BIOLOGY

MR 1 9 'C79

The person charging this material is re-

sponsible for its return to the library from

which it was withdrawn on or before the

Latest Date stamped below.

Theft, mutilation, and underlining of books are reasons

for disciplinary action and may result in dismissal from

the University.

To renew call Telephone Center, 333-8400

UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN

L161— O-1096

FIELDIANA: BOTANY

A Continuation of the

BOTANICAL SERIES

of

FIELD MUSEUM OF NATURAL HISTORY

VOLUME 31

The Library of the

JAN 1 8 1979

FIELD MUSEUM OF NATURAL HISTORY

CHICAGO, U. S. A.

. b

TABLE OP CONTENTS

PAGE

1. Two New Species of Palms from Nicaragua. By S. F. Glassman ... 1

2. Tropical American Plants, VI. By Louis 0. Williams 11

3. Agriculture, Tehuacan Valley. By C. Earle Smith, Jr 49

4. Flora, Tehuacan Valley. By C. Earle Smith, Jr 101

5. Preliminary Studies in the Palm Genus Syagrus Mart, and Its Allies.

By S. F. Glassman 145

6. Tropical American Plants, VII. By Louis 0. Williams 165

7. Supplement to Orchids of Guatemala. By Donovan S. Correll .... 175

8. Preliminary Notes on Scrophulariaceae of Peru. By Gabriel Edwin . . 223

9. New Species in the Palm Genus Syagrus Mart. By S. F. Glassman . . 233

10. Tropical American Plants, VIII. By Louis 0. Williams 247

11. Notes on the Flora of Costa Rica, I. By William C. Burger 273

12. A New Eurystyles from Nicaragua. By Alfonso H. Heller 279

13. New Species in the Palm Genus Syagrus Mart. By S. F. Glassman . . 285

14. A Revision of the Family Geastraceae. By Patricio Ponce de Leon . . 303

15. Studies in American Plants. By Dorothy N. Gibson 353

16. Two New Nicaraguan Juglandaceae. By Antonio Molino R 357

17. Studies in the Palm Genus Syagrus Mart. By S. F. Glassman .... 363

18. Tropical American Plants, IX. By Louis O. Williams 401

TROPICAL AMERICAN PLANTS, VIII

LOUIS 0. WILLIAMS

Chief Curator of Botany, Field Museum of Natural History

FIELDIANA: BOTANY

VOLUME 31, NUMBER 10

Published by

FIELD MUSEUM OF NATURAL HISTORY

APRIL 24, 1967

PUBLICATION 1025

Printed with the Assistance of

The Edward E. Ayer Lecture Foundation Fund

Library of Congress Catalog Card Number: 61-17730

PRINTED IN THE UNITED STATES OF AMERICA

BY FIELD MUSEUM PRESS

Tropical American Plants, VIII

The notes which make up this paper are the results of studies of

the families involved in the next number of the Flora of Guatemala,

in the determination of material from Mexico, Central and South

America sent in by others and in the study of our own collections

made in Central America.

Large collections of plants made by D. E. Breedlove in the state

of Chiapas, Mexico, and in adjacent Guatemala are of the finest that

it has been my pleasure to study; Roy W. Went's collections from

Costa Rica have been rewarding as have those of Dr. K. Lems from

a very interesting part of the same country.

Field work of our own group in Central America — every country

was worked in by at least one of our group and often by as many as

five of us in 1965 and again in 1966 — and travel to study in other her-

baria has been supported by grants from the National Science Foun-

dation.

I saw all the material of the Sapotaceae and Symplocaceae con-

tained in U. S. National Herbarium and the herbaria of Arnold Arbo-

retum and Gray Herbarium during the course of revision of these

families for Flora of Guatemala. I appreciate the privilege of having

studied those collections.

APOCYNACEAE

Aspidosperma capitatum L. Wms., sp. nov.

Arbores usque ad 28 m. altae laticem sanguineum exudens. Ramulis crassiuscu-

lis, ferrugineo-pilosis; folia alterna, coriacea, anguste oblongo-elliptica vel oblongo-

ovata, obtusa, subtus fusco- vel ferrugineo-pilosa; inflorescentae subumbellatae

multiflorae in capitulae congestis aggregatae; calyx ca. 3 mm. longus, lobi anguste

triangulares vel lanceo-triangulares; corolla viridio-flavescens, lobi lineari-lanceo-

lati, acuminati, tubus angulatus, chartaceus; fructus ignotus.

Large rain forest trees to 28 m. tall and 0.45 m. in diameter, exuding reddish

latex. Branches terete, densely black pubescent and becoming glabrous, densely

ferruginous pubescent on the new growths; leaves alternate, coriaceous, narrowly

249

250 FIELDIANA: BOTANY, VOLUME 31

oblong-elliptic to oblong-ovate, obtuse, acutish at the base, 12-22 cm. long and

6-11 cm. broad (leaves if present in inflorescence smaller), the lateral veins at

nearly right angles to the mid-vein, about 30 pairs and with a faint nerve between

each prominent pair, blade glabrous or glabrescent above except densely ferrugi-

nous-pilose on the mid-nerve, short pilose pubescent below, densely so on the

nerves, the margin obscurely revolute and prominently pubescent; petiole terete

or angled, densely ferruginous-pilose, about 2 cm. long; inflorescences axillary at

the upper nodes, densely dark ferruginous-pilose, the primary peduncle about 5 cm.

long, the secondary peduncles subumbellately disposed and about as long as the

primaries, bracts or much reduced leaves often present, flower-clusters capitate

(probably exceedingly compact cymes) each with some 25-50 flowers and the in-

florescence with several hundred flowers; calyx about 3 mm. long, the lobes about

2 mm. long, imbricate, narrowly triangular or lance-triangular, acute, densely

pilose-puberulent outside, glabrous within; corolla greenish-yellow, about 8-10 mm.

long, the tube and lobes equal, glabrous, lobes linear-lanceolate, acuminate, 4-

5 mm. long, tightly spiralled after anthesis, corolla tube somewhat angled and

chartaceous, with the anthers inserted at or above the middle, 4-5 mm. long;

fruits not seen.

PERU: "Chontaquiro Colorado", bosque humedo, Estacion Ex-

perimental Agropecuario, Rupa-rupa, Dept. Huanuco, alt. 680 m.,

die. 11 1962, Abelardo Gutierrez R. 95 [tree number TM-68] (F, type;

WIS; NY; G; others).

In Dr. Woodson's revision of Aspidosperma (Ann. Mo. Bot. Gard.

38: 119-204. 1951) this species fits into the Series Nobiles and seems

to be most closely allied to A. nobile Muell. Arg. from south-central

Brazil, and less so to the Venezuelan A. steyermarkii Woodson. The

congested capitate cymes of the terminal division of the inflores-

cences (from whence the specific name), the ferruginous pilose mid-

nerve on the upper surface of the leaves, details of the flowers as well

as the areas of distribution separate this species.

