Journal of the Botanical Research Institute of Texas
J. Bot. Res. Inst Texas ISSN 1934-5259
The BRIT Press
andd
of botanical research and scientific discoveries
for the twenty-first century.
The BRIT Press— bringing out the best in botanical
science for plant conservation and education.
Direction and Coverage
The BRIT Press considers original research papers
concerned with classical and modern systematic
Bibliographical
Citation abbreviation for the
Journal of the Botanical Research Institute of Texas is
J. Bot. Res. Inst Texas following the principles
iternational Standard Serial No. (ISSN 1934-5259)
Frequency of Publication
J. Bot. Res. Inst. Texas is published semiannually
(summer/fall) as one volume
by the Botanical Research Institute of Texas.
Subscription Prices (2010)
$43. Personal (Individual/Family)
$100. USA (Institutional)
$110. Outside USA (Institutional)
Subscriptions/back Volumes
J. Bot. Res. Inst. Texas and Sida, Contr. Bot
Publications Assistant
Botanical Research Institute of Texas
Fort Worth. Texas 76102-4025, USA
Electronic mail: orders@brit.org
COMPOSITION
rhorngraphics, Plano, Texas; rlrchorn@verizon.net
Printing
Prepress production and printing in the United States of
America by Millet the Printer, Dallas, Texas
Cover Illustration
of Liatris aestivalis originally used on
BRIT's anniversary poster 2001.
Summer gayfeather flowers mid
Jul-Aug(-Sep) and is endemic to
Sida 19:768.2001.
Botanical illustration b;
Linny Heagy ©2001.
BRIT
PRESS
Table of Contents
SYSTEMATICS
Agalinis flexicaulis sp. nov. (Orobanchaceae: Lamiales), a new species from northeast Florida
John F. Hays
A new Lomatium (Apiaceae) from the Ochoco Mountains of central Oregon
Richard Helliwell
Potentilla uliginosa (Rosaceae: Rosoideae), a new presumed exiinct species from Sonoma
County, California
Barry C. Johnston and Barbara Ertter
Two new taxa of Scutellaria section Resinosa (Lamiaceae) from northern Arizona
S.L. Rhodes and T.J. Ayers
Salvia coriana sp. nov. (Lamiaceae), a new species from a cloud forest in western Guatemala
Taylor Sultan Quedensley and Mario E. VEuz PErez
A new species of Calceolaria (Calceolariaceae) from disturbed paramos in south Ecuador
Pamela Puppo
Poa unispiculata, a new gynodioecious species of cushion grass from Peru with a single
spikelet per inflorescence (Poaceae: Pooideae: Poeae: Poinae)
GERR.I Davidse, Robert J. Soreng, and Paul M. Peterson MISSOURI BOTANICAI
Una nueva especie de Fevillea (Cucurbitaceae: Zanonieae) de Costa Rica ^
Armando Estrada Ch. and Daniel SantamarIa A.
A new species of Angelonia (Plantaginaceae) from Mexico
AUG 1 6 2010
Kerry Barringer
Folia taxonomica 17. Dilkea (Passifloraceae)
three new species
Christian Feuillet
GARDEN
2. Conspectus of the species of the Gt
library
19
27
33
37
45
51
55
Miliusa wayanadica (Annonaceae), a new species from Western Ghats, India
M.K. Ratheesh Narayanan. P. Si janapai . N. Ami Komar. N. Samdiiaran, am. \1 Sivadasan 63
Stylidium darwinii (Stylidiaceae), a new trigger plant from Western Ghats of Karnataka, India
Sachin A. Punekar and P. Lakshminarasimhan 69
Studies in Capparaceae XXVII: six new taxa and a new combination in Quadrella
Xavier Cornejo and Hugh H. Iltis 75
Studies in Capparaceae XXVIII: The Quadrella cynophallophora complex
Hugh H. Iltis and Xavier Cornejo 93
Studies in the Capparaceae XXIX: synopsis of Quadrella, a Mesoamerican and West Indian genus
Hugh H. Iltis and Xavier Cornejo 1 1 7
Ten new Myrtaceae from eastern and northeastern Brazil
Revision of Lobelia sect. Speirema (Campanulaceae: Lobelioideae)
Revision of Lobelia sect. Plagiobotrys (Campanulaceae: Lobelioideae)
Thomas G. Lammers 169
RehabilitaciOn y lectotipificacibn del genero Tessiera, su relacibn con Diphragmus y Staelia
(Rubiaceae: Spermacoceae): una nueva combinacibn y un nuevo sinbnimo
Roberto M. Salas and Elsa L. Cabral 181
Planaltina nuevo g£nero de la tribu Spermacoceae (Rubiaceae), endemico del planalto central
de Brasil y una nueva especie del Estado de Goias, Brasil
Roberto M. Salas and Elsa L. Cabral
193
New names and combinations in the flora of Colorado. XIII
W.A. Weber and R.C. Wittmann
Validation of Exochordeae (Rosaceae)
James L. Reveal
The Dominican amber fossil Lasiambix (Fabaceae: Caesalpinioideae?) is a Licania (Chrysobalana<
Kenton L. Chambers and George O. Poinar Jr.
A new combination in Lagotis (Plantaginaceae)
David F. Murray, Reidar Elven, and Kanchi N. Gandhi
Neotypification of Hechtia podantha (Bromeliaceae)
Adolfo Espejo-Serna, Nancy MartInez-Correa, and Ana Rosa Lopez-Ferrari
Four nomenclatural changes in Viola (Violaceae)
R. John Little and Landon E. McKinney
vation of Vanilla (Orchidaceae) in Amazonian wetlands
of Madre de Dios, Peru
Ethan Householder, John Janovec, Angel Bala
Renan Valega, Helena Maruenda, and Eric Chi
Phylogenetic analyses of the Gaylussacia fro
Michael T. Gajdeczka, Kurt M. Neubig, Waltei
and Kent D. Perkins
Population genetic analysis of Argemone pleiacantha subsp. pinnatisecta (Sacramento Prickly
Poppy, Papaveraceae) and re-evaluation of its taxonomic status
Sandy D. Cervantes, Phil Tonne, Rajanikanth Govindarajulu, Patrick J. Alexander, and
Huinga Maceda, Jason Wel
dosa complex (Ericaceae: Vaccinieae) based on
>. Judd, W. Mark Whitten, Norris H. Williams,
Occurrence of anisophylly and anisoclady within the Am
Donald B. Pratt and Lynn G. Clark
Observations on Buckleya (Thesiaceae) in China
David E. Boufford, Chengxin Fu, Qiu-Yun (Jenny) Xiang, ai
Comments on a revision of Celtis subgenus Mertensia ((
Celtis pallida
James Henrickson
Croton bigbendensis Turner (Euphorbiaceae) revisited
funpeng Zhao
idaceae) and the recognit
t of Beautempsia (Capparaceae) a
311
FLORISTICS, ECOLOGY, AND CONSERVATION
Flourensia ilicifolia (Compositae: Heliantheae), nuevo registro para Durango y segunda zona
de distribucibn para la especie
M. Socorro GonzAlez-Elizondo, I. Lorena LOpez-Enriquez, Jose A. Villarreal Quintanilla,
JovAn Aleman Medrano, Jaime Sanchez Salas
Diversity and abundance of orchids in a Peruvian cloud forest
Rebecca Repasky Luke, John P. Janovec, Eric Christenson, John E. Pinder III, and Keri McNew Barfield
The ferns and lycophytes of a montane tropical forest in southern Bahia, Brazil
Fernando B. Matos, AndrE M. Amorim, and Paulo H. Labiak
Carex oklahomensis (Cyperaceae) new to Alabama, Georgia, and Louisiana, and additional
records for Mississippi
Charles T. Bryson and Paul E. Rothrock
Plants new to Florida
Richard P. Wunderlin, Bruce F. Hansen, Alan R. Franck, Keith A. Bradley, and John M. Kunzer
Vascular flora and edaphic characteristics of saline prairies in Louisiana
Christopher S. Reid, Patricia L. Faulkner, Michael H. MacRoberts, Barbara R. MacRoberts, and
Marc Bordelon
Floristic composition and potential competitors in Lindera melissifolia (Lauraceae) colonies
in Mississippi with reference to hydrologic regime
Tracy S. Hawkins, Daniel A. SkojacJr., Nathan M. Schiff, and Theodor D. Leininger
Vascular flora of the Old Mulkey Meeting House State Historic Site, Monroe County, Kentucky
Ralph L. Thompson and Ronald L. Jones
Plant communities of selected embayments along the mid- to mid-upper Ohio River floodplain
Joseph S. Ely and Dan K. Evans
Checklist of the vascular plants of Allegheny County, Pennsylvania
Cynthia M. Morton and Loree Speedy
New vascular plant records for South Dakota
Penstemon oklahomensis (Scrophulariaceae) in Texas
Jeffrey N. Mink, Jason R. Singhurst, and Walter C. Holmes
Annotated checklist of the vascular flora of the Jack Gore Baygall and Neches Bottom Units of the
Big Thicket National Preserve, Tyler, Jasper, and Hardin counties, Texas
Larry E. Brown, Barbara R. MacRoberts, Michael H. MacRoberts, Warren W. Pruess,
I. Sandra Elsik, and Stanley Jones
The vascular plants of Mowotony Prairie: a small remnant coastal grassland in Brazoria County, Texas
D.J. Rosen
The vascular flora of Kerr Wildlife Management Area, Kerr County, Texas
Jason R. Singhurst, Laura L. Hansen, Jeffrey N. Mink, Bill Armstrong, Donnie Frels Jr., and
Annotated checklist of the vascular plants of Fort Hood, Texas
Laura L. Hansen
First report of Persicaria hispida (Polygonaceae) from North America north of Mexico (Texas)
Daniel E. Atha and William Carr
313
317
333
347
349
381
391
435
467
471
473
489
497
523
559
12, 32, 50, 68, 74, 92, 116, 212, 214, 216, 224, 244, 270, 280, 294, 302, 308,
AGAL1NIS FLEXICAULIS SP. NOV. (OROBANCHACEAE: LAM1ALES),
A NEW SPECIES FROM NORTHEAST FLORIDA
John F. Hays
t Johns River Water Management District
7775 Baymeadows Way, Suite 102
Jacksonville, Florida 32256, USA
RESUMEN
Field work in preparation for the author’s forthcoming treatment of Agalinis Raf. in volume 17 of the Flora of
North America, in collaboration with Judith Canne-Hilliker of the University of Guelph, Ontario, has resulted
in the discovery of an unknown species of Agalinis.
Annual, hemiparasitic herb, roots fibrous, haustoria present, host plant(s) unknown, 2.5-9.0 dm tall,
often tinged purplish along upper surfaces of stem faces, remaining greenish when promptly pressed, turning
purplish-maroon if allowed to air dry. Stems spreading-ascending, single from the base, the lower branches
on larger plants decumbent-ascending, to 6 dm. Stem sub-terete below, becoming strongly four-angled
above, stem faces glabrous, callous angles glabrous to minutely scabridulous, rounded, trichomes antrorse,
but both perpendicular and retrorse trichomes occasionally present. Leaves opposite, often sub-opposite
or alternate on ultimate branches, spreading to spreading-ascending above. Axillary fascicles absent. The
leaves at mid-stem 7-13 mm long, linear to linear-spatulate (the lowermost leaves sometimes narrowly el-
liptic), 0.8-1. 5 mm wide, stiff but fleshy, u-shaped in cross section, the upper surface slightly concave when
fresh, becoming involute or folded upon drying, tips obtuse to acutish, silicified. Upper surfaces of leaves and
margins scabrous with silicified trichomes, noticeably so upon drying, lower surfaces glabrous save some
trichomes on midveins. Inflorescence indeterminate, paniculate-racemose, racemes and panicles remote,
panicles sometimes as much as 5 dm removed from the main stem on larger plants. Flowering branches
with 3 to 8 floriferous nodes, ultimate flowers solitary, not paired (though often appearing as such). Pedicels
slender, clavate, terete, spreading, glabrous, often two-toned, purplish-maroon above and green below, 4-12
mm at anthesis, to 20(±22) mm in fruit, lowermost pedicels of raceme decidedly longer than those above
in flower and fruit. The pedicel indistinct from base of calyx in flower. Bracts of ultimate, fully-developed
flowers 0.5-4 mm long, linear-spathulate, ascending-appressed with the pedicel. Calyx at anthesis long
campanulate to campanulate-turbinate, glabrous, tube 2. 2-2.7 mm long, often two-toned, upper surface
purplish-maroon, lower surface green; lobes 5, minute, acute, deltoid to subulate, 0.2-0.5 mm long, pu-
J. Bet. Res. Inst. Texas 4(1): 1 - <
Journal of the Botanical Research Institute of Texas 4(1)
rolla lobes spreading, shorter than tube, surface of all lobes glabrous externally and internally above the
sinuses. The two upper lobes 3-4 mm long, ±3 mm wide, apex truncate, emarginate, or sometimes erose-
cordate, strongly pink-ciliate, often reflexed above the middle of the lobes, lanose across the base of lobes
with pink trichomes to 1.2 nun long. The three lower lobes 4-5 mm long, 2. 5-3.0 mm wide, apex truncate,
emarginate, or sometimes erose-cordate, ciliate, the central lobe with shorter cilia. Stamens didynamous,
abaxial filaments 3 9-4.2 mm long, pubescent basally, adaxial fliaments 2. 4-3.0 mm long, glabrous. An-
thers 4, apices rounded, bases mucronate to apiculate, white when fresh, valvular surfaces lanose, abaxial
anthers elliptic, held mostly perpendicular to the filaments, apices often connivent by a dense entanglement
of trichomes 1.2-1. 8 mm long, adaxial anthers linear-oblong, held parallel to filaments, 1. 5-2.0 mm long.
Style 5-7 mm long, exsert, glabrous, white-translucent, stigma ±1 mm long, white, papillate. Capsule long
obovoid to oblongish, broadly 4-sided, ± as wide as broad, 3. 8-4.5 mm long at maturity, apex golden brown
when mature, greenish-tan beneath calyx tube, apex truncate, appearing broad-shouldered along margins
of callous ridge, base broadly cuneate, conspicuously-reticulate veined on the upper one-third of mature
capsules, with scattered, minute, silica-filled trichomes often on surface of the capsule above the adhering
calyx; callous ridge on capsule perpendicular to the septa, extending from the apex transversely downward
to the base along opposite sides. Seeds triangular to quadrangular, lustrous golden-brown, minute, ±0.5
mm long, testa reticulate. Chromosome number unknown.
Flowering from late September through early November, fruiting into late November. Occurring in
mesic to hydric prairies or longleaf pine savannas dominated by wiregrass ( Aristida spp.) and the remnants
of such communities (Table 1). Currently known only from five populations in southern Bradford County,
Florida. Apparently endemic. It is, however, to be expected in the following portions of adjacent counties:
the northern one-third of Alachua County, SW Clay County, NW Putnum County, and the southern half
of Union County.
Additional collections examined: FLORIDA, all Bradford County: Type locality, 17 Oct 2009, Hays 3453 (FLAS, USF); CR 18, 2.1 mi
W of jet. of CR18 & SR100 on south side of road in remnant moist prairie/savanna, ca. 29‘>51,9.30"N, 82'5'57.04"W, 35 + plants in flower
+ plants in flower and early fruit on S side of road in longleaf pine savanna, 24 Oct 2009, Hays 3455 (BRIT, MO, USF); CR 221, 0.3 mi S
of jet. of HY301 and CR221, ca. 29°53'12.98"N, 82°8'17.06"W, 50 + plants in late flower and fruit along margin of regenerating longleaf
pine savanna, 31 Oct 2009, Hays 3456 (FLAS, MO); SR100 west, 7.6 mi W of jet. of SR100 & HY301, and 0.8 mi E of jet. of CR 235 and
SR100, ca. 29°58'44.41"N, 82°14'0.13"W, 60 + plants in fruit in remnant prairie vegetation beside regenerating longleaf pine plantation,
S side of road, 1 Nov 2009, Hays 3457 (BRIT, FLAS, MO, NY, USF).
Common name. — Hampton False Foxglove; Sprawling False Foxglove. The common names are taken from the
small town around which most of the original collections were made, and the obvious decumbent, flexous
habit of larger branches on mature specimens.
Etymology.— -The specific epithet “flexicaulis" was chosen to aptly describe the sprawling, decumbent
branches so characteristic of the secondary and tertiary branches of the larger specimens of this species.
Agalinis flexicaulis is yet another elusive, but distinctive species of North American Agalinis that seems
to melt into its surrounding habitat (Hays 2002). This species is unique among North American members
of the genus because of its weakly ascending main stem, which becomes more drooping and lax as it ma-
tures, as well as the remote panicles (Canne-Hilliker & Kampny 1991; Pennell 1929). The secondary stems
(especially the lower ones) are often 4 dm or more long, decumbent-ascending as the larger branches often
droop toward the ground in the middle and then slightly arch upward toward their tips, supported by sur-
rounding vegetation (mostly grasses) to expose the flowering branches. As mature capsules develop, the
main stem leans toward one side and the long secondary flowering branches are near, on the ground, or
entangled within surrounding vegetation, which often grows over the branches as the season progresses.
Latent flowers often appear to arise from the ground on severed stems. The plants often need to be pulled
gently, branch by branch, from the vegetation supporting them to avoid breaking the stems or losing leaves,
flowers, or fruit. It is also common to see plants with decidedly secund branching, although not exclusively so.
Bradford County is part of the Northern Highlands physiographic province, with much of the county
rarity of A. jlexicaulis. The author has studied Agalinis for over 12 years in Florida alone and was more than
surprised at his “discovery” within a well known botanical area less than 20 miles from the University of
Florida in Gainesville. This emphasizes not only the need for an additional survey for Agalinis jlexicaulis, but
underscores the fact that there are additional plant species to be described within the remaining intricate
Below is a key to the long-pediceled species of Agalinis to aid in identification.
KEY TO THE LONG-PEDICELED SPECIES OF AGAUN1S IN NORTHEAST FLORIDA
. Leaves alternate, reddish-purple, strongly fascicled and appearing whorled, succulent; plants of xeric
. Leaves opposite (sometimes sub-opposite on the flowering branches), green to stramineous, fascicles absent
or rarely developed; plants of various habitats.
2. Plants perennial from a slender root stock, up 1 .5 m tall; stems glabrous, often leaning; leaves and pedicels
strongly ascending; plants of saturated or inundated soils; dried plants turning black
2. Plants annual, roots fibrous; the stems less than 1 m tall; leaves and pedicels various; plants of xeric to wet
soils (but not in permanent bodies of water); plants drying stramineous to dark green, but not drying
black.
3. Stems strongly 4-angled, winged along principal stems; leaves linear to linear-spatulate, 0.6-2.5 cm
long, 05-1 .5 mm wide.
4. Main stem erect, the branches stiffly ascending, puberulent to scabrous throughout; stems stramin-
eous, drying the same; leaves 1-2.5 cm long, 0.5-1 .5 mm wide, erect-ascending; corolla 12-1 7 mm
long, pink, with two yellow guidelines present (sometimes faint), with purple spots in the
throat; capsule globose, rounded at summit, surface smooth at apex A. <
4. Main stem weakly ascending, glabrous-puberulent, the lowerm
ing upon or growing within surrounding vegetation, the secondary branches up to 5 dn
stems green, suffused with purplish-maroon, drying green to purplish-green; leaves 0,7-1
0.8 to 1 .4 mm wide, spreading; corolla 8-1 1 mm long, rose-purple, yellow guidelines ab
sionally with purplish spots in the throat; capsule obovoid to oblong, the apex truncate
with distinctively raised veins on the calyx and capsule, giving the appearance of sma
clearly 4-angled; leaves filiform, li
a glabrous within; pedi-
3. Stems sub-terete to striate-angled, m
0.6-3.5cm long, 0.5-3.5 mm wide.
5. Two upper lobes of the corolla flattened or arched over the stamens; core
cels 10-30 mm long at anthesis, at least some of the pedicels abruptly up
more noticeable when dried.
6. Leaves ascending-spreading, 1 -4 mm wide, linear to narrowly lanceolate; two upper lobes of the
corolla arching over the stamens, about equaling the lower lobes in length; corolla glabrous ex-
ternally; bracts approaching or equaling pedicels in length; plants erect, often profusely branched,
appearing bushy, to 8 dm tall ■■ A.
6. Leaves widely spreading, horizontal to the stem, or slightly reflexed at mid-stem, filiform, 0.5-1. 0
mm wide; two upper lobes of the corolla flattened over the stamens, half as long as the three
lower lobes; bracts much shorter than pedicels; plants sprawling, widely branched, rarely over 3
5. Two upper lobes of the corolla reflexed; corolla pubescent witf
lobes; pedicels 3-30 mm long at anthesis, not abruptly upturn
7. Corolla 1 4-1 8 mm long, pedicels on v\
is long as the subtending bracts _
cross the bases of the two upper
n long, slender,
ACKNOWLEDGMENTS
The author wishes to thank Judith Canne-Hilliker for her critique of the manuscript, as well as for her
continued, invaluable insights in to world of Agalinis. In addition, the author also thanks Noel Holmgren
for his review and insights, Linny Heagy for yet another excellent line-drawing of Agalinis, and Guy Nesom
for the Latin diagnoses.
A NEW LOMATIUM (APIACEAE)
FROM THE OCHOCO MOUNTAINS OF CENTRAL OREGON
Richard Helliwell
L Forest Service, Umpqua National Forest
2900 Northwest Stewart Parkway
Roseburg, Oregon 97471, USA
rhelliwell@fs.fed.us
ABSTRACT
RESUMEN
INTRODUCTION
A new species of Lomatium Raf. with decidedly thick glaucous foliage and broad overlapping oval to ovate
leaflets was discovered in 1994 on the Ochoco National Forest in the Ochoco Mountains of Crook County
in north-central Oregon. The initial site consisted of approximately 200 plants but since then six additional,
much larger, populations have been documented on nearby Bureau of Land Management administered land.
All known populations are within eleven air kilometers of each other in an isolated area that is difficult to
access in early spring when the plants are in flower due to persistent snowdrifts in the canyon bottoms.
Lomatium ochocense Helliwell & Constance ex Helliwell, sp. nov. (Fig. 1). Type: UNITED STATES. Oregon. Crook
Co.: Long Ridge, Ochoco Mountains, T16S, R21E, S11NE1/4, 4540 ft (1360 m), 7 Apr 1995, Richard Helliwell 2 313 (holotype: UC;
Herba perennis humilis acaulescens, glabra et glauca, caule subterraneo (pseudoscapo); radix ovoidea-globosa irregulariter, pagina furva;
Low, acaulescent, glabrous and glaucous perennial herbs, 4-8 cm tall, with a prominent pseudoscape 2-6
cm long. Roots irregularly ovoid-globose, 1-3 cm in diameter, black. Leaves usually 2-5 (the outermost
usually much reduced), ascending, pinnate, oblong-oval to triangular-ovate, 1-6.5 cm long, 0.6-2.5 cm
broad; petioles up to 3 cm long, abruptly and narrowly sheathing basally; leaflets pinnate or pinnatifid, 3 or
5 (-7), overlapping, oblong to ovate, fleshy, glaucous and blue-green, glabrous, lower-most pair often remote
and 3 terminal ones ± confluent basally, lobes oblong to ovate, acute to rounded apically. Inflorescence
compact compound umbels; peduncles 1-2, 4-6 cm long, spreading horizontally; involucre 0; rays 2-6,
unequal, spreading; fertile pedicels 1-6, 5-10 mm long. Umbellets 10-15-flowered; mature pedicels ca 4
mm long; involucel dimidiate; bractlets ca 8, lanceolate to ovate, 2.5-3 mm long, 0.6-1 mm broad. Flowers
perfect, andromonoecious; sepals triangular-lanceolate, green, 0.1-0.2 mm long or absent; petals yellow,
oval-ovate, abruptly acuminate, 1-2 mm long; stamens maturing unequally, filaments terete, 0.2-1.7 mm
long, anthers yellow, turning white, 0.4-0.7 mm long; styles 0.8-1.2 mm long, glabrous, yellowish-green;
ovary glabrous, green. Fruit elliptic, 5-8 mm long, 3.5-4.5 mm broad, glabrous; dorsal ribs filiform, lateral
ribs narrowly thin-winged, narrower than the body; vitae small, 2-3 in intervals, usually 4 on commisure;
seed face nearly plane.
J. Bot. Res. Inst. Texas 4(1): 7 -11. 2010
Journal of the Botanical Research Institute of Texas 4(1)
dorsal and ventral view. h. Fruit cross-section.
2306 (UO immS
Relationship. — The massively thickened root of Lomatium ochocense clearly places it within the tuberous
group of lomatium, an informal group initially recognized by Coulter and Rose (1888), retained by Marcus
E. Jones (1908) as sect. Cous and most recently monographed by Schlessman (1984). Subsequent genetic
work by Solstis and Novak (1997) have determined this group to be polyphyletic having evolved from at
least four lineages. Although L. ochocense is nearly as diverse genetically as the common L. cous (S. Watson)
J.M. Coult. & Rose and L. macrocarpum (Nutt, ex Torr. & A. Gray) J.M. Coult. & Rose (Gitzendanner &
Soltis 2001), its phylogenetic relationships within the genus remain to be determined. Although not mono-
phyletic, this remains a useful artificial group within this large and diverse genus since the tuberous roots
that are glaucous blue-green. Among non-tuberous sp
been noted. This subalpine endemic of the Wallowa 1
foliage however it is distinctly caulescent, has mostlj
ortheast Oregon also has pale, glaucous
s, and arises from a multicipital, woody
appear
DISTRIBUTION, ECOLOGY, AND PHENOLOGY
on lava tablelands known locally as “scablands”. These scablands are dissected by deep V-shaped canyons
that are timbered with Pirns ponderosa P. Lawson & C. Lawson and Pseudotsuga menziesii (Mirb.) Franco
with Juniperus occidentals Hook, abundant along the margins of timber and scattered sporadically across
the scabland wherever fracturing in the basalt or lenses of deeper soil allow it. Soils on the scablands are
classified as loamy-skeletal, mixed, superactive, frigid Lithic Argixerolls in the Tweener Series http://www2.
ftw.nrcs.usda.gov/osd/dat/T/TWEENER.html. Bedrock is hard, highly fractured basalt of the Picture Gorge
formation of the Columbia River group (Baldwin 1964). Lomatium ochocense is restricted to areas where there
typically saturated through the winter but dry out rapidly by late spring. Elevations of known populations
are between 1300 and 1400 m.
Lomatium ochocense occurs in plant communities dominated by Artemisia rigida (Nutt) A. Gray and Poa
secunda J. Presl. Other associates include Allium macrum S. Watson, Lewisia rediviva Pursh, Lomatium cous,
Lomatium hendersonii (J.M. Coult. & Rose) J.M. Coult. & Rose, Lomatium macrocarpum, Achnatherum hender-
sonii (Vasey) Barkworth, and Sedum stenopetalum Pursh. The Lomatium species provide the most conspicuous
cover across these scablands in early spring with L. ochocense flowering from mid-March to late April.
cm of L. ochocense, there are at least eight other species of Lomatium in the immediate vicinity. Similar cases
of congeners growing together in close proximity have been called “genus communities” (Diver 1936). The
ability of similar species to coexist in nearly identical habitats without apparent production of hybrids affords
testimony to their reproductive isolation. Lomatium ochocense, L. cous, and L. hendersonii are united by their
massive corm-like roots, low acaulescent habit, and yellow flowers but clearly maintain their distinguishing
Across these open scablands the Lomatium species have segregated themselves by often subtle microsite
characteristics. Lomatium cous is the most common of the species. It is present on every scabland opening and,
together with L. macrocarpum, seems to have the least specific habitat requirements. Nevertheless, despite
considerable overlap, these two species exhibit significant spatial segregation. In sympatric populations in
this area, L. cous usually dominates the broad tops of these tablelands while L. macrocarpum occupies the
gently sloping sides. Lomatium ochocense displaces L. cous in many areas where there is exposed bedrock in the
form of parallel ripples although other areas of apparently identical habitat in the vicinity lack L. ochocense.
This rock feature appears to be unique to the narrow range of L. ochocense and is strikingly evident on aerial
photographs. The soil depth where L. ochocense occurs is typically 5-9 cm with the root seated in fractures
at the bedrock/soil interface. Lomatium hendersonii is largely confined to narrow bands of unconsolidated
rock called “rock stripes” that commonly lie in parallel formation across the top or side of scablands. In
many cases the root is seated in a substrate of loose rock with little or no soil.
Three other Lomatium species were observed sharing scabland habitat in the immediate vicinity although
they were not intermingled with L. ochocense. In slightly deeper or less rocky soil L. piperi J.M. Coult. &
Rose occurs and in slightly moister areas, particularly those with depressional microrelief, L. leptocarpum
(Torr. & A. Gray) J.M. Coult & Rose replaces the other Lomatium species. A short-statured ecotype of L.
nudicaule (Pursh) J.M. Coult. & Rose is present in slightly deeper soil around the perimeter of the habitat
Additional Lomatium species that were noted in other than scabland habitat in the vicinity include L.
grayi (J.M . Coult. & Rose) J.M. Coult. & Rose on vertical rock outcrops; L. dissectum (Nutt. & Torr. & A.
Gray) Mathias & Constance in talus; L. triternatum (Pursh) J.M. Coult. & Rose under a timbered canopy;
L. vaginatum J.M. Coult. & Rose in an inclusion of heavy red soil; and L. donnellii (J.M. Coult. & Rose) J.M.
Coult. & Rose in deeper, disturbed soils.
The scabland Lomatium species are also slightly separated by phenology although all flower early. Lo- j
matium hendersonii and L. piperi flower by mid-February and perhaps even earlier. Lomatium cous appears to
slightly precede L. ochocense but both were in flower by the second week of March in 1995. Lomatium mac-
rocarpum and L. leptocarpum began flowering by early April in 1995 leaving only L. nudicaule yet to flower.
Note that these dates differ significantly from published dates in regional floras (e.g., Cronquist 1961).
It is notable that thickening of the taproot is a consistent feature of the species that are confined to open
lithosols. Lomatium ochocense, L. cous, and L. hendersonii all have large irregularly shaped tuberous taproots.
Lomatium piperi has a globose taproot while L. leptocarupum and, to a lesser extent, L. macrocarpum both
have tuberous thickenings unevenly distributed along their taproot. Lomatium nudicaule, which is more com-
monly found on deeper soils throughout its range, has a stout evenly thickened taproot. The other species
have roots that are much less thickened.
Etymology. — For the Ochoco Mountains.
Conservation status. — Although Lomatium ochocense appears to be a narrow endemic, the bulk of the
known populations lie within a BLM wilderness study area, population sizes are mostly very large, and the
potential threats to this species are few. One occurrence is adjacent to a popular recreation area but the
rocky nature of the habitat minimizes the potential for adverse impacts from recreationalists. The entire
area lies within cattle grazing allotments but the specific areas where L. ochocense grow receive insignificant
livestock use due to the inherently sparse forage. Several plants were observed to have been browsed, appar-
ently by mule deer that browse the scablands in early spring. Although the species had not yet been validly
published, the U.S. Fish and Wildlife Service formally proposed that L. ochocense be considered a “species
of concern” to ensure its long-term management in August of 2009. Likewise, the plant is considered an
“imperiled plant” by the Forest Service (http://www.fs.fed.us/wildflowers/rareplants/profiles/critically_im-
periled/lomatium_ochocense/index.shtml), and the name is found on the USDA PLANTS database (http://
plants.usda.gov/java/profile?symbol=LOOC2).
Comments. — It should be noted that the name Lomatium ochocense was first proposed in 1995 soon after
the species was discovered. With the assistance of the late Lincoln Constance (1909-2001) a manuscript
was prepared and state and federal agencies notified of this rare species. Over the intervening years L. ocho-
cense has been the subject of genetic research resulting in the name being published, without description or
designation of type, as a nomen nudum in Gitzendanner and Soltis (2001). The scientific name proposed by
Constance and myself, the detailed description, and the authorship of the name is retained in recognition
of Dr. Constance contribution to this effort. However, as he cannot defend the text, I assume responsibility
KEY TO THE TUBEROUS SPECIES OF LOMATIUM IN OREGON
;s thick, glaucous, blue-green, glabrous; leaflets broadly overlapping, oval-ovat
. Leaves often glabrous but never also thick, glaucous and blue-green, leaflets typically remote except some-
times distally, variously shaped.
2. Petals white, anthers and stylepodia purple.
3. Mericarps minutely granular-roughened or bristly, the papillae 0.05 mm long, visible at 1 5x; acaulescent
BOOK REVIEW
Stewart McPherson. 2008 Glistening Carnivores: The Sticky-Leaved Insect-Eating Plants. (ISBN 978-
0-9558918-1-6, hbk.). Redfem Natural History Productions, 61 Lake Drive, Hamworthy, Poole, Dorset
BH15 4LR, England, United Kingdom. (Orders: www.redfemnaturalhistory.com, sales@redfemnatu-
ralhistory.com, +44-1202-686585). £29.99 (US $46.15), 392 pp„ 271 color photos, 1 b&w figure, 7
maps, 6 1/4" x 9 1/4".
E 4“ Street, Fort Worth, Texas 76102-4025, U.S.A.
From the publisher: “The seven genera of sticky-leaved inse
them. Each produces shimmering leaves lined with glisl
J. Bot. Res. Inst. Texas 4<1): 12. 2010
POTENTILLA ULIGINOSA (ROSACEAE: ROSOIDEAE),
NEW PRESUMED EXTINCT SPECIES FROM SONOMA COUNTY, CALIFORNIA
Barry C. Johnston
USDA Forest Service
216 N. Colorado Street
Gunnison, Colorado 81230-2197, U.SA
bcjohnston@fs.fed.us
Barbara Ertter
Johnston and Ertter, Potentilla uliginosa, a new species from California
16 Journal of the Botanical Research Institute of Texas 4(1)
sp. (nominal P hickmanii)” in the event that living plants are rediscovered or “recruited from a (hopefully)
dormant persistent seed bank.” It is also possible that additional populations remain to be discovered in the
mountains of northwestern California. Tantalizing in this regard is a 1929 collection ( Applegate 5735, UQ
identified as P. millefolia Rydb. from Deer Creek near Selma, Josephine County, Oregon, which approaches
P uliginosa in stature and occurs outside the otherwise known range of P. millefolia (i. e., meadows from
south-central Oregon to the east side of the Sierra Nevada in California and Nevada).
Cunningham Marsh itself is one of three perennial wetland complexes in southern Sonoma County
that share a suite of disjunct, endemic, and regionally rare wetland species (Baye 2005; Best et al. 1996;
Rubtzoff 1953), the other two being Pitkin Marsh and the lesser known Perry Marsh. All three are rela-
tively oligotrophic (acidic and nutrient-poor) with permanent fen-like water systems (Baye 2005). As such,
they harbor numerous disjunct species more characteristic of northern bog-like habitats, leading Rubtzoff
(1953) to interpret the marshes as relictual floristic “islands.” The main “poster-child” plant is the federaUy
endangered Pitkin Marsh lily ( Lilium pardalinum Kellogg subsp. pitkinense (Beane & Vollmer) M.W. Skin-
ner), which is endemic to Pitkin and Cunningham marshes. While more conservation attention has been
paid to Pitkin Marsh, which has the largest number of significant species, a conservation easement was
established in Cunningham Marsh in 1998, in response to a proposed subdivision on adjacent lands (Baye
2005). Whether the easement site includes the historical locality of P. uliginosa, or whether that locality has
already been converted to agricultural uses or subdivisions, has yet to be determined.
Plant species that are presumed extinct are more commonly described in the paleobotanical literature,
and the unavailability of living or recently collected material of Potentilla uliginosa has hampered the kind
of studies (e.g., molecular analysis, common garden experiments) that would otherwise be appropriate for
taxonomic recognition of morphologically anomalous plants known from single populations. The morpho-
logical anomalies are nevertheless sufficient in their own right to justify recognition of the Cunningham
Marsh cinquefoil as a distinct species. This action clarifies conservation efforts not only of P. uliginosa but
also P. hickmanii, which is endangered under both federal regulations (Clark 1998) and California state
regulations (California 2004). In particular, formal recognition of P. uliginosa supports the strong recom-
mendation in the Cunningham Marsh conservation easement area management plan (Baye 2005) against
“nominal reintroduction” efforts using P. hickmanii from San Mateo or Monterey counties.
Relationships. Potentilla uliginosa is a distinctive species that belongs to a morphologically and ecogeo-
graphically defined species cluster in Potentilla sect. Multijugae that also includes P. hickmanii, P. millefolia, P.
multijuga, and P. plattensis. As already noted, the new species has heretofore been included within P. hickmanii,
differing most notably in its larger dimensions (e. g„ stems 2.5-5.S dm vs. 1.5-2.5 dm). The leaflets of P.
uliginosa are more comparable to those of P. millefolia and P plattensis in being divided 70-90% to the midrib
into (3-)7-10(-15) narrowly oblanceolate to linear segments (vs. divided only 50-65% to midrib into 2-5
somewhat broader segments in P. hickmanii). In contrast to the irregularly pinnate leaflets of P. uliginosa, those
of P. millefolia tend to be palmately lobed, while those of P. plattensis are more regularly pinnately lobed, with
the terminal leaflet relatively distinct from (vs. confluent with) the lateral leaflets. Plants of P. uliginosa tend
to be larger (2.5-5.5 dm vs. 0.5-2. 5 dm) than either P millefolia or P plattensis, and are rather in the size
range of P. multijuga (2. 0-3.5 dm). Potentilla multijuga also differs from P uliginosa in having fewer leaflets
(7-17 vs. [13— ]17-21[— 251) that are more shallowly incised (only 30-40% to the midrib). Likewise presumed
extinct, P. multijuga is known only from a handful of collections in the 1890’s from Ballona Marsh near the
current site of Los Angeles International Airport in Los Angeles County, California. The name P. multijuga
has been proposed for conservation with a conserved type (Enter & Reveal 2008), since the original type
has proven to be Horkelia cuneata Lindl. (Sojak 1996).
As previously hypothesized (Enter 1993b), the species cluster to which Potentilla uliginosa belongs has
evidently radiated among seasonally or permanently saturated wetlands of western North America, resulting
in numerous isolated populations. Some of these populations have speciated into narrow mdprnic* includ-
ing species that are federally and state endangered (P. hickmanii), candidates for federal listing (P basaltica
Johnston and Ertter, Potentilla uliginosa, a new species from California 17
Tiehm & Ertter), or presumed extinct (P multijuga, P. uliginosa). Several other anomalous populations may
prove distinct upon further analysis, notably an outlier of P. millefolia in Reese River Valley in Lander County,
Nevada. In our understanding, the recently described P diversifolia Lehm. var. madsenii S. L. Welsh & N.
D. Atwood, reported from a single collection from Kane County, Utah (Welsh et al. 2008), also belongs to
this complex as a variant of P plattensis ; it is currently on the “Need Data List” for the Rare Plants of Utah
(Utah Native Plant Society Rare Plant Committee 2009). Molecular analyses of the complex are currently
underway by Matt Guilliams, a graduate student at the University of California at Berkeley.
ACKNOWLEDGMENTS
Grateful thanks to William A. Weber and the late David J. Rogers for encouragement and guidance to the
senior author; to Charles F. Quibell, who guided the junior author to Cunningham Marsh in 1987 and
provided follow-up information from the field notes of Milo S. Baker; to Deb Trock, for checking label data
and J.T. Howell’s field notes at CAS; to Anita Cholewa, for tracking down Congdon’s specimen at MIN; and
to Jim Shevock, for checking label data of specimens at CAS. Special appreciation to Richard Whitkus, for
bringing our attention to the Cunningham Marsh conservation easement site vegetation management plan,
and for help with and permission to quote from Milo S. Baker’s field notes (housed at the North Coast of
California Herbarium [NCC, formerly ROPAJ of Sonoma State University). The junior author is indebted to
the staff of the University of California Botanical Garden for maintaining her living collections, including P
hickmanii and P. millefolia. The manuscript benefited greatly from the constructive reviews by Arnold Gerry)
Tiehm and Richard Lis.
A portion of this paper was part of the senior author’s thesis. Ongoing support for research on Potentilla
and other Rosaceae from the Lawrence R. Heckard Endowment Fund of the Jepson Herbarium to the junior
author is gratefully acknowledged.
REFERENCES
Baye, P. 2005. Vegetation management plan: California Department of Fish and Game 'Cunningham Marsh'
conservation easement site, Sonoma County, California. California Native Plant Society, Milo Baker Chapter,
(available at http://www.cnpsmb.org/)
Best, C, J.T. Howell, W. Knight, I. Knight, and M. Wells. 1 996. A flora of Sonoma County. California Native Plant Society,
Sacramento.
Caufornia Department of Fish and Game 2005. The status of rare, threatened, and endangered plants and animals
of California 2000-2004. Published online at http://www.dfg.ca.gov/wildlife/nongame/t_e_spp/new_tejpt.
Clark, J.R. 1998. Endangered and threatened wildlife and plants: Final rule listing five plants from Monterey
county, CA, as endangered or threatened. Fed. Reg. 63(155):4310CM31 16.
Ertter B. 1993a. Potentilla. In: J.C. Hickman, ed. The Jepson manual: higher plants of California. University of
California Press, Berkeley. Pp. 964-969.
Ertter B. 1993b. The puzzling potentillas. Fremontia 21(1)25-29.
Ertter B. and J.L Reveal 2008. Proposal to conserve the name Potentilla multijuga (Rosaceae) with a conserved
type. Taxon 57:312-313.
Jepson, W.L. 1 936. A flora of California. Vol. II. Capparidaceae to Cornaceae. Associated Students Store, University
of California, Berkeley.
Johnston, B.C. 1 980. Studies of population variability leading to a new classification of Potentilla Sect. Multijugae
(Rosaceae). Ph. D. Thesis, University of Colorado, Department of Environmental, Population, and Organismic
Biology, Boulder, CO.
Munz, PA 1 959. A California flora. University of California Press, Berkeley.
Rubtzoff, R 1 953. A phytogeographic analysis of the Pitkin Marsh. Wasmann J. Biol. 11:1 29-2 1 9.
SojAk, J. 1996. Notes on Potentilla (Rosaceae). XIV. Type specimens in the Lehmann herbarium. Preslia 68:
Utah Native Plant Society Rare Plant Committee 2009. 2009 UNPS Rare plants of Utah list-IV. Need Data List. Sego
Lily 32(6):1 7.
Welsh, S.L, N.D. Atwood, S. Goodrich, and LC. Higgins. 2008. A Utah flora, Fourth Edition. Brigham Young University,
Provo, UT.
TWO NEW TAXA OF SCUTELLARIA SECTION RESINOSA (LAMIACEAE)
FROM NORTHERN ARIZONA
S.L. Rhodes and TJ. Ayers
Department of Biological Sciences
P.O. Box 5640, Northern Arizona University
Flagstaff, Arizona 8601 1-5640, USA
slr3 14@gmail.com; tina.ayers@nau.edu
RESUMEN
Scutellaria sect. Resinosa Epling is well represented in the southwestern U.S. and throughout northern and
central mainland Mexico. Epling (1942) defined sect. Resinosa based on the presence of entire, oval or rounded
leaves, violet-blue corollas, corolla tube and galea less than 22 mm long, calyx compressed into an erect and
transverse scutellum, and subglobose mericarps. Turner’s (1994) revision of sect. Resinosa provided evidence
for a broadly distributed S. potosina Brandegee, which occurs from the Sierra Madre Oriental, Mexico north
and west to central Arizona. The species was comprised of two subspecies: subsp. potosina, a morphologically
diverse group of six varieties; and subsp. platyphylla Epling, represented by the type variety only. Scutellaria
potosina was separated from the other 18 species in sect. Resinosa by the presence of tessellate mericarps
and a pubescence of glandular and eglandular hairs. Turner described four new varieties of S. potosina in
his treatment of Scutellaria sect. Resinosa based on differences in habit, leaf morphology, and vestiture.
Scutellaria potosina var. platyphylla was later elevated to specific status by Turner and Delprete (1996) based
on mericarp ornamentation, foliage vestiture, and distribution.
Collections made in Yavapai and Coconino counties, Arizona, previously identified as S. potosina, ap-
peared to be intermediate between S. potosina and S. platyphylla (Epling) B.L. Turner & P. Delprete. These
collections represent two different entities from four localities on the eastern rim of the Kaibab Plateau in
Coconino County and three localities in the Upper Verde River drainage in Yavapai County. Both entities have
a distinctive short stature and prolific branching from the root crown not seen in either S. potosina or S. platy-
phylla. However, they appeared to differ significantly in vestiture when viewed under a dissecting microscope.
All of thel2 recent collections have been made since 1993 except for one specimen from the Upper
Verde watershed collected in 1984 ( Hodgson 2887, DES). This specimen was not included by Turner in his
1994 taxonomic treatment nor his subsequent morphological studies (Turner & Delprete 1996).
A SEM study was undertaken to visualize micromorphological characters needed to verify the taxo-
nomic rank of the new collections and their placement within sect. Resinosa. Leaf vestiture and mericarp
size, color, and shape of the two new taxa fit well within the sect. Resinosa, but a SEM study was needed to
Journal of the Botanical Research Institute of Texas 4(1)
visualize individual hair types and papillae on the mericarps, both important species level characteristics!
in Scutellaria (Olmstead 1990; Turner 1996).
The Kaibab Plateau populations are unique in that the leaves lack glandular and eglandular hairs alto- 1
gether but are densely covered with sessile glands on both the upper and lower surfaces. Leaves of known
varieties of Scutellaria potosina in the U.S. and northern Mexico also lack glandular hairs but var. tessellata
(Epiing) B.L. Turner has scattered eglandular hairs and a few sessile glands on the upper and lower leaf
surfaces, while var. grahamiana B.L. Turner is nearly glabrous, with few sessile glands and eglandular hairs.
The mericarp surface-sculpturing of the Kaibab Plateau populations is similar morphologically to var. tes-
sellata. Both have short flattened papillae with epidermal cells that have conical or domed apices without
apical callosities (Fig. 1). The holotype and one topotype of var. grahamiana lacked mericarps and none was
found during field work conducted in 2000. Var. grahamiana may require outcrossing, lack its pollinator, or i
may possibly represent a sterile hybrid of unknown origin. Based upon field observations and morphological
study, including SEM of leaves and mericarps (Table 1), the Kaibab Plateau population is here described as
a new variety of S. potosina, differing chiefly in habit and a vestiture of sessile glands only.
The Upper Verde populations are densely covered with sessile glands and glandular hairs similar to S.
platyphylla, although those found on S. platyphylla are long, ca. 50 microns, while those on the Upper Verde
plants are only ca. 20 microns (Fig. 1, C and D). Mericarp surface sculpturing of the Upper Verde plants
also resembles that found in S. platyphylla (Turner 1996). Both have short conical papillae that are domed
at the apex and flask-shaped epidermal cells. The Upper Verde populations have very pronounced apical
callosities, some of which are extended into a narrow elongated or curved tip, like those of S. platyphylla
(Fig. 1, 1 and J). Based upon field observations and the morphological study, including SEM of leaves and
mericarps (Table 1), the Upper Verde populations are here described as a new variety of S. platyphylla, dif-
fering chiefly in habit and a vestiture of dense short eglandular hairs. The recognition of this new taxon
within S. platyphylla, following Turner and Delprete (1996), is preferred until the species boundaries in sect.
R esinosa have been more thoroughly investigated.
Scutellaria potosina '
TAXONOMIC TREATMENT
kaibabensis S. Rhodes & T. Ayers, var. nov. (Fig. 2, A-E; Fig. 3, A-C). Tm U.S.A.
:ast run of the Kaibab Plateau, North Canyon Trail #4, 36°25'04"N 1 12°04T7"W (NAD 27), 2133 m (7000
Perennial rhizomatous herbs, 12-16 cm high. Stems simple, erect or ascending from a woody caudex,
yellow-green, with scattered sessile glands, hirtellous to glabrous, hairs eglandular, retrorse; older stems
persistent. Leaves sessile to subsessile, elliptic to obovate, 9-15 mm long, 5-7 mm wide, ca. 2x as long as
wide, margins entire, apex rounded, adaxial surface glabrous except for dense sessile glands, abaxial surface
with a few scattered short (<10pm) hairs in addition to sessile glands. Flowers axillary in upper stem leaves;
calyx densely glandular; corolla 9-13 mm long, tube diameter at throat 3.9-6.0 mm, outer surface with
sessile glands and short gland-tipped hairs, limb deep violet-blue, throat white with purple spots. Mericarps
1.0-1. 5 mm long, black, with an obscure apical callosity on some papillae.
Plants of dry sandy soil, often near ephemeral drainages; east side of Kaibab Plateau and western
House Rock Valley (Fig. 4); ponderosa pine-white fir, pifion-juniper, and interior chaparral associations
(1250-)1950-2606 m [(4100-)6300-8550 ft]. Flowering May-August.
Scutellaria potosina var. kaibabensis is endemic to the eastern edge of the Kaibab Plateau in northern Arizona.
The closest known populations of S. potosina are of var. grahamiana in southwestern Graham County, which occurs
below 4,000 ft in elevation and var. tessellata in Pima, Cochise and Santa Cruz counties of southern Arizona, which
is found between 3,500 and 5,500 ft. Var. kaibabensis occurs further northwest than any other known populations
in sect. Resinosa, and 250 miles north of the nearest populations of S. potosina. This variety is the sole representative
23
doseup showing vestiture.
Journal of the Botanical Research Institute of Texas 4(1)
found in the Colorado Plateau Floristic area, all other varieties are in the Apachean or Chihuahuan Floristicl
areas (McLaughlin 1986, 1989). The compact growth-form and larger corollas make this taxon particulariji
striking when in flower (Table 1). Variety kaibabensis is densely clothed with sessile glands but is otherwise!
glabrous or possesses just a few short hairs. The lack of eglandular hairs is similar to var. grahamiana, butl
the number and density of sessile glands, and architecture can easily separate the two.
219, on ridgetop; 36*24'35"N, 112e03'28"W (NAD 27) 2606 m (8550 ft), 21 Jun 1999, Rhodes 9954 (AS C); Marble Canyon, South Canyon*
trail, ca. 5 km upstream from the Colorado River 36»29'02"N, 111“54'07"W, 1244 m (4180 ft); 18 May 1996, Stevens s.n. (ASC); Kaibabl
National Forest, Cocks Comb, off FSR 445A, ca. 7 mi W of FSR445, 36°23'37“N, 112*00'41*W (NAD 27) 1920 m (6300 ft); 20 Jun 1998,*
Hodgson 11076 (ASU), 11089 (ASU, DES); North arm of Nankoweap Canyon between Sieber and Marion Points, below top of the Tapeats J
Sandstone, 12S 415000E 4017380N (NAD 27) 5100 ft, 20 May 2008, Rink 7091 (ASC); South Canyon, ca. 0.2 mi downstream from whet®
the trail drops to the bottom of the canyon, growing along rocky ephemeral drainage channel, 12S 417366E 4036917N (NAD 84) 4408*
ft, 6 May 2007, Christie 1418 (ASC).
Scutellaria platyphylla var. occidentalis S. Rhodes & T Ayers, var nov. (Fig. 2, F-G; Fig. 3, D-F) Type:
■' '■ '1 1 * ■ mi 'A ..I Mu'. SO,, n FF: s: < v, OI'V'K 1IJ J-.VHVW 1 tiSS in 0440
ft), 10 Jul 1999, S. Rhodes 9964 (holotype: ASC; isotyfes: ARIZ, ASU).
Simile Scutellaria platyphylla sed cum cauks simplici; foliis subsessilis; et vestis glandula sessilis et pili glanduliferi brevi.
Perennial rhizomatous herbs, 12-16 cm high. Stems simple, erect or ascending from a woody caudex, yellow- 1
green with scattered sessile glands, densely hirtellous, hairs gland-tipped; older stems persistent. Leaves 1
sessile to subsessile, elliptic to obovate, 11-19 mm long, 5-8 mm wide, ca. 2x as long as wide, margins entire, 1
apex rounded, both leaf surfaces hirtellous with scattered sessile glands, short (20-30 microns) gland-tipped*
hairs, and a few sparsely scattered longer (>30 microns) hairs. Flowers axillary in upper stem leaves; pedicels 1
2. 5-3.0 mm; calyx with dense sessile glands and gland-tipped hairs; corolla 7-11 mm long, tube diameter 1
at throat 4.3-5.6 (6.3) mm, outer surface with minute, sparse hairs and scattered glands, limb dark violet, M
aging blue; throat white with purple spots. Mericarps 1.0-1.5 mm long, greenish-black with a prominent, I
often elongate apical callosity on the papillae. Plants of dry sandy soil; Upper Verde River drainage, northern 1
Yavapai County (Fig. 4); Pifton-juniper associations. 1646-1950 m (5400-6400 ft). Flowering May. j
Scutellaria platyphylla var. occidentalis is endemic to the Upper Verde River drainage of northern Yavapai I
County, where it is locally abundant. These populations occur as a northwestern range extension of the |
current distribution of S. platyphylla. Epling (1942) makes reference to a specimen collected near Ashfork I
(Tourney 385, ARIZ) in the exsiccatae for S. tessellata, but noted that it was anomalous with “pubescence like !
that of S. resinosa, but glandular” in his discussion of S. potosina subsp. platyphylla. This collection is the I
earliest specimen referable to this taxon. Variety occidentalis also has smaller corollas, a mixture of sessile I
glands, eglandular, and gland-tipped hairs, but the hairs are consistently a fraction of the length of the hairs |
found in var. platyphylla.
>27)1
2 7 km SE of Rock Butte, Limestone Canyon, dry canyon bottom. N3871380, E370420, T18NJ
450 m (4760 ft), 23 Jun 1994, M. Baker 11469 (ASU).
; 9 Jun 2001 , Rhodes 215 (ASC); Junior Mesa Wilderness Area, c^. ^^^r^ri^334.9635<-N. ]
S9. 1954 m (6410 ft), 20 Jun 1992, Baker 9598 (ASU); Juniper Mts, 50 yards S of FSR 7 and Mud Spring Ri j
May 1984, Hodgson 2887 (DES); Ash Forks, 26 Jun 1892, J.W. Tourney 385 (ARIZ).
KEY TO THE SCUTELLARIA SECJ. RESINOSA TAXA
1 Plants densely pubescent on stems and leaves; pubescence of gland-tipped hair:
2. Stems rarely branched above middle; gland-tipped hairs ca. 0.1 mm long; plant
County y y
2. Stems many-branched above middle; gland-tipped hairs 0.4-0.I
Maricopa, and Pinal counties
1 . Plants with few hairs on leaf surfaces, pubescence of eglandular hairs
I long: plants of Apache, Gila,
var platyphylla
Journal of the Botanical Research Institute of Texas 4(1)
SALVIA CORIANA SP. NOV. (LAMIACEAE),
A NEW SPECIES FROM A CLOUD FOREST IN WESTERN GUATEMALA
Taylor Sultan Quedensley Mario E. V4!iz Perez
4. Upper corolla lip 1 0-1 2 mm long, lower lip short or n
4. Upper corolla lip 2.5-3 mm long, subbequal
. Corollas white, blue, or purple.
5. Corolla tube 4-8 mm long.
. Calyx in flower 7-8 mm long, densely pubescent _
8. Corollas light purple; shrub S. pur
Distribution and habitat. — Salvia coriana is known from only two adjacent localities at approximately 200C|
meters on the northwestern slopes of Pico Zunil. Pico Zunil (14°46'N; 91°27'W) is a mountain with a
elevation of 3,542 meters (Williams 1960; Gall 1983; Quedensley & Bragg 2007). Pico Zunil is part of the|
Sierra Chuatroj range formed by the Zunil ridge that extends from south to north perpendicular to the Pacific!
coast between the departments of Quetzaltenango and Solola (Fig. 2). The Sierra Chuatroj, located between |
Rio Samala to the west and Rio Nahuala to the east, is part of a volcanic belt that extends 120 km from
south of Guatemala City to the Mexican border. The two localities include the roadside to the hot springs
Aguas Amargas, and trails through dense forest with rocky outcrops above the hot springs. Associated trees I
included Billia hippocastanum, Bocconia arborea, Leandra subseriata, Oreopanax xalapensis, O. peltatus, Podacha- 1
enium eminens, Urera caracasana, and Wigandia urens. Shrubs included Alloispermum integrifolium, Aphelandra j
schiedeana, M onochaetum subtriplinervium, Montanoapteropoda, Piptothrix areolaris, Tibouchina longisepala, and!
Vemonia arborescens. Common herbs were represented by Ageratum rugosum, Fleischmannia pycnocephaloidesM
Heterocentron subtriplinervium, Nasa triphylla subsp. rudis, and Salvia purpurea.
Phenology.— Observed in flower in late December and early January.
This species appears to be morphologically related to Salvia recurva, but S. coriana is a liana, the leaves!
are smaller, the upper and lower corollas lips are shorter in length, and the corollas are blue. Salvia recurva!
has not yet been reported from the volcanic belt of Guatemala and in that country is known only from th«J
Department of Huehuetenango at elevations above 2500 meters (Standley & Williams 1970). Morphology!
suggests that Salvia coriana is in the subgenus Calosphace (Benth.) Benth., and within the section Dusenostachp M
Epl. According to Standley and Steyermark (1946), 45 species of Salvia are found in Guatemala, and 312 spe- 1
cies are located in adjacent Mexico (Ramamoorthy & Elliot 1993). Original coflections of S. coriana included j
only two individual plants at the two reported localities. A return visit to the site in Dec 2009 revealed to j
the authors a large population of S. coriana above Aguas Amargas growing high up into cloud forest trees!
and shrubs. Pico Zunil has been documented to consist of relatively high numbers of endemic Guatemalan!
species, and two so far (Ageratina zunilensis: Asteraceae and Stanmarkia spectabilis: Melastomataceae) are |
considered to be endemic to the Sierra Chuatroj (Nash & Williams 1976; Almeda 1993; Quedensley 2007)1
More field work in the volcanic belt of Guatemala is required to illustrate the distribution of S. coriaM, 1
especially in the region of Pico Zunil.
I>bo'TYPhS j“l^^EMALA* Quetzaltenango. Municipio de Zunil: 5 km SE of Xela, N W slopes of Pico Zunil dense cloud forest on trail j
F, TEW.disturClTfe^T885’ 2094 me’ 5 Jan 20°6' N14°45'22 0" W91°29'21.4" T. Sultan Quedensley &J.E. York 4810 (BIGU. I
ACKNOWLEDGMENTS
th^GraduTte Coir^t^aSt^TI|V*<*e(f ^r°m ^ Un'vers*ly Nebraska at Omaha (UNO) Department of Biology
uate College, the College of Arts and Sciences, and the University Committee on Research. Fundinf
Quedensley and Veliz, Salvia coriana, a new species from Guatemala
a. A. Flower, side view. B. Flower, front view (Photos by T.S. QuedensleyJ
Journal of the Botanical Research Institute of Texas 4(1)
Fig. 2. Type locality of Sah/ai c
Journal of the Botanical Research Institute of Texas 4(1)
BOOK REVIEW
Judith Larner Lowery 2007. The Landscaping Ideas of Jays. (ISBN 978-0-520-24164-0, pbk). University;
of California Press, 2120 Berkeley Way, Berkeley, California 94704-1012, U.S.A. (Orders: ww w.ucpress.
edu or Califomia-Princeton Fulfillment Services, 1445 Lower Ferry Road, Ewing, New Jersey 08618, |
U.SA-, orders@cpfsinc.com, 1-800-777-4726). $25.95, 280 pp., 20 color plates, 6" x 9".
(S. Inst. Texas 4<1): 32. 2010
A NEW SPECIES OF CALCEOLARIA (CALCEOLARIACEAE) FROM
DISTURBED PARAMOS IN SOUTH ECUADOR
Pamela Puppo
Research Center in Biodiversity and Genetic Resources (CIBIO)
University of Porto, Campus Vairao
R. Monte-Crasto, 4485-66 1 Vairao, PORTUGAL
pj>uppo@hotmail.com
ABSTRACT
RESUMEN
Calceolaria L. is the largest genus of the Calceolariaceae, comprising about 250 species distributed from central
Mexico to southern Argentina. Calceolaria was formerly part of the Scrophulariaceae, but was raised to the
family level together with Porodittia G. Don and Jovellana Ruiz & Pav. by Olmstead et al. (2001). Andersson
(2006) found that Porodittia was nested within Calceolaria, leaving only two genera in the family.
Calceolaria is easily recognized by its yellow bilabiate flowers with saccate lower lip and two stamens
(Molau & Sanchez-Vega 1986). Most of the species have a patch of glandular trichomes that produce oil.
This structure is known as elaiophore (Vogel 1974) and is located in an infold of the lower lip in the corolla.
Calceolaria species with an elaiophore are pollinated by specialized oil-gathering bees (Vogel 1974; Molau
1988; Rasmussen 1999; S^rsic 2004; Cosacov et al. 2009) while species lacking this structure are pollinated
by common bees or bumblebees (Molau 1988; Stoic 2004; Cosacov et al. 2009).
Calceolaria is subdivided in 3 subgenera and 22 sections (Molau 1988). Subgenus Calceolaria is the most
frequent in the Neotropics, while subgenus Cheiloncos and Rosula are mainly distributed in the southern
temperate. Many sections in Molau (1988) have been shown to be polyphyletic (Andersson 2006; Cosacov
etal. 2009).
I here describe a new species collected in Loja, southern Ecuador that I detected during general iden-
tifications at the Missouri Botanical Garden in 2005.
Calceolaria molaui Puppo, sp nov (Fig. IP t.ti to apor i ,,uJ imihur.no Zuml-u 3W m. h Nov 2000. pm
Subshrub, stems strigose with ramified brown glandless hairs. Leaves decussate, petiolate; blades subco-
J. Bot Res. Inst Texas 4(1): 33 - 36. 2010
Journal of the Botanical Research Institute of Texas 4(1)
Puppo, Calceolaria moiaui, a new species from Ecuador
thecae equal, deflexed; filaments 1.4-1.5 mm long. Style 4-4.5 mm long. Capsules conical, 5 mm long,
puberulous, glutinous.
Distribution and ecology.— This species is known only from the type locality in Loja. It was collected
in disturbed paramo with patches of Andean forest at 3400 m. The area in which Calceolaria moiaui was
collected is near the border with Peru, it is therefore expected to occur in northern Peru as well.
Etymology. — The species is named after Ulf Molau, who has contributed to the knowledge of Calceolaria
for several years.
DISCUSSION
lowing the last treatment of Calceolaria (Molau 1988), C. moiaui is placed in section Lehmannina Pennell
ause of its subshrub habit, subcoriaceous leaves, anthers totally dehiscent, and especially its hirsute
umentum. The species of this section are distributed from northern Colombia to southeastern Peru,
wing in humid habitats usually above 2000 m (Molau 1988). This section however, has also appeared as
yphyletic in phylogenies presented by Andersson (2006) and Cosacov et al. (2009).
Although C. moiaui is known only from the type specimen, it has a particularly combination of characters
t makes it a clearly distinct species. The plant is covered with branched trichomes, the leaves are subco-
;eous and rounded at base, the cyme bracts are absent, the corolla lacks an elaiophore, and the anthers
deflexed. This is the first time that branched trichomes are described for Calceolaria thus constituting a
Other species from section Lehmannina present in Ecuador are: C. cataractarum Molau, C. frondosa
Molau, C. lehmanniana Kraenzl., C. martinezii Kraenzl. and C. pedunculata Molau, all of these from northern
and central Ecuador (Molau 1988; Jorgensen & Le6n-Yanez 1999). Calceolaria cataractarum has elliptic or
lanceolate leaf blades and the flower has an elaiophore. In C. lehmanniana the leaves are also elliptic or lan-
ceolate but the corolla is white. In C. moiaui the leaves are ovate, the corolla is yellow, and the elaiophore is
absent. Calceolaria frondosa has herbaceous leaves less than 2 cm long, and the elaiophore is present while
in C. moiaui the leaves are subcoriaceous larger than 2 cm, and the elaiophore is absent. Calceolaria marti-
nezii has lanceolate leaves, an upfolded lower lip that closes the corolla, and ascending thecae whereas C.
moiaui has ovate leaves, an open corolla, and deflexing thecae. Finally, C. pedunculata has glabrous leaves
and triangular sepals while C. moiaui has hairy leaves and the sepals are ovate.
Species from this section present in Northern Peru are C. luteocalyx Edwin and C. hirsuta Molau. Cal-
ceolaria luteocalyx has herbaceous leaves, yellow-green or brightly yellow sepals more than 9 mm long, the
corolla is closed and the elaiophore is present. Calceolaria hirsuta has also herbaceous leaves, and the sepals
are glabrous abaxially. Both species are restricted to the department of Amazonas in Northern Peru. Calceo-
laria moiaui has subcoriaceous leaves, green sepals less than 5 mm long densely pubescent on the abaxial
margins, open corolla and elaiophore absent.
Other species possibly related to C. moiaui are C. phaeotricha Molau and C. fusca Pennel from section
Salicifoliae. Calceolaria phaeotricha is restricted to a small area of the Cordillera Oriental in Ecuador while C.
fusca occurs from South Colombia to North Peru. Calceolaria phaeotricha has simple trichomes only at the
base of the petioles and upper branches. It has glabrous, glutinous leaves with pinnate venation, the cyme
bracts are present, and the anthers are pale yellow. Calceolaria moiaui is covered with branched trichomes,
has hirsute, non glutinous leaves with reticulate venation, the cyme bracts are absent, and the anthers are
brown. Calceolariafusca has a closed corolla and elaiophore present while C. moiaui has an open corolla and
Conservation status. — In Ecuador, the paramos are highly disturbed due to anthropogenic activities such
as agriculture, provoked fires, cattle, reforestation with alien species, among others (Quizhpe et al. 2002).
Calceolaria moiaui was collected in the southern paramos and should be consider as a critically endangered
species (CR) under the B2ab(iii) criteria of the IUCN (2001).
POA UN1SP1CULATA, A NEW GYNODIOECIOUS SPECIES OF CUSHION GRASS
FROM PERU WITH A SINGLE SPIKELET PER INFLORESCENCE
(POACEAE: POOIDEAE: POEAE: POINAE)
Gerrit Davidse
Robert J. Soreng and Paul M. Peterson
John S. Lehmann Curator of Gras
Department of Botany
National Museum of Natural History
Smithsonian Institution
Washington, DC 2001 3-701 2, USA
sorengr@si.edu, peterson@si.edu
ABSTRACT
:s of Poa by having ir
RESUMEN
A curious grass from Cerro de Pasco, Peru, has remained unidentified for many years. Renewed attempts to
identify this little, puna cushion grass, have led us to the conclusion that it is a new species of Poa L. Two
of the coauthors (RJS and PMP) traveled to Cerro de Pasco, Peru (the type locality) to obtain more material
for study. It is unique among the American species of the genus because its inflorescences are reduced to
a single spikelet.
Poa unispiculata Davidse, Soreng & P.M. Peterson, sp. nov. (Figs. 1A-V; 2). Type: Peru Pasco: Cerro de Pasco, 4 km
Plants herbaceous perennials, gynodioecious herbs forming low dense cushions to at least 19 cm in diam-
eter and 1-1.5 cm thick; culms 1.5-4 cm long, decumbent, ascending, extensively branching intravaginally
and rooting adventitiously from the lower nodes; prophylla prominent, about as long as adjacent sheaths,
membranous, 2-keeled, glabrous but the keels densely covered with more or less erect prickle-hairs. Leaves
strongly distichous, glabrous, 20-30 per culm (6-15 blades present at flowering, old blades disintegrating
before the sheaths frequently with irregular fibers from their bases), crowded but more or less equitant along
the culm, each at 1-2 mm above the next of the same rank; sheaths long persistent, smooth, glabrous, very
pale, strongly 3-nerved at the base, 5-nerved near the apex but the marginal nerves notably less developed
than the central and middle nerves; ligules to 1.5 mm long in lower leaves, graduated upwards to 3.1 mm
long, prominent in the uppermost leaves, scarious-hyaline, smooth or sparsely scabrous below, glabrous,
decurrent along the margins of upper sheaths, obtuse and prominent, and shallowly praemorse to acute
with lateral dents flanking apex: blades, 3-8 mm long, 0.5-0.7 mm wide, divergent at nearly right angles
at maturity, folded or open V-shaped in vivo, folded when dry, pale green, smooth, glabrous, 7-9-veined,
J. Bot. Res. Inst Texas 4(1): 37 - 44. 2010
Journal of the Botanical Research Institute of Texas 4(1
Davidse et al„ Poa unispiculata, a new species from Peru
Fig. 2. Habit of Poa unispiculata [P.M. Peterson, RJ. Soreng & K. Romaschenko 20382 (US)]. A bisexual individual in anthesis, in vim, excavated and laying
the margins slightly involute and sparsely to regularly scaberulous, bases flattened and to 1.6 mm wide
with age, abaxially smooth and shiny to slightly lustrous, with veins expressed, adaxially dull, smooth,
with low narrow costae, apex blunt, the blade keel running in a nearly straight line (slightly upturned for
ca 0.2 mm) into the apex, the margins being more downturned to meet the keel. Inflorescence composed
of a single erect spikelet, barely exerted above the foliage; without nodal scars below the solitary spikelet;
peduncles 5-6 mm long, smooth, glabrous, slightly thickened at the apex and laterally compressed. Spike-
lets slightly dimorphic, bisexual or functionally unisexual and pistillate, smooth, glabrous, 2(-3)-flowered
(the 3rd, slightly reduced in all parts), often with a short rachilla extension to 0.4 mm long; glumes much
shorter than the spikelets, keeled, about as broad as long, subequal, herbaceous in the central area, and
broadly scarious-hyaline margins and lobes flanking the apex, edges smooth, apex obtuse to truncate or
emarginate or retuse with a broad, U-shaped, Q.2-0.5 mm deep notch between the slightly flared, acute
lobes; rachilla smooth, glabrous, terete, intemode between 1st and 2nd floret mostly obsolete or up to 0.4
mm long, between the 2nd and 3rd florets ca. 1 mm long; callus of lemma not well-developed, blunt, gla-
brous, a narrow thickened rim at the base of the lemma; lemmas strongly laterally compressed and keeled,
smooth, glabrous, thinly herbaceous in the lower 1/2-2/3, scarious-hyaline in the upper 1/3 and narrowly
so along the margins to the base, apex and margins smooth, the lateral nerves extending ca. 1/2 way to
apex; palea slightly shorter to nearly as long as the lemma, smooth, glabrous, scarious-hyaline except
the thinly herbaceous keels and adjacent inter nerve and margins in the lower part, 2-nerved and 2-keel
between the keels 0.5-0.7 mm wide in the sulcus, incised up to 1/3 the length, the 2 terminal lobes slightly
flared in the distal %; lodicules 2, with or without a slender to 0.5 mm long lateral lobe nearly equaling the
main lobe, smooth, glabrous; ovary glabrous, styles 2, ca. 1 mm long, terminal, bearing stigmatic hairs to
¥■ w §■
Journal of the Botanical Research Institute of Texas 4(1)
near the base, densely plumose, lateral branches simple; bisexual spikelets: lower glumes 1.3-1.6 mm long,!
1-3-nerved; upper glumes 1.6-2.0 mm long, 1-3-nerved (both glumes 3-nerved in Peterson, Soreng &Ro-
maschenko 20382); lemmas 4.9-5.5 mm long, 5(-7)-nerved, intermediate nerves faint, apex obtuse, slightly!
flared; lodicules 13-1.5 mm long, 0.3-0.5 mm wide, narrowly lanceolate, attenuate; stamens 3, filaments |
erect, apically exerted, anthers 2.3-2.6 mm long, glabrous; (pistil as in pistillate flowers but slightly smaller,!
and no caryopsis observed); pistillate spikelets: lower glumes 1.3— 1.5 mm long; upper glumes 1-5— 1.6 mm |
long; lemmas 3.3-4.6 mm long, 3-nerved, in the lower part becoming slightly thicker at caryopsis maturity!
apex obtuse to truncate; lodicules 0.8-1.5 mm long, to 0.7 mm wide, asymmetrically elliptical; staminodesi
3, 1.0-1. 5 mm long including anthers, anther sacs 0.3-0.4 mm long, rudimentary, sterile; caryopses 1.3-1.51
x 0.4-0.5 mm, fusiform, brown, translucent, falling with the lemma and palea but not adherent to them, a
the adaxial (hilum) surface shallowly sulcate, the apex obtuse, the base acute; embryo ca. 1/3 as long as thel
caryopsis; hilum ca. 1/4 as long as the caryopsis, broadly elliptical.
Habitat. — Puna, 4380-4400 m, rare (ca. 10 clumps seen in 2007) in dense thatch, on low hummocks, 1
and around the perimeters of shallow moist depressions in gentle, heavily and closely grazed, terrain, with |
Aciachne sp., Alchemillapinnata Ruiz & Pav., Calamagrostis minima (Pilg.) Tovar, Carex sp., Didymodon sp ., Pofl j
aequigluma Tovar, and P. humillima Pilg. Other plants collected by J. Solomon at this location were: Aloinrik ]
cucullifera Steere, Alternanthera lupulina Kunth, Astragalus sp., Baccharis caespitosa (Ruiz & Pav.) Pers. var. 1
caespitosa, Barbula replicata Taylor, Bromus sp., Draba pickeringii A. Gray, Festuca rigescens (J. Presl) Kunth, !
Gentianella muscoides (Gilg) T.N. Ho & S.W. Liu, Senecio nutans Sch. Bip., Senecio spinosus DC., Senecio wer-f
nerioides Wedd.
Comments— Poa unispiculata is morphologically similar to P. perligulata Pilg., but differs by having a j
more compact habit with multiple equitantly overlapping leaf sheaths with short, tightly ranked, distichous |
blades, presence of one spikelet per inflorescence, and gynodioecy with long anthers in bisexual plants. We
know of only one other species of Poa that consistently has a single spikelet per inflorescence (see discus- j
sion). Apical notching of spikelet bracts is rare in the genus, but also appears in P {sect. Anthochloa (Neesfil
Meyen) Soreng & L.G. Gillespie] lepidula (Nees & Meyen) Soreng & L.J. Gillespie.
4382 m. 10 Mar 2007. perfect flowered plants only. P.M. Peterson, R.J. Soreng & K. Romaschenko 20382 (B, CONC, CORD, K. LPB, MO, ]
DISCUSSION
Poa unispiculata is reasonably placed in the “punapoa” complex of Poa (Soreng et al. 2003) which is now!
considered to include dwarf alto-andinean species with reduced inflorescences that are frequently diclinous!
or pistillate only and apomictic (Negritto et al. 2008). Poa unispiculata, like P. perligulata and some other j
punapoa species, has smooth palea keels and is thus problematical for separation from members of the Cata-|
brosa complex [Catabrosa P. Beauv., Catabrosella (Tzvelev) Tzvelev, Paracolpodium (Tzvelev) Tzvelev, HyaloprtW
(Tzvelev) Tzvelev] with multiple flowered spikelets. However, P. unispiculata has strongly keeled lemmas withl
occasionally up to seven nerves, and lacks epidermal papillae; whereas in the Catabrosa complex there are
usually three or rarely five nerves, papillae are usually present on blades and culms, and the prophylla (pale*!
homologues) keels are smooth (versus densely covered with more or less erect prickle-hairs in P unispiculataM
and variously scabrous in other punapoa). In analyses of nuclear ribosomal and plastid DNA data punapoaj
species resolve among species of P subgen. Poa supersect. Homalopoa (Dumort.) Soreng & L.G. Gillespie!
whereas the Catabrosa complex stands well apart from Poa (Gillespie & Soreng 2005; Gillespie et al. 2008 j
2009). Within a broad survey of Poa and allied genera, preliminary ETS and ITS nrDNA sequence data place!
the new species nearest to P. aequigluma or P. perligulata in a clade with P. lepidula and P. gymnantha P%|
(L.J. Gillespie pers. comm. Feb. 2010).
Poa is the largest genus of grasses (Gillespie & Soreng 2005) and is well known for its diverse breeding J
systems (Anton 1978; Anton & Connor 1995; Negritto & Anton 2000; Soreng 1991; Soreng & Keil 2003)1
Although bisexual species are the most common, species exhibiting dioecism, gynodioecism, and gynomol
noecism are well known in both North and South America (Soreng 1991, 2007). A few andinean species
are known only from populations composed of plants with pistillate spikelets, e.g., P. perligulata Anton &
Connor 1995; Negritto & Anton 2000; Negritto et al. 2008).
Our judgment of breeding system of the new species is based on only three separate plants available to
us for study from the type collection, and the one new collection of 10 plants (which we collected parts of).
Poa unispiculata is judged to be a gynodioecious species because two of the cushions collected in the original
gathering produced only pistillate spikelets and the third only bisexual spikelets. All individuals plants of
the newer gathering had bisexual spikelets with anthers well exerted (Fig. 2), and no pistillate plants were
found. No fruit were found in the bisexual spikelets from either gathering. Apparently all these individuals
were collected from a single population from an area possibly less than 30m2. The breeding system obviously
needs confirmation and additional field studies are necessary. It could be that plants have bisexual flow-
ers early in the season, and that a proportion of individuals shift to pistillate flowering later in the season
(Soreng & Keil 2003). It is assumed that each cushion as collected is an individual plant that has attained it
compact growth, it is impossible to be sure that no additional individuals were mixed within the cushions.
To compensate for reduction in reproductive capacity of each culm harboring a single spikelet, a single mat
of P. unispiculata can potentially produce several hundred flowering shoots per flowering season (e.g., the
US isotype has approximately 200 flowering culms in one mat of 14 cm diam.).
Anton and Connor (1995) listed 14 other gynodioecious species then known from the Americas.
The breeding system of four of the North American “partially gynodioecious” species was reclassified as
sequentially gynomonoecious (Soreng & Keil 2003) and another as subdioecious (Soreng 2007), which is
considered intermediate between gynodioecism and dioecism. Of the South American species listed, four
belonging to Poa subgen. Andinae were transferred to a new genus, Nicoraepoa Soreng & Gillespie (Soreng &
Gillespie 2007) [quite likely P. s tepparia Nicora belongs here also], and one was reclassified as gynomonoe-
cious, while yet another species was recharacterized as gynodioecious (Negritto & Anton 2000). Therefore,
only four cases of gynodioecism exist in North America [P. leibergii Vasey, P. stebbinsii Soreng (which can also
produce staminate inflorescences), P. cusickii Vasey subsp. cusickii and P chambersii Soreng (both of which
are gynodioecious in part of their ranges and dioecious over the rest), and two exist in South America (P.
cabreriana Anton & Ariza and P. lilloi Hack.; Negritto & Anton 2000)]. So, actually gynodioecism is quite
rare in Poa.
It is well known that dioecious species of Poa sect. Dioicopoa E. Desv. differ noticeably in the size of the
spikelets, glumes, and florets (Giussani 2000; Soreng 2007; Soreng and Peterson 2008); pistillate spikelets
have fewer and larger florets, and, if pubescent, have substantially more and longer hair. Differences among
pistillate and perfect-flowered individuals of gynodioecious species of Poa have not been noted before, but
i P. unispiculata. In this species the two spikelet types are slightly sexually dimorphic. The
of the bisexual florets typically have an additional 1 (or 2) faint intermediate nerves. Besides the very obvious
differences in the stamens, the pistils of the pistillate spikelets are slightly larger. Fruit set was abundant
in the pistillate plant, but no fruit were observed in the bisexual plants, presumably because most of its
spikelets were pre-anthesis or at anthesis. Possibly the pistils are infertile in the plants with fertile anthers
(our bisexual plants), in which case the species is dioecious, but this would be an exceptional pattern in Poa,
as pistils are rudimentary or at least more obviously reduced in species known to be dioecious. Solanaceae
have minimally reduced sex organs in some dioecious species, but pollinator attraction and reward factors
are in play there (Martineab et al. 2006) that do not apply to wind-pollination. In the bisexual plants the
florets are longer and broader, and thus appear more prone to wind dispersal than those in the pistillate
plants, which seems contrary to expectation unless they actually do produce seed.
As far as we are aware, 3-nerved lemmas are not known elsewhere in the genus (except in some spe-
cies from New Guinea; Veldkamp 1994), but in Poa unispiculata this seems to be a clear specialization in an
Journal of the Botanical Research Institute of Texas 4(1)
otherwise faintly 5(-7)-nerved species, as judged from the lemmas of the bisexual spikelets. This charade®
state (often used to characterize genera and tribes) is clearly secondarily derived.
Another curious characteristic of Poa unispiculata is the flared scarious-hyaline margins of the glumes *
which extend upward as lobes beyond the herbaceous keel and back, leaving a U-shaped notch (Fig. 1. Bj|
E-H), and the deep narrow notch between the somewhat flared and scarious-hyaline distal lobes of the palea.
These are reminiscent of the flared scarious-hyaline margins of glumes, lemmas, and paleas, and notches of#
various depth in those, in Poa lepidula. The latter species was formerly treated as the sole species of the genus |
Anthochloa, but is now accepted as Poa sect. Anthochloa, and closely related to the punapoa group (Gillespie I
et al. 2007, 2008). The flared scarious-hyaline margins are an obvious adaptation for wind dispersal. S
In Poa, inflorescence size varies widely from highly branched panicles with hundreds of spikelets as in J
large plants of P. occidentalis Vasty or P. palustris L. to highly compact inflorescences with very few spikelets I
example of the ultimate extreme in panicle size reduction. It appears most closely related to a group of dimintt®
tive, alto-andinean species from the Andes of Peru, Bolivia, and northern Argentina and Chile ( Poa group
punapoa; in particular, P. humillima, P. aequigluma, P. chamaeclinos Pilg., and P. perligulata , the later three of I
which are generally recognized as strictly pistillate; Soreng et al. 2003). Poa humillima is gynomonoecious#
(Anton 1978) and differs in its more clearly naviculate leaf blades apices, abaxially smooth leaf blades, short |
branched, compact, ovate panicles 0.5-l(-1.5) cm long with less than 10 spikelets; single bisexual proximal 1
florets with (1— )2 pistillate distal florets; broader, subflabellate upper glumes; shorter lemmas (2-2.8 mm J
long); scaberulous glume and lemma apical margins (x 50); all spikelets (2-)3-flowered; and fertile anthem J
ca. 1 mm long. Poa aequigluma, P. chamaeclinos, P. perligulata all have relatively longer glumes (subequal to!
the floret pair in P. aequigluma and P. perligulata ), pistillate spikelets only (as do a few other South Americanl
species, and a several North American species at least in part), and only two florets per spikelet. Recently!
discovered populations of P aequigluma from Ancash, Peru, at the northernmost end of the species range, j
have bisexual spikelets with normal looking anthers about 3 mm long ( Peterson , Soreng, LaTorre & Rojas Fox
21550, 21593, 21684). These latter three species also share the obsolete or nearly obsolete rachilla intemode j
between the proximal two florets with P. unispiculata. The leaf arrangement differs in these three species j
from that in P. unispiculata. All vegetative branching is intravaginal as in P. unispiculata, but leaves of the ’
other species are fewer in number, the sheaths more elongated, and the leaf blades, which are mostly longer, j
diverge at approximately the same level (versus equitantly).
Inflorescences reduced to one or a few spikelets are relatively rare in Poaceae, but according to a search j
on the World Grass Species website (Clayton et al. 2002) are known in at least 73 other species in more than
30 genera, of which eight are annuals and 66 are perennials. However, among those species that produce J
inflorescences with single spikelets, the solitary condition is usually facultative. Individuals of most of these |
species also produce inflorescences with more than one spikelet.
We know of only a ten grass species that always have an inflorescence consisting of a single true spikeletl
( Anomochloa Brongn. and Streptochaeta Schrad. ex Nees can be excluded from our sample by not having
fully evolved “grass spikelets”). This extremely rare condition is known in one other species of Poa, an ahi
pine, cushion grass, P inconspicua Veldkamp, from the Cartestensz Mountains of New Guinea (Veldkamp j
1979, 1994). This species belongs to the remotely related P sect. Pauciflorae Pilg. ex Potztal (LJ. Gillespie,
unpublished DNA data), nine species of which have a single floret per spikelet, and seven others have 1-21
floretsper spikelet (Veldkamp 1994). Poa pygmaea Buchanan, another unrelated, alpine cushion grass, from
ca. 45° S in New Zealand, comes close in having inflorescences with 1-3 spikelets (Edgar & Connor 2000; j
Gillespie et al. 2009). One spikelet per inflorescence occurs rarely in depauperate specimens of P. leibergh ;
Scnbn [(l-)6-17(-22) spikelets] from the Pacific Northwest in North America. Obligate unispiculate inM
rescences are additionally known in two of the three species of the genus Aciachne Benth. (Pooideae, Stipeae;
A. Jlagellifera Laegaard and A. pulvinata Benth.) cushion plants of the high Andes and Costa Rica (p*ramo del
Cerro Chimpd). In these two species the fully developed spikelet is subtended by one or two scars which
Davidse et al., Foa unispiculata, a new species from Peru
may represent minute, vestigial spikelets. In the third species, A. acicularis Laegaard, inflorescences with
two or rarely three spikelets may occur on the same plant with predominantly single-flowered inflorescences
(Laaegaard, 1987). Obligate unispiculate inflorescences have also been reported in one species of Rytidosperma
Steud. (Danthonioideae, Danthonieae), R. oreoboloides (F. Muell.) H.P. Linder, a cushion species from also
alpine plants (Veldkamp 1979). All of the above taxa with strictly one spikelet per inflorescence are from
high elevation grasslands. Lygeum Loefl. ex L. (Pooideae, Lygeeae) with only the single species of low, arid
is derived from the fusion of two spikelets that is subtended by a somewhat larger, loose spathe. The above
genera have non-Kranz anatomy and presumably C3 metabolism. A few other species with single spikelet
inflorescences are known or presumed to have C4 metabolism (Zuloaga 1987; Aliscioni et al. 2003). Within
Distichlis Raf. (Chloridoideae, Cynodonteae) both of the species formerly treated as Monanthochloe Pilg. ex
Potztal, D. littoralis (Engelm.) H.L. Bell & Columbus, coastlines of the southern United States and Mexico,
and D. acerosa (Griseb.) H.L. Bell & Columbus, saline flats of northern Argentina (Bell & Columbus 2008),
occur at sea level or in low plains up to ca. 350 m. The dwarf alto-andinean species Distichlis humilis Phil, has
1-4 spikelets, but can be distinguished from our species by the 7-11 veined lemmas. Panicum cupressifolium
A. Camus (Panicoideae, Paniceae), which has single spikelets per inflorescence, occurs at moderately high
elevations in the mountains of Madagascar (massif de LAndringitra, 2000-2500 m) in a rocky/scrubby zone.
Evidently, obligately single-spikelet inflorescences have evolved in grasses most commonly open habitats,
in plants with C3 metabolism, in alpine-like situations of tropical to subtropical latitudes.
ACKNOWLEDGMENTS
We thank the National Geographic Society Committee for Research and Exploration (grant number 8087-06)
for field and laboratory support; the Smithsonian Institutions, Restricted Endowments Fund, the Scholarly
Studies Program, Research Opportunities, Atherton Seidell Foundation, and Biodiversity Surveys and Inven-
tories Program, all for financial support; Alice R. Tangerini for illustrating the new species; Lynn Gillespie for
sharing preliminary DNA results; Liliana Giussani and anonymous reviewers for providing helpful comments
on the manuscript; and the following colleagues who facilitated or assisted with fieldwork: Asuncion Cano
Echevarria, Nancy Refulio Rodriguez, Konstantin Romaschenko, and Dorita Susanibar Cruz.
REFERENCES
Aliscioni, S5., L. M. Giussani, F.O. Zuloaga, ano E.A. Kellogg. 2003. A molecular phytogeny of Panicum (Poaceae: Pan-
iceae): tests of monophyly and phylogenetic placement within the Panicoideae. Amer. J. Bot. 90:796-821 .
Anton, A.M. 1 978. Notas crfticas sobre gramineas de Argentina. III. Contribucidn al conocimiento de la sexualidad
en Poa. Bol. Acad. Ci. (Cdrdoba): 52:267-275.
Anton, A.M. and H.E. Connor 1995. Floral biology and reproduction in Poa (Poeae: Gramineae). Austral. J. Bot.
43:577-599.
Bell RL and IT. Columbus. 2008. Proposal for an expanded Distichlis (Poaceae, Chloridoideae): Support from
molecular, morphological, and anatomical characters. Syst. Bot 33:536-551.
Clayton, W.D., K.T. Harman, and H. Williamson. 2002 onwards. World Grass Species: Descriptions, Identification, and
Information Retrieval, http://www.kew.org/ta/grasses-db.html. Accessed 15 May 2006.
Edgar E. and H.E. Connor 2000. Gramineae. In: FI. New Zealand 5:1-650. Manaaki Whenua Press, Lincoln.
Gillespie, LJ. and RJ. Soreng. 2005. A phylogenetic analysis of the bluegrass genus Poa based on cpDNA restriction
site data. Syst. Bot. 30:84-105.
Gillespie, LI, A. Archambault, and R J. Soreng. 2007. Phylogeny of Poa (Poaceae) based on trnT-trnF sequence data:
major dades and basal relationships. Aliso 23:420-434.
Gillespie, U, RJ. Soreng, R.D. Bull, S.W.L Jacobs, and N.F. Refuuo-RoorIguez. 2008. Phylogenetic relationships in subtribe
Poinae (Poaceae, Poeae) based on nuclear ITS and plastid tmT-trnL-trnF sequences. Botany 86:938-967.
UNA NUEVA ESPECIE DE FEVILLEA
(CUCURBITACEAE: ZANONIEAE) DE COSTA RICA
Armando Estrada Ch.
Daniel Santamana A.
e, una nueva especie de Costa Rica
Estrada and Santamaria A., Fevillea narae, una nueva especie de Costa Rica
AGRADECIMIENTOS
Los autores desean agradecer a la reserva y albergue “Los Campesinos” y a la comunidad de Quebrada Arroyo
de Tarrazu, por el apoyo logistico brindado y por su afan de proteger y usar sosteniblemente los recursos
naturales de su localidad. A Barry Hammel y Michael Grayum por la revision y aportes al manuscrito; a
Carlos O. Morales (USJ) por su ayuda en la elaboracion de la diagnosis latina y a Claudia Aragbn por las
ilustraciones realizadas. A los revisores del articulo, Richard P. Wunderlin y otro revisor anbnimo por las
valiosas observaciones realizadas. A Rafael Chacbn por su compaiiia y asistencia en las giras de campo.
REFERENCIAS
Jeffrey, C. 2001. Cucurbitaceae. En: Stevens, W.D., C. Ulloa, A. Pool, and O.M. Montiel, eds. Flora de Nicaragua.
Monogr. Syst. Bot. Missouri Bot. Gard. 85:688-71 7.
Monro, AK. 2009. Fevillea L (Cucurbitaceae). Flora Mesoamericana. http://mobot.mobot.org/cgi-bin/search_vast
(Consultada el 23 febrero del 2009).
Robinson, G.L and R.P. Wunderun. 2005. Revision of Fevillea (Cucurbitaceae: Zanonieae). Sida 21:1971-1996.
BOOK REVIEW
Rick Bennett and Susan Caua (eds). 2009. A Rare Botanical Legacy: The Contributions of Ruby and
Arthur Van Deventer. (ISBN 978-1-59714-116-1, cloth hbk.). Heyday Books, PO. Box 1945, Berkeley,
California 94709, U.S.A. (Orders: www.heydaybooks.com, 1-510-549-3564, 1-510-549-1889 fax).
$35.00, 154 pp., 100+ paintings and photographs, 9 1/4" x 12 1/4".
ity Library (retired), Fort Worth, Texas, l
A NEW SPECIES OF ANGELONIA (PLANTAGINACEAE) FROM MEXICO
Kerry Barringer
Brooklyn Botanic Garden
1000 Washington Avenue
Brooklyn, New York 1 1225-1099, USA
kbarringer@bbg.org
ABSTRACT
i fruits. It is geographically isolated on the lin
RESUMEN
Angelonia Humb. & Bonpl. (Plantaginaceae) is a genus of about 25 species growing mainly in seasonally
dry lowlands from the Caribbean and southern Mexico to Argentina. A few species are cultivated and have
escaped in tropical regions worldwide. The species have distinctive flowers with a pair of shallow, saccate
nectaries behind the lower lip. These contain elaiophores, mats of glandular hairs that produce a fatty oil
(Vogel 1974). Centris bees, who collect the oil as a larval food, are the principal pollinators.
All of the Angelonia species that grow in Mexico, Central America, and the Caribbean are in Section
Angelonia (Barringer 1982). Species in Section Angelonia have a shallow, depressed palate at the base of the
median corolla lobe that is a landing platform for pollinating bees. There is a small, cylindric, apically bifid
tooth on the outer edge of this palate. The exact function of this tooth is not known, but pollinators ap-
pear to hold on to it while visiting flowers (Vogel 1974). These species appear to be derived from the South
American species, which have a greater diversity of palate and tooth morphologies (Barringer 1982).
An undescribed species of Angelonia grows along the northern tip of the Yucatan Peninsula in fields and
well as the smallest fruits.
Angelonia parviflora Barringer, sp. nov. (Fig. IV i,m i m.ikI.i nud m ik-an-d marshy
Annual herb; roots fibrous, branching from a short taproot; stem erect, to 20 cm tall, glabrous or glabrescent
with glandular trichomes, slightly 4-angled, sometimes branching from near the base. Leaves opposite, sessile,
lanceolate, 3-4 cm long, 0.8-1 cm wide, green, glandular-pubescent, membranaceous, not gland-dotted,
the base narrowed, the margin serrate, the apex acute. Inflorescence a terminal raceme, slightly glandular-
pubescent; pedicels 1 cm long, glandular at the base, bracteolate or ebracteolate; sepals ovate, 2 mm long,
1 mm wide, glabrous, the margin opaque, the apex acute; corolla purple; tube 3 mm long, 5 mm deep, the
sacs 1-2 mm deep, with two pads of glandular trichomes on the forward surface within, the upper lobes
obovate, 3 mm long, 2-3 mm wide, ciliate; lateral lobes obovate, 3 mm long, 2 mm wide, ciliate, abaxially
glabrous, median lobe oblong to obovate, 4 mm long, 1-3 mm wide, adaxially glandular-pubescent, abaxi-
J. Bot Res. Inst. Texas 4(1): 51-5
Journal of the Botanical Research Institute of Texas 4(1)
Fk.1./
K. Habit. B. Corolla. C. lateral view of flower.
ally glabrous, palate crateriform, 3 mm long 2 mm wide, ciliate; horn 1 mm tall, bifid; stamens 4 mm long,
the filaments glandular-pubescent, the thecae ellipsoid, sessile, divaricate; ovary and style glabrous. Capsule
globose, 4-5 mm diam., thin-walled, matte; fruiting pedicel slightly recurved, 0.8-1.0 cm long; seed light
brown, obpyramidal to obconic, 0.5-1. 5 mm long. 0.5-1.5 mm wide, the exotesta reticulate, the longitudinal
Distribution and Habitat. — Mexico (Yucatan & Quintana Roo). Seasonally wet, open grasslands and
thorn scrub on stony, pitted land. (Fig. 2).
Vernacular names. — This is one of the species called “Xac-xvi,” “Xacxiu” “Xakxiw” or “Chi-Bilam xiw,”
“boca de la vieja These names have also been used to refer to A. angustifolia and A. ciliaris.
Phenology. — Blooming from March to October.
Representative specimens examined: MEXICO. Quintana Roo: 4 km S of Puerto Morelos, Tittez & Cabrera 3276 (F, MO, NY, US);
and May 6743 (NY, MEXU, MO, XAL); 10-15 km N of Chuehuemi, Darwin 2402 (TULANE); Progreso, Floree s.n. (F); Progreso, Gaumer
Merida road, Progreso, Jul 1938, Lundell & Lund ell 8028
(MO, US); near Sisal, Schott 917 (F); Merida, Souza 256 (US); Progreso, Steere
Angelonia parviflora is most easily distinguished from other Angelonia species by its small flowers and fruits.
The flowers are 5 mm or less wide across the mouth and only 3 mm deep. The nectary sacs are relatively
shallow at 1 mm deep. The fruits are 4-5 mm in diameter, about half the diameter of most other species. In
addition, the species can be distinguished by the long, sparse glandular trichomes on the pedicels but these
wear off and are usually present only at the base of the pedicels in older flowers and fruits.
Specimens of Angelonia parviflora have been identified as A. angustifolia Bentham, a species which is
native to other parts of Mexico (Fig. 2), but which is often found in cultivation. Angelonia angustifolia differs
in having glabrous pedicels, acuminate sepals 3-5 mm long, and capsules 7-10 mm in diameter. Its flowers
are much larger; more than 7 mm across the mouth and 3-5 mm deep. Angelonia ciliaris B.L. Rob. can be
found in southern Mexico, Belize, and Guatemala (Fig. 2) (Standley & Williams 1973) but, in addition to
having larger flowers and fruits, that species has distinctive pubescent stems, leaves and pedicels.
The species of Angelonia growing in southern Mexico, Guatemala, and Belize can be difficult to dis-
tinguish, because they all have lanceolate leaves and blue or white flowers. South American species, not
included in many keys for the region, have spread from cultivation making identifications more difficult.
The following key distinguishes the species, both native and cultivated, that grow in the region.
KEY TO THE SPECIES OF ANGELONIA IN SOUTHERN MEXICO, GUATEMALA, AND BELIZE
. Stem or inflorescence glandular-pubescent.
2. Leaves to 4 cm long; upper corolla lobes 2-3 mm '
2. Leaves more than 4 cm long; upper corolla lobes r
3. Pedicels 2 per axil; sepals acute. E. Brazil (cultiva
3. Pedicels! per axil; sepals acuminate.
4. Capsules 10 mm diam.; lateral and median ci
Colombia and Venezuela (cultivated)
4. Capsules 6-8 mm diam.; lateral and median coroll
lanceolate; horn 2 mm long. E. Brazil (cultivated) —
. Stem or inflorescence eglandular pilose to glabrous.
5. Sepals acute, cuspidate.
6. Pedicels pubescent; corolla tube 6-7 mm deep; s
Guatemala, and Southern Mexico
?. South America (cultivated).
s abaxially glandular-pubescent, t
FOLIA TAXONOMICA 17. DILKEA (PASSIFLORACEAE) 2.
CONSPECTUS OF THE SPECIES OF THE GUIANAS WITH THREE NEW SPECIES
Christian Feuillet
Department of Botany, MRC-166
Smithsonian Institution
P.O. Box 37012,
Washington, D.C 20013-7012, U.SA
ABSTRACT
The genus Dilkea Mast, can be separated easily from the other three genera of Passifloraceae from the west-
ern hemisphere. It has tetramerous flowers, versus pentamerous in Passiflora L. an androgynophore null to
shorter than the ovary, versus well developed longer than the ovary in Ancistrothyrsus Harms, and 1 style
with 4 branches 1/3 to 1/2 from base, versus 4 free styles in Mitostemma Mast. (Feuillet & MacDougal 2007).
An introduction to Dilkea and a key to the subgenera can be found in Feuillet (2009).
means “not present in the collections studies.” In the descriptions, morphology and measurements between
parentheses pertain to organs still growing.
A. Subgenus Dilkea
The species of Dilkea subg. Dilkea are sometimes low shrubs or treelets or mostly shrubs becoming lianas,
climbing with tendrils that are trifid at apex. If the tendrils do not find a support, they are early deciduous,
but otherwise become rigid and similar to twigs with respect to their diameter. The grabbing segment rolled
around the support is usually swollen. In contrast with species of subg. Epkia Feuillet, those of subg. Dilkea
have branches with internodes gradually increasing in length from the branching point, and their leaves are
more or less regularly spaced, although rarely two are sub-opposite.
Subgenus Dilkea includes all the species described previous to 2009, but (see Feuillet 2009) not D.
johannesii var. parvifolia Hoehne. Killip (1938) placed D. ulei Harms in the synonymy of D. johannesii Barb.
Rodr. and everybody since then agreed with him. Although he suggested the genus might be monotypic,
Killip kept four separate species— D. acuminata Mast, from Amazonas (Brazil), D. johannesii from Amazonas
and Pars (Brazil), D. retusa Mast, from Amazonas (Brazil) and Loreto (Peru), that he selected as the type of
the genus, and D. wallisii Mast, from Pari (Brazil, Loreto (Peru), and Amazonas (Venezuela) - and added
one of his own, D. parviflora Killip from Loreto (Peru). Since then, the taxonomic history of Dilkea has been
a series of various regroupings and a couple of new species. In 1968, D. magnifica Steyerm. was described.
Holm-Nielsen et al. (1988) recognized D. johannesii, D. parviflora, and D. retusa (including D. acuminata, D.
magnifica, and D. wallisii). In 1991, Cervi described D. margaritae from Mato Grosso (Brazil) and compared
it to D. acuminata and D. johannesii. Tillett (2003) recognized as good species D. acuminata (including D.
johannesii var. parvifolia, D. magnifica, and D. parviflora), and D. retusa differing by flower aestivation, but did
J. Bot. Res. Inst Texas 4<1): 55 - 62. 2010
Fig 1 Dilkea darkei A Stem with leaves and tendrils; B. Stem with leaves and a fruit; C. Nodes showing the base of a branch and numerous bud scales;
D. Fruit; E. Fruit open showing one of the seeds in the aril sac; F. Seed with only the base of the aril sac. A from H.D. Oarke 1848 (US); B-F from type H.D.
Oarke 2987 (B, E-F: US; C-D: MO).
Fk. 3. Dilkea exilis, paratype de Granville 1252 (CAY)
Feuillet, New species of Dilkea from the Guianas
Distribution. — Dilkea exilis is known from northern and western Amazonia from the Oyapock River,
the border between French Guiana and AmapS (Brazil), to Para and Amazonas (Brazil).
Dilkea exilis can be recognized easily from other species of subg. Epkia by the combination of the fol-
lowing characters. Its leaves have petioles up to 3 cm long with a short pulvinus and blades with a thin
Etymology.— The specific adjective exilis, meaning thin in Latin, refers to the leaf blades of that species
e: FRENCH GUIANA:
BRAZIL. Mato Grosso: Rio Juruen
Distribution. — Dilkea lecta is kno
western French Guiana, and from Mato Gross<
6. Dilkea vanessae Feuillet, J. Bot. Res. Inst. Tt
so (Brazil).
:xas 3:602, fig. 5, 25 Nov 2009. Typi
, V. Hequet 1000 (holotype: US; isotype: CAY)
e made in the rainforest, on Pic K
l Guiana. They were fruiting in N
KEY TO THE SPECIES OF DILKEA I
itinuous, leaves scattered along the branches _
. Leaf blades membranous or chartaceous^
. Leaf blades coriaceous.
3. Leaf blades elliptic, 5-1 1 x 4-7.5 cm
(5— )7— '
. Petioles reduced tc
y, leaf blades narrow-obianceolate, 9-14.5
. Leaf blades membranous toe
0-1 5 main veins each si
ACKNOWLEDGMENTS
I would like to thank Paul Hiepko (B) and Shawn Krosnick (RSA) for their reviews of this paper. This work
could not have been completed without the help of the curators of the herbaria AAU, CAY, MO, NY, and P
who hosted me in their institution or made available to me the material in their care. This is number 164 in
the Smithsonian’s Biological Diversity of the Guiana Shield Program publication series.
REFERENCES
Cervi A.C. 1991. A study on Brasilian Passifloraceae, Dilkea margaritae A.C. Cervi, spec nov. Candollea 46:61 -63.
Feuillet, C. 2009. Folia Taxonomica 1 6. Epkia, a new subgenus of Dilkea (Passifloraceae), and five new species from
western Amazonia and the Guianas. J. Bot. Res. Inst. Texas 3:593-604.
Feuiuet C. and J.M. MacDougal 2007 [sent to me 19 Dec 2006], Passifloraceae. In: Kubitzki, K. The families and
genera of vascular plants IX: 270-281. Springer, Berlin.
Journal of the Botanical Research Institute of Texas 4(1)
Houvi-Nielsen LB., P.M. Jorgensen, and J.E Lawesson. 1 988. 1 26. Passifloraceae. In: Harling, G. and L. Andersson (edsj,
Flora of Ecuador 31:1-129.
Kilup E.P. 1 938. The American species of Passifloraceae. Publ. Field Mus. Nat. Hist, Bot. Ser. 19:1-613. |
Tillett S.S. 2003. Passifloraceae. In: Berry, P.E., K.Yatskievych, and B.K. Holst, eds. FI. Ven. Guayana, vol. 7:625-66®
M1LIUSA WAYANADICA (ANNONACEAE),
A NEW SPECIES FROM WESTERN GHATS, INDIA
M.K. Ratheesh Narayanan, P. Sujanapal, and N. Anil Kumar
MS Swaminathan Research Foundation
Wayanad -67312 1, Kerala, INDIA
N. Sasidharan
M. Sivadasan
Kerala Forest Research Institute
Peechi, Thrissur-680653, Kerala, INDIA
King Saud University
Riyadh 1 1451
KINGDOM OF SAUDI ARABIA
ABSTRACT
RESUMEN
INTRODUCTION
The genus Miliusa Lesch. ex A. DC. is currently recognized with 40 species mostly of Austral-Asiatic ranges
from India, Sri Lanka, and Bhutan to Australia through the Malayan islands (Mols & Kefcler 2003). Ten
species are reported from the Austro-Malesian area (Mols & KeKler 2003), 33 species from continental Asia
(Tanawat, pers. comm.), and three species from Sri Lanka (Huber 1985). In India the genus is represented
by 15 species distributed mainly in western and northeastern regions and Andaman and Nicobar Islands
(Mitra 1993). Western Ghats is an important center of Miliusa in Peninsular India. Six species of Miliusa are
reported from Kerala (Sasidharan 2007). Ramamurthy (1983) reported five species from Tamil Nadu. Recently,
a new species was described from the Kalakkad-Mundanthurai Tiger Reserve of Tamil Nadu (Murugan et
al. 2004). Five species of Miliusa are known from Karnataka (Saldanha 1984; Sharma et al. 1984). Among
the 15 species in India, four species are endemic to Western Ghats (Mitra 1993; Murugan et al. 2004). This
includes the red listed Miliusa nilagirica Bedd. (IUCN 2006).
Wayanad district in Kerala is unique. The entire area is along the Western Ghats with altitudes ranging
from 700 m to 2100 m msl. During our floristic exploration in the forests of Wayanad, some specimens of
Miliusa collected from the hilltops of Meppady and Kalpetta Forest Ranges were strikingly different from
other species of Miliusa, due to their recurved petals. On critical study with the available literature and
comparison with authentic specimens, it is confirmed that the material does not match with any previously
described species. Therefore it is described and illustrated here as a new species.
Finet and Gagnepain (1906) recognized two sections in the genus viz, sect. Saccopetalum (Benn) Finet &
Gagnep. and sect. Miliusa. Mols and Kefiler (2003) did not refer to these sections in their revisionary studies
of Miliusa in the Austro-Malesian area. However, Huber (1985) followed this categorization in his treatement
for A revised Handbook to the Flora of Ceylon. These sections were solely based on the number of ovules. Section
Saccopetalum is characterized by a minimum number of four ovules and sect. Miliusa is characterized by a
maximum of two ovules. Earlier these two sections were recognized as separate genera. Following Finet and
Gagnepain (1906), our species belongs to the sect. Miliusa due to the 1-2-ovuled carpels and is resembles
to Miliusa indica with intermediate characters of M. nilagirica. Van Heusden (1992) studied morphology,
BOOK REVIEW
toy Green. 2007. Managing Water: Avoiding a Crisis in California. (ISBN 978-0-520-25327-8, pbk.)
University of California Press, 2120 Berkeley Way, Berkeley, CA 94704-1012, U.S.A. (Orders: orders®
cpfsinc.com or wwwucpress.edu, 510-642-4247, 510-643-7127 fax). $27.95, 336 pp„ 6" x 9". 1
e of Texas, 500 E j
the publisher: “Dorothy Green is founding president of Heal the Bay and among the founders of the Los Angeles and San Gabriel
s Watershed Council, of which she is also president emeritus. She has chaired the California Water Policy (POWER) conference
71
Journal of the Botanical Research Institute of Texas 4(1)
obovate, retuse to subretuse at apex, glandular hairy, usually 1-3-nerved; anterior segments much smaller, 1
0.4 x 0.15-0.2 mm, ovate to deltoid, subacute to acute at apex, 1-nerved; labellum very small, hyaline,*
usually 0.2-0.3 mm, rarely 0.5 mm long, subulate or subdeltoid, acuminate to subacute at apex, 1 -nerved®
Column much exserted ca. 2.5 mm long, geniculate, swollen and pink in middle, rest hyaline. Stamens 2,1
lateral, sessile, connate with style; anthers 2-celled, extrorse, anther thecae ca. 0.3 x 0.15 mm, ovoid, pale I
yellow, dehiscing apically by a slit, pollen grains ovoid. Ovary inferior, ca. 5 mm long, 2-celled, with many *
ovules; stigma entire, greenish, hairy. Capsules linear, up to 1 cm long, dehiscent. Seeds very small, ca. 0.1*
mm long, ovoid, brownish.
Stylidium darwinii Punekar & Lakshmin. is allied to S. kunthii Wall, ex DC., S. inconspicuum Slooten,*
a species endemic to Java (Slooten 1954; Bean 2000), and S. tenellum Sw. Stylidium darwinii differs from & I
or pink), anterior corolla segments ovate to deltoid (vs. oblong to lanceolate) and posterior corolla segments*
retuse to subretuse at apex (vs. emarginate). Stylidium darwinii differs from S. inconspicuum in having smaller*
leaves, 2-3 x 1.5-1. 8 mm (vs. bigger leaves 4-8 x 2-5 mm), corolla sparingly glandular hairy (vs. corolla*
glabrous), appendages clear on the throat (vs. appendages on the throat hardly visible), posterior corolla*
segments retuse to subretuse at apex (vs. bifid) and anterior corolla segments subacute to acute at apex (vs.*
bilobed or entire, obtuse at apex). Stylidium darwinii differs from a most akin species S. tenellum in having §
anterior corolla segments ovate to deltoid (vs. lanceolate to oblong-lanceolate), posterior corolla segments*
smaller, 0.8-1 x 0.5-0.7 mm (vs. longer, 2. 1-3.3 x 1. 1-1.3 mm), retuse to subretuse at apex (vs. deeply bifid),*
column ca. 2.5 mm long, swollen in the middle (vs. 4-5 mm long, of uniform width throughout).
Bean (2000)— in the revision on the Stylidium subg. Andersonia (R.Br. ex G. Don) Mildbr. — treated many
taxa as synonyms under Stylidium tenellum such as Stylidium roseum Kurz (Type: Chittagong, Bangladesh), 5*
tenellum var. minimum C.B. Clarke (Type: Chota Nagpur at Hazaribagh) and Epilobium tonkinense H. L£v. (Type: 1
Tonkin near Quang-Yen, Vietnam). These synonyms are nothing but a size variant of S. tenellum especially I
considering the height of the plant and number of flowers, but none of them differ in floral morphology. The I
newly proposed species S. darwinii is compared with that of Bean’s description for S. tenellum which covers!
the entire length of variation of S. tenellum and its synonyms. Mildbraed (1908), Erickson (1958) and Mitra |
et al. (1998) also treated S. roseum as a synonym of S. tenellum, while Mitra et al. (1998) in their treatise on*
the genus Stylidium of India for the first time considered S. tenellum var. minimum as a synonym of S. tenellum. 1
The only available classification of Stylidium is that of the German botanist Mildbraed (1908). On the*
basis of habit and floral morphology S. darwinii is currently best placed in subgenus Andersonia (R.Br. ex Gf
Don) Mildbr. until a formal revision of the infrageneric taxonomy of Stylidium is completed.
Distribution. This species is known only from Anshi National Park, Uuara Kannada (North Kanara)|
District of Karnataka State, India.
Habitat and Ecology.— Rare in the outskirts of evergreen forest near roadsides at an elevation of 530 fll
above mean sea level in moist gravelly soil near monsoon puddles in association with Drosera burmannii, D*
indica, Eriocaulon achiton, E. lanceolatum, E. truncatum, E. xeranthemum, Fimbristylus spp., Lindernia cmUttm
L. hyssopioides. Lobelia alsinoides, Utricularia caerulea, U. minutissima and U. naikii. Insect caterpillar probably J
a Lepidopteran was observed feeding the flower parts.
Flowering and Fruiting.— September-November.
Etymology. The specific epithet “darwinii" is after Charles Robert Darwin, who is well known for his |
publication of Origin of Species by Means of Natural Selection.
IV: U':„ s: INDIA. Maharashtra. Uttara Kannada District: Anshi National Park, 15”00.67 N, 07473 68 E 530 m, 28 Sep 2004,
Punekar 388 (MACSG); same locality, 24 Nov 2009, Punekar 698 (AHMA. MACSG)
KEY TO THE INDIAN SPECIES OF STYUD1UM
1 . Leaves strictly cauline, scattered all over stems.
2 Posterior corolla segment retuse to subretuse at apex; anterior corolla segmen
its ovate to deltoid S.<
2. Posterior corolla segment deeply bifid at apex; anterior corolla segments lanceolate to oblong-lanceolate
. Leaves mostly in terminal rosette, with some scattered along stems
ACKNOWLEDGMENTS
The authors are grateful to the Directors of Agharkar Research Institute, Pune and Botanical Survey of India,
Kolkata for facilities. One of us (SAP) is thankful to the Joint Director, Botanical Survey of India, Western
Circle, Pune, and K.P.N. Kumaran, Agharkar Research Institute, Pune for encouragement and Karnataka
Forest Department for granting permission for undertaking field studies in Anshi National Park, Karnataka.
Help rendered by Katherine Challis and Wilmot-Dear (Melanie Thomas), Royal Botanic Gardens, Kew, and
M.K. Vasudeva Rao, Ex-Joint Director, Botanical Survey of India, Pune for the Latin diagnosis is gratefully
acknowledged. Thanks are also due to Lourdes Rico, Royal Botanic Gardens, Kew for the Spanish abstract.
We thank Juliet Wege, E.S. Santhosh Kumar, and an anonymous reviewer for providing helpful suggestions.
Thanks are also due to Robert Vogt and Marion Cubr, Botanic Garden and Botanical Museum Berlin-Dahlem,
Freie Universitat Berlin, Berlin, Curator and V.P. Prasad, Indian Liaison Officer, Royal Botanic Gardens, Kew,
Adele Smith, Royal Botanic Garden, Edinburgh, Curator and Sovanmoly Hul, Museum National d’Histoire
Naturelle, Paris, M.S. Mondal, and S.K. Srivastava, Central National Herbarium, Howrah, for sending the
scanned type images and microfiche of Stylidium tenellum and its synonyms and various other species.
REFERENCES
Bean, A.R. 2000. A revision of Stylidium subg. Andersonia (R.Br. ex G. Don) Mildbr. (Stylidiaceae). Austrobaileya
5:589-649.
Erickson, R. 1958. Trigger plants. Perth: Paterson Brokensha Pvt Ltd.
Mildbraed, J. 1908. Stylidiaceae. In: Engler, A., ed. Das Pflanzenreich. Vol IV, 278, Heft. 35. Wilhelm Engelmann,
Leipzig.
Mitra, S., S. Bandyopadhyay, and A.K. Sarkar. 1 998. Study on the genus Stylidium Sw. ex Willd. (Stylidiaceae) of India.
J. Econ. Taxon. Bot. 22:643-646.
Punekar S.A. 2007. An assessment of floristic diversity of Anshi National Park, Karnataka, India. Ph.D. Thesis, Uni-
versity of Pune, Pune (unpublished).
Santapau, H. and A.N. Henry. 1973. A dictionary of flowering plants of India. CS.I.R., New Delhi.
Slooten, D.F. van. 1 954. Stylidiaceae In: Steenis, C.G.GJ, ed. Flora Malesiana, Ser. 1 , 4:529-532. Djakarta: Noordhoff-
BOOK REVIEW
STUDIES IN CAPPARACEAE XXVII:
SIX NEW TAXA AND A NEW COMBINATION IN QUADRELLA
Xavier Cornejo
Hugh H. litis
The New York Botanical Garden
200th St. and KazimiroffAve.
Bronx, New York 10458-5126, USA
rnejo@nybg.org; xcomejoguay@gmail.cc
Department of Botany
University of Wisconsin
430 Lincoln Drive
Madison Wisconsin 53706, USA
swos@charter.net; tscochra@facstaff.wisc
ABSTRACT
the Yucatan Peninsula; (6) Q. morenoi Cornejo & litis, a new species described from Nicar;
southwestern Mexico to northern Costa Rica; and (7) Q. morenoi forma hastate litis, a juv<
e Capparaceae, the gent
RESUMEN
Dominicana; (2) Quadrella dre
ro Quadrella (DC.) J. Presl (Cappara
Quadrella (DC.) J. Presl (Capparaceae s.s.), is an American genu
and from Florida and the Bahamas through the West Indies I
Cornejo 2010 a, b). The six new taxa we discovered during ou
paraceae for Flora Mesoamericana (in prep.); are described as
1- Quadrella alaineana Comejo & litis, sp. nov. (Fig. 1) Tvr
Azua, sandy soil, in thickets, 22 Aug 1964 (fl), Bro. Basilio Augisto Lavai
is that now comprises 25 species, 9 subspecies
ribbean Sea, from western Mexico to Panama
to northern Colombia and Venezuela (litis &
ir revision of Quadrella for a treatment of Cap-
; follows.
>e: DOMINICAN REPUBLIC [Hispaniola], Azua: Monte Rio
stre 1730 (Hotorrrc: JBSD; isotypes: NY, W1S).
