Vegetationsbilder 23 (1932) 3, p. 6. With plate.

Transvaal Agric. Journ. VII (1908-9), P. 570.

Passifloraceae Africanae Bot. Jahrb. XIV (1892), 374-87.

Uber die Verwertung des antom. Baues fiir die Umgrenzung und

Hinteilung der Passifloraceae. Bot. Jahrb. 15 (1893), 548-633.

Passifloraceae. Natiirliche Pflanzenfam. First Ed. (1895) IIT.

Abt. VI und VIA. 69-94.

Ditto. Nachtrag u. Register zu Teil II-IV, p. 255 (1895).

Zur Morphologie der Ranken und Bliitenstande bei den Passi-

floraceen. Bot. Jahrb. 24 (1898), 163-78.

Passifloraceae. Engler’s Pflanzenwelt Afrikas III (1921).

Passifloraceae. Natirl. Pflanzenfam. 2nd Ed. Vol. 21 (1925),

470-507.

A Contribution to the knowledge of the Transvaal Passifloraceae.

Ann. Trans. Mus. (1917), 235-7.

Synops. Monogr. II (1846).

Beitrag zur der Kenntniss Afrik. Passifl. Bot. Jahrb. 15 (1893).

Beiblatt. 33 pp., 1-3.

Beitrag zu der Flora Deutsch 8.W.A. ITI.

Gramm. of Bot. 1888 (1821).

The toxicology of plants in South Africa.

S.A. 1934. p. 310-15.

The Medicinal and Poisonous plants of S.A., E. & S. Livingstone

Edinburgh, 1932, p. 122.

Prodr. fl. Penins. Ind. Or. I, 353 (1834).

Central News Agency.

545

Plate 1.—A. fruticosa Burtt Davy. At Naauwpoort, near Pietersburg, Transvaal.

[Photo by H. Lang.

Flate 2.—A. Fruticosa Burtt

Davy.

546

At Naauwpoort, near Pietersburg, Transvaal.

[Pholo by H. Lang.

547

‘buvy “A fig coy)

*[eRASULIT, “‘Sanqsiojolg vou

‘

qIOOMAMNVEN FV

Aarqg 3

Ing vsouids

—"¢g 998

548

Plate 4.—A. glauca Schinz. From Fairy Glen, near Pretoria. The light portion and

below this was underground,

|Photo by H. King.

549

Plate 5.—Flowers of A. glauca Schinz. [Photo by H. Lang.

Plate 6.—Fruits of A. digitata (Harv.) Engl. [Pholo by H. Lang.

550

one?

Plate 7.—Two tubers of A. digitata (Harv.) Enel., found side by side at Baviaanspoort, outside

Pretoria. The one was partly exposed and green; the other was underground.

[Photo by H, King.

55]

Plate 8.—A photo of the type specimen of A. digitata (Harv.) Engl.

{Photo by courtesy of Kew Herb.

552

“Bs

j en ik. «+f ie

Pacre fone y “

Ve )

Ver amen A ran,

i re beg Mh :

Moan big (de nen ey Lees ae,

Plate 9.—Photo of one of the type specimens of Clemanthus senensis Kl. ([A. Senensis

(KI.) Engl.]. These specimens were collected at “ Rios de Senna” and therefore was

very likely the same locality where Kirk later collected his specimens (plates 12-14)

which Masters described as A. senensis. Mast.

[Photo by courlesy of Berlin Herb.

503

J. is suenihte al,

ine ee Y Ze» HUNG ii

Plate 10.—Photo of one of the type specimens of Clemanthus senensis KI.

[A. senensis (K1.) Engl. ].

[Photo by courtesy of Berlin Herb.

5d4

HERB. AUST \FRIC, R SCHLECHTER

Awa Bohan. Han |

acc. ea l¥. 196% 5

I AD,

e ee

Plate 11.—Photo of the type specimen of 4. Buchananivi Harms.

[Photo by courtesy of Berlin Herb.

595

Plate 12.—Photo of one of the specimens collected by Dr. Kirk at Senna,

Zambezi, and from which Masters (Fl. Trop. Afr. Vol. 2) described A. senensis

Mast. Practically all lobes are entire. [Photo by courtesy of Kew Herb.

}

596

fype Specimen.

TEES}

Plate 13.—Photo of one of the specimens collected by Dr. Krik at Senna, Zambezi,

and from which Masters (I. Trop. Vol. 2) described A. senensis Mast. Practically all the

median and side lobes are lobed.

[Photo by courtesy of Kew Herb.

557

lea. Eopeditien

bs 3 “a

dn ee

Plate 14.—Photos of one of the specimens collected by Dr. Kirk at Senna,

Zambezi, and from which Masters (FJ. Trop. Afr. Vol. 2) described A. senensis

Mast. The median lobes are only slightly lobed. z

[Photo by courtesy of Kew Herb.

508

Flora Africae australis. ——————}

Lanepkylla Harry

Dr brso)

: . at. Lap ervens ( AXS SE,

oi GH pe rin

, Habitat rempublic. Transvaal. |

District Lydenburg,

A, yea eT | nani ica CMR | : Z be

Let 18 24, 3 AUG. 1 Wi, Pr F. Wilms ee) | : [ats bot. By

Plate 15.—Photo of the type specimen of A. stenophylla Harms.

Photo bu courtesy of Berlin Herb.

509

ObrgAled alls F hamlieny

td 1d y

Bek gard or

Sn HOR Pmt,

Plate 16. Plate 17.

Plate 16.—A photo of Mundy in Bolus Herb. 4700, 2, from Ohrigstad valley. It is the type specimen

of A. angustisecta Burtt Davy. In morphology of floral structures it resembles A. digitata and in leaf form

it links up with other Lydenburg specimens (plate 17, etc.).

Plate 17.—A photo of Barnard 153 (1), showing 8-1 and 5-lobed leaves (entire).

(Photos by H. King.

560

Plate 19.

Plate 18.

Plate 18.—A photo of Barnard 155, showing all 5-lobed leaves (entire). A. stenophylla Harms

(plate 15) fits in here.

Plate 19.—A photo. of Barnard 155 (collected leaves). All Barnard specimens are from Sukukuni,

not very far from Lydenburg.

[Photos by H. King.

Plate 20. Plate 21.

Plates 20 and 21.—These are photos of other Barnard specimens showing various degrees cf lobing

in the direction of the typical A. digitata.

[Photos by H. King.

WERIARIEM TRANSVAAL MUSEEM.

No, = SOLO

Rdarwins Unptaha (Hard

‘ Kaew

Prawns Paya PR RR

Plate 22. Plate 23.

Plate 22.—This is a photo of another Barnard specimen showing various degrees of lobing in the direction of

the typical A. digitata.

Plate 23.—A photo of Schweickerdt in Trans. Mus. Herb. 30164, 3. In leaf-shape it resembles the previous.

The inflorescence is more elaborate. {Photo by H. King.

563

Plate 24. “=e

Plate 25.

Plate 24.—A photo of Liebenberg 3362. This links up with plate 23.

Plate 25.—A photo of Liebenberg 3366. This is from the same locality as that of plate 24. but from a

different habitat, showing no lobing. {Photo by H. King.

564

Plate 26.—A photo of a specimen from the same tuber as Liebenberg 3366 (plate 25) but grown at

the gardens of the Division of Plant Industry, at Pretoria. The specimen shows characteristic lobing.

It is identical to Liebenberg 3362 when grown at the D.P.1. gardens, but the latter was not used as it would

not have made such a good photo,

| Photos by H. King.

565

1 er

Mordecen dy bln Meas.

a AC ESE 3 ry

Plate 27. Plate 28.

Plates 27 and 28.—Photos of Galpin 677 (2 sheets), showing variation of the lobing. These link

up with plate 26.

[Photos by H. King.

566

btm Sed

a Pe

Herb. 7645, 9.

These link up with plate 28. Compare these with the photo of the type specimen (plate 8).

[Photos by H. King.

567

ip,

Be 1 pores eee

nae t i HERBARIOM i

Pie # THE TRANSVAAL MUSEUM,

t ‘ }

| Hi at

H t

ba |

1Oeli 4, e

Wee ll ares aye

Plate 31.—Photo of Scheuble in Trans. Mus. Plate 32.—Photo of Bremekamp in Trans. Mus.

Herb. 10908, 3. Herb. 28575, 4.

These link up with plate 18 on the one hand and with plate 25 on the other hand. The link between

plates 32 and 35 is not very clear but is due to the fact that Licbenberg 3301 was omitted to reduce the

number of plates.

[Photos by H. King.

568

WARE GOVERNMERT HERBARIOM,

4002, 4

i Maw (ht Yellen At ® '

Ficus?” Seabled ;

M Cisic oy orem Mewndtraad |

t Bartclen 10S .

MMA AGO ee lx a ;

WN face ee A oe foarsncs Re

Plate 33.—A photo of Thorneroft 15, 3. This links

up with Liebenberg 3366 and 3362 (plates 24 to 26).

[Photo by H. King.

569

TEED ARIOM 4

Plate 34. Plate 35.

Photo of Obermeyer in Trans. Mus. Herb. 29287, 2

Photo of Obermeyer in Trans. Mus. Herb. 30349. 2

These link up with plate 33 and is where A. senensis Masters (plates 12, 13 and 14) fits in.

Compare the flowers on plate 34 with those on plates 33 and 23.

570

Pham Fu shdmaee dimen

Uy Ry ats Gu ha

Mauerabnnt se) hare FY

Determnioavit AG, hice dude wee,

Plate 36.—Photo of Barnard 229, 9. This links up with plates 34 and 35

and this is clearly also where 4. Buchananii Harms (plate 11) fits in.

571

A REVISION

of the

SOUTH AFRICAN SPECIES OF HYPERICUM

H. C. Bredell.

The first reference to a South African species of Hypericum appears in Thunb. Prod.

Fl. Cap. 139 (1800) where a concise description of H. aethiopicum Thunb. is given. In DC.

Prod. I, 550 (1823) another species H. Lalandii Choisy is enumerated and is also described

in the Fl. Trop. Afr. 1: 155 (1868). In Fl. Cap. 117 (1859-1860) only three species were

enumerated and a few years later Wood & Evans described H. natalense from Natal (Journ.

Bot. XXXV: 487 (1897). In Bull. Herb. Boiss 179 Sér II, VIII, 1908, H. Wilmsii Rob.

Keller appears in the author’s key as a new species recorded from the Transvaal. Keller

probably overlooked the description of H. natalense and described the same plant as H.

ood in Engl. Jahrb. LVIII 193 (1923). The most recent revision of Tropical species

“ppeared in Journ. of Bot. 329 (1927), where Good included the two South African shrubby

species H. leucoptychodes Steud. and H. Roeperianum Schimper.

In the present revision only the South African species of the genus have been dealt

7ith but as some species extend southwards from Tropical Africa, accounts of other species

from this region have also been studied and compared with the South African plants.

(Unfortunately most of the work has been carried out on dried specimens, fresh material

of most species not being available at the time.) The large number of specimens investi-

gated gave a good idea of the variability of most of the species and the search for constant

specific characters in closely allied species proved to be a difficult task.

General Notes and Distribution.

South African species of Hypericum are recorded from South West Africa, the Cape

Province extending inland and along the east coast as far north as Tropical Africa. Most of

the species show an overlapping in distribution but others appear to be restricted to certain

areas with the same climatic conditions. The variations and distribution of the different

species are not only interesting from an ecological point of view but have proved useful in

determining some of the species. In the accompanying maps the distribution of each

species is shown. .

572

Of all the species examined H. Lalandii Choicy (Fig. 1) shows the greatest range of

variation in shape and structure of the vegetative parts. In almost every account of this

species the authors refer to small plants + 10 cm. high with very short leaves. According

to Keller some forms are unbranched whereas others are branched from the middle. He

also noticed the variations in size and shape of the leaves and showed that the typical form

has the stems unbranched, medium-sized leaves (1 em. long, 1-2 mm. broad) and flowers

1-5 cm. in diameter. Most of the specimens studied, fitted these earlier descriptions but a

number of plants were different from the typical form. In plants branched from the middle,

the leaves were usually larger, somewhat rigid in texture and more or less glaucescent.

The specimen with the largest leaves Payn 33 collected at Tsolo (Tembuland) had some

leaves which measured 3-7 cm. long and 1-7 cm. broad.

The varieties lanceolata and latifolia described by Sonder and the variety lanceolatum

described by Keller are chiefly distinguished from the typical form by variations in leaf-

size. Many of the specimens fitted these descriptions very well, but a number of successive

transitional forms between the short, and long-leaved, narrow, and broad-leaved forms

rendered it impossible to separate these extreme forms and to group them into the above

varieties (Fig. 1, a-h). In the variety macropetala described by Sonder, the petals are twice

as long as the sepals a character which after careful measurement and comparison of the

relative length of the sepals and petals in almost all the specimens, does not seem to be

constant in any group of similar plants within the species. In the writer’s opinion all

plants, exhibiting these varying characters would-be better considered as forms of the

typical H. Lalandit with the variation in léaf-size probably due to the influence of a

changed habitat ; some growing in open and some in sheltered places. It is interesting to note

that most of the broad-leaved specimens were collected near and along the east coast of

Natal. A few specimens recorded from the eastern Transvaal which differed remarkably

from the rest in having rigid glaucescent leaves, with the stems and leaves densely and

prominently yellowish pellucid-dotted have been described as a new variety transvaalense.

These characters are very constant and no intermediate forms were found.

The species is widely distributed from the southern Cape, along the east coast and

inland as far north as Southern Rhodesia and is also recorded from South West Africa.

H. aethiopicum Thunb. (Figs. 2, 7 and 9) may be distinguished from H. Lalandw by

its round stems, ovate leaves and black-dotted sepals and petals. The variety glawcescens

described by Sonder on characters such as dwarf erect stems, glaucescent closely-set black-

dotted leaves longer than the internodes, cannot be upheld because all these characters

proved to be very variable in the great number of specimens examined. Moreover, the

glaucescent leaves present in a few specimens do not justify a separation of these from the

others because this character cannot be correlated with any other constant feature. This

species extends south of north eastern Cape (Matatiele) as far as the Riversdale district.

H. Sonderi Bredell sp. nov. (Fig. 6) is closely allied to H. aethiopicum but differs from

it in having spotted stems and no stalked gland-like protuberances along the margins of

some bracts and sepals. All plants exhibiting the above characters have been previously

named H. aethiopicum. In the description of H. aethiopicum Sonder refers to the toothed

sepals as being an outstanding characteristic of the species. It is also interesting to note

that there is almost no overlapping in the distribution of the two species. H. Sonderi

extends southwards only as far as the north eastern Cape (Aliwal North) and H. aethiopicum

is not known to occur north of Matatiele. A few specimens collected in the Pietersburg

district, Transvaal, and which I have described as a new variety, transvaalense of

H. Sonderi exhibit characters which link with H. aethiopicum and H. natalense. The

flowers resemble those of H. aethiopicum in almost every detail but the leaves are much

more like those of H. natalense being slightly membranous but larger.

573

H. Wilmsii R. Keller of which I have seen the type, shows an affinity with H. aethiopicum

(Figs. 8 and 10) but the plants are usually small and more or less procumbent. Characters

such as the oblong rounded sepals and the presence of black dots along the margins of the

sepals and petals also links the species with H. natalense. This species occurs on mountains

in the northern Cape, Basutoland and Transvaal. In the Flora Capensis Sonder referred

Drége 7530 in Herb. Sond. collected at the Cape without precise locality, to H. humifusum

Linn., a typical European species. He also states that the Cape specimens of this species

are taller than usual but not otherwise different from the species. At Kew Dr. H. G.

Schweickerdt very carefully compared Drége 7530 with the typical H. humifusum, and

came to the conclusion that Drége’s plant is not H. humifusum and that it is the same species

as Dieterlen 1222 cited under H. Wilmsii in this revision. The European plant does there-

fore not occur in South Africa.

According to Medley Wood and Maurice Evans who described H. natalense (Figs. 3;

4, 8, 10) this species “ has much the appearance of H. aethiopicum for which it has doubtless

been mistaken ; but it differs in having more numerous stems from the root . . . leaves

which are subsessile and not amplexicaul, flat not revolute edges, sepals which are not

lanceolate or acute ; in the absence of black dots from the sepals, petals and anthers and

by its 5 styles and 5-celled capsule.” These observations were based on the type Medley

Wood 4034 of which I have seen two duplicate specimens. After a careful examination of

these it was found that some leaves showed revolute margins, at least in the dry state, that

a few black dots are present along the margins near the apex of at least some petals and

sepals, that the anthers are furnished with black dots and that the number of styles and

chambers in the ovary may vary from 3 to 5. All the characters were confirmed in sub-

sequent examinations of a number of specimens which undoubtedly are the same as Medley

Wood 3034. These characters probably had been overlooked by the authors. Robert

Keller overlooked the description of H. natalense and described a new species H. Woodii

from Medley Wood 3034. In the description of H. Woodii Keller showed that the number

of styles and chambers in the ovary may vary from 3 to 5 and remarks that the species

“nimmt . . . innerhalb der Subsectio Homataenium durch die Veranderlichkeit in der

Zahl der Fruchtblitter eine Sonderstellung ein”. Although the species appears to be mostly

confined to the province of Natal it occurs south as far as the Kentani district.

A specimen collected at Camperdown by Franks (Government Herb. Natal 12968)

resembles specimens of H. natalense in most respects but the leaves are obovate with a shor

but distinct petiole. I have described this specimen as a new variety petiolatum (Fig. 4).

H. leucoptychodes Steud (Fig. 5) and H. Roeperianum (Fig. 11) Schimper differ from

the other South African species of Hypericum in being shrubby with large flowers and

relatively large leaves. Good in his account on the shrubby species of Hypericum of Tropical

Africa pointed out that the best characters for primary classification are those of leaf-

features, particularly the type of venation as seen on the under side of the leaf. H.

leucoptychodes may therefore be distinguished from H. Roeperianwm in having the tertiary

venation slightly conspicuous giving the effect of small striae and pellucid veins; the

tertiary venation in the leaves of H. Roeperianum is closely reticulated so as to form small

but distinct meshes. According to Good the styles are almost united to the tip in H.

Roeperianum. I have seen Eyles 795 and Teague 214 cited by him under H. Roeperianum

and in neither of these could partly free styles be found. Subsequent investigations of a

- great number of specimens belonging to this species showed that the styles are always connate

to the tip, with the stigma distinctly 5-lobed. Of the two species, H. leucoptychodes appears

to have the wider range in South Africa, extending southwards from the tropics as far as

north-eastern Cape. In South Africa H. Roeperianum is only recorded from the Lydenburg

district of the Transvaal.

574

ACKNOWLEDGMENTS.

I am greatly indebted to the Chief, Division of Plant Industry, Dr. I. B. Pole Evans

and the Principal Botanist, Dr. E. P. Phillips for granting me the privilege to undertake

this revision and for the valuable assistance and guidance which I have received from the

latter. Dr. H. G. Schweickerdt was kind enough to compare certain specimens with the types

at Kew and to advise me on some new South African species described by Rob. Keller.

I have to acknowledge with thanks the loan of specimens from the Transvaal Museum,

Pretoria ; the Herbarium of the South African Museum, Capetown; the Herbarium of

of the Albany Museum, Grahamstown ; and the Natal Herbarium, Durban.

The following symbols accompanying the citations indicate the different herbaria

where the specimens are kept.

A. Albany Museum, Grahamstown.

C. The Herbarium of the South African Museum, Capetown.

P. National Herbarium, Pretoria.

G. Galpin’s Herbarium in P.

M. Marloth’s Herbarium in P.

N. Natal Herbarium, Durban.

T. Herbarium of the Transvaal Museum, Pretoria.

DESCRIPTION OF THE GENUS.

Shrubs or perennial herbs from a woody base. Leaves opposite, sessile, amplexicaul or

shortly petiolated, entire, mostly pellucid-dotted, with or without reticulated veins.

Stipules 0. Inflorescence a loose or compact cyme or flowers terminal on short lateral

branches. Flowers yellow, bisexual, regular. Sepals 5, equal or subequal, acute or rounded

at the apex, pellucid-dotted, with or without glandular stalked protuberances along the

margins. Petals 5, twisted in bud, distinctly veined with black spots on the surface and

margins or along the margins only. Stamens indefinite. Filaments thread-like, free or

connate at the base into 3-5 distinct or indistinct groups. Anthers with or without black

or yellow spots. Ovary superior, sessile, ovate to subround in outline, 3-5 chambered,

with few to many ovules on parietal placentas; styles 3-5, free, or connate to the tip :

stigmas small, capitate. Fruit a capsule with few to many seeds,

515

KEY TO THE SPECIES.

A! Stems 4-sided or 4-ribbed ; sepals and petals without black dots.

al Leaves and stems very prominently and distinctly yellowish pellucid-dotted ; leaves linear-

lanceolate, glaucescent. H. Lalandii var. transvaalense.

a? Leaves and stems variously dotted but not as in var. transvaalense ; leaves linear to ovate-

lanceolate, glaucescent or not. 1. H. Lalandii.

A? Stems round and smooth ; sepals and petals usually black-dotted.

b! Herbs or half-woody plants; flowers small, stamens connate at the base into 3 or 4 irregular

groups.

mo! Scr pals linear or linear-lanceolate, acute to sub-acute, equal; petals always black-dotted

on surface and along the margins.

d! Bracts and sepals without marginal stalked bodies; stems usually spotted,

occasionally without spots.

e! Leaves membranous, prominently dotted on abaxial surface; tertiary

venation very distinct to form small meshes. H. Sonderi var. transvaalense.

e? Leaves not membranous, variously dotted; tertiary venation absent or if

present not conspicuous on both surfaces. 3. H. Sonderi.

d? Bracts and sepals (at least some) with marginal black-tipped protuberances ;

stems spotless. 2. H. aethiopicum.

c? Sepals oblong to obovate, rounded at the apex, unequal; petals usually with a few

black marginal spots near the apex.

f! Leaves sessile, elliptic to elliptic-oblong.

g Plants 30-40 cm. high, erect ; leaves 1-5-2 cm. long.

5. H. natalense.

g? Plants less than 20 cm. high, more or less procumbent; leaves

usually less than 1 em. long. 4. H. Wilmsu.

f? Leaves shortly petiolated (especially on young branches), obovate.

H. natalense var. petiolatum.

b? Bush or shrub; flowers large; stamens connate at the base into 5 distinct groups opposite

the petals.

h! Styles connate to the tip with the free branches 1-2 mm.

long; leaves 1-5-3 « 0-3-0-9 em., narrowly lanceolate.

6. H. leucoptychodes.

h? Styles connate to the tip with the stigma 5-lobed; leaves

4-7 X 1-5-2-5 em., elliptic-lanceolate.

7. H. Roeperianum.

1. H. Lalandii Choisy. in DC. Prod. 1: 550 (1823); Fl. Cap. 1: 118 (1859-1860); FI.

Trop. Africa 1: 55 (1868); Bull. Herb. Boiss. Sér II, VIII: 187 (1908); Engl.

Jahrb. LVIIL: 197 (1923); Enel. & Prantl. Nat. Pflantzenfam. ed. 2, XXI: 181

(1925). H. Lalandii Choisy. var. lanceolata, var. latifolia and var. macropetala of Sond.

in Fl. Cap. 1: 118 (1859-1860); H. Lalandii var. lanceolatum Rob. Keller in Bull.

Herb. Boiss. Sér IT, VIII: 187 (1908).

Semiherbaceous, 6-55 em. high. Stems from a woody underground rootstock, erect

or decumbent below, branched or unbranched, 4-ribbed, glabrous. Leaves sessile, ascending,

0-5-3-6 em. long, 0-2-1-5 em. broad, linear to ovate-lanceolate or oblong-elliptic, obtuse

to sub-acute, entire, with numerous more or less translucent dots, 1- to 7-nerved, glabrous.

Flowers few to many in terminal cymes. Bracts 1-9 mm. long. Sepals 2-10 mm. long,

1-5-3 mm. broad, limear-lanceolate, subacute. Petals 3-5-12 mm. long, 1-5-5 mm. broad,

oblong to oblong-spathulate, distinctly veined. Stamens indefinite, 2-5-7 mm. long ;

filaments thread-like, free or connate at the base into a few irregular groups; anthers

0-25-1 mm. long. Ovary 1-5-8 mm. long, 1-5 mm. broad, 3- to 4-chambered : ovules

many ; styles 3-4, free, 1-4-5 mm. long; stigmas capitate.

576

Sournh West Arrica.—Waterberg Plateau: Boss in T. 34997! 34998! 34999 !:.

Cape Province.—Caledon distriet—Sir—Lowry’s Pass, farm Knoshoek, Marloth

4866! P. Mosselbay district: Vryersberg, May, Mur 2043! P. George district: In

humid places, Schlechter 2373! A, C; Paterson 1236! A; > Marloth 2545! P. Knysna

district: Hackerville, March, Breyer in T 23914!. Humansdorp district: Zitzikamma,

May, Fourcade 173! A:~ Flats Ratelsbosch, Nov., Fourcade 539! A; Britton 1167! A.

Uitenhage district : In Van Stadesbergen and on swarded places on the hills of Adow and

Krakamma, Sept._Nov., Zeyher 361! A, C. Bathurst district: Trapps Valley, Dec.,

Daly 588! A. Albany district: Grahamstown, Hill slopes, Nov., Galpin 381! P; Mac

Owen 192! A, N. and in Herb. MacOwanianum ; Daly and Sole in T.12953! ;. Hill other

side of Douglas Reservoir, April, Daly and Sole 141! A. Stockenstrom district: Katberg,

Galpin 2072! P. Kast London district: Ovuton, very common in lands, growing in sandy soil,

Dec., Hilner 379! A; June, Rattray 66! A. Komgha district: Komgha, among rocks,

Flanagan 795! C, P. Kentani district: Sheltered damp places, Pegler 117! P in part.

Port St. Johns district : Port St. Johns, Wager in P 21072!. Tsolo district : Tsolo, Payne

33! A. Flagstaff district: Fort William, Dec., Tyson 2828! A, Steynsburg district :

Zuurberg, April, Schonland 3212! P, 3213! A. Maclear district: Maclear, Murray 25! A.

Matatiele district : Cedarville, Mvenyani, near stream among rocks, Nov., Bandert 46! A.

Griqualand without precise locality, Tyson 1230! C. Prior in P 21065! without locality.

ORANGE FREE StaTe.—Fouriesburg district : Farm Dunelm, on mountain side in wet

soil, Potts in Grey Univ. Coll. Herb. 3111! P. Bethlehem district: Bethlehem, railway

enclosure in village, grassy and sandy places, common, Dec., Phillips 3215! P: Bothaville

district : Bothaville, common in damp places, Jan., Goossens 1216! P. Heilbron district :

Viljoensdrift, Jan., Rogers 4825! A, T.

BasuTOLAND.—Leribe district : Leribe, slopes, banks, damp spots, Dieterlen 678! P ;

south slopes of Leribe Plateau, Phillips 792! 905! 914! C.

NaraL.—Pinetown district: Durban, in humid places, Medley Wood 187! P and in

N 852!. Polela district: Bulwer, Jan., Bayer 356! N. Pietermaritzburg district : Mount

Ashly, Mogg 6375! P; Tweedie, Mogg 1192! P; Impolweni, Nov., Rump in N 20327! ;

Lidgetton, Jan., Mogg 6670! P, 6719! P; Allerton, Dec., Mogg 6536! P; Dimock-Brown

275! N. Umvoti district : Greytown; Nov., Wyle in N 21697 ! 21698 ! 20456! T 34130!

34131 !. Hshowe district: Entumeni, Oct., Wylie in N 9253!. EHntojaneni district :

Ulundi, Jan., Hvans 432! 19602! N; Melmoth, Dec., Mogg 6052! 4533! P. Estcourt

district : Giant’s Castle, Dec., Symons 290! T. Bergville district : National Park, Dec.,

Galpin 10177! P. Utrecht district: Farm Glen, Atholl, near Charlestown, Jan., Smith

5697! P.

SwaZILAND.—Mbabane district: Dalriach Mbabane, in swampy places, Dec., Bolus

11702! A.

TRANSVAAL.—Wakkerstroom district : Jan., van Dam in T 24643!. Ermelo district :

Nov., Collins in T 12201!; on farm Nooitgedacht, not frequent, Dec., Henrici 1348!

and 1551! P. Heidelberg district: Dec., Leendertz 1030! T. Potchefstroom district :

Losberg, Theron 768! T. Krugersdorp district: Krugersdorp, Jenkins in T 9229!.

Johannesburg district : Jeppestown Ridge, Nov.—Dec., Gilfillan 6087! A, P in part, 6088 !

A, P; Houghton Estate, in low grassy places, April, Bryant C 30! P; Turffontein, Bryant

D 10! D 34! P; Jenkins in T 10310!. Benoni district: Bradfield T 225! P; Gilmore

2186! P. Rustenburg district: Groenkloof, van Dam in T 10283!; Vlakfontein, ten

miles west of Koster, in vlei, Feb., Lvebenberg 126! P. Pretoria district : On open grassy

veld below the Wonderboom, Jan., Smith 2291! P, 2252! P; Oct., Leendertz 3730! T;

Oct., Pole Evans 408! P. Middelburg district: Klein Olifants River, along river banks,

Nov., Young A 91! T; Hewitt in T 10443!; Dec., Gilfillan 7163! A. Carolina: district :

Rademacher in T 7284!. Belfast district: Leendertz 2688! T; Jenkins in T- 6800!.

577

Barberton district: Umlomati Valley, Feb., Galpin 1284! A, P; Berlin, Godwan River,

Jan., Hofmeyer in P 21070!; Kaapsche Hoop, Gilmore 2229! P; Nelspruit, Nov., Rogers

in T 2389! A, T.. Pilgrims Rest district: Sabie, Tweefontem Experimental Area, Feb.,

Wagner A 84! P; Pilgrims Rest, Dec., Smuts and Gillett 2318! P. Lydenburg district :

Graskop, in vlei ground, damp and marshy places, not common, Jrvin 3! P; Nov., Wilms

in T 5836 !; Mac-a-Mac Falls, Jan., Burtt Davy 5368! P.. Waterberg district : Nylstroom,

van Dam in T 19690! P; Naboomspruit, Mosdene, in grassy, sandy glades, Galpin R

257! P; Moorddrift, Leendertz 2148! T; Palala River, Breyer in T 21448!. Pietersburg

district: Damp places, Shiluvane, Jwnod 4289! T, P; Woodbush, Mogg 14705 ! P, 14675 !

P; Rehman 6340! P; Wager in T 23087!; van Dam in T 25641!. Zoutpansberg dis-

trict: Vlei plant, soil very moist, common, McCallum in P 21068!; farm Zoutpan, vlei

behind homestead, Obermeyer, Schweikerdt and Verdoorn 204! P, T.

SouTHERN RuopeEsia.—Matopos : Eyles 3767! C. Bulawayo: Nov., Eyles and Johnson

1098! A, 1100! A. Marandellas: Grass plots, Govt. Farm, Nov., Rattray 343! P.

H. Lalandii Choisy var. transvaalense Bredell var. nov. Caules foliaque valde prominenter

et pellucido punctati.

TRANSVAAL.—Standerton district : Near Volksrust, Gray in P 22267!. Ermelo district :

Feb., Leendertz in T 7779!; Jan., Collins in T 6345!; Govt. School, Nel 35!.P; Farm

Nooitgedacht, in veld, common, Feb., Henrici 1538! P. Bethal district: Hoggeveld,

Trichardsfontein, Feb., Rehmann 6608! P. Middelburg district: Witbank, Zondags-

fontem, Dec.—March, Thode A 2789! P, N; near Witbank station, Dec., Gilfillan 7162!

Pan Part:

2. H. aethiopicum Thunb. in Thunb. Prod. PI. Cap. 138 (1800) ; DC. Prod. 1 : 552 (1823) ;

Engl. & Prantl. Nat. Pflanzenfam. ed. 2, XXI: 179 (1925). H. aethiopicum Thunb.

var. glauceseens Sond. in Fl. Cap. 1: 118 (1859-1860).