Mr. Gutierrez says that the tree is used for lumber, to obtain

coloring matter and for tannin, and that the wood is hard and of

commercial quality.

I assume that the corolla lobes are spiralled to the left before

anthesis, as they are after anthesis. Flowers on the specimens are

past anthesis.

BIGNONIACEAE

Tynnanthus macranthus L. Wms., sp. nov.

Arbuscula vel arbores usque ad 3 m. vel ultra. Folia opposita, bifoliolata;

foliola elliptico-ovata, acuta vel breviter acuminata, subtus lepidota; inflorescen-

tiae breves, pauciflorae; calyx anguste campanulatum, subtruncatum, lobis per-

parvis; corolla bilabiata et prope basim divisa, leviter arcuata; stamina 4, didyma.

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 251

Shrubs or small trees to 3 m. or more tall, twigs somewhat flattened at the

nodes, scurfy-pubescent. Leaves opposite, bifoliolate, petiole about 1 cm. long;

leaflets elliptic-ovate, acute or shortly acuminate, densely scurfy-pubescent below,

nearly glabrous above, 6-8 cm. long and 3-4.5 cm. broad; petiolules about 1 cm.

long, somewhat canaliculate above; inflorescences short, few-flowered axillary di-

chasia hardly exceeding the subtending leaves, scurfy-pubescent; calyx narrowly

campanulate, subtruncate with minute lobes, 7-9 mm. long and 4-5 mm. in diam-

eter, scurfy-pubescent as is the 4-6 mm. long pedicel; corolla bilabiate, about 3 cm.

long, narrow and divided to near (6-8 mm.) the base, slightly arcuate, scurfy-

pubescent outside, less so within, the upper lip bilobed, the lobes short, about

4 mm. long, obtuse or somewhat acute, the lower lip trilobed, the lobes 6-8 mm.

long, the lateral lobes similar, obliquely ovate, obtuse, the middle lobe broadly

rhombic, obtuse, 6-7 mm. long and as broad; corolla tube 6-8 mm. long, cylindric;

stamens 4, didymous, connivent under the upper lobe of the corolla, the longer

pair adnate just below the sinus on the lower lip of the corolla, the shorter pair

adnate on the upper lip; anthers 2-celled, widely explanate; style about 20 mm.

long, slightly flattened at the stigmatic apex, about as long as the shorter pair of sta-

mens; disc inconspicuous but surrounding the base of the ovary; ovary puberulent .

COSTA RICA: Flowers white, tree to 3 m., roadside 8 km. south of

San Miguel, province of Heredia, alt. 800 m., 9 July 1964, Roy W.

Went 42 (F, type; OKL).

This species is easily distinguished from the other species of Tyn-

nanthus by the relatively long narrow corolla. The species seems to

have most of the characters of this genus and I find no other into

which it fits as well. The corolla is relatively long and narrow and

divided more nearly to the base than any other tropical American

member of the family known to me.

BURSERACEAE

Dacryodes kukachkana L. Wms., sp. nov.

Arbores polygamodioeciae vel polygamae. Folia 4-5-jugata; foliola elliptico-

oblonges, acuminatia, subcoriacea; inflorescentiae axillariae, paniculatae; calyx

cupulatus, trilobatus, lobi truncati; corolla trilobata, lobi valvati, ovati, acuti;

fructus ovoideus, vernicosus.

Large polygamodioecious or polygamous trees of the dry tropical forest. Branch-

lets densely puberulent, becoming glabrous and grayish with age; leaves alternate,

15-20 cm. long, 4-5-jugate, imparipinnate or apparently even pinnate; petioles

10-15 cm. long, canaliculate and slightly winged at the base, angled above; leaflets

opposite, mostly elliptic-oblong, acuminate, unequal at the base, subcoriaceous

when mature, with 10-12 pairs of lateral veins and conspicuously reticulate-veined

on both sides when mature, petiolules 4-5 mm. long; inflorescences axillary, panicu-

late, mostly about 10 cm. long and much exceeded by the leaves; flowers fragrant;

calyx cupulate, trilobate, truncate, the lobes about 0.5 mm. long and fused into

the hypanthium-like, thickened base; corolla trilobate, the lobes valvate, ovate,

acute, free or slightly connate at the base; stamens obdiplostemenous, 6, affixed to

252 FIELDIANA: BOTANY, VOLUME 31

the disk and nearly sessile; disk fleshy, surrounding the ovary and fused with the

calyx; fruits ovoid, bright green, vernicose, about 20 mm. long and 15 mm. in

diameter.

PERU: "Copal, lacre." Flores con olor; bosque seco tropical,

Leoncio Prado, distrito Honoria, prov. Pachitea, depto. Huanuco,

4 Sept. 1963, Rafael Lao Magin 79 (type, F; WIS; others) ; same local-

ity, frutos color verde claro, 24 Oct. 1963, Rafael Lao Magin 92.

BRAZIL: Tree 80 feet tall, trunk 10 inches diameter, municipality

Sao Paulo de Olivenca, state of Amazonas, 11 Sept. -26 Oct. 1936,

Krukoff8194.

Related to Dacryodes peruviana (Loesner) Lam and to D. olivifera

Cuatr. from both of which it is distinguished by the smaller leaflets

and smaller fruits.

The two specimens collected by Mr. Lao Magin are from the same

tree (tree number 105-P) . Wood specimens were collected from the

tree. The wood specimens contain 2.96 per cent of silica, perhaps the

second highest known silica content for any known American wood

according to Dr. B. Francis Kukachka, for whom this tree is named.

The leaves of this tree are unusual in that certain of them avail-

able from the type tree appear even pinnate. In each case, however,

it appears that there may have been an initial which did not develop.

The specimens in University of Wisconsin herbarium and in U. S.

Forest Products Laboratory collections show odd pinnate leaves.

The Krukoff collection has one leaf that is odd pinnate, the other

imparipinnate. Dr. Krukoff's collections show some slight differ-

ences from the type collection — fewer lateral veins in the leaflets and

the reticulate venation between them less conspicuous. The speci-

men is in fruit only and it is, of course, possible that it does not belong

to the species described here. The description is drawn from the

Peruvian material.

CELASTRACEAE

Crossopetalum parvifolium L. Wms., sp. nov.

Frutices vel herbae suffruticosae. Inflorescentiae axillares, cymosae, pauci-

florae; flores perparva; sepala cupulata, 4-lobata, lobi suborbiculares, plusminusve

0.5 mm. longes; petala obovata vel oblongo-obovata, obtusa; discus cruciformis;

fructus biloculare.