Shrubs ca. 2 m tall, slender and much branched, evergreen, pubescent throughout, with dense, short, deli-
cately branched candelabra trichomes with yellowish brown centers and whitish arms. Leaves alternate to
spirally arranged; petioles 0.6-1.5 cm, canaliculate; blades conduplicate when young, at maturity ovate to
elliptic or slightly obovate, 3.4-8 x 1.3-3.7 cm, thinly coriaceous or chartaceous, dull, green to brownish,
densely white-stellate and glabrescent, leaving adaxially abundant, microscopic dark to light dots, light green
or brown abaxially, with abundant short candelabra trichomes not soft to the touch; apex acute to obtuse
or rounded and minutely apiculate, sometimes emarginate, the base rounded, truncate to inconspicuously
retuse; lateral veins 7 to 10 pairs, upwardly arching, the tertiary veins laxly reticulate, both prominent
J- B«t Res. Inst. Texas 4(1): 75 - 91. 2010
76
1.5-2.7 cm, «
7-8.5 x 4-5
°f Q- 1
Iressleri, a detritophilous, humilectic, Panamanian endemic. A. Dense, leafy crown of this single-stemmed treeletformsa'
nd animal detritus decomposes into humus soon invaded by adventitious, mycorrhizal roots from the stem. Panama: Col
ar Agua Clara. B. A 4- to 6-flowered, sub-umbellate-corymbose raceme, of which all flowers open at once. C Close-up
small white spots and lepidote abaxially, the apex abruptly acuminate to caudate-acuminate or gradually
long-acuminate, the contracted tip (l-)2-7 cm, often ending in a hair-like extension of the midrib, the bas*
narrowed to 4-8 cm, then rounded or truncate-retuse to subcordate or rarely cordate; lateral veins 9 to lj
(to 20), upward curved and strongly ascending. Inflorescences paniculiform, initially probably supra-axiM
in, and projecting from, the leafy crown, but at maturity cauliflorous and projecting sideways to downward
branch, that is basally unbranched, bractless, sterile and terete for 1-20 cm, distally and laterally bears up
to ca. 50 distichous, subulate, somewhat complanate and upward-curved persistent bracts, 3-4 mm long
and mostly arranged 4-6(-15) mm apart, with each bract subtending a densely brown lepidote subumbel-
late raceme, 2-5(-8) cm, on pedicels (0.5-)0.7-1.3 cm, and 2 to 8 flowers near the tip, all of them opening
at once; flower buds spherical, 4-6 mm diam. Up to 50 racemes are produced sequentially, alternately left
and right, and, if not fertilized and developing into fruit, are soon deciduous. Sepals 1-seriate, triangular,
cream or light-yellow, somewhat cucullate and strongly curved inward, divergent at anthesis (Stapf, pers.
comm.), brown lepidote abaxially, glandular adaxially; stamens ca. 16 to 20, the filaments clavate, (4-)7-8
mm, glabrous to lightly stellate at base, the anthers ca. 1-1.2 mm, basifixed; gynophores absent, or at most
only 1-3 mm, the ovaries ca. 1-3 mm, densely lepidote-stellate. Infructescences with pedicels 0.7-1.6 cm,
the gynophores suppressed, the fruit pendulous, linear, 10-35 x 0.8-1. 1 cm, strongly constricted between
the (4 to) 12 to 17 seeds, tapering at both ends, densely brown or brownish green lepidote without, bright-
orange within, at maturity dehiscent, splitting lengthwise along one suture, the seeds dangling from slender
funicles, oblong, purplish or black, ca. 13-14 x 5 mm, immersed in an orange endocarp and surrounded
by a bright orange aril; embryo green.
In its detritophilous habit, with the leaves densely arranged in pseudoverticils at the end of the branches
collecting humus, the main stem with adventicious roots, and the flowers without androgynophores, Qua-
drella dressleri resembles Q. antonensis, the more common and similarly detritophilous allopatric species in
Panama. However, Quadrella dressleri differs from the latter by the larger leaf blades, (25-)30-70 x 8-20
cm (vs. f 12-J15-35 [-40] x 3-10 [-11] cm), and by its paniculiform inflorescences, bearing to 2nd order
bracteate determinate, soon deciduous few-flowered racemes (vs. inflorescences racemose, multi-flowered,
elongated, indeterminate), the persistent bracts of the inflorescences (Fig. 2A, vs. bracts soon deciduous);
and the longer filaments, 7-8 mm (vs. 4-5 mm). In addition, a very different floral behavior has also been
Journal of the Botanical Research Institute of Texas 4(1)
observed between the two species in the field. In Quadrella dressleri, the relatively few flowers of each short
raceme are blooming at the same time (Fig. 2 B), but Q. antonensis has acropetal racemes, each sequentially
opening only 1 to 3 flowers at a time of some 100 or more flowers per inflorescence (Stapf, pers. comm.;
see also SSnchez 2001). On account of the stellate pubescence, and the long fruits bearing seeds with green
embryos, this new species is placed in Quadrella subg. Breyniastrum (DC.) litis (litis & Comejo 2010b). |
cloud forests of the Caribbean slopes of eastern Panama, north of the continental divide and east of the
Panama Canal, in the provinces of Colon, San Bias and Panama, at 50-500(-850) m.
Phenology. — Flowering and fruiting apparently throughout the year, but on the basis of collections here
examined, Quadrella dressleri seems to flower mostly between March and November, and is in fruit more
so from September to January (April). The diurnal flowers have a very slight fragrance; between ca. 9:00 to 1
10:00, they are visited by beetles of Chrysomelidae Latreille (Stapf, pers. comm.).
Germination. — In its natural habitat as well as in laboratory testing, the seeds of Quadrella dressleri
germinate between 3 to 4 days (Stapf, pers. comm.).
Vernacular names. — PANAMA. Basurera (Spanish, for trash basket), a name also used for Q. antonensis
(Stapf, pers. comm.).
Il/CN. — Due to the detritophyllous habit, Quadrella dressleri is not able to live in optimal conditions
or even survive in open places after deforestation (S&nchez 2001). This new species is assigned to EN
BlabGii).
Etymology. — The species is named for Robert Dressier, the great neotropical Orchidologist and expert on
Panamanian Flora, who first noticed the distinctive characters of this remarkable species (Dressier 1985). :
Trail, S Mori & M. Crosby 6337 (MO). Comarca de San Bias: Cerro Brewster, G. de Nevers, A. Henderson. H. Herrera. G. McPherson & L.
Nergala, G. de Nevers, H. Herrera 6-5. Charnley 6528 (MO); quebrada affluente de Rto Diable, Cordillera frente a Isla NarganS, C. Galdames,
E. Montenegro. C. Chung. H. Herrera 1558 (PMA, SCZ); Cerro Habu, K. Sytsma, T. Antonio, R. Dressier 2658 (MO); trail to Cerro Obu [Habu], ,
G. de Nevers. H. Herrera & E. Gemado 7970 (MO). Panama: El Llano-Carti rd., 10 km from Inter-Amer. hwy, S. Mori &J. Kallunki 2318 }
(MO, PMA, W1S); El Llano-Carti rd., 10-12 km from jet. w. Inter-Amer. hwy, S. Mori & J. Kallunki 2855 (WIS); ca. 12 km N of El Llano, El
Llano-Carti rd.. R. Dressier 5920 (WIS [2]); ca. 13 km N from Pan-Amer. hwy, B. Hansen&S. Mori 2996 (WIS [2, wood at USDA, MAD]); |
Kuna Yala, Nusagandi, sede de campo de PEMASKY, R. Paredes, R. Foster, R. Perez, S. Aguilar, Z. Batista, R. Mihalik, A. Salywon 899 (F, J
PMA, SCZ); Gorgas Memorial Labs, Camp. Quatro, 5-10 km NE of Altos de Pacora, S. Mori & J. Kallunki 3409 (MO, US).
3a. Quadrella incana (Kunth) litis & Comejo subsp. incana, Novon 17:452. 2007. (Figs. 4 A-D, 5 A,
For a complete description of this species see litis & Comejo (2007: 452).
Habitat and Distribution.— In dry woodlands, from southwestern to eastern Mexico, where very common,
rarely into southeastern Texas, east to Guatemala and rarely into Honduras (litis & Comejo 2007: 453). |
3b. Quadrella incana var. triangularis Comejo & litis, var. nov. (Figs. 5 C, D, 9 lower map) Type MEXICO
Beach, ca. 8 km NW of Chamela, 19°33'N 105°08’W, 2 m, 22 Jun 1984,
Sepals triangular, flat, minute, 1-1.5 x 1-1.3 mm (Fig. 5 C-D), ascendent at anthesis, each subtending a
H.H. litis & S. Wisniewski 29194 (holoiype WIS [fl]; isotype: MEXU [stl).
sepals. D. Flower at postanthesis.G-1. Quadrellamorenoi. G. Showing large, solitary, straight, green capparoid embryos, enclosed by a thin, fragile testa
(I) enclosed by a ± thickened fruitcoat. A, B: Moreno & Robleto 22899 (WIS). C, D: litis & Wisniewski 29194 (from the holotype, WIS). E, F: Johnson 4370
(WIS). G: Espinosa 462 (WIS). H-fc King 239 (WIS). J, L Quern 2983 (WIS). K: Pringle 7291 (WIS).
Journal of the Botanical Research Institute of Texas 4(1)
splitting lengthwise along one suture, exserting the orange-arillate seeds on the bright orange pulpy endocarp J
of their valves; seeds cochleate-reniform, 5-8 x 4-5 mm.
Quadrella lindeniana is vegetatively very similar to the allopatric Q. indica, but differs by having distinct
tively smaller flowers (Fig. 7 B, C), with petals only 5-7 x 4-5 mm (vs. 8-14 x 6-8 mm), filaments only 7-10
mm (vs. [15-118-30 mm), gynophores only 2-7 mm (vs. [12-117-40 mm) in flower, and 3-9 mm (vs. 12— 45 1
mm) in fruit, ovaries 2-3 mm (vs. 3-5 mm) and fruits narrower, 4-7 mm (vs. [6— 17— 12[— 14] mm). In addi-|
tion, the inflorescences of Quadrella lindeniana are truly corymbose or even umbellate, having fewer flowers|
([1]4 to 10) than those of Q. indica, which has the inflorescences distinctively more racemose-umbellate tM
racemose-corymbose, bearing more flowers (8-14[-20]).
Habitat and Distribution. — Quadrella lindeniana is restricted to Mexico’s Yucatan Peninsula, where iffl
occurs from the landward mangrove fringes in the far north, through dry thorn scrub to subdeciduousf
lowlands and subevergreen forests in the south, often growing on coastal dunes or rocky soils, and persisting;;
in disturbed forests. It is allopatric to the similar but widespread Q. indica, which ranges from the Mexican
states of Sinaloa and Nayarit to Costa Rica, mostly in the dry Pacific coastal areas, and from the central West ,
Indies to Colombia and Venezuela (Fig. 6).
Phenology. — Flowering from April to June, and fruiting from July into September.
Vernacular names.— MEXICO. Negrita (J.D. Shepherd 180), salvo ( Zamora 4795).
1UCN. — Although Quadrella lindeniana is restricted to the Yucatan Peninsula, the many fertile collecl
tions suggest however, that there are still a reasonable number of healthy populations of this new species,
thus Q. lindeniana deserves a lower priority, LC.
Etymology. The earliest collections of Quadrella lindeniana were apparently made by a youngjeanjules'j
Linden (1817-1898) from the “environs de Campeche,” when, in May of 1838, he collected the species in.
full bloom with the comment (from the French), that the flowers were white and fragrant. Labelled by hand
simply as Capparis indica Linden 999" in the Kew and Florence Webb herbaria (now included at FI), but as j
85
87
Journal of the Botanical Research Institute of Texas 4(1)
Cornejo and litis. Six new taxa and a new combination in Q
a unique, translucent, thin, papery or membranous testa; by inflorescences borne on much longer peduncles
(Fig. 8, vs. 0.4-2. 5(-3) cm in Q. incana ); and by the greater number (7 to 16 vs. 3 to 8) and denser pattern
of the lateral veins. This new species is placed in Quadrella subgen. Breyniastrum (litis & Cornejo 2010 b),
on account of the stellate pubescence covering the plant throughout and its seeds with green embryos with
Quadrella morenoi existed unrecognized for 189 years (since 1821) as part and parcel of the superficially
similar Q. incana, so in Flora of Guatemala (Standley & Steyermark 1946), and the Flora Arborescente de Costa
Rica (Zamora 1989), but also in the Flora de Nicaragua (litis 2001), where nevertheless the often solitary seeds
and green color of the embryos were first mentioned but not discussed.
Habitat & Distribution. — Quadrella morenoi occurs scattered from southern Mexico (Michoacan) to
southwestern Costa Rica (Fig. 9, upper map), but is especially common in Central Nicaragua. Only at Tehu-
antepec in southern Oaxaca, Mexico, this new species overlaps parapatrically the range of the more common
Q. incana. Quadrella morenoi ranges from sea level to 930 m in deciduous thorn scrub, gallery forests and
semi-deciduous tropical dry forests and woodlands. It persists in secondary vegetation and grazed areas.
Vernacular names.— MEXICO (Oaxaca): Arnica (Salas et al. 1310); mata gallina ( Cervantes 2168).
Phenology. — Flowering mostly from February to July (rarely in November), and fruiting mostly from
May into September (rarely in February).
Seed morphology. — Unlike the one to six seeds of Quadrella incana fruits and those of the related Q.fer-
ruginea (L.) Rafinesque complex of the Greater Antilles (Rankin & Greuter 2004), which have their white
embryo enclosed by a smooth, relatively thick, hard testa typically cleomoid in shape (i.e., with the seed
cleft deeply invaginated between the claws of the cochleate-reniform seed; Fig. 5 J-L), the one to two seeds
of Q. morenoi (Fig. 5 G-I) have an essentially straight and non-cleomoid green embryo covered by a unique,
removed. While the embryos of Quadrella incana are whitish or cream, with their very thin cotyledons doubly
conduplicate, the deep-green cotyledons of Q. morenoi are thicker, somewhat irregularly but simply folded
one into the other.
Etymology. — The species is named in honor of Pedro Paul Moreno, the energetic Nicaraguan botanist
at HNMN, an avid explorer of the Nicaraguan flora and collector of the type specimens.
IUCN.— The numerous and relatively recent collections of Quadrella morenoi indicate that this species
deserves a lower concern, LC.
Journal of the Botanical Research Institute of Texas 4(1)
f. Type: NICARAGUA: vie. Hac. San Jacinto, I
, 14 Oct 1991 (st), H.H. Utis&P. Anderson 3
Juvenile branches of a few New World Capparaceae occasionally produce radically different, hastate leaves,
in the case of Q. morenoi f. hastata, with the elongated, 4-15 cm long and 0.4-0.7 cm wide, slender and!
narrowly-oblong central lobe abruptly diverging at base into two, ± equal, laterally-projecting, rounded basal •,
lobes, these together 1.2-4 cm wide from side to side, and with the whole leaf with as many as 17 main!
lateral veins on each side of the midrib.
It is of great interest that, while the eight stamens and short ovoid fruits of Quadrella morenoi suggest|
a close relationship to the eight staminate, short- and ovoid-fruited Q. incana and the other species of Qud-1
drella subg. Intutis, its green embryos rather than white, and the frequent occurrence of hastate, juvenile!
stump sprout leaves (which never occur in Q. incana), are more congruent with the similarly eight staminate, |
green-embryoned Q. lundellii (Standley) litis & Cornejo, a co-member of Quadrella subg. Breyniastrum, for
which, despite its elongate capsules, its hastate juvenile leaves provide strong reinforcing evidence for this
odd, putative relationship. Another species of subg. Breyniastrum with eight stamens, Quadrella steyermarkm
(Standley) litis & Cornejo, nevertheless, does not produce juvenile hastate leaf blades, but rare as it is, they •
may yet turn up!
Finally, the occurence of these peculiar leaf forms in Quadrella morenoi is reminiscent of the analogous|
situation in the distantly related Cynophalla hastata (Jacq.)J. Presl. However, it is important to note that there
is no close relationship whatever between these two genera of American Capparaceae.
Paratypes: MEXICO. Michoacan: 4 mi NW of ApatzingSn, R. McVaugh 17924 (MICH, WIS); 13 km W of ApatzingSn J. Rzedomki 2230$
(ENCB, MEXU, WIS). Oaxaca: Chivela, C. Mell 35 (US, US fragm. at WIS); Mun. Santo Domingo Tehuantepec, Ejido El Lim6n, A. R eyes-
Garcia ,J. Gordon & I. Sdnchez 3305 (MEXU). GUATEMALA. Zacapa: vie. Zacapa, P. Standley 74179 (F), P. Standley 74624 (GH, F [21, NH
P Standley 71969 (F [21, US). NICARAGUA. Boaco: 1 km E de Santa Cruz, W. Stevens 22916 (WIS). Hac. San Antonio, carr. a BoaquiW,|
P Moreno 21545 (MO). COSTA RICA. Guanacaste: Parque Nac. Santa Rosa, Sendero Carbonal, A. Ferndndez 296 (INB).
ACKNOWLEDGMENTS
Many thanks to the Missouri Botanical Garden and its Director of Latin American Botany, Olga Marthli
Montiel, for partial financial support to the authors for a treatment of Capparaceae for Flora Mesoamericanai j
to Molly Harker for her collaboration with the 1BUG specimens of Quadrella incana; to Sharyn Wisniewski!
life companion of the second author and field companion on the 1984 collecting expedition as well, who j
gathered the type collection of Quadrella incana van triangularis; to Pamela Anderson, who guided the second
author on his Nicaraguan excursion; to Maria (Sanchez) Stapf (STRI), for sharing her field observations <|
Quadrella dressleri; and to Victoria Hollowell, Gordon Tucker and a anonymous reviewer, for many helpfoM
stylistic comments on the manuscript. Our special appreciation, however, goes to Kandis Elliot, artist ®
residence in the UW Botany Department and Claudia Lipke, head of the Photo laboratory at the UW BotaflH
Department, for preparing the excellent illustrations, which have greatly enhanced this publication. ||
Journal of the Botanical Research Institute of Texas 4(1)
BOOK REVIEW
S.M. Walters and E. A. Stow. 2009. Darwin’s Mentor: John Stevens Henslow, 1796-1861. (ISBN 9781
0-521-11799-9, pbk.). Cambridge University Press, 100 Brook Hill Drive, West Nyack, New York!
10994-2133, U.S.A. (Orders: www.cambridge.org, 1-845-353-7500). $36.99, 338 pp. including 77J
pp. of appendices, endnotes, bibliography, and index, 6 3/4" x 9 5/8".
it Res. Inst Texas 4(1): 92. 2010
STUDIES IN CAPPARACEAE XXVIII:
THE QUADRELLA CYNOPHALLOPHORA COMPLEX
Hugh H. litis
University of Wisconsin
430 Lincoln Drive
Madison Wisconsin 53706, USA
Xavier Cornejo
The New York Botanical Garden
200 * St. and KazimiroffAve.
Bronx, New York 10458-5126, USA
ABSTRACT
plex, variously branched or fused trichomes, was recently resuscitated from the breakup of the New World Capparis L„ s.l. Within Quadrella,
RESUMEN
Quadrella (DC.) J. Presl (1825) es un genero (Capparaceae) de las Indias Occidentales y Mesoam£rica, recientemente resucitado de Cap-
Quadrella, el complejo lepidoto o peltado Q. cynophallophora (L.) Hutchinson estd compuesto por dos grupos de especies y subespecies
The almost exclusively West Indian-Mesoamerican genus Quadrella (DC.) J. Presl (1825), emended herein,
is one of only two stellate to lepidote or peltate pubescent New World genera of Capparaceae with dehis-
cent fruits. Quadrella, a genus of 25 species, is characterized by a valvate calyx with four sepals in a single
scries. These may be 1), either very small and triangular with an open calyx aestivation long before anthe-
sis in Quadrella subg. Breyniastrum (DC.) litis and half of the species of Quadrella subg. Intutis (Raf.) litis,
( Quadrella domingensis [Sprengel ex DC.] litis & Cornejo, Q. ferruginea [L.] litis & Cornejo, and Q. incana
[Kunth] litis & Cornejo); 2), or with a closed calyx aestivation, i.e., temporarily fused and closed over the
petal buds until or nearly to anthesis in subg. Quadrella and in the remaining species of Quadrella subg.
Intutis (Q. angustifolia [Kunth] litis & Cornejo, Q. alaineana Cornejo & litis & Q. singularis [R. Rankin] IltisJ
& Cornejo) [litis & Cornejo 2010]).
Our view of the systematics of the Quadrella cynophallophora (L.) Hutch, complex, a taxonomical and 1
nomendaturally difficult group, is to place heavy emphasis not only on morphology of leaves, flowers and
fruits but also on the mostly allopatric geographic distribution of its taxa. Thus, our systematic viewpoint i
is the exact opposite of that of the first author’s former major professor, the late Robert E. Woodson, who
in his treatment of Capparaceae for the Flora of Panama (1948) lumped all four of the then recognized taxa
into one, all-encompassing Capparis cynophallophora without any further discussion. We, on the other hand,
recognize the morpho-geographic reality and split this complex into several allopatric species and subspecies
(Fig. 1), based on either Quadrella cynophallophora or Q. isthmensis. An exception lies in the Q. jamaicensis ^
J. Presl-Q. cynophallophora geographic overlap in Hispaniola, and rarely with a Q. jamaicensis plant seen in
the Bahama Islands, both of which are recognized herein as separate species based on their characteristic
leaf shapes.
TAXONOMIC TREATMENT
Quadrella (DC.) J. Presl in Berchtold &J. Presl, Prir. Rostlin 2:260. 1825.
1. Capparis subg. Quadrella (1
►e designated by litis & Cornejo 2<
KEY TO THE QUADRELLA CYNOPHALLOPHORA COMPLEX
1. Ovaries at anthesis 3-6.5 mm; sepals 5-1 0(-1 1) mm; receptacle in fruits (2— )3— 5 (-6) mm wide; stamens 16
to 35 (-40); leaf blades mostly lustrous above, usually uniformly lepidote or peltate, green or brownish-green
beneath; Florida, through the West Indies to Trinidad and Barbados, disjunct to Mexico's Yucatan
Peninsula (Cozumel Island and Quintana Roo) I. Cynophallophora alliance (five species) i
2. Flowers large, with petals 8-1 7 mm and sepals (5— )7— 1 1 mm; stamens with simple or few-branched stel-
late trichomes at filament bases; anthers (2-)3-4 mm; southern Florida, West Indies, and in Mexico from
3. Gynophores conspicuous, 10-50 mm in flowers, to 80 mm in fruits; filaments 20-50 mm; petals 10-17
mm; common, the West Indies and southern Florida.
4. Leaf blades 3-8(-1 0) x 1 — 4(— 4.5) cm, elliptic to narrowly obovate or oblong, rarely lanceolate (linear-
lanceolate [to 1 7 cm] in juvenile leaves or stump sprouts), stiffly coriaceous and revolute, lustrous
above; petioles 1 -2 cm; seeds 5-8 x 4-6 mm, packed into mostly barely torulose capsules; through-
out the West Indies and southern Florida, in Jamaica mostly in coastal lowlands.
5. Leaf blades ± elliptic, mostly 5-1 0 x 2-6 cm, with the apex acuminate or acute, sometimes obtuse
to rounded but then often apiculate and not emarginate; central and eastern West Indies (Baha-
mas and Flispaniola to Puerto Rico, Trinidad and Barbados, sympatric with the following in ' '
Hispaniola) 1. Quadrella cynophallophora (L.) Hutchinson
5. Leafblades mostlyoblong -r I |. .i,.| ..-lln-rr m,. .,ri, m ,\irh th- ape- usu-
ally emarginate (notched) to rounded: southern Florida, Cuba, coastal Jamaica, Cayman and Swan
r preceding species in Flispaniola, rarely in the Bahamas 2. Quadrella
jamaicensis (Jacq.) J- WjjM
4. Leaf blades 7-1 5(— 20) x 4-9 cm, ± broadly elliptic, ± chartaceous, rather opaque above, on petioles
(1 ,5-)2-4 cm; seeds 7-1 5 x 6-9 mm, spaced out in strongly torulose capsules; highlands of Jamaica ^
3. Gynophores in flowers and fruits highly reduced, to 5 mm or less or lacking; filaments 6-1 5 mm; petals
8-9 mm; endemic to Mexico (Cozumel Island and Quintana Roo, northeastern Yucatan Peninsula)
icate, with petals 6-7 mm and
lepidote-stellate (many-branched) trichomes a
eastern half of Flispaniola (Haiti and eastern-mo:
1. Quadrella quintanarooensis litis & Com# ;
nm; stamens with densely lepidote-rr
bases; anthers 1-2 mm; very rare, scattered
:an Republic) 5. Quadrella gonaie
(Helwig) Hutd
. Ovaries at anthesis 6.5-1 0 mm; sepals (9-)1 0-1 8 mm; receptacle in fruits 5-1 0 mm wide; stamens ca. 30 to
60; leaf blades opaque above (except Quadrella isthmensis subsp. mexicana), silvery lepidote-peltate beneath,
with scattered darker bronze hairs giving the lower surface a peppered appareance; Panama and Costa Rica,
disjunct westward to the central Yucatan Peninsula (northern Guatemala and Belize), and north to the
Caribbean coast of Mexico's Yucatan I. Isthmensis alliance (one species and three subspecies)
6. Petals densely lepidote without, sepals tomentose within; eastern Mexico (northern Yucatan and Campeche)
south into Guatemala and Belize, disjunct to Costa Rica and lowlands of Panama.
cm, on gynophores 4-8 cm and pedicels (1 ,3-)2-4.5 cm; leaf blades dull above when dry; Costa Rica
and lowlands of Panama 6. Quadrella isthmensis subsp. isthmensis (Eichler) Hutchinson
7. Flowers smaller, with petals 9-1 3 x 6-7.5 mm, stamens 2-4M.5) cm and gynophores 1 -4.5 cm; fruits
8. Quadrella isthmensis subsp. glabripetala Cornejo & litis
la. Quadrella cynophallophora (L.) Hutchinson, Gen. Fl. Pi. 2:309. 1967. (Figs. 1-3). Bas.onym: Capparis
t vnophallophont L . S|. PI cd I -'•04 17 V. l,u IAMAIC.V Lcll Iwnd spa mien V.if’/’iMs?. llnrius i. Iill..riumis Hh (ih mrm.
designated here, BM-628728).
Shrubs or trees to 5(-10) m tall, to 40 cm dbh, rusty-brown peltate-lepidote throughout, the branchlet tips
and peduncles flattened. Leaves ± coriaceous, evergreen, dark-green, often lustrous and glabrous above,
densely covered with light golden brown (sometimes with scattered darker, rusty-brown) peltate-lepidote
hairs beneath, ± elliptic to obovate, rarely lanceolate, acuminate or acute to rounded, often apiculate, cuneate
to widely rounded at base (excluding the linear [to 15 x 2 cm] juvenile or stump sprout leaves), (3-)5-10 x
1.5-4.6 cm, the midrib sulcate above, the secondary nerves inconspicuous; petioles 0.5-2(-3.5) cm, canali-
culate. Inflorescences terminal in leaf axils at ends of branches, subcorymbose, 1- to 7-flowered, the linear
bracts ca. 2 mm, soon caducous; peduncles and stout 7-40 mm pedicels flattened, rusty-brown, densely
lepidote; flower buds ovoid to lanceolate, pointed, longitudinally ± 4-ribbed; sepals at anthesis reflexed,
ovate to lanceolate, 5-1 1 x 3-6 mm, densely rusty lepidote without, tomentose to tomentulose within; nectar
scales flat, 1— 2(— 3.5) mm; petals divergent at anthesis, creamy white, soon fading (as do the stamens) to a
pink or dark purplish violet, elliptic-obovate, about as long as the sepals, usually densely-lepidote without,
glabrous within; stamens 20-35(-40), the filaments 2.5-5 cm, densely pilose with simple, usually white
trichomes at the very base, the anthers 2-3.5 mm; gynophores 1.5-5 cm, glabrous, on androgynophores 1-1.5
mm; ovaries linear-cylindric, 3-7 mm, densely peltate-lepidote, the stigma truncate. Fruits linear-cylindric
± torulose siliquiform capsules, 4-40 x 0.4-1 cm, brown, densely lepidote, ± irregularly rupturing along
(usually) one suture, the valves turning inside out to expose their bright orange to scarlet pulpy endocarp
and the embedded bright orange to reddish arillate seeds, these often dangling from the replum by a thin
funicular thread; gynophores (l-)1.6-8 cm, glabrous, clearly demarcated from fruit valves; receptacular disks
in fruit, (2-)3-5(-6) mm diam.; pedicels 1.5-4 cm, lepidote; seeds (2— >5 to 40, oblongoid, 5— 8[-10 mm in
Hispaniola] x 4-6 mm, with a bright orange aril and thin testa, the embryo green throughout, much folded.
The lectotype of Quadrella cynophallophora was selected by Fawcett and Rendle (1914a: 142) with the
statement that “ Capparis cynophallophora L„ Sp. Pi. 504 is based on the plant (sic!) Capparis 2 of Hortus Clif-
fortianus 204: Linnaeus merely repeats the diagnosis from the earlier work.” However, unfortunately, Capparis
2 in the Hortus Cliffortianus Herbarium (Fig. 2) is a mixed collection, with the left hand specimen a branch
of Quadrella cynophallophora labeled “ Capparis 2” [and which we explicitly designate as the lectotype of that
species], and the right hand branch clearly belonging to Quadrella jamaicensis (Jacq.) J. Presl (small, narrow,
emarginate leaves, etc. . .), and simply another collection of that common coastal Jamaican species. No mat-
ter what Linnaeus (Anonymous 1933: 204) may say in Hortus Cliffortianus (“Capparis. . .foliis, ovalibus. . .”),
he cites as his only reference Plukenet “Aim. 126. 1. 172, f. 4” with “. . .foliis subrotundis. . .” from Barbados,
this a drawing with typical Q. cynophallophora leaves.
h the left hand specimen the lertotype of Quadrella cynophal-
lophora, and the right hand specimen a typical branch of Quadrella jamaicensis. Legend at bottom of left side, *p. 204 Capparis 2" refers to page 204
s “Capparis jamaicensis Jacq." on the top line and “Capparis
Habitat and Distribution. — Central and eastern West Indies, from the Bahamas to Hispaniola (where
it intergrades with Quadrella jamaicensis) and to the southern coast of Puerto Rico and through the Lesser
Antilles to Trinidad and Barbados (Gooding et al. 1965), mostly in coastal, seasonally dry, evergreen wood-
lands, but also in a great variety of habitats, and often cultivated as hedges.
Phenology. — Flowering from April to January, fruiting throughout the year.
Vernacular names.— BARBADOS: Black Willow (E. Gooding 39, W1S); DOMINICAN REPUBLIC: Frijol
(Valeur 403, WIS), Frijolillo (J. Schiffino 167, WIS), Olivo (J. Schiffino27, W1S).
Selected spec, mens US Virgin Islands: s. Croix. Big fountain. .4 Ruksecke, (MU). St Thomas. Water Bay. £ggm 417 (CAS). St John.
QUi>ner’ R Aceved°-Rodrigu'Z et al 1821 (W[S) Puerto Rico: El Tuque, near Ponce, H. Alain & P. Uogcr 31206 (UPR); Salinas
Foresial Guanu a. H /In, VMHW.B- Dominican Republic: m, trade Nciha Prov lndependcncia. Cerros del Cao. T Zunon. rt ul 24814
''A lN IVnui'ul.i de Barahona. Prov Pcdcrnalcs. S km Mir del puortode Cabo Ro|o do Alcoa Expl Co . T Zanom&J hmentel 26337 (WISi
Guadeloupe: Grande Terre. L'Auter Bord distr., E of Le Moule, G. Proctor 19916 (WIS). Antigua: St. Paul, Shirley Heights, G. Webster
D598 (WIS). St. Maarten*: near top of Sentry Hill, A. Staffers 4526 (A). Barbados: Newton, Christ Church. E. Gooding 39 (WIS).
lb. Quadrella cynophallophora (L ) Hutchinson, f. linearifolia litis, t
Virgin Islands: St. John, East End Quarter, Hansen Bay, 0-10 m, dry scrubby coastal
Rodriguez, A Reilly & M. Davis 1808 (holotwe: NY, NY photocopy at WIS).
Distribution. — This form is found aero
2a. Quadrella jamaicensis (Jacq.) J. 1
the range of Quadrella cynophallophora.
Shehbaz in Flora of the Lesser Antilles-Dicotyledoneae 4(1):296. 1988).
Shrubs to small trees, peltate-lepidote throughout [except for the narrower, smaller leaves, very similar ttfl
Quadrella cynophallophora]. Leaf blades mostly oblong to oblong-elliptic, 3-9(-ll) x 1-3.5 cm, with apex emar- 1
ginate or notched to rounded, hard, shiny above, margin revolute; petioles 0.7-1 .6 cm. Seeds 5-7 x 4-5 mm. |
Habitat and Distribution. — This common coastal species is found in the United States in Florida (froml
Cape Cafiaveral on the east coast and St. Petersburg Beach on the west coast, south to Key West), Cuba!
coastal Jamaica, and the Cayman and Swan Islands, intergrading with Quadrella cynophallophora on His*
paniola, and rarely in the Bahamas.
Phenology. — Flowering from March to August, fruiting from June to October.
Vernacular names.— FLORIDA. Jamaica caper (D. Caldwell 8779, WIS), Black wattle. Black willow. CUB«
Carbonero, Ciguarayo, Mostaza, Mostacilla, P[enis] de perro ( Roig & Mesa 1945), Moruro Prieto (Bro. Lediw
& T. Roig 11429, WIS).
Due to the ubiquitous confusion with the common Quadrella cynophallophora and the montane Q.
siliquosa on Jamaica, we have chosen to cite only a few representative collections of Q. jamaicensis across itS
distributional range.
Selected specimens: U.S.A. FLORIDA: Monroe Co.: Florida Keys, Key West, A. Curtis 204 (BM, BKL, K, UT). St. Lucie Co.: ca. 2 miSof
i Co.: Sarasota, Longboat Key, O. Lakela &D. L
1, 1J) Isle of Pines: Base of Sierra de Casas, Bro.
r 28686 (USF). CU1
in <S- E. Killip 2261 ((
). Hillsborough
t al. 4 875 (US®
ilan, ]. Jack 571®
Clough 2 (IJ). BAHAMAS. Long Is
tear Calabash Bay, R. Thorne
This strikingly different form is evidently a stump-sprout <
blades, common in this species and, in a homologous foi
but apparently lacking in Q. siliquosa. Quadrella jamaicei
tab. 327, fig. 6 p.p., the leafy branch only).
Distribution.— This form occurs across the range of Quadrella jai
rile, juvenile branch with narrow elongate tim
Iso in Quadrella cynophallophora f. linean/offitf
. longifolia was illustrated by Plukenet (16%
Fkl 3. Leaf blades shape variability in Quadrella cynophallophora on St. Maartens (of the former Dutch West Indies). Upper left: Stump sprout leaves,
Quadrella cynophallophora f. linearifolia litis, all others, typical Q. cynophallophora, one leaf/collection (ex herb. Utrecht).
Capparis torulosa Sw., Prodr. 81. 1788. Quadrella torulosa (Sw.) J. Presl, Prir. Rostlin 2: 261 . 1825. Pleuteron torulosa (Sw.) Raf., Sylva Tellur.
109. 1838. Type: Jamaica, s.d. (fr), C. Wight s.n. (neotype, designated here, S-05-9708, S photo at WIS).
Large shrubs or trees to 15 m tall, to 40 cm dbh, with dense bushy foliage. Leaves ± broadly elliptic, acute
to abruptly short acuminate to almost rounded to the tip, broadly cuneate to rounded at base, 7-15(-20) x
4-9 cm, thinly coriaceous with mostly flat, occasionally revolute margins, glabrous above, peltate-lepidote
beneath; petioles (1.5-)2-4 cm. [Juvenile leaves or stump sprouts not long and linear, as in both Q. cynophal-
lophora and Q. jamaicensis ]. Inflorescences small terminal corymbs or if lower, subterminal in the leaf axils;
peduncles complanate-angulate, 3-6.5 cm, each bearing 3 to 7 flowers; pedicels 1-2 cm. Sepals lanceolate,
7-10 x 5 mm, densely rusty lepidote without, tomentose within; nectar scales flat, one each projecting from
the sepal inner base, ca. 1-2 mm; petals 9-13 x 5-6 mm, divergent at anthesis, creamy white, densely lepidote
outside, glabrous within; filaments ca. 3-5 cm, anthers ca. 3 mm; gynophores 2-5.5 cm, ovaries ca. 4 mm.
Fruits disctinctly torulose siliquiform capsules, (3.5-)10-36 x ca. 1-1.2 cm, on gynophores (3.5-)5-8 cm;
Pedicels 1.2-2 cm; receptacles ca. 5 mm; seeds (l-)5-15(-20) or more per fruit, 7-15 x 6-9 mm, separately
spaced out, covered by a red aril embedded in red endocarp.
Both Quadrella siliquosa and Q. torulosa are based on the same polynomial, Breynia 2. Arborescens,
foliis ovatis utrinque acuminatis, siliqua torosa longissima of P. Browne, Hist. Jam. 1756: 246, pro parte [excl.
Plukenet, 1696: tab. 327, fig. 6; and in agreement with Eichler, 1865: 270 and Fawcett & Rendle 1914a: 143,
Fk. 4. Holotype of Quadrellajamaicensis f. longifolia [Swartz s.n., S): a juvenil
Quadrella jamaicensis: Jamaica coastal
Fkl s- Leaf b,ades of Quadrella on Jamaica: A. the coastal Q. jamaicensis. B. the montane Q. siliquosa.
which is Quadrella jamaicensis f. longifolia , a linear lanceolate stump sprout or juvenile form of Quadrella
jamaicensis]. For the lectotype of Capparis siliquosa L., Al-Shehbaz (1988: 296) selected “(1) siliquosa,” Lin-
naean Herbarium sheet 664.8. Unfortunately, this sheet has two specimens attached to it, neither if which
is Q. siliquosa. Since Linnaeus, in his original description of C. siliquosa (1759), cited Browne’s (1756: 246,
°P- cit.) and then later (Sp. Pi. ed. 2, 1762; ed. 3, 1764) added “Planta forte” and also its resemblance to the
Preceding C. cynophallophora and “Habitat in Jamaica” (all four of these items aplicable only to C. siliquosa ),
102
and small ovaries. B. Mature silique.
103
we must reject Al-Shehbaz’ choice of LINN-664.8. In this mixed collection, the right-hand specimen, labeled
petioled, thus not to be confused with the longer petioled Q. siliquosa). The left-hand specimen, labeled “2.
cynophallophora,” is a typical Q. jamaicensis with emarginate, notched leaves. We have therefore chosen as
a neotype Harris 9488, cited above (Fig. 6), about which there can be no argument as to its identity.
One of the pleasurable surprises of this revision was the rediscovery of the taxon now named Quadrella
siliquosa (L.) litis & Cornejo (encompassing Capparis torulosa Sw., its ancient synonym). This is a broadly
elliptic-leaved mesophytic tree, endemic to the montane woodlands and savannas of Jamaica, apparently
closely related to Quadrella cynophallophora s.s. All the lowland, mostly coastal, plants in Jamaica belong
to Q. jamaicensis, a shrub or small tree with uniformly small, hard, shiny, oblong to oblong-elliptic, most
often emarginate leaves (Figs. 2, 5A), widespread from Florida to Cuba and occasionally Hispaniola, where
it overlaps and intergrades with the more easterly Q. cynophallophora. The name Quadrella siliquosa must
now be restricted to the large shrubs to trees of the Jamaican uplands, with much larger and thinner, more
broadly elliptic leaves and almost rounded to broadly acute or abruptly short acuminate apices (Figs. 5B, 6),
leaf size and shape presumably all adaptations to the cooler, moister, and shadier conditions of their upland,
woodland habitats, as are their longer (2-4 cm) petioles. Although some specimens from the eastern range
of Q. cynophallophora may have broad leaves, these tend to be more ovate or obovate-elliptic and are borne
on shorter petioles (Figs. 2, 3), while those of Q. siliquosa are always symmetrically elliptic (Figs. 5B, 6).
The rediscovery of the name siliquosa was slow in coming. A review of many photocopies and a few
old specimens of these upland plants in the Wisconsin Herbarium pointed to their frequent identification
with the name Capparis torulosa by botanists of the early nineteenth century. This name, in turn, alerted us
to the rather infrequent, strongly torulose siliquiform capsules with more distantly spaced seeds, which, in
turn, separated from each other by conspicuous constrictions, helped explain Swartz’s (1788) choice of his
very apt epithet. Swartz’s reference then led us to Browne’s (1756) Breynia2, a Jamaican tree different from
the shrubby coastal Breynia 1 (=Q. jamaicensis), which in turn led us to the Linnaean Capparis siliquosa, also
based on the same polynomial description. Swartz (1788: 81) and Browne (1756: 246) emphasized the very
long (“longissima”), torulose [“torose”] fruits. Finally, DeCandolle’s (1824: 252) statement “in fruticetis mon-
tosis Jamaiceae. . .” clinched the identification of Q. siliquosa as the correct name, even though the immediate
continuation of Candolle’s sentence, “. . .et Barbados. . .,” pointed to an easy misidentification with some of the
broad-leaved and more lanceolate-elliptic collections of Q. cynophallophora from that most easterly island,
where apparently that species is cultivated extensively.
For the past hundred and fifty years, however, all authors, from Grisebach (1864), Eichler (in Martius,
1865: 270 (both under C. jamaicensis ]), Fawcett & Rendle (1914a: 143; 1914b: 231-232) to Al-Shehbaz (1988:
296) and Rankin & Greuter (2004), reduced both Q. siliquosa and C. torulosa to synonyms of Q. cynophallophora
without further comment, even though on all bio-geo-ecological and morphological grounds the species
are amply distinct. Only C.D. Adams, in his Flowering Plants of Jamaica (1972: 305-306), clearly notes both
the ecological and morphological differences between Quadrella jamaicensis and Q. siliquosa (both under C.
cynophallophora s.l., as applied to Jamaica).
Habitat and Distribution.— Jamaica, in upland woodland forests and often on limestone, from 500 to
2300 feet.
Phenology. — Flowering from (January) June to September, and fruiting from September to July.
Vernacular name.— JAMAICA: Zebra wood (Wright s.n., S).
In recent years, Quadrella siliquosa has rarely been collected and we are afraid, considering the human
Population explosion in Jamaica, and the dramatic local destruction of forests, that this interesting endemic
species is slowly drifting into extinction. As handsome and bushy-leaved as that tree is reported to be, it surely
deserves protection, if nothing else at least if need be by cultivation, or better still in a nature preserve.
Specimens examined: JAMAICA. Parish not specified: 1850, R.
■ ■■■ i ; v : a..' , v. vVM.chi > •• k
s 11780 (BM, CAS, GH.K
105
linear-cylindric siliquiform capsules, 6-32 x 0.5-0.9 cm, brown, densely lepidote without, bright orange
wit^in, ± torulose, on very short and thick, 1-5 mm gynophores and 0.5-2.2 cm pedicels; seeds 5-8 x 4-6
rnm> covered by an orange aril, the embryo green.
Despite the hundreds of thousands of tourists that now visit Cozumel Island and Tumul each year, it
was not until the 1980s, when the tourist industry opened up the region, that Edgar Cabrera and associates
(MEXU), discovered this interesting local endemic, which, apparently an offshoot of C. cynophallophora s.s.
and generally so identified, differs from it by the highly reduced to essentially absent gynophores in flowers
or fruits (Fig. 7B).
Journal of the Botanical Research Institute of Texas 4(1)
a rare endemic of Hispaniola.
In their variable size, acuminate outline, and the occasional peppered appearance of the peltate pubes-
cence on their underside, the leaves of Quadrella quintanarooensis are similar to the northern collections of j
the nearby but allopatric Q. isthmensis subsp. mexicana, which may well be due to some past introgression j
that may have influenced its morphology.
By its very short or absent gynophore, this species resembles somewhat the distinctive, silvery peltate
Quadrella odoratissima (Jacq.) Hutchinson (one of only three species of Quadrella sect. Quadrella not included I
in the Q. cynophallophora complex) distributed from southern Mexico to northern Colombia and Venezuela,
the Dutch West Indies and Trinidad. However, Quadrella quintanarooensis differs from Q. odoratissima by its j
longer, quadrangular, not spherical calyx, longer rusty brown fruits, wider leaf blades with flat not revolute j
margins when dry, and light brown pubescence.
Habitat and Distribution. — In open, dry, highly seasonal tropical forests and woodlands (selva baja \
mediana, selva mediana subperennifolia [Rzedowski 19781), often with Pseudophoenix sargentii H. Wend-
land ex Sargent (Palmae), Manilkara Adanson (Sapotaceae), Coccoloba P. Browne (Polygonaceae), Beaucarnet j
Lemaire (Agavaceae) and other xerophytes, on or near beaches at sea level to 50 m on Cozumel Island and :
the adjoining mainland of Quintana Roo, apparently not uncommon locally but restricted to this very small |
area on the northeastern coast of the Yucatan Peninsula of Mexico (Fig 9).
Phenology.— Flowering from April to July and fruiting from July through October.
107
1 km N Xel-ha, E. Cabrera &H.de Cabrera 7170 (MEXU, MO); 3 k
1 Tellez, E. Cabrera & L. R
6. Quadrella isthmensis (Eichler) Hutchinson, Gen. FI. Pi. 2: 308. 1967.
Shrubs to slender trees 3-10 m tall, rusty brown peltate-lepidote throughout, the branchlet tips and peduncles
flattened. Leaves thinly coriaceous, ± evergreen, dark green and glabrous above, densely covered with light
golden or silvery (often peppered, with golden or rusty brown) peltate-lepidote hairs beneath; leaf blades
conduplicate when young, elliptic, oblong elliptic to oblanceolate, sometimes rather abruptly acuminate into
a sharp drip tip apex, usually cuneate to widely rounded at base, (excluding the linear, juvenile or stump
sprout leaves) (6-)8-21(-26) x (2-)4-8.5 cm, with strongly impressed midrib on upper surface, secondary
nerves rather inconspicuous on both sides; petioles (0.5-)l-4 cm, canaliculate. Inflorescences leafy corymbs,
with 1 to 5 peduncles 0.5-10 cm, borne leaf axils at branch apices, each corymb 1 to 7-(to 13) flowered, linear
bracts 2-6 mm, usually soon caducous; pedicels stout, usually ± flattened, 0.5-4.5 cm, bright rusty brown,
lepidote; sepals valvate, before anthesis totally enclosing petals to form an ovoid to lanceolate, pointed, ±
longitudinally ribbed bud, sepals after anthesis reflexed, ovate to lanceolate, (9-)10-18 x 5-8 mm, densely
rusty lepidote without, tomentose to tomentulose within; nectar scales triangulat to deltoid and flat, one
each projecting from inner sepal bases; petals divergent at anthesis, creamy white, soon fading (as do the
stamens) to a purplish violet, elliptic-obovate, (9-)10-20 x 6-10 mm, glabrous within; stamens ca. 30 to
60, 20-70 mm, densely white pilose at base, anthers 3-5 mm; gynophore (10-)15-80 mm, glabrous, on an
androgynophore 1-2 mm, ovaries linear cylindric, 6.5-10 mm, densely peltate-lepidote, stigma truncate.
Fruits linear cylindric capsules, (6-)10-60 x 0.4-1. 2 cm, dark rusty brown lepidote throughout, ± torulose,
± irregularly rupturing along (usually one) suture, the valves turning out to expose their pulpy bright orange
to scarlet inner wall and embedded in it the bright orange to red arillate seeds, these often dangling by a thin
funicular thread; gynophores 15-80 mm, glabrous, clearly demarcated from the fruit valve, receptacular
disks in fruits 5-10 mm diam.; pedicels 1.5-4 cm, lepidote; seeds (3 to) 6 to 40, oblong, 6-8 mm, with
bright orange aril, the embryo oblong, green throughout.
6a. Quadrella isthmensis subsp. isthmensis (Eichler) Hutchinson, Gen. Fl. Pi. 2:308. 1967. (.Figs. 9C,
11). Basionym: Capparis isthmensis Eichler in Martins, Fl. Bras. 13:269. 1865 Type: COSTA RICA: “Habitat ad Costa Rica et Veraguas
Amencae Centralis,” without date, C. Hoffmann & Warszcwicz 21 7 (lectotype, designated here B, B fragm. at M, WIS).
Leaf blades (6-)8-21(-26) x (2.5-)4-8.5 cm, matte above (when dry), on petioles to 4 cm. Inflorescence
peduncle to 10 cm; floral bracts linear, 3-6 mm, usually soon deciduous in very young buds; flower buds
ovate (in Panama) to mostly lanceolate (in Costa Rica), often with sharply angled longitudinal sutures; sepals
ascendent to reflexed; petals ovate to elliptic, 12-18 x 7-10 mm; stamens 4-7 cm, gynophores 4-8 cm;
fruits often very long, 10 to 60 cm, light to copper-brown, with elongate gynophores 4-8 cm, and pedicels
(1.3-)2-4.5 cm; seeds 7-10 x 4.5-7 mm.
The lectotype of Capparis isthmensis is designated here because in the protologue, Eichler only cited
“Habitat ad Costa Rica et Veraguas Americae Centralis: C. Hoffmann et Warszewicz ," but the number of col-
lection was not specified.
Typical Quadrella isthmensis is characterized by the large, elliptic, acuminate leaves with drip tips and
often relatively long capsules, as well as by a silvery, pronouncedly bronze peppered, peltate-lepidote pubes-
cence, especially on the underside of the leaves. The ovaries and petals are longer than those of Quadrella
cyanophallophora or Q. siliquosa; and the gynophores in flower and fruit are longer than those of the other
subspecies in Quadrella isthmensis.
Quadrella isthmensis subsp. isthmensis presents no special taxonomic problems, except for occasional
Plants from the arid southern beaches and islands off Panama’s Pacific coast, which tend to have smaller,
thicker leaves, this apparently a response to local aridity. While these plants resemble Quadrella cynophal-
109
really do not know!), take a seed and, after swallowing the desired aril together with the thin testa (which,
with a little manipulation easily slip off the distasteful acrid, now naked embryo), spits out the embryo, which
then could start to photosynthesize and grow into a seedling soon after it hits the ground (litis, unpubl ).
A chance reading of the “ Capparis ” treatment in Linnaeus’ Pjlanzensystem (1779), the rare 16-volume
German translation from the Latin 13th edition of Systema Naturae (1764) (Anonymous 1933), has this to
say about the seed germination of “Capparis cynophallophora L.” (by which the author meant not the densely
peltate-pubescent Quadrella cynophallophora, but the glabrous Cynophalla Jlexuosa (L.) J. Presl (= Capparis
Jlexuosa (L.) L.), [judging from its thick, stiff, smooth leaves with “glandulis axillaribus LINN. Syst. Veg. p.
405 Sp. pi. 721 [should be 722!], ...which on both surfaces are completely smooth...,” reflecting the hor-
rible nomenclatorial confusion dating back to old Linnaeus himself between two of the most widespread
and common Neotropical species that are also the two most common Capparids grown by the British and
Dutch aristocrats, a confusion lasting well into the 20th Century (e.g.. Pax and Hoffmann 1936:178):
Habitat & Distribution.— This, the typical subspecies is found in Costa Rica, from sea level (in Corcovado
National Park, Osa Peninsula) to 900 m (in the Cordillera de Tilaran, Rio San Luis valley below Monteverde
Cloud Forest Reserve), generally on the Pacific slope, in wet or moist tropical forests, and east to Panama,
in drier habitats in the Pacific lowlands. In Panama, this subspecies is often found at or near sea level on
beaches or behind mangroves, apparently not overlapping the range of the montane Quadrella isthmensis
subsp. glabripetala in the Cordillera Central. Surprisingly, Q. isthmensis is not known to occur in Nicaragua
even though, under the name of Capparis cynophallophora, it was optimistically listed as a species to be
expected (litis 2001: 570).
Phenology. — Flowering from January through April (September to October), fruiting from June to De-
Vernacular name.— COSTA RICA, camaron bianco (white shrimp), (J. Ledn 953, F; Zamora 1989).
Specimens examined. COSTA RICA. Puntarenas: forest near Palmar Norte, P. Allen 6639 (F); Res. Biol. Monteverde, Cordillera de
HI inn i \ ! i |, _4’, | | > | > l| p hill! pi. mil I m rde. H H /It.' etui. U'.U' >0341 iWb' Mon-
teverde, 10 km SO, on road to Inter American hwy, W. Haber & W. Zuchowski 9933 (1NB, MO, WIS); 9 km W Monteverde on road to Inter
American hwy, W. Haber & W. Zuchowski 9248 (CAS. CR, 1NB [2], MO, VDB, WIS). W. Haber & W. Zuchowski 9761 (INB, MO, WIS); 4
km W of Monteverde, Cuencas del Lagarto y Guacimal, W. Haber & N. Obando 12128 (INB); trail from Playa Manuel Antonio to Puerto
Escondido, M. Grayum & R Sleeper 5921 (MO, WIS); Miramar, Cerro Zapotal, Quebrada Seca, L. Gdmez, et al. 23994 (MO, WIS); Carara
Reserve, M. Grayum & R. Warner 5707 (MO), sendero Quebrada Bonita a Bijagual, sitio Lomas Pizote, R. ZMga & Q. Jimenez 14 (MO,
WIS); Canton Golfito JimOnez, Rio Piro y Quebrada Coyunda, A. Chacdn 1039 (MO {21, WIS 121); Alto Carbonera, cerro Osa, cabeceras
de Quebrada Sombrero, G. Herrera 4344 (INB, MO, WIS); Osa, 3 km N de Playa Piro, Puerto JimOnez, Q. Jimenez et al. 659 (MO); Osa
Peninsula, Corcovado Nat Park Sirena R. Liesner2976 (MO, WIS), P. Delprete 5152 (TEX), R. Liesner 2825 (MO); entre Senderos a Rio
Claro y a Los Patos, G. Fonseca 56 (INB, MO); R. Aguilar 2477 (CR. INB, WIS); 0. Tellez 4247 (MEXU. MO); A. Chacdn 86-05 (WIS); S.
KnaPP 2186 (MEXU, MO, WIS); W. Alverson 1837 (WIS); G. Maass 90 (INB, WIS); A. Gentry &OTS class 48484 (MO, WIS); C. Kernan 61
(MO);J. Saborio 83 (MO, WIS); G. Fonseca 14 (INB, MO); Res. For. Golfo Dulce, Rinc6n, L. Angulo 498 (INB). San Jose: Z . P. La Cangreja,
Cenos de Puriscal, Santa Rosa de Puriscal J. Morales & Q. Jimenez 3317 (INB, MO); P*rez Zeledon, Tinamaste, Finca de Los Suizos, A.
Estrada et al. 1584 (MEXU); Rio Naranja, versant Pacifique, A. Tonduz 7656 (BR, WIS fragm.) Guanacaste: Hojancha de Nicoya, J. LeOn
953 CF); Zona Protectora Nosara, Res. For. Monte Alto, L. Gonzdlez & F. Hidalgo 2917 (INB ); J. Morales 8711 (INB); Cerro San Jos*, near
H. Granadilla, C. Dodge & W. Thomas 6459 (F, MO); 5.4 mi. W of Tilaran on road to Laguna de Arenal, D. Stone 2168 (DUKE); J. Walker
■ :'uil PS uirdhs^l \\<!,no-n\'-T\ ' ' •" ; : M' '
AUjuela: San Ramdn, Finca San Gerardo, A. Carvajal 117 (MEXU, MO [21, WIS); near Atenas, in monte Aguacate, 0rsted 3168 (C); San
M'UUvl dc Sun F'.im. mi or.llus d. I Hu, B„ r,rn ., 4 R.ei.C' 2I'>52 ‘ ,F. NY'. Yin Pedro de V.n Ramon. A B».- ws NY > Sun., ago. cu.mm.
de Sun Francisco de '.an Pj.m.n 1 (W. I NY sun.uc dc x,n A NY- PANAMA. Veraguas: shore ot
Ensenada Santa Cruz, N tip of Coiba Island, R. Foster 1621 (DUKE, F, MO, PMA), R. Foster 1644 (DUKE, F, MO, PMA); lsla Coiba, camp.
p'aya de la Salina, borde de la playa, S. Castronejo et al. 7186 (MA, MEXU); Par. Nac. Coiba, Islajicardn, C. Galdames, et al. 4000 (BM, MA,
Mo) Canal Zone: Madden Dam, J. Ebinger 867 (F, PMA); Boy Scout Rd., Madden Dam area, D. Porter et al. 4054 (MO, WIS), J. Dwyer &
LaBfltto« 8830 (F, MO); T. Elias 7509 , 7510 (MO); Fort Clayton, 1 mi. bey. Madden Dam Bridge. M. Correa & R. Dressier 354 (DUKE, MO,
110
Journal of the Botanical Research Institute ofTexas4(1)
Subspecies of Quadrella isthmensis on Yucatan
The disjunct occurrence of Quadrella isthmensis subsp. mexicana on the Yucatan Peninsula is problematic, -
both in its geography, morphology, and relationships. Ranging from tropical seasonally dry, high forests to 1
Guatemala and Belize, where very rare, all the way to the dry forests beaches and edges of mangroves tot |
to the north in Mexico, the plants follow somewhat of a cline, from rather typical but, somewhat smaller f
leaves from around Lago Peten Itza and elsewhere in Peten Province and in Belize, to smaller, stiffer, more
narrowly lanceolate-elliptic leaves in the dry deciduous forests of northern Yucatan (Fig. 9A).
Mexican specimens of Quadrella isthmensis subsp. mexicana were initially identified as Capparis cynophd- 1
lophora, the leaves of which resemble somewhat the leaves of our Mexican subspecies, were it not that the
latter species shares at anthesis the larger ovaries with Q. isthmensis and the characteristic bronzed dotted,
peppered pubescence on the underside of the leaves. In addition, the closest Quadrella populations in Mexico
are those of the endemic Q. quintanarooensis, and are apparently vegetatively closest to Q. cynophallophm |
Thus, since we are faced with an isolated, somewhat morphologically differentiated population ca. 1000 ktol
disjunct to the east from the main population of Quadrella isthmensis in Costa Rica and Panama, we have
6b. Quadrella isthmensis subsp. mexicana Cornejo & Utis, subsp. nov. (Figs. 9A, 10). Type: Mfixico. Yucatan:
1916 (fl). C. Caumer & sons 233-H MO h. mi BM, F. K, NY, US).
petalis (9-13 mm longis), stamina (2-41-4.51 cm longis), gynophoris (1-4 cm longis) et fructis (ad 23 cm longis).
Leaf blades 6-15 x 1.5-6 cm, often lustrous above when dry, on petioles to 1.7(-2.3) cm. Inflorescence I
peduncle 0. 7-5.5 cm, floral bracts linear, ca. 2 mm, soon deciduous; flower buds ovate, with softly angled j
longitudinal sutures; flowers with petals lanceolate to somewhat elliptic, 9-13 x 6-7.5 mm; stamens 2— 4[— 4.5l j
cm, gynophores 1-4 cm. Fruits relatively shorter than the typical subspecies, to 23 cm, on usually shorter £
gynophores, (l-)2— 4<— 5.5) cm, and pedicels 1-2 cm; seeds 6-8 x 4-5 mm.
Habitat & Distribution. — This new subspecies, one of the successional elements from the mangrove’s j
landward borders in Yucatan ( Lira et al. 368), ranges inland through selvas medianas subperennifolias south I
to high forests in Guatemala and adjoining Belize.
Phenology.— Flowering from (February) April through May, fruiting from (June) August through Sep- f
Vernacular name.— MEXICO, YUCATAN: caimito-che (Enriquez 544, MEXU).
IUCN. — Because of the numerous collections gathered in recent years, this new subspecies deserves* j
lower concern, LC.
BOOK REVIEW
Carol KaesukYoon. 2009. Naming Nature: The Clash Between Instinct And Science. (ISBN: 978-0-393
06197-0, hbk.). WW Norton & Company, Inc., 500 Fifth Avenue, New York, New York 101 10, U.sXI
(Orders: www.wwnorton.com, 1-800-233-4830, 1-212-869-0856 fax). $27.95, 352 pp., 6 1/2" x9jfc
STUDIES IN THE CAPPARACEAE XXIX: SYNOPSIS OF QUADRELLA,
A MESOAMERICAN AND WEST INDIAN GENUS
Hugh H. litis Xavier Cornejo
Journal of the Botanical Research Institute of Texas 4(1)
Table 1. Morphological comparison of Quadrella subgenera Quadrella, Breyniastrum and Intutis.
oblong, lanceolate tc
16 to 60
usually tufted or Stella
few times to peltate
open when young
open or covering the petals in bur
triangular to linear, or oblong to
usually 8, ca. 16 fin 0 singularis)
Calyx characters in New World capparid taxonomy have been important. Used first by de Candolle |
(1824) for closed vs. open calyx aestivation, then by Eichler (1865) for his stellate- to peltate-pubescent sub-
genera, and since then by subsequent workers (e.g., Hutchinson 1967). As an example we may take valvate
calyx aestivation, the condition in which the four sepals develop in a single series with their parallel margins |
adjoining one another to tightly appressed or even reduplicate, but with the margins never overlapping.
The calyx may then a), either stop growing from early on, to be rapidly exceeded by the corolla bud and
thus ending up with small, ± triangular sepals producing a calyx aestivation that is open (“aperta,” Robert
Brown 1826: 220), this the hallmark of Quadrella subg. Breyniastrum (DC.) litis and half of the species in|
Quadrella subg. Intutis (Raf.) litis; or b), continue to grow and enclose the corolla bud and result with a calyx |
aestivation that is closed until shortly before anthesis, this the hallmark of Quadrella subgenus Quadrella.
The valvate calyx is one of the primary unifying characters of Quadrella, with the closely adjoining sepals
meeting exactly but never overlapping. Although, this character is present in Quadrella subg. Quadrella and
half of the species of Q. subg. Intutis (Q. angustifolia [Kunth] litis & Cornejo, Q. alaineana Cornejo & litis awl
Q. singularis [R. Rankin] litis & Cornejo), where the relatively large sepals result in a valvate or reduplicate
calyx with closed aestivation until anthesis, in the majority of species (subg. Breyniastrum and the remaining
three species of subg. Intutis), the calyx aestivation is open, with the very short, triangular to linear sepals
not enclosing the petal bud. Admittedly, while calyx aestivation is distinctive, nevertheless those calyx types
are valvate, and the differences are subtle and not significant enough to break up the genus Quadrella into
two genera, as has been done, whether at the sectional level by de Candolle (1824) or at the generic level by
Hutchinson (1967).
The complex lepidote-peltate pubescence that unites Quadrella subg. Quadrella is clearly a specialization
of once stellate or tufted pubescence types fused into lepidote or peltate, multicelular scales as they occur in
great diversity in Quadrella subg. Breyniastrum. Nevertheless, comparing Q. odoratissima (Jacq.) Hutch. andQ.
cynophallophora (L.) Hutch, of Quadrella subg. Quadrella, their respective peltate trichomes are so distinctive
that we must conclude that they most probably evolved independently in these two taxa. Similarly, we can
separate out the stellate or peltate pubescent groups from those that are characterized by stellate-tufted (c g -
Quadrella subg. Breyniastrum) or the stellate-candelabra types of pubescence (e.g., Quadrella subg. Intutis).
Much like the basic ubiquituous (n = 8) chromosome number, these are conservative characters not easily
subject to change (litis & Cornejo 2007b).
Stellate to peltate pubescence evolved, especially in Quadrella, into an astonishing diversity of morpho-
logical tnchome types that obliterate the sectional division (de Candolle 1824) imposed by the closed vs.
open aestivation of the calyx. Pubescence varies from simple stellate hairs to complex, stemmed, bottlebrush
hairs m subg. Intutis, m a continuum from ca. 20 plus, simple, unbranched but tufted hairs on a short stem
through complex intermediates to semi-peltate or peltate trichomes in subg. Breyniastrum (e.g., Q- indica [L||
119
litis & Cornejo), to end in the massive shield-shaped peltate hair types of subg. Quadrella, reminiscent of
the indumenta in the Brassicaceae Burnett and Croton L. of the Euphorbiaceae Juss. (Webster et al. 1996).
Close examination of trichomes types has not been studied in Capparaceae since Vesque in the late 1800s
(Vesque 1882) and recently by Rankin & Greuter (2004, as Capparis sect. Breyniastrum [DC.] Eichler), for
Quadrella subg. Intutis.
The number of stamens in Quadrella varies from eight to ca. 60, with eight-staminate species scattered
throughout subgen. Breyniastrum (three of 13 species), and predominant in subgen. Intutis (all species except
Q. singularis). The eight-staminate condition, otherwise infrequent in Capparaceae, is considered here to be
the basic primitive condition often linked with stellate pubescence (see above). The multi-staminate condition
may reflect a derived state due to “dedoublement” (chorosis) of eight staminal intials and a specialization
that in its extreme is often linked to peltate pubescence, e.g., in Quadrella isthmensis (Eichler) Hutch.
In Quadrella, the seed is often surrounded by a bright red or orange aril (or sarcotesta and/or pulp), but
exactly what that structure is, and from what it is derived, we must leave to the morphological insights of
others.
One character related to seed dispersal is the nature of the testa. In Quadrella subg. Intutis, seeds have
a hard testa that encloses a white or cream embryo that is presumably ingested and then expelled with the
feces. In contrast, in species of subg. Quadrella and Breyniastrum, the testa is relatively thin and in some
species easily removed to expose a naked green embryo that, apparently acrid and bad tasting, is soon spit
from the New World across ^Atlantic via sail ship unsuccessful.
Even though only sparingly investigated, the number of chromosomes has turned out to be, as in many
angiosperm groups, one of the most significant characters defining the stellate-pubescent Capparaceae (litis
& Cornejo 2007b). In Quadrella, the chromosome number of n * 8 (or 2 n= 16, from root tips) is herewith
reported for Q. antonensis (Woodson) litis & Cornejo, Q. isthmensis (Eichler) Hutch, subsp. isthmensis, and
Q. odoratissima (Jacq.) Hutch. In addition to this cytological information, the following other Neotropical
genera of Capparaceae, all characterized by having a stellate or peltate pubescence, have the same chromo-
some number: Atamisquea emarginata (DC.) Miers ex Hook & Am. (Kers 2003), Beautempsia avicennifolia
(Kunth) Gaudich. ( litis & litis E-20, WIS), Capparicordis crotonoides (Kunth) litis & Cornejo ( litis & Utis E-15,
WIS), Colicodendron yco Mart. (Mori et al. 11218, WIS), C. scabridum (Kunth) Seem. (Utis & Utis 243, WIS),
Neocalyptrocalyx longifolium (Mart.) Cornejo & litis (Mori et al. 11205, WIS), and Preslianthus pittieri (Standi.)
Utis & Cornejo (Judziewicz 4555, WIS). On the other hand, the generic type of Capparis, Capparis spinosa L.
has reportedly 2n = 38 (Al-Turki et al. 2000), with base numbers that are neither n = 8 nor 16.
TAXONOMIC TREATMENT
Quadrella (DC.) J. Presl in Berchtold & J. Presl, Prir. Rostlin 2:260. 1825. Basionym: Capparis sect. Quadrella dc„
Shrubs to small trees, pubescent throughout, with abundant lepidote-peltate to variously stellate or to-
mentose tufted hairs, with several trichomes types often intermixed at least when young, especially on the
underside of the leaves. Leaves various, lanceolate to narrowly to broadly elliptic or oblong. Calyx valvate,
anthesis (subg. Quadrella and Q. angustifolia, Q. alaineana and Q. singularis, of subg. Intutis), or small, trian-
gufar to linear and with an open aestivation, exposing the corolla from early on (subg. Breyniastrum and
the Gaining species of subg. Intutis). Disk of 4 separate and erect episepalous scales arranged on a flat
receptacle; stamens 8 to 60. Capsules linear-cylindric, often torulose, to 60 cm (in subgenera Quadrella and
Breyniastrum, except Q. morenoi Cornejo & litis), or fruits short and ovoid to shortly oblongoid, to 3(-10, Q.
angustifolia) cm (in subg. Intutis), dehiscent (rarely tardily so) along one or both sutures; valves fleshy with
reddish pulp; seeds alternating from both repla but usually arranged in a single row, oblong with shallow or
120
obsolete testa invagination, the embryo straight or slightly curved, green (subgenera Quadrella, Breyniastrm) I
or white (subg. Intutis).
Quadrella was originally established by de Candolle (1824) as a neotropical section of Capparis L.S.L
In Capparis sect. Quadrella DC., de Candolle accepted seven Capparis species (numbered 92-98), but not ;
C. cynophallophora. However, under species 95, C. breynia L., he does list in synonymy Breynia indica L. |
(= Quadrella indica [L.l litis & Cornejo) and C. cynophallophora (=Q. cynophallophora ), evidently confusing Q. I
cynophallophora and Q. indica.
The elevation of Capparis sect. Quadrella to generic status occurred a year later (1825), and with it the
transfer of seven species in Volume 2 of a very rare Czech work, written in the rare Bohemian language,
O Prirozenosti Rostlin by the world traveler and amateur botanist, Friedrich Berchtold (1781-1876), and j
his junior co-author, Jan Swatopluk Presl (1791-1849). The latter, the trained taxonomist of the pair, is the |
one to whom the genus and new combinations are credited. In Flora Brasiliensis, Eichler (1865) reduced
Quadrella to a subgenus of Capparis, but 102 years later, Hutchinson (1967), in his generic realignment of j
Capparaceae, recognized Quadrella as a valid genus.
Hutchinson (1967) designated Quadrella cynophallophora as the lectotype of the genus Quadrella, with I
Quadrella jamaicensis in synonymy under the former species. However, Quadrella cynophallophora was not
cited by Presl when the genus was established, therefore Hutchinson’s lectotypification of Quadrella is not j
valid, because was not based on Presl’s original material for the name. We designate Quadrella jamaicensis |
the neotype for this genus, which was previously cited by de Candolle (1824), in Capparis sect. Quadrella, by j
Eichler (1865) under Capparis subg. Quadrella, by Presl (1825) under Quadrella, and by Hutchinson (1967) | j
as a synonym of his lectotype Quadrella cynophallophora.
Inocencio et al. (2006) state that Quadrella crotonoides (Kunth) Presl ( =Capparicordis crotonoides [Kunth] j
litis & Cornejo) is the type of Quadrella. However, in the cited literature, there is no such designation of this I
species as type for this genus. Their typification was not explicitly cited with the phrase “designate here" (as I
must be for accepted lectotypifications on or after 1 Jan 2001, in accordance to the Art. 7.11 of the Vienna |
Code [McNeill et al. 2006]), therefore Quadrella crotonoides is not the type for this genus. Finally, in our j
concept of Quadrella, which follows most of the species cited by de Candolle and Presl, and all of Eichler’s J
and Hutchinson’s, Quadrella crotonoides doesn’t belong to the genus Quadrella at all (see rejected names). I
A molecular work (Hall 2008) has demonstrated that the genus Quadrella is related to some Neotropical j
taxa of Capparaceae characterized by the presence of stellate to peltate indumenta, such as: Capparicordis Iltk jj
& Cornejo, Calanthea (DC.) Miers, Colicodendron Martius, M orisonia L, and Steriphoma Spreng.; and that those j
genera, jointly with all remaining Neotropical species also formerly placed in Capparis, represent a separate
lineage that is not closely related to that of the Old World Capparis, including C. spinosa L., the generic type J
That molecular evidence is also consistent with the morphology of all Neotropical genera of Capparaceae j
(including Quadrella), in which, in contrast, it is evident the absence of the following characters that are j
present in Capparis spinosa: 1) A pair of retrorse stipular spines; 2) flowers with one galeate (helmet-shape) I
sepal; 3) only a single nectary gland with its apex directed towards the interior of the flower; and 4) two
of the four petals having an irregular shape, with their asymmetrical bases laterally connate to each other
and folded, forming a petaloid hood-like structure that partially envelops and protects the solitary nectary
gland (Cornejo & litis 2009a). Thus, those morphological and molecular differences make it impossible to
continue using the name Capparis for the Neotropical species traditionally assigned to that generic name- j
A key to Quadrella and related Neotropical genera with stellate to lepidote-peltate indumenta is provided in I
Cornejo & litis (2009b).
key to the subgenera of quadrella
1 . Indument of terminal branches and leaf blades (at least abaxially) of peltate-lepidote trichomes; calyx with
cose aestivation, the sepals lanceolate to oblong or ovate, covering the corolla until anthesis
' — I. Quadrella (DC) J Presl subg. Quadrella
tm
ally 8, or ca. 1
122
Journal of the Botanical Research Institute of Texas 4{1)
and Costa Rica, and disjunct to the central Yucatan Peninsula (northern Guatemala and Belize) north to
the Caribbean coast of Mexico's Yucatan State.
7. Petals completely glabrous without, sepals tomentulose within; very rare, on or near the crest of the
Cordillera Central of western Panama, at 500-1 200 m Quadrella isthmensis subsp. gl
7. Petals densely lepidote without, sepals tomentose within; eastern Mexico, from northern Yucatan and
Campeche south into Guatemala and Belize; and, disjunct, to Costa Rica and the lowlands of Panama.
8. Flower larger, with petals 1 2-1 8 x 7-1 0 mm, stamens 4-7 cm, and gynophores 4-8 cm; fruits to 60
cm, on gynophores 4-8 cm, and pedicels (1 ?• - 1 2 - 4.5 cm. mature leal blades (5-18-21 (-26) cm,
opaque above (when dry); Costa Rica and lowlands of Panama Quadrella isthmensis subsp. i
8. Flowers smaller, with petals 9-13 x 6-7.5 mm, stamens 2-4H1.5) cm and gynophores 1-4.5 cm;
fruits to 23 cm long, on gynophores 2-4(-5.5) cm and pedicels 1 -2 cm; mature leaf blades 6-1 5 cm,
often lustrous above (when dry); northern and central Yucatan Peninsula of Mexico to adjacent
Guatemala and Belize
it ridged; peltate indumentum golden to silvery; gynophores absent _
_ Quadrella odoratissima
la. Quadrella cynophallophora (L.) Hutch., Gen. 1
Quadrella cynophallophora is a widespread, polymorphic species, ranging from the Bahamas through
the Dominican Republic (Hispaniola) and the Lesser Antilles to Trinidad and Barbados (in the latter two
apparently known mostly from cultivated or escaped plants), generally at low elevations near the ocean, but
to 400 m in Dominican Republic. Locally rarely sympatric with the similar Quadrella jamaicensis (Jacquin)
J. Presl, in the Bahamas and the Dominican Republic, it can be distinguished by its mostly pointed, acute
Quadrella cynophallophora occurs in central and eastern West Indies, from the Bahamas to Hispaniola
(where intergrades with Quadrella jamaicensis) and the southern coast of Puerto Rico and through the Lesser
Antilles to Trinidad and Barbados (Gooding et al. 1965), mostly in coastal, seasonally dry, evergreen wood-
lands, but also in a great variety of habitats, often cultivated as hedges.
t, A Reilly & M. Davis 1808 (hoi
istic linear leaf blades.
. Quadrella gonaievensis (Helwig) Hu
s.d., E Ekman 8483 (lectotype, B, B
Quadrella gonaievensis is a rare, highly localized but scattered endemic of Haiti and adjacent western parts
of the Dominican Republic. This species has usually been ignored completely, or listed in synonymy under
Q. cynophallophora (Urban 1920-21; 239; Al-Shehbaz 1988; Rankin & Greuter 2004). However, Quadrella
gonaievensis differs of Q. cynophallophora by having small, delicate flowers, with petals 6-7 mm (vs. ca. 9-10
mm); stamens with densely lepidote-radiate or lepidote-stellate, many branched (vs. densely pilose with
simple) trichomes at the filament base; and smaller anthers, 1-2 mm (vs. 2-3.5 mm).
3. Quadrella isthmensis (Eichler) Hutch., Gen. FI. PI. 2:308. 1967. basionym:
217 (tECToiYre, designated by litis & Cornejo 2010a, B, B fragm. at M, W1S).
3a. Quadrella isthmensis (Eichler) Hutch, subsp isthmensis
Chromosome number.— 2n - 16 [IPazy, from i
t tipi. Panama: Canal Zone, 7
e, 10 cm dbh, visited
s and Cornejo, Synopsis of Quadrella
123
small bees, trail along Rio Petit Pie, near road to Fort Sherman, on Limestone, upland Panama, S. Mori &
J. Kallunki 5011 (WIS).
Quadrella isthmensis subsp. isthmensis is found in Costa Rica, from sea level (Osa Peninsula) to 900 m
(Cordillera de Tilaran), generally on the Pacific slope, and in coastal mostly Pacific lowlands of Panama, mostly
in wet or moist tropical forests, apparently not overlapping the range of Q. isthmensis subsp. glabripetala.
3b. Quadrella isthmensis glabr ipctala
Restricted to higher elevations and the crest of Cordillera Central in Panama (Cornejo & litis 2010a).
atan (Lira et al. 368, MEXU) inland to “selva
ti forests in Guatemala and adjoining Belize.
subperennifolias” (Rzedowski 1978), south to
Quadrella jamaicensis is a
teristic of coastal areas from central Florida (from Cape Canaveral on the east
on the west coast), south to Key West. It also occurs in Cuba, Jamaica, and t]
overlapping and intergrading with Quadrella cynophallophora in Hispaniola, b
& Cornejo 2010a).
ihis strikingly different-appearing form is a stump sprout
leaf blades, common in this species and, in a homologous foi
This form is distributed across the range of the species.
5. Quadrella odoratissima (Jacq.) Hutch., Gen. FI. Pi. 2:308. 1967. Basionym:
nbr 1:57, t. 110. 1797. Type: VENEZUELA: Caracas, “In caldario, floret Martio & Aprili’
a of the Lesser Antilles-Dicotyledoneae 4(1): 1988).
in Quadrella cynophallophora f. I
LB onpland39(w
Chromosome number. — 2 n = 16 (!Pazy, IPrzywara, from root tips]. Venezuela: Distrito Federal, shrub 3 m,
tropical low deciduous forest, carretera vieja Caracas-La Guaira, 320 m, 4 Feb 1979, P. Berry & T. Plowman
3363 (MO, WIS).
Quadrella odoratissima is a common and very distinctive silvery-peltate evergreen species with beautiful
flowers and sessile siliques, characteristic of arid tropical dry forests, from northern coastal Colombia and
Venezuela into the Lesser Antilles, and, in Mesoamerica, from Panama to southern Mexico.
Capparis intermedia is herein synonymized under Quadrella odoratissima because both describe the same
sPecies, giving priority the older name.
6- Quadrella quintanarooensis litis <Sr Cornejo, J. Bot. Res. Inst Texas 4: 104. 2010. Type: MEXICO. Quintana Rod
Journal of the Botanical Research Institute of Texas 4(1)
), E. Cabrera & -H.de Cabrera 3495 (holoiype: MEXU, MEXU
uadrella quintanarooensis occurs in dry, open, tropical forests and woodlands on or near beaches at se
n Cozumel Island and the adjoining Mexican mainland of Quintana Roo (litis & Cornejo 2010a).
is f. longifolia ). Breynia L
[L.] Hutch., L grisebachii [Eichler] H
Z. hastata Jacq., C. linearis Jacq., <
ajacq. (>Q. ferrugine
Type: Pleuteron breynia (L.) Raf. [=Quadrella indica (L.) litis & Cornejo] , 1
The logical choice of Hutchinson (1967) of Breynia indica ( =Quadrella indica) as the subgenus lectotype |
(for Capparis subg. Breyniastrum) preserves traditional useage, while Capparis ferruginea L. (=Quadrdk I
ferruginea), posteriorly designated as lectotype by Rankin & Greuter (2004: 262, for Capparis sect. Breyni- j
astrum), introduces an extraneous element belonging to a different taxonomic group that is characterized I
by a capsular fruits with few seeds containing white embryos surrounded by a ± hard testa, which we now I
segregate in Quadrella subg. Intutis (Raf.) Utis; see below.
Hutchinson (1967), proposed Unnaeobreynia as a new generic name for Breynia L., a rejected illegal I
homonym, because Breynia J.R. Forster & G. Forster (Euphorbiaceae) is a previously established nomen I
conservandum. He repeated most of the species included in Eichler’s broad but heterogeneous concept of I
Capparis subgenus Calanthea (DC.) Eichler, intermixing species with different calyx structure, dehiscent 1
and indehiscent fruits and with seeds of green and white embryos, including L. pulcherima (Jacq.) Hutch, j
(= Calanthea pulcherrima [Jacq.] Miers, the type of Calanthea [DC.] Miers, a valid South American small genus j
of Capparaceae characterized by the flowers with widely spaced, linear-ligulate to oblong sepals exposing
the valvate corolla from early bud on and fleshy amphisarca, bearing seeds with uncommonly very thick
cotyledons [Cornejo & litis 2008, 2009]); four species of Quadrella subg. Breyniastrum: L. admirabilis (Standi.) J
Hutch. (=Q. lundellii [Standi.] litis & Cornejo), L. asperifolia (K. Presl) Hutch. (=Q. asperifolia [K. Presl] Hhs
& Cornejo, L. pringlei (Briq.) Hutch. (=Q. pringlei Briquet), and L. tonduzii (Briq.) Hutch. (=Q. indica. ); and the
following taxa of Quadrella subg. Intutis: L. domingensis (=Q. domingensis [Spreng. ex DC.] litis & Cornejo),
L. grisebachii (=Q. domingensis subsp. grisebachii [Eichler] litis & Cornejo), L. ferruginea (=Q. ferruginea [U
litis & Cornejo), and L. incana (=Q. incana [Kunth] litis & Cornejo).
Pleuteron breynia (- Quadrella indica) is herein designated the lectotype of Pleuteron because Rafinesque,
in t e protologue, referred to it as the “main type.” Pleuteron is synonymyzed under Quadrella subgen-
,beCaUSe thie ™enti°ned selected lectotype has stellate pubescence, flowers with calyx valvate
naming green embryos.
Quadrella in its peltate or stellate pubescence and cap- J
ng in small triangular sepals. These, valvate from very
on much exceeded by the corolla, thus leading to an open calyx aestivation
elongate capsules are filled with sticky red pulp and several to many seeds,
Quadrella subg. Breyniastrum is similar to subg.
ar fruits, but with its retarded calyx growth result
ly on in development
125
these covered by a red or orange aril or endocarp(?) and a thin testa that easily is slipped off the green em-
bryos by the vertebrate dispersers. Twelve species are assigned to this subgenus.
Despite the calyx differences between subg. Quadrella (closed calyx aestivation) and subg. Breyniastrum
(open calyx aestivation), the two sections seem very closely related indeed, considering their similar fruit,
pulp, and embryo colors, the nature and color of their seed arils and seed coats, their variable tufted to stel-
late to peltate-lepidote pubescence and finally their distributional biogeographic coherence.
Distribution. A New World group centered on Mexico and Central America.
KEY TO THE SPECIES OF QUADRELLA SUBG. BREYNIASTRUM
I. Detritophilous plants, main stem with adventitious roots and spiralate leaves, densely arranged at the end
of the stem or branches in pseudoverticils that collect humus; androgynophores and gynophores absent.
2. Shrubs or branched treelets; leaf blades (1 2-) 1 5-35(-40) x 3-1 0(- 1 1 ) cm; inflorescence a persistent raceme,
bracts soon deciduos; filaments 4-5 mm, petals 5-7 mm; Panama, west of the canal Quadrella i
2. Unbranched or sometimes to 2-branched treelets; leaf blades 30-70 x 9-1 9 cm; inflorescence paniculiform
bearing to several few-flowered 2nd order bracteate racemes, soon deciduous if not pollinated, raceme
bracts persistent, 3-4 mm; filaments and petals 7-8 mm; Panama, east of the canal. Quadrel
I . Plants not detritophilous, main stem without adventitious roots, leaves alternate to laxly-spiralate, arranged
along the branches; androgynophores and gynophores present.
3. Gynophores 2-10 mm, staminal filaments 5-10 mm.
4. Petals 4-8 x 2-5 mm, anthers 1 .5-2.8 mm.
5. Flowers with 8 stamens.
6. Leaf blades (2.5-)3-8.5(-10) x 1-4(-5) cm; petals 4.5-6 mm, fruits broadly ellipsoid to obovoid.
x OB-1.1
-27 x 5-10.5 cm; petals 7-8 n
r, torulose, 11-31 >
7-10 mm, gynophores golden lepidote;
7. Leaf blades 2-5.5(-8) cm wide, petioles (15-)20-50(-7'
gynophores lepidote-stellate; fruits 7-1 5 mm wide, seeds 8-1 2 nr
4. Petals 1 2-1 3 x 8-9 mm, anthers 3-4 mm
3. Gynophores 12-40 mm, staminal filaments 12-30 mm.
Quadrella calciphila
s with rounded to subcordat
lorescences corymbose racemes, peduncles 1-2 mm wide, ecostate; stamens 15-j
tonspicuous in floral bud; leaf blades with usually cuneate base.
. Leaf blades smooth above, lepidote-peltate beneath; ovaries densely covered t
lepidote-peltate trichomes _
_ Quadrella lundellii
1 0. Leaf blades ± scabrous <
- Leaf blades without a reddish brown, si
• Quadrella antonensis i Woodson' Hits & Cornejo, comb nov Bx i jrra,,>jn,,>n,-,M\voodion.Ann Missouri
But Card. 35:90. 1948. Type: PANAMA. Cooe: hills N El Valle. 13 Jan 1942. PH. Allen 2948 (Hournre: MO).
Chromosome number.— 2n = 16 [IPazy, IPrzywara, from root tip]. Panama: Code, shrub 2 m, in cloud forest,
ca- 3 km NE of El Valle, 2 Nov 1961, S. Mori & J. Kallunki 2961 (WIS).
A locally common shrub or treelet with subverticillate, overlapping, cordate based leaves that form a
trash basket invaded by adventitious roots (detritophily) and home to a mini-botanical garden of mosses,
fems, pepperomias and a mini-zoo of amphibians and diverse invertebrates, such as Collembolas. The plant
127
\. Capparisfurfuracea Sesse & Mocmo, FI. Mex. ed. 2
. Type: MEXICO. Quintana Roo: Is
Quadrella indica is a common species, widespread from Pacific coastal w<
and from the Dominican Republic and Puerto Rico through the West Ir
7. Quadrella lindeniana Cornejo & litis, J. Bot. Res. Inst. Texas 4:83. '
Quadrella lindeniana is restricted to Mexico’s Yucatan Peninsula, where it occurs through dry thorn scrub
to subdeciduous lowlands and subevergreen forests in the south, often growing on coastal dunes or rocky
in disturbed forests (Cornejo & litis 2010).
Koepper & KA. Wagner 8130 (holotype: F, F photo 051607; isotypes: S, US).
Quadrella lundellii is a rare and local, in pockets of aridity near the coast, from southern Mi
This species has the ability to produce stump sprouts and juvenile branches with strc
(ditto for species 10, Q, morenoi).
Quadrella mirifica is a gorgeously golden, softly and velvety pubescent shrub, especially on the leaf margins,
extremely rare and endemic to the tropical forests in the vicinity of the Canal Zone of Panama. It is simi-
lar in inflorescence structure by virtue of the elongate filiform peduncles of the pendulous umbels to the
sometimes sympatric Q.filipes. Its stellate pubescence corresponds to that seen in two other species as Q.
steyermarkii and Q. asperifolia.
10a. Quadrella morenoi Cornejo & litis, J. Bot. Res Inst Iexas 4:86 20 1 0 I Nicaragua Boaco km 7 carr. al
Quadrella morenoi is scattered from southern Mexico (Michoacan) to southwestern Costa Rica, but is especially
common in Central Nicaragua. It ranges from sea level to 930 m, in deciduous thorn scrub, gallery forests
and semi-deciduous tropical dry forests and woodlands, persists in secondary vegetation and grazed areas,
often on rocky and volcanic soils, and overlaps parapatrically the range of the similar but more common Q.
incana (of subg. Intutis) from southern Mexico to Guatemala and rarely into central Honduras, mostly in the
seasonally arid climates of the Pacific slope, with their sympatry apparently only at Tehuantepec in southern
Oaxaca, Mexico (Cornejo & litis 2010).
10b. Quadrella morenoi I hastata Hus, J. Bot. Res. Inst. Iexas 4.90. 2010 T,-, nkarau.Wk ii„ S.m u .mo
l] Quadrella pringlei vBriqueO 1 1 1 1 > Cornejo,
Canyon, ca. 3000 ft, 18 May 1894 (fr), C. Pringle 4639 (h
Capparis pringlei Briquet, Ann. Conserv. & Jard.
2:310. 1967. Type: MEXICO. Oaxaca: Tomellin
photo at WIS, BKL, BM, BR, CM, F, GH, GOET,
Quadrella pringlei is distributed from southern Mexico to Guatemala with disjunct local populations in Ni-
caragua (Sierra de Managua) and Costa Rica (below Monteverde). Due to the similar appareance, Quadrella
pringlei is often confused with Q. indica, but has somewhat broader leaves, these ± spirally disposed on sub-
verticillate petioles highly variable in length on the same branch. The flowers of Q pringlei are smaller and
less attractive with shorter petals and stamens and with shorter, subsessile, fewer-seeded capsules.
12. Quadrella steyermarkii (Standi.) litis & Cornejo, comb. nov. Basionym Capparis steyermarkii Standi Publ Field
Mus Nat Hist . Bot. Ser. 22 140 N-tO hn l >t ATFMALA I ' m m Rio Duke. Iviw Livingston & o mi up river, on N sick, ne.it
Quadrella steyermarkii is a rare Guatemalan rainforest endemic, similar in its obovate to obovate-elliptic leaf
shape, but with shorter petals (7-8 x 3-5 mm vs. 12-13 x 8-9 mm) to the sympatric Q. calciphila, also Ml
subg. Breyniastrum. Quadrella steyermarkii is exceedingly similar in leaf pubescence to Quadrella mirifica, but
the pubescence is soft and velvety underneath, similar to that of Q. asperifolia.
III. Quadrella subg. Intutis (Raf.) litis, comb, et stat. nov. Basionym: Intutis Raf., Sylva Tellur. 108. 1838, p
Capparis amygdalina Lam., =Quadrella indica). Type: Intutis ferruginea (L.) Raf. (^Quadrella ferruginea [LI lids & Cornejo)
Shrubs to small trees, stellate throughout, with small, subsessile, densely stellate to candelabroid leaves,
small flowers with calyces with open aestivation, the sepals linear to triangular, far exceeded by the corolla
in bud, or calyces with closed aestivation, the sepals oblong to lanceolate, covering the petals in bud nearly
to anthesis; stamens usually 8, or ca. 16 (in Q. singulars ;); fruits ovoid to obovoid capsules, 1.5-3(-10 cm in
Q. angustifolia), often tardily dehiscent, 1 to 6-seeded, with red pulp, splitting open with 2 to 4 recurved thin
valves, exposing red arillate seeds dangling from funicles, the seeds with white or cream embryos. Whether
the tissue around the seed is arillate in origin or part of the pulp or mesocarp is an unanswered question.
In Quadrella incana it appears as if there is no pulp and the seed is enclosed by its own aril. This subgenus
comprises 6 species, 6 subspecies and one variety.
When Hutchinson (1967) validated the genus Quadrella, he cited the Rafinesquean Intutis in synonymy
and lectotypified both with Quadrella cynophallophora. However, similar to the lectotypification of Quadrella
(op. cit.), Hutchinson’s lectotypification of Intutis is not valid because Quadrella cynophallophora was not
mentioned by Rafinesque in the protologue of Intutis, and is not original material for the name. Capparis
ferruginea (-Quadrella ferruginea) is herein designated the neotype of Intutis because was the first syntype
designed by Rafinesque in the protologue. Hutchinson (1967) placed Quadrella ferruginea within his het-
erogeneous Linnaeobreynia (= Quadrella subg. Breyniastrum), but Q. ferruginea is kept in Quadrella subgenus
Intutis because the eight staments and seeds with white embryos.
Quadrella angustifolia, the first syntype cited by Rafinesque in the protologue of Octanema, is the
herein designated lectotype of Octanema because of its flowers with eight stamens and seeds with a white
embryo. Quadrella angustifolia and Q. incana, both species placed by Rafinesque in Octanema were included
by Hutchinson (1967) within Linnaeobreynia.
Distribution.— Quadrella subgenus Intutis ranges from southern NorthAmerica (Texas) through Mexico
to northern Honduras, and in the West Indies from Cuba to Hispaniola.
KEY TO THE SPECIES OF QUADRELLA SUBGENUS INTUTIS
1 . Calyx with open aestivation, sepals triangular to linear, distinctively shorter than petals from buds.
ii i fi
litis and Cornejo, Synopsis of Quadrella
2. West Indies species.
3. Branches and leaves, especially beneath, with a dense, ± rusty cow
gular, 1-25 mm, distinctively smaller to half of the petals length.
4. Leaf blades lanceolate; sepals linear-triangular; the common subspecies of Jamaica and Dominican
Republic Quadrella ferruginea subsp.
4. Leaf blades spathulate-lanceolate; sepals triangular; Cuba Quadrella ferruginea subsi
3. Branches and leaves covered by yellowish white trichomes, the leaves beneath with a mixed tomentum
of stellate and dendritic hairs; sepals linear, (2— )3— 5(— 6) mm, two third to as long as the petals.
5. Leaf blades above covered with soon deciduous sessile stellate trichomes, shaped like a sea-urchin
ora stout bottlebrush, and with persistent, sharply pointed conical protuberances
domingensis subsp. do
5. Leaf blades above initially dense, stellate, yellow, velvety pubescent, becoming glabrous, smooth
and shiny, sharply pointed conical protuberances on the leaf blades absent Quadrella do
subsp. i
2. Continental species, from southeastern Texas, south to Guatemala and rarely into northern Honduras.
6. Sepals linear to filiform, 2-5 x 05-0.9 mm; staminal filaments 6-9 mm.
7. Leaf blades lanceolate or narrowly elliptic to oblanceolate or obovate.
7. Leaf blades broadly obovate to rhomboid; mostly
i. Sepals triangular, 1-1.5 x 1-1.3 n
rella incana subsp. incana
scattered across the range of the
cana subsp. yucatanensis
isco Quadrella
_ Quadrella singularis
I. Calyx with closed aestivation, sepals oblong, covering the petals in bud nearly to anthesis.
8. Leaf blades narrowly oblong to oblong-lanceolate; petioles 1 -5 mm; calyx valvate-reduplicate in flowerbud;
southern Mexico _Quadrella ai
8. Leaf blades ovate to elliptic or slightly obovate; petioles 6-15 mm; calyx valvate but not reduplicate in
flowerbud; West Indies.
9. Stamens 8, filaments ca. 8 mm; gynophores 6-9 mm; Dominican Republic — __ Quadrella
9. Stamens ca. 16, filaments 12-14 mm; gynophores 12-1 8 mm; eastern Cuba _
1. Quadrella alaineana Cornejo & litis, J. Bot. Res. Inst. Texas 4:75. 2010. Ty
; Dominican Republic (Cornejo & litis 2010).
(iktotwe, designated here, P).
Quadrella angustifolia is a rare, highly local endemic in two super arid desert pockets inland from the Pacific
coast in the southern Mexican states of Michoacan and Guerrero. This species has xerophyllous, linear
leaves and thick-walled, often tardily dehiscent siliquiform capsules, the latter a condition which induced
Eichler (1865) to place it in Capparis subg. Colicodendron. The species is placed in Quadrella because of its
calyx with closed reduplicate valvate aestivation and eight stamens. The capsules are similar to those of Q.
asperifolia and Q. indica in subg. Breyniastrum, which suggests that our classification of Quadrella is somewhat
arbitrary. Molecular biology is needed to understand the true relationships.
3 Quadrella domingensis (Spreng. ex DC.) litis & Cornejo, comb, i
3a- Quadrella domingensis (Spreng. ex DC.) litis & Cornejo subsp domingensis
Quadrella domingensis subsp domingensis is a local taxon on Hispaniola, with the type collection (fide Urban
1920-21) from Azua, Santo Domingo. It is characterized by having persistent, sharply pointed conical
Protuberances above the leaf blades (Rankin & Greuter 2004).
3b Quadrella domingensis subsp. grisebachii (Eichler) litis & Cornejo, comb. nov. Ba
130
4a. Quadrella ferruginea (L.) litis & Cornejo subsp. ferruginea
Quadrella ferruginea subsp. ferruginea is a common subspecies of Jamaica The lanceolate leaf blades and
linear-triangular sepals characterize this susbpecies (Rankin & Greuter 2004).
4b. Quadrella ferruginea subsp cubensis (R Rankin) litis & Cornejo, comb. nov. Bask
Quadrellaferruginea subsp. cubensis is a local endemic of eastern Cuba. It is characterized by the spathulate-
lanceolate leaf blades and triangular sepals (Rankin & Greuter 2004: 274).
Quadrella incana is
shape, Quadrella in
the only species of subg. Intutis in the Mesos
ana is a rather uniform species as for flowers ;
liform embryo surrounded by a hard brittle I
flora. Except for its variable teaf i
s. It can be identified by its white
: leaves with fewer (3 to 8) lateral
5a. Quadrella incana (Kunth) litis & Cornejo subsp. incana
Quadrella In earn subsp. incana is widespread from Texas, eastern and southwestern Mexico to Guatetnab i
{Contreras 11084 ) and barely into Honduras (Molina & Molina 26028 ) (litis & Cornejo 2007a).
5b. Quadrella incana var. triangularis Cornejo & II
ca. 8 km NW of Chamela, 19°33’N 105°08'W, 2 m, 22 Jun 1
Quadrella incana var. triangularis is known by only
distinctive smaller (1-1.5 x 1-1.3), triangular sepal-
variety (Cornejo & litis 2010).
tis, J. Bot. Res. Ins. Texas 4:80. 2010. '
984, H.H. Otis & S. Wisniewski 29194 (holotype:
two collections from Chamela (Mex
> and smaller filaments (ca. 4 mm),
Type: MEXICO. Jausco:
WIS; isotype: MEXU)-|
' A- Lun^ett 7452 (holotype: MICH; i
tna subsp. yucatanensis is n
Ins. Texas 4:83. 2010. Basionym: Capp**
: Chichen ItzS, off Kaua road, 8 Jun 1938 (b)M
TEN NEW MYRTACEAE FROM EASTERN AND NORTHEASTERN BRAZIL
Marcos Sobral
Departamento de Ciincias Naturais UFSJ
Praga Dorn HeMcio 74-36301-160
SaoJoao del-W MG, BRAZIL
marcos_5obral@hotmail.com
ABSTRACT
RESUMO
The family Myrtaceae is represented in Brazil by about 1000 species (Landrum & Kawasaki 1997:508).
Species of the family occur in all vegetation types in Brazil, but they seem to be especially frequent in the
coastal rainforest (Floresta AtlSntica) that ranges from northeastern to southern Brazil, as was recorded in
some inventories (e.g., Mori et al. 1983; Lima & Guedes-Bruni 1997; Thomaz & Monteiro 1997). During the
examination of exsiccates kept at some Brazilian herbaria, I had the opportunity of studying some species,
mainly from the Atlantic Forest domain, that I consider here as undescribed.
1- Eugenia L
Eugenia is characterized mostly by the presence of tetramerous flowers with two locules and several ovules
per locule and seeds with completely fused cotyledons without any visible hypocotyl. The genus is pantropi-
cal, and presents about 350 species in Brazil (Landrum & Kawasaki 1997).
11. Eugenia azeda Sobral, sp. nov. (Figs. 1-3).
Aeronautica, 5°54'30"S, 35°13'30"W, 15 Dec 19<
Species Eugeniae ligustrinae foliis subsessilibus fructis majoribus recedit.
Shrub to small tree 2-9 m. Twigs terete, gray when dry, the most distal ones to 1 mm in diameter, glabrous
very minutely stngose with simple erect trichomes to 0.1 mm, falling off in longitudinal stripes; internodes
7-20 mm. Leaves subsessile, with petioles 1-2 x 1.5 mm, frequently evident only when the leaf is viewed
abaxially; blades elliptic to ovate-elliptic, 20-50 x 10-22 mm, discoloured when dry, the abaxial face paler;
glandular dots smaller than 0. 1 mm in diameter, 10-30 per square milimeter, ocasionally evident adaxially but
clearly visible and darker than the surface abaxially; apex obtuse or rounded, sometimes widely acute; base
°btuse to rounded, sometimes very slightly auriculate; midvein sulcate adaxially and prominent abaxially;
!ateral veins 6-8 pairs, markedly prominent adaxially and faintly so abaxially, leaving the midvein at angles
of about 45 degrees; marginal vein to 1 mm from the revolute margin. Inflorescences terminal, with two
Journal of the Botanical Research Institute of Texas 4(1)
135
flowers on an auxotelic axis 1-4 x 0.4-0.8 mm that
produces adult leaves after anthesis; bracts at the base
of the axis lanceolate-obovate, 4 x 0.8-1 mm; pedicels
8-25 x 0.2-0.3 mm, with simple erect trichomes to
0.1 mm; bracteoles missing, probably deciduous before
anthesis; flower buds obovate, glabrous, to 5 x 4 mm,
the ovary markedly distinct from the calyx; calyx lobes
four, oblong-lanceolate, 5-6 x 2 mm, concave, gla-
strongly reflexed and concealing the ovary at anthesis;
petals obovate, glabrous, white, to 8 x 5 mm; stamens
about 50, 5-6 mm, the anthers elliptic, to 0.6 x 0.3
mm, eglandular; staminal ring to 3 mm in diameter,
occasionally with scattered trichomes to 0.2 mm; style
4-6 mm, the stigma punctiform, minutely papillose;
per locule. Fruits globose, yellow when ripe, to 30 mm
in diameter when fresh, about 20 mm when dry, with one seed, this slightly reniform, to 10 x 8 mm, with
a gray, easily detachable testa and embryo with fused cotyledons and hypocotyl not visible.
Habitat, distribution and phenology.— This species is a shrub or small tree from coastal forests (resting^
and seasonally deciduous forests) of the municipalities of Natal and Pamamirim, at the northeastern state
of Rio Grande do Norte, growing at altitudes near the sea level. Flowers were collected in November and
December, and fruits in December and February.
Conservation. — According to IUCN criteria (IUCN 2001), Eugenia azeda can be considered as endangered
(EN), fitting criteria B1 ab(iii): its presently known area of occurrence is smaller than 5000 km2 (criterion
(criterion b(iii)), since Natal, the capital of the state of Rio Grande do Norte, and neughbouring municipali-
Affinities. — This species is close to Eugenia ligustrina (Sw.) Willd. (for description see Berg, 1857-1859:343),
from which it is set apart by the following characters:
. Blades cuneate at base, with evident petioles to 4 m
. Blades rounded at base, subsessile, the petioles to 2 mi
Etymology. — The epithet is derived from the local name for the species, “ubaia-azeda.” The vernacular name
being Eugenia pyriformis Cambess. (Sobral 2003), and “azeda” is the Portuguese word for “bitter”— the fruits
are widely appreciated locally for their bitter taste.
Paratypi v BRAZIL. Rio Grande do Norte, mun.: Natal, 8 Nov 1951, Aharenga 32 (BHCB, RB); 26 Feb 2001, Cestaro s.n. (BHCB); 10
Dec 2006, Sobral & Cestaro 10887 (BHCB, BRIT, MBM); 10 Dec 2006, Sobral & Cestaro 10888 (BHCB, BRIT, K, MBM, RB).
1 .2. Eugenia vakuganana Sobral. sp nov iFig.4* r.rt brazil e up .. > m . •„ . k.,,.. v..i..ic>iu v.lh.i
Biol6gica de Santa Lilcia, 28 Jan 1999, L Kbllnumn, E. Bausen & W Pizziolo 1 737 (holotype: MBML; isotypes: BHCB, BRIT). J
Tree 4-7 m. Plants glabrous, except for gray trichomes to 0.1 mm on the ovaries. Twigs gray, complan-
ate, to 2 mm in diameter, the internodes 15-20 mm. Leaves with petioles 6-8 x 1-1,7 mm, drying black;
blades lanceolate to lanceolate-oblong, 90-140 x 25-40 mm, pale light green and concolored when dry,
with 10-15 glandular dots per square milimeter, these smaller than 0.1 mm in diameter and easily visible
on both sides; apex acute or acuminate to 10 mm; base cuneate; midvein prominent on both sides, more
70-80 degrees; marginal veins two, 4-5 mm and 0.5-1 mm from the margin, the margin itself revolute.
Inflorescences axillary, fasciculiform, with 4-8 flowers crowded in an axis to 3 x 2 mm; pedicels absent or
occasionally developed in fruits, then to 2 x 1 mm; bracteoles widely triangular, to 1.5 x 1-1.5 mm, densely
covered by convex glands to 0.1 mm in diameter, flower buds to 3 x 2 mm, the ovary densely pilose and
clearly distinct from the glabrous calyx lobes, these rounded, glabrous, to 2 x 2 mm, with glands as on the
bracteoles; petals rounded, to 2 x 2 mm in buds; stamens about 60, to 2 mm in bud, the stammal ring t '
1 .5 mm in diameter; style to 2.5 mm in bud; ovary with two locules and 6 ovules per locule. Fruits elliptic.
20-26 x 16-20 mm, yellow when ripe, glabrous, crowned by calyx lobes to 3 x 3 mm, with one seed, this
elliptic, to 22 x 12 mm, with an easily detachable chartaceous, gray testa and embryo with two fused, in*
distinguishable cotyledons and no evident hypocotyl.
Habitat, distribution and phenology.— This species is a small tree from montane forests, growing at alti-
tudes of about 600 m above sea level. It is presently known only from the municipality of Santa Teresa, in
the eastern Brazilian state of Espirito Same. Flowers were collected in October and fruits in January. ’
Conservation.— According to IUCN criteria (IUCN 2001), Eugenia valsuganana can be considered as
endangered, fitting criteria B1 ab(iii): its known area of occurrence is smaller than 5000 km2 (criteriu®
138
Bl), it grows in a severely fragmented habitat (criterium a) that presents a continuing decline in extension
(criterium b(iii», since only 18% of the area of Santa Teresa still retains its original vegetation (Mendes &
Padovan 2000:16).
Affinities. — This species is appare
1857-1859:575), a species collected in
by the following characters:
1 . Blades light brown when dry, to 130 x 60 mm, the ratio length/width about 2:1; midvein adaxially sulcate;
flowers without sharp contrast between ovary and calyx lobes, both equally pilose Eugenia plicatoc<
1 . Blades pale light green when dry, to 140 x 40 mm, the ratio length/wrlth J - 3 5 l,n iidvem adaxially prominent
or biconvex; flowers with ovary pilose and calyx lobes glabrous Eugenia valsugi
Etymology. — Alludes to the locality of the type collection.
) Eugenia plicatocostata O.Berg (for c
a and Espirito Santo, from which it c
Myrcia is an American genus with about 400 species, characterized by mostly paniculiform inflorescences, |
pentamerous flowers with bi- or trilocular ovaries with two ovules per locule and embryos with well-
developed hypocotyl and cotyledons (Landrum & Kawasaki 1997). It is one of the four genera of subtribe
Myrciinae as provisionally accepted by Lucas et al. (2007) — Calyptranthes Sw., Gomidesia O.Berg, Marlierea
Cambess. and Myrcia. The separation between these genera relied mostly in calyx characters: Calyptranthes '
and Marlierea present fused calyx lobes (opening through a calyptra in the first genus and tearing in lobes in
the second) and Myrcia and Gomidesia free calyx lobes ( Gomidesia being distinguished from Myrcia through
its sinuose opening of the anthers). Such characters are feasible, and indeed the results of Lucas et al. point
to this assemblage as more phylogenetically close than was thought before, since in her cladograms there is r
no consistent separation of these genera. So, 1 have decided here to take Myrcia in a wider circumscripticoj
naming under it three species that might also be included under Marlierea (Myrcia crassa, Myrcia mucugensis
and Myrcia pseudomarliered). Indeed Kawasaki & Holst (1994:141) made a similar decision when they de-
scribed Myrcia rupta, a species that otherwise would be assigned to Marlierea.
It is worth noting that among these genera Calyptranthes has nomenclatural priority over Myrcia and
the others (McVaugh 1956). Moving all species of Myrcia to Calyptranthes would be quite undesirable as
was adequately pointed by McVaugh (1968), so, it is convenient to consider the possibility of proposing the
conservation of Myrcia over Calyptranthes.
2.1. Myrcia crassa Sobral, Sp nov. ^Figs. 5-61. Tvrr BRAZIL E.mkih. Santo, mun.: Santa Teresa. 25 Oct 2000. V Dcmunat
Small tree 1.5 m to tree 18 m. Twigs glabrous, to 2-5 mm in diameter, subterete; intemodes to 60 mm. Leaves
with petioles 8-20 x 2,2-3,2 mm; blades obovate, elliptic or ovate-elliptic, 100-200 x 60-110 mm, stiff,
discoloured when dry; glands from 0.05 to 0.1 mm in diameter, about 10 per square milimeter, sometimes
visible adaxially but more easily observed against light; apex widely acute to obtuse, sometimes abruptly |
acuminate to 10 mm; base cuneate or obtuse; midvein sulcate adaxially and prominent abaxially; laterals
veins 15-20 pairs, prominent on both sides, leaving the midvein at angles 60-70 degrees; marginal veins
two, 2-3 mm and 0.5 mm from the margin. Inflorescences paniculiform, with 30-60 flowers and two to
three levels of ramification, the first branch occasionally with two contiguous ramifications, the axis 40-12°
x 2-4 mm; flowers sessile, densely covered with brown simple appressed trichomes to 0.1 mm; bracteoks
ovate-lanceolate, 1-1,5 x 0,8-1 mm; flower buds obovate, 3-3,5 x 2 mm, with four rounded lobes visible at
the apex, at anthesis tearing the calyx tube somewhat irregularly, 1, 2-1,5 x 1,5-2 mm; petals two to three,
rounded, 2-3 x 2-3 mm; stamens 60-70, 3-4 mm, the anthers elliptic, to 0.5 x 0.3, eglandular; staminal ring
139
FlG' S Myrcia cr™<> - image of holotype.
Journal of the Botanical Research Institute of Texas 4(1)
description see Berg 1857-1859:145 and
Legrand & Klein 1971:479), from which it is
distinguished by the following characters: \
t width of the twigs and the stiffness
2 (BHCB, MBML).
Tree to 6 m high. Bark brown or orange brown, exfoliating; twigs glabrous, 1-2 mm in diameter, terete*
exfoliating longitudinally, with the same color as the bark, the internodes 25-50 mm. Leaves with petioles
canaliculate, 5-7 x 1.2-1.5 mm; blades lanceolate or elliptic, 110-150 x 20-45 mm, 3-5 times as long as
wide, sometimes markedly bullate, discoloured when dry, with white, dibrachiate, asymmetrical trichomes
0.3-0.4 mm, scattered and curled, occasionally absent adaxially and more dense and appressed abaxially,
glands more visible abaxially, with less than 0.05 mm in diameter and about 10 per square milimeter,
apex acuminate to 10-15 mm; base cuneate; midvein sulcate adaxially and strongly prominent abaxially;
lateral veins 17-20 pairs, sulcate or plane adaxially and prominent abaxially, leaving the midvein at angles
2 mm and 0. 2-0.6 mm from the margin, the margin
60-70 degrees! smaller ??]; marginal '
Sobral, New Myrtaceae from Brazil
142 Journal of the Botanical Research Institute of Texas 4(1)
itself with a yellow thickening to 0.2 mm. Inflorescences drying dark
purple or black, purple [when fresh], axillary or occasionally terminal,
paniculiform, pyramidal, ramified two to four times, with dibrachiate I
trichomes like those of the leaves, the main axis 90-150 x 0.7-1 mm
below the first branching, the internodes 10-15 mm, with 1-6 suc-
cessive branchings, these nearly perfectly opposite one to the other,
apparently bearing terminal flowers; pedicels in strict sense possibly 1
absent, the last branches of the inflorescences simulating pedicels, 1-5 1
x 0.3 mm, occasionally with glabrous, linear [bracts] to 1 x 0.2-0.3|
mm; bracteoles the same form and size of hypsophylls; flower buds
white, obovate, 2-2.5 x 1.7-2 mm, glabrous; calyx-lobes five, visible
in bud, glabrous or with scattered trichomes to 0.1 mm adaxially and
cilia to 0.1 mm, unequal, three widely triangular, 0.8-1 x 1 mm, two
hemispheric, 1.5-1.8 x 2 mm, strongly reflexed at anthesis; petals
white, orbicular, 2 x 2.5 mm, glabrous or with scattered trichomes to J
Fig. 8. Myrcia floridissima - detail of inflores- 0.1 mm abaxially, reflexed at anthesis; stamens 60-70, white, 6-7 mm,
cences [franca 571; scale: 10 mm). the anthers hemispheric, 0.2-0.3 x 0.3, opening through longitudinal
slits, eglandular; staminal ring to 3 mm in diameter, with scattered
white trichomes to 0.1 mm; hypanthium 0.5-0.7 mm deep; style 6 mm, the stigma slightly papillose; ovary i
2-locular, with two ovules per locule. Fruits unknown.
Habitat, distribution and phenology. — Myrcia floridissima is a tree from rainforests (Floresta Atlantica) of
eastern Minas Gerais, where it was collected at altitudes about 200 to 500 m above sea level. This species
is presently known only for the state park of Rio Doce, in the municipalities of Marlierea and Dionisio, in
the eastern portion of the southeastern Brazilian state of Minas Gerais. Flowers were collected in April and
September.
Conservation. — Considering the presently available information on this species, it can be considered
as an endangered one (EN; IUCN 2001), since it fits criteria B1 ab(iii), that is, its presently known range of
occurrence is smaller than 5000 km2 (criterium Bl) it grows in a severely fragmented habitat (Aguiar et al.
2005:121) (criterium a) and its habitat presents a continuing dechne in extension (criterium b(iii)), due ttj
intense land use in southern Bahia, such as urban expansion and extensive cocoa plantations.
Affinities. This species resembles Myrcia pertusa DC., presently known for the Brazilian state of Ama-
zonas and from Peru (for description see McVaugh 1958:656), and the sympatric Myrcia lacunosa (O.Berg)
N.Silveira (for description see Berg, 1857-1859:545). Myrcia pertusa is set apart by the presence of simple
trichomes, smaller petioles (to 2 mm), longer bracteoles (to 5 mm), subequal, rounded and strigose calyx
lobes and flowers with 250-300 stamens, while M. lacunosa is distinguished by its shrubby or semi-scandent
habit, twigs and inflorescences with simple erect trichomes, petioles to 2 mm, bracteoles to 5 mm and densely
strigose flowers with triangular calyx lobes.
Etymology.— The specific epithet is derived from the Latin superlative for “flowered,” alluding to the
profusely branched inflorescences of the type specimen.
, Mumbaca, 19°48'18"S, 42°32'35"W, 6 Sep 2004,
2.3. Myrcia mucugensis Sobral, sp. nov. (Figs. 9-10). Type: BRAZIL. Bahia, mun.; Mucuge, 9 Sep 1988, R. Krai & M.C
Shrub 1.5-2 m tall; cortex gray, longitudinally rugose, glabrous, not exfoliating; twigs subterete to complan-
ate mid moderately keeled, to 2 mm wide, densely covered with gray or brown arachnoid simple trichomes
^•Wywan
s - image of isotype.
more dense than the adaxial one, the |
trichomes on the adaxial side occasionally V
deciduous in adult leaves, glands virtually Ij
invisible, generally evident against light,
smaller than 0.05 mm in diameter and
ca. 10 per square milimeter; apex acute
to widely acute; base cordiform; midvein
slightly sulcate adaxially and prominent
abaxially; lateral veins 6-7 pairs, invis-
ible on both sides, leaving the midvein at
angles about 60 degrees; marginal vein
not evident, the margin itself strongly
revolute for at least half the width of the
completely the abaxial side. Inflorescences J
axillary, capituliform, as densely covered
with brown trichomes as the twigs, with
1 12-15 x 1. 5-1.8 mm wide at the
with generally t
Fig. 10. Myrcia muaigensis - detail of inflorescence (scale: 10 mm).
8-12 sessile flowers crowded at the apex of a s
apex, the axis occasionally with two ramificati
bracts at the base of the ramifications or sub
x 1 mm; bracteoles mostly abortive, sometin
0.2 mm; flower buds globose-obovate, dens
1. 5-1.8 mm, with four subequal triangular-c
opening regularly or tearing very slightly thi
5, suborbicular, white, to 1 x 1.5 mm, with scattered brown dibrachia
50, 3-4 mm, the connective with at least one apical gland ca. 0.05 mm, the anthers orbicular, 0.
opening through longitudinal slits; staminal ring glabrous, 1-1.5 mm in diameter; hypanthiur
deep, glabrous; style 5 mm, the stigma minutely capitate and papillose; ovary 2-locular, with
per locule. Fruits globose, to 2 mm in diameter, unripe, either crowned with calyx lobes or wit
es present, linear, to 1.2 x 0.3 m
ly covered with brown arachnoid trichomes to 0.2 mm, 2 x
rate calyx lobes 1 x 1-1.2 mm, with trichomes on both faces,
staminal ring, occasionally falling after anthesis; petals 4 o&j
Habitat, distribution and phenology.— So far This species has been collected only in “campos rupestres
(rocky open habitats)of the municipality of Mucuge at central Bahia, at altitudes 800-1000 m above sea level.
Conservation. — Although Myrcia mucugensis is known from only one municipality, I could not find ad-
ditional informations relative to the present conditions of its habitat; considering this lack of information,
this species can be considered as data deficient (DD), as is the case when “there is inadequate information
to make a direct, or indirect, assessment of its risk of extinction based on its distribution and/or population
status” (IUCN 2001).
Affinities. — This species is apparently related to Myrcia densa (DC.) Sobral, a species also collected in
campos rupetres of central Bahia (for description see Berg (1857-1859:69), under Aulomyrcia densa (DC>j
0. Berg). Both species can be distinguished by the characters in the following couplet:
1 . Leaves with petioles to 3 mm; blades glabrous or with scattered trichomes, cuneate at base, the margin not
revolute; inflorescences paniculiform, two to three times ramified, the secondary branches at least 5 mm
long; flowers with five calyx lobes Mvrcia densa
145
. Leaves sessile; blades densely pilose at least abaxially, the base markedly cordate, the margin strongly revo-
i:,t~ ri iMti„ ii . .. -ill',.) ill. -I 1.1,1 i I- inflorescences capitulilbrm, occasionally with only one second-
ary branch to 1 .5 mm; flowers with four calyx lobes. Myrda i
Trees 3-15 m tall. Branches pendulous, terete or slightly flattened, ocasionally weakly channeled longitudi-
nal^ 4_5 mm in diameter, with dense simple red-brown erect trichomes to 8 mm long (occasionally deciduous
in older twigs), occasionally with straight to slightly curved interpetiolar scars; internodes 40-100 mm long.
Leaves sessile or shortly petiolate, the petiole when present to 1 x 4 mm, frequently evident only abaxially;
blades ovate-lanceolate, 130-200 x 58-80 mm, 2-2.5 times as long as wide, slightly discoloured when dry,
lighter abaxially; glands not evident on the surfaces but visible against light, to 0.1 mm in diameter, about
15 glands per square milimeter; apex acute or acuminate in 3-5 mm; base rounded, frequently with the
margins folded in dried specimens; midvein sulcate adaxially, with red, brown or gray simple trichomes to
1 mm, strongly prominent abaxially, there with erect brown trichomes to 5 mm; lateral veins 18-25 pairs,
sulcate adaxially and with scattered trichomes as on the midvein, prominent abaxially, [ ] with trichomes as
the midvein but these 0.5-1 mm; marginal veins two, 2.5-6 mm and 1-1.3 mm from the margin. Inflores-
cences with dense brown erect trichomes to 0.5 mm, axillary or terminal, pendulous, mostly capitate, with
3-10 flowers crowded at the apex of an axis, occasionally racemiform or paniculiform and then mostly with
one level of ramification the main fertile axis 40-120 x 1-2.5 mm below the first branching, the secondary
axes when present 2 at a same point, to 5 x 1.5 mm; bracts triangular, 10-12 x 5 mm, abaxially with brown
appressed trichomes to 1.2-2 mm; pedicels absent; bracteoles triangular as the bracts, 9-10 x 4-6 mm, vis-
ibly carenate, abaxially with trichomes as on the bracts, persisting after anthesis; flower buds globose, 8-10
x 9 mm, densely and evenly covered with simple red brown trichomes 1-2 mm; calyx lobes five, triangular
to widely triangular, somewhat unequal in size, 4-5 x 5-7 mm, abaxially with trichomes as on the rest of
the flower bud and adaxially with gray appressed trichomes to 0.2 mm; petals rounded, to 9-10 x 10 mm,
stamens about 200, to 6 mm, the anthers linear-rectangular, to 1 x 0.3 mm, opening longitudinally with
a very slight sinuosity, without evident glands; staminal ring 5-6 mm in diameter; hypanthium absent or
mostly 0.5 mm deep; style to 10 mm, stigma slightly capitate; ovary 2-locular, with 2 ovules per locule.