Semiherbaceous, 10-40 em. high. Stems few to many from a persistent rootstock,

erect, branched, round and smooth, glabrous, without black dots. Leaves sessile, 0-5-2-4.em.

long, -03-1-7 cm. broad, oblong to elliptic-ovate, rounded at the apex, entire, prominently

veined ; tertiary venation absent, if present meshes conspicuous on upper surface only.

Flowers in terminal lax or compact cymes.. Bracts 3-7-7 mm. long, ovate to linear-

lanceolate, acute or obtuse and with stalked, black-tipped protuberances along the margins.

Sepals 4-5-7-5 mm. long, 3-5-5 mm. broad, oblong to linear-lanceolate, acute and with

stalked black-tipped protubzrances along the margins. Petals 0-6-1-5 cm. long, 3-5-0-5 mm.

broad, oblong to oblong-spathulate, black-dotted along margins and on the surface.

Stamens indefinite, 4-5-9 mm. long, filaments thread-like, free or usually connate at the

base into 3 or more irregular groups ; anthers 0-5 mm. long with or without black spots.

Ovary 2-5-4 mm. long, ovate in outline, 3- or very occasionally 4-chambered : styles 3 or

occasionally 4, free; 2-5-6 mm. long; stigmas capitate:

Carre Province.—Riversdale district : Corente Riv., Muir 855!C, P; 4774!C. George

district: Nov., Prior in P 21066!. Knysna district: Gully, N.E. of Royal Hotel facing

west, Dec., Schonland 3498! A; March, Breyer in T 23357!; Plettenbergs Bay, without

name of collector in C 29506!. Humansdorp district: Assegaaibos, Nov., Marloth 10931!

P. Untenhage district: Oct., Zeyher 149! C, 360! C in part. Albany district : Grahams-

town, Howisons Poort, McOwan 397! A, N; grassy slopes Elandskloof, Galpin 382! A, P.

Bedfort district: Dec., Bennie 213! A. Fort Beaufort district: Adelaide, Great Wuinter-

berg, Jan., Ford 11410! P. Stockenstroom district: Katberg, Nov., Sole 383! A. Victoria

East district: Victoria Hast, Rattray 130! P. Kingwilliamstown district : Grassy places

at foot of perie, Flanagan 2142! C, P. Cathcart district, In sandy spots along the Kabousie

Riv., Dec., Flanagan 794! A, C: Fairford, Nov., Cotterrell 96! A. Herschel district :

Sterkspruit, May, Hepburn 380! A. Griqualand without locality, Feb., Tyson 1376! P in

part and Kaffraria without locality, Cooper 238! P.

578

3. H. Sonderi Bredell sp. nov., affinis H. aethiopico Thunb. sed foliorum margine haud

dentato-glandulo, caulibus plerumque nigro-punctatis differt.

Herba perennis; caules 10-45 em. alti, pauci vel plures, erecti, leviter ramosi, teretes

levigati, nigropunctati vel haud- nigropunctati. Folia sessilia, 0:5-2-5 cm. longa,

0-3-1-5 cm. lata, oblonga vel late ovata, punctis nigrisve luteis, apice rotundata, nervis

prominentibus ; nervi tertii plus minusve conspicuosi vel absunt. Flores terminales in

cymis laxis vel compactis dispositi. Bracteae 3-8 mm. longae, 1-5-2 mm. latae, ovatae vel

lineari-lanceolatae, acutae. Sepala 4-5-8 mm. longa, 1-5-3 mm. lata, lineari-lanceolatae,

acuta, nigro-punctata. Petala 0-8-1-5 cm. longa, 2-6 mm. lata, oblonga vel oblongo-

spathulata, nigro-punctata. Stamina indefinita, 5-9 mm. longa; filamenta filiformes

libera vel basi polydelpha; antherae 0-5 mm. longae nigro-punctatae vel impunctatae.

Ovarium 2-5-4-5 mm. longum, ovatum, 3-loculare ; styli 3, liberi 3-6-5 mm. longi; stig-

mata capitata.

Care Province.—Aliwal North district: Doctors Drift, Gerstner 137! P in part.

Matatiele district : Cedarville, on hillside, Nov., Bandert 102! A. Griqualand Kast without

precise locality, Tyson 1376! C, N.

ORANGE FREE StraTeE.—Harrismith district: Platberg slopes, in grass, Putterill in C

16861!. Bethlehem district: Witzies Hoek, Feb., Junod in T 17321!.

BasuToLAND.—Leribe district: Dveterlen 358! C, P; Phillips 921! C.

Naraut.—Pinetown district: Isipingo, Medley-Wood 12475! T, N; Amanzimtoti,

June, Forbes 647!; Bothas Hill, Oct., Medley-Wood in N 6423!, 8004! C, C 11461! N, P.

Camperdown district : Schlechter 3270! T, P, A. Pietermaritzburg district: Sept., Rump

in N 20899! and 20316!. Impendhle district : Impendhle, Nov., Levett 83! N; Deepdale,

in grass, Feb., Mawrice-Evans 92! N. Estcourt district : Estcourt, Mogg 3516! P in part.

Mooi Riv., Meteor ridge, Oct., Mogg 3235! P; Giant’s Castle, Symons 132! T. Umyvoti

district: Greytown, Nov., Wylie in N 21699!. Entojaneni district: Melmoth, Sept.,

Forbes 733! N. Nongoma district: Nongoma, Nov., Gerstner in N 22257!. Bergville

district : National Park, Drakensberg, Nov., Oliver 390! P ; Mont-aux-Sources, near Tugela

Drift, Nov., Schweikerdt 823! T; Tugela Valley, grassland, common, Feb., Bayer and

McClean 212! P; Grantleigh, woody herb, King 9! P; grassy slopes, Galpin 11426! P.

Kliprivier district : Ladysmith, Nov., Rogers in T 5146!. New Castle district : Charles-

town, Boscobello, Jan., Jenkins in T 12486!. Utrecht district : Riversmoor, Viljoensdrift,

Jan., Parkhouse in P 21064!; Wahl in T 15386!; grassy places, Spitzkop, Emangweni,

Oct., Thode in Herb. Univ. Stellenbosch 9326! and in A.

SwazILANnp.—Hlatikulu district : Hlatikulu, Stewart 146! C, T.

TRANSVAAL.—Piet Retief district: Grassy slopes, Oct., Galpin 9638! P; Jenkins in

T 10953!. Wakkerstroom district: Jan., Beeton 208! C;: Dec., van Dam in T 24642! ;

Ermelo district: Spitskop, Dec., Pott 5100! T. Carolina district: Dec., Rademacher in

T 7470!. Barberton district: Plastron, Oct., Holt 74! P; on summit, Saddleback Mts.,

southern slopes, Oct., Galpin 1116! A; Galpin 1028! G. Belfast district : Machadodorp,

orassy fields, Nov., Galpin 12522! P; Jenkins in T 10301! ; Rietvlei, Crocodile Riv., June,

Smuts 29! P; Waterval Boven, Nov., Rogers 18396! P; Waterval Onder, Jan., Jenkins

in T 6760!. Johannesburg district : Dec., Leendertz 1737! T ; Hutton in A 1039!. Venters-

dorp district : Goedgedacht, Dolomite Fountain in black loam, occasional Sutton 502! P.

Rustenburg district: Vlakfontein, 10 miles west of Koster in red loam, Feb., Liebenberg

170! P. Pretoria district: Ashbury Fountains, Nov., Smith 1381! P; Koedoespoort,

Smith 1529! P; on banks of Aapies Riv., near zoo, Smith 207! P ; Derdepoort, Feb., Quin

in P 12953!; near stream, occasional, straight stem from underground rootstock, Dec.,

Verdoorn 16! P; Fountains Valley, in grass on east side of magnetite koppie, Nov., Ver-

doorn 580! P; Waterkloof in valley near stream, Jan., Verdoorn 77! C, P; Wolwekloo’,

western hillside, Feb., Mogg 15929! P; Aapies River, Rehmann 4350! P; Meintjeskop,

Feb., Burtt-Davy 3958! P ; small shrub, with straight stems from an underground rootstock,

579

Oct., Phillips 3038! A, P; south-western slopes of Wonderboompoort, Nov., Pole Evans

63! P; Fountains Valley, Repion 65! P ; Leendertz 221! T in part, in T 3734! and T 4173! ;

Nov., Forbes 810! N. Lydenburg district: Farm Swagershoek, very common, Jan.,

Obermeyer in T 27901!. Pilgrims Rest district: Mauchsberg, Sabie, Dec., Smuts and

Gillett 2320! P; Olifants River, van Niekerk in T 7561!. Pietersburg district : Woodbush,

Mountain Home Farm, Dec., Mogg 14711! P; Duiwelskloof, in slopes of valley in burnt

veld, July, Galpin 9401! P. Zoutpansberg district: Growing on hillsides, Pigeon Hole,

Oct., McCallum 17! P; ravines, dry places, Feb., Junod 4288! P, T. Transvaal without

precise locality Pocock 36! A.

H. Sonderi Bredell var. transvalense Bredell var. nov. Folia membranacea infra

prominente lepidota supra et infra reticulata nervis tertils conspicuis.

TRANSVAAL.— Pietersburg district : Woodbush, Mountain Heme Farm, erect “herb ”,

common in moist places in grassland, Dec., Mogg 13996! P; Shiluvane, Junod 4290! T, P.

Wager in T 7223! without locality, im part.

4. H. Wilmsii Rob. Keller in Bull. Herb. Boiss. Ser. II. VIII: 179 (1908).

Semiherbaceous, less than 20 cm. high. Stems many from a persistent tap root, more

or less procumbent, branched, round and smooth, glabrous. Leaves sessile, 0-5-1-5 cm.

long, 0-2-0-6 cm. broad, elliptic-oblong, rounded at the apex, entire, 1- to 3-nerved ;

tertiary ventation inconspicuous or absent. Flowers in few-flowered cymes at the end of

the branches. Bracts leaf-like, obtuse. Sepals 4-5-6 mm. long, 1-5-2-5 mm. broad, oblong,

rounded at the apex, distinctly vemed wiih black dots along the margins and yellowish

spots on the surface. Petals 5-8 mm. long, 2-2-5 mm. broad, oblong to oblong-spathulate,

sparsely black-dotted along the margins near the apex. Stamens indefinite, usually less

than 30; filaments threadlike, free or connate at the base into a few irregular groups,

4-5 mm. long; anthers 0-25-0-5 mm. long, each with a black dot. Ovary 2-5-4 mm.

long, ovate in outline, 4-chambered ; styles 4 or very occasionally 3, free, 1-5-2 mm. long ;

stigmas capitate.

Care Province.—Murraysburg district: Tyson 413! A. Queenstown district :

Mountain sides, Nov., Galpin 1629! P. Aliwal North district : Elandshoek, moist ground

near water course, Oct., Bolus 153! C, P.; Doctors Drift, Gerstner 137! P, in part. Herschel

district: Majubanek, near Sterkspruit, Dec., Hepburn 92! A.

BasuToLanD.—Mafeteng district: Slope of Ha-Muya-pela Mt., Likhoele, Jan.,

Dieterlen 1222! P; bank of Catai Riv., Ha-Ma-Khonofane, Nov., Dieterlen 1293! P.

TRANSVAAL.—Lydenburg district: Near town, Jan., Wilms 136! (fragment of the

type in P).

5. H. natalense Wood & Evans in Journ. Bot. XXXV: 487 (1897); Engl. & Prantl.,

Nat. Pflanzenfam. ed. 2, XXI: 177 (1925); Burtt Davy, Flow., Pl. and Ferns of

the Transv. and Swaziland 1, 251 (1926) H. Woodi Rob. Keller in Eng. Jahrb.

LVIII: 193 (1923).

Semiherhaceous, 15-40 cm. high. Stems 1 to many from a persistent underground

rootstock, erect, much branched, round and smooth, glabrous. Leaves sessile, 0-5-2 cm.

long, 0-3-1-5 em. broad, elliptic to oblong-elliptic, rounded at the apex, entire, 3- to 7-

nerved ; tertiary venation distinct, with the meshes small and punctate. Flowers terminal.

Bracts leaf-like, obtuse. Sepals 4-5-7 mm. long, 1-5-4 mm. broad, oblong to oblong-

spathulate or obovate, rounded at the apex, distinctly veined, densely dotted with yellowish

spots. Petals 5-9 mm. long, 2:5-3-5 mm. broad, oblong to oblong-spathulate, distinctly

veined with occasional black spots along the margins near the apex. Stamens indefinite,

3-5-7 mm. long ; filaments threadlike, free or connate at the base into 3 or 4 irregular groups ;

anthers 0-25-1 mm. long, each with a black dot. Ovary 2-5-4 mm. long, ovate in outline,

3-, 4- or 5-chambered ; styles 3, 4 or 5, free, 2-3 mm. long; stigmas capitate.

580

Care Province.—Kentani district: Damp valleys, Oct., Pegler 117! A, P in part.

Tsolo district : Payne 14! A.

NataL.—Pietermaritzburg district: Balgowan, Oct., Mogg 5610! P; April 3809! P;

Allerton, Dec., Mogg 6567! P. Lions River district: Howick Falls, Noy., Rogers in T

5212!; St. Ives, Dargle Road, Oct., Mogg 5679! P. Estcourt district: Near bank of Mooi

River, Oct., Medley-Wood 4034! N, A. Umvoti district: Greytown, Oct., Wylie in N

21576!, in N 21577!, in N 22378! and in T 34129!.

H. natalense W. & E. var. petiolatum Bredell var. nov. Folia petiolata obovata ; petioli

0-5-1 mm. longi.

NataL.—Camperdown district: Franks in N 12968!.

6. H. leucoptychodes Steudel ex Richard in Tent. Fl. Abyss. 1: 96 (1847); Engl. & Prantl.

Nat. Pilanzenfam. ed. 2, XXI: 176 (1925); Journ. Bot. LXV: 330 (1927). dH.

lanceolatum Lam. ex Oliver in Fl. Trop. Afr. 1: 156 (1868); Burtt Davy in Fl. Pl.

and Ferns of Transy. and Swaziland 1: 251 (1926) and of Hutchinson. and Dalziel

‘in Fl. West. Trop. Africa 1: 230 (1927).

Woody shrub, 4-8 ft. high. Stems much-branched, round or almost so, glabrous ;

internodes very short. Leaves sessile, 1-5-3 cm. long, 0-3-0-9 cm. broad, narrowly lanceolate,

sub-acute, narrowed towards the base; venation regularly but sparsely pinnate; the

tertiary venation somewhat conspicuous and often having the effect of small striae. Flowers

large and terminal. Bracts 0-4-1-5 em. long, 3-7 mm. broad, ovate to ovate-lanceolate,

acute or obtuse. Sepals 6-9 mm. long, 3-5-6 mm. broad, ovate to broadly-ovate, obtuse,

with minute stalked protuberances on the margins. Petals 1-6-3-7 em. long, 0-5-1-5 em.

broad, oblong-spathulate, distinctly veined and occasionally with stalked bodies along the

margins. Stamens indefinite, 0-8-1-5 cm. long, filaments thread-like, connate at the base

into 5 very distinct groups opposite the petals ; anthers 0-5-1 mm. long, occasionally with

yellow gland-like bodies between the thecae. Ovary 4-8 mm. long, ovate to broadly ovate

in outline, 5-chambered ; styles 5, 5-9 mm. long, connate almost to the tip, with the free

branches 1-5-2 mm. long; stigmas capitate.

oO?

Care Province.—Umzimkulu district: Edge of Malowe forest, near Clydesdale,

March, Tyson 3044! T, C, P.

Natat.—Richmond district: Richmond, May, Medley-Wood 1995! N. Nkandhla

district : Qudeni Forest, Sim 2942! N; Gout. Forester in Forest. Dept. Herb. 8637!; Kotze

113! P: Gerstner in N 23007!.

SwazILaAnD.—Mbabane district: Jan., Rogers 11483! A, P.

TRANSVAAL.—Barberton district : T/orncroft 342! P and in T 3929!; banks of streams

upper Moodies-and also Devil’s Kantoor, Galpin 658! C, A, P; Kaapsche Hoop, shrub,

whole plant strongly scented, Oct., Thode 1565! N; Dec., Gilmore 2248! P. Lydenburg

district: Schlechter 3942! A, C, N, P, T; farm Swagershoek, Jan., Obermeyer 238! T;

on bank of stream and edge of forest, common, Aug., Keef 1108! P; Sabie Falls in kloof

and along streams, June, Burtt Davy 442! P, N. Pietersburg district: Haenertsburg,

van der Merwe in P 21060!; Murray 730! P; Corlat 142! P; Broederstroom farm, Nov.,

Hodgson in P 11416!; Magoebaskloof, March, Potts 3652! P; Pole Evans in P 21061! ;

New Agatha, June, McCallum in P 21060!: Woodbush, Kratzenstein, July, Hoffman

98!T; Jenkins in T 7181!; van Dam in T 25632!. Zoutpansberg district : Ntabini Forest,

near Louis Trichardt, June, Galpin 10314! P; Nov., Obermeyer 1124! T; Laastgevonden,

growing along rivers where the soil is damp, Sept., Koker 18! P; Cyprus Poin, edge of woods,

Junod 4207! P.

SourHern Ruopesta—Inyananga: Common, Henkel 2568! P. Umtali: July,

Pardy 5105! P. Stapleford, Oct., Rattray 126! P.

581

Sa —,

CPR AS

ENS DSssy

ay.

==:

wy,

SOD

o ast

Nese

als

WETS

saat

Sor

QE5

Fig 10,

582

7. H. Roeperianum Schimper in Tent. Fl. Abyss. 1: 96 (1847); Engl. & Prantl., Nat.

Pflanzenfam. ed. 2, XXI: 176 (1925); Journ. Bot. LXV: 331 (1927). H. Quar-

tunianum Rich. ex Oliver in Fl. Trop. Africa 1: 156 (1868).

Woody shrub, 5-12 ft. high. Stems erect, slightly branched, almost smooth, glabrous.

Lexves sessile, 4-7 cm. long, 1-5-2-5 cm. broad, elliptic-lanceolate, obtuse to sub-acute,

cuneate at the base, dark on the upper surface, much paler beneath; venation almost

reticulated, with the meshes small and punctate. Flowers large, terminal. Bracts 4-5-7 mm.

long, 2-5-3 mm. broad, lanceolate to ovate, obtuse or sometimes acuminate, black-dotted

or with gland-like bodies along the margins. Sepals 5-6-5 mm. long, 2-5-4 mm. broad,

ovate to broadly-ovate, obtuse, with black dots and stripes near the margins and minute

stalked bodies along the margins. Petals 2-3 cm. long, 1-1-5 cm. broad, oblong-spathulate,

partly fleshy, with occasional black dots and stalked gland-like bodies along the margins.

Stamens indefinite, 1:4-1-7 mm. long; filaments thread-like, connate at the base into 5

distinct groups opposite the petals ; anthers 1 mm. long, occasionally with white gland-like

bodies between the thecae. Ovary 5-6-5 mm. long, ovate to sub-round, 5-chambered ;

style 1, 0-6-1 cm. long ; stigma distinctly 5-lobed.

TRANSVAAL.—Lydenburg district: Mariepskop, Nov., Fitzsimons and van Dam in

T 26311!; April, van Son in T 32921!.

SouTHERN Ruopesta.—Umtali: July, Pardy in Govt. Herb., Salisbury 5092! and in

P. Odzani Riv. valley, Teaque 214! C. Vumba Mts., margins of forest, “ Cloudlands ”’,

July, Galpin 9247! P. Melsetter, June, Rump in N 23094! ; common shrub, large, showy,

June, Eyles 795! C.

EXPLANATION OF FIGURES.

Fig. 1, a-h, ditferent shapes and sizes of leaves of H. Lalandii; Fig. 2, abaxial surface

of leaf of H. aethiopicum; Fig. 3, abaxial surface of leaf of H. natalense; Fig. 4, leaf of

H. natalense var. petiolatum; Fig. 5, abaxial surface of leaf of H. leucoptychodes ; Fig. 6,

sepal of H. Sonderi; Fig. 7, sepal of H. aethiopicum showing the stalked gland-hke pro-

tuberances along the margins; Fig. 8, sepal of H. natalense and H. Wilmsii; Fig. 9, petal

of H. aethiopicum; Fig. 10, petal of H. natalense and H. Wilmsii; Fig. 11, abaxial surface

of leaf of H. Roeperianum.

penne

UNION OF SOUTH AFRICA

H. Lalandii.

T wovewouse ;

ihe oF Lio

. vy

SS eninct ALSERT ~ wiLicowmon

H. Lalandii var.

Transvaalense

ih H. Roeperianum

1 5 208800 or 4-8 Mites tet taek.

(

‘

-———

GORDONIA

| gsuTMERtanod |

LaAinGseur

52 — ;

nook

Se Bera

teh eae re

UNION OF SOUTH AFRICA

vu a N A & 4 a

Ill H. Sonderi

’ j REST

: ose ae c i Sa5 Ps f

H. aethiopicum , T at

if sgt A f Mteesis

one eager: as cr '

ee ee ee

| OP crap

be Wes ita

momscee

“One

T wooanovuce

e _ awe,

: Ze sreoes Sey Se

He «0 SP F Bow ‘=

, . 4 ae

A ys

‘ene

‘ wihtowson

ae int uNronoace

UNION OF SOUTH AFRICA

Ill H. Wilmsu

ash H. Natalense

H. Sonderi var.

Transvaalense

‘7 Thur

f i Swasitza sl

batanss 22h

_ WILLISTON

KENHAROT

? .

| SUTHERLAND

583

THREE SPECIES OF STRYCHNOS WITH |- SEEDED FRUITS.

I. C. Verdoorn.

When naming specimens of Sérychnos with 1-seeded fruits (rarely 2-seeded) sent in

by Mr. W. E. Marriott of Durban, a few interesting points came to light. As a consequence

it is now possible to define the three species concerned more completely. The specimens

were from trees marked Nos. I, I] and IV. At first only fruiting material was sent and

flowers of Nos. I and II were received later when they: developed in November. Unfor-

tunately the flowers of No. IV were not obtained.

In specimen No. I (Fig. 1) the leaves are up to 6 cm. long and 2-5-3 cm. wide, more

or less elliptic; buds short, ovate; perianth-tube short, sparsely hairy in the throat ;

lobes quite glabrous ; fruit dark green, oblong, equal-sided at the base ; seed shaped like

a coffee-bean being deeply channelled down one side and rounded on the other.

Specimen No. II (Fig. 2) has leaves up to 4-2 cm. long and 2-5 cm. broad, more or

less obovate ; buds oblong ; perianth-tube as long as the lobes ; lobes distinctly bearded ;

fruit light green to yellow and reddish, oblong or globose oblique at the base ; seed oblong

globose, slightly compressed, not channelled down one side.

Specimen No. IV (Fig. 3) has leaves 4-6 cm. long and 2-3-5 em. broad, ovate, long

acuminate, folding along the midrib; upper surface shiny; fruit globose, distinctly

stipitate ; seed globose, slightly compressed (flowers not seen).

This plant with the stipitate fruit and the long-acuminate, folded leaves did not agree

with the description of any known South African species.

With regard to No. I, as far as the leaves and flowers were concerned it appeared to

be S. Henningsii as described in FI. Cap. Vol. 4 pt. 1, 1051 (1909). ‘The seed, however, is

here described as “ globose or ovoid’ whereas the coffee-bean-like seed of No. I is very

distinctive. To elucidate this point the specimens in all South African herbaria were

examined. This showed that leaves, flowers and fruits characteristic of No. I are always

correlated with the coffee-bean shaped seed. Such specimens were usually identified as

S. Henningsii. This species was originally described on leaf characters only and the type

specimen is housed in the Berlin Herbarium, Dahlem, Searching other botanical literature

for mention of the species it was found in Wood’s Natal Plants and Sim’s Forests and Forest

Flora of the Cape Colony. In the former the description and drawing of leaves, flowers

584

W Tis

oe”

j- \

\N BETTY CONNELL

Fig. 1.—Strychnos Henningsti Gilg (Marriott 1).

586

KE Z LY

lu le

BETTY CONNELL. y)

Fig. 2.—Strychnos decussata (Pappe) Gilg (=S. Atherstonei Harv,) (Marriott 11).

586.

BETTY CONNELL.

Fig. 3.—Strychnos micans Sp. Moore (Marriott IV).

587

and fruits are those associated with the coffe-bean seed. The seed itself is not figured but

is described as ‘‘ globose ” and a note added that only | fruit had been seen. In the latter

publication for the first time the characteristic seed was clearly figured and described.

The figure agrees also in the other characters associated above with such seeds. From this

it was concluded that the description of the seed of S. Henningsvi in the Flora Capensis

was inaccurate and that the error was repeated by J. Medley-Wood.

The material of No. I was later taken to the Royal Botanic Gardens, Kew and there,

through the courtesy of the Director and the Director of the Botanisches Museum, Berlin-

Dahlem, the type material of S. Hennings Gilg (2 leafy specimens collected at Durban)

was borrowed. Dr. H. G. Schweickerdt, Botanist for South Africa at Kew, examined these

types and reported that the leaves had the general characteristics of those usually placed

in S. Henningsw (and which had been proved to be correlated with the coffee-bean shaped

seed) but were on the whole larger. He matched one or two leaves on the Marriott specimens

with those of Beyiich I (1.e. one of Gilg’s syntypes) and on Wood 6672, which is a duplicate of

the specimen Medley-Wood described and figured in Natal Plants, Dr. Schweickerdt wrote

‘Leaves marked a, } and ¢ are an excellent match of those on Beyrich 1”. He noted too

that on Bachmann 1745 (the other syntype) the collector gives a native name

“Umnonono ” for the tree and it is the same as mentioned by Medley-Wood. The leaves

being on the whole larger can probably be explained by their having been collected when

the trees were not in flower or fruit ; such leaves often being larger than those found on

flowering or fruiting branches. Thus Marriott I can now be named Strychnos Hennings

Gilg with some degree of certainty.

Regarding Marriott No. II, according to floral characters given in the Flora Capensis

this is S. Atherstonei Harv. It agrees well with the type description and figure in Harvey’s

Thesaurus Capensis. The seed as mentioned above is oblong-globose, slightly compressed

and with no sign of the longitudinal channel, only a slight indentation in the centre of one

of the slightly flattened surfaces (the mark of the hilum or attachment). The drawing of

the seed in the type plate is accurate and the description “ peltate ”’ is used in the botanical

sense meaning “ attached by the surface”. Through the courtesy of Prof. H. H. Dixon

of Trinity College, Dublin, I was enabled to examine photographs of the material which

Harvey had worked on and these supported the above identification. In studying Harvey’s

notes, however, it is evident that, according to the international rules of nomenclature,

the specific epithet must be changed. Prior to Harvey’s work Pappe described a plant

Atherstonea decussata. Harvey, when describing Strychnos Atherstonei cites this name in

the synonymy of his species saying that he could not find any tangible character on which

to separate the plant generically from Strychnos. There is no doubt that Pappe’s plant

is the same species as Harvey’s and since the first valid specific epithet must be used it is

necessary to employ the combination Strychnos decussata (Pappe) Gilg, published in Engl.

Bot. Jahrb. 28, 121 (1901). Marriott No. II therefore is named as such.

Dealing now with Marriott No. IV which, as mentioned, does not agree with any

description of a South African species, it was matched in Pretoria with fruiting specimens

collected in the Hlatikulu Forest (Boocock in National Herbarium 23199). Later it was

matched at Kew with unnamed fruiting specimens from tropical Africa (Swynnerton 1071

from Chirinda Forest and Dawe 531 from the Kibala Forest). These 4 specimens possessing

stiped fruits also have similar leaves which are long-acuminate, shiny on the upper surface

and have a tendency to fold along the mid-rib. A further search through the material

of Strychnos at Kew brought to light a young flowering specimen (Swyrnerton 125) which

also has these characteristic leaves. It is the type of Strychnos micans Sp. Moore. The

characters and the locality seem to point to its being conspecific with the other specimens

mentioned. The description disclosed another character which supports this opinion. The

flowers of S. micans are described as being 4-merous (most other species are 5-merous in

this genus) and the remains of the calyx on all the fruits showed 4 distinct lobes. It seems

588

justified to assume, therefore, that Marriott IV is Strychnos micans Sp. Moore, and if this

is so the distribution records of the species have been considerably amplified. The following

table summarises the distinguishing characters of the three species investigated :—

STRYCHNOS HENNINGSII G/LG

Fruit oblong-ovoid, not oblique at

base, not stipitate; seeds ob-

long-ovoid, deeply grooved

down one side; hilum midway

in groove.

Leaves up to 6 em. long and 2-5-

3 cm. wide, elliptic, widest

about the middle and gradu-

ally narrowing to the apex or

somewhat acuminate; tertiary

veins obvious on upper surface.

Flowers 5-merous; corolla-tube

glabrous or sparsely bearded in

throat; lobes glabrous.

S. DECUSSATA (PAPPE) GILG

(S. Atherstonei Harv.).

Fruit globose or ovoid, oblique at

the base, not stipitate; seeds

ovoid, slightly compressed ;

hilum in centre of slightly com-

pressed surface.

Leaves up to 4-2 em. long and 2-5

em. wide, obovate, widest

above the middle, rounded at

the apex or widely and bluntly

acuminate ; tertiary veins not

obvious on upper surface.

Flowers 5-merous, corolla-tube

glabrous ; lobes densely bearded

within.

S. MICANS SP. MOORE

Fruit globose, not oblique, stipi-

tate; seeds globose, slightly

compressed ; hilum in centre

of slightly compressed sur-

face.

Leaves up to 6 cm. long and

2-3-5 em. wide, ovate, long

acuminate, widest below the

middle, long accuminate fold-

ing along the mid-rib; tertiary

veins usually obvious.

corolla-tube

throat; lobes

Flowers 4-merous,

bearded in

glabrous.

589

THE FLORA OF TRISTAN DA CUNHA:

ev .S. CAREISEE, EXPEDITION; _ 11937.

By R. A. Dyer.

Through the courtesy of the British Admiralty, the Union Government was invited

early in 1937 to nominate a meteorologist and a botanist to join an expedition to Tristan

da Cunha on H.ML.S. Carlisle.

The main, objects of the expedition were to make a record of conditions in general on

the island, with particular reference to the health and teeth of the islanders, to take provisions

and mail to the islanders, and to bring away on holiday the Rev. H. Wilde, who had served

on, the island for three years. During the past 100 years various scientific expeditions

have visited the Tristan group of islands, but none made a very lengthy stay. The 1937

visit of H.M.S. Carlisle was no exception in this respect.

Acknowledgments.

The presence of a botanist in the expedition was due indirectly to a suggestion

by Mr. P. A. Snell, informally referred to as the Ambassador for Tristan da Cunha at Cape

Town. I am personally indebted to Dr. I. B. Pole Evans and Dr. K. P. Phillips for my

nomination, as botanist. There was scarcely a month in which to prepare for the trip and

at the time of notification I had very little knowledge about the Island. I am especially

grateful to Captain Marrick, the officers and several members of the company of H.M.S.

Carlisle for assistance in various respects which made my efforts more profitable than they

may well have been in the circumstances. In view of the opinion of the officers that

unusually fine weather was experienced except for a small gale, I may state, in part defence

of my results, that I was not born to the sea.

The Director of the South African Museum, Capetown, allowed me, with the assistance

of Miss 8. Garabedian, to consult the botanical specimens collected on Tristan da Cunha

by Messrs. Bonomi and Kaytel, and later the Director forwarded on, loan to Pretoria im-

portant literature. The Director, Royal Botanic Gardens, Kew, arranged for the comparison

of my material with the Kew collection and put me in touch with Mr. H. N. Dixon, who

has very kindly supplied a complete report on the mosses. The assistance of these and several

of my colleagues is greatly appreciated. I shall refer later to the invaluable aid rendered

by the islander Arthur Rogers.