Suffruticose herbs or shrubs. Branchlets 4-5-angled and sulcate, with increas-

ing age merely striate or nearly smooth, lenticellate on older wood; leaves opposite,

alternate, or rarely ternate, linear to linear-lanceolate, acute or shortly acuminate,

obscurely denticulate or entire, blade 5-15 mm. long and 1-4 mm. broad, petiole

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 253

to 1 mm. long; inflorescences axillary, cymose, few-flowered, mostly about 2 cm.

long, the rachis subfiliform; flowers minute, red; sepals cupulate, lobes 4, sub-

orbicular, glabrous or sparsely hirsute, about 0.5-0.7 mm. long; petals 4, obovate

or obovate-oblong, obtuse, about 1-1.5 mm. long, imbricate in bud; stamens in-

serted on the margin of the cruciform (4-lobed) disc, the filaments subglandular,

short (±0.5 mm.), the anthers in anthesis about 0.3 mm. long and a little broader;

style very short or none, stigmas 2; fruit apparently 2-celled, narrowly obovate,

about 8 mm. long, slightly asymetrical.

MEXICO: Flowers red, wooded slopes near crest of ridge, 2 miles

south of Tuxtla Gutierrez along the road to Villa Flores, Municipio

de Tuxtla Gutierrez, alt. 2,800 feet, 16 October 1965, Breedlove &

Raven 13327 (F, type; DS; others).

Closely related to Crossopetalum gonocladum (Urban) Rothm.

(basionym, Rhacoma gonoclada Urban) which is known to me from a

few specimens from Hispaniola. The present species is distinguished

from that of Hispaniola by several not very impressive characters —

larger leaves, longer inflorescences, larger flowers, a cruciform disc,

branching less fastigiated. The geographical range is very different

and, to the best of my knowledge, some 800 miles exists between the

Mexican locality and those in Hispaniola with no collections of either

species known between. Dr. Urban described the ovary of C. gono-

cladum as 4-locular, ours seems to be bilocular.

Crossopetalum parvifolium and C. gonocladum seem quite differ-

ent from the other species of the genus known to me. The generic

limits in this family in the neotropics are anything but sharp and I

have no intention of becoming involved.

Wimmeria acuminata L. Wms., sp. nov.

Arbores usque ad 12 m., glabri. Folia lanceolata vel elliptico-lanceolata, acu-

minata, serrulata, longe petiolata; fructus trialatus, suborbicularis, plusminusve

1.5 cm. altus et 2 cm. latus.

Trees to about 12 m. tall. Twigs glabrous, terete or the newer ones somewhat

angled. Leaves lanceolate or elliptic-lanceolate, acuminate, attenuate to the peti-

ole, serrulate or sometimes nearly entire near the base, the teeth obscurely gland-

tipped or not, subcoriaceous, long petiolate, the lateral veins leave the mid-nerves

at an acute angle and trend toward the apex of the blade, blade 3.5-8 cm. long and

1.3-2.4 cm. broad, the petioles mostly about 1 cm. long; inflorescence a few-

flowered cyme, 1-2 cm. long and much shorter than the subtending leaves; flowers

unknown; fruit trialate, broadest at the middle, suborbicular in outline, notched

at both base and apex, about 1.5 cm. high and 2 cm. broad, individual wings about

1.5 cm. long and 0.7-0.8 cm. broad.

MEXICO: Tree 40 feet tall, Chorreadero de Tuxtla, 5.6 miles east

of Chiapa de Corzo along Mexican Highway 190, Municipio de Chi-

WIMMERIA1ACUMINATA L. WMS

FIG. 1. Wimmeria acuminate. A, Flowering branch, natural size. B, Flower,

X 5. C, Flower dissected to show disc, X 5. D and E, Fruits, natural size.

254

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 255

apa de Corzo, state of Chiapas, alt. 2,500 feet, 20 February 1965,

Breedlove 9090 (type, F; DS).

Wimmeria acuminata is related to W. concolor Schlecht. & Cham.

It may be distinguished by several details — the consistently longer

petioles, the veins of the leaf which leave the mid-nerve at a more

acute angle and trend toward the apex rather than the margin, the

shorter fruits which tend to be suborbicular rather than elliptic-

oblong or broadly elliptic. The recently described W. sternii Lun-

dell from Panama, and which we have gotten in Nicaragua (Williams,

Molina & Williams 23859} is a related species which may be distin-

guished by its fruits and much smaller leaves.

ELAEOCARPACEAE

Elaeocarpus serratus L. Sp. PI. 515. 1753.

COSTA RICA: Trees in pasture and along roads near sea level, Gol-

fito in the Golfo Dulce region from the surroundings of the airfield,

14 Sept. 1964, Lems 5349.

To the best of my knowledge this is the first report of this tree in

Central America as a naturalized plant. It is especially curious in that

it comes from a region where Paul Allen lived and botanized for sev-

eral years yet he does not report it in his The Rain Forests of Golfo

Dulce.

Dr. Lems (in lit.) reports that but a single tree about 7-9 m. tall

was seen. This grows along a path on the south side of the airport

and was being used as a fence post.

ERIOCAULONACEAE

Eriocaulon molinae L. Wms., sp. nov.

Herbae parvae graciles caespitosae palustres usque ad 15 cm. altae. Folia

linearia, acuta; inflorescentiae capitatae, usque ad 20-30-flores; flores trimeres,

fern, sepala oblanceolata, petala elliptico-lineares, masc. sepala connata trilobata,

petala in tubo connata; stamina 6.

Small slender dioecious or possibly monoecious palustrine herbs to 15 cm. tall

with short stems or none, with tufted leaves and with few to several capitate in-

florescences each borne on a long slender bractless peduncle. Leaves pale green,

linear, acute or short acuminate, with conspicuous septiform lines contrary to the

mid-nerve, glabrous, 2-3.5 cm. long and 1-2 mm. broad; inflorescence capitate,

borne on long slender peduncles to 15 cm. long, heads at anthesis about 4-5 mm.

broad and with 20-30 flowers; pistillate flowers trimerous, each subtended by a

cucullate, acute, elliptic, nigrescent bract about 1-1.5 mm. long, sepals 3, oblan-

ceolate, acute or acuminate, somewhat cucullate, short unguiculate, nigrescent,

about 1.5 mm. long, petals elliptic-linear, acute, free, hyaline, styles 3, slender,

256 FIELDIANA: BOTANY, VOLUME 31

ovary trilobate and trilocular, about 0.3 mm. long at anthesis; staminate flowers

trimerous, subtended by nigrescent, elliptic bracts about 1 mm. long, calyx united

to form a tridentate bract open along one side, the tips ciliate, about 1.2 mm. long,

the petals united into a tube about 1.2 mm. long, the 6 stamens borne on short

filaments from the mouth of the tube, non-functional style globose.

HONDURAS: Abundante en sabanas palustres cerca de la quebrada

El Chorrito, Cerro de Hule, 20 km. al sur de Tegucigalpa, Depto.

Morazan, alt. 1,500 m., 27 Oct. 1966, Molina 18500 (type, F; EAP;

others) ; en pantano, colinas empantanadas, area de pino y roble entre

Las Flores y San Juan del Rancho, Depto. Morazan, alt. 1,500 m.,

25 Nov. 1948, Molina 1740; scarce, shallow open bogs, San Juan del

Rancho, eastern base of Cerro de Uyuca, Depto. Morazan, alt.