Fruits globose, smooth, only immature collected, to 20 mm in diameter, densely covered with trichomes as
the flowers; seeds one or two per fruit, incompletely developed.
Habitat, distribution and phenology— Myrda pendula was collected in rainforests in the municipalities of
Ibirapitanga and Camamu, at altitudes from 600 to 700 m above sea level; flowers were collected in March
and May and fruits in February.
Conservation.— According to 1UCN criteria (1UCN 2001), Myrda pendula can be scored as endangered
(EN), fitting criteria B1 ab(iii), that is, its presently known range of occurrence is smaller than 5000 km2
(criterium Bl), it grows in a severely fragmented habitat (Aguiar et al. 2005:121) (criterium a) and its habitat
presents a continuing decline in extension (criterium b(iii)), due to intense land use m southern Bahia, such
as urban expansion and extensive cocoa plantations.
Affinities. — This species is apparently related to Myrda crocea (Veil.) Kiaersk. (for description, see Berg
(1857-1859:533), under Gomidesia crocea (Veil.) O.Berg), another species from atlantic rainforests that
grows from Bahia to Sao Paulo; these species can be distinguished by the characters given in the following
■; fruits longitudinally sulcate
. Myrcia crocea
Journal of the Botanical Research Institute of Texas 4(1)
| HE!
HERBARIO-
Sobral, New Myrtaceae
147
1 inches pendulous; plants with trichome
row; fruits smooth
Etymology. — The epithet is derived fro
branches of this species in nature.
fWvres BRAZIL. Bahia. Mini. Ibirapitanga
tin word for “pendulous,” alluding to the position of t
l5- Myrcia pseudomarlierea Sobral, sp. nov. (Figs. 15-14). Type: brazil i
emrada ae^quenl i ii i r u ii n i L- .1 — iI:i'kih.i wi 9 S km. ranul .1 esqwnla do P.«
« 39°6-W, 23 Oct 2003, P Fioschi, A. Amorim, S. Smt'Ana, J.L PavOo &C.W.Esptyl 747
Species Marhcreae ajjini proxima, a qua foliis majoribus scssilibus, basi cordati et apicis non
Journal of the Botanical Research Institute ofTexas4(1)
Sobral, New Myrtaceae from Brazil
r slightly
130-320
Trees 4-20 m. Branches and leaves glabrous. Twigs t
1 flattened, 4-7 mm in diameter; internodes 30-100 mm I
sile; blades lanceolate, obovate-lanceolate or ovate-lane
x 55-150 mm, 2-2.5 times as long as wide, discoloured when dry,
the abaxial surface lighter; glands not visible against light, moderately
H excavated in the adaxial side and occasionally barely prominent at the
abaxial one, less than 0.1 mm m diameter, 10 to 15 glands per square
milimeter; apex acute or rounded; base cordate or rounded, frequently
folded or breaking in dried material; midvein sulcate or sometimes flat
adaxially, strongly prominent and darker than the rest of the surface
— — - abaxially; lateral veins 22-25 pairs, evident and moderately prominent
on both sides, with about the same number of thinner secondary lateral
veins; marginal veins two, 2.5-5 mm and 1-1.5 mm from the margin.
Inflorescences with dense and frequently deciduous glistening brown
trichomes to 0.2 mm, terminal, paniculiform, with three to four levels
tis flattened, to 150 x 3-4 mm below the first branching, the secondary
mm; bracts, pedicels and bracteoles absent; flower buds obovate, to 4 x 3
mes to 0.2 mm; calyx lobes four, fused between them except apically and
regularly, strongly unequal, one of them larger and rounded, sometimes
1 5-1.7 mm, with evident glands; petals absent; stamens 90-100, to 4 mm, the anthers subquadrate, 0.3 x
0.2-0.3 mm, opening through longitudinal slits, the connective bearing one evident gland; staminal ring
2-3 mm in diameter; hypanthium 1.5-2 mm deep; style 6-7 mm, stigma punctiform, softly papillose; ovary
2-locular, with 2 ovules per locule. Fruits globose, red when ripe, 12-15 mm in diameter; seeds one per fruit
(in three examined fruits) subreniform, 12-15 x 12 mm, the testa light brown, shining, easily detachable,
embryo with two folded cotyledons surrounded by an evident hypocotyl.
Habitat, distribution and phenology. — This species grow in coastal rainforests of the municipalities of
Camamu, Itacare and Uruquca in southern Bahia, at altitudes 100-500 m above sea level. Flowers were
collected in October and December; fruits in March and May.
Conservation.— Through IUCN criteria (IUCN 2001), this species could be considered as endangered
(EN), fitting criteria B1 ab(iii), since its presently known area of occurrence is smaller than 5000 km2 and
its habitat is severely fragmented and suffering a continuing decline (for explanations on criteria and refer-
ences, see comments under Myrcia pendula).
Affinities.— This species resembles Marlierea qffinis (O.Berg) D.Legrand (for description see Berg
(1857-1859:149), under Eugeniopsis affinis O.Berg), from which it can be distinguished by the characters in
the folowing couplet:
’■ Leaves with petioles to 16 mm long and blades to 200 x 60 mm, with acuminate apex and cuneate base;
flowers with calyx lobes rounded ' Marlierea affinis
]- Leaves sessile, to 320 x 1 50 mm, with acute apex and cordate base; flowers with unequal calyx lobes, o
them somewhat calyptriform .. —
Etymology. — The epithet is derived from the Greek work meaning “false and
resemblance of this plant to a species presently assigned to Marlierea.
alluding to the dost
al. 282 (CEPEC). Man. Itacare: APA de Itacai
5 (CEPEC); estrada que liga a torre da Embrai
- Benton 12857 (CEPEC); mata parcialmente p «
1991, Amorim et al. 399 (CEPEC). Man. Urup
i-101/ltacart, a 5,8 k
Journal of the Botanical Research Institute of Texas 4(1)
(Figs. 13-16). Typ,
mm. Twigs terete or slightly flattened, 3-4 mm in diameter, scarcely or not longitudinally striate, occasionally
with straight to slightly curved interpetiolar scars; intemodes 40-80 mm long. Leaves petiolate, the petiole
2.5-4 x 2-2.5 mm, frequently evident only abaxially; blades lanceolate, oblong-lanceolate or ovate-lanceolate, 1
130-220 x 30-55 mm, 4-4.5 times as long as wide, discoloured when dry, lighter abaxially; glands visible^
against light, moderately prominent abaxially, to 0.1 mm in diameter, 6 to 10 glands per square milimeter;
apex acuminate (acumen 15-20 mm) or acute; base cordate, rounded or slightly auriculate above the petiole;
midvein biconvex adaxially, prominent abaxially; lateral veins 16-27 pairs, evident and moderately promi-
nent at least abaxially, with about the same number of much thinner secondary lateral veins; marginal veins
simple trichomes to 0.3 mm, terminal, paniculiform, with two to three levels of ramification, the main fer-
tile axis 80-130 x 2.5-3 mm below the first branching, the secondary axes 2 at a same point, 20-40 x 1-2
mm; bracts mostly absent, when present linear-lanceolate, those at the base of the first branching to 10 x
2 mm, those of the higher branchings to 5 x 0.5 mm; pedicels absent; bracteoles lanceolate, to 2 x 0.3-0;5fj
mm, deciduous at anthesis; flower buds obovoid or pyriform, to 5 x 3 mm, with simple, ochraceous or gray
persistent trichomes to 0.2 mm densely covering the ovary, the calyx lobes with somewhat clearer and less
dense trichomes to 0.3 mm, the ovary markedly wrinkled in dried specimens; calyx lobes five, triangular, •'
slightly unequal, 1-1.5 x 2 mm; petals rounded or elliptic, 3-6 x 2-5 mm; stamens 40-50, to 6 mm, the
angula
mm> openinj
in diameter; hypanthium to 1 mm deep; style to 8 mm; stigma pun
per locule. Fruits glabrous, globose, purple or black when ripe, 10-1
fruit, to 10 mm, the testa easily detachable, embryo with two folder
hypocotyl, this with simple white trichomes to 0.3 mm near the proximal part.
Habitat, distribution and phenology. — Myrcia pseudospectabilis grows in coastal rainforests of
Bahia, at altitudes from 170 to 510 m above sea level; flowers were collected in November and fr
February to July.
finally, eglandular;
form; ovar
otyledons
ing about 2 mm
ith 2 c
Conservation. — Tl
(EN),fi
cies can be considered, according to IUCN criteria (IUCN 2001) as endangered
i), since its presently known area of occurrence is smaller than 5000 km2 and
its habitat is severely fragmented and declining (for explanations on criteria and references, see comments
under Myrcia pendula).
Affinities. This species is apparently related to Myrcia spectabilis DC., a species widespread in coastal
Brazil from Bahia to Santa Catarina (for description see Berg (1857-1859:12), under Gomidesia spectabilis
(DC.) O.Berg); both species can be set apart by the following characters:
1. Tree to 15 m tall or more; twigs with an interpetiolar callose scar; leaves
3 200 x 70 nr
. Tree to 6 m tall; twigs devoid of inter
ar; leaves with blades to 220 x 55 nr
Etymology.— The epithet is derived from the Greek word for “false” an
blance between this species and Myrcia spectabilis.
rami vicinal 5.5 km i direita para a Lagoa Encantada, 14°35'10"S, 39o07'2rW, 19 Feb
de Serra GrLde pa E ^ ^ mUniclpio’ da 0nCa, ca 4 5 km da entrada,
“toSZT® 2 A|” 2004’
1. 2726 (CEFEO. Mun. !
tim et al. 854 (CEPEC); e
*S, 39°06’54"W, 18 Mar
l- U363 (BHCB, CEPEC). »
Sobral, New Myrtaceae from Brazil
155
* * Plink espinhacensis - image of holotype.
157
:r; hypanthium 1-1.5 mm deep;
vules per locule. Fruits unripe,
0. 3-0.4 x 0.3 mm, without visible glands; staminal ring to 3 mm in
style 7-8 mm; stigma punctiform and papillose; ovary bilocular, \
elliptic, 7-8 x 5 mm, 8-ridged longitudinally, 1-seeded; embryo wit!
without visible hypocotyl.
Habitat, distribution and phenology. — This species is a tree from montane gallery forests or dry forests
(“carrascos") from the Espinhago Range in central Minas Gerais, growing at altitudes from 1000 to 1400 m.
It is presently known from two municipalities Flowers were collected in October and January and unripe
fruits in November. Curiously, the flowers present a strong scent of cooked beans (R. Mota and P. Viana,
pers. comm.).
Conservation. — Plinia espinhacensis is known for two localities distant about 300 km from each other,
but there is neither information relative to the condition of its habitat nor about the occurrence of the species
along this range. So, this species can be considered as data deficient (DD), since IUCN (2001) suggests this
tion based on its distribution and/or population status.”
Affinities. — This species is close to Plinia complanata M.L.Kawasaki & B.Holst, described for the Brazil-
ian atlantic coastal forest, from which it is kept apart by the characters in the following couplet:
1 . Branches evidently flattened; petioles to 1 0 mm; coastal atlantic forests of the state of Sao Paulo — Plinia complanata
1. Branches cylindrical; petioles to 6 mm; montane gallery forests from the Espinhago Range at the state of
Minas Gerai:
Etymobgy. — The epithet refers to the cent]
Gerais, the Serra do Espinhago (Rapini e
2001), t
Concei?ao do Mato De
ACKNOWLEDGMENTS
To the staff of the herbaria cited, for their constant good will in sending duplicates and helping in many other
respects; to the many collectors of the exsiccates studied here; to Andre Amorim (CEPEC) for his fortunate
suggestion for naming Myrcia pendula and to Luiz Antonio Cestaro (UFRN), for his kind help and friend-
ship during my stay in Natal looking for Eugenia azeda; to Rafaela Forzza and Erika von Sohsten Medeiros,
from RB, for their valuable help in the preparation of the images. Although not participating directly in this
Paper, I am very indebted to Eve Lucas (K), Fiorella Mazine (ESA), and Marcelo Souza (R), for sharing their
knowledge of Myrtaceae and their opinions with me. I am also indebted to Barney Lipscomb (BRIT) and Leslie
Landrum (ASU), who improved this paper with their careful revisions and kindly correcting my English.
Agu'a* AR, A.G. Chiarello, S.L. Mendes, and E.N. de Matos. 2005. Os corredores central e da Serra do Mar na mata
atlantica brasileira. In Galindo-Leal, C. & I.G. Camara, eds. Mata atl§ntica, biodiversidade, ameagas e perspectivas.
Belo Horizonte, Fundagao SOS Mata Altantica / Conservagao Internacional. Pp. 1 1 9-1 32.
B^-O-C 1857-1859. Myrtaceae. In Martius, K.F.P. von (org.) Flora Brasiliensis 14(1): 1-656.
^ B„ L.R. Landrum, and F. Gr.fo. 2003. Myrtaceae. In Berry, P„ K. Yatskievych & B. Holst, ed. Flora of the Venezu-
elan Guayana 7:1-99.
,UCN (International Union for the Conservation of Nature). 2001 . IUCN red list categories and criteria: Version
31 • IUCN, Species Survival Commission, Gland, Switzerland, 32p. (on-line version at http7/www.iucnredlist.
org/rnfo/ categories_criteria2001 .html.)
Kawasak., M.L. and B. Holst. 1994. New species and a new combination in Myrtaceae from northeastern South
America. Brittonia 46:1 37-143.
Kawasaki, M.L and B. Holst 2002 Two new species of Plinia (Myrtaceae) from coastal forests of Brazil. Brittonia
54:94-98.
REVISION OF LOBELIA SECT. SPEIREMA (CAMPANULACEAE: LOBELIOIDEAE)
Thomas G. Lammers
Department of Biology and Microbiology
University of Wisconsin Oshkosh
Oshkosh, Wisconsin 54901, U.S.A.
ABSTRACT
e L.Jangiana, L. longipedicellata, yL.m
INTRODUCTION
The 415 species assigned to Lobelia L. (Lammers 2007a, 2010) make it the largest of the 29 genera in the
Campanulaceae subfamily Lobelioideae (Lammers 2007b). Its infrageneric classification was revised recently
(Lammers 2010), with recognition of 18 sections. Among those newly recognized was L. sect. Speirema, to
which were assigned five species of southeastern Asia: L. brevisepala, L.fangiana, L. longipedicellata, L. montana,
and L. reflexisepala. In the course of treating these species for the Flora of China, problems were encountered
with their circumscription and characterization. As a result, this taxonomic revision was undertaken, the
fourth in a series of such accounts for sections of Lobelia (Lammers 2000, 2004, 2007c).
TAXONOMIC HISTORY
The type of Lobelia sect. Speirema is L. montana, which was published in 1826 by Carl Ludwig Blume, direc-
tor of the Buitenzorg Botanic Garden at what is today Kota Bogor, Indonesia. The species’ description was
drawn up from specimens collected in the mountains of Java by the garden’s founder, Caspar Reinwardt.
Blume’s diagnosis was woefully brief, describing only the branched procumbent stems, serrate leaves, and
solitary axillary flowers; no mention was made of the size of the plant, details of the flowers, or fruit type.
As a result of this brevity, the species remained poorly known for some time. It was not mentioned at all
by Don (1834) or Presl (1836), while Candolle (1839) was obliged to list it among “species minus notae aut
generis dubii.” Generic placement remained contentious and the species would peregrinate among four other
genera before finally coming home to Lobelia.
In 1844, Justus Hasskarl, the assistant curator of Buitenzorg, reassigned Lobelia montana to the genus
Pratia Gaudich. Though he did so without comment, one assumes that he had observed (either in nature
Pratia and Lobelia.
Within Pratia, Hasskarl specifically assigned this species to an unranked taxon, Bernonia Endl. (Endli-
cher 1838; Brizicky 1969), which comprised Asian members of the genus with a bilabiate corolla. Meisner
(1839) segregated this taxon from Pratia and accorded it generic rank as Bernonia (Endl.) Meisn.; however,
no species-rank combinations were ever effected under this name. Candolle (1839) felt it inappropriate for
> Bot. Res. Inst. Texas 4< 1 1: 1 59 - 168. 2010
Journal of the Botanical Research Institute of Texas 4(1)
two genera to commemorate a single person (the honoree of both Pratia and Bernonia was a French naval
officer named Prat-Bernon), so he substituted the name Piddingtonia A. DC. for the latter. As Candolle was
unsure of the identity and generic placement of Lobelia montana (see above), it remained for Friedrich Miquel
in 1857 to transfer that species to Piddingtonia. At the same time, he described a related species from Java,
P. patens, and two years later, Hasskarl added another, P. cyanocarpa. The name Piddingtonia is illegitimate
under Art. 52.1 (McNeill et al. 2006), and subsequent authors treated it as a synonym of Pratia (Bentham
1876; Baillon 1885; Hemsley 1886; Schonland 1889; Wimmer 1943).
About this same time, Joseph Dalton Hooker and Thomas Thomson were “obliged to found a new genus
[that they named Speirema ] upon a remarkable and hand:
they took to be conspecific with (“nullo modo differt”) Lobelia n
articulate exactly why this species a
was largely a matter of habit. While I
species then assigned to Piddingtonia
ana of distant Java. The authors did not
:commodated in Pratia or Piddingtonia, but one assumes it
ir plant was a tall herb with divaricate branches a foot long, all other
1 Pratia (cf. Hooker 1844, 1852) were much smaller plants with pros-
ognition of Speirema did not extend beyond the initial publication,
ler Pratia , commenting that it differed solely by its taller habit and
i demoted it to sectional rank, a classification adopted by Hemsley
:cumbent stems. However
Bentham (1876) discussed it briefly u
larger flowers. Baillon (1885) agreed a
(1886) and Schdnland (1889).
In his monograph, Wimmer (1943, 1953, 1968) divided Pratia into two sections: P. sect. Pratia (includ-
ing Piddingtonia) for the small usually prostrate or decumbent species, and P. sect. Colensoa (Hook, f.) Bail!
(including Speirema) for the tall usually ascending or erect species. The latter section included not only the
types of Speirema and Colensoa Hook. f. [P. physaloides (A. Cunn.) Hemsl. of New Zealand], but an additional
ten baccate species of similar habit from southeastern Asia and the Neotropics.
Because the fleshy fruit of Pratia was the sole feature that consistently distinguished the genus from
Lobelia, a given species of Pratia often had more in common with certain species of Lobela than it did with
congeners. Such patterns often are an indication that the one genus is polyphyletic, stemming from disparate
elements of the other. Consequently, taxonomists in the mid-Twentieth Century (summarized by Lammers
1993) began to question the wisdom of recognizing Pratia. This movement began with regional authors
transferring their species of Pratia to Lobelia ; Moeliono (1960) did so for the Asian species, including L
montana. Eventually, the two genera were fully merged by Murata (1995) and Lammers (1998). Wimmer’s
Pratia sect. Pratia became Lobelia sect. Pratia (Gaudich.) J. Murata [assigned to L. subg. Mezleria (C. PrtsD
E. Wimm.], while P. sect. Colensoa became L. sect. Colensoa (Hook, f.) J. Murata [assigned to L. subg. Tupa
(G. Don) E. Wimm.]. Baccate fruit was no longer a distinguis
capsular species.
also included
In the most recent revision of Lobelia (Lammers 2010), L. sect. Colensoa was pared down to its type, L
physaloides. Most of the other species placed there by Wimmer and Murata were reassigned to L. sect. Rhyn-
chopetalum (Fresen.) Benth. and L. sect. Tylomium (C. Presl) Benth., while L. bomeensis (Hemsl.) Moeliono was
segregated as L. sect. Plagiobotrys Lammers. Lobelia montana and four similar Asian species (L. brevisepda,
L. fangiana, L. longipedicellata, and L. rejlexisepala ; cf. Lammers 2007a) then became L. sect. Speirema. )
MATERIALS AND METHODS
Morphological data were gathered from approximately 150 specimens (including types) deposited in 15
herbaria (see Acknowledgments) and analyzed via traditional taxonomic methodology (Leenhouts 1968;
Qualls 1986; Vogel 1987; Maxted 1992; Watson 1997; Winston 1999). Once taxa had been discerned in
this fashion, they were compared to type specimens to determine the correct name under the International
Code of Botanical Nomenclature (McNeill et al. 2006). Decisions on rank for the taxa were made in light of the
definitions of species and subspecies I have employed previously (e.g., Lammers 1991, 2005, 2007a). |
RESULTS AND DISCUSSION
This study supports the continued recognition of only three of the five species originally assigned to Lo-
belia sect. Speirema (Lammers 2010): L.fangiana, L. longipedicellata, and L. montana. Lobelia brevisepala and
L. reflexisepala are here added to the synonymy of the last for the following reasons. First, the calyx lobe
features used to distinguish them from sympatric L. montana are not likely to function in the enforcement
of reproductive isolation. Second, these features are merely extremes in a continuum of variation; there are
no gaps in the range of calyx lobe lengths nor in their posture. In fact, although widely distributed species
often evince geographically correlated patterns of variation that make the recognition of subspecies possible,
such was not the case with L. montana (cf. Bamesky & Lammers 1997; Thompson & Lammers 1997). The
geographic origin of a specimen cannot be predicted on the basis of morphology.
This study also revealed that Lobelia deleiensis, treated as a synonym of L. montana (Haridasan &
Mukherjee 1988; Lammers 2007a), differs consistently from that species in size of the corolla and staminal
column and in color of the anther tube. Among Lobelioideae, differences of this sort often are correlated
with pollinator differences and thus may be indicative of reproductive isolation (Wood 1961; Young 1982;
Lammers & Freeman 1986; Lammers 1991, 1995, 2000, 2009; Sazima et al. 1994; Thompson & Lammers
1997; Muchhala 2003, 2006). For this reason, L. deleiensis is here accorded specific rank.
Lobelia sect. Speirema (Hook, f.)
Plants perennial (hemicryptophytes or chamaephytes; L. deleiensis a rhizomatous geophyte), 0.3 2 n
terrestrial. Stems herbaceous or suffruticose, 2-8 mm diam., branched (commonly from base and sometimes
above) or unbranched, erect, ascending, arching, or sprawling, glabrous (rarely scabrous or puberulent);
latex acrid, viscous, white. Leaves alternate, simple, exstipulate, dorsiventral, pinnately veined (dillemd),
petiolate (sessile in L.fangiana); lamina elliptic, oblong, lanceolate, oblanceolate, or ovate, chartaceous or
subcoriaceous (fleshy in L. longipedicellata), glabrous (rarely puberulent when young, or sparsely scabrous
on midrib below); margin callose-toothed, flat (revolute in L. longipedicellata ); apex caudate, cuspidate, or
acuminate; base rounded, obtuse, cuneate, or rarely attenuate; petiole (when present) winged, much shorter
than the blade. Flowers tetracyclic, perfect, zoophilous, chasmogamous with a specialized method of pro-
terandrous secondary pollen presentation, resupinate, epigynous, zygomorphic, pedicellate, solitar*
wils of little-reduced or unreduced leaves (in L. fangiana often supplemented by a termir
bracteate anauxotelic raceme); pedicels ascending, spreading, sigmoid, or incurved, Vw-,+ um6 « »«-
subtending leaf or bract, glabrous or puberulous, ebracteolate (in L. longipedicellata often with a pair of
linear bracteoles in the lower third). Calyx synsepalous, radially symmetric, adnate to the ovary, forming
an appendicular hypanthium Ve-tt as long as corolla, glabrous (sometimes the tube short-pubescent on
the nerves); base acute obtuse or rounded; lobes 5, valvate, lanceolate, triangular, or linear, V5-3 times
as long as the hypanthium, erect, spreading, recurved, or rarely reflexed, the margin entire (in L. fangiana
sometimes with 1-3 teeth per side), flat (revolute in L. longipedicellata), the apex acute or acuminate. Corolla
early-sympetalous, valvate, bilaterally symmetric, sub-bilabiate with 2 dorsal lobes and a trifid ventral lip,
dark violet, red-purple, bluish lilac, greenish, or cream-colored, the lip often contrastingly marked; tube
straight, cleft to its base on the dorsal side, about as long as broad to M times longer than broad, pubescent
Within and sometimes without; dorsal lobes linear, linear-spatulate, or linear-triangular, a little shorter than
*e tube to 4% times longer than the tube, apex acuminate; ventral lobes connate for Vs-X their length, free
Portion elliptic or lanceolate, apex caudate, acuminate, or acute. Stamens 5, antisepalous, connate distally,
i-flowered
164 Journal of the Botanical Research Institute of Texas4(1)
the axils of little-reduced or unreduced leaves; pedicels 24-55 mm long, ascending, spreading, sigmoid, or
incurved, ¥5-% as long as its subtending leaf, glabrous, ebracteolate. Hypanthium globose or campanulate
(rarely obconic), 3-6 x 3-5 mm, Vfe-Vio as long as corolla, glabrous; base obtuse or rounded (rarely acute).
Calyx lobes linear-triangular or linear, 2-14 x 0.5-1 .4 mm, 4/^-2% times as long as the hypanthium, spread-
ing, recurved, or rarely reflexed; margin entire; apex acuminate. Corolla dark violet to bluish lilac, the lip
often paler, or striped, margined, or blotched with white, 15-26 mm long, long-pubescent within; tube
6-13 x 2.5-4 mm, l%-4> A times longer than broad; dorsal lobes linear, 9-17 x 0.9-1.2 mm, equaling to
2% times longer than the tube, the apex acuminate; ventral lobes 7-15 x 2-5 mm, connate for Vs-3/? their
length, the free portion elliptic, apex caudate. Filament tube 9-12 mm long, glabrous; anther tube light gray,
2-2.8 mm diam., its dorsal surface sparsely short-pubescent at least toward apex; dorsal anthers 4.8-7 mm
long, %-% as long as the filament tube; ventral anthers 3.8-6 mm long, bearded at apex with tufts of white
hairs 1-2 mm long. Berry violet to black-purple, globose, 6-15 x 7-19 mm. Seeds amber-colored, broadly
ellipsoid, slightly compressed, 0.5-0.8 x 0.3-0.5 mm; testa shiny.
Distribution, habitat and phenology. — In three disjunct areas in southeastern Asia; the first overlaps the
geographic distribution of Lobelia deleiensis and L. longipedicellata. (1) Northeastern India (Arunachal Pradesh)
and Tibet to northeastern Burma. Thickets and openings in wet broadleaf and evergreen cloud forests at
1300-2135 m. Flowering July through November, fruiting August through April. (2) Southeastern Yunnan
and northern Vietnam. Thickets and openings in wet broadleaf and evergreen cloud forests at 1300-2135 m.;
Flowering July through November, fruiting August through April. (3) The Cameron Highlands of the Malay
Peninsula, and the Indonesian islands of Sumatra and Java. Wet montane rain and cloud forests, subalpine
scrub forest, and meadows at 1000-3000 m. Flowering and fruiting primarily during the monsoon seasons:
December through April and July through September. Map: Moeliono (1960), fig. 18 (solid line).
leones.— Moeliono (1960), fig. 17; Wimmer (1968), fig. 14b [as Pratia wardii] ; Academia Sinica (1979),
fig. 153 1-2 [as P. montana]; Lian (1979), fig. 2 [as P rejlexa]; Lian (1983), pg. 170 [as P. brevisepala], pg. 172
[as P. rejlexa]-, Lian (1985), fig. 266 [as P rejlexa]; Haridasan & Mukherjee (1988), fig. 13; Murata (1992) figs.
25, 35, 39-40.
li-leuntja. Sundanese: djanghe leuweung.
Vernacular names. — Chinese: shan zi chui cao. Indonesi;
djonghe rende.
Discussion— Although type material of Javanese Piddingtonia cyanocarpa and P patens was not available
for study, these names are included in synonymy with confidence, following Moeliono (1960). Both were
compared to P. montana by their authors, and in neither case have the alleged distinctions from that spe-
cies held up following examination of a far larger body of material. As such, they are a parallel case to the
synonymization of Chinese Lobelia brevisepala and L. rejlexisepala (see above).
The name Pratia montana f. variegata was bestowed on plants with distinctive bicolorous corolla pig-
L Judging from collectors’ notes on specimens, floral color is quite variable within Lobelia montana,
does
t perce
>rrelated with geography nor with other morphological features, it
!« lacked fruit and so the possibility of a relationship to L. montana
1, it was compared (as were L. deleiensis and L. longipedicellata) toll
(1968) who, in transferring the
ted it as a distinct
. Rhynchopetalum
erectiuscula H. Hara, a species of L.
species to Pratia, first noted that it
species by virtue of the peculiar CeigentOmliche-) ventral Up oflts corolla, though it is dfficult to detect
anything in his description of the novelty that differs from his earlier description of P. montana (WimroO
194». Haridasan and Mukherjee (1988) .mated the name as a synonym of P. montana. commenting tW
the distinguishing characters are too meagre.” The present study supported their conclusion; the type ol
L. wardii could be matched by any number of specimens confidently assigned to L. montana.
mac * **
y9370A&j
2 llllllll 1*44
REVISION OF LOBELIA SECT. PLAGIOBOTRYS
(CAMPANULACEAE: LOBELIOIDEAE)
Thomas G. Lammers
170
urnal of the Botanical Research Institute of Texas 4(1)
Wimmer (1943, 1953, 1968) expanded the circumscription of Pratia sect. Colensoa to include 11 ad-
ditional species from southeastern Asia and the Neotropics. All were large baccate plants like the first two
members, but bore solitary axillary flowers or terminal racemes instead of lateral racemes.
Moeliono (1960) transferred the Malesian species of Pratia (including P. borneensis) to Lobelia, arguing
that fruit type was an inadequate basis upon which to distinguish genera among Campanulaceae. However,
he did not concern himself with infrageneric classification and the precise relationships of the species within
its new home were unstated. Murata (1995) resolved this uncertainty by transferring the sectional epithet
from Pratia to Lobelia, creating L. sect. Colensoa (Hook, f.) J. Murata. He also expanded the circumscription ;
of the section by adding a dozen species with capsular fruit from Brazil, tropical Africa, and tropical Asia
to the baccate species included by Wimmer.
In the recent revision of Lobelia (Lammers 2010a), all of the species added to L. sect. Colensoa by Wim-
mer and Murata were removed and reassigned to L. sect. Rhynchopetalum (Fresen.) Benth., L. sect. Speirem
(Hook, f.) Lammers, and L. sect. Tylomium (C. Presl) Benth. Furthermore, L. borneensis was judged to not be
closely related to L. physaloides A. Cunn., and was likewise removed from L. sect. Colensoa; as it could not be
accommodated in any other section (see below for details), it was segregated as monotypic L. sect. Plagiobotrys.
Although the type of Lobelia borneensis was collected in Sarawak, the majority of specimens come from
farther east and north, in Sabah, particularly from the vicinity of Mt. Kinabalu. For many years, the species
was considered endemic to Borneo. It was not until well into the Twentieth Century that plants identified 5
as this species were discovered on the Indonesian islands of Sulawesi (Lam 1945), and Flores in the Lesser
Sunda Islands (van Steenis 1967).
The only other name referable to Lobelia sect. Plagiobotrys is Pratia borneensis var. grandiflora. Kew
botanist Otto Stapf based this on a specimen which George Darby Haviland collected on Mt. Kinabalu in
March-April 1892 (Stapf 1894; van Steenis-Kruseman 1950; Beaman et al. 2001). Though recognized by
Gibbs (1914), it was relegated to synonymy under the species by all subsequent authors (Wimmer 1953; J
Moeliono 1960; Beaman et al. 2001; Lammers 2007a).
MATERIALS AND METHODS
Morphological data were obtained from about 75 specimens in a dozen herbaria (see Acknowledgments)
and analyzed via traditional taxonomic methodology (Leenhouts 1968; Qualls 1986; Vogel 1987; Maxted
1992; Watson 1997; Winston 1999). As I have done with other insular taxa (e.g., Lammers 2007c, 2009),
I sought correlations between morphological variation and geographic distribution. Populations of Lobelia
borneensis are known from just four discrete areas on three islands: (1) the Malaysian state of Sarawak on the
island of Borneo at 425-1200 m, (2) the Malaysian state of Sabah on the same island at 975-3500 m; (3) the
Indonesian island of Sulawesi at 900-1800 m, and (4) the Indonesian island of Flores in the Lesser Sunda
Islands at 1600-1800 m. As in the prior studies, the initial hypothesis was that each area would harbor*
Therefore, when analyzing the data, I asked two questions. First, were all specimens from a given
area relatively homogeneous, with most characters evincing a continuous pattern of variation? If so, the
hypothesis would be supported; if instead, several characters consistently showed correlated gaps in that
patterns of variation within an area, the hypothesis would be refuted. Second, were the plants with a gi«n
correlated suite of morphological features restricted to a single area or island? If so, the hypothesis would
be supported; if not, the hypothesis would be refuted. Once taxa had been discerned in this fashion, they
were compared to the nomenclatural type specimens to determine the correct name under the IntemOti^
Code of Botanical Nomenclature (McNeill et al. 2006). Decisions on rank for the taxa were made in light of
the definitions of species and subspecies I have employed previously (e.g., Lammers 1991, 2005, 2007a).
RESULTS AND DISCUSSION
noted above, all previous authors have allied Lobelia borneensis tob
Relationships to other species. — As i
physaloides, the type of L. sect. Colensoa. These are the only two species of Lobelia with lateral racemes, and they
further share a moderately robust habit, large serrate petiolate leaves, corolla lobes several times longer than
the tube, and baccate fruit. Nevertheless, I find it difficult to accept a close relationship between them.
The value of seed-coat structure in understanding relationships among Lobelioideae has been amply
demonstrated (McVaugh 1936, 1940a, 1940b; Murata 1992, 1995; Buss et al. 2001; Lammers 2010). Varia-
tion in testal structure correlates well with variation patterns of other morphological features, as well as
with phylogenies based on molecular data (Knox et al. 2006, 2008a, 2008b; Antonelli 2008; Givnish et al.
2009) . It is largely testal structure that fails to support a close relationship between Lobelia borneensis and
L. ph ysaloides.
In Lobelia physaloides, the testa is striate, conforming to Murata Type D. This is by far the most common
pattern in the genus, characterizing more than half its species. Its distribution in the more comprehensive
molecular phylogenies (Knox et al. 2006; Antonelli 2008) supports Murata’s (1992, 1995) hypothesis that it
is the ancestral condition within the subfamily, from which each of the other types evolved independently.
In contrast, the testa of L. borneensis is reticulate, conforming to Murata Type B. This pattern is far less
common, occurring elsewhere only in the 43 species of amphi-Pacific L. sect. Hypsela (C. Presl) Lammers
and in the sole species of L. sect. Jasionopsis Lammers, endemic to the West Cape region of South Africa.
On this basis alone, one would hypothesize that L. borneensis and L. physaloides are not each other’s closest
In addition to seed-coat structure, the two species differ in other morphological features (Lammers
2010) . In Lobelia borneensis, the stem is herbaceous to suffruticose, scrambling or straggling, and up to 3 m
long; that of L. physaloides is suffruticose at base and fleshy above, erect to ascending, and no more than 1
m tal1- Though both bear lateral racemes, those of L borneensis are elongate, bearing as many as 35 flowers
over the course of their development, while those of L. physaloides are subcorymbose, with rarely as many
as 20 flowers. The pedicels in the former are bibracteolate below the middle, while those of the latter have
several bracteoles along their length. The corolla of L. borneensis is 10-28 mm long with dorsal lobes only
2/3-y4 as long as the ventral and laterally reflexed; that of L. physaloides is 30-45 mm long with the dorsal
lobes equaling the ventral and erect. The tips of the anthers in L. borneensis are bearded with tufts of flat-
tened hairs, while those of L. physaloides are nude or nearly so. Finally, the seeds of the former are cuboidal
or quadrate and thus quite angular, while those of the latter are ovoid and terete. These numerous differ-
ences fail to support a close relationship between L. borneensis and L. physaloides, and support removal of
the former from L. sect. Colensoa.
As noted above, some of the other baccate Asian species formerly assigned to Pratia sect. Colensoa have
been removed to Lobelia sect. Speirema (Lammers 2010a, 2010b). These plants do resemble L. sect. Plagiobotrys
in habit and in size and general structure of the flowers, and in an earlier draft of the revised classification, it
was planned to assign L. borneensis there. However, these species differ in typically bearing solitary axillary
bowers on ebracteolate pedicels, in having tufts of filiform hairs on the apex of the two ventral anthers, and
ln the striate-reticulate (Murata Type C) seed testa. In light of these differences, it seems best not to assign
L borneensis to L. sect. Speirema.
In %ht of their similarity in testal structure and fruit type, one might suggest that L. borneensis is
referable to L. sect. Hypsela. These plants do resemble L. borneensis in their blue, purple, magenta, pink, or
white corollas with the tube straight or nearly so and shorter than the lobes; many members have petiolate
leaves, bibracteolate pedicels, and baccate fruit. However, these plants are far smaller, less than 0.4 m tall,
°hen with stems prostrate or decumbent. Flowers are solitary in the axils of the upper leaves (rarely aggre-
Sated into a terminal raceme) and much smaller, with the corolla just 2-15 mm long. The corolla lobes are
“‘onomorphic or rarely somewhat dimorphic (vs. dimorphic) and the ventral anthers typically bear just a
Smgle lon8 bristle each at apex. Though the seeds of both are reticulate, those in L. sect. Hypsela are ovoid
t0 °blong and terete or rarely slightly compressed. For these reasons, L. borneensis was not assigned to L.
SCCt‘ hypsela, but instead segregated as a monotypic section.
172
Molecular data (Knoxet al. 2006, 2008a; Antonelli 2008) indicate that Lobelia physaloides is derived from
the endemic Australian taxon L. sect. Holopogon Benth.; like L. sect. Hypsela, this section consists of small
often decumbent plants, rarely more than 0.4 m tall. One might hypothesize that in a parallel fashion, the
larger L. bomeensis evolved from smaller species referable to L. sect. Hypsela. Though L. borneensis has been
sequenced (Givnish et al. 2009), it has yet to be included in the same cladistic analysis with L. physaloides
or exemplars of L. sect. Hypsela. Doing so would test the hypothesis presented here.
Species and subspecies. — Careful analysis of the data supported the initial hypothesis: it is possible to
determine consistently on the basis of morphology alone whether a specimen identified as Lobelia borneensis
came from Sarawak, Sabah, Sulawesi, or Flores. Furthermore, in one of these areas (Sabah), two sets of
populations can be distinguished on this basis, one occurring between 975-2900 m elevation, the other at
3000-3500 m. As a result of this finding, the originally monotypic L. sect. Plagiobotrys is here divided into
five taxa: three allopatric on Sulwesi, Flores, and Sarawak, the other two parapatric in Sabah.
Most of the differences that serve to distinguish the five taxa involve floral characters, particularly
the size, color, and posture of the corolla and staminal column. Such differences often are correlated with
pollinator differences and thus may reflect mechanisms of reproductive isolation (Young 1982; Lammers &
Freeman 1986; Lammers 1991, 2000, 2009; Thompson & Lammers 1997). As such, they are particularly
The high-elevation plants on Sabah are the most distinctive of the five taxa, with flowers that are far
more darkly pigmented and markedly larger: the ranges of values recorded for quantitative aspects of the
hypanthium, corolla, and staminal column (see key below) show in almost all cases a definite gap between
this taxon and the remainder. Such differences suggest to me that there is little or no gene flow between
the two, despite their parapatry, i.e., they are reproductively isolated. As such, each merits specific rank.
Although the four remaining taxa can be distinguished morphologically, the differences are less pronounced,
with the ranges of quantitative floral features showing various degrees of overlap. This suggests to me that
they are isolated primarily by geography and thus best treated as conspecific subspecies (cf. Lammers 1991,
2004, 2005, 2007c). As a result, L. sect. Plagiobotrys as recognized here comprises two species, one of which
is divided into four allopatric subspecies.
As noted above, only two validly published names are referable to taxa in this section. The type of
Pratia borneensis is a representative of the Sarawak taxon; as such, the more widespread species takes the
name Lobelia bomeensis, and its Sarawak subspecies is denoted by an autonym. The type of P. borneensis var.
grandiflora represents the low-elevation Sabah taxon; the new combination L. bomeensis subsp. grandiflon
is here created for it. The species endemic to higher elevations in Sabah is here christened L. origenes, while
appropriate names are likewise bestowed on the Sulawesi and Flores subspecies of L. bomeensis.
TAXONOMY
Plants perennial (hemicryptophytes or chamaephytes), 0.7-3 m tall, terrestrial. Stems herbaceous to suf-
fruticose, 2-5 mm diam., branched, straggling or scrambling, moderately leafy, glabrous or sparsely short-
pubescent toward apex; latex acrid, viscous, white. Leaves alternate, simple, exstipulate, dorsiventiA
pinnately veined (dillenid), petiolate; lamina broadly to narrowly elliptic, narrowly oblong, lanceolate, or
ovate, chartaceous to subcoriaceous, glabrous adaxially, glabrous abaxially or short-pubescent on the veins;
margin callose-serrate, callose-serrulate, or callose-denticulate, flat; apex acuminate to caudate; base curte-
ate, obtuse, or rounded; petiole much shorter than the blade. Flowers tetracyclic, perfect and proterandrouS,
zygomorphic, epigynous, pedicellate and resupinate, 5-35 in a terminal pedunculate bracteate raceme that
appears lateral due to overtopping by a branch; axis elongate, short-pubescent; bracts linear, glabrous#
short-pubescent; pedicels ascending, spreading, or deflexed, %-3 times as long as its bract, bibracteok#
: 7-25 i
, or oblate, 2-5 x 2-5 1
1.5-3.5 i
1.5-4
1.2-1. 8 i
Journal of the Botanical Research Institute of Texas 4(1)
Distribution, habitat, and phenology. — Endemic to the eastern shoulder of Mt. Kinabalu on northern
Borneo, in the Malaysian state of Sabah. Evergreen broadleaf forest, 3000-3500 m; flowering and fruiting
July-August.
Etymology. — From the Greek opetyevT^, “mountain-born,” in reference to its hypothesized origin as a
high-elevation derivative of L. bomeensis.
Discussion.— These specimens were originally identified as Lobelia borneensis, and the type was cited
under that name by Beaman et al. (2001). That species also occurs on Kinabalu, but only at elevations of
975-2900 m; it would seem to be the progenitor of L. origenes. The pronounced gap in corolla size and color
and in filament column length, in addition to the elevational separation, suggest that the two are reproduc*
lively isolated by pollinator differences.
ACKNOWLEDGMENTS
The curators and staffs of the following institutions are gratefully acknowledged for allowing study of
specimens in their care, either via loans or during personal visits: B, CAS, F, HAST, K, L, M, MO, MSC,
NY, OSH, and S. The figures were expertly delineated by Alex Sanford, a recent graduate of the University
of Wisconsin Oshkosh.
REFERENCES
Antonelli, A. 2008. Higher level phytogeny and evolutionary trends in Campanulaceae subfam. Lobelioideae
molecular signal overshadows morphology. Mol. Phylogen. Evol.46:1-18.
179
&AMAN, J.H, C. Anderson, and R.S. Beaman. 2001 .The plants of Mount Kinabalu 4. Dicotyledon families Acanthaceae
to Lythraceae. Natural History Publications (Borneo), Kota Kinabalu.
Buss, C.CJ.G. Lammers, and R.R. Wise. 2001 . Seed coat morphology and its systematic implications in Cyanea and
other genera of Lobelioideae (Campanulaceae). Amer. J. Bot. 88:1301-1308.
Gb& Ll 1914. A contribution to the flora and plant formations of Mount Kinabalu and the highlands of British
North Borneo. J. Linn. Soc., Bot. 42:1 -240 + pi. 1 -8.
Gnnish,TJ„ K.C. Miuam, A.R. Mast, T.B. Paterson, TJ. Theim, A.L. Hipp, J.M. Henss, J.F. Smith, K.R. Wood, and KJ. Sytsma.
2009. Origin, adaptive radiation and diversification of the Hawaiian lobeliads (Asterales: Campanulaceae).
Proc Roy. Soc B 276:407-41 6.
Hemsiey,W.B. 1886. Pratia borneensis, Hemsl. Hooker's Icon. PI. 16.pL- 1532 + 1 unnumbered p. text.
Knox, EB, P.B. Heenan, A.M. Muasya, and B.G. Murray. 2008a. Phylogenetic position and relationships of Lobelia
glaberrima (Lobeliaceae), a new alpine species from southern South Island (New Zealand). New Zealand J.
Bot 46:77-85.
Knox, EB., A.M. Muasya, and N. Muchhala. 2008b. The predominately South American clade of Lobeliaceae. Syst.
Bot 33:462-468.
Knox, EB., AM. Muasya, and P.B. Philupson. 2006. The Lobeliaceae originated in southern Africa. In S.A. Ghazanfar
and HJ. Beentje, eds. Taxonomy and ecology of African plants, their conservation and sustainable use. Royal
Botanic Gardens, Kew. Pp. 2 1 5-227.
Um, H. J. 1 945. Contributions to our knowledge of the flora of Celebes (Coll. C. Monod de Froideville) and of
some other Malaysian islands. Blumea 5354-599.
Lammers, T.G. 1991. Systematics of Clermontia (Campanulaceae: Lobelioideae). Syst. Bot. Monogr. 32:1-97.
Lammers, T.G. 1993. Chromosome numbers of Campanulaceae. III. Review and integration of data for subfamily
Lobelioideae. Amer. J. Bot. 80:660-675.
Lammers, T.G. 2000. Revision of Lobelia sect. Tupa (Campanulaceae: Lobelioideae). Sida 19:87-1 10.
Lammers, T.G. 2004. Revision of Lobelia sect. Homochilus (Campanulaceae: Lobelioideae). Sida 21: 591-623.
Lammers, T.G. 2005. Revision of Delissea (Campanulaceae: Lobelioideae). Syst. Bot. Monogr. 73:1-75.
Lammers, T.G. 2007a. World checklist and bibliography of Campanulaceae. Royal Botanic Gardens, Kew.
Lammers,T.G. 2007b. Campanulaceae. In: K. Kubitzki, ed.The families and genera of vascular plants, Vol. 8, Asteridae.
Springer-Verlag, Berlin. Pp. 26-56.
Lammers'T'G- 2007c. Revision of Lobelia sect. Galeatella (Campanulaceae: Lobelioideae). J. Bot. Res. Inst Texas
1:789-810.
^mmers-T.G. 2009. Revision of the endemic Hawaiian genus Trematolobelia (Campanulaceae: Lobelioideae).
Brittonia 61:126-143.
^ekJ.G. 2010a. Revision of the infrageneric classification of Lobelia L (Campanulaceae: Lobelioideae). Ann.
Missouri Bot. Gard. (in press).
Lammers' T-G. 2010b. Revision of Lobelia sect Speirema (Campanulaceae: Lobelioideae). J. Bot Res. Inst. Texas
4:161—170.
l>MM£rs.T.G. and CE. Freeman. 1 986. Ornithophily among the Hawaiian Lobelioideae (Campanulaceae): evidence
floral nectar sugar compositions. Amer. J. Bot. 73:1 61 3-1 61 9.
^HajTS' Rw- 1 968. A guide to the practice of herbarium taxonomy. Regnum Veg. 58:1-60.
n- 1 992. Towards defining a taxonomic revision methodology. Taxon 41 :653-660.
F.R. Barrie, H.M. Burdet, V. Demouun, D.L. Hawksworth, K. Marhold, D.H. Nicolson, J. Prado, P.C. Suva, J.E. Skog,
LWl£RSEMA' AND N J- Turund (eds.). 2006. International code of botanical nomenclature (Vienna Code) adopted
ty the Seventeenth International Botanical Congress, Vienna, Austria, July 2005. Koeltz, Konigstein.
:AUGH' R- 1936. Studies in the taxonomy and distribution of the eastern North American species of lobelia.
hodora 38:241-263, 276-298, 305-329, 346-362 + pi. 435-436.
7*' R- 1940a. A key to the North American species of Lobelia (sect Hemipogon). Amer. Midi. Naturalist
: G G. G. .
rehabilitaciOn Y LECTOTIPIFICACI6N del genero tessiera, su relaciOn
CON DIPHRAGMUS Y STAEL1A (RUBIACEAE: SPERM ACOCEAE):
UNA NUEVA COMBINAClON Y UN NUEVO SINONIMO
Roberto M. Salas
Institute) de Botdnica del Nordeste
IBONE-CONICET. Casilla de Correo 209
3400. Facultad de Ciencias Exactas,
NaturalesyAgrimensura (UNNE)
Corrientes, ARGENTINA
robertoymanuels@gmail.com
Elsa L. Cabral
Facultad de Ciencias Exactas, Naturalesy
Agrimensura (UNNE), Instituto de Botdnica
del Nordeste, IBONE-CONICET
Casilla de Correo 209, 3400
Corrientes, Argentina
ecabral@agr.unneedu.ar
ABSTRACT
introducciOn
En el transcurso del estudio de los generos americanos de la tribu Spermacoceae s. str. (Robbrecht 1988), se
analizaron tres taxones con estrecha afinidad historica y morfoldgica, a fin de establecer sus correspondientes
llmites taxondmicos: Stadia Cham. & Schltdl., Tessiera DC. y Diphragmus C. Presl.
Stadia fue descripto por Chamisso & Schlechtendal (1828) basado en una especie que vive en el sur de
Brasil; S. thymoides Cham. & Schltdl. Los autores definieron al gdnero esencialmente por las caracteristicas
del frut0: “Capsula membranacea, bilocularis, bivalvis, disepimento integro persitente, valvulis linea a dis-
^Pimento declivi basi scissis seorsum deciduis " Solo dos aflos despuds, Candolle (1830) agrega dos especies
182
nuevas de Brasil. El numero de especies de Staelia se incrementa para ese pais cuando Schumann (1888)
en las Rubibceas de Flora Brasiliensis, describe cuatro especies nuevas. En esa obra este autor agrupa a las
especies brasileflas en tres secciones: sec. Tessiera (DC.) K. Schum. ( =Tessiera ), con dos especies; sec. Staelia
con siete especies y una nueva seccibn, Anthospermopsis K. Schum. con una nueva especie. Esta clasificadda
se mantuvo hasta que Kirkbride (1997) eleva la sec. Anthospermopsis al rango genbrico: Anthospermopsis (K.
Schum.) J.H. Kirkbr.
En el transcurso de la revisibn de Staelia se comprobb que desde su descripcion hasta el presente se lo
vinculb con varios gbneros afines, los que fueron incluidos o segregados de Staelia de acuerdo al criterio de
Tessiera fue descripto por Candolle (1830) en base a dos especies: T. lanigera DC. (Brasil) y T. lithosper-
moides DC. (Mexico). El autor resaltb desde la descripcion original su afinidad con Staelia por la dehiscencia
de los frutos. A este gbnero lo definib por los caracteres de dehiscencia del fruto: “ Capsula bilocularis bivalvis,
dissepimento integro ovali deciduo, valvulis concavis calycis dentibus coronatis seorsum deciduis, verticaliter secus
dissepimentum seeds ” Con respecto al tipo de las dos especies, el material de Tessiera lanigera de Brasil, con-
sultado en el herbario de Viena (W), tiene datos de localidad y fecha coleccionado por Pohl, sin embargo d
de T. lithospermoides de Mexico, carece de datos de localidad precisa. El autor menciono tambibn en el pro-
tblogo de la especie que el material original pertenecia al herbario Haenke (actualmente PR). Con respecto
a la fecha de coleccibn, se puede estimar entre los aflos 1791-1792, durante la expedicibn por Ambrica de
Haenke en la nave del capit&n M alaspina. En ese periodo Haenke visitb y colectb en numerosos estados de
Mexico (Nayarit, Guerrero, Oaxaca), razon por la cual se hace dificil precisar la localidad tfpica de dicha
especie (Sternberg 1825; Slavik & Ceska 2006).
Don (1834), en su estudio de gbneros de Rubiaceae, reconocib a Tessiera y amplib su descripcibn “Stigma
2-lobed. Capsule 2-celled, 2-valved, with an entire oval deciduous dissepiment; valves concave, crowned by
the teeth of the calyx, falling asunder from being cut vertically along the dissepiments.” En la misma obra
tambibn senalb algunos caracteres para diferenciar las dos especies conocidas hasta ese momento, segun
este autor Tessiera lanigera presenta toda la planta con un indumento lanoso blanquecino mientras que T.
o glabrc
Diphragmus C. Presl fue descripto por Presl (1844), basado en D. scaberC. Presl. En la descripcibn el autor
indicb que el material estudiado fue colectado en Mbxico probablemente por Haenke. Se considera la posibk
validez del coleccionista, porque la coleccibn de Haenke fue la base de los estudios de Presl, publicadosen
gran parte en Reliquiae Haenkeanae (Presl 1825-1830; Skobdopolova &r Stbpanek 2002). Presl trabajb durante
mucho tiempo con la coleccibn de Haenke en el herbario PR, donde tambibn se encuentra depositado el
material tipo que fue consultado por Candolle (1830). El material tipo de Diphragmus scaber no fue local-
izado en los herbarios de Praga, PR ni en PRC, lo que hace suponer que Presl ignorb la existencia de Tessiera
de Candolle y uso para su descripcibn el tipo de Tessiera lithospermoides. Despues de analizar numerosas
colecciones identificadas con este nombre, se determinb que Diphragmus scaber y Tessiera lithospermoides son
conspecificos.
Autores como Endlicher (1838), Dietrich (1839) y Steudel (1841) aceptaron como validos a los gbneros
hasta aqui tratados, Staelia y Tessiera, reconocieron las especies descriptas hasta ese momento, sin embargo
desconocian a Diphragmus.
Con respecto a Diphragmus y Tessiera, fueron mantenidos hasta que Hooker (1873) los incluyb como
sinbnimos de Spermacoce L. El seftalo que las caracterlsticas del fruto de Diphragmus no son suficientes pat*
mantenerlo como gbnero aparte y que el indumento lanoso de Tessiera se puede observar en otras especies
de Spermacoce. Sin embargo no mencionb la consulta de los ejemplares tipo de cada taxbn para soportar
los cambios propuestos. Siete aiios despubs, Baillon (1880) propuso una nueva delimitacibn para la tribu
Spermacoceae, donde respetb la propuesta de Hooker, sobre la sinonimia de Spermacoce e incluyb como
sinbnimo tambibn a Staelia. El amplio concepto de Baillon con respecto a Spermacoce tambibn incluye como
sinbnimos M itracarpus Zucc. ex Schult. & Schult. f., Hypodematium A. Rich., Diodia L., Dasycephala DC 1
s and Cabral, Nueva especie y novedades nomendaturales de Tessiera
183
Octodon Thonn. A pesar de la relacidn seflalada por Hooker y Baillon, existen marcadas diferencias en los
frutos de Tessiera y Diphragmus con los de Spermacoce.
Schumann (1888), propuso una nueva delimitacibn para la tribu Spermacoceae, concepto que i\ mismo
defends en sucesivos trabajos (Schumann 1888, 1891, Delprete et al. 2005). En esta postura el contradijo
y discutid la propuesta de Hooker y la de Baillon, que incluyeron numerosos taxones como sindnimos de
Spermacoce. Schumann, no menciond a Diphragmus, revalidd a Staelia e incluyd Tessiera como una seccidn
de Staelia.
Con respecto a Tessiera, Kuntze (1902) siguid el criterio de Schumann (1888), lo reconocid en la cir-
cunscripcidn de Staelia, mientras que para Diphragmus, siguid la propuesta de Hooker y lo mantuvo bajo la
sinonimia de Spermacoce, sin realizar la combinacidn correspondiente. Standley (1931), fue el primero en
reladonar y transferir a Diphragmus como sindnimo de Staelia, realizd la nueva combinacidn con la especie
mexicana, Staelia scabra (C. Presl) Stand. ( =Diphragmus scaber). El seftald que en la mayoria de las especie;
del genero Staelia, la porcidn central del fruto es persistente y que la dehiscencia es transversal-oblicu.
cerca de la base. Ademas el comentd, que la planta mexicana presenta diferencias en la dehiscencia, qu<
en Diphragmus “no es transversal-oblicua como en Staelia.” A pesar de esto Standley realizd la nueva com
binacidn, porque observd que el tabique intercarpelar persiste visible entre las brdcteas, como ocurre ei
Staelia. El criterio de mantener a Diphragmus scaber bajo el nombre de Staelia scabra fue seguido en alguno
trabajos de especies mexicanas (Dominguez-Licona & Ochoterena 2006; Lorence 1990, 1999; Lott 1985
Nava et al.1998; Villasenor 2004).
Si bien histdrica y morfologicamente Tessiera y Diphragmus estuvieron m&s relacionados con Staelia, autore
contemporaneos como Delprete & Cortds-B (2006) o Dessein (2003) mantuvieron a estos gdneros bajo 1
sinonimia de Spermacoce y no de Staelia, coincidiendo con la propuesta original de Hooker (1873).
Durante el estudio de las Rubidceas de Mexico, Borhidi & Lozada (2006) realizaron un andlisis compara
tivo entre la especie de Staelia de Mexico y las de Brasil y encontraron diferencias en la morfologia del frutc
como lo habia observado Standley (1931). Dichos autores sostuvieron que las caracteristicas de flor, fruto
no mencionan la consulta del tipo de D. scaber. En ese trabajo tambien describieron una nueva especie D.
hexasepalus Borhidi & Lozada-Perez y en el mismo afto, Borhidi (2006) describid una nueva variedad de D. j
scoter var. minor Borhidi & Lozada-Perez.
Con respecto a Tessiera se estudiaron los ejemplares tipo de las dos especies de Candolle, T. lanigera
depositado en W y T. lithospermoides de PR y GOET. A partir de est
que habian observado diferencias en el indumento entre la especie brasilefta y la mexicai
ademas diferencias en caracteres florales, carpoldgicos, seminales
observaciones concluimos que ambas especies pertenecen a distintos generos.
En este trabajo coincidimos con la propuesta de Borhidi & Lozada de agrupar a las especies mexicanas en
“n genero separado de Staelia. Por todo lo expuesto se rehabilita a Tessiera y se lectotipifica el genero, basado
en el Art. 9.2 del ICBN. Se considera a Diphragmus como sindnimo nomenclatural de Tessiera, debido a que
fue descripto a partir del tipo de T. lithospermoides depositado en PR, por lo tanto Diphragmus es.un nombre
i'egitimo (ICBN Art. 14.4). Se redescribe a Tessiera y a las dos especies que lo integran. Se realiza una nueva
combinaciOn y se reduce a Diphragmus scaber var. minor a la sinonimia de T. lithospermoides. Se presentan
microfotografias de polen y semilla y se ilustra T. lithospermoides.
relaciOn intergenErica tessiera-staeua
Sl * considera el vinculo previo del genero Tessiera con Staelia, se pueden establece
diferencias;
^ es un genero sudamericano con 17 especies (Salas & Cabral 2010). Presenta el frut(
cencia septifraga, longitudinal y oblicua separado en tres piezas, dos apicales o valvas, cad
Persistente. La dehiscencia de su fruto comienza longitudinalmente desde el ipice hasta 1
issiguientes
litad y desde
Salas and Cabral, Nueva especie y novedade
185
s 4-5,5(-6) mm long., c
188
> Botanical Research Institute of Texas 4(1)
Fk. 3. Tessiera lithospermoides. A-F. Variation intraespedfica de la semilla. A. Cara dorsal de la semilla. B-C. Detalle de la exotesta. D. Cara dorsal *
la semilla. E-F. Detalle de la exotesta. Tessiera hexasepala. 6-1. Granos de polen. G. Vista polar de grano de polen. H. Vista ecuatorial. I. Detalle
exina del apocolpio. [A-C: J.C Soto Nunez 6850 (MEXU), D-F: £ Palmer 397 (K), G-H. M. Elorsa3613 (MEXU)].
Salas and Cabral, Nueva especie y novedades nomendaturales de Tessiera 191
hiscente y otro dehiscente y flores con corola muy pequefta, estambres y estilo inclusos (Bacigalupo & Cabral
1999a). Cabral & Bacigalupo (1999) despues de analizar el tipo de T. pubescens, prefieren considerar a este
nombre como sindnimo de Borreria remota (Lam.) Bacigalupo & E.L. Cabral por sus frutos con dehiscencia
septifraga con ambos mericarpos dehiscentes y estambres y estilo exertos.
h-ssu-ra lanigera DC. 4:573. 1830. Tiro BRASH Distrito Fiiim Brasilia, props- Mana da Souza.'J E. PM 5085 (holotipo:
Comentarios.— Se excluye a esta especie por presentar caracteres morfoldgicos y polinicos diferentes a lo
establecido para el genero Tessiera (ver relaciones intergendricas y claves de generos afines). Salas & Cabral (en
este volumen) ubican a este taxon como parte de un nuevo genero distinto de Tessiera y Staelia, denominado
Planaltina l=Planaltina lanigera (DC.) R.M. Salas & E.L. Cabral.]
AGRADECIMIENT OS
Agradecemos a los curadores de los Herbarios PR y GOET, Ota Sida y Jochen Heinrichs por el envio de
fotografias de los tipos consultados. A Laura Simdn por la confeccidn del habito y el pasado a tinta de los
detalles. Al Dr. Luciano Paganucci de Queirdz por la lectura critica del manuscrito y sugerencias nomencla-
turales. A los Drs. Massimiliano Dematteis y Elnatan B. Souza por las valiosas sugerencias. A los revisores,
quienes enriquecieron notablemente la version final del manuscrito. A todas las personas que colaboraron
en la elaboraci6n del manuscrito.
REFERENCIAS
Baogalupo, N.M. & E.L Cabral 1 999b. Sobre la identidad de dos especies de Lamarck, Spermacoce leavis y Sper-
macoce remota (Rubiaceae, Spermacoceae). Darwiniana 37:333-334.
Bailon, H.E 1880. Spermacoce. Hist. PI. (Baillon). 7:262-266.
Borhidi, AL& LP. Lozada. 2006. Estudios sobre Rubiaceae Mexicanas VIII. Diphragmus C. Presl (Spermacoceae) un
genero endemico olvidado de la flora de Mexico. Acta Bot. Hung. 48:43-45.
Borhidi AL 2006. Diphragmus. In: A.L. Borhidi, ed. Rubiaceas de Mexico. Akademiai Kiado, Budapest. Pp. 188-190.
Bremekamp, C.E.B. 1 966. Borreria leavis. In: A. Pulle, ed. FI. Suriname 4(1 ):289-290. Leiden: EJ. Brill.
Cabral E.L & N.M. Bacigalupo. 1999. Estudio de las especies americanas de Borreria Series Leavis (Rubiaceae,
Spermacoceae). Darwiniana 37:259-277.
Canoo^-A.R 1830. Staelia, Tessiera. Prodromus 4:573-574. 1830.Treuttel & Wurtz, Paris.
Chamisso, LA. and D.F.L Schlechiendal 1828. De plantis in expeditione speculatoria RomanzofFiana observatis.
Linnaea 3(4)338-366.
Delpretl pG. and R. Cortes-B. 2006. A synopsis of the Rubiaceae of the states of Mato Grosso and Mato Grosso do
Sul, Brazil, with a key to genera, and a preliminary species list. Rev. Biol. Neotrop. 3:13-96.
DtLpRfTE' pG.,T.M. Schuster, and P. Hiepko. 2005. An annotated translation of Karl Schumann's (1888): 'Ober einige
verkannte oder wenig gekannte Geshlechter der Rubiaceen Sudamerikas” ['About some misunderstood or
tale known taxa of South American Rubiaceae! with notes on the Rubiaceae type specimens kept at the
Berlin Herbarium. Bot. Jahrb. Syst. 126: 3-69.
k*®* s- 20°3. Systematic studies in the Spermacoceae (Rubiaceae), Doctoral Thesis. Katholieke Universiteit
Leuven. Leuven, Belgium.
?*** DNF- 1839. Staelia, Tessiera. Syn. PI. 1 :481 , 483. B.F. Voigtii, Vimariae.
Dom*hjez-Licona, E. and H. Ochoterena. 2006. The family Rubiaceae at the Chamela Biological Station (IBUNAM),
Jalisco, Mexico. Third International Rubiaceae Conference - Programme & Abstract. Scripta Bot. Belg. 40:31 .
GRLS- 1 834- CXCVt-Sfoe/io, CXVIl-Tess/era. Gen. Hist. 3:632. J.G. Rivington, London.
S.L 1838. Spermacoceae. Gen. PI. 6-7525-530. Fr. Beck, Wien.
7®*^ J-D- 1873. LXXXIV. Spermacoceae. In: Bentham, G.& J. D. Hooker, ed. Genera plantarum 2:142-148. Lovell
London.
PLANALTINA NUEVO GfiNERO DE LA TRIBU SPERMACOCEAE (RUBIACEAE),
ENDEMICO DEL PLANALTO CENTRAL DE BRASIL Y UNA
NUEVA ESPECIE DEL ESTADO DE GOlAS, BRASIL
Roberto M. Salas
Institute de Botdnica del Nordeste
IBONE-CONICET
Casilla de Correo 209, 3400
Facultadde Ciencias Exactas Naturales y
* Agrimensura (UNNE)
Corrientes, ARGENTINA
Naturales y Agrimensura (UNNE)
Instituto de Botdnica del Nordeste
IBONE-CONICET
Casilla de Correo 209
3400, Corrientes, ARGENTINA
ecabral@agr.unne.edu.ar
RESUMEN
ABSTRACT
INTRODUCClON
Hgfoero Stadia Cham. & Schltdl. fue descrito por Chamisso & Schlechtendal (1828), basado en S. thymoides
r®- &Schltdl., del sur de Brasil. Se lo diferencio de los demas generos de la tribu Spermacoceae s. sir.,
P ^Tobado, por la corola infundibuliforme y por el fruto capsular con dehiscencia oblicua. Can-
a y describio dos nuevas especies: S. galioides DC. y S. rejlexa DC. del Estado
obra el autor describio al genero Tessiera DC., basado en T. lanigera DC. y T.
d°k(l830)rec(
‘k Goias, Brasil Fn lo „
-on. m id nuaiua uora ei auior aesci
^pcrmoides DC., de Brasil y Mexico, respectivamente.
y Te^era°reS Endlicher (1838)’ Dietrich (1839) Y Steudel (1841) acePta
essiera ^aj° la sinonimia de Spermacoce L., por considerar que el indume
status gen£rico de Stadia
p, Hooker (1873) colocb
noso, caracteristico de la
197
^ r romparativo {fotos MEB) de los g^neros staelia- y planaltina- A-F- staelia- A- Valvas del ^ B- Sept0 intercarpelar y base de
H sZ a Intema de valva- D- Vista ventral de semilla. E. Vista polar de grano de polen. F. Detalle de la exina. G-N. Tessiera. G. Valvas del fruto.
'meJCarpelar- '• wsta intema del septo intercarpelar. J. Vista ventral de semilla. K. Vista dorso-apical de semilla. L. Detalle de la cubierta
grano de polen. N. Detalle de mesocolpio. 0-Z. Planaltina. 0. Valvas del fruto. P. Septo intercarpelar. 0. Vista intema^dd
Sir"3' V' Wsta dorso'aPica| de semilla. W. Detalle de la cubierta seminal. X. Vista ventral de semilla. Y. Vista polar de grano de polen. Z.
E-Letal. 168 (CTES)]. G-N. Tessiera lithospermoides. [G-l, X, L Painter, £ 397 (IQ; J, M, N. Soto
’'"'^mynddiana [Holotipo, Hatschbadi, G. &J.M. Silva 60266 (CTES)].
198
» (Fig. 2, Y-Z).
200
F* 3’ Planaltina BP**” Watschbadt, 6. &JM Silva 60036 (CTES)]. A. Habito. B. Vaina estipular. C. Flor. D. Fruto dehiscente. L Septo in
6. Vista interna de valvas del fruto. H. Cara ventral de la semilla. I. Cara dorsal de semilla. J. Corte longitudinal de semilla.
SalasandCabral, Revision c
203
ANEW VARIETAL COMBINATION, TYPIFICATION, AND NOMENCLATURAL
COMMENTS IN THE NYCTAGINACEAE FOR THE INTERMOUNTAIN FLORA
Richard Spellenberg
Biology Department, MSC3AF
New Mexico State University
Las Cruces, New Mexico 88003-8001, U.S.A
ABSTRACT
RESUMEN
The final volume, vol. 2A, of the Intermountain Flora now largely in preparation by Noel and Pat Holmgren
of the New York Botanical Garden will have a treatment by me for the Nyctaginaceae for the region. This
area encompasses all of Utah, most of Nevada, and parts of some of the adjacent states. In the course of pre-
paring that work, I have here selected neo- or lectotypes for some names that mostly apply to plants within
the Intermountain region, have provided one new nomenculatural combination, and included some notes
on other problems in typification, particularly noting inadvertent lectotypifications.
A NEW VARIETAL COMBINATION
Hputocalw carneus Greene pcdunculatus
St- Geoige, 2700 ft, 26 Apr 1894, M.E. Jones 5101 (lectotype: US!, inadvertently lectotypified in Galloway, Brittonia 27:335. 1976
("1975”). His date 1844 [for 1894] is a typographical error.).
% (1969) warned of using low weight regressive characters to define taxa. Such apparently has hap-
I*ned in Tripterocalyx, where the reduced, whitish green perianth (7-18 mm) has been largely used to
«nne Tripterocalyx micranthus (Torr.) Hook, sensu lato relative to other taxa in the genus, particularly T.
Mrneus (Perianth [18— ]25— 29 mm). The reduction of size and coloration in the perianth likely has been in
response to increased autogamy. Douglas (2008) showed that at least one population of the large-flowered T.
«"ieus, well to the south of the small-flowered forms, is self-compatible and probably capable of autogamy.
of r0Ughout *e range of large-flowered T. carneus are occasional small-flowered variants. Thus, reduction
0 Penanth length in response to vagaries in pollination may be easily accomplished.
As interpreted here, in the northern part of its range Tripterocalyx carneus enters into a more or less
reduction in perianth length. More northern populations have shorter flowers relative to southern
Potions. This occurs primarily in northeastern Arizona, southern Utah, and southwestern Colorado
r°ugh the phase that has been called T. wootonii Standi, as delimited by Galloway (2003). Even farther
nh’ the small flowered plants have been called T. micranthus (Galloway 2003). Early on M.E. Jones noted
208
Journal of the Botanical Research Institute of Texas 4(1)
these plants to be different, distinguishing them a
these northern plants are indistinguishable or distir
but are readily distinguishable from T. micranthus (
1 . Fruit body papery-spongy (can be easily scraped and br
tie T. micranthus var. pedunculatus M.E. Jones. In fruit
uished only with difficulty from T. carneus to the south,
2. Fruit body indurate (not easily damaged by scraping with needle), ribs prominent T. carneus var. pedunculatus
The two deceptively similar taxa have nearly separate geographic ranges. Tripterocalyx micranthus is a taxon
primarily of the western Great Plains, barely reaching northeastern and central-eastern Utah. Tripterocalyx
carneus var. pedunculatus is a taxon of the arid Intermountain Region, from sw. and wc. Utah to eastern Utah,
sw. Colorado, and ne. Arizona. The two are closely adjacent or sympatric in Uintah and Grand cos., Utah.
NEOTYPES AND LECTOTYPES
Abronia crux-maltae Kellogg, Proc. Calif. Acad. Sci. 2:71, fig. 16.
1 19°41'46.4"W, WGS-1
4, growing with Tetradymia canescens on sand flats, flowei
e: NEVADA. Ormsby (Co.: Eagle Valley,
rtrite, 19 May 2008, Arnold Tiehm 15570
In 1863 Kellogg described a novel Nyctaginaceae brought to him from the Carson Valley, Nevada, by A.
A. Veatch. He named the plant Abronia crux-maltae [now generally recognized as Tripterocalyx crux-maltae
(Kellogg) Standi.] because the shape of perianth limb was reminiscent of a Maltese cross. He provided
a thorough written description, contrasting the new species with A. cycloptera A. Gray (nom. nud.\ not-
ing the 4-merous perianth, but he included an erroneous illustration (fig. 16) that might be several other
Tripterocalyx or Abronia, but which is incorrect for T. crux-maltae. This is a fairly crude line drawing of a
stem with a flowering head bearing flowers with a 5-merous (or slightly greater) perianths blooming syn-
chronously, as in Abronia. None of the perianths is drawn as 4-lobed, a characteristic of T. crux-maltae. The
most distinctive feature of T. crux-maltae, the fruit, is not illustrated. He cited no specimen at the time, and
one has not been found at CAS or elsewhere. Galloway (1975) observed “no type cited,” and Tiehm (1996)
stated “collector not named [type not extant].” Curran (1885) cited the name but listed no type; from other
examples from this publication it is apparent that Kellogg might not have always made specimens or, if he
1906 San Francisco fire.
A neotype is here selected from a fine and ample collection made specifically for this purpose from the
general vicinity of the Carson Valley.
a Choisy vai
as, Sep 1883, Havard 61 (u-ctotytc, here de
Notes. — This specimen was seen by Watson with sketches of the flower on the sheet. Ii
“type ” The taxon is now recognized at the species level by Spellenbeig (2003) as proposed by Standley (1909, 1911, 1918).
Watson (1998) was discussing variation in Boerhavia spicata Choisy, within which he coined names for two
varieties, B. spicata var. palmeri S. Wats, and B. spicata var. torreyana. Contrasting plants from the Southwest,
Watson wrote (p. 70), “The form of Texas, New Mexico, and Arizona that has been referred to this speries
is usually more glandular than the var. Palmeri, the leaves thickish and scabrous, and the perianth al
a line long. It may be distinguished as var. (?) Torreyana." These taxa have been c
(2002). Watson cited a type for the var. palmeri, but he did not for var. torreyana.
specie
a type in later pub
reyana (S. 1
) Stand]
;sed in Spellenbeig
? Standley raised
. He also did n«
Oxybaphus glaber S Wats
. Naturalist 7:302. 1873. 1
209
When Watson was nearing the end of his tenure at US he described 0. glaber, citing only the type collection,
without date, and no specific holotype. Watson (1873), in the introduction to a series of new names, indicated
the collections of Mrs. E.P. Thompson were made in the second summer of 1871-1872.
COMMENTS ON INADVERTENT LECTOTYPES
Occasionally authors cited specimens in taxonomic treatments that resulted in inadvertent lectotypification
undercurrent international rules of nomenclature. Those that I have encountered while working with the
Abronia bakeri Greene (Pi. Baker. 3:32. 1901). Greene proposed the name Abronia bakeri for a sand verbena
from western Colorado. In the description he cited three syntypes, Baker 13 from Montrose, Baker 89 from
Deer Run, and Baker 92 from Grand Junction. He indicated no preference as to which might be a holotype
in the publication or on the specimens at NDG. Galloway (1975, p. 343) placed A. bakeri in synonymy in
A. elliptica A. Nelson (a disposition with which I concur), writing, “Type: Colorado. Grand Junction, 1,400
m, 11 Jun 1901, C. F. Baker 92 (MO 1746238!), inadvertently lectotypifying the name. Barbara Hellenthal
(NDG) kindly examined the Abronia bakeri syntypes, and noted that the labels are written out in Greene’s
hand; the labels of distributed specimens are in the hand of someone else. The specimens seen by Greene
also have habitat information, whereas the distributed specimens do not. The NGD 15722! Baker 92 has
added information, “borders of mesas,” and is an isolectotype.
Abronia robusta Standley (Contr. U.S. Natl. Herb. 12:324, plate 42. 1909). Beginning only a short time after
its description, A. robusta has been considered a synonym otA.fragrans Nutt, (e.g., Standley 1918). In 1909
Standley, in commenting on his new species, wrote, “The type material in the herbarium of the Missouri
Botanical garden consists of 4 sheets collected on sand hills near Monahans, Ward County, Texas, May 10,
1901, by H. Eggert.” Eggert’s collections were without collection number. Galloway (1975) cited MO 1746889
as the type, thereby lectotypifying the name (p. 345). James Solomon kindly checked the MO collection,
and noted that MO 1746889, 1746890, 1746891 were all seen by Standley, these representing the lectotype
and two isolectotypes. The fourth specimen alluded to by Standley is apparently a record-keeping error.
Two other specimens at MO were annotated by him; an Eggert collection from near Colorado, Mitchell Co.,
^ 5 June 1900 (MO 1746888), and an Eggert specimen (annotated by Standley as A. robusta ?) from
*ould be considered in typification of the name.
Allionia glandulifera A. Nelson (Bot. Gaz. 34: 364. 1902). Nelson (1902) described this species, gave a
general range, and cited no specimens. Reed (1969) consistently cited US specimens as “type” when they
"ere part of a type gathering, inadvertently lectotypifying many names that he placed in synonymy, as he
f for this name (page 173), the “Type from sunny canyon. Head of Woods Creek, Albany Co., Wyoming,
Ug- 10, 1900, A. Nelson 8048 (US-434056).” Isolectotypes are also at MO!, NY!, RM! The name is considered
a synonym of UirabiMs linearis (Pursh) Heimerl var. decipiens (Standi.) Welsh.
/ar. glabra Choisy (in DC., Prodr. 13(2):435. 1849). This name is supported by two
cum et ad S,Fernando (Berlandier! PL mexic. exsicc. n. 577 et 2236).” Later Stand-
i species level as Wedelia glabra (Choisy) Standi., transferring that name m 1930 to
Journal of the Botanical Research Institute of Texas 4(1)
Allionia glabra (Choisy) Standi., thereby creating a later homonym of Allionia glabra (S. Wats.) Kuntze. He
immediately corrected this error, coining the name A. choisyi Standi, in 1931. This name has been applied
to the taxon in the United States since that time.
In 1909 Standley inadvertently lectotypified A. incamata var. glabra. In this discussion of the species
he wrote (p. 332), “The description is based upon plants collected in the Mesilla Valley, New Mexico, which
seem well to match portion of the type collection preserved in the Bernhardi Herbarium.” Below he wrote
under specimens examined, “MEXICO: Environs de Mexico (City), Berlandier, type collection; ...” He does
not mention the San Fernando Collection. In the preface to his treatment (p. 304) he notes that he had the
privilege of examining specimens from a number of major herbaria, among them “Missouri Botanical Gar-
den, including the Engelmann and Bernhardi herbaria; ...” At MO is a specimen, #577, from the Bernhardi
herbarium which Standley saw, annotated by him as “cotype ,” which by his explanation and citation in 1909
becomes a lectotype. In 1918 (p. 201) he further supports this selection of this specimen by noting the type
locality of the species is “Near the City of Mexico, Mexico.”
Turner (1994) lectotypified the species on Mexico, Tamaulipas, San Fernando, w/o date, Berlandier 816
(pi. exs. 2236), selecting G-DC as the lectotype, which he apparently did not see, and GH as an isolectotype,
which he saw. Turner’s lectotypification appears unnecessary.
NEED FOR A NEOTYPE
Allionia incarnata L. (Syst. Nat., ed. 10. 2: 890. 1759) needs a type. No type was specified for this name.
Turner (1994) lectotypified Allionia incamata by citing for the holotype Loefling s.n., Sucre: near Cumana,
Venezuela (LINN). Through communication with C. Jarvis (BM), who is involved with the Linnean typifica-
tion project (http://www.nhm.ac.uk/research-curation/research/projects/linnaean-typification/), I learned
that no such specimen exists. Loefling’s material has apparently been lost. Thus, Allionia incarnata needs
a neotype. Turner’s map of the distributions of A. incarnata and A. choisyi Standi, in South America shows
choisyi in Haiti. Both species are common in North America, plants resembling A. choisyi are found in the
Carribean region, including the northern coast of South America, and both species are again common from
southern Peru and Bolivia southward.
North American herbaria that I have surveyed have insufficient material to select a representative
neotype that is part of a widespread collection. I have no information from South American herbaria. From
material that I have seen, all specimens from Venezuela (which are from the coastal region) would better fit
the North American concept of Allionia choisyi. These plants have relatively low-convex fruits with several
long, slender, and comparatively thin teeth. This also applies for all specimens I have seen from the Caribbean
islands. The concept of widespread and variable A. incarnata as understood in North America may depend
on a careful collecting effort around Cumana, Venezuela, in an effort that would attempt to determine the
variation occurring at the type locality prior to the selection of a neotype.
ACKNOWLEDGMENTS
I thank the following for help and advice with various nomenclatural and typification problems. Kanchi
Gandhi provided advice in numerous instances, as did John Strother and Noel Holmgren. James Solomon,
Ron Hartman, Barbara Hellenthal, and C. Jarvis checked for critical specimens at MO, RM, NDG, and L.
respectively. Pat Holmgren provided critical literature from NY. I have attempted to follow the capable guid-
ance of these individuals; nevertheless any errors remain mine.
REFERENCES
Curran, M.K. 1 885. List of the plants described in California, principally in the Proceedings of the California Acad-
emy of Sciences, by Albert Kellogg, Dr. H.H. Behr, and Mr. H.N. Bolander; with an attempt at their identification.
Bull. Calif. Acad. Sd. 1:1 28-1 51 .
212
Journal of the Botanical Research Institute of Texas 4(1)
BOOK REVIEW
Jonathan Silvertown. 2009. An Orchard Invisible: A Natural History of Seeds. (ISBN: 978-0-226-75773-5,
hbk.). The University of Chicago Press, 1427 East 60th Street, Chicago, Illinois 60637, U.S.A. (Orders:
www.press.uchicago.edu, 1-773-702-7000, 1-773-702-9756 fax). $25.00, 224 pp., 21 halftones,
x8
I. Bot Res. Inst. Texas 4(1): 212. 2
NEW NAMES AND COMBINATIONS IN THE FLORA OF COLORADO. XIII
W.A. Weber and R.C. Wittmann
University of Colorado Museum
UCB265, Boulder, Colorado 80309, U.SA
Ik following nomenclatural changes are proposed for the fourth edition of the two
Colorado Bora, Eastern and Western Slope (2010).
BORAGINACEAE
Oreocarya revealii WA. Weber & R.C. Wittir
Etymology. — Named in honor of James L. Reveal.
guide,
CARYOPHYLLACEAE
Gastrolychnis hitchguirei (Bocquet) WA. Weber & R.C. Wittmann, comb. nov. Silene hitchguirei Bocquet, Candol-
REFERENCES
) flora, Eastern Slope, A
i press). University Pre;
214
Journal of the Botanical Research Institute of Texas 4(1)
BOOK NOTICE
DoikiHsJ. Futuyma, H. Brai'i i v Shakir, and Daniel Simp.i ki oil (cn-i 20i>> Annual Knu « ol I ( oIo^n. I \()lu
tion, and Systematics. (ISSN 1543-592X; ISBN 978-0-8243-1440-8, hbk.). Annual Reviews, Inc., 4139
El Camino Way, PO. Box 10139, Palo Alto, CA 94303-0139, U.S.A. (Orders: www.AnnualReviews.oig,
science@annualreviews.org, 800-523-8635, 650-493-4400). $84.00 indiv., 731 pp., 7 5/8" x 9 3/8".
d Ecology of Species Range Limits— Jason P Sexton, Patrick J. h
.. Lewis, Jon Lloyd, Stephen Sitch, Edward T.A. Mitchard, Witt**
le Reproductive Biology of Sclerai
27. Effects of Natural Enemy Biodiversity on the Suppr
Julie A.Jedlicka, Sara C. Bothwell, Carlo R. Moreno
VALIDATION OF EXOCHORDEAE (ROSACEAE)
James L. Reveal
Cornell University
LH. Bailey Hortorium
Department of Plant Biology
41 2 Mann L ibrary
Ithaca, New York 14853, U.SA
jlr326@cornell.edu
ABSTRACT
The name Exochordeae (Rosaceae) was proposed by Schulze-Mentz in Syllabus der Pflanzenjamilien (1964)
without a description in Latin and thus it was not validly published. An earlier name existed for the three
genera now assigned to Exochordeae, (Exochorda Lindl., Lindleya Kunth, nom. cons., and Oemleria Rchb.) but
that name, Osmaronieae Rydb., was based on Osmaronia Greene, a superfluous substitute for Oemleria, and
thus the Rydberg name, like Greene’s generic name, is not legitimate.
Exochordeae Schulze-Mentz ex Reveal, trib. i
ACKNOWLEDGMENTS
|^am g^eful to Luc Brouillet of the University de Montreal for his comments and to Barney Lipscomb for
REFERENCES
Schuue-Mentz, G.K. 1964. Rosales. In: H. Melchior, ed. A. Engler's Syllabus der Pflanzenfamilien mit besonderer
ferurksichtigung der Nutzpflanzen nebst einer Obersicht fiber die Florenreiche und Florengebiete der Erde.
« Band. Angiospermem. Ubersicht uber die Florengebiete der Erde. Gebruder Borntraeger, Berlin, Germany.
Pp. 193-242.
Journal of the Botanical Research Institute ofTexas4(1)
BOOK REVIEW
Glenn Keator. 2009. California Plant Families: West of the Sierran Crest and Deserts. (ISBN: 978-0-
520-23709-4, hbk.). University of California Press, 2120 Berkeley Way, Berkeley, California 94704-1012,
U.S.A. (Orders: www.ucpress.edu or Califomia-Princeton Fulfillment Services, 1445 Lower Ferry Road,
Ewing, New Jersey 08618, U.S.A., orders@cpfsinc.com, 1-800-777-4726). $27.50, 224 pp., b/wline
drawings, 7" x 10".
the DOMINICAN AMBER FOSSIL LAS1AMBIX (FABACEAE: CAESALPINIOIDEAE?)
IS A LICANIA (CHRYSOBALANACEAE)
George 0. PoinarJr.
Kenton L. Chambers
218 Journal of the Botanical Research Institute of Texas4f1)
subfamily Caesalpinioideae. The latter was based on our effort to reconcile the simple pistil, presence of
a hypanthium, reduced number of stamens placed unilaterally, and drupaceous fruit with some common
tropical family.
Dominican amber has been dated as late early Miocene to early middle Miocene (15-20 Ma; Iturralde-
Vincent & MacPhee 1996; Graham 2003). It is interesting that within the correct family Chrysobalanaceae,
Licania dominicensis shows a mixture of putatively apomorphic traits, such as the stamens short, reduced
in number, and inserted on one side of the flower, with plesiomorphic ones like the petals present and the
hypanthium (receptacle of Prance 1972, 1989) regular and campanulate, with the ovary positioned basaUy.
This combination occurs in numerous modern species in the sections of Licania mentioned above, helping
to differentiate the genus from relatives such as Chrysobalanus, Couepia, and Hirtella (Prance 1972). W
This verified record of Licania from the Neogene of Hispaniola may be significant to studies of the
spread, by dispersal or vicariance, of tropical forests northward into the Caribbean region, Central America,
and North America (Graham 2003; Wendt 1993).
ACKNOWLEDGMENTS
We thank Ghillean T. Prance for his advice and suggestions concerning the similarity of the fossil to par-
ticular modem species.
REFERENCES
Graham, A. 2003. Historical phytogeography of the Greater Antilles. Brittonia 55:357-383.
Iturralde-Vincent, M.A. and R.D.L MacPhee. 1 996. Age and paleogeographic origin of Dominican amber. Science
273:1850-1852.
Nicolson, D.H. 1991. Chrysobalanaceae. In: Nicolson, D.H. ed. Flora of Dominica. Part 2: Dicotyledoneae. Smith-
sonian Inst. Press, Washington. Pp. 64-65.
Poinar, G.O., K.L Chambers, and A.E. Brown. 2008. Lasiambix dominicensis gen. and sp. nov., a eudicot flower in Do-
minican amber showing affinities with Fabaceae subfamily Caesalpinioideae. J. Bot. Inst. Texas 2:463-471.
Prance. G.T. 1972. Chrysobalanaceae. FI. Neotropica 9:1-409.
Prance, G.T. 1989. Chrysobalanaceae. FI. Neotropica 95:1-267.
Wendt, T. 1 993. Composition, floristic affinities, and origins of the canopy tree flora of the Mexican Atlantic slope
rain forests. In: Ramamoorthy, T.R, R. Bye, A. Lot, and J.A. Fa, eds. Biological diversity of Mexico: origins and
distribution. Oxford Univ. Press, New York. Pp. 595-680.
A NEW COMBINATION IN LAG0T1S (PLANTAGINACEAE)
David F. Murray
Reidar Elven
|| University of Alaska
Museum of the North
Fairbanks, Alaska 99775-6960, USA
dfmurray@alaska.edu
Natural History Museum
University of Oslo
Oslo, NORWAY
reidar.eiven@nhm.uio.no
Harvard University Herbaria
Cambridge, Massachusetts 02138-2021
gandhi@oeb.harvard.edu
The mainly Asian genus Lagotis is represented in North America by L. glauca Gaertn., in which there are three
taxa, the diploid subsp. glauca in the areas surrounding the northern Pacific coasts, the tetraploid subsp. minor
(Willd.) Hult6n (based on Gymnandra minor Willd.) in northeastern European Russia and northwestern Asia
east to northern Yakutia (the Sakha Republic), and a diploid taxon in northeastern Asia and northwestern
North America for which we publish the new combination subsp. lanceolata (Hulten) D.F. Murray & Elven
(based on Lagotis glauca var. lanceolata Hulten). Each subspecies is distinct in its main range.
In North America, subsp. glauca occurs along the coastline of the Aleutian Islands, Alaska Peninsula,
Kodiak Island, the Pribilof Islands, St. Matthew Island, and intermittently north along the mainland coast
to Cape Prince of Wales, Seward Peninsula.
Subspecies lanceolata in North Arne rica occurs farther north and eastward along the Arctic Coastal Plain,
Arctic Foothills, and Brooks Range southward to the mountain ranges of interior Alaska, arctic and interior
Yukon, and westernmost N.W.T. However, there is a narrow zone along the coast of the Bering Sea where
transitional forms occur north to the Seward Peninsula, and thus the rank of subspecies has been chosen.
InAsia, this subspecies occurs in Chukotka westward to the Kolyma River and southward to Karaginsky
kland, northern Kamtchatka, and the mountains northwest of the Okhotsk Sea.
The two North American subspecies can be distinguished in the following key.
wsa teaf tildes broadly obovate-oblanceolate, apex rounded or obtuse, sometimes subacute, margins
, crenate-dentate; filaments ca 1 .5 mm or shorter 1 a. Lagotis glauca subsp. glauca
• leaf blades narrowly oblanceolate, apex obtuse or subacute, sometimes acute, margins entire or
'ttantty dentate to serrate; filaments 2 mm or longer 1 b. Lagotis glauca subsp. lanceolata
^ names have been applied to the northern, amphi-Beringian plant in the last 50 years: Lagotis glauca
2 *P- minor or L. minor (e.g., Hultfn 1968a, 1968b; Petrovsky 1980; Ivanina 1991; Cody 1996) and subsp.
^ °r L stelleri fe g-. Gjaerevoll 1967; Porsild & Cody 1980). The application of L. minor and subsp. minor
J^roneous as this name belongs to the European and Asian tetraploid. The names based on Gymnandra
^which may apply to this plant are problematic for other reasons.
G-ymnandra stelleri has been published twice, but probably based on the same
; type. Sprengel
r NEOTYPIFICATION OF HECHTIA PODANTHA (BROMELIACEAE)
Adolfo Espejo-Serna1, Nancy Marti nez-Correa, and Ana Rosa Lopez-Ferrari
oUAMIZ
Departamento de Biologia
Divisidn de Oencias Biol6gicasyde la Salud
Universidad Autdnoma Metropolitana
Unidad Iztapalapa. Apda Postal 55-535
09340 MEXICO, D.F.
ABSTRACT
RESUMEN
In his contribution to the Bromeliaceae for the Monographiae Phanerogamarum of C. de Candolle, Mez (1896)
proposed seven new species of Hechtia using mainly material from Mexico.
In the aforementioned work, Mez described Hechtia podantha citing a single collection of material grown
intheSchonbrunn garden, which was collected and deposited in the herbarium of the Department of Botany
of the Naturhistorisches Museum Wien in Vienna (W) and was cited in the protologue as follows: “Patria
absque dubio Mexico. (Descript, ex specimine sicco hort. Schoenbrunn., in herb. Vindob. conserv.).” In 1935,
in his contribution to the work of Engler, Das Pflanzenreich, Mez mentioned that the living material had
disappeared from the Schonbrunn garden: A Warfruher im Schonbrunner Garten in Kultur, ist aber nicht mehr
wrhanden. However, he added to the examined material, one specimen from Tehuacan, Puebla ( Liebmann
s n - 1841) and two other collections from Pachuca, Hidalgo (Pringle 6932, Aug 1898, and Pringle 11188,
% 1902). Moreover, the original specimen at W was apparently destroyed during World War II, or at least
* is not actually found in the collections (W. Till, pers. comm.) (cf. also Merrill 1943, p. 490). Therefore,
dK type material of Hechtia podantha is missing. There is a photograph of the type deposited at the Field
Museum (negative 29960, F), in which original label appears with the following text: HRB. MUSEI PALAT.
VINDOB. /N°. /Hechtia/ C[ulta]. h[orto], V[indobonensis]. [11852. However, there are no known duplicates
of the original collection at the herbarium and the living material cultivated in Vienna has also been lost.
Currently, the interpretation of this species is based on the original description and in the circumscrip-
tion made by Mez in Engler Das Pflanzenreich (1935) and by Smith and Downs (1974) in their treatment of
the genus for Flora Neotropica.
During a taxonomic revision of the Hechtia podantha complex (Martinez-Correa 2008), we found that
tnis name was applied to different plant populations with similar morphological characteristics. The inap-
propriate application of the podantha epithet has been the result, at least in part, from the lack of type material
lh3t a,lows Ae questionable identification of the species, coupled with the lack of suitable specimens.
F<* the above cited reasons, we believe it is important to review the nomenclatural status of this spe-
0165 and designate a neotype to provide the correct application of the name. Of all the material reviewed by
Jj^e selected specimens collected by C.G. Pringle (6932) in the state of Hidalgo and deposited in several
/ 3na’ Whose features match those described for Hechtia podantha in the protologue, and also with the
tograph of the original material deposited at F (negative 29960).
Inst Tons ^1): 22! -223.2010
, 2. 1-8. 8 cm long, 2.3-6.1 c
3.5-5.8mm
1.2-3.1 mm
ACKNOWLEDGMENTS
We would like to thank Fernando Chiang and Walter Till for their critical revision of the manuscript. The
comments one anonymous reviewer are appreciated. Also to the curators and staff of the following herbaria
for providing access to their collections: B, BM, CHAP, CHAPA, CICY, ENCB, F, GH, IEB, K, LY, MEXU, MA,
JJICH, P, UAMIZ, UC, US, VT, W, WIS, WU, XAL, and Z. Some results of this work were derived from the
raduate Project of the second author. We thank CONACyT for the grant 202259 in support of this work.
% we would like to thank Dra. Alma Orozco Segovia responsible of the project “Mecanismos fisiologicos
mducidos por el priming natural, relacionados con la tolerancia de algunas especies de plantas a diferentes
for the gram to N. Maninez-Correa (11253).
REFERENCES
M^-CorrEA, N. 2008. Sistematica del complejo de especies de Hechtia podantha Mez (Pitcairnioideae,
omeliaceae). Dissertation, Universidad Autonoma Metropolitans Iztapalapa.
E D‘ 1943- Destruction of the Berlin Herbarium. Science 98(2553):490-491 .
1896. Bromeliaceae. ln:CDC. Monographie Phanerogamarum 9549-550.
1935‘ Omeliaceae. In: Engler, A. Pflanzenreich IV.32 (Heft 100, 3):32 1-480.
’ and RJ. Downs. 1 974. Pitcairnioideae (Bromeliaceae). Flora Neotropica Monograph 1 4(1 ):1 -658. Hafner
BOOK NOTICE
k Gadgil and Diana M. Liverman (eds). 2009. Annual Review of Environment and Resources. (ISSN
1543-5938; ISBN 978-0-8243-2334-9, hbk.). Annual Reviews, Inc., 4139 El Camino Way, RO. Box
10139, Palo Alto, CA 94303-0139, U.S.A. (Orders: www.AnnualReviews.org, science@annualreviews.
org, 1-800-523-8635, 650-493-4400). $78.00 indiv., 445 pp„ 7 5/8” x 9 3/8”.
Crop Yield Gaps: Their Importance, M
^merging Threats to Human Health fr
FOUR NOMENCLATURAL CHANGES IN VIOLA (VIOLACEAE)
Landon E. McKinney
2380 Crestbrook Drive, Apt. 8
Crescent Springs, Kentucky 41017, U.S. >
ldmckinney@fuse.net
RESUMEN
Viola epipsila Ledeb var repens luivz ex 1 stat nov B » ym Viola repens Turcz
Schwein. 1822. Viola epipsila subsp. repens (Turcz. ex Trautv. & C.A. Mey.) W Becker, Bot. Centralbl. Beiheft 34(2):406. 1917.
The acaulescent, rhizomatous, V epipsila var. epipsila occurs in Europe and northwest Asia. Viola epipsila var.
repens occurs in Europe, eastern Asia, Alaska, and western Canada. Variety repens differs from var. epipsila
by its smaller size, its glabrous, abaxial leaf surfaces (Baird 1942), by its distinctly acute leaves, and larger
Dowers (Sorsa 1968). Baird and Sorsa noted that bracteoles of V. epipsila occur above the middle of the pe-
duncle, a character used to differentiate it from V. palustris. The Violaceae treatment in the recent Flora of
China (Yousheng et al. 2008) treated V. repens as a synonym of V. epipsiloides A. Love & D. Ldve. However,
other authors do not recognize V. epipsiloides and consider V. repens as a subspecies of V. epipsila (Cody 20Q0;
Douglas 2000; Karlsson et al. 2008).
Viola pedatifida G. Don var. brittoniana (Pollaid) R.J. Little &L.E. McKinney, stat. nov. Bas.onym: Viola brittoniana
Pollard, Bot. Gaz. 26:332. 1898. Viola pedatifida G. Don. subsp. brittoniana (PoUard) L.E. McKinney, Sida, Bot. Misc. 7:22. 1992.
this violet was originally described as V. atlantica by N.L. Britton (1897), but the name was invalid due to
an earlier homonym. Britton considered this violet as simulating V. pedatifida apparently making him the
Drstto consider the affinity between var. brittoniana and V. pedatifida. Both are homophyllous species and
«« quite similar in many other characters leading L.E. McKinney (1992) to follow Britton’s lead and to fully
^blish the relationship between these two taxa. Viola pectinata is considered to be a sporadic form and
« Russell (1965) suggested, is likely due to genetic dimorphism. Additional studies are necessary before
considering V. pectinata as anything other than a dimorphic form.
V,ola praemorsa Douglas ex Lindl. var. flavovirens (Pollard) R.J. Little, stat. nov. Basionym: Viola flavovirensPol-
65:2579. 1987^ ^ ^ ^ '*** subsp -/!aV0VlrenS (PoUard) Canad J
Piperand Beattie (1914) and subsequent authors treated this taxon as synonymous with V. praemorsa ssp.
?r msu Baker (Fabijan et al. 1987). The size of the basal and cauline leaves are among the diagnostic
Ures seParating the three varieties of V. praemorsa. The basal and cauline leaves of var. flavovirens are
^lly *on8er (basal to 17 cm long; cauline to 14.8 cm long) and wider (basal to 6.7 cm wide; cauline to
to m wide) than vars. linguaefolia or praemorsa (basal to 8.5 cm long, to 3.7 cm wide; cauline to 8 cm long,
nCm wide)- Fabijan et al. (1987) reported that Baker and Clausen annotated herbarium sheets with
rcvis. rens as a subspecies or variety of praemorsa indicating that they recognized the taxon. However, no
011 °^’ts taxonomic status had been published prior to Fabijan (Fabijan et al. 1987).
DIVERSITY, NATURAL HISTORY, AND CONSERVATION
OF VANILLA (ORCHIDACEAE) IN AMAZONIAN WETLANDS
OF MADRE DE DIOS, PERU
Ethan Householder, John Janovec, Angel Balarezo Mozambite,
Javier Huinga Maceda, Jason Wells, and Renan Valega
Botanical Research Institute of Texas
500 E 4* St, Fort Worth, Texas 76102-4025, U.SA
jehouseholder@gmail.com
' Helena Maruenda Eric Christenson
DepartamentodeCiencias-Quimica 4503 21 stAve. West
A Pontificia Universidad Catdlica del Peru Bradenton, Florida 34209, U.SA
Av. Universitaria 1801, San Miguel, Lima 32, PERU
ABSTRACT
INTRODUCTION
VaniHa Plum. <
1 100 species of e
i pantropical genus of the Orchidaceae comprising more tha
Ptyicand terrestrial lianas. It is noteworthy among the genera of the Orchidaceae tribe Vamlleae as tne
species-rich, widespread genus (Stern & Judd 1999). Based on recent molecular studies the vamlioid
recognized as an ancient lineage that deserves to be treated as a subfamily of the Orchidaceae
7** et & 2003; Cameron 1999). Such attributes highlight Vanilla as key to understanding the orchid fam-
l y’ ar8uably the most species-rich group of plants on the planet, calling for further studies of the diversity
^ natural history of this genus.
Taxonomic investigations are scarce or incomplete and the genus is plagued by lack of collections, mis-
“'“ttfications of existing collections, poor understanding of species concepts, and a conflicting synonymy
228
Journal of the Botanical Research Institute of Texas 4(1)
(Chevailer 1946; Correll 1946). Two major revisions of the genus by Rolfe (1896) and Porteres (1954) are
largely out of date. Recent molecular studies (Bouetard et al. 2010; Soto-Arenas & Dressier 2010) have greatly
improved our understanding of the phylogenetic relationships between species; however, these are still tenta-
tive and incomplete. Soto-Arenas and Cribb (2010) have recently revised the infrageneric classification of the
genus, lamenting that our knowledge of Vanilla is far from adequate to produce a good, modern revision.
Exacerbating taxonomic problems, there is a deficit of natural history studies and the observations that
come from them, such that even basic information concerning dispersal mechanisms and pollinator relation-
ships remains enigmatic. Several characteristics of the group make detailed investigation, documentation,
and collection difficult. First, Vanilla species often occur in low-density populations that are hyper-dispersed
in local areas of their limited geographic ranges (Soto-Arenas et al. 2003). This has contributed to the lack
of herbarium collections and basic natural history data. Second, plants are usually encountered without
flowers as many species require pendant growth in bright light to initiate budding. Third, the succulent
nature of the plants renders them very slow to dry and botanical collectors often avoid them as logistically
problematic. Finally, Vanilla species often have delicate, ephemeral flowers that do not preserve well on
herbarium specimens.
From an economic perspective, cured Vanilla fruit provide the world’s major source of natural vanilla
fragrance and flavoring. In 1995, fruits from Vanilla planifolia ranked as the leading U.S. spice import on par
with black pepper (Buzzanell & Gray 1995). World production estimates in 2001 of approximately 2300
metric tons (Loeillet 2003) are generally restricted to developing tropical countries with a cheap labor force
(Koekoek 2005; Anonymous 2003). The crop is important to thousands of small-scale farmers worldwide,
with highest production coming from Madagascar and Mexico (Hermans and Hermans 1995). However,
the narrow genetic base of commercial Vanilla planifolia, due to strictly vegetative propagation, and its grow-
ing susceptibility to viral and fungal pathogens suggest that the industry has not fully explored the genetic
resources harbored by potentially disease-resistant wild species (Besse et al. 2004; Grisoni et al. 2004; Bory
et al. 2008).
Modern studies of the diversity and natural history of Vanilla are needed in the Amazon. Amazonian
species of Vanilla have been poorly collected and are little known in comparison with Central American
species (Soto-Arenas, pers. comm. 2007). During ongoing studies of the diversity, natural history, and
conservation of wetland plants and ecosystems in Madre de Dios, Peru, by the Botanical Research Institute
of Texas (BRIT), six Vanilla species have been encountered. The goal of this paper is to provide an updated
report of the diversity of Vanilla species in Madre de Dios wetlands, with comparisons to the historical
collection record and knowledge of the genus in the Amazon. Most species are newly recorded for Peru.
Morphological descriptions and a key are provided along with notes and observations to document current
knowledge of the distribution, habitat, and natural history of each species. We follow this with a discussion
MATERIALS AND METHODS
Study Site
All studies were carried out along the Madre de Dios River watershed in the Department of Madre de Dios,
located in southeastern Peru and corresponding to the headwaters of the southwestern Amazon (Fig-D
The region is widely recognized as a hotspot of biological and cultural diversity that is partially protected
by a series of large conservation areas, including Manu National Park, Bahuaja-Sonene National Park, the
Tambopata Reserved Zone, Los Amigos Conservation Area and the Amarakaeri Indigenous Reserve.
study region includes flat lowland habitat ranging in elevation from 300 to 350 m classified as humid tropi-
cal forest. Wetlands with hydromorphic soils, generally dominated by palms, are a patchy, but ubiquity
aspect of the landscape in topographical depressions and in the vicinity of seepages (Kahn 1987; Kahn
Mejia 1990). Along current river floodplains wetlands are a prominent habitat, becoming quite extend
Householder etal.. Diversity, natural history, and conservation of Vanilla
229
d- ^nnual average rainfall ranges between approximately 2,000 and 3,500 mm. Rainfall is unevenly
^ ted throughout the year, with greater than 80% falling between October and April. Average daily
nature ranges from 21 to 26°C. Southerly cold fronts from Patagonia, known as “friaje s,” are com-
r°m June through August, rapidly decreasing temperatures by 10°C or more in a matter of minutes;
230
Journal of the Botanical Research Institute of Texas 4(1)
sustained for a period of several days, temperatures as low as 10°C have been reported. Climate data since
2001 is made available by the Amazon Conservation Association through the weather module of the Atrium
Biodiversity Information System at BRIT (http://atrium.andesamazon.org).
Specimen and Data Collection
Between July 2005 and August 2009, approximately 70 km of line and trail-based transects were surveyed
in 28 separate wetlands throughout Madre de Dios. Transects varied in length according to wetland size,
ranging from 4 km in the largest wetlands, to 1 km in the smallest. All transects were oriented strategically
to cross a diversity of wetland vegetation formations. Accessible populations of the most common Vanilla
species were frequently monitored for phenological patterns. Opportunistic observations were made to
document the activity of bees visiting flowers and fruits. If possible, photos and specimens of pollinators
were collected for subsequent identification. Pollination rates for all species were estimated at the end of
the flowering season by counting the number of empty floral bracts and fruits. Descriptions of floral scents
were facilitated by enclosing the entire flower in a plastic film container for five minutes to intensify the fra-
grance. Plant specimens were collected in duplicate sets and deposited at the San Marcos Herbarium (USM)
in Lima, Peru, and the BRIT Herbarium. Fresh flowers and fruits from each collection were photographed
and preserved in 90% alcohol for descriptive studies.
RESULTS
Six species of Vanilla were encountered and documented in the wetlands of Madre de Dios. The following
key is based primarily on morphological characteristics that are most useful in the field. Several species are
notoriously variable in their morphology but proper identifications are not difficult even in sterile condition
using a combination of vegetative and ecological characteristics. Climbing individuals attached to a host
are generally much easier to key out and identify in the field than are scrambling vines. The key is followed
by morphological descriptions of each species with notes about nomenclature, distribution, habitat, and
;ter, short (< 5 m in length) vine or epiphyte, racemes terminal a
1 cuspidate apex, generally held horizontal to ground, always t<
5. Leaves sessile, entirely green, internode length < than leaf length, sepals yellow .
Plant a vine, occasionally reaching up to 10 m. Stems thin, brittle, ca. 0.4-0.7 cm in diameter, internodes
8-10 cm long, one to many-branched. Aerial roots short, thin, ca. 0.2 x 3 cm, unbranched. Terrestrial roots
emerging from lower nodes and entering superficially into substrate, branched underground. Leaves alter-
nate, produced from each node opposite to roots, distichous, thin and membranaceous; blade elliptic, ca
3.7-11 cm wide and 7.5-22 cm long, green, apex acute-attenuate, margins entire. Inflorescences axilla^
racemose, sessile, 2-20 flowered, fleshy, green; bracts triangular, ca. 1 cm long, green; pedicel ca. 0.3 c®
wide and 3 cm long (pedicel of auto-pollinating individuals may be longer because fruit development beg®5
232
on raceme, slightly fragrant (like freshly cut grass or trunk slash); buds green; sepals elliptic, undulate, ca.
1.7 x 4 cm, green, apex acute, margins keeled; petals elliptic, often strongly recurved, undulate or not, ca.
1.2 x 4 cm, green, apex acute; lip stiff, white, glabrous, forming triangular opening when fresh, attached to
column on dorsal side for 0.4 cm of length, when flattened broadly pentagonal in shape, distinctly three-
lobed; middle lobe 3 cm long, adaxial surface with distinctive yellowish channel formed from two broadly
raised longitudinal keels, abaxial surface with distinct longitudinal furrow; lateral lobes white, ca. 1.8 cm
long, stiff; callus absent; column 2.4 cm long, 0.4 cm wide, stiff, forks 0.4 cm from tip of stigma, upper fork
holds anther ca. 0.3 cm beyond stigma; rostellum absent; anther ca. 0.2 x 0.2 cm with red-orange margins;
stigma ca. 0.05 x 0.2 cm, with four nipple-like projections. Fruit dehiscent, cylindric, ca. 1 x 18 cm, green
and brittle when immature, splitting longitudinally along sutures, splitting valves turn dark brown, pliable,
leathery, revealing seeds within a somewhat oily mesocarp, with disagreeable fragrance when moist.
Notes. — This species has a complicated nomenclature, with no type specimen and several synonyms.
The oldest name is V. guianensis Splitg. and thus conserved. However, it may also be found as V. acuta Rolfe,
V. latisegmenta Ames, and V. surinamensis Rchb. f. (Soto-Arenas & Cribss 2010). Soto-Arenas and Cribb
(2010) place it in subgenus Vanilla, subsection M embranaceae, a group characterized by its thin leaves, con-
reproductive axes (Soto-Arenas 2003).
Habitat and Distribution.— This species is relatively common throughout the Amazon basin (Soto-Arenas,
pers. comm.). Although in our study region it occurs more abundantly in association with M. Jlexuosa wet-
lands, it is the only species not strictly restricted to wetlands. For example, we have documented individu-
als growing in managed banana plantations, suggesting that it is a fast-growing, colonizing species. In M.
Jlexousa wetlands with deep histosols and acidic waters it generally prefers areas with abundant, low-lying,
shrubby trees common to wetlands, such as Hex sp. (Aquifoliaceae), Tapirira guianensis (Anacardiaceae), and
Tabebuia insignis (Bignoniaceae). The leaf litter of these common, non-palm wetland trees may be important
Natural History. — Flowering phenology is somewhat sporadic; various individuals flowered in January,
April, May, and November during 2005-2006. The number of flowers per individual ranged from 5 to about
50, depending on plant size. Flowers often open in triplets and remain open for three to seven days as ob-
served for other membranaceous species (Soto-Arenas et al. 2003). Due to longer flower longevity and rapid
rates of anthesis a large floral display often accumulates on reproductive branches. The most gregariously
flowering branches are almost always those that are pendent, hanging from the main vegetative axis.
Vanilla guianensis is apparently self-pollinated at early anthesis as the stigma and anther grow to contact
one another. Pollination rates are approximately 78% (Table 1). No potential pollinators have been observed,
but Soto-Arenas (2003) suggests that the stiff flowers and relatively closed throat of V. guianensis may be
adapted to large, strong pollinators such as carpenter bees ( Xylocopa spp). However, no potential pollinators
red nez
the developing
al days. While
fibrous stems ■
Unlike other Vanilla species, the flower and column of V. guianensis rapidly absci
fruit. Once mature, dehiscing fruits have a rather unpleasant, putrid odor for one
such odors may be attractive to some animal dispersers, none, however, have been observed near fruits
r remain oily and pliable for several weeks and when dry become leathery. The thin
lily propagate vegetatively in artificial settings.
2. Vanilla riberoi Hoehne (Fig. 2b)
Plant a vine, generally loosely climbing on understory trees. Stems dark green, succulent, ca. 0.5-0.8 cm in
diameter, internodes 10-15 cm long. Aerial roots produced from each node, short, stout, white, unbranched.
Terrestrial roots emerging from lower nodes and entering superficially into substrate, branched and exten-
sive underground. Leaves alternate, produced from each node opposite to roots, distichous, held at 0 to
degrees below horizontal, succulent; blade narrowly lanceolate-linear, ca. 1.5-2 cm wide and 8.5-14
234
ca. 1 to 20-flowered; bracts triangular, ca. 1.5 cm long, green; pedicel green, persistent on maturing fruit.
Flowers resupinate, tender, fleshy, span ca. 15 cm wide, arranged spirally, fragrant or not, remaining at-
tached to ovary if successfully pollinated; buds green; dorsal sepal oblanceolate, reflexed when fully open,
ca. 1.5 x 9.4 cm, yellow; lateral sepals oblanceolate, reflexed, ca. 1.5 x 8.5 cm, yellow; petals oblanceolate,
reflexed, ca. 1.1 x 9 cm, yellow, with thickened midvein on dorsal surface; lip somewhat saccate when fresh,
fused with column for ca. 5 cm, uniformly yellow except for a characteristic pale, triangular discoloration
on ventral surface of opening, flabellate and almost triangular when flattened, margins slightly revolute,
pleated, unguiculate, the claw caniculate with dark yellow rows of trichomes running longitudinally between
lip margin and callus, ca. 2.2 cm long and 0.3 cm wide,; callus well-developed, ca. 0.8 x 0.6 cm, penicillate
with laciniate scales; column ca. 6.3 cm long, 0.4 cm wide, hirsute on ventral surface near rostellum gradu-
ally becoming glabrous; rostellum broad, thick, yellow. Fruit dehiscent, ca. 2.6 x 15 cm, fleshy, triangular
in cross-section, green and brittle when immature, splitting longitudinally along sutures into unequal valve
at maturity, split ends turn pliable, leathery and brown-black in color, revealing seeds embedded within an
extremely oily mesocarp and emitting strong vanillin fragrance.
Notes. — Vanilla pompona subsp. grandiflora is a member of the subgenus Xanata, section Xanata. New
molecular data suggest that the V. pompona of Mexico is nested within the variable V. grandiflora, but the
older name, V. pompona, is retained (Soto-Arenas pers.comm.). The species complex is quite variable and
several disjunctions are evident within its widespread distribution (Soto-Arenas & Dressier 2010). The
original species described by Schiede, now V. pompona Schiede subsp. pompona, is a narrow endemic on the
Mexico/Guatemala border and all records of its existence in South America are wrong.
Distribution and Habitat. — Vanilla pompona subsp. grandiflora is widely distributed from Honduras to
Bolivia (Soto Arenas pers. comm.). In Madre de Dios it is mostly restricted to open, flooded savannah habi-
tat within Mauritiaflexuosa wetlands, characterized by deep, permanently saturated histosols and minimal
water level fluctuations. While the vine is robust and resilient to frequent disturbances caused by falling
palm fronds, its roots are vulnerable to even short periods of inundation. Within its habitat it is generally
restricted to small hummocks formed by the accumulation of M.flexuosa frond mulch. It often forms dense
populations over extensive areas where conditions are favorable.
Natural History. — On a given inflorescence flowers open singly every two to three days. Flowers open
as early as 3:00 am and generally last 12-15 hours, wilting with the afternoon heat. Individuals tend to
flower more profusely immediately following small-scale (5 C° or less are sufficient) temperature reduc-
tions (i .e.Jriajes). During these cool weather spells, up to two flowers may open per inflorescence and wt
have observed displays of over 30 open flowers on a single individual. Flowering responses to temperature
seem to be facilitated by determinant racemes with buds almost fully developed prior to the initiation of the
flowering season. Several of the most fully developed, basally located buds may be poised to open during
unpredictable cool spells. The response to cool temperatures is general and immediately following /rfoje
the entire population may not flower for one to several days while apically located buds develop. The maxi-
mum number of flowers observed on a single individual throughout an entire season was 200. Flowering
phenology is bimodal, with two rather short flowering episodes of approximately 30 days each in April and
September/October. Significantly more individuals flower during the September/October season, probably
in response to the long dry season commencing in June/July.
The flowers produce a “sweet” fragrance, but this is sometimes barely perceptible and the potency varies
between individuals. Particularly fragrant flowers attracted two species of Euglossine bees, Euleama meriana
and Euglossa imperialis. Several individuals of both species may repeatedly circle a single fragrant flower for
several minutes, often demonstrating antagonistic behavior. The smaller, and generally more numerous Eu&
imperialis is the more aggressive species, however its small size probably precludes it as an actual pollinator.
We have observed only Eul. meriana to pollinate flowers on two occasions. In both cases a solitary bee entered
the floral tube for two to four seconds and with no delay departed along a straight line track. Collection of
fragrance of floral origins has never been observed, suggesting that flowers are deceptively pollinated (s#
Lubinsky et ai. (2006) for a more detailed description).
236
the drawings and type specimens are not useful (Soto Arenas, pers. comm.). These represent a diverse group
of poorly collected Amazonian species and field work is necessary to resolve species concepts. Soto-Arenas
and Cribb (2010) place it within subgenus Xanata, section Xanata.
Habitat and Distribution. — Only two fertile collections possessing flowers are known and they are both
from the Peruvian Amazon (Soto Arenas, pers. comm.). In our region V. cristato-callosa occurs in M.jlexum
wetlands characterized by deep histosols where it exhibits a patchy distribution, occurring in small popula-
tions within localized areas. Generally, the species is more tolerant of lower light conditions than other local
Vanilla species.
Natural History.— Flowering occurs in mid-January to April, during the height of the rainy season. In
agricultural settings where resource limitations are reduced the flowering season is greatly extended several
months. In this case, the terminal racemes prevent upward growth and the vine remains vertically stunted.
Ephemeral flowers open singly over an interval of two to four days. Pollination rates are approximately 6%
(Table 1). The pollinator is unknown.
Fruits of reproductive individuals in their natural habitat often occupy a terminal position on the inflo-
rescence, suggesting that once pollinated, continued flower production ceases. Fruits mature in approximately
six months. Green, brittle fruits turn pliable and oily during dehiscence, a transformation that begins at
the tips of newly separated valves and progressively moves upward. In this species a gelatinous substance
is sometimes present at the confluence of the valves as they separate. Dehiscent fruits are oily with a strong
vanilla aroma, somewhat spicy and cinnamon-like. On several accounts we observed a few individuals of a
single species of metallic green Euglossa sp. visiting old, completely opened fruits and demonstrating typical
fragrance collection behavior (Roubik 1989). The species exhibits a degree of myrmecology, as ants that
feed on a sugary exudates secreted at the abscission line between bud and ovary are often present during
bud and raceme development.