A large number of duplicates of the botanical specimens collected on the expedition

has been distributed to the more important oversea herbaria, the Royal Botanic Gardens,

Kew, having been presented with the first set.

Recapitulation.

Tristan da Cunha is the largest of a small group of islands of volcanic origin in the

centre of the South Atlantic Ocean just south of the latitude of the Cape of Good Hope

and about 2,000 miles distant from it. It was discovered in 1506 by a Portuguese Admiral

who gave it his name. Inaccessible Island and Nightingale Island are within a radius of

about 16 miles. Tristan da Cunha is roughly 21 miles in circumference and 6,500 feet

590

high. Except for the present settlement plateau, $-1 mile wide and 50-100 feet in altitude,

the island rises precipitously from the sea to an altitude of between 2,000 and 3,000 feet ;

thence upwards to the summit the rise is more gradual. Vegetation practically ceases

above 4,000 feet, due to the lack of soil. The peak of the island is rarely seen owing to the

presence of mist almost throughout the year. The rainfall is well distributed and although

no exact data are available, the total annual fall must be anything between 50-100 inches

on, the settlement plateau.

Although over 300 years elapsed after its discovery in 1506 to the date from which

permanent colonisation took place, the Tristan, group of islands had been, the home of various

shipwrecked sailors for short periods during the long interval.

The flora of Tristan da Cunha was investigated first in January, 1793 by Aubert du

Petit Thouars during a five days’ call for watering on the way to the Mascarene Islands.

He published his results in 1811 in two papers, “ Description, abregée des isles de Tristan d’

Acugna ” and “ Esquisse de la flore de Visle de Tristan d’Acugna ” being parts of his “ Me-

langes de Botanique et de Voyages.” This was followed by an excellent account in 1818

by Captain Dugald Carmichael who landed on the island in 1816. He had the advantage

of a longer stay than Thouars and his account of the vegetation serves as a reliable basis

for comparison with present-day conditions.

H. N. Moseley, Naturalist to the H.M.S. Challenger expedition (1873-76) published

what may be looked upon as an interim report on the flora in 1875. MacGillivray and

Milne collected earlier in 1852 but their report was not published. Hemsley, in 1885,

reviewed the data accumulated up to that date with particular reference to Moseley’s report.

Subsequent to this, Gane, Bonomi (1904) and Keytel collected botanical specimens ;

those of Keytel gathered 1908-1909 were the basis of a paper by Phillips, 1913. The first

set of both Bonomi’s and Keytel’s specimens are in the South African Museum, Capetown.

The Report in 1912 on the S.Y. Scotia expedition, 1902-4, which called only at Gough

Island, 220 miles to the south, added a little to our knowledge. The botany of the island

is mentioned also by Macklin as a result of the visit in 1922 of the “ Quest ” on Shackleton’s

last voyage, and by Mrs. Rogers, wife of Rev. Rogers, missionary on the island, 1922-25.

The most recent publication on the flora of Tristan da Cunha is by Christophersen,

1937, who deals principally with a collection of plants made im 1934 by Hinar Siggeson.

Dr. Erling Christophersen himself sailed to the Tristan group in November, 1937, in charge

of a scientific expedition. Unlike previous expeditions his was planned to extend over a

period of several months and should result in the accumulation of most valuable data in

connection with controversial subjects such as the origin and main affinities of the flora.

Notes on Origin and Affinities of the Flora.

A brief reference to some of the more important views on the origin and present affinities

of the flora may be of interest. Although Sir Joseph Hooker did not land at Tristan da

Cunha while engaged on, his scientific expeditions in the Antartic about the middle of last

century, he considered its flora in the light of Carmichael’s results. Dealing with the An-

tartic flora as a whole, he pointed out in guarded terms that “the many bonds of affinity

between the three southern Floras—the Australian, Antarctic and South African—indicate

that these may all have been members of one great vegetation, which may once have covered

as large a Southern area as the Huropean does a Northern.” Referring to the flora of Tristan

da, Cunha he found that of the 28 flowering plants then recorded only one species of Phylica

and one Pelargoniwn, amounting to one-fourteenth of the whole, are Cape forms, whilst

seven others, or one-quarter of the flora, are either native of Fuegia or typical of South

American botany ; and the ferns and lycopodia exhibit a still stronger affinity. He further

591

pointed out that there are some points in which the vegetation of Tristan da Cunha resembles

that of St. Helena and Ascension. This is through the Phylica sp. and Pelargonium sp.

which link the flora with that of the Cape also. However, such important families at the

Cape as Proteaceae, Rutaceae, Oxalidaceae, Crassulaceae, Ericaceae, Restiaceae are entirely

absent from St. Helena, Ascension and the Tristan da Cunha group, which is in marked

contrast to the affinity of the South Western Cape flora with that of Australia.

Schonland, when dealing with the subject of the origin of the Angiospermus Flora of

South Africa does not mention the flora of the Tristan group specifically. He does, however,

make several pertinent remarks when, discussing Hooker’s work on the Antarctic flora. “The

theory which I formulated is briefly this : At the times when these common types developed,

probably in Lower Cretaceous times, possibly even in Jurassic times, there was still a direct

land connection between Australia and South Africa, that some of these types were carried

to South America by a later connection, with Australia and that on the other hand Tropical

Africa was able to exchange types of animals and plants with Tropical America up to early

Tertiary times, and that in this way a second connection between the South African flora

and the flora of America was established.”

Hemsley, writing earlier, considered that the flora of the Tristan group consists of 3 or 4

distinct elements, “no one of which sufficiently predominates to justify the assumption that

it is essentially the original flora—the facts indicate the former existence of a floral region

distinct from America and Australia, and, if not African, at least more closely allied to that

than to any other.” Personally I consider the weight of evidence against Hemsley’s con-

clusion. He himself concedes that there are probably equally weighty facts against it.

Christophersen, while expressing the view that our knowledge of the phanerogams of

Tristan da Cunha is insufficient for phytogeographic discussions of any consequence, draws

attention to the relatively low number of endemic species, and “ the high number of species

of South American or/and New Zealand—Australian distribution, as compared with that of

South African distribution.” Since the expression of these views he has had the advantage

of several months’ study of the flora on the island and his conclusions are now awaited

with interest.

History of Flora since colonisation.

Before giving the results of my work, I propose to recapitulate briefly from the records

of Carmichael, Moseley and Macklin in order to trace the changes which have taken, place

in the indigenous flora of Tristan da Cunha since the first days of permanent colonisation.

Captaim Dugald Carmichael joined the expedition which embarked at the Cape on the

2nd November, 1816, in order to put into effect the decision of the British Government

to take possession of the Tristan da Cunha group ofislands. Carmichael remained on Tristan

da Cunha four months and studied the “natural production of the island.” He records

that the steep face of the mountain was covered with brushwood mixed with fern and long

grass, which, for the most part, hid its native ruggedness. The only plant which approaches

to the size of a tree is Phylica arborea. At that time it occupied, not only the whole plain

(present settlement plateau), but also all parts of the face of the mountain where the roots

could “insinuate themselves into the crevices of the rock.’ Under favourable conditions

the Phylica grows up to about 20 feet in height, with a crooked twisted stem, 1-1} feet in

diameter. The only other woody plant is the shrublet Empetrum rubrum.

Of the herbaceous plants Carmichael points out that the most remarkable is the

gigantic grass Spartina arundinacea : “* It overruns the whole of the island from the upper

edge of the tableland down to the sea-shore, accommodating itself to all soils and situations.

It springs up in large close tufts, which, when full grown, are borne down by their own

weight and lean upon each other in such a manner that a person may roll himself over them

without any danger of sinking. Its stems grow to the length of six or seven feet, and are

592

of a solid, almost ligneous, texture, and are covered with a profusion of leaves. This grass

makes an excellent and durable thatch, and the young leaves are eaten by horses and oxen.”’

He continues, “ the wild celery (Apiwm australe) grows in abundance over all the low ground

and attains a great size, its stem sometimes measuring upwards of three inches in diameter.”

Of Acaena sarmentosa Carm., Carmichael says, ‘‘it overruns the low ground, is of no apparent

utility, but an intolerable nuisance to such as have occasion to walk over the ground where

it grows. Its fruit 1s a sort of bur, which, on the slightest touch, fixes itself on one’s clothes,

and falling in a hundred pieces, covers him all over with an unseemly crust of prickly seeds,

not to be got rid of without infinite labour.”

Carmichael was the first botanist to make an ascent to the summit of the mountain.

He noted that the mountain is less precipitous after the initial stiff climb (approximately

2,500 feet), the ground very wet and “studded with tufts of rushy plants that gave way

under the slightest pressure.’ Here also were extensive patches of the fern Blechnum

tabulare (Lomaria robusta) “ the stems of which ” to use his words, “ trail along the ground

crossing each other in an intricate manner. This beautiful fern is more generally scattered

than most of the others, being found in all moist places, from the table-land down to the

plain. The trunk grows to the length of four or five feet, sometimes erect, but usually lying

on the ground with its apex only upright.”’

At about 4,000 feet the soil practically ceases, and consequently vegetation also, except

for a few mosses.

In addition to the general account of the vegetation, Carmichael gives a systematic

list of the flora in which ecological notes are included. In these supplementary notes he

states that the grasses Agrostis ramulosa and A. media are found on the high part of the

mountain forming the chief part of the herbage.

Even in Carmichael’s time the northern extremity of the present settlement plateau

was denuded of its natural vegetation. By setting fire to the grass the trees were scorched

and killed, otherwise it formed an “ impenetrable copse.”

When Moseley arrived at the settlement 57 years later, the distribution of the flora

had already been considerably modified. Unfortunately, owing to unfavourable weather

conditions, he had only six hours on the island. When he had reached an altitude of about

400 feet, he was recalled to the ship. “ Shrubs ” he says, referring to Phylica arborea “ com-

mence at about 400 feet elevation. There are no trees in this locality, since they have been

cut down for firewood, but there is still plenty of wood on the island.” Spartina arundinacea

he found dotted about among the other herbage in rounded tufts of pale bluish green and

Nertera granadensis (=N. depressa) very abundant creeping everywhere amongst the grass.

His observations on the flora of Nightingale and Inaccessible Islands proved it to resemble

very closely that of Tristan da Cunha. That of the former two islands was at that time,

and still is, much less influenced by human, interference.

Macklin, a medical doctor, made an ascent of the mountain behind the settlement in

approximately the same direction as Moseley. He went considerably higher, however,

found abundant vegetation, numbers of ferns, including the tree fern Blechnum tabulare,

and the tussock grass, Spartina arundinacea, mosses, lichens and the island tree. He also

found the “ dogeatcher ” plant Acaena sarmentosa referred to by Carmichael.

On, another occasion, Macklin made an excursion south west from the settlement, past

the potato patches, to the cattle and sheep pasturage. To get round the bluff at the end it

was necessary for the party to climb up to 2,000 feet. He observed that the vegetation

in the gullies is very luxuriant and the grass being sheltered from the wind “ grows lush and

long.” Over a precipice one saw below a long, grassy plain on the “* back of the island

where numerous semi-wild cattle graze.” Going higher they found masses of tree fern and

tussock grass and island tree.

593.

Outline of Botanical Programme.

After leaving Simonstown on 23rd February, 1937, H.M.S. Carlisle, under Captain

Marrick, anchored off Tristan da Cunha about 9 a.m. on Sunday 28th. In the morning the

personnel and equipment were landed and a camp established on the island (Fig. 1). During

the day the area near the beach landing stage, and the margin of the settlement plateau

leading from it, were botanised for a distance of several hundred yards. Owing to the

small size of many of the plants, progress in collecting was comparatively slow.

On, the following day, Ist March, a small expedition, consisting of two marines, Arthur

Rogers, an islander, and myself, set out with the object of climbing as high as possible up the

mountain. We approached the cliffs up a gully about a mile south of the settlement. This

must be taken into consideration when reading the floristic data which follow later. At an

altitude of 1,500 feet, before the steepest part of the climb was negotiated, rain began to

fall and the two marines were left in a rock shelter. With the aid of Arthur Rogers, I

climbed 1,000 feet higher, often with a very precarious hold. At about 2,500 feet, with

clothes drenched and limbs numb with cold, it seemed prudent not to aspire to greater

heights. Collecting specimens was difficult and photography well nigh impossible: both

would certainly have been unobtainable at that locality without Arthur Rogers’ assistance.

The morning of the next day, 2nd March, was occupied mostly in sorting out and put-

ting into press the saturated material collected on the mountain. In addition, with the

ald of a marine, a small collection of marine algae was obtained. The most important

species to the islanders is Macrocystis pyrifera Ag. ““ kelp,” which grows in great quantity

in, deep water round the island. The long branches float near the surface checking to a

certain degree the force of the waves and preventing the formation of “ breakers ’ except

under very rough conditions. It is used as a fertiliser on the potato fields.

An opportunity was made during the afternoon to plant the following grasses taken

from the Rietondale Experiment Station, Pretoria :

Acroceras macrum, Stapf

Digitaria swazilandensis, Stent (Swaziland finger grass.)

Echinochloa pyramidalis, Hitche. and Chase (Limpopo grass.)

Pennisetum purpureum, Schum. (Napier fodder.)

Echinochloa sp. (Antelope grass.)

Digitaria scalarum, Chiov. (Dunn’s finger grass.)

Digitaria sp. (Pongola River finger grass.)

Pameum repens, Linn.

Panicum coloratum, Linn. (Makarikari strain.)

These were planted in short rows in, Arthur Rogers’ garden, with the advice that the

grasses which showed promise should be planted out on a more extensive scale in order to

improve the pasturage.*

* A preliminary report received from Dr. Christophersen (August, 1938), states that the results up

to the time of his departure from the island were not promising. A recent letter from Arthur Rogers

states that they are now mostly dead.

594

In addition to the grasses mentioned, the following trees and shrubs, supplied by the

Director of Forestry, were taken to the island for trial. They were planted under the

supervision of Mr. N. Ayson, a farmer resident at the Cape, who accompanied the expedi-

tion, to give advice on agricultural matters in general.*

Timber Trees. Common Name. Number.

Cupressus macrocarpa, Hartw.............. | Monterey Cypress..................-.-:- 50

Eucalyptus cornuta, Labill............-.... Wate a ayeiicuiils ct Ad wees RG eee ae 50

Eucalyptus diversicolor, F. Muell........... AEG RG I ee es SRG Heh ee A Oy te ole 50

iucalhypiusigom plocepialany) Ce teres r etka ieee eer rece eer ee 50

Eucalyptus Lehmann, Preiss.............. ISHAM. here as ai Sener eee cere eke ereiee 50

IO GHCCT COUCH, MAW, 5 c0d0c009ec0K8G0C Belhamibraine ae ita testa felae a reas 10

IOS CHDGMUS, IDOE sococcosocaecesoocoss |! lbaeWenny INO. 505 n5c0ccosccccdu0vccCD6S 50

inusshalepenstsm Ulli eet rei tcre AMleppogPine). jy saspicpaciat irene ce nee: 50

IPS CHNCHIGIOSIS Ch SNe osgc0c0 see0sccese Chmeiny lisievael 12S; so 600006 000000000008 50

IZUUS (ESO, SOIC codvocccccccacodc0s Cluster"Pineee.ecnises cel. cn eee ae 50

= ; | if

2 Hedge Plants.

iHakearsalignass kent hither eee eerie Hiakear rt jortstevevcets enemies ei cauan ear menor 40

Lept rspermum laevigatum, F. Muell......... Amsiarabinin Miyl@oaccoscto sooo escaeccsuc 100

Myoporumunsulane’ is Bisel oe IManatokals st, eects mene eta cee 100

700

}

A quantity of “ worms,’ which constitute a pest in the potato fields, attacking both

the aerial growth and tubers, was received from Mr. Green. Some of these survived the

jouroey to Pretoria but none developed further than the crysalis stage. A moth very like

one of the common “ cut worm ” moths was collected in the grass sward near the settlement.

Special permission was granted by the Captain for me to remain that night on the island.

A few sods of grass, including a root of Spartina arundinacea were collected on the morning

of the 3rd, but no further botanising was done as we were due on board at 7.30 am. A

few living Roane were also taken on board and are at present in cultivation at the National

Herbarium, Pretoria.

Present Conditions.—Details of Botanical Work.

The landing beach is composed of pulverised, black, voleanic rock. The plants nearest

high tide level are, Mariscus congestus C. B. Cl., Rumex frutescens Thouars and Cynodon

Dactylon Pers. On the steep slope are found forms of Scirpus Thouarsianus Schult. Blech-

num penna-marina Kuhn, a small hardy fern, Chrysanthemum leucanthemum L., Plantago

lanceolata L., Sporobolus capensis Kunth and Gnaphalium luteo-album L. The moss Pols Yy-

trichum juniperinus Hedw. was found in plenty near boulders on the plateau extending

to about 1,500 feet altitude. Associated with it on the plateau was an abundance of Nertera

granadensis Druce (=N. depressa) densely matted together with two other small runners,

Hydrocotyle capitata Thouars and H. asiatica L. Hmpetrum rubrum Vahl, and Blechnum

australe L., which were formerly common in this area, are now rare and poorly developed.

At the head of the waterfall below the settlement is Holews lanatus L. Near the stream

is found a variety of plants including, in addition to most of those mentioned above, Plantago

major L. form ?, Cotula australis Hook., Polygonum aviculare L., Cerastium caespitosum

Gillib., Veronica ‘serphyllifolic L., Rumex ‘Acetosella Ls. and Oxalis corniculata L.

* Dr. Christophersen states that the progress made by the trees up to the time of his departure

from. the island was promising and a recent report from Arthur Rogers is also encouraging.

595.

In a small area from which sods had been removed not long previously, the following

small pioneer annuals were collected : Senecio vulgaris L., Juncus bufonius L. form, Cyperus

tenellus L.f. and Poa annua L.

Many of the plants listed above have been introduced to the island, mostly uninten-

tionally, and their presence in quantity emphasises the drastic changes which have taken

place in the vegetation, due to the influence of man. The present settlement plateau which

was, in Carmichael’s time, covered by an “ impenetrable copse ” is now covered by a short

dense grass sward. It has been grazed off short and very little of it was in flower at the

time of our visit. It appeared that Vulpia bromoides and Agrostis sp. were dominant over

a considerable area. Between the garden plots the grass Poa pratensis was found with a

certain amount of P. annua, only the latter, however, being collected in flower. What I

took to be Trifolium repens L. was frequent in the sward but no flowers were seen, to verify

the identification.

The excursion up the mountain furnished further evidence of the effects on the vegeta-

tion, of heavy stocking. The short grass sward, characteristic of the settlement plateau,

extended almost unbroken up the gully to an altitude of about 500 feet. At this stage

scattered specimens of Pelargonium grossularioides Ait., Acaena sarmentosa Carm. and

Empetrum rubrum Vahl occur. Much of the surface of the steep inclines is broken into

long parallel steps, the initial causes of which are the combined effects of rain and tramping

of animals. The soil is banked up and held in check by a dense mixed growth in which

grasses and the fern Blechnum penna-marina are most conspicuous. On rocky portions

too steep to afford a footing for animals, the vegetation is dense, but the absence of Phylica

arborea Thouars trees is a notable feature as compared with the early records of its abund-

ance.

At an altitude of about 1,000 feet a boulder outcrop sheltered a solitary small specimen

of Blechnum tabulare, the island “ tree fern’ which formerly dominated parts of the settle-

ment plateau. Another fern Dryopteris aquilina C. Chr. was also first seen here. On

the neighbouring rocky cliffs, regeneration growth of Phylica arborea was prominent, but

not a large one was to be seen.

The gully is cut off abruptly at about 1,500 feet altitude by a rock shelter. From this

point the climb is mostly up a series of semi-perpendicular rock faces. The character of

the vegetation changes radically. The grass sward of the lower slopes gives way to a dense

growth in which ferns and mosses are dominant. Among the ferns are Elaphoglossum

succisifolium Moore, fairly rare, Hymenophyllum aeruginosum Carm and Asplenium obtu-

satum Forst. var. crassum C. Chr. which is often found on, otherwise bare rock faces with the

roots penetrating the crevices in the weathered surface. Scirpus Thouarsianus is, for a short

distance, frequently found in this association. Mosses, including the previously unrecorded

genus Sphagnum and undescribed species are dealt with later. The endemic species Lyco-

podium diaphanum Sweet, grows commonly amongst mosses and liverworts. The abund-

ance of these, often several matted together, is particularly interesting.

Owing to the less disturbed conditions on the mountain slopes, as compared with the

settlement plateau, species such as Hydrocotyle capitata and Nertera granadensis, are more

luxuriant. They also grow intertwined with mosses. At about 2,000 feet the first plant

of the island celery, Apiwm australe Thouars, was recorded. Hmpetrum rubrum becomes

common, and more luxuriant. Blechnuvm tabulare has elongated stems and is dominant on

many ridges and also in depressions (Figs. 6 and 7). While forcing a passage through a

dense community of Blechnum tabulare, I noted on the margin of the cliff several flowers

of a miniature member of the Compositae, Lagenophora nudicaulis Dus. (=L. Commersonii

596

Cass.). Its small rosettes of leaves rested on a cushion of different coloured mosses and

liverworts in which the slender rhizomes were embedded. The minute fern Hymenophyllum

peltatwm Deav. occurs in the same area on the stumps of Blechnum tabulare.

The most arduous and dangerous part of the climb is negotiated below about 2,200 feet.

Although the mountain rises steeply above this, it is not interrupted so much by perpendicu-

lar rock faces. There is an evident accumulation of soil. Grass becomes rank with scattered

tufts of Cyperaceae. The only grass in flower was Holcus lanatus L., which is not indigenous.

The Cyperaceae collected include Carex Thouars Carm., Scirpus virens Boeck., S. sulcatus

Thouars, Uneinia breviculmis Carm. var. rigida Kiik.

During the ascent only one small plant of Spartena arundinacea Carm. was recognised.

Small shrubs of Phylica arborea appeared to be frequent on certain slopes at about 2,300

feet, but owing to the poor visibility it was practically impossible to distinguish forms

even at a short distance. However, except for the eradication of the old trees of Phylica

arborea, the indigenous flora on the upper slopes of the mountain has not been, greatly

changed from its natural state.

Arising out of previous statements, a note on, soil erosion seems desirable.

Soil Erosion.

The subject of soil erosion on Tristan da Cunha has not been mentioned specifically

in earlier literature. That soil erosion has taken place from time immemorial goes without

saying. The first settlers built their homes largely of boulders which had fallen from the

cliffs in the course of natural erosion. The vegetation was sufficiently dense and firmly

established on, the rocky slopes and plateau to retard very considerably the progress of

erosion. With the advent of man and his domesticated animals and the consequent deple-

tion of the vegetal covering, the progress of erosion was accelerated. The settlers removed

the island tree (Phylica arborea) for firewood ; the stock reduced the flora by grazing and

trainping. The absorption power of the soil surface being thus reduced the “ run-off”

became proportionately greater.

Within the memory of middle-aged islanders, erosion along certain watercourses has

assumed extensive dimensions. In Fig. 4 of the stream, which affords the settlement with

a permanent water supply, the effect of erosion is clearly seen. The steep-sided watercourse

is deep in comparison with the small regular flow of water. A water course to the south

of the settlement is much deeper and broader yet it does not have a perennial flow of water.

I have no knowledge of the area specially set apart for the grazing of the untamed or “ wild ”

catiile.

Owing to the high rainfall, regeneration of the vegetation under normal conditions is

comparatively rapid. Within the immediate vicinity of the settlement area, which is undu-

lating or gently sloping to the seaward margin of the plateau, a short dense sward is dominant.

This effectively withstands tramping and prevents the development of sheet erosion.

Erosion, however, is likely to loom more seriously on, the Tristan, horizon, than it has up to

now, if precautionary measures are not adopted to control grazing on the steep slopes ad-

jacent to the settlement plateau. There are practically no sizeable specimens of Phylica

arborea in, the neighbourhood. The regeneration growth of small shrubs should certainly

be protected and not on any account be rooted up for firewood.

597

BOTANICAL SPECIMENS COLLECTED ON EXPEDITION.

Algae.

Very little time was available for collecting “ sea weeds ” but in spite of this, with the

aid of a marine, the following species were obtained on the rocks below the settlement.

The material was identified at the Royal Botanic Gardens, Kew, and the majority of them

are new records.

Phaeophyceae.

Macrocystis pirifera (Twrn.) Ag. No. 3617.*

Splachnidium rugosum (L.) Grev. No. 3605.

Rhodophyceae.

Callymenia Harveyana J. Ag. No. 3610.

Dipterosiphonia sp. possibly new. No. 3616.

Epymenia obtusa (Grev.) Ktz.? No. 3612.

Gymnogongrus polycladus (Ktz.) J. Ag. Nos. 3606, 3613.

Tridaea ciliata Ktz. Nos. 3608, 3609.

Chlorophyceae.

Cladophora fiagelliformis (Suhr.) Ktz. No. 3615.

Enteromorpha sp. No. 3611.

Ulva lactuea LZ. Nos. 3614, 3618.

No. 3607, indeterminable.

Fungi.

Psalhota sp. (Agaricaceae, ‘‘ mushroom ’’).

Only one immature plant was collected just prior to departure. Islanders stated that

the ‘‘ mushrooms ” were common, at certain seasons during which time they were eaten.

Lichens.

Owing to lack of time, this group was almost entirely neglected. One species collected,

however, although widely distributed elsewhere, is a new generic record for the island.

Cladonia pyxidata (L.) Hoffm. Associated on an old stump of Blechnum tabulare

with several species of moss about 2,200 feet alt., No. 3579.

Parmelia sp. On Blechnum australe stump, about 2,000 feet. alt. No. 3564.

Parmelia sp. possibly P. saxatilis (L.) Ach. On rocks about 1,000 feet alt. No. 3550.

Hepatics.

Several specimens were collected and identifications are still awaited from oversea.

* Collector’s numbers are those of the writer.

598

Mosses.

By H. N. Drxon.

The mosses collected by R. A. Dyer on Tristan da Cunha were received under seventeen

numbers, but some of these included several species, and the total amounted to 25 species.

Hitherto some 40 species have been recorded from the island.

Of these 25 species, 12 are novelties for Tristan da Cunha, a considerable proportion.

Apart from the four new species, the chief interest lies in the geographical distribution of

the newly recorded plants. These may be tabulated as follows :—

New Species.

Name. Affimty.

B. gracillima Mitt. Kerguelen.

B. magellanica Schimp. Fuegia, New

Zealand.

B. tenuirete Dus. Guaitecas.

Mascarene or S. American species.

Blindia brachystegia.......:..........

Bryum flaccidifolaum...... «22.22:

Porotrichum atlanticum.............--

Eurlynchium crassicostatum ined*

New Records.

Name. Distribution.

Sphagnum amblyphyllum, Russ........ Wide.

Rhacomatrium lanuginosum (Hedw.), Wide.

Brid.

Dicranoweisia antarctica (C. M.), Par..

Dicranum aciphyllum (H. £. & W.).....

Fuegia ; Campbell Is.; New Zealand.

Temperate 8. America; Falklands ;

S. Georgia ; Antarctic.

Dicranoloma imponens (Mont.), Par...

Dicranoloma Harioti (C. M.) Par.....,..

Orthostichopsis subimbricata (Hampe),

Fuegia.

Mascarene Islands.

Broth.

Leptodontium interruptum (Mitt.), Broth. Amsterdam I.; New Zealand.

The distribution, of these newly recorded species agrees on the whole very well with

what is already known of the general affinities of the bryophytic flora of Tristan da Cunha ;

that is to say a small but appreciable percentage is endemic, a considerably larger propor-

tion sub-antarctic and particularly Fuegian; a very small number distinctly associated with

the New Zealand flora, and a still smaller but very marked association with the flora of the

Mascarene Islands. One of the most marked features is the almost entire absence of any

association with the general African continental flora.

As all the specimens were gathered at or about the same locality, viz. about 2,200 feet

altitude under conditions always moist, and at the same date, Ist March, 1937, it has not

seemed worth while to repeat these data.

* T have just studied the very large collection of mosses made by the Norwegian Scientific

Expedition, 1937-38, which is of remarkable interest. I find that the Hurhynchium of which you

collected two fragments is a new species (Dixon, 1/4/39). :

599

Sphagnum amblyphyllum, Russ. No. 3591. Mr. W. R. Sherrin has examined this,

and writes that he does not find it differs from the widely spread plant of the North Temperate

zone. The pores are rather fewer than usual, but not more so than may be found in S.

amblyphyllum. The genus is new to Tristan da Cunha.

Ditrichum conicum, Mont. No. 360lb. A few stems only.

Pseudodistichium atlanticum, Dix., forma elata. No. 3586g.

Blindia brachystegia, Dix., sp. nov.

Caules circa 3 cm. alti, caespites sat densos, sericeos, olivaceovirides instruentes. Folia

parum conferta, erecto-patula, sicca parum mutata, 2.5-3 mm. longa, e basi angusta, con-

voluta, lanceolata sensim tenui-subulata, Integra, apice saepe paucidenticulata. Costa

angusta, ad basin 30-40 pw lata, tenuis. Cellulae anguste lineares, infra sensim longiores,

parietibus firmis ; alares magnae, aetate rubrae, auriculas paullo dilatatas, magnas, decur-

rentes formantes.

Dioica : Flores turgidi, antheridiis magnis. Perichaetia turgida, bracteae internae e

basi lata, convoluta, raptim in subulam contractae. Sefa perbrevis, cygnea, crassiuscula.

Theca subglobosa, pachydermica, exothecii cellulis irregularibus, valde incrassatis, apud

orem seriebus pluribus transverse rectangularibus, parietibus horizontalibus fortiter incras-

satis: Operculum altiuscule conicum, rectum. Annulus O. Sport 20-28 p. Peristomir

dentes lanceolati, intense purpurei, fortiter trabeculati.

Hab. alt. 700 m., Tristan da Cunha, 1 Mar., 1937. No. 3594.

Very near to the Kerguelen B. gracillima, Mitt., and scarcely differing in the vegetative

characters, but markedly distinct in the fruit ; the lid there is conico-rostrate, with a long,

curved beak, while here it is shortly conical. The spores here also are larger, in B. gracillima

being 18-22 p.

B. magellanica differs in the denser foliation, wider leaf base and longer lid.

Dicranoweisia antarctica (C. W.), Par. No. 3601e.

New to Tristan da Cunha.

Dicranum aciphyllum, H./. & W. Nos. 3592e, 3596. With young and very old

capsules, which is noteworthy, as it is a dioicous species.

New to Tristan da Cunha.

Dicranoloma imponens (Mont.), Par. Sparse in one or two gatherings.

New to Tristan da Cunha.

Dicranoloma Harioti (VC. M.), Par. In several gatherings, and in fair quantity. No.

3586c is a form with strongly falcate leaves, the forma falcata of Roivainen (Syn. Dicranum

fuegianum, Dus.). The leaves are usually much shorter in this form than in most of the

typical gatherings.

New to Tristan da Cunha.

Campylopus introflexus, Brid. No. 3553b.

Leptodontium interruptum (Mitt.), Broth. Nos. 3589, 3592b and mixed among other

gatherings. A very interesting plant, as it is only known from Amsterdam Island and from

New Zealand, where it appears to be rare.

New to Tristan da Cunha.

Rhacomitrium crispulum, H. f. & W. No. 360la. This is the R. membranaceum of

Mitten, but it certainly cannot be separated from the widely spread, subantarctic R. crispulum.

600

Rhacomitrium lanuginosum (Hedw.), Brid. Nos. 3695c, 3597.

New to Tristan da Cunha.

Macromitrium fimbriatum (P. Beawv.), Schwaegr. No. 3584.

Bryum flaccidifolium, Dix., sp. nov.

Pseudotriquetra. Humile, sordide olivaceo-viride, molle, flaccidum. B. neodamensi

Itzigs. affine; differt folus mollioribus, angustioribus, minus concavis, apice obtuso sed

minus rotundato, vix cucullato ; cellulae parietibus multo tenuioribus, limbo minus definito,

costa debiliore.

Fructus ignotus.