1,500 m., 10 Dec. 1949 and 30 Oct. 1951, Standley 25028 and 29141;

heads blackish, common in swampy swale and meadows near Ojo-

jona, Dept. Marazan, alt. 1,350 m., 6 Dec. 1966, Williams & Correll

29289.

The species seems to be most closely allied to E. mexicanum Mol-

denke of the species known from Central America and Mexico. The

species being described is more delicate with differences also in detail

of floral structure. We have known this species for many years but

have misdetermined it as E. seemanii Moldenke, a dimerous species

known from lowland Panama.

FLACOURTIACEAE

Abatia borealis L. Wms., sp. nov.

Arbores debiles usque ad 5-7 m. Folia opposita, lanceolata vel ovato-lanceo-

lata, acuta, serrulata, subtus stellato-floccosa; inflorescentia racemosa, multiflora;

corolla 0; calyx 4-lobatus, lobi lanceolati vel lanceolato-oblongi; stamina numerosa.

Small weak forest trees to 5-7 m. tall. Twigs glabrous or glabrescent, the

younger ones somewhat flattened; leaves opposite, lanceolate to ovate-lanceolate,

acute, serrulate, thinly to densely stellate- floccose below, sparsely pubescent with

short, simple or bipartite hairs below, the blade 6-15 cm. long and 1.5-5 cm. broad;

petioles stellate pubescent, 1-2.5 cm. long; inflorescence a terminal many-flowered

spike-like raceme up to about 20 cm. long and 2-3 cm. in diameter; flowers green-

ish-yellow to orange, fragrant, on pubescent pedicels about 5-7 mm. long; corolla

none; calyx 4-lobed, the tube short, the lobes lanceolate or lanceolate-oblong, acute,

stellate pubescent outside, ornamented at the base inside with elongated filament-

like but sterile processes 5-6 mm. long and 1.5-2 mm. broad; fertile stamens many,

inserted at the base of the calyx below the filament-like processes, exceeding the

calyx in length at maturity; ovary densely pubescent; style exceeding the calyx.

COSTA RICA: Flowers yellow, fragrant. Weak tree to 5 m. tall in

cut-over montane cloud forest area, Cordillera de Talamanca about

20 km. north of San Isidro de General along Pan American Highway,

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 257

province of San Jose", alt. 2,800 m., 29 Jan. 1965, Williams, Molina,

Williams & Gibson 28507 (F, type; EAP; others) ; fl. jaunes, arbre de

5-7 m. de haut, Cerro de las Vueltas, I 1897, Pittier 10511; flowers

greenish yellow, scarce small tree in open forest, Cerro de las Vueltas,

province of San Jose, alt. 2,700-3,000 m., 29 Dec. 1925-1 Jan. 1926,

Standley & Valerio 43506; orange panicles, shrub 5m., along Pan

American Highway in oak forest from Cerro de la Muerte to near

San Isidro del General, alt. about 2,500 m., 23 Jan. 1965, Lems 5182

(F, NY); Terraba, Leon 1119; flowers yellow, honey-scented tree

25 feet [tall], alt. 8,500 feet, 6 Sept. 1923, Lankester 674.

This plant was collected first by Pittier in 1897. Capt. Donnell-

Smith named the plant Abatia parviflora R. & P. and Pittier con-

curred in the determination at a later date. Standley, who had

collected the plant (1925-1926), used the same name in The Flora of

Costa Rica.

Dr. Sleumer annotated our Andean material of the genus years

ago and certainly confused two quite distinct species under the name

A. parviflora. One of those is much like the present species and it

may be for this reason that Standley accepted the Andean name in

The Flora of Costa Rica.

LECYTHIDACEAE

Gustavia cincta (Cuatr.) Williams & Kukachka, comb. nov.

Eschweilera cincta Cuatr., Fieldiana, Bot. 27, 2: 83, fig. 1951.

While Dr. Cuatrecasas has placed this Colombian species in Esch-

weilera it appears to us to be rather a Gustavia, — which is borne out

by the anatomy of the wood. Anatomically, Gustavia differs rather

markedly from Eschweilera. The most prominent difference, observ-

able with a hand lens, is the reticulate parenchyma which character-

izes Gustavia as opposed to the concentric arrangement of parenchyma

in Eschweilera.

The illustration cited above is inaccurate in that the leaves are

shown to be scattered rather than crowded at the apex of the annual

growths, a character to which Dr. Cuatrecasas called attention in his

description.

LEGUMINOSAE

Pithecellobium leucocalyx (Britt. & Rose) Standl. Field Mus.

Bot. 4: 308. 1929. Samanea leucocalyx Britt. & Rose, No. Am. Fl.

23:34. 1928.

258 FIELDIANA: BOTANY, VOLUME 31

GUATEMALA: "Cenicero." Tropical moist forest, El Rosario,

Depto. Pete"n, alt. 125 m., 18 April 1966, McClay & Clara 41.

This specimen was collected in the "Projecto Evaluacidn Forestal"

being carried on in the Pete"n lowlands in Guatemala. It is interesting

to note that the common name given this tree is the same as that for

the much better known P. saman.

In a recent letter Dr. B. Francis Kukachka writes that anatomi-

cally the genus is "a real nightmare and displays as much variability

as the entire Mimoseae. I believe that it would be impossible to

write a generic description of its anatomy that would not hold equally

well for the subfamily." Dr. Kukachka points out, however, that it

would be possible to write generic anatomical descriptions of some of

the genera that have been segregated from Pithecellobium.

ORCHIDACEAE

Cranichis trilobata L. Wms., sp. nov.

Herbae terrestres, erectae, graciles, ca. 10 cm. altae. Folia elliptica vel late

elliptica, acuta, glabra, sessiles vel subsessiles; inflorescencia racemosa, pauciflora;

sepalum dorsale anguste elliptico-oblanceolata, obtusa; sepala lateralia ovato-lan-

ceolata, acuta, 3-5-nervia; petala longa unguiculata, lobi basium laminorum promi-

nentes; labellum unguiculatum, unguis callosum, lamina apice trilobata basi leviter

concavum; columna generis.

Small terrestrial herbs about 10 cm. tall and with a rosette of sessile or sub-

sessile leaves. Stem slender, strict, with two or three sheathing scales, glandular

puberulent above becoming glabrescent below; leaves elliptic to broadly elliptic,

acute, glabrous, thin, sessile or essentially so, 2.5-5 cm. long and 1-2.2 cm. broad;

flowers white, disposed in a few-flowered raceme; dorsal sepal narrowly elliptic-

oblanceolate, obtuse, 3-nerved, obscurely ciliate-glandular, about 5 mm. long and

1.5 mm. broad; lateral sepals completely free, symmetrical, about 5 mm. long

and 3 mm. broad; petals on long unguiculate claws attached near the middle of

the column, strongly asymmetrical with a large basal lobe on the outer side of the

blade, the petals connate at their tips and connate also with the tip of the dorsal

sepal, claw about 1.5 mm. long, the blade about 3 mm. long and 2 mm. broad across

the basal lobe; lip attached near middle of column, unguiculate with a fleshy claw

bearing a cucullate callus closed at its base, the blade strongly trilobate at the apex

and broadest across the apex, obcordate in general outline and obscurely saccate

at the base of the blade, the fleshy claw about 1 mm. long, the blade about 2 mm.

long and as broad at the apex; column about 3 mm. long, the lobes of the clinan-

drium rounded; pedicellate ovary glandular pilose, to about 10 mm. long.