5. Vanilla bicolor Lindl (Fig. 2c).
ternodes 8.5-10.5 cm long. Aerial roots produced from each node, attached to host, thin, ca. 3-100 cm in
length, sometimes few-branched at apical end, tannish in color. Leaves alternate, produced from each node
opposite to roots, distichous, drooping from long petioles; petiole slightly caniculate, ca. 1 cm long, dark
red-brown; leaf blade elliptic to slightly ovate, ca. 2.5-4.5 cm wide and 5-11 cm long, discolorous, with
pale-green central area and pale red-brown outline along margins, succulent, texture smooth, apex acute,
margin entire. Inflorescences axillary and terminal, racemose, fleshy, green, sessile, one to 20 per individual
ca. 1 to 10-flowered; bracts cymbiform, ca. 0.3 cm wide and 0.6 cm long, color red-brown, non-floriferous
bracts generally two; pedicel ca. 3 cm long and 0.2 cm in diameter, dark red when young, maturing to green
Flowers resupinate, fleshy, tender, 10 cm across, arranged spirally on raceme, non-fragrant, remaining at-
tached to ovary if successfully pollinated; buds deep red-brown; dorsal sepal oblanceolate, ca. 0.7 x 7.5 cm,
orange-brown; lateral sepals oblanceolate, ca. 0.9 x 7.5 cm, orange-brown; petals oblanceolate-linear, a
0.6 x 6.5 cm, very pale orange-brown; lip trumpet-like when fresh, fused with column for three quarters of
its length, exterior ringed white around mouth fading to yellowish-orange towards interior, margins entire,
flabellate and almost triangular when flattened, unguiculate, the claw of soft, yellow tomentose-papillo*
hairs near apex merging with short yellow pubescence in interior and callus; callus diffuse, ca. 0.7xft3
cm, with thick yellow tomentose; column straight, ca. 4.2 cm long and 0.2 cm wide, ventral surface with
lanuginous-hirsute hairs for about a third of its length distally, then glabrous; rostellum extremely thin,
narrow, pale white; anther ca. 0.3 x 0.4 cm. Fruit dehiscent, cylindric, ca. 10 cm long, when immature
brittle and brownish red or green, splits longitudinally along sutures, splitting valves turning black,
leathery, revealing seeds embedded in a scarcely oily mesocarp, non-fragrant.
Notes. — Vanilla bicolor has been generally known by its later synonym, V. wrightii Rchb. F. (Christen**
1995). It differs from all other American Vanilla (except for the membranaceous group [V guianensisl), toth^
Householder et al., Diversity, natural history, and conservation of Vanilla 237
the penicillate callus is not compact, but rather diffuse. Soto-Arenas and Cribb (2010) placed it in subgenus
Xanata section Xanata.
Distribution and Habitat. — This species has a wide distribution, previously being collected in Cuba,
Ecuador, British Guyana, and Venezuela (Soto-Arenas & Cribb 2010). This is the first documentation of V.
bicobr in Peru, significantly extending its range to the south. In our region it is a common epiphyte of Mau-
ritiajlexuosa. Long, slender roots reach deep into the slender pocket formed between the sheathing petiole
bases of palm fronds and the trunk. It is largely restricted to specific wetland habitat characterized by deep
histosols and acidic waters. The species is occasionally encountered as a short terrestrial vine in very open
areas with high insolation. These uncommon individuals reach maximum heights of 3 to 4 m well below
canopy level and never gain epiphytic status in M. flexousa crowns. Terrestrially rooted individuals likely
result from the clonal propagation of fallen epiphytes rather than seedling germination and represent the
exception, rather than the rule, thus we maintain its classification as a true epiphyte, at least in our region.
That said, at Kaieteur National Park, Guyana, the species is known to scramble across rocks at ground
Natural History.— The flowers of V. bicolor are non-fragrant, open singly at approximately five-day inter-
vals, and last less than 24 hours. Flowering phenology is fairly aseasonal. We have observed flowers during
all 12 months of the year, although there seems to be a slight peak in August during the height of the dry
Pollination rates are extremely high, falling within the range of 45 to 75% (Table 1). Nc
have been observed at or near flowers and bagging experiments suggest that the plant is self-pollinating.
> to be achieved by the secretion of excess stigmatic fluids that creates a mixture of
e and pollen grains. The rather thin, narrow rostellum is an ineffective barrier between the pollen-
ixture and the stigmatic surface, greatly facilitating self-pollination through excessive stigmatic
Fruiting phenology is aseasonal. Mature fruits, which are quick-drying and non-oily, can be observed
throughout the year. The thin, leathery capsule opens from below over the course of several days. The splitting
halves turn black as they dry, however they remain pliable for some time. Seeds are not held in a particularly
°tly matrix as in other fragrant species. The mechanism of seed dispersal of V. bicolor is unknown. Lack
of fragrance, oils, or sugars, presence of extremely small seeds, and its canopy habitat are all suggestive of
dispersal by a mixture of wind turbulence and gravity. As the fruits dehisce and dry, the seeds are exposed
to strong winds and then fall downward. Seedlings are abundant, germinating almost exclusively in the
openings between palm trunks and the sheathing petiole bases of palm fronds. This species is not
ProPagated vegetatively in artificial settings.
6 VaniUa Palmarum Lindl. (Fig. 21).
Obligate epiphyte on M auritiaflexuosa. Stems succulent, ca. 0.5 cm in diameter, green, internodes 5-6 cm
or climbing individuals, one to many-branched. When reproductive often forming a long shoot with slightly
’minishing leaf size and internode length. Aerial roots produced from each node, either short (less than 2
Jto) and cup-like on reproductive shoots or long and thin on vegetative shoots. Leaves alternate, produced
bom each node opposite to roots, distichous, sessile, succulent; blade elliptic to ovate, ca. 3.2-6 cm wide
-16 cm long, color light green, texture smooth, apex acute to acuminate, margin entire. Inflorescences
f t y terminal, racemose, fleshy, green, ca. 5 to 15-flowered; bracts either foliaceous or cymbiform;
Jiaceous bracts located at inflorescence base, elliptic, ca. 3 cm long, apex acute, generally two, occasion-
* 7 "on-Aoriferous; cymbiform bracts above, ca. 0.5 cm long, green, always floriferous; pedicel green, but
^®ewhat Offish at extreme base. Flowers resupinate, tender, fleshy, arranged spirally, flowers lightly
6agrant- training attached to ovary if successfully pollinated; buds green; dorsal sepal elliptic, ca. 1.2 x
cmCm’ ^1W- aPex acute; lateral sepals slightly falcate with bend beginning 2.5 cm from base, ca. 1.3 x 5.6
’ yellow; Petals elliptic, ca. 1.6 x 5.5 cm, yellow, apex acute, longitudinally ribbed on dorsal surface; lip
largely pentagonal when flattened, uniformly yellow, unguiculate, the claw of 6-7 longitudinal ridges formed
by conspicuous swollen tissues, margins undulate; callus absent; column slightly convex, ca. 4 cm long and
0.25 cm wide, hirsute on ventral surface near rostellum; rostellum somewhat thick, yellow. Fruit dehiscent,
triangular in cross-section, green and brittle when immature, ca. 12 cm long, splitting longitudinally in
two unequal valves at maturity, valves turning dark brown, pliable, leathery, revealing seeds embedded in
a scarcely-oily mesocarp, non fragrant.
Notes. — Vanilla palmarum belongs to subgenus Xanata, section Xanata. It differs from all other species
within the group in lacking a penicillate callus (Soto-Arenas & Cribb 2010).
Habitat and Distribution. — Vanilla palmarum is widespread in Madre de Dios, occurring in the small
but ubiquitous Mauritia Jlexuosa- dominated wetlands characterized by hydromorphic substrates with a
high mineral component. The species has been observed in wide (approximately 200 m) spring-fed streams
with sandy bottoms, on unconsolidated clays on the edges of sping-fed blackwater lakes, and in rain-fed
depressions on gleysols. In all cases M. Jlexuosa was monodominant. Similar to V. bicolor, the vining herbs
are epiphytic in M. jlexuosa crowns, sending long, slender, branched roots deep into the sheathing petioles.
Although restricted to the same host, V. palmarum and V. bicolor never occur in mixed populations, the latter
restricted to M. Jlexousa wetlands with thick, accumulated organic material.
This species is also known from the Guianas where it also grows in the crowns of Mauritia Jlexuosa. It
is also reported to occur commonly in northeastern Brazil and in Mato Grosso, Brazil, where is grows on
Syagrus coronata (Arecaceae), a common arborescent palm of drier ‘cerrado’ formations (Toscano de Brito,
pers. comm.). Other possible hosts include species of the palm genera Orbigyna, Attalea, and Scheelea ( Soto-
Arenas & Cribb 2010).
Natural History. — Flowers open singly on the racemes at intervals of three to four days. Flowering phe-
nology is bimodal, with peaks in January to February and October to November. The flowers are slightly
fragrant but no potential pollinators have been observed near flowers. Pollination rates are high at 76%,
apparently resulting from self-pollination (Table 1). Despite a seemingly more effective barrier provided by
stigmatic fluids induce pollen grains to leak onto the extreme lateral margins of the stigmatic surface. Pol-
lination rates of flowers located basally on racemes are significantly higher than flowers located apically.
This may indicate a degree of control of the self-pollination mechanism, perhaps induced by high resource
requirements of fruit maturation and mediated by the allocation of limited resources to competing vegeta-
tive and reproductive needs (Chiariello & Gulman 1991). The fruits are non-fragrant and dehisce over the
course of several days. Although significantly thicker than V. bicolor the fruit mesocarp is not particularly
oily and the two valves rapidly dry within weeks, turning leathery. Seedlings are abundant in the canopies
of M. Jlexuosa palms as in V. bicolor. Reproductive adults high in the canopy probably seed the very special-
ized germination microsites through combination of wind turbulence and gravity.
DISCUSSION
Wetlands and Vanilla
This work represents the accumulation of about seven years of data from field investigations of Vanilla orchids
in the wetlands of Madre de Dios. Early in our investigations we directed our attention to wetland habitat as
Vanilla showed distinct preferences for hydromorphic conditions. In the area, the genus generally occurs in
association with Mauritia Jlexuosa, or the “aguaje” palm as it is known locally in Peru. The palm is indicative
of hydromorphic soils in topographical depressions and often forms monodominant stands, or “aguafila
(Kahn 1988). Substrates, vegetation structure, and flora vary according to water source and hydrological
parameters, but all M. Jlexuosa formations are characterized by: (1) very low sediment loads during rela-
tively brief periods of inundation, (2) relatively minor water level flucuations (less than 50 cm), and (3)1°*’
canopies with greater light penetration into the subcanopy (Kahn et al. 1993). Relatively higher insolation
and stable hydrological parameters favor the establishment of vining Vanilla species, which have precarious
seedling stages in the understory and, as adults, superficial root systems vulnerable to prolonged flooding.
Under these conditions Vanilla often occurs in dense populations and is one of the most conspicuous ele-
ments of wetland vegetation. For example, in a related study concerning the distribution patterns of Vanilla
inregional wetlands, approximately 60% of points placed systematically throughout 70 km of line transect
in 28 wetlands, contained at least one species of Vanilla, although up to four species occurred at the same
point within a 30 m radius. Vanilla pompona subsp. grandiflora alone was present in over 40% of points and
has population densities exceeding 250 individual ramets per hectare (Householder, unpublished data).
While we have collected over 50 species of orchid associated with Maurita wetlands, surprisingly few
other orchid taxa are found on M. Jlexuosa trunks, probably the result of the challenges of frequent distur-
bances created by falling palm fronds, slow growth of epiphytic orchids (Zotz 1995), and lack of suitable
germination sites on the bare, smooth Mauritia trunks (Nieder et al. 2000). Within the Orchidaceae, the
vining habit is unique to the genus Vanilla (Dressier 1993), and the habit’s role in palm-dominated wet-
lands is noteworthy. All of the epiphytes are obligate to M.flexousa and their vining habit allows individu-
als to maintain their canopy positions by occupying new fronds as senescing ones are sloughed. Unlike
the epiphytes, Vanilla climbers are not restricted to Mauritia hosts, however the palm’s shear dominance
does make it their most common phorophyte. Climbing vines with terrestrial roots are able to maintain or
quickly regain their elevated positions in the canopy after frequent disturbances caused by falling fronds.
This level of adaptability may be extremely important in a constantly changing environment, especially if
seed germination is limited (Price & Marshall 1999). In short, the success of Vanilla in Mauritia-dominated
wetlands is likely a reflection of the shared ecological inclinations of both palm and orchid and the peculiar
growth form of the orchid that provides resilience under a constant disturbance regime.
Fragrance and Bees Interactions
Three of the six Vanilla species of Madre de Dios possess pleasantly fragrant fruits. The fragrant condition
is associated with succulent fruits with an oil-rich mesocarp and frequent visits by fragrance-collecting
Euglossine bees. Hanging fruits slowly split into two separating valves over the course of days to weeks,
depending on the species and environmental conditions (fruits in full sun seem to split faster). The split-
ting valves turn pliable and dark brown to black in color. The opening of the suture is usually preceded
ty a noticeable band of yellow discoloration that moves upwards as the valves separate. The intensity of
fruit fragrance during dehiscence, even on the same fruit, is quite variable from day to day suggesting that
the complex enzymatic reaction responsible for producing f:
influenced. The separate valves can remain pliable and me
Progressively leathery as they dry out on the vine.
The dispersal mechanism(s) of Vanilla remains enigmatic; however several authors have suggested Eu-
fene bees or bats as potential dispersers (Madison 1981, Soto-Arenas 2003, Lubinsky et. al 2006). We
°«*rved Trigona, and possibly Euglossine bees, to form sticky seed packets which they often fly away with,
nglossine bees are well-known scent collectors; however the Trigona bees may have been more interested
mthe nutritional value of the oils. Herbivory, possibly due to vertebrate mammals, seems to be extremely
rBt 11131 said> the seeds are easily rubbed off and are extremely sticky due to a thin covering of oil. This
C0Venng may aid in epizoochorous forms of seed dispersal by any visitor, insect or vertebrate.
The adaptive value and evolutionary significance of fruit fragrance remain unexplored. Fruit fragrance,
aSsociated with an oily mesocarp, is a common trait amongst New World Vanilla (Soto-Arenas 2003) and
Uwas clearly the main attractant for the Euglossine bee visitors. However, we were surprised by the ap-
j^t specificity of bees to fruits. Vanillin, one of the most important aromatics in the commercialization
mia fruit, is a general attractant to dozens of Euglossine species in southeastern Peru (Pearson and
v SSler 1985)- Thus, the few species of bees lured to fragrant fruits of V. pompona subsp. grandiflora and
^ato-callosa suggests that it is the less abundant aromatics or modifiers other than, or in combination
J ' Vanillm that may be responsible for the observed specificity (Hills et al. 1972). Such phenomena have
n well documented in orchid flowers pollinated by Euglossine bees (Adams 1966; Gerlach 2010) where
240
specific pollinator relationships, floral fragrance, and reproductive success are clearly linked (Dodson
1962; van der Pijl & Dodson 1966; Dodson et al. 1969; Hills et al. 1972). However, links between disperser
specificity and reproductive success are less clear, at least in Vanilla. Curiously, in the case of V. pompom
subsp. grandiflora, both flowers and fruits seem to attract the same Euglossine species, presumably due to
the presence of similar aromatic compounds. The potential parallels between flower and fruit fragrance raise
intriguing questions concerning the possible role of pollinator interactions in evolution of fruit fragrance of
Vanilla. Thus, while we suspect that seed dispersal by bees (Madison 1981; Lubinsky et al. 2006), bats, or
other animals is entirely possible, and that fragrance may indeed play an important role, we are unsure of
the role that these disperser relationships may have played in the evolution of fruit fragrance.
Vanilla bicolor is expected to have fragrant fruits (Soto-Arenas et al. 2003). However, the relatively dry,
non-oily fruits are odorless. The lack of fragrance is curious given its basal position within the clade of fra-
grant Vanilla (Soto-Arenas et al. 2003). Lubinsky (2006) and Dressier (1989) suggest that succulent, possibly
fragrant fruits may be the ancestral condition in the genus. Madison (1981) suggests that a transition from
bee dispersal (possibly associated with fruit fragrance) to wind dispersal (non-fragrant), would be a simple
evolutionary step within the family. In any case, the evolution of fruit fragrance within the genus would be
much enlightened by mapping its occurrence or disappearance on a phylogeny. However, neither detailed
information concerning natural history nor a complete phylogeny of the genus exists.
Distribution Patterns
Vanilla is often considered to be a genus of habitat specialists, often restricted to small geographic ranges and
high levels of endemism (Soto-Arenas et al. 2003). Dispersal mechanism(s) remain enigmatic, but Vanilla is
notable for its smooth, crustose seeds with a well-developed endosperm (Garay 1986). These traits are highly
atypical of the family with the majority of the 20,000-30,000 species possessing tiny, dust-like, highly cor-
rugated seeds with little or no endosperm neatly adapted for wind dispersal (Dahlgren and Clifford 1983).
Several authors, using evidence from observational (Madison 1981), morphological (Cameron and Chase
1998), and molecular (Nielsen and Siegismund 1999) studies, suggest that Vanilla is efficiently dispersed by
animals only at local, narrow scales. This is consistent with our current understanding of endemism and
restricted distribution patterns of neotropical Vanilla species (Soto-Arenas et al. 2003).
example, Vanilla bicolor has previously only been collected in northern South America (Venezuela, Guyana,
Ecuador) and the extensive populations discovered in southern Peru are highly disjunct. Vanilla palmarm
has a similar disjunct distribution, known from Mato Grosso, French Guyana, and now southern Peru.
Either these widespread species have more effective long-distance dispersal mechanisms than imagined, or
their previously continuous ranges have been fragmented by vicariance events in the geological past. In the
latter case, the apparent association between Amazonian Vanilla and hydromorphic conditions is intriguing
in light of palynological (Rull 1998; Hoorn 2006) and paleogeographical (Hoorn 1993; Hoorn 1994) studies
suggesting that semi-aquatic environments of lakes and Mauritia-dominated swamps were extensive dur-
ing the deposition of the upper Pebas Formation in the early-mid Miocene. Indeed, an ancient, extensive
wetland system covering over one million km2, in the heart of the modern Amazon basin (Wesselingh et
al. 2002) would have been a cradle of speciation for wetland-associated organisms and a means of range
expansion for others (i.e.. Vanilla). With Andean orogeny and the formation of the modern Amazon Riyer-
many wetland-associated organisms may have become progressively isolated and fragmented, pushed to the
peripheries of their original distributions displaced by the highly pulsating, highly depositional environ-
ment that characterizes most Amazonian rivers. The extent to which this has influenced current species
distribution, evolution, and diversity of Amazonian Vanilla , however, is limited to conjecture, especialty
without molecular evidence.
Conservation
Our studies of the flora of the wetlands of Madre de Dios has led to the documentation of nearly 600 specks
244
BOOK REVIEW
Brian Lofun and Shirley Loflin. 2(
with flaps). Texas A&M Uni\
77843-4354, U.S.A. (Ordei
photos, 103 maps, 5 3/4" x <
Cacti: A Field Guide. (ISBN 978-1-60344-108-7, flexbc
John H. Lindsey Building, 4354 TAMU, College Station, I
tiupress.com, 1-800-826-8911). $24.00, 312 pp., 483 c
PHYLOGENETIC ANALYSES OF THE
GAYLUSSAC1A FRONDOSA COMPLEX (ERICACEAE: VACCINIEAE)
BASED ON MOLECULAR AND MORPHOLOGICAL CHARACTERS
Michael T. Gajdeczka, Kurt M. Neubig, Walter S. Judd
Department of Biology
220 Bertram Hall
P.O. Box 118526
W. Mark Whitten, Norris H. Williams, Kent D. Perkins
University of Florida Herbarium
379 Dickinson Hall
P.O.Box 110575
Florida Museum of Natural History
University of Florida
Gainesville, Florida 32611, USA
ABSTRACT
RESUMEN
246
Journal of the Botanical Research Institute of Texas 4(1)
Gaylussacia Kunth is a New World genus of 53 species of mainly understory shrubs occurring in mesic to
xeric woodlands and shrublands, as well as acidic bogs. The genus is differentiated from the closely-related
Vaccinium L. by ten-locular ovaries, drupaceous fruits containing ten pits, the presence of resin glands on the
leaves (except for G. brachycera (Michx.) A. Gray), and the lack of staminal spurs (Duncan & Brittain 1966;
Luteyn et al. 1996; Palser 1961; Wood 1961). However, the phylogenetic relationships between members of
the two genera are still uncertain (Kron et al. 2002b). Vaccinium was shown to be non-monophyletic as tradi-
tionally circumscribed, with Gaylussacia nested within it (Kron et al. 2002a). These results are corroborated
by the ambiguous generic identity of G. brachycera, which, in addition to being eglandular and similar in
form to Vaccinium species, has been shown to be sister to the rest of Gaylussacia (Floyd 2002). Finally, some
GayIussacia(Vander Kloet & Dickinson 1992).
Gaylussacia usually has been divided into three sections (following Sleumer 1967): section Vitis-idm
Drude (only G. brachycera), section Gaylussacia (ca. 47 species, mainly South American; plants mostly ever-
green, with stalked glands on the leaves), and the North American section Decamerium Torr. & A. Gray (ca.
five species; plants deciduous, with sessile glands on the leaves). The last is the focus of this study. All three
sections are represented in eastern North America: G. brachycera, four of the species of section Gaylussada
(G. mosieri Small, G. dumosa (Andr.) A. Gray, G. orocola (Small) Camp, G. bigeloviana (Femald) Sorrie &
Weakley; see Sorrie & Weakley, 2007, although the latter two entities are often included within an expanded
G. dumosa), and all five species of section Decamerium [G. frondosa (L.) Torr. & A. Gray, G. tomentosa (A.
Gray) Pursh ex Small, G. nana (A. Gray) Small, G. ursina (Curtis) Torr. & A. Gray, and G. baccata (Wang.) K.
Koch], For more information on the history and systematics of these groups see Camp (1935), Wood (1961),
Luteyn et al. (1996), Floyd (2002), and Sorrie and Weakley (2007).
A phylogenetic analysis of Gaylussacia had not been carried out prior to that of Floyd (2002). In her
morphology-based analysis, which included most of the species of the genus, the monophyly of sects. Decame-
rium and (of course) Vitis-idaea were supported, but sect. Gaylussacia was paraphyletic. All three traditional
sectional divisions were weakly to moderately supported by cpDNA ( tmL-tmF ) data, as well as in analyses
combining morphological and molecular data, but nrlTS sequences alone did not support the monophyly
of either sect. Decamerium or sect. Gaylussacia. Molecular data were lacking for most species of sect. Gflyfos-
sacia and relationships within this large clade remain poorly understood. Floyd’s (2002) analysis also did
not clarify the placement of G. brachycera; neither did it fully clarify species delimitations and relationships
within sect. Decamerium.
Camp (1941) recognized three subsections within sect. Decamerium, two of which, subsects. Baccate
and Ursinae, are monotypic, i.e., Gaylussacia baccata and G. ursina, respectively. The remaining three taxa,
i.e., G. frondosa, G. tomentosa and G. nana, were placed in his subsection Frondosae. The taxa within the
Frondosae group have had a confusing taxonomic history and have been variously circumscribed. Elliott
(1821) and Chapman (1889) recognized only G. frondosa. Radford et al. (1964) recognized only G. frondosa,
but included two varieties: i.e., G. frondosa var. tomentosa A. Gray and var. frondosa; Wunderlin and Hansen
(2003) did the same, recognizing only var. tomentosa as occurring in Florida. Gray (1878) was the first au-
thor to recognize G. frondosa var. tomentosa, and he also recognized G. frondosa var. nana A. Gray. Harpe
(1906) treated the members of subsect. Frondosae as a single species with three varieties, i.e., G. frondosa
var. frondosa, var. tomentosa, and var. nana. Sleumer (1967) also treated the subsection as a single sped*
but recognized G. frondosa var. polycodioides Camp and f. glaucophylla Camp (both pertaining to northern
plants usually treated within var. frondosa). Small (1897, 1933), Camp (1935, 1941), Wood (1961), Dune*”
and Brittain (1966), and Luteyn et al. (1996) treated the subsection as three separate species (G. frondosa, 0-
tomentosa, and G. nana). Most recently, Floyd (2002) treated the subsection as one species with three variet-
ies. However, she suggested that var. nana perhaps should be considered a separate species, citing a posa*
Gajdeaka et al. Phylogenetic analysis of the Gaylussacia
247
resulting in poor resolution in her cladograms. She recommended more research in order to resolve species
field observations (over many years, by W.S. Judd) of the notable and seemingly consistent morphologi-
cal differences between the sympatric Gaylussacia tomentosa and G. nana in Florida motivated the present
study (Fig. 1). The purpose of this study is to resolve the phylogenetic relationships within this subsection,
i.e., the G.frondosa complex, by focusing on plants identified as G.frondosa, G. tomentosa and G. nana. We
sought to determine appropriate species limits within the G. frondosa complex, based on phenetic, evolu-
tionary, diagnostic, and apomorphic species concepts (Davis & Nixon 1992; Donoghue 1985; de Queiroz
2007; Judd et al. 2007; Mishler 1985; Mishler & Theriot 2000; Wheeler & Platnick 2000; Wiley & Mayden
2000).
MATERIALS & METHODS
Taxon sampling and field work. — Voucher material representing the Gaylussacia frondosa complex was col-
lected by M.T. Gajdeczka and W.S. Judd from numerous localities in Florida, Georgia, South Carolina, and
North Carolina in the spring and summer of 2007 in order to estimate genetic diversity (Table 1; all deposited
at FLAS). Gaylussacia ursina and G. baccata were also collected, and were selected as outgroups based on
the traditional classification of section Decamerium (Camp 1941) and the recent phylogeny of Floyd (2002).
Leaf material for DNA extraction of field-collected specimens was preserved in silica gel. DNAs derived
from herbarium collections also were included in the study to supplement the number of evolutionary units
in the analyses, especially of G. frondosa, as well as outgroup taxa. One collection of G. dumosa (of section
Morphology.— Potentially phylogenetically informative morphological characters were selected after
careful consideration of the pattern of variation seen in material at the University of Florida Herbarium of
the Florida Museum of Natural History (FLAS), field observations (of M.T. Gajdeczka and W.S. Judd), and
previous taxonomic work on the genus (Duncan & Brittain 1966; Luteyn et al. 1996; Floyd 2002). Seventy
qualitative and quantitative morphological characters were initially assessed; all above ground organs were
examined, including both vegetative and reproductive features, with an emphasis on density, distribution,
and form of the hairs (both glandular and non-glandular). From the initial list of characters observed, 20
were selected as potentially phylogenetically informative and were included in the matrix for analysis (see
Tables 2, 3).
Unicellular hair presence, density and length and resin gland presence, density and width have been
considered especially useful in distinguishing species of Gaylussacia (Floyd 2002; Luteyn et al. 1996), and
these were included in the analyses. Most of the initial hair and gland measurements could be grouped
m character-clusters, which showed identical or very similar patterns of variation among taxa. In such
^single character (from each group of highly correlated characters) was selected for inclusion in the
Phylogenetic analyses. However, measurement of the length of the longest hair and average hair length of
tlle adaxial leaf surfaces (chars. #5 and 6; see Table 2) were both included in order to take into account the
°bserved variation in the range of hair length between specimens.
Many morphological characters used in the analyses (Table 2) were readily divisible into discrete states,
Voiding arbitrary decisions relating to state delimitation (Stevens 1991). However quantitative characters
P^dmore problematic. Variation in these characters (e.g., chars. #1, 3, 6, 7, 8, 11, 12, and 13; see Table 2)
** assessed by means of bar graphs, and the states of those included in the analyses were delimited by more
w less discrete 8aPs (e g-. Fig. 2, char. #9). The most problematic characters in terms of state delimitation
j*re non-glandular hair length on the abaxial leaf surface (#8), and the length (#33) and width (#34) of the
jongest blades; these characters showed nearly continuous distributions, but were included in the analyses
^use they have been stressed as taxonomically important in the Gaylussacia frondosa complex (Duncan
nttain 1966), and because the overlap in values between taxa was limited. Many characters could not
mcluded in the analyses because they showed too much infraspecific variation or varied continuously
Gaylussacia Kunth is a New World genus of 53 species of mainly understory shrubs occurring in mesic to
xeric woodlands and shrublands, as well as acidic bogs. The genus is differentiated from the closely-related
Vaccinium L. by ten-locular ovaries, drupaceous fruits containing ten pits, the presence of resin glands on the
leaves (except for G. brachycera (Michx.) A. Gray), and the lack of staminal spurs (Duncan & Brittain 1966;
Luteyn et al. 1996; Palser 1961; Wood 1961). However, the phylogenetic relationships between members of
the two genera are still uncertain (Kron et al. 2002b). Vaccinium was shown to be non-monophyletic as tradi-
tionally circumscribed, with Gaylussacia nested within it (Kron et al. 2002a). These results are corroborated
by the ambiguous generic identity of G. brachycera, which, in addition to being eglandular and similar in
form to Vaccinium species, has been shown to be sister to the rest of Gaylussacia (Floyd 2002). Finally, some
Vaccinium species have ovaries that are pseudo-10-locular and, therefore, resemble the 10-locular ovaries of
Gaylussacia (Vander Kloet & Dickinson 1992).
Gaylussacia usually has been divided into three sections (following Sleumer 1967): section Vitis-idm
Drude (only G. brachycera), section Gaylussacia (ca. 47 species, mainly South American; plants mostly ever-
green, with stalked glands on the leaves), and the North American section Decamerium Torr. (sc A. Gray (ca.
five species; plants deciduous, with sessile glands on the leaves). The last is the focus of this study. All three
sections are represented in eastern North America: G. brachycera, four of the species of section Gaylussacia
(G. mosieri Small, G. dumosa (Andr.) A. Gray, G. orocola (Small) Camp, G. bigeloviana (Fernald) Sorrie &
Weakley; see Sorrie & Weakley, 2007, although the latter two entities are often included within an expanded
G. dumosa), and all five species of section Decamerium [G. frondosa (L.) Torr. & A. Gray, G. tomentosa (A.
Gray) Pursh ex Small, G. nana (A, Gray) Small, G. ursina (Curtis) Torr. & A. Gray, and G. baccata (Wang.) K.
Koch], For more information on the history and systematics of these groups see Camp (1935), Wood (1961),
Luteyn et al. (1996), Floyd (2002), and Sorrie and Weakley (2007).
A phylogenetic analysis of Gaylussacia had not been carried out prior to that of Floyd (2002). In her
morphology-based analysis, which included most of the species of the genus, the monophyly of sects. Decam-
rium and (of course) Vitis-idaea were supported, but sect. Gaylussacia was paraphyletic. All three traditional
sectional divisions were weakly to moderately supported by cpDNA (tmL-tmF) data, as well as in analyses
combining morphological and molecular data, but nrlTS sequences alone did not support the monophyly
of either sect. Decamerium or sect. Gaylussacia. Molecular data were lacking for most species of sect. Gayhts-
sacia and relationships within this large clade remain poorly understood. Floyd’s (2002) analysis also did
not clarify the placement of G. brachycera; neither did it fully clarify species delimitations and relationships
within sect. Decamerium.
Camp (1941) recognized three subsections within sect. Decamerium, two of which, subsects. Baccatae
and Ursinae, are monotypic, i.e., Gaylussacia baccata and G. ursina, respectively. The remaining three taxa,
i.e., G. frondosa, G. tomentosa and G. nana, were placed in his subsection Frondosae. The taxa within the
(1821) and Chapman (1889) recognized only G. frondosa. Radford et al. (1964) recognized only G. frondosa,
but included two varieties: i.e., G. frondosa var. tomentosa A. Gray and var. frondosa; Wunderlin and Hansen
(2003) did the same, recognizing only var. tomentosa as occurring in Florida. Gray (1878) was the first au-
thor to recognize G. frondosa var. tomentosa, and he also recognized G. frondosa var. nana A. Gray. Harper
(1906) treated the members of subsect. Frondosae as a single species with three varieties, i.e., G. fronds
var. frondosa, var. tomentosa, and var. nana. Sleumer (1967) also treated the subsection as a single specks,
but recognized G. frondosa var. polycodioides Camp and f. glaucophylla Camp (both pertaining to northern
plants usually treated within var. frondosa). Small (1897, 1933), Camp (1935, 1941), Wood (1961), Duncan
and Brittain (1966), and Luteyn et al. (1996) treated the subsection as three separate species (G. frondosa, 6
tomentosa, and G. nana). Most recently, Floyd (2002) treated the subsection as one species with three varitt-
ies. However, she suggested that var. nana perhaps should be considered a separate species, citing a possi»
lineage sorting or hybridization event as being responsible for its incongruent placement in her nrlTS an®
cpDNA trees. Yet, as noted by Floyd, such problems could also merely be due to inadequate data, poss^f
247
resulting in poor resolution in her cladograms. She recommended more research in order to resolve species
limits within the Frondosae group.
Field observations (over many years, by W.S. Judd) of the notable and seemingly consistent morphologi-
cal differences between the sympatric Gaylussacia tomentosa and G. nana in Florida motivated the present
study (Fig. 1). The purpose of this study is to resolve the phylogenetic relationships within this subsection,
Le., the G.frondosa complex, by focusing on plants identified as G.frondosa , G. tomentosa and G. nana. We
ijpqght to determine appropriate species limits within the G. frondosa complex, based on phenetic, evolu-
tionary, diagnostic, and apomorphic species concepts (Davis & Nixon 1992; Donoghue 1985; de Queiroz
2007; Judd et al. 2007; Mishler 1985; Mishler & Theriot 2000; Wheeler & Platnick 2000; Wiley & Mayden
MATERIALS & METHODS
Taxon sampling and field work. — Voucher material representing the Gaylussacia frondosa complex was col-
lected by M.T. Gajdeczka and W.S. Judd from numerous localities in Florida, Georgia, South Carolina, and
North Carolina in the spring and summer of 2007 in order to estimate genetic diversity (Table 1; all deposited
at FLAS). Gaylussacia ursina and G. baccata were also collected, and were selected as outgroups based on
the traditional classification of section Decamerium (Camp 1941) and the recent phylogeny of Floyd (2002).
leaf material for DNA extraction of field-collected specimens was preserved in silica gel. DNAs derived
from herbarium collections also were included in the study to supplement the number of evolutionary units
in the analyses, especially of G.frondosa , as well as outgroup taxa. One collection of G. dumosa (of section
iflylussacia) was included as a more distant outgroup.
Morphology. — Potentially phylogenetically informative morphological characters were selected after
careful consideration of the pattern of variation seen in material at the University of Florida Herbarium of
the Honda Museum of Natural History (FLAS), field observations (of M.T. Gajdeczka and W.S. Judd), and
Previous taxonomic work on the genus (Duncan & Brittain 1966; Luteyn et al. 1996; Floyd 2002). Seventy
qualitative and quantitative morphological characters were initially assessed; all above ground organs were
examined, including both vegetative and reproductive features, with an emphasis on density, distribution,
and form of the hairs (both glandular and non-glandular). From the initial list of characters observed, 20
were selected as potentially phylogenetically informative and were included in the matrix for analysis (see
Tables 2, 3).
Unicellular hair presence, density and length and resin gland presence, density and width have been
considered especially useful in distinguishing species of Gaylussacia (Floyd 2002; Luteyn et al. 1996), and
thus these were included in the analyses. Most of the initial hair and gland measurements could be grouped
mt° character-clusters, which showed identical or very similar patterns of variation among taxa. In such
^ses, a single character (from each group of highly correlated characters) was selected for inclusion in the
Phyogenetic analyses. However, measurement of the length of the longest hair and average hair length of
^Jdaxial leaf surfaces (chars. #5 and 6; see Table 2) were both included in order to take into account the
e variation in the range of hair length between specimens.
Many morphological characters used in the analyses (Table 2) were readily divisible into discrete states,
avoiding arbitrary decisions relating to state delimitation (Stevens 1991). However quantitative characters
P"*ed more problematic. Variation in these characters (e.g., chars. #1, 3, 6, 7, 8, 11, 12, and 13; see Table 2)
or , ass*"SSed by means of bar graphs, and the states of those included in the analyses were delimited by more
^discrete gaps (e.g., Fig. 2, char. #9). The most problematic characters in terms of state delimitation
l ^-gkndular hair length on the abaxial leaf surface (#8), and the length (#33) and width (#34) of the
^gest blades; these characters showed nearly continuous distributions, but were included in the analyses
&Bnt?they ^ bCen stressed as taxonomically important in the Gaylussacia frondosa complex (Duncan
k'ncl^11 and because the overlap in values between taxa was limited. Many characters could not
U ed m the analyses because they showed too much infraspecific variation or varied continuously
across taxa (and thus could not be delimited into states). Characters not observed for particular species were
scored as missing values, as were situations where a character was considered “not applicable” in a particular
taxon, and coded as T (Table 2).
DNA extraction, amplification and sequencing. — Leaves and flowers dried in silica-gel were used
DNA extraction of field collected material. A modified version of the 2x CTAB procedure (Doyle &
1987) was used, as described in Whitten et al. (2007) with the addition of proteinase K (5 units) to d*
extraction buffer instead of 2-mercaptoethanol. Precipitated DNA pellets from extractions of field-collected
material were resuspended in 200 pL of Tris-EDTA (TE) buffer. Small fragments of leaf tissue were removed
A Crooks 526
A Dorr 1909
M. Gajdeczka 47
M.Gajdeczka48
Clay Co., FL
Sevier Co., TN
Oconee Co, SC
Transylvania Co, NC
Transylvania Co, NC
FJ985235
FJ985195
FJ98521 1
250
Table 2. Morphological characters and character states included in the phylogenetic analyses of Gaylussacia.
1. Plant height: n < 0.7 m(0);n ^ 0.7 m (1).
2. Ratio of average length of lower secondary branches to length of main stem: n < 7 (0); n > 7 (1).
3. Length of longest non-glandular hair on branch: n < 0.5 mm (0); n >0.5 mm (1).
4. Number (density) of non-glandular hairs on the adaxial leaf surface per standardized 2.56 mm2 region: none (0); 0 < n
<10(l);n^10(2).
5. Length of longest non-glandular hair on adaxial leaf surface: 0 < n < 025 mm (0); n >025 mm (1 ).
6. Length of average non-glandular hair on adaxial leaf surface; 0 < n < 0.1 75 mm (0); 0.1 75 < n < 0.322 mm (1); n 2 0322
mm (2).
7. Number (density) of non-glandular hairs on the abaxial leaf surface per standardized 1 .32 mm2 region: 0<n <6.5(0); 65
in<40(1);n^40(2).
8. Length of longest non-glandular hair on abaxial leaf surface: 0 < n < 0275 mm (0); 0275 <, n < 0.45 mm (1); n 2 0.45 mm©
9. Number (density) of resin glands on abaxial leaf surface per standardized 4.0 mm2 region: 0 < n < 1 9 (0); 1 9 < n < 65 (1);
1 0. Glaucousness of leaves: none (0); slight (1); moderate (2); strong (3).
1 1 . Length of blade (average of five longest leaves per specimen): n < 41 .1 mm (0); 41 .1 £ n < 70 mm (1);n^.70mm(2).
1 2. Width of blade (average of five longest leaves): n < 203 mm (0); 203 £ n < 30 mm (1 ); n ^ 30 (2).
1 3. Petiole length (average of five longest leaves): n < 1 .35 mm (0); n> 1.35 mm(1).
14. Bract type: leaf-like (0); small (1; states coded and scored as per Luteyn et al. 19%, and for this character and #16, 17, It?
and 1 9, our observations, when these were possible, always matched these scorings).
1 5. Resin glands on adaxial leaf surface: present (0); absent (1).
1 6. Fruit color: black (0); blue (1 ; states coded and scored as per Luteyn et al. 1 9%).
1 7. Corolla color: white (0); greenish-white to pinkish white (1); red (2; states coded and scored as per Luteyn et al. 1996).
18. Corolla shape: cylindrically urceolate (0); broadly urceolate (1; states coded and scored as per Luteyn et al. 1996).
1 9. Pilose hairs on the filaments: present (0); absent (1; states coded and scored as per Luteyn et al. 1 9%).
20. Exudate around perimeter of resin glands on leaves: present (0); absent (1).
Gajdeaka et al.r Phylogenetic analysis of the Gaylussacia frondosa complex
251
US Judd 1679
HGajdeaka 17
M. Gajdeaka 19
HGajdeaka20
M.Gajdeaka21
M. Gajdeaka 24
M. Gajdeaka 30
M. Gajdeaka 32
M. Gajdeaka 33
M Gajdeaka 36
HGajdeaka37
At Gajdeaka43
M. Gajdeaka 58
M. Gajdeaka 16
M. Gajdeaka 22
H Gajdeaka 23
H Gajdeaka 31
Gajdeaka 38
M Gajdeaka 60
M. Gajdeaka 61
M Godfrey 8484583
kH Pittman 05090108
<MS eckingerJr.271
J-B, Nelson 16,473
#G ajdeaka56
*A tojdeaka57
WS Judd 786
Smith J563
U Dorr 1909
B Crooks 526
M-Gajdeaka47
Gajdeaka 48
^Gajdeaka 50
WS,Judd7i6
EW. Wood 4222
BVascott405
^ Judd 1657
*Gajdeaka18
G. tomentosa
G. tomentosa
G. tomentosa
G tomentosa
G. tomentosa
G. tomentosa
G. tomentosa
G. tomentosa
G. tomentosa
Gtomentosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
G. frondosa
CTon number represented by three herbarium sheets, and a few leaves do show slight glaucousness.
herbarium specimens for DNA extraction. Anticipating a lower DNA concentration, the herbarium
DNa Uons Were resuspended in only 120 pL of Tris-EDTA (TE) buffer. In all extractions the resuspended
m ^eP“rified using Qiaquick columns (Qiagen, Valencia, California, USA) and Buffer PE, then eluted
AUPCRr
5j , . reactions were carried out using Sigma Jumpstart Taq polymerase and reagents (Sigma-Aldrich,
tlieran5’ MiSSOuri’ USA)- Initially a survey of phylogenetically informative regions were carried out using
-^mal nrlTS region and seven plastid DNA regions (trnL-F, rpl32-trnL, trnQ-rpsl6, tmH-psbA,
• psbD-trnT and trnV-ndhC ) based on suggestions of phylogenetically useful regions in Small et al.
252
Journal of the Botanical Research Institute of Texas 4(1)
(1998) and Shaw et al. (2005, 2007). Two representative field-collected specimens of each of the three in-
group taxa ( Gaylussaciafrondosa , G. tomentosa, and G. nana) and one sample each of G. baccata and G. ursina
were included in the survey. The reaction mixture for amplification of nrlTS from field collected specimens
included 7.0 pL of betaine (5 M), 12 pL H2O, 2.5 pL 10X buffer, 2.0 pL MgCh (25 mM), 0.5 pL dNTPs (10
pM), 0.5 pL of each primer (10 pM), 0.5 pL template, and 0.2 pL Taq polymerase (25.5 pL total). In order
to improve amplicon concentration, volumes were adjusted to 1.0 pL template and 11.0 pL H2O (25.7 ill
total). Reaction mixtures for the amplifications from herbarium specimens totaled 24.7 pL (18 pL H20, 2.5
pL 10X buffer, 2.0 pL MgCh (25 mM), 0.5 pL dNTPs (10 pM), 0.5 pL each of forward and reverse primer
(10 pM), 0.5 pL template, and 0.2 pL Taq polymerase).
The following PCR protocol was used for the initial survey: an initial bake at 94°C for 3 min; then 30
cycles of (1) denaturation at 94°C for 30 s, (2) annealing at 60°C for 30 s, and (3) extension at 72°C for 2
min; and a final extension at 72°C for 3 min. Primers 17SE and 26SE from Sun et al. (1994) and the PCR
program in Whitten, et al. (2007) were used for amplification and sequencing of the nrlTS region from field
ITS A/ITS C and ITS B/ITS D and the protocol from Blattner (1999) were used, and the reaction mixture was
doubled to 50.4 pL (30 pL H2O, 6 pL MgCh (25 mM), 5 pL 10X buffer, 4 pL template 2 pL each of forward
and reverse primer (lOpM), 1.0 pL dNTPs (lOpM), and 0.4 pL Taq polymerase).
Phylogenetic analyses. — Herbarium specimens, based on the first author’s own field work or from
the collections of FLAS, and representing populations of all the entities within the Gaylussacia Jrondosa
complex, were employed as terminal taxa in the phylogenetic analyses. These analyses served as the basis
not only of the assessment of phylogenetic relationships within Gaylussacia section Decamerium but also
for hypotheses of species circumscription (based on an application of the phylogenetic/apomorphic species
concept; Donoghue 1985; Mishler 1985; Mishler & Theirot 2000).
PAUP* 4.0bl0 (Swofford 2002) was used to construct most parsimonious trees for three data sets: (1)
a morphological analysis of 38 specimens (i.e., terminal taxa, representing populations; 26 of these field-
collected and 12 herbarium/FLAS collections) and 20 characters, and 19 of these parsimony informative
(see Table 2), (2) a nrlTS analysis of 47 specimens and 827-857bp (31 field-collected specimens) or 648-664
bp (FLAS herbarium material, 16 specimens) but only 12 parsimony-informative sites (and 19 total variable
characters), and (3) a combined “total evidence” analysis, i.e., morphology + nrlTS, for a pruned data set of
31 specimens (24 field-collected, and 7 herbarium), and 48 variable characters, with 31 of these parsimony
informative. All characters were equally weighted and unordered. The morphological analysis used a maxi-
mum parsimony (MP) heuristic approach, 500 random-addition replicates, TBR, MaxTree = 200 per replicate,
and MulTrees on. Relative support for clades in all analyses was evaluated using a fast-heuristic bootstrap
analysis (1000 replicates in the morphological analysis, 100 replicates in the nrlTS and combined analyses).
The nrlTS and analysis used a MP heuristic approach, 100 random-addition replicates, TBR, MaxTree = lOptf
replicate, and MulTrees on. The combined analysis was similar but used MaxTree = 10,000. Gaps were coded
as missing data; indels were not coded as characters. Morphological character state changes were mapped onto
the morphological cladogram and traced onto the total evidence MP cladogram using the PAUP* output in
MacClade 4.05 (Maddison & Maddison, 2005). The molecular and morphological analyses resulted in trceS
with no strongly supported incongruent patterns, so were combined (see “total evidence” analysis, above).
RESULTS
Morphological analysis. — A heuristic search using maximum parsimony yielded 3360 most parsimonious
trees (MPTs) with a length of 65 (Consistency Index [Cl] = 0.462, Retention Index [RI] = 0.829, Reseat
Consistency Index [RC] = 0.383). The strict consensus tree is shown in Figure 3. Gaylussacia tomentosa, 0-
nana, G.frondosa, and G. ursina form a clade with moderate support (bootstrap [BS] = 78%) in all trees (Fig
3A). The G.frondosa complex, i.e., G. tomentosa, G. nana, and G.frondosa, also forms a clade (BS * 85%; Fig
3). Within the G. frondosa complex, populations of G. tomentosa form a clade (without bootstrap suppo^’
253
^thmTter t0 3 dade comPrisinS the populations of G.frondosa and G. nana (without bootstrap support).
Within 6 G jfr°nd0Sa + G‘ nana clade, Ae populations of G. nana form a clade (BS = 58%), which is nested
*^%)3b>ara^^etiC G'-frondosct- The terminals representing populations of G. ursina also form a clade (BS
Ut G' baccata is unresolved at the base of the Decamerium clade.
254
nrlTS analysis. — None of the plastid regions showed phylogenetically informative variation, so after
preliminary work these regions were not pursued. However, the 47 sequences of the nrlTS region did show
useful variation and, therefore, were studied further. A MP heuristic search (100 replicates, MaxTrees « 10
per replicate) of the nrlTS sequences resulted in a total of 1000 MPTs with a length 33 (Cl = 0.970, RI »
0.992, RC = 0.962). The strict consensus tree is shown in Figure 3B. The monophyly of Gaylussacia nana
(BS = 56%) was weakly supported. A G. tomentosa + G. frondosa clade (BS = 95%) was strongly supported.
A third clade in the strict consensus tree groups G. baccata and G. ursina (BS = 69%).
Combined analysis.— The combined total evidence analysis included all specimens that had both
successfully sequenced nrlTS regions and morphological data. A MP heuristic search resulted in a total of
1000 MPTs of length 89 (Cl - 0.663, RI = 0.878, RC = 0.582). A representative tree is shown in Figure 4,
with the strict consensus and bootstrap consensus values mapped. Clades Gaylussacia nana (BS = 91%) and
G. tomentosa + G. frondosa (BS * 87%) are strongly supported and are sister in the strict consensus tree, thus
supporting the monophyly of the G. frondosa complex (BS = 62%). All specimens of G. tomentosa form a
monophyletic group in the strict consensus tree (but without BS support). Specimen Gajdeczka 43 (MG43),
probably a hybrid (see below), and the operational taxonomic units (OTUs) representing G. frondosa forma
paraphyletic grade in the G. tomentosa + G. frondosa clade, and G. frondosa forms a clade only in some of the
equally parsimonious trees. The G. tomentosa + G. frondosa + G. nana clade is sister to a strongly supported
G. ursina clade in the SC (BS = 95%). A G. baccata clade (BS = 100%) is sister to the G. ursina + G. tomentosa
+ G. frondosa + G. nana clade (BS * 59%).
Morphological synapomorphies. — Mapping morphological characters onto a randomly chosen MPT
from the combined (or total evidence) analysis showed that most clades were supported by at least two mor-
phological synapomorphies. The monophyly of Gaylussacia nana in the combined MPT was supported by six
character states: short non-glandular hairs on both adaxial and abaxial surfaces of the leaf (chars. #5, 6, and
7, Table 2), short and narrow blades (#1 1 and 12) and short petioles (#13). The monophyly of G. tomentosa was
supported by dense non-glandular hairs on both adaxial and abaxial leaf surfaces (chars. #4 and 7), as well
as long hairs on the branches (#3) and on the abaxial leaf surface (#8). Gaylussacia tomentosa specimens also
were less glaucous than those of G. frondosa and especially G. nana (char. #10, see also Fig. 1). Gaylussacia
frondosa (when monophyletic) was poorly supported by morphological synapomorphies: most G. frondosa
specimens are characterized by sparse or absent hairs on the adaxial leaf surface (char. #4) and medium-
length leaf hairs on the abaxial surface (char. #8), although the variation pattern is homoplasious. In addition,
Gaylussacia frondosa tends to have fewer unicellular hairs on its twigs than either G. tomentosa or G. nana.
Specimens of Gaylussacia ursina formed a monophyletic group supported by the average length of the
non-glandular hairs on the adaxial leaf surface (char. #6) (as opposed to the “longest hair” measurement),
long abaxial leaf hairs (#8), the longest and widest blades of all taxa in the analysis (#11 and 12), usually
nonglaucous leaves (#10), and sparse glands on the abaxial leaf surface (#9). The monophyly of G. baccata
was supported by five unique synapomorphies: very dense abaxial leaf glands (char. #9), a red corolla (#17).
a cylindrical^ urceolate corolla (#18), and presence of glands on the adaxial leaf surface (#15). Additional
derived features that distinguished it were sparse adaxial leaf hairs (char. #7), medium length abaxial leaf
hairs (#8), black coloration of fruits (#16), presence of filament hairs (#19), lack of glaucousness (#10), aDC*
narrow leaf blades (#12).
The Gaylussacia tomentosa + G. frondosa clade was supported mainly by features that differentiate it
from G. nana: tall plant height (char. #1), long adaxial leaf hairs (#5 and 6), dense abaxial leaf glands (#30),
long and wide leaf blades (#11 and 12), and long petioles (#13). The monophyly of the G. tomentosa + c
frondosa + G. nana clade was supported by dense abaxial leaf hairs (char. #30), leaves with slight to intense
glaucousness (#10; in many specimens, but lost in some, especially G. tomentosa), blue fruits (#16), and gla-
brous filaments (#19). Finally, the monophyly of the G. tomentosa + G. frondosa + G. nana + G. ursina cla*
was supported by a lack of adaxial leaf glands (char. #15), greenish-white to pinkish-white coloration (#17)
of the broadly urceolate corolla (#18), and resin glands lacking exudate (#20).
256
Putative hybrids. — Various specimens are putatively hybrids between G. tomentosa and G. nana (Orze D
& Bridges 19346, Herring 1548, Gajdeczka 21, Gajdeczka 43, Gajdeczka 75, and Gajdeczka 76) or between G.
frondosa and G. nana (Judd 3118) and all are deposited at FLAS. Most of these specimens were excluded from
analysis because of the intermediate nature of the data. The interspecific hybrid status of these specimens is
supported by both DNA and morphology. The putative G. tomentosa x nana hybrids are noteworthy in having
leaves that are usually glaucous (and thus similar to G. nana) but also have long unicellular hairs on their
abaxial surfaces (like G. tomentosa); their leaves often are also smaller then is characteristic of G. tomentosa
(especially so in Gajdeczka 21), again like G. nana. The specimen Judd 3118 has small, abaxially glaucous
leaves (similar to those of G. nana), but their adaxial surfaces are essentially glabrous (like G. frondosa), and
at all of the variable nucleotide positions among these three species. For example, at base 163 in our ITS
hybrids exhibit a strong polymorphic signal for both C and T. Additional polymorphic sites (substitutions)
are found at sites 207, 344, 358, 679, 682, as well as two indels (insertion/deletion) from bases 704-711 and
a single base indel at 739. Because we did not clone these PCR products to separate the different copies of
ITS, we cannot be certain of the distribution of these characters and their association with either parental
species. However, it has been shown that PCR-mediated recombination can cause false or mixed signal of
the original parental types (Lahr & Katz 2009). Nonetheless, we are confident that these sequences represent
interspecific hybrids because of the high association of the polymorphic nucleotide positions only at the
variable sites among different species and at every variable site among the species in question (see Soltis et
al. 2008) and because of the intermediate morphology discussed above.
Taking all three analyses into account, it is clear that Gaylussacia nana and G. tomentosa are cladospecies
(Donoghue 1985; Mishler 1985; Mishler & Theriot 2000), i.e., each of these species is a monophyletic group
of populations (see Donoghue 1985; Mishler 1985; Mishler & Theriot 2000). Thus, both should be recognized
as distinct species. We note that they are largely sympatric and are easily distinguished by the morphologi-
cal characters listed above. The monophyly of G. nana is supported by all analyses, with strong support in
the combined analysis (see Fig. 4, BS = 91%). Diagnosable features for G. nana included short hairs on both
adaxial and abaxial leaf surfaces, short and narrow blades, and short petioles (and compare these with
features of G. tomentosa above). Recognition of these two species is also justified from the perspective of the
diagnostic species concept (Wheeler & Platnick 2000; Davis & Nixon 1992), and the phenetic/taxonomie
species concept (Judd et al. 2007; Smith 1994) as evidenced by the data presented in Duncan and Brittain
(1966). Finally, they must surely represent distinct evolutionary lineages (de Quieroz 2007; Wiley 1978) as
= 24 in G. tomentosa).
Given that Gaylussacia tomentosa and G. nana are largely sympatric, and very frequently co-occurring,
it is not surprising that we found a few morphologically intermediate specimens— likely hybrids— that also
exhibited polymorphic ITS sequences (possibly as a result of their possessing ITS sequences representative
of the parental species). The frequency of such plants appears to be quite low (i.e., 7 specimens out of nearly
200 observed); most individuals (both in the field and herbarium) are easily determined. More research is
necessary on the breeding system of these plants, and the ploidy levels of both the parental species and the
putative hybrids needs to be assessed. Despite the occurrence of these putative hybrids, we believe that G.
nana and G. tomentosa are largely reproductively isolated (possibly by the difference in chromosome number),
thus fitting the biological species concept (Mayr 1969).
The results presented here provide strong support for the treatments of Small (1933), Camp (1935, 1941).
Wood (1961), Duncan and Brittain (1966) and Luteyn et al. (1996), all of whom distinguished G. i
G. tomentosa at the species level. Recognition at the varietal level, as proposed or adopted b
Chapman (1889), Harper (1906), Radford et al. (1964), Sleumer (1967), and Floyd (2002) is considered
inappropriate. A monophyletic Gaylussacia tomentosa is present in the nrlTS and combined strict consensus
trees, albeit without BS support, although the morphological analysis does not resolve G. tomentosa into a
elade in all trees (see Fig. 3). Twelve morphological characters distinguish G. tomentosa from G. nana (Table
2, chars. #2-9, 11-13). Five of the characters also separate G. tomentosa from G.Jrondosa (chars. #2, 3, 4, 6
and 7); these vary infraspecifically to a greater extent within G. tomentosa and G./rondosa than in G. nana,
except branch hair length (char. #3, Table 3). Finally, the difference in ploidy distinguishes G. tomentosa,
which is the only tetraploid in the subsection (Ldve 1976; Luteyn et al. 1996), providing a distinctive apo-
morphic feature for this taxon.
The taxonomy of Gaylussacia jrondosa is more problematic. Gaylussacia frondosa has no support as
a monophyletic group in any of the strict consensus trees, yet the monophyly of this taxon is supported
in some of the most parsimonious trees resulting from the combined analysis (Fig. 4). The two character
states that could be considered diagnostic for G. frondosa are the absence of adaxial leaf hairs except for
the midvein and margin (char. #4; Tables 2, 3) and medium-length abaxial leaf hairs (#23). However, their
pattern of variation is homoplasious. Thus, G. frondosa lacks clear morphological autapomorphies, which
likely contributes to its lack of support in the nrlTS and morphological analyses. Although placed as a close
relative of G. tomentosa in the combined analyses, we note that these two taxa can be clearly differentiated
by chromosome number, n = 12 in G. frondosa and n = 24 in G. tomentosa (Love, 1976; Luteyn et al., 1996).
In addition, G. frondosa has a more northern distribution (to NH and MA; Luteyn et al. 1996; Camp 1935)
than does G. tomentosa, so in large parts of their ranges they are allopatric. It is commonly reported that
G. Jrondosa can often reach heights of 2-3 m (Luteyn et al. 1996; Camp 1941), while G. tomentosa seldom
reaches heights above 1.5 m (Luteyn et al. 1996). Camp (1941) also mentioned the highly clonal habit of G.
tomentosa and G. nana, as opposed to the more spread out, less rhizomatous habit of G. frondosa.
Most putative Gaylussacia tomentosa x nana hybrids were not included in the nrlTS analyses due to their
polymorphic sequences, however, Gajdeczka 43 was included (Figs. 3, 4) as its ITS sequence is fairly clear,
flus specimen (as discussed under results) is morphologically intermediate between G. tomentosa and G.
"®w (e.g., it has intermediate-length hairs on the abaxial leaf surface, and it may have been slightly glaucous)
and its nrlTS sequence shows a slight indication of polymorphism at the variable nucleotide positions (i.e.,
small secondary peaks are present at the polymorphic sites). The ITS sequence of this specimen was coded
“smgthe major peaks at these sites, and thus it was placed in the G. tomentosa/frondosa clade (instead of
theG- nana clade) in the ITS tree (Fig. 3B). Had these sites been coded as polymorphic its position in the
cladogram would have been unresolved (within the G. frondosa complex). Its more isolated placement in the
^consensus of the combined analysis (Fig. 4), i.e., sister to the remaining members of the G. tomentosa
* G.Jrondosa clade, results from the interplay between the ITS sequence (with characters arbitrarily coded
as those of G. tomentosa) and the presence of certain morphological characters of G. nana, e.g., shorter hairs
<m the abaxial leaf surface. Gajdeczka 43 is placed in the G. tomentosa clade (Fig. 3) in the morphology-based
T*?. due to the more numerous apomorphies of this species (and indeed, this specimen is phenetically
®^t similar to G. tomentosa). Another putative hybrid, Gajdeczka 21, was included in the morphology-based
genetic analysis, and this plant, like Gajdeczka 43, was placed in the G. tomentosa clade, likely as a
** 1 its exPressing more synapomorphies of G. tomentosa than G. nana (as coded in our matrix). Of
conKe’ SUcJl placements are influenced by the necessarily somewhat arbitrary delimitations of some of the
^phological character states (see methods). The nrlTS sequence of Gajdeczka 21 allows us to confidently
^Pothesize that this specimen represents a hybrid individual, and its placement with G. tomentosa acces-
tn the morphological strict consensus tree (Fig. 3A) does not cause us to doubt that it is very likely of
origin. Such placements of hybrid individuals are to be expected (McDade 1990, 1992).
baccata and G. ursina were considered outgroups in this study, and as such, their intraspe-
^taxonomy was not assessed (Floyd 2002). In addition, their species limits have not been controversial.
inthe1S SUPPOrted 3S 3 cladosPecies in the total evidence trees (Fig. 4). Strangely, G. baccata is paraphyletic
Morphology-based trees (forming the basal branches of the cladogram).
The nrlTS analysis resulted in some unexpected results, e.g., Gaylussacia baccata and G. ursina formeda
clade (Fig. 3B), although the resolution of the analysis was too poor to shed light on subsectional relationships,
likely as a result of insufficient phylogenetically informative characters. Floyd’s (2002) nrlTS topology was
similarly unexpected (i.e., sect. Decamerium was shown as polyphyletic), and her nrlTS data was incompatible
with her tmL-tmF data (as assessed by a partition homogeneity test). Possible lineage sorting or hybridiza-
tion events affecting the nrlTS region might be contributing to the observed irregularities (Floyd 2002), and
it should be noted that Floyd coded indels as additional characters, while we did not. Understanding the
discrepancies between nrlTS sequences and other data sets used for hypothesizing phylogeny in Gaylussacia
requires additional study.
Our morphological ingroup topography also shows discrepancies with the nrlTS and combined topog-
raphies, i.e., Gaylussacia frondosa groups with G. nana (rather than G. tomentosa ) in the morphological trees.
Three character states are synapomorphies of the G. frondosa + G. nana clade in the morphological tree: habit
(#2), short non-glandular hairs on the stems (#3), and moderately dense hairs on the abaxial leaf surface (#7).
Floyd’s (2002) morphological analysis places a G. tomentosa + G. nana clade (BS - 73%) sister to G. frondosa;
G. tomentosa and G. nana also group together in Floyd’s tmL-tmF topology. Her analysis, however, did not
include any morphological character states that were shared by G. nana and G. frondosa, but also differed
from those of G. tomentosa. Our characters #2 and 3 are not treated in Floyd’s study, while character #7 is
coded similarly for G. frondosa (but the state is shared with G. nana), and we note that it is not unexpected
delimitations. Also in her analysis, several character states relating to hair density (glabrous twigs, adaxial
leaf surfaces and leaf margins, few hairs on abaxial leaf surface) are unique to G. frondosa, and thus separate
it from G. tomentosa and G. nana. Twig hair density was measured in our study (and G. frondosa does have
the least dense hairs), but this character was excluded from our analyses because it follows a similar pattern
of variation as other hair density characters. It is possible that glabrous to sparsely pubescent twigs could
represent a synapomorphy for the populations of G. frondosa.
Finally, the morphological (BS = 85%) and combined (BS = 62%) data support the recognition of a
monophyletic Gaylussacia frondosa complex (i.e., subsect. Frondosae) as it is commonly treated (Camp 1941;
Sleumer 1967; Floyd 2002). In the majority rule consensus tree of the nrlTS analysis (not shown), a G. bac-
cata + G. ursina clade is sister to a G. tomentosa + G. frondosa clade in 52% of trees (with G. nana sister to all
four species). In Floyd (2002) a G. tomentosa + G. frondosa + G. nana clade was evident in most analyses.
In conclusion, within Gaylussacia sect. Decamerium subsect. Frondosae, it is clear that the largely sym-
patric G. nana and G. tomentosa each represent distinct species (regardless of the species concept applied),
while the evidence for the distinctiveness of G. frondosa is more problematic — although our analyses provide
some preliminary evidence that it, too, should be recognized as specifically distinct (Fig. 4). An identification
key for these species is available in Luteyn et al. (1996). More work, however, needs to be done to determine
relationships within the complex, as the analyses are incongruent in the placement of G. frondosa, the rep-
resentatives of which are more closely related to G. nana in the morphological analysis and are more closely
related to G. tomentosa in the nrlTS analysis. We also recommend additional investigations on the ploidy
level of members of subsect. Frondosae, especially the putative hybrids.
ACKNOWLEDGMENTS
This work was conducted as the senior thesis of the first author, who thanks the co-authors for their assis-
tance in the project. We thank Gretchen M. Ionta for assistance in using the scanning electron microscope »
compare the leaf surfaces of Gaylussacia tomentosa and G. nana, and Momingside Nature Center (Gainesville.
Alachua Co.) for permission to collect material. We thank Jennifer Floyd and an anonymous reviewer for
their helpful comments on the manuscript. The first author also acknowledges the Howard Hughes Medical
Institute (HHMI) Science for Life program for providing support for summer laboratory and field work in
connection with this research.
259
REFERENCES
Buttneh FR. 1999. Direct amplification of the entire ITS region from poorly preserved plant material using re-
combinant PCR BioTechniques 27:1 180-1186.
Gw,W.H. 1935. Studies in Ericales I.The genus Gaylussacia in North America north of Mexico. Bull.Torrey Bot.
Oub 62:1 29-1 32.
Camp,W.H. 1941. Studies in the Ericales: a review of the North American Gaylussacieae; with remarks on the
origin and migration of the group. Bull. Torrey Bot. Club 68:53 1-551.
Chapman, A.W. 1889. Flora of the southern United States. Ed. 2. Ivison, Blakeman, and Company, New York.
Dwb,JJ.anoK.C Nixon. 1992. Populations, genetic variation, and the delimitation of phylogenetic species. Syst.
Sol. 41:421 -435.
kQueiroz, K. 2007. Species concepts and species delimitation. Syst. Biol. 56:879-886.
Donoghue, MJ. 1985. A critique of the biological species concept and recommendations for a phylogenetic
alternative. Bryologist 88:1 72-1 81 .
Dowt JJ. and J.L. Doyle. 1 987. A rapid DNA isolation procedure for quantities of fresh leaf tissue. Phytochem.
M.E. Brittain. 1966. The genus Gaylussacia (Ericaceae) in Georgia. Bull. Georgia i
1-15.
Duncan, W.H.
24:13-26.
Ewott, S. 1821 . A sketch of the botany of South Carolina and Georgia. J.R. Schenck, Charleston, South Carolina.
Floyd, J.W. 2002. Phylogenetic and biogeographic patterns in Gaylussacia (Ericaceae) based on nuclear DNA,
and chloroplast DNA variation. Syst. Bot. 27:99- 115
tov,A. 1878. Manual of the botany of the northern United States. Ivison, Blakeman,Taylorand Company, New York.
Hwet dM. 1906. A phytogeographical sketch of the Altamaha Grit Regions of the Coastal Plain of Georgia.
Annals of the New York Academy of Science. Ann. New York Acad. Sci. 1 7:1 -41 4.
too,W.S.,C.S. Campbell, E.A. Kellogg, P.F. Stevens, and MJ. Donoghue. 2007 [2008]. Plant systematics: a phylogenetic
approach, ed. 3. Sinauer Assoc., Sunderland, Massachusetts.
•NKA,WS. Judd, P.F. Stevens, D.M. Crayn, A.A. Anderberg, P.A. Gadek, CJ. Quinn, and J.L Luteyn. 2002a. Phylogenetic
classification of Ericaceae: molecular and morphological evidence. Bot. Rev. 68:335-423.
, KA, EA. Powell, and J.L Luteyn. 2002b. Phylogenetic relationships within the blueberry tribe (Vaccinieae,
ricaceae) based on sequence data from matK and nuclear ribosomal ITS regions, with comments on the
Placement of Satyria. Amer. J. Bot. 89:327-336.
fflDj.G.ANDLA Katz. 2009. Reducing the impact of PCR-mediated recombination in molecular evolution and
^vironmental studies using new-generation high-fidelity DNA polymerase. Biotechniques 47:857-866.
jr/ 1976‘ l0PB Chromosome number reports HI. Taxon 25:341-346.
7’J'L'W.S. Judo, S.P. Vander Kloet, U. Dorr G.D. Walwce, KA Kron, P.F. Stevens, and SE. Clements. 1 996. Ericaceae
^^southeastern United States. Castanea 61:101-144.
M«r f WP AND D R' Madd,son- 2005. MacClade 4.08. Sinauer Associates, Sunderland, Massachusetts.
^ • 1969. Principles of systematic zoology. McGraw-Hill Book Co., New York, USA.
I*11990- Hybrids and phylogenetic systematics. I. Patterns of character expression in hybrids and their
McDadU11005 f°r dad'StiC ana|ysis- Evolution 44:1 685-1 700.
4&l329-^HybridS anCl phy,09enetic systematics. Il.The impact of hybrids on dadistic analysis. Evolution
morphological, developmental, and phylogenetic basis of species concepts in bryophytes.
^^.D. and E. Theriot. 2000. The phylogenetic species concept sensu Mishler and Theirot: monophyly,
HPhy, and phylogenetic species concepts. In: Q.D. Wheeler and R. Meier, eds. Species concepts and
tHe0ry: 3 debat& Columbia University Press, New York. Pp. 44-54.
UnitpH c 1 ’ Studies 0<r floral morphology in the Ericales. V. Organography and vascular anatomy in several
wet» States species of the Vacciniaceae. ” ■ -
Bot. Gaz 123:79-1 11.
Radford, A.E., H.E. Ahles, and C.R. Bell. 1 964. Guide to the vascular flora of the Carolinas. University of North Carolina
Chapel Hill, North Carolina.
Shaw, J., E.B. Lickey, J.T. Beck, S.B. Farmer, W. Liu, J. Miller, K.C. Siripun, C.T. Winder EE. Schilling, and R.L Small 2005. The
tortoise and the hare II: relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis
Amer.J. Bot 92:142-166.
Shaw, J., E.B. Lickey, E.E. Schilung, and R.L Small 2007. Comparison of whole chloroplast genome sequences to
choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. Amer.J.
Bot. 94:275-288.
Sleumer, H. 1 967. Die Gattung Gaylussacia H.B.K. Bot. Jahrb. 86:309-384.
Small J.K. 1 897. The genus Gaylussacia in the southern states. Pp. 441 -445. In: Shrubs and trees of the southern
states - II. Bull.Torrey Bot. Club 24:437^45.
Small J.K. 1 933. Manual of the southeastern flora. University of North Carolina Press, Chapel Hill, North Carolina
Small R.L., J A Ryburn, R.C. Cronn, T. Seelanan, and J.F. Wendel. 1 998. The tortoise and the hare: choosing between
noncoding plastome and nuclear Adh sequences for phylogeny reconstruction in a recently diverged plant
group. Amer. J. Bot. 85:1 301-1315.
Smith, A.B. 1 994. Systematics and the fossil record: documenting evolutionary patterns. Blackwell Science, Oxford
UK; Cambridge, Massachusetts.
Soltis, D.E., E.V. Mavrodiev, JJ. Doyle, J. Rausches and P.S. Sans. 2008. ITS and ETS sequence data and phylogeny
reconstruction in allopolyploids and hybrids. Syst. Bot. 33:7-20.
Sorrie, B.A. and A.S. Weakley. 2007. Notes on the Gaylussacia dumosa complex (Ericaceae). J. Bot. Res. Inst. Texas
1:333-344.
Stevens, P.F. 1991. Character states, morphological variation, and phylogenetic analysis: a review. Syst. Bot
16:553-583.
Sun, Y., DZ. Skinner G.H. Liang, and S.H. Hulbert. 1994. Phylogenetic analysis of Sorghum and related taxa using
internal transcribed spacers of nuclear ribosomal DNATheoret. Appl. Genetics 89:26-32.
Swofford, D.L 2002. PAUP* Phylogenetic analysis using parsimony (*and other methods). Version 4.0b1 0. Sinauer
Associates, Sunderland, Massachusetts.
Vander Kloet, S.P. and TA Dickinson. 1 992. The taxonomy of Vaccinium section Hemimyrtillus. Bot. Mag., Tokyo
105:601-614.
Wheeled Q.D. and N.I. Platnkx 2000. The phylogenetic species concept (sensu Wheeler and Platnick). In: Q.D.
Wheeler and R. Meier, eds. Species concepts and phylogenetic theory: a debate. Columbia University Press,
New York. Pp. 55-69.
Whitten, W.M., M.A. Blanco, N.H. Williams, S. Koehler G. Carnevali, R.B. Singe* L. Endara, and K.M. Neubig. 2007. Molecular
phylogenetics of Maxillaria and related genera (Orchidaceae: Cymbidieae) based on combined molecular
data sets. Amer. J. Bot. 94:1 860-1 889.
Wiley, E.0. 1 978. The evolutionary species concept reconsidered. Syst. Zoo. 27:1 7-26.
Wiley, EO. and R.L. Mayden. 2000. The evolutionary species concept. In: Q.D. Wheeler and R. Meier, eds. Species
concepts and phylogenetic theory: a debate. Columbia University Press, New York. Pp. 70-89.
Wood, C.E., Jr. 1 961 . The genera of Ericaceae in the southeastern United States. J. Arnold Arbor. 42:1 0-80.
Wunderlin, R.P. and B.F. Hansen. 2003. Guide to the vascular flora of Florida. University Press of Florida, Gainesville,
Florida.
POPULATION GENETIC ANALYSIS OF ARGEMONE PLEIACANTHA SUBSP.
P1NNATISECTA (SACRAMENTO PRICKLY POPPY, PAPAVERACEAE) AND
RE-EVALUATION OF ITS TAXONOMIC STATUS
Sandy D. Cervantes
Rajanikanth Govindarajulu
Department of Biology Natural Heritage New Mexico Department of Biology
Hew Mexico State University UNM Biology Department New Mexico State University
LasQuces, New Mexico 88003, U.SA Albuquerque, New Mexico 871 31, USA LasCruces, New Mexico 88003, US
Patrick J. Alexander
Department of Biology
New Mexico State University
Las Cruces, New Mexico 88003, USA
C. Donovan Bailey
Department of Biology
New Mexico State University
Las Cruces, New Mexico 88003, USA
Author for Correspondence - dbailey@nmsu.e(
ABSTRACT
RESUMEN
INTRODUCTION
^understanding of species has changed dramatically throughout the history of botanical nomenclature.
J ,0Ugh disagreement persists, contemporary species concepts broadly agree in viewing species as mor-
IdatefCally and/°r 8enetically discontinuous groups of populations (e.g., Nixon & Wheeler 1990) that are
)n d t0 each other through common evolutionary history, with reproductive isolation playing a key role
gating discontinuity in sexual taxa (e.g., Dobzhansky 1935; Mayr 1942). This understanding derives
part from a fusion of ideas from systematics, paleontology, cytology, and genetics, which became
of Sot aSthC modern synthesis” (Huxley 1942). Previous authors, especially prior to Darwin’s On the Origin
oj^es, largely vkwed specks ^ temporaUy unchanging but not necessarily morphologically discontinu-
■ tuies that were not connected through evolutionary history. Many treated species and infraspecific
merdy as tools for naming natural variation, not as fundamental units of evolution.
262
non-specialists to assume that “species” in contemporary discussions reflect the application of a species
concept that is consistent with modern evolutionary biology. However, the majority of the roughly 2.5 mil-
lion described species (e.g., May 1988) and infraspecific taxa were published prior to the On the Origin of
Species, much less works of the modem synthesis. Few authors for these taxa were aware that the species
rank would one day represent a critical boundary in evolutionary biology.
Ideally the limits of all species would be determined consistently, based on contemporary species
concepts, but the scale of such work and limited global investment in the field makes this level of study
unlikely in the foreseeable future (e.g., Heywood 2001; Scotland et al. 2003). The historical and potentially
arbitrary application of rank at and below the level of species may not impact many taxa. However, over-
interpretation of ranks can be an acute problem in conservation, with critics often suggesting that threat-
ened subspecies and varieties are less than “species” and therefore unworthy of protection. When taxa are
threatened with global extinction, scientific assignments based on modern concepts of species are critical
for our understanding of the taxon and for downstream conservation decisions (see Desalle & Amato 2004;
Holsinger & Gottliebo 1991; Van Dyke 2008). This study was conducted to evaluate the species status of
Argemone pleiacantha Greene subsp. pinnatisecta G.B.Ownbey (Sacramento Prickly Poppy) is a federally
listed endangered member of Papaveraceae, known from a few small populations on the western slope of
the Sacramento Mountains in Otero County, New Mexico (Fish & Wildlife Service 1989). In the most recent
monograph of Argemone, Ownbey (1958) described the taxon from his own specimens and an 1899 Wooton
collection. This geographically restricted (Fig. 1) taxon was distinguished from other Argemone pleiacantha
Greene by the presence of simple bud prickles, paler yellow latex, and sparingly prickly capsules (Fig. 1).
In the introduction to the monograph, Ownbey (1958) explicitly provided a species concept consistent
with numerous contemporary concepts. His definition focused on distinctive morphological traits and either
geographic isolation or failure to intergrade when occurring in sympatry with other Argemone. The mor-
phology and geographic isolation of A. pleiacantha subsp. pinnatisecta fits this definition; however, Ownbey
subsequently recognized the taxon as a subspecies without any discussion of his reasoning. With the ex-
ception of Stunners’ (1958) overlooked shift in rank to A. pleiacantha Greene var. pinnatisecta (G.B.Ownbey)
Shinners, other taxonomic assessments have not been made since Ownbey’s description.
Stunners’ shift in rank to variety might easily be misconstrued as his viewing the taxon as a lesser entity
than Ownbey. Similar interpretations, commonly applied by non-scientists in arguments against protecting
subspecific taxa (e.g., Wilcove et al. 1993), demonstrate the danger of over-interpreting the differential use of
ranks (particularly subspecies and varieties) as well as the risk of excluding plant varieties from the formal
language in the Endangered Species Act (Wilcove et al. 1993). In actuality, Shinners’ taxonomic modification
simply reflected his opinion that “One only uses subspecies on the relatively uncommon occasion when it
is desired to label a group of varieties.” (Shinners 1958) and had nothing to do with isolation or uniqueness
of the group.
With the total number of established A. pleiacantha subsp. pinnatisecta dwindling well below 1000
(Sivinski 1999; Tonne 2008), the unclear rank appropriate for the taxon (species, variety, or subspecies), and
threats to its viability coming from aspects of reproductive biology (Sivinski 1992; Tonne 2008) as well a
variety of human-related factors (e.g., water withdrawal and right-of-way development, flooding, off-roadinfr
grazing, highway maintenance (Lightfoot and Sivinski 1994; Tonne 2008)), the taxonomic assignment of
these plants is of considerable interest.
The objective of this study is to evaluate the genetic structure and distinctiveness of Argemone pleiacantha
subsp. pinnasecta population systems through the application of an AFLP-based (Vos et al. 1995) molecular
assessment of population-level genomic variation and to use this information to address the taxonomic statu5
of these plants. AFLPs are randomly sampled genetic loci that, in combination with appropriate metho®
of analysis, have proven powerful in developing objective fine-scale assessments of population-level**
264
Journal of the Botanical Research Institute of Texas 4(1)
species-level variation in many plant and animal groups (e.g. Bacon & Bailey 2006; Baskauf & Burke 2009;
Duminil et al. 2006; Martinez- Ortega et al. 2004; Routtu et al. 2007). For this study, AFLP profiles from
A. pleiacantha subsp. pinnatisecta populations were compared to all geographically proximate species of Ar-
gemone. Results relating to the genetic isolation and differentiation of A. pleiacantha subsp. pinnatisecta are
discussed to assess whether the taxon is a species based on scientifically accepted concepts of the species,
or if these represent geographic, but not genetically distinct, populations of proximate species.
MATERIALS AND METHODS
Sampling. — Multiple collecting trips to Argemone localities in New Mexico were made between June and
August of 2007. Argemone pleiacantha subsp. pinnatisecta were collected from 12 localities representing four
canyon systems that run east to west across the Sacramento Mountains (Alamo, San Andres, Dog, and La
Luz/Fresnal Canyons - Fig. 1 and Appendix 1). Argemone pleiacantha subsp. pleiacantha was collected from
Kingston and Hillsboro, A. polyanthemos (Fedde) G.B.Ownbey from San Augustin Pass near Las Cruces, and
A. squarrosa Greene from the eastern Sacramento Mountains (Fig. 1 and Appendix 1). The latter three taxa
were sampled because they are the only other species of Argemone geographically proximate (within 150
miles) to A. pleiacantha subsp. pinnatisecta and because they are among a number of possible close relatives to
A. pleiacantha subsp. pinnatisecta identified in a phylogenetic analysis of Argemone (Schwarzbach & Kadereit
1999). With the exception of the endangered A. pleiacantha subsp. pinnatisecta, which is sufficiently repre-
sented by specimens in the NMSU Dept, of Biology Herbarium (NMC - Thiers 2009), a voucher specimen
was collected from each locality and deposited in NMC (Appendix 1).
DNA Extraction and AFLP Amplification.— DNA samples were extracted from each individual us-
ing the DNA extraction protocol of Alexander et al. (2007) eluting DNA into 10 mM Tris. DNA quantity
and quality were evaluated on 0.7% agarose gels with a 100 bp DNA mass ladder standard (New England
Biolabs). The restriction ligation (RL) and preselective amplifications followed a modified Vos et al. (1995)
AFLP approached used by Bacon and Bailey (2006) and marketed by Applied Biosystems (“Plant Mapping
Protocol” - P/N 402977 rev. E). In short, 50 ng of genomic DNA was digested overnight at 37 °C with IX T4
Ligase buffer (NEB), 0.046 M NaCl, 0.046 M BSA, 1 pM Msel Adapter pair, 10 pM EcoRI Adapter Pair, 1 U
Msel, 5 U EcoRI, and 67 U of T4 ligase (NEB). RLs were diluted to a final volume of 200 pL with 0.1 X TE.
Preselective and selective amplifications included 1.5 mM MgCh 0.1 M Tris-HCl pH 8.3, 0.5 M KC1, 0.25
pM of each primer, and ca. 2 U Taq in a 20 pL reaction containing 4 pL of dilute RL or preamplification
product. Preselective amplifications applied single selective bases on each primer (A on EcoRI and C on Msel)
and selective primer combinations included EcoRI-AC/Msel-CTA and EcoRI-TC/Msel-CTA. Preselective and
selective amplification cycling followed the ABI Plant Mapping Protocol. Selective amplification reactions,
labeled with 5’FAM on the EcoRI primer, were run on a 3100 sequencer (Applied Biosystems) with a ROX500
standard (Applied Biosystems).
Data Analysis. — AFLP profiles were extracted from raw sequence files and converted to comparative allele
presence/absence tables using GeneMapper 4.0 (Applied Biosystems). Alleles used in the analyses ranged
from 100-500 bp. Runs on single individuals were considered to have failed if the number of fragments
amplified was below the mean and standard deviation of fragments amplified across the population. In almost
all cases, these failed runs correlated with low quality DNA and generated few or no peaks.
Two approaches were implemented to assess the number of genetically distinct clusters of individuals
supported by the AFLP data irrespective of previously conceived notions of species or population limits. Fir*'
a principle coordinate analysis (PCO) employing Euclidean distances was run in MVSP ver. 3.131 (Kovach
Computing Services). The first two coordinates were plotted to display the degree of differentiation among
groups. This visual approach was augmented by the Bayesian statistical analysis presented by Pritchard et
(2000). The application of STRUCTURE vers. 2.3.1 (Pritchard et al. 2000) tested relative likelihood supp^
in the data for K genetic clusters (K=l-8) and the assignment of each individual to specific clusters under
each value of K. The scoring of AFLP patterns in STRUCTURE followed the recommendation of Evanno*
265
aL (2005) in treating the unobserved alleles as missing data (i.e., for an individual with a presence allele at
locus X, it is not possible to directly infer if the sample is homozygous dominant [1,11 or heterozygous [1,0J).
STRUCTURE analyses included 10,000 burn-ins and MCMC replicates for each run, 10 replicate runs for
each value of K, use of the admixture model, and allele frequencies set to independent, as recommended by
Evanno ttal. (2005). Other parameters were set to the software defaults. The inferred number of clusters
best supported by the data was further tested through the application of AK (K=l-8) as applied by Evanno
ttal. (2005).
RESULTS
Of the 93 individuals from which DNA was extracted, 63 were successfully amplified using both selective
primer combinations (30 A. pleiacantha subsp. pinnatisecta, 16 A. pleiacantha subsp. pleiacantha, 11 A. squar-
m, and 6 A. polyanthemos). Across these samples, AFLP reactions employing selective primer combinations
MI-AC/Msel-CTA and EcoRI-TC/Msel-CTA amplified 221 and 203 loci, respectively. Each primer combi-
nation amplified one fixed presence allele (“private”) in every accession of A. pleiacantha subsp. pinnatisecta
that was absent in all other sampled accessions.
Argemone squarrosa is considered polyploid based on chromosome counts from one locality (Ownbey
1958). When analyzed with diploid taxa in distance based analyses the increased number of fragments could
generate artifactual results of concern to this study. However, individuals from the locality of A. squarrosa
sampled for this study produced numbers of AFLP fragments well within the range for all other accessions
amplified, reducing concerns with variance in fragment number. It is possible that there are both diploid
and polyploid populations of the species and that we sampled diploid individuals.
Interspecific Analyses: PCO analysis (Fig. 2) identified three highly differentiated clusters. All accessions
of A. pleiacantha subsp. pinnatisecta were recovered in a single cluster and the taxon showed greater differ-
entiation from all other accessions than the other three taxa showed among one another. Argemone squarrosa
®d A. polyanthemos accessions displayed the lowest degree of differentiation between taxa (Fig. 2).
Results of the Ln P(D) in STRUCTURE supported Ks3 (Table 1). Pritchard et al. (2000) and Evanno
(2005) have documented likelihood values continuing to rise after passing the “true” K. This appears
» be the case here, were the likelihood of K=3 is considerably higher than the values for K=1 or 2, but K
continues to rise slightly and fluctuate above K=3. The application of Evanno et al.'s (2005) method found a
10 fold reduction in AK (from 15.95 to 1.61) between K=3 and K=4 (Table 1), identifying strong support for
aK=3 conclusion (Evanno et al. 2005).
In all replications of the MCMC method with K>1, representatives of A. pleiacantha subsp. pinnatisecta
*®^ssigned to a single unique cluster, adding credence to the conclusion that these individuals represent
^inct group. With all values of K>1, no individual of A. pleiacantha subsp. pinnatisecta had less
®*®963% assignment to the same single cluster and no individuals from the other taxa show greater than
• % assignment to the A. pleiacantha subsp. pinnatisecta cluster (Table 1).
!nl|aSPeCifiC Different'ation: The sampling available for this study largely precluded a comprehensive
^ of intraspecific variation for A. pleiacantha subsp. pinnatisecta. Nonetheless, some preliminary analyses
tun to test for signs of potential population differentiation. A PCO analysis restricted to A. pleiacantha
JP Pinnatisecta accessions displayed weak differentiation (Fig. 3) between accessions from canyons in
difb°nhern 3nd southern portions of the range of the subspecies (see Discussion). The weakness of this
entiation is clear from the results in STRUCTURE, which failed to reject K=1 (Table 2) for the taxon,
^cannot be tested for K=1 and was not applied.
T DISCUSSION
Wa7c Status of the Sacramento Prickly Poppy.— The presence/absence pattern of randomly se-
al^ LP loci amplified from A. pleiacantha subsp. pinnatisecta and populations of other Argemone known
teas°nable proximity of A. pleiacantha subsp. pinnatisecta, are consistent with the Sacramento Prickly
PCO case scores
; - number of distinct groups applied, Ln - log probability, AK - a
mean Ln D(K)
SDLnD(K)
5PP CLUSTER -of min.
-12945.75 -10190.29 -8810.68
1.40 67.06 81.46
NA 2053 15.95
NA 0.963 0.985
-8748.87 -8439.63
1.61 1.42
0.985 0.99
-8534.8 -8504.54
353.11 199.96
0.974 0.982
SPP individual
SPP CLUSTER -max.
assignment of any
non-SPP individual
0.062
0.012 0.005
Poppy being genetically cohesive and reproductively isolated. In particular, the identification of two fixed
(private) allelic differences present in all sampled A. pleiacantha subsp. pinnatisecta, which are absent from
the other samples, suggests that the taxon has been reproductively isolated from these other Argemone for
sufficient time to have: 1) developed unique fixed genetic traits not found in other taxa, or 2) that other
geographically and phylogenetically proximate Argemone have lost traits that were once common to more
species. Furthermore, the fixed allelic differences at two of the 424 loci are not the only differentiating sign*1
loci found in the genetic dataset. Isolation at the level of species is further identified by the sum of alW*
frequencies investigated through PCO (Fig. 2) and STRUCTURE (Table 1).
Fixed allelic differences are the explicit delimiting factor in the “phylogenetic species concept” (Davis
& Nixon 1992; Nixon & Wheeler 1990), are consistent with the principles of the more widely known “bio-
logical species concept” (Dobzhansky 1935; Mayr 1942), and the overall pattern corroborates Ownbey’sfpg-
9, 1958) suggestion that that the most important form of speciation in Argemone is “. . . geographic isolation
leading to the accumulation of genetic differences in isolated populations.” Thus, for the Sacramento Pricfoy
Poppy the combination of genetics, morphology, and geography identify extensive reproductive isolation
Cervantes et al.. Genetic analysis of Argemone pleiacantha subsp. pinnatisecta
PCOcase scores
~^T
ALa Luz/Fresnal Canyons
▼Dog Canyon
^Andres Canyon
♦Alamo Canyon
'•’eanlnD(K) -3161.3 -3166.8 -3256.5 -3382.1 -3389.5 -3299.1 -3689.3 -3253.6
0.81 2.47 281.40 521.27 704.81 510.24 848.12 340.44
from other proximate Argemone as well as continued intraspecific contact. These patterns are consistent with
SPeci^c s^tus under widely accepted concepts of eukaryotic sexual species.
Intraspedfic Differentiation. — Analyses of the available A. pleiacantha subsp. pinnatisecta data set did not
^K=l for the taxon. However, the PCO analysis of A. pleiacantha subsp. pinnatisecta (Fig 3) provides evi-
for weak divergence of populations found in different canyon systems. The Fresnal and La Luz canyon
^essions tend to cluster in one cohort and the three remaining canyon systems into another. Clearly there
“strong population structure on the level of species differentiation, but these preliminary results are of
cance to future management practices. Most importantly, resource managers should not assume that
variation in the Sacramento Prickly Poppy is randomly distributed across the species’ range.
and Future Research.— Ownbey’s (1958) reasoning for having described the Sacramento
^ y Poppy at the subspecific rank contradicts both his own description of what constitutes a species
isolat h 8enetiC data generated from this study- Argemone pleiacantha subsp. pinnatisecta is geographically
conce ’ m°rphologlcall>r distinct, and genetically unique. These features are consistent with contemporary
sPecies and are applied here as the scientific evidence behind an elevation in taxonomic
ks^LStUdy incorP°rated taxa growing within 150 miles of Argemone pleiacantha subsp. pinnatisecta in
^ ^wester" “sky island” system. This distance was selected based on an evaluation of taxa that could
1V3b y interbreed with A. pleiacantha subsp. pinnatisecta and what was feasible for the study. Future
assessments of species limits within Argemone will hopefully address the limits of all species in the group. Of
potential relevance to the understanding of A. pleiacantha subsp. pinnatisecta may be the comparison of three
morphologically similar but geographically disjunct taxa (A. arizonica G.B. Ownbey, A. gracilenta Greene,
and A. pleiacantha subsp. ambigua G.B. Ownbey). Since these taxa are only known from localities at least 300
miles distant, across inhospitable habitat, recent historic or contemporary interbreeding with A. pleiacantha
subsp. pinnatisecta is largely inconceivable, but more inclusive comparative studies may shed light on the
origin of A. pleiacantha subsp. pinnatisecta and aspects of the phylogeographic history of Argemone. Further
analyses of infraspecific genetic variability and geographic differentiation incorporating greater population
sampling for A. pleiacantha subsp. pinnatisecta should also be carried out to develop an appropriate fine scale
TAXONOMIC TREATMENT
Argemone pinnatisecta (G.B. Ownbey) S.D. Cervantes & C.D. Bailey, comb, et stat. nov. Argemone pleiacantha
Ownbey) Shinners, Southw. Naturalist 3:213—214. 1958. Type: U.S.A. New Mexico: Otero Co.: 9.6 mi W of Cloudcroft, 6600 ft, U
APPENDIX 1
Sampling Information. For each taxon different collecting localities include location, centroid of population latitude, centroid
of population longitude, collector and number, date, and number of samples taken.
Argemone pleiacantha subsp. pinnatisecta— Sacramento Mtns. Fresnal Canyon, 32.9698,-105.9010, Phil Tonne & Bob Siv-
inski, 2 Samples. Sacramento Mtns. Dog Bench, 32.7550, -105.8872, Phil Tonne & Bob Sivinski, 8 Samples. Sacramento Mtns.
Fresnal Canyon, 32.9548, -105.8748, 14 Aug 2007, Phil Tonne & Bob Sivinski, 2 Samples. Sacramento Mtns. Fresnal Canyon,
32.9665, -105.8978, 14 Aug 2007, Phil Tonne & Bob Sivinski s.n„ 1 sample. Sacramento Mtns. Dog Bajada, 32.7503, -105.9191,
14 Aug 2007, Phil Tonne & Bob Sivinski s.n., 5 samples. Sacramento Mtns. La Luz, 32.9817, -105.9257, 14 Aug 2007, Phil Tonne
& Bob Sivinski s.n„ 1 Sample. Sacramento Mtns. Fresnal Canyon, 32.9480, -1 05.8428, 1 4 Aug 2007, Phil Tonne & Bob Sivinski sn,
4 Samples. Sacramento Mtns. La Luz, 32.981 7, -105.9257, 14 Aug 2007, Phil Tonne & Bob Sivinski s.n„ 2 Samples. Sacramento
Mtns. San Andres Canyon, 32.7826, -1 05.901 3, 1 5 Aug 2007, Phil Tonne & Bob Sivinski s,n„ 2 Samples. Sacramento Mtns. San
Andres Canyon, 32.7826, -105.901 3, 1 5 Aug 2007, Phil Tonne & Bob Sivinski s.n., 1 Sample. Sacramento Mtns. Fresnal Canyon,
32.9476, -1 05.8553, 1 4 Aug 2007, Phil Tonne & Bob Sivinski s.a, 4 Samples. Sacramento Mtns. Upper Alamo Canyon, 32.8537,
-1 05.8348, 1 4 Aug 2007, Phil Tonne&Bob Sivinski s.n., 8 Samples. A. pleiacantha subsp. pleiacantha— Hillsboro, NM, 32.8249,
-107.5421, 19 Jun 2007, Sandy Cervantes 2, 14 Samples. Kingston, NM, 32.9187, -107.6873, 19 Jun 2007, Sandy Cervantes U?
samples A. polyanthemos— Organ Mtns, leaf samples taken from plants along 110 -collected from 0.5 mi W of White Sands
Missile Range entrance, 32.4381,-106.4866, 26 Jun 2007, Sandy Cervantes 3, 1 3 samples. A. squarrosa - Lincoln, NM 33.5340,
-1 05.4964, 27 Jun 2007, Sandy Cervantes 4, 1 3 Samples.
ACKNOWLEDGMENTS
This research was completed in partial fulfillment of undergraduate honors thesis requirements (S.D. Cer-
vantes) at New Mexico State University. We are grateful to Robert Sivinski for collecting leaf material under
permit TE820730-0 and Fabian Michelangeli for his assistance with the Spanish abstract. We also thank Julie
McIntyre, Brook Milligan, Tom Parchman, Richard Spellenberg, Lillis Urban, and an anonymous reviewer
for helpful discussions on the topic and/or manuscript. Funding for this project came from the Howa
Hughes Medical Institute funded NMSU Undergraduate Scholars Program and NSF grants DEB0817033&
EF0542228 (to CDB).
REFERENCES
Alexander RJ, G. Rajanikanth, C. Bacon, and CD. Bailey. 2007. Rapid inexpensive recovery of high quality plant DNA
using a reciprocating saw and silica-based columns. Molec. Ecol. Notes 7:5—9.
Bacon, C.D. At® CD. Bailey. 2006. Taxonomy and conservation: a case study from Chamaedorea altemans. AM-
Bot. 98:775-763.
Baskauf, CJ. and J.M. Burke. 2009. Population genetics of Astragalus bibullatus (Fabaceae) using AFLPs. J. Heredity
100:424-431.
270
Journal of the Botanical Research Institute of Texas 4(1)
BOOK REVIEW
Henry Connor and John Fountain. 2009. Plants That Poison: A New Zealand guide. (ISBN 978-0-478-
09398-8, pbk.). Manaaki Whenua Press, PO. Box 40, Lincoln 7640, Gerald Street, Lincoln, Canterbury,
New Zealand. (Orders: www.mwpress.co.nz or www.balogh.com (US orders), +64 3 321 9749, +64 3
321 9997 fax, 1-217 355 9413 US fax). $NZ 29.99, $US 45.00, 112 pp., color photos, 7 1/2" x 10*.
i- tat Res. Inst Tens 4(1): 270. 2010
OCCURRENCE OF ANISOPHYLLY AND ANISOCLADY
WITHIN THE AMARANTHACEAE
Lynn G. Clark
272
Journal of the Botanical Research Institute of Texas4(1)
Amaranthoideae
Alternate/Opposite
Alternate/Opposite
Gomphrenoideae 2
Pseudoplantageae 2
Gomphreneae 2
Alternate/Opposite
Opposite
Opposite
Opposite
and Amaranthinae in the Amaranthoideae whereas opposite, decussately arranged leaves are diagnostic
for Gomphrenoideae (Townsend 1993). The character is polymorphic within Aervinae, occasionally being
polymorphic within a population of a single species (Townsend 1993).
While collecting Amaranthaceae in Texas for molecular phylogenetic research on the Chenopodiaceae-
Amaranthaceae alliance, it was noted that several genera of Gomphrenoideae (e.g., Guilleminea Kunth,
Tidestromia Standi., and Alternanthera Forssk.) appeared to have alternate leaves, an unexpected condition
within the subfamily. Further examination of the collected specimens revealed that the taxa in question
have opposite leaves, but that the taxa superficially appear alternate-leaved due to a combination of aniso-
phylly and anisoclady. A morphological survey of the family was conducted to determine the prevalence,
phylogenetic distribution, and potential significance of these characters within the Amaranthaceae.
MATERIALS AND METHODS
Taxon Sampling. — Herbarium specimens from 24 genera and 52 species of the Amaranthaceae, includ-
ing members of both subfamilies and all tribes except the monotypic Pseudoplantageae (Gomphrenoideae),
were surveyed. Both subtribes of Amarantheae were also sampled. Multiple species were sampled fro®
Charpentiera Gaudich and the large genera Alternanthera , Amaranihus L., Gomphrena L., I resine P- Browne,
and Pfa/fia C. Mart. (Table 2).
Character Sampling. — Phyllotaxy, equality of leaf size between a pair of leaves at a single node, an
branching pattern were examined directly from herbarium specimens using a dissecting scope and a m '
fied Pohl’s solution (Pratt & Clark 2001) where necessary as follows (Table 3).
• Phyllotaxy was examined from all taxa and recorded as alternate or opposite.
• Equality of leaf size at a single node is inapplicable to alternate-leafed taxa, and was examine d
opposite-leafed taxa. Leaf equality was recorded as isophyllous when both leaves of a pair at an
were of equal size, or as anisophyllous when one leaf of the pair was larger and better developed
the second.
Leaf equality was quantified by measuring the lengths of leaf pairs from three nodes and calc“^
the size ratio using the average leaf sizes in opposite-leaved taxa using herbarium specimens (
ISC, and MO) or digitized computer images (GH, K, and NY). Measurements on digitized images
made only when leaf pairs were unambiguous, a situation that was difficult to measure on
lous taxa. The length ratios were calculated (Table 4) and analyzed using t-Tests assuming bothequ* *
and unequal variances with JMP 8.0.1 statistical software (SAS 2002).
• Branching pattern was observed from all taxa and recorded as either isocladic or anisocladic I
opposite leaves, anisocladic branching was recorded for those taxa in which only one bud at a n0<"®^
oped into a branch. Isocladic branching was recorded when both buds at a node developed into b .
tail Specimens examined. ASTC= Stephen F. Austin State University Herbarium. BPM= Borsch, Pratt, and Muller, GH= Gray Herbarium, ISC= Iowa State Ada Hayden
Herbarium, K= Kew Botanical Garden, M0= Missouri Botanical Garden Herbarium, NY= New York Botanical Garden. Numbers in parentheses indicate number of genera/
species per indicated taxon (Townsend 1993)
Amaranthaceae (69/780)
Amaranthinae (12/92)
Amaranthus blitoides S. Wats.
Qtamissoaaltissima (Jacq.) Ki
Charpentiera obovata Gaud.
Charpentiera ovata Gaud.
Aervinae (38/241)
Achyranthes bidentata Blume
Aervajavonica (Burm. F.) Juss.
Calicoremacapitata (Moq.) Hex
Pandiaka heudelotii (Moq.) Benth. & Hook.
Ptilotus obovatus (Gaudich.) F. Muell.
Pupalia lappacea (L.) Juss.
£e!osieaei5/761
Cehsiaargentea L.
feeringia polysperma (Roxb.) Moq.
Hermbstaedtia glauca Moq.
Heuropetalum sprucei (Hook. F.) Standley
Amarantheae (50/333)
Ames, Iowa, Pratt 200 (ISC)
Ames, Iowa, Pratt 199 (ISC)
Canary Islands, Kunkel 12484 (MO);
Canary Islands, Bramwell 1326 (MO)
Bolivia, Nee 40597 (ISC)
Oahu, Hawaii, Perlman & Lau6125 (MO)
Maui, Hawaii, Sohmer6594 (MO)
Ames, Iowa, Pratt 201 (ISO
Pakistan, Ajab&Ashraf 1254 (MO)
South West Africa, Giess, Volk, & Bleissner 6206 (MO)
Oahu, Hawaii, Degeners.n. (ISC)
Burundi, Lambinon 78/84 (MO)
Australia, Conn 2285 (MO)
Ghana, Schmidt, Amponsah, & Welsing 1881 (MO)
Ames, Iowa, Pratt 222 (ISQ
Taiwan, Shu-Hui Wu 1 153 (MO)
South Africa, Esterhuysen 240 (MO)
Costa Rica, Jimdnez & Soto 98 1 (MO);
Costa Rica, Haber &Zuchowski 9397 (MO)
Argentina, s.c s.n. w
Arequipa, Peru, Pennell 13131 (NY)
Nacogdoches, Texas, Banks 2046 (ASTC)
Brazil, Tsugaru & Sano B-223 (NY)
Alpine, Texas, BPM3433 (ISC)
Puerto Rico, Luquilb, Liogier&Liogier 31898 (NY)
Asuncion, Paraguay, Morong 40 (NY)
Chamber Co, Texas, Jones 1623 (ASTC)
Montgomery Co, Texas, Raines 258 (ASTC)
Carlsbad, New Mexico, BPM 3449 (ISC)
Hamilton, Texas, Stanford 1337 (ASTC)
Concordia Parish, Louisiana, Thomas, Martin,
Scarborough, & Slaughter 106,565 (ASTC)
St. John, Virgin Islands, Acevedo-Rodriguezet al. 2913 (NY)
South Padre Island, Texas, BPM 3444 (ISC)
Mcintosh Co., Georgia, Duncan 20458 (ISQ
Colombia, Dawe 527 (K)
Mato Grosso do Sul, Brazil, Lindman A2497 { NY)
Ames, Iowa, Pratt 228 (ISQ
Ixiamus, Bolivia, Cardenas 191 1 (NY)
Minas Gerais, Brazil, Pirani etal. CRCR8686 (NY)
St. Thomas University, Virgin Islands,
Acevedo-Rodriguez 1 1372 (NY)
Paraguay, Hassler7491 (NY)
mill imiii mi iiiililiilliliiiiiiiiilliiiiiii
276 Journal of the Botanical Research Institute ofTexas4(1)
Twiif 4. leaf length ratios of equal and unequal leaf pairs in Gomphrenoideae.
Alternanthera bettzickiana
Alternanthera brasiliana
Alternanthera ficoidea*
Alternanthera philoxeroides
Alternanthera sessilis
Gomphrena albi flora
Gomphrena globosa
Gomphrena lutea
Gomphrena pangens
Gomphrena serrata
Iresineangustifolia
Iresineargentata
Iresine diffusa
Gossypianthus lanuginosa
Guillemineadensa
Tidestromia lanuginosa
Average Leaf Ratio
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Isophyllous
Anisophyllous
Anisophyllous
Anisophyllous
Anisophyllous
Anisophyllous
•Average was calculated from only two leaf pairs
0.99:
0.98:
0.93:
0.93:
1.00:
0.90:
0.98:
0.92:
0.93:
0.97:
0.95
0.93
0.89
0.97
0.96:
0.86:
0.90:
0.97:
used in taxonomic treatments, despite the fact that they may be of some taxonomic utility. The unreP°rt^
yet widespread presence of anisophylly and anisoclady within the Amaranthaceae was somewhat surp
ing, but underscores the great need for critical morphological examination of the Amaranthaceae as we
for its sister family the Chenopodiaceae. ^
Molecular phylogenetic analyses of the Chenopodiaceae-Amaranthaceae alliance (Pratt 2003; Ka
2003; Muller & Borsch 2005) have recovered a strongly supported monophyletic Gomphrenoideae ^
relationships of the Gomphrenoideae (opposite leaves) with opposite-leaved taxa of the Aervineae are
certain. Strict consensus places many of the Aervineae as sister to Gomphrenoideae, although rela”° s
with the Aervineae-Gomphrenoideae clade are currently unresolved (Pratt 2003; Kadereit 2003; MU^
Borsch 2005; Sage et al. 2007). Anisophylly and anisoclady were restricted to the Gomphrenoideae i®
survey of the Amaranthaceae. Because the characters have been previously unreported, their presence ^
a taxon cannot be ruled out based on prior literature.
adaptive significance of anisophylly and anisoclady within the Gomphrenoideae must also take into account
the correlation of these characters with photosynthetic system.
ACKNOWLEDGMENTS
We thank an anonymous reviewer who provided comments that greatly improved our paper. The authors
gratefully acknowledge the Iowa State University Ada Hayden Herbarium (JSC), Herbarium of Kew Gardens
(K), Herbarium of the Missouri Botanical Gardens (MO), Gray Herbarium (GH), Herbarium of the New York
Botanical Garden (NY), and the Herbarium of the Stephen F. Austin State University Biology Department
(ASTC). We also thank our illustrator Ms. Sarah Kudron, and Anna Gardner and Dr. Neal Cox for technical
help with the illustrations.
REFERENCES
Angiosperm Phylogeny Group. 2003. An update of the Angiosperm Phylogeny Group classification for the orders
and families of flowering plants: APG II. Botanical Journal of the Linnean Society 141: 399-436.
CiInoud, P., V. Savolainen, L.W. Chatrou, M. Powell RJ. Grayer and M. W. Chase 2002. Molecular phylogenetics of
Caryophyllales based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences. Amer. J. Bot.
Dengleh N. 1999. Anisophylly and dorsiventral shoot symmetry. Int. J. PI. Sci. 160:S67-S80.
DtWNiE, S.R., D.S. Katz-Downie, and K J. Cho. 1 997. Relationships in the Caryophyllales as suggested by phylogenetic
analysis of partial chloroplast ORF 2280 homolog sequences. Amer. J. Bot. 84:236-253.
Hbwggon, J. 1987. A taxonomic reevaluation of Gossypianthus and Guilleminea (Amaranthaceae). Sida 12:
307-337.
toxAtrr, G,T. Borsh, K. Weising, and H. Fretac. 2003. Phylogeny of the Amaranthaceae and Chenopodiaceae and
the evolution of C4 photosynthesis. Int. J. PI. Sci. 164:959-986.
K&hn-U. 1993. Chenopodiaceae. In: K. Kubitzki ed.The families and genera of vascular plants. Springer-Verlag,
Berlin, Germany. Pp. 253-281.
^J-R. and J.H. Rettig. 1 994. Gene sequence data. In: Behnke, H.D. andTJ. Mabry eds. Caryophyllales evolu-
tion and systematics. Springer-Verlag, Berlin, Germany. Pp. 235-246.
Mok*n. M. and N. Dengler 1 988. Vascular architecture in isophyllous and facultatively anisophyllous species of
Pentadenia (Gesneriaceae). Amer. J. Bot. 75:1485-1494.
M(tHer, K. and T. Borsch. 2005. Phylogenetics of Amaranthaceae based on matK/trnK sequence data— Evidence
from parsimony, likelihood, and Bayesian analyses. Ann. Missouri Bot. Gard. 92:66-102.
Psatt* d-b- and LG. Clark. 2001 . Amaranthus rudis and A. tuberculatus— One species or two? J. Torrey Bot. Soc.
128282-296.
e.Ph.D.
Dissertation, Iowa State University, A„,CJ.
***»<- K.R. and S. Clemants. 2003. Amaranthaceae. In Flora of North America Editorial Committee eds. Flora
ol North America vol. 4 Magnoliophyta- Caryophyllales , part 1, Oxford University Press, New York City, USA.
Rp* 405—456.
S^F JL Sage R.W. Pearcy, and! Borsch. 2007. The taxonomic distribution of Ca photosynthesis in Amaran-
maceae sesu stricto. Amer. J. Bot. 94:1 992-2003.
^-fiwsos, A. and N. Dengler 1 988. Leaf development in isophyllous and facultatively anisophyllous spec.es
(Gesneriaceae). Amer. J. Bot. 75:1472-1484.
Inc. JMP version 8.0.1, SAS Institute Inc, Cary North Carolina, USA.
T;H- 1934. Amaranthaceae. In: Engler, A. and K. Prantl, eds. Die Naturlichen Pflanzenfamlien 16c, 2nd ed.
^J^mann, Leipzig, Germany. Pp. 7-85.
GC. 1 993. Amaranthaceae. In: K. Kubitzki ed.The families and genera of vascular plants. Spnnger-Verlag,
^Germany. Pp. 253-281.
280
BOOK REVIEW
John Alcock. 2009. When the Rains Come: A Naturalist’s Year in the Sonoran Desert. (ISBN 978-0-
8165-2762-5, pbk.). University of Arizona Press, 355 S. Euclid Ave. , Suite 103, Tucson, Arizona 85719,
USA. (Orders: www.uapress.arizona.edu, 1-800-426-3797). $21.95, 334 pp., Ill color photographs.
8 1/2" x 9 1/2".
OBSERVATIONS ON BUCKLEYA (THESIACEAE) IN CHINA
Harvard University Herbaria
22 Divinity Avenue
Cambridge, Massachusetts 2 138-2020, USA
Qiu-Yun (Jenny) Xiang
Botany Department
North Carolina State University
Campus Box 761 2
Raleigh, North Carolina 27695-76 12, USA
Lab ofSystematics and Evolutionary Botany and Biodiversity
College of Life Sciences
Zhejiang University
Hangzhou 310058, Zhejiang, CHINA
Lab ofSystematics and Evolutionary Botany and Biodiversity
College ofUfe Sciences
Zhejiang University
Hangzhou 3 10058, Zhejiang, CHINA
CHINESE ABSTRACT
tt : 6 1908^BucWeya henryi Diels£*tt3fc , lanceolata (Siebold & Zucc.) Miquel-
( B. angulosa S B. Zhou &
n5B*#mB.lanceolatam&o*XmmB£#m#mft7ii&,
xm® : *®mm ; mm ; %% ; mx* ;
^nns 3 class field trip in 2008 with students from Zhejiang University, Hangzhou, China, and North
Molina State University, Raleigh, North Carolina, U.S.A., we visited a number of sites in southern Anhui
** m ZheJiang provinces. One of the most floristically interesting areas was Qiyun Shan, in Xiuning Xian,
Anhm Province. Among the rich assemblage of plants on Qiyun Shan, was a most interesting shrub of the
Pj5 Buddeya T orr. (Thesiaceae. Although Buckleya has been placed traditionally in the Santalaceae, it will
. pUced in Thesiaceae by D. Nickrent (manuscript submitted) in the forthcoming treatment of the genus
1111 Floraof North America; also Nickrent et al. in press).
upon returning to the United States, the collection of Buckleya from Qiyun Shan was compared with
T*cunens of B iancedata (Siebold & Zucc ) Miquel (Fig. 1) from Japan and with specimens of B. henryi
k from China, since it did not fit descriptions in the monograph by Carvell and Eshbaugh (1982) or any
J* $pecies of Buckleya in the Flora of China (Xia & Gilbert 2003). The plants from Qiyun Shan resembled
^Panese B. lanceolata in the longitudinal grooves of the mature fruit and in the broader and somewhat
^-acuminate leaves, but they resembled B. henryi in having terminal and axillary inflorescences. Buckleya
^ froni J3Pan has only terminal inflorescences, but B. henryi from China sometimes produces both
“ ^ terminal inflorescences (Carvell & Eshbaugh 1982). The trichomes of B. lanceolata are reported
to be both uniseriate and multicellular while B. henryi has minute conical and/or papillate trichomes (“hairs”)
on the vegetative parts (Carvell and Eshbaugh, 1982). Some specimens of B. lanceolata from Japan, however,
do not have the long trichomes, for example, H. Ohashi & Y. Ueno 8557 (A) from along the Takase River in
Namie-machi, Futaba-gun in Fukushima Prefecture. In fact, the Ohashi & Ueno specimen, collected north
of Tokyo, closely resembles our Anhui collections vegetatively. In many ways the plants of Buckleya from
Anhui appear to be hybrids between the Japanese B. lanceolata and the Chinese B. henryi. After making these
observations we became aware that plants from Tianzhu Shan, Anhui, and matching the characteristics of
our find on Qiyun Shan, were described and named B. angulosa by S.B. Zhou & X.H. Guo (2004).
Housed in the Harvard University Herbaria (A, GH) are other specimens from southern Anhui that
resemble the plants we saw in the field. Specimens of Buckleya from farther north in Henan and from the
Qinling (and Funyu) Mountains, however, have more slender, elliptic leaves and the fruiting sepals are
smaller, but at a quick glance and without careful measurements, both characters appear to be variable.
The plants from Anhui have leaves that are broader toward the base, or at least below the middle, and the
*pex is acuminate-caudate, but not as obviously acuminate-caudate as in B. lanceolata from Japan. Most
plants of B. lanceolata have uniseriate, multicellular trichomes on the vegetative parts, particularly near the
base on the lower surface of the leaves, plus minute conical projections (trichomes). Buckleya henryi lacks
multicellular trichomes, but has the minute conical projections. Although the plants from Anhui resemble
the Chinese B. henryi in pubescence, several specimens of B. lanceolata from Japan also have similarly shaped
Itaves and similar pubescence. Pubescence and leaf shape are therefore not 100 percent reliable features for
distinguishing the species of Buckleya in Asia.
The illustration of Buckleya angulosa provided by Zhou and Guo (2004) matches our collections (Fig. 2) I
wtU in showing the axillary inflorescences and elliptic sepals on the fruit. Although they provide a diagnostic
to separate the Chinese species of Buckleya, B. henryi is neither mentioned in their key to the species
m China nor in the narrative of their paper. The narrowly elliptic leaves in their illustration of B. angulosa,
•'wever, resemble more closely those of B. henryi than they do those of our collection from Qiyun Shan.
°w specimens also lack the peculiar apical projection on the leaves, shown in the illustration of Zhou and
Guo’ and which appear sporadically on all Asian species of Buckleya.
Buckleya henryi may not have been considered to be a distinct species by Zhou and Guo (2004), since it
w« placed in synonymy under B. lanceolata by Tam (1988) in the treatment of the genus in Flora Reipublicae
*¥daris Sinicae volume 24. Xia and Gilbert (2003), however, who also recognize two species of Buckleya in
B graebneriana Diels and B. henryi, noted in their treatment in the Flora of China that B. henryi has
h*® confused with the Japanese B. lanceolata.
j ln examining specimens of Buckleya from Japan and China, one cannot fail to speculate on how the
| might have been treated had they all occurred in China, or all in Japan. Would separate species have
j been ^cognized, or even distinct taxa at an infraspecific rank? The characteristics used to separate them are
** fway$ clear and considerable plasticity can be seen in leaf shape, pubescence and length of the sepals
J^efruit. Although the differences in most individuals of B. henryi and B. lanceolata can easily distinguish
^-careful study is needed to determine if clear, discontinuous distinctions occur in the individuals that
1 **yy combine characteristics of the two species. Might these seemingly intermediate plants be examples
^hybridization where the offspring of crosses between species (B. henryi xB. lanceolata = B. angulosa?)
r®1 together sympatrically during the Pleistocene have maintained themselv
^extinction of one of the parents there?
Carvell and Eshbaugh (1982) speculated that plants with both axillary am
y lnceslral in Buckleya, with terminal inflorescences being derived. Of the four species recognized by
. ' *ey considered B. henryi to be most similar to the ancestral form, with B. lanceolata being a direct
SS* The molecular analysis by Li et al. (2001) supports the relationships hypothesized by Carvell
Z ; T8*1- Now, with the recognition of B. angulosa and the complexities it adds to the understanding of
^ b«t biogeographically interesting and important genus, a comprehensive review of the taxonomy
unland after
a Anhui Yiiininn Yian Hivim Qhan f) F Rnufford. CX. fU> 0-
Zhao 405 15 (A).
COMMENTS ON A REVISION OF CELTIS SUBGENUS M ERTENS1A
(CELTIDACEAE) AND THE RECOGNITION OF CELTIS PALLIDA
James Henrickson
Mexico, trans-Pecos and south Texas) into central Mexico (Baja California Sur, Sonora, Sinaloa, Chihuahua,
Coahuila, Durango, Zacatecas, San Luis Potosf , Nuevo Le6n, Tamaulipas, Queretero, Hidalgo, Mexico, Ve-
racruz, Puebla, Oaxaca, and Chiapas), also in Lee County, Florida, and in South America, in north-central
Argentina and adjacent Paraguay (Hunziker & Dottori 1976).
In their revision of Celtis subgenus Mertensia, Berg and Dahlberg (2001) recognized six species, four
restricted to dry or deciduous gallery forests in South America and two occurring in both North and South
America. They recognized a very broad, variable and wide-ranging C. iguanaea that includes lianas, shrubs
and small trees with leaves varying greatly in size, marginal toothing and vestiture. They list 26 heterotypic
synonyms for the species, but they recognize four peripheral variants as separate species based on differ-
ences in style branches (C. laxensis C.C. Berg), vestiture [C. brasiliensis (Gardner) Planch., C. chichape (Wedd.)
Miq. in Mart.], and young-stem coloration (C. orthocanthos Planch.) as well as thorn structure etc. (Berg&
Dahlberg 2001).
Berg and Dahlberg (2001) placed the more xeric shrubs with smaller leaves into Celtis ehrenbergim
(epithet named in 1847) listing 13 heterotypic synonyms of the species, including the South American C. tab
(1848) and the amphitropically distributed C. pallida (1858). It is in this that I disagree with the authors.
CELTIS PALLIDA VS. CELTIS IGUANAEA
Celtis pallida is very different from C. iguanaea, differing in growth habit, the structure of the supranodal
thorns, the distribution of domatia on the lower (abaxial) leaf surface, size and structure of the leaves, and
the size of flowering styles, mature fruit and pyrenes.
Celtis pallida is an erect, much-branched shrub with notably zig-zagged stems; stem nodes usually
produce a pair of straight, sharp-tipped, divergent, lignified thorns (l-)5-20(-55) mm long directly above
the nodes. Thorns over 5(-10) mm long bear 1-3 nodes and the thorn nodes typically develop single or
fasciculate leaves or even lateral branches. When the thorns are very long they can have secondary thorns
above their axillary leaves and thus may mature into branches. But very short thorns can lack nodes, and
thorns can sometimes be solitary above the nodes, and some branches can lack throns altogether. The abaxial
domatia are usually restricted to the lower midrib axils with the first pair of secondary veins, and the domatia
may become swollen and green in color, and they also sometimes occur in the mid-leaf portion — there are
none in the axils of the secondary to tertiary vein axils (except in occasionally very large leaves). Leaves
range from entire to crenate-serrate, 10-25(-43) mm in length, 6.5-14(-33) mm in width. Inflorescences
are small, axillary dichasial cymes consisting mostly of staminate flowers, with only 1-2 bisexual flowers.