Hab. alt. circa 700 m., Tristan da Cunha, 1 Mar., 1937. No. 3587.

Near B. neodamense, but more flaccid, with narrower leaves, obtuse, but not rotundate

at apex, and scarcely cucullate, with thinner-walled cells and weaker nerve. It is also very

near B. tenuirete Dus. from Guaitecas, but as figured by Cardot the leaves there are more

ovate and slightly narrowed to the obtuse apex, while here they are distinctly elliptic in

outline.

Eustichia longirostris (Brid.), C. M. No. 3586h.

Bartramia sp. No. 360ld. A single stem, of no doubt Section Vaginella, but with the

sheathing leaf base tapering upwards and passing gradually into the lamina; the cells

lowly tubercular almost to base.

Breutelia tenuifolia (Muitt.), Par. Nos. 3592, 3598b, and among other gatherings, in

fact generally mixed with most of the specimens.

Ptychomnion densifolium (Brid.), Jaeg. Nos. 3595a, 3695d. In several other gatherings

also, and showing much variation in size, but without manifesting any tendency to inter-

grade with other related species.

Orthostichopsis subimbricata (Hampe), Broth. No. 3586f. A very small quantity.

New to Tristan da Cunha, and only known from Madagascar.

Porotrichum atlanticum, Diz., sp. nov.

Frons unica, gracilis, inventa, 4 c.m. alta, parce divisa, divisionibus sat regulariter

pinnatis, ramis complanatis, vix 1 cm. longis, 1 mm. latis, iterum parcissime brevissime

ramulosis. Stipes rigidus, foliis infimis e basi arctissime adpressa brevissime cuspidatis,

cuspide rigide patente acuta; supra magis magis foliaceis, membranaceis, patulis. Folia

caulina complanata, sicca estriata, 1-5 mm. longa, valde asymmetrica, ovata, breviter

acutata, acuta, uno margine plano, altero inflexo, superne argute, haud grosse, subdistanter

inaequaliter denticulato. Costa tenuiuscula, circa ? folii longitudinem attingens. Cellulae

inferiores lineares, breviusculae, supra sensim breviores et latiores, superiores omnes breyiter

rhomboideae, 2-3 x 1. Folia ramea multo minora, subcomplanata, costa debiliore, cellulis

paullo brevioribus, cetero subsimilia.

Cetera ignota.

lab. Circa 700 m. alt., Tristan da Cunha, 1 Mar., 1937, No. 3592f.

The position of this (whether in Porotrichum or Thamnium) is rather doubtful. P.

valdiviae (C. M.) Mitt., from Chile, closely resembles it, but has narrower leaves, and all the

stipes leaves foliaceous, etc. The Mascarene Pinnatella tamariscona (Hampe) Broth. is

much like it in leaf form, but the habit is very different, and the leaves broader and rounder

in outline. ;

601

Thuidium curvatum (Mitt.). Nos. 3595f, 3599. A rather remarkable form (3595c),

is much more robust and rigid than, most of the specimens, the leaves all larger and less

incurved when, dry as well as when moist ; the stem leaves especially with long, rigid subulae,

widely divaricate when dry and slightly recurved. Most of the specimens of the species

that I have seen from the island are very much alike, small and slender with all the leaves

(stem and branch) small and similarly incurved-catenulate when dry. If the Tristan plant

is truly distinct, then this form is well worth a varietal name. But I strongly suspect that

it is not specifically different from the New Zealand and South American 7’. furfurosum,

which in, its slender forms [7'. sparsum (H.f. & W.) Jaeg.] very closely resembles, to say the

least, the Tristan da Cunha plant. If this is the case, then the form referred to is but a

slight divergence from the type compared with the many and varied forms which 7. fur-

furosum exhibits.

Drepanocladus uncinatus (Hedw.) Warnst. Nos. 3598a, 3589, etc.

Eurhynchium crassicostatum Dir. ined. Nos. 3593, 3592d.

New to Tristan da Cunha.

Hypnum cupressiforme (Hedw.) Brid. No. 3586b. A curious form, most of the

leaves being straight, and with small alar cells ; in this condition it is very unlike the typical

plant. A few leaves here and there, however, are falcate, and have the larger, characteristic

lower alar cells, thus revealing its identity.

Polytrichum juniperinum Hedw. No. 3553.

Ferns.

No new records were made in, this group.

Hymenophyllum aeruginosum, Carm. Frequent at the base of larger ferns, e.g. Blech-

num tabulare, and amongst moss, not readily observed owing to small size, 2,000—2,400

feet alt., No. 3558.

H. peltatum, Desv. On old stump of Blechnum tabulare mixed with moss, possibly

frequent but less readily observed than H. aeruginoswm, about 2,200 feet alt., No, 3582.

Polypodiaceae (Alphabetical list).

Asplenium obtusatum, Forst. /. var. erassum C. Chr. Scattered on bare rock faces

without soil, and occasionally growing in leaf mould with mosses, 1,500-2,000 feet: alt,

No. 3561. :

Blechnum australe, 1. Rare on steep rocky slope from landing beach to the settle-

ment, No. 3522, frequent elsewhere in gullies according to islanders.

Blechnum penna-marina, Kuhn. Frequent on steep and precipitous slopes and rock

faces, often forming dense consocies or associes with Hlaphoglossum laurifolium ; prevents

excessive soil erosion in parts by the binding effect on the soil of its dense rhizomatous

growth, 30-1,500 feet alt., No. 3521.

Blechnum tabulare, Kuhn. (the island “ tree fern”). Rare between, rocks at 1,000 feet

alt. and nearly stemless ; at 2,000 feet alt. it becomes dominant in parts on ridges and in

some depressions, about 3-4 feet tall, sheltering a wealth of moss, lichen and H ymenophyllum

spp., sterile fronds more rigid than, the South African form and the pinnae with somewhat

revolute margins, No. 3555. In Thouars’ and Carmichael’s time Blechnum tabulare was

common, on the settlement plateau with Phylica arborea the “ island tree.”

’

602

Dryopteris aquilina C. Chr. Occasional to frequent, stemless or with short thick stem,

among rocks, some fronds 23-3 feet long, 600-1,000 feet alt., No. 3554.

Elaphoglossum laurifolium, Moore. Fairly rare on steep rocky slopes with moss, etc.,

6-15 ins. tall, about 2,000 ft. alt., No. 3557.

E. succisifolium, Moore. Frequent on rockfaces in dense associes with Blechnum

penna-marina, 6-12 ins. tall, 1,000-1,500 ft. alt., No. 3556.

Gymnogramma cheilanthoides, Sw. Apparently rare, growing with moss and other

small. ferns, about 2,000 ft. alt., No. 3578.

Polystichum adiantiforme, J. Sm. Apparently rare, on steep rocky slope with moss

and fern, about 2,000 ft. alt., No. 3562.

Vittaria stricta, Carm. (Vittaria vittarioides C. Chr.). Apparently rare, on steep

rocky slope with moss and fern, about 2,000 ft. alt., No. 3559.

Lycopodiaceae.

Lycopodium diaphanum Sw. Frequent on steep slope amongst moss and other short

dense growth, long lateral branches penetrating moss and rooting at the nodes, 2,000-2,300

ft. alt., No. 3563.

Flowering Plants.

Among the phanerogams the following six are recorded for the first time. Almost

certainly all of them were introduced by man.

Cynodon Dactylon (L.) Pers.

Cyperus tenellus L.

Juncus bufonius L. form.

Polygonum aviculare L.

Plantago major L. form.

Veromea serpyllifolia L.

These were all growing within a radius of a few yards on the margin of the settlement

plateau between the settlement and the beach landing stage. At least two were collected

previously by Keytel, but owing to his poor material, they were not definitely identified.

Keytel’s specimens, tentatively named Scirpus cernuus and Nertera assurgens are, respec-

tively, Cyperus tenellus and Veronica serpyllifoia. In view of Christophersen’s

remarks about the poor condition of some of Hinar Siggeson’s specimens, it 1s possible that

he also may have collected some of the above-mentioned species in an unrecognisable state.

Of these newly recorded species the most important one economically is Cynodon

Dactylon. Yt has taken a firm hold on the sand above high water level near the landing

stage and is also spreading on the margin of the settlement plateau. Owing to its aggressive

character 1t will, in due course, no doubt make a bid for dominance over the settlement

plateau. It is likely to prove an asset in the work of preventing and checking soil erosion,

and at the same time is a useful addition to the fodder plants on the island.

Gramineae.

Agrostis simulans Hemsl. Apparently frequent in sward of the settlement area but

not many specimens in flower, probably owing to grazing, No. 3534. In association, with

other species of grass not in flower, forming dense sward on small ridges of mountain slope,

leading up gully, up to 1 foot high with protection, 500-1,000 feet. alt., No. 3551.

603

Mr. C. EK. Hubbard, Kew, reports—* Nos. 3534 and 3551 probably represent the same

species as Agrostis simulans Hemsley from St. Helena. The degree of development of

the awn, on the lemma appears to be very variable ; in some cases it may be represented by

a minute point only, whereas in others it is comparatively long. Agrostis Helenae Steud.

may be an earlier name for the species but the type has not been examined.”

Cynodon Dactylon Pers. (First record.) Producing vigorous growth with strong

runners, frequent on steep slope from landing beach above high water level and on margin

of settlement plateau, No. 3540.

Holcus lanatus L. (Yorkshire Fog). Scattered on slope from the beach but not much

in flower: dense clump in flower at top of the waterfall below the settlement, No. 3525.

Up mountain about 1 mile south of settlement associated with several species of Cyperaceae,

9-18 inches high, 2,300-2,400 feet alt., No. 3570.

Poa annua L. Frequent in a small area near the margin of the settlement plateau

and elsewhere, No. 3530.

Poa pratensis 1. Not in flower, sod containing it collected in the settlement area and

grown, at the National Herbarium, Pretoria, where it flowered in October, 1937, No. 3625.

Spartina arundinacea Carm. (Tussock grass, thatch grass.) An isolated young tuft

18 inches high, about 1,000 feet alt., No. 3576. Cultivated in Fred Rogers’ cues for

repairing thatch, 6-7 feet high, No. 3577.

When Thouars and Carmichael arrived it overran the whole of the island from the

upper edge of the settlement area down, to the sea shore.

Sporobolus capensis Kunth [S. mdica Auct. non (L) R. Br.]. Densely tufted on, slope

from landing beach, frequent, about 15 inches tall, No. 3541. On margin of settlement

plateau, about 20 inches tall, No. 354la. The difference in growth of the two collectings

was apparently due to the better soil condition, of the latter.

Vulpia bromoides Gray. In damp depressions below the settlement towards the beach,

common, in, the sward but not much in flower, a grass of fine texture, culms 4-8 inches

high, No. 3533.

Cyperaceae.

Carex Thouarsii Carm. Occasional in dense tufts on margin of “ tree fern’ Blechnum

tabulare consocies with other species of Cyperaceae and the grass Holeus lanatus, about

18 inches high, 2,300-2,400 feet alt., No. 3572.

Cyperus tenellus L. (first record). Near margin of cliff below the settlement ; one of the

pioneers in a small area where turf had been removed, annual about 2 inches high, No.

3523.

Although recorded here for the first time, it was collected by Bonomi (1904) and probably

Keytel (1908-9) as discovered by comparison with specimens in the South African Museum,

Capetown. These specimens in the S.A. Museum were placed tentatively under Scirpus

cernuus Vahl. var. subtilis. In view of this it would be advisable to check the identification

of earlier collectings from Tristan da Cunha named as S. cernwus in other herbaria.

Mariscus congestus (. B. Cl. (Cyperus congestus Vahl.). Frequent on black, sandy

volcanic soil above landing beach, with grass, forming tufts with short rhizomes, 9-18 inches

high, No. 3542.

Scirpus sulcatus Thouars. Scattered in dense growth of other species of Cyperaceae

and grass, near “ tree fern” Blechnum tabulare consocies, in, tufts 9-18 inches high, 2,300-

2,400 feet alt., No. 3574. si

heen?

604

S. Thouarsianus Schult. Frequent in dense tufts and mats on rock faces, with little

S. Thouarsianus Schult. var. bicolor Hemsl. On rock faces and in tufts with short

runners in grass near boulders on the settlement plateau towards the beach, frequent, No.

3543.

S. virens Boeck. Apparently comparatively rare in dense growth of other species of

Cyperaceae and grass, near “ tree fern,” Blechnum tabulare consocies ; in tufts about 9

inches high, 2,300-2,400 feet alt., No. 3573.

Until the above identification was received from Kew, I did not associate this specimen,

very closely with §. Thouarsianus. My material of the latter has only a few, and those

poor, spikelets, which look very distinct from the comparatively large dense ones of my

No. 3573 (S. virens). The habit of the plants is also different and it is only on detailed

examination that the points of similarity become evident. Whether it should be considered

as a variety of S. Thouarsianus as done by Hemsley is one of the problems which was to

engage Dr. Christophersen’s special attention. In addition to the differences in habit

the dissected specimens of my 3573 usually had two style branches united only near the

base and three stamens: those of 3571 had the style branches further united and only two

stamens. The glumes of the latter were narrower and more thickened down the back.

Uncinaria brevicaulis Thouars var. rigida Kiik. Occasional in grass with other cyper-

aceous growth, near margin of “tree fern’ Blechnwm tabulare consocies, in tufts 18 inches

—2 feet high, 2,300-2,400 feet alt., No. 3575.

Juncaceae.

Juncus bufonius L. forma (new record). On margin of settlement plateau towards

sea, ploneer on small cleared area, apparently an annual, No. 3524.

Juncus tenuis 2. On margin of settlement plateau towards sea, in grass near stream,

occasional to frequent in small tufts, No. 3526.

Polygonaceae.

Polygonum aviculare, /. (introduced; new record). Near margin of cliff below the

settlement, evidently only recently introduced, No. 3529.

Rumex Acetosella 2. On margin of cliff below the settlement in short grass and on

stream bank, rather dwarf, frequent but not much in flower, No. 3519. Mixed with grass,

moss, etc. on mountain slope, up to 18 inches high ; 2,000 feet alt., No. 3580.

Rumex frutescens Thouars. Near sea shore and near the settlement stream, frequent,

with tough, short or straggling stem up to about 1 foot long, No. 3518.

Caryophyllaceae.

Cerastium caespitosum Glib. Occasional near margin of the settlement plateau in

grass, No. 3532. 4

Rosaceae:

Acaena sarmentosa Carm. (=A. sangwisorba Auct.). Procumbent sub-shrub in ~

grass with Hinpetrum rubrum, fruiting stage past, 500—1,500 feet alt., No. 3549.

Leguminosae.

Trifolium repens 1. Noted in grass, not in flower and not collected.

> Oxalidaceae.

Oxalis corniculata L. On margin of the settlement plateau, occasional in grass, No.

abe. @

e €>

605

Geraniaceae.

Pelargonium grossularioides (L.) Aw. (=P. acugnaticum Thouars), Frequent in

grass with Empetrum rubrum and Acaena sarmentosa, evidently grazed somewhat by stock,

procumbent, fl. pink, 500-1,000 feet alt., No. 3547.

Empetraceae.

Empetrum rubrum Vahl (Island berry used for pies). Small Hrica-like shrub with

spreading branches, rare on margin of cliff below the settlement, where it was formerly

common, ; not in fruit, No. 3520. Frequent on steep slope in grass with Acaena sarmentosa,

in fruit, berries sour, red to black on same plant, 500 feet alt., No. 3548. In dense grass

near Blechnum tabulare associes, berries very plentiful, 2,300 feet alt., No. 3567.

Rhamnaceae.

Phylica arborea Thovars (the island tree). Shrub up to about 6 feet high (formerly

up to 20 feet in this area), now only regeneration growth on rocky slope, No. 3552.

Umbelliferae.

Apium australe Thouars (island celery). On, steep slopes in dense fern and moss growth,

1-14 feet high (formerly abundant and more luxuriant) about 2,000 feet alt., No. 3568.

Hydrocotyle asiatica L. Runner in short grass near margin of the settlement plateau,

occasional to frequent in patches, No. 3537.

H. capitata Thouars. Runner with short internodes, in dense patches with grass and

Scirpus Thouarsianus var. on margin of the settlement plateau, No. 3538. In dense moss

and Lycopodium growth, much more luxuriant than specimens collected near the settlement,

about 2,000 feet. alt., No. 3566.

Scrophulariaceae.

Veronica serpyllifolia L. (new record but collected previously —Keytel 1822 in S. Afr.

Museum, Capetown). Small diffuse herb in grass near stream and occasional as a pioneer,

on, moist bare soil with Plantago lanceolata and Cynodon Dactylon, minute mauve flowers,

No, 3536.

Plantaginaceae.

Plantago lanceolata 1. Common on margin of cliff below the settlement area and on

steep slope from the landing stage, varying considerably in height according to the situation,

No. 3546.

P. Major L.? On margin of the settlement plateau, in small area, dwarf herb 2-4 inches

high, apparently perennial, No. 3531. The specimens probably represent a form of this

variable species although considerably smaller than plants growing in other countries under

favourable conditions.

Rubiaceae.

Nertera granadensis (L. /.) Druce (=N. depressa Banks and Sol.). Frequent on margin

of settlement plateau towards the landing beach, small runner rooting at nodes, forming

dense patches in grass and near boulders, with many red berries, No. 3539; mixed with

moss on mountain side, not in fruit as near coast and not in dense mats, about 2,000 feet

alt., No. 3581.

The specimen/collected by Keytel, tentatively named as N. assurgens Thouars, is Ver-

omea serpyllifolia L. te

606

Compositae.

Chrysanthemum leucanthemum L. Frequent on slope leading from beach landing

place and along the settlement watercourse above waterfall; leaves either nearly entire

or deeply pinnate, plant from almost stemless up to 1 foot high, No. 3544.

Cotula australis Hook. Occasional on margin of grass sward near the settlement stream,

No. 3528.

Gnaphalium luteo-album L. Frequent in parts on margin of cliff below the settlement

plateau, 3-12 inches high, No. 3545.

Lagenophora nudicaulis Dus. (=L. Commersonu Cass.) Delicate herb producing

slender runners amongst mosses and Nertera granadensis, flower mauve, evidently rare,

only one small colony found, 2,300 feet alt., No. 3569. The discovery and collection of this

minute rarity was one of the “ highlights ” of the climb in the drenching rain.

Senecio vulgaris 2. Pioneer annual 3-9 inches high in small clearing on the settlement

plateau, near cliff, No. 3527.

Sonchus oleraceus L. was observed in a garden but not collected.

COMPOSITION OF FLORA.

Phanerogams.

(Flowering Cryptogams.

Plants.)

Indi- Intro- os , Liver- ¢ 2

Prior to “ Car-

lisle’’ expedi-

HODooo000000¢ 42 32 30 38 11 12 2 19 186

Present record .. 42 38 30 50 11? 13 3 25? | +212

SUMMARY OF RESULTS.

) Several new records of ‘“‘ sea weeds’ (Algae) were made.

) One lichen was a new generic record for the island.

3) Among the mosses collected were four novelties and eight additional new records for the island.

)

Six flowering plants were recorded for the first time, all of which were probably accidentally introduced

to the island within recent times.

(5) Soil erosion is on the increase on the settlement plateau and the problem of combating it should be

seriously considered.

(6) Selected species of grass and trees from the Union Department of Agriculture and Forestry were

planted on the island as an experiment.

(7) Composition of flora (see table above).

(8) Illustrations.

Carmichael, Dugald.............-

Christophersen, E................

du Petit-Thouars, Aubert.........

Cees 1D, Mhosccoosecaneookoeno007

Inlenmslay VWo Ihe coanpns0ccoss000

Hooker, Sir Joseph.............-.

iMoseleyan blew Neracriieercciercer rca

Officers of H.M.S. “ Carlisle ”’......

TMS, WI caedeeancaas eH soes O10

IROL, INOS Mocseoocodoaceonoce

SchonlandsS-sceeeeeee ee eee eerie

Wild, Commander Frank..........

607

BIBLIOGRAPHY.

(SEE Nore UNDER GANE.)

** Some account of the Island of Tristan da Cunha and of its Natural

Productions,” Trans. Linn. Soc., Vol. 12, 1818.

Biographical Notice of the late Capt. Dugald Carmichael, F.L.S. in

Hook. Bot, Misc. 3, 1833.

“Plants of Tristan da Cunha.” Sci. Res. Norw. Antarct. Exped.

1927-28, No. 16, 1937.

“ Esquisse de la flore de Visle de Tristan d’ Acugna,’

de Botanique et de Voyages, 1811.

in Mélanges

Tristan da Cunha. London, 1932 (including a comprehensive

bibliography, mostly of a non-botanical nature).

“ Report on the botany of the Bermudas and various other islands

of the Atlantic and Southern Oceans,”’ pt. 2, Rep. Sci. Res. Voy.

’ Challenger, Botany, Vol. 1, 2, 1885.

Botany of the Antarctic Voyage, 1847.

* Notes on Plants collected in the Islands of the Tristan d’Acunha

group, Journ. Linn. Soc. Bot., Vol. 14, 1874.

* Tristan da Cunha in 1937,” The Geog. Mag., Vol. 6, Nov., 1937.

“A list of the phanerogams and ferns collected by Mr. P. C. Keytel

on the island of Tristan da Cunha, 1908-1909,’ Ann. S. Afr. Mus.,

Vol. 9, 1913.

The Lonely Island, London, 1926.

~ A Study of some facts and theories bearing upon the question of

the origin of the Angiospermous Flora of South Africa,” Trans.

S. Afr. Phil. Soe. 18, 1907.

Shackleton’s Last Voyage. London, 1923 (including botanical

notes by Dr. Macklin.

Report on the Scientific Results of the Voyage of S.Y. Scotia during

the years 1902, 1903, 1904, Botany 3, 1912.

ILLUSTRATIONS.

(PHOTOGRAPHS BY THE AUTHOR.)

(1) Settlement of Edinburgh on N.-W. of Tristan da Cunha.

(2) Approach to the settlement from landing beach: this area was covered by an “ impenetrable copse ”

in 1816, according to Carmichael : island cattle and young bull landed by H.M.S. Carlisle : erosion

in background.

(3) Landing beach, with severely eroded slopes in the background.

(4) Settlement stream with steeply eroded banks: in background New Zealand flax (Phormium tenax

Forst.) cultivated in a garden for thatching.

(5) Cottage roof thatched with Spartina arundinacea, pieces of turf along the top to prevent leakage

during storms: (New Zealand flax is cultivated and also used for thatching, see Fig. 4).

(6) Blechnum tabulare on precipitous slopes at about 2,400 feet alt. (heavy rain at the time).

(7) Bleehnum tabulare, as in Vig. 6, near view, with Arthur Rogers.

i an te. cKeOG

Boa ae as ee

duit: pene oe

Ve ip

i His Derg peold Vicker) Hy mic | ry om Fuel

AWE ni Seti : = i a i

Lae ee ae

eres: pak We 2 -

od) ane sins Pacaee Pro ct sai : : “4 : ; i :

; , 3 , iv soe

Meares: os 0 | +8 Ww ede. bat i! it Base

» Gale 1 bead gic aut ip wei ae nl Fi é. ioral

he ba eee hiteh

open liana et win pier shaw’ i a id ds gaia "a

Aa oe | dele Yigedie ile dramcte linedivldegiy, fly

sr sesh ih malig: the Kidige beter: call 7

. aeear crcl filet. Tel ghagl Woa.

san Syne: tt het hike abe jo Sha nen wom e

ae i ue Ce ne ev biditoinss ii |

SeeaL eet bs Pest dl fue)

: ert ms anit sins daz ig Ram

a G: ak digs OO Cas ooae

609

“eyunyy Vp UvYSTIT, JO

“MON

uo YSanquipy Jo quour9]y}eg—"T “SIT

LAS

—

Fig 2.—Approach to settlement from landing beach: this area was covered by an “impenetrable

copse”’ in 1816, according to Carmichael : island cattle and young bull landed by H.M.S.

Carlisle: erosion in background.

Fie. 3.—Landing beach, with severely eroded slopes in the background.

Fig. 4.Settlement stream with steeply eroded banks: in bac

tenax Forst.) cultivated in a garden for thatching.

Fig. 5.—Cottage roof thatched with Spartina arundinacea, pieces of turf along the top to prevent

leakage during storms: (New Zealand flax is cultivated and also used for thatching, see

Fig. 4).

613

A REVIEW OF THE GENUS ADROMISCHUS

LEMAIRE.*

By C. A. Smira.

[Indexed at end of the Article. ]

The genus Adromischus was first described by Lemaire (Jard. Fleur. II. Misc. 59) in

1852, two species—A. robustus and A. mucronatus—being described by him for the first

time, and nine species, originally described by their authors under Cotyledon Linn., were

transferred [according to Berger (1)] by Lemaire to his new genus. These species all had in

common a dwarf habit and a spicate or racemose arrangement of the usually erect flowers

round the upper part of the elongated and scape-like inflorescence-rhachis in 1—2—3-

flowered cymules, characters which these species do not share with typical species of Coty-

ledon Linn. (emend.) and which were utilised by Harvey (?) about 10 years later, when he

monographed the known South African species of the genus for the first time, for his Sect.

Spicatae. Under this section Harvey (?) described Cotyledon Zeyheri Harv., and redescribed

C. hemisphaerica Linn., C. cristata Haw. and C. mammillaris Linn. f., the incorrectness of his

treatment of these species being noted further on under their respective heads. Five

species, which really also belong to this section, were further enumerated by Harvey (?)

under the “‘ Imperfectly known and doubtful species,” and he seems to have been quite

ignorant of Lemaire’s paper, for he does not refer to the two species described by Lemaire

under Adromischus, nor to this generic name. Harvey certainly appears to have regarded

the characteristic inflorescence features as having no more than sectional value.

The same state of affairs seems to have obtained among all later authorities, for example

Dr. S. Schonland in his several papers on the South African species of Cotyledon. No

reference to Lemaire’s article is found until the late Professor Alwin Berger monographed

the Crassulaceae (1), vol. 18. Here Berger resuscitated Lemaire’s generic name and gave

the transferences of eight additional species, seven of which had been described under the

generic name Cotyledon in recent times, so that the genus Adromischus Lem., as circum-

scribed by Berger (4), then stood accredited with 15 species, Lemaire’s two species being

held by Berger to be conspecific with two previously described plants. In the present

paper, the writer has added a further 15 to Berger’s total, either as hitherto undescribed

species, or as new transferences, or new names, together with critical notes on some of the

older species, as well as including three from other German authors and not previously

referred to.

* This paper was completed in the year 1938, and references to some of the later authors, e.g. von

Poellnitz, have been inserted by the Editor. (See also Appendix.)

{ The latest area to yield an interesting series of species new to botanlcal science, is South-West Africa,

where Dr. Kurt Dinter has been responsible for the discovery of these novelties. Thus the present article

has included four from that Territory, but it would appear from current publications that correct identi-

fication of the species (judging from the descriptions published) is as rare as imaginative nomenclature

for the species.

614

As re-circumscribed for purposes of this paper, the genus presents the following out-

standing characteristics :

1. The stems are succulent in all species, though variously developed, being frequently

very dwarf [A. nanus (N. E. Br.), A. eristatus (Haw.), A. rupicolus *], usually about 10 cm.

high, seldom up to 90 cm. (A. kleinioides) and rarely obsolescent [A. humilis (Marl.), A.

Schaeferranus (Dtr.)] though one is described as a semi-shrub (A. montiwm-Klinghardtii).

As a rule they are unbranched, though a number have a few very short and stubby

or podgy branches [A. hemisphaericus (Linn.), A. umbraticolus, A. rotundsfolius (Haw.)],

while at least one [A. caryophyllaceus (Burm. f.)], is very distinctly branched, and are

erect and suffrutescent, being very rarely prostrate and rooting at the nodes [A. mammillaris

(Linn. f.)].

In almost all species the stems are devoid of a hairy ndumentum, but A. cristatus (Haw.)

and A. clavifolius (Haw.) are unique in the possession of numerous short and clo.ely inter-

woven reddish- to rusty-brown rather coarse and curly “aerial roots’? which have water

absorptive properties, according to Marloth. Such species as A. hemisphaericus (Linn.),

A. festivus, etc., are frequently found with tufts of adventitious roots arising from the

cauline nodes, but these should not be confused, at least as to function, with the aerial

roots of the two former species. A true hairy indumentum occurs only in such species as

A. Schaeferianus (Dinter), A. leucothriz, A. Zeyheri (Harv.), in which the leaves and in-

florescence are pubescent or hispid, while in A. cristatus (Haw.) and A. clavifolius (Haw.)

only the leaves are softly pubescent. In nearly all species of the genus, also, the younger

parts at least are densely covered with a whitish bloom caused by a waxy secretion, this

no doubt playing an important part in reducing transpiration losses.

The leaves, which are invariably thick and fleshy, are either alternate and closely

crowded (or scattered), or opposite—see examples in sectional groups below—or rosulate

[A. humilis (Marl.)], and, in several of the species, are beautifully spotted or blotched with

purple to chocolate-brown flecks and dots—A. festwus, A. Cooperi (Baker), A. Marianae

(Marl.), A. tricolor, A. fusiformis (Rolfe), A. maculatus (Salm Dyck), the spots or blotches

usually flowing into an irregular and larger blotch under the apical margin. (Here may

incidentally be noted that this blotching of the leaves is by no means so variable as some

authorities have assumed.) In shape the leaves vary from flat and obovate-cuneate, e.g.

A. rhombifolius (Haw.), A. rotundifolius (Haw.), A. sphenophyllus, to orbicular or sub-

orbicular or reniform, e.g. A. rupicolus, A. nanus (N. KE. Br.), to fusiform or spindle-shaped,

e.g. A. kleinioides, A. fusiformis (Rolfe), A. tricolor, ovoid- to subglobose and terete in cross-

section, e.g. A. mamillaris (Linn. f.), semiglobose [A. hemisphaerrcus (Linn.)], oblong or

oblong-elliptic and semiterete in cross-section [A. Marianae (Marl.)], sometimes with a

deep sulcus on the upper face (A. leucothriz). In all these the leaves are always distinctly

sessile, but, on the other hand, A. festivus, A. pachylophus, A. Zeyher: (Harv.), A. cristatus

(Haw.) and A. clavifolius (Haw.), share the unique property of having their much thickened

terete to semiterete leaves narrowed from about the middle or lower third of their length

into a distinct much thinner terete “ petiolar’’ portion, the apices being flattened and

crisped or undulate. In A. Cooperi (Baker), and A. pachylophus again, the apical part of

the subcylindric leaves is flattened and expanded into an ovate or ovate-rotundate broader

part, which is also a unique character inthe genus, and to which the specific epithet of the

latter refers.

The inflorescence invariably partakes of the nature of a typical spike or true raceme,

though a hard and fast line cannot be drawn between groups of species on this score alone,

since the flowers may be sessile even when mature, but become distinctly pedicelled in the

fruiting stages. In their arrangement the flowers appear more usually to be singly disposed

* Specific names cited without an author’s name are new, and are accompanied by descriptions or

validating references further on.

615

along the rhachis, which is most generally unbranched, but 3-flowered sessile or subsessile

cymules occur in the lower half of the inflorescences of such species as A. kleinioides, A.

triflorus (Linn. f.)—the specific epithet here being self-explanatory—A. Alstoni (Schonl.

and Bak. f.), and A. sphenophyllus, typically so in the first two, and from 1-3 in the last

two, whereas in A. maculatus (Salm Dyck) the number varies from 1-2 (so far as seen),

though here the 1-flowered condition would appear to be the more usual.

Two well marked conditions of flower colour occur in the genus, the Little Namaqua-

land and Little Karoo species having striated green corolla-tubes with salver-shaped limbs

which are either bright scarlet above and below or white above and scarlet to wine red on

the lower face, while the more northerly species (i.e. those found north and, in the N.E.,

just south of the Orange River), have purple corolla-tubes and dull purple limbs which

become completely reflexed over the apical part of the former, the throat being very

generally of a richer and deeper purple.