MEXICO: Flowers white, steep rocky slopes with Quercus along

Mexican highway 190 in the Zinacantan paraje of Mectajoc, Muni-

cipio of Ixtapa, state of Chiapas, alt. 3,500 feet, 17 Aug. 1965, Breed-

love 11856 (type, F; DS) ; terrestrial, flowers white, steep, sandy, moist

ravine banks in open Pinus-Quercus woods with grass groundcover

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 259

on highway 195, 3 km. N of junction on highways 190 and 195

(16°50'N; 92°50/W), state of Chiapas, alt. ca. 1,100 m., 9 Aug. 1965,

Roe, Roe & Mori 1088 (F, WIS); flowers white, steep, rocky, open

roadside bank adjacent to maize fields, with shrubby Crataegus,

Compositae, Leguminosae and some ferns and mosses, Cumbre de

Acultzingo near top of ridge overlooking valley, 4 km. south of Acult-

zingo on highway 150 at km. 283 (18°40'N; 97°20'W), state of Vera

Cruz, alt. 1,700 m., 17 Aug. 1965, Roe, Roe & Mori 1279 (F, WIS).

The relationship of this species is with Cranichis schaffneri Reichb.

f. and less so to C. gracilis L. Wms. Cranichis trilobata is distin-

guished easily from C. schaffneri by the prominently trilobate apex

of the lip and its fleshy claw with cucullate callus. The long basal

lobes of the petals are also distinctive as are the sessile or nearly ses-

sile leaves.

It is curious that these collectors found this species at three local-

ities between 9 and 17 Aug. 1965 and that it has not turned up pre-

viously. The species must be fairly common at least for a short time

during the year and it is curious that it has been missed.

Pleurothallis microchila L. Wms., sp. nov.

Herbae parvae epiphyticae usque ad 15 cm. altae. Caules secondarii elongati,

superne alati. Folia ovato-lanceolata, acuta, coriacea. Inflorescentia uniflora vel

fascicula pauciflora, pedunculi filiformes, uniflores. Sepalum dorsale anguste ob-

longo-lanceolatum, actum; sepala lateralia adnata, anguste oblongo-ovata; petala

linearia, obtusa; labellum perparvum, rhombico-ovatum, tricallosum, obscure

pubescens.

Small epiphytic herbs up to 15 cm. tall. Secondary stems elongate, 5-12 cm.

long, covered with 2-4 scarious sheaths which appear alate on the two angles above,

brownish ; leaves ovate-lanceolate, acute, coriaceous, somewhat conduplicate at the

base, 6-10 cm. long and 2-3 cm. broad; inflorescence a single long pedunculate

flower or a fascicle of 2-3 long pedunculate flowers, borne from the axil of the leaf

in an inconspicuous sheath 6-8 mm. long; the peduncle subfilamentose and with a

tubular bract below the middle, exceeded in length by the subtending leaf; flowers

red or white with red markings; dorsal sepal narrowly oblong-lanceolate, somewhat

acute, about 10 mm. long and 3 mm. broad; lateral sepals completely adnate one

to another, narrowly oblong-ovate, obtuse, about 10 mm. long and 4 mm. broad;

petals linear, obtuse, subterete and canaliculate above, narrowest toward the base;

lip very small, rhombic-ovate, obtuse, with two lateral angles above the base, disc

tricallose, one median one starting above the base, two lateral submammilate ones

at the basal angles, the lip with obscure crispate pubescence, about 1.3 mm. long

and 0.9 mm. broad.

GUATEMALA: Epiphyte, wet mountain forest near Aldea Fraterni-

dad, between San Rafael Pie de la Cuesta and Palo Gordo, west fac-

ing slope of the Sierra Madre mountains, department of San Marcos,

260 FIELDIANA: BOTANY, VOLUME 31

alt. 1,800-2,400 m., 10-18 Dec. 1963, Williams, Molina & Williams

25671 (typeF).

Closely related to Pleurothallis nelsonii Ames which is known from

nearby Chiapas in Mexico and to the very similar P. arietina Ames

of Costa Rica. Superficially, the present species is very similar to

these two but it is easily distinguished by the very different form of

the lip which is much smaller than in the related species.

A sterile specimen (Steyermark 36266) may well be this species.

It had been questionably determined by Dr. Donovan S. Correll,

during his studies for the Orchidaceae of Guatemala, as Pleurothallis

arietina Ames.

Sobralia pleiantha Schltr. Fedde Repert. 3: 79. 1906; Ames,

Bot. Mus. Leafl. Harv. Univ. 4: 47-51, t. 1936.

An interesting Sobralia and, even though it is relatively little

known, it must be exceedingly common for we have seen it every-

where in the rain-forest of the El General Valley in Costa Rica. It is

difficult to collect except when the big forest trees are being felled,

for it occupies the high branches of these. One of our collections will

be widely distributed.

COSTA RICA: Flowers cream, large epiphytic clumps on forest

trees, rain-forest area long Rio Cacao above Pan American High-

way, El General Valley, province of Puntarenas, alt. 900 m., 30 Jan.

1965, Williams, Molina, Williams & Gibson 28671 .

We saw the plant in one place, quite abundant, on rocks in a large

clearing, which forms part of the "yard" of Dr. Alexander Skutch's

home, "El Quizarra", in Costa Rica.

Sobralia xantholeuca Hort. ex Williams Orch. Grow. Man.

ed. 6:576. 1885.

MEXICO: Flowers yellow; epiphyte; steep, heavily- wooded slopes

at Lagos de Colores, 25 miles east of La Trinitaria, Municipio La Tri-

nitaria, Chiapas, alt. 5,200 feet, 27 May 1965, Breedlove 10044-

Charles Lankester told me he had seen this species in Chiapas and

on his sight identification I added Mexico to the range of the species

in "An Enumeration of Orchidaceae of Central America, British Hon-

duras and Panama." This excellent specimen is the first I have seen

from Mexico.

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 261

PHYTOLACCACEAE

Gallesia integrifolia (Spreng.) Harms, Pflanzenf. ed. 2. 16c:144.

1934.

PERU: "Ajos", El Caucho, Depto. & Pcia. de Tumbes, March

(flowers) to May (fruits) 1963, Jose Vargas A. 13 & 20 (F; WIS;

G; NY;Serv. For.; Lima).