The two style bases are flattened, ±1 mm long, to 0.7 mm wide, each producing two equally short, papillate
divergent lobes at their tips with the papillae extending down the margins of the broad style bases. Mature
fruits are orange, usually 6-7(-8) mm in diameter when dried and have a rounded, white pyrene to 4 nun
long with a low, alveolate-reticulate sculpturing.
In contrast, Celtis iguanaea can be a tree or shrub with spreading liana-like branches or a liana *
typically produces a solitary, sharp-tipped, decurved, woody thorn 2-8(-10) nun long above each node,
these are rarely paired, rarely to 25 mm long and when large may rarely bear a leaf and secondary decurv
thorn at the tip. Inflorescences often develop at the base of the thorn or from the stem below the thorn base
and are considerably larger than those in C. pallida. Domatia are often poorly developed at the base o t ^
midrib at the axils of the basal secondary veins, but they occur well up the midrib and also along the pa1
secondary veins in the outer axils with the tertiary veins (this however, may be a feature of leaf size),
in tropical areas are often oblong, to 100-150 mm long, 45-60+ mm wide, pinnately veined above t
base, with caudate tips, cuneate to subcordate bases, entire or crenate-serrate toothed margins, but indntf
areas of Mexico they are more ovate to oblong-ovate, (15-)33-100 mm long, (10-)20-50(-85)
and can be well within the size range of those of C. pallida. Leaves can also become thick, coarse, strong
scabrous, and have fewer domatia in arid regions. Styles are similar in structure to those of C. pallida,
to 3-4.5 mm in total length. Mature fruit are orange to orange yellow, edible, 7-12 mm in diameter.
Henri ckson, Comments on a revision of Celtis subg. f
when mature the rounded pyrene is white, 5-7 mm long, with irregularly raised margins around the coarse
alveolate reticulations.
THE TYPE OF CELTIS EHRENBERGIANA
Klotzsch (1847) cited a single collection by C. Ehrenberg in his description of Momisia ehrenbergiana Klotzsch,
thebasionym of Celtis ehrenbergiana ( C . Ehrenberg 1114). The type collection was borrowed from the Berlin
Herbarium (B) (Fig. 1). It has Ehrenberg’s original label “Arbor seu frutex, ubique in loc. calid. las ajuntas
(sic), Montezuma, Jan. 1840, Mexico, C. Ehrenberg 1114.” (=Tree or shrub, common in warm sites. . .)
The las ajuntas noted is actually Los Adjuntos, a small mining settlement along the Rio Montezuma that
separates Querdtaro from Hidalgo near a small contemporary village of Maconi, Querdtaro, northwest of
Zimapan, Hidalgo. The locale is shown on the 1937 map of Quer&aro in the Atlas Geografico de las Estados
UnidosMexicanos (Anonymous 1932-1937).
The type specimen of C. ehrenbergiana (Fig. 1) is rather sparse, consisting of two branches, no flowers,
and one sub-mature fruit in the packet. The leaves are ovate, 27-57 mm long, 22-37 mm wide, rounded,
or in one leaf, with an obtuse point at the tip, rounded at the base, subentire to crenate with up to 8 teeth,
with veins impressed above, with the mid and secondary and some tertiary veins raised, brownish-yellow,
the larger 0.3-0.5 mm wide beneath. The leaves are coarse and thickened (as seen in leaves over a year in
age) the blades are hirtellous-hirsute and coarsely scabrous with hairs to 0.2-0.3 mm long (to 0.5 mm long
along the leaf margins). There are domatia in the axils of the four secondary veins that branch from the base
•o upper-mid-section of the midrib, and between the secondary and larger tertiary veins on the abaxial leaf
surface. The sub-mature dried fruit measured 10.2 x 8.7 mm, the immature pyrene is estimated to be ± 5
strongly decurved, ± 3 mm long, but the tip has beenbroken off (Fig IB).
The upper epidermis of the type of C. ehrenbergiana has a very bullate appearance in part due to the
cystoliths, but the epidermal cells can be seen with 40x magnification and they have a distinctive radiating
Pattern from the base of the erect, stiff hairs. This pattern is not found on all specimens, particularly those
with young, thin leaves, but 1 have found several other Mexican specimens with the same epidermal pattern.
AU of these specimens have been placed within C. iguanaea based on other characteristics, with the single
Curved thorn being the best character to distinguish the species from C. pallida.
The TEX-LL herbaria has many specimens of Celtis from central Mexico (Veracruz, Mexico, Morelos,
H'dalgo and San Luis Potosf) that have leaves of the same general size and shape, with crenate leaf margins,
jjdthat have curved non-node-bearing thorns, and large fruit like those of the type of C. ehrenbergiana.
* ^ of C. ehrenbergiana is not an isolated specimen, rather is representative of populations of the drier
Pwhons of interior Mexico.
1 visited the Los Adjuntos site in June of 2004 to collect the Celtis species in the area. The Los Adjuntos
2^ has low xeric subtropical scrub, strongly disturbed fields, pastures and disturbed woodlands with
^famesiana, Mimosa spp., Neopringlea, Karwinskia, Colubrina, Vallesia glabra, Capparis incana, Maclura
^ Celtis pallida ( Henrickson 23639, TEX) was frequent in disturbed open areas. We then went
mto the mountains and came down another road that crosses the Rfo Montezuma about 18 km to the
**“h at presa de Zimapdn (which forms Laguna de Zimapan) on a road that leads to Zimapdn, Hidalgo.
east of [he dam, we again found shrubs of Celtis pallida ( Henrickson 23648, TEX) and further on a second
J?* Celtis caudata Planch. (Henrickson 23645, TEX) of subgenus Celtis. No specimens similar to the type
^ ehrenbergiana were found on the trip. However, it can be stated that C. pallida near type locality of C.
°«*rgiflna is characterized by paired nodal thorns bearing nodes, small leaves with domatia only along
, “hdrib and other character-states typical of C. pallida elsewhere and it does not look anything like the
ehrenbergiana near its type locality. On the other hand, the type of C. ehrenbergiana clearly must be
'^red a dryland phase of C. iguanaea, based on the diagnostic character-states shared with that species.
rh«efore Celtis ehrenbergiana (Klotzsch) Liebm. 1851 (based on Momisia ehrenbergiana Klotzsch 1847)
:uma, in Queretaro joins three other named species described
from Mexico that are synonyms of C. iguanaea (Jacq.) Sarg. (1895) [based on Rhamnus iguanaeus Jacq. (1760)].
These include: M ertensia laevigata H.B.K. (1817) [= C. laevigata (H.B.K.) Spreng. non C. laevigata Willd. (1814)],
from the state of Campeche; C. anfractuosa Liebm. (1851) from Veracruz; and C. platycaulis Greenm. (1903)
from Morelos (see Berg & Dahlberg 2001).
CELTIS TALA VS. CELT1S PALLIDA
With the removal of C. ehrenbergiana, there remains in Berg and Dahlberg’s C. ehrenbergiana two species that
have been previously recognized as distinct: Celtis tala Gillies ex Planch, (described in 1848) and C. pallida
(described in 1858) (Hunziker & Dottori 1976; Berg & Dahlberg 2001). If the two taxa were considered
conspecific as treated by Berg and Dahlberg (2001), then C. tala would be the oldest available name for the
taxon. So I present here a brief analysis of these two species.
Celtis pallida, as noted above, occurs both in North America and in South America in northern Argentina
and adjacent Paraguay. Celtis tala is confined to north and east Argentina, and adjacent southern Uruguay,
and south Brazil (Hunziker & Dottori 1976; Romanczuk & Martinez 1978; Dottori & Hunziker 1994).
Taxonomic history.— Sprengel (1825) in the 16th [17th] edition of Linneaus’ Systema Vegetabilium, described a
Cdtis spinosa Sprengel from Brazil (location unknown) characterized by flexuous stems and paired curved spines.
Planchon (1848) described Celtis tala (spines paired, straight, leaves small, serrate to subentire) from
ing that the weak diagnosis was not instructive and that the taxon was perhaps better ignored.
In his later treatment in DeCandolle’s Prodromus, Planchon (1873) once again ignored Sprengel’s C.
finosa, but recognized five varieties of C. tala that differed in leaf shape and serration, young stem and leaf
vestiture, and thorn number and shape. His var. gilliesiam of Argentina and Paraguay was considered the
iriety (vel typical var. weddelliana
r- pallida from I
r. chichape fro
Baehni (1937) resurrected Sprengel’s C. spinosa and placed the later-described C. tala within C. spinosa
m his treatment of South American Celtis. He also recognized Planchon’s var. weddelliana as a variety of C.
sPinosa. Baehni’s treatment was followed by Cabrera (1945), Digilio and Legnane (1966) and Dawson (1967)
in their various flora treatments. This also led to M.C. Johnston (1957) making the combination C. spinosa
VaT pallida (Ton). M.C. Johnston for North American material.
Romanczuk (May 1976) in her study of Argentian Celtis, rejected Baehni’s concept of C. spinosa, rem-
Slated C tal“ as a species, but recognized a second species with paired, straight node-bearing spines as C.
*em ana (Planch.) Romanczuk. She further recognized a third species, C. sericea Romanczuk, differing
,°m her c weddelliana in having broader, crenate leaves, with the lower surfaces velutinous-sericeous (vs.
entire, ovate to narrowly ovate leaves with lower surfaces ± uniformly pubescent). Lectotypes and
hol°types were designated for these taxa.
Hunziker and Dottori (June 1976) also rejected the use of C. spinosa, showing it was attributable to C.
,gUanaea- They restored the names C tala and C. pallida and presented complete synonomy and typified
2y °f the syn°nyms, and in a post script (after having seen Romanczuk’s 1976 paper) took Romanczuk
976) t0 for her typifications. They also listed all collections seen from North and South America. They
Titrated the val^ of pyrene characters for distinguishing C. pallida from C. tala, with C. pallida having
Reticulate sculpturing vs a much higher and variably raised reticulate sculpturing for the pyrenes of
: ^ and Provide pyrene illustrations of both taxa. They also noted that the only difference found between
2 a»d South American C. pallida taxa involved pyrene size with the North American plants having
7dylarger pyrenes ( 3.3-4.8 x 2.6-3.8 mm) than those in Argentina [(2.4-)2.9-3.8 x (2-)2.4 3.4 mm],
*Wition they described a new variety, C. pallida var. discolor Hunziker & Dottori, that differed from the
r'mal var'ety in having bicolored leaves, green and sparsely pilose above and whitish and densely seri-
Us ^neath, and with vein axils having small to very small domatia. In their postscript they retypified
Romanczuk’s C. weddelliana, noting that it corresponded to their C. pallida var. pallida, and they equated
their C. pallida var. discolor to Romanczuk’s C. sericea. Dottori (1976), in a separate paper, contrasted foliar
morphology of C. tala and C. pallida, with special reference to the domatia.
In 1978 Romanczuk and Martinez presented a treatment of Celtis of Argentina, recognizing 6 species
including C. tala and C. pallida, the latter with subsp. pallida and subsp. sericea (here using their epithet
at the rank of subspecies rather than Hunziker’s var. discolor ). They presented data on pollen stainability
for the six species of Argentina. They also presented phytochemical data obtained from two-dimensional
paper chromotography of secondary compounds obtained from dried herbarium specimens using standard
butanol-acetic acid-water (6:1:2) vs. 2 % acetic acid runs. Data were given as R.f. values in BAW vs. acetic
acid runs, and the color of the spots in U.V. light vs. U.V. light in the presence of ammonia. While the specific
compounds corresponding to these spots were not identified, C. tala and C. pallida gave different patterns of
spots, while C. iguanaea, C. spinosa and C. pubescens gave almost identical patterns. The Argentina specimens
of C. pallida were distinctive, but the signatures of subsp. sericea from Argentina were quite similar to those
of North American subsp. pallida. Romanczuk (1987) also contributed the treatment of Celtis to the flora
Hustrada de Entre Rios (Argentina) where she used the name C. pallida subsp. sericea.
In summary, C. tala and C. pallida are closely related and share certain characteristics: (1) paired, straight,
divergent, supranodal thorns; (2) relatively small, ovate to narrow, entire or distally weakly serrate-dentate
leaves with three prominent basal veins and a gap between these and the more distal pinnate secondary
veins; (3) domatia mainly occurring along the base of the leaf between the primary-secondary vein axils but
also sometimes more distally along the midvein; (4) leaf epidermal cells with striate cuticles (Romanczuk
& Martinez 1978); (5) and small glomerate inflorescences consisting mostly of staminate flowers, with few
bisexual flowers. There are also consistent differences between the species as indicated in the following key
modified from Hunziker and Dottori (1976) and following the nomenclature accepted in the Catdlogo de to
Plantas Vasculares de la Republica Argentina II (Zuloaga & Morrone 1999) and used in Dottori and Hunzikei
(1994).
. Thorns paired, straight, bearing nodes and leaves, (6-)15-35(-46) mm long [l
2-5C-12) mm long lacking nodes]; pyrenes generally (2.4-)2.9-4.8 mm long,
dally reticulate (rarely smooth), the reticulae low, flat surfaced;.
2. Leaves concolorous, hairs few above, more common below, or the hairs c<
then large and erect (none appressed); basal leaf domatia large, inflated; '
5-12 cm in basal diameter; sw. United Stat
i on both surfaces and
(1-)15-25(-35) mtall,
t. Argentina, rare in adjacent Paraguay.
Celtis patch *
2. Leaves generally bicolored, with a dense whitish indument of short appressed hairs on the lower surface,
upper leaf surface with few hairs or glabrescent; basal leaf domatia small, sometimes inconspicuous; shrubs
to small trees 1 .4-5 m tall, 8-1 5 cm in basal diameter; n. Argentina. Celtis pallida var
. Thorns paired, straight, lacking nodes and leaves, (1.5-)3~15(-20) mm long, (nodes sometimes present on
sucker shoots); pyrenes (3.6-)4.8-5.8(-6) mm long, 3-5(-5.6) mm wide, the reticulae surfaces irregularly raised;
s (3-H
i basal d
As the above key shows, Celtis pallida is a shrub-small tree 1.5-5 m tall with basal trunks 3-15 dm in
ameter, the paired nodal thorns bear nodes and leaves, the fruit pyrenes are 2.4-4.8 x 2-3.8 mm m ^
and the pyrene surface is marked with a low flat-surfaced reticulation. In contrast Celtis tala grows to
large tree 4-10(-20) m in height with basal trunks 3-6(-10) dm in diameter; the paired nodal thorny
not form nodes and leaves, and the fruit pyrenes are slightly larger, 3.6-6.0 x 3.0-5.6 mm, with the
late patterns irregularly raised (see comparative illustrations in Hunziker and Dottori (1976), R°maI’
and Martinez (1978), Burkart (1987). But the problem with these species is the variation found in lea ^
outline, toothing, and the development of thorns. In Texas-Mexico C. pallida, leaf and thorn
greatly, along with leaf indentation, vestiture (sericeous, hirtellous to glabrous) and in well develop*
shoots the paired supranodal thorns are well developed, with conspicuous nodes on the thorns but
collections or branches of the same specimen thorns are highly reduced, often unequal in size, an W1
293
nodes. If one looks only at the vegetative aspects of C. pallida vs. C. tala, they appear to blend together. But,
the pyrene differences are consistent. And the trends in thorn nodes are real.
The above shows that there are differences between C. tala and C. pallida that have been used by
published catalogue of the vascular plants of Argentina (Zuloaga & Morrone 1999), though in most recent
unpublished enumeration of the flora obtained from the Instituto de Bot&nica Darwinion the above taxa are
all placed within C. ehrenbergiana following Berg and Dahlberg (2001).
In contrast to the paper of Berg and Dahlberg (2001), I conclude that C. ehrenbergiana should be con-
sidered a northern-most representative and a synonym of the widespread C. iguanaea (Jacq.) Sarg. and not
related to the C. tala-pallida complex. Furthermore, based on the characteristics expressed by Hunziker and
Dottori (1976) noted in the above key, I support the recognition of C. tala and C. pallida as distinct species.
AH the North American material should be referred to as Celtis pallida Torr. var. pallida in deference to the
recognition of the South American C. pallida var. discolor and var. pallida. Of course, the one South American
taxon can also be recognized as C. pallida subsp. sericea Romanczuk vs. C. pallida subsp. pallida. I also feel
that the final disposition of C. tala vs. C. pallida and its varieties should involve a thorough molecular study
and/or be made by botanists with extensive field experience in Argentina and not by botanists without
first-hand knowledge of the taxa.
ACKNOWLEDGMENTS
1 thank C.C. Berg and one anonymous reviewer for valuable comments on an earlier draft of the manuscript.
REFERENCES
An°wmous. 1 932-1 937. 1 500,000 map of Quer^taro. 1 937. In: Atlas geografico de las Estados Unidos Mexicanos,
Secretarfa de Agricultura y Fomento, Direccion de Geografia Meteorologia y Hydrologfa.
1937. Les Celtis Sudamericans. Candollea 7:1 89-214.
^ cc AND S.V. Dahlberg. 2001 . A revision of Celtis subg. Mertensia (Ulmaceae). Brittonia 53:66-81 .
Ca8*,a a-L 1945. Las Ulmaceas de la Republics Argentina. Revista Argent. Agron. 13:54-55.
Dawson, g. 1 967. Ulmaceae. In: A.I. Cabrera. Flora de la Provincia de Buenos Aires, I.N.T.A., Buenos Aires, 4(3):7-1 2.
^A. and P.R. Legname. 1 966. Los arboles indigenas de la provincia deTucuman. Op. Lilloana 15.
Dott3r|' NM 1976. Morpholigla foliar en Celtis talayC. pallid con special referencia a los domacios. Kurtziana
89:63-80.
***»• N- and A.T. Hunziker. 1 994. Celtidaceae Link. In: Flora Fanerogamica Argentina. 2:5-1 1 .
A.T. and N.M. Dottori. 1 976. Contribucion al conocimiento sobre los talas (Celtis, Ulmaceae) de Argentina,
c°n especial reference a la region Mediterranea. 1. Sobre una especie austro-sudamericana y sobre otra de
disyunta. Kurtziana 9:1 03-140
1 957. Notes. Southw. Naturalist 2:1 72,
1847- Celtideae. Linnaea 20:536-542.
J£ 1848. Sur les Ulmacles. Ann. Sci. Nat. Ser 3, 10:244-314.
1873. Ulmaceae. In: De Candolle, Prodromus Syst. Nat. Regni Veg. 17:168-210.
1-5//.
'9»7. Celtis L In: A. Burkart, ed. Flora ilustrada de Entre Rios (Argentina) 3:1 5-22.
^^es, S.L, W.T. Barker and L.M. Schulz.1996, Ulmaceae Mirbel. Flora N. Amer. 3:368-380.
1825. Celtis. Systems Vegetabilium 16, 1:931-932.
1858- Botany of the boundary. In: W.H. Emory, United States and Mexican boundary survey 2.
And O. Morrone (eds.). 1999. Celtidaceae 1526-528. In: Catalogue de las plantas vasculares de I,
^Wica Argentina II. 2 vols. Monogr. Syst Bot. Missouri Bot. Gard. 74:1-1269.
294
BOOK REVIEW
Jane S. Smith. 2009. The Garden of Invention: Luther Burbank and the Business of Breeding Plants.
(ISBN 978-1-59420-209-4, hbk.). Penguin Press, Penguin Group (USA) Inc., 375 Hudson Street, New
York, New York 10014, U.S.A. (Orders: www.penguin.com). $25.95, 345 pp., 5 1/2" x 8 1/4".
ith does an incredible job of detailing th
From the publisher: “Jan
os Angeles Times Book Prize for Science and Technology. She has s
J. Bot Res. Inst Texas 4<1): 294. 2010
CROTON BIGBENDENSIS TURNER (EUPHORBIACEAE) REVISITED
James Henrickson
INTRODUCTION
Croton dioicus Cav. [C. neomexicanus Mull.Arg.] (Euphorbiaceae) is a wide ranging, distinctive, dioecious
Perennial of Croton section Drepadenium (Raf.) Mull.Arg. (Webster 1993). The species is characterized by
dichotomous styles, valvate, eglandular sepals, petals absent, moderately short petioles, and leaves
with a distinctive silvery canescence formed by fringed peltate scales 0.3-0.5 mm in diameter. The species
occurs from west-central and trans-Pecos Texas and adjacent New Mexico, southward into arid regions of
Mexico in the states of Chihuahua, Coahuila, Nuevo Le6n, Tamaulipas, Veracruz, Durango, Zacatecas, San
iuis Potosi, Hidalgo, Tlaxcala, Mexico, Puebla, and Oaxaca.
Turner (2004) distinguished from C. dioicus, a separate species, with the epithet “bigbendensis,” char-
aaerized as being a larger, bushier plant, with longer upper-stem internodes and linear-lanceolate leaves
5-7 times as long as wide. He noted that the species is restricted to areas along the Rio Grande in Hudspeth,
Kcsidioand Brewster counties of Texas (with one specimen cited from adjacent Coahuila, Mexico) becoming
fominant in sandy habitats. He further noted that the species does not occur with C. dioicus , and where they
** near contact” they do not appear to intergrade. He considered his new species to be a good cryptic
or "biological” species. All the characteristics used to delineate the proposed species were quantitative; no
dilative characteristics were noted or implied.
Unfortunately, in his paper he cites two herbarium specimens as holotypes, one pistillate (Turner 22-
fA) and one staminate ( Turner 22-204B ), each given a separate number. Though collected at the same time
^ the same population (a single gathering), the International Code of Botanical Nomenclature (McNeill
" 4 2006> does not allow two separate specimens with two separate numbers to be designated as separate
?»ypes [ICBN Art. 9.1-”A holotype of a name or a species or infraspecihc taxon is the one specimen or
Ration used by the author, or designated by the author as the nomenclature type (emphasis mine)] and
(Kanchi Gandhi, pers. comm.;
one sheet of a single specimen
KSNAn.
alidly published and is considered an
t Name Index). A holotype
8-3) if clearly noted,;
than one plant in a single gathering (Art 8.2), but
296
the designation of two separately numbered specimens, mounted on separate sheets, each named asholotypt
(Turner 2004, pp. 81 & 82), renders the specimens syntypes (Art. 9.4, ex. 1) and therefore no holotype exists
and the name is not validly published.
As his proposed new taxon occurs within the Chihuahuan Desert, my area of interest, I analyzed
characteristics used by Turner (2004) to distinguish the two taxa using specimens he annotated in the
TEX-LL herbarium. I initially measured total stem length and maximum internode length as an indication
of plant size, leaf blade length and width as well as petiole length from one larger leaf on each specimen. 1
also recorded soil type (clay, gypsum, or sandy-well drained) for each collection when noted on the label
Petiole length and total stem length, while quite variable, were not further used because the petiole length
generally correlated with leaf size and the total stem length displayed in specimens too often reflects the
portion collected, not total plant size. Data on soil type were very spotty from specimens outside west Texas
so this was not used in the analyses but is discussed below. While criticism can be raised from taking
measurements from herbarium specimens, these are the specimens used to make taxonomic decisions and,
while imperfect, they represent samples of a species and they are readily available.
METHODS
Data were collected from all specimens at TEX-LL (n= 296) with 41 from trans-Pecos Texas identified by
Turner as “Croton bigbendensis” plus six of my specimens that are clearly referable to that taxon, 72 from
typical Croton dioicus in trans-Pecos Texas (west of the River), 98 from C. dioicus from cis-Pecos Texas (east of
the Pecos River) and adjacent New Mexico, and 85 from C. dioicus from Mexico. For the quantitative analysis
measurements of longest internode (in mm) present on each specimen, as well as the longest leaf length and
its width (in mm) and these were used to formulate a leaf length/width ratio.
RESULTS
Measurements of maximum intemode length in mm (abscissa) graphed against a ratio of leaf length/leaf
width indicating leaf shape (ordinate) are shown in Figure 1. Collections of putative “Croton bigbendensis
are distinguished by open circles; those of Croton dioicus are indicated by solid shapes with those from
trans-Pecos Texas indicated by solid circles, from cis-Pecos Texas by solid squares, and those from Mexico
by solid triangles.
The means and standard deviations of four characteristics recorded from the "bigbendensis specimens-
from the specimens from each of the three areas of C. dioicus and from all combined C. dioicus are present
in Table 1. While the means of leaf and intemode lengths are distinctive, the large standard deviations
indicate the wide variation found in Croton “bigbendensis.” .
The graph (Fig. 1) shows that while many specimens of “bigbendensis” are characterized by longmtemw®
and/or narrow leaves, the measurements blend into those of C. dioicus — there are not two distinct cUSl ,
of data points indicating well defined character differentation, and a fourth (10 of 41) of the “bigben ^
data points are nested within those of C. dioicus. The specimens of the proposed taxon “bigbendensis ^
leaf 1/w ratios averaging 5.9, ranging from 3-14.4, and maximum intemode lengths averaging 50.
ranging from 31-98 mm. Specimens assignable to C. dioicus, in contrast, have leaf 1/w ratios averaging^
ranging from 1.4-6.8, and maximum internode lengths averaging 30.4 mm, ranging from 11-72 mm.
features exhibit strong quantitative overlap. , ^
Throughout the geographical range of Croton dioicus there is much variation in plant habit, leaW
and leaf and internode size (Fig. 1). In growth form plants range from low, woody-based subshrub ^
erect new growth, to plants that die back to the base and produce erect-ascending shoots that eventua y
and give rise to upright lateral shoots during the growth season. In most plants the basal leaves are ar£*
and the terminal, younger leaves are progressively smaller and often narrower. Leaf shape ranges t ^
broadly ovate, elliptical, oblong-ovate, oblong-elliptic, oblong-lanceolate, linear-lanceolate, and
from central Texas and northeastern Mexico have broader, more ovate, moderate- to large-size
Henrkkson, Croton bigbendensis revisited
297
Croton dioicus
• trans-Pecos Texas
■ cis-Pecos Texas
* Mexico
Maximum intemode length (in mm)
^IMeans and standard deviations of measurements taken of the longest leaf (length, width, and lenght/width ratio) and the longest intemode of each specimen annotated
*T*gbendensisT by Turner, and each specimen of Croton dioicus from trans-Pecos Texas, cis-Pecos Texas, and Mexico. All C. dioicus specimens are combinedfor the
^ standard deviation of"dioicus all collections".
Of desert habitats often have more numerous, smaller leaves, but in years of good rainfall they may
Urge leaves and strong terminal growth. Overall, the leaves of C. dioicus in trans-Pecos Texas are
^•elliptical to elliptical ovate but are otherwise very similar to those of “bigbendensis.”
Uglue 2 shows the distribution of plants of Croton dioicus and “bigbendensis" in trans- Pecos Texas based
annotated by Turner (2004). Croton dioicus is indicated by solid circles and well differentiated
a.e. those with narrower leaves and/or long intemodes) are shown as open circles, and the
T**01 Ihose specimens annotated as "bigbendensis" by Turner, but nested withC. dioicus in Figure 1, are
*>y open squares. This map differs from that of Turner (2004) as it shows the two taxa growing
299
from Puebla also had narrow leaves with 1/w ratios of 3.9-6.3, but internodes 27-55 mm in maximum length.
Some specimens from Nuevo Leon, also compare well with the proposed “bigbendensis,” e.g., Urbatsch 1019
(TEX), with internodes to 41 mm, leaves 37 x 5.2, (1/w ratio 7.1), showing that the characteristics used by
Turner to establish his new taxon were by no means restricted to the Big Bend area of Texas.
The one characteristic that was not possible to analyze is that of growth habit. Most Croton dioicus in
Texas is suffrutescent-suffruticose, woody at the base, with herbaceous stems above forming bushes to about
2-3 dm tall, with the size reflecting water resources of the season. Plants of “bigbendensis” are typically
larger. Turner (2004) notes the hight of “bigbendensis” as 4-5 dm. However, Lott et al. 5588 [Presidio Co.,
Texas (TEX)] notes growth habit as a single-stemmed shrub to 1 m tall, Turner 24-351 [Hudspeth Co. Texas
(TEX)] includes a photograph of a bushy plant ca. 1 m tall and wide. 1 have seen plants in the foothills of
the Chianti Mts. (Presidio Co., Texas) with long lax stems that are initially erect but that eventually fall over
and produce erect sucker shoots to form low rounded plants about 3 dm tall but 12 dm in diameter.
Regarding substrate, most of the specimens of “bigbendensis” are from sandy or otherwise well drained
soils, but others are from clay-limestone flats and hillsides and still others are from roadsides etc. Label data
is missing o
: specimens c
e soil t
» several collections fi
Brewster Co., Texas that were noted coming from clay flats, or limestone substrates-hillsides.
Two collections made by Turner in 2004 in gypsum soils in northeastern Culberson County, Texas
(Fig 2) illustrate the problem with Turner’s interpretation of the taxon. Turner 24-409 (TEX) is a pistillate
plant considered to be C. dioicus that has oblong-ovate leaves with rounded tips, 26.5 x 8.2 mm (1/w ratio
3.2) and intemodes to 42 mm long. It is growing with/near a staminate specimen (Turner 24-402 ) (TEX)
that has lanceolate, acute-tipped leaves, 31 x 6.5 mm (1/w ratio 4.7) with internodes to 46 mm long that
is identified as “bigbendensis.” The two specimens appear identical in stature, coloration etc. and both are
m flower. But while the specimen he considers to be “bigbendensis” is well out of its normal range (Fig. 2),
the specimen of C. dioicus bears a hand- written note by Turner “growing with C. bigbendensis (24-402); no
intermediates seen.” 1 see this as just variation within a population of C. dioicus.
Turner’s Species Concepts : Turner (in Turner 2001; Turner et al. 2003; Turner 2004) advocates the
logical species concept implying that his species have some degree of biological (i.e., reproductive) isola-
tion sensu Mayr 1969, 1992. At the same time, in his personal discussions, he speaks of the importance
tto-geo-naorphologic al factors (or morpho-eco-geographic integerity, Turner 2001) in species recognition
“which species may differ ecologically i.e., occur in different habitats, or in different soil types (gypsum,
»nd) or be geographically isolated (disjunct) and/or have some morphological discontinuity or a combina-
of these features. His primary rank for infraspecific taxa is variety, with subspecies reserved only for
dustering groups of varieties if the need arises sensu Holmgren (1994) and Turner and Nesom (2000). He
^ that varieties are subunits of a species that are geographically separate and generally morphologically
dlstinguishable, but in Turner’s concept they must show intergradation with related taxa to be varieties: if
kseesno intergradation, they are considered distinct species (Henrickson 2004).
The concept that varieties must show intergradation is where Turner departs from others. He considers
“s interpretation as modem in contrast to Old-School definitions where varieties are used for minor varients
ofa species as defined in Correll and Johnston (1970) and Diggs et al. (1999). But I think it is safe to say that
m~L > today recognize varieties (or subspecies) as subunits of a species that have a geographical
t without such regard to the presence or absence of intergradation. And then
15 separate question of whether all variation found in wide-ranging species needs to be recognized
^laturally. But Turner’s varietal concept does explain why he publishes so many new species.
11 “expected that all outbreeding species will exhibit some degree of variation, and that m penphera
^“tions, selection would favor those individuals best adapted to any new biotic or abiotic environmental
Over time, all populations tend to move toward adaptation to their current environment. In
fiTv^0115’ these chanSes can be enhanced, sometimes in a non-adaptive way, by genetic drift (Grant
I Nation of an outlying population differs in substrate or in moi!
^PonentfStue:
1%3)-Ifthek
e availabili
expect changes reflecting adaptation to this new environment. But even if adaptive characteristics become
fixed in an outlying population, does this make the population a distinct species?
In this particular instance, Croton dioicus of northern trans-Pecos and adjacent portions of central Texas
mostly occurs in clay flats in which much of the rainfall runs off quickly, and the plants tend to be short,
low rounded specimens adapted to a seasonally severe environment. In areas of sandy-gravelly substrates,
the water can penetrate deeply, providing a reservoir of water for later development. Rhyolitic substrates can
form coarse gravels-sands that also accumulate rainwaters. Other sandy habitats in this region occur along
the Rio Grande. An increase in moisture availability would favor those plants that could grow larger and
produced more flowers and fruits over a season. This would not require any novel gene-based changes, bot
rather would simply favor those plants within a population that produce more growth-regulating substances
(auxins etc.) that would in turn promote more robust plants with longer internodes and steins and leaves,
characteristics that would not be adaptive in drier sites. Since the southern portion of the southwestern
trans-Pecos Texas has volcanic (not limestone) mountains, and has other sandy habitats along the Rio Grande,
the sandy-gravelly habitats tend to be geographically separated.
With Turner’s eco-geo-morphological concept, he finds ecological, geographic and morphological dif-
ferences between the clay- vs. sandy-substrate populations. But he notes that his new species is a biological
isolation. His map (Turner 2004) shows the two taxa as completely allopatric, but using his own specimens
as mapped here in figure 2 (some added after his publication), they are not completely allopatric except in
the sands along the Rio Grande in Presidio Co., Texas.
What we are dealing with in “bigbendensis,” appears to be a derivative series of populations in well
drained soils that differ only in quantitative characteristics from those on primarily clay substrates. In
characteristics of vestiture, flower, and fruit, and time of anthesis, they are identical to the widespread C
dioicus. Quantitative characters can be used to establish species if there are no or reduced overlapping values
(Luckow 1995). But as shown in Fig. 1, the values of C. “bigbendensis” merge strongly with those of C. dioicus
making recognition at the species rank, in my opinion, untenable. As noted above, measurements of 10 o
the 41 specimens recognized by Turner as “bigbendensis” are nested well within the specimens of C. dioicus
(Fig. 2) and others are closely peripheral. Even if these were to be considered C. dioicus and “bigbendensis
were redefined as only those plants with longer internodes and/or narrower leaves, “bigbendensis" would
still consist of an amalgamation of plants, some with narrower leaves and others with longer internodes,
that would not separate from the variation in C. dioicus.
There is another problem associated with such quantitatively defined “taxa.” In times of good raii&
the plants can show strong growth with long leaves and internodes. But under poor growth conditions,
plants may be stunted with shorter internodes and smaller leaves and some of the plants nested wlthin_
dioicus in figure 2 may be just that. Also, the first-formed leaves of a growth season are typically broad*
than those formed later in the season.
But while the data do show the existence of plants with narrower leaves and longer internodes in
southern margin of trans-Pecos Texas, the plants are not clearly distinguished from C. dioicus. Due t0
blending of the quantitative characteristics used to define “bigbendensis,” using Turner’s own criteria 0*
and Nesom 2000), the taxon would be considered a variety of C. dioicus. But, I see no reason to rec0gI^
bigbendensis” at any rank, but to incorporate it’s characters into the known variation of the wide ranging ^
poorly studied Croton dioicus. As it stands, Croton dioicus is a very distinctive species, easily recognize ^
its range by substantial characters. I see no value in recognize peripherial populations that differ m
quantitative characters as equal species. Certainly in this instance there is no evidence that “bigben e ^
is a good biological species. But we must remember that even with field observations, without expen
data, such taxonomic decisions are just opinion.
302
Journal of the Botanical Research Institute of Texas 4(1)
BOOK NOTICE
Robert D. Dorn. 2010 (March). The genus Salix in North America North of Mexico. Published by the
author. (Orders: digital version available for free at www.lulu.com/content/8538913; printed copy avail-
able in US only from Mountain West Environmental Services, 8481 Road 39, Lingle, Wyoming 82223,
U.S.A.; e-mail: linglebird@yahoo.com). $7.00 (for printed copy), 59 pp., 8 1/2" x 11".
THOMAS WALTER’S SPECIES OF MELANTHIUM (LILIACEAE)
Department of Botany
University of Florida
Gainesville, Florida 326 1 1, U.SA
ABSTRACT
RESUMEN
In the 1780s, Thomas Walter owned and operated a rice plantation on the Santee River, now in Berkeley
County, South Carolina. With a classic education and a keen and inquisitive mind, and guided by his few
books by Carl Linnaeus, Walter observed the plants he found around him and prepared an inventory. His
Flora Caroliniana, published in 1788, was the first American flora to follow Linnaeus’s sexual system of clas-
sification and binomial nomenclature.
Walter kept no herbarium (Ward 2007b), and his brief descriptions, in Latin, have been in many cases
difficult or impossible for later workers to associate with the plant species he intended. Because of the early
date of his book, many of Walter’s names are prior to names given the same plants by Michaux, Pursh, Nut-
tall, and other early authors. In many cases Walter’s names are recognized as prior and are in modern use.
Others lurk in obscurity, potentially of nomenclatural importance, but overlooked or neglected because of
tl»e uncertainty of their identification. Though no complete analysis of Walter’s 400+ new names has yet
published, authors have from time to time dipped into Walter’s Flora with the objective of matching
bis names with known species. Notably, Hitchcock (1905) has addressed the grasses, Blake (1915) vanous
species, Dayton (1952) the pines, and Wilbur (2002) the oaks.
The present task is to examine andidentify the sevennames that Walter used for species within the genus
^ianthium. These species are now seen as representing six distinct genera: Amianthium Gray, Chamaebnum
w'Udenow, Helonias Linnaeus, M elanthium Linnaeus, Tofieldia Hudson, and Zxgadenus Michaux. Historically,
d*e six genera have been included within the Liliaceae. Five of these genera, with others, are often now
^gnized to represent a distinct family, the Melanthiaceae. Tofieldia has been segregated into a separate
Tofieldiaceae. As such most of the included species have been given careful study, with their
Oology and relationships now well understood (Zomlefer 1997a, 1997b; Zomlefer et al. 2001, 200®.
a °*e name formed by Walter, however, has at times been misinterpreted or improperly dismissed. As
Elliott (1817), Walter’s M elanthium hybridum was understood to be the “broad-petaled Melanthm.
'*Vcrafrum) of the Southern Appalachians. Fernald and Schubert (1948: 193) confirmed applicant
J*. Zimmerman (1958: 281) also accepted the name (as a Vemtrum), though simply as “ex char
JV8* by HE. Ahles in the regional Carolina flora (Radford et al. 1968). However Bodkin (1979 .stated
to be “misapplied” and *not in accordance with the rules of nomenclature.” Zomlefer (1997a:
^5) followed Bodkin’s lead in rejecting Walter’s name (though placing the name within Veratrum).
a of the
304
Bodkin and Utech (2003: 78), under M. latifolium Desrousseaux (1797), again dismissed Walter’s name as
“misapplied” without stating where the name should be correctly assigned.
But if one believes that Melanthium hybridum was not a figment of Walter’s imagination, his name
cannot be ignored. With the assumption that Femald and Schubert (1948) and Zimmerman (1958) could
not have been far afield in their identification, Walter’s plant must have been one of the few species of
Melanthiaceae to occur in the Southern Appalachians and adjacent coastal plain. His description excluded
other related genera, and the orbicular, crinkled tepals of M. hybridum (imperfectly described by Walter as
“ plicato-undulatis ”) contrast sharply with the elliptic, plane tepals of M. vi rginicum (which Walter knew as
M. monoicum ). As to Walter’s name not being in accord with “the rules of nomenclature,” the statement is
made without substantiation. Walter’s M. hybridum (1788) is prior to Desrousseaux’s M. latifolium (1797).
There appears to be no impediment to recognition of Melanthium hybridum Walter as the correct name for
the broad-petaled Melanthium of the Southern Appalachians.
Walter’s Santee River home was on the South Carolina coastal plain. His plants, as noted in the introduc-
tion to his Flora, came from within a 50-mile radius. The Scottish horticulturist, John Fraser, with his travels
to the higher Appalachians and into central Georgia (Ward 2006, 2007b), had access to a wider array of
species. It has long been acknowledged that some of Walter’s plants reached him via Fraser (though perhaps
fewer than sometimes stated). The Fraser fir (Abies fraseri), Fraser magnolia (Magnolia fraseri), and showy
lady-slipper (Cypripedium reginae), all montane or upper piedmont species, are outstanding examples.
The identifications given here are subject to an unavoidable level of uncertainty. The basis for each
identification is initially Walter’s Latin diagnosis, which too often is brief and ambiguous. Additional guid-
ance can at times be obtained from early authors with knowledge of the Carolina flora. Three decades after
Walter’s death, Elliott, with his Sketch of the Botany of South Carolina and Georgia (1816-1824), often was
perceptive of his predecessor’s intentions. Britton clearly tried to account for all the Walter names that oc-
curred within the range of his Illustrated Flora (Britton and Brown 1896-1898). Index Kewensis, comp
by Jackson (1893-1895), listed essentially every Walter name, though too frequently reflecting Jackson's
unfamiliarity with American plants.
Perhaps the most powerful aid to correct identifications of Walter’s names is modem knowledge
present plant distribution within the Carolinas. The maps prepared by Radford and colleagues (Radfo **
al. 1968) permit tabulation of species likely to have been known to Walter on the Carolina coastal plain-
Even so, many species may have become much less frequent in subsequent years, possibly through climatt
change and certainly through habitat modification. Thus species that would seem attainable only y wa7
Fraser may possibly have once grown in Berkeley County or environs.
Notations in Walter’s book at times carry unpublished information. His Flora— as an original pub*
cation— is rare. A more widely available facsimile edition was published in 1946. This facsimile present
the sequence of the original copy’s owners, as ascribed on the title page, from James Macbride in | .
Charles Sprague Sargent around 1900. One of the owners (the handwriting has not been identified- ) *^
marginal notes to certain pages, such as pages 125 and 126 regarding Melanthium. These notations,
mum, record the thoughts and speculations of an attentive early botanist.
Of Walter’s seven species of Melanthium, only three (Radford et al. 1968) occur with any frequency^
the Carolina coastal plain— his M. virginicum (= Zigadenus glaberrimus ), M. Muscaetoxicum (=
muscaetoxicum ), and M. racemosum (= Tofieldia racemosa). His other four— M. hybridum (= Veratrum ny ^
M. monoicum (= Veratrum virginicum ), M. spicatum (= Helonias bullata), and M. dioicum (=
luteum)— are rare or absent within his range, and best explained as having come by way of Fraser,
lack of opportunity to know these species in the field may have contributed to his unclear and so
conflicting descriptions. (Another species, Stenanthium densum, is also frequent on the coastal p a
The species are recorded below in the sequence given by Walter (1788). Each entry eonsists
scientific name and Latin diagnosis (in italic ) provided by Walter; the modem name, conforming to
305
ment of the present writer; and pertinent comments of the present writer. Walter’s division of Melanthium
into two sections is recorded, with his diagnostic details (in italic). Marginal notations of a previous owner
of the original volume are also transcribed (in italic). Specimens noted to be in the Fraser/Walter herbarium
are now held by the Natural History Museum, London (BM); they were collected in 1787 by John Fraser and
seen at least in part by Thomas Walter (Ward 2006, 2007b).
UST OF SPECIES
Sectional characters: Petalis unguiculatis imprimis albis demum obscuro-rubris feminibus semi ovatis.
Walter’s name: Melanthium virginicum (p. 125) Linnaeus, Sp. Pi. 339. 1753; misapplied.
Walter’s description: petalis planis maculis duabus Jlavis notatis, floribus plerumque hermaphroditis.
Marginal notation: “hie Zigadenus glaberrimus Michx.”
Modern name: Zigadenus glaberrimus Michx.
Comments: A first assumption was that Walter’s Melanthium virginicum is simply the Melanthium virginicum of
Uimaeus (1753). However, Walter’s description better fits Zigadenus glaberrimus, a frequent species in eastern
South Carolina. The perfect flowers and two distinct yellow glands on each tepal (“ maculis duabus Jlavis") are
diagnostic. Specimen 71-A of the Fraser/Walter herbarium (BM) is probably this; it was labeled “Melanthium
virginicum” by Walter. True Melanthium virginicum, a rare species, was also known to Walter, but under an-
other name (M. monoicum, q.v.). Misidentifications by Walter of Linnaean names are not uncommon (Wilbur
2002; Ward 2007b).
Walter’s name: Melanthium hybridum Walter (p. 125)
Walter’s description: petalis plicato-undulatis [i]mmaculatis,jloribus masculis etfoemineis mixtis.
Modern name: Melanthium hybridum Walt. [= Melanthium latijolium Desr. in Lam.; Veratrum kitijolium
(Desr. in Lam.) Zomlefer]
Comments: Rare in South Carolina, frequent in western North Carolina; likely a Fraser discovery. No specimen
that can be considered Walter’s type is known. [Desrousseaux (1797) cited a Fraser collection, possibly seen
^ Walter, in the Lamarck herbarium, Paris (photo, GH), but made it the type of his Melanthium latijolium .]
A neotype for M. hybridum has been selected elsewhere (Ward 2008b). The justification for restoration of
Walter’s name is provided above.
Walter’s name: Melanthium monoicum Walter (p. 125)
Walter’s description: petalis planis, maculis 2 luteis,Jloribus injerioribus masculis majoribus, paniculis lateralibus;
ntperioribus Joemineis racemo terminali.
Marginal notation: “Melanthium virginicum Michx. ? / [Zygadenus (deleted)) Mx
1 ' me Melanthium virginicum L. [= Veratrum virginicum (L.) Ait ]
Comments: Rare in South Carolina (two counties), frequent in North Carolina; likely a Fraser discovery,
liter’s unusually lengthy diagnosis is adequate for Melanthium virginicum; its flowers are often imperfect.
J®lype is known, and no specimen has been identified in the Fraser/Walter herbarium (BM). Because
a ters name will remain in synonymy, no neotype is needed.
StCTK)NAL characters: Petalis sessilibusjeminibus ov:
W«-Ter’snamp- x
description: scapo thyrsifero.Jloribus hermaphroditis, petalis imprin
M JINALNotat,on: “Helonias [word deleted] / erythrosperma Michx."
ame Amianthium muscaetoxicum (Walt.) Gray [= Zigadenus
i: Frequent throughout. Walter’s name has been uniformly considi
epal col
n Walter (p. 1
n viridibus.
• >uiMuimum muscaetoxicum v»vaiu./ i- * <WaU > ^
^mments: Frequent throughout. Walter’s name has been uniformly considered to apply to the plant presently
^8 his epithet. The ^ „ . I
^ specimen is known. A neotype has been selected elsewhere (Ward 2008a), thereby confirming
ter's eP«het in this usage However a marginal note (immediately following, by a previous owner of
306
the volume from which the 1946 facsimile was copied) well indicates the uncertainty that Walter’s descrip-
tions left behind. Another species, frequent on the coastal plain and surely known to Walter, appears not
mentioned by him. This is Stenanthium densum (Desr.) Zomlefer & Judd [= Zigadenus densus (Desr.) Fern.;
Amianthium angustifolium (Michx.) Gray; Tracyanthus angustifolius (Michx.) Small], a plant quite similar to A
muscaetoxicum but separated (often to generic rank) by its small single tepal glands (vs. tepal glands absent in
A. muscaetoxicum ). As the previous owner suggested (below), Walter may have confounded the two species.
Marginal notation (at page bottom, below all text): “Which of these is the Helonias angustifolia Michx. /It is
confounded with M. muscaetoxicum or omitted.’’
Walter’s name: Melanthium spicatum Walter (p. 125)
Walter’s description: spica nutante, flor. hermaph. radice fibrosa, fol. caulinis subovatis.
Modern name: Probably Helonias bullata L
Comments: If correctly identified, this species is very rare in western North Carolina, South Carolina, and
northern Georgia; it is likely a Fraser discovery. Specimen 58-C (BM) appears to be Helonias bullata; it shows
a distinctive short-spicate inflorescence, although the leaves are atypically narrow. The specimen was num-
bered “579” by Fraser, and labeled “ Helonias bullata ?” by Walter. Since Walter must have relied on Fraser
material for his description, it is unclear why he would have used Linnaeus’s name on the label and formed
a new name in his text. Thus there must be doubt as to identity of Walter’s plant. Melanthium spicatum was
suggested by Index Kewensis (Jackson 1893-1895) to be Xerophyllum setifolium Michx. [nowX. asphodeloides
(L.) Nutt.]. But Walter stated cauline leaves to be “ subovatis ,” while X. asphodeloides leaves are linear, almost
acicular. If specimen 58-C should be confirmed as Helonias bullata, the probability that Walter relied on this
or other material of the same Fraser collection would justify designation of 58-C as lectotype of Melanthium
spicatum Walt. [Rationale for such typification is given by Ward (2007a).] However, the status of that epithet
as a synonym of Linnaeus’s prior name makes this designation unnecessary.
Walter’s name: Melanthium dioicum Walter (p. 126)
Walter’s description: petalis sublinearibus.
Marginal notation: “ Veratrum luteum Linn. /Helonias pumila, Jacquin.”
Modern name: Chamaelirium luteum (L.) Gray [= Veratrum luteum L.]
Comments: Rare on South Carolina coastal plain, common inland. Walter’s diagnosis is exceedingly brief
(two words), but the narrowly oblanceolate tepals of Chamaelirium luteum correspond well, and usage by
other authors has been consistent. No type specimen is known. Linnaeus’s prior epithet makes designation
of a neotype unnecessary.
Walter’s name: Melanthium racemosum ? Walter (p. 126)
Walter’s description: racemo termiali, pedunculis trifloris, floribus calyculatis, petalis albis obovatis, anthe ns
ovatis, luteis, erectis.
Marginal notation: “ Narthecium pubens / Michx.”
Modern name: Tofieldia racemosa (Walt.) BSP.
Comments: Frequent in eastern South Carolina. The inflorescence with mostly 3-flowered nodes
culis trifloris ”) is distinctive of Tofieldia racemosa, as are the obovate white tepals. There seems no reason
for Walter’s indication of doubt; the name Melanthium racemosum was original with him. Michaux (l»w
used Melanthium racemosum differently, perhaps for Veratrum virginicum L. as suggested by Index Kewensis
(Jackson 1893-1895). Other authors have been consistent in recognizing the transfer of Walter’s epM*
to Tofieldia; none has suggested invalidity on the basis that the query indicates Walter’s failure to a
his new name (McNeill et al. 2006, Art. 34.1). No type specimen is known. A neotype has been
elsewhere (Ward 2008b).
ACKNOWLEDGMENTS
Ik author is grateful to Janet R. Sullivan (NHA) and Kanchi Gandhi (GH) for their careful evaluation of an
earlier version of this manuscript. The studies of Wendy B. Zomlefer (GA) on the Melanthieae have been most
helpful, as has her review of the manuscript. Christine M. Housel (ABT) has provided the Spanish abstract
with speed and grace. And a nod of thanks must go to Charles E. Jarvis (BM) for the example he has set by
his long-continued efforts to typify the far more numerous names by Carl Linnaeus.
REFERENCES
Biake, S.F. 1915. Some neglected names in Walter's Flora Caroliniana. Rhodora 1 7:129-137.
Bodkin, N.L 1979. Melanthium hybridum Walt.: A misapplied name. (Abstr.) Virginia J. Sci., ser. 2, 30:54.
Bookin, NL and RH. Utech. 2003. Melanthium. I n: Flora North America, vol. 26. Pp. 77-79,
Bruton, N.L. and A. Brown. 1 896-1 898. An Illustrated Flora of the Northern United States, Canada and the British
Possessions. New York. 3 vols.
Dayton, WA. 1952. Some notes on United States tree names. Rhodora 54:67-79.
Desrousseaux, LA. J. 1797. In: Lamarck, Encycl. Method. Bot. 4:25.
Ewott, S. 1 81 6-1 824. A Sketch of the Botany of South Carolina and Georgia. J. R. Schenck, Charleston.
fwwiD, M.L and B.G. Schubert. 1 948. Studies of American types in British herbaria. Part IV: Some species ofThomas
Walter. Rhodora 50:190-208, 21 7-229.
Hitchcock, AS. 1905. The identification ofWalter's grasses. Rep. (Annual) Missouri Bot. Gard. 16:31-56.
Jackson, B.D., ed. 1893-1895. Index Kewensis. Cambridge.
Annaeus, C. 1 753. Species Plantarum. 2 vols. Stockholm.
^Neill, J, F.R. Barrie, H.M. Burdet, V. Demoulin, D.L. Hawksworth, K. Marhold, D.H. Nicolson, J. Prado, P.C. Silva, J.E. Skog,
NJ-Turland and J. Wiersema (eds.). 2006. The International Code of Botanical Nomenclature (Vienna Code), July
2005. Regnum Veg. 1 46:1 -568.
M^haux, A. 1803. Flora Boreali-americana. 2 vols. Paris.
Rw°®< A.E., H.E. Ahles, and C.R. Bell 1 968. Manual of the vascular flora of the Carolinas. Univ. of North Carolina
^ss, Chapel Hill.
Walter, T. 1788. Flora Caroliniana. London.
Ward, D.B. 2006. Thomas Walter typification project, I: Observations on the John Fraser folio. Sida 22:
fill-1118.
Wa®' d-B. 2007a. Thomas Walter typification project, II: The known Walter types. J. Bot. Res. Inst. Texas 1 :
ward, D.B. 2007b. The Thomas Walter herbarium is not the herbarium ofThomas Walter. Taxon 56:91 7-926.
Wa®. D.B. 2008a. Thomas Walter typification project, V: Neotypes and epitypes for 63 Walter names of genera
D through Z. J. Bot. Res. Inst. Texas 2:475-486.
D B- 2008b. Thomas Walter typification project, VI: Neotypes for an additional 1 8 Walter names. J. Bot. Res.
hst Texas 2:1279-1 283.
2002. Thomas Walter's oaks from the coastal region of South Carolina. Rhodora 1 04:1 34-1 50.
J-H- 1 958. A monograph of Veratrum. Ph. D. dtss, Univ. of Wisconsin, Madison.
^ W.B. 1997a. The genera of Melanthiaceae in the southeastern United States. Harvard Pap. Bot 2:
133-177.
^ W.B. 1997b. The genera of Tofieldiaceae in the southeastern United States. Harvard Pap. Bot. 2:
179-194
N.H. Wiluams, W.M. Whitten, andW.S. Judd. 2001 . Generic circumscription and relationships in the tribe
^arithieae (Liliales, Melanthiaceae), with emphasis on Zigadenus : evidence from ITS and trnL-F sequence
^•^er.J. Bot 88:1 657-1 (
BOOK REVIEW
Tom Standage. 2010. An Edible History of Humanity. (ISBN 978-0-8027-1991-1 , pbk.). Walker & Company,
175 Fifth Avenue, New York, New York 10010, U.S.A. (Orders: www.amazon.com or MPS Distribution
Center, 16365 James Madison Highway, Gordonsville, Virginia 22942-8501, U.S.A., 1-888-330-8477).
$16.00, 288 pp„ 5 1/2" x 8 1/4"
way Standage brings t
i the affect food has
e, politics, and war.
I.Bot. Res. Inst. Texas 4(1): 308. 2010
ACRONYMS FOR RECENTLY PROPOSED ANGIOSPERM FAMILIES
Daniel L. Nickrent
Department of Plant Biology
Southern Illinois University
Carbondaie, Illinois 62901 -6509, U.S.A.
nickrent@plant.siu.edu
ABSTRACT
Nine additional
RESUMEN
Standardized abbreviations have been proposed for all vascular plant families (Kiger & Reveal 2000, 2009).
The complete list of 3-letter acronyms is being maintained and updated by the Colorado Native Plant Society
(see http://www.conps.org/pdf/Plant%20Keys/ACROS.pdf). Two recent publications have appeared that
proposed new angiosperm family names (Nickrent et al. 2010; Schaeferhoff et al. 2009). Given that these
I thank Neil Snow (BISH) for encouraging me to propose the Santalales acronyms as well as for pointing out
that Microteaceae also requires one.
*** RW and J.L Reveal 2000. A comprehensive scheme for standardized abbreviation of usable plant-family
names and type-based suprafamilial names. Huntia 1 1:55-84.
^ R W. and J.L Reveal. 2009. A comprehensive scheme for standardized abbreviation of usable plant-family
names and type-based suprafamilial names, http://www.plantsystematics.org/reveal/pbio/fam/famabbr.
html [Accessed 3 May 2010].
Nckr£^ D.L V. Malecot, R. Vidal-Russell, and J.P. Der 2010. A revised classification of Santalales. Taxon 59:538-558.
Sow&ERHOff. B-, K.F. Mgller, and! Borsch. 2009. Caryophyllales phylogenetics: disentangling the Phytolaccaceae
and Molluginaceae and description of Microteaceae as a new isolated family. Willdenowia 39:209-228.
acronyms are seeing wide usage among herbarium curators and others,
. The proposed acronyms are:
important to now link each
Amphorogynaceae (AMP) Nanodeaceae (NAD)
Cervatesiaceae (CER) Strombosiaceae (STB)
Comandraceae (COM) Thesiaceae (THI)
Coulaceae (COU) Ximeniaceae (XIM)
Microteaceae (MCT)
Ximeniaceae (XIM)
ACKNOWLEDGMENTS
REFERENCES
Journal of the Botanical Research Institute of Texas 4(1)
BOOK REVIEW
Paul Martin Brown. 2008. Lady’s-slippers in Your Pocket: A Guide to the Native Lady’s-slippers Or-
chids, Cypripedium, of the United States and Canada. (ISBN 1-58729-655-1, pbk .). University
of Iowa Press, 119 W Park Road, 100 Kuhl House, Iowa City, IA 52242-1000, U.S.A. (Orders:
uiowapress.org, 1-800-621-2736, 1-800-621-8476 fax). $9.95, pocket guide, selection of color images,
plant descriptions, and basic key, 4 1/8” x 9” folds out to 16 3/4" x 16 7/8".
Paul Martin Brown. 2008. Ladies’-tresses in Your Pocket: A Guide to the Native Ladies’-tresses Orchids,
Spiranthes, of the United States and Canada. (ISBN 1-58729-656-X, pbk.). University of Iowa Press,
119 W Park Road, 100 Kuhl House, Iowa City, IA 52242-1000, U.S.A. (Orders: www.uiowapress.org,
1-800-621-2736, 1-800-621-8476 fax). $9.95, pocket guide, selection of color images, plant descrip-
tions, and basic key, 4 1/8” x 9” folds out to 16 3/4" x 16 7/8".
76102-4025,
I. Bot R«. Inst Texas 40): 110. 3
CORRIGENDA:
THE REINSTATEMENT OF BEAUTEMPSIA (CAPPARACEAE) AND A KEY TO THE
GENERA OF NEOTROPICAL CAPPARACEAE WITH VARIOUSLY STELLATE TO
PELTATE INDUMENTA
Xavier Cornejo Hugh H. litis
312 Journal of the Botanical Research Institute ofTexastfl)
BOOK REVIEW
Charles R. Hart, Barron Rector, C. Wayne Hanselka, Robert K. Lyons, and Allan McGinty 2008. Brush &
Weeds of Texas Rangelands. (ISBN 0-9721049-4-1, pbk.). Distributed for Texas AgriLife Extension
Service by Texas A&M University Press, John H. Lindsey Building, Lewis Street, 4354 TAMU, College
Station, Texas 77843-4354, U.S.A. (Orders: http://agrilifebookstore.org, www.tamu.edu/upress, 1-800-
826-8911). $25.00, 203 pp, 406 color images, preface, 97 maps, plant descriptions, and general index,
5 1/2" x 8 1/2".
FL OURENSIA ILICIFOUA (COMPOSITAE: HELIANTHEAE), NUEVO REGISTRO
PARA DURANGO Y SEGUNDA ZONA DE DISTRIBUCION PARA LA ESPECIE
M. Socorro Gonzalez-Elizondo and I. Lorena Lopez-Enriquez
CIIDIR IPN Unidad Durango, Sigma 1 1 9 Fracc 20 de Noviembre II
34220 Durango, Dgo., MEXICO
herbario_didir@yahoo.com.mx
Jos^ A. Villarreal Quintanilla Jovan Aleman Medrano and Jaime Sanchez Salas
DepartamentodeBotanica
UmrsidadAutdnoma Agraria Antonio Narro
Buenavista253 15 Saltillo, Coah., MEXICO
Escuela Superior de Biologla
Universidad Judrez del Estado de Durango
Av. Universidad s/n. Fracc. Filadelfia
GdmezPalacio 35070, Dgo., MEXICO
RESUMEN
HMmiaiUcifdia Brandegee (Compositae), conocida previamente sdlo para
SMtstt del estado de Durango, Mexico. Se presenta una clave para identific
ABSTRACT
ia ilicifolia Brandegee (Cot
RourensiaDC. (Compositae: Heliantheae), comprende alrededor de 30 especies americanas distribuidas prin-
cipalmente en regiones aridas y semiaridas, con distribucion disyunta (Rzedowski y Calderon de Rzedowski
2008), del suroeste de Estados Unidos a Centroamerica (Strother 2006) y de Peru a Chile y Argentina. Son
arbustos, subarbustos o pequenos arboles aromaticos, usualmente glutinosos, de hojas alternas, simples.
Gbezuelas con involucro de bracteas en 2 a 5 series; receptaculo con paleas rodeando a los aquenios; flores
bguladas presentes o ausentes, flores del disco hermafroditas, amarillas; aquenios sericeos o a veces glabros,
con vilano de (1)2(4) aristas, rara vez ausente.
En el presente trabajo se registra una extension de la distribucidn de Hot
“ presenta una clave (adaptada de Dillon 1984) con el fin de facilitar el recono
Rwrensia de Coahuila y Durango. Flourensia ilicifolia se describio en 1906 de c
Coahmla (Blake 1921; Dillon 1984)- desde entonces solo se conoce de algunos
l0«Mad y ha sido considerada como endtmica para el sur de Coahuila (VilUsetior 1991; vmarreal-
Wanffla 2001; Villaneal-Quintanilla y Encina-Domlnguez 2005). La especie se registra ahora para la
*°* El Samoso, al oriente del estado de Durango. A pesar de que la distancia entre la localidad de Durango
)’ hs localidades conocidas en Coahuila no rebasan los 140 km en llnea recta, se considera de interns dar a
‘""xrreste nuevo registro que amplla su distribucidn hacia el occidente de la regidn fisiografica Pliegues
Oversales (Villarreal Quintanilla et al. 1996) a trav^s del Desierto Chihuahuense. Con esto, el area de
^emismo de F. ilicifolia pasa de ser local a regional (Rzedowski 1991). Otra especie conocida para la Sierra
* ^rras yotras serranias de Coahuila (Villarreal-Quintanilla y Encina-Domlnguez 2005) que ha sido re-
^^mente reportada para el oriente de Durango es Agave parrasana Berger (Gonzalez Elizondo et al. 2009),
611 este caso la especie esta restringida a las partes alias de las sierras.
P^rensia ilicifolia y Agave parrasana pertenecen a un contingente de especie
a de distribucion
315
1. Cabezuelas radiadas (con flores liguladas).
6. Hojas 15-2.5 cm de largo, 0.5-1 cm de ancho, pecfolos hfspido-pilosos por lo menos en la base; cabezu-
-I i | y, .rim In' r li,.|il tnl .1 Flourensia microphylla (A Gray) S.F. Blake
6. Hojas (2— )5— 1 2(-1 5) cm de largo, 1 -4.5 cm de ancho, pecfolos no hfspido pilosos; cabezuelas jovenes con
involucro glabrescente o glabro.
7. Subarbustos con tallos erectos poco ramificados; ligulas 1 3-21 , 1 0-1 6 mm de largo x 3-4 mm de acho;
flqres del disco 40-50 Flourensia pringlei (A. Gray) S.F. Blake
7. Arbustos muy ramificados desde la base; ligulas (10-)13(-15), 20-30 mm de largo x 5-7 mm de acho;
flores del disco 50-60(-1 50).
8. Bracteas externas del involucro 9-13 mm de largo; hojas 6-115 cm de largo, 2-45 cm de ancho,
estrigosas; plantas de bosque de pino-encino, Coahuila y Nuevo Leon Flourensia monticola
M.O. Dillon
8. BrScteas externas del involucro 18-20 mm de largo; hojas (5— )8— 1 2(— 1 5) cm de largo, 2-3 cm de
ancho, puberulentas; plantas de matorral submontano, Chihuahua, Coahuila y Durango Flourensia
pulcherrima M.O. Dillon
De las ocho especies incluidas en la clave, F. monticola se encuentra en la Sierra Madre Oriental y las restantes
seencuentran en el Desierto Chihuahuense o en serranias dentro de esa region, que en su parte de Coahuila
y Durango parece representar el centra de diversificacidn del genera. Ademas de estas especies, se conocen
para Mexico las siguientes: Flourensia collodes (Greenm.) S.F. Blake, de bosque tropical caducifolio en Oaxaca y
Chiapas; F. glutinosa (Robinson et Greenm.) S.F. Blake, de Puebla y Oaxaca (Dillon 1984); F. laurifolia DC., de
matorral tamaulipeco, matorral submontano y bosque tropical caducifolio del sur de Tamaulipas a Guanajuato,
Querftaro e Hidalgo y F. resinosa (Brandegee) S.F. Blake, de la zona arida queretano-hidalguense, la porcidn
surdela zona arida Chihuahuense (Rzedowski 1978; Rzedowski y Calderon de Rzedowski 2008).
AGRADECIMIENTOS
Agradecemos al M. en C. Jorge A. Tena el apoyo con trabajo de gabinete. A los revisores del manuscnto,
Dr. John L. Strother y Dr. Jose Luis Villasenor sus sugerencias para mejorarlo; a los curadores y personal de
losherbarios ANSM, CIIDIR y MEXU por facilidades para consulta de colecciones, y a la COFAA y EDI del
Instituto Politecnico Nacional por los estimulos a la investigation.
REFERENCIAS
^.F. 1921. Revision of the genus Flourensia Contr. U.S. Natl. Herb. 20:393-409.
°lU)N-M.0. 1984. A systematic study of Flourensia (Asteraceae, Heliantheae). Fieldiana, Bot, n.s. 16:1-66. 1991.
^ez Elizondo, M. S„ M. Gonzalez Euzondo y J. Rzedowski. 2000. Dos nuevas especies de Viguiera (Compositae)
Estado de Durango, Mexico. Acta Bot. Mex. 53:35-48.
Gonzalez Euzondo, M„ R. GalvAn Villanueva, I.L Lopez Enriquez, L Resend* Rojas y M.S. Gonzalez Elizondo. 2009. Agaves
-^gueyes, lechuguillas y noas- del Estado de Durango y sus alrededores. CIIDIR Unidad Durango Institute
Politecnico Nacional-CONABIO. Durango, Dgo. . .
1972. Contribuciones a la fitogeograffa florfstica e histbrica de Mexico. III. Algunas tendencias en la
dWribucion geografica y ecologica de las Compositae mexicanas. Ciencia (Mexico) 27:123 132.
1978. Vegetation de Mexico. Editorial Limusa. Mexico, D.F.
RaDOWs«. J- 1991. El endemismo en la flora fanerogamica mexicana: una apreciacibn analitica preliminar. c a
801 Mex. 15:47-64
**** J y G. Calderon de Rzedowskl 2008. Compositae, tribu Heliantheae (g^neros Acme/fa - JeM). Flora del
cJ'°ydere9iones adyacentes. Fasciculo 157.
^2006 .Floorensia. In: Flora of North America North of Mexico. Vol. 21. Ast«,d« part
* ^eraceae, part 3, eds. Flora of North America Editorial Committee, New York: Oxford University Press, e
fcnea http-y/www ef|0ras ora/florataxon aspx?flora _id=1 &taxon id=1 1 2885 [1 1 Junio 2009]
J.A. 2001 Flora de Coahuila. Listios florfsticos de Mdxteo XXIII. Instltuto de Biologla. Uni-
^sidad Nacional Autbnoma de Mexico.
316
Villarreal-Quintanilla, JA, J. Valdes Reyna y J.L Villasenor R. 1 996. Corologfa de las Asteraceas de Coahuila, M6dca
Acta Bot. Mex. 36:29-42.
Vularreal-Quintanilla, JA y JA Encina-DomLnguez. 2005. Plantas vascu lares endemicas de Coahuila yalgunas areas
adyacentes, Mexico. Acta Bot. Mex. 70:1-46.
Villasenor, J.L 1991. Las Heliantheae endemicas a Mexico: Una guia hacia la conservacion. Acta Bot. Mex.
15:29-46.
DIVERSITY AND ABUNDANCE OF ORCHIDS IN A PERUVIAN CLOUD FOREST
Rebecca Repasky Luke1
m Christian University
TCU Box 298930
North, Texas 76129, USA
Department of Biology
Texas Christian University
TCU Box 298930
Fort Worth, Texas 76129, USA
John P. Janovec Eric Christenson
mica! Research Institute of Text
500 East 4th St.
Worth, Texas 76102-4060, USA
jjanovec@brit.org
Keri McNew Barfield
Botanical Research Institute of Texas
Fort Worth, Texas 76102-4060, USA
ABSTRACT
RESUMEN
* Orchidaccae is one of the largest families of flowering plants (Stebbins 1981; Gentry 1988; Gravendeel
» 2004) with at least 1,000 mostly epiphytic species occurring in Peru alone (Ibhsh et ah 1996). The
J‘gh Peruvian species diversity of orchids, especially that of epiphytic orchids in Andean cloud forests has
n orchid from the Peruvia
318
Journal of the Botanical Research Institute of Texas 4(1)
was first characterized in 1760 (Rolfe 1916). In 1911, Hiram Bingham described the Machu Picchu site and
commented on the orchids occurring there (Christenson 2003). This high orchid diversity occurs due to
a mixture of conditions including (1) a cool, moist climate favorable to epiphytes, (2) the diversification of
isolated populations on separate mountain ranges (Tremblay et al. 2005), and (3) the ability of orchid specks
to easily hybridize (Tremblay et al. 2005).
Although this diversity has been recognized, there is limited quantitative information on the species
richness and spatial distribution of orchids across sites and along elevation gradients in the Peruvian Andes.
More is known about how these factors affect epiphytes in general than orchids in particular. These montane
forests harbor a high diversity and abundance of epiphytic plant species (Grubb et al. 1963; Comelissen &
Ter Steege 1989) with maximum species richness and endemism occurring in wet aseasonal forests on fertile
soils at elevations of 1500 to 2000 masl (Gentry & Dodson 1987; Iblish et al. 1996; Kessler, 2002; Kuperet
al. 2004; van der Werff & Cosiglio 2004; Kromer et al. 2005). Species richness declines above 2000 m aslbut
whether this decline indicates some climatic optimum or merely reflects the decline in available land mass
above this elevation remains unclear (Kromer et al. 2005). Orchid diversity generally follows the trends for
all epiphytes (Kessler 2002; Kuper et al. 2004; Kromer et al. 2005), but the pattern may vary among orchid
genera (Kessler 2002).
Detailed studies of orchid diversity and distribution at cloud forest elevations are limited to two studies
in southeastern Peru. One was a 12-month study in a 143.5-hectare section on the Machu Picchu Historical
Sanctuary (MPHS) near the Winay-Wayna structures (WW; Zambrano et al. 2003a). The other was a 12-month
inventory of 16 100-m2 plots in the Manu National Park (MNP; Zambrano et al. 2003b). Both studies report
more than 170 orchid species, with limited overlap with the extensive species list compiled by numerous
investigations at the MPHS (Christenson 2003). These two studies suggest a diverse orchid flora composed
of numerous species that may have limited spatial distributions. Although these studies document high
species diversity at each site and high site-to-site variability in species composition, neither study assesses
the relative abundances of the various orchid species. Comparison of the variability in species compositions
among these studies is also complicated by potential problems in nomenclature that are difficult to resolve
because of the lack of access to reference specimens, digital imagery, or detailed collection data.
The present extent of our understanding of Andean orchid diversity is similar to those that existed fat
western Amazonian tree floras before the advent of quantitative studies of species abundances. Based on early
compilations of species lists at different sites, the forests could be depicted as small-scale mosaics of relatively
unpredictable species composition. As quantitative studies were performed following the recommendations
of Phillips and Raven (1996) and others, patterns of structure involving some widely dispersed, commo y
occurring, and relatively abundant taxa could be recognized (Pitman et al. 2001; Macia & Svenning 2005)-
Although some studies have not provided strong support for the existence of such widespread comm
ity, formation of a species community, of abundant tree species (e.g., Phillips et al. 2003; Tuomist0
2003; Vormisto et al. 2004), they have demonstrated that testing for the presence, or the absence, o
communality requires abundance as well as species composition be documented.
This study had two main goals. First, to initiate quantitative assessments of orchid diversity
abundance in cloud forests similar to those employed for Amazonian tree flora, and second to comPar*
results to other detailed studies of cloud forest orchid diversity in southeastern Peru to determine poss
communalities in abundant taxa. The specific objectives of the study were to: (1) determine orchid spec®
richness; (2) estimate the relative abundances of species; and (3) question whether the abundant o
species at this site were also abundant, or at least present, at other research sites in the region,
sity and abundance. If elevations and habitats have large effects, they could contribute to heterogeneity
species compositions and abundances.
Lukeetal., Diversity and abundance of Orchids
METHODS
The design of this study emulates the general protocols recommended by Phillips and Raven (1996) for
assessing tree diversity in the Neotropics. These protocols share procedures designed to quantify diversity
on a local scale by obtaining: (1) a more complete inventory of plant species composition; (2) an assess-
ment of the spatial distribution of species; and (3) basic phenological information. These objectives are
accomplished by conducting surveys to document the species composition and by monitoring of perma-
nent plots for species abundances. For orchids, these objectives were addressed between September, 2005
and August, 2006 through monthly, systematic collections of the orchid species present and the monthly
monitoring of permanent plots to measure diversity, distribution, and abundance of the species present.
Because the study involves the application of new protocols to quantify orchid abundance and diversity,
an evaluation of these protocols with recommendations for potential modifications are also discussed.
Study site. — Research was conducted at the 560-ha Wayqecha Cloud Forest Research Station (WCFR;
13°10'40"S, 71°36'20"W) located 60 km northeast of Cusco in southeastern Peru (Fig. 1). The station borders
Manu National Park and is owned and operated by the Amazon Conservation Association (ACA) of Wash-
ington, DC and its sister organization in Peru, the Asociacibn para la Conservacibn de la Cuenca Amazbnica
(ACCA). Elevations at WCFR range from 2200 to 3200 m, and the primary natural vegetation is upper
montane forest (Young & Lebn 2000). The forest is continually saturated with rain and fog. Temperatures
average 1PC with little seasonal variation. Precipitation ranges from < 10 mm in the months of June and
July to > 100 mm in the months of January, February, and March.
Systematic plant collection.— To document the orchid species occurring at WCFR, approximately
10 days of each month were devoted to collecting, photographing, and preserving specimens of flowering
orchids. Every orchid with a unique vegetative or reproductive morphology was considered to be a poten-
tial new morphospecies unless it could be positively identified as belonging to a previously collected type.
Although this practice resulted in replicate collections of some orchid species, it minimized the number of
missed species.
One or more specimens were collected for each potentially new morphospecies. At least one specimen
was ^posited in the herbarium of the Museo Nacional Mayor de Historia Natural (USM) in Lima, Peru.
Another specimen was deposited in the herbarium of the Botanical Research Institute of Texas (BRIT) in
Fort Wonh, Texas, and additional specimens were dispersed to orchid experts at other herbaria. Flowers
were also collected and preserved in 80% alcohol and 20% glycerine and deposited at BRIT.
Each potential morphospecies was assigned an identification number, and pertinent data were re-
corded including: (1) date of collection; (2) location as determined by a Garmin Map76C global positioning
Sysiem unit (GPS, nominal accuracy < 15 m); (3) relative flower size (1 to 5, 5 being the largest); (4) color
°f the flower (particularly lip and column colors); and (5) habit (terrestrial or epiphytic). The habitat of the
individual was also classified as: (1) tall cloud forest with tree heights > 15 m; (2) short cloud forest with
tree hei8hts < 15 m; and (3) grassy areas with no trees or few trees with heights < 3 m. Grassy areas likely
from disturbances, such as logging, road construction, or fire over the past 60 years (Young & Lebn
°°0; Lozan<> et al. 2006). To aid in species identification, *8 mega-pixel digital photographs were taken for
^ Potential morphospecies. Collection data and digital photographs are accessible through the Atrium
iversity Information System at BRIT (http://atrium.andesamazon.org).
upon completion of the field collections, specimens of potential morphospecies were compared in
* ^barium and judged to be either a unique, valid taxon, or to be conspecific with another collected
^ospecies. For those morphospecies judged to be conspecific, the collected specimens were combined
“t0 a single morphospecies. All morphospecies collections were reviewed by taxonomic experts to confirm
^morphospecies status and to begin the process of species identification and description.
, Quantitative data collection.— To analyze the effects of elevation and habitat differences on orchid
'^rs‘iy, abundance, and distribution, 47 permanent, 5 x 5 m plots were established across the elevation
Wknt wi[hin WCFR. Because of steep gradients and dense vegetation, the plots were established at ran-
BRAZIL
dom locations along pre-existing trails passing through the characteristic forest and grassland habitats
the area. The latitude, longitude, and elevation of each plot were measured by GPS. Habitats were classified
as tall cloud forest, short cloud forest, or grassy areas, as described above. , ^
From September 2005 through August 2006 the plots were monitored between the 9th and the 1 ^
each month for the occurrence of potential morphospecies in flower. For the first occurrence ofaPote"
morphospecies in a plot, the species was catalogued using its identification number from the general c0“
and the number of individuals with flowers, total number of flowers, and whether the orchid was occu ^
as a terrestrial or epiphytic form was recorded. If a potential new morphospecies was first encounte
the plots, it was added to the general collection using the previously described procedures to col edt
able specimens located outside of the plot. If the orchid only occurred in the plot, only digital phoiograp^
were taken to assist in species identification. For subsequent monthly observations of a morphospec‘es
plot, the number of individuals with flowers and the total number of flowers were recorded. Genera
model tests of the effects of elevation and habitat type on the species richness and density of orchids were
performed using the Statistical Analysis System (SAS; Cody & Smith 2006).
Limited resources and field conditions prevented each individual plant from being tagged, and it is
possible that individuals counted as flowering in one month could also be counted as flowering in subse-
quent months. For this reason, the term density as used in subsequent discussions is a minimum estimate
of density taken as the maximum number of individuals of that morphospecies observed flowering in that
plot in any one month.
Rarefaction analysis. — To evaluate how well the plot sampling could estimate the species richness
determined by systematic collections, rarefaction curves (Cotelli Colwell 2001) were computed using
the non-parametric species richness estimators Chao 1 (Chao 1984) and Chao 2 (Chao 1984, 1987). These
procedures use the occurrence of rarer species in the samples to adjust the observed number of species for
the number of species that were likely missed in the sampling. The Chao 1 procedure defines rareness as
being represented by only a few individuals. The Chao 2 procedure defines rareness as being observed in
only a few plots. Chaol and Chao 2 estimators and their 95% Cl were computed from 10,000 randomiza-
tions of plot sequences using the software Estimates 8.0 (Colwell 2006). Other richness estimators were
Systematic collections.—
were identified from 49 gen
RESULTS
?m the 341 potential morphospecies collected, 239 unique morphospecies
. Over 134 morphospecies have been identified as previously described spe-
e (Chri
lining 102 morphospecies r
islynat
& Galan de Mera 2008). The
: identified as previ
new to science. It is important to note that the as yet unidentified taxa reported here are morphospecies
based on structural characteristics which may not be supported by subsequent genetic analysis. However,
it is also possible that subsequent genetic analyses may separate apparently similar orchids into separate
species. The number of morphospecies per genus ranged from 41 for Stelis to <2 species in 32 other genera.
The number of morphospecies in the other common genera Epidendrum, Maxillaria, and Pleurothallis were
three, six and nine, respectively. A species list organized by genus is presented in Appendix SI .
Monthly collections were important for adequately assessing orchid diversity as most species flowered
fors3 months. Monthly importance is best illustrated by the flowering phenology data from the plot stud-
ies where only six of the 17 species flowering in September 2005 were among the 65 species observed to be
flowering in February, 2006. Only six of the 65 species flowering in February were among the 23 species
Do^g in August, 2°°6. ^ k d- t equal
National intervals of 175 m, there were similar numbers of plots in each interval (Fig. 2). All three habitats
Were Present in each interval
Orchid densities and distributions.-The minimum number of individuals recorded in all of « he plots
>n of the study was 2565, which is the sum of the maximum number recorded in any month
0r a species in each plot. Minimum numbers of individuals per plot showed little variation with elevation
habitat type (F = 0.33; df = 5, 41; P > 0.10; Appendix S2). Mean (± Standard Delation; SD) densittes^ for
|*7’ Actively. Densities declined with increasing elevations at a rate of 0.38 individuals per m2 per 100
®°f elevation, but this rate was not significantly different from 0. Densities were approximately normally
Abated (Kohnogorov-Smirnov test of normality; Conover 1971).
distribution of orchid species among plois.-Only 128 of .he 239 species were observed in plots
**"* “* 12-month study. Because almost all orchid individuals observed in plots (lowered durtng the
£*• ,h“ nnderrepresentation was due to inadequate sampling of the spatial heterogeneity of the orchtd
J'W' »»mber of species per plot ranged from three to twenty-five with a median of nine species per
Mosl of orchid species (72%) occurred as epiphytes.
For terrestrial species, there were significant differences in the number of terrestrial species per pb*
among habitats (F = 6.15; df = 2, 43; P < 0.01) and at different elevations (F = 14.09; df = 1, 43; P < 0.01). the
mean (± SD) number of terrestrial species per plot for tall cloud forest was 1.8 (± 2.5). This was significantly
(P< 0.05) less than the means of 4.0 (± 2.7) species per plot in short cloud forest and 3.4 (± 2.6) species pet
plot in grassland. There was a decrease of 0.81 (SE = 0.22) terrestrial species per plot for each 100 m increa*
in elevation. The two terrestrial orchids with the highest frequency of occurrence, Epidendrum secundum
Elleanthus sp., occurred in 17 of the 47 plots.
For epiphytic species, there were also significant differences in the number of species per plot among
habitats (F = 7.29; df = 2, 43; P < 0.01) and at different elevations (F = 19.12; df = 1, 43; P < 0.01). The tin®'
ber of epiphytic species per plot for short cloud forest (10.1 ± 4.5) was significantly greater than that for
tall cloud forest (6.3 ± 3.7) and the grass areas (6.2 ± 4.0). For each 100 m increase in elevation, there
a decrease of 1.4 (SE = 0.3) epiphytic species per plot. The epiphytic orchid with the highest frequency
occurrence was Pleurothallis acuminata which occurred in 20 plots.
There was considerable variation among species in abundances and frequencies of occurrence. T ere
27 species with an abundance of one individual and 12 species with an abundance of two. The majority
orchid species had mean densities < 2 per m2. Only nine species had mean densities > 2 per m2. These ^
listed in order of decreasing abundance, Pachyphyllum sp.l, Pleurothallis aff. vestigipetala, Stehs brenrac ^
Elleanthus sp. 1, Stelis grandibracteata, Maxillaria alpestris, Epidenrum secundium, Pleurothallis acummuW^
Pachyphyllum pectinatum. Most (77%) species were observed in < 5 plots (Fig. 3). Forty-nine species
found in only one plot, and 22 species occurred in only two plots. Only six species were observed m >
Lukeetal., Diversity and abundance of Orchids 323
NUMBER OF PLOTS
°f the 10 plots. These, listed in order of decreasing frequencies of occurrence, were Pachyphyllum sp. 1 , Pleu-
r°thdlis acuminata, Stelis breviracema, Epidendrum secundum, Elleanthus sp. 1, and Pachyphyllum pectinatum.
Rarefaction Analysis.— Both the Chao 1 and Chao 2 procedures consistently underestimated the
^cies richness of 239 observed in the systematic collections (Fig. 4). The Chao 2 procedure, which defines
r»eness based on incidences of occurrence, was consistently greater than the Chao 1 procedure, which
defines rareness based on the abundance of individuals. The final estimate for the Chao 2 procedure, based
00 randomizations of all 47 plots was 183 species with a 95% Confidence Interval (95% Cl) of 155 to 240
Species- The final estimate for the Chao 1 procedure was 158 species with a 95% Cl of 140 to 206 species.
Flowering Phenology.-The percent of species in the plots that were in flower ranged from 13% in
Member to 51% in February (Fig. 5), and there was little differences in percents among habitat types or
SCross Ovations. The mean (± SD) density of flowers also ranged from a minimum of 0.65 (± 1.1) per m2
September to a maximum of 25.7 (± 37.6) per m2 in February. Peak flowering occurred during months
lh> 100 mm of rainfall, and minimum flowering occurred in months with < 5 mm of rainfall. The me-
aian lowering duration was three months and most (80%) species flowered for less than six months. The
j*°P°mon of flowering species shared between successive months was S 0.40 (Fig. 6), and there were no
jounced peaks in the proportions shared that indicated major subsets of species exhibiting synchronous
Journal of the Botanical Research Institute of Texas 4(1)
DISCUSSION
The cloud forests at WCFR harbor a rich diversity of orchids with 130 species and 109 morphospecies, <*
239 unique taxa, collected and documented during the 12-month sampling period. Most of the orchids wp*
relatively rare. Over 75% of them were either found only during the systematic collections or only occunw
in s 2 of the 47 plots. Those that did occur in plots were often represented by < 3 individuals. Only nffl*
of the species could be described as being abundant due to their occurrence in more than one-third of
plots or their mean densities being > 2 individuals per m2. These included the identified species Epidendnu*
secundum, Maxillaria alpestris aff., Pachyphyllum pectinatum, Pleurothallis acuminata, Pleurothallis vesti^*T
aff., Stelis breviracema, and Stelis grandibracteata and the morphospecies Elleanthus sp. 1 and Pachyphy™*
sp. 1. Elleanthus sp. 1 and Epidendrum secundum were primarily terrestrial species. The Pachyphylh‘m
Pleurothallis species were primarily epiphytes. The remaining three abundant species occurred with si
abundances in both terrestrial and epiphytic growth forms. ^
Similar intensive, 12-month assessments of orchid species richness in the Peruvian Andes have
made near the Winay-Wayna ruins (WW; Zambrano et al. 2003a) at MPHS which is about 100 k®
of WCFR and in a small portion of Manu National Park (MNP; Zambrano et al. 2003b) to the northe
WCFR. In addition to these intensive analyses of smaller sites, Christenson (2003) has compiled a
list for the 35,952-ha MPHS based on collections since that site’s discovery in 1911. Comparing the
compositions of these sites and WCFR is complicated because: (1) the MNP and WW studies include .
chid species from elevations above and below the cloud forest; (2) a number of unidentified or undesc
morphospecies still exist; and (3) variation in the use of nomenclature and taxonomic classifications
study sites and researchers. Despite these complications, three general trends in species richness and
position are apparent. ^
First, the intensive studies of small areas have found comparably large numbers of orchid species-
Luke et al„ Diversity and abundanc
60
SONDJ FMAMJJA
MONTHS
WCFR (Christenson 2003). The number of species at each of these <
been reported for all of MPHS. As many as 50 species of epiphytic o
forest elevations (Kromer et al. 2005).
it of the species o
ied sites belong to the large genera Epidendrum
Epidendrum is the most numerous genus at the MNP and WW sites, but Stelis is the most numerous
•T at WCFR- Epidendrum is the second most diverse genus at WCFR, but Stelis is less diverse than the
°ther three genera at the MNP and WW sites.
Third, although the important genera appear to be similar across sites, the species representing these
®tnera differ among sites. No species list is provided for the WW site, but Zambrano et al. (2003b) listed the
63 species that occurred at MNP which are also known to occur at MPHS. Apparently, the remaining 149
Spreies at MNP could not be matched to MPHS species. Of those 63 species common to MNP and MPHS,
r*28 occurred at WCFR. Thirty-six of the 63 species common to MNP and MPHS were members of
Vwndrum, Maxillaha, Pleurothallis and Stelis, but only 17 of these occurred at WCFR.
r ^are other differences in species compositions between WCFRand MPHS. Of the 79 genera reported
both WCFR and MPHS (Christenson 2003) only 43 occur at both sites, and 25 of these are represented
CFR hy only a single species. For those genera represented by numerous species at WCFR, there were
Wdendrum species, 16 Maxillaria species, and six Pleurothallis species that are not reported from MPHS.
"t®***1 species that occur at WCFR but not at MPHS may be discovered as progress is made on accurate
Rations of the numerous morphospecies in the difficult Stelis. In addition, a number of WCFR species
additions to the flora of Peru or previously undescribed species (Christenson & Repasky 2008).
PROPORTION OF SHARED SPECIES
326 Journal of the Botanical Research Institute ofTexas4(1)
MONTHS
Fis. 6. The proportions of species flowering in one month that were also flowering in the following month.
At present, the species lists from MPHS and the sites of intense study are analogous to the earl^
species lists for the Amazonian rainforests. They suggest a small-scaled and as yet unpredictable stru ^
of the orchid community. The only quantitative data on abundance which might imply commonality
structure are from WCFR, and two aspects of these data do not suggest a simple communality amony
sites. First, only four of the seven named abundant species at WCFR, Epidendrum secundum,
alpestris, Pachyphyllum pectinatum, and Pleurothallis acuminata, were among the 28 species that are c0
to WCFR, MNP and MPHS. Second, Pachyphyllum sp. 1 and Elleanthus sp. 1, two of the abundant spec*’
at WCFR, were not readily-identifiable or familiar species. If they were similarly abundant at other s
may be expected that they would be among the commonly-known and readily-recognized specie*
This lack of correspondence of abundant species among sites implies local variation in the a u" ^
orchid species. Large variation in orchid species composition between sites only kilometers aPa
been reported by Kuper et al. (2004) where 70 % of the species occurring at one site are not presen
Lukeetal., Diversity and abundance of Orchids 327
neighboring site, and perhaps such local variation should be expected in a largely epiphytic group where
occurrence may be related to the species composition, size, height or age of the host surfaces (Catling &
Lefkovitch 1989; Andersohn 2004; Arevalo & Betancur 2006; Trapnell & Hamrick 2006; Burns 2007).
The results of the plot studies also support the possibility that local variation in forest structure and
composition may influence orchid species compositions and abundances. Although the total number of
orchid individuals appears to be similar across elevation and habitats at WCFR, species richness decreases
at higher elevation and within stands of taller cloud forests. Although more forest structure may occur in
taller cloud forests, fewer epiphytic species occur in these forests than in shorter stature cloud forests. Taller
distinct orchid floras at WCFR were obfuscated by the large proportion of orchid species which occurred
in < 5 plots, the possibility exists that such orchid-forest flora associations may occur and contribute to
site-to-site variability in species composition and abundance. Unfortunately none of the studies at MNP,
WW or WCFR have collected or reported data on tree species, tree densities, or other aspects of canopy
architecture, which might account for variations in orchid diversity, abundance, and distribution.
These results document considerable local variation in the orchid floras of Peruvian cloud forests and
suggest that this level of variation may extend to the abundant members of the flora. There is little evidence
that the abundant orchid species at WCFR are also abundant, or perhaps even present, at other sites. Until
similar analyses of species richness and orchid abundances are available from other sites, the issue of whether
there is less site-to-site variation among the abundant members of the orchid flora than that for the flora in
general will remain unresolved.
implications of the results and experiences for the design of similar studies
Varying sampling protocols using plots have been employed or recommended for analyzing orchid floras.
Zambrano et al. (2003a) used rectangular 25 m2 plots at WW, and Zambrano et al. (2003b) used 2 m x 50
m Ptas at MNP. Gradstein et al. (2003) suggested 20 m x 20 m plots for sampling the < 10-m tall understory
of tropical rain forests. The results and experiences of the present study with 5 m x 5 m plots suggest three
Potentially useful modifications for studies attempting to determine the species composition, species rich-
ness’ and relative abundance of cloud forest orchids.
First, plots alone are unlikely to be sufficient to determine the species composition of the study area.
Atoough almost all orchid stems with the plots at WCFR flowered and were assigned to morphospecies, the
^contained less than 60% of the species observed in the systematic collection. Zambrano et al. (2003a)
^PPfement their plot analyses with orchid collections from the area surrounding the plots. The rar-
rftttion analyses also suggest that plots alone are unlikely to be effective in estimating the total number of
^des present.
Second, the assessments of species composition and abundance in plots must be performed on a monthly
r5' is too little overlap in species flowering in successive months to permit less frequent sampling,
^use a single orchid flower may persist for as few as six to eight days (Tremblay et al. 2006), it is possible
1 ^en more frequent than monthly sampling may be required.
ihird, the plot design and manner of data collection used here can be modified to reduce disturbance
J^Prove efficiency. It was necessary to enter the 5 m x 5 m plots to assess species types and abundances.
care> such intrusion may affect the plots composition through time. A design using a rectangular
r*Wllh a width of 2 m would allow data collection with little need for intrusion. Given the large disparity
*een the abundance of rare and common species, enumerating the exact abundance of individuals in
'enW Plots may not be as informative as assessing relative abundance in many plots. A more rapid assess-
relative abundance using an ordinal scale where, for example, 0 indicates no individuals present,
J:1Cates < 3 individuals, 2 indicates 3 to 10 individuals and so forth, would be sufficient to establish a
nt of rare to abundant species. A plot composed of a linear arrangement of five 2 x 2 m subplots with
^estimated abundance being the median ordinal score of the five subplots could be more rapidly sampled
provide sufficiently useful data from more plots in less time.
||SZ
Lukeetal., C
329
mungoschraderi aff.
notylioglossa
procurrens
rotundilabia
32 Prescottia
33 Prosthechea
34 Pterichis
35 Rusbyella
36 Sauroglossum
37 Scaphyglottis
42 Sudamerlycaste
43Telipogon
santiagocastroviejoi
48 Unknown genus
miiHHiiitiE
ACKNOWLEDGMENTS
We extend our special appreciation to Sy Sohmer, Director of the Botanical Research Institution of Texas,
along with the board, administration, development, herbarium, and staff, for institutional support and infra-
structure during all phases of our work in the Andes-Amazon region of southeastern Peru. Rebecca Repasky
thanks the TCU Biology Department for supporting her graduate research and education experience. She also
thanks her parents, James and Karen Repasky for support and encouragement, and especially for allowing
her to live alone in a Peruvian cloud forest during the 12 months of this project. She thanks her husband
for his constant support in finishing the writing of this publication. We are grateful to Barney Lipscomb
and the staff of the BRIT Press for seeing this manuscript through from first draft to final publication. We
thank Harold Koopowitz and an anonymous reviewer for their critical feedback on an earlier draft of the
paper. We thank specific BRIT staff members for their input during research and writing, including Mathias
Tobler, Renan Valega, Amanda Neill, Jason Best, and Tiana Franklin. We are grateful to Adrian Forsyth for
encouraging and supporting, over the years, our studies of the biological diversity of the Andes-Amazon
region of southeastern Peru. We are grateful to the staff of the Amazon Conservation Association and their
Wayqecha Cloud Forest Research Station for logistical support during our field research. We thank Martin
Ccana Avendaflo, Lucio Ccoyo Cordova, and Madeleine Perez Quijano de Janovec for assisting, through
rain or shine, in various phases of field research. We owe special gratitude to the former Instituto Nacional
deRecursos Naturales (INRENA) of Peru for research, collection, and export permits. We also appreciate
support and cooperation from Joaquina Alban, Asuncion Cano, Betty Millan, Miguel Chocce, and other
colleagues of the San Marcos Herbarium in Lima, Peru. Lastly this project would not have been possible
without generous support from the Gordon and Betty Moore Foundation, the U.S. National Science Foun-
dation Biotic Surveys and Inventory Program (grant number DEB-0717453), the Discovery Fund of Fort
Wonh, Texas, and the Beneficia Foundation.
REFERENCES
Anohsohn, C. 2004. Does tree height determine epiphyte diversity? Selbyana 25:1 01 - 1 1 7.
Akvalo, R. and J. Betancur 2006. Vertical distribution of vascular epiphytes in four forest types of the Serranfa de
Chiribiquete, Colombian Guayana. Selbyana 27:175-185.
Burns, K.C. 2007. Network properties of an epiphyte metacommunity. J. Ecol. 95:1 1 42-1 151.
Cwung, RM. and LP. Lefkovitch 1 989 Associations of vascular epiphytes in a Guatemalan cloud forest. Biotropica
2135-40.
A- 1984. Non-parametric estimation of the number of classes in a population. Scand. J. Stat. 1 1:265-270.
9^ A- 1987. Estimating the population size for capture-recapture data with unequal capability. Biometrics
Jenson, E. 2003. Machu Picchu: Orchids. PROFONAPE, Lima, Peru.
°*istenson, E. and R. Repasky. 2008. A new species of Stellilabium from southern Peru. Orchid Digest 72:1 70-1 72.
W* and J.K. Smith. 2006. Applied statistics and the SAS programming language, Fifth Edition. Pearson
™ceHaltNJ.
RK- 2006. Estimates 8.0 User's Guide. http//viceroy.eeb.uconn.edu/estimates.
1971. Practical nonparametric statistics. John Wiley and Sons, Inc. New York.
LofiNaissEN, J.H.C. AND H. Ter Steege. 1 989. Distribution and ecology of epiphytic bryophytes and lichens in dry
evergreen forest of Guyana. J. Trap. Ecol. 5:1 3 1 -1 50.
jJgAK 1988. Changes in plant community diversity and floristic composition on environmental and geo-
9raptiical gradients Ann Mjssouri Bot Gard 75;i _34.
^A.H. andCH. Dodson. 1987. Diversity and biogeography of neotropical vascular epiphytes. Ann. Missouri
** Gard. 74205-233.
^ N. and RK. CawELL 2001 . Quantifying biodiversity: Procedures and pitfalls in the measurement and
^Parison of species richness. Ecol. Lett. 4:379-391.
THE FERNS AND LYCOPHYTES OF A MONTANE TROPICAL FOREST
IN SOUTHERN BAHIA, BRAZIL
Fernando B. Matos
Andre M. Amorim
Centro de Pesquisas do Cacau - CEPEC
Universidade Estadual de Santa Cruz - UEl
Departamento de Ctencias Bioldgicas
IMus, 45.662-900, Bahia, BRAZIL
aamorimm@terra.com.br
Paulo H. Labiak
Universidade Federal do Parand - UFPR
Departamento de Botdnica
Caixa Postal 19031
Curitiba, 81531-980, Parand, BRAZIL
RESUMO
INTRODUCTION
Bahla one of the richest states in Brazil in terms of numbers of plant species and contains representative
*** of almost all of the phytogeographic domains in the country (Brazao & Araujo 1981). According to
Harley and Mayo (1980), two factors that contribute greatly to this species richness are the topography and
lhe tension of the state, which covers an area of approximately 560,000 km* (larger than France, with
a 5H0OO km*). This ample territory likewise contains considerable climatic diversity, principally as one
***** from the coast to the continental interior. Humid Atlantic Forests are found in the eastern part of
the state, while semi-arid regions (including the “caatinga” [dryland], “cerrado” [savanna], riparian forests,
*aS0nal forests, and “campos rupestres” [open, rocky mountain vegetation]) occupy more than 50% of the
* (Giulietti «
l. 2006).
Although this floristic richness has attracted t
19th century (for a more detailed account, see Urban, 1906), few collections have been made in the humid
forests of southern Bahia.
Remnants of the Atlantic Coastal Forest in southern Bahia are known to be among the most important
sites for conservation of biodiversity in the world. Indeed, these forests seem to hold higher levels of specks
richness and plant endemism than any other part of the Brazilian Atlantic Rain Forest (Mori et al. 1981;
Thomas et al. 1998, 2008; Martini et al. 2007). Southern Bahia is also considered one of the most remark-
able Pleistocene forest refuge areas, and a center of diversity for several major groups of organisms (Mori et
al. 1983; Whitmore & Prance 1987; Gentry 1992).
Most botanists studying this region have directed their efforts towards angiosperms, with only rare
studies of ferns and lycophytes, although the plant lists prepared by Mori et al. (1983) and Amorim et al.
(2005, 2008) are notable exceptions. Additional recent taxonomic information has also become available
concerning specific groups of ferns in Fernandes (2003), Labiak and Prado (2003, 2005a, 2005b, 2005c,
2007), Matos et al. (2009, 2010), Prado (2000), and Sundue and Prado (2005). In addition to species lists of
the ferns and lycophytes of southern Bahia State, Paciencia and Prado (2004, 2005a, 2005b) have described
some of the ecological aspects of the area.
Considering these survey limitations and the fact that essentially all of the floristic studies undertaken
in southern Bahia have focused on the diversity of lowland vegetation, it is clear that the regional flora con-
tinues to be insufficiently studied. In an effort to further our floristic knowledge of the Brazilian Atlantic
Forest, the present study examined the fern and lycophyte species in a fragment of montane humid forest
The Serra Bonita Mountain Range occupies an area of approximately 7,500 ha in the municipalities of Ca-
macan and Pau-Brasil, 130 km from the city of llheus and 526 km from the state capital, Salvador (Fig. D
The Serra Bonita Private Reserve is located in the central region of this range (15°23'S x 39°33'W)
and was founded to help protect regional biodiversity. The reserve occupies an area of approximately 2,000
hectares at elevations that vary from 300 to 1,080 m (Amorim et al. 2009).
Roeder (1975) described the climate of the region as warm and humid, with an average annual rainfall
of 1 ,500-1 ,800 mm that is well distributed throughout the year. Average annual temperatures vary between
23 and 24°C, the total annual evapotranspiration potential is 1,200-1,300 mm, and the relative humidity
reaches 80%. There is no specific information currently available for precipitation and temperature wit
the study area itself, although there are obvious climatic variations between the highland and lowland areas,
where Roeder’s earlier collections (1975) were made.
Both the degree of preservatic
- physi
tructure of the vegetatio
lin the reserve vary
_ „ act forests while others have forests in different stages of regeneration
(areas that were harvested for timber or cleared for agriculture). Additionally, a road leading to a mountaintop
transmission tower cuts through part of the Reserve, resulting in the eradication of the original vegetation
in that area (Amorim et al. 2009).
Recent research has shown that the few remaining moist forest fragments in the region shelter
numbers of rare species, including two endangered primates: Cebus xanthosternos (yellow-breasted capuc
and Leontopithecus chrysomelas (golden-headed lion tamarin) (Vitor O. Becker, pers. comm.). PrellI°_
floristic studies have identified ca. 700 species of angiosperms (Amorim et al. 2009), with several spec**
new to science, including a very notable bromeliad (Amorim & Leme 2009).
However, despite its remarkable regional biodiversity, the harvesting of native trees for timber coni
to be a major contributor to forest destruction in this area.
Matos et al.( Ferns and lycophytes in southern Bahia, Brazil
Journal of the Botanical Research Institute of Texas4(1)
able for this area are some ferns and lycophytes that were collected by Talmon Soares dos Santos (in 1969,
1971, and 1979), Scott Alan Mori (in 1979), Andre Mauricio Vieira de Carvalho (in 1983), and Wm. Wayt
Thomas (in 1994 and 2002). Complete sets of all of these collections are deposited in the CEPEC herbarium,
and duplicates have been sent to NY and UPCB.
While duplicates of some collections were sent to specialists of other institutions for identification, the
great majority of the plants were identified by the first author using well-documented collections at CEPEC,
NY and UPCB (among others) and specific bibliographies.
The ferns and lycophytes are arranged alphabetically in the checklist by family, genus, and then species.
The fern families were circumscribed according to the system proposed by Smith et al. (2006). Authors of
species names are abbreviated according to Pichi Sermolli (1996). Occasionally, a specimen could be reliably
identified only to genus but was clearly distinct from any of the identified material; these specimens were
included in the list as “sp.” In the present work, the concept of Atlantic Forest was treated in a broader sense
(Oliveira-Filho & Fontes 2000), with the result that the areas of distribution of some species extended into
the continental interior, sometimes reaching neighboring countries (Argentina and Paraguay).
A total of 182 species in 67 genera and 23 families of ferns and lycophytes from the Serra Bonita Reserve
are presented here (Appendix 1). The ferns comprised 173 species in 64 genera and 21 families, while the
lycophytes included nine species in three genera and two families. Only three species remained unidenti-
fied ( Ctenitis sp., Elaphoglossum sp., and Thelypteris sp.), but they are distinctly different from all of the other
species in their respective genera.
The most diverse fern families were Polypodiaceae (35), Dryopteridaceae (28), Pteridaceae (19), and
Hymenophyllaceae (16). Together, these four families comprise about 57 percent of the species found in
Serra Bonita. The most species-rich genera were Asplenium (12), followed by Elaphoglossum (11), Thelypterts
(10), and Diplazium (9). ' '
Several specimens collected during this study have been identified as new species or are currently being
described by specialists, including: Asplenium truncorum, M egalastrum indusiatum, and Thelypteris beckeria*
(Fig. 2C-E). £
Terpsichore asplenifolia (Fig. 2F) was recorded for the first time in Brazi
junct distribution between the Venezuelan-Guyana Shield and the Serra do 1
Brazil (Labiak & Prado 2007). Forty-three additional species were recorded for the first time for Bahia or
northeastern Brazil (comprising the states of Alagoas, Bahia, Ceara, Maranhao, Paraiba, Pernambuco, Pi*
Rio Grande do Norte, and Sergipe), representing 24.2% of the total number of species found in the rescn*
These new records are indicated with asterisks in the checklist (Appendix 1). In general, these new reco
represent species characteristic of montane areas of the Atlantic Forest of southeastern Brazil, and are
expected to occur in the other states of northeastern Brazil. Polypodium dulce, Pteris schwackeana,
decomposita, S.Jlexuosa, and Serpocaulon levigatum have been previously reported from other states in no
eastern Brazil, but are new records for the Atlantic Rain Forest of southern Bahia.
In terms of habitat preferences, many representatives of the local flora are typically associate
disturbed environments (e.g., roadsides, forest edges, forest regrowth, cabrucas [cocoa plantations
native forests] , and other anthropogenic sites), and include: Adiantum latifolium, Asplenium auritum, B
brasiliense (Fig. 3), B. occidental Dennstaedtia globulifera, Dicranopterisflexuosa, GleichenellapectinataM
Hemicmitis tomentosa, Hypolepis aff. repens, Lycopodiella cemua, Lygodium volubile, Osmundastrum cinna
Pityrogramma calomelanos, Pleopeltis pleopeltifolia, Pteridium arachnoideum, Sticherus bifidus, 5. lanugos
Thelypteris opposita, and T. serrata.
ortheastern
atedwith
The great majority of the ferns and lycophytes identified in the present study (more
c chai
an 70%). h****
ats, 49% of the species inventoried were exclusively terrestrial (89 spP*
including herbacec
(9 spp.), and 2% w
type of habitat.
Ferns are the third most species-rich group of epiphytes in the New World, after orchids and bromeliads
(Gentry & Dodson 1987). The last two groups are represented in the Serra Bonita Reserve by 62 and 25
species respectively (Amorim et al. 2009), thus making the ferns the second most diverse group of epiphytic
plants in the area (59 spp.).
Several authors (e.g., Moran et al. 2003; Mehltreter 2008) have observed that some epiphytic fern spe-
cies demonstrate notable specificity in relation to the type of substrate (i.e., phorophytes) on which they can
grow. The arboreal ferns of the family Cyatheaceae stand out among the phorophyte supports available in
tropical humid forests, as they generally have a dense mass of adventitious roots surrounding their trunks,
which can store considerable quantities of water and nutrients due to their high porosity (Mehltreter 2008).
In Serra Bonita, five epiphyte species were observed growing exclusively on this type of substrate (notably
on Alsophila setosa ), including Asplenium truncorum, Pecluma truncorum, Polyphlebium angustatum, Terpsichore
asplenifolia, and Trichomanes polypodioides. However, most epiphytic species inventoried did not appear to
depend exclusively on these ferns and were collected on a wide variety of angiosperms.
Among the terrestrial species, it is clearly noticeable the preference of many narrowly distributed spe-
cies to the forest understory, occurring exclusively in well-preserved forest remnants. The high diversity of
epiphytic ferns, and the large number of terrestrial species in the forest understory, show the importance
of well-preserved environments not only for the richness of species in numerical terms, but also for the
maintenance of the narrowly distributed species populations.
Geographical distribution and endemism
According to Tryon (1972), Brazil is one of the principal centers of fern and lycophyte diversity and endemism
in the Neotropical region, with about 40% endemism.
The number of species endemic to the Atlantic Rain Forest that were encountered in the Serra Bonita
Reserve is quite high (59 species, or 32.5% of the total), especially when compared with other studies un
dertaken in the Atlantic Rain Forest domain (e.g., Labiak & Prado 1998; Salino et al. 2005; Schwartsburd
& Labiak 2007; Amorim et al. 2008).
appear to constitute the northern limit of occurrence for many ende
Forest. Some of these species can be considered characteristic of the montane regions of southeastern
southern Brazil (e.g., Adiantum mynssenae, Asplenium scandidnum, Cyathea rufa, Huperziaflexibilis, H.
Micropolypodium achilleifolium, Pleopeltis pleopeltidis, Pteris angustata, Terpsichore reclinata, Thelypteris r
and Vittaria scabrida, among others).
Paradoxically, the southern region of Bahia represents the southern limit of distribution of some specif
endemic to the Atlantic Forest, as they also occur in the “Brejos de altitude” (high elevation dryland foreSt^
of northeastern Brazil. Two examples of this unusual situation among the taxa recorded for the Serra Bon1
Reserve are Cyathea praecincta and Megalastrum eugenii.
Although ferns and lycophytes are usually capable of dispersing over large areas, mountain chains
been found to represent frequent barriers to their migration (Tryon 1986; Moran 1995, 2008). In
favoring high levels of endemism, mountains can also influence the taxonomic richness of a give11 TeP^
Entire families can sometimes be considered typical (but not exclusive to) high elevation regions andha ^
fundamental role in the floristic characterization of those areas (Moran 1995; Paciencia 2008). Acco %
these authors, the main examples of these families are Cyatheaceae, Dryopteridaceae, Hymenophy **
Lycopodiaceae, and Polypodiaceae. In the same sense, many tropical genera such as Diplazium and E qr
lins of southern Bahia
d lycophytes
n the Atlantic
Dincidence that they show the
greatest species richness in the Serra Bonita Reserve.
It is also important to note the existence of species with disjunct distributions between the Atlantic
and Amazon (and/or Andean) forests. According to Tryon (1986), a given fern species can be considered
disjunct when the distance between its populations reach about 1,000 km or more, even though there are
adequate environments for their occurrence in the intervening regions. Doryopteris sagittifolia (Pteridaceae),
for example, occurs in the mountains of eastern Brazil but has also been reported in isolated populations in
various countries in northern South America (Guyana, French Guyana, Suriname and Venezuela). Terpskhort
asplenifolia (a species widely distributed from southern Mexico to Bolivia and the Antilles) occurs disjunctly
in Brazil, and the Serra Bonita Reserve is the site of the only known population in the country (Labiak &
Prado 2007). Also noteworthy are Elaphoglossum gardnerianum, Hymenophyllum caudiculatum and Pleopeltis
monoides, Brazilian endemics showing disjunct distributions between the Atlantic and Amazon forests; 58
additional species encountered in our study occur in both southern Bahia and in Amazon forests.
Some Brazilian species have wide geographical distributions throughout many parts of the world (e.g.,
Adiantum latifolium, Asplenium auritum, Hymenophyllum polyanthos, Lycopodiella cernua, and Pityrogramma
number of species shared by both the Atlantic and Amazon forests (a total of 61 species) is still quite notable,
however, especially in view of the fact that these two extensive forest formations are separated by a “diagonal
of drought” — a corridor of highly seasonal and dry formations (e.g., the “Caatinga” of northeastern Brazil,
the “Cerrado” savanna of Central Brazil, and the “Chaco” of Paraguay, Argentina and Bolivia) (Prado &
Gibbs 1993) that “isolates” the Atlantic Rain Forest from other rain forests in South America (Andean and
Amazon forests).
Although reported by many authors (e.g., Andrade-Lima 1953, 1966, 1969; Bigarella et al. 1975; Moriet
al. 1981; Rizzini 1997), the disjunction observed between species of the floras of northern and eastern Brazil
must be evaluated with some caution, as the differentiation between long distance dispersal and vicariance
in the historical biogeographical patterns of fern distribution requires still more intensive examination (Wolf
et al. 2001).
Invasive species
In spite of the relatively recent introduction of Macrothelypteris torresiana, Nephrolepis brown ii
and Thelypteris dentata into the tropical Americas, these species are already found in many areas °f Sou*
America (Mickel & Smith 2004; Hovenkamp & Miyamoto 2005). These species are primarily associated
with ruderal environments in the Serra Bonita Reserve (e.g., along forest edges, disturbed areas, and areas
under cultivation) and apparently are not yet seriously competing with the local flora for habitats.
The low number of unidentified species encountered in the present study (three) is due, in large part, to
the collaboration of many specialists in identifying this material (see Acknowledgments). Identification was
especially difficult in very diverse or little-studied genera such as Ctenitis, Elaphoglossum, and Thelypt*™'
each of which had one species of unknown identity.
Floristic relationships
Preliminary analyses of similarities between the flora of southern Bahia and other diverse
America (Matos 2009) indicate that the montane forests of southern Bahia are especial _
“southern” elements of the Atlantic Forest. The large number of new records reported here supports this^
as the majority of these species are characteristic of the montane regions of southern Brazil. The same c
floristic relationships between the montane forests of southern Bahia and the mountains of southeastern B
has likewise been observed for the angiosperms (Amorim et al. 2009). On the other hand, the low elevatio^
forests of southern Bahia appear to be more closely related to the Amazon forests or to northern arej^
characterized by
Matos et al.. Ferns and lycophytes in southern Bahia, Brazil
Conservation
One of the greatest modern challenges to mankind is reconciling economic growth (which usually implies
increased utilization of natural resources) with global biodiversity conservation, and preservation efforts must
be optimized by establishing criteria that guarantee the most favorable possible balance between costs, social
needs and the efficient functioning of established conservation areas (Crozier 1997; Crandall et al. 2000).
Among the notable examples of diversity observed in the region, some species are noteworthy by their
Appendix 1) represent approximately 25% of the total number of ferns and lycophytes encountered in the
Serra Bonita Reserve.
The montane forests of the Serra Bonita Reserve offer ideal environmental conditions for the establish-
ment and growth of ferns and lycophytes, harboring almost all of the taxonomic orders proposed by Smith
ttal. (2006) (the only exceptions being the Psilotales, Equisetales and the heterosporous ferns [Salviniales],
although the last can be found at lower elevations in the region). The conservation importance of the Serra
Bonita Reserve therefore lies not just in the number of species, but also in the phylogenetic diversity it pre-
APPENDIX 1
Checklist of the ferns and lycophytes from the Serra Bonita Reserve, southern Bahia, Brazil. The asterisk before species names
gjtothefirstrecoi.j f . ,, »ri r.-n -io rF-.hn -a i "■ I- ' fer to species newly described, based on material
:-V ; ,-vn-.H.h hi I -urn- ■ i o;.|l-.-.;r, r. v.- i. r;ll . AA - Andre M Arn,,-im FF = Fabric m M Ferreira.
JM = Fernando B. Matos, ML = Mardel M. Lopes, PL = Paulo H. Labiak, and WT - Wm. Wayt Thomas. Abbreviation for the
habitat of the species: EP = Epiphytic HE = Hnir.i-ec- pnvtir RU - Ru(* m,,i and TE = Terrestrial.
FERNS AND LYCOPHYTES
Anemiaceae
Sw.— FM 625, RU/TE
^iaphyllitidis (L) Sw.-FM 1 789, TE
Blechnum acutum (Desv.) Mett.— FM 292, HE/TE
Blechnum brasiliense Desv.— FM 614, TE (Fig. 3)
Blechnum occidentale L— FM 3 1 6, TE
'Blechnum organense Brade— FM 1064.TE
Salpichlaena volubilis (Kaulf.) J.
n.— FM1366.TE
^ich. ex Willd. — FM 1538.TE
‘X F£e — FM 448, EP
kunzeanum Klotzsch ex Rosenst.— PL 3730, TE/
Sgil 11562, EP/TE
mpteropus Kaulf.— FM 1567, EP
^scandidnum Kaulf. — FM 299, EP
mserra Langsd. & Fisch.— AA 5435, EP
m t!iWetrum N. Murak. & R.C.Moran— FM 313, RU
H toncorum F.B. Matos, Labiak & L. Sylvestre— FM
7, EP (Fig. 2)
Alsophila setosa Kaulf.— FM 447, TE
Alsophila stembergii (Sternb.) D.S. Conant— FM 1 384, TE
Cyathea corcovadensis (Raddi) Domin— FM 1550,TE
Cyathea delgadii Sternb. — FM 1 382, TE (Fig. 3)
Cyathea glaziovli (F6e) Domin— FM 636, TE
Cyathea phalerata Mart— FM 619,TE
Cyathea praecincta (Kunze) Domin— FM 1088,TE
'Cyathea rufa (F6e) Lellinger— FM 1 103.TE
Dennstaedtia dissecta (Sw.) T. Moore— FM 295, TE
Dennstaedtia globulifera (Poir.) Hieron.— FM 1 529, TE
Dennstaedtia obtusifolia (Willd.) T. Moore— FM 1063.TE
Histiopteris incisa (Thunb.) J. Sm.— PL 371 5, TE
342
Hypolepis aff. repens (L) C. Presl— FM 1522, TE
Pteridium arachnoideum (Kaulf.) Maxon — FM 1368.TE
Dryopteridaceae
'Ctenitis aspidioides (C. Presl) Copel — FM 1332,TE
Ctenitisdistans (Brack.) Ching— FM 1372.TE
Ctenitis submarginalis (Langsd. & Fisch.) Ching — FM 264, TE
Ctenitis sp.— FM 1 787, TE
Cyclodium heterodon (Schrad.) T. Moore— FM 1 095, TE (Fig. 3)
Cyclodium meniscioides (Willd.) C Presl— FM 321 ,TE
Didymochlaena truncatula (Sw) J5m. — FM 1 383, TE
Elaphoglossum sp.— FM 1329, EP
'Elaphoglossum decoratum (Kunze) T. Moore — FM 445, RU
'Elaphoglossum gardnenanum (Kunze e>
446, EP
Elaphoglossum glabellum J. Sm.— FM 607, RU
Elaphoglossum insigne (Fee) Brade— FM 620, EP
Elaphoglossum lingua Brack. — FM 1 060, RU
'Elaphoglossum peltatum (Sw.) Urb. — FM 1584, EP (Fig, 2)
'Elaphoglossum rigidum (Aubl.) Urb.— FM 1331, EP
'Elaphoglossum strictum (Raddi) T. Moore— JP 449, EP
'Elaphoglossum vagans (Mett.) Hieron.— FM 1583, EP
'Elaphoglossum villosum (Sw.) J. Sm.— FM 1572, EP
Lastreopsis amplissima (C. Presl) Tindale — FM 284, TE
a guianensis (Aubl.) Ching— FM 1047, HE
-JP
n— FM
Prado & Labial
°Megalastn
1 365, TE (Fig.