The leaves of several (if not all) species are capable of rooting from the base [A. wmbra-

ticolus, C. A. Sm., A. rupicolus, A. Marianae (Marl.)], and in course of time will give rise to

a new plant thus affording a ready method of propagation.

It would seem that at least some of the species of the genus are toxic to stock, an un-

determined species from Namaqualand [probably A. Alstoni (Schonl. & Bak. f.)], being

reported from that area as causing the so-called ‘*‘ krimpsiekte ”’ in goats, while experimental

feeding tests carried out with A. umbraticolus C. A. Sm. at Onderstepoort Veterinary

Research Laboratories (near Pretoria), have shown this species to be fairly toxic with thé

development of cotyledonotoxin poisoning symptoms (Cotyledonosis), closely resembling

those seen in animals poisoned by species of “ true ’’ Cotyledon.

Many of the species have already been accurately figured in colour (sub. gen. Cotyledon),

though several of the plants represented by these rather excellent plates have been erron-

eously identified, as a glance through the synonymy given under each species below (e.g.

A. sphenophyllus, A. kleinioides) will show, and it is not surprising to note that several of

the older “‘ Flora Capensis”’ species, in the light of critical examination, are at present

known only from the type gathering [A. Zeyheri (Harv.), A. filicaulis (E. & Z.), A. trigynus

(Burch.), A. triflorus (Linn. f.)]; some only from description [A. rhombifolius (Haw.)],

the type being apparently non existent ; some only from a figure [ A. hemispahericus (Linn.),

A. Cooper: (Baker), A. kleinioides, A. fusiformis (Rolfe)], of which the original specimens

were apparently never kept, while yet other species have only been re-discovered once or

thrice [A. maculatus (Salm Dyck), A. caryophyllaceus (Burm. f.), A. mammillaris (Linn. f.),

A. hemisphaericus (Linn.), the first in 1908, 1926 and 1930, the second in 1878, the third

in 1926 and 1930, and the last in c. 1824].

Many of the older species were prime favourites in European hot houses and all these

were described in the first instance from cultivated plants, the earliest known species (as

to actual date of publication) bemg A. hemisphaericus (Linn.), which was in cultivation

in Sherard’s famous garden at Eltham, near London, prior to 1737, when it was accurately

figured and described by Dillenius, Linnaeus subsequently (1762) naming the species from

this plate, while A. caryophyllaceus (Burm. f.) was figured by the elder Burmann in 1738,

and named from this plate by the younger Burmann in 1768. There is however, a great

deal of indirect evidence for believing that the elder Burmann figured the plant before

Dillenius figured his, Burmann’s figure being practically no more than a copy of a plate

executed in colour at the Cape during the time of the elder v. d. Stel’s governorship, i.e.

prior to 1699, and contained in the famous Codex Witsenius, which was never published,

but came to Burmann’s hands via the Commelins. Thus he quotes that the plant “A

Casp. Commel. in Catal. MSto ad Cod. Wits. vocatur Sedum Africanum montanum, foliis

orbiculatis, floribus parvis, variegatis; & in Cod. Wits. Sedum Africanum montanum,

minus, folio rotundo, flore ex rubro & albovariegato, ubi & dicitur quod in montibus

616

crescat inter rupium fissuras, florens Novembri.”” He also gives synonyms from Boerhaave

and Tournefort, as well as ‘“‘ Cotyledon Africanum, foliis oblongis, floribus umbellatis,

fibrosa radice, Oldenl. (sphalm. Olendenl.) Catal. Plant. Afric. p. 27,” while Plunkenet’s

(Mant. p. 169) “ Sedum Africanum, angustis longioribus folis, Jasmini floribus umbellatum,”’

also cited, obviously foreshadows a later epithet of the specific name Cotyledon jasmini-

flora by which the species was described by Salm Dyck (see p. 628).

It is not, of course, possible always to judge of the correctness of synonymy involving

the pre-Linnaean phrase names, but there is ample reason to believe that the Dutch authors

cited above, at least, were all dealing with the same species, living material of which had

been collected at the Cape, most probably by Oldenland during one of his excursions to the

eastern part of the Colony after plants, and sent by Governor Simon v. d. Stel to the Dutch

gardens at Amsterdam and Leiden. From these historical gardens plants were also sent

to many other gardens by way of exchange, both Plukenet and Tournefort, for example,

receiving South African plants on a number of occasions of which there are actual records.

It thus seems reasonably safe to assume that, in the particular case under review, all the

authors cited had the same plant in mind. This being the case, then it follows that his-

torically, A. caryophyllaceus (Burm. f.) is by far the oldest recorded species of the genus.

The Haworthian species of the genus (described by Haworth as species of Cotyledon,

of course), were all described from specimens sent by James Bowie from the Cape to and

cultivated at Kew between the years 1815-25, while Salm Dyck, who will be remembered

for his monumental volumes on the “‘ Aloes”’ and “‘Mesems,” also had specimens from the

Cape about the same time, principally from Ecklon, in addition to receiving some material

from his correspondent Haworth, and cultivated these in his famous succulent garden on

the Continent. Then there was a lapse in the introduction of new species of the genus until

Thomas Cooper visited South Africa to collect plant material for William Saunders of

Reigate in the early 60’s. Three species of the genus were later figured for Saunders’ “ Re-

fugium Botanicum ”—A. Cooperi (Baker), A. sphenophyllus, A. maculatus (Salm Dyck)—

from specimens sent to Reigate by Cooper, but it is only of comparatively recent date that

such species as A. eristatus (Haw.) and A. clavifolius (Haw.), etc., are again coming into

favour in Kurope, a fine collection of species of the genus being in cultivation in the Royal

Botanic Gardens, Kew (1930).

In general, unless exceptionally well dried, or accompanied by detailed notes on all

parts of the fresh plant, accurate identification of the species from herbarium or other dried

material is extremely difficult and in some cases well-nigh impossible. To obviate this

difficulty in the future preservation of dried specimens should be duplicated by spirit ma-

terial, as well as by accurate, wholly or partly coloured figures of the complete plant, to-

gether with notes made from the fresh plant of those features which are liable to disappear

in the preservative fluid. The writer has found a system of “ nature prints ’ made from

the fresh leaves very useful in reconstructing their shape from the dried material. For

this purpose cross-sections are cut at short intervals from one or more leaves with an old

razor blade or sharp penknife, the exposed surface being inked over at every successive cut

and carefully pressed on to a slip of paper in the exact order of the sections, this being

supplemented by a “print” from a median longitudinal section of one or more leaves,

care being naturally exercised not to exert undue pressure on the section in making each

“print.” The outlines are then very carefully inked over in india ink.

It is quite obvious, of course, that the system of “ nature prints ” could be used very

effectively in succulent genera such as Aloe, Haworthia, Orassula, Cotyledon, Euphorbia,

Trichocaulon, etc., and that where polymorphism in the leaves occurs, a representative

series of ‘‘ prints’ could be made and attached to the sheet on which the dried specimen is

afterwards mounted.

617

In maintaining Adromischus Lem. (emend.) as a genus distinct from Cotyledon Linn.

(emend.), chief reliance is placed upon the characters presented by the inflorescence, in which

the spicate to racemose arrangement of the 1-3-flowered cymules (as opposed to the panicled

inflorescence of typical species of Cotyledon), is correlated with the shape of the corolla-

tube, which is either somewhat ventricose above the base or narrowly cylindric, and the

fusion of the segments so as to form a salver-shaped spreading or at length reflexing 5-

toothed limb (rarely 5-lobed). Schonland (°), in his last paper on the genus Cotyledon,

maintains Harvey’s two original sections—PanicuLATAE (—Cotyledon Linn., emend.) and

SpicaTAE (—Adromischus Lem., emend.)—but comes to the erroneous conclusion that the

branching of the inflorescence of Cotyledon caryophyllacea Burm. f. (loc. cit. 151) bridges

over the gap between the two sections. The arrangement of the flowers along the inflores-

cence-rhachis and its few racemose branches is, however, that of a typical raceme, and the

structure of the corolla-tube and limb is that of other species of Adromischus Lem. (emend.),

i.e. Cotyledon § Spicatae Harv. Branching of the inflorescence-rhachis also occurs, for

example, in such species as A. trigynus (Burch.), A. wnbraticolus C. A. Sm., A. sphenophyllus,

A. caryophyllaceus (Burm. f.), etc., but in each case the flowers, apart from their structure,

are always spicately or racemosely arranged and not panicled at the apex of the branches.

The known species of Adromischus Lem. appear to be readily capable of being grouped

in two subdivisions, for which purpose (a), the inflorescence parts, and (b), the disposition

of the leaves may be utilised :

(a) Owing to the mceonstaucy of the character presented by the type of inflorescence

and the disposition of the flowers along the rhachis (1-2-3 at a node, though in some the

3-flowered condition remains constant), neither of the two characters could be employed

satisfactorily for subdividing the genus. The nature of the limb of the corolla, however,

is far more useful. Comparison, for example, of the corollas of such species as A. kleiniordes,

A. rotundifolius (Haw.), A. maculatus (Salm Dyck), A. caryophyllaceus (Burm. f.), and A.

mammillaris (Linn. f.), with those of A. rupicolus, A. nanus (N. E. Br.), A. wmbraticolus

or teeth on the corolla-lobes, but these are absent or at most very much reduced in the

latter group. This difference is also correlated with colour differences. Thus in the first

group the corolla-tube is green and slightly ventricose above the base, and the limb white

with (or without) rosy to pale scarlet flushes along the middle of each lobe, and usually

scarlet to rosy-red below, or the limb entirely reddish to wine red or rusty red-brown. In

the second group the corolla-tube is purple or purplish-mauve and cylindric, with a deep

purple or purply-mauve throat, the limb being similarly coloured on both faces, though

paler along the margins.

(6) A more obvious and conveniently described character for subdivision of the genus

into two well marked groups is found in the disposition of the leaves on the stem, a con-

sideration of which leads to the followimg :

I. Alternifolii *, Sect. nov.—Leaves alternate, usual closely crowded, though occasion-

ally loosely scattered along the stem or its branches, rarely subrosulate.

e.g. A. fusiformis (Rolfe), A. hemisphaericus (Linn.), A. mammillaris (Linn. f),

A. rotundifolius (Haw.), A. kleinioides, A. caryophyllaceus (Burm. f.), ete.

II. Oppositifolii, Sect. nov—Leaves opposite and decussate.

e.g. A. maculatus (Salm Dyck), A. Cooperi (Baker), A. festwus, A. Bolusii (Schonl.),

A. Marianae (Marl.), ete.

For further subdivision each of the two sections may be divided on the shape presented

by the leaves in cross-section, a character already employed by Berger (1), who did not,

however, employ sectional or subsectional names. Thus he separated those species with

* In some of the species belonging to this section, the leaves are apparently sub-opposite, but then the

one leaf is always much shorter than the one sub-opposite to it.

618

flattened leaves as one group, and all the other species known to him fell into another group

which he further subdivided according as to whether the leaves had a semiterete or terete

cross-section. Utilismg the characters here noted under each of the two sections above

proposed, it follows that two subsections of each section are distinguished from one another

on the same character, as will be seen from the following :

Sect. ALTERNIFOLII :

A. Platyphylli, Subsect. nov.Leaves flattened (i.e. breadth m anterior half much

exceeding the thickness, as seen 1m cross-section).

e.g. A. rupicolus, A. rotundifolius (Haw.), A. humilis (Marl.), A. nanus (N. EK. Br.),

A. umbraticolus C. A. Sm.

B. Heterophylli, Subsect noy.—Leaves terete or more or less semicircular (i.e. breadth

more or less equal to the thickness, as seen in cross-section).

e.g. A. cristatus (Haw.), A. clavifolius (Haw.), A. mammillaris (Linn. f.), A.

pachylophus, A. fusiformis (Rolfe), A. leucothrix, A. kleinioides, A. hemisphaericus

(Linn.), etc.

Sect. OPPOSITIFOLII :

A. Planifolii, Subsect. nov.—Leaves flat (i.e. breadth in anterior half much exceeding

the thickness, as seen in cross-section).

e.g. A. Bolus (Schonl.), A. Alstoni (Schonl. & Bak. f.), A. sphenophyllus, A.

maculatus (Salm Dyck), A. triflorus (Linn. f.), ete.

B. Crassifolii, Subsect. nov.—Leaves terete or semiterete (i.e. breadth more or less

equal to the thickness, as seen in cross-section).

e.g. A. festivus, A. Marianae (Marl.), A. Cooperi (Baker), etc.

Using the above as a basis, the species falling under each subdivision may very readily

be distinguished from one another by utilising such characters as are afforded by imdu-

mentum, blotching of the leaves, flower colour, ete. This article is not offered as a revision

of the whole genus, so that several species mentioned in the above general survey are not

mentioned further on again, the following notes referrig only to such species as (in the

author’s opinion) were wrongly interpreted, and such as may be regarded as hitherto un-

described, the tentative key being supplied for further discrimination between the species

dealt with, and to indicate the réle leaves and flowers play in the distinctions drawn.

In order to facilitate references to specimens dealt with, the herbarium in or from which

a particular specimen has been examined is indicated by the following abbreviations, but

those not seen, but cited, have the herbarium names only slightly abbreviated :

Pa, National Herbarium, Division of Plant Industry, Pretoria.

K, Herbarium of the Royal Botanic Gardens, Kew.

These abbreviations are inserted in brackets after each citation. This has the advantage

of indicating to others where types are preserved, and tends to eliminate confusion in the

interpretations of specific names assigned to the specimens examined.

619

KEY TO THE SPECIES.

Leaves opposite and decussate :

Leaves markedly flattened on both sides :

Corolla-tube green ; limb white or pale rosy :

Leaves blotched or spotted :

Leaves minutely dotted in the upper half.

Leaves with large blotches all over.

Leaves neither spotted nor blotched :

Leaves thickest in middle and upper half.

Leaves thickest in the lower half :

Flowers in 1-flowered cymules only.

Flowers typically in 3-flowered cymules :

‘ Corolla-lobes ovate, obtuse, white above.

Corolla-lobes deltoid acuminate, rosy above.

Corolla-tube never green, but usually dull purple-mauve ; limb pale

purple-mauve or mauve.

Leaves not flattened on both sides :

Leaves terete or subterete :

Apex of leaf expanded into a deltoid-ovate part broader than the

leaf itself; limb of corolla

wine-red and papillose in the throat.

Apex of leaf not as above ; limb of corolla white or rosy :

Leaves constricted at the base into a short petiolar portion, and

flattened at the apex, ashy-grey between the large blotches.

Leaves fusiform, the apex not flattened, green between the blotches.

Leaves oblong, flattened and subconcave above, convex below, thus

semiterete in cross-section.

Leaves alternate and scattered or crowded :

Leaves markedly flattened on both sides :

Leaves closely spotted, especially in the upper half, and with firm

white cartilagineous margins.

Leaves not as above :

Corolla-tube green; limb white to rosy or deep maroon above,

scarlet or maroon below ;

Plants acaulescent, tuberous-rooted, with rosulate leaves..

Plants distinctly caulescent, fibrous-rooted, with the leaves

crowded below the apices.

Corolla-tube never green, usually purplish-brown to -mauve; limb

mauve to purple on both sides :

Corolla-tube somewhat curved.

Corolla-tube straight :

Leaves subrosulate ; stems obsolescent.

Leaves linear-oblong to elliptic-oblong sometimes crowded, but

scattered ; stems welldeveloped :

Leaves ‘‘ ovate-cuneate or suborbicular ”’

Leaves oblong to oblong-cuneate or obovate-cuneate.

Leaves not as above :

Leaves flattened or subconcaye or subconvex above, but always

markedly rounded below :

| Leaves with a distinct indumentum :

. Leaves elongate and deeply sulcate on the upper face, closely

covered with rigid white bristly hairs.

Leaves not as in the former, “ almost spherical,”’ pubescent.

Leaves glabrous, at most with a waxy bloom :

Leaves semiglobose, papillose, with acute margins, at most 1-5

em. long.

Leaves obovate or spathulate to oblong-elliptic, epapillose and

glossy green, with rounded margins, up to 3 cm. long.

mAs

3. A

. A. rhombifolius.

maculatus.

sphenophyllus.

* A. Bolusit.

2. A.

20. A.

26. A.

13s A.

triflorus.

alston.

rupicolus.

Cooperi.

15. A. festivus.

12. A.

21. A.

23. A.

22. A.

DeeAr

ip Ae

IOs Z

5. A.

Us Zc

* The specific names not numbered are not referred to in the text.

tricolor.

Marianae.

nNanus,

humilis

. rotundifolius.

A. procurvus.

A. saxicolus.

tigynus.

umbraticolus.

. leucothrix.

. Schaeferianus.

hemisphaericus.

caryophyllaceus.

620

Leaves more or less terete :

Leaves abruptly narrowed into a much thinner ‘“ petiolar”’ part in the

lower third :

Stems densely covered with rusty- to red-brown aerial roots :

Leaves flabelliform with a flattened crisped apex, and nearly as

broad as long in the thick part. 17. A. cristatus.

Leaves elongate and subcylindric, slightly crisped at the narrow

flattened apex, the thicker part very much longer than broad. 18. A. clavifolius.

Stems at most with nodal tuits of adventitious roots :

Leaves glabrous, expanding into a broad ovate apical part much

wider than the leaves. 14. A. pachylophus.

Leaves pubescent, at most with a flattened but not expanded ;

apex. : 16. A. Zeyheri.

Leaves without a definite “ petiolar”’ part, at most only insensibly

tapering at the base:

Leaves 1-2, “ almost spherical ”’. A. sphaerophyllus.

Leaves numerous :

Stems prostrate and rooting at the nodes, the vegetative parts

very like those of Kleinia radicans. ll. A. mammillaris.

Stems (where developed) erect :

Stems simple or many, tall and over 10 cm. high : [hardtii.

“ Semi-shrub, with many stems; flowers greenish-red.” A, montium-kling-

Plants not as above ; stems simple and elongated :

Corolla limb pallid ; leaves flecked with purple. A. fusiformis.

Corolla limb deep maroon to red-brown on both sides ;

leaves unspotted. 10. A. kleinioides.

Stems very dwarf (or almost to absent) :

Leaves oblong, narrowed to the base. 9. A. Marlothii.

Leaves fusiform (‘‘ tereti-acuminata ”’), tapering at both ends. 8. A. filicaulis.

Leaves “ pea-shaped,”’ with red dots. A. Kethackw.

1. A. maculatus (Salm Dyck) Lem. ex Berger.

Of this species there is a very fine coloured figure of a complete plant in the collection

of drawings at Kew, dated “ February 15, 1824,’ when i was made from “a typical plant

received from Salm Dyck. Compared with an authentic specimen in Haworth’s Herbarium

at Oxford. Oct. 31,1901. N.E. Br.” In the above collection there is also a very careful

drawing of Haworth’s specimen, showing a complete inflorescence and two leaves, and

against the former Haworth noted “ Kew Sept. 28, 1824,” while against the two leaves

he noted ‘ Hot ironed. Hx. horto, Apr. 1827.’ The inflorescence of Haworth’s specimen

thus in all probability came from the specimen “ received from Salm Dyck,” and from the

extremely close match of the illustrations, there can be little doubt that they all represent

the same species, viz. Cotyledon * maculata Salm Dyck [Obs. Bot. in Cat. Hort. Dyck. 5

(1820), ex Haw., Rev. Pl. Succ. 21 (1821)]. Now the figure of C. maculata in Saund., Ref.

Bot. I. t. 35 (1869), agrees exactly with the figures already cited, except that the flowers are

always in pairs (one of which is generally a bud) in the lower part of the inflorescence, but

are singly disposed at the nodes in the apical part, whereas Haworth’s specimen and the

figures just cited all show the sessile flowers to be singly disposed along the rhachis, 1.e. in

the form of a simple spike. Specimens, again, collected by the writer at Robertson in

April, 1926, flowering at the Division of Plant Industry in December of the same year,

and undoubtedly referable to this species from their close match with the Kew plate, also

showed single flowers at the nodes of the inflorescence rhachis, as will be seen in the accom-

panying Figure 1.

* or convenience, the species in these notes are cited in the text hereafter by their old name under

Cotyledon (C.).

621

OID

A aed

Lee. 1G26-

Fig. 1._-A. maculatus (Salm Dyck) Lem. ex Berger. See text.

622

Thus it would appear that the plant may show a variation of from 1-2 (apparently

never more) flowers at the nodes, with the 1-flowered nodal condition as the more typical.

Such variation in what must be regarded as an originally 3-florous cymule is by no means

uncommon in the genus, occurring, for example, in A. sphenophyllus, the next species but

one. However, in selecting the type of the name C. maculata Salm Dyck, the choice lies

between the coloured plate made at Kew and Haworth’s specimens, since Salm Dyck does

not appear to have kept a specimen of the plant originally described by him. Haworth’s

specimens are made up of parts introduced, as noted above, at different dates, and the leaves

“ex horto ” were (in view of the different labelling) certainly not obtained from the same

plant as the inflorescence. The coloured plate in the Kew collection should therefore be

selected as representing the typical plant, the more so since it was made from specimens

named and sent as C. maculata by Salm Dyck himself.

With reference to the ‘‘ Refugium Botanicum”’ plate (t. 35), identified with this species

above, Schonland and Baker fil. (4): state that it ““may represent a spotted

variety of C. rhombifolia Haw.” as typified (on their authority) by t. 36 of the same work.

The latter, however, cannot be regarded as that species (see No. 3). Then, again, in his

last paper on the genus Cotyledon Linn. (sensu Fl. Cap.), Schonland (3), regards “C.

maculata Salm Dyck” as a doubtful synonym of “ C. rhombifolia Haw.”, stating: “ It is

. a little doubtful whether the plant he [Baker] figured [Ref. Bot. t. 35] as C. maculata

Salm Dyck is really that species,” but does not advance any reasons for this statement,

though he further suggests (3) that ‘“‘the true C. maculata Salm Dyck may be identical

[sic!] with C. trigyna Burch.” In both statements Schonland erred through erroneously

identifying specimens of C. nana N. EH. Br. as C. maculata Salm Dyck, from which it differs

conspicuously in the structure of its flowers, though agreeing with C. trigyna Burch. mn

habit and floral characters.

For convenience Salm Dyck’s original diagnosis may be here inserted :

maculis atro-rubentibus notatis. Floribus spicatis, subalternis.” Schonland and Baker fil.

(loc. supra cit.) giving the following notes made from Haworth’s specimens : “ Leaves few,

obovate or obcordate emarginate, apex obtuse with a short acumen, margin cartiligmeous

sometimes undulate, base cuneate, 4-2-4-:5 cm. long, 2-3-3-3 ecm. broad at the broadest

point which is about one fourth of the total length from the apex. Flowers sessile, solitary,

numerous, erecto-patent, alternate, arranged in a lax spike. Peduncle terminal, terete,

of a purplish colour. Calyx-lobes short (about 1 mm. long). Corolla tubular, somewhat

ventricose above the calyx, + 8 mm. long; lobes erect or erecto-patent, -- 3 mm. long,

acute.” With this description to go by, especially that part relating to the flowers, it is

difficult to understand how the aboye errors cropped up in Dr. Schonland’s paper.

The following represents the revised synonymy for the species :

A. maculatus (Salm Dyck) Lem. in Jard. Fleur. II. Misc. 60 (1852), ex Berger in Engl.

and Prantl, Nat. Pflanzenfam, Vol. 18, a. 416 (1930).

Cotyledon maculatus Salm Dyck, Obs. Bot. 5 (1820), ex Haw., Rev. Pl. Succ. 1821 ;

Kekl. & Zeyh., Enum. No. 1973 (1836) ; Harv. in Harv. & Sond., Fl. Cap. II. 378 (1861-62) ;

Baker in Saund., Ref. Bot. I. t. 35 (1869); Schonl. & Bak. f. in Journ. Bot. Vol. 40. 92

(1902).

C. alternans Salm Dyck ex Haw., Suppl. Pl. Suce. 26 (1819); non Willd. (1799).

A. mucronatus Lem. in Jard. Fleur. II. Misc. 60 (1852), ex Ind. Kew. & Berger (loc. cit.).

C. hemisphaerica Harv. in Harv. & Sond., Fl. Cap. II. 376 (1861-62), partim; non

Linn. (1762).

623

SoutH WestERN REGIon—Worcester distr.: “In aridis in convalle flum. Hex River,

prope De Doorns, 1600 ped., Jan. 1908,” Bolus 13044! (K.). Robertson distr. : Near Robert-

son Station, on the western slope of a rocky hill, associated with Gasteria carinata Haw.

in shady places, April 1926, flowering at Division Plant Industry, Dec. 1926, Smith in Nat.

Herb. 8875 ! (Pa).

CENTRAL ReEGIon—Oudtshoorn distr.: In a poort on shale of the Bokkeveld series,

between Oudtshoorn and Montagu Pass, April 1930, van Nouhuys! s.n. (Pa).

2. A. triflorus (Linn. f.) Berger.

The type specimen of this species was collected by Thunberg “ prope Zekorivier ”

[Zeekoerivier] in the Clanwilliam district, during Dec—Jan. 1774-75, and a much more

detailed and amplified description given by him (Fl. Cap. Ed. Schult. 396: 1823) than was

first supplied by the younger Linnaeus (1781), and is here quoted for purposes of reference :

“ Caulis carnosus, crassus, herbaceus, glaber. Folia: inferne sessilia, crassa, obtusissima,

subtruncata, inferne attenuata, basi teretiuscula, approximata, erecta, subincurva, glabra,

bipollicaria, ultra pollicem lata, pallida viridia. Omnino referunt folia Cotyledonis orbi-

culatae. Flores subterni, spica longa, spithamea, rhachis angulata. Bractea sub singulo

flore minuta. Perianthium monophyllum, 5-dentatum, erectum, viride, brevissimum.

Corolla 1-petala, tubulosa; tubus cylindricus, striatus, viridi-rufescens, subunguicularis,

glaber : Limbus 5-partitus patens : Laciniae ovatae, obtusae, intus albae, extus rufescentes,

lineam longae. Filamentae 10, tubo inserta, paulum adnata, subulata erecta, viridia,

tubo breviora. Antherae ovatae, minutae, flavae. Nectarii squamae 5, ad fundum ger-

minis, subexcisae, albidae. Germina supera 5, subulata, glabra, viridia. Stigmata acuta.

Capsula 5, subulata.”

So far as is known, the species appears not to have been found since Thunberg’s time,

though a specimen described by Lemaire in 1852 (loc. infra cit.) as Adroimischus robustus

Lem., is regarded by Berger (loc. infra cit.) as conspecific with Thunberg’s plant, though it

is not known from what locality Lemaire had his plant(s). Both Salm Dyck (Obs. Bot. 6:

1820) and Haworth (Rev. Pl. Succ. 19: 1821) described what they took to be C. triflora

Linn. f., but were guided probably more by the 3-florous condition of the cymules in the

inflorescence of their plants, which these exhibited i common with Thunberg’s specimen,

and their erroneous identifications were perhaps natural in view of the inadequate diagnosis

of C. triflora given by both the younger Linnaeus (loc. infra cit.) and Thunberg (Prodr. PI.

Cap: 83: 1794), which they must have consulted, since the latter’s “* Fl. Cap. Ed. Schultes ”

did not appear till 1823 (see also note under the next species, No. 3).

Then Harvey (?), for no clear reason, reduced C. triflora Linn. f., the type specimen of

which he had seen, under C. hemisphaerica Linn., though he cites the typical plant of the

latter as figured in Dill. Hort. Eltham. t. 95, f. 111 and DC. Hist. Pl. Grass. t. 87! The

leaves of the former species, however, differ so profoundly in shape size and cross-section

from those of the latter, that Harvey’s reduction is wholly unwarrantable (see also note

under No. 5).

A. triflorus (Linn. f.) comes closest to A. sphenophyllus (the next species), but differs

from this in several characters, such as size and leaf-shape, colour of flowers and shape of

its corolla-lobes, smaller and less excised nectarial scales, which taken in conjunction with

its ““ western” distribution, as opposed to the “ south-western” distribution of the other,

have led the writer to keep the two species apart. The following embraces the synonymy

treated above :

A. trifiorus (Linn. f.) Berger l.c. (416).

Cotyledon triflora Linn. f., Suppl. 242 (1781) ; Murr., Syst. Ed. xiv. 429 (1784) ; Thunb.,

Prod. 83 (1794); & FI. Cap. Ed. Schultes, 396 (1823) ; non auct. alior.

C. hemisphaerica Harv. |.c. 376, partim ; non Linn. (1762).

624

A. robustus Lem. in Jard. Fleur. II. Mise. 60 (1852), ex Berger L.c.

WEstTERN ReGIon—Clanwilliam Div.: Near the Zeekoerivier, Dec._Jan., 1774-75,

Thunberg. (Type in Herb. Thunb., Upsala).

3. A. sphenophyllus C. A. Sm., nom. nov.

A dried specimen of the type gatherig of the species, figured as C. rhombifolia Haw.

in Saund., Ref. Bot. I. t. 36 (1869), is in Herb. Kew., having been presented by W. W.

Saunders in 1877. In this the inflorescence is simple and the flowers singly disposed at the

nodes. Cooper’s original specimen, i.e. the wild plant (Cooper 2338! C. B. S., sine loc.

exact.) is also in Herb. Kew., and the inflorescences mounted on the sheet show one with

1-, and the other with 1—3-flowered cymules. Neither, however, represents the typical

form of Haworth’s species (v. seq.), but they both match a fine coloured illustration (in the

collection of drawings at Kew) of a plant which is, however, without inflorescence “ received

[as C. triflora] from the Prince of Salm in the year 1823,” and against which N. EK. Brown

has noted : “ This quite agrees with the leafy part of the specimen of C. triflora in Haworth’s

Herb. at Oxford. Compared Oct. 30, 1901.” There can be no doubt that the specimen

in Haworth’s herbarium was sent him by Salm Dyck at the same time that the latter sent

the specimen to Kew, and that they came from the same original gathering. This is further

borne out by the fact that both Salm Dyck and Haworth describe “ C. triflora Linn. f.”, and

their descriptions agree very closely, while there can be no doubt that also in inflorescence

details the latter species [i.e. C. triflora Salm Dyck (non Linn. f.)] agrees with those in C.

rhombifolia Baker (non Haw.), though the inflorescence of the former in Haworth’s her-

barium shows signs of having been injured by mechanical or biotic agencies. From the

descriptions and specimens available, however, it would appear that the cymules may be

1—3-flowered in this species.

Since neither of the two specific names may validly be applied to the species under

consideration, the following new name (with details of synonymy) is proposed for it under

Adromischus :

A. sphenophyllus C. A. Sm., nom. nov.

Cotyledon triflora Salm Dyck, Obs. Bot. 6 (1820); Haw., Rev. Pl. Suce. 19 (1821) ;

Schonl. & Baker f. l.c. (91), non Linn. f. (1781).

C. rhombifolia Baker in Saund., Ref. Bot. I. t. 36 (1869) ; Schonl. & Baker f. lc. (92) ;

Schonl. in Rec. Alb. Mus. Vol. 3, 154 (1915), excel. syn. ; non Haw. (1825).

A. rhombifolius Berger l.c. (416) ; non C. rhombifolia Haw. (1825).

“co. B. 8.” (sine loc. exact.) : Cult. spec. e Hort. Saund. leg. Cooper! Type (K) ; Cooper

2338! Syn-type (K).

CentTRAL REeGion—Willowmore distr.: On hillside near Willowmore, anno 1931,

Steyn ! s.n. (Pa).

N.B.—The specimen referred to under ‘* C. rhombifolia Haw.” by Schonl. & Baker f.

(loc. swpra cit.) as having “‘ flowered at Grahamstown in the Spring of 1898’ no doubt

belongs here.

4. A. rhombifolius (Haw.) Lem. ex Berger.

This species was originally described by Haworth (loc. infra cit.) from a non-flowering

mspecien which was apparently never kept, since there is no plant so named in Haworth’s

herbarium at Oxford. From his description, however, there can be little doubt that the

625

specimen figured and described by Baker under the name C. rhombifolia in Saund., Ref.