As far as I know these collections from the Pacific department of

Tumbes are the first record of the tree out of the Amazon basin. The

garlic-like odor remains overpowering after three years, justifying the

common name.

The wood is said to be used in carpentry and for railroad ties and

beams.

SABIACEAE

Meliosma dives Standl. & Steyerm. Field Mus. Bot. 23: 61. 1944.

MEXICO: Bank of Rio Hondo 4 miles north of Jitotol, municipio

de Jitotol, state of Chiapas, alt. 5,500 feet, 12 Feb. 1965, Breedlove

8961.

The specimen cited is a good match for the type. The species is

new to the Mexican flora.

The genus Meliosma is a difficult one in Mexico and Central Amer-

ica possibly due to the fact that there are relatively few good speci-

mens available for study in herbaria.

Meliosma ira (Liebm.) L. Wms., comb. nov. Lorenzanea ira

Liebm. Vid. Medd. Kjoebenhavn 71. 1850.

This Costa Rica Meliosma has been confused by Standley and

others with Meliosma glabrata (Liebm.) Urban, yet the two seem to

be fairly distinct and easily separated on leaf venation, fewer veins

in M. ira; the inflorescence in M. glabrata (isotope and photograph of

specimen which was in Berlin) is borne on old wood while those of M.

ira are in the axils of leaves on the current year's growth; M. glabrata

comes from relatively low country (Turrialba, some 600 m.), while

M. ira is a mountain species (Naranjo to Tilaran and Poas, perhaps

elsewhere in Costa Rica, similar mountain area in Nicaragua) .

COSTA RICA: Brenes 211+30; Skutch 3683; Austin Smith H135, 144,

A3 85, H445, A460, A514, A529, A678, H535, NY 700, NY 1355, 4195,

10009; Standley 33680; Williams & Williams 24587.

262 FIELDIANA: BOTANY, VOLUME 31

A very mature specimen from the Cordillera Central de Nicaragua

seems to be this species, — Williams, Molina & Williams 23664.

SAPOTACEAE

Bumelia pleistochasia Donn.-Sm. Bot. Gaz. 18: 4. 1893 ;B. lei-

ogyna Donn.-Sm., I.e., syn. nov.; B. persimilis Hemsl. sensu Cron-

quist, Journ. Am. Arb. 26: 449. 1965.

Dr. Cronquist in his review of the genera Dipholis and Bume-

lia has placed Bumelia persimilis Hemsl. in Bumelia and referred

there as synonyms several names, including three used on Guate-

lan material. These are B. pleistochasia Donn.-Sm., B. leiogyna

Donn.-Sm. and B. guatemalensis Standl.

I have had occasion to study the isotype specimens of Bumelia

persimilis Hemsl. (Botteri 989} from Mexico. I believe that these

specimens should be referred, in synonymy, to Dipholis salicifolia (L.)

A. DC. Most, if not all, of the remaining specimens which Cronquist

has annotated with the name Bumelia persimilis Hemsl. are Bumelia

and I believe should bear the name B. pleistochasia Donn.-Sm.

Cronquist places Bumelia subsessiliflora Hemsl. as a subspecies of

B. persimilis. It would seem, therefore, that this name might be

taken up as the basic name for this species in its broad sense. I have

at hand a photograph and a fragment of an isotype (G) of the species

and, in addition, two fine specimens collected by Howard Scott Gentry

(7294 & 7308) in Sinaloa which seem to belong with B. subsessiliflora.

If so, then the fruits which are truncate, and even a bit concave at the

apex, indicate that the species is quite different from Bumelia pleisto-

chasia which extends from south Mexico through Central America to

Venezuela.

The type specimen (US) of Bumelia guatemalensis Standl. Trop.

Woods 4: 9. 1925, which Cronquist placed as a synonym of B. per-

similis Hemsl. (sensu Cronquist) , is undeterminable. The species will

be excluded in Flora of Guatemala.

MASTICHODENDRON. This name was published by Cronquist

(Lloydia 9: 245. 1946) to contain some American plants formerly

referred to Sideroxylon L., a genus based on S. inerme L., an African

plant, which Cronquist thinks, apparently with reason, not congeneric

with the American S. foetidissima Jacq.

Dr. H. J. Lam proposed provisionally that the oldest valid syn-

onym, "Mastichodendron Jacq. in Hedw. Gen. (1806)" should be

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 263

taken up for certain American species. This name was not prop-

erly published.

In Index Nominum Genericorum the name is given as "Masticho-

dendron (Engler) Lam, Rec. Trav. Bot. Ne'er!. 36: 521. 1939." Dr.

Lam made no such proposal nor was there a basionym given which

would justify this assumption.

Mastichodendron foetidissimum (Jacq.) Cron. var. gaumeri

(Pittier) L. Wms. comb. nov. Sideroxylon gaumeri Pittier. Contr.

U. S. Nat. Herb. 13: 460. 1912. Mastichodendron foetidissimum

subsp. gaumeri Cronquist, Lloydia 9: 248. 1946.

This change is merely to reflect the subspecific category used in

Flora of Guatemala.

NlSPERO Aubreville. Prof. Aubreville (in Adansonia 5 : 19, 1. 1 .

1965) has published this name for one of the sapotaceous plants of

Central America. The name itself would indicate that it was in-

tended for the plant commonly known in many parts of Central

America as nispero. This plant will appear in a forthcoming part of

the Flora of Guatemala as Manilkara achras (Miller) Fosberg. The

illustration given by Prof. Aubreville to illustrate the name ''Nispero

achras (Miller) Aubr." is the common zapote of the region, Pouteria

mammosa (L.) Cronquist.

The generic name Nispero is a nomen subnudum; the specific name

is improperly published (but I assume that the basionym is Sapota

achras Miller) ; the combination is used to identify the illustration of

a plant that no Central American ever called nispero.

Pouteria durlandii (Standl.) Baehni, Candollea 9: 422. 1942.

Lucuma durlandii Standl. Trop. Woods 4: 5. 1925. Paralabatia dur-

landii AubreVille, Adansonia ser. 2. 5: 21. 1963.

The type of this species from department of Pete"n, Durland s.n.

(US), is essentially sterile and Baehni has annotated it as Pouteria ?

durlandii (Standl.) Baehni. Cronquist did not see the type, or did

not annotate it. Although we are following his treatment of the spe-

cies for Flora of Guatemala it is my feeling that because the type of

this species is essentially sterile and that of Pouteria izabalensis is

sterile there can be no assurance as to what they represent and per-

haps would be best be treated as nomina dubia.

Dr. B. F. Kukachka advises, in lit., that the wood anatomy of

this species is suggestive of the genus Ecclinusa.

264 FIELDIANA: BOTANY, VOLUME 31

Pouteria izabalensis (Standl.) Baehni, Candollea 9: 347. 1942.

Lucuma izabalensis Standl. Trop. Woods 4: 6. 1925.