* Megalastrum umbrinum (C Chr.) A.R. Sm. & R.C. Moran— FM
1076,TE
Otfersia cervina (L) Kunze— FM 1 096, RU/TE
* Polybotrya speciosa Schott— ML 788, HE
Stigmatopterisprionites (Kunze) C Chr.— FM 1548.TE
Dicranopteris flexuosa (Schrad.) Underw.— FM 1804, TE
Gleichenella pectinata (Willd.) Ching— FM 652, TE (Fig. 3)
Sticherus bifidus (Willd.) Ching— FM 657, TE
Sticherus lanuginosus (F£e) Nakai— FM 654, TE
Hymenophyllaceae
Abrodictyum rigidum (Sw.) Ebihara & Dubuisson— FM 301,
Hymenophyllum caudiculatum Mart— FM 303, EP
'Hymenophyllum elegans Spreng. — PL 371 1, EP
Hymenophyllum hirsutum (L) Sw.— FM 646, EP
Hymenophyllum polyanthos (Sw.) Sw. — FM 645, EP
Polyphlebium angustatum (Carmich.) Ebihara & Dubuisson—
FM 1068, EP
Polyphlebium diaphanum (Kunth) Ebihara & Dubuisson — FM
650, EP
Polyphlebium hymenophylloides (Bosch) Ebihara & Dubuis-
son—FM 649, EP
Trichomanes elegans Rich. — FM 1 371 , TE (Fig 3)
Trichomanes pilosum Raddi — FM 1 337, EP
Trichomanes polypodioides L— FM 302, EP
Vandenboschia collariata (Bosch) Ebihara & K. Iwats.— FM
304, RU
* Vandenboschia rupestris (Raddi) Ebihara & K. Iwats.— FM
1555, RU
Lindsaeaceae
Lindsaea lancea (L.) Bedd. — FM 1 376, TE
'Lindsaea quadrangularis Raddi — FM1$51,TE ,
Lomariopsis marginata (Schrad.) Kuhn — FM 1 074, HE
Nephrolepis brownii (Desv.) Hovenkamp & Miyam.— FM
610,TE
Nephrolepis cordifolia (L) C. Presl — FM 1 523, TE |
Nephrolepis pendula (Raddi) J. Sm— FM 1588, EP/TE
Nephrolepis rivularis (Vahl) Mett. ex Krug— FM 1589, EP/TE
Lycopodiaceae
Huperzia acerosa (Sw.) Holub — FM 319, EP/RU
* Huperzia flexibilis (Fee) B. 0llg. — FM 1571, EP
* Huperzia heterocarpon (Fee) Holub — FM 1581, EP/RU j
Huperzia mandiocana (Raddi) Trevis— FM 1 553, EP (Fig. 2)
Huperzia martii (Wawra) Holub — FM 1777, EP
Lycopodiella cernua (L) Pic Serm.— FM 653, TE
Lygodiaceae
Lygodium volubile Sw. — FM 265, TE
Danaea geniculata Raddi — FM 1335.TE
Eupodium kaulfussii (J. Sm.) J. Sm.— FM 1067,TE
Oleandraceae
Oleandra articulata (Sw.) C. Presl— FM 440, EP
Ophioglossaceae
Ophioglossum palmatum L — FM 1 375, EP
Osmundaceae
Osmundastrum cinnamomeum (L) C. Presl— PL. 3731 , |
Polypodiaceae
* Campyloneurum acrocarpon Fee— WT 1 4230, EP
Campyloneurum aff. angustifolium (Sw.) Fee— FM 41 ,
* Campyloneurum decurrens (Raddi) C. Presl— FM 290, j
* Campyloneurum nitidum (Kaulf.) C. Presl— FM 273,
Cochlidium serruhtum (Sw.) LE. Bishop— FM
Lellingeria suspensa (L) A.R. Sm. & R.C. Moran— FM >
'Melpomene melanosticta (Kunze) A.R. Sm. & R-C- Morans*
6497, EP
Microgramma acatallela Alston — FM 1 577, EP ___
Microgramma geminata (Schrad.) RM Tryon & Af. Tryon"
FM315‘EP . !
Microgramma tycopodioides (L.) Copel.— FM 1 U** T
Microgramma percussa (Cav.) de la Sota— FM 1 544,
Microgramma tecta (Kaulf.) Alston — FM 327, EP ^
Microgramma vacciniifolia (Langsd. & Fisch.) Cope .
343
Evans) M. Kessler & A.R. S
EP/RU
Pecluma plumula (Humb. & Bonpl. ex Willd.) M.G. Price— FM
283, EP/RU
Pecluma recurvata (Kaulf.) M. G. Price— FM 1545, EP/RU
Pecluma robusta (Fee) M. Kessler & A.R. Sm. — FM 1381, EP
* Pecluma truncorum (Lindm.) M. G. Price— FM 1 536, EP
Phlebodium pseudoaureum (Cav.) Lellinger — FM 280, EP/
RU (Fig. 3)
Pleopeltis astrolepis (Liebm.) E. Foum. — FM 1 525, EP
Pleopeltis macrocarpa (Bory ex Willd.) Kaulf.— FM 1 527, EP
Pleopeltis monoides (Weath.) Salino — FM 281 , EP/RU
'Pleopeltis pleopeltidis (Fee) de la Sota— FM 1528, EP
Pleopeltis pleopeltifolia (Raddi) Alston— FM 656, EP
Polypodium chnoophorum Kunze — FM 1 526, EP
* Polypodium duke Poir. — FM 1 093, RU/TE
Serpocaulon catharinae (Langsd. & Fisch.) A.R. 5m.— FM
291, EP
* Serpocaulon levigatum (Cav.) A.R. Sm,— FM 629, EP
Serpocaulon meniscifo/ium (Langsd. & Fisch.) A.R. Sm.— FM
1059, EP/RU
Serpocaulon triseriale (Sw.) A.R. Sm.— FM 278, EP/RU/TE
‘Terpsichore asplenifolia (L) A.R. Sm— FM 1 566, EP (Fig. 2)
'Terpsichore reclinata (Brack.) Labiak— FM 1338, EP
'Terpsichore taxi folia (L.) A.R. Sm.— FM 307, EP
Pteridaceae
Miantopsis radiata (L.) Fee— FM 1369. TE
Miantumabscissum Schrad.— FM 1373,TE
Montumdolosum Kunze— FM 1560.TE
i.— FM1388,TE
1786.TE
Tkryopteris collina (Raddi) J.
'jtywteris rediviva Fee — FF 1294,TE
rWerissagittifolia (Raddi) J. Sm.— WT 13030, RU
J"*"* tonentosa (Lam.) Raddi— FM 270, RU/TE
gramma calomelanos (L.) Link— FM 275, TE
'Waenium lineatum (Sw.) J. Sm— FM 263, EP
wnangustata (F<§e) CV. Morton— FM 637, TE
Pteris decurrens C. Presl — FM 1 541 , TE
Pteris deflexa Link.— FM 660, TE
Pteris denticulata Sw.— FM 1387,TE
*Pteris schwackeana H. Christ— FM 1091,TE
Pteris splendens Kaulf.— FM 1 539, TE
*Vittaria scabrida Klotzsch ex F4e — FM 632, EP
Saccoloma inaequale (Kunze) Mett. — FM 1070.TE
Tectaria incisa Cav. — FM 267, TE
Triplophyllum hirsutum (Holttum) J. Prado & R.C. Moran— FM
Macrothelypteris torresiana (Gaudich.) Ching— FM 274, TE
Thelypteris sp.— FM 1559.TE
°Thelypteris beckeriana F.B. Matos, A.R. Sm. & Labiak — FM
1 558, TE (Fig. 2)
Thelypteris dentata (Forssk.) EP. St John— FM 1 53 1 , ' TE
*Thelypteris leprieurii (Hook.) R.M.Tryon — FM 1532,TE
Thelypteris macrophylla (Kunze) C.V. Morton— FM 1 561, TE
* Thelypteris opposita (Vahl) Ching — FM 1524.TE
Thelypteris polypodioides (Raddi) C.F. Reed— FM 1 380, TE
(Fig. 3)
* Thelypteris raddii (Rosenst) Ponce — FM 288, TE
Thelypteris serrata (Cav.) Alston — FM 61 2, TE
Thelypteris vivipara (Raddi) GF. Reed — FM 1 781, TE
Woodsiaceae
Diplazium ambiguum Raddi — FM 658, TE
Diplazium asplenioides (Kunze) C. Presl— FM 296, TE
Diplazium lechleri (Mett.) T. Moore— FM 622, TE (Fig. 3)
Diplazium leptocarpon F£e— FM 1081 , TE
Diplazium lindbergii (Mett.) Christ — FM 1083.TE
Diplazium mutilum Kunze— FM 436, TE
Diplazium plantaginifolium (L) Urb. — FM 1385.TE
ACKNOWLEDGMENTS
We thank the National Science Foundation (DEB 0516233), the Beneficia Foundation, The National Geo-
f 3phic Society (DEB 7785-05), CNPq (Edital Universal 474648-4), and CAPES (master’s scholarship to
tonando B. Matos) for financial support for this research. We are extremely grateful to Vitor O. Becker and
emira O. Souza, the owners of the Serra Bonita Reserve, for their hospitality and help. While we take full
tesponsibility for all names listed here, we are indebted to the many botanists who assisted us by identifying
^ens; these include Alan R. Smith (Thelypteridaceae), Alejandra Vasco ( Baphoglossum ), Claudine Myns-
(THplazium), Jasivia Gonzales (Gleicheniaceae), Jefferson Prado (miscellaneous), John T. Mickel (Anemia
^baphoglossum), Maarten J.M. Christenhusz (Danaea), Michael Sundue (miscellaneous), Regina Y. Hirai
and Robbin C Moran (miscellaneous). We also thank the personnel of the following herbaria for
Tragus to examine material and for loans: ALCB, CEPEC, FUEL, HB, HBR, HUEFS, MBM, MBML, NY,
’ SP> UpCB, and US. This paper is part of the first author’s Master thesis at the Universidade Federal
, RAX. Borges, R.O. Pei
i Bahia, Brazil
345
Lawk, RH. and J. Prado. 2005c. As especies de Terpsichore A. a $m. e Zygophlebia L. E. Bishop (Grammitidaceae)
do Brasil. Acta Bot. Bras. 1 9:867-887.
Lahak, P.H. and J. Prado. 2007. New records of pteridophytes from Bolivia and Brazil. Amer. Fern J. 97:1 1 3-1 23.
Martini, A.M.Z., P. Fiaschi, A.M. Amorim, and J.L. Paixao. 2007. A hot-point within a hot-spot a high diversity site in
Brazil's Atlantic Forest. Biodivers. Conserv. 1 6:3 1 1 1 -3128.
Matos, F.B. 2009. Saimambaias e licbfitas da RPPN Serra Bonita, munidpio de Camacan, sul da Bahia, Brasil. Unpubl.
Master thesis, Universidade Federal do ParanS, Brazil.
Matos, F.B., P.H. Labiak, and LS. Sylvestre. 2009. A new Brazilian species of the genus Asplenium L. (Aspleniaceae).
Amer. Fern. 199:101 -1 05.
Matos, F.B., A.R. Smith, and P.H. Labiak. 201 0. A new species of Thelypteris (Thelypteridaceae) from southern Bahia,
Brazil. Brittonia 62:149-152.
Mboeter, K. 2008. Phenology and habitat specificity of tropical ferns. ln:TA Ranker and C.H. Haufler, eds. Biology
and evolution of ferns and lycophytes. Cambridge University Press, New York. Pp. 201 -221.
Mickel, J.T. and AR. Smith. 2004. The pteridophytes of Mexico. Mem. New York Bot. Gard. 88:1-1 055.
Moran, R.C. 1995. The importance of mountains to pteridophytes, with emphasis on neotropical montane
forests. In: S.P. Churchill, H. Baslev, E. Forero, and J.L Luteyn, eds. Biodiversity and conservation of neotropical
montane forests. The New York Botanical Garden, New York. Pp. 359-363.
Moran, R.C. 2008. Diversity, biogeography, and floristics. ln:T.A. Ranker and C.H. Haufler, eds. Biology and evolu-
tion of ferns and lycophytes. Cambridge University Press, New York. Pp. 201-221 .
Moran, R.C., S. Kumas, and M. Carlsen. 2003. Low-trunk epiphytic ferns on tree ferns versus angiosperms in Costa
Rica. Biotropica 35:48-56.
Mown, R.C, J. Prado, and P. Labiak. 2009. Megalastrum (Dryopteridaceae) in Brazil, Paraguay and Uruguay. Amer.
Fern J. 99:1 -44.
Morl SA, B.M. Boom, and G.T. Prance. 1981 . Distribution patterns and conservation of eastern Brazilian coastal
forest tree species. Brittonia 33:233-245.
Mow, SA, B.M. Boom, A.M. Carvalho, andT.S. Santos. 1 983. Southern Bahian moist forests. Bot Rev. 49:1 55-232.
^bra-Fiiho, A.T. and M AL Fontes. 2000. Patterns offloristic differentiation among Atlantic forests in southeastern
Brazil and the influence of climate. Biotropica 32:793-810.
Paoencia, M.LB. 2008. Diversidade de pteridofitas em gradientes de altitude na mata atlantica do Estado do
Parana, Brasil. Unpubl. Ph.D. thesis, Universidade de Sao Paulo, Brazil.
Paoencia, M.L.B. and J. Prado. 2004. Efeitos de borda sobre a comunidade de pteridofitas na Mata Atlantica da
fegiaode Una, sul da Bahia, Brasil. Rev. Bras. Bot 27:641-653.
p*onoa, M.LB. and J. Prado. 2005a. Effects of forest fragmentation on pteridophyte diversity in a tropical rain
forest in Brazil. Plant Ecol. 180:87-104.
*3enqa' m |-B. and J. Prado. 2005b. Distribuigao espacial da assembfeia de pteriddfitas em uma paisagem frag-
rnentada de Mata Atlantica no sul da Bahia, Brasil. Hoehnea 32:103-1 1 7.
^Sermolli, R.E.G. 1996. Authors of scientific names in Pteridophyta. Royal Botanic Gardens, Kew.
D£ ^ PE. Gibbs. 1 993. Patterns of species distribution in the dry seasonal forests of South America. Ann.
Missouri Bot. Gard. 80:902-927.
>J- 2000. A new species of Adiantum (Pteridaceae) from Bahia, Brazil. Brittonia 52:210-212.
J&HC.T. 1997.Tratado de fitogeografia do Brasil. 2nd Edition. Ambito Cultural Edi^oes Ltda, Rio de Janeiro.
**** M. 1975. Diagnostic socioeconomic da Regiao Cacaueira, vol. 4: Reconhecimento dimatoldgico. Co-
^ Executiva do Plano da Lavoura Cacaueira and Instituto Interamericano de Cfencias Agrfcolas - OEA.
*** Bahia, Brasil.
T0, A’ S M- Silva, VAO. Dittrich, and R.M. Britez. 2005. Flora pteridofftica. In: M.C.M. Marques and R.M. Britez, eds.
natural e conservagao da llha do Mel. Editora UFPR, Curitiba. Pp. 85-101 .
PB. and P.H. Labiak. 2007. Pteridofitas do Parque Estadual de Vila Velha, Ponta Grossa, Parang Brasil.
34:159-209.
CAREX OKLAHOMENSIS (CYPERACEAE) NEW TO ALABAMA, GEORGIA, AND
J; LOUISIANA, AND ADDITIONAL RECORDS FOR MISSISSIPPI
Charles! Bryson
Paul E. Rothrock
USDA-ARS
Southern Weed Science Research Unit
RO. Box 350
Stoneville, Mississippi 38776, U.SA
charks.bryson@ars.uida.gov
Randall Environmental Studies Centei
Taylor University
236 WReade Avenue
Upland, Indiana 46989- 1 00 1 USA
plrothroc@tayloru.edu
RESUMEN
tomlands, marshes, right-of-ways, and wet meadows in Arkansas, Delaware, Kansas, Illinois, Indiana,
Kentucky, Maryland, Mississippi, Missouri, North Carolina, Oklahoma, Tennessee, Texas, and Virginia
(Mackenzie 1931; Steyermark 1963; Bryson et al. 1992; Bryson et al. 1994; Jones and Reznicek 1995; Bryson
etal. 1996; Standley 2002). Recent records from northern and eastern states indicate C. oklahomensis may
be increasing its range (Standley 2002), and human disturbance and other activities seem to be the primary
(actors in its dispersal (Bryson & Carter 2008). Previously, C. oklahomensis populations in Mississippi and
Tennessee were associated with highway and airport runway construction and the use of contaminated hay
and seeds for erosion prevention and grass maintenance (Bryson et aL 1992; Bryson et al. 1994; Bryson et
* 1996)- Some populations of C. oklahomensis have persisted for several years and then disappeared fol-
towing drought or herbicide usage along highway right-of-ways (Bryson 2007), and others have persisted
for longer periods (Bryson, pers. obs.).
Herein, we report the first records of C. oklahomensis from Alabama, Georgia, and Louisiana and new
«>unty records for Mississippi. It is highly likely that C. oklahomensis was introduced during highway and
reservoir construction or maintenance in contaminated hay, grass seeds, or machinery at these sites, and it
B likely that C. oklahomensis will continue to disperse in contaminated hay and grass seeds or on construc-
tion, maintenance, and mowing equipment. Presently it does not seem to pose a major weed threat.
Voucher specimens. ALABAMA r«, \ > idunville. W ol Hwy US 231/431 jus. N of Beaverdam Creek. 34°50 376'N.
M2 & Bryson (JSU, NY, SWSL, VDB, VSC, herb. Bryson); 17 May 2007, Bryson 22248 (SWSL, VDB,
£ l ** of Hwy US 76/SR 52, 0.6 mi E jet. with SR 286 to Dawnville, 2.0 mi E jet. with north
8293'N, ®4. 87904° W. 17 May 2009, Rothrock 4829 (GA, MICH, MOR, VSC, SWSL, herb. Brys
01 165 N of Monroe, 16 May 1997 Thomas 154042 (herb. Bryson; verified by Bryson, 1997). :
j«- Hwy US 78 and US 178; just W of ramp onto Hwy US 78 from Hwy 178, 34°37.085'
knQVSC’ SWSL’ herb‘ B,Yson) Marshall Co.: Potts Camp, W of Bypass Street, just N of Hwy US
^ri8592,w' 2 jun 2o°9, Brys°n 2mi
. 33°28.662'N, 089°40.0:
JnAlabama> C. oklahomensis grew around the edges of a small man-made lake on a dark red clay soil with few
ermixed rocks just above the water-line. At this site, C. oklahomensis was associated with Carex annectens
‘cknell, C. longH Mackenzie, C. lurida Wahlenberg, C. triangularis Boeckeler, C. tribuloides Wahlenberg, C.
VU^“Undea Michaux Juncus sp., and a number of Poaceae. Small populations of two to 20 clumps were dis-
348
persed around the edge of the lake in open areas. Although the area is mowed periodically, C. oklahomensis
plants were observed again in May, 2009 (Bryson, pers. obs.). The Georgia population had over 50 plantsof
C. oklahomensis in a wet pasture/shallow marsh that receives road verge runoff. Associated species included
Agrimonia sp., Carex annectens, C. aureolensis Steudel, C. leavenworthii Dewey, C. lupulina Willdenow, Eleocharis
obtusa (Willdenow) Schultes, E. palustris (L.) Roemer & Schultes, Lolium pratense (Hudson) S. J. Darbyshire
(Festuca elatior L.), Juncus spp. including/ effusus L., Ludwigia palustris (L.) Elliott, and Ranunculus sardous
of C. oklahomensis were discovered in an unplanted wildflower area (Thomas, pers. comm.). At the sites in
Mississippi, C. oklahomensis grew in wet ditches along the highway right-of-ways on wet silt or clay soils
Common associates in Mississippi were Carex annectens, C. aureolensis, C. longii, C. lurida, C. vulpinoufca,
Eleocharis obtusa, Juncus spp., and various Poaceae species. The Mississippi populations of C. oklahomensis
ranged from fewer than 50 plants at the Montgomery County site to several hundred plants at the Marshall
County site. Lack of extensive seasonally wet roadside ditches at each of the Mississippi sites may be a limit-
ing factor for additional dispersal, unless mowing equipment transports achenes to suitable habitats several
hundred meters from the observed populations.
ACKNOWLEDGMENTS
We thank Nancy B. Bryson for help with field work and Richard Carter (VSC), Wade Steinriede (USDA-ARS,
Oxford/Stoneville, MS), Robert F.C. Naczi (NY), and Dan Spaulding (Anniston Museum of Natural History)
for reviewing the manuscript.
REFERENCES
Bryson, C.T. 2007. Carex opaca new to North Carolina. Castanea 72:1 23.
Bryson, C.T. and R. Carter 2008. The significance of Cyperaceae as weeds. In: R.F.C. Naczi and B.A. Ford, eds. Sedges
uses, diversity, and systematics of the Cyperaceae. Monogr. Syst. Bot. Missouri Bot. Gard. 108. Pp. 15-101-
Bryson, GT., J.R. MacDonald, R. Carter and S.D. Jones. 1 996. Noteworthy Carex, Cyperus, Eleocharis, Kyllinga, and
Oxycaryum (Cyperaceae) from Alabama, Arkansas, Georgia, Louisiana, Mississippi, North Carolina, Tennessee,
and Texas. Sida 1 7:501-518.
Bryson, CT., J.R. MacDonald, and R. Warren. 1 994. Notes on Carex (Cyperaceae), with C. godfreyi new to Alabama
and C. communis and C. scoparia new to Mississippi. Sida 16:355-361 .
Bryson, C.T., R.F.C Naczi, and S. McDaniel 1992. Notes on noteworthy records of Carex (Cyperaceae) from
southeastern United States. Sida 15:125-135.
Jones, S.D. and A.A. Reznkzek. 1 995. Carex conjuncta (Cyperaceae) verified for Arkansas, and notes on the range
Carex oklahomensis. Sida 16:772-774.
Mackenzie, K.K. 1931. Cyperaceae-Cariceae. N. Amer. Flora 1 8:1 -1 68. . ;
Standley, LA 2002. Carex section Vulpinae. In: Flora of North America Editorial Committee, eds. Flora of Noftn
America north of Mexico. Oxford University Press, New York and Oxford. 23:273-278.
Steyermark, JA 1963. Flora of Missouri. Iowa State Univ. Press, Ames.
PLANTS NEW TO FLORIDA
Richard R Wunderlin, Bruce F. Hansen,
and Alan R. Franck
Institute for Systematic Botany
Department of Cell Biology, Microbiology and Molecular Biology
University Of South Florida
4202 East Fowler Avenue, BSF 218
Tampa, Florida 33620-5200, USA
Keith A. Bradley
The Institute for Regional Conservation
22601 SW 1 52 Avenue
Miami, Florida 33170, USA
John M. Kunzer
Tomoka State Park
2099 North Beach Street
Ormond Beach, Florida 32174, USA
ABSTRACT
RESUMEN
Work toward a third edition of the Guide to the Vascular Plants of Florida to be published in 2010 revealed that
® the six years since the publication of the second edition (Wunderlin & Hansen 2003), over 100 species
are new to the state. Of these, about 75% are non-native to Florida. About 70% of these have been previously
reported for Florida in a variety of publications (e.g., Abbott & Carlsward 2004; Anderson 2007; Bradley
2006; Darst & Gholson 2004; Davis et al. 2006; Kunzer et al. 2009), Flora of North America (2003-2009), and
several recent taxonomic revisions. Although cited in the online Atlas of Florida Vascular Plants (Wunderlin
& Hansen 2009), this paper documents the first report of 34 taxa as occurring in Florida. Twenty-six of
these taxa are the first reports for the continental United States. Of the total, 33 taxa are non-native and 25
°f the non-native species are considered to be escaped from cultivation. A recent paper by Pemberton and
Liu (2009) demonstrates that horticultural introductions have been an important source of non-native plants
naturalizing (reproducing outside of cultivation) in the state, a trend that will clearly continue. Over half
°f the species reported here were found in the two most populous counties in Florida (Miami-Dade Co.-13
spP- and Broward Co.-6 spp.). It is likely that some taxa reported here for first time may be listed as invasive
antics in the Florida Exotic Pest Plant Council (FLEPPC) list (2009) in the future.
Araacaria heterophylla (Salisb.) Franco (Araucariaceae). This species, endemic to Norfolk Island in the
*uth Pacific, is cultivated in tropical and subtropical areas of the Old and New World for its wood and as
30 ornamental, and occasionally becomes naturalized. It is commonly cultivated as an ornamental in central
** Southem Florida. This is the first report of it naturalizing in the continental United States.
'n'Kher spcunien? Broward Co ■ Pm,' Mm I Rid ■< N mini Area 1 mi N of Oran*/ Drive on Golden Shoe Road, ca 1 mi N of Cooper
,V T5,'s R-HP ,,, u , ,r , , j. hie parent iree. 17 Ma 007. Ho. M Monroe
C*': Pine Key. Mourn. TOO m E of the intersection of Fern Avenue and Guava Lane, ca. 27 air
sHiiim
Pentas lanceolata (Forssk.) Defiers (Rubiaceae). Native to northern Africa, it is commonly cultivated in the
tropics and subtropics worldwide as an ornamental where it has sometimes become naturalized. This is the
first report of the species naturalized in the continental United States.
k. The Orchid Jungle, E side of SW 157 Ave. at theoretical 268 St., disturbed
42°W, 23 Feb 2002, Bradley 2192 (FTG)
Philodendron hederaceum (Jacq.) Schott var. oxycardium (Schott) Croat (Araceae). Native to Mexico, this
species is frequently cultivated as an ornamental. It was planted along nature trails in Hattie Bauer Ham-
mock when it was commercially operated as the tourist attraction Orchid Jungle. Now a county preserve, it
has spread aggressively throughout undisturbed portions of the rockland hammock there. This is the first
report of it naturalized in the continental United States.
tcuiVr specimen Miami-Dade Co.: Hull* Bauer Hammock, b side of SW 1ST Ave a. theoretical 268 St . abundant m rockland ham-
mock throughout property where it spread from cultivation along nature trail, 24 Jun 2009, Bradley 2675 (FTG, USF).
Pimenta dioica (L.) Merr. (Myrtaceae). Native to the West Indies, Mexico, and Central America, it is frequently
cultivated in the Old and New World tropics as a food flavoring (allspice) and sometimes for medicinal use.
It is naturalized in Puerto Rico, the Virgin Islands, Bermuda, Hawaii, and other Pacific Islands. Commonly
cultivated in southern Florida, it has escaped into multiple rockland hammocks and disturbed sites in
Miami-Dade County. This is the first report of it naturalized in the continental United States.
Miami-Dade Co.: Meissner Hammock, S of theoretical SW 300 St., 0.1 mi E of 202 Ave., frequent inside western
edge of rockland hammock, 19 Jun 2009, Bradley 2681 (FTG, USF).
Pithecellobium bahamense Northr. (Fabaceae). Native to the central Bahamas (Andros, New Providence,
Exuma, and Long Island) and northern Cuba, it is here reported for the continental United States. The Monroe
County keys population consists of over 200 plants in a small, dense colony in an undisturbed pine rockland
with no evidence of cultivation. This appears to be a native population. It was also found by Paula Cannon
and Ben Edwards in 2006. Although experimental evidence is lacking, Bameby and Grimes (1997) propose
this taxon to be a hybrid between R keyensis Britton and P. histrix (A. Rich.) Benth. based on intermediate
morphology and sympatric distribution. Of the supposed parental species, only P. keyensis occurs in Florida.
Pithecellobium bahamense is readily distinguished from P. keyensis in Florida by having stipular spines and
the kaflet s 1-2.5 cm long (versus spines lacking and the leaflets 3-7 cm long in P. keyensis).
specimen. Monroe Co.: Big Pine Key, National Key Deer Refuge; ca. 180 m W of Key Deer Blvd., ca. 460 m N of Big Pine Street,
°PP«ite parking lot for Watson Hammock Nature Trail, 18 Oct 2007, Bradley & Barry 2662 (USF).
Po'yalthia suberosa (Roxb.) Thwaites (Annonaceae). Native to Asia, it is widespread as a cultivated orna-
mental in the Old and New World tropics. This is the first report of it naturalized in the continental United
***■ In addition to the localities below, it is also well established within rockland hammock habitat at
Smpson Park in the City of Miami.
'
^nmock P«k, S of Schoolhouse Road endpoint, 25.38360°N, 80.27593*W, 14 Dec 2004, Possley & Prieto 72 ( FTG).
kdermachera sinica (Hance) Hemsl. (Bignoniaceae). Native to subtropical Asia, it is cultivated widely as
T/0usePlant and outdoors in the Old and New world tropics as an ornamental. It is naturalized in Hawaii.
***** first report of it naturalized in the continental United States.
2^ SPeCime" Co.: Fire suppressed pine rockland at SW comer of SW 296 St. & 197 Ave., frequent in understory,
747°N, 80.510248°W, 24 Jun 2009, Bradley 2677 (FTG, USF).
R0USStlia humilis (SW.) Urb. (Urticaceae). This weedy species native to the West Indies, Mexico, and Central
^nca ^ here reported for the first time from continental United States. Our specimens were determined
^Plicate by David E. Boufford (A).
Broward Co.: IV mhrokc r.irk 42o Marine Dr . 31 Dec 2003. lUSF'i Coconut Creek Maple Swamp
mi SE of West Dixie Bend, T48S, R42E, Sec 8, 17 Sep 2008, Howell 1074 (USF).
Senna atomaria (L.) H.S. Irwin & Bameby (Fabaceae). Native to Mexico, the West Indies, Central America,
and northern South America (Columbia and Venezuela) and recently introduced in the Old World for agro-
forestry purposes. It is rare in cultivation in Florida and the origin of material at Dismal Key is uncertain;
possibly a recent natural range extension. This is the first report of it in the continental United States.
Solanum umbellatum Mill. (Solanaceae). A weedy species native to Mexico, Central America, South America,
and the West Indies, it is related to S. erianthum D. Don, a widespread species native to peninsular Florida
(Roe 1967). This is the first report of the species naturalized in the continental United States.
Voucher specimens Miami-Dade Co.: edge ot l.irm held 0 25 mi W of K.omc Avc <SW 177 Au - on \ c.l* .4 s\\ jss-i -m,ll[vp
lation of a few plants, 25o30,00.3*N, 80°28'54.05"W, 16 Dec 2008, Bradley 2665 (FTG, USF); Silver Palm Groves Hammock Preserve,
Chromolaena odorata, Bidens alba, and other weeds, ca. 850’NofSW232 St.,ca. 70’ W of theoretical 142 Ave., Easting 558056, Northing
2826371 (UTM17N, WGS84), 3 May 2006, Bradley & Green 2646 (FTG, USF).
Stellaria parva Pedersen (Caryophyllaceae). A species native to South America and introduced in Texas
and Louisiana (USDA, NCRS 2009), large colonies of it were recently discovered in Colt Creek State Park,
Polk County.
Voucher specimens. Polk Co.: Colt Creek State Park, ca. 3.27 km SSE of FL 471 bridge over the Withlacoochee River, ca. 4.5 km NNE of
the junction of FL 471 and US 98, edge of pasture near hydric hammock, 28°17T3"N, 82°2'21"W, 3 Apr 2008, Kunzer et d. 2317<USF);
Colt Creek State Park, ca. 2.6 km SSE of FL 471 bridge over the Withlacoochee River, ca. 5.3 km NNE of the junction of FL 471 and
US 98, trail in ecotone between pasture and cypress [Taxodmm] swamp, 28°17’38"N, 82°2'23"W, 15 Apr 2008, Kunzer & Hansen 2383
(USF); Colt Creek State Park, ca. 2 .6 km SSE of FL 471 bridge over the Withlacoochee River, ca. 5.3 km NNE of the junction of FL 471
and US 98, pasture edge adjacent to hydric hammock, 28°17,13"N, 8202'21"W, 15 Apr 2008, Kunzer & Hansen 2390 (USF); Colt Creek
State Park. ca. 2.6 km SSE of FL 471 bridge over the Withlacoochee River, ca. 5.3 km NNE of the junction of FL 471 and US 98, pasture
edge adjacent to Toxodium swamp, 28°17T3'N, 82°2,21,W, 15 Apr 2008, Kunzer & Hansen 2392 (USF).
Trema orientals (L.) Blume (Ulmaceae). Native to Asia, Africa, and Australasia, it is cultivated in the New
and Old world tropics for a variety of uses and is naturalized in Hawaii. A population was found in Miami-
Dade County by Bradley and Roger Hammer in 1996 at the USDA Subtropical Horticultural Research Station.
Plants were formerly cultivated at the Station where it was introduced from Mozambique. It subsequently
spread into a fire suppressed pine rockland. This is the first report from the continental United States. In
2009, it was found along the historical eastern shoreline of Lake Okeechobee in Martin and Palm Beac
Counties. Whether the source of this population was from the Miami-Dade plants is unknown.
Voucher specimens. Miami-Dade Co.: USDA Subtropical Horticultural Research Station, Chapman Field, Old Cutler Road just
per Creek, small trees naturalized along edge of a degraded pine rockland fragment, recruits frequent here, 25.641 266°N,80796M^
27.09336°N, 80.65745°W, 18 Oct 2009, Bradley 2M70TG). Okeechobee Co.: a few plants along western edge of US 441, 0.2 mi N of
Martin Co. line, with Schinus terebinthifolia, 27.12389°N, 80.67926°W, 18 Oct 2009, Bradley 2580 (FTG).
Trichosanthes cucumerina L. (Cucurbitaceae). Native to tropical Asia and Australasia, it is cultivated foj
food, medicine, and as an ornamental in tropical areas of the Old and New World. This is the first report
Triplans melaenodendron (Bertol.) Standi. & Steyerm. (Polygonaceae). Native to Mexico and Central Arn^1
is occasionally cultivated as an ornamental. This is the first report of it naturalized in the continental United St*
Wundertin et al., Plants new to Florida
7, Bradley 2670 (FTG,
South America, the West Indies, and the continental United states in Texas. It also occurs in Louisiana
(USDA, NRCS, 2009). It is reported here for the first time as naturalized in Florida.
ACKNOWLEDGMENTS
The authors thank Mike Barry, James R. Burkhalter, Stephen Dickman, George Gann, Steven Green, Pierre
Guertin, Patricia L. Howell, Reese H. Kessler, Chris Lockhart, Randy Mejeur, Ray Miller, Brent Nicholas,
Nancy O’Hare, Stella Park, Kelly Peterman, Shirley Petty, Jennifer Possley, Jimi Sadie, Carmel vanHoek,
Steve Woodmansee, and P. York for specimens. We also wish to thank Alice Bard (Florida Department of
Environmental Protection, Division of Parks) for the collecting opportunity in Colt Creek State Park. We
also thank Elizabeth Gandy and an anonymous reviewer for their helpful suggestions.
Abbott, J.R. and B.S. Carisward. 2004. Noteworthy collections: Florida. Castanea 69:324-327.
Abrams, L 1923. Illustrated flora of the Pacific States: Washington, Oregon, and California. 1:1-568. Stanford
University Press, Stanford, CA.
Anderson, L.C. 2007. Noteworthy plants from north Florida VIII. J. Bot. Res. Inst. Texas 1 :741 -751 .
Barneby, R.C. and J.W. Grimes 1 997. Silk tree, guanacaste, monkey's earring: a generic system for the synandrous Mi-
mosaceae of the Americas. Part II. Pithecellobium, Cojoba, and Zygia. Mem. New York Bot. Gard. 74(2):1-149.
Brmwy, KA. 2006. Thelypteris sancta (L) Ching, New for Florida and the continental United States. Amer. Fern J.
96:112-114.
Darst, M. and A.K. Gholson. 2004. Noteworthy collections: Florida. Castanea 69:328.
Davis, S.B., W.S. Judd, and K.D. Perkins. 2006. Noteworthy collections: Florida. Castanea 71:333-334.
foROA Exotic Pest Plant Council (FLEPPC). 2009. Florida Exotic Pest Plant Council's 2009 list of invasive plant spe-
cies (vwvw.fleppc.org).
^ of North America Editorial Committee. 2003-2009. Flora of North America north of Mexico. Oxford University
Press, New York NY & Oxford UK.
*** J-M., LC. Anderson, R.P. Wunderun, and J.R. Burkhalter 2009. New and noteworthy plants from Florida. J. Bot.
^ Inst. Texas 3:333-339.
McDon*d. M.W., M.I.H. Brooke^ and P.A. Butcher 2009. A taxonomic revision of Eucalyptus camaldulensis (Myrta-
ceae). Austral. Syst. Bot. 22:257-285.
Pem6ERT0n> R.W. AND H. Lu, 2009. Marketing time trends predicts naturalization of horticultural plants. Ecology
90:69-80.
BoUE 1967. Revision of Solanum section Brevantherum (Solanaceae) in North and Central America. Brittonia
19353-373.
°^A- NRCS. 2009. The PLANTS Database (http://plants.usda.gov). National Plant Data Center, Baton Rouge,
Wun*Run, R.P. and B.F. Hansen 2003. Guide to the Vascular Plants of Florida, 2«* ed. University Press of Florida,
Gainesville, fl
******* R-P. and B.F.Hansen. 2009. Atlas of Florida vascular plants (http//www.florida.plantatlas.usf.edu/) [S.M.
“"dry and K.N. Campbell (application development), Florida Center for Community Design and Research.]
nstitute for Systematic Botany, University of South Florida, Tampa.
356 Journal of the Botanical Research Institute of Texas^l)
BOOK REVIEW
Luis Digo GOmhz y Alba L. ArbeiAez (authors) and WD. Stevens, Olga Martha Montiel y Amy Pool (eds) and Aha
L. ArbelAez (ilustradora) . 2009. Flora de Nicaragua, Helechos (Tomo IV). (ISBN 978-1-930723-87-0,
hbk.). Monogr. Syst. Bot. Missouri Bot. Gard. Vol. 116 (ISSN 0161-1542). Missouri Botanical Garden
Press, P.O. Box 299, St. Louis, Missouri 63166-0299, U.S.A. (Orders: www.mbgpress.org). $109.00,
348 pp., distribution maps, 151 line drawings, 8 1/2" x 11".
VASCULAR FLORA AND EDAPHIC CHARACTERISTICS OF
SALINE PRAIRIES IN LOUISIANA
Christopher S. Reid and Michael H. MacRoberts and
Patricia L Faulkner Barbara R. MacRoberts
Louisiana Department of Wildlife and Fisheries
2000 Quail Drive
Baton Rouge, Louisiana 70808, USA
Bog Research, 740 Columbia
Shreveport, Louisiana 71104, USA
Herbarium, Museum of Life Sciences
Louisiana State University-Shreveport
Shreveport, Louisiana 71 1 15, USA
Marc Bordelon
USDA - Natural Resources Conservation Sen/ice
2263 Hall Street
Ringgold, Louisiana 71068, USA
ABSTRACT
ikk^l««4(1):357 - 379.2010
Journal of the Botanical Research Institute ofTexas4(1)
INTRODUCTION
Saline prairies are typically small-scale natural grassland communities that occur on soils with unique chemi-
cal and physical properties (Louisiana Natural Heritage Program 2009b). Saline prairies occur in the West
Gulf Coastal Plain of central and northwestern Louisiana, eastern Texas, and southern Arkansas (Louisiana
Natural Heritage Program 2009b; NatureServe 2009), in the Arkansas River Valley of western Arkansas
(Pittman 1993), and in the Mississippi River Alluvial Plain of eastern Arkansas (Arkansas Multi-Agency
Wetland Planning Team 2001). In Louisiana, pre-settlement aerial coverage of saline prairies is estimated to
have been fewer than 800 ha, with only 10 to 25 percent currently remaining (Lester et al. 2005). Threats
to saline prairies include development and maintenance of roads and utility corridors, grazing practices,
incompatible forestry practices, invasive species (e.g., feral pigs), oil and gas drilling, and vehicular traffic
and recreational use (Lester et al. 2005). The community is currently considered globally imperiled with
a rarity ranking of G1G2 (NatureServe 2009). Largely due to the presence of the federally-listed Geocarpm
minimum, saline prairies are being studied floristically and edaphically (Mclnnis et al. 1993; Keith et al
2004; Lester et al. 2005; Arkansas Natural Heritage Commission 2006; Diggs et al. 2006; MacRobertset
al. 2009b). The purpose of our study is to add descriptive floristic and edaphic information to the growing
body of literature on saline prairies.
PHYSICAL SETTING, GEOLOGY, AND SOILS
Louisiana saline prairies occur on fluvial terraces adjacent to active floodplains. The soils formed in loamy
late Pleistocene sediments (Soil Survey Staff 1991; Soil Survey Staff 1998) that correspond to the Prairie
Terrace formation (Snead & McCullough 1984). The Prairie Terrace consists of reworked sediments that
eroded from adjacent higher Pleistocene terraces and/or Tertiary formations (Huner 1939). the Prairie
Terrace is the youngest and lowest in elevation of the Pleistocene terraces, and possibly formed during
interglacial periods when sea level was high, thus reducing stream gradients and allowing deposition of
fine sediments (Murray 1948). Three soil series are known to support saline prairies in Louisiana: Bonn,
Brimstone, and Lafe. The following are taxonomic names for these soil series: Bonn series: fine-silty, mixed,
superactive, thermic Glossic Natraqualfs; Brimstone series: fine-silty, siliceous, superactive, thermic, Glosac
•afe series: fine silty, mixed, active, thermic, Glossaquic Natrudalfs (Soil Survey Staff 20061
mples of natric soils, also called alkali, sodic, sodium, or solonetz soils (Horn 1962; Hornet
Natric soils
type of argillic
ized by high levels of exchangeable Na 15%) and
,re often associated with
These a
al. 1964; Pettry et al. 1981; Soil Survey Staff 1
illuviated clays) called a natric horizon that is
by columnar/prismatic or blocky structure (Soil Survey Staff 2006). 1
and or semi-arid environments where low precipitation and high evaporation demand result in incomplete
leaching of salts. Accumulation of Na in soils in humid environments has been attributed to interruption
of leaching by impervious subsoil horizons (Smith 1937; Pettry & Switzer 1998). In natric horizons, soil
hydraulic properties are degraded due to the dispersal of clay particles by Na*. Dispersion of clay part**5
produces a natric horizon that is dense, compact, and slowly permeable to air and water (Smith 1937, Horn
1962, Hassett & Banwart 1992; Pettry & Switzer 1998). Physical properties of natric soils can reduce *»
water storage and transport capacity resulting in droughty conditions (Rengasamy et al. 2003). The nat
horizon functions like a claypan by perching water during wet periods, preventing upward movernc*
water during dry periods, and preventing plant roots from exploiting water deeper in the soil (Horn •
Pettry & Switzer 1998). Thus, soils are often either extremely dry or waterlogged, a condition descri
xerohydric (NatureServe 2009). Hydrolysis of Na and formation of compounds such as NaOH and
result m strongly alkaline pH in natric horizons (Horn 1962). Natric soils have been the subject of consult
study due to the effects of their adverse physical properties on agricultural crop production (e.g., Sham
et al 1980; Rengasamy & Olsson 1991; Rengasamy et al. 2003; Vukadinovid & Rengel 2007).
In studies of natric soils in eastern Arkansas and in Illinois, the weathering of Na-rich feldspars
oess parent material was implicated as the source of Na (Wilding et al. 1963; Horn et al. 1964). 1’""
in Arkansi
s seems possible in saline prairies in central
r above large amounts of Jurassic Louann salt
and Horn et al. (1964) suggest that possible sources of Na ii
incomplete leaching of salts adhering to flood-deposited sedi
(Andrews 1960; Ingram 1991).
Depth to natric horizon is apparently important in determining soil physical and chemical properties
and associated vegetation. In a study of natric soils in Arkansas, Horn et al. (1964) recognized soil groups
based on depth to natric horizon, with the most extreme category being represented by the Lafe series,
which in their study had a natric horizon within ~ 25 cm of the surface. Lafe soils were characterized by
drought-tolerant species such as Aristida L. spp., Opuntia spp., and Quercus stellata (Horn et al. 1964).
The general appearance of a saline prairie is often pasture-like (Fig. 1). Saline prairies in Louisiana are
open, usually with a short to medium-height grass canopy. The density of grasses is variable but is sparse
tlh fairly continuous herbaceous cover as “broad flats.” Broad flats account for the largest portions of saline
prairies, the other landscape features usually being inclusional. Wet depressions are often embedded features
in northwestern Louisiana saline prairies. These depressions tend to be shallow and support wetland plant
species. Small “slicks” nearly devoid of vegetation occur where Na has been brought to the surface or where
the natric horizon is very near the surface or exposed (Fig. 2). The soil surface within slicks often has a
cryptogamic crust composed of algae, mosses, and lichens. Smaller slicks may be completely covered by this
crust' whik on larger slicks it may occur only along the margins. Pimple mounds (Fig. 3) are often scattered
about and support trees and shrubs of various density, but are sometimes almost entirely open and grassy.
Pimple mounds are enigmatic landscape features. Cain (1974) postulates that pimple mounds may have
been formed by rill erosion occurring in a former climate characterized by alternating periods of drought
and heavy rainfall. A more recent study provides evidence that pimple mounds are relict nebkhas, formed
by deflation and subsequent deposition of soil by wind during prolonged droughts of the late Holocene
Friod (Seifert et al. 2009). Seifert et al. (2009) point out that pimple mounds often coincide with claypan
»ils, which might have accentuated the effects of drought.
in 2005, we began searching for saline prairies mainly in Caddo and DeSoto parishes in northwestern
^misiana. The Louisiana Natural Heritage Program database had no occurrence records of saline prairies
that region of the state at that time despite the relative commonness of Bonn silt loam in Caddo and De
parishes (Soil Survey Staff 1980; Soil Survey Staff 1991).
Using the soil surveys, Bonn soil map units were identified, and then located on 7.5' topographic maps
aerial photographs. Saline prairies and other inclusional prairies and barrens are often symbolized
k ifregular openings on topographic maps. The signature on dormant season (“leaf off”) aerial photos is
>sh gray, indicating that the herbaceous layer is composed of warm season species (dead at time photo
t^en) rather than cool season (these would appear pinkish or reddish) (Fig. 4). Pimple mounds and
3*ssions are often evident on aerial photos (Fig. 4). We examined soil surveys of parishes adjacent to areas
wing Bonn soil map units as well and scanned the correct landscape position looking for irregular
gs. Apparently Bonn soils or related soil series were overlooked in Bienville and Sabine Parishes.
^ TlDe^ sahne prairies occur in these parishes on Guyton soils or map units containing the Guyton series
^ompfex with other soils (Soil Survey Staff 1997; Soil Survey Staff 2001). Once potential saline prairies
^ ntified, ownership of the sites was determined, and after gaining permission for access, the sites
(lou V1SUed f°r ground inspection. Since 2005, 10 new saline prairie occurrences have been documented
Wd fed113 Natural Heritage Program 2009a). Seven of these were discovered in Bienville, Caddo, De Soto,
Pj ^iver Parishes using the process described above. Three saline prairies were confirmed in Sabine
a ter being informpH of thpir ootpntial orrnrrpnrp hv a landowner and a colleague. Locations of extant
360
362
saline prairies in Louisiana are presented in Figure 5. Cursory surveys were conducted at all sites to assess
habitat quality and search for rare species. Two sites were more intensively studied and are the primary
focus of this paper.
Barron Road Saline Prairie in southern Caddo Parish (32°21'13"N, 93°47'32"W) was discovered in July,
2005. This 6 ha site is situated just above the floodplain of Boggy Bayou, which empties into Wallace Lake.
The prairie elevation is in the range of 46 to 49 m above sea level. The prairie is surrounded on its upland
side by oak-pine-hickory woodlands and on its lowland side by bottomland hardwood forest that occupies
the Boggy Bayou floodplain. Pimple mounds, measuring between 12 m and 33 m in diameter and <1 to
1.5 m high, are scattered over the prairie. Shallow wet depressions and slicks are occasional. The prairie
has several roads through it and has been damaged by off-road vehicle use. At one time, establishment of
loblolly pine (Pinus taeda L.) was attempted (S. Evans, pers. comm.); however, survival was extremely poor
everywhere but on the pimple mounds, where even-aged, spindly trees have persisted.
Dalton/Dickson Saline Prairie, located in northern De Soto Parish (32°18I28''N, 93°48'24"W) about 5
km SSE of Barron Road Saline Prairie, consists of a complex of numerous small to large openings totaling 53
ha. It is positioned above the floodplain of Cypress Bayou, which also feeds into Wallace Lake. Most of the
prairie openings are within 50 to 52 m in elevation. Pimple mounds and small wet depressions are scattered
about. Much of this prairie complex is undisturbed and of very high quality. Portions do, however, experi-
ence off-road vehicle traffic. Establishment of rye grass (Lolium perenne L.) in one of the larger openings was
attempted. Some rye grass persists in small scale patches.
The current climate of northwest Louisiana generally provides warm and humid summers and mild
(Climate-Zone.com 2004).
FLORISTIC STUDY
Barron Road Saline Prairie was visited and voucher specimens collected every two to three weeks (except
during the winter) between July 2005 and July 2006. Dalton/Dickson Saline Prairie was surveyed frequently
during 2006 to search for rare plants, which were collected along with some additional species. A more
thorough collection of plants from Dalton/Dickson Saline Prairie was made during regular visits from March
to October 2007. Occasional collecting was carried out into 2009. Our collecting effort included species
from all landscape features including broad flats, slicks, wet depressions, and pimple mounds. Scienti c
names follow Kartesz and Meacham (1999), Diggs et al. (2006), and USDA, NRCS (2009), in most cases.
Voucher specimens are deposited at LSU and LSUS. To determine whether a species is native or exotic we
consulted USDA, NRCS (2009). Using our checklist of species found at the study sites (native species only)
and data from Flora of North America (1993-2007), Turner et al. (2003), NatureServe (2009), and USD i.
NRCS (2009), we plotted the North American distribution of species by state, region, or province to deter
mine geographic pattern and affinity.
SOIL COLLECTION AND ANALYSIS
In order to sample the intact soil profile, soil cores were taken at three relatively undisturbed saline P®**
sites: Dalton/Dickson Saline Prairie, Rambin Bayou Saline Prairie (ca 12 km south of Dalton/Dickson)
Upper Weyerhaeuser Saline Prairie (a.k.a. Saline Creek Prairie) in Winn Parish, ca 145 km to the east
Roberts et al. 2009a). At each prairie, one soil core was extracted from broad flat, slick, and pimple "V
landscape positions at subjectively determined locations within these sites. Upper Weyerhaeuser ^ _
has no pimple mounds; thus only two cores were taken there, from broad flat and slick sites. The samp^
were separated into horizons with each horizon morphologically described at the time of collection ^
sample from each horizon of each core was collected and stored in a plastic bag for further analyses-
samples from each horizon of each core were analyzed to determine basic chemical and physical attn
Particle size analysis, pH, electrical conductivity (EC), cation exchange capacity (CEC), and sodium a
ration (SAR) were determined by the Louisiana State University Coastal Wetlands Soils Characteriza-
twnLab (Wet Soils Lab). Particle size analysis was conducted for particles less than 2.0 mm. Soil samples
analyzed for sand content (0.05-2.0 mm), silt content (0.002-.05 mm), and clay content (<0.002 mm).
^ was separated by sieving. Clay content was determined using the pipette method. For pH and EC,
samples were mixed in a 1:2 ratio (soil weight: water volume) and EC and salinity was read with an EC
“Kter and pH with a pH meter. The 1:2 soil: water method for determining EC is a relatively inexpensive
** duick method for getting an initial salinity estimate (Burt 2004). To get a more accurate measurement
0fEC> we re-sampled broad flat and slick soils at Rambin Bayou and Upper Weyerhaeuser Saline Prairies
®d analyzed each horizon for EC using the saturated paste method. For measurements of CEC and SAR,
NH4°Ac was used as an extractant. CEC was determined by distillation and SAR by Inductively Coupled
Hasma Mass Spectometry (ICP). Soil analyses performed by the Wet Soils Lab follow the Soil Survey Labora-
tory Methods Manual (Soil Survey Staff 1996). Soil nutrient analyses were performed by LSU Soil Testing and
Analysis Laboratory (LSU STPAL) using Mehlic 3 solution as an extractant (Mehlic 1984) and values
ICporiedare mehlic-extractable quantities.
praline praise flora of both sites together included 219 species and infraspecific taxa distributed among
' ***** and 65 families (Appendix 1). Barron Road Saline Prairie had 130 species and Dalton/Dickson
*7* Prairie supported 193. The two sites had 104 species in common. Poaceae was the most diverse family
39 species. Cyperaceae and Asteraceae followed with 22 and 20 species, respectively. Exotic species
accounted for 7.8% (17 taxa) of the total flora. Species rare at the global or state level totaled 20 (9.1%) (Table
2). The rarest globally are Geocarpon minimum, which is federally- listed as threatened and globally imperiled
(G2), and Schoenolirion wrightA, which is globally rare (G3) (NatureServe 2009). Geocarpon minimum was very
abundant at Dalton/Dickson Saline Prairie and rare at Barron Road (see MacRoberts & MacRoberts 2007,
2008, 2009a for monitoring data). Our field work resulted in the discovery of two state records, Lecheasan-
sabeana and Saxifraga texana (Reid et al. 2007; Reid et al. 2008), and the rediscovery of three taxa that were
previously historical for Louisiana: Carex arkansana, M inuartia drummondii, and Gratiolaflava (MacRoberts
et al. 2007; Reid et al. 2008). Table 1 lists characteristic species observed on each landscape position within
the saline prairie complexes including broad flats, slicks (including broader sparsely-vegetated prairie),
pimple mounds, and wet depressions (including broader-scale wet prairie).
The distribution of saline prairie species documented in our study is presented in Figure 6. Percent-
ages of saline prairie species in states and regions decline gradually to the east and northeast and more
steeply to the north and west beyond Texas and Oklahoma. One hundred percent of the taxa recorded at
our study sites are found in Texas. Arkansas and Oklahoma followed Texas in percentages with 97 % and
92 %, respectively. Alabama and Mississippi both have 84% of saline prairie species reported herein.
Pedon descriptions may be found at the following web address: (www.wlf.louisiana.gov/soil-descriptions).
In most cases, natric horizons are designated by an “n” suffix (e.g. Btn) in the pedon descriptions. However,
the “n” suffix in horizon designations is not used with A and E horizons even if pH and SAR values qualify
them as natric. The A and E horizons of slick cores from Dalton/Dickson and Rambin Bayou Saline Prairies
were natric. This suggests that Na is being “wicked” to the surface in water and is concentrated at the surface
by evaporation. A natric horizon in the slick core from Upper Weyerhaeuser Saline Prairie occurred at a
depth of 28 cm. A natric horizon in the Rambin Bayou broad flat core was encountered at 13 cm, while in
Dalton/Dickson and Upper Weyerhaeuser broad flat cores natric horizons first occurred at 46 and 25 cm,
respectively. No natric horizons were encountered in the Dalton/Dickson pimple mound core. At Rambin
Bayou, a natric horizon occurred at a depth of 48 cm in the pimple mound core.
Particle size distributions by landscape position are presented in Table 3. Broad flat and slick soils art
a marked increase in clay content from E to Btn horizon in broad flat and pimple mound soil cores from
Rambin Bayou Saline Prairie, and broad flat and slick cores from Upper Weyerhaeuser Saline Prairie (Table 31
Mehlic-extractable Ca, Mg, and Na were elevated in the broad flat and slick cores compared to levels o
these elements in the pimple mound cores (Table 4). Natric horizons are characterized by high pH and high
SAR values (Table 4). Pimple mound soil horizons for Dalton/Dickson Saline Prairie are acidic throughout
while at Rambin Bayou, horizons were acidic until a depth of 48 cm where a natric horizon was encountered
(Table 4). In broad flat cores, the upper horizons (A and E) were acidic to neutral, with subsurface honzons
being alkaline (Table 4). As mentioned above, slick cores from Dalton/Dickson and Rambin Bayou
natric at the surface, the A and E horizons being alkaline and with high SAR values. Saturated paste E
values for each horizon of broad flat and slick cores from Rambin Bayou and Upper Weyerhaeuser Sai
Prairies are presented in Table 4. For Upper Weyerhaeuser broad flat and slick cores, no horizon had an
EC value that reached 2.0 mS cm1. For Rambin Bayou, the broad flat Btnl and Btkn/El horizons had
values of 2.204 and 2.27 mS cm1, respectively. The El and Btkn/El horizons in the Rambin Bayou she
core had respective EC values of 2.303 and 2.39 mS cm1. Electrical conductivity values of below 2.0
cm-1 are non-saline (Soil Survey Staff 1993). The horizons with EC values above 2.0 mS cm1 fall in the “W
slightly saline” category (Soil Survey Staff 1993).
DISCUSSION AND CONCLUSIONS
The only other comprehensive floristic study of a saline prairie that we are aware of was complete^
MacRoberts et al. (2009a). Their study site was Upper Weyerhaeuser Prairie in north-central Loui^
Aristidalongespica
s
I§
yi
;|33 3S 8§
1 1 1 lillil It li i lit
tal., Vascular flora of saline prairies in
367
(!993) reported 136 taxa. Of these, 46 taxa are also found in our study sites. Species reported from sand-
stone outcrops that were conspicuous at Dalton/Dickson and Barron Road Saline Prairies include Aristida
^spica, A. oligantha, Croton willdenowii, Dichanthelium aciculare, Evolvulus sericeus, Fimbristylis puberula,
Hedeoma hispida, Phemeranthus parvijlorus, Sabatia campestris, and Schoenolirion wrightii (MacRoberts &
Roberts 1993).
Coastal and calcareous prairies are characteristically dominated by perennial grasses including An-
J°Pogon gerardii Vitman, Schizachyrium scoparium, Sorghastrum nutans (L.) Nash, and Panicum virgatum L..
(MacRoberts et. al. 2009b; Allen et al. 2001). Of these, S. scoparium is a component of the saline prairie
• Though often conspicuous on pimple mounds, S. scoparium was not a dominant species of broad flats
® °ur study sites. In our study sites, the most conspicuous grasses in broad flats, which accounted for the
rgest area of any landscape position, were Aristida longespica, A. oligantha, and Sporobolus vaginiflorus var.
^^nis. These species are fall-flowering annuals. They are widespread and occur in a variety of habitats
toduding disturbed areas, dry fields, roadsides, and waste places (Grass Phylogeny Working Group and
2003; Peterson et al. 2003). These species were conspicuous in undisturbed parts of our study sites,
^aps these disturbance-tolerant species were abundant in our saline prairies because they are exploiting
aVailable growing space provided by relatively sparse vegetative cover and taking advantage of seasonally
368
Table 3. Soil particle size distribution and organic matter (OM%) for each horizon of soil cores taken from broad flat, slick,
and pimple mound landscape positions from Barron Road, Dalton/Dickson, and Upper Weyerhaeuser Saline Prairies in
Position Site
Upper Btn/EI
Weyerhaeuser Btng/EI
Btng/E2
Depth (cm) Sand Silt
0-13 39.2 503
13-25 27.3 47.9
25-46 26.1 44.7
46-79 22.6 45.8
79-104 30.3 43
104-132 37.1 41.2
132-158 43.1 39.1
0-8 50.8 44.8
8-13 35.3 49.1
13-69 31.2 41
99-223 64.9 21.9
0-13 34 62.1
13-25 31 61.5
25-53 17.1 62.4
53-86 20.7 55
86-135 22.6 54.8
135-190 5.9 65
190-213 11.8 56.8
Clay
10.5
24.8
293
31.6
21.7
17.8
15.6
277
27
3.9
7.5
24.3
226
29.1
31.4
Dalton/
Dickson
BE 8-28 6.1
Btng/EI 28-51 5.1
Btng/E2 51-89 5.5
Btkng/E3 89-127 6.9
Btkng/E4 127-160 9.1
Dalton/
Dickson
Bt/EI 61-84
Bt/E2 84-119
Btgl 119-157
A.E1.E2 0-23
E3 23-48
Btnl, Btn2 48-111
Btn, El 111-144
Btn/E2, 8C 144-221
57.6 37 5.4
53 41.1 5.9
53.4 40.7 5.9
53.5 43 36
49.3 43.9 6.8
44.9 38.4 167
543 42.1 3.7
48.5 49 26
34.9 47 18.1
29.9 41.3 28.8
59.09 236 1736
Class Matter W
Silt Loam 2.76
Sandy Loam
Loam/Clay Loam
Silty Clay Loam
4.27
2.75
2.32
2.12
3.42
1.39
lOt
187
175
2.87
Sandy Loam
Sandy Loam
Loamy Sand
Clay Loam
Sandy Loam
Sandy Loam
326
297
1.95
161
326
Silty Clay Loam
Silty Clay Loam
353
3.12
2.95
3.78
Sandy Loam
Sandy Loam
Sandy Loam
Clay Loam
Sandy Loam
0.91
088
185
137
0.75
3.49
156
5333338 53333
h
I II I I I I S-SISSiS
3 -isSSSSg s3^r2
£ 1 3S1SS1* aSISS
A $mi*i §1111
A illiilf 2llgl
J liillllillll
iSIllil ssSil
, = a i I * a <aiSl
11 ll
If .....
A conspicuous floristic component of saline prairies consists of late-winter and spring ephemerals
that complete their reproductive cycle while there is adequate moisture. During winter and spring months,
moisture is abundant in the upper soil horizons as rainfall is plentiful and evapotranspiration is low. These
ephemeral species usually disappear by May or June (earlier in some cases). Examples of such species in-
clude Anagallis minima, Crassula aquatica. Coreopsis tinctoria, Geocarpon minimum, Gratiolaflava, Houstonia
SPP-. Krigia dandelion, Krigia occidentals, M inuartia drummondii, Minuartia muscorum, Phacelia glabra, Sabatia
cmpestris, Schedonnardus paniculatus, and Scutellaria parvula.
As the summer sets in, soils of the broad flats and slicks become drier and harder as water in the A and
E horizons is depleted by evapotranspiration. Drought tolerant species such as Iva angustifolia, Opuntia sp„
Mvdbilis albida, and Phemeranthus parviflorus are often the only plants in broad flats and slicks that are in
good condition during the hottest summer months.
The pimple mound flora includes species that would be expected in dry to sub-xeric sandy soils of pine
or hardwood dominated woodlands elsewhere in the West Gulf Coastal Plain. Examples of such species
include Alophia drummondii, Andropogon temarius, Digitaria cognata, Ilex vomitoria, Panicum brachyanthum,
Quercus margaretta, Schizachyrium scoparium, and Vaccinium arboreum. Pirns taeda seedlings and saplings
were relatively frequent on pimple mounds of our study sites. Mature R taeda were rare. Some seedling and
sapling mortality was noted but the cause was not diagnosed. Drought stress is a possibility. Mature trees
on pimple mounds were most frequently Quercus margaretta, and often these trees were of large diameter,
short height, and were twisted and possibly very old.
The soil analyses confirmed the natric horizon characterizations previously described for both the
Bonn and Brimstone soil series (Soil Survey Staff 2006; Soil Survey Staff 1998; Soil Survey Staff 1991; Soil
Survey Staff 1980). Soil textural data from the extracted cores underscored the differences between landscape
Positions. Pimple mounds were notably sandier than the broad flats and slicks, while the silt component
dominated the broad flat and slick cores (Table 3).
(T HI11* SUrfaCC horizons in our study sites’ with the exception of slicks, were acidic with low SAR values
ab e ^ Obligate halophytic species are lacking in our study sites. Based on personal observation at our
study sites and elsewhere, Sporobolus pyramidatus could be considered halophytic. In our study areas it
^restricted to and common on slicks. Peterson et al. (2003) report S. pyramidatus to occur in “disturbed
301 . roadsides, railways, coastal sands, and alluvial slopes in many plant communities.” Iva angustifolia is
a *ry conspicuous herb on broad flats. Habitat accounts for Iva angustifolia by Gandhi and Thomas (1989)
rather (2006) indicate that it grows in disturbed areas, waste places, and overgrazed pastures, with
^“tention of saline or natric soils. Since this species colonizes areas impacted by oilfield brine (personal
rvation), it might be considered a facultative halophyte. It appears that the vegetation of saline prairies
«rgely controlled by the influence of the natric subsurface horizons on soil moisture content and avail-
ty rather than chemical characteristics in the plant rooting zone.
Kej ,The community name “saline prairie” has been used by Louisiana Natural Heritage Program (2009b),
“ « aL (2004), Lester et al. (2005), MacRoberts et al. (2009a) and Mclnnis et al. (1993), among oth-
comh me S°ilS ^ high levds °f dissolved salts> the concentration of which is measured by electrical
but rUCtlvi[y (Hassett & Banwart 1992). Natric soils do not have excessive concentrations of dissolved salts
n,Jalh,er have high concentrations of exchangeable Na. While salinity will negatively impact sensitive
ICS of natric soils are not necessarily present in saline soils (Hassett
Jws, the adverse physical character!
^Banwart 1992). The sodium adsorr
GoilSui
lrvey Staff 1993). '
red the standard r
1 sodicity
s of calcii
^sium ions that tend to moderate the adverse effects of Na. Sodium ac
ourb C°ml>ined widl alkaline pH, identify a natric horizon (Table 4). For n
4) D^dflat and slick cores, SAR values exceed the threshold range, and i
3Ce Water- When dry they can be very hard and resist wetting. In both cases, plants will have difficulty
372
obtaining soil water. Our analysis of soil cores for saturated paste EC shows that a few subsurface horizons
from one site are very slightly saline according to Soil Survey Staff (1993) and soils at the other study site
are completely non-saline. Hassett and Banwart (1992) consider soils to be saline if EC is 4 mS/cm or higher,
well above EC values in our data. Thus, applying the term “saline” to the natural community name maybe
misleading. Arkansas Multi-Agency Wetland Planning Team (2001) avoids this inaccuracy by applying the
name “Alkali Flat,” which accounts for the natric soils and generally flat topography.
The floristics and edaphic traits of saline prairies are the principal factors that define these natural
community systems. Based on the International Vegetation Classification (IVC) system, NatureServe (2009)
places saline prairies and the associated saline oak woodlands within the South-Central Saline Glade eco-
system (CES203.291). This glade ecosystem is broken down into several vegetation associations with four
prairie associations and one woodland association known from Louisiana. The most frequently identified
association in Louisiana is the Artistida longespica Poir. - Schizachyrium scoparium - Diodia teres Saline Her-
baceous Vegetation (CEGL008419), and it seems to be most closely matching association for saline prairies
in northwest Louisiana. This association was described based principally on saline prairies in Arkansas,
and therefore some differences in plant species occur for the Louisiana prairies. For example, Schizachyrium
scoparium, Diodia teres, and Sabal minor while present, are not dominants in Louisiana examples we have
studied.
The historical role of fire in saline prairies is not known. Sparse vegetative cover may have reduced
the role of fire in these systems (NatureServe 2009). Extreme soil properties and droughty conditions are
apparently sufficient to maintain prairies in our study sites, which have no threat of woody encroachment.
While prairies studied by us appear to be edaphically extreme enough to prevent woody encroachment and
to maintain a grassland community, fire may have been very important in adjacent communities and thus
important in a broader landscape context.
Studies of floristics and edaphics of additional saline prairies throughout the range of the community
would expand information on geographic variation and enable refinement of the natural community clas-
sification. Quantitative sampling of vegetation and edaphic variables in random or systematically placed
plots would enable correlations of vegetation to soils on a fine scale. This type of procedure may reveal
important factors determining the vegetation composition, density, and coverage in saline prairies that are
There is a wet-variant saline prairie that is sometimes adjacent to the saline prairie community described
in this paper (Louisiana Natural Heritage Program 2009b). These prairies occur on lower landscape po-
tions, experience regular inundation, and support halophytes more characteristic of coastal localities such
as Atriplex cristata Humb. & Bonpl. ex Willd., Distichlis spicata (L.) Greene, Heliotropium curassancum -
and Solidago sempervirens L. (Smith 1996). In Louisiana, Spartina pectinata Bose ex Link is almost entirely
restricted to wet-variant saline prairies. Wet-variant saline prairies are in need of similar baseline florist*
and edaphic study.
APPENDIX 1. CHECKLIST OF SPECIES
Species recorded from Barron Road Saline Prairie (B) and Dalton/Dickson Saline Prairie (D). Key to voucher sPecim^^
Christopher Reid, whose specimens are deposited at LSU; MM = B.R. and M.H. MacRoberts, whose specimens
at LSU and LSUS. If there is no mention of landscape position, broad flats is intended. Species of slicks are treated as
ring in broad flats in this list. Table 1 lists characteristic species of each landscape position, including slicks. Rr“ '
denoted with a dagger (t). State- or globally-rare species are summarized in Table 2, where conservation status ra
provided. Exotic species are denoted with an asterisk (*).
Acanthaceae
Justicia ovata (Walt.) Lindau var. tanceolata (Chapm.) R. W.
Long— D, wet depressions, MM 7816
Ruellia humilis Nutt-B D, broad flats and pimple mounds
,s ranks a*
t depressions, *
CR 5931, MM 7245. 7447, 7954
Allium canadense L. var. mobilense (Regel) M.
MM 7354, 7384 (observed but not collected a
Metal., Vascular flora of saline prairies in Louisiana
373
Hothoscordum bivalve (L) Britt.—
Kooperia drummondii Herbert
} MM 7292, 7604
3 D, CR 5655, MM
fPterocaulon virgatum (L.) DC. — D, pimple mounds,
5927
Pyrrhopappus carolinianus (Walt.) DC. — B D, CR 7018, N
7416
Symphyotricum pratense (Raf.) Nesom— B, MM 7283
flhu scopallinum L.— B, pimple mounds, MM 7260
Daucuspusillus Michx.— D, pimple mounds, CR 7001
limciadium pinnatum (DC.) Mathias & Constance— B D,
broad flats and wet depressions, CR 61 29, 6982, MM 7403,
7414,7421,7749,7774,7799
t&mhim nuttallii (DC.) Britt. — B D, pimple mounds, broad
flats, aid wet depressions, CR 6174, MM 7422, 7444,
7800
kpocarpus aethusae Nutt, ex DC— B, MM 7440
Aquifoliaceae
ktmitoria Ait.— B, pimple mounds, CR 7199, MM 7261
Polypremum procumbens L. — D, pimple mounds, MM 8050
Callitriche pedunculosa Nutt. (=C. nuttalliilorr.) — B, MM 7306
Campanculaceae
Lobelia appendiculata A. DC.— B D, pimple mounds, CR 7000,
MM 7400
Triodanis perfoliata (L) Nieuwl. — B D, CR 5734, MM 7350,
iple mounds, MM 7450,
flats edges (upslope), CR
^ MM 7464
\jjj* retusa (Lam.) Sch. Bip.-D, pii
Gray— B D, CR 6996, MM 7337,
brr. & A. Gray) Greene— 8 D, CR
iMinuartia drummondii (Shinners) McNeill — B D, CR 5721,
MM 7308, 7586, 761 5
IMinuartia muscorum (Fassett) Rabeler— B D, CR 5712, 5720,
601 3, MM 7309, 731 7, 7331 , 761 2, 7723
Cistaceae
iHelianthemum rosmarinifolium Pursh— 8 D, pimple mounds,
CR 5770, 5776, MM 7431, 7401, 7739, 7804
iLechea san-sabeana (Buckley) Hodgdon — B D, pimple
mounds, CR 6064 (liX-LL), 6179, MM 7738, 7824, 7827,
7828,7832,7834
bechea tenuifolia Michx.— B D, broad flats and pimple mounds,
CR 61 76, MM 7829, 7822, 7823, 7951
r.&A.Gray — BD,CR
5662, 6126, MM 7250, 7462, 7356, 7446, 8052
Hypericum gentianoides (L.) B5.P. — D, MM 7874
Hypericum hypericoides (L) Crantz— B, pimple mounds, MM
nd pimple
mounds, CR 7016, MM 7348, 7602, 7730
Tradescantia occidentals (Britt.) Smyth— B D, CR 6075, MM
7346,7347,7755,7603
Per Z Utt ~B D> CR 6005' MM 730°, 7598
«« i Michx.— d, pimple mounds, CR 5835
«ytosa Nutt, ex DC— B D, pimple mounds, CR
7351, 7378
Cyp
MM 781 1,7770
Cyp
7775
376
Schizachyrium scoparium (Michx.) Nash — B D, broad flats and
pimple mounds, CR 5990, MM 7271, 8051
Setaria pan/iflora (Poir.) Kerguelen — B D, CR 6758, MM 7266,
7952
*Setaria pumila (Poir.) Roem. & Schult. subsp. pallidefusca
(Shumach.) B.K. Simon— B, MM 7443
Sphenopholis filiformis (Chapm.) Scribn.— D, CR 6079
Sporobolus compositus (Poir.) Merr. var. macer (Trin.) Kartesz &
Gandhi— D, pimple mounds, CR 5932, 6752, 6760
Sporobolus pyramidatus (Lam.) A.S. Hitchc.— B D, CR 5654,
5660, 5780, 5928, 6125, MM 7277, 7437, 7873
Sporobolus vaginiflorus (Torn ex A. Gray) Alph. Wood. var.
vaginiflorus — B D, CR 5659, 5987, MM 7272
Steinchisma hians (Ell.) Nash — B D, broad flats and wet depres-
sions, CR 61 38, MM 7268, 7254, 7773, 7888, 7956
Tridens flavus (L.) Hitchc. var. fiavus — D, broad flats edges,
CR 6755
Tridens strictus (Nutt.) Nash— B D, wet depressions, CR 5985,
MM 7269, 7891, 7942
Urochloa platyphylla (Munro ex C. Wright) R. D. Webster— B D,
MM 7275 (observed but not collected at D)
Vulpia octoflora (Walt) Rydb.— D, CR 61 33, 6998
Polygonaceae
*Polygonum aviculare L— B, CR 5665
Polygonum hydropiperoides Michx.— D, wet depressions, CR
5761, MM 7771, 7871
Polygonum ramosissimum Michx.— B D, MM 7445 7821
7876
Rumex hastatulus Baldw. — B D, CR 5717, 6143 MM 7298
7592
Claytonia virginica L.— 8 0, CR 5997, MM 7294, 7610
Nutt) — B D,?R 5593, 5725, MM
Potamogeton diversifolius Raf— D, wet depressions MM
7736
3 D, MM 7307 (observed b
niKiiivnetaroiiniana wait. — D, CR 5998
Delphinium carolinianum Walt. — D, pimple mounds CR
6992
Ranunculus fascicubris Muhl. ex Bigelow— D, pimple mounds,
* Ranunculus sardous Crantz — B D, CR 6991 , MM 7318
Rubus louisianus Berger (= R. argutus Link)— B, pirr
mounds, MM 7372
Rubus trivialis Michx. — B, pimple mounds, MM 7426
Diodia teres Walt— B D, CR 5662, MM 7801, 7869, 7943
Galium tinctorium (L) Scop.— B, MM 7345, 7386
Houstonia micrantha (Shinners) Terrell — B D, CR 5999, 1
Zanthoxylum clava-herculis L. — D, pimple mounds, CR 7202,
Sideroxylon lanuginosum Michx.— B D, broad flats and pimple
mounds, CR 6142, MM 7264, 7591, 7725 i;
Saxifragaceae
Lepuropetalon spathulatum Ell. — B I
not collected at D)
t Saxifraga texana Buckley — D, CR 6001 , MM 81 23
Scrophulariaceae
Agalinis heterophylla (Nutt.) Small ex Britt— D, MM 8046
fGratiola flava Leavenworth— B D, CR 5724, 6004, 6009, MM
7285,7312, 7584
Gratiola virginiana L.— D, wet depressions, MM 7733
Nuttallanthus canadensis (L) D. A. Sutton— B D, CR 6082,69ft
MM 7295, MM 7600
Smilax rotundifolia L. — B, pimple mounds, MM 7276
Solanaceae
Solanum carolinense L. — B, pimple mounds, MM 7417
Ulmaceae
Ulmus alata Michx.— B D, pimple mounds, CR 6132, 72»
MM 7259
Ulmus crassifolia Nutt. — D, pimple mounds, MM 7764
Valerianaceae
Valerianella radiata (L.) Dufr. — B D, CR 6076, MM 73 .
Verbena halei Small— 0, CR 6999, 1
17819,7879
ACKNOWLEDGMENTS
We thank landowners Ralph and Becky Dalton and George Dickson for allowing our surveys on Dak
Dickson Saline Prairie. Sidney Evans and James Taylor allowed us to survey Barron Road Saline Tt»
Alan Boyd of Weyerhaeuser Company facilitated our research on Upper Weyerhaeuser Prairie. Barry _
of Hancock Forest Management allowed our work on Rambin Bayou Saline Prairie. Beverly AnderS°n
also owns a portion of Rambin Bayou Saline Prairie, kindly allowed access. Robert Love and Tommy
377
informed us about saline prairies in Sabine Parish and accompanied us on field surveys to those sites and
we appreciate their help and interest. We thank Haus Cordray of the De Soto Parish Tax Assessor’s office for
providing landowner information by phone and fax on many occasions. Garland Weidner kindly provided
laid ownership for Caddo Parish on several occasions. Charles Allen vetted some of our grasses, particu-
larly Dichanthelium. Lucas Majure examined material of Opuntia from our study sites and advised us on the
systematics of this group and we appreciate his input. Janice Swab determined several difficult Juncus for
us and we appreciate her help. Lowell Urbatsch assisted with determinations of difficult Asteraceae and
Juncaceae. Diane Ferguson assisted with determinations of Cyperus. We thank Charlie Henry and Mitchell
Mouton for extracting soil cores for us. Michael Lindsey provided advice and information pertaining to
measuring electrical conductivity and we appreciate his assistance. Manoch Kongchum and staff performed
many of the soils analyses and provided us with detailed information pertaining to methods. Nicole Lorenz
kindly prepared Figure 5. Paul Heinrich directed us to some helpful geological references. Stephen Faulkner
viewed an early draft of the manuscript and provided helpful comments. We appreciate thorough reviews
of the manuscript by Wayne Hudnall, David Rosen, and David Weindorf.
REFERENCES
tows, D.L 1 960. The Louann Salt and its relationship to Gulf Coast salt domes. Gulf Coast Assoc. Geol. Soc
Trans. 10215-240.
to^CM, M. Vidrine, B. Borsari, and L. Allain. 2001 . Vascular flora of the Cajun Prairie of southwestern Louisiana.
Proc 1 7th NA. Prairie Conf. 1 7:35-41 .
toNSAs Multi-Agency Wetland Planning Team. 2001 . Wetlands in Arkansas: Alkali Flat, (http://www.mawpt.org/
wetlands/classification/community_types.asp?communityType=Alkali+Wet+Prairie) (accessed 18 March
2010).
tonsAs Natural Heritage Commission. 2006. Warren Prairie Natural Area, http://www.naturalheritage.org.
ed. 2004. Soil Survey Laboratory methods manual. Soil Survey Investigations Rpt. 42, Version 4.0. USDA.
Natural Resources Conservation Service.
tyllH. 1974. Pimple mounds: a new viewpoint. Ecology 55:1 78-1 82.
^'Zonlcom. 2004. http://www.climate-zone.com/climate/united-states/louisiana/shreveport/.
Lipscomb, M.D. Reed, and RJ. O'Kennon. 2006. Illustrated flora of EastTexas. Sida, Bot. Misc. 26:1-1 594.
i ^ Now America Editorial Committee. 1 993-2007. Flora of North America north of Mexico, Vols. 1 -26. Oxford
Univ. Press, NY.
R D- Thomas. 1989. Asteraceae of Louisiana. Sida, Bot. Misc. 4:1-202.
Phytogeny Working Group and K.W. Allred. 2003. Aristida. In: Flora of North America Editorial Committee,
Nora of North America North of Mexico, Vol. 25: Magnoliophyta: Commelinidae (in part): Poaceae, part
^42. Oxford Univ. Press, NV.
•U AND W.L. Banwart. 1 992. Soils and their environment. Prentice Hall, Englewood Cliffs, NJ.
Wmf 1962‘ Saiine‘a,kali soils in Arkansas. Arkansas Farm Res, 1 1(2):10.
^ EM. Rutledge, H.C. Dean, and M. Uwson. 1 964. Classification and genesis of some solonetz (sodic) soils
e*tern Arkansas. Soil Sci. Soc. Proc. 28(4):688-691 .
>939. Geology of Caldwell and Winn Parishes. Department of Conservation, Louisiana Geological Survey,
G' ;■ ’991. Salt Tectonics. In: Goldthwaite, D. ed. An introduction to Gulf Coast geology. New Orleans
C?al Society, New Orleans, Louisiana. P. 31-60.
y ,an°C.M. Meacham, 1999. Synthesis of North American flora. Chapel Hill, North Carolina. CDROM.
1i(Kj|l| S'nghurst, and S. Cook. 2004. Geocarpon minimum (Caryophyllaceae), new to Texas. Sida 21:
^sen, P.L Faulkner CS. Reid, and I.E. Maxit. 2005. Louisiana comprehensive wildlife conservation
^•Louisiana Department of Wildlife and Fisheries, Baton Rouge.
Soil Set
380
BOOK REVIEW
Allan E. Bessette, David B. Harris, and Arleen R. Bessette. 2009. Milk Mushrooms of North America: A
Field Identification Guide to the Genus Lactarius. (ISBN 978-0-8156-3229-0, hbk., alk. paper).
Syracuse University Press, 621 Skytop Road, Suite 110, Syracuse, New York 13244-5290, U.S A (Orders:
supress@syr.edu, 1-800-848-6224, 315-443-5545 fax). $110.00, 256 pp., 263 color illustrations,
glossary, index, 7" x 10".
J. Bot. Res. Inst. Texas 4<1): 380. 2010
^ FLORISTIC COMPOSITION AND POTENTIAL COMPETITORS IN
LINDERA MEUSSIFOLIA (LAURACE AE) COLONIES IN MISSISSIPPI
WITH REFERENCE TO HYDROLOGIC REGIME
Tracy S. Hawkins1-2, Daniel A. Skojac Jr.3, Nathan M. Schiff, and Theodor D. Leininger
USDA Forest Service
Center for Bottomland Hardwoods Research
Stoneville, Mississippi 3 1 776, U.SA
ABSTRACT
RESUMEN
INTRODUCTION
Species
^melissifolia (Walter) Blume is a dioecious shrub endemic to the southeastern United States. This member
*l»aceae, commonly called pondberry, was collected in South Carolina and described by Walter (1788)
®Uurus wtissaefolia. By the mid-20th century, Steyermark (1949) found that few herbarium specimens
“dbeen collected subsequent to Walter’s (1788) description, with only three collections having been made
18 ^ first half of the twentieth century. Herbarium studies by Steyermark (1949) throughout the United
^confirmed the historical presence of L. melissifolia in Alabama, Florida, Georgia, Louisiana, Missouri,
T^Grolina, and South Carolina. Although Gattinger (1901) made reference to L. melissifolia in Tennessee,
^er®lark(1949) believed this report probably referred “...to misidentified pubescent Lindera Benzoin, since
Wtnentic material of L. melissaefolia has been found in the herbarium of the University of Tennessee... *
7^’ (1924) excluded L. melissifolia from “Shrubs of Indiana” citing only one historic report with
joucher specimen. Based on the number of preserved L. melissifolia, Steyermark (1949) concluded that
^Cies may be one of the rarest shrubs in the United States.
J^atly, extant populations of L. melissifolia are present in Alabama, Arkansas, Georgia, Mississippi,
Zj*"’ North Carolina, and South Carolina. It is believed to be extirpated in Louisiana and Florida. In the
2** Coastal Plain (South Carolina and Georgia), L. melissifolia grows along the margin of seasonally
J*" depressional wetlands dominated by Nyssa biflora Walter and Taxodium ascendens Brongn. (Alenc &
^ 2005). The only population known to occur in Alabama grows along the edge of a forested depres-
fill
382
Journal of the Botanical Research Institute of Texas 4(1)
2005), In the Lower Mississippi Alluvial Valley (LMAV), L. melissifolia populations in Mississippi, Arkansas,
and Missouri grow in periodically flooded, bottomland hardwood forests underlain by hydric soils. Forest
canopy composition among the sites in the LMAV is similar, composed primarily of trees designated as
facultative wetland or obligate wetland species (Hawkins et al. 2009a). Liquidambar styraciflua, Acer rubrum,
Q. lyrata , Q. nuttallii, and Q. phellos are important canopy components; however, the relative importance of
these and other canopy species varies between sites (Hawkins et al. 2009a).
Lindera melissifolia populations consist of spatially segregated, unisexual colonies (Hawkins etal. 2009b).
Colony sizes range from approximately 20 stems to >1,000 stems (Morris 1987; Devall et al. 2001; Hawkins
pers. obs.). In the LMAV, populations are strongly male-biased, with male to female colony ratios ranging
from 7:1 (Wright 1994) to 19:1 (Hawkins et al. 2009b).
In both male (pollen-bearing) and female (seed/fruit-bearing) L. melissifolia, anthesis occurs in bte
spring, often during flooded conditions, and precedes leafing out. Small, yellow flowers are produced on
axillary inflorescences (Fig. 1A), followed by production of green drupes on female plants. Approximately
90 days following anthesis, drupes contain a single, fully developed seed (Connor et al. 2007) and upon
maturation drupes are bright red (Fig. IB, 1C). Drupes are dispersed from fall to early winter (Smith etal.
2004) and fruit pedicels may remain on the plant until the following spring. Female L. melissifolia appear
to invest heavily in sexual reproduction (Connor et al. 2007); however, seedlings are rarely observed in
naturally occurring populations (Wright 1990). Vegetative propagation of ramets from rhizomes appears to
be the predominant form of reproduction (Wright 1990, 1994).
During the first one to three years following seedling and/or ramet emergence, L. melissifolia plants
are morphologically very similar to L. benzoin (L.) Blume. In fact, during Steyermark’s (1949) investigation
of the species, he noted numerous misidentified herbarium specimens of L. melissifolia resulting from this
close similarity. Although leaf pubescence and leaf size sometimes differ between the two species, these
characteristics may be dependent on season and locality. On the other hand, the angle of the lateral veins
in the leaf blade may be used to distinguish the congeners. In L. melissifolia, the lowest two pairs of lateral
veins of the leaf blade diverge at a 45°-50° angle from the midrib, and distal to this, successive veins diverge
at approximately 35° (Steyermark, 1949). In other words, the lower two lateral veins are not parallel to the
successive upper veins. In contrast, all lateral veins of L. benzoin leaves diverge from the midrib at the same
angle (35°-45°); therefore, are parallel (Steyermark 1949). Both Nuttall (1818) and Steyermark (194#*r
scribed the fruits of L. melissifolia as “larger” than those of L. benzoin. The senior author has found the seeds
of the congeners to be consistently reliable for definitive identification of fruiting plants. While seeds of L
melissifolia are spherical and light brown to yellowish brown those of L benzoin are oval and dark biow»
(Fig. ID).
In 1986, L. melissifolia was listed under the Federal Endangered Species Act of 1973 (U.S. Fish and
Wildlife Service 1986). Recent research has provided some insight into the ecophysiology (Hawknwjw^
2009b; Aleric & Kirkman 2005) of this species, and forest types and forest structure associated wi*T
melissifolia in the LMAV have been reported (Hawkins et al. 2009a). Wright (1990) suggested speck*
Brunnichia, Rubus, and Smilax as main competitors of L. melissifolia in northeast Arkansas, and that peri**
flooding served to minimize competitive interactions. Beyond Wright’s (1990) initial report, very
is known about the microhabitat or biotic interactions in L. melissifolia colonies. In an effort to P^
concise information regarding the microhabitat and general ecology of L. melissifolia, three native, disjo**
populations in Mississippi were monitored for three years. The objectives of our study were to 1) asse®
vascular plant checklist of groundcover species growing within L. melissifolia colonies, 2) identify P*#**
competitors of L. melissifolia, and 3) monitor hydrologic regime for L. melissifolia colonies.
L, Lindera melissifolia colonies
L. melissifolia, and (D) seeds
384
Journal of the Botanical Research Institute of Texas 4<1)
METHODS
Study Sites
Two of the study sites are in Sharkey County, Mississippi. The North Delta National Forest site (NDNF) is a
25-ha tract of Delta National Forest that has been under management by the U. S. Forest Service since 1938
(Devall & Ramp 1992). The South Delta National Forest (SDNF) site is a 10-ha tract of forest approximately
9 km southeast of NDNF. The third study site (BC) is in Bolivar County, Mississippi, and is a privately owned
30-ha forest fragment surrounded by agricultural fields (see Hawkins et al. 2009a for map and detailed
descriptions). Soil association for the three sites is Dowling (very fine, smectitic, nonacid, thermic Venic
Endoaquepts)-AUigator (very-fine, smectitic, thermic Chromic Dystraquerts)-Sharkey (very-fine, smectitic,
thermic Chromic Epiaquerts), an association characterized by poorly drained, fine-textured clayey surface
soils and subsoils formed from Mississippi River alluvium (Rogers 1958, Scott & Carter 1962). Mean annual
temperature for Sharkey and Bolivar counties is approximately 18.0°C, and total annual precipitation ranges
from 1228 mm to 1319 mm (Rogers 1958, Scott & Carter 1962).
Data Collection and Analysis
In autumn 2003, rectangular plots were established for selected L. melissifolia colonies at BC (Ncoiony * 6), NDNF
( Ncoiony = 10), and SDNF (Ncoiony = 1). In 2000, SDNF was habitat for numerous L. melissifolia colonies, many
with >200 stems (GSRC 2000). However, during reconnaissance of this area we found only one remaining
colony and were able to establish only one study plot. The perimeter of each plot was positioned lm beyond
the outermost L. melissifolia stems of a colony. A 1.2 m wooden stake (5 cm x 5 cm) was established at each
of the four comers of the rectangular plot. Colony sizes were variable, and thus plot areas varied with colony
size. Within each plot, lm x lm quadrats were marked with pin flags along the diagonals of the rectangular
plot; therefore, the percent of plot area sampled was the same among plots.
In June 2004, May 2005, and June 2007, L. melissifolia stems in each lm x lm quadrat within the {Jots
were counted, and groundcover species were identified and stems were counted. Common plants were
identified in the field by TSH and DAS. When field identification was problematic, plants in question were
top-cropped and taken to the lab for identification using Radford et al. (1968), supplemented by Godfrey
and Wooten (1979, 1981).
From 10 November 2004 to 22 May 2007, plots were visited bi-weekly. When plots were not Hooded,
two soil samples were collected from each plot and placed in individual 141 cm3, hermetically sealed, mewl
containers. Soil samples were taken directly to the lab, weighed to the nearest 0.01 g, dried in an oven at
70°C for 48 hr, then weighed again. Percent moisture for each soil sample was calculated by dividing the
difference of initial (wet) and final (dry) weights by the initial weight and multiplying the quotient by W-
During flooded conditions, water depth was measured to the nearest 0.1 cm at each plot comer post and
the mean (± SE) used to represent water depth for the plot.
At NDNF, there were substantial differences in water depth among some of the L. melissifolia colonies
therefore, mean water levels for each sampling date at each colony were compared using a one-way analysis
of variance (ANOVA; SAS 2001). For colonies where mean water level was not significantly different (p
0.2421), data were pooled producing one mean ± standard error for each of three groups of colonies.
RESULTS AND DISCUSSION
Hydrologic regime for forests with Lindera melissifolia populations is often described within the broad defini-
tion of “periodically flooded”. For populations in Mississippi, flooding generally occurred in
vever, flooding events and flood duration were found to vary among sites. The BC
d annual flooding events of comparable depth and duration in 2005, 2006, and 2 ^
t this site is artificially controlled by the lan
-spring. !
refore, hydrologic regime i
isistent from year to year. .
Flood duration and initial time of flooding varied with year at NDNF. Flooding occurred m #0*
not in 2007 (Fig. 2B). Although water depth differed among some of the ten monitored c
2006, but i
NDNF, this had no influence on time and duration of individual colony inundation (Fig. 2B). Difference in
water depth among colonies is the result of surface topography (e.g. sinks or sloughs). Lack of difference in
time and duration of flooding between colonies results from rapid rising and dropping of water. In contrast
to NDNF and BC, SDNF did not experience flooding in 2005, 2006, and 2007 (Fig. 3).
When L. melissifolia colonies were not flooded, soil moisture content was comparable among the three
sites, ranging from 20% to 30% throughout the year, with the exception of mid- to late-summer, when soil
moisture contents dropped as low as 15% to 18% for BC and NDNF (Figs. 2A and 2B); and as low as 10%
at SDNF (Fig 3).
The combined annotated list of vascular plants (including L. melissifolia ) for the three study sites includes
70 species in 57 genera in 45 families (Appendix 1) and of these, 9 species have the potential to become
weedy or invasive (SWSS 1998). The contribution of L. melissifolia to overall colony composition ranged from
approximately 20%-40% at NDNF and BC, and 5%-15% at SDNF (Fig. 4). Other species growing within the
colonies are typical of bottomland hardwood forests in this area of the LMAV, and tend to reflect hydrologic
regime at each study site. Forty-nine percent of the 69 species identified as growing in association with L.
melissifolia (Appendix 1) have a wetland indicator status of FACW and a 67%-99% probability of occurrence
in a wetland area (USDA, NRCS 2008). However, the presence of Callicarpa americana at SDNF is atypical
for bottomland forests in the LMAV and represented a county record first reported in 2007 (Skojac et al.
2007). Lack of inundation at SDNF (Fig. 3) has possibly allowed establishment of C. americana, as well as
other species, such as Asplenium platyneuron and Phytolacca americana, that generally are not found in forests
prone to flooding.
Of the 69 species growing in association with L. melissifolia, few appeared to pose an immediate
above-ground competitive threat and the ratio of stem density for these plants to L. melissifolia stem density
remained relatively stable throughout our study (Fig. 4). The predominant growth habit in L. melissifolia
colonies was vines (Fig 4). Wright (1990) considered Brunnichia ovata as a plant with potential to be an ag-
gressive competitor of L. melissifolia; however, we observed early-summer emergence of B. ovata, as well as
Toxicodendron radicans, after L. melissifolia plants had flowered and leafed out. Both B. ovata and T. radicans
remained prostrate throughout the season and did not compromise light capture by L. melissifolia leaves,
nor twine around or climb L. melissifolia stems. On the other hand, Smilax spp. (S. bona-nox, S. glauca, S.
rotundifolia, S. tamnoides) and Vitis spp. (V. aestivalis, V. palmata, V. rotundifolia ) have potential to become
strong competitors, by remaining upright throughout the year, and in some cases, using L. melissifolia stems
for above-ground support.
Many of the associated species in L. melissifolia colonies in Mississippi do not appear to have direct
competitive impact on L. melissifolia populations. However, species with a vining growth habit and/or those
with potential to become weedy or invasive (Appendix 1), should continue to be monitored. The response
of these species to natural or anthropogenic disturbances has the potential to alter competitive interactions
within these L. melissifolia populations.
APPENDIX 1
Plants included in this checklist are compiled in alphabetical order by family within two major groups (Monilophytes
Angiosperms). Genera and species are alphabetical within each family. Scientific nomenclature and common names
The Plants Database (USDA, NRCS 2008). Family circumscriptions for monilophytes follow The Plants Database (USDAN
2008), and for angiosperms, APG (Stevens 2008). Plants with potential to become weedy or invasive (SWSS, 1 998) aredeno^
with an asterisk (*) before the species name. Species wetland indicator status for the Southeast Region (AL AR,
NC,SC,TN) is noted as: obl = obligate (99% probability of occurrence in wetlands); facw = facultative wetland (67%'W*^
ability of occurrence in wetlands, but occasionally found in non-wetlands); fac = facultative (equal probability <*
m wetlands or non-wetlands); facu = facultative upland (occurrence usually in non-wetlands, occasional w
and ni - unable to determine wetland indicator status based solely on genus. Locality data are noted as: (1) - BotlVM-
MS; (2) - North Delta National Forest, Sharkey County, MS; and (3) = South Delta National Forest, Sharkey County.
388
Journal of the Botanical Research Institute of Texas4(1)
| ■ Forbs/Herbs KS Trees/Shrubs □ Vines □ Pondberrv
-NDNF 1 SDNF 1 BC ■
per unit area (stems/m2).
Melothria pendula L, Guadeloupe ci
Cyperaceae
Carex crus-corvi, Shuttlw. ex Kunze, r
Carex louisianica LH, Bailey, Louisiar
Carex tribulokles Wahlenb., blunt brc
Gleditsia triacanthos L, honeylocust, fac, (13
Fagaceae
Quercus lyrata Walter, overcup oak, obl, (U)
Quercus nigra L., water oak, fac, (3)
Quercus pheltos L, willow oak, facw, (1,23)
Quercus texana Buckley, Texas red oak, obl, (1
Uquidambar styraciflua L, sweetgum, fac, (1,
Carya aquatica (Michx. f) Nutt, water hickor
*Cocculus carolinus (L.) DC., Carolina coralbead, fac, (2,3)
Moraceae
Forestiera acuminata (Michx.) Poir., swamp privet, obl (2)
Fraxinus pennsylvanica Marsh, green ash, facw, (1 33)
Fraxinus profunda (Bush) Bush, pumpkin ash, obl, •
Ludwigia glandulosa Walter,
Orchidac
?s ovalis Lindl, October lady's tresses, n
*Passifiora lutea L, yellow passionflower, fa
Phytolaccaceae
* Phytolacca americ
i L„ American pokeweed, f*
389
'Sr unnichia ovata (Walter) Shinners, redvine, facw, (1,2,3)
Polygonum virginianum L, jumpseed, fa c, (2)
d leatherflower, facw, (2)
Smilax glauca Walter, cat greenbrier, fac, i
Smilax rotundifolia L. roundleaf greenbrk
Smilax tamnoides L, bristly greenbrier, fai
Styracaceae
Styrax americanus Lam, American snowbell, facw, (2)
Ulmaceae
i supplejack, facw, Planera aquatica J.F. Gmel, water elm, obl, (2)
Ulmus americana L, American elm, facw, (1 ,2,3)
Rubus trivialis Michx,
Boehmeria cylindrica (L.) Sw, false nettle, facw, (23)
buttonbush, obl, (2)
Callicarpa americana L, American beautyberry, facu, (3)
Violaceae
Viola sp. L, violet, ni, (3)
*Ampelopsis arborea (L) Koehne, peppervine, fac, (133)
*Parthenocissus quinquefolia (L.) Planch, Virginia creeper.
fac, (1,23)
Vitis aestivalis Michx. summer grape, fac, (2)
Vitis palmata Vahi, catbird grape, facw, (1,2)
Vitis rotundifolia Michx, muscadine, fac, (23)
ACKNOWLEDGMENTS
Ihe authors thank Greg Comer, Stephanie Skojac, and Theran Stautz for their assistance in the field, U. S,
Fish and Wildlife Service for permits, the U. S. Army Corps of Engineers for underwriting the cost of this
research, and Drs. Charles Bryson, Emile Gardiner, and Edward W. Chester for review of an earlier draft
°f this manuscript.
REFERENCES
Alewc' K-M. and LK. Kirkman. 2005. Growth and photosynthetic responses of the federally endangered shrub,
Lindera melissifolia (Lauraceae), to varied light environments. Amer. J. Bot. 92:682-689.
Cqnn^ K, G5. Schaefer, J. Donahoo, M. Devall, E. Gardiner! Hawkins, D. Wilson, N. Schiff, P. Hamel, and! Leininger. 2007.
Development, fatty acid composition, and storage of drupes and seeds from the endangered pondberry
(Undera melissifolia). Biol. Cons. 137:489-496.
DtAM.CC. 1924. Shrubs of Indiana. Indiana Department of Conservation, Indianapolis. Pub. No. 44.
Dewu» MS. and P.F. Ramp. 1992. U.S. Forest Service research natural areas and protection of old growth in the
Nat. Areas. J. 1 2:75-85.
N- Schiff, and D. Boyette 2001 . Ecology and reproductive biology of the endangered pondberry, Undera
Melissifolia (Walt.) Blume. Nat. Areas J. 21 :250-258.
I1*®* A- 190! The flora of Tennessee and a philosophy of botany, respectfully dedicated to the citizens of
ennessee.„ Press of Gospel Advocate Publishing Co, Nashville, TN.
R k- and J.W. Wooten. 1 979. Aquatic and wetland plants of the southeastern United States. Monocotyle-
5*** University of Georgia Press, Athens.
^ UK. and J.W. Wooten. 198! Aquatic and wetland plants of the southeastern United States. Dicotyledons,
diversity of Georgia Press, Athens, GA.
South Research Corporation (GSRQ. 2000. Final survey report: reevaluation of pondberry in Mississippi. Prepared
^ US. Army Corps of Engineers, Vicksburg District, Gulf South Research Corp, Baton Rouge, LA.
DS. Skojac, B.R. Lockhart, T.D. Leininger, M.S. Devall, and NJVI. Schiff. 2009a. Bottomland forests in the
UJvver Mississippi Alluvial Valley associated with the endangered Undera melissifolia. Castanea 74:105-1 13.
VASCULAR FLORA OF THE OLD MULKEY MEETING HOUSE STATE
HISTORIC SITE, MONROE COUNTY, KENTUCKY
Ralph L. Thompson
Ronald L. Jones
Hancock Biological Station
. Murray State University
• ; Murray, Kentucky 40271, USA
Berea College Herbarium, Dept, of Biology
, Berea College, Berea, Kentucky 40404, USA
- ■ - ralph_thompson@berea.edu
Department of Biological Sciences
Eastern Kentucky University
Richmond, Kentucky 40475, USA
ron.jones@eku.edu
RHSUMEN
^Clwk; plantas vasculares, plantas invasivas, las comunidades de plantas, bosque de roble-nogal americano, bosque mesofftico.
INTRODUCTION
°W Mulkey Meeting House State Historic Site, hereafter Old Mulkey, is the only state park in Kentucky to
the name and commemorate the history of a church. The historic site was designated to preserve
■ r*ld Mulkey Meeting House, the oldest wooden building of its kind in Kentucky and the oldest meet-
,T0Use west of the Appalachians. Old Mulkey is located 0.6 km from the city limits of Tompkinsville
1446 (Old Mulkey Road) in south-central Monroe County, Kentucky, at latitude 36.679722°N
>de 85 705556°W. Monroe County is a southernmost Kentucky county bordering Clay County,
,l^Ssee (Fig- 1). Old Mulkey became a part of the Kentucky State Parks system on November 8, 1931, as
‘ ^ulkey Meeting House State Shrine (Kentucky State Parks 2009).
Mulkey is comprised of 32.0 hectares in the unglaciated Eastern Highland Rim of the Interior
** based Keys et al. (1995) and Woods et al. (2002). Old Mulkey encompasses a continuum of
m ‘orest vegetation, woodlands, and anthropogenically disturbed (semi-natural and modified) vegeta-
Journal of the Botanical Research Institute of Texas4(1)
393
tion. The terrain consists of upland ridges, rolling to steep hillsides, rock outcrops, drainage gullies, toe
slopes, a narrow ravine, and an alluvial terrace. The topography then transitions to near level slopes with
an admixture of forest woodlands bordering a large, open grassy yard. Elevations range from 230 m at the
junction of the Mill Creek floodplain to 274 m on an upland area adjoining KY 1446. The historic site features
the Old Mulkey Meeting House, a Revolutionary War patriot and pioneer cemetery, a Visitor Center, an Old
Visitor Center, a maintenance building, a picnic and playground area, and a large mowed yard primarily of
graminoid herbaceous species.
A descriptive floristic study was undertaken of the Old Mulkey Meeting House State Historic Site dur-
ing the 2008-2009 growing seasons. An extensive herbarium search (BEREA, EKY, KY, WKU) revealed
a paucity of Monroe County plant specimens from a few incidental collections. Our objectives were to: 1)
document the vascular plants with representative voucher specimens, 2) describe the plant communities
present with characteristic species, 3) determine Monroe County distribution records, 4) discuss geology,
soils, and climate of the physical site, 5) prepare an annotated species list with county records, origin (native
or exotic), invasive plant pest status, vernacular name, plant communities, relative abundance, and collection
number(s) for each taxon, and 6) establish a baseline inventory for further botanical study in Monroe County.
THE STUDY SITE
History
Old Mulkey Meeting House represents both the Pioneer Era and the Great Religious Revival Era in the
development of Kentucky history in the early 1800s. Philip Mulkey led a small group of Baptist settlers to
Kentucky in 1773 from what is now North Carolina and South Carolina. The congregation established the
Mill Creek Baptist Church in 1797 at the current site of the Old Mulkey Meeting House. This group was a
part of the Religious Restoration Movement, known as “The First Great Awakening,” that swept the south after
the Revolutionary War partly to declare separation from the established church in England. John Mulkey,
brother to Philip, became the first minister of the church in 1797 (Rogers 1960).
church. It measured 15.24 m x 9.14 m and was built with 12 corners representing the 12 Apostles and three
doors symbolic of the Holy Trinity. It had five windows, a puncheon floor, clapboard shutters, hand-driven
bogles, chinked and daubed walls, and split-log peg-leg benches (Kentucky State Parks 2009).
On November 18, 1809, the congregation of 200 members split into two factions over differences in
John Mulkey’s religious doctrine. After this schism, 150 members remained with John Mulkey and retained
the church building, which then became known as “Old Mulkey” or “Old Mulkey Meeting House.” The 50
remaining members of the congregation left and established a new Baptist church in Tompkinsville (Rog-
ers I960). John Mulkey was a well-respected preacher in the Baptist Association for 53 years and delivered
2° JO, 000 sermons before he died at Old Mulkey on December 13, 1844 (Rogers 1960). After Mulkey’s
“ath> rehgious services continued to be held at Old Mulkey until 1856, after which the existing congrega-
disbanded and its former members moved to the Church of Christ in Tompkinsville. Old Mulkey was
ndoned following the dissolution of the church (Kentucky State Parks 2009).
ln the 1870s, Old Mulkey Meeting House was renovated and was used for various religious meetings
"rd events until 1910, when it was again abandoned. In 1925, the Honorable Joe H. Eagle, U.S. Represen-
Utlve and Monroe County native, and Reverend William Thomas of Tompkinsville initiated a successful
^Paign to rebuild and restore Old Mulkey to its 1804 form (Fig. 2). Through their efforts and others, the
Mulkey Meeting House and cemetery were designated a Kentucky state shrine by the Kentucky State
ar Commission on November 8, 1931 (Kentucky State Parks 2009).
. cemetery at Old Mulkey is the final resting place of 76 Old Mulkey pioneer parishioners, many of
^°m were Revolutionary War soldiers. Among those gravestone markers are Hannah Boone (1746-1828)
& Lieutenar»t Nathan Breed, James Chism, John Giss, Joseph Gist, William Howard, Obadiah How-
ar - philip Mulkey, John Newton Mulkey (son of John Mulkey), Edward Pediford, General Samuel Wilson
is, Flora of Old Mulkey Meeting House State Historic Site
Mississippi shaly siltstones, greenish-gray to light gray weathering to yellowish-gray of the Salem and
Warsaw Limestones are present on gentle side slopes and hillsides to the upper elevations from 256 m to
252 m (Witkind 1971). These calcareous shales are not distinguishable from the shaly siltstone beds of the
underlying Fort Payne Formation and often form rock outcrops. From 252 m to 275 m, Salem and Warsaw
Limestones underlie the upland areas of Old Mulkey to the eastern boundary with KY 1446. Exposed lime-
acme outcrops consist of medium to dark gray, thick-bedded, fine to medium carbonate grains with fossil
fragments and tightly cemented oolites (Witkind 1971).
The soils of the Garmon Association form an almost continuous band from the spring-fed ravine to the
steep 20-60% west-trending hillside slope adjacent to Mill Creek. Garmon Shaly Silt Loams on 12-20%
slopes lie along the ravine slopes and upper hillside slopes near the northern boundary (Mitchell & Latham
1982). These residual soils are medium acidic to neutral in reaction, moderately deep to 100-152 cm deep,
well-drained, and are composed of calcareous shaly limestone bedrock. Lowell Silt Loams of 2-6% slopes,
are found around the old cemetery and picnic area and also lie on small upland benches and ridges on
6-12% slopes south of the Visitor’s Center. These interbedded limestone, shale, and siltstone residual soils
are strongly acid to neutral in reaction, well-drained, gently sloping and are less than 122-152 cm deep to
bedrock (Mitchell & Latham 1982). Waynesboro Loams of 6-12% slopes and 12-20% slopes are present in
much of the uplands around the Visitor’s Center and along the KY 1446 roadside shoulders. These residual
soils are strongly acid or very strongly acid in reaction, well-drained, and 61-203 cm deep on moderately
steep upland slopes. Waynesboro Clay Loams of 6-12% to 12-30% slopes lie on hillside slopes at the
southeastern boundary area. These residual, well-drained soils are strongly acidic to very strongly acidic in
reaction and 96-203 cm deep (Mitchell & Latham 1982).
^ continental climate of south-central Kentucky is temperate, humid mesothermal with little water defi-
aarey, warm to hot summers, and mild to cold winters (Trewartha & Horn 1980). Kentucky climatic data
(1971-2000) are from the weather station at Summer Shade, Metcalfe County, ca. 35 km northwest of the
0W Mulkey site. Mean annual precipitation is 128.0 cm with March the highest at 13.0 cm and October the
est at 8.1 cm, and includes a mean annual snowfall of 27.7 cm with 20.8 cm in January and February.
gh1 annual temperature is 13.8°C with January the coldest month at 1.4°C and July the warmest month
* 4'8°C. Median length of the growing season is 183 days with April 17 the median last spring frost and
^obcr 18 the median first fall frost (Kentucky Climate Center 2009).
20C»,ar PlamS WCre collecte(T from 17 all day field trips during the growing seasons of 2008 (10 trips) and
„ . 7 triPs)- The master voucher specimen set for this floristic study is deposited at the Berea College
Hcrbari
a (BEREA). Vernacular nai
. -1 Chester et al. (2009) or USDA, NRCS (2010). The
rr,s for origin of taxa are from Thompson (2008). Nomenclature of taxa and classification of families is
W k °n^ones (2005), except in the case of a few woody plants where the nomenclature follows Clark and
fordetT^ (2008)' J°nes (2005) and Gleason and Cronquist (1991) were used
in identifies
^ d on Thompson (2007). A single relative abundance or occurrence value (rare, scarce, infrequent,
S(v 10na^’ frequent, abundant) is assigned for each taxon and is inclusive for all plant communities. Inva-
Pest plant status (severe threat, significant threat, lesser threat) is from the Kentucky Exotic Plant Pest
^Hist (KY-EPPC 2009). Relative species richness was derived from the species-area equation (S =
a known ^ ^ Wade and Tfr°mPson (1991). A predicted species richness (S) may be calculated by inserting
wn value in hectares (A) and a deviation value calculated from actual number of species divided by
number of species.
A plant community is defined as an assemblage of associated species with a definite floristic composition
under relatively consistent habitat conditions with a uniform physiognomy (Thilenius 1968). Our floristic
plant communities are based on Thilenius’s concept: the structural physiognomy, species composition of the
characteristic or diagnostic canopy, subcanopy, and herbaceous species, and habitat diversity (combination
of physiographic terrain, soils, geology, and topographic-moisture relief).
RESULTS AND DISCUSSION
Taxonomic Summary
Vascular flora of Old Mulkey consists of 415 species in 266 genera from 105 families (Table 1). In several
cases, an infraspecific category, i.e., subspecies or variety, is given for a species, but in no instance is there
more than one infraspecific category per species. Taxonomic distribution includes 10 species of Polypo-
diophyta, five species of Pinophyta, and 400 species of Magnoliophyta (301 Magnoliopsida, 99 Liliopsida).
Eighty-three (20.00%) are naturalized or introduced exotics (Table 1). Ninety-six (23.13%) were woody
and 319 (76.87%) were herbaceous species. A total of 319 species (76.87%) are Monroe County distribution
records, which is indicative of an inadequately collected county. No rare state-listed taxa were encountered
based on KSNPC (2009). The families containing the most species are the Asteraceae (51), Poaceae (46),
Fabaceae (25), Cyperaceae (24), Rosaceae (17), Scrophulariaceae (11), and Lamiaceae (10). Care* is the larg-
est genus with 21 taxa.
The presence of the recently-described, facultative calciphile, Carex superata, is significant. This rare
sedge is growing on the lower slope of a small mesic ravine in Garmon calcareous shaly silt loam soils ad-
jacent to a spring-fed creek at Old Mulkey. In Kentucky, this taxon is near its northern range limit; it has
only been documented from three counties bordering Tennessee; Cumberland, Monroe, and Todd Counties
Species Richness
Using the species-area curve for floras in the Mixed and Western Mesophytic Forest Regions (Wade &
Thompson 1991), the predicted species richness is 402 species for a 32.0-ha site. The actual species rich
ness at Old Mulkey is 415 species. The relative species richness from the actual species number dividedby
the predicted species gave a deviation value of +1.03%. This species richness value is indicative of an ample
survey for the Old Mulkey study site.
Invasive Species
At Old Mulkey, 43 (51.80%) of the naturalized species are invasive plant pests based on the Kentucky
Exotic Pest Plant Council list (KY-EPPC 2009). Eighteen invasive species are classified as “severe threat
plants. The worst woody invasive. Vinca minor , forms extensive ground cover over approximately 4.0 ha
of the Mesophytic Forest community. Lonicera japonica is found throughout both Mesophytic Forest
Oak-Hickory Forest. Pueraria montana var. lobata is entangled in localized areas of the Oak-Hickory F
Among the “severe threat” species, Ailanthus dtissima , Celastrus orbiculatus, Euonymus alatus , U&istrumsfr
ense, Lonicera maackii, and Rosa multiflora, are growing in a small restricted area of Oak-Hickory Forest^
forest woodland ecotones east of the Visitor’s Center and adjacent to KY 1446. Thirteen “significant th
invasives and 12 “lesser threat” invasives are present within the boundaries of Old Mulkey. The
the invasives are widespread herbaceous species in culturally disturbed habitats, e.g., mowed yar ^
historic site and the KY 1446 roadside shoulders contiguous with the eastern boundary of Old Mulkey
Oak-Hickory Forest. Two introduced exotic trees, Koelreuteria paniculata and Pyrus calleryana, are P'anKU
at the Old Visitor’s Center.
The consequences of not controlling woody naturalized invasives will ultimately continue to i»P*
the native flora, vegetation, and habitats at Old Mulkey Meeting House State Historic Site audits
Invasive plant species have the deleterious effects of displacing and out-competing native species,
the pattern of progressive plant succession, influencing decomposition processes, and disrupting
cycles (Poindexter & Thompson 2009). Among the “severe threat” taxa, Celastrus orbiculatus, Euony**
397
Magnoliophyta
Uliopsida
Magnoliopsida
105
255
201
266
21
62
83
2.41
1.20
23.86
72.53
100.00
Lonicera maackii, and Vinca minor ; have extended their geographical range based on the maps of Clark and
Weckman (2008). All severe threat woody invasive species, except for Lonicera japonica and Vinca minor,
could be controlled because of the small areas of establishment at the current time. Treatments for effectively
removing and destroying aggressive invasive woody pest plants at Old Mulkey are highly recommended
through foliar application, hack and spray, and/or cut/paint stem-stump applications of herbicides (e.g.,
thdopyr with 2.4-D and glyphosate).
nant communities are described with physical features and characteristic species in the predominant Meso-
phytic Forest and Oak-Hickory Forest types. Plant communities are further discussed within these two major
orest community types in context including ecotonal forest woodland border, intermittent creek drainage
area, and culturally disturbed areas. The anthropogenically derived plant community includes the historic
site mowed yard, cemetery, playground, picnic area, and the roadside shoulder and ditch of asphalt-paved
W 1446 (Old Mulkey Road).
Mesophytic Forest— The Mesophytic Forest community encompasses the largest part of the forested Old
ulkey site, and it forms a mosaic throughout with the Oak-Hickory Forest community on upland hillside
f°pes’ midslopes, lower slopes, side slopes, and drainage gullies on the central to western and northern
^ries. Mesophytic Forest continues through a mesic west-facing ravine with a natural spring-fed creek,
•j*ine slopes, and toe slopes, and it ends in an alluvial floodplain terrace near the junction with Mill Creek at
extreme western boundary. The closed-canopy is comprised primarily of very shade tolerant to partially
e tc,lerant trees. Diagnostic canopy trees in general order of abundance are Fagus grandifolia, Acer sac-
, Quercus alba, Q. rubra, Liriodendron tulipifera, Carya cordiformis, Aesculus flava, Fraxinus americana, Q.
enbergii , Morus rubra, Nyssa sylvatica, Ulmus rubra, and Tiiia americana. In species composition, this forest
tnunity is comparable to a combination of the Fagus grandifolia-Acer saccharum-(Liriodendron tulipifera)
est Alliance and the Fagus grandifolia-Quercus rubra- Quercus alba Forest Alliance of the Land Between
Acicfc ^ Nati°nal Recreation Area (NatureServe 2004). Evans (1991) characterizes this forest type as the
ie Mesophytic Forest. Several large mesophytic trees, Acer saccharum, Fagus grandifolia, Liriodendron
•Pi/era, Quercus alba, Q.falcata, Q. muhlenbergii, and Q. rubra, are 0.70 m dbh or greater. Trees measuring
r 10 m dbh are Fagus grandifolia and Quercus alba.
> rk develoPed tail-shrub layer and woody vines are present particularly within the spring-fed ra-
Utx racteristic taxa are Asimina triloba, Bignonia capreolata, Corylus americana. Hydrangea arborescens,
^n<*era benzoin, Menispermum canadense, Sambucus canadensis, Smilax hispida, Staphylea trifolia,
m acertfolium, and Vitis rotundifolia. The most abundant invasive woody plant, Vinca minor, forms
ground cover in several upland and lowland mesic areas.
n t e Mesophytic Forest community, the species richness increases with the presence of several spring
. * species especially within the spring-fed ravine. Indicator perennials include Arisaema triphyl-
Sarum canadense, Carex amphibola, C. cumberlandensis, C. kraliana, C. oligocarpa, C. rosea, Erythronium
anurn’ Galium triflorum. Iris cristata, M aianthemum racemosum, Pachysandra procumbens, Poa autumnalis.
Dip 1
(s, Flora of Old Mulkey Meeting House State Historic Site
Desmodium nudiflorum, Dichanthelium acuminatum, D. boscii, Galium circaezans, G. pilosum, Goodyera pubes-
crn, Qxalis violacea, Phryma leptostachya. Podophyllum peltatum. Polygonum virginianum, Sanicula canadensis,
Scutellaria elliptica, Solidago caesia, Tipularia discolor, and Viola palmata. Ferns include Asplenium platyneuron,
Botrychium dissectum, B. virginianum, Phegopteris hexagonoptera, Polystichum acrostichoides, and Woodsia ob-
A unique stand of Oak-Hickory Forest is found on a steep, west-trending shaly hillside above Mill Creek
« the southwestern boundary. Dominant canopy trees and shrubs include those characteristic of the Oak-
Hkkory Forest community. Some indicator species found only on the thin soil, calcareous shaly outcrops
at the upper and midslope are Carex communis. Coreopsis major, Dodecatheon meadia, Heuchera americana,
Porteranthus stipulatus, Saxifraga virginiensis, and Solidago sphacelata. Other characteristic species include
fyrostis perennans, Arabis laevigata, Asplenium platyneuron, Carex albicans, Houstonia purpurea, Paronychia
canadensis, Solidago caesia, and Sphenopholis nitida. The exposed soils of the lower slope are colonized by
Cystopteris bulbifera, Eurybia macrophylla. Hydrangea arborescens, Lindera benzoin, and Sambucus canadensis.