Bot. I. t. 36, is distinct in habit and shape of its leaves from Haworth’s plant. Fresh speci-

mens from between Oudtshoorn and Montagu Pass, and recently examined by the writer,

agree perfectly with Haworth’s description, so far as this goes, and serve still further to

distinguish Baker’s plant from the species under discussion. Unfortunately it has not been

possible to give an amended description of the species from the fresh material available,

but this defect may be rectified at a later date.

A. rhombifolius (Haw.) Lem. in Jard. Fleur. II, Mise. 60 (1852), ex Berger, l.c. 416,

quoad nom. sol.

Cotyledon rhombifolia Haw. in Phil. Mag. 1825, 33 ; DC., Prod. Vol. 3, 398 (1828) ; Harv.

lc. 878 ; Schonl. l.c. 154, partim, et excl. syn.

CENTRAL REeGIon—Oudtshoorn distr.: Between Oudtshoorn and Montagu Pass, 7

miles from North Station, in Bokkeveld series on outcrops of Table Mountain Sandstone,

April 1930, van Nouhuys! s.n. (Pa).

Originally described from specimens sent from the Cape to Kew by James Bowie in

1825, but there is no record so far known of where he collected his specimens, nor has a single

specimen been kept.

5. A. hemisphaericus (Linn.) Lem. ex Berger.

This species has apparently never been correctly interpreted by monographers of the

South African species of Cotyledon Linn. (sensu Fl. Cap.), except perhaps by Berger. It

was originally based by Linnaeus on a Dillenian figure—Hort. Eltham. t. 95, f. 111 (1738),

where it is named as “* Cotyledon Capensis, folio semiglobato.”” According to Druce and

Vines (Dillen. Herb. 165: 1907), there is no specimen of the species in the Dillenian Her-

barium at Oxford, nor is there such a specimen either in Herb. Cliff. or Herb. Linn. in Lon-

don, but Dillenius’ figure (which thus represents the type figure) is accompanied by such

an excellently drawn up description of the plant that, in view of the statement in the first

sentence above, this description is here given in full, the more so since the original work

may not readily be accessible to others: “‘ Cauliculi carnosi lenti sunt, non recta protensi,

sed incurvi & pleurumque tortuosi, laeves, spadicei, variis lineis cinereis, nunc rectis, nune

transversis, nunc inaequali ordine connexis distincti, quibus hine inde folia singularia, in

summitato vero plura temere apposita sunt, levi tactu decidentia colorem, tenuibus punctis

undique notata, crassa; inferius, seu ad basim cauliculorum, rotundiora, superius seu

versus summitatem magis plana, lenius nempe parte interiori elevata, exteriori vero, ut in

illis, protuberante & pulvinata, succulenta, sapore acerbo & adstringente praedita.

A palmari ad dodrantalem & pedalem subinde nascitur altitudinem, & caules carnoso-

lignosos, magis ramosos, acquirit, sed longo temporis spatio indiget, ut adhanc altudinem

perveniat ; lente enim nascitur, & facile ob succositatem putrescit.

* Porro flores aegre fert, nec eos una vice amplius vidisse memini, eosque imperfectos :

scapus erat dodrantalis subteres, glaber, obsolete virescens, secundum cujus longitudinem

gemmae quaedam tenues, in summitate vero flores oblonges sex septemve nascebantur,

quinque lineis impressis notati, virescentes, in extremitate purpurascentes, calyce mono-

phyllo, in quinque laciniae diviso excepti, & hujus ope scapo annexi, tenui ligula ab basim

praedita, qui quales suturi essent, dicere non habeo, cum apertos videre non contingerit.

. . . Haec Augusto mense observare licuit.”

Two other published figures of the above species, both in colour and both correctly

identified specifically, are known to the writer: Roth, Bot. Abhandl. & Beob. t. 6 (1787)

and DC., Hist. Pl. Grass. t. 87 (1799-1829).

626

Roth’s figure is excellent for the habit and leaves of the plant and typical of the species

as represented by the type figure. He figures the flowers, however, as being borne in sessile

pairs among the apical leaves on the short and podgy lateral upper branches of the stem,

and does not give a detailed description of the plant in the text, but in the index to the

plates in the work he describes the flowers as “in capituli speciem collecti, quorum bini hic

[i.e. in the plate] conspiciuntur,” and “ initia forte spicatum, quae tamen absque mutatione

molis perierunt.”” The young inflorescence is frequently (especially in cultivated specimens)

injured by aphids, and so gives rise to various teratalogical forms, such as suppression of

the main axis of the inflorescence, oppositely borne 1—6-flowered cymules, etc. Hence

the condition figured by Roth.

De Candolle’s exquisite plate again is perfect for the species, though not illustrating

the characteristic habit so well as Dillenius’ or Roth’s figure does, but showing the shape,

colouration and papillose texture of the leaves exceptionally well, these being further de-

scribed as “sparsa, sessilia, ovato-rotunda, subtus valde convexa et inde semiglobata,

subobtusa, pinguia, glabra, furfure punctata.”

In the Kew collection there is a fine coloured illustration of the type plant of C. cunei-

formis Haw. (Phil. Mag. 1828, 185), which is noted as “ Received from the Cape of Good

Hope in 1823 from Mr. Bowie.” Though not in flower at the time of figuring, the plant

agrees in all essential vegetative characters with the three plates of C. hemisphaerica Linn.,

and there can be very little doubt that Haworth’s plant is conspecific with the latter.

None of these plates presents any difficulty since they clearly represent one and the

same plant, yet so consistently has the species under review been misinterpreted by various

authors that one can only conclude that (excluding for the moment Haworth’s and Roth’s

figures as being less readily accessible) the Dillenian and De Candollean plates were never

properly referred to. Thus Schonland (*) reduces “ C. triflora Linn. f.” and “ C. rotundi-

folia Haw.”, both of which he had wrongly interpreted, under “ C. hemisphaerica Linn.”,

also stating (loc. cit. 153) that C. nana N. E. Br. “ evidently [sic !] belongs to this species

though it has only a one-flowered peduncle,” and this in spite of the very different and dis-

tinctive type of flowers produced by the latter and the conspicuous blotching of its leaves.

Of Haworth’s C. rotundifolia there is an exact drawing in the Kew collection of drawings

(a photo of the type specimen is given in Journ. Bot. Vol. 40, t. 435), and comparison of this

taken in conjunction with the existing descriptions of the species, indicates clearly the very

distinct specific differences between C. rotundifolia Haw. and C. hemisphaerica Linn. The

differences in foliage characters may best by illustrated by cross-sections as shown in the

following figure :

Kia. 2.—A. Median longitudinal section through the fresh leaf of A. hemisphaericus (Linn.), with

transverse sections in the regions X—Y and P—Q indicated by C ; B. Median longitudinal section through

the fresh leaf of A. rotundifolius (Haw.), with transverse sections in the regions X,—Y, and P,—Q, indi-

cated by D. The base is marked by an asterisk.

627

[C. triflora Linn. f. similarly differs from C. hemisphaerica Linn. in having leaves which

are flat, i.e. their thickness in the anterior half (as seen in cross-section) much exceeds the

breadth, apart even from the fact that they are opposite in that species and its flowers are

arranged in 3-flowered cymules along the inflorescence rhachis. |

Harvey (1) also reduces C. triflora Linn. f. (of which he had actually seen the type, and

which he cites as of “ Thunb. Fl. Cap. p. 396”) under C. hemisphaerica Linn., and in this

he may have been followed by Schonland, but from the previous paragraph it is evident

that this reduction is quite untenable. From the fact that Harvey describes the leaves as

“ flat, 1-2 inches long, 1-14 inches wide,” as well as other details, it would appear that he

based his description of ‘‘ C. hemisphaerica Linn.” for the greater part on the type specimen

(Thunberg’s) of C. triflora Linn. f. and Ecklon & Zeyher’s specimens of C. maculata and C.

rhombifolia. Baker had also, as far back as 1869, pointed out that Harvey’s synonymy fer

C. Lemisphaerica Linn. was erroneous, correctly laying stress on the shape, relative size and

colouration of the leaves of the latter species, of which he had himself seen live cultivated

specimens as well as de Candolle’s fine plate above noted, though Baker, like Harvey,

omitted to note the all important fact that in C. hemisphaerica Linn. the leaves are alternate,

and opposite in the other three species regarded as conspecific with it.

The writer has also examined the type plant of C. nana N. H. Br. (see No. 12) at Kew

and cannot endorse Dr. Schonland’s statement previously quoted. In habit and foliage

it approaches C. maculata Salm Dyck, but its leaves are alternate, while its floral characters,

as noted above, readily remove it from the group to which C. kemisp..ewiea Linn. belongs.

The revised synonymy for the latter species would then be as follows :

A. hemisphaericus (Zinn.) Lem. in Jard. Fleur. Il, Misc. 60 (1852), ex Berger, l.c. (416).

Cotyledon hemisphaerica Linn., Sp. Pl. Ed. ii. I. 614 (1762); Roth, Bot. Abhandl. &

Beob. t. 6 (1787) ; Willd., Sp. Pl. II. 756 (1799) ; DC., Hist. Pl. Grass. t. 87 (1799-1832) ;

Dryand. in Ait., Hort. Kew. Ed. u. III. 109 (1812); Haw., Syn. Pl. Succ. Ed. Germ. 116

(1819); Harv. lc. (376), pro minime parte, excl. syn. Thunb., E. & Z., et spec. Zeyh. et Drege ;

Baker in Saund., Ref. Bot. I. sub. t. 36 (1869), in obs.; Schcnl. lc. 152, pro minime purte,

et excl. syn. Linn. f. et Haw.

C. cuneiformis Haw., in Phil. Mag. 1828, 185.

Cotyledon capensis foliis semiglobosis Dill., Hort. Eltham. t. 95, f. 111 (1738).

The writer has so far not seen a single dried or living specimen of tnis interesting species,

which is historically the eldest in the genus, and no one appears to have re-discovered the

plant during the last 70 cdd years.

E. & Z. (Enum. 307: 1836) quote their No. 1970 as “ C. hemisphaerica Linn.’’, and as

being collected “inter saxa . . . laterum montis “ Leeuwenberg’ (Cap.),’’ but the writer

has not seen these specimens, and believes these to belong to A. rotundifolius (Haw.), the

next species, of which specimens have been collected in the same locality (=Lions Head).

6. A. rotundifolius (Haw.) C. A. Sm., comb. nov.

There is an accurate drawing of Haworth’s type in the collection of drawings at Kew,

a fair photographic reproduction of the type being also given by Schonl. & Bak. f. (+). These

authors state (loc. cit. 91) that this species is “‘ probably not specifically distinct from C.

hemisphaerica Linn.”’, but that “ the leaves are broader, branches less erect, and the caudex

thicker ” than in the latter. The last two characters are of doubtful value, since the branches

in typical specimens of C. hemisphaerica Linn. are frequently spreading and the caudex up

to 3 cm.in diam. The leaves, however, afford the best and most ready characters for dis-

tinguishing between these two species (see Fig. 2, and note under previous species), those

of the latter being semiglobose and not obovate to rotund and flat.

628

The following notes made from Haworth’s type by the above two authors will serve to

supplement Haworth’s description: “ Leaves subrotund, apex rounded, margin some-

times somewhat undulate, 2-5-3-5 cm. long, 2-2-5 em. broad, glabrous. Flowers sub-

sessile, solitary or rarely in twos, patent or erecto-patent, spicato-racemose. Calyx-lobes

deltoid, short, about 1 mm. long. Corolla tubular, tube + 1-1 cm. long, lobes ovate,

subacuminate, finally reflexed or subreflexed.’”’ In addition may be given the following

details drawn up from fresh specimens cited below and collected by the writer: ‘‘ Succulent

perennials growing socially in close masses between rocks and in fissures of rocks and other

crevices, usually in shady situations. Stems several from the crown of the rather shallow

growing fibrous root system, or single, usually very short and stout, with or without short

stubby branches which bear the leaves, rigid, terete, with numerous more or less concave

protuberances marking the old leaf bases and thus giving the stems an irregular outline,

leafy only in the upper third, and covered with a thin chartaceous greyish skin, glabrous.

Leaves alternate, scattered, very fleshy, suborbicular to obovate or obovate-cuneate, always

rounded at the flat or slightly crinkled apices, thickest at the flattened expanded base,

becoming thinner towards the apex, greyish-green in colour and unspotted.”

A. rotundifolius (Haw.) C. A. Sm., comb. nov.

Cotyledon rotundifolia Haw., in Phil. Mag. 1827, 273; Schonl. & Bak. f. lc. 91;

R. A. Dyer in Bot. Mag. t. 9368 (1934).

C. hemisphaerica Harv. l.c. 376, partim ; Schonl. |.c. 152, partim ; non Linn (1762).

C. Bolusw Schonl, l.c. 59.

A. Bolusw (Schonl.) Berger l.c. 416.

A. hemisphericus, Jacobsen, Succ. Pl. (Engl. trans.), 17 (1935); non Lem.

South WESTERN RecG1on—Cape distr.: Rocky crevices on west side of Lion’s Head,

above Capetown, Wolley Dod 2279! (K). Stellenbosch distr.: In fissures of rocks and in

crevices between rocks on the western slopes of the Hottentots Holland Mountains at Sir

Lowry’s Pass, near the tunnel, March 1931, Smith 6000! (Pa) et spec. cult. (Pa).

Here may probably also be referred Cooper 3628! (C. B. S., sine loc. exact.), preserved

im Herb. Kew.

This is the only species of the genus so far known, to reach the Cape South-West (see

also last paragraph under the preceding species).

? A. caryophyllaceus (Burm. f.) Lem.

This is one of the earliest known species of the genus, being fairly well figured and

described by the elder Burmann in his Rar. Pl. Afr. p. 39, t. 17, published in 1738, a year

after Dillenius published his figure of C. hemisphaerica, and specifically named by the younger

Burmann in 1768 (see also p. 615). It still remains, like the latter, one of the rarest of South

African plants, haying apparently been collected only twice since 1738. In 1818 Salm

Dyck had it from the Cape and described it as Cotyledon jasminiflora Salm Dyck (Obs.

30: 1820), under which name Haworth also received it about the same time from Salm Dyck,

neither recognising the much earlier name for the species in Burmann’s C. caryophyllacea.

And it was not until ¢, 1878 that Bolus rediscovered the plant “ in fissuris ruprum in monte

Tandjiesberg, prope Graaff-Reinet,” and for the first time definitely identified the wild

plant with that figured by Burmann nearly a century and a half earlier. Schonland and

Baker fil. were the first to point out, however, as a result of examining Haworth’s specimens

of C. gasminiflora Salm Dyck in his herbarium at Oxford (1902), that the latter plant was

conspecific with C. caryophyllacea Burm. f., both these specific names being referred to by

Harv.(?) under the heading of ‘“‘ Imperfectly known and doubtful species ” of Cotyledon.

629

In as much as no complete description has as yet been drawn up from living specimens

of the species, Burmann’ s original account is here given in full for purposes of ready reference

and convenience : “ Cotyledon foliis ad genicula plurimis planis, oblongis ; floribus gemellis,

erectis, Caryophyllaeis. Ex plurimis fibrillis tenuibus, flexuosis, intricatis, nigricantibus,

ex uno centro prorumpentibus oritur caulis unus alterve, qui in ejus summo folia gerit

plurima collecta, & ex uno quasi centro provenientia, quae sunt plana, obliqua, ex tenuiori

basi in latum apicem desinentia, crassa, venosa, glauca, limbo purpureo ; ex horum centro

erigitur caulis florifer tenuis, rotundus, viridis, qui ultra dimidium divaricatur, & in ejus

summo gerunt flores utcunque gemellos, singuli tamen suo petiolo proveniunt; suisque

calicibus tenuibus, oblongis, quinquefidis continentur; flores hi sunt tenues, oblongi,

tubulosi, forma fructus Caryophylli arboris referentes, in summo quinquefidi, segmentis

planis, angustis, acutis carneis, in singulo segmento linea rubra distinctis, seu variegatis.

Post hos sequuntur fructus teretes. in quinque loculamenta perpendiculariter sissi, quin-

queloculares, in singulo loculo continentes semina minima subrotunda, fusca’’. In addition

to the above the following notes, made by Schonland and Baker fil.(*+) from authentic speci-

mens in Haworth’s herbarium, will serve to amplify Burmann’s description: “ Stem rather

thick, branching, 6-0 cm. long, suffrutescent. Leaves fleshy, oblanceolate or oblong-

spathulate (convex above, rounded below, thickish), green, shining, 1-3-3-0 cm. long,

and -9-1-3 cm. broad at the broadest part, obtuse. Scape 13-15 em. long, 4-6-flowered,

sometimes the scape branches, and branches reach 5-5 cm. long, ascending. Pedicels

sometimes rather short, thickened obclavate, 3-5 mm. Flowers erect, with a green tube

and a revolute purple and white limb. Calyx lobes triangular, acute, 1-5 mm. long. Co-

rolla tube 1-4 cm. long, lobes ovate acute, nearly mm.long. Stamensincluded. Squamae

longer than broad ”.

The jomt authors of these notes (cit. 2 f.) regard this plant to be allied to C. hemis-

phaerica Linn. “in the structure of its flowers”, but the flower is almost exactly that of

C. rotundifolia Haw. (the previous species), C. maculata Salm Dyck, etc., which all certainly

have the same fundamental floral structure as that of C. hemisphaerica Linn., but the flowers

are much larger in size and of quite different colouration than those of this species. On the

whole this is one of the most well marked species in the genus, with a possible affinity, as

far as habit and floral characters are concerned, with C. rotundifolia Haw. The following

synonymy must supplant that so far given by authors :

A. caryophyllaceus (Burm. f.) Lem.in Jard. Fleur. II., Misc. 60 (1852) ex Berg. l.c. 416.

Cotyledon caryophyllacea Burm. fil., Prod. Fl. Cap. 13 (1768); DC., Prod. III. 398

(1828) ; Harv. l.c. 378, Schonl. & Baker f. I.c. 93; Schon. l.c. 151.

C. jasminiflora Salm Dyck, Obs. 38 (1820); Haw., Rev. Pl. Succ. 20 (1821); DC

Prod. IIT. 398 (1828) ; Harv. l.c. 378.

A. jasminifiorus (Salm Dyck) Lem. in Jard. Fleur. I. Misc. 60 (1852), ex Berger l.c.

416.

C. foliis ad genicula plurima planis, oblongis ; floribus gemellis, erectis, caryophyllaeis,

Burm., Rar. Afr. Pl. 39, t. 17 (1738).

c. B. S.”’ (sine loc. exact.): Cult. spec. e Hort. Dyck.! Type, in Herb. Haw. Oxon.

CentTraL Recion—Graafi-Reinet distr.: In rocky fissures on the slopes of the Tand-

jiesberg, near Graaff-Reinet, anno 1878, Bolus 758! (K).

Bolus states that the plant is extremely rare.

630

Cotyledon mammillaris Auct. non Linn. |.

Schonland (?) considers the plant figured in Bot. Mag. t. 6020 as representing the typical

C. mammillaris Linn. f., but, in as much as Thunberg (FI. Cap. Ed. Schult. 397) describes

the stem as “repens, radicans, . . . crassitie dimidia calmi...’”’ and the leaves as

““secunda, verticellata, instar mammillae ... unguicularia ...”, the corolla-tube as

“viridis... unguicularis” and the limb as “ plicatus .. . albido-purpureus .. .”’,

there can be no doubt that the Botanical Magazine plant (description below) is quite distinct

from the species described by Linneaus fil. and is thus without a valid name, unless either

of the two synonyms cited by Schonland (loc. swpra cit.), viz. C. filicaulis HK. & Z. and C.

Marlothii Schonl., may be resuscitated for the Botanical Magazine plant, this depending

on whether these plants are conspecific with the latter.

A comparison of Harvey’s (”) description of C. mammillaris Linn. f. with that of the

type of this name as given by Thunberg (loc. swpra cit.) immediately shows that Harvey

must have described almost exclusively from FH. & Z. 1975, the type of C. filicaulis BK. & Z.,

cited by Harvey as a synonym of the former, and Zeyher 2897. Thus Harvey gives “ stems

very short or scarcely any ; leaves crowded round the apex, or scattered along the short

stem .. . 14-2 inches long’, and these details certainly exclude the specimens cited from

C. mamillaris Linn. f. Moreover, the two species come from very different botanical areas—

the former from the Khamiesberge in Little Namaqualand, and the latter from the Oudt-

shoorn div. in the southern limits of the Central Region, and from the latter, as well as from

the Robertson district the writer has seen a number of fresh specimens which agree perfectly

in every detail with Thunberg’s description of his type specimen of C. mammilaris Linn f.

but certainly not with the Ecklon and Zeyher type material. The latter also differs very

markedly from the Botanical Magazine plant in its very much dwarfer habit, much shorter

racemose inflorescence in which the flowers are borne in 1-flowered cymules at the nodes,

and in the colour of its flowers. Hence C. filicaulis H. & Z. ranks as a species by itself which,

under Adromischus Lem. will bear the following name (with details of synonymy) :

8. A. filicaulis (H. & Z.) C. A. Sm., comb. nov.

Cotyledon filicaulis HK. & Z., Enum. 307 (18386).

C. mammillaris Harv. l.c. 377, pro majore parte, sed excl. syn. Thunb., et DC.,

non Linn. f.

WesteRN Recion—Namaqualand Minor: Sides of the Khamiesberge, Ecklon and

Zeyher 1975, type (Herb. Sond.) ; near Springbokkuil, Zeyher 2897! (Herb. Sond.).

C. Marlothii Schonl.(?) was described by its author from specimens gathered at Laings-

burg (Central Region) by the late Dr. R. Marloth in 1902. In habit and leaf shape it comes

nearest to C. filicaulis KE. & Z., from which it differs among others in the shape of its leaves,

differing also from C. mammillaris Linn. f. in the same characters already noted for the for-

mer, as well as in habit. In this character, too, it resembles C. hemisphaerica Linn., but

differs from this in its fusiform terete leaves. From the Botanical Magazine plant it differs

by its very much shorter racemose inflorescence in which the flowers are singly borne at the

nodes and differently coloured, and by its much dwarfer habit. Berger (1) was therefore

correct in regarding C. Marlothii Schonl. as a distinct species under Adromischus Lem. :

9. A. Marlothii (Schonl.) Berger l.c. 416.

Cotyledon Marlothu Schonl. 1.c. 59.

C. mammillaris Schonl. l.c. 153, in part. non Linn. f.

CenTRAL REGIon—Laingsburg distr.: Near Laingsburg, July 1902 (flor. in Hort. Alb.

Mus., Feb. 1903), Marloth 1520! Type (Herb. Alb. Mus.).

The Botanical Magazine plant thus appears to be distinct from all three species treated

above, and must therefore under International Rules have a new name, for which the fol-

lowing, with details of synonymy, is proposed :

631

10. A. kleinioides C. A. Sm. nom. nov.

Cotyledon mammillaris Hook. f. in Bot. Mag., Vol. 99, t. 6020 (1873); Schonl. l.c. 153,

partim ; non Linn. f.

Stem up to 50 cm. high and 2 cm. thick, ascending erect or ascending, sparingly branched,

with decurved tips, glabrous. Leaves alternate and scattered, spindle-shaped, acute,

narrowed to a broad base, up to 6 cm. long and 1-3 em. thick, terete, glaucous-green, gla-

brous. Spike up to 30 cm. long, at length pendulous, glabrous in all parts; flowers in

3-flowered nodal cymules in the lower part, with only the middle flower evident in the

younger stages of development in the upper part, the lateral flowers strongly divergent.

Corolla-tube dull yellowish-green or brown, up to 1-5 cm. long; limb dull reddish-brown,

on both surfaces, up to 6 mm. in diam., spreading-reflexed, with the lobes undulate and

furnished with long apical subulate processes. Nectarial scales minute, orbicular and

notched.

WESTERN ReGIon—Namaqualand Minor (without precise locality or collector).

This exceedingly handsome species is thus far apparently known only from the fine

coloured plate in the Botanical Magazine, the original specimen not being kept.

11. A. mammillaris (Linn. f.) Lem. ex Berger.

It is difficult to understand how this species, even though considered only from Thun-

berg’s fairly detailed description, could have been confused with the three preceding (see

also notes under these), so that for purposes of ready comparison with the notes made under

them, the original description of the type by Thunberg (though first shortly described and

named by the younger Linnaeus) is here given in full: ‘‘ Caulis repens, radicans, carnosus,

teres, glaber, crassitie dimidia calmi, ramosus, cinereus. Yolia subpetiolata, secunda,

verticellata, instar mammillae, utrinque attenuata, obtusa, carnosa, unguicularia, cinerea.

Pedunculus longus, filiformis, spithameus. Flores patentes, subpedunculati; pedunculi

breves. Tubus cylindricus, angulatus, viridis, glaber, unguicularis. Limbus 5-lobatus,

plicatus, patenti-reflexus, albido-pupureus, vix lineam longus. Filamenta 10, quorum 5

longitudine tubi et 5 breviora, tubo inserta, capillaria, albida. Antherae minutae, ovatae,

pallidae. Stigmata 5, truncata. Styli 5, subulati, longitudine staminum, breviorum.

Capsulae quinque.”

A. mammillaris (Linn. f.) Lem. in Jard. Fleur. II, Misc. 60 (1852), ex Berger, 1.c. 416.

Cotyledon mammullaris Linn. f., Suppl. 242 (1781); Thunb., Prod. 84 (1794); & FI.

Cap. Ed. Schult. 377 (1823) ; DC., Prod. Vol 3, 398 (1828) ; non Haw. (1821).

CENTRAL ReGIon—Oudtshoorn distr. : “ Olifantsbad ’’, Dec., Thunberg! Type (Herb.

Upsala) ; mm a poort on Bokkeveld series, between Oudtshoorn and Montagu Pass, April

1930, van Nouhuys! s.n. (Pa). Robertson distr.: On karoo-like hills, near Robertson,

March, Galpin 10334! (Pa). Ladismith distr.: On hills near Ladismith, Dec. 1926, Lie-

benberg 620! (Pa).

N.B.—The plant (Herb. Norm. Austro-Afr. 1860!) distributed by Macowan as “ C.

mammuillaris L. f.”, is A. Marianae (Marl.) Berger.

The leaves of the above species are unspotted, Berger (loc. cit.) erroneously placing the

species in his key under the group with richly spotted leaves, and so close is the resemblance

of the plant in habit and foliage to some specimens of Kleinia radicans (Thunb.) Haw.

[Phil. Mag. Vol. 62. 381 (1823)] and K. gonoclada DC. (Compositae), that distinction between

non-flowering specimens of these species is nearly impossible. In the fresh condition,

however, the species are readily distinguished by the turpentine-like flavour of the broken

leaves of the two species of Kleinia and the pale greyish-green longitudinal band which marks

their “ midrib.” These characters are not met with in A. mammillaris (Linn. f.).

632

The inflorescence described in detail by Haworth (Rev. Pl. Succ. 21: 1821) as belong-

ing to C. mammillaris is clearly that of another species of Cotyledon. This inflorescence,

which Haworth had “ ex horto regio Kewense . . . in Junio 1819”, showed the following

outstanding features: ‘‘ Flores terminales in racemo 3-4-floro parum paniculato .

Pedunculi graciles 6-12 lineares, erecti nutantesve, cum calycibus uti corollis, ramenatceo-

pubescentes . . . Corolla ventricosa . . . sordide fulvescens, laciniis 5 subrevolutis acutis

.. . Filamenta 10, sordide flavescentia, lente villosa . . .”—characters which agree

perfectly with those observed in the inflorescence of typical Cotyledon ventricosa Burm. f.

12. A. tricolor C. A. Sm., sp. nov.

Planta perennis, succulenta, in omnibus partibus glabra. Caules breves (ad 3 cm.

alti, ut videtur), crassi, simplices. Folia opposita et decussata, oblongo-cylindrica ad

oblongo-elliptica, teretes, ad basin molliter angustata, apice subobtusa, ad 6 cm. longa et

7 mm. lata, carnosa, in medio parte crassissima, cinereo-viridia sed maculis purpureo-

brunneis omnino notata. Inflorescentia terminalis, spicata, simplex, vel 1—2-ramosa ;

rhachis rigida, erecta, ad apicem subcernua, ad 25 cm. longa (pedunculus inclusus) ; ramuli

ascendentes, breves, pauciflori. Flores in cymulis 1-floris laxe dispositi, bracteis patentibus

lanceolato-subulatis membranaceis. Calyx carnosus, viridis; dentes ovato-deltoidei,

acuminati, ad. 1-5 mm. longi, saepe post fructus persistentes et spinescentes. Tubus

corollae cylindricus, rectus, obscure 5-angulatus, ad 1-5 cm. longus, viridis; lobi ovato-

lanceolati, acuminati, superne albi vel purpureis suffusi, inferne rubri; limbus patens,

deinde subreflexus. Ov.ria oblique-ovata, m stylo subulato longe angustata. Squamae

nectarii obovato-cuneatae, e marginatae, plus minusve dentibus calycis aequantes.

WestERN ReGion—Clanwilliam distr.: On dry hills near Brandvlei, 1,200 ft., Jan.

1896, Schlechter 9933! Type (Pa).

13. A. Cooperi (Baker) Berger.

First described from material collected by Thomas Cooper on the Zuurberg Range

(Uitenhage distr.) in 1860, and so far apparently known only from the excellent type figure

in Saunders’ Refugium Botanicum, Vol. I. t. 72 (1869), made from Cooper’s specimens which

flowered at Reigate. It is certainly one of the most distinct in the genus, being characterised

by its terete or subterete leaves which are markedly blotched all over and uniquely flattened

into a broader ovate-rotundate (“‘ spathulately dilated’, Baker) spotted apical portion,

and by the beautiful wine-red corollas with little papillae on the upper face of the basal

parts of the lobes. Yet Schonland and Baker f. (loc. infra cit.) suggest that it “ may only

be a varitey of Cotyledon maculata Salm Dyck”! of which incidentally Schonland had

anything but the correct conception (see note under No. 1). It is far more nearly allied to

the next species and A. festivus C. A. Sm., the next but one, but from the former it is readily

known by its spotted opposite leaves, while the latter differs in the curious apical portion

of its alternate leaves, as well as in floral characters—wider and shorter corolla-tube,

differently coloured parts, and absence of papillae.

A. Cooperi (Baker) Berger, |.c. 416.

Cotyledon Cooperi Baker in Saund., Ref. Bot. I. t.772 (1869); Schonl. & Bak. f. Lc.

91; Schonl. l.c. 153.

South Eastern Coastan Reeton—Uitenhage distr.: On the Zuurberg Range, anno

1860, Cooper! sn. Type (ic. col. tant. vidi).

633

14. A. pachylophus C. A. Sm., sp. nov.

Cotyledon Cooperi var. immaculata Schonl. & Bak.f. lc. 92.

Planta perennis, succulenta, in omnibus partibus glabra. Aadices ab eis origionibus

tuberosi, in ramulis fibrosis angustati. Caules breves (ad 3-5 cm. alti) crassique (ad 1-3

em. diam.), simplices vel ad apicem bifurcati ramulis brevissimis crassis, ad apices foliosi.

Folia 4-6 alterna et subdecussata, laxe disposita, oblongo-cylindrica sed ad basin in parte

breve terete crasso abrupter angustata instar -petiolata, et tertia parte superiore in apice

ovato-rotundata ad 3-5 cm. lata obtusa immaculata depressissima et expansa, ad 7 cm.

longa, carnosissima, glauco-viridia, immaculata. Jnfeorescentia laxe spicato-racemosa,

ad 30-flora; rhachis simplex vel ramosa, erecta, ad 30 cm. longa (pedunculus inclusus).