The type of this species, Whitford & Stadmiller 35 (US) , from the

department of Izabal, Guatemala is sterile and the species should

never have been described from it. In Flora of Guatemala we shall

follow Cronquist's interpretation of the taxon (although he did not

see the type) with considerable reluctance. See note on P. durlandii

above.

Pouteria km del Hi (Standl.) L. Wms., comb. nov. Sideroxylon

lundellii Standl. Publ. Carnegie Inst. Wash. 461: 79. 1935.

This species has been treated as a synonym of P. belizensis

(Standl.) Cronquist by Cronquist. However, the type of Standley's

Lucuma belizensis is sterile and not determinable with any degree of

certainty. The type of Sideroxylon lundellii Standl. from Pete"n

(Lundell 2767) is a fine specimen.

SYMPLOCACEAE

The Symplocaceae consists of a single genus, which is said to be

composed of about 300 species, occurring in America, Asia, and Aus-

tralia. The species are mostly tropical but extend into temperate

regions on both sides of the equator in America.

The last and only monographic treatment of Symplocos is by

Brand (Pflanzenr. IV. 242: 1-100. 1901). This is entirely inade-

quate for Central America today. Brand cited fewer than 30 speci-

mens for Mexico and Central America — these distributed among ten

species, of which seven were endemic to Mexico, two to Guatemala,

and one in Costa Rica.

There are 11 species of Symplocos in the manuscript for the Flora

of Guatemala, which includes British Honduran plants, one of these

without a formal name because of inadequate material, and two

which are described as new below. In the other Central American

countries there are the following species of Symplocos. El Salvador

has one species; Honduras, two species; both described below, and

material available indicates that there are at least two more; Nica-

ragua, none is recorded but there are two species there, one of which

seems to be undescribed ; Costa Rica, six species recorded of which four

or perhaps five are distinct, one more, or perhaps two, may be unde-

scribed. In the Republic of Panama there is recorded one species.

A revision of the American species should be well worthwhile.

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII

265

FIG. 2. Symplocos bicolor. A, Branch,

C, Fruit, X 5.

natural size. B, Flower, X 3.

I have seen the types of all of the species of Symplocos reported

for Guatemala except S. hartwegii and S. limoncillo — the first seems

to offer no problem and is quite clear, and although I have a frag-

ment of the type of S. limoncillo it is not adequate to place the spe-

cies with certainty.

Symplocos bicolor L. Wms., sp. nov.

Arbores parvae usque ad 10 m. Folia elliptica vel elliptico-obovata, acumi-

nata, obscure repando-serrulata vel integra; inflorescentia pauciflora, cymosa; calyx

parvus, lobi ovati, aliolati; corolla glabra, lobi oblongi, obtusi; stamina usque ad 30;

fructus ovoideus.

Trees to 10 m. tall and 20 cm. in diameter, the twigs sparsely strigose pubes-

cent at first but soon glabrous, angulate but soon terete, grayish. Leaves elliptic

to elliptic-obovate, acuminate, obscurely repand-serrulate or entire, the two sur-

faces distinctly bicolored, the blades mostly 5-13 cm. long and 2-5 cm. broad, the

petioles somewhat fleshy, 0.5-1 cm. long; inflorescences short, few-flowered axillary

266 FIELDIANA: BOTANY, VOLUME 31

cymes; the rachis puberulent, mostly less than 1 cm. long; flowers white, fragrant,

small, mostly 7-9 mm. long; calyx lobes broadly ovate to suborbicular, obtuse,

glabrous, or nearly so, but somewhat ciliolate, about 1 mm. long; corolla divided

to below the middle, 5.5-7 mm. long, the lobes, oblong, obtuse, glabrous, reflexed

at anthesis, 4-6 mm. long; stamens about 30 in 2-3 series, the filaments connate

and forming a short corona, adnate with the corolla up to the throat of the tube,

about as long as the corolla; style about 6 mm. long, glabrous, the stigma subcapi-

tate, lobate; the fruits small, ovoid, glabrous, about 6-10 mm. long and 5-7 mm.

broad.

GUATEMALA: La Libertad and vicinity, 10 Jan. 1935, Mercedes

Aguilar H. 217; La Libertad, Pete"n, 2 June 1933, Lundell 357; wet

thicket, shrub 2 m., fruit purple black, near Puerto Barrios, Izabal,

25 April-6 May 1939, Standley 73114.

BRITISH HONDURAS: Maskal, 14 April 1934, Schipp 1253; tree 6

inches in diameter, flowers white, fragrant, on high ridge, Big Eddy

Ridge, Stann Creek Valley, 12 Jan. 1941, Gentle 3486; tree 2 inches

in diameter, flowers white, on wooded island, pineland, Mullins River

Road, Stann Creek District, 12 Jan. 1955, Gentle 8538; thicket near

Manatee Lagoon, 3 Jan. 1906, Peck 246; upright growing tree in

jungle, 30 feet tall, 4 inches in diameter, Mullins River Road,

29 March 1929, alt. 50 feet, Schipp 116; tree 30 feet tall, 7 inches in

diameter in forest, flowers white, sweetly perfumed, Mullins River

Road, alt. 100 feet, 15 Dec. 1931, Schipp 862 (type, F); tree 30 feet

tall and 4 inches in diameter, in forest, Camp 32, British Honduras-

Guatemala survey, alt. 2,100 feet, 17 April 1934, Schipp 1268.

This small flowered species has been one of the "components" of

a complex called S. martinicensis but seems to be adequately distinct.

It is easily distinguished by the small glabrous flowers, the small

fruits, and the distinct difference in intensity of the color of the two

leaf surfaces.

Symplocos molinae L. Wms., sp. nov.

Arbuscula vel arbores 2-8 m. altae vel ultrae. Folia elliptico-oblanceolata vel

oblongo-obovata, breviter acuminata vel obtusa, coriacea, glabra; inflorescentia

axillaris, pauciflora, racemosa; calyx glabris, lobi rotundati, ciliati, corolla anguste

campanulata, glabris, lobi oblanceolati, obtusi, leviter crenulato-lanceolati; stam-

ina 30-50; fructus ovoideus, glaber.

Shrub or small tree with smooth pale bark 2-8 m., rarely to 12 m. tall, the

branches glabrous, brownish or reddish. Leaves elliptic-oblanceolate to oblong-

obovate, short acuminate to obtuse, coriaceous, glabrous, lateral veins not promi-

nent but at about 45° angle from mid-nerve, blade 4-12 cm. long and 2.5-6 cm.

broad, the petiole fleshy, mostly 6-8 mm. long, inflorescences short, axillary, few-

flowered racemes; flowers white; calyx glabrous, small, the lobes rounded, ciliate,

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 267

about 1 mm. long and as broad; corollas mostly 8-10 mm. long, narrowly campanu-

late, divided to near the base, glabrous, the lobes oblanceolate, obtuse, minutely

crenulate-lacerate at the apex; stamens 30-50, bi-triseriate, the filaments connate

to near their apices and forming a corona about as long as the corolla, lightly adnate

to the corolla to the middle or above; style 7-9 mm. long, glabrous; ovary glabrous;

fruits ovoid, glabrous, 6-10 mm. long and 3-5 mm. in diameter.