Ecotonal Woodland Border. — A forest ecotonal woodland border of Oak-Hickory Forest and Mesophytic
Forest surrounds the Culturally Disturbed community of the historic site. The flora consists of an admixture
of serai woody and herbaceous species from the Oak-Hickory Forest, Mesophytic Forest, and the Culturally
Disturbed yard. This secondary successional woodland community has an assemblage of shade-intolerant
to very shade-intolerant woody species. Pirns virginiana, Rhus copallina, Robinia pseudoacacia, Rosa Carolina,
Ruhus argutus, SmUax bona-nox, S. glauca , Toxicodendron radicans, Ulmus alata, and Vacdnium corymbosum,
are present along with the saplings of several intermediate to shade-intolerant species of Carya and Quercus.
Suffrutescent plants include Chimaphila maculata and Hypericum stragulum. Herbaceous species occupying
this community include Agrimonia rostellata, Antennaria plantaginifolia, Carex albicans, Desmodium glabellum,
Dichanthelium commutatum, D. laxiflorum, Elephantopus caroUnianus, Erigeron annuus, E. philadelphicus, Galium
opanne, Hypericum punctatumjuncus tenuis, Luzula echinata, Oxalis violacea, Ruellia caroliniensis, Salvia lyrata,
Scutellaria nervosa, and Viola sororia var. sororia. The omnipresent Lonicerajaponica and Toxicodendron radicans
are interspersed along these ecotonal forest woodland edges among numerous other woody and herbaceous
A small open, xeric woodland border of Oak-Hickory Forest has a herbaceous floristic composition
“f several prairie species descriptive of a prairie-like barren. This mid-seral stage community is found on a
south-trending hillside edge near the spring and southwest of the Old Mulkey Meeting House adjacent to the
*®°wed yard. Characteristic perennial graminoids are Andropogon virginicus var. virginicus, Carex hirsutella,
nthelium commutatum, D. laxiflorum, Elymus glabriflorus, and Tridens Jlavus. Broad-leaved annual and
P^nnial prairie-like forbs include Asclepias tuberosa, Aureolaria virginica, Chamaecrista nictitans, Croton mo-
"“"diogynus, Desmodium rotundifolium, Diodia teres, Erigeron strigosus, Euphorbia corollata, Hieracium gronovii,
#ahi/lora, Lespedeza intermedia, L. procumbens, M anfreda virginica, Packera anonyma, Penstemon canescens,
'hgonemoralis, Spiranthes tuberosa, and Vernonia gigantea.
The secondary succession prairie-like barren community has been severely influenced by continual
”’°Wlng of ad the perennial prairie species during the last two years. The mowing practice has continually
expanding the established mowed yard area into the woodland prairie-like border. This close mowing
j^Ke is effectively destroying the herbaceous prairie species by allowing encroachment of undesirable
utahzed grasses from the yard and an increase in embankment erosion. It is recommended that this
e-like barrens community should not be mowed at all to allow for natural plant succession.
^Culturally Disturbed Community.— Anthropogenic affects on the flora diversity at the Old Mulkey site
occurred numerous times throughout its history. The Culturally Disturbed community includes plants
IJ/'Wng the extensive mowed yard, cemetery, playground and picnic area, asphalt and gravel roadsides,
%qu gr°Und by a maintenance building, and the ruderal KY 1446 roadside shoulder and ditch con-
with the ecotonal woodland border. Anthropogenic influences have created a weedy vegetation of
exotic and native annuals, some biennials, and several perennial herbs of the Asteraceae, Brassicaceae,
iceae, Lamiaceae, Poaceae, ;
Ion, Digitaria sanguinalis, Festuca arundinacea, Pasj
principal grasses in the mowed yard. Naturalizec
Cerastium vulgatum, Draba verna, Lespedeza stipulac
Poa annua , Sherardia arvensis, Stellaria media, Tara
id Polygonaceae, among other families. Cynodon dacty-
ilum pubiflorum, Poa compressa, and P. pratensis are the
annual and perennial herbs include Cardamine hirsute,
x, L. striata, Medicago lupulina, Ornithogalum umbellatum,
lcum officinale. Trifolium dubium, T. repens, and Veronica
include Ambrosia artemisiifolia, Carex cephalophora, C. glaucodea, Claytom
virginica, Dentaria multifida, Euphorbia maculata, E. nutans, Gamochaeta purpurea, Houstonia caeruleajuncus
tenuis, Krigia virginica, Luzula echinata, Oxalis stricta, and Viola sororia var. sororia. The introduced Narcissus
pseudonarcissus has persisted in the cemetery among an intense ground cover of Vinca minor. A few natural-
ized herbaceous plants only found at one ground-disturbed site around a maintenance building near the
Visitor’s Center are Amaranthus spinosus, Anagallis arvensis var. arvensis , Cerastium glomeratum, Chenopodim
album, Fatoua villosa, Lolium perenne var. aristatum, Mentha xpiperita, Secale cereale, Thlaspi alliaceum, Tritirum
The KY 1446 mowed roadside shoulder and ditch are colonized by many of the same ruderal natural-
ized and native taxa present in the culturally disturbed yard. Characteristic forbs include Allium vineak,
Ambrosia artemisiifolia, Bidens bipinnata, Carduus nutans. Chrysanthemum leucanthemum, Cichorium intybus,
Coronilla varia, Daucus carota, Lactuca saligna, Melilotus alba, Sonchus asper, Verbascum blattaria, and V. thap-
sus. Grasses include Andropogon virginicus var. virginicus, Bromus commutatus, Dactylis glomerata, Digitaria
sanguinalis, Eleusine indica, Festuca arundinacea, Paspalum pubiflorum, Setaria pumila. Sorghum halepense, and
Tridensflavus.
ANNOTATED LIST OF SPECIES
The annotated list is arranged alphabetically by family, genus, and species in the Polypodiophyta, Pinophyta,
and Magnoliophyta (Magnoliopsida and Liliopsida). Each taxon entry has a symbol code preceding the scien-
tific name for Monroe County record (o), naturalized exotic species (*), invasive plant species (**), introduced
exotic species (4), and native planted species (t). A vernacular name, plant community type(s), an inclusive
relative abundance value, and representative voucher specimen number(s) ensue for every taxon. After the
vernacular name, a code for plant communities follows: OF (Oak-Hickory Forest), MF (Mesopl
WB (Forest Woodland Ecotone Border), IC (Intermittent Creek Drainage), and CD (Cultura
Community). These communities are listed in sequence of importance for a characteristic species present
11 plant comm
;:R (Rare)— 1-4
", S (Scarce) — 5-10 plants or colonie
lifficult to find in
A relative abundance value for e
nies, very difficult to find in one or t' _ __ H|
few locations; I (Infrequent)— 11-30 plants or colonies, scattered in some locations; O (Occas
plants or colonies, widely scattered in several locations; F (Frequent)— 101-1000 plants or colonies, eas-
ily found in numerous locations; and A (Abundant)— greater than 1000 plants or colonies, diagnostic or
dominant species throughout many locations. An italicized voucher specimens) collection number by
first author in a year-number (e.g., 08-558; 09-757 ) format ends each taxon entry.
POLYPODIOPHYTA
?uron (L) BSP, Ebony spleenwort; OF; O;
Polystichum acrostichoides (Michx.) Schott, Christmas
MF, OF; F; 08-752
» Woodsia obtusa (Spreng.) Torr., Blunt-lobed cliff
R; 08-868
Ophioglossaceae
nCystopteris bulbifera (L) Bernhardi, Bulblet bladder fern; OF;
S; 08-651 S; 09-776
Cystopteris protrusa (Weath.) Blasdell, Southern bladder fern- ^chium virginianum (L) Sw, I
MF; R; 08-738, 09468 ' 08-285
eferrc#
r OF, MR 1
; CD; F;
; MF, OF; 0;
406
oParietaria pensylvanica Muhl. ex Willd, Pennsylvania
MF,0F;0; 08-555
aPumila pumila (L) A. Gray, Canada dearweed; I
a Verbena urtidfolia L, White vervain; IQ I; 08-779
Violaceae
n Hybanthus concolor (T.F. Forst) Spreng., Green violet; MF;
I; 08-668
n Viola palmata L, Palmate-lobed violet; OF, MF; 0; 08-255
d sedge; M
Michx., Muscadine grap
;t grape; MF; 0; 09-230
& Narcissus pseudonarcissus L, Daffodil; CD; O; 09- i 1
) Schott, Green-dragon; I
via, Common blue violet; CD, WB;
l violet; MF; I; 09-2 13
aparthenocissus quinquefolia (L) Planch., Virginia creeper; OF,
a Carex blanda Dewey, Eastern w
08- 350,09-241
Carex cephalophora Muhl. ex Willd, Oval-leaved sedge;®,
m;S, 08-370, 09-231
aCarex communis L.H. Bailey, Fibrous-rooted sedge; OF; 0;
09- 239
Carex cumberlandensis Naczi, Krai, & Bryson, Cumberland
sedge; MF, OF; F; 08-375, 09-237
aCarex digitalis Willd. var. digitalis, Slender woodland sedge
OF; I; 08-365
aCarex frankii Kunth, Frank's sedge; 1C; Ft; 08-662
aCarex giaucodea Tuck., Blue sedge; OF, WB; 0; 08-450, 09-
d Carex granularis Muhl. ex Willd., Limestone meadow sedge
MF; S; 08-457, 09-234
Carex hirsutella Mack., Fuzzy sedge; OF, WB; 0; 08-376, 09-
232
Carex kraliana Naczi & Bryson, Krai's sedge: MF; 0; 08-391
aCarex laxiflora Lam. var. laxiflora, Broad loose-flowered sedge
OF, MF; I; 09-238
aCarex lurida Wahlenb., Yellow-green sedge; 1C; R; 08437
aCarex oligocarpa Schkuhr ex Willd., Richwoods sedge; MF,
I; 08-358 • — ^
Carex planispicata Naczi, Flat-spiked sedge; MF, OF; Q »
Carex rosea Schkuhr ex Willd., Rosy sedge; OF, MF;g 08-446,
09-236
Carex superata Naczi, Reznicek, & Ford, Willdenow’s sedge
MF;R; 08-378
aCarex texensis (Torn ex Bailey) L.H. Bailey, Texas sedge; Mr,
S; 09-242 VJ§8|
Carex willdenowii Schkur. ex Willd., Willdenow's sedge;
aCypems^hinatus (L) Alph. Wood, Globe flatsedge;®*
08-739
Cyperus strigosus L, Straw-colored flatsedge; 10 S-,0&77t
aKyllinga pumila Michx., Low spikesedge; CD; R; 08-89/
a**Dioscorea polystachyaTurcz., Chinese
Wild yam; MF, OF; O; 09-228
Bethlehem; CK®
atic dayflower; IQ O; 08-926 iridaceae
rarf-crested iris; Mf; ft
^row-leavedblue^
Cyperaceae
Carex albicans Willd. ex Spreng, var. albicans, White-tinged
sedge; OF, WB; I; 08-357, 09-69
a Carex albursina Sheld, White bear sedge; MF; S; 09-240
Carex amphibola Steud. var. amphibola, Eastern narrow-leaved
sedge; MF, OF; F; 08-355
d&Wiegand.Srft"*
aJuncus tenuis Willd., Path rush; CD; F; 08-610
aLuzula echinata (Small) FJ. Herm., Hedgehog w
FW, CD; F; 08-13, 08-271
, for the Fig.
BOOK REVIEW
PLANT COMMUNITIES OF SELECTED EMBAYMENTS
ALONG THE MID- TO MID-UPPER OHIO RIVER FLOODPLAIN
Joseph S. Ely1
Department of Biology
University of Central Missouri
Warrensburg, Missouri 64093, U.5A
ely@ucmo.edu
Marshall University
Huntington, West Virginia 25701, L
evans@marshall.edu
ABSTRACT
RESUMEN
INTRODUCTION
The importance of wetlands to society has been recently, within the last 30 years, brought to the attention
of policy makers (Hyberg & Riley 2009; Junk &r Wantzen 2004; Mitsch et al. 2000). In 1974, the U.S. Fish
and Wildlife Service (FWS) established the National Wetlands Inventory (NWI) Program (Tiner 2009). The
charge of this program was to conduct a nationwide inventory of wetlands in order to provide FWS biologists
with information on the distribution of wetlands for wetland conservation efforts (Tiner 2009). In 1986, the
Federal Government mandated, through the Emergency Wetlands Resources Act, that the FWS complete
mapping and digitizing of wetlands data for the Nation, and to distribute and archive the data (Dahl et al.
2009). The NWI program developed a wetland classification system (Cowardin et al. 1979) which is the cur-
rent federal standard of the USFWS and the Federal Geographic Data Committee (Tiner 2009). The original
purpose of the National Wetlands Inventory program was to map, on a large scale, our nations’ wetlands.
The current products of the program are a wetlands geospatial database that can be used to generate maps
and statistics about the status of the Nation’s wetlands and the other products include national wetlands
status and trends reports (Tiner 2009).
Overall the NWI program has been an incredible success but there are limitations. According to the
2009 NWI Status Report, the most frequent complaint about the program is that the data are too old for
many applications and that the NWI data should be updated more frequently (Tiner 2009). Further, there
is the issue of incomplete data, only 61% of the nation’s wetlands have been mapped and entire wetland
areas and wetland margins were or are not detected with the remote sensing that took place in the 1970s
and 1980s (Cowardin & Golet 1995).
The regulation of rivers and natural water-ways for flood control, agriculture and urban development,
and river navigation affects the hydrological connectivity or flood pulse of the river with floodplain and
riparian wetland ecosystems (King et al. 2009; Reid and Brooks 2000). Specifically, the flood pulse concept
(Junk et al. 1989) states that the dynamic nature of the hydro-geomorphological condition is the primary
force that is responsible for the existence, productivity, and interactions of biota in river-floodplain systems
(Junk et al. 1989; Junk 1999; Junk & Wantzen 2004).
The Ohio River is formed by the confluence of the Allegheny and Monongahela Rivers at Pittsburg ,
Pennsylvania. Navigation on these rivers has evolved from the use of bark and dugout canoes of the Native
Americans to mass river transportation by barges. In its original condition, navigation along the Ohio Rivtf
was difficult and hazardous due to natural fluctuations of the river throughout the year (US Army
Engineers 1979). The first river modernization project, started in 1955, replaced nineteen moveable locks
and dams. The purpose of the modernization project was to convert river navigation from a local commeitt
to an ‘interstate highway’ (Johnson 1992). The hydrology of the river has changed from an open c^n
to a series of ‘reservoirs’ between dams. The water depth increased profoundly in low-lying areas a ja
to the river. These changes in the river channel, flow and navigation water pool elevation resulte J"
loss of many wetlands but it also created wetlands or embayments. Originally these areas were peri
inundated throughout the growing season but after the modernization projects they were pe
flooded. Embayments are most often formed along permanent or intermittent stream channels. .
Although the U. S. Fish and Wildlife Service (2002) identified approximately 2225 ha of undi*^
embayments within its’ planning area of the Ohio River Island Wildlife Refuge and the agency reC°^
that embayments are significant habitats for fish, mussels, amphibians and reptiles, waterfowl an ®
mals. Embayment plant communities have not been inventoried except for a few isolated studies per o
through agents of the U.S. Army Corps of Engineers. Koryak (1978) examined a recently formed embay
after the completion of the Hannibal Locks and Dam. The potential for embayment aquatic and eme^
plant community development is high and there is strong evidence to indicate that embayments
Ely and Evans, Plant communities and associations of selected embayments
413
navigational pools support productive plant communities with high biological diversity (Koryak 1978). There
is little known about embayment plant communities along the Ohio River floodplain other than large scale
wetland mapping by the National Wetlands Inventory Program. The purpose of this study was to survey and
identify embayment plant communities, assess potential gradients, and to provide a local and regional charac-
terization of embayment that typically occur along the mid to mid-upper Ohio River floodplain. River.
METHODS AND MATERIALS
Study Site Description
Potential study sites were selected from U.S.G.S. and U.S. Army Corps of Engineers navigational topo-
graphical maps (U.S. Army Corps of Engineers 1966, 1978), Ohio River Navigational Charts (U.S. Army
Corps of Engineers 1989), and color aerial photographs (obtained from the U.S. Army Corps of Engineers,
Huntington, WV). These resources were used to assess wetland size, type, associated habitat, disturbance
levels, and ease of access (Ely 1993). Each study site was delineated from the surrounding uplands based
on hydrological conditions, vegetation, soils, and slope.
The study sites (embayments) are located in the Greenup and Gallipolis or R.C. Byrd navigational pools
(Fig. 1) between river km 434 and 549. The 115 km stretch of the Ohio River occurs between Gallipolis
Ohio and Greenup Kentucky. The original Gallipolis Locks and Dam was a part of the 46 movable lock and
dam series and maintained a maximum navigational water depth of 2.75 m (Shows & Wooley 1989). This
was replaced with non-removable roller-gates that increased navigational pool elevation by 5.8 m (Johnson
1992; Jones 1914). Sites selected in this navigation pool were at least fifty-three years old in 1991 and include
Chickamauga Creek (river kilometer 434, 8.3 ha), Crab Creek (river kilometer 444, 2.1 ha) and Teen’s Run
(river kilometer 449, 4.0 ha).
The current Greenup Locks and Dam replaced four of the original 46 dams and it was operational in
1962 (Johnson 1992). This increased the navigational depth more than six meters (Jones 1914; US Army
Corps of Engineers 1979). Sites located in the Greenup navigational pool include Franklin’s or Chandler’s
Run (river kilometer 549, 3.2 ha), and Ginat’s Run (river kilometer 546, 9.3 ha), (Fig. 1) and these sites were
at least twenty-nine years old at the time of the study.
The study area lies within the mixed mesophytic forest of the Allegheny Plateau (Braun 1950) and are
1°cated within the Western Allegheny Plateau of the Eastern Temperate Forest Ecoregions of North America
(Alan et al. 2007a, 2007b). Three of the five study sites are located within the Monongahela Transition
Zone and the other two sites occur in the Ohio-Kentucky Carboniferous Plateaus with elevations ranging
from 157 to 165 m above sea level. Mixed mesophytic and mixed oak forests were the original forest of the
Western Allegheny Plateau (Alan et al. 2007a, 2007b). Today, urban, residential, and agriculture develop-
mem dominate the floodplains. The average January and June temperatures are 1.6°C and 24°C, respectively
(Alan et al. 2007a, 2007b). Braun (1950) described the bottomland hardwood forest of the mixed mesophytic
region as an Acer saccharinum L. (silver maple) and Populus deltoides Bartram ex Marsh, (eastern cottonwood)
community with contributions of Salix spp., Platanus occidentals L. (sycamore), Liquidambar styraciflua L.
feweetgumX and Betula nigra L. (river birch). It is believed that these five study sites represent the features of
typical embayment for this region of the Ohio River floodplain.
s»mpling
Payment site was visited monthly during summer and fall of 1990 and 1991. The vegetation at each
embayment site was divided into three layers or strata: trees (overstory), sapling and shrub (understory),
*md herbaceous (groundcover). Submerged aquatic plants were sampled with the herbaceous cover because
1 ey often integrate. In addition, trees, woody saplings and shrubs, and ground cover species were treated
^ separate quantitative studies. Sampling plots in the tree and shrub and sapling layers were randomly
ed while a stratified random sampling of herbaceous plots was used (see below). Original identification
Prifcedures followed Gleason and Cronquist (1991), Strausbaugh and Core (1978), Cusick and Siblerhom
Journal of the Botanical Research Institute of Texas 4(1)
415
(1977), Britton and Brown (1970), but the current nomenclature follows the Flora of North America North
of Mexico (Flora of North America Editorial Committee 1993+). Voucher plant specimens were collected
and are deposited in the Marshall University Herbarium, Huntington, WV (MUHW; Ely 1993).
Trees in this study were defined as woody plants that were equal to or greater than 8.0 cm in diameter at
breast height (1.4 m) or DBH. Trees were identified to species and DBH recorded to the nearest tenth of a centi-
meter. Saplings and shrubs were defined as woody plants less than 8.0 cm DBH but equal to or greater than 1.0
m tall. Their height was visually estimated to the nearest meter and recorded.
Herbaceous species were composed of herbaceous plants, woody plant seedlings less than 1.0 m tall, and
species of woody vines. Woody plants that exceeded 1.0 m in height but were too dense and/or prostrate to
record in the usual manner were also included. Such species included in this category were Rosapalustris Marsh,
(swamp rose) and Cornus amomum Mill, (silky dogwood).
Herbaceous plot sampling occurred along two to five transect lines at each study site. The number and
placement of transect lines depended on embayment size. For example, the larger sites were evaluated with four
to five 100 m transect lines whereas smaller sites were evaluated with two. In general, transect lines reached
from the edge of the wetland to open water. Plots measuring 100 m2 were placed along each transect line. Eight
1 m2 quadrats were randomly selected within each 100 m2 plot. For each species, the percent cover values were
visually estimated and recorded in each quadrat. Species area curves were established and used as a guide to
obtain an adequate number of plots.
Data Analysis
Overstory, understory, and herbaceous communities were assessed through the calculation of relative
importance values. The overstory DBH values were converted to basal area (BA) for each species. Density,
frequency, BA (overstory), estimated height (understory) values were used in determining relative impor-
tance values. Relative importance values for herbaceous plant species was derived from relative frequency
of occurrence and percent estimated cover values for each taxon.
Herbaceous species percent cover values were converted to a modified Daubenmire cover class scale
(Daubenmire 1959). The midpoint values of each cover class were used in subsequent analyses. The seven
cover classes were: 1: 0 to < 1%, 2: 1 to 4%, 3: 5 to 24%, 4: 25 to 49%, 5: 50 to 74%, 6: 75 to 94% and 7: 95
to 100%.
Herbaceous species diversity was evaluated using species richness, evenness (Pielou 1966) and Shannon-
w«ner (Shannon 1948) indices. Index values were calculated based on a quadrat by quadrat basis and means
and standard errors were calculated for each study site. Coefficient of Conservatism (Wilhelm & Ladd 1988)
and wetland indicator status probability values (Reed 1988) were determined for each native species. Because
the Coefficient of Conservatism values for the same species differed significantly (P < 0.05) between the states of
Ohio and West Virginia, average values among the two states were used in order to make the data comparable.
Non-native species were also identified among embayment sites. The indices, WI, CC, cover values, non-native
cover values were used to assess the uniqueness and similarities of each site and to identify potential gradients.
n ^chdon, herbaceous species and species composition among study sites were compared using Multi-Response
Permutation Procedures (MRPP) and indicator species were calculated using Indicator Species Analysis (ISA).
Herbaceous species cover data were classified into groups using agglomerative hierarchical cluster analysis
anc* potential gradients were identified with Non-metric Multidimensional Scaling, NMS, an indirect ordina-
ls*1 procedure. In order to define groups, we calculated a Cluster Analysis with Sorensen’s distance measure
the rCnSen 1948^ baSCd °n tbe midPoints of the Duabenmire cover class data using flexible beta with = -0.25 in
' Program PC-ORD (McCune & Mefford 1999). Although Soreseris distance measure has been under scrutiny
gendre & Legendre 1998) for its’ use with community data analysis, our justification for use is t hat it resulted
the east amount of chaining among sample plots with cluster analysis and the lowest stress with the NMS
mation (see below). Chaining is the phenomenon of the addition of single sample quadrats to existing
P^oups that result in undefined clusters. Indicator species for each of the cluster analysis defined groups were
identified using Indicator Species Analysis, ISA (see below). The wetland probability indicator status (Reed
1988), WI, was determined for each native species belonging to each group or community as defined by
cluster analysis. The categories include Obligate+, Obligate, Obligate-, Facultative Wet+, Facultative Wet,
Facultative Wet-, Facultative+ Facultative, Facultative-, Facultative Upland+, Facultative Upland, Facultative
Upland- or Upland. In addition, habitat (aquatic, mudflat, or emergent) and dominant species height (tall,
medium or short) were determined for each group.
An ISA was used to determine the statistically significant indicator species (as defined by indicator value
or IV) in each cluster grouping. This procedure combines information on abundance and frequency data
to determine the ‘faithfullness’ of a particular species to a particular group or cluster. Faithfulness can be
defined as the constancy of presence or how likely a specific species will be found in a cluster group (Mc-
Cune & Grace 2002). In order to explore the relationship of herbaceous species and sample data to potential
environmental gradients we use Non-metric multi-dimensional scaling (NMS). Although direct measurement
of environmental variables was not a part of this study, we used an indirect approach using Species Richness,
Shannon-Wiener Diversity (H’), embayment age and river location (expressed in river kilometer), species cover
(total and non-native), wetland indicator status probability values (Reed 1988) and Coefficient of Conservatism
values (Wilhelm & Ladd 1988; Rentch Sr Anderson 2006; Andreas et al. 2004). Strengths of association between
these values and the NMS axes scores were calculated using Kendall’s Tau, t, values (Zar 2010). Joint biplots
All NMS analyses were conducted using Sorensen’s distance measure. Monte-Carlo tests were used to
determine statistical significance of each NMS solution. The percent of variation represented and the coef-
ficient of determination for each axis, the final stress value, and associated P-values are reported.
Cluster analysis, NMS, MRPP and ISA were performed using PC-ORD version 4.41 (McCune Sr Mefford
1999). An alpha (a) or significance level of 0.05 was used for NMS and ISA. A Bonferroni correction to the
o-level (Sokal & Rohlf 1995) was made for the MRPP based on the number of embayment sites.
RESULTS
Species Composition
A total of 257 vascular plants species was encountered among study sites during the course of the study
period; there were 3 fern and fern allies, 84 Monocots and 170 Dicots representing 77 families and 201
genera. The largest number of species occurred in the Poaceae (28, 10.85%), Cyperaceae (27, 10.47%), and
Asteraceae (25, 9.69%). Polygonaceae (11, 4.26%) and Fabaceae (9, 3.49%) had a limited contribution to
species numbers. The number of non-native species across sites was 40 (16.72%). The average Coefficieni
of Conservatism value was 3.51 (± 0.02).
The number of species encountered within and outside of sampling plots at each embayment site ranged
from a low of 119 (Teen’s Run) to a high of 157 (Ginat’s Run). Ginat’s and Teen’s Run had the largest number
woody species encountered in the tree (11 each) and sapling and shrub communities species (n = 13
14, respectively) during this study (Table 3). _ .
Herbaceous species richness and diversity was significantly higher at Chickamauga Creek (X - 3.35*0.
than other sites (Fig. 2i, ii). There were no differences in Species Richness among remaining embayment*
Ginat’s Run had a significantly higher diversity value (X = 1.10±0.06) than Teen’s Run (X = 1.00±0.04),
there were no differences among the remaining embayments (Fig. 2ii). Although there were no differs**
in total cover between Chickamauga (X = 193.88±9.47) and Crab Creek (X = 176.94*9.09), both sites
higher values than the remaining study sites (Fig. 2iii). Coefficient of Conservatism was lowest at _
mauga Creek (X = 3.35*0.09) while Crab Creek (X = 3.59*0.04) and Chandler’s Run (X = 3-56±0^ ,*
the highest (Fig. 2iv). The mean wetland probability values among sites were highest at Crab Cre*
95.43*0.62) and Chandler’s Run (X = 94.04*0.73) (Fig. 2v) while Chickamauga Creek (X = 84.49*1 .lo)
the lowest. Both Chandler’s (X « 19.42*3.99) and Teen’s Run (X = 18.13*3.28) had the highest non-
species cover (Fig. 3vi) and Ginat’s Run (X * 0.46*0.20) had the lowest non-native.
417
419
420
Local Community Composition and Structure
Tree, Sapling and Shrub and Herbaceous Communities. — The tree community at the site level reflected
similarities in species composition and importance that was observed at the regional scale with subtle differ-
ences. Although all sites were dominated by Acer saccharinum, Platams occidentalis was a subdominant at Crab
Creek and both R occidentalis and Fraxinus pennsylvanicum were also important at Teen’s Run (Table 1).
The sapling and shrub communities varied more than the tree community among sites. Acer saccharinum
was an important sapling and shrub species among all sites except for Ginat’s Run. Alnus serrulata dominated
the community at Ginat’s Run and was a sub-dominant at Chandler’s Run. Fraxinus pennsylvanicum domi-
nated the sapling and shrub community at Crab Creek and was a subdominant at Chickamauga Creek (Table
2). Cornus amomum was dominant and subdominant at Teen’s Run and Chickamauga Creek, respectively.
Lindera benzoin was an important understory tree at Crab Creek. Amorpha fruiticosa L. (false indigo) was of
minor importance at Chandler’s Run (Table 2). Five communities were identified among embayment sites.
Similar to the tree community, Ginat’s and Teen’s Run’s the sapling and shrub community had the highest
species represented among all sites, 13 and 14, respectively.
The MRPP for the testing of differences in herbaceous species cover and sample data were moderately to
well defined among embayment sites (T * -83.12, 8 * of 0.39, A = 0.21, P < 0.01). The herbaceous plant com-
munity consists of Sparganium eurycarpum, Leersia oryzoides, Sagittaria latifolia, Persicaria sagittata, and Persicaria
punctata (Table 3). The floating aquatic community would be characterized by Lemna minor / Spirodela potynky
while the submerged aquatic layer consisted of Myriophyllum spicatum and Najas minor that was interspersed
with Elodea nuttalli. Relative importance and indicator values indicated that the herbaceous community was the
most variable community at the site level. The emergent marsh community at Chickamauga Creek was charac-
terized by Persicaria punctata, Sagittaria latifolia, Commelina virginica L. (dayflower), Carex tribuloides Wahlenb.,
Leersia oryzoides, and Nelumbo nucifera Gaertner, Crab Creek emergent herbaceous community was dominated
by Sparganium eurycarpum, Leersia oryzoides, and Sagittaria latifolia. One floating and emergent plant community
consisting of Lemna minor L. (lesser duckweed) / Spirodela polyrrhiza was found in an old oxbow channel at Crab
Creek. Overall the herbaceous layer at Chandler’s Run was characterized by a Persicaria sagittata, Leersiaoryztifa
Sparganium eurycarpum. Hibiscus moscheutos L. (crimsoneyed rosemallow), Persicaria punctata, and Sagittaria
latifolia emergent community (Table 3). Ginat’s Run’s emergent marsh consisted of Leersia oryzoides, Sagittaria
latifolia, Sparganium eurycarpum, Persicaria sagittata, Onoclea sensibilis L. (sensitive fern), and Juncus ejfusus L
(soft rush). Sparganium eurycarpum, Sagittaria latifolia, and Iris pseudacorus L. (yellow flag) characterized Teens
Run herbaceous community.
Classification of Herbaceous Plant Communities
The results of the cluster analysis and indicator species analysis (Fig. 4) show twelve defined groups with
22% of the information (Wishart 1969) remaining at that level. The result of the Indicator Species Analysis
(ISA) was significant (P < 0.05). ISA indicator species values or IV are reported along with the Relative Im-
portance Values (RIV) for each defined group or community (Table 4). The wetland indicator status, height,
embayment site, and number of quadrats, n, for each community are also included.
There were two obligate+ aquatic communities consisting of a submerged community and a
emergent community. The submerged community was dominated by Najas minor and Myriophyllum sPlca\^
and these species were indicators (Table 4, Fig. 4-community G). This community was only
Chandler’s Run (n = 8). The short emergent community (Table 4, Fig. 4, community H) with a °ro
and indicator species of Nelumbo lutea was interspersed with the submerged aquatic Myriophyllum sp**»
Similar to the former community, this community was observed only at one embayment, Ginat’s Run
6)-
There were three obligate- wetland communities that consisted of one medium and two tall eme$P
communities. The medium emergent community was dominated by Persicaria punctata and P.
(Michx.) Small (swamp smartweed) and these species were the top indicators for that community J
Fig. 4, community A). This community was distributed at Chandler’s Run (n = 13), Chickamauga Cree
By and Evans, Plant communities and associations of selected embayments
Journal of the Botanical Research Institute of Texas 4(1)
Ely and Evans, Plant communities and associations of selected embayments
values in bold represent species that were faithful to a particular community and RIV bold values are those species that contributed to 5% or more to that community type.
Boehmeria cylindrica
052
Cotnusamomum
Cuscutagronovii
Cyperusodoratus
Etymus villosus
Galium tinctorium
Itersia oryzoides
Undemiadubia
lysimachia nummularia
Myriophyllum spicatum 68.0 0.29
Hajas minor 10o.o 0.52
0.0
0.81 058 1.36 4.81
0.02 0.00 035 0.08
033 038 1.71 4.89
1855
237
20.92
9) and Crab Creek (n - 1). The tall obligate- emergent community was characterized by Hibiscus moscheW
and Persicaria sagittata were the indicator and dominant species with subdominants of Leersia oryzoides i n
Sparganium eurycarpum (Table 4, Fig. 4, community B). This community occurred at Chandler’s Run G» ■
43), Ginat’s Run (n - 14) and Crab Creek (n = 2). The other obligate- tall emergent community (Table 4, ng
4, community E) was characterized by a non-native species, Iris pseudacorus , and a shrub or understoiy t
species, Comus amomum with limited contributions of Sagittaria latifolia. This community was distri ut
primarily at Teen’s Run (n « 18) with one sample found at Chandler’s Run. ^
There were three obligate emergent communities defined among embayments. Ludwigia palustris was
dominant and indicator species for a short or mudflat emergent community (Table 4, Fig. 4, community
with limited contributions of Leersia oryzoides and minor contributions of Persicaria punctata was de®1
Chickamauga Creek (n = 2) and Crab Creek (n = 10). The short emergent community (Table 4, Fi& 4.
munity K) was dominated by Sparganium eurycarpum and its subdominant of Sagittaria latifolia with l«n^
contribution of Persicaria punctata. Both of the former two species were indicators. Although Teen (
the most number of samples (n - 61), this community was also distributed among three other em ayu»“
sites and include Chandler’s Run (n = 10), Crab Creek (n = 20), Ginat’s Run (n = 11). The medium 0&W
emergent community (Table 4, Fig. 4, community L) was dominated by the only indicator species,
oryzoides, with limited contributions of Sparganium eurycarpum, Sagittaria latifolia, and Persicaria
This community was primarily at Ginat’s Run (n = 59) but was found among all sites. Although C an
427
Run and Crab Creek had nearly equal numbers of samples (20 and 21, respectively), both Chickamauga
Creek and Teen’s Run sample numbers were small (1 and 4, respectively).
Four Facultative Wet+ communities were comprised of an equal number of tall and medium height
emergent species. Two indicator species (Carex tribuloides and Commelina virginica) shared dominance at one
of the medium emergent community (Table 4, Fig. 4, community D) and was distributed among Chickamauga
Creek (n « 31), Ginat’s Run (n = 6) and Teen’s Run (n » 2). Onoclea sensibilis was the dominant and indica-
tor species at a tall facultative wet+ emergent community (Table 4, Fig. 4, community C), Carex stipata and
Typha latifolia L. (broad-leaved cat-tail) were also indicators. The latter community was found at Crab Creek
(n ■ 10) and Chickamauga Creek (n = 2). Leersia oryzoides, Nelumbo nucifera, and lmpatiens capensis were co-
dominants and indicators at a tall facultative wet+ emergent community (Table 4, Fig. 4, community F) that
was distributed among four sites: Chandler’s Run (n = 9), Chickamauga Creek (n = 22), Ginat’s Run (n « 13)
and Teen’s Run (n = 14). The other tall emergent community (Table 4, Fig. 4, community J) was dominated
byjuncus effusus with limited a contribution of Schoenoplectus tabemaemontani (C.C. Gmelin) Palla (soft-
stem bulrush). Both species along with Justicia americana were indicator species but the latter species was
of minor importance to the overall community and occurred primarily at Teen’s Run. This community (J)
was characteristic of Ginat’s Run (n - 17), but was also found at Teen’s Run (n = 4), Chandler’s Run (n « 1)
and Crab Creek (n = 1). Non-metric Multi-dimensional Scaling (see below) show that the ordination procedure
separated the cluster analysis defined herbaceous plant communities (Fig. 4B i, ii, iii) among three axes.
Herbaceous Community Composition and Potential Gradients
Non-metric multi-dimensional scaling (NMS) was used to explore the relationship of herbaceous species
and samples among embayment sites and cluster analysis defined groups for each site. The NMS resulted
in three axes or dimensions with the number of runs for the Monte-Carlo simulations were 150. The final
stress was 11.6. Ecological community data often have final stress values between 10 and 20%. Values in the
lower half of this range represent reliable solutions (McCune & Grace 2002). The total proportion of variance
represented by NMS was 84.6%. The first axis accounted for majority of the variation (64.0%), the second axis
accounted for 13.6% and the third accounted for 7.1%. The NMS ordination axis 1 (Fig. 5i) separated Chickamauga
Creek, Crab Creek and Ginat’s Run from Chandler’s Run and Teen’s Run. The first axis was positively associated
with embayment age (r = 0.58), non-native species (r = 0.42) and negatively associated with river kilometer (t
* -0.55) and a weak negative association with Coefficient of Conservatism (CO (t - -0.22), richness (x * -0.09)
and diversity (t = -0.11). The MRPP for the testing of differences in herbaceous species cover and sample data
were weakly defined among age of embayment sites (T . -60.18, 6 = 0.47, A * 0.21, P < 0.01). The second axis
separated Chandler’s Run and Teen’s Run from the other sites (Fig. 5ii). This axis was positively associated with
Wetland Indicator Status Values (t = 0.45). Richness (x = -0.41) and diversity (x = -0.36) had a negative association
with the second axis. The third axis separated all sites except for Chickamauga Creek and Crab Creek (Fig. 5iii)
and was positively associated with embayment site (x = 0.64), embayment size (x = 0.27) and river kilometer (x
* 015). Species cover (x = -0.18), richness (x = 0.16), diversity (x = -0.12), CC (x = -0.12) and Wetland Indicator
Status Value (t = -0.09) were negatively associated with this axis. The MRPP for the testing of differences in
herbaceous species cover and sample data were moderately defined according to river kilometer (site) among
embayment sites (T = -79.83, 8 = of 0.41, A = 0.18, P < 0.01).
DISCUSSION
Floodplains have been tradition ally ‘hotspots’ of community and species diversity (Gopal & Junk 2000) due
^ characteristics of the flood pulse processes (Junk & Wantzen 2004; Junk 1997; Middleton 2002). In
ition, local species richness and diversity in small wetland areas are typically high (Flinn et al. 2008)
ause of the diversity of habitats along hydrological gradients that result in defined niches (Silvertown
« al. 1999. Silvertown et al. 2001). There are limitations of the current study and include the inferred
environmental gradients from floristic quality index values (Wetland Indicator Status and Coefficient of
Conservatism), limited distribution of study sites along the Ohio River, and pseudoreplication (Hurlbert
428
Journal of the Botanical Research Institute of Texas 4(1)
NMS Axis 2
1984). However, the authors feel that this study provides baseline data, plant community classification,
potential gradients, and species information for future work along the Ohio River and other major navigable
rivers where embayments may have been formed. We document species composition, structure, diversity,
plant communities and potential gradients of embayment wetlands that occur along the mid- to mid-upper
Ohio River floodplain.
There were 257 species of vascular plants recorded from approximately 27 ha of embayment habitat.
Our results are in accord with other wetland studies of the region. For example, Stark (1993) reported 335
species from Green Bottom Wildlife Management Area while West and Evans (1982) recorded 252 vascular
plant species along the Kanawha River. Both of the latter studies encompassed a combined study area of
approximately 400 ha. There were 360 species recorded in a classical study of the floodplain vegetation
and environment along the Raritan River, NJ, but encompassed the entire floodplain from terrace to ter-
race (Wistendahl 1958). In another example, a total of 629 vascular plants were recorded along a 660 mile
the Wabash and Tippecanoe Rivers by Lindsey et al. (1961).
Sixty-four of the 257 species observed (34.6%) occurred at all embayment sites. However, these species
are common and widely distributed elements of the region. The remaining 192 species have a more limited
distribution with 36 (5.3%) at four sites, and 23 (9.0%) at three sites. Of those species most restricted in
distribution, forty-five taxa (17.4%) occurred at two sites, and the largest number of taxa, 90 (34.6%), were
limited to a single site.
Embayment sites capture approximately 25% of the species that are common elements of the floodplain.
However, more than half the number of species encountered across the sample area was restricted to one
or two sites. This suggests that these wetlands have common elements at large broad scales with distinct
elements at small scales.
The woody species that defined the embayment bottomland hardwood forests were common elements
of the Ohio River riparian zone. The Acer saccharinum, Fraxinus pennsylvanicum, Platanus occidentalism and
Salix nm community dominated the forests and our results (Table 1) are consistent with other riparian
studies along the Ohio River floodplain and other large rivers (Clagg & Mills 1978; Evans 1980, 1977a,
1977b; Furry & Evans 1979- Koryak 1978; Liu 1991; Stark 1993). The forests encountered in this study would
be considered an Acer saccharinum forest alliance as defined by The Nature conservancy (Weakley et al. 1996).
Gmat’s and Teen’s Run embayment sites had the greatest tree species diversity among all sites and is probably
due to the degree of disturbance or lack of it. Ginat’s Run forest is secluded from surrounding agriculture
fields and development and has had minimal disturbance while Teen’s Run site has had significant amount
of disturbance in the past with numerous homesteads and agriculture. In addition, Ginat’s Run’s forest were
more even-aged than the remaining embayment sites that were characterized by high recruitment of Acer
wcharimm as indicated by the density of trees in the smallest size class (Ely 1993).
Although Acer saccharinum was the dominant tree species, it had a limited distribution among the
shrub and sapling layer (Table 2). Alnus serrulata was the dominant shrub at Ginat’s Run with very limited
contributions of other species including Acer saccharinum. Cornus amomum and Salix nigra were co-dominants
w'(h Acer saccharinum at Teen’s Run. Acer saccharinum along with Lindera benzoin were sub-dominants to
Fro*mus pennsylvanicum at Crab Creek. The shrub alliances (Weakley et al. 1996) among embayments were
more ^erse than the tree layer. Chickamauga Creek’s shrub community would be considered a Cephalanthus
Kadentahs shrub alliance while Ginat’s Run would be an Alnus serrulata alliance. Chandler’s Run shrub layer
would be mix of Alnus serrulata and Salix nigra and the shrub community at Crab Creek would be considered
a Undera benzoin mixed alliance with Betula nigra and Asimina tribba (L.) Dunal (pawpaw). Finally, Teen’s Run
a iance is characterized by Cornus amomum, a Midwestern element.
At the local scale (embayment site) herbaceous species composition, richness and diversity show that
1 re were differences among sites (Fig. 3i, ii). In addition, sites with significantly higher species richness
iTld diversity tended to have the lowest wetland indicator and coefficient of conservatism values (Fig. 3i,
430
ii, iv, v). Two embayment sites have non-native species cover values approaching 20% while the remaining
sites are below 10% (Fig. 3vi). Chickamauga Creek and Teen’s Run embayment sites would be considered the
most disturbed of the five sites. Chickamauga Creek site is surrounded by the city of Gallipolis, Ohio, and
adds to the level of disturbance along the edge of that site. Teen’s Run while in a remote location consists of
former farmsteads and agricultural fields. There were two non-native species that were important components
of several communities at Chickamauga Creek and Teen’s Run, Nelumbo nucifera and Iris pseudacorus L. (yellow
flag), respectively, both species were indicators and important component of the plant communities (Tables 3
and 4; Fig. 4) Nelumbo nucifera was previously reported by M. L. Roberts in 1973 (Cusick & Silberhom 1977),
Another non-native species, Amaranthus cruentus L. (purple amaranth), was also recorded for the second
time from this site; however, this taxon is cultivated as an ornamental and often escapes cultivation. Lopez
and Fennessy (2002) found that the Coefficient of Conservatism values tend to be low in Ohio wetlands that
are surrounded by agriculture fields and urban centers. This is consistent with Cohen et al. (2004) findings
of the effect of landscape development of intensity on Coefficient of Conservatism values in depressional
marshes of Florida. Floodplains and riparian zones are generally vulnerable to non-native and invasive
species under normal conditions due to the dynamic nature of riparian systems but more so under altered
hydrological conditions (Nilsson & Berggren 2000).
The NMS ordination of species and samples were associated with age and river kilometer (Fig. 4i; axis
1), wet gradient or water depth (Fig. 4i; axis 2, as indicated by wetland indicator status values and distance
of sample plots from the edge of the embayment to open water), size and site (Fig. 4ii, iii; axis 3). Non-native
species cover, total cover, Coefficient of Conservatism, Richness, and Diversity were associated moderately
to the ordination scores. Furthermore, MRPP was able to detect differences in species and species cover
values among embayment sites, age, and river kilometer location. These results indicate that embayment
location, size, and age are factors that determine the distribution and abundance of plant species.
Of the herbaceous plant communities we defined, there were four groups that were distinct and had
limited integration with the other herbaceous communities. These include two aquatic and two emergent
communities. Two communities of lotus (Nelumbo nucifera and N. luted) are considered distinct and unique
for this region of the Ohio River (Table 4 and Fig. 4). The Nelumbo lutea community was an important com-
ponent of the emergent-submerged aquatic community at Ginat’s Run and was unexpected for the unglaciated
region of the state (Alison Cusick, pers. comm.; Cusick & Silberhorn 1977). The tall emergent community at
Chickamauga Creek is the only known community along the Ohio River that has a dominant and indicator
species of Nelumbo nucifera, an introduced taxon. The remaining eight defined communities are composed
of common elements of the region and are not unexpected. Communities are typically composed of species
that are distributed along an integrated series of aquatic, wetland and upland habitats and are consistent with
riparian forests of the Allegheny Plateau, Pennsylvania (Williams et al. 1999).
The purpose of this study was to classify and describe plant communities of selected embayment sites
along the mid to mid-upper Ohio River floodplain. We classified five Acer saccharinum bottomland hard-
wood forest communities, five sapling and shrub and twelve herbaceous plant communities. Herbaceous
plant species and communities were further analyzed through the use of indirect ordination procedtf^
and non-parametric tests in order to describe potential environmental gradients. We also have provi
baseline information on embayment plant communities along this stretch of the Ohio River that alloWV.
future comparative studies. Future embayment work needed includes surveys of the flora, fauna
classification, characterization of nutrient cycling, productivity analyses, soil classification, succession*1
relationships and hydrological surveys.
We extend <
tience, and
W. Keller ai
ACKNOWLEDGMENTS
ir gratitude and appreciation to Donald C. Tarter and Ralph T. Taylor for their assistance^*
teir review of the original Masters’ manuscript. Special thanks go to David J. Gibson,
l Lori A. Dameron for providing constructive comments and edits on drafts of the manu
J. Bot Res. Inst. T«as 4(1): 434. 2
CHECKLIST OF THE VASCULAR PLANTS
OF ALLEGHENY COUNTY, PENNSYLVANIA
Cynthia M. Morton
Section of Botany
S Carnegie Museum of Natural History
4400 Forbes Avenue
Pittsburgh, Pennsylvania 15213, USA
mortonc@carnegiemuseums.org
Loree Speedy
Carnegie Museum of Natural History
4400 Forbes Avenue
Pittsburgh, Pennsylvania 152 13, USA
EESUMEN
INTRODUCTION
hen though Pennsylvania lacks checklists for most of its counties, the vascular flora known thus far is quite
l|Se- Rhoads and Klein (1993) reported 3318 taxa of vascular plants for the state, including 2076 native
1242 introduced. Approximately 27 of Pennsylvania’s 67 counties have floras completed, but some are
"“published theses and surveys and therefore unavailable to the public. The exact number of genera and
^des Ae flora of Allegheny County was heretofore poorly known because the most recent checklist
been compiled over 50 years ago. This paper is an updated checklist providing the baseline information
^nfedto monitor environmental changes and guide conservation decisions.
‘^gbeny County, Pennsylvania, is located in the southwestern part of the state (Fig. 1). It is bordered by
, rand Armstrong Counties to the north, Westmoreland County to the east, Washington and Beaver
^hes to the west.
**al annual precipitation in Allegheny County is approximately 36 to 40 inches. Average seasonal
7™ « 30 inches annually in the southern valley areas to nearly 50 inches at the higher northern eleva-
^ Snow cover of 1 inch or more occurs about 20 days each year in the south and about 45 to 50 days m
Jbe average date of the last 32 degree temperature in the spring is usually around April 29th and the
*e season approximately October 15th. The length of the growing season, from the last killing
^ng to the first killing frost in autumn, is about 170 days (USDA 1981).
The soils of the county consist of eight main series: Gilpin-Upshur-Atkins association, Gilpin-Weikert-
Atkins association, Culleoka-Weikert-Newark association, Gilpin-Wharton-Upshur association, Dormont-
Guemsey-Culleoka association. Urban land-Philo-Rainsboro association, Urban land-Dormont-Culleo
association, and Strip mines-Guemsey-Dormont association. These soils associations can be grouped mto
two general kinds of landscapes: areas that are unaltered and areas altered by urban development and strip
mines (USDA 1981).
The first five associations listed above are from areas that are unaltered by urban development and strip
mines and make up about 74 percent of the county. .
Gilpin-Upshur-Atkins association is located mainly on steep and very steep sides of valleys but a
includes narrow, nearly level flood plains. These soils are moderately-deep to deep and well-drained,
soils are underlain by red and gray shale on uplands and deep, poorly drained soils on flood plains.
Morton and Speedy, Flora of Allegheny County, Pennsylvania
437
soil association makes up about 15 percent of the county. The area is mostly wooded; due to flooding and
1 steepness of the terrain, other land uses are limited.
I Gilpin-Weikert-Atkins association is located mainly on steep and very steep sided valleys but also in-
cludes narrow, nearly level flood plains. These soils are shallow to moderately deep, and well-drained. They
are underlain by gray shale on uplands and deep, poorly drained soils on flood plains. This soil association
makes up about 5 percent of the county. The area is mostly wooded; due to flooding and steepness of the
| terrain, other land uses are limited.
I Culleoka-W eikert-Newark association is also located mainly on steep and very steep sided valleys and
includes narrow, nearly level flood plains. These soils are shallow to moderately deep, and well-drained.
1 They are underlain by shale and limestone on uplands and deep, poorly drained soils on flood plains. This
| soil association makes up about 4 percent of the county. The area is mostly wooded; due to flooding and
the steepness of the terrain, other land uses are limited.
I Gilpin-Wharton-Upshur association is on undulating to hilly uplands and is highly dissected by small
streams and drainage ways. These soils are moderately deep to deep and moderately to well-drained. These
| soils are underlain by red and gray shale on uplands. This soil association makes up about 32 percent of
I the county. Much of this area has been cleared. Some of the cleared areas are under continued suburban
1 development while others are used for recreation, are farmed or remain idle. Artificial drainage is needed
in many areas while other areas are susceptible to landslides.
I Dormont-Guernsey-Culleoka association is on undulating to hilly uplands and is highly dissected by
I small streams and drainage ways. These soils are moderately deep to deep and moderately to well-drained.
I They are underlain by shale and limestone on uplands. This soil association makes up about 18 percent of
| the county. Much of this area is under continuous suburban development while other cleared areas are used
for recreation, contain scattered farms or remain idle.
The Urban land-Philo-Rainsboro association. Urban land-Dormont-Culleoka association, and Strip
mmes-Guemsey-Dormont association make up about 20 percent of the county and contain areas that are
^ered by urban development and strip mines.
Urban land-Philo-Rainsboro association is mainly on nearly level bottom land adjacent to major streams.
; ^Iese soils are deep and moderately well-drained on flood plains and terraces. This soil association makes
UP about 6 percent of the county. Most of this area is used for residential, commercial, and major industrial
^velopments and for major highway and railroad routes. In general some areas are prone to flooding.
Urban land-Dormont-Culleoka association is in the southern half of the county. It is on nearly level to
% uplands that are dissected by small streams and drainage ways. The soil is moderately deep to deep
moderately to well-drained. The urban land is underlain by shale while the uplands are underlain by
«®estone. This soil association makes up about 9 percent of the county. This association is used mainly for
m^an development. Most land is limited by a seasonal high water table.
Strip mines-Guernsey-Dormont association is in the southern half of the county. It consists of rolling
,0 very steep areas that have been strip mined. The soil is deep and moderately well-drained. The strip
®lnes are underlain by shale and limestone on uplands. This soil association makes up about 5 percent of
county. Much of the land remains idle and vegetation is sparse (USDA 1981).
Allegheny County consists of a hilly region that is situated in the Allegheny Plateau physiographic
re^on- The Allegheny Plateau is mainly covered by hardwood forest. The county is bounded partially by
the Monongahela and Youghiogheny Rivers to the southeast and partially by the Allegheny River to the
“^heast. Elevations range from a high of 1,200 feet to a low of 82 feet above sea level.
Allegheny County is known for the three major rivers that flow through it: the Allegheny and the
T^ongahela converge at Pittsburgh to form the Ohio River. An additional river, the Youghiogheny, meets
Monongahela at McKeesport, 10 miles south of Pittsburgh. Allegheny County contains seven major
j^ersheds, located in the subbasins of the lower Allegheny River, the Monongahela River and the Ohio
438
The Lower Allegheny River watershed drains a total area of 324 square miles. Its major waterways include
the lowest portion of the Allegheny River and Deer Creek, which originates north of Allegheny County.
The watersheds of the Monongahela subbasin in Allegheny County include the Lower Youghiogheny
River, the Middle Monongahela River and Turtle Creek. The Lower Youghiogheny River watershed has a total
drainage area of 478 square miles and its major waterway in Allegheny County is the lower Youghiogheny
River, which enters the county in the southeast. The Middle Monongahela River watershed has a total drain-
age area of 509 square miles and its major waterway in Allegheny County includes the lower portion of
the Monongahela River. Turtle Creek watershed has a total drainage area of 202 square miles and its major
stream is Turtle Creek. This watershed drainage area includes forest, farmland, industry, abandoned mined
lands, as well as urban and suburban residential communities.
The Ohio subbasin contains the following watersheds: Chartiers Creek, Raccoon Creek and Upper
Ohio River. Chartiers Creek drains a total of 296 square miles and enters Allegheny County in the south-
west. Raccoon Creek drains 327 square miles and creates a portion of the western boundary of Allegheny
County. Forty miles of the main stem of Raccoon Creek and 30 to 40 miles of tributaries are degraded by
mine drainage. The Upper Ohio River watershed has a total drainage area of 209 miles; Sewickley Creek
and the upper portion of the Ohio River are its major water ways.
Watershed conditions need to be evaluated to detect if biodiversity is increasing or decreasing. Com-
parison studies between years and sites will be extremely important in mining areas because they will reflect
the interactions of many of the other environmental indicators, and these comparisons provide measurable
effects for management decisions (Pennsylvania DEP Watershed Notebook, 2006).
History
Allegheny County was the first county in Pennsylvania to be given a Native American name. During the
colonial era various native groups, such as the Iroquois, Lenape, Shawnee, and Mingo, settled in the area.
Very little however is known about the native inhabitants of the region prior to European contact. The first
known European settlers to enter the area were the French in 1749. The French claimed the Ohio Valley
and all of Western Pennsylvania for Louis XV of France. Both the French and the British desired control
over the rivers because most of the towns were developed along rivers. In 1754, George Washington forged
an expedition into the region which ended in defeat by the French and Indians at Fort Necessity. In 1758
the British won the French and Indian War and settlers came back into the region. In the 1770s both Penn-
sylvania and Virginia claimed the region that is now Allegheny County. Pennsylvania administered most
of the region as part of Westmoreland County. Virginia considered everything south of the Ohio River and
east of the Allegheny River to be part of its Yohogania County and governed it from Fort Dunmore. The
overlapping boundaries, multiple governments, and confused deed claims soon proved unworkable. The
claims were settled in 1780 when the Mason-and-Dixon Line was extended to the original limit of Penn’s
land grant, giving Virginia frontage on the Ohio River, but ceding the land north of the Mason-and-Dixon
Line to Pennsylvania.
Allegheny County was created in September 24, 1788 from portions of Washington and Westmoreland
Counties. The county’s current borders were settled by the 1800s.
In the late 1700s farming played a critical role in the growth of the area. During this period a surplus
of grain occurred due to transportation difficulties and the farmers began to distill the grain into whiskey.
The production of whiskey significantly helped the economy of the farmers until the 1790s when a whiskey
excise tax was imposed. This started the Whiskey Rebellion when the farmers who depended on whiskey
income refused to pay. After a series of demonstrations by farmers, President George Washington arrived
with troops to stop the rebellion.
The county’s strategic location and wealth of natural resources spurred its commercial and industrial
growth in the 19* century. The blast furnace, erected by George Anschutz about 1792, was the forerunner
of the iron and steel industry that for more than a century was the county’s economic mainstay. Pittsburgh.
439
the county seat, would later be labeled the Steel Capital of the World. The Pennsylvania Canal and the
Portage Railroad, both completed in 1834, opened vital markets for trade and shipping. As the 19th century
progressed, Pittsburgh became one of the nation’s greatest industrial cities and was a leading producer of
glass, iron and textiles. During the American Civil War the county became a supplier to the Union and the
iron industry and its shipyards gained wealth. After the American Civil War, great numbers of European
immigrants moved into the area. Transportation improvements facilitated growth, and by 1900 nine railroad
lines entered Pittsburgh. The Army Corps of Engineers constructed an extensive series of locks and dams
during this time which improved shipping. World War II and the industrial demand it created temporarily
boosted Pittsburgh’s economy.
By the late 1970s, the steel industry had virtually disappeared. Over 100,000 steel and steel-related jobs
were eliminated between 1978 and 1983. By the mid-1980s many of the region’s manufacturing plants had
gone out of business or left the area. Since then the county has successfully diversified its economy based
on services, especially medical, financial, corporate, and educational, rather than steel. In 2008, Pittsburgh
has been ranked in the top ten among U.S. cities hosting headquarters of Fortune 500 corporations. These
include major corporations such as U.S. Steel Corp., H.J. Heinz, PNC Financial Services Group, and PPG
industries (Alan Enterprise 2007).
The county is divided into 130 municipalities with the majority of the population residing in the urban
and suburban areas. The first census taken in 1790 documented that Allegheny County had a population
of nearly 10,309. As of 2000, the population was 1,281,666.
According to the U.S. Census Bureau, the county has a total area of 1,929 km2 (745 sq mi); 1,891 km2
(730 sq mi) of it is land, and 38 km2 (15 sq mi) is water (U.S. Census Bureau 2009).
Major Collectors
Collecting in Allegheny County for the Carnegie Museum Herbarium began in the early 1800s and has
continued to the present. Most of the collectors have been members of the Botanical Society of Western
Pennsylvania and the staff of the Section of Botany at Carnegie Museum.
Botanists who have made significant contributions to the knowledge of the Allegheny County are W.A.
Tanol, W.E. Buker, O.E. Jennings, J. Bright, and B.L. Isaac andJ.A. Isaac.
METHODS
checklist was compiled by searching the herbaria of the Carnegie Museum of Natural History (CM).
Herbarium collection information was obtained from the Morris Arboretum (MOAR). Other herbaria that
"ere examined for specimens include the Academy of Natural Sciences of Philadelphia. In addition, fieldwork
*as conducted targeting underrepresented areas of the county. Most of the collections were made during
k late 1940s and early 1950s but several specimens date back as far as 1873 and 1875. The majority of
specimens are deposited at the Carnegie Museum of Natural History (CM). Rhoads and Block (2000)
*as ttle Primary source for plant identification. For generic and species names, we have generally followed
Synthesis of North American Flora (Kartesz 1999). Authorities are abbreviated for the majority of taxa
Wording to Brummitt and Powell (1992).
RESULTS AND DISCUSSION
^ list includes the names of all native and naturalized species known to occur in Allegheny County. It
“Eludes a total of 142 families, 628 genera, and 1533 species. The five families with the largest number of
^csare Asteraceae, Poaceae, Rosaceae, Cyperaceae and Fabaceae. Carex, Viola, Crataegus, Symphyotrichum
“^Rubus are the five largest genera. This checklist recognizes 10 species of Lycophytes, 44 species of Poly-
^jophytes, 5 species of Equisetophytes, 12 species of Gymnosperms and 1462 species of Angiosperms.
316 apProximately 400 non-native species that have been introduced mainly from Europe and Eurasia.
03 species have global or state conservation status ranking as defined by The Nature Conservancy.
Of the 103 plants that have global or state ranking, one has G2 status, and four have G3 status. The
remaining 98 plants are either a G4 or G5 status, indicating, respectively, an apparently secure or secure
condition globally.
Potamogeton tennesseensis (Tennessee pondweed) has a status of G2 or an imperiled status. This species
contains floating leaves 2 to 4 cm long and 5 tol3 mm wide. Submersed linear leaves are 1 to 3 veined with
prominent lacunar bands, whereas the floating leaves have 9 to 23 veins. The stipules are fused to the leaf
bases for Vi or less of their length, or are free. The achenes are orbicular with a winged keel. The small size
of P. tennesseensis, in combination with the acute floating leaves, renders the species reasonably distinctive;
the species can however be confused with an immature specimen of Potamogeton epihydrus.
With a status of G3 or vulnerable status, is Scutellaria saxatilis (rock skullcap). These plants are slender
and decumbent with glabrous or eglandular stems. The leaves are glabrous, petioled, ovate-shaped and
rounded at the base. The margins have fewer than 10 rounded teeth on each side. This species has a distinc-
tive prominent protuberance on the upper corolla. The flowers are in terminal few-flowered racemes and
have bracts underneath the calyx. This species can be confused with Scutellaria ovata, but this species has
much longer leaves and is not decumbent.
Trifolium reflexum (buffalo clover) has a status of G3G4 or vulnerable to apparently secure status. These
plants are 1 to 5 cm tall and are not stoloniferous. They have globose heads that are 2.5 to 4.5 cm in diameter
and each flower has a pedicel at least 2 mm long. The corolla has a red or white standard and white wings
and keel. T. reflexum has calyx teeth that are twice or more than twice as long as their calyx tube. This species
can be confused with Trifolium hybridum, T. pratense, or T. stoloniferum (Vincent 1991); however, T. reflexum
is distinctive in having long calyx teeth, long pedicels, and by lacking stolons.
With a ranking of G3 or vulnerable status, Crataegus pennsylvanica (Pennsylvania hawthorn) is a tree of
up to 10 meters with slender thorns that become stout with age. Their leaves are broadly ovate and sharply
toothed with 4 to 6 pairs of shallow lateral lobes. When young the leaves are densely hairy on the underside,
but the hairs become more infrequent with age. The upper surface of the leaves is scabrous. The flowers are
stalked with bases of the stalks densely hairy, the sepals have glandular teeth, and the stamens are about
10 in number. Their anthers are ivory-colored. The fruits are about 12 mm in diameter.
Delphinium exaltatum also has a global ranking of G3 or vulnerable status. Delphinium exaltatum con-
tains alternate leaves palmately divided into 3 to 5 parted segments, 15 cm long, silvery-green abaxially
and pubescent above and below. The basal lobes of some of the larger leaves are divided again and make the
blade appear 5-lobed. The flowers are in terminal racemes. The 4 purplish-blue petals are dimorphic with
the upper two petals expanded at the base and forming a spur. The lateral petals are reflexed in the apica
half with long white hairs. The stamens are approximately 30 in number with yellow pollen. The 5 sepals
are irregular. The species can be confused with Delphinium tricorne ; however D. tricorne is a smaller plant
and flowers much earlier than D. exaltatum, in May to early June. By July, when D. exaltatum is beginning
to flower, D. tricorne already has set fruit. Most of the collections are from rich shaded woods and on rocky
limestone bluffs. This species is quite conspicuous and may be subject to casual picking or may be dug or
gardens.
There are 8 taxa in the Allegheny County flora that are listed by the Pennsylvania Department of Ag-
riculture (2007) as noxious weeds. It is therefore illegal to propagate, sell or transport the following ta?a
the Commonwealth: Cannabis sativa (marijuana), Cirsium arvense (Canadian thistle), Cirsium vulgare (
or spear thistle). Datura stramonium (jimsonweed), Lythrum salicaria (purple loosestrife), Pueraria montam
var. lobata (kudzu), Rosa multiflora (multiflora rosa), and Sorghum halepense (Johnsongrass). Other spec »
considered serious invasives in Pennsylvania’s native ecosystems are: Aegopodium podagraria (bishops gw
weed), Ailanthus altissima (tree-of heaven), Akebia quinata (chocolate-vine), Alliaria petiolata (garlic ’
Ampelopsis brevipedunculata (amur peppervine), Berberis thunbergii (Japanese barberry), Berberis vulgaris (
pean barberry), Bromus tectorum (cheat grass), Celastrus orbiculatus (oriental bittersweet), Elaeagnus um ^
(autumn olive), Euonymus alatus (winged euonymus), Hesperis matronalis (dame’s rocket), Ligustrum vu
Morton and Speedy, Flora of Allegheny County, Pennsylvania
(common privet), Lonicera japonica (Japanese honeysuckle), Lonicera maackii (Amur honeysuckle), Lonicera
morrowii (Morrow’s honeysuckle), Lonicera tartarica (Tartarian honeysuckle), Miscanthus sinensis (Chinese
silver grass), Ornithogdum nutans (drooping star-of-Bethlehem), Omithogalumumbellatum (star-of-Bethlehem),
Postinaca sativa (wild parsnip), Paulownia tomentosa (princesstree), Phalaris arundinacea (reed canary grass),
Phragmites australis (common reed). Polygonum cuspidatum (Japanese knotweed), Rhamnus cathartica (Euro-
’ jean buckthorn), Rubus phoenicolasius (wine raspberry), Ulmus pumila (Siberian elm) and Viburnum opulus
var. opulus (highbush-cranberry) (DCNR 2004).
While this checklist may be incomplete, it is the most comprehensive list of Allegheny County species
presently available.
p ANNOTATED CHECKLIST OF THE SPECIES OF ALLEGHENY COUNTY, PENSYLVANLA
global: state ranking [Synonyms] and non-native source. We follow the state (S) and global (G) ranking
systems developed by The Nature Conservancy (1996 version). The global numbers are designated from 1
(critically imperiled) to 5 (secure). Other notations include SH, which denotes historical occurrence, and
$*, which indicates reported without persuasive documentation. Synonyms are included for names not in
common usage in the state or regional manuals. In cases where there was more than one specimen pres-
at in the Carnegie Museum Herbarium collection, recent collections of current collectors for the western
Pennsylvania region were cited.
Families, genera, and specific and infraspecific taxa are arranged alphabetically within vascular plant
poups Angiosperms, Equisetophytes, Gymnosperms, Lycophytes, and Polypodiophytes.
."tea americana (L.) Vahl, 1 994 WA. Zanol 3074
keraceae
tocampestre L, 2002 B.L. Isaac and J A Isaac and S.P. Grund
; and McPherson 14353 Europe
** negundo L. var. negundo, 1999 B.L Isaac and T. Herman
rubrum L.v;
1SCW58
tenjbrum l var. trilobum Torr. & A. Gray ex K. K
“ ccharum Marsh, var. saccharum, 1 998 S A. Thompson
B.L Isaac and JA. Isaac 15158
Lam, 1936 O.E Jennings sit- '
^aceae
F* omericanus (Raf.) Raf., 1989 S.A. Thompson 6862
51 Scorns calamus L. var. americanus (Raf.) H.D. Wulff.]
^ calamus L„ 1943 e. Mason s.n
„ 1991 B.L Isaac and JA. Isaac and
E Allen 3511
Sagittaria tof/fo//oWilld., 1975 J.M. Eggers 14
Sagittaria rigida Pursh, 1 940 O.E. Jennings s.n.
Amaranthaceae
Amaranthus albus L, 1986 AW. Cusick 25787 [Amaranthus
albus L. var. pubescens (Uline & Bray) Fern.]
Amaranthus arenicola I.M. Johnston, 1946W.E. Buker s.n.
Amaranthus blitoides S. Wats. s.n„ Exotic
Amaranthus blitum L. var. emarginatus (Moq. ex Uline & Bray)
comb., 1 993 W.A. Zanol 2066 [ Amaranthus lividus L.]
America
Amaranthus graecizans L, 1 91 3 J. Bright s.n. Mediterranean
Amaranthus hybridus L„ 1 990 W.A. Zanol 339
Amaranthus hypochondriacus L., 1891 JA Shafer s.n. Africa
js retroflexus L., 1968 W.E. Buker s.n. Trc
rs tuberculatus (Moq.) Sauer, 1 922 J. Bright s.
Raf, 1 991 WA. Zanol 825 [Alismaplanta-
“vuor lca L. var. parviflorum (Pursh) Torr.]
W australis (J.G. Sm.) Small, 1922 AB. Lord s.n. [Sagit-
eayelmanniana J.G. Sm. ssp. longirostra auct non
Rhus aromatica Ait. var. aromatica, 1 947 W.E. Buker s,n.
Rhus copallinum L, 1910 AE Ortman s.n.
Rhus glabra L., 1945 LK. Henry s.n.
Rhus typhina L., 1 998 K. McGowan and M. Scanlon 220
Toxicodendron radicans (L) Kuntze ssp. negundo (Greene) Gillis,
1 998 K. McGowan and M. Scanlon 1 79
Toxicodendron vemix (L) Kuntze, 1 91 5 O.E Jennings s.n.
Aegopodium podagraria L, 2001 B.L Isaac and J. Isaac and C.
Chuey and S. Grund 13560 Eurasia [Aegopodium poda-
graria L var. variegatum Bailey]
Aethusa cynapium L, 1916 N. McCallum s.n. Eurasia
Anethum graveolens L„ 1 945 W.E. Buker s.n. Europe
Angelica atropurpurea L„ 1933 C.M. Boardman s.n. [Angelica
atropurpurea L. var. occidentals Fassett]
Angelica venenosa (Greenway) Fern., 1945 W.E. Buker s.n.
Chaerophyllum procumbens (L.) Crantz var. procumbens, 1 989
C.W. Bier s.n.
Cicuta maculata L var. maculata, 1 991 WA Zanol 805
Conium maculatum L, 1 998 S A Thompson and B. Isaac and
J. Isaac and F.H. Utech 14381 Europe
Cryptotaenia canadensis (L) DC., 2001 B.L Isaac and J. Isaac
and C. Chuey and S. Grund 1 3599
Daucus carota L., 2002 H. George s.n. Eurasia
Erigenia bulbosa (Michx.) Nutt., 1922 O.E. Jennings s.n. G5:S4
Foeniculum vulgare P. Mill., 1889 J.A. Shafer s.n. Southern
Europe
Heracleum maximum Bartr, 1997 SAThompson and M. Sally
13513 [Heracleum lanatum Michx.]
Hydrocotyle americana L, 1 930 W.R. VanDersal 1 73 1
Osmorhiza claytonii (Michx.) C.B. Clarke, 1998 SAThompson
and B. Isaac and J. Isaac and F.H. Utech 14327
Osmorhiza longistylis (Torr.) DC, 1998 SAThompson and B.
Isaac and J. Isaac and F.H. Utech 14328
Oxypolis rigidior (L) Raf., 1 91 2 J. Bright s.n. G552
Pastinaca sativa L, 1 997 SA Thompson and B.L. Isaac 1 3645
Eurasia
Sanicula canadensis L. var. canadensis, 1998 SA. Thompson
and B.L Isaac and J.A. Isaac 15150
Sanicula marilandica L. 1 969 W.E. Buker s.n.
Sanicula odorata (Raf.) K.M. Pryer & L.R. Phillippe, 2002 Bi-
Isaac and JA Isaac and S.P. Grund and McPherson 14356
[Sanicula gregaria Bickn.]
Sanicula trifoliata Bickn, 1997 SAThompson and M. Sally
Taenidia integerrima (L) Drude, 1 946 W.E. Buker s.n.
Thaspium barbinode (Michx.) Nutt, 1994 F.H. Utech andTD.
Jacobsen and R.W. Kiger 94-2265
Thaspium trifoliatum (L) A. Gray var. aureum Britt, 1997 Bi-
Isaac and J A Isaac 9603
Zizia aptera (A Gray) Fern, 1 969 W.E. Buker s.n.
Zizia aurea (L) W.DJ. Koch, 1 991 WA Zanol 479
oaca, 1938 RJ.Templeton and W.C Grimm
Ilex verticillata (L) A. Gray, 1 944 O.E. Jennings s.n.
Arisaema dracontium (L.) Schott, 1 942 O.E. Jennings s.n.
Arisaema triphyllum (L) Schott ssp. pusillum (Peck) Huttleston,
1940L.K. Henry s.n.
Arisaema triphyllum (L.) Schott ssp. stewardsonii (Britt) Hut-
tleston, 1 945 O.E. Jennings s.n.
Arisaema triphyllum (L.) Schott ssp. triphyllum, 1998 S.A
c and J.
„ 1991 WA. Zanol
425
Aralia hispida Vent, 1 908 O.E. Jennings s.n.
Aralia nudicaulis L, 1998 SA. Thompson and B. Isaac and J.
Isaac and F.H. Utech 14320
Aralia racemosa L ssp. racemosa, 1997 SAThompson and
B.L. Isaac and R. Goto and T. Collins 14205
Aralia spinosa L, 1997 SAThompson and B.L. Isaac 14199
Eleutherococcus sieboldianus (Makino) Koidzumi, 2002 B.L
Isaac and JA Isaac 14373 Japan [Eleutherococcus penta-
phyllus auct. non (Sieb. & Zucc.) Nakai]
Panax quinquefolius L, 1912 J. Bright 22
Panax trifolius L„ 1 920 O.E. Jennings and LC. Denise s.n.
Apocynum androsaemifolium L„ 1 991 WA Zanol 779
Apocynum cannabinum L„ 1 998 SAThompson and B.L Isaac
and JA Isaac 15099
Apocynum x floribundum Greene (pro sp.), 1917 B.H. Patterson
s.n. [Apocynum medium Greene]
Vinca minor L, 1998 K. McGowan and M. Scanlon 8 Europe
Asarum canadense L, 1945 O.E. Jennin
dense L var. acuminatum Ashe]
Endodeca serpentaria (L) Raf, 1937 O.E. Jennings s.n. tAristo-
lochia serpentaria L]
Isotrema macrophyllum (Lam.) CF. Reed, 2001 J A Isaac 1 3621
[Aristolochia macrophylla Lam.]
Asdepiadaceae
Asclepias exaltata L, 1945 L.K. Henry s.n.
Asclepias incarnata L ssp. incarnata, 1998 K. McGowan and
M. Scanlon 309
Asclepias purpurascens L, 1 998 SAThompson 1491 1
Asclepias quadri folia Jacq, 1 967 LK. Henry s.n.
Asclepias syriaca L, 1998 SA. Thompson and B.L teacand
J.A. Isaac 15207 [Asclepias syriaca L var. kansana (vaw
Palmer & Steyermark]
Asclepias tuberose L ssp. tuberosa, 2001 S.N. Costabile 16
Asclepias viridiflora Raf, 1 949 E. Mason s.n.
Matelea obliqua (Jacq.) Woods, 1896 WJ. McAdams s.
Achillea millefolium l_ 2001 B.L Isaac and J. Isaac and C Chuey
and S. Grund 1 3607 Europe .
Ageratina altissima (L) King & H.E. Robins, var. altis^^
S.P. Grund and M.L. Speedy 2898 [f "
Houtt]
Ilex montana Torr. & A Gray ex A Gray, 1 926 H.W. Graham s.n.
443
Mennaria howellii Greene ssp. neodioica (Greene) Bayer,
T 1992WAZanol 1221
Mennaria howellii Greene ssp. petaloidea (Fern.) Bayer,
|p;.)915 B.H. Patterson [Antennaria neodioica Greene ssp.
petaloidea (Fern.) Bayer & Stebbins]
Mennaria parlinii Fern. ssp. fallax (Greene) Bayer & Stebbins,
II 1925 J. Bright s.n.
Antennaria parlinii Fern. ssp. parlinii, 1950 L News
tennaria plantaginifolia (L.) Richards var. parlinii (Fei
I Cronq.]
Richards., 1946 D. Benz<
IAn-
, 1 986 A.W. Cusick 25792 [Bidens
polylepsis Blake]
Antennaria virginica Stebbins, 1 949 W.E Buker s.n. G4:S3 [An-
tennaria virginica Stebbins var. argillicola Stebbins]
Anthemis cotula L, 1 994 WA. Zanol 2441 Europe
Araium lappa L, 1 998 K. McGowan and M. Scanlon 353 Eurasia
j| . [Arctium vulgare (Hill) Evans]
Arctium minus (Hill) Bernh, 1994 WA Zanol 3469 Eurasia
Amoglossum atriplicifolium (L) H.E. Robins, 1 91 2 J. Bright 27
m- \Cacalia atriplicifolia L]
lA/nogbssum reniforme (Hook) H.E. Robins, 1 91 9 Of. Jen-nings
m.G45l [Cacalia muehlenbergii (Schultz-Bip) Fern.]
Artemisia annua L, 1 986 AW. Cusick 25790 Eurasia
Artemisia biennis Willd, 1889 J. Furgeson s.n. Europe
Artemisia ludoviciana Nutt. ssp. ludoviciana, 1967 W.E. Buker
;; [Artemisia ludoviciana Nutt. var. gnaphalodes (Nutt.)
Torr. & A Gray]
Artemisia vulgaris L, 1998 SA Thompson and B.L. Isaac and
Cirsium vulgare (Savi) Ten, 1 997 J.A Isaac and B.L Isaac 10178
Eurasia
Conoclinium coelestinum (L) DC, 1997 S A Thompson and B.L
Isaac 14127G553 [Eupatorium coelestinum L]
Conyza canadensis (L) Cronq, 1 998 SA. Thompson and B.L.
Isaac and JA. Isaac 15206
Coreopsis lanceolata L, 2001 B.L Isaac and J. Isaac and C Chuey
and S.Grund 13604
Coreopsis tinctoria Nutt var. tinctoria, 1965 N.R. Farnsworth
Crepis capillaris (L) Wallr, 1 935 O.E. Jennings s.n. Europe
Doellingeria infirma (Michx.) Greene, 1919 O.E Jennings s.n.
[Aster infirmus Michx.]
Doellingeria umbellata (P. Mill.) Nees var. umbellata, 1966
R.N. Blomster WP-651 [Aster umbellatus P. Mill. var. um-
bellatus J
Dyssodia papposa (Vent.) AS. Hitchc, 2004 B.L Isaac and JA.
Isaac 18551
Echinacea purpurea (L) Moench, 1 994 WA Zanol 3499
Eclipta prostrate (L.) L, 1 994 WA. Zanol 3457
Erechtites hieraciifolius (L) Raf. ex DC. var. hieraciifolius, 1994
WA Zanol 3752
Erigeron annuus (L) Pers, 2001 B.L Isaac and J. Isaac and C
Chuey and S. Grund 1 3566 [Erigeron annuus (L) Pers. var.
discoideus (Victorin & Rouss.) Cronq.]
Erigeron philadelphicus L var. philadelphicus, 1 998 K. McGowan
JA Isaac 15171 EuH
Misperennis L, 2002 B.L. I;
"tens bipinnata L„ 1997
gw cemua L, 1 993 WA Zanol 2072
frondosa L„ 1 998 K. McGowan and M. Scanlon 389
tripartita L, 1 993 WA. Zanol 21 50
vulgata Greene, 1 944 LK. Henry s.n.
m°nia Weroides (L.) L'Her var. asteroides, 1988 E. York 355
Erigeron pulchellus Michx. var. pulchellus, 1 982 W.E. Buker s.
: 14335 Europe Erigeron strigosus Muhl. ex Willd. var. strigosus, 1998
and B.L. Isaac McGowan and M. Scanlon 280
Eupatorium altissimum L„ 1997 SA Thompson and J.
r. eupatorioides, 1945
■«a/w eupatorioides (L.) Shinners
E Mason s.n.
tendula officinalis L„ 1966 W.E Buker s.n. Southern Europe
1 992 G.H. Pollack 56 Europe
964 N.R. Farnsworth WP-972 Europe
Eupatorium perfo/iatum L, 1 998 K. McGowan and M. Scanlon
363
Eupatorium serotinum Michx, 1 996 LL Smith and R Hawk 75
Eupatorium sessilifolium L, 1945 LK. Henry sit
Eurybia divaricata (L.) Nesom, 2000 R.B. Coxe s.n. [Aster
divaricatus L]
Eurybia macrophylla (L.) Cass, 1 945 LK. Henry s.n. [Aster
{(ntaureac\
bntaurea r
_ L var. radiata DC.]
;**tourea stoebe L. ssp. micranthos (Gugler) Hayek, 2000 R.B.
-T*® in. Europe [Centaurea biebersteinii DCJ
intVbus L„ 2002 H. George s.n. Europe
°r5im arvense (L) Scop, 1998 SA. Thompson 14912 Eur-
■p» discolor i
(Muhl. ex Willd.) Spreng, 1921 O.E. Jennings
m Mlchx- 1950 D-H. Krouse s.n.
C^Pumilum (Nutt) Spreng, 1991 WA. Zanol 876
Eurybia schreberi (Nees) Nees, 1991 WA. Zanol 866 [Aster
schreberi Nees]
Euthamia graminifblia (L) Nutt, 1 998 B.L Isaac and JA. Isaac
1 1657 [Solidagograminifolia (L) Salisb.]
Eutrochium fistulosum (Barratt) E. Lamont, 1998 K. McGowan
and M. Scanlon 288 [Eupatorium fistulosum Barratt]
Eutrochium maculatum (L) E Lamont var. maculatum, 1990
W.A. Zanol 225 [Eupatorium maculatum L. var. macu-
Eutrochium purpureum (L) E Lamont var. purpureum, 1 994 WA
Zanol 3456 [Eupatorium purpureum L. var. purpureum]
Galinsoga quadriradiata Cav, 1997 SA. Thompson and B.L
Isaac 1 3869 Central and South America
Gamochaeta purpurea (L) Cabrera, 1902 J.A. Shafer s.n.
[Gnaphalium purpureum L var. purpureum]
Morton and Speedy, Flora of Allegheny County, Pennsylvania
445
Sonchus arvensis L. ssp. uliginosus (Bieb.) Nyman, 1994 WA
gjkKhus oleraceus L, 2001 B.L. Isaac and J. Isaac and C. Chuey
and S. Grund 1 3589 Europe
mrphyotrichum cordifolium (L.) Nesom, 1 997 SA Thompson
and BL Isaac 14166 [Aster cordifolius L]
Jmtphyotrichum drummondii (Lindl.) Nesom var. drummondii,
1912 B.H, Patterson s.n. G5:SH [Aster drummondii Lindl.]
fymphyotrichum laeve (L) A. & D. Love var. laeve, 1944 LK.
[pHenry s.n. [Aster laevis L. var. laevis]
fyriphyotrichum lanceolatum (Willd.) Nesom ssp. lanceolatum
^yar. interior (Wieg.) Nesom, 1921 M. Knauz s.n. [Aster lan-
ceolatus Willd. ssp. interior (Wieg.) A.G. Jones]
bfynphyotrichum lanceolatum (Willd.) Nesom ssp. lanceolatum
t m lanceolatum, 1990WAZanol 1 19 [Aster simplexWUd
i :Wt. simplex]
fymphyotrichum lateriflorum (L) A. & 0. L6ve var. horizontal
(Desf.) Nesom, 1 944 LK. Henry s.n. [Aster lateriflorus (L)
Britt var. pendulus (Ait.) Burgess]
tymphyotrichum lateriflorum (L.) A. & D. L6ve var. lateriflorum,
1998 SA Thompson and B.L. Isaac and JA Isaac 15174
P [Aster lateriflorus (L.) Britt]
anglia
, 1998 S
c and J.
> novae-angliae L]
fymphyotrichum oblongifolium (Nutt.) Nesom, 1965 N.R.
Farnsworth WP-981 [Aster oblongifolius Nutt.]
fynphyotrichum patens (Ait.) Nesom var. patens, 1914 B.H.
Patterson s.n. [Aster patens Ait.]
m¥votrkhum phlogifolium (Muhl. Ex Willd.) Nesom, 1922
1 ^ght s.n. [Aster phlogifolium Muhl. ex Willd.]
$tyhyotrichum pilosum (Willd.) Nesom var pilosum, 1 998 BJ_
Isaac and J.A. Isaac 1 1650 [Aster pilosus Willd.]
' Wphyotrichum pilosum (Willd.) Nesom var. pringlei (A. Gray)
1 966 R.N. Blomster WP-652 [Aster pilosus Willd. var.
(A A. Gray) S.F. Blake]
Ffyphyotrichum praealtum (Poir.) Nesom var. angustiorf Wieg.)
N^om, 1 998 B.L Isaac and JA Isaac 1 1 658 G5:S3 [Aster
Praealtus Poir. var. angustiorWteg]
Wphyotrichum praealtum (Poir.) Neson
Zanol 379 [Aster praealtus PoirJ
W^wyotrichum prenanthoides (Muhl. e: ,
^Thompson and B.L. Isaac and JA Isaac 15173 [As
Prenontho/des Muhl. Ex Willd.]
Wphyotrichum puniceum (L) A & D. Love var. puniceu
l^OWA Zanol 370 [Aster puniceus L]
^wwtrichum shortii (Lindl.) Nesom, 1998SA.Thomps<
BL Isaac and J .A. Isaac 15172 [Aster shortii Lindl.}
Wwyotrichum undulatum (L) Nesom, 1949 E. Mason s
l^ferundu/atum L.]
wPhyotrichum urophyllum (Lindt) Nesom, 1966 R.N. Bloi
sterWP-659 [Aster urophyllus Lindl.]
Return parthenium (L) Schultz- Bip, 2001 J.A. Isaac 1 36
rJro* [ Chrysanthemum parthenium (L) Bernh.]
r^erum vu!gare L 1 ^ W A 2862 Europe
Taraxacum officinale G.H. Weber ex Wiggers, 1 998 S AThomp-
son and M. Burroughs 14395 Eurasia
Tragopogon dubius Scop., 1 997 B.L Isaac and J A Isaac 961 0
Europe
Tragopogon lamottei Rouy, 1965 N.R. Farnsworth WP-1 134
Europe [Tragopogon pratensis L]
Tussilago farfara L., 1 998 B.L Isaac and M. Rosenberger 10210
Eurasia
Verbesina alternifolia (L) Britt, ex Kearney, 1 997 SA Thompson
and J.E. Rawlins 13890
Vernonia gigantea (Walt) Trel., 1998 K. McGowan and M.
Scanlon 342
Vernonia noveboracensis (L) Michx., 1 990 WA Zanol 245
Xanthium strumarium L, 1997 SA Thompson and B.L Isaac
Azollaceae
Azolla caroliniana Willd., 1 996 J. Wagner and P. Youngand s.n.
IcGowan and M. Scanlon
Berberis thunbergii DC, 1998 K. McGowan a
291 Japan
Berberis vulgaris L, 1939 W.C. Grimm and B
Caulophyllum thalictroides (L.) Michx., 1997 S A Thompson
and M. Sally 13497
Jeffersonia diphylla (L.) Pers., 1 945 H.D. Gill s.n.
Podophyllum peltatum L., 1 998 K. McGowan and M. Scanlon
Alnus serrulata (Ait.) Willd, 2004 C. Tracey and RJ. Gruszka and
M B. Steisslinger 2004-0030 [Alnus serrulata (Ait) Willd. var.
subelliptica Fern.]
Betula alleghaniensis Britt, 1 998 SA Thompson and B.L Isaac
and J.A. Isaac 15145
Betula lenta L, 1997 SA Thompson and B.L Isaac 14198
Betula papyrifera Marsh, var. papyrifera, 1 982 W.E. Buker s.n.
Carpinus caroliniana Walt ssp. virginiana (Marsh.) Furlow, 1 998
K. McGowan and M Scanlon 219
Corytus americana Marsh, 1998 S A Thompson and B. Isaac
and J. Isaac and F.H. Utech 14346
Corylus cornuta Marsh, var. comuta, 1 922 J. Bright s.n.
Ostrya virginiana (P. Mill.) K. Koch, 1 998 K. McGowan and E
Isaac and M. Scanlon 28
Campsis radicans (L) Seem, ex Bureau, 1906 JA Shafer s.n.
Catalpa bignonioidesWalt., 1997SAThompson and B.L Isaac
and M. Sally 13449
Catalpa speciosa Warder ex Engelm, 1998 S.A. Thompson
and B.L Isaac 14941
Boraginaceae
Buglossoides arvensis I
. Johnston, 1970 W.E. Buker so.
.Speedy
Sisymbrium altissimum L„ 1 993 F.W. Wright Jr. 1 06 Eurasia
fymbrium officinale (L.) Scop. 1945 W.E Buker s.n. Europe
EpSi isymbrium officinale (L.) Scop. var. leiocarpum DC.]
Jhbspi arvense L, 1 994 W.A. Zanol 2432 Europe
mis glabra L, 1946 W.E. Buker s.n. [Arabis glabra (L.)
kddleja davidii Franch., 1997 SA Thompson and B.L Isaac
* 13632 China
Cactaceae
Opuntia humifusa (Raf.) Raf, 1 944 O.E. Jennings m GS53
ssp. heterophylla, 1 924 J. Bright
(ollitriche palustris L„ 1913 J. Bright s.n. [Callitriche verna L.
IpMcte terrestris Raf. Emend Torn, 1 964 LK. Henry s.n.
Bright s.n.
i americanum (L.) Small, 1998 K. McGowan
and M. Scanlon 317 [ Campanula americana L]
Mia cardinalis L, 1 925 J. Bright s.n.
Mia inflata L, 1998 K. McGowan and M. Scanlon 265
Mia siphilitica L, 1 998 SA Thompson and B.L Isaac and
JA Isaac 15202
Mia spicata Lam. var. spicata, 1982 SA. Thompson and
■ W.E Buker 460
^xtonis perfoliata (L.) Nieuwl., 1991 WA. Zanol 589
knnabaceae
wulusjaponicus Sieb. & Zucc, 1997 S.A. Thompson and
81 Isaac 141 26 Asia
lupulus L. var. lupuloides E. Small, 1 932 G.K. Jennings
1|. O.E. Jennings s.n.
a Chod., 1946 W.E. Buke
I Mill., 1945 W.E. Buker and M. Henr
is Bartr. ex Marsh., 1945 LK. Henry s.i
a,oica L, 1912 J. Bright s.n. [Lonicera dioica
9^cescens(Rydb.) Butters]
jA^Jopon/coThunb., 2001 S.N. Costabile 12 Asia
T'onaackii (Rupr.) Herder, 2002 B.L. Isaac and JA
Lonicera morrowii A. Gray, 2002 5.P. Grund and M.L
2899 Japan
Lonicera sempervirens L., 1 945 W.E. Buker s.n.
Lonicera tatarica L., 2002 B.L. Isaac and J.A. Isaac 14337
Eurasia
Sambucus nigra L. ssp. canadensis (L.) R. Bolli, 1998 SA
Thompson and B.L Isaac and JA. Isaac 1 5085 [Sambucus
canadensis L.]
Sambucus nigra L ssp. nigra, 1 91 1 D.R. Sumstine s.n. Exotic
[Sambucus nigra L var. laciniata L]
Sambucus racemosa L var. racemosa, 1 987 F.H. Utech 87-045
[Sambucus racemosa L ssp. pubens (Michx.) House]
Symphoricarpos albus (L) Blake var. laevigatus (Fern.) Blake,
1 920 O.E. Jennings and L.C. Demoise s.n.
Symphoricarpos orbiculatus Moench, 1958 M. Reynolds s.n.
Triosteum aurantiacum Bickn. var. glaucescens Wieg., 1913
O.E. Jennings s.n.
Steyermark, 1 998 B.L Isaac and JA. Isaac 1 1 756
Triosteum perfoliatum L, 1 923 O.E Jennings s.n.
Viburnum acerifolium L, 1 998 S.A. Thompson and B.L. Isaac
Viburnum dentatum L var. dentatum, 2002 M.L Young and
B.R. Swiech 8
Viburnum dentatum L var. lucidum Ait., 1998 S A. Thompson
and B.L. Isaac and JA. Isa
and M.A. Rosenberger 1 41 14 Asia
Viburnum lantana L, 1998 SA Thompson and B. Isaac and J.
Isaac and F.H. Utech 1 4367 China
Viburnum lentago L„ 1998 S A Thompson and B.L Isaac and
JA Isaac 15100
Viburnum opulus L. var. opulus, 1997 S.A. Thompson and B.L
Isaac and M A. Rosenberger 14118 Eurasia
Viburnum plicatum Thunb., 2002 B.L. Isaac and J.A. Isaac
14371 Asia
Viburnum prunifolium L, 1997 B.L Isaac and MA Rosenberger
Viburnum rafinesquianum J A Schultes, 1 920 O.E. Jennings s.n.
Viburnum sieboldii Miq, 1 998 SA Thompson and B. Isaac and
J. Isaac and F.H. Utech 1431 9 Japan
Agrostemma githago L., 1 970 W.E Buker s.n. Europe
Arenaria serpyllifolia L var. serpyllifolia, 1997 SA Thompson
13439 Europe
Cerastium arvense L. ssp. arvense, 1888 J. McCandless s.n.
Cerastium arvense L. ssp. strictum (L) Ugborogho, B.H. Pat-
sp. vulgare (Hartman) Greuter
& Burdet, 2001 B.L Isaac and J. Isaac and C. Chuey and S.
Grund 13556 Eurasia [Cerastium vulgatum L]
Cerastium glomeratum Thuill., 1 998 K. McGowan and M.
Scanlon 192 Eurasia
Cerastium nutans Raf. var. nutans, 1 947 E. Mason s.n,
Dianthus armeria L., 1998 K. McGowan and M. Scanlon 236
1947 L.K.
Journal of the Botanical Research Institute of Texas 4(1)
Trifolium aureum Pollich, 1994 F.H. Utech and T.D. Jacobsen
and R.W. Kiger 94-2279 Eurasia
Trifolium campestre Schreb, 2001 B.L Isaac and J. Isaac and C.
Chuey and S. Grund 13584 Europe
Trifolium dubium Sibthorp, 1962 CJ. Poth s.n. Europe
Trifolium hybridum L, 2001 S.N. Costabile 3 Eurasia
Trifolium incamatum L, 1 949 LK. Henry s.n. Europe
Trifolium pratense L, 2003 C. Tracey 2003-0014 Europe
Trifolium reflexum L. 1886 J A Shafer s.n. G3G4:SX
Trifolium repens L„ 1998 K. McGowan and M. Scanlon 168
>. americana, 1 897 /
Gentiana clausa Raf., 1 932 O.E. Jennings s.n.
Gentianopsis crinita (Froel.) Ma, 1 886 J.D. Shafer s.n.
Obolaria virginica L, 1891 B.H. Patterson s.n.
Sabatia angularis (L) Pursh, 1 945 W.E. Buker s.n.
Geraniaceae
Erodium cicutarium (L) L'Her. ex Ait., 1 992 S A Thompson
10042 Europe
Geranium carolinianum L var. carolinianum, 1 998 SAThomp-
son and M. Burroughs 14408
Geranium maculatum L., 1998 K. McGowan and M. Scanlon
Vida caroliniana Walt., 1 946 W.E. Buker s.n.
Vida sativa L. ssp. sativa, 1923 J. Bright s.n. Europe
Vida villosa Roth ssp. villosa, 1 944 LK. Henry s.n. Europe
Vigna angularis ( Willd.) Ohwi & Ohashi, 1 994 SA Thompson
Castanea dentata (Marsh.) Borkh., 1 977 H. Shockey s.n.
Castanea mollissima Blume, 1 991 WA Zanol 496 Asia
Castanea pumila (L) P. Mill var. pumila, 1 920 C.W. Parker s.n.
Fagus grandi folia Ehrh., 1969 F.B. Schell s.n.
Quercus alba L, 1998 S A Thompson and B.L Isaac 14978
Quercus b/co/or Willd., 2002 B.L Isaac and JA. Isaac 14349
Quercus coccinea Muenchh., 1998 B.L Isaac and JA Isaac
11671
Quercus ilicifolia Wangenh, 1932 CM. Hepner s.n.
Quercus imbricaria Michx., 1998 S.A. Thompson and B.L
Quercus macrocarpa Michx. var. macrocarpa, 2002 B.L Isaac
and JA. Isaac 14362
Quercus muehlenbergii Engelm, 1941 O.E. Jennings s.n.
Quercus palustris Muenchh., 1998 SA. Thompson and B.L.
Quercus prinoides Willd., 1 957 J. Grom s.n.
Quercus prinus L., 1 964 P.B. Monk s.n.
Quercus rubra L, 1998 S A Thompson and B.L Isaac 14979
Quercus velutina Lam, 1 998 B.L Isaac and JA Isaac 1 1 643
Quercus x leana Nutt, (pro sp.), 1934 E.H. Graham s.n.
Quercus xruncinata (A. DC.) Engelm., 1937 H.H. Chisman son.
Fumariaceae
Adlumia fungosa (Ait) Greene ex B.S.P., 1945 W.E. Buker sji.
Capnoides sempervirens (L) Borkh, 1 900 JA Shafer s.n. [Cory-
dM. Scanlon
Geranium sanguineum L„ 2002 B.L. l<
, 1998 K. McGowan and M. Scanlon
Liquidambar styraciflua L, 1935 L.F. Allabach s.n.
Hippocastanaceae
Aesculus flava Ait., 1960 EJ. Mason s.n.
Aesculus glabra Willd. var. glabra, 2001 B.L Isaac and J
and C. Chuey and S. Grund 1 3568
Aesculus hippocastanum L, 1 991 G.H. Pollack 48 Eurasi
Hydrangea arborescens L, 2002 B.L. Isaac and J.a. is<m
14382
Philadelphus coronarius L, 1 997 SA Thompson and B.L. Isaac
14257 Eurasia
Fumaria officinalis L ssp. officinalis. 19% A.W. Cuslck 33095 WropWumappendiculatum K
Hydrophyllum canadense L, 1 992 WA Zand 1 237
Hydrophyllum virginianum L var. virginianum, 2002 B.L Is
andJA Isaac 14365
Phacelia purshii Buckl, 1937 LK. Henry s.n.
458
Phalaris arundinacea L., 1 992 WA. Zanoi 11 61
Phalaris canadensis L, 1 947 W.E. Buker s.n. Europe
Phleum pratense L., 1 998 S A Thompson 14916 Europe
Phragmites australis (Cav.) Trin. ex Steud. s:
WA Zanoi 2136
Poa alsodes A. Gray, 1 998 K. McGowan and M. Scanlon 1 97
Poa annua l„ 1998 SA Thompson and B.L. Isaac and JA
Isaac 151 14 Eurasia
Poa compressa L., 2001 B.L Isaac and J. Isaac and C. Chuey
and S.Grund 13548 Europe
Poa cuspidata Nutt, 1991 WA Zanoi 468
Poa palustris L, 1 944 O.E. Jennings s.n.
Poa pratensis L. ssp. pratensis, 1998 K. McGowan and M.
Scanlon 159
Poa sylvestris A. Gray, 1 998 K. McGowan and M. Scanlon 1 05
Poa trivialis L, 1998 K. McGowan and M. Scanlon 216 Eu-
rope
Puccinellia distans (Jacq.) Pari, 1 985 A.W. Cusick 24315
Leersia oryzoides (L) Sw, 1 944 LK. Henry s.n.
Leersia virginica Wiild, 1 998 SA Thompson and B.L Isaac and
JA Isaac 15124
Leptochloa fusca (L.) Kunth ssp. fascicularis (Lam.) N. Snow,
1 986 A.W. Cusick 25780
Lolium perenne L. ssp. multiflorum (Lam.) Husnot, 1994 WA
Zanoi 3493 Europe [Lolium multiflorum Lam.]
Lolium perenne L ssp. perenne, 2001 B.L Isaac and J. Isaac and
C. Chuey and S. Grund 1 3546 Europe
Schizachyrium scoparium (Michx.) Nash var. scoparium, 1 945
W.E. Buker s.n.
Secale cereale L, 1 884 J A Shafer s.n. Eurasia
Setaria faberi Herrm, 1994 W.A. Zanoi 3488 Eastern Asia
Setaria italica (L) Beauv, 1924 J. Bright s.n. Eurasia
Setaria pumila (Poir.) Roemer & JA Schultes ssp. pumila, 1 954
O.E. Jennings s.n. Europe [Setaria glauca (L) Beauv.]
Setaria verticillata (L) Beauv, 1979 F.H. Utech 79-335 Eurasia
Setaria viridis (L) Beauv. var. viridis, 1929 W.R. VanDersal 321
Spartina pectinata Bose ex Link, 1 945 W.E. Buker s.n.
Sphenopholis intermedia (Rybd.) Rybd, 2001 B.L. Isaac and J.
Isaac and C. Chuey and S. Grund 1 3542
Sphenopholis nitida (Biehler) Scribn, 1925 J. Bright s.n.
Sphenopholis obtusata (Michx.) Scribn, 1 928 J. Bright 466
Sporobolus compositus (Poir.) Merr. var. compositus, 1 902 B.H.
Patterson s.n. [Sporobolus asper (Beauv.) Kunth]
Sporobolus neglectus Nash, 1 895 B.H. Patterson s.n.
Sporobolus vaginiflorus (Torr. ex A. Gray) Wood var. vaginiflorus,
1998 SA Thompson and B.L Isaac and JA Isaac 15209
Tridens flavus (L) AS. Hrtchc. var. flavus, 1997 SA Thompson
and B.L. Isaac 14157
Triticum aestivum L, 1992 WA Zanoi 1 1 14 Eurasia
Vulpia octoflora (Walter) Rydb. var. glauca (Nutt.) Fern, 1890
B.H. Patterson s.n.
Phlox divaricata L. ssp. divaricata, 1 991 WA Zanoi 520
Phlox maculata L ssp. maculata, 1 923 M. Stewart s.n.
Phlox paniculata L, 2001 S.N. Costabile 20
Phlox subulata L. ssp. subulata, 1991 WA Zanoi 507
/.A. Zanoi 480
Polygala ambigua Nutt, 1 955 W.E. Buker s.n. [Polygala vertidl-
lata L. var. ambigua (Nutt) Wood]
Polygala sanguinea L, 1985 M. Macdonald and K. Stewart
LSCW146
Polygala verticillata L. var. isocycla Fern, 1909 B.H. Patterson
Polygala verticillata L
erticillata, 1939W.G. Burig s.n.
Fallopio convolvulus (L.) A. Love, 1968 N.R. Farnsworth s.n.
Europe [Polygonum convolvulus L]
Fallopio japonica (Houtt.) R. Deer, 1998 K. McGowan and
M. Scanlon 320 Japan [Polygonum cuspidatum Sieb. &
Zucc.]
Fallopiasachalinensis (F. Schmidt ex Maxim.) R. Deer, 1997 SA
Thompson and B.L Isaac and MA Rosenberger 14124 Asia
[Polygonum sachalinense F. Schimdt ex Maxim.]
Fallopio scandens (L) Holub, 1 998 SAThompson and B.L Isaac
and J.A. Isaac 15123 [Polygonum scandens L]
Persicaria amphibia (L) S.F. Gray p.p., var. emersa (MichxJ
Hickman, 1969W.E. Buker s.n. [Polygonum amphibium
var. emersum Michx.]
Persicaria arifolia (L.) Harolds, 1 919 O.E. Jennings s.n. [Polygo-
num arifolium L.] _ ___
Persicaria hydropiper (L) Opiz, 1927 W.C Grimm 225 Europe
[Polygonum hydropiper L.]
Persicaria hydropiperoides (Michx.) Small, 19120.E. Jenm
s n. [Polygonum hydropiperoides Michx.]
Persicaria lapathifolia (L) S.F.Gray, 1 990 W.A. Zanoi 400 [Potfr
num lapathifolium L] .
Persicaria longiseta <de Bruyn) Modenl <e,: !001 *
and J. Isaac and C. Chuey and S. Grund 13552 ( W
num caespitosum Blume var. longisetum (de Bruyn) AN-
Persicaria maculosa S.F. Gray, 1998 SA. Thompson 14929
Europe [Polygonum persicaria L] mdia
Persicaria orientalis (L.) Spach, 1968 W.E. Buker s.n.
[Polygonum orientate L.]
Persicaria pensylvanica (L.) Small, 1 990 WA Zand 250 VWT
num pensylvanicum L. ssp. pensylvamcum ]
Persicaria perfoliata (L.) H. Gross, 2002 B.L Isaac and
1 4333 Asia [Polygonum perfoliatum L] 1990
Persicaria posumbu ( Buch.-Ham. ex D. Don) • ' vaf-
W.A. Zanoi 269 Asia [Polygonum caespitosum
>8 SAThompson and Bi-
Persicaria punctata (Ell.) Small, 1
Morton and Speedy, Flora of Allegheny County, Pennsylvania
Graebn, 1994 F.H. Utech and T.D. Jacobsen and
Kiger 94-2289
PDtentilla recta L, 1 998 S A Thompson 14917Europe
Pirns cerasus L, 1 922 J. Bright s.n. Eurasia
Piunus mahaleb L, 1 997 F.H. Utech 97-1 62 Europe
Pmnus padus L, 1978 N.D. Richmond s.n. Europe
Piunus pensylvanica L. f. var. pensylvanica, 1926 SA Auld s.n.
Piunus persica (L.) Batsch, 1979 E.W. Wood and D.E. Boufford
and P. Singleton 421 1 Asia
jfimspumila L var. depressa (Pursh) Gleason, 1 900 JA Shafer
jp in.G5T5:S1
, 2002 B.L Isaac and JA
virginiana, 1998 SA. Thompson and
B.L Isaac 14963
TfjWcanthacoccinea M. Roemer, 1 997 SAThompson and B.L
Isaac 14167 Exotic
Posa canina L, 1 996 LL Smith and R. Hawk 44 Europe
Posa Carolina L. var. Carolina, 2003 CM. Bowers ss\.
Posa blanda Aiton var. blanda, 1884 J.A. Shafer s.n. [Rosa
subblanda Rydb.]
Posaeglanteria L„ 1 955 E. Mason s.n. Europe [Rosa rubiginosa
L]
Posa multiflora Thunb. ex Murr., 2002 M.L. Young and B.R.
Swiech 7 Asia
Posa palustris Marsh., 1 945 LK. Henry s.n.
Posa setigera Michx. var. tomentosa Torr. & A. Gray, 1913 T.E.
Kirch s.n.G5:SU
virginiana RMill. var. virginiana, 1918 B.H. Patterson s.n.
! G551
Phodotypos scandens (Thunb.) Makino, 2001 RL Isaac and J.
'»ac and C. Chuey and S. Grund 13614 Japan
Pobus aliegheniensis Porter var. allegheniensis, 2002 B.L Isaac
and JA. Isaac and S.P. Grund and McPherson 1 4352
J'txjs caesius L„ 1998 B.L Isaac and JA Isaac 11722 Europe
deamii Bailey, 1 905 O.E. Jennings s.n.
“ " .,1998 K. McGowan and M. Scanlon
jtow hispidus L., 1 944 O.E. Jennings s.n.
idaeus L. ssp. strigosus (Michx.) Focke, 1 925 CK. Henlen
393 Exotic [Rubus strigosus Michx.]
ladniatusW\Wd, 1 997 SAThompson and B.L Isaac and
MA Rosenberger 14106 Europe
T" 'wdatus Berger, 1 944 LK. Henry s.n.
: wuspensilvanicus Poir., 2002 B.L. Isaac and JA Isaac and S.P.
? Gfund and McPherson 14351
wous pergratus Blanch., 1998 K. McGowan and M. Scanlon
■ 274 [Rubus orarius Blanch.]
*™sPhoenicolasius Maxim, 1979 D.E Boufford 21 187 Asia
Rubus occidentals L, 1 998 K. McGowan and M. Scanlon 286
Rubus rosa Bailey, 1 918 O.E. Jennings s.n.
Sanguisorba canadensis L, 1 91 2 J. Bright s.n.
Sorbaria sorbifolia (L.) A. Braun, 1 901 JA Shafer s.n. Asia
Spiraea alba Du Roi var. alba, 1 945 LK. Henry s.n.
Spiraea alba Du Roi var. latifolia (Ait.) Dippel, 1 91 3 J. Bright s.n.
[Spiraea latifolia (Aiton) Borkh.]
Spiraea prunifolia Sieb. & Zucc, 1 946 W.E. Buker s.n. Asia
Spiraea tomentosa L, 1990 WA. Zanol 206
Waldsteinia fragarioides (Michx.) Tratt. ssp. fragarioides, 1991
WA. Zanol 528
Cephalanthus occidentals L, 1 996 S. Grund 1 674
Diodia teres Walt, 1 982 W.W. Thomas and W.E. Buker 2975
Galium aparine L, 1 998 K. McGowan and M. Scanlon 1 84
Galium asprellum Michx, 1 944 LK. Henry s.n.
Galium boreale L, 1921 O.E. Jennings s.n.
Galium circaezans Michx, 1 944 L.K. Henry s.n.
Galium concinnum Torr. & A. Gray, 1 991 WA Zanol 81 8
Galium lanceolatum Torr, 1 932 C.M. Hepner s.n.
Galium mollugo L, 1931 J. Bright 5404 Eurasia
Galium obtusum Bigelow ssp. obtusum, 1902 B.H. Patterson
7 B.H. Patterson s.n.
Galium pilosum Ait, 1937 F.R. Fosberg 14320
Galium tinctorium (L) Scop, 1932 CM. Hepner s.n.
Galium triflorum Michx, 1 998 SA. Thompson and B.L Isaac
14966
Galium verum L var. verum, 1 961 V. Phelps s.n. Eurasia
Houstonia caerulea L, 1998 SAThompson and B. Isaac and
Houstonia canadensis Willd. ex Roemer & JA Schultes, 1950
W.E. Buker s.n.
Houstonia longifolia Gaertn, 1946 W.E. Buker s.n.
Mitchella repens L, 1 889 M.T. Shafer s.n.
Sherardia arvensis L, 1949 LK. Henry s.n. Eurasia
Ptelea trifbliata L. ssp. trifoliata var. trifoliata, 2002 B.L Isaac and
J.A. Isaac 14348 G5:S2
Zanthoxylum americanum P. Mill, 1900 J A Shafer s.n.
Populus alba L, 1979 E.W. Wood and D.E. Boufford and L.
Schram 4240 Eurasia
Populus deltoides Bartr. ex Marsh, ssp. deltoides, 2001 B.L Isaac
and J. Isaac and C Chuey and S. Grund 1 3610
Populus grandidentata Michx, 1999 B.L Isaac and T. Herman
11781
Populus nigra L, 1966 P. Monk s.n. Eurasia
Populus tremuloides Michx, 1 997 SAThompson and B.L Isaac
and M. Sally 13480
Populus xjackiiSatg., 1940 O.E. Jennings s.n.
Salixalba L, 1991 WA. Zanol 454 Eurasia
Salix bebbiarta Sarg, 1 924 J. Bright s.n.
Salix caroliniana Michx, 1 908 O.E. Jennings sj>. G5S1
Salix discolor Muhl, 1 992 WA Zanol 973
Salix eriocephala Michx, 2002 S.P. Grund and M.L. Speedy
Morton and Speedy, Flora of Allegheny County, Pennsylvania
Sotiychium matricariifolium (A. Braun ex Dowell) A. Braun,
I 1982 E.G. Allen s.n.
f totrypusvirginianus (L.) Holub, 1964W.D. Loub WP-880 [Botry-
; chium virginianum (L) SwJ
Sceptridium multifidum (Gmel.) Nishida exTagawa, 1933 L.
| Lincoln s.n. [Botrychium multifidum (Gmel.) Trev.]
■ Saptridium oneidense (Gilbert) Holub, 1 937 O.E Jennings s.n.
R Sotiychium oneidense (Gilbert) House]
Osmunda cinnamomea L, 1991 G.H. Pollack 47
Osmunda claytoniana L., 1955 L.K. Henry s.n.
Osmunda regalis L. var. spectabilis (Willd.) A. Gray, 1946 W.E.
Polypodium virginianum L, 1 945 W.E. Buker s.n.
Pteridaceae
Adiantum aleuticum (Rupr.) Paris, 1 998 B.L Isaac and J.A. Isaac
Phegopteris hexagonoptera (Michx.) Fee, 1998 B.L. Isaac
and JA Isaac 11668 [Thelypteris hexagonoptera (Michx.)
Weatherby]
Thelypteris noveboracensis (L) Nieuwl., 1998 B.L. Isaac and
J. A. Isaac 1 1670
Thelypteris palustris Schott var. pubescens (Lawson) Fern., 1 922
O.E Jennings s.n.
Polypodium appalachianum Haufler & Windhar
1940 R.W.
ACKNOWLEDGMENTS
Our thanks go to the Wild Resource Conservation Fund of the Pennsylvania Department of Conservation
and Natural Resources for partial funding of the field work. We would also like to thank Venice Bayrd, Jerry
G. Chmielewski, and one anonymous reviewer for their helpful comments. We are also thankful to Amanda
Jutetic for data entry.
REFERENCES
BfiUMMrrr, R.K. and C.E. Powell 1 992. Authors of plant names. Royal Botanic Gardens, Kew.
°CNR. 2004. Invasive plants of Pennsylvania, httpy/www.dcnr.state.pa.us/forestry/wildplant/invasive.aspx
Pennsylvania Department of Agriculture. 2007. Pennsylvania noxious weed control list. httpy/www.agriculture.state.
. P4us/agricu Itu re/cwp/view.asp?q= 1 27683
Pennsylvania DEP Watershed Notebook. 2006 httpy/www.dep.state.pa.us/dep/ deputate/ watermgt/WC
Aean Enterprise. 2007. History of Allegheny County, Pennsylvania. httpyMww.alleghenycounty.net
^ARTE5Z, JT. 1 999. A synonomized checklist and atlas with biological attributes for the vascular flora of the United
States, Canada and Greenland. First edition. In: Kartesz, J.T. and CA Meacham. Synthesis of the North American
flora, Version 1.0. North Carolina Botanical Garden, Chapel Hill.
*'HQADS> A.F. and W.M. Klein, Jr 1 993. The vascular flora of Pennsylvania: annotated checklist and atlas. American
Philosophical Society, Philadelphia, Pennsylvania.
Rhoads, A.F. and T.A. Block 2000. The plants of Pennsylvania. University of Pennsylvania Press, Philadelphia,
Pennsylvania.
The Nature Conservancy (1996 version). Element Ranking List - Pennsylvania Natural Heritage Program.httpy/
^w.naturalheritage.state.pa.us/PlantsPage.aspx
^ Census Bureau. 2009. State and county quickfacts. httpy/quickfacts.census.gov/qfd/states/42/42125.html
US°A. 1981. Soil survey of Allegheny County, Pennsylvania. USDA Soil Conservation Service. US Government
Printing Office, Washington, D.C.
Vncent' M A 1991. Trifolium reflexum L. (Buffalo Clover: Leguminosae) in Ohio, its history and present status.
1/4" x 8 3/4".
NEW VASCULAR PLANT RECORDS FOR SOUTH DAKOTA
Gary E. Larson
CA Taylor Herbarium
Department of Biology & Microbiology
South Dakota State University
Box 2207B
Brookings, South Dakota 57007-0595, U.SA
ABSTRACT
RESUMEN
Field and herbarium research conducted at South Dakota State University have led to the discovery of twelve
vascular plant taxa not previously recorded for the state. These are species omitted from prior treatments
covering the South Dakota flora either entirely or in part, including Saunders (1899), Over (1932), Rydberg
(1932), Dorn (1977), Great Plains Flora Association (1986), and Van Bruggen (1996). Publication of these
new records serves to update distributional accounts of these species, such as those provided by the USDA
NRCS PLANTS Database (http://plants.usda.gov/), and is intended to augment general knowledge of South
Dakota’s vascular flora. Voucher specimens documenting these findings are deposited in the C. A. Taylor
Herbarium at South Dakota State University (SDC), with duplicates traded to other regional herbaria, in-
cluding North Dakota State University (NDC), the Rocky Mountain Herbarium (RM), and the University of
Kansas (KANU). The newly discovered taxa for South Dakota and their documentation are as follows:
Vra interrupta (L.) P. Beauv. (Poaceae), dense silkybent. Apera interrupta is an annual grass introduced
from Europe that typically occurs in disturbed habitats. It is naturalized in all states west of the Great Plains
*nd in British Columbia (USDA NRCS 1999) and has otherwise been recorded at scattered locations in eastern
North America. Barkworth et al. (2007) describe the grass as “a weed in lawns, grain fields (especially winter
wheat), sandy open ground, and roadsides.” The South Dakota record is likely attributable to introduction
K a contaminant in grass seed since it occurred in a recently planted area of the abandoned Gilt Edge Gold
Mine in the northern Black Hills. The Black Hills State University Herbarium Database (2009) includes two
fccords of this species from Crook Co., WY, and thus a South Dakota occurrence is not unexpected and
presents an eastward extension of its range.
v°ucher specimen: SOUTH DAKOTA. Lawrence Co.: 44.33°, -103.66°, uncommon on east-facing slope of waste rock repository cap,
“C0nd swson after seeding, 28 Jun 2005, Gary E. Larson 11268.
'z°Ha mexicana Presl (Azollaceae), Mexican mosquitofem. This is a small, surface-floating aquatic fern
7* Propagates vegetatively to form floating mats much like common duckweed. This species is otherwise
n°wn from Nebraska and Minnesota, and so this record represents a northward and westward extension
j k range. It may be introduced into our range on a seasonal basis by migrating waterfowl.
Journal of the Botanical Research Institute ofTexas4(1)
Carex leavenwor thii Dewey (Cyperaceae), Leavenworth’s sedge. According to Kaul et al. (2006), C. leaven-
worthii has been recorded for Cedar County, Nebraska, across the Missouri River from Clay County, South
Dakota, and so this occurrence represents a minor northward extension of its range.
Voucher specimen: SOUTH DAKOTA. Clay Co.: 42.77°, - 97.12°; Myron Grove, old floodplain forest remnant on Missouri River
floodplain, shaded understory; common ( Gary E. Larson 11300, 19 Jun 2008).
Carex pseudocyperus L. (Cyperaceae), cypresslike sedge. Apparent hybrids with bottlebrush sedge (C.
hystericina Muhl. ex Willd.) were also noted at the collection site. The South Dakota record represents a
southward extension of the previously documented range of C. pseudocyperus.
Galium circaezans Michx. var. hypomalacum Fern. (Rubiaceae), woods bedstraw. This finding is a lo-
cal northward range extension from northern Nebraska where the species is found in woodlands along the
Niobrara River and its tributaries (Kaul et al. 2006).
Eriochloa villosa (Thunb.) Kunth (Poaceae), hairy cupgrass. This introduced agricultural weed from East
Asia has spread into eastern South Dakota from farther east in the Corn Belt and has become more com-
mon and widespread since the original 1983 collection. Northward and westward expansion in the state
is expected.
Voucher specimens: SOUTH DAKOTA. Deuel Co.: 5 mi NW of Toronto; weed in cultivated field, August 1983, LeonJ. Wrage s.n. Min-
nehaha Co: 3 mi E, 2 mi N of Garretson in field, 14 July 1986, Mark Peterson 96. Brookings Co.: South Dakota State Untv. campus.
Glaux maritima L. (Primulaceae), sea milkwort. Sea milkwort is otherwise frequent in North Dakota and
scattered in western Nebraska where it favors alkaline seeps.
Voucher specimen: SOUTH DAKOTA. Perkins Co.: Grand River National Grassland; T 23N, R 11E, Sec. 35, SW ‘A; sandstone bank
of North Fork of Grand River, sandy textured soil; uncommon, 14 Jun 2002, Christopher Kopp 207.