Flores sessiles vel subsessiles, in cymulis 1-floris laxe dispositi, bracteis patentibus deltoideo-

acuminatis membranaceis. Calyx subcylindricus, carnosus, glauco-viridis; dentes ovato-

acuminati, ad 2 mm. longi. T'ubus corollae cylindricus, rectus, ad 1 cm. longus, viridis sed

superiore dimidia parte pallido rubro suffusus; lobi ovati, acuminati, ad 4 mm. longi,

epapillosi, pallidi rubri. Ovaria 4-5, obliqueter oblonga, in stylo subulato longe angustata,

stigmatis capitatis. Sqeuamae nectarii obovato-cuneatae, emarginatae, dentibus calycis

in dimidio parte aequantes.

CENTRAL REGION—Graaft-Reinet distr.: Mountain side near Graafit-Reinet, Nov. 1897,

Rattray! s.n. Type (cult. Hort. Alb. Mus. annis sequentis). Queenstown distr.: On

Bowker’s Kop, near Queenstown, 3,800 ft., Nov. 1896, Galpin 2209! (Pa).

Galpin’s material here cited showed one of the uppermost leaves on each of the speci-

mens on his sheet, and these were distinctly smaller than the dimensions given by Schonland

and Baker f.(*) for the leaves of their material. In other details, however, Galpin’s speci-

mens agreed very well with the rather scanty details given by the authors cited, and the

writer has little doubt as to the conspecificity of the Rattray and Galpin specimens. Both

localities should again be combed for additional material for purposes of verification, and

for further amplifying the description. In the above, inflorescence details and basal parts

were taken from Galpin’s specimens and the leaves only described from the Rattray plant.

This species is most closely allied to the previous one, but from this it is readily known

by its unspotted alternate leaves of which the apical crest is far more markedly expanded,

and by the epapillose bases of the corolla lobes.

15. A. festivus C. A. Sm., sp. nov. (Fig. 3).

Planta perennis, succulenta, in omnibus partibus glabra, sed omnino pulvo albescente

(Anglice “ bloom ” dicta) tenuiter obtecta. Caules robustus, ad 4 cm. alti, sed speciminibus

cultis multo altior et saepe ad basin foliorum nodis radices adventitiones emittentes. Folia

opposita decussataque, ovoideo-cylindrica, gradatim et obtuse ad basin tertia parte inferiore

in parte breve multo angustiore terete circa 5 mm. longe angustata instar crassiter petiolata,

deinde abrupter in amplio basi amplectente expansa, molliter ad apicem deltoideum vel

rotundatum vel obtusum saepe crispulatum maculatum depressissimum attenutaa, ad

5 cm. longa et 1-5 cm. lata, carnosissima, teretia vel subteretia, cinereo-viridia, maculis

purpureo-brunneis pulchriter omnino notata (maculae frequenter infra apicem mergentes),

patenti-ascendentia et saepe incurvata. Inflorescentia laxe spicata, simplex; rhachis

erecta, ad 35 cm. longa (pedunculus inclusus). Flores in cymulis 1-floris laxe dispositi,

braceteis deltoideo-ovatis acuminatis membranaceis. Calyx viridis, carnosus, cupuliformis ;

dentes deltoidei, subacuminati, ad 1-5 mm. longi. Tubus corollae cylindricus, viridis vel

brunneo-viridis, ad 8 mm. longus ; lobi ovato-acuminati, ad 2 mm. longi, superne alba vel

pallido roseo suffusi, inferne rubri. Ovaria semi-ovoidea, longe in stylo acuminato angus-

tata; stigmata capitata. Squwamae nectarii obovato-cuneatae, emarginatae, calycis denti-

bus subaequantes.

634

Fic. 3.—A. festivus C. A. Sm. (see text).

635

CENTRAL ReGIon—Graaff Reinet distr.: Near Graaff Reinet, on slopes of rocky hill

above the reservoir, April 1926 (flowering at the Division of Plant Industry, Dec. 1926),

Smith in Nat. Herb. 8876! Type (Pa).

The accompanying plate (Fig. 3) was made from the type specimen at the time of

flowering (Dec. 1926).

The type agrees exactly with a photograph of a specimen cultivated by the late Dr.

N. E. Brown in his conservatory ay Kew during July 1920, and sent by him to the Division

of Plant Industry, Pretoria, sometime after the specimen had flowered in January of the

following year, together with the name which is here taken up for the first time.

It is most closely allied apparently to A. Cooperi (Baker), but may be readily known

from this by the shape and colouration of the leaves and their different apices, as well as

the absence of the peculiar papillae on the corollas of its ally.

16. A. Zeyheri (Harv.) von Poellnitz in Cactus Journ., 1958, 68.

This species was regarded by Berger (1) as conspecific with C. cristata Haw., probably

following on a note by Schonl. & Baker f.(?). The absence, however, of the very charac-

teristic shaggy reddish aerial roots from the stems and the pubescent peduncles and corollas

easily serve to distinguish it from the latter. The stems of the type plant are also “ half-

recumbent, rooting at the nodes’ (Harv.), a condition not observed in C. cristata (Haw,

so that the plant will bear the following name under Adromischus Lem. :

A. Zeyheri (Harv.) von Poellnitz l.c.

Cotyledon Zeyheri Harv. |.c. 397; Schonl. & Baker f. l.c. 91; Schonl. lc. 155.

SoutH WESTERN REGIoN—Swellendam distr.: “ Rocky places on the Kenko Rivier,

east of the Buffeljachts Rivier’, Zeyher 2571! Type (Herb. 8. Afr. Mus.).

Schonl. and Baker f. also cite Rattray (sine num.) from Graaft-Reimet and Schonland

709 from “rocky places near Grahamstown ” under the above species, stating “ they have

been compared with Zeyher, no. 2571, . . . and also with the type of C. cristata Haworth

(Phil. Mag. 1827, 274), and we think these two species should be united ”’.

The differences between the two species have been indicated above, and the writer has

no hesitation in referring the two specimens just cited to A. cristatus (Haw.) ; v. seq.

17. A. cristatus (Haw.) Lem. ex Berger.

See also note under the previous species for distinguishing features between it and A.

Zeyheri (Harv.), and under the following species for other differences between it and A.

clavifolius (Haw.), with both of which it has been confused. Thus, as previously noted,

Schonland and Baker f. (*) unite it with A. Zeyheri (Harv.)—as species of Cotyledon. How-

ever, so doubtful was Schonland himself of this that in his last paper on Cotyledon (°) he

resuscitates C. Zeyheri Harv., but unites C. clavifolia Haw. with C. cristata Haw., keeping

up the latter name. The following are the full details of synonymy and citations for the

species under Adromischus Lem. :

A. cristatus (Haw.) Lem. im Jard. Fleur. II, Misc. 60 (1852), ex Berger, |.c. (416), exel.

syn. Zeyh. Jacobsen, Succ. Pl. (Eng. trans.), 17 (1935), incl. fig. 2 sed excl. syn. C. Zeyhera

eles”

Cotyledon cristata Haw. in Phil. Mag. 1827, 274; DC., Prod. Vol. 3, 399 (1828) ; Eckl.

& Zeyh., Enum. 307 (1836) ; A.P. & A. DC. in Mem. Soc. Phys. Hist. Gen. Vol. 7 t. 7 (1836),

bona ; Harv. l.c. 376; partim, excl. syn. Haw. ; Schonl. l.c. 155, partim. et exel. syn.

Haw. ; Marl., Fl. S. Afr. Vol. 2, 1-20, t. 9-5 (1925); Phill. in Flow. Pl. 8. Afr. Vol. 9, t.

325 (1929). C. Zeyher Schonl. & Baker f. l.c. 91, excl. spec. Zeyh. ; non Harv. (1861-62).

636

SourH-HAsreRN CoastaL Recion—Uitenhage distr. : (sine loc. exact.), Bowie! s.n. Type

in Herb. Haw., Oxon.: dry hills on the Zwartkops Rivier, HE. & Z. 1974, partim (Herb.

Sond.) ; Port Elizabeth distr.: At Redhouse on dry hills, April 1915, Mrs. Paterson 442 !

(Pa).

CenTRAL Recron—Albany distr.: In rocky places, near Grahamstown, Schonland

709! (Herb. Alb. Mus.). Graaff-Reinet distr.: At Graafi-Reinet, cult. spec. plur. e Hort.

Div. Pl. Ind. in Nat. Herb. 7945! et viv. spec.! (Pa); dry hills near Graaff-Remet, Nov. 1897,

Rattray! s.n. (Herb. Alb. Mus.) ; occasional along basal slopes of mountain, near Graaff-

Reinet, March 1930, Galpin 10275! (Pa).

Haworth’s type specimen is preserved in his herbarium at Oxford, and is represented

in the Kew collection of drawings by an exceedingly fine accurate coloured plate (No. 264)

made from specimens sent by Bowie from the Cape (undoubtedly from the Uitenhage divi-

sion) to Kew in 1823, the plate beg prepared when the plants flowered in September 1826.

That the original of the plate and Haworth’s specimen both came from the type gathering

is evident from the date of flowering cited and the fact that Haworth notes that his specimen

flowered at Kew (Sept. 1826), N. E. Brown noting that he compared the plate with Ha-

worth’s type and found them to agree perfectly (Oct. 1901). Moreover, there is ample

reason to believe that, in view of the rarity of the material sent by Bowie, a single specimen

was figured and this same specimen passed on to Haworth.

This plate has also been very carefully compared with the original of the “ Flowering

Plants ” coloured figure and found to match exactly, both also agreeing in every respect

with A. P. & A. De Candolle’s fine plate cited above.

18. A. clavifolius (Haw.) Lem. ex Berger.

This species was regarded by Harvey (*) as doubtfully conspecific with the previous

Baker f. (*) ‘had united C. Zeyheri Hary., Schonland (?) reversed this by excluding the last

species, but united C. clavifolia Haw. with the former under that name. This latter step

is by no means justified by reference to the type material and excellent figures, for they

show most marked foliage and even floral differences. This was already noted by Haworth

(Phil. Mag, 1827, 274) where he states “ Priori (i.e. C. cristata Haw.) valde affinis at abunde

distincta videtur. Tolia subtrientalia, plusquam duplo angustiora, petiolo [svc /] incurvo,

vix puberulo, ramentis caulinis forte paucioribus ; cum eodem modo florendi; tubo sub-

incurvo robustiore, viridi ; laciniis intus albis, extus (uno latere) purpureis, et basi undulato-

sublobulatis ut in priori”. There is no specimen of C. clavifolia Haw. in Haworth’s her-

barium at Oxford, but there is a very fine coloured plate of a plant so-named in the collec-

tion of drawings at Kew, and, as explained under the previous species, there is every reason

to believe that the plate represents Haworth’s type, or at least an identical specimen, the

material coming from the same source as the former species and flowering at Kew at the

same time, Sept. 1826. This coloured plate must therefore, in the absence of a type specimen

be chosen as the type.

The original specimen of the plate given in “ Flowering Plants ” (cited below) agrees

perfectly with this plate, having been compared with it at Kew by the writer, and, as the

original collections from which tt. 325, 328 of this publication were made, are still (1932)

in cultivation at the Division of Plant Industry, Pretoria, where they had then been growing

in the same greenhouse under similar conditions for nearly 6 years, typical specimens of the

two Haworthian species were available for comparison. Thus the longer ovoid-elliptic

to subeylindric terete dark green “ body” of the leaves of C. clavifolia Haw., with their

flattened but scarcely or not crisped and more or less abruptly constricted apices, contrast

markedly with the much shorter dorso-ventrally somewhat compressed obovoid-cuneiform

637

to flabelliform paler green “ body ” of the leaves of C. cristata Haw. with their flattened and

conspicuously crisped or undulate scarcely or not at all constricted apices. Their floral

differences can best be seen by comparison of the two plates already cited.

Schonland and Baker f. (4) have suggested that in leafshape “it is more nearly allied

to C. Cooperi”’, but in this character it is far more nearly allied to A. festivus C. A. Sm.,

which has the same type of leaf apex. In the possession of the peculiar “ petiolar ” leaf

base, however, A. clavifolius (Haw.) is far more nearly related to A. cristatus (Haw.) and

A. Zeyheri (Haw.), with which it also agrees in indumentum and inflorescence characters,

so that these two species would seem to be its closest affinities. The following would be the

synonymy and citations under Adromischus Lem. :

A. clavifolius (Haw.). Lem. in, Jard. Fleur. II, Mise. 60 (1852), ex Berger, l.c. 416,

f. 199, B—E (1930) ; Jacobsen, Succ. Pl. (Eng. trans.), 16. excl. ref. fig. 6, 1, quae est spec.

distincta.

Cotyledon clavifolius Haw. in Phil. Mag. 1827, 274; DC., Prod. Vol. 3, 399 (1828) ;

Schonl. & Bak. f. l.c. 92; Schonl. l.c. 155.

C. cristata Harv. |.c. 376, pro parte ; non Haw. (1827).

C. Schonlandii Phill. in Flow. Pl. 8. Afr. Vol. 9, t. 328 (1929).

“A. van der Heydeni Hort.” ex Berger (loc. sypra cit.).

A. Schonlandii (Phill.) von Poellnitz m Cactus Journ. 1938, 68.

SoutH HasteRN Coastal Recion—Uitenhage distr. : (sine loc. exact.), Bowie! Type

(tc. col. ined. tant. Herb. Bot. Reg. Kew. vidi) ; on dry hills on the Zwartkopsrivier, Dec.,

Eckl. & Zeyh. 1974, partim (Herb. Sond.).

CENTRAL ReGion—Graafi Reinet distr.: Near Graaff Reinet, cult. spec. plur. e Hort.

Div. Pl. Ind. in Nat. Herb. 7944! (Pa, K).

In Harvey’s description of C. cristata Haw., principally drawn from H. & Z. 1974

(Herb. Sond.), he states “leaves 1-1} inches long . . . varying in breadth from 2 to 6-8

lines; the narrower forms answer to the description of Haworth’s C. clavifolia”’. Both

these two species grow in the same locality, and, though the writer has not seen the Herb.

Sond. specimens of E. & Z.’s gathering, there can be no doubt that this number is composed

of a mixture of the two plants, a point sufficiently evident from Harvey’s observations.

K. & Z. (Enum. 307: 1836) only list “ C. cristata Haw.” under their number 1974, but

without descriptive detail.

19. A. leucothrix C. A. Sm., sp. nov. (Plate I.)

Planta nana, succulenta, perennis. Cavles crassi, ad 4 cm. alti et 3 cm. diam. eis

basibus tuberosis sed gradatim ad apicem foliosum angustati e parte supra basin nudo

ad 6 mm. diam. subcylindrico, simplices vel ramis e basi brevibus paucis, laeves, glabri.

Folia alternata et circum apices caulium vel ramorum dense aggregata, lineari-oblonga ad

oblanceolata vel elliptico-oblonga, vix basin angustata, sed gradatim ad apicem obtusum

angustata, ad 3-5 cm. longa et 5 mm. lata, ascendenti-erecta vel inferiores patentia, car-

nosissima, saturato-viridia et immaculata, prope basin subteretia, profunde in facie superiore

sulcata sed inferire convexa, omnini crinibus hispidis albidis dense obtecta ut in foliis Cras-

sulae mesembrianthemoidis (Haw.) D. Dietr. Inflorescentia simplex, pauciflora, racemoso-

spicata ; rhachis erectus vel ascendenti-erectus, debilis, in inferiore parte dense sed breviter

crinibus hispidis albidis obtectus. Flores non visi. Carpellae modo generis.

638

CENTRAL ReEGIon—Ladismith distr. : On the eastern slope of a hill along the Ladismith-

Calitzdorp road, Dec. 1926, Liebenberg! s.n. Type (Pa).

The above species differs from all the other known species of the genus in the conspicuous

white bristly unicellular hairs on the deeply sulcate leaves and lower part of the peduncles,

thus giving these organs a distinct hoary appearance.

Described in greater part from notes made on the fresh type plant shortly after planting

at Pretoria (Dec. 1926), and from the original photograph taken by the writer at that time

and reproduced as Plate 1.

A later examination (April 1931) of the same plants (i.e. after they had been

in cultivation in a warm green house for nearly five years) showed no variation in the density

and character of the hairs on the leaves and peduncles, but the following dimensions will

indicate the influence of green house conditions on a plant coming from an area with a mean

annual rainfall of 15 inches. : Stems up to 15 em. high and 1-5 em. thick, with the bulbous

swollen base 6 cm. in diam. Leaves up to 11 cm. long, 1 cm. wide and 8 mm. thick, and still

with the deep sulcus on the upper face.

Apart from the character of the leaf indumentum, the species also exhibits other leaf

characters (shape, cross-section, the deep groove) which make it unique in the genus, since

none of the other species shows any character like it, but the nearest approach to the type

of leaf in the above species is met with in Cotyledon Wallichii Harv., some narrow leaved

forms of C. ventricosa Burm. f., C. Dinteri Bak. f., and C. cacalioides Linn. f., though, of

course, the leaves of none of these in the fresh state show the grooving or indumentum

which characterises the Adromischus. The non-flowering plant also bears a strong super-

ficial resemblance to Crassula mesembrianthemoides (Haw.) D. Dietr. =(Cr. trachysantha

HK. & Z.), but from this it is readily known by its succulent (not shrubby) stems, alternate

(not opposite) leaves and their shape (not triangular in cross-section).

20. A. Alstoni (Schonl. & Bak. f.) C. A. Sm., comb. noy.

Described from a specimen collected by G. Alston in Namaqualand and cultivated by

Dr. Schonland in the Albany Museum Herbarium garden, where it flowered in Jan. 1901.

It is very closely related to A. triflorus (Linn. f.), from which it appears to differ in its longer

but narrower leaves, in the colouration of the flowers, and in the shape of the corolla-lobes.

Unfortunately, the type of A. triflorus (Linn. f.), Thunberg’s Zeekoerivier specimen, was

not available for examination, so the characters in the following table under this name

were taken from Thunberg’s excellent description made obviously (in greater part at least)

from notes on the fresh plant at the time of collection :

A, Alstoni. A. triflorus.

Folia ad 7 cm. longa et 2-5 cm. lata. Folia ad 5-5 em. longa et 2-5 cm. lata.

Corollae tubus viridis, ad 1-5 em. longus. Corollae tubus “ viridi-rufescens, subun-

guicularis.”

Corollae lobi deltoideo-acuminati, pallide Corollae lobi “ ovatae, obtusae, intus albae,

rosel vix 2 mm. longi. extus rufescentes, lineam longi.”

Thunberg’s description does not include the leaf shape of his plant, though he says :

“folia. . . . obtusissima, subtruncata, inferne attenuata, basi teretiuscula”’, and from

the dimensions above given, this would indicate an obovate leaf long-cuneate to the base,

Alston’s specimens also having the leaves obovate and long cuneate to the base and rounded

at or only very slightly tapering to an obtuse apex. In the latter, too, the flowers are

sometimes singly disposed along the rhachis, or in 3-flowered cymules, whereas in the Thun-

berg specimens, the 3-florous condition seems consistently to obtain, at least in the maturer

basal half of the inflorescence.

639

The species appears to have been omitted by Berger (1), hence the following new com-

bination is proposed for it under Adromischus Lem. :

A. Alstoni (Schonl. & Bak. f.) C. A. Sm., comb. nov.

Cotyledon Alstoni Schonl. & Bak. f. l.c. 93; Schonl l.c. (154).

A. Cooperi Jacobsen, succ. Pl. (Engl. trans.) 16 (1935) excl. syn.; non Berger.

WESTERN ReEGION—Namaqualand Minor (sine loc. exact.), anno 1900, G. Alston! s.n.

Type (Herb. alb. Mus., K).

Of this plant the writer has only seen the dried specimens taken from the type collection

at the time of flowering. Jacobsen (loc. supra cit.) described a plant which appears beyond

doubt to be A. Alston.

21. A. Marianae (Marl.) Berger.

This is another of the species confined to the western area (Clanwilliam distr.) of the

Cape Province, and is well characterised by its oblong greyish-green fleshy leaves which

show a more or less semicircular cross-section and are conspicuously marked with purple-

brown flecks and blotches over their whole surface. The fine coloured plate of the plant

cited below also shows a rooting leaf, which had originally broken off from a fresh specimen

and left lying on a shelf in the late Dr. R. Marloth’s laboratory. Here, after a period of

several months, the leaf developed adventitious roots and new leaves at its base.

A. Marianae (Marl.) Berger, l.c. 416.

Cotyledon Marianae Marloth in Trans. 8. Afr. Phil. Soc. Vol. 18, 47 (1907); Schonl.

lc. 153; Marl., F]. S. Afr. Vol 2, 1. 14. t. 3, f. A (1925), bona.

WEstTERN RecIoN—Clanwilliam distr. : Stony slopes of a mountain near Clanwilliam,

1,000 ft., March 1898, Leipoldt in Herb. Norm. Austro-Afr. Cent. XIX, 1860! (K, Pa).

Leipoldt’s specimen was erroneously distributed by MacOwan in 1899 as “ Cotyledon

mamumillaris Linn. f.”, but agrees with every detail of the fine plate of the type plant, of

which, however, the writer has not seen the dried material.

22. A. humilis (Marl.) Berger.

One of the most distinct species in the genus, with the following most outstanding

characters : A dwarf succulent perennial without any sign of a stem. Root tuberous below

the crown. Leaves flat, rosulate, alternate. Peduncle very short, laxly 2-5-flowered, the

7-flowered condition only found in cultivated specimens. Covolla-tube yellowy-green and

shghtly tapering to the base ; limb stellately spreading or at length reflexed, deep purple

to maroon.

A. humilis (Marl.) Berger, \.c. 417.

Cotyledon nana Marl. in Trans. Roy. Soc. 8. Afr. Vol. 2, 33 (1910); non N. EH. Br. (1902).

C. humilis Marl. apud. Schonl. I.c. 151 and FI. 8. Afr. Vol. 2, 1. 16, 17. t. 3, f. D (1925),

bona.

CENTRAL REGIoN—Beaufort West distr. : On the Nieuweveld Mtns., Dec. 1909, Marloth

4689! Type (Pa).

So far only recorded from this locality, which in the early half of last century was one

of the fruitful collecting grounds of Marloth’s countryman, the equally energetic J. F.

Drége, but he appears to have missed the plant altogether.

640

23. A. nanus (NV. HL. Br.) von Poellnitz (Pls. II., IIT.)

This species was first described (as C. nana N. K. Br.) in 1902 from a plant sent “in

1899 by Prof. MacOwan from South Africa to Kew ” where it is still alive and flourishing

(1930), being for some time cultivated in the late Dr. Brown’s conservatory at Kew, and

there photographed in Sept., 1920. An authenticated photo of the plant (Plate II) was

shortly afterwards sent to the Division of Plant Industry, Pretoria, but the original descrip-

tion appears to lack some of the details which are evident even in the photo, such as the

subdistichous arrangement of the leaves which are among the smallest in the genus.

The locality from which MacOwan originally obtained his plant (the type) is not known,

but almost certainly it came from Middelburg (Cape), whence he had plants sent him in

1898-99. That this part of the central region may be involved is further indicated by the

fact that a plant collected by Dr. T. R. Sim in January, 1902 at Hanover, in the district

adjoining Middelburg, agrees in every essential detail as far as vegetative parts are concerned

with N. E. Brown’s type. The latter, however, at the time it was described, bore only a

very short (“ 4 inch long”) peduncle, with a single, apparently terminal, flower, whereas

Sim’s specimen shows the typical raceme found in the species belonging to the northern

group (cf. p. 615). That the inflorescence of the type plant is not a normal one is amply

borne out by the fact that even in the wild state, this group is frequently seen to produce

abnormal floral features, a condition even more frequently met with in cultivation, where

soil, water, or biotic factors may even retard flowering for several years! Some, or all of

these factors seem to have operated in the type plant, since it has not flowered once again

so far as observation at Kew has shown during the last 30 years, and even its leaves remain

far smaller than those of wild specimens which are beyond doubt conspecific. As Brown

also cautiously observes (loc. infra cit.), “ the one-flowered peduncle may not be a constant

character, as other species of the group have a spicate [or racemose] inflorescence’. The

following description, which will serve to amplify the original in further leaf and inflorescence

detail, was drawn largely from flowering specimens (seen in Plate III) collected by the writer,

and from Sim’s excellently dried material, parts of the original incorporated being indicated

in inverted commas :

Plant a dwarf succulent perennial, glabrous in all parts, but the leaves and inflorescence

covered with a thin waxy bloom. Stem at most 2 cm. high, stout and fleshy, very slightly

and inconspicuously 2- (or not at all) branched at the apex. Leaves closely crowded at

the apical part of the stem (or crown of the rootstock in the subacaulescent forms), alternate

(frequently apparently opposite), distichous, or subdistichous owing to overcrowding at

the apices, broadly oblong-elliptic to orbicular or reniform, broadly rounded at, and never

tapering to the base and apex (the latter very rarely, and then only casually, subacute),

up to 1-5 cm. wide and about as much long, erect or ascending-erect, very thickly fleshy,

with the thickest part (up to 3-5 mm.) in the middle, convex on either side, whence narrowly

elliptic or oblong-elliptic in cross-section, thinnest towards the apical margin, the margin

itself white and cartilagineous, greyish-green in colour, invariably spotted with numerous

purple-red to purply-brown flecks (especially in the upper half), the blotches flowing into a

continuous irregular blotch under the apex. Inflorescence single and simple, terminal

and up to 25 em. long (or more “), including the nude lower peduncular part, laxly racemose

in the upper half or third. Flowers generally spreading on their short pedicels and sub-

secundly arranged when opening, erect or ascending-erect after fertilisation and in the bud

stage. Pedicels up to 5 m.m. long in the fruiting stages, arising in the axils of much reduced

ovate acute membranous spreading bracts, terete, more or less insensibly widening into the

calyx. Calyx fleshy, with narrowly ovate-deltoid, very acute up to 2 mm. long teeth.

Corolla tube cylindric, ‘‘ slightly clavate”, straight, obtusely 5-angled, up to 1-2 cm. long,

‘reddish-brown with ... greenish-angles”, or purplish-brown, and purplish in the

throat ; lobes reflexed, ovate-acute, up to 2 mm. long, “ rosy-purple with whitish margins

641

and with the mouth of the tube at the sinuses somewhat membranous and whitish”. ‘“‘ Sta-

mens included ; anthers ochreous-yellow”’. Ovaries 5, oblong-ovoid, up to 8 mm. long,

and tapering insensibly into their subulate styles. Nectarial scales obovate to oblong-

obovate, deeply notched at the apex.

The following are the details of synonymy and citations for the species which appears

to have been missed by Berger (Engl. & Prantl., Nat. Pflanzenfam. XVIII, a. 416: 1930) :

A. nanus (NV. FL. Br.) von Poellnitz comb. nov. in Desert Plaut Life 227 (1938).

Cotyledon nana N. K. Br. in Gard. Chron. Ser. 3, vol. 30, 280 (1901) ; non Marl. (1910).

“C. B. 8.” —(sine loc. exact., sed forsan in ditione Middelburg dicta), anno 1899, Spec.

cult. leg. MacOwan! Type (Hort. Reg. Bot. Kew.).

Centra Div.: Hanover distr.: Hillside at Hanover, Jan., 1902, Sim in Herb. Galpin

5975! Lecto-type (Pa).

ORANGE FREE StaTE—Fauresmith distr.: Upper half of western slope of hill on the

Fauresmith Botanical Reserve, east of the town, April, 1927, Spec. cult. leg. Smith! (Pa).

In view of the uncertainty attendant upon cultivation and the fact that no dried speci-

men of the original type is preserved in Herb. Kew., the writer would suggest that Sim’s

specimen above cited be chosen as the type of the specific name, the specimen being per-

fectly complete in all details.

The following are the associated plants shown in the photograph reproduced as Plate ITI,

reading from left to right : Crassula obvallata L., Adromischus nanus (N. E. Br.) von Poelln.,

* Kleinia radicans (Thunb.) Haw., lying prostrate in the foreground, Crassula obvallata

Linn., with the dried remains of the previous season’s inflorescence, Huphorbia aggregata

Berg. (spiny), Haworthia tessellata Haw., with its fruiting inflorescence lying over the Crassula

just obliquely above it, ¢ Cotyledon toxicaria C. A. Sm., with a dried inflorescence on it, and

{ Eustachys paspaloides (Vahl) Lanza & Matti in the right background. The dried objects

in the foreground are leaves of the Haworthia and the Cotyledon, together with 3 old capsules

of the latter.

24. A. procurvus (NV. HL. Br.) C. A. Sm.

Known so far only from a single cultivated specimen, the type being preserved in Herb.

Kew., being chiefly characterised by its curved corolla-tube. In a genus of some 30 species,

all with straight corolla-tubes, an unusual character such as a curved corolla-tube is of special

interest, but, in as much as many of the species are subjected to teratalogical modifications

resulting from mechanical or biotic (e.g. aphid) injury, the curvature of the corolla tube in

this particular case may have been due te such injury. In the meantime, until experimental

evidence has been obtained in this direction, and for the purpose of calling attention to the

plant, its specific rank is here maintained under Adromischus Lem., from which it was

omitted by Berger (Engl. & Prantl, Nat. Pflanzenfam, Vcl. 18, a. 416: 1930).

A. procurvus (NV. FL. Br.) C. A. Sm., comb. nov.

Cotyledon procurva N. EB. Br. in Kew Bull. 1912, 276; Schouni. lc. 154.

“C. B. S.°—(sine loc. «xact.), Spec. cult. Hort. Bot. Reg. Kew! (K). Type.

It is most closely allied to the next two species, from which, however, it must, at least

for the present, be excluded by its curved corolla tube, if not on foliage characters also.

* Erroneously cited as of “‘ DC.” by Harv. l.c. (317).

+ C. decussata Phill. & C. A. Sm. in Flow. Pl. S. Afr. Vol. 8, t. 289 (1928); non Sims (1824).

{ Eustachys (Chloris) petraea Auct. ; non Chloris petraca Swartz.

642

25. A. trigynus ( Burch.) von Poellnitz.

Known only from a few specimens collected in Griqualand West by Burchell in Dee.

1811, but they are incomplete as to leaves, these having apparently been lost, so that the

species requires to be recollected in the type locality (v. infra), when a fuller description of

the vegetative as well as floral parts may be drawn up, and its specific identity determined

in terms of the preceding and the next species. The specific epithet “trigyna”’ is inap-

propriate and misleading, since in the type specimens the number of carpels vary from

3-4-5, with the last number probably the more usual as in the other species of the genus.

Thus Baker fil. & Schonland (°) also observe that in the flower dissected by them from the _

type, the normal number of carpels was found. In the matter of descriptive detail, the

following, a copy of Burchell’s original field label, represents all that is known as to leaf

characters :

“ 1898—

Cotyledon trigyna B. folia cuneato-ovata (vei suborbiculata), glabra com-

planata carnosa. Acaulis. Flores scapo elongato simplici (rarissimo bifido),

basi nudo alterni erecti. Corolla cylindrica purparascens limbo albo brevi reflexo.

Faux purpurea. Capsula tres.

Dec. 14, 1811.

At Klaarwater, in the kloof near the Burying Ground.”

Burchell’s final published description was no more than a re-arrangement of the char-

acters already noted on his field label, with the addition “ flores erecti alterni ’, and to these

the above two authors added “ Calyx lobes lanceolate subacuminate, + 2mm.long. Corolla

tubular, much longer ; the tube (sphalm. “ calyx”) + 1-3 cm. long ; lobes reflexed or sub-

reflexed, about 2 mm. long, acute.”

A. trigynus (Burch.) von Poellnitz in Fedde Rep. 44, 60 (1938).

Cotyledon trigyna Burch., Trav. Vol. 2, 226 (1824); DC., Prod. Vol. 3, 398 (1828) ;

Harv. l.c. 378; Schonl. & Bak. f. l.c. 91; Schonl. lc. 153.

GRIQUALAND WEst—Hay distr.: At Klaarwater (=Griquatown) “in the Kloof near

the Burying Ground ’’, Dec. 1814, Burchell 1898! Type (K & Herb. DC).

Those plants, chiefly from the Transvaal Highveld, which have been identified as this

species, belong to the next but one.