HONDURAS: Flowers white, small tree to 8 m. on slopes of Mt.

Uyuca, Dept. Morazan, alt. 1,160 m., 7 Aug. 1946, Williams & Mo-

lina 10249 (type F; EAP).

In addition to the type, there are the following specimens, all from

the pine-oak highlands of central Honduras, the departments of

Morazan, El Paraiso, or Choluteca: Molina 1346, 1488, 10766, 14196;

Standley 13737, 14356, 21687, 21698, 21816, 23830, 28501, 29023;

Standley & Molina 4145; Williams 18252; Williams & Molina 11927,

12241, 12246, 13297, 14002, 19027; Valerio 2133, 3129.

This plant is a common one in the pine-oak highlands of central

Honduras and should be found in the similar regions of adjacent

Nicaragua. It belongs in the complex of species in Central America

that have been referred to both S. martinicensis and S. limoncillo,

neither of which seems to be in Honduras. The type of inflorescence,

details of flowers and color, shape (and coloration) of leaves, along

with a limited but logical range, help to distinguish the plant.

I am pleased to name this plant for Prof. Molina, superb field

companion for 20 years.

Symplocos vernicosa L. Wms., sp. nov.

Arbores, usque ad 15 m., folia ovales, elliptico-ovata vel elliptico-obovata,

breviter acuminata, supra vernicosa; inflorescentia abbreviata, axillaris, racema

vel cyma pauciflora; calyx strigillosum, lobi late rotundati, ciliolati; corolla sub-

campanulata, lobi oblongi, apices rotundates, obscure lacerates; stamina 40-45,

triseriata; fructus ingotus.

Trees mostly 5-15 m., the branchlets appressed pubescent, soon glabrous, red-

dish. Leaves oval to elliptic-ovate or elliptic-obovate, obscurely serrulate, abruptly

short acuminate, the upper surface very dark green and vernicose, the lower surface

paler, pubescent on the mid-nerve below, blade 4.5-10 cm. long and 2-3.5 cm.

broad, the petiole to 1 cm. long, fleshy, puberulent dorsally; inflorescence much

abbreviated, axillary few-flowered raceme or cyme; flower pink, waxy; calyx 2-

2.5 mm. long, short strigillose pubescent dorsally but the margins glabrous, edges

ciliolate, the lobes broader than long, rounded, about 1 mm. long and 1.5 mm.

broad; corolla 8-10 mm. long, subcampanulate, glabrous, divided to near the base,

the lobes oblong, the apex rounded and obscurely lacerate, reflexed in an thesis;

stamens 40-45, triseriate, the filaments connate and forming a corona with the an-

thers borne at or near the apex, adnate with the corolla up to about the middle;

style slightly shorter than the corolla, glabrous or sometimes pubescent at the base;

268 FIELDIANA: BOTANY, VOLUME 31

the stigma slightly enlarged and 3-4-lobate; fruits oblong-ovoid, 10-12 mm. long

and 5-7.5 mm. in diameter.

MEXICO: In forest, tree 7-8 m., flowers rose, Pinabeto, Moto-

zintla, Chiapas, alt. 2,586 m., 9 May 1945, Matuda 15483.

GUATEMALA: Chichen (or Coban), Alta Verapaz, alt. 1,600 m.,

March 1908, Tuerckheim II 2159.

EL SALVADOR: Corolla rose-lavender at the base shading to white

distally, tree 5-5.5 m. tall, cloud forest, east slope of Los Esesmiles,

lat. 14°21'N., long. 89°09/W., Dept. Chalatenango, alt. ca. 2,250 m.,

11 March 1942, Tucker 999 (MICH; US).

HONDURAS: Flowers lavender, tree 60 feet, upper rocky slopes and

summit of Cerro Sta. Barbara, dept. Sta. Barbara, alt. 2,750 m.,

5-6 April 1951, Allen, Armour & Chable 6081 ; berries blue, Rancho

Quemado, road to El Rosario, San Juancito, Dept. Morazan, 3 March

1952, Carlson 2636; cloud forest, tree 50 feet high with spreading

branches, La Laguna, San Juancito [Dept. Morazan], 4 April 1932,

Edwards P-46; flowers pink, tree 8 m., common, mountains above

San Juancito, Dept. Morazan, alt. 2,000 m., 22 Feb. 1949, Merrill,

Williams & Molina 15632; arbol 5 m., bosque de nubes de Calaveras,

Cordillera de Opalaca, Dept. Intibuca, alt. 1,700 m., 10 Abril 1956,

Molina 6427; paderones humedos del Huis, kilometro 110 sobre carre-

tera Marcala, Dept. La Paz, alt. 1,600 m., 11 Abril 1956, Molina

6488; flores rosadas, arbol 2-8 m., frequente en el bosque nebuloso

La Vuelta del Caite, noreste de San Juancito, Montana La Tigra,

Dept. Morazan, alt. 2,300 m., 17 Marzo 1957, Molina 7734; arbol

8-12 m., frequente, bosque mixto en Banos La Esperanza, Dept.

Intibuca, alt. 1,800 m., 6 abril 1957, Molina 7946; flowers pale laven-

der, tree 4 m. in cloud forest, near summit San Juancito Mountains,

Dept. Morazan, alt. 2,000 m., 22 March 1951, Williams 17431 (type

F); tree 5 m., cloud forest area, in mountains above San Juancito,

Dept. Morazan, alt. 2,000 m., 6 Nov. 1947, Williams & Molina

13369; flowers rose, tree 8 m., in cloud forest area, mountains above

San Juancito, Dept. Morazan, alt. 2,000 m., 20 Feb. 1948, Williams &

Molina 13677, 13740; arbol 8m., comun el la floresta de nubes de la

Montana de la Tigra, suroeste de San Juancito, Dept. Morazan, alt.

2,000 m., 5 Feb. 1950, Williams & Molina 17084.

This is another component of the complex that has been called

Symplocos martinicensis-S. limoncillo and, of those which I have seen

in the field, one of the most attractive.

WILLIAMS: TROPICAL AMERICAN PLANTS, VIII 269

This widely distributed montane species, so far as I know, is

found only in wet, cool forests rather than in the drier pine-oak forest

regions. All of the specimens which I have seen are at elevations of

1,600 meters or more.

The single specimen from Mexico is not in the best condition but

it seems to belong here.

The species seems to be most closely related to S. molinae, de-

scribed above, which occupies a part of the same geographical area,

but occurs down in the pine-oak region. The larger pink flowers, the

more pubescent calyx and inflorescence, and difference in color of

the two leaf surfaces help to distinguish this species.

Publications 1023, 1024 and 1025

UNIVERSITY OF ILLINOIS-URBANA