Lechea stricta Leggett ex Britton (Cistaceae), prairie pinweed. Prairie pinweed is a psammophilous spe-
cies found widely scattered in the eastern Nebraska Sand Hills (Kaul et al. 2006) and in Sheyenne National
Grasslands of southeastern North Dakota and Bowman County of southwestern North Dakota (Great Plains
Flora Association 1977).
W lu*i spo. nr.< n Mil III Marshall Co.: I 12*N R V)\V v, t> <E '•* model. uelv toh..nih «•!
and T 128N, R 59W, Sec. 19, SE %; moderately to lightly grazed sand hills prairie; dry sandy sites, temporarily wet in spring, uncom
22 Aug 1996, Gary E. Larson 11767 &■ Eric Fairlee.
Lipocarpha micrantha (Vahl) G. Tucker (Cyperaceae), smallflower halfchaff sedge. Upocarpha micrantha
(formerly Hemicarpha m.) is otherwise known from eastern and central Nebraska and with disjunct records
from Cass County, ND and Becker County, MN (Larson 1993).
sand hills and wet depressions; margin of moist depression, in cow path with Cyperus sauarrosus; occasional, 16 Aug 1996, Gary E 10**
11702 &■ Eric Fairlee, and T 127N, R 61W, Sec. 14. SW *; level to rolling non-nVtive pasture on loamy fine sand. Ughtly grazed; localK
common in moist or drying soil in shallow depressions, usually where grazed, 22 Jul 1997, Eric Fairlee 196 & Stacy Sererding
Monotropa uniflora L. (Monotropaceae), Indianpipe. This nongreen saprophyte has been otherwise fou
i (1977) ir
PENSTEMON OKLAHOMENSIS (SCROPHULARIACEAE) IN TEXAS
Jeffrey N. Mink
Department of Biology
Baylor University
Waco, Texas 76798-7388, USA
Jeffrey_Mink@baylor.edu
Jason R. Singhurst
Wildlife Diversity Program
Texas Parks and Wildlife Department
4200 Smith School Road
Austin, Texas 78704, USA
Walter C. Holmes
Department of Biology
Baylor University
Waco, Texas 76798-7388, USA
ABSTRACT
> the flora of Texas. Included is a key to discriminate the white-flowered Penstemon
RESUMEN
tostemon Mitchell (Scrophulariaceae) is the most widespread genus of Scrophulariaceae (sensu lato) in North
America, with the largest number of documented species (Fernald 1950). At least 270 species (Lodewick &
J-afewick 1999; Nold 1999) have been described in North America distributed from Alaska to Guatemala,
■jeanan (1981) recognized 22 species in the Great Plains, while Clinebeil and Bernhardt (1998) suggested
Jtono fewer than 24 species of Penstemon are native to the midwestern American prairies from the upper
issippi Valley southward. Twenty-two species of Penstemon are currently recognized in Texas by Turner
ttaL (2003), with 17 species limited in distribution to the central and western portion of the state.
Recent botanical study in northeast Texas has resulted in the discovery of Penstemon oklahomensis Pen-
wkidl is now documented as new to the state.
Mink &J.R. Singfrurst 14514 (BAYLU).
Gambill Goose Refuge is an area of tallgrass prairie that contains mima r
jat and zonation components of alfisol blackland prairies known to increa
ns et al. 1975). Characteristic non-woody vegetation of the refuge included Schizachyrium scoparium,
fo*erWn’ TriPsacum dactyloides, Tradescantia hirsutiflora, Aphanostephus skirrhobasis, Asclepias oenotheroides,
“’’Pfoia bracteata, Brickellia eupatoriodes var. texana, Castilleja purpurea var. purpurea. Coreopsis tinctoria, Croton
^.atUS’ pycnostachya, Marshallia caespitosa, Penstemon digitalis, P. laxiflorus, Psoralidium tenuiflorum,
umilis, Salvia azurea, and Tephrosia virginiana. The most abundant woody plant in this prairie was
americanus, while others, such as Berchemia scandens, Comus drummondii, Diospyros virginiana,
Serotina, Juniperus virginiana, Rhus copallina and Zanthoxylum clava-herculis were principally limited
areas. Other rare or unusual species were Calapogon oklahomensis, Manfreda virginica ssp. lata
** Anthem lacera.
jR^stemon oklahomensis is well-marked by a white corolla with a constricted mouth. The species was
ered to be endemic to Oklahoma (Pennell 1935; McGregor &r Barkley 1977; McGregor & Barkley 1986).
32 J|fareSt 0klahoma record of collection is southeast of Bokchito, Bryan County, which is approximately
1 es northeast of the Lamar County, Texas, record.
472
Journal of the Botanical Research Institute of Texas 4(1)
Three other species of Penstemon in northeast Texas have white flowers: R digitalis Nutt, ex Sims, P.
laxiflorus Pennell, and P. tubaeflorus Nutt. The following key, adapted from Pennell (1935), Correll and
Johnston (1970), and Diggs et al. (1999), may be used to distinguish among the four species.
KEY TO THE WHITE-FLOWERED PENSTEMON OF NORTHEAST TEXAS
1 . Floor of corolla lined internally with reddish or purplish nectar guide streaks.
2. Corolla 20-25 mm long; nectar guides with pleats evident; staminode prominently exserted and densely
bearded with yellow-orange hair Penstemon laxiflorus
2. Corolla 23-30mm long; nectar guides prominently rounded and ventricose; staminode included and
lightly bearded with whitish or yellow hair Penstemon digitalis
1. Floor of corolla unlined internally.
3. Corolla 1 7-23 mm long; tube funnelform to trumpet-shaped from base, the lobes spreading to form an
open and prominent orifice; staminode included, lightly bearded with whitish or yellow hair Penstemon
tubaeflorus
3. Corolla 25-30mm long; tube narrow at base, abruptly enlarged to the closed orifice (anterior three-lobes
upcurve tightly against the united posterior two-lobed lip so as to close orifice of the corolla throat);
staminode exserted, densely bearded with deep yellow hairs Penstemon oklahomensis
REFERENCES
Coluns, O.B. 1975. Range vegetation and mima mounds in north Texas, J. Range Managem. 28:209-21 1.
Collins, O.B., F.E. Smeins, and D.H. Riskind. 1 975. Plant communities of the Blackland Prairie ofTexas. In: M.K. Wali, ed.
Prairie: a multiple view. University of North Dakota Press. Pp 75-88.
Correll D.C. and M.C. Johnston. 1970. Manual of the vascular plants ofTexas. Texas Research Foundation, Renner,
Texas.
Cunebell R.R. II and P. Bernhardt. 1 998. The pollination ecology of five species of Penstemon (Scrophulariaceae) in
the tallgrass prairie. Ann. Missouri Bot. Gard. 85:1 26-1 36.
Diggs, G.M. Jr, B.L. Upscomb, and RJ. O'Kennon. 1999. Shinners & Mahler's illustrated flora of north central Texas.
Sida, Bot. Misc. 16: Botanical Research Institute ofTexas, Fort Worth, TX.
Fernald, M.L. 1 950. Gray's manual of botany, 8th edition. American Book Company, NY.
Freeman, C.C. 1981. A biosystematic study of the genus Penstemon (Scrophulariaceae) in the Great Plains. Un-
published Master's Thesis, Kansas State University, Manhattan.
Lodewick, K. and R. Lodewick. 1999. Key to the genus Penstemon and its related genera in the tribe Cheloneae
(Scrophulariaceae). Published by the authors, Eugene, OR
McGregor R.L (Coord.) and T. M. Barkley (ed.). 1977. Atlas of the flora of the Great Plains. Iowa State University
Press, Ames.
McGregor RL (Coord.) and T. M. Barkley (ed.). 1986. Flora of the Great Plains. University Press of Kansas, Law-
Nold, R. 1999. Penstemons. Timber Press, Portland, OR.
Pennell F.W. 1935. Scrophulariaceae of eastern temperate North America. Acad. Nat. Sci. Philadelphia Monogr.
1, Wickersham Printing Co, Lancaster, PA.
Turner B.L., H. Nichols, G. Denny, and O. Doron. 2003. Atlas of the vascular plants ofTexas, Vol. I. Sida, Bot. Misc 24:
Botanical Research Institute ofTexas, Fort Worth.
ANNOTATED CHECKLIST OF THE VASCULAR FLORA OF THE JACK GORE
BAYGALL AND NECHES BOTTOM UNITS OF THE BIG THICKET NATIONAL
PRESERVE, TYLER, JASPER, AND HARDIN COUNTIES, TEXAS
Larry E. Brown
Spring Branch Science Center Herbarium
8856 Westview Drive
Houston, Texas 77055, USA
larry-theplantman@att.net
Warren W. Pruess & I. Sandra Elsik
f Department of Ecology & Evolutionary Biology
Rice University
Houston, Texas 77005, USA
Barbara R. MacRoberts &
Michael H. MacRoberts
Bog Research, 740 Columbia
Shreveport, Louisiana 71 104, USA
and Herbarium, Museum of Life Sciences
Louisiana State University in Shreveport
Shreveport, Louisiana 71115, USA
Stanley Jones
Botanical Research Center Herbarium
P.O. Box 6717
Bryan, Texas 77805, U SA
ABSTRACT
RESUMEN
midades (de estas solamente hay 734 especies, esto debido
les (12,1%) son exdticas.
INTRODUCTION
Jtos is the 8th part of an inventory, based on available voucher specimens, of the vascular flora of the Big
d* £t Nati0nal Preserve in southeastern Texas. In previous papers, we provided annotated checklists of
Btt Hickory Creek, Turkey Creek, Big Sandy Creek, Lance Rosier, Beech Creek, Loblolly, and Menard Creek
nits (MacRoberts et al. 2002; Brown et al. 2005, 2006a, 2006b, 2008a, 2008b, 2009). In this paper, we
Provide an annotated checklist of the Jack Gore Baygall and Neches Bottom Units since they are continuous
occupy the same habitat.
SITE AND METHODS
^ Big Thicket National Reserve is located in the West Gulf Coastal Plain in southeastern Texas (see Diggs
a • 2006) for literature and description). It consists of units ranging from 222 to 10,000 hectares and of
_^0,100 hectares scattered over seven counties (Fig. 1). The Jack Gore Baygall and Neches Bottom Units
le the Neches River in Tyler, Jasper, and Hardin counties and total 5,379 hectares (Peacock 1994).
Mude extends from 30°47.933'N at the B.A. Steinhagen Lake dam at Town Bluff south to 30°15.055'N at
ure with the Beaumont Unit. Longitude extends from 94°10.723' at Highway 2937 in Hardin County
‘°94°03.660 at Sally Withers Lake in Jasper County. These units are located on Deweyville, Lissie, and
^ern alluvial soil series (Deshotels 1978; Aranow 1981). The Jack Gore Baygall and Neches Bottom are
^8 the most diverse units in the Preserve in that they contain upper-slope, mid-slope, and lower-slope
as well as floodplain hardwood forest, wetland baygall, and cypress-tupelo swamp (Harcombe &
Ir*5 1979; Marks & Harcombe 1981).
474
Elevation ranges from a high of 33 meters in the small portion of the Jack Gore Baygall Unit adjacent
to Highway 2937 to 3 meters at the junction of the Neches River Corridor Unit with the Beaumont Unit
Like all units of the Big Thicket Biological Preserve, excluding the recent floristic inventories of the
above cited units, the Jack Gore Baygall and Neches Bottom units had not previously been the subject o a
detailed floristic inventory.
Because our purpose is to produce a list of taxa know to occur in the Jack Gore Baygall and Neches Bot
tom units, a vouchered specimen was considered to be the only evidence acceptable for inclusion in t e
In all, 1304 + herbarium specimens form the data for this report. Larry Brown annotated all specimens.
Larry Brown collected 716 specimens during 25 field days beginning in June 2008 and ending t®
November 2009. Stanley Jones collected 338 specimens in the mid 2000s. Geraldine Watson collected 5
Brown etal., Flora of the Big Thicket National Preserve
specimens over several years in the late 1970s and early 1980s. The MacRoberts collected 31specimens.
Sandra Elsik and Warren Pruess together collected 105 specimens and Warren Pruess singly collected 57
specimens. Collection totals for the few others are 12. All collectors and collection numbers are designated
by appropriate abbreviations below.
In most cases, the current family nomenclature follows Chester et al. (2009) and A.S. Weakley (2007).
Species nomenclature for the most part follows Jones et al. (1997), Kartesz (1999), and Diggs et al. (2006).
Incases of multiple collections, no more than four are listed for each taxon.
RESULTS
The following is an annotated list of the Jack Gore Baygall Unit and Neches Bottom Units taxa. Abbrevia-
tions used are given below:
Acanthaceae
Justiciaovata (Walter) Lindau var. lanceolata (Chapm.) R.W.
Long, WP 144
faellia caroliniensis (J.F. Gmel.) Steud., SJ 14204; LB 34575
mdliahumilis Nutt., GW 2953; LBRH 34645
pedunculata Torr. ex A. Gray, LB 34606
Aceraceae (see Sapindaceae)
Adoxaceae
ni9ra L subsp. canadensis (L) Bollii, SJ 14306
¥^rnum dentatum L„ SJ 14132; LBRH 33905
tmimum nudum L. var. nudum, SEWP 460; LB 35106
+*Alternanthera sessilis (L) R. Br. ex DC, LB 343 1 0
Froelichia floridana SJ 1441 1
Dysphanis ambrostoides (L) Mosyakin & Clemants, SJ 14516;
LB 35034
Rhus copallinum L, SEWP 222
Toxicodendron pubescens Mill, LB 34595
Toxicodendron radicans (L) Kuntze, WP 142; SJ 14621
Toxicodendron vemix (L.) Kuntze, LBJL 34036, 34031
Asiminaparviftora (Michx.) Dunal, GW 2948; LBRH 33900; LBSH
34191; LB 34610
Agavaceae
f buisianensis Trel, LB 34235
Afemataceae
Minodorus cordifolius (L) Griseb, SJ 14288; LBSH 34723
Bptotfe htifoiia Willd, LBSH 34184
*Qttariapapillosa Buchenau, WP 105
tyittaria platyphylla (Engelm.) J.G. Smith, LBJL 34018
AHiaceae
B^conodense L. var. mobilense (Regel) Ownbey, LB
? othoscordum bivalve (L.) Britton, SJ 14429
Jfcngiaceae
M^nbarstyraciflua L, LBSM 33728
(Mart) Griseb, LBSH 34104
Apiaceae (see Araliai
Centella erecta (Lf.) Fernald, MM 8012; SEWP 1%
Chaerophyllum tainturieri Hook. var. dasycarpum Hook, ex !
Watson, LBSH 34393 (J)
*Cyclospermum leptophyllum (Pers.) Sprague ex Britton &
Wilson, WP 131
Eryngium prostratum Nutt ex DC, SEWP 206; LBSH 3471 3
Limnosciadium pumilum (Engelm. & A. Gray) Mathias ,
Constance, LB 34577
Ptilimnium capillaceum (Michx.) Raf„SJ 14194a, 14134, 1419-
LB 34589
Sanicula canadensis L, WP 1 36; LBSH 33826
•Torilis nodosa (L) Gaertn, LB 34520
Apocynaceae
Asclepias ampkxicaulis Michx, SJ 1 43 1 1
Asclepias obovata Elliott, SEWP 1 88; GW 2960
476
Asclepias perennis Walter, SJ 14337; LBSH 34954
+Gonolobussuberosus (L) R.Br, var. granulatus (Scheele) Krings
&0.Y. Xiang SJ 14146, 14305; LB 34529, 34986
Trachelospermum difforme (Walter) A. Gray, SJ 14340; LB
35088
Aquifoliaceae
Ilex ambigua (Michx.) Torr., SJ 14436, 14422, 14618; LBSH
34942
Ilex coriacea (Pursh) Chapm, WP 1 61 ; LB 35 1 04
Ilex decidua Walter, LBSM 33720: LB 34257
llexlongipes Chapm. exTrelease, LBRH 33893
Ilex opaca Alton, WP 1 56; LB 35075
Ilex vomitoria Aiton, SJ 1 4502
Arisaema triphyllum (L) Schott, GW 717
Araliaceae
Aralia spinosa L, SJ 14421 ; LBJL 34028
Hydrocotyle umbellata L, SJ 14366; LB 3421 1 , 3431 (
Hydrocotyle verticillataThunb., LB 34527
Aristolochiaceae
Aristolochia reticulata Jacq, SEWP 1 89
Aristolochia serpentaria L„ LB 19218; SJ 14341
Asdepiadaceae (see Apocynaceae)
Asplenium platyneuron (L.) Britton, Sterns & Poggenb., SJ
3217
Asteraceae
Acmella repens (Walter) Rich., SJ 1 431 9; LB 34975
Ambrosia artemisiifolia L, MM 8038; LB 34237
Ambrosia psilostachya DC, LBJL 34010
Ambrosia triflda L„ LB 3421 7
Artemisia ludoviciana Nutt., GW nl 8
Baccharis halimifolia L, GW 3297; LBJL 3401 5
Berlandiera pumila (Michx.) Nutt. var. scabrella Nesom & B.L
Turner, SJ 14190a; LBSH 33830; LB 34597
+Bidens aristosa (Michx.) Britton, SJ 14376; LBRH 34060;
LB35071, 35184
Bidens bipinnata L, LBJL 34014
Bidens laevis (L) Britton, Sterns & Poggenb, LBSH 34097
Boltonia diffusa Elliott, SEWP 191; LB 35068
Bradburia pilosa (Nutt.) Semple, SJ 14418; LB 35019
Chrysopsis mariana (L) Elliott, SEWP 295; LBRH 33888; LB
34224
Cirsium horridulum Michx, WP 080
Conoclinium coelestinum (L) DC, MM 8024; SJ 14321
Conyza bonariensis (L.) Cronquist LB 341 59
Conyza canadensis (L) Cronquist var. canadensis, SEWP 21 3;
LBJL 34027
Conyza canadensis (L) Cronquist var. pusilla (Nutt.) Cronquist
SJ 14414
Coreopsis tinctoria Nutt, SJ 14228
Croptilon divaricatum (Nutt.) Raf, SEWP 390; LBRH 33899
Croptilon rigidifolium (EB. Smith) E.B. Smith, LBSH 34951
Echinacea sanguinea Nutt, SJ 14210
Eclipta prostrata (L) L, SEWP 311; LBSM
Elephantopus carolinianus Raeusch, MM 8021
Elephantopus nudatus A. Gray, LBSH 341 82a; LB 35103
Elephantopus tomentosus L, MM 801 1 ; SJ 14420; LBRH 33885;
LBJL 34040
Erechtites hieraciifolia (L) Raf, MM 8070
Erigeron annuus (L.) C.P. Person, LB 34988
Erigeron philadelphicus L, WP 1 22; LB 34294
Erigeron strigosus Muhl. ex Willd, SJ 14205; LBJL 34001
Erigeron tenuis Torr. & A. Gray, WP 081
Eupatorium capillifolium (Lam.) Small, LBRH 33884
Eupatorium lancifolium (Torr. & A. Gray) Small, SEWP 220; LBSH
33828; LB 35004
Eupatorium perfoliatum L, LBSM 33765
Eupatorium x pinnatifidum Elliott, WRC 12489; LBRH 34051;
LB 35158
Eupatorium rotundifolium L. var. rotundifolium, MM 8017; SJ
14500; LBSH 33827
Eupatorium semiserratum DC, LBSM 33757a; LBRH 33909; LBJL
34023; LBSH 34188
Eupatorium serotinum Michx, LBSM 33718a; LB 34217a
Euthamia leptocephala (Torr. & A. Gray) Greene, LB 34519
* Facelis retusa (Lam.) Sch.-Bip, LB 3461 5
Gaillardia pulchella Foug, SEWP 197; LB 35100
Gamochaeta antillana (Urb.) Anderb, LB 34306
•Gamochaeta coarctata (Willd.) Kerguelen, SEWP 456
) Cabrera, SJ 1
H 34411;
LB 34540
Gamochaeta pensylvanica (Willd.) Cabrera, LB 34306a
Helenium amarum (Raf.) Rock var. amarum, WP 1 60
Helenium flexuosum Raf, LB 34619
Helianthus angustifolius L, SJ 14378; LB 34220
Heterotheca subaxillaris (Lam.) Britton & Rusby, LBSH 34086
Hieracium gronovii L, GW 2954; LBRH 33906
Hymenopappus artemisiifolius DC, SJ 1 4281
*Hypochaeris microcephala (Sch. Bip.) Cabrera, LB 34546
+lva asperifolia Less. var. angustifolia (Nutt, ex DC.) B.L. Turner,
SEWP 468; LBJL 33997
Iva annua L, SEWP 467; LBJL 34009
Krigia cespitosa (Raf.) K.L Chambers var. cespitosa, LB 34550
5 125
Krigia wrightii (A. Gray) K.L Chambers e>
LBSH 34396 (J)
Lactuca canadensis L„ LBRH 33891
Lactuca floridana (L.) Gaertn, SEWP 309
Uatris acidota Engelm. & A. Gray, SJ 14396
i, LB 34314;
s (Wall
elegans, SEWP 280; LB
35005
:. var. glabrata (Rydb.) Gaiser, GW
3507, 2975; LBSH 34928, LB 34993
+Liatris squarrosa (L) Michx. var. squamosa, LBSH 34928; LB
34993, 35006
Mikania scandens (L) Willd. LBSM 33741
Packera obovata (Michx. ex Willd.) Weber & A. Love. GW 3323,
LB 34278
Pityopsis graminifolia (Michx.) Nutt. SEWP 1 93; LBSH 34180
Pluchea camphorata (L) DC, LBSM 33762
Pluchea foetida (L) DC, MM 8030; SJ 14384; LBJL 34033
133764
Journal of the Botanical Research Institute ofTexas 4(1)
Ipomoea lacunosa L, LB 35090, 35113
Ipomoea pandurata (L) G. Mey., LBSH 34943
Jacquemontia tamnifolia (L) Griseb., SEWP 304
Cornaceae (see Nyssaceae for the genus Nyssa)
Cornus florida L, SEWP 3005; LB 34289; WP 083 (SBSC)
Comus fbemina Mill., LB 34231
Crassulaceae (see Penthoraceae for the genus Penthorum )
Cayaponia quinqueloba (Raf.) Shinners, SJ 14522; LBSH
34710
•Citrullus vulgaris Schrad. ex Eckl. & Zeyh, LB 35161
Xucumis melo L., LB 351 14
Melothria pendula L„ LBSM 33736
Cupressaceae
Juniperus virginiana L var. virginiana, LBJL 34005
Taxodium distichum (L) Rich. var. distichum, LBSH 33845
Cuscutaceae (see Convolvulaceae for the genus Cuscuta)
Cyperaceae
Bulbostylis barbata (Rottb.) Clarke, SJ 14428
Bulbostylis ciliatifolia (Elliott.) Fernald var. coarctata (Elliott) Krai,
SJ 14129, 14469; LBRH 33902; LB 35036
Carexabscondita Mack., SJ 14127; LB 34579; LBSH 34648 (J)
Carex atlantica L.H. Bailey subsp. capillacea (L H. Bailey)
Reznicek, LBJL 33995
Carex brevior (Dewey) Mack, ex Lunell, LBSH 34395 (J); LB
34528
Carex cherokeensis Schwein., LB 34978
Carex debilis Michx., LB 34545
Carex digitalis Schkuhr ex Willd. var. floridana (L.H. Bailey) Naczi
& Bryson, SJ 14595
Carex flaccosperma Dewey, LBSH 34394 (J); LB 34549
Carex frankii Kunth, LB 34581
Carex glaucescens Elliott, LBSH 34037
Carex porii L H. Bailey, NIX 15988; SJ 14473; LBSM 33756
Carex leptalea Wahlenb. subsp. harperi (Fernald.) W. Stone, SJ
14145; LBJL 33996
Carex nigromarginata Schwein var. floridana (Schwein.) Kuk,
SJ 13708; LB 34283
Carex reniformis (L.H. Bailey) Small, LB 34536, 34586
Carex tribuloides Wahlenb. var. sangamonensis Clokey, SJ
Carex verrucosa Muhl, SEWP 376, 375
Cladiumjamaicense Crantz, LBSH 34667
Cyperus acuminatus Torn & Hook, ex Torn, LB34602
Cyperuscompressus L, SJ 14486
Cyperus croceus Vahl, SJ 14487, 145124; LBJL 340342; LBRH
34637
Cyperus elegans L, SEWP 463; LB 35101
* Cyperus entrerianus Bockeler, LBSM 33743
Cyperus erythrorhizos Muhl., SJ 14440, 14457; LB 35085,
35089
•Cyperus esculentus L, SJ 14441; LBSH 34163
Cyperus flavescens L., LBRH 34059
Cyperus grayoides Mohlenbr., SJ 1 4479
Cyperus haspan L, SEWP 369; LBSM 33760
Cyperus hystricinus Fernald, SJ 14471, 14491; LB 34221
•Cyperus iria L., SJ 1 4494
Cyperus odoratusV ahl, SJ 14442; LBSH 34078, 34163a, 34970
Cyperus polystachyos Rottb, SJ 14443; LBSH 34162
Cyperus pseudovegetus Steud, SJ 14344; LBSM 33748; LBSH
Cyperus retroflexus Buckley, SJ 14490; LBSH 34958; LB 35016
Cyperus retrorsus Buckley, SJ 14330; LBSM 33740; LBJL 34030
Cyperus squarrosus L., SJ 14451a
Cyperus strigosus L„ SJ 14379
Cyperus surinamensis Rottb, SJ 14444; LBSH 34100; LB
35112
Cyperus virens Michx, SJ 1 43 1 6
Eleocharis microcarpa Torn, MM 8022; LBSH 33831
Eleocharis montana (Kunth) Roem. & Schult, LBSH 34183
Eleocharis obtusa (Willd.) Schult, SJ 14314
Eleocharis tuberculosa (Michx.) Romer & Schult, SEWP 374
Fimbristylis autumnalis (L) Roem. & Schult, SJ 14446
Fimbristylis dichotoma (L) Vahl, SJ 14365
* Fimbristylis miliacea (L.) Vahl, NIX 16005 (SBSC); LB 35031
Fimbristylis vahlii (Lam.) Link, DK 1 61 4; LBSH 34962
Fuirena breviseta (Coville) Coville, SJ 14395
Kyllinga pumila (Torn) Spreng, SEWP 312; LBSH 33820; LB
34563,35030
Rhynchosporacaduca Elliott, LBRH 34630; LBSH 34664
Rhynchospora colorata (L) H. Pfeiff, LB 34660
Rhynchospora corniculata (Lam.) A. Gray; MM 8025, SJ 14474,
14318, 14149
Rhynchospora elliottii Elliott, SJ 1 4390
Rhynchospora globularis (Chapm.) Small var. globularis, SEWP
446; LB 34599
Rhynchospora glomerata (L) Vahl, SJ 1 4405, 1 4355; LBJL 34043;
LBSH 34938
Rhynchospora grayi Kunth, SJ 14480a, 14480
Rhynchospora inexpansa (Michx.) Vahl, SJ 14404; LBSM
33759
Rhynchospora latifolia (Bald, ex Elliott) Thomas, SJ 14391
Rhynchospora mixta Britton, LB 1 921 8b, 1 9221; SJ 141 50
Rhynchospora recognita (Gale) Krai, LBRH 34644; LBSH
34665
Schoenoplectuscalifornicus (Meyer) Sojak,SJ 14153
Scirpus cyperinus L, LBJL 34000
+Scirpus divaricatus Elliott, GW 3495; SEWP 320; LBSM 33766;
LB 34980
Scleriaciliata Michx. var. ciliata, LB 34600
Scleria oligantha Michx, LB 34552
Scleria pauciflora Muhl. ex Willd, SEWP 294
Scleria triglomerata Michx, SJ 14357
Cyrillaceae
Cyrilla racemiflora L, SJ 14230
Dennstaeditaceae
Pteridium aquilinum <L) Kuhn var. pseudocaudatum (Clute)
Heller, GW 2953; LBSH 341 79
479
• Drosera brevifolia Pursh, LB 34596
Dtoseracapillaris Poir., SJ 14398
Dryopteridaceae (see Woodsiaceae for Athyrium and Ono-
deaceae for Onoclea)
Wtyopteris ludoviciana (Kunze) Small, GW 678; SJ 14285; LBJL
^ 34021
Wystichum acrostichoides (Michx.) Schott, LBSM 33751
2919
Dalea villosa (Nutt.) K. Spreng. var. grisea (Torr. & A. Gray)
Barneby, SJ 14209
Desmanthus illinoensis (Michx.) B. L Rob. & Fernald, SJ 14427
Desmodium ciliare (Willd.) DC., SJ 13967; GW 3313b;
LBSH, 33824, 34089
+Desmodium fernaldii B.G. Schub, LB 35021
Desmodium glabellum (Michx.) DC, LB 35061
BSH 33840, 34927; LB
jjquisetum hyemale L, GW 3352
i viscosum (L) Torr, SJ 1 4289
Vaccinium arboreum Marsh, SJ 14503
Vaccinium corymbosum L, SJ 14358
Wanium elliottii Chapm, LB 34296
Euphorbiaceae (see Phyllanthaceae for Phyllanthus)
kalypha gracilens A. Gray var. gracilens, SJ 14413; LB 34991
kaiypha virginica L„ LBSH 34167; LB 34233
'Caperoniapalustris A. St-Hil, LB 35109
Chamaesyce humistrata (Engelm. ex A. Gray) Small, LBRH
33892
Chamaesyce maculata (L.) Small, LBSH 34192, 34949, 34969
Chamaesyce nutans (Lag.) Small, GW 295 1 ; LB 35062
Chamaesyce serpens (Kunth) Small, LBJL 34016; LBSH 34662
OHdoscolus texanus (Muell. Arg.) Small, LB 34229
b°ton capitatus Michx, MM 801 5; SJ 14482
W**on glandulosus L. var. glandulosus, SEWP 286; LBSH
34932
willdenowii Webster, SJ 14478; LB 34238, 35029
morbiacorollata L„ MM 8033; SJ 14220; LB 35152
yonihotgrahamii Hook, GW 446
^tianiafruticosa (Bartr.) Fernald, SJ 14126
Mngiasylvatica L„ SJ 14495
s/no///7 Sh i n n e rs, LB 34591
Bivens L, SJ 1 41 95; LB 35076
^ourtici folia Michx, LBRH 33883; LB 35188
Wicasebiferum (L.) Small, LBSH 33832
F*baceae
'kxhynomene indica L, LB 34209, 35111
'eptocarpus Torr. & A. Gray, LB 34297
g*®9** virginianum (L) Benth, SJ 14184; LBRH 3
gwco/wdens/s L. var. canadensis, LB 34561
fascicuiata (Michx.) Greene, SJ 14187
nictitans (L) Moench, LBSH 33850
E ^Mariana L„ SJ 14424; LBSM 33727
35083
Desmodium lineatum DC, LB 35063
Desmodium obtusum (Muhl. ex Willd.) DC, LBSH 34090
Desmodium paniculatum (L) DC, GW 3416, 2957; LBSM
33735,33772
^Desmodium viridiflorum (L) DC, LBSH 33852; LB 35033; LBRH
34065; PRm(T)
Erythrina herbacea L„ WP 1 59
+Galactia regularis (L) Britton, Sterns, & Poggenb, SJ 14416;
LBRH 33901 ;SEWP 454; LBJL 3401 7
Gleditsia aquatica Marshall, SJ 14448; LBSH 34400 (J)
Glottidium vesicarium (Jacq.) Harper, SEWP 300; LB 34076
*lndigofera suffruticosa Mill, LBSH 34725
*Kummerowia striata (Thunb. ex Murray) Schindl, LB 34264
+Lackeya multiflora (Torr. & A. Gray) Fortuna, LP. Queiraz& C.P.
Lewis, LBSM 33730; LB 34590; LBRH 34633
Lespedeza hirta (L) Hornem. subsp. hirta, SJ 14199, 14475;
LBSH 33836; LB 34236
Lespedeza repens (L) Bart, LBSM 33737
*Medicago lupulina L, WP 084
Mimosa hystricina (Small) Turner, LB 34543
Mimosa strigillosa Torr. & A. Gray, SJ 1 4409
Neptunia pubescens Benth, LBSH 34661
*Pueraria montana (Lour.) Merrill var. lobata (Willd.) Van der
Maesen& Almeida, SJ 14335
Rhynchosia difformis (Elliott) DC, SJ 14498; LBSH 34722
Rhynchosia lati folia Nutt, ex Torr. & A. Gray, SJ 14123; LB
35018
Rhynchosia reniformis DC SEWP 285; LB 34557
Senna occidentals (L) Link, WP 153
Sesbania drummondii (Rydb.) Cory, LBSH 33846
Sesbania herbacea (Mill.) McVaugh, LBRH 33910
Strophostyles helvola (L) Elliott, LBSH 34076a
Strophostyles leiosperma (Torr. & A. Gray) Piper, LBSH 34716
Strophostyles umbellata (Willd.) Britton, MM 8037; SEWP 301;
GW 341 7; LBSH 34935
Stylosanthes biflora (L) Britton, Sterns & Poggenb, MM 8035;
SJ 14218; LB 35008
Tephrosia onobrychoides Nutt, SJ 14383
Tephrosia virginiana (L) Pers, SJ 14222; LB 34594
*Trifolium campestre Schreb, LB 34298, 34574
* Trifolium repens L, WP 1 23
Vida ludoviciana Nutt subsp. ludoviciana, LB 34295
Vicia minutifhra A. Dietr, LB 34256, 34585; LBSH 34405 (J)
Wisteria frutescens (L) Poir, SJ 14568, 14343
Journal of the Botanical Research Institute of Texas 4(1)
* Cotoneaster sp, LBSH 34652 (J)
Crataegus marshallii Eggi, SJ 13714
Crataegus opaca Hook. & Am. ex Hook., SJ 13710; LBRH
34629
Crataegus viridis L, LB 34293
+Aronia arbutifolla (L.) Pets. SJ 14287
+*Potentilla indica (Andrews) T. Wolf, WP 109; LB 34262
Prunus caroliniana (Mill.) Alton, SEWP 204; LB 35069
Prunus serotina Ehrh., SEWP 303; LB. 35189
Rubus argutus Link, WP 076
Rubus triviaiis Michx, SEWP 450
Physalis heterophylla Nees, LB 34228
Physalis pubescens L., SJ 1 445 1 ; GW 2956; LBSH 34095
Solanum carolinense L, SJ 14201
J 14203; LB 35167 V!
Sphenodeaceae
*Sphenoclea zeylandica Gaertn., LB 34983
Styraceae
Halesia diptera Ellis, LBSM 33723; SJ 14140, 14339
Styrax americana Lam., SJ 14382, 13715; LBSH 33843; LB
Diodia teres Walter, SJ 1421 3; LBSH 34946
Diodia virginiana L, LBSM 33732; SJ 14229
* Galium aparine L, LB 34265
Galium pilosum Aiton, SJ 1 421 5
Galium tinctorium L, LB 34307; LBSH 34403 (J)
Galium uniftorum Michx., SEWP 1 98
Mitchella repens L, WP 1 24; LB 35 1 5 1
Oldenlandia boscii (DC.) Chapm., LBRH 34054
Oldenlandia uniflora L, SEWP 381 ; LBSH 34096, 34941
*Richardia brasiliensis Gomes, SJ 1441 7
*Richardia scabra L., LBRH 33907; LB 35023
?r„ LBSM 33763; SJ 13712
Polypremum procumbens L, SEWP 461 , 203; SJ 1 41 97
Thelypteridaceae
Thelypteris kunthii (Desv.) Morton, MM 8032; SJ 143C
829
.. subsp. parviflorus (I
e Malvaceae for Tilia)
Zanthoxylum clava-herculis L„ LBSH 34950
Salicaceae
Salix nigra Nutt., WP 074
Salviniaceae
* Sahrinia minima Baker, LBSH 34080; LB 351 1 8
Typhaceae
Typha lati folia L,WP 162
Ulmaceae (see Cannabaceae for Celtis)
Planera aquatica (Walter) J. F. Gmel., WP 1 55
Ulmus alata Michx, SEWP 289
Ulmus americana L, SJ 14303
Sapindaceae
Acerbarbatum Michx, SJ 14302; LB 34096
Acer rubrum L, LBSM 33731
Saururuscernuus L, LBSH 3384, 34185
Scrophulariaceae (all of our taxa are in the following fami-
lies: Orobanchaceae ( Agalinis and Aureolaria ); Linderniaceae
( Lindemia ); Phrymaceae ( Mimulus ); Plantaginaceae (Bacopa,
Gratiola, Mecardonia, Micranthemum, Nuttallanthus, Veronica,
Urticaceae
Boehmeria cylindrica (L) Sw„ LBJL 34037
Valerianaceae
Valerianella radiata (L) Duff, WP 082; LB 34551, 34582
Verbenaceae (see Lamiaceae for Callicarpa and Vitex)
*Glandularia pulchella (Sweet) Trons, SEWP 1 90
Phyla lanceolata (Michx.) Small, LB 35162
Phyla nodiflora (L.) Greene, GW 2934
*Verbena brasiliensis Veil, SJ 1 4202; GW 2932
Verbena halei Small, GW 2922; LB 34532
*Verbena rigida Spreng, LBJL 3401 2
Verbena scabra Vahl, GW 2959; LB 351 64
and Scoparia)
Selaginellaceae
Selaginella apoda (L) Spreng, PR s.n.
Smilacaceae
Smilax bona-nox L, LBSH 34407 (J), 34653 (J); LB 34984
Smilax glauca Walter, SEWP 21 7, 207; LBSH 3441 3
Smilax laurifolia L, LBJL 34005
Smilax pumila Walter, MM 801 9; SEWP 278
Smilax rotundi folia L, SEWP 215
Smilax smallii Morong, SEWP 216; LB 34230, 34258; LBSH
34944
Smilax waited Pursh, SJ 14561; LB 35025
Violaceae
Viola lanceolata L„ LB 34291
Viola sororia Willd. var. missouriensis (Greene) McKinney, LB
Viola sororia Willd. var. sororia, LB 34548
Viola palmata L, LB 34276
Vitaceae
Ampelopsis arborea (L) Koehne, WP 143; LBRH 33908
Parthenocissus quinquefolia (L) Planch, SJ 14304
Vitis aestivalis Michx, SEWP 211; LBRH 34636; LB 34616 ^
Vitis cinerea (Engelm.) Engelm. ex Miliardet var. cinerea,
14460a, 14408
485
486 Journal of the Botanical Research Institute of Texas 4(1)
(Turner et al. 2003). This legume vine, bearing rather large leaflets, was common on sandy soil along the
Neches River in Hardin County. Our three collections are from sites that we were able to access.
ley 2007). ’ V Y W
Oxalis dillenii. See Nesom (2009a) regarding this name and a significant revision of the other caulescent Oxalis.
Oxalis debilis. See Nesom (2009b) regarding this name and a significant revision of the other acaulescent Oxahs.
Glyceria striata. Turner et al. (2003) mapped this common species of the more northern states largely in
the Edwards Plateau with Lamar and Washington counties the only stations east of the Plateau. The Hardin
Wagon Road Trail which is off of a pipeline right of way in the only portion of the Jack Gore Baygall Unit
that contacts Highway 2937.
Potentilla indica. Formerly named Duchesnea indica (Andrews) Focke, molecular studies indicate it is better
placed in Potentilla (Erickson et al. 1998).
Pseudognaphalium luteoalbum. The only record east of the Edwards Plateau for this invasive species
was Liberty County (Brown et al. 2007). Our Hardin County collection, from a draw-down site along the
Neches River, is the second record east of the Plateau.
DISCUSSION
There are 133 families and 379 genera for the 734 (645 native) species and 739 taxa (650 native) recorded
for the Jack Gore Baygall and Neches Bottom Units. Seven families: Poaceae (114 taxa), Asteraceae (95 taxa),
Cyperaceae (69 taxa), Fabaceae (52 taxa), Euphorbiaceae (21 taxa), Lamiaceae (15 taxa), and Fagaceae (13
taxa) account for 51% of all taxa collected. The genera with 10 or more taxa are Cyperus (21 taxa), Dichan-
thelium (19 taxa), Carex (17 taxa), Quercus (11 taxa), and Juncus (10 taxa). As are all floras, our list is incom-
plete but we believe that we have collected over 90 % of the taxa. Of the eight units inventoried so far, the
following 28 taxa were found only in the Jack Gore and Neches Bottom units: Alternanthera philoxeroides,
Torilis nodosa, Bidens laevis, Croptilon rigidifolium, Pseudognaphalium luteoalbum, Solidago stricta, Tetragonothera
ludoviciana, Vemoniagigantea, Scirpus divaricatus, Impatiens capensis, Cucumis melo, Schoenoplectus califomicus,
Equisetum hyemale. Astragalus leptocarpus, Desmodiumfemaldii, Desmodium lineatum, Limnobium spongia, Glims
radiatus, Hexalectris spicata, Spiranthes laciniata, Saccharum brevibarbe var. brevibarbe, Sporobolus vaginiflorus,
Potamogeton pusillus, Eichhornia crassipes, Potamogeton pusillus, Anemone berlandieri, Richardia brasiliensis, and
Vitex negundo var. intermedia.
ACKNOWLEDGMENTS
This study was supported in part by an All Taxa Biological Inventory grant (ATBI) from the Big Thicket As-
sociation of Saratoga, Texas. We appreciate Michael W. Palmer, Eric L. Keith and on anonymous reviewer for
their careful reading of the paper. We also thank Frank Gregg and Mary Pritchett of the Outdoor Nature Club
of Houston who have mounted 9,000+ plants from the currently surveyed Big Thicket Units. This volunteer
work occurred most months of the year and only for a few hours on Wednesday mornings. The first author
also wishes to thank his wife. Ruby Brown, for some computer help for our Big Thicket papers.
REFERENCES
Aronow, S. 1981. Notes on the geological units: Big Thicket National Preserve. Big Thicket National Preserve,
Beaumont, TX.
Brown, L.E., B.R. MacRoberts, M.H. MacRoberts, PA Harcombe, W.W. Pruess, LS. Elsik, and D. Johnson. 2005. Annotated
checklist of the vascular flora of the Turkey Creek Unit of the Big Thicket National Preserve, Tyler and Hardin
counties, Texas. Sida 21:1807-1827.
I. Pruess, I.S. Elsik, and S.D. J
a of the Big Sandy Creek U
THE VASCULAR PLANTS OF MOWOTONY PRAIRIE:
A SMALL REMNANT COASTAL GRASSLAND IN BRAZORIA COUNTY, TEXAS
DJ. Rosen
Department of Biology
Lee College
Baytown, Texas 77522-0818, U.SA
drosen@lee.edu
ABSTRACT
1 previously reported a species rich flora for Nash Prairie, a 120 ha remnant of undisturbed Texas coastal
grassland (Rosen 2007). During the course of that research, I simultaneously surveyed Mowotony Prairie, a
trailer but equally rich and unique coastal grassland. Mowotony Prairie is a 42 ha remnant coastal grassland
also Seated on the Kittie Nash Groce (KNG) Ranch in Brazoria County, Texas (N29°16'16.0,'W95o40,19.5";
% 1, Fig. 2). Like Nash Prairie, Mowotony Prairie has similar topographic features, disturbance history,
“anagement, and use. The purpose of this paper is to provide an annotated checklist of the vascular
Plants of Mowotony Prairie and add additional taxa to the native flora of Texas coastal grasslands. Nash
^ Mowotony Prairies and other remnant coastal grasslands are threatened by the expanding Houston
^tropolitan Area (Fig. 1).
CHECKLIST
^lies are arranged alphabetically, beginning with monocots, and followed by eudicots. Genera, species,
infraspecific names are arranged alphabetically under families and their classification generally follows
et al. (1997). Dichanthelium (Hitchc. & Chase) Gould is treated seperately from Panicum L. Nativity
native to the United States) is based on Correll and Johnston (1970). Non-native species are indicated
tyan asterisk (*). Native coastal grassland taxa not reported from Nash Prairie (Rosen 2007) are indicated
^superscript dagger 0. Finally, endemic (with distribution limited to grasslands of the Upper Texas
0r the greatest extent of their range occurring therein), rare (of limited range), or regionally rare
Wdom occurring or of previously unknown occurrence in grasslands of the Upper Texas Coast) species are
493
m
Journal of the Botanical Research Institute of Texas 4(1)
494
Michx.,3275
Scrophulariaceae
*. Agalinis fasciculate (Elliott) Raf., 4379
Agalinis viridis (Small) Pennell, 4432
Buchnera americana L., 2598
Castilleja indivisa Engelm, 3289
Mecardonia acuminata (Walter) Small, 4436
f Nuttallanthus texanus (Scheele) D. Sutton, 3273
Tabu 1. Taxonomic summary of vascular plants of Mowotony Prairie.
Families
Total
This research resulted in collections of 199 species of vascular plants representing 41 families and 129 gen-
era (Table 1). The four families with the most species were Poaceae (45), Asteraceae (38), Cyperaceae (30),
and Fabaceae (19). Species-rich genera included Carex (7 spp.), Cyperus (8 spp.) Juncus (6 spp.), Panicum (5
spp.), and Paspalum (6 spp.). Non native species ( Cyperus haspan, Briza minor, Cynodon dactylon, and Paspalum
urvillei) accounted for 2% of the total number of species.
Of the 195 native species that occur at Mowotony Prairie, 41 are not known from Nash Prairie (Rosen
2007). One family, Buddlejaceae, is not known from Nash Prairie (Rosen 2007). The native vascular plant
taxa reported here combined with those reported from Nash Prairie provides a documented 63 families and
331 species for remnant coastal grasslands of the Upper Texas Coast.
Nine endemic, rare, or regionally rare species found at Nash Prairie also occur at Mowotony Prairie:
Cooperia traubii, El eocharis compressa var. acutisquamata , E. wolfii, Scleria muehlenbergii, Juncus elliottii var.
elliottii, Boihriochloa exaristata, Sporobolus silveanus, Rudbeckia texana, and Thalictrum texanum. An addi-
tional interesting characteristic of Mowotony Prairie is the local abundance and dominance of uplands by
Sporobolus silveanus. This species is endemic to the southeastern United States (Louisiana, Oklahoma, Texas)
where it grows in blackland prairies, wet to mesic pine woodlands and adjacent glades and barrens (Gould
1975; Flora of North America Editorial Committee 2003). Diamond and Smeins (1985) described a novel S.
silveanus-Carex meadii grassland type from the northern end of the Blackland Prairie region of Texas. This
occurrence of seemingly the same or similar community over 500 km south of sites where it was discovered
and described by Diamond and Smeins (1985) might indicate that it was more widespread, and much of its
original extent has been destroyed.
ACKNOWLEDGMENTS
I am grateful to Larry Brown, Michael Palmer, and Chris Reid for reviewing this manuscript. Robert Freck-
mann, Robert Krai, and Guy Nesom annotated difficult specimens. Special thanks to Susan and Peter Conaty
for the tireless pursuit of a means for permanent conservation of Mowotony and Nash Prairies. Adam Young
prepared figure 2.
REFERENCES
CoftRELL, D.S. and M.C. Johnston. 1 970. Manual of the vascular plants ofTexas. Texas Research Foundation, Renner.
Diamond, D.D. and F.E. Smeins. 1985. Composition, classification, and species response patterns of remnant ta
grass prairies in Texas. Amer. Midi. Naturalist 1 13:294-308.
BOOK REVIEW
Jennifer P Mathews. 2009. Chide: The Chewing Gum of the Americas, From the Ancient Maya to Wil-
liam Wrigley. (ISBN 978-0-8165-2821-9, pbk.). The University of Arizona Press, 355 S. Euclid Avenue,
Suite 103, Tucson, Arizona 85719, U.S.A. (Orders: www.uapress.arizona.edu, 1-800-426-3797). $17.95,
142 pp., b&w photographs and figures, 7" x 8 1/2*.
h a 9-page opening section (acknowlec
THE VASCULAR FLORA OF
KERR WILDLIFE MANAGEMENT AREA, KERR COUNTY, TEXAS
Jason R. Singhurst Laura L. Hansen Jeffrey N. Mink
Journal of the Botanical Research Institute of Texas 4(1)
mission (now Texas Parks and Wildlife Department) in June 1950 with funds made available by the Pittman-
Robertson Wildlife Restoration Act. The initial purpose of KWMA was to serve as a wildlife research and
demonstration area where biologists could study and evaluate wildlife and habitat management practices.
During the 1960s, the objectives of the KWMA were expanded to include maintaining optimal productiv-
ity of range land, thus maximizing monetary return, while sustaining maximum wildlife resources. This
objective permitted the initial habitat manipulation, particularly the clearing of large areas of mature Ashe
juniper (Juniperus ashei) for both range and wildlife habitat enhancement. In 1989, more flexible multiple-
use goals, which included research, demonstration, education, preservation, conservation, and recreation,
were adopted and are currently used as a management guide.
MATERIALS AND METHODS
The checklist is largely based upon examination of specimens collected between 1955 and 2001 which
are deposited in the Baylor University Herbarium (BAYLU). Additional specimens from the University of
Texas Herbarium (TEX and LL), Kerr WMA Herbarium (acronym KWMA used within), and the S.M. Tracy
Herbarium (TAES) were also examined. Field studies were conducted from 2006 through spring 2009,
with emphasis on finding species expected to be present, but not yet vouchered. These specimens were
also deposited at BAYLU.
Nomenclature generally follows that of Correll and Johnston (1970), with updates and corrections as
needed from Hatch et al. (1990), Jones et al. (1997), and NRCS, USDA (2010).
The vegetational analysis compared species richness of floristic inventories of various areas (see Table 1)
of Texas. Documented species lists were compared against Arrhenius’ (1921) model subsequently adapted by
Williams and Lutterschmidt (2006) in order to determine the adequacy of the KWMA sampling effort. Spe-
cies richness and geographic area of KWMA, nine selected sites, and the state of Texas, were log-transformed
into a database. A statistical relation of species richness as a function of geographic area produces a theoreti-
cal slope (z) and intercept (d) based on this formula: S = dAz (Arrhenius 1921). A linear function is created
from this log-transformed data and the slope determines a theoretical value of species fidelity equated per
unit area and thereby an empirical measure of sampling effort. Arrhenius (1921) first fit a model to data on
increasing species number with increasing size of area sampled. Relationship between species and area partly
arises because of increasing likelihood of habitat diversity with increasing area sampled (Diamond 1988).
Arrhenius explicitly stated that his power formula, S = S(A) = dAz, was empirical and should be regarded
as an approximation whose existence was entirely dependent on agreement with data from lists of flora
that he had obtained. Because his formula calculated an average number of species occurring in an area,
he also contemplated the problem of establishing a stochastic model for species richness in smaller land
parcels consumed by a larger land mass. In order to relate area to species occurrence, Arrhenius assumed
that any individual of any species of this smaller area must have an equal opportunity of occurrence in the
larger area and thus probability could be expected. However, expectation in occurrence contrasts sharply
with the difficulties of explaining variance by this equation (Ugland et al. 2003). It is suggested that a dis-
turbance regime, or lack of one, is a significant contributor to relationship exceptions and variance. Similar
approaches have been taken more or less independently by several authors who examined the distribution
of individuals and presence/absence pattern of species (e.g. Gleason 1922; Hurlbert 1971; Heck et al. 1975;
Brewer & Williamson 1980; Coleman 1981; Ney-Nifle & Mangel 1999; Williams & Lutterschmidt 2006). AH
of the proposed formulae may be regarded as variants of Arrhenius’s (1921) original model. For an historical
review of species-area curves, see McGuinness (1984).
The Sorenson coefficient (1948; also known as “quotient of similarity,” was used as a community
similarity index to compare KWMA to both Mason Mountain WMA and Enchanted Rock State Natural
Area (ERSNA) to quantitatively assess the best floristic comparison. Numbers of species within each of the
three areas were cross-checked for commonality of species occurrence and used as: CG « 2c / si + S2, where
si and sz are species number in communities 1 and 2, respectively, c is the number of species common to
Singhurst et aL, Flora of Kerr Wildlife Management Area
v plants and associated geographic a
a from published and unpublished irr
>ries in Texas, USA.
Amistad NRA
ig Lake Bottom WMA
Enchanted Rock SNA
Fairfield Lake SRA
] Fort Hood Military Res.
|6us Engeling WMA
I McLennan County
707
459
497
719
985
5524
498.96
413.09
351.06
76455
57.4242
17.0182
6.7234
5.9697
45.1060
265313
12363422
21.6859
26975428
6779403
Poole (unpub.)
Fleming et al. 2002
Singhurst et al. i
Singhurst et al. t
P&th communities. The value of CCs ranges from 0 (when no species are common to either community) to
The Sorenson coefficient is an adaptation of Jaccard’s (1902) coefficient of community originally stated
Ip CG = c / (si + si) - c and was originally utilized to accompany data consisting of presence or absence of
S species. The following caution should be noted: for a given amount of similarity between communities, the
Kmilarity indices (Sorenson and Jaccard) do not necessarily express the same quantitative values. Thus,
| both express similarity between communities, but should not be compared against each other. Assessment
Roverlapping plant associations and groupings applying similarity indices are attempts to quantify niche
1 overlap, an arena of significant disagreement among contemporary ecologists (Looman & Campbell 1960;
, Hurlbert 1978; Abrams 1980; Wallace 1981; Hurlbert 1982; Abrams 1982; Ungland et al. 2003).
relations (NatureServe 2008) and land use classes were circumscribed and mapped for KWMA utilizing
1 1996 digital orthophoto aerial photography and ERDAS Imagine 8.7 software (Leica Geosystems 2008).
DESCRIPTION OF STUDY AREA
KWMA consists of 2635.5 ha (6514.9 acres) located 35.5 km (22 miles) west of Kerrville, Texas. The
topography, soil types, and vegetation of KWMA are representative of the surrounding Edwards Plateau
Ecological Region. Soils are generally rocky and shallow, covering a substratum of limestone. Topography
is gently rolling to hilly with occasional draws (a shallow, open, natural, drainage) and small canyons. An-
9ual rainfall from 1951 to 1986 averaged 64.7 cm (25.48 inches), with the wettest months being April, May,
: June, August, September, and October. KWMA is drained by the North Fork of the Guadalupe River which
also forms part of the southern boundary. Most drainages are intermittent. Several small springs and the
} Guadalupe River provide the only natural permanent water sources. Elevation varies from 588.3 m (1930
kt) to 682.8 m (2240 feet), with average elevation being 609.6 m (ca. 2000 feet). With respect to manage-
ment at KWMA, practices used are designed to encourage perennial bunch grasses and maintain a high
diversity of herbaceous annuals and perennials and include prescribed burning (especially winter bums),
KWMA supports a diversity of native wildlife species. These include white-tailed deer ( Odocoileus vir-
frianus), Rio Grande wild turkey (Meleagris gallopavo intermedia), javelina ( Pecari tajacu), eastern cottontail
rabbit ( Sylvilagusfloridanus ), black-tailed jackrabbit ( Lepus califomicus), northern raccoon ( Procyon lotor),
^iped skunk ( Mephitis mephitis), and Virginia opossum ( Didelphis virginiana), all being abundant. Larger
Predators such as coyote ( Canis latrans) and bobcat ( Lynx rufus) are uncommon. Historically, mountain lion
concolor), American black bear (Ursus americanus), and gray wolf ( Canis lupus) inhabited the area, but
500
Journal of the Botanical Research Institute of Texas 4(1)
all have been extirpated. Hahn (1951) reported that black bear was present in this vicinity as late as 1905
and the last recorded killing of a gray wolf was at the head of the North Fork of the Guadalupe River in 1913.
KWMA has recorded 191 species of resident and migratory birds and 29 species of herpetofauna. Exotic
wildlife species intentionally introduced to the area are axis deer (Axis axis), aoudad sheep ( Ammotragus
lervia), and sika deer ( Cervus nippon ). Also present is the feral hog (Sus scrofa), which probably originated
from escaped stock.
Three federally listed endangered species occur on KWMA. These include two birds, the black-capped
vireo (Vireo atricapillus ) and the golden-cheeked warbler ( Dendroica chrysoparia ) and one plant, the Tobusch
fishhook cactus ( Sclerocactus brevihamatus ssp. tobuschii).
RESULTS AND DISCUSSION
Twenty-one natural plant community associations (NatureServe 2008) and three land use classes were
determined to be present in KWMA. Two natural associations, Sawgrass-Spikesedge-Beakrush-Black
Bogrush-Aparejograss Herbaceous Vegetation Association and Ashe Juniper - Bastard Oak - Plateau Live Oak
Woodland are described as new associations for Texas. The other associations are of common occurrence in
the Edwards Plateau vegetational area. Land use classes include developed, old field and reservoirs. For the
purpose of organization, the plant community association descriptions are separated into system categories
that include uplands, canyons, cliff faces, floodplain, springs, seeps, and aquatic types, and land use classes.
In general, the associations are discussed from north to south. All references to geology are based upon the
Llano [Map] Sheet, University of Texas Bureau of Economic Geology, 1981.
NATURAL TERRESTRIAL ASSOCIATIONS
Upland Types
Plateau Live Oak / Curly-mesquite Woodland Vegetation Association (Allard 1990, Diamond 1993)
occurs on limestone with clay soils in the Edwards Plateau. Normally it is found on flat to moderately roll-
ing terrain of 0-5% slope. It comprises 60.7 ha (150.1 ac) and is developed on the lower Cretaceous Segovia
Member of the Edwards Limestone Formation. This association is concentrated in the northwestern portion
of KWMA (Fig. 1). The vegetation is dominated by Quercus fusiformis and grasslands or grassy openings
with Bouteloua curtipendula (both varieties), Hilaria belangeri, and Schizachyrium scoparium. Other important
components in the understory include Condalia hookeri, C. spathulata, Juniperus ashei, Quercus buckleyi, Q-
sinuata var. breviloba, Q. stellata var. stellata, Rhus lanceolata, R. trilobata, and Ulmus crassifolia.
Post Oak - Blackjack Oak / Little Blnestem Woodland Vegetation Association (Diamond 1993;
Hoagland 2000) occurs over shallow soils on limestone mesa tops in the Cross Timbers and Prairies, Edwards
Plateau, and Post Oak Savanna Ecoregions in Texas as well as Cross Timbers in Kansas and Oklahoma.
Land form is flat to rolling with 0-5% slope. Approximately 70.7 ha (174.7 ac) of KWMA consists of this
formation, which is developed on the lower Cretaceous Segovia Member of Edwards Limestone Formation.
This association is found in the northern portion of KWMA (Fig. 1). The area is dominated by Quercus stel-
lata var. stellata, Q. marilandica, and Schizachyrium scoparium and varies from open woodland to savanna.
Shrub species include Diospyros texana, Cylindropuntia leptocaulis, and Smilax bona-nox. Dominant grasses
and forbs include Andropogon gerardii, Berlandiera betonicifolia, Carex planostachys, Cheilanthes tomentosa, Coc-
culus carolinus, Cyperus rotundus. Heterotheca subaxillaris, Hypericum drummondii, Leptochloa dubia, Matelea
gonocarpos, and Sorghastrum nutans. Generally, there is low forb diversity.
Plateau Live Oak - Post Oak Savanna Vegetation Association (Diamond 1993; Hoagland 2000)
occurs over shallow soils on limestone mesa tops in the Edwards Plateau and Post Oak Savanna Ecoergions
in Texas and in the Quartz and Wichita Mountains in Oklahoma. The association is found on flat to roll-
ing terrain with 0-5% slope on the lower Cretaceous Segovia Member of Edwards Limestone Formation.
Approximately 120.3 ha (297.3 ac) of the association occur in the northern and eastern portion of KWMA
(Fig. 1). Dominant plants are Quercus fusiformis, Q. stellata var. stellata, Diospyros texana, and Schizachyrium
xoparium. Shrub species include Mahonia trifoliolata and Opuntia engelmannii var. lindheimeri. Dominant
grasses include Bothriochloa laguroides, Bouteloua curtipendula (both varieties), B. pectinata, B. rigidiseta, Maria
belangeri, Muhlenbergia reverchonii, and Panicum virgatum. A moderate diversity of forbs is present.
Plateau Live Oak / Little Bluestem Woodland Vegetation Association (Diamond 1993) occurs
011 gently sloping to nearly flat slopes in the eastern Edwards Plateau and Cross Timber Ecoregons in Texas
a*l in Quartz and Wichita Mountains in Oklahoma. This association consists of about 739.5 ha (1827.4
ac) found on flat to rolling upland landscape with 0-5% slope. It is best developed on the lower Cretaceous
Segovia Member of Edwards Limestone Formation. The association is concentrated in the central and
southwestern portion of KWMA (Fig. 1). The area is dominated by Quercus fusiformis and Schizachyrium
502
Mahonia trifoliolata. Dominant grasses include Aristida purpurea (several varieties), Bouteloua curtipendula
(both varieties), B. pectinata, B. rigidiseta, and Nassella leucotricha. An extensive diversity of forbs can occur
Nuttall’s Stonecrop - (Ozark Dropseed, Poverty Dropseed) - Wright’s Spikemoss Limestone
Outcrop Vegetation Association (NatureServe 2008) consists of 246.9 ha (610.2 ac) of exposed limestone
surfaces in the Edwards Plateau. It is typical of flat upland landscape with 0-1% slope and developed on the
lower Cretaceous Segovia Member of Edwards Limestone Formation. The association is concentrated in the
southwest and southeast portion of KWMA (Fig. 1). This association is characterized by shallow pothole
depressions that accumulate soils or serve as ephemeral pools. It is dominated by Sedum nuttallianum, Sporobo-
lus ozarkanus, S. vaginiflorus, and Selaginella wrightii. Nostoc commune Vaucher (Cyanophyta, Nostocaceae)
is common during wet periods. Bare rock is occupied by scattered patches of crutose and foliose lichens
and bryophytes (Musci). Typical ferns are Cheilanthes tomentosa and Pellaea wrightiana. Other characteristic
vegetation includes annuals (ephemerals) and species adapted to bare rock xeric conditions. These include
calycosum, Chaetopappa asteroides, Cooperia pedunculata, Crassula aquatica, Croton monanthogynus, Drabacu-
neifolia, Echinochloa walteri, Erioneuron pilosum, Hedeoma drummondii, Hedyotis crassifolia, Heteranthera dubia,
Juncus marginatus, Ludwigia repens, Nothoscordum bivalve. Paronychia lindheimeriana, P. virginica, Phemeranthus
aurantiacus, Plantago virginica. Polygala lindheimeri, Scutellaria drummondii. Verbena canescens, and Veronica
peregrina.
Ashe Juniper - Bastard Oak - Plateau Live Oak Woodland Vegetation Association occurs over
shallow soils on limestone mesa tops in the Edwards Plateau. This association is found on flat terrain with
0-2% slope and consists of 173.8 ha (429.6 ac) developed on the lower Cretaceous Segovia Member of Edwards
Limestone Formation. It is found along drainages throughout the management area (Fig. 1). The vegetation
is dominated by Juniperus ashei and Quercus sinuata var. breviloba, with densities varying from open to closed
canopy woodlands. When canopy cover is dense, Quercus sinuata var. breviloba is sometimes limited to the
understory. Other components in the understory include Cercis canadensis var. texensis, Diospyros texana,
Forestiera pubescens, Fraxinus texensis. Ilex decidua, Lonicera albiflora, Mahonia trifoliolata, Quercus bucUeyi, Q.
fusiformis, Rhus trilobata, R. virens, Sophora secundiflora, Toxicodendron radicans, Ulmus crassifolia, Ungnadia
speciosa, and Yucca rupicola. Herbaceous cover is generally sparse, especially with dense canopies, and may
include Carex planostachys, Commelina erecta, Galactia texana, Matelea edwardsensis, M. reticulata, Lespedeza
texana, Rhynchosia senna, Sporobolus compositus, and Tragia ramosa.
Little Bluestem - Sideoats Grama - Texas winter-grass herbaceous Vegetation Association is a
midgrass grassland and characteristic of uplands over relatively deep soils in the Rolling Plains of Texas, but
also in the central and western Edwards Plateau (Diamond 1993). This association is found on flat to rolling
terrain with 0-5% slope. It includes 306.8 ha (758.1 ac) on the lower Cretaceous Fort Terrett and Segovia
Member of Edwards Limestone Formation in the western and southern portion of KWMA (Fig. 1). The area
is dominated by Schizachyrium scoparium, Bouteloua curtipendula (both varieties), and Nassella leucotricha.
Trees speceis include Juniperus ashei and Prosopis glandulosa. Shrub species include Opuntia engelmannii var.
lindheimeri, and Ziziphus obtusifolia. Typical grasses include Aristida purpurea (several varieties), Bothriochloa
barbinodis, Bouteloua barbata, Digitaria californica, Hilaria belangeri, Panicum obtusum, Sorghastrum nutans, and
Tridens muticus. An extensive diversity of forbs can occur in this association.
Ashe Juniper - (Buckley Oak, Plateau Live Oak, Vasey Shin Oak, Bastard Oak) Woodland
Vegetation Association (Diamond 1993) consists of woodlands over shallow soils on limestone slopes
in the Edwards Plateau. This association is found on flat to rolling terrain with 0-5% slope and consists
of 413.3 ha (1021.2 ac) on the lower Cretaceous Fort Terrett and Segovia Member of Edwards Limestone
Formation. The association is scattered throughout the KWMA (Fig. 1). Dominants are Juniperus ashei with
mixtures of Quercus buckleyi, Q. fusiformis, Q. pungens var. vaseyana, and Q. sinuata var. breviloba. Tree species
include Celtis laevigata var. reticulata and Fraxinus texensis. Shrub species include Diospyros texana, M ahonia
trifoliolata, Nolina texana, Rhus virens. Toxicodendron radicans, and Yucca rupicola. Herbaceous flora include
Bouteloua curtipendula (both varieties), Carex planostachys, Lespedeza texana, and Schizachyrium scoparium.
Plateau Live Oak - Buckley Oak / Bastard Oak - (Ashe Juniper) Woodland Vegetation Associa-
tion (NatureServe 2008) consists of woodlands occurring over shallow soils on limestone mesa tops in the
Edwards Plateau. This association is found on flat to rolling terrain with 0-5% slope and consists of 713.1
ha (1762 ac) on the lower Cretaceous Fort Terrett and Segovia Member of Edwards Limestone Formation.
The association is scattered throughout the management area (Fig. 1). It is dominated by Quercusfusiformis,
Q. buckleyi, and Juniperus ashei and characterized by a mixture of other trees and shrubs such as Celtis laevi-
gata var. reticulata, Cercis canadensis var. texensis, Forestiera pubescens, Frangula caroliniana, Fraxinus texensis,
Bex decidua, Lonicera albiflora, Opuntia engelmannii var. lindheimeri, Quercus sinuata var. breviloba, Sideroxylon
1 anuginosum, Rhus trilobata. Toxicodendron radicans, Ulmus crassifolia, and Ungnadia speciosa. Herbaceous spe-
cies include Carex planostachys, Chaerophyllum tainturieri, Limnodea arkansana, and Nassella leucotricha.
Canyon Types
Chinquapin Oak - Arizona Walnut - Slippery Elm / Frostweed Forest Vegetation Association
(Diamond 1993) occurs on mesic limestone slopes in the Edwards Plateau. This association is found on
moderate to steep slopes (5-20% slope) at KWMA. It consists of 49.5 ha (122.3 ac) and is developed on
the lower Cretaceous Fort Terrett Member of Edwards Limestone Formation. This association is found in
the southeastern portion of KWMA (Fig. 1). The vegetation is dominated by Quercus muehlenbergii, Juglans
major, Ulmus rubra, and Verbesina virginica. Other tree species include Carya illinoinensis, Fraxinus texensis,
Quercus buckleyi, Moms rubra, and Ulmus crassifolia. Subcanopy species include Moms microphylla and Juni-
perus ashei. A diversity of shrubs and woody vines may include Aesculus pavia var. pavia, Comus dmmmondii,
Garrya ovata ssp. lindheimeri, Frangula caroliniana. Ilex decidua, Parthenocissus quinquefolia, and Viburnum
rufidulum. Herbaceous species include Adiantum capillus-veneris, Carex edwardsiana, C. planostachys, Galium
texense, Brickellia cylindracea, Chaetopappa effusa, Desmodium paniculatum, Geum canadense, Packera obovata,
and Tripsacum dactyloides.
Cedar Elm - Chinquapin Oak - Arizona Walnut Woodland Vegetation Association occurs on
mesic limestone slopes in the Edwards Plateau. This association is found on moderately steep slopes (5-15%)
at KWMA and includes 1.1 ha (2.7 ac) on the lower Cretaceous Fort Terrett Member of Edwards Limestone
Formation. This association is found in the southeastern portion of KWMA (Fig. 1). The vegetation dominated
\ by Ulmus crassifolia, Quercus muehlenbergii, and Juglans major. The understory consists of Bromus pubescens,
Carex planostachys, Chasmanthium latifolium, Elymus virginicus, Panicum virgatum, and Verbesina virginica.
Lacey Oak - Ashe Juniper Woodland Vegetation Association (Diamond 1993) occurs on rocky
canyon slopes or on shallow soils that have developed over limestone in the southern and southwestern por-
tion of the Edwards Plateau. This association is found on moderately to steep slopes (5-35%) at KWMA and
includes 52.8 ha (130.4 ac) on the lower Cretaceous Fort Terrett and Segovia Member of Edwards Limestone
Formation. The association is concentrated in the eastern and western portion of KWMA (Fig. 1). The area
is dominated by Quercus laceyi and Junipems ashei. Other important canopy components include Quercus
fakleyi, Q.fusiformis, Q. muehlenbergii, and Ulmus crassifolia. Understory shrubs include Comus dmmmondii,
Hex decidua, Ptelea trifoliata, and Ungnadia speciosa (Riskind and Diamond 1988).
Cliff Face Types
Wand Butterfly-bush - Mexican buckeye / American Columbine - Dutchman’s pipe Shrubland Vegeta-
tion Association (NatureServe 2008) occurs on limestone rim rock along mesic canyons m the southern por-
tion of the Edwards Plateau. This association is found on very steep slopes to vertical cliff faces and compnses
*2 ha (10.3 ac) on the lower Cretaceous Fort Terrett Member of Edwards Limestone Formation. It is limited
to the southeastern portion of area (Fig. 1). Typical dominants are Buddleja racemosa ssp. incana, Ungnadia
speciosa, Aqudegia canadensis, and Aristolochia serpentaria. Other shrubs include Ageratina havanensis, Croton
504
fruticulosus, Eysenhardtia texana. Moms microphylla, Nolina lindheimeri, N. texana, Parthenocissus quinquejolia,
Petrophytum caespitosum, Rhus trilobata, R. virens, Toxicodendron radicans, and Yucca mpicola. Characteristic
herbaceous species included Acalypha phleoides, Aristolochia coryi, Asplenium resiliens, Brkkellia cylindracea,
Chamaesyce villifera, Cheilanthes alabamensis, C. horridula, Desmodium psilophyllum, Lespedeza texana, Li mm
rupestre, Pellaea atropurpurea, P. ovata, Perityle lindheimeri, Polygala lindheimeri, Phyllanthus polygonoides, and
NATURAL AQUATIC ASSOCIATIONS
Floodplain Types
Plateau Live Oak - Netleaf Hackber ry Woodland Vegetation Association (Diamond 1993) occurs along
dry to mesic flood plains of streams on the Edwards Plateau, South Texas Plains, and Trans Pecos Ecoregions.
This association consists of 81.5 ha (201.5 ac) developed on the lower Cretaceous Segovia Member of Edwards
Limestone Formation. It is limited to northwestern portion of KWMA (Fig. 1). The association is dominated
by Quercusfusiformis and Celtis laevigata var. reticulata. Other trees and shrubs include Juglans major, Diospyros
texana, Fraxinus texensisjuniperus ashei, and Ulmus crassifolia. Forb diversity is generally low.
Netleaf Hackberry - Little Walnut / Green Sprangletop Shrubland Vegetation Association
(Diamond 1993) occurs along dry to intermittent streams on the Edwards Plateau and Chihuahuan Desert.
This association comprises 81.2 ha (200.7 ac) in the northwestern part of KWMA (Fig. 1) developed on
the lower Cretaceous Segovia Member of Edwards Limestone Formation. Vegetation is dominated by Celtis
laevigata var. reticulata and Juglans microcarpa. Other woody plants include Diospyros texana, Fraxinus texen-
sis, Sideroxylon lanuginosum ssp. albicans, and Smilax bona-nox. Characteristic herbs can include Bothriochloa
barbinodis var. barbinodis, Bouteloua curtipendula (both varieties), and Leptochloa dubia.
Sycamore - Black Willow Woodland Vegetation Association (NatureServe 2008) occurs along
periodically scoured flat-bedded limestone on relatively flat terrain along creekbeds and riverbeds in the
Edwards Plateau and adjacent areas. It consists of small narrow strips typically not more than 10 m wide
in moist to wet gravelly soils. This association consists of 2.5 ha (1.0 ac) along the shores of the North Fork
of the Guadalupe River developed on the lower Cretaceous Fort Terrett Member of Edwards Limestone
Formation and is limited to the southeastern portion of KWMA (Fig. 1). This association is dominated by
Platanus occidentalis and Salix nigra, often as scattered small trees since this association receives frequent
catastrophic floods. Another tree species that infrequently occupies this community is Populus deltoides. A
aris nedecta. Cevhalanthus occidentalis, and
poorly developed shrub layer included Amorpha jru
Juglans microcarpa. Herbaceous species
Spring, Seep, and Aquatic Types
Switchgrass - Bushy Bluestem - Jamaica Sawgrass Herbaceous Vegetation Association (NatureServe
2008) occurs along periodically scoured flat-bedded limestone shores of perennial streams on the Edwards
and Stockton Plateaus. Terrain is relatively flat. The association is comprised of herbaceous flora that is rooted
in cracks and in soil mats along the edges and minor shelves along the river’s edge of the North Fork of the
Guadalupe River. This association consists of 1.3 ha (3.2 ac) developed on the lower Cretaceous Segovia
Member of Edwards Limestone Formation and is limited to the southeastern portion of KWMA (Fig. D-
The dominants include Panicum virgatum, Andropogonglomeratus, and Cladium mariscus ssp .jamaicense. This
as scattered individuals, and may include Platanus occidentalis, Salix nigra, Juglans microcarpa, Baccharis ne-
glecta, and B. salicifolia. Herbaceous flora includes Eleocharis caribaea, E. montevidensis, Eupatorium serotinum,
Fuirena simplex, Indigofera lindheimeriana, Ratibida columnifera, R hynchospora corollata, R. nivea, SchiZflchyrium
scoparium, and Solidago altissima.
Sawgrass - Spikesedge - Beakrush/Whitetop - Black bogrush - Aparejograss Herbaceous Veg-
etation Association occurs along spring and seep influenced herbaceous wetlands along creeks and rivers
of the Edward Plateau. Sites are dominated by sedges, grasses, and other herbaceous flora. This association
505
includes 4.1 ha (10.0 ac) along the shores of the North Fork of the Guadalupe River in the southwestern
portion of KWMA (Fig. 1) developed on the lower Cretaceous Segovia Member of Edwards Limestone
Formation. Dominants include Cladium mariscus ssp. jamaicense, Eleocharis rostellata, E. montevidensis, E.
caribea, Rhynchospora capillacea, R. nivea, and R. colorata, Schoems nigricans, and M uhlenbergia utilis. The
substrate includes well developed marl clays and gravels over limestone bed rock. Sites occur as braided
mats of vegetation in mucky soils originating at spring sources (spring heads on slopes or bases of limestone
bluffs). The sites can extend into the floodplain of the river as a matrix of patchy islands of well developed
muck with small streamlet channels braiding through the vegetation patches. Seepage slopes are spongy.
High concentrations of calcium carbonate cake layers can be present. Significant herbaceous plants often
associated with this wetland community include Adiantum capillus-veneris, Andropogon glomeratus, Carex
edwardsiana, C. microdonta, C. muhlenbergii, Bidens laevis, Boehmeria cylindrica, Centella asiatica, Epipactus
gigantea, Fimbristylis puberula, Fuirena simplex, Helenium microcephalum var. microcephalum, Juncus texanus,
Ludwigia repens, Lythrum ovalifolium, Mitreola petiolata. Nasturtium officinale. Lobelia cardinalis, Muhlenbergia
1 indheimeri, M. reverchonii, Dichanthelium acuminatum var. lindheimeri, D. oligosanthes var. scribnerianum,
Panicum virgatum, Pluchea odorata, Sorghastrum nutans, Symphyotrichum praealtum, Paspalum pubifiorum,
Ihelypteris ovata var. lindheimeri, Utricularia gibba, Verbena scabra, and Verbesina lindheimeri. The few woody
species that occur in these wetlands include Cephalanthus occidentals, Lindera benzoin, Platanus occidentals,
and Salix nigra. Unusual species include Rhynchospora capillacea, Rudbeckiafulgida, and Scleria verticillata.
Seep Muhly - Tall Grama - Little-tooth Sedge Herbaceous Vegetation Association (NatureServe
2008) occurs on ephemeral moist, open, rocky slopes on the Edwards Plateau. This association includes
0.2 ha (0.4 ac) developed on the lower Cretaceous Segovia Member of Edwards Limestone Formation and
is limited to the southeastern portion of KWMA (Fig. 1). The dominants include M uhlenbergia reverchonii,
Bouteloua pectinata, and Carex perdentata. Other herbaceous species Agalinis edwardsiana, Aristida purpurea
(several varieties), Calylophus berlandieri, Carex planostachys, Centaurium beyrichii, Desmanthus velutinus,
Galphimia angustifolia, Hedeoma drummondii, Heliotropium tenellum, Liatris mucronata, Marshallia caespitosa,
Melampodium leucanthum, Oenothera triloba. Paronychia virginica, Pediomelum hypogaeum, Polygala alba, P. lind-
heimeri. Salvia texana, Spiranthe s magnicamporum, Stenaria nigricans var. nigricans, Stillingia texana, Tetraneuris
scaposa, Thamnosma texana, Thelesperma simplicifolium, Vernonia lindheimeri, and Wedelia texana.
Southern Maidenhair - (Lindheimer’s Maidenfern, Kunth’s Maidenfern) Herbaceous Vegeta-
tion Association (NatureServe 2008) occurs on cliff faces and lower slopes of forested box canyons on the
Edwards Plateau, usually in narrow horizontal bands where seepage from exposed limestone or water from
perennial or nearly perennial creeks consistently provides greater moisture than is available on adjacent
slopes. This association consists of 4.2 ha (10.3 ac) developed on the lower Cretaceous Fort Terrett Member
of Edwards Limestone Formation in the southeastern portion of KWMA (Fig. 1). The area is dominated
^ Adiantum capillus-veneris and Thelypteris ovata var. lindheimeri. Other prominent herbaceous species are
fypactus gigantea, Hydrocotyle verticillata, Dichanthelium acuminatum var. lindheimeri, Samolus valerandi ssp.
Purviflorus, and Silphium radula.
American Water-willow - Coastal Water-hyssop Edwards Plateau Herbaceous Vegetation
Association (NatureServe 2008) occurs on relatively permanently watered streams which flow over flat-
bedded limestone strata on the Edwards Plateau. This association consists of 0.3 ha (0.6 ac) developed on
lbe lower Cretaceous Segovia Member of Edwards Limestone Formation. It is present in the southern por-
bon of KWMA (Fig. 1). The dominants include Justicia americana and Bacopa monnieri. Other herbaceous
flora includes Cyperus spp., Fuirena simplex, EleocharS caribaea, E. montevidensis. Lobelia cardinalis, Ludwigia
PdustrS, Rhynchospora colorata, and R. nivea.
Und use associations
developed land (Fig. 1) consists of about 61.5 ha (152 ac) that are heavily impacted by human use, such as
|)awns and parking areas near buildings, roads, etc. Typically, these areas are occupied by early successional
°ra hat vary from season to season and frequency of disturbance. Herbaceous flora include Capsella bursa-
506 Journal of the Botanical Research Institute ofTexas4(1)
pastoris, Erodium cicutarium, Geranium texanum, Glandularia bipinnatifida var. bipinnatifida , Krigia caespitosa,
Lamium amplexicaule, Lepidium virginicum, Plantago virginica, Stellaria media, and Veronica peregrina.
Old field vegetation ( Fig l) consists of 64.7 ha (.150.9 act of mostly non-native grasses and dominated
by Bothriochloa ischaemum. Herbaceous vegetation is quite variable and includes plants such as Asclepias
viridiflora, Bromus unioloides, Conyza canadensis, Cucurbita foetidissima, Glandularia bipinnatifida var. bipin-
natifida, Gutierrezia texana, Hordeum pusillum, Lygodesmia texana, Melilotis officinalis, Ratibita columnifera,
Ruellia nudiflora. Salvia farinacea, Sida abutifolia, Solanum spp., Verbascum thapsus, Verbena officinale, and
Vulpia octoflora.
Ponds and wells (Fig 1> consists of 0.1 ha (0 a ac' ol most K marginal and shallow pond- and well-
dominated by emergent rushes, sedges, and grasses. Herbaceous vegetation includes Agalinis homolantha,
Ammannia coccinea, Cyperus erythrorhizos, C. strigosus, Echinochloa walteri, Eleocharis montevidensis, Juncus
interior, and J. marginatus. Other associated flora includes Eclipta prostrata, Helenium elegans, Lindemia dubia,
VEGETATIONAL ANALYSIS
Figure 2 demonstrates an adequate sampling indicated by the Kerr WMA datum point above the slope of
the determined function, moreover data for both MMWMA and ERSNA place those sampling efforts above
this slope-line suggesting these proximal locales can be compared legitimately. The Sorenson coefficient
designates MMWMA as more analogous in floristic composition than ERSNA and the statistic for Jaccard’s
coefficient of community is in agreement with that floristic similarity (Table 2).
Summary Data of the Flora
The vascular flora of KWMA consists of 719 taxa (species and below). Families represented by the larg-
est number of species are Asteraceae (with 113 species), Poaceae (109), Fabaceae (38) and Euphorbiaceae
(34). Apparently, the drier nature of KWMA is more agreeable to the Euphorbiaceae and less so for the
Cyperaceae, which is the fourth largest family within Texas. The Cyperaceae is represented by 27 species in
KWMA, which includes four species of Carex and nine species of Cyperus, while in Texas as a whole these
two genera have 96 and 56 species respectively. Other large families include Lamiaceae (22), Apiaceae (17),
and Scrophulariaceae (15).
Of the 719 species reported for KWMA, 65 (9.04%) are considered to be non-native. This compares
favorably with other wildlife management areas in Texas. For example, Mason Mountain WMA has 8.5 %
of its flora introduced (Singhurst et al. 2007), while Gus Engeling WMA, of the Post Oak Savannah of the
eastern part of the state, has 6.34 % introduced species (Singhurst et al. 2003). Diggs et al. (1999), reports
17.7% of the flora treated in their work (north central Texas) as non-native. The easily noticed trend is that
species present. Several of the non-natives present at KWMA may have been intentionally introduced as
part of land management plans.
The following species, subspecies, and varieties documented to occur in KWMA, are considered en-
demic to the state of Texas (Correll & Johnston 1970; Carr 2008): Agalinis edwardsiana, Argythamnia simulans,
Astragalus wrightii, Brickellia eupatorioides var. gracillima, Buddleja racemosa ssp. incana, Carex edwardsiana,
Chaetopappa bellidifolia, C. effusa, Clematis texensis, Euphorbia roemeriana, Galactia texana, Matelea edwardsensis,
Monarda punctata ssp. punctata var. intermedia, Nolina lindheimeriana, Parthenocissus heptaphylla, Pediomelum
hypogaeum var. scaposum, Penstemon triflorus ssp. triflorus. Phlox roemeriana, Physaria densiflora, P recurvata,
Sclerocactus brevihamatus ssp. tobuschii, Tradescantia edwardsiana, Tragia nigricans, Verbesina lindheimeri, Val-
erianella stenocarpa, Vitis monticola, and Yucca rupicola.
Following are comments on unusual or interesting plant distributions.
R hynchospora capillacea. — In Texas, this species is recorded only from Kerr Co. The nearest known
occurrence is 355 miles to the north in the Soper Bog (Railroad Bog) in Choctaw County, Oklahoma. The
507
508
Journal of the Botanical Research Institute of Texas 4(1)
Boltonia diffusa. — In Texas, distributed in the Pineywoods, Poat Oak Savanna, and Gulf Coast Prairies
and Marsh Ecoregoins. In central Texas, this species has been dcoumented in Hays County, with this this
being the western limit of the species.
Following is a list of western species that reach their eastern limits in Kerr County. In general, these spe-
cies occur in the Trans-Pecos vegetational region, the area of the state west of the Pecos River, occasionally
referred to as “far west Texas.” These records may only indicate a need for more intensive field studies in
the western part of the Edwards Plateau.
Petrophytum caespitosum. — Also recorded for Real Co., on the west border of Kerr Co. and Frio Co.,
Echeandia flavescens. — This is the first record east of the Trans-Pecos. Turner et al. (2003) mapped the
species as occurring in Brewster, Jeff Davis, and Presidio cos.
Dyssodia papposa. — A Trans-Pecos and Panhandle (Rolling Plains and High Plains) species with one
record in Live Oak Co. of the Rio Grande Plains.
Pseudognaphalium canescens. — Also present as a disjunct in Llano Co., about 75 km to the northeast of
Kerr Co.
Viguiera stenoloba. — Also known from the Rio Grande Plains, especially those counties adjoining the
Rio Grande [River],
ANNOTATED CHECKLIST OF THE FLORA
The annotated checklist is divided into pteridophytes, gymnosperms, and angiosperms, which are subdivided
into monocots and dicots. Family, genus, and species are arranged alphabetically beneath each major head-
ing. Collectors and collection numbers are referenced as following: CAM = C.A. McMaham, DH = Donnie
Harmel, FG = Frank Gould, KW = Kerr Wildlife Management Area staff, MM = Morton May, JS = Jason
Singhurst, LH = Laura Hansen, LS = Laura Sanchez (earlier name for LH), S&R = D. Seigler & W. Renold, TD
= Timery Debore, TT = Thomas Trinzie, and AC = Amy Choy. An asterisk (*) denotes an introduced species.
Common names are included to facilitate ease of use by persons unfamiliar with botanical names.
PTERIDOPHYTES
8473; 10592; 10627
Selaginellaceae
Selaginella wrightii Hieron.; Wright's spikemoss; JS 1 0044,
LS4106
Marsilaceae
Marsilea vestita Hook. & Grev. ssp. vestita; water clover fern;
JS 10228
Thelypteridaceae
Thelypteris o\,atn P F r i, >hn ,r «n il r,- t < >,AF p
Lindheimer's maidenhair fem; JS 9971; LS 3836
Pteridaceae
Adiantum capillus-veneris L.; southern maidenhair fern; JS
8500; 9970; LS 3843
Astrolepis integerrima (Hook.) Benham & Windham; hybrid
cloakfern; LH 5860
Astrolepis sinuata (Lag. ex Sw.) Bentham & Windham ssp.
sinuata ; bulb lip fern; JS 1 81 30
GYMNOSPERMS
Cupressaceae
Juniperus ashei Buchh.; Ashe juniper; JS 18004; DH s
8288; LS 381 2
Juniperus pinchotii Sudw.; Pinchot's juniper; JS 17205
Taxodium distichum (L) Rich.; bald cypress; JS 17207
LS 3948
Cheilanthes horridula Maxon; rough lip fern; LH 5863
Cheilanthes tomentosa Link; woolly lip fern; JS 1 8003
Pellaea atropurpurea (L) Link; purple cliffbrake fern; JS 1 0054;
LS3947
ANGIOSPERMS -Monocots
Agavaceae
Agave americana L.; American century plant; JS 18005
Dasylirion texanum Scheele; Texas sotol; JS 1 0563 __
Nolina lindbeimeriana (Scheele) S. Wats.; devil's shoestring,
Pellaea ovata (Desv.) Weatherby; ovate leaf cliffbrake; JS
8771; 10055
Pellaea wrightiana Hook.; Hook's cliffbrake fern; JS 8490,
10058
JS 10574
Nolina texana S. Wats.; sacahuista; LH 4773
Yucca reverchoniiJtel; Plateau yucca; JS 10649
Yucca rupicola Scheele; Texas yucca; JS 1 0372
510
Fisch. & Mey.) Celarier & Harlan; King Ranch bluestem; JS
1 0006; KW s.n.; LS 3768; LH 5310
Bothriochloa laguroides (DC.) Herter ssp. torreyana (Steud.)
Allred & Gould; silver bluestem; JS 10007; 10667; KW s.n.;
FG 8278; LS 3765; 4407
Bouteloua barbata Lag.; sixweeks grass; JS 1 0226 a
Bouteloua curtipendula (Michx.) Torr. var. caespitosa Gould
& Kap.; sideoats grama; JS 10584; FG 8289; MM 5514;
LS 3907
Bouteloua curtipendula (Michx.) Torr. var. curtipendula ; side
oats grama; JS 9979; FG 8289; LS 3906
Bouteloua hirsuta Lag.; hairy grama; FG 8262; 8451; 8452;
LS 3918
Bouteloua rigidiseta (Steud.) Hitchc. var. rigidisetajexas grama;
JS1 0397; FG 8284; LS 3762
Bouteloua pectinata Featherly; tall grama; LS 4543
Bouteloua trifidalhurb:, red grama; FG 8453; 8281 ; LS 4047
Bouteloua uniflora Vasey; Neally grama; FG 8255; MM 5520
*Bromu$ arvensis L.; Japanese brome; JS 1 0404; JS &TD 1 0728;
LS 3124; 3724
*Bromu$ cartharticus Vahl; rescue grass; JS 10391; FG 8307;
LS 3723
Bromus pubescens Muhl. ex Wild.; hairy brome; JS &TD
10729
tctorum ; cheat grass; JS &TD 1 0727;
Buchloe dactyloides (Nutt) Engelm.; buffalo grass; JS 10473;
FG 8303; LS 3731
Cenchrus spinifex Cav.; common sand bur; JS 9953; FG 8280;
LS4536
Chloris andropogonoides Fourn.; slimspike windmill grass; FG
8295; LS 3766
Chloris cucullata Bisch.; hooded windmill grass; JS 10677
Chloris subdolichostachya Muell.; short-spike windmill grass;
LS4406
Chloris verticillata Nutt.; tumble windmill grass; JS 10424; LS
3784; 3911; 41 14
Xynodon dactylon (L.) Pers.; bermuda grass; JS 10676; LS
3751
*Desmazeria rigida (L) Tutin; cat grass; JS 1 0630; LS 3774
Dichanthelium acuminatum (Sw.) Gould & C.A. Clark var.
acuminatum ; woolly rosette grass; LS 3844
Dichanthelium acuminatum (Sw.) Gould & C.A. Clark var.
lindheimeri (Nash) Gould & CA Clark; Lindheimer pan-
icgrass;JS 15971
(Nash) Gould; Scribner's rosette grass; LS 3797
ichanthelium pedicellatum (Vasey) Gould; cedar
grass; FG 8262; 8454
Digitaria californica (Benth.) Henr.; Arizona cottontop; KW s.n.
Digitaria ciliaris (Retz.) Koeler; southern crabgrass; LS 4090
Digitaria cognata (Schult.) Pilger; fell witchgrass; FG 8252
Digitaria pubiflora (Vasey) Wipff; Carolina crabgrass; LS 41 1 3;
*Echinochtoa cobna (L.) Link; Jungle rice; JS 10456; FG 8298;
LS 3924; 4091; 4566
*Echinoch!oa waited (Pursh) A. Heller; Coast cockspur; JS
10675
Elymus canadensis L. canadensis ; Car
*Eragrostis cilianensis (All.) Vign. ex Janchen.; stinkgrass; JS
1 0461 ; FG 8467; 8309; LS 405 1 ; 4068
Eragrostis curtipedicellata Buckley; gummy lovegrass; LS
Eragrostis intermedia Hitchc; plains lovegrass; FG 8287; KW
s.n.;LS3909;4108
Eragrostis lugens Nees; mourning lovegrass; FG 8287
Eragrostis pectinacea Michx.) Nees ex Steud.; tufted lovegra
JS 102266
*Eragrostis superba Peyr.; Wilmann's lovegrass; LH 5319
Eriochloa contracta Hitchc.; prairie cupgrass; FG 8300
Eriochloa sericea (Scheele) Munro ex Vasey; Texas cupgrass;
KW s.n.; LS 3798
Erioneuron pilosum (Buckley) Nash; hairy erioneuron;JS 10403;
JS &TD 1 0725; LS 3756
Hilana belonged (Steud.) Nash.; common curly mesquite; JS
1 0043; FG 8243; LS 391 4
Hordeum jubatum L.; foxtail barley; FG 8274
Hordeum pusillum Nutt.; little barley; JS 1 0471 ; LS 3727
*Hordeum vulgare L; barley; JS 1 0443
Leersia oryzoides (L) Sw.; rice cutgrass; JS 18014
Leptochloa dubia (Kunth.) Nees; green sprangletop; JS 10445;
1 061 0;KWs.n.;LS 4077
Leptochloa panicea (Retz.) Ohwi ssp. mucronata (Michx.)
Nowack; red sprangletop; JS 1 0468; KW s.n.
Limnodea arkansana (Nutt.) LH. Dewey; Ozark grass; JS &TD
1 0755; KWs.n.;LS 3729
* Lolium perenne L.; perennial ryegrass; JS 10459; 10797; LH
5028
Melica nitens (Scribn.) Nutt, ex Piper; three flower melic; JS
1 0450; 1 0590; KW s.n.; LS 3870
Muhlenbergia lindheimeri Hitchc; Lindheimer muhly; JS 10005;
10062; LS4118
Muhlenbergia reverchoniiVasey & Scribn.; seep muhly; JS&TD
1 0733; KW s.n.; MM 55 1 5; LS 4079
Muhlenbergia schreberi J.F. Gmel.; nimblewill muhly; JS
10577
Muhlenbergia utilis (Torr.) Hitchc; aparejograss; JS 15961
Nassella leucotdcha (Trin. & Rupr.) Pohl; Texas winter-grass; JS
1 042 1 ; 1 0578; FG 8270; LS 3728
Panicum capillare L; common witchgrass; JS 10408
•Panicum coloratum L.; Klein grass; LS 3846; 3942; 4568
Panicum dichotomiflorum Michx.; fell panicum; JS 10224
Panicum diffusum Sw.; spreading panicum; JS 10579a
Panicum hallii Vasey var. hallii; Halls panicum; JS 10677;FG
8249; LS 4050; 441 6; 4532
Panicum hians Ell.; gaping panicum; JS 1 801 5
Panicum obtusum Kunth; vine mesquite; JS 10600; FG 8272;
LS 3908; 3956
Panicum virgatum L; switchgrass; JS 9985; FG 8470; LS 3927
•Paspalum dilatatum Poir.; dallisgrass; JS 10446; KW s.n; LS
Paspalum distichum L; knot grass; LH 5313
j;JS&TD1
512
Chaerophyllum tainturieri Ho
1 0706; KWs.n.;LS 3780
Cicuta maculata L.; spotted
ban; spadeleaf; JS 10013; 10616; LS
•ri Hook. var. tainturieri; chervil; JS &TD
k;JS&TD 10712;
) King 8
Xonium maculatum L; poison hemlock; JS 1063
*Cyclospermum leptophyllum (Pers.) Sprague e
Wilson; slim-lobe celery; LS 3858
Daucus pusillus Michx.; southwestern carrot; J!
Hydrocotyle v
10602
Polytaenia texana (J
//ex dec/dua Walt.; <
FG 8761 ;LS 3854; 3935
Aristolochia serpentaria L.; c
3841; 4221
Asteraceae
Achillea millefolium L; westei
]eratina havanensis (Kunth) King & H.E. Robins.; Havan
snakeroot; JS 8475; 10594; LS 3869; 4561
nbrosia artemisiifolia L; common ragweed; JS 10050
nbrosia confertiflora DC,- field ragweed; FG 8201; LS 4531
S8<R928
nbrosia psilostachya DC,- western ragweed; JS 18037; L
17; LS Amphiachyris
S (DC) Nutt.
Hydrocotyle prolifera Kellogg; whorled water pennywort;
LS3863
•rticillata Thunb.; water pennywort; JS 10014;
>se) Mathias & Constance;
Ptilimnium capillaceum (Michx.) Raf.; herbwilliam; JS 10418
Sanicula canadensis L; Canada sanide; JS 10444
Spermolepis divaricata (Walter) Raf. ex Ser.; forked scaleseed;
LH 5019
Spermolepis inermis (Nutt, ex DC.) Mathias & Constance;
spreading scaleseed; JS&TD 1 071 1; LS 3779; 3790; 3859
* Torilis arvensis (Huds.) Link; hedge parsley; JS 10400; KW
5Ji.;LS 3735
Apocynaceae
Apocynum cannabinum L.; dogbane; JS 10669
Asclepias aspemla (Decne.) Woodson; trailing milkweed; JS
1 0371;TT & AC 147; LS 4056
Asclepias oenotheroides Cham. & Schltdl.; hierba de Tiyores;
JS 10678; LS 4549
Asclepias texana A. Heller; Texas milkweed; JS 10057; 10561;
LS4104
Asclepias verticillata L.; whorled milkweed; JS 10652
Asclepias viridiflora Raf.; green antelope horn; JS 10677; KW
Cynanchum racemosum (Jacq.) Jacq. var. unifarium (Scheele)
Sundell; talayote; LH 5023
Funastrum crispum (Benth.) Schitr.; wavyleaf twinevine; JS
Matelea biflora (Raf.) Woods.; two-flower milkvine; JS 18033;
LS4058
Matelea edwardsensis Correll; plateau milkvine; JS 18034
Matelea gonocarpos (Walt.) Shinners; angular-fruit milkvine;
A. Johnst.; Cory dutchman's pipe; JS 10053;
Aphanostephus ramosissimus DC. var. ramosissimus; plains lazy
daisy; FG 8271 ; 8304; LS 3801 ; LH 4779; 5020
Aphanostephus skirrhobasis (DCJTrel. var. skirrhobasis; Arkan-
sas lazy daisy; JS 10420
Artemisia ludoviciana Nutt.; Louisiana sagewort; JS 18039;
LS4548
Baccharis neglecta Britton; Roosevelt-weed; LS 41 19; 4540
Baccharis salicina Jon & A. Gray; seepwillow; KW s.n.
Baccharis texana (Torr. & A. Gray.) A. Gray; prairie weed; JS
8487; 9950
Berlandiera betonicifolia (Hook.) Small; Texas greeneyes; JS
10380; 10644
Bidens bipinnata L.; Spanish needles; JS 10819
Bidens laevis (L.) Britton, Sterns & Poggenb.; smooth beg-
gartickjJS 15971; LS 4557
Boltonia diffusa Elliot; smallhead doll's daisy; JS 1 5969
Brickellia cylindracea A. Gray & Engelm.; brickell bush; LS
r; false boneset;JS 9983
Xarduus nutans L. var. macrocephalus (Desf.) Boivin; musk-
thistle; JS 1 0556; LS 3742; LH 5323
•Carduus tenuiflorus W. Curtis; slender bristlethistle; JS &TD
Centaurea americana Nutt.; American basket flower; JS
"Centaurea melitensis L; Malta centaurea; JS &TD 10720; KW
s.n.; LS 3741
Chaetopappa asteroides Nutt, ex DC.; common least daisy;
JS 9952; KW s.n.
Chaetopappa bellidifblia (A. Gray & Engelm.) Shinners; hairy
least daisy; JS 1 0423; KW s.n.; LS 3787; LH 4776; 4991
Chaetopappa effusa (A. Gray) Shinners; spreading least daisy;
JS 8495
Chaptalia texana Greene; nodding lettuce; JS 10646
Chrysactinia mexicana A. Gray; damianita; LS 3802
Cirsium ochrocentrum A. Gray; yellow spine thistle; JS 10455
Cirsium texanum Buckley; southern thistle; JS 10556; KW
s.n.; LS 3760
Cirsium undulatum (Nutt.) Spreng.; wavyleaf thistle; JS
*• Cirsium vulgare (Savi) Ten.; bull thistle; LH 5027
Conyza canadensis (L) Cronq. var. glabrata (A. Gray) Cronq,
Canada fleabane; JS 1 8040; KW s.n.; LS 4073
515
Chenopodium simplex (Torr.) Raf.; maple-leaf goosefbot; LS
Hypericum drummondii (Grev. & Hook.) Torr. & A. Gray; Drum-
mond St. John's wort; JS 1 8074
Hypericum mutilum L.; dwarf St. John's wort; JS 1 8075
Convolvulaceae
•Convolvulus arvensis L.; field bindweed; JS 1 0384
Convolvulus equitans Benth.; gray bindweed; JS 10673; LS
3778
Dkhondra carolinensis Michx.; grass pony weed; JS 10390;
I LS4210
Evolvulus sericeus Sw.; silver dwarf morning glory; JS 18076;
| LS3916
; Ipomoea cordatotriloba Dennst. var. cordatotriloba ; sharp-pod
I- morning glory; LS 3958
Ipomoea hederacea Jacq.; ivy-leaf morning glory; JS 8489
Ipomoea lindheimeri A. Gray; Lindheimer's morning glory;
Ipomoea purpurea (L.) Roth; Mexican morning glory; JS
p 10411
Cornaceae
iComus drummondii CA Mey.; roughleaf dogwood; JS 1 0589;
p KW s.n.; LH 5867
Crassulaceae
vCrassula aquatica (L.) Schoenl.; crassula; JS 18077
jr Sedum nuttallianum Raf.; Nuttall's stonecrop; JS 10463; LS
II 3794
Cucurbitaceae
I Cue urbita foetidissima Kunth. in H.B.K.; buffalo gourd; JS
I 10462; LS 3752
I bervillea lindheimeri (A. Gray) Greene; Lindheimer globeberry;
I JS 18078
Cuscutaceae
^Cwcuto sp.; dodder; JS 1 8079; LH 6256
Ebenaceae
I Diospyros texana Scheele; Texas persimmon; JS 18080; KW
in; LS 4201
Ericaceae
; kalypha ostryifolia Riddell; hop-hornbeam copperleaf; LS
ft 3770
kalypha phleoides Cav.; Lindheimer copperleaf; JS 10625;
If fG 8214; LS 3788
V kalypha radiansTon:, round copperleaf; JS 10060
tyythamnia humilis (Engelm. & A. Gray) Muell. Arg. var.
cury; KW s.n.; LS 3750
\l. Ingram; Plateau wild mercury; JS
I 10614; LH 4973
tenardia myricifolia (Scheele) S. Wats.; brush myrtlecroton;
Qurrtaesyce angusta (Engelm.) Small; blackfoot euphorbia;
JS 10662; KW s.n.; LS 3809; 4093
Chamaesyce fendleri (Torr. & A. Gray) Small; spurge; JS 10395
Chamaesyce glyptosperma (Engelm.) Small; ridgeseed eu-
phorbia; JS 10619
Chamaesyce maculata (L) Small; spotted spurge; JS 10037;
10653
Chamaesyce missurica (Raf.) Shinners; spurge; JS 99630
Chamaesyce nutans (Lag.) Small; eyebane; JS 9976; LS 3772;
4089
Chamaesyce prostrata (Aiton) Small; prostrate euphorbia; JS
9978; LS 4070; 4535; 4563; LH 5309
Chamaesyce serpens (Kunth) Small; mat spurge; JS 10761;KW
S.n.;LS 3771; 4045
Chamaesyce serpyllifolia (Pers.) Small; thymeleaf euphorbia;
JS 10653
Chamaesyce villifera (Scheele) Small; hairy euphorbia; JS
Croton capitatus Michx.; woolly croton; JS 1 0016
Croton fmticubsus Engelm. ex Torr.; bush croton; JS 10660;
LS 3840
Croton lindheimerianus Scheele; three-seed croton; JS
10017
Croton monanthogynus Michx.; one-seed croton; JS 10052;
FG 8293; LS 3937
Croton texensis (Klotzsch) Muell. Arg.; Texas croton; JS 10047;
1 0660; LS 4067
Euphorbia cyathophora Murray; wild poinsettia; KW s.n.; LS
4063
Euphorbia dentata h
.; toothed spurge; JS 10609; LH
10025;LS4102
Euphorbia roemeriana Scheele; Roemer euphorbia; JS 8477;
LS3791
Euphorbia spathulata Lan
4775
G.L. Webster; maidenbush;
abnormis Baill. var. abnormis; Drummond's leaf
Phyllanthus polygonoides Nutt, ex Spreng.; knotweed leaf
flower; JS &TD 1 0746; FG 8250; LS 3758
1 0564; LS 3805
Tragia amblyodonta (Muell. Arg.) Pax. & K. Hoffm.; dogtooth
noseburn; JS 10659
Tragia bentonici folia Nutt.; betony noseburn; JS 10376
Tragia nigricans Bush; dark noseburn; JS 10382
Tragia ramosa Torr.; catnip noseburn; JS 10634; FG 8254; LS
3795
Acacia greggii A. Gray; catdaw acacia; KW s.n.
Acacia roemeriana Scheele; roundflower catdaw; JS 18082;
LH4784
Amorpha fruticosa L; indigo bush amorpha; JS 18083
Astragalus distortus Torr. & A. Gray var. engelmannii (Sheldon)
M.E. Jones; bent pod milk vetch; JS &TD 10742
Astragalus nuttallianus DC.; small-flowered milk vetch; JS &
TD1 0748; KWs.a;LH 4761
Journal of the Botanical Research Institute of Texas 4(1)
516
Astragalus plattensis Nutt.; Platte River milk vetch; JS 10432
Astragalus wrightii A. Gray; Wright's milk vetch; LH 4764
Cercis canadensis L var. texensis (S. Wats.) M. Hopk.; Texas red-
bud; JS 8763; JS &7D 1 0734; KW s.n.; LS 3804; 41 95
Dalea aurea Nutt, ex Pursh; golden dalea; JS 1 0571; LS 3922
Dalea frutescens A. Gray; black dalea; KW s.n.
Dalea lasiathera A. Gray; purple prairie clover; JS 18084; LS
4078
Dalea nana Torn & A. Gray; dwarf dalea; JS 1 0645; LS 4542
Desmanthus acuminatus Benth.; sharp-pod bundleflower;
LS3903
KW s.n.;FG 825 1;LS 3904
Desmodium paniculatum (L.) DC; panicled tickdover; JS
10657;LH 5314
Desmodium sessilifolium (Torn) Torn & A. Gray; sessileleaf
tickdover; JS1 0628; KWs.n.
Eysenhardtia texana Scheele; Texas kidneywood; JS 10027;
1 0620; KWs.n.;LS 3852
Galactia texana (Scheele) A. Gray; Texas milkpea; JS 10040;
10657
Indigofera miniata Ortega; scarlet pea; JS 1 8085
Lathyrus pusillus Ell.; low peavine; LH 4780
Lespedeza texana Britt; Texas bush clover; LS 3926
Lespedeza virginica (L.) Britt.; slender lespedeza; JS 1 0429
Leucaena retusa Benth. in A. Gray; little-leaf leadtree; JS 1 0049;
DH s.n.; LS 3925
Lotus unifoliolatus (Hook.) Benth. van helleri (Britton) Kartesz
& Gandhi; Heller’s bird's-foot trefoil; JS 1041 0
Lupinus texensis Hook.; Texas bluebonnet; JS & TD 1 0737;
LS3792
*Medicago lupulina L; black medic; LS 3734
*Medicago minima (L) Bartalina; small medicago; JS & TD
10747; LS 3725
*Melilotus officinalis (L) Lam.; yellow sweetdover; JS 10605;
ir. biuncifera (Benth.) Barneby;
k. Gray; fragra
4782
Mimosa nuttallii (IX. ex Britton & Rose) B.L Turner; catdaw
sensitive brier; JS 18086
Pediomelum hypogaeum (Nutt ex Torn & A. Gray) Rydb. var.
scaposum (A. Gray) Mahler; subterranean Indian bread-
root; LH 4783
Prosopsis glandulosa Torn; mesquite; JS 10472; KW s.n.; LS
3919
Rhynchosia senna Gillies ex Hook. var. texana (Torn & A.
Gray) M.C. Johnst; Texas snoutbean; JS 18087; KW s.n.;
LS3781
Senna roemeriana (Scheele) Irwin & Barneby; two-leaf senna;
JS 1 0028; 10381; FG 8291 ; LS 3754
Styphnolobium affine (Tom & A. Gray) Walp.; Eve's necklacepod;
JS1 8088; KWs.n.;LS 4043
Vicia carolinianum Walt; Carolina vetch; JS 10430
Vicia ludoviciana Nutt.; deer pea vetch; JS & TD 10749, KW
s.n;LS4211
Quercus buckleyi Nixon & Dorr; Buckley oak; JS 9956; KW s.n.;
LS4194
Quercus fusifbrmis Small; Plateau live oak; JS &TD 10714;KW
s.n.;LS4202
Quercus laceyi Small; Lacey oak; JS 10631; DH s.n.; KW sji;
LS 4111
Quercus marilandica Muenchh.; blackjack oak; JS 10671; KW
s.n.;LS4044;4197
Quercus muehlenbergii Engelm.; chinquapin oak; JS 10039;
KW s.n.; LS 4222
Quercus pungens Liebm. var. vaseyana (Buckley) C.H. Mull.;
Vasey shin oak; JS 9992
Quercus sinuata Walter var. breviloba (Torn) GH. Mull.; bastard
oak; JS 9957; 1 061 8; JS & TD 1 071 3; KW s.n.; LS 41 99
Quercus stellata Wang. var. stellata .; post oak; JS 10670; KW
s.n.;LS4196
Corydalis curvisiliqua Engelm. ssp. curvisiliqua-, scrambled
eggs; LH 4765
Corydalis micrantha (Engelm.) A. Gray; scrambled eggs; JS &
TD10687;KWs.n.
Garryaovata Benth s p lindnemun (Ton.) Dahlmg iIMjs*
JS 8765; KW s.n.
Gentianaceae
Centaurium beyrichii (Torn & A. Gray) B.L. Rob.; mountain pink;
JS 1 0393; 1 0026a; 1 0632; FG 8259; LS 441 5
Centaurium calycosum (Buckley) Fernald; Buckley centaury;
JS 1 0581 ;LH 5022
Eustoma exaltatum (L) Salisb. ex G. Don ssp. russellianum
(Hook.) Kartesz; showy prairie gentian; JS 1 81 32; L$ 4413
Geraniaceae
•Erodium cicutarium (L) L'Her. ex Aiton; alfilaria; JS &TD 10705;
KWs.n.;LS3733
Erodium texanum A. Gray; Texas filaree; JS &TD 10704; JS
1 0453; LH 5861
Geranium carolinianum L; Carolina geranium; JS 10457
Geranium texanum (Trel.) A. Heller; Texas geranium; LS4209
Hippocastanaceae
Aesculus pavia (Sarg.) Cornell var. pavia ; red buckeye; JS &
TD 10717
Hydrophyllaceae
Nama jamaicense L; Jamaican weed; JS 1 0659; LH 4757
Nemophila phacelioides Nutt.; large-flower nemophila; JS
TD 10694
Phacelia congesta Hook.; spike phacelia; JS 10552; KW sjv;
LH4975
Juglandaceae
Carya Ulinoinensis (Wang.) Koch; pecan; JS 9988; KW s.n.,
4224
Juglans major (Torn) A. Heller; Arizona walnut; LS 4087
Juglans microcarpa Berland.; little walnut; JS 9962; 106 *
DH s.a; LS 3830
Juglans nigra L; black walnut JS 1 0562; KW s/i.
Singhurst et al.(
517
Krameriaceae Mitreola petiolata (J.F. Gmel.) Torr. & A. Gray; lax hornpod; JS
Krameria lanceolata Torr.; trailing krameria; JS 1 8089 1 8091 ; LS 4099
Hedeoma acinoides Scheele; slender hedeoma; JS 10399;
LH4766
Hedeoma drummondii Benth.; Drummond hedeoma; JS
1 0396; FG 8254; LS 4046
Hedeoma nana (Torr.) Briq.; low hedeoma; JS 9977
Hedeoma reverchonii (A. Gray) A. Gray var. reverchonii j rock
; hedeoma; JS 10565; 10575
Hedeoma reverchonii (A. Gray) A. Gray var. serpyllifolia (Small)
Irving; thymeleaf hedeoma; LS 3808; 391 7
•Lamium amplexicaule L; henbit; JS & TD 1 0685; LH 5579
•Marrubium vulgare L; common horehound; JS 10560; FG
8271; LS 3740
Monarda citriodora Cerv. e
n beebalm; FG 8276;
Lythraceae
Ammanniacoccinea Rottb.; Valley redstem; JS 18092
Lythrum ovalifolium Koechn.; low loosestrife; JS 10063
Malpighiaceae
Galphimia angustifolia Benth.; narrowleaf thryallis; JS 18093;
FG8211;LS3860
Malvaceae
Abutilon fruticosum Guill. & Perrottet; Texas Indian mallow; JS
1 0658; FG 8256a; LS 3799; 4538
Callirhoe pedata (Nutt, ex Hook.) A. Gray; finger poppy mal-
low; JS 10438
Rhynchosida physocalyx (A. Gray) Fryxell; buff petal; LS 3776
Sida abutifolia Mill.; spreading sida; JS 10038; 10647; 10800;
FG 8247; LS 3789
Sida lirtdheimeri Engelm. & A. Gray; showy sida; JS 10061;
Monarda punctata L. ssp. punctata var. intermedia (E.M. Mc-
Clint. & Epling) Waterf.; spotted beebalm; JS 1 0434
*Nepeta cataria L.; catnip; JS 10629
Physostegia angustifolia Fernald; Edward's lionsheart; JS 1 0569;
10663; LS 441 2; LH 4983
Salvia farinacea Benth.; mealy cup sage; JS 10383; FG 8257;
Salvia reflexa Hornem.; lanceleaf sage; FG 821 5
Salvia roemeriana Scheele; cedar sage; JS 8480; LS 3867
Scutellaria drummondii Benth,- Drummond skullcap; JS 1 0401 ;
JS &TD 10710; KW s.n; LS 3782
Scutellaria ovata Hill. ssp. ovara; eggleaf skullcap; JS 10035
Stachys crenata Raf.; shade betony; LH 5862
Teucrium canadense L var. canadense; American germander;
JS 10597; LS 3930
Trichostema brachiatum L.; flux weed; JS 18090; LS 4074;
4080; LH 6254
Warnockia scutellarioides (Engelm. & A. Gray) M.W. Turner;
prairie brazoria; JS 1 0426; KW s.n.; LH 4990
Undera benzoin (L.) Blume; spicebush; JS 10583
Lentibularaceae
Utricularia gibba L.; cone-spur bladderwort; JS 1 0001
Unaceae
Unum berlandieri Hook. var. berlandierf, flax; JS 9982
Unum rupestre (A. Gray) Engelm. ex A. Gray; rock flax; JS
10377; LS 4059
LH 5555
^tzelia oligosperma Nutt, ex Sims.;
LS4408
mentzelia;JS 10572
cknettle;JS 8768; 10554;
; chicken-thief; JS 10452;
H J. Thomp. & Zavort.;
Loganiaceae
Buddleja racemosa Torr. ssp. incana
; butterfly-bush; JS 8474
man; wand
10658
Sphaeralcea angustifolia (Cav.) G. Don. var. angustifolia ; leaf
globe mallow; KW s.n.
Menispermaceae
Cocculus carolinus (L) DC.; Carolina snailseed vine; JS 8491;
1 001 2; KW s.n^ LS 3953
Molluginaceae
Glinus radiatus (Ruiz & Pavon) Rohrb.; spreading sweetjuice;
JS 18094
Mollugo verticillata L; green carpetweed; JS 1 8095
Moraceae
Morus microphylla Buckley; Texas mulberry; JS 9990; 10551;
KWs.n.;LS4219
Morus rubra L; red mulberry; JS 10593
Nyctaginaceae
Boerhavia diffusa L.; scarlet spiderling; JS 18096
Mirabilis albida (Walt.) Heimerl; four-o'clock; KW s.n.; LS 3796;
4066; 4545; 4559
Mirabilis nyctaginea (Michx.) MacMill; heartleaf four-o'clock;
JS 10557
•Mirabilis jalapa L; man/el of Peru; LS 4530
Mirabilis linearis (Pursh.) Heimerl; linearleaf four-o'clock; JS
10373; JS &TD 10731;LH 6258
Forestiera pubescens Nutt,- elbowbush; JS 8766; 10636; KW
s.n,- LS 4200
Forestiera reticulata Torr.; netleaf foresteria; JS 18097
Fraxinus texensis (A. Gray) Sarg.; Texas ash; JS 1 8098
Menodora longiflora A. Gray; showy menodora; LS 4054
Calylophus berlandieri Spach ssp. berlandieri; halfshrub sun-
drops; JS 18099
Ludwigia peploides (Kunth. in H.B. K.) P.H. Raven; floating
primrose-willow; JS 18100
Ludwigia repens J.R. Forst; roundleaf seedbox; JS 18101
i (Torr.) Norn
LS 3745
Galium virgatum Nutt.; southwest bedstraw; LS4213
fboustonia pusilia Schoepf ; tiny bluet; JS 1 811 9
*Sherardia arvensis L; spurwort; JS & TD 1 0690
. Stenaria nigricans (Lam.) Terrell var. nigricans; fineleaf bluets;
JS 8488; 10374; FG 8263; KW sjk LS 3807
.; hoptree; JS 10820
Jhamnosma texana (A Gray)Torr.; dutchman's breeches; JS
f 18120; LS 3810
Zanthoxylum hirsutum Buckley; pricklyash; JS 18121; KW s.n.
Populus deltoides Bartram ex Marsh, ssp. deltoides; cot-
i tonwood;JS 18122
; Salix nigra Marsh.; black willow; JS 1 81 23; LS 4223
Sapindaceae
Sapindus saponaria L. var. drummondii (Hook. & Arn.) L.D.
Benson; western soapberry; JS 1 001 1; 1 0591
Ungnadia speciosa EndL; Mexican buckeye; JS&TD 10167;
KWs.n.;LS4215
Sapotaceae
\ Sderaxylon lanuginosum Michx. ssp. albicans (Sarg.) Kartesz &
; Gandhi; gum bully; JS 9993; 1 0633; KW s.n.; LS 3944
Scrophulariaceae
Agalinis edwardsiana Pennell; Plateau gerardla; JS 18124
Agalinis homalantha Pennell; San Antonio false foxglove;
' JS 18125
Bacopa monnieri (L) Pennell; coastal water hyssop; JS 1 81 26
1 Buchnera americana L; American bluehearts; JS9966; 10616;
| LS3862
leucospora multifida (Michx.) Nutt; narrow leaf conobea; JS
10061, 10221, 10599, 1 5967; LS 3857
: Undemia dubia (L) Pennell var. anagaiiidea (Michx.) Cooperr;
clasping false pimpernel; JS 10465
Maurandella antirrhiniflora (Humb. & Bonpl. ex Willd.) Rothm.;
snapdragon vine; JS 8492; LS 3827
Mecardonia procumbens (Mill.) Small; prostrate water hyssop;
JS 10666; LS 3856
Nuttalianthus canadensis (L) DA Sutton; Texas toadflax; JS &
TD 10709; KWs.n.
Penstemon cobaea Nutt; foxglove; JS 10550
Penstemon triflorus A. Heller ssp. trifhrus; Heller penstemon;
JS 18127
* Verbascum thapsus L; flannel mullein; JS 10608; FG 8256;
LH5025
"Veronica agrestis L; wayside purslane; JS 10607
"Veronica arvensis L; common speedwell; LS 371 8
Veronica