26. A. rupicolus C. A. Sm., sp. nov. (Fig 4.)

Cotyledon trigyna Marl., Fl. S. Afr. Vol. 2, 1. t. 2, f. D (1925) ; non Burch. (1824).

C. rhombifolia var. spathulata N. E. Br. ex Marl. (loc. cit. in icon.).

Planta humilis, succulenta, perennis, in partibus omnibus glabra. Caules ad 4:5 cm.

alto, crassi, simplices vel ramis brevissimis crasissimis, circum apices dense foliosi, et parte

inferiore frequenter prostrati, tum ad 2-5 cm. crassi et terra semi-obtecti. Folia opposita

decussata, late oblongo-elliptica ad suborbicularia, vix ad basin late amplectentem angus-

tata, apice rotundata, rarissime emarginata vel mucromata, ad 5 cm. longa et 2 cm. lata,

carnosissima, superne inferneque convexa, ad 4 mm. in medio crassa, sed ad apicem et

margines cartilagineas angustiora, viridia, semper irregulariter maculis saturartoris viridibus

maculata, sed maculis purpurascentibus vel purpureo-brunneis sub marginibus apicis dense

collectis. Inflorescentia ex planta singula 1-2, simplex, racemosa, ad 50 cm. longa (pedun-

culo nudo incluso) ; rhachis subflexuosus vel rectus, plus minusve rigidus sed prope apicem

643

subcernuus, brunneus vel purpureus, teres. Flores longe pedicellati, postanthesin sub-

secundi, sed post florentem ascendentes. Pedicelli ad 1 cm. longi, post florentem ad 4 cm.

elongati, debiles, teretes, pedunculo concolorosi, bracteis ad 1 mm. longis ovatis carnosis.

Calyz viridis, carnosus; dentes ovato-lanceolati, acutissimi, ad 2 mm. longi, carnosi. Tubus

corollae cylindricus, rectus, ad 1-5 cm. longus, obtuse 5-angulatus, rubro-purpureus,

fauce purpureus vel saturate ruber ; lobi ovati, acuti, ad 2 mm. longi, reflexi, pallidissimi

purpurei vel in senioribus albescentes, inferne saturate purpurei, marginibus undulatis vel

crispulatis. Stamina 2-seriata, plerumque 4 exserta, alia inclusa; filamenta saturate

purpurea, plerumque 4 paululum altiora in tubus corollae quam aliis inserta; antherae

ovoideae, albescentes. Ovaria 4-5, oblongo-ovoidea, ad 1 cm. longa, ad basin paululum

cuneata, in stylo subulato ad apicem angustata ; styli sub antheras staminum exsertorum

breviter excurvati. Squamae nectarii obovatae, manifester emarginatae.

CenTRAL RecGIon—De Aar distr.: On the slope of a hill to N.W. by W. of De Aar,

in rocky crevices almost right under stones, Nov. 1926, Liebenberg 263! (Pa).

ORANGE FREE StaTE—Fauresmith distr.: On the Fauresmith Veld Reserve, near the

crest of the hill among rocks and in crevices of rocks, 4,650 feet, Dec. 1937, Smith 5203 !

Syn-type (Pa), & in eod. loc., Jan., 1928, Smith 5603! Type (Pa).

WESTERN TRANSVAAL—Marico distr.: Among rocks at Zeerust, 4,000 feet, Dec. 1926

van der Merwe 37! (Pa).

This species, figured in the accompanying illustration (Fig. 4), is locally known as

“plakkie” or “bontplakkie’, and invariably, unlike A. nanus (N. HE. Br.) von Poellnitz

grows socially in crevices and in the shade of rocks, with the long graceful inflorescences

projecting well above their immediate rocky environment and so readily accessible to insect

visitors. The plant is common on the hillsides, usually in the upper half of the western

slopes near the crest, round the Fauresmith in the district, and, judging from the above,

appears to be one of the few with a relatively wide distribution.

On the Fauresmith Reserve, the plant is very generally found among rocks under

taller shrubs and bushes such as Rhus Burchellii Sond., * Ehretia rigida (Linn. f.) Druce,

Rhigozum obovatum Burch., along with other such dwarf social succulents as Kleinia radicans

(Thunb.) Haw. and Haworthia tessellata Haw. mixed with it, or in shady places not under

other plants and associated with other succulents such as Ewphorbia aggregata Berger,

Cotyledon toxicaria C. A. Sm. (see p. 641), Crassula obvallata Linn. and Mesembryanthemum

saxicolum t (L. Bolus) N. E. Br., Stapelia flavirostris N. E. Br. Marloth’s fine coloured plate

cited above agrees in all essential detail with a partly coloured plate made from the type

specimen of which Fig. 4 is the rendering in monochrome.

27. A. umbraticolus, C. A. Sm. (Pls. IV, V.)

This species was first described in connection with toxicological experiments carried

out at Onderstepoort (near Pretoria) by Dr. D. G. Steyn (see p. 615), and as the Journal

containing the publication may not be readily accessible to other workers in the field of

systematic botany, the description is given here in full :

Plant a succulent sometimes acaulescent perennial, glabrous in all parts, but covered

on the leaves and inflorescence parts with a thin white waxy bloom. Stem (where present)

stout and well-developed, up to 12 cm. high and 2 cm. thick, erect, simple or usually dicho-

tomously branched in the upper half, terete, closely leafy toward and round the apices,

with knobby excrescences marking the old leaf scars on the lower nude part, and covered

* E. hottentotta, Burch. of Thiselt.-Dyer, Fl. Cap. IV, 2-5 (1904).

‘ j{ Ruschia saxicola, L. Bolus. The above determination was made for the writer by the late Dr-

Ab E. Brown at Kew from Dr. L. Bolus’ type number, also collected by the writer on the Fauresmith

Reserve.

644

with a thin greyish- or ashy- to yellowish-green skin. Branches resembling the stems, but

thinner, up to 3 cm. long, simple or again shortly branched. Leaves alternate, occasionally

pseudopposite at the apices of the branches, decussately arranged or somewhat scattered,

oblong to obovate-cuneate, gradually narrowed from near, and toward, the base, usually

rounded at the apex (or casually abruptly narrowed to a subacute, often in the younger

stages, slightly crisped apex), up to 5 cm. long and 2 cm. wide, erect or ascending-erect,

frequently (the lower) incurving-erect, very fleshy and up to 4 mm. thick in the middle

near the base, gradually thinner towards the apical margin, green, often purple-red flushed

at the apex, but never spotted, convex on the outer, and convex or flattened on the inner

face in the upper half, convex on both surfaces in the lower half, and hence elliptic to oblong-

elliptic in cross-section. Inflorescence apparently terminal or axillary, simple or with

2-3 branches, laxly racemosely-flowered in the upper half, nude, or with a few

scattered and much reduced sterile scale-like bracts in the lower half; rhachis rigidly erect

or subcernuous near the apex, up to 35 cm. long, terete, brownish-purple. Flowers sub-

secund when open, and spreading, erect after fertilisation and in the bud stage. Pedicels

up to 6 mm. long prior to, and | cm. long during the fruiting stage, most usually with 1-3

much-reduced lanceolate-ovate bracteoles, the lower occasionally with a sessile non-maturing

flower from the upper bracteole, terete, and insensibly widening into the calyx, arising from

the axils of much reduced ovate to ovate-lanceolate acute up to 1-5 mm. long bracts. Calyx

purply-brown, fleshy ; teeth ovate-deltoid, acuminate, up to 2-5 mm. long, fleshy and con-

vex on the outer face and adpressed to the corolla. Corolla-tube cylindric, straight, obtuse ly

5-angled and marked with 5 evident sulci between the angles, up to 1 cm. long, purple to

mauve-purple on the outer face deep mauve to maroon-coloured in the throat, scarlet to

deep coral-red in the bud; lobes ovate-deltoid, acuminate, up to 2 mm. long, purple to

mauve, thin and delicate; the limb at first spreading but at length reflexed. Filaments

filiform, imserted as in the former species, purplish-mauve at the apex. Anthers ovoid,

creamy-white or purple-flushed before dehiscence. Ovaries 4-5, oblong-ovoid, up to 5 mm.

long, tapering into the subulate green styles. Nectarial scales oblong, up to 1-5 mm, long,

very slightly notched at the apex, pale green.

A. umbraticolus, C. A. Sm., in Onderstepoort Journ. Vet. Sc. & An. Ind. Vol. I, 174

(1933).

Cotyledon trigyna Burtt Davy, Fl. Transv. & Swaz. Vol. I, 142, 143 (1925), pro majore

parte, sed. excl. syn. et non-Tvl. exsice. ; non Burch. (1824).

TRANSVAAL HigHveLp—Pretoria distr. : On the Magaliesberg at Silikaatsnek, m rocky

crevices in shady places along the northern slopes, 5,000 ft., Nov. 1926, Smith 3432! Type

(Pa) ; at Wonderboom, along rocky ledges and in crevices of precipitous sides of cliffs in the

northern entrance to the Poort, 4,650 ft., Dec. 1925, Smith 1766! (Pa); On the farm “ Wit-

fontein ’”’, along upper half of northern slopes of the Magaliesberg, about 2 miles west of

Wonderboom Poort, 4,600 ft., July 1933, Smith 6272! (Pa); at Pretoria, along northern

slopes (upper half) of Meintjeskop range, below the crest near the old Fort in rocky crevices,

4,800 feet, Sept. 1925, Smith 6934! (Pa), and in rocky fissures and crevices in rich humus

on the same range below the Reservoir, 4,800 ft., Nov. 1926, Smith 3456! (Pa), & cod. loc.,

June 1931, Smith & Ward 3! (Pa). Witwatersrand distr.: Rocky crevices on a koppie

at Braamfontein, near Johannesburg, 6,000 ft., Nov.—Dec. 1898, Gilfillan 60! (Pa).

A very common plant on the ranges round Pretoria and along the northern slopes of the

Magaliesberg, invariably found growing socially (like the last species) in rocky crevices and

fissures, with the vegetative parts in shade of other plants (Plate IV) and the rootsystem

under rocks or often firmly wedged into the crevices, the inflorescence being exposed to the

open where the flowers are more readily accessible to insect visitors. Fig. 5 shows a typical

specimen :

645

TO ce

ees

Seem

‘

Fig. 4.—A. rupicolus C. A. Sm. Sketched

from the living plant (the type) collected by

the writer (Smith, 5603). See text.

Cross-Sec™

Leaf.

flower—

NYe vecurved

y reflexéd

3 ee lobes,

aie one

Wa. 5.—A. wmbraticolus, C. A. Sm. Sketched from the

living plant, the type. See text.

647

In the Meintjeskop localities the plants were found forming practically pure stands

under dwarf arborescent specimens of Strychnos pungens Sol. (Plate IV), Vangueria infausta

Burch., Combretum Gueinzii Sond., and very commonly also under tangled masses of Lan-

dolphia capensis Oliy., or in other cases associated with one or very generally more of the

following: Crassula argyrophylla Diels (commonly so), Euphorbia Schinzii Pax, Salacia

Rehmanni Schinz, Kalanchoe paniculata Hary., Aloe Davyana Schonl., A. transvaalensis

O. Ktze., Becium angustifolium Benth., Pachystigma Zeyheri Sond., Lannea edulis (Sond.)

Engl., Leonotis microphylla Skan and Cotyledon leucophylla, C.A. Sm. Huernia Loesneriana

Schltr.

The accompanying photo (Plate IV). taken by the writer in Nov. 1926, shows Smath

3456 growing under a specimen of Strychnos pungens Sol., with trailing branches of Landolphia

capensis Oliv. (to the left and top right of the phceto), the numerous erect peduncles with

their long racemes forming an interlaced mass above the plants.

The next photo (Plate V) shows the plant in more open situations due to removal during

the previous year or two of the sheltering Landolphia ita Oliv. (seen to left and bottom

right of photo) by nocturnal hunters for the notorious but legendary “ Kruger millions ”

(note the hole in the centre).

This species is of interest in that the pedicels bear 1—3 bracteoles along the lower pedicels,

from one of which (the upper) a sessile and generally abortive flower may arise, thus indicating

an approach to the 3-florous condition seen in other species of the genus already noted (see

p. 617), though in the former the ee condition prevails, and it is a feature of further

interest to note that the bracteoles with barren flowers were seen only in cultivated specimens

thus far.

The leaves of this species also behave like those of A. Marianae (Marl.), 1.e. when they

drop or are broken off from the parent plant desiccation does not follow immediately, but

adventitious roots are developed at the base, followed in time by tufts of leaves, the genesis

of one or more new plants. Both the two leaves seen in the foreground in Plate V had

started to root in this manner, the one on the right having also produced new leaves.

28. A. saxicolus, (. A. Sim.,

Planta nana, succulenta, perennis, in partibus omnibus glabra, locis apricosis crescens.

Radix plerumque multus incrassatus et succulentus, ad 2 cm. diam., basi nodulis rotundatis.

Caules e corona radicis 1-3 (vel plus), breves (ad 2-5 em. alti), erassi (ad 1 em. diam.), ad

apices dense foliosi, vel caules O, foliis tum circum coronam rotundatam aggregatis. Folia

dispersa, vel summa pseudopposita, dense congesta et subrosulata, limeari-oblonga ad

oblongo-elliptica, plus minusve ad hasin et apicem auctum equaliter angustata, ad 3-5 cm.

longa et 1 cm. lata, carnosissima et ad 5 mm. in medio crassa, superne inferneque convexa,

viridia, immaculata, pallide rosea in superiore parte. Inflorescentia simplex : rhachis

rigide erectus, ad 25 cm. longus, simplex. Flores pedicellati, in cymulis 1- floris dispositi,

post anthesin patentes et subsecundi, postquam impregnationem ovariorum erecti. Pedi-

celli ad 5 mm. longi, vel 7 mm. in fructu, teretes, saepe bracteola membranacea subulata

onusti. Calyx cinereo-viridis, carnosus ; dentes lanceolato-deltoidei, acuminati, ad 1-5 mm.

longi. Tubus corollae cylindricus, rectus, obtuse 5-angulatus, ad 8 mm. longus, purpur-

ascens, in fauce purpureus vel saturate ruber ; lobi ovato-deltoidei, acuminati, ad 1-5 mm.

longi, purpurei, sed pallidioribus marginibus, deinde reflexi. Filamenta filiformia, ut more

sectionis in tubo corollae inserta, ad apices purpurea, infra viridescenti-lutea. Antherae

ovoideae, post anthesin lutescentes vel purpureae. Ovaria 5, oblongo-ovoidea, ad 5 mm.

longa, in stylo viride subulato angustate. Squamae nectarii oblongo-cuneatae, ad 1 mm.

longae, apice breviter emarginatae, pallido-virides.

648

TRANSVAAL HIGHVELD At the southern entrance to Baviaanspoort,

on a Magaliesberg, growing oie on eee ledges and in crevices filled with black hu-

mus, 4,500 ft., Now. 1926, Smith 3424! Syntype (Pa); at Premier Mine, in open places on

a koppie to east of the mine, growing in rocky depressions and crevices, June 1931, Smuath

& Ward 5! Type (Pa).

The plant is closely related to the previous species, but from this it is readily known

by its rootstock, the acute leaves (only very casually obtuse), as well as by their shape, and

its consistently dwarf habit. Found only in one locality at the Premier Mine associated

with Crassula argyrophylla Diels, Euphorbia truncata N. B. Br., FE. Schinzii Pax, Lopho-

lacna corivifolia (Sond.) Phill. & C. A. Sm., Selaginella rupestris Spreng., Parinarium capense

Harv., and the very dwarf forms of Burkea africana Hook.

SPECIES EXCLUSA.

The Cotyledon described by Haworth (Misc. Nat. 180: 1903) as C. caespitosa Haw.

with * Habitat ad Cap. Bon. Sp.” is not a Cape plant at all, but, as later corrected by Haworth

himself (Syn. Suppl. Succ. Ed. Germ. 117: 1819), is a Californian plant, said to be con-

specific with C. linguaeformis (Ait. Hort. Kew. Ed. 2. Vol. 3, 109: 1812), which was obtained

from California (see Saund., Tee Bot. I. t. 69: 1869). It belongs to a genus now excluded

from Cotyledon Linn. (emend.).

BIBLIOGRAPHY.

) Engler and Prantl, Nat. Pflanzenfam, Vol. 18. (1930.)

) Harvey and Sonder, Flora Capensis, Vol. 2 (1861-62).

3) Records of the Albany Museum, Vol. IIT (1915).

) Journal of Botany, Vol. 40 (1902).

Yea

649

Flate I.—A leucothrix C.A. Sm. Type plant.

botybdon nama, WV & fn,

j re i 40%

ee ect be

Plate I1.—A. nanus (N. E. Br.) ven Poellnitz Type plant photographed

by Miss. Brown, Sept., 1920 (see text).

Plate I11.—A. nanus (N.E. Br.) von Poell., photographed in its native habitat with a group

of associated succulents by the writer in April, 1927. (See text.)

Plate [V.—-A. wmbraticolus C.A. Sm., in its natural habitat. (See text.)

651

NEnagromee, te ao

Plate V.—A. wmbraticolus C.A. Sm., with associated plants. Photo taken by the writer (Nov.,

1926), slightly to the right of the one shown in plate IV. (See text.)

INDEX.

Ad romIschusAISLONI Stee eee eee ls neeinee Ga RiGee ee ee sieinie seis ieee 615, 618, 638, 639

/s\ sil BY) UST 1g. S auch ce oroe G0 ONO TIE CCR CRE ICIP ERE eee ena eS 617, 618, 628

Ne CHOP MACTUSs o =o coousoanbodc0 deonnsan00b00c0ds oo Sanoade 614, 615, 616, 617, 628, 629

JAge@E MUONS sone co aro aaucbodoo mers ttoleredtlc o ceicicic ica cenn erties 614, 616, 618, 636, 637

NACHO TSM ngs ogane ace onsneemaade swam mae rene arenes 614, 615, 616, 617, 618, 632, 635, 639

AMS Cristatuseee ete ate See Olena oe Aatioteoistaba sans eidmin Seam vn 614, 616, 618, 635

INMATES TIVILIS eee ee IC TE ere ees een iene mnemea ear seus enol 614, 617, 618, 632, 633, 634, 637

INE SERGE UIE. OLE See dob ao OOP OOE OS6 & SRD TOD BE ae OO ne en AEE ene 615, 630

INS TRIO YGINS 5.0.5 cooeo mood oo Oc 614, 615, 617, 618

AV hemisphaen Cust crear cin ee ones oleae sa saelle a ora ae 614, 615, 617, 618, 625, 627, 628

NS dinrtaa Tae boeaclce.a Saad 6 ba bo OE CUO OUT OE OOO EIA eae naan ate 614, 618, 639

ARs ASIMIMULLOLUS erreur ecae eareteer LRA POC HeRe oreo ore ssretciey ketrevckeletaie fey sverel sens shisbonsvstioseene 629

y NOMA KCCSTIE Te ca ea ey Bape ys ato tr st les ton NST BOS Gar OI HOR ACEC NCHCL aR CRC aE CREE RRC a 620

pe \ eet tal leyitravoy (alee ren eietetare GHRERO Oo. bd ol Gaueid Cais oe en TE Pee 614, 615, 617, 618, 631

PAN MLETICO GTR, caspeatiarars tesla eve lel iene meets fede been eoreneres sb css fnscaste shoe siti /e,eunave\ sabe 614, 618, 637, 649

A eT ACUL AGUS) js sreucy rence SS Ree easiness onary Ped Siva eee enuaile eto eos 614, 615, 616, 617, 618, 620, 622

YA Bris cavaicab aot Wn ots aoe A coh Sete rere ena seo stlone ow itelicoes SLs Sh nee eR TS 614, 615, 617, 618, 631

PNP ENTE ENE ae ae ls eee tin le a oe ok eer eR ee 614, 615, 617, 618, 631, 639

ACs ar Ophir ctr tine tener Cran uae re pI pay Sci Ein aeae sctjalevac cits 630

Ae WX AMMAR 555 oo coo DO oo ODO DOO DODO DCO COOO aS OO sn DU DOO EObOSO0 614, 620

JING SHORES doraneoG ne ob od OE .o0\s on DIO aH ROIS RT ORCI Ce 613, 622

HBOS oF Tinh ler ce ey steering SE CL Het REECE al Otc Cee eae 614, 617, 618, 640, 641, 649, 650

Ave pac hiyLOpIUSi peter ersvernaces velsio sre sells) ou cio nearer oIreR TAC See creivyale siale vlamisie + ne ee 614, 618, 633

AN, FOROCUIATIERS 06 codoMiba ASeolbe Seow OaS.0 COU cols ad boon Oo MDeosDomaG oeDO eal 617, 641

PARET KOMP ILOMUSE lorctaceeere ai ernvetclstal ete colo Pe UCT OrS ROE Tose ee re tanG esi oe 614, 615, 624, 625

NE Dito) OVE Sis Bord opamie aan Bie ol cigars b Eat buc 6 cro ucloio bio bbe Sicko RIC cm Oo 613, 623

ABT OLUNGIFO]ING sre cys, Gok ene sees ete asides eee ieee Tete ere ein arers a8 614, 617, 618, 627, 628

AB TUPICOLUS arayeietei sete tveviersicysusiesoicverone) el ova ereteloeretorch ere tekstereiet eres alee 614, 615, 617, 618, 642, 645

pA SEU COMUISE iy aero cos ieee HSESONE TOE Ge CI OM ea eto e I ee hs 647

Fale WSLG NEN) CAEN EUS a iaty eedrOras exehend iG Cy SORE CIGNA Ueto OUNCE nO OS aDiG Rein 614

INS STO avay ol ey aa btm ser stete eee enter tipo claire nearer cca RD ONCE eee eC OE ae 637

Are SPH ASLO PM US aye cys jee) arse clers Ges erties oe cis signe tiers resereels siceo eysier) steie ye ausnenaane vo 620

Ne KOAEMOO MAUI G 6 ooeadcdacacdooobobooODenDoDéueoDga 614, 615, 616, 617, 618, 622, 623, 624

YN Ge ALPLAOLO) (6) Hy Gate OH tile ICE ie AN ois UP US RCH GOL CCR ROUT NG ETE NOLO Roo nr eRERER 614, 632

PNG AHBU NOTES Gee onets Ga ch Stee RTOs OPED Gs BO OLGNOTO OT CURATORS OTRO Te ECA etc CIN rane lor eae rn 615, 618, 638

TNE SONA OU Sal eareig Sid ero at Rnaeel pao evel Us biey cici SIRE RIO cS Oe Boy anole Stn een ene ae 615, 617, 642

ASS UMD LAbICOLUS\sictonsais evetays iataysveveieses sucvevetcaeisyo:euarey oisvel 614, 615, 617, 618, 643, 644, 646, 650, 651

AV ViaDy demmbley cl entteyrsicreipess clits teen eeysnereecies repeal alsa oor ey Banvenevarecce. se ceiane 637

LAAVAGH ANS tense ona Woon Bune aoe oOOR i Up COU CEL CO OBO re tere aetna 614 615, 635

CotyledonmAlstomilya a= ardor certs aysters len(noevesuetere cis) clayceeiecs.s sce Gleers tes. + acayeyos 639

(Ck OIIGISA Te orn oie Slemo biol o> Cabias amo te co BiOl ao > ECE ee OE marc ence eee 622

(Oss TBXOVIESI 3 4.5m areicto'cio, cick 6: O.e owiehe et Raa OB chEo 0 5/0 Shee escent RI aCe 628

Cia calioid ester rare re NT CP ay CCL EE NOT ae oa eiea ev nMeu ee ron tneaeue ouererers 638

CE CEEOL opera cla Gadks bal a6 ycbgnonbodo 66 oT One aoe ree oe 648

Cicaryopnyillaceasmrr ery retrieve tine foo cottes essa Meare e eae sd eusens ayes 628, 629

Ce iclawifoliaacee scree ee oon 0 ORO A Geo od ee Sots, Syotancovegeretases 635, 637

COCO penienerecerey pes tater parser ner pote ene sie hice cieleiing slerene aiteelee Duseies 632

(CS) CHM Bigs cod ua 656-05 pel 0 TOO TOTS oso BIS Oe RETO ECR eee open earns 613, 635, 637

(Co GHAI OMNIS, cc 0.0.6 bo oS 0 Fee OO gO OU 6.5 Uo dR ON OR eC Cone CCT ene aicrcaeeae 626

CMidecussatabvamr ae rire eer yore eerie ciao: =: scare e: oral Sie avail auanas eeu eretrane 641

CLAIM das codes VomlocUPaUd be O COGS COO EOD Tete MTR onc D ree nar tcee 638

OG}, IMG 6.5 a cme.do cco able One COO eC On TERE On EO a enon Cree 630

@iwpbemisphaenicarmerrtwrcace cise oem vacsc terres eave Sit 613, 622, 623, 626, 627, 628, 629, 630

(OL Inter isigon 6 060 4 6 okos DOCU BEDS BOR SORTA Tb ORO RTE E Daten ice eee 639

CA] ASO OLA re eee RTS Gira eR ee oes Ho ie ows dee SLelare ae 616, 628, 629

Camleucophivl ame werecer ett neysk can icy sen tactserte vas deiisie ciate iene yhpsve is, Stiles ards 647

(Ch NAT AC ONT s 6 odovooeaeD Ob Ome Oe aols 6 6 SUERTE Occ ere Nero aie ere eee 648

Oh STO CULENHTE Son beets roan GLa ea ere Oy eee eR a ae 620, 622, 627, 629, 632

(Ck linenanil iu 3.3 opooon asin on be alae b POSS cOO 5G CoE ce CIO DICT OLR One 613, 630, 631

(Cs WTB ENED. o\5'5 radioed SOO NOB a 5 OO OI Urs :015 Be EERE TEER Ory eae 639

(Cy ibid oiittihs..5 b Sebae oa Ocdoc CODOIO GD G6 Ue OO OE RODS Ee ED Ieee eet 630

654

(ORR It) 11: enon EA oS Aicic cio tac od BUCO 6 Gb meee cnauomesaseooe ts 622, 627, 639(

OF 9) OLN Nan: ince IA ERIM On en One eee otc madsola eae BOL Mans ec Ris clogd aa aiclacs G 0,0 5 641

C: rhombifolia ssc, asleep ces hcvors cose eee Ce as TR Ce ee 624, 625, 627

(ONCE OTE, WEIR SOUND. 5 e540 c9c0000 0000800000 an000Gadn0000G0000008 642

CS 2otundifoliancncc 5. see eee ens Oe ee 626, 628, 629

Cia Schonlandiky 2 s.-= ores eC se eae OO eS eee oe ee 637

Gi Once be Mereisimet acc. i clc A Oca lero arate o RE ne Ring Mian Se Oue oe ae ENA SSS 641, 6432

isp tei OP 85, eRe aid Be cocoa eos Stee ah LE. ra Se 623, 624, 627

Ces trl oyMaris cakes. folore Seen oo eeu sale wie ~ Soe Seven Otys SrSer enemies a aieereae 642, 644

GRR OMIA OS Nokes Meu eter MBean Ose bo eaee roma a IGG: BIRO Tine Cob axieolna oO Mose 632. 638

Gee AVE) UGK) clit! Sees Seceeeerees Siete eee etecroeNeen neurnAR Creer cuidln oc cata amin tec 0 oes oes 638

CSV ASN CBs toy ESOS 7 cin ea hy ae Renee oT eee eee manera dais cen Ge roids 613, 635, 636

OTHER SPECIES CITED IN THE TEXT.

LALO Cp AD A Way ATE Bley, cae stint sievoestoessas eve ors cadbsea Osi” OTe SEL Ni Rese RUB Rea ue ees eer eam Ra 647

ATS CLANS VALE SIS oa wteress eg vnve. cus, cesnrge enacts Ieee OSCE Ae RR eee 647

Beeman Sustilo WMM. jeer ic se latest eae acaesey ane toot rclaveh eeeieaeCeh satin: Seer eee aera Reel ree ase 647

IBiin ech ae AUER) eit ee iene eine atin Gisitnt oo Coole eam Wa ee wooo mete oue eS 648

Chloris petrale as. ices che nei as ww cinema area eRe re Ran cer ere Re eI ete 641

Comibre tina: 1G we trzais go: s-xeyer. a= cere en oes econ ek A ee or eet Re 647

Crassulasarcynopbivilll an cj... us cwserseies ate onea turers rerer emer aro rie tartar sie meee ene Teeny sete 647, 648

Ceamesemibnianthemoidesy. svnaaaqe oc nue aac aeen Ome oer A eae eee 638

Crm Ol vialllait altreca caer: tere. oh asses nuance te cit arty HERO Ce AO eee ee 641, 643

Cente aieb ysaumib atte cacy fs is, lexes syetrelen/sfaye an esaemoun oor ees ee URLs as a ee Ee ee ees 638

hretiahottentotta sacri cece eaten ie ra ern ao ere enero one 643

HY p SHON eg vege p tins cnmicse nye geo sie maar cee eyes mia oc cect ine ete emenee 64:

HuUphouibiamagereratar:..v.c cn acy eee ae eke Mec ae eremienciers eee meee 641, 643

1 De SSKO AURAL OSPR S ish te mR Re ene ont dee Ne MES poe ak ee GU once oie Book Sa vate ce 647, 648

Arse BRUNA CA GAYS searogeite. ic -aeehe tes esa erev en on otacere eis eye ee oe aye oearee me ee eee aa 648

Hustachys spaspeloides'..%,. aces cscs crs oe aisles sats Porat em Core: 641

IBY AAC LEC ab snce vaweteyel sccscin aati ous ace Syiei ate ts cae eave he eo nae LOR oe EERO Ne) eT Nero CER ee 641

Gasteriat carina ta’sc., af supa anencs acesare omer meee ore a eee dare Anon ar ene tee 623

Elanmorbhia: “ibessellaital sce meet occ esp sey ao ee OTe aa eee He ae NPT I 641, 643

Euerni a TOCSCMELLAM As ahoalern pe tiny eee coat SER cick Sere TERT Cora Ieee 647

Kalanchoespaniculatare nee ener eee ree SERN ete teres Bean a a aes 647

WWlenntay como clades iccni is g.t.seccm aeons cene ease ees eeatten sev ieee e susetae ie et ces tem ee ae 631

Hees eAGUT CENTOS eat spac oR cy ore oa ee Ee rn ee ee ee 631, 641, 643

and Ola Ca pensisicec cre a coe se serine ae emcees olunteer mlaveae aT eR RRR aR eRe Se 647

Tamm eae dni sisi ercrececss os-Aacyiceie cccceecte ee wees ee cosa oie elton eee tee te meena 647

Meonotismmlcrop lye iss 5 we aicieresacen <iesteeemarererererae ols cise eee Meera RM RES 647

Lopholaenarqonutoliais: mo racine secgy ans oe Oe bio cee ae oo IIo 648

NESOMO VA MENEMNUTMN SASHCOMNTMN co ocococoo cess oo uscansecooaonoccocaouseeaod 643

RachystigmanZeyReris wis snie docs sins cite cum emia cesta Gee eee eee aac eres 647

Prapima rium (Ca pense zrve succor aieiess cae eroieioceae teensy ereneny stares eieteternhe fe sotra ener Roe 648

RhicozumMobovatumia arc ws eio ers ere ieee te ee eee 643

aus! Burchelit jas) s:;. cece eee Se coe ee Seon oe eee ares 643

RUS CHa SARLC OLA sacrosanct rea ea ook oie ch cA eee ee TS ne Cee eee eo ene 643

Salacia, Wehuma mimi Seco gs ce, teary hae cieian ae eae eet Peo oe RR ACIS 647

Selaginellay mupestrishe wc mis ier ectekn es ern alos IE one nO eR ee eee 648

Stapelia:. flavanostrise yc saacecetree iieini-cnnc OG ee OR ea Ce ere TO 643

SUL: CMOS: UNG CMS). 5 cece osmeeeteicdeatnecseetate siete ce sae ce eeeaece ene teeten eet serene ner oneices 647

Mam euenia: Tntaustar. cuca ccc epee neers ei cacra Sree ere eee ore oreo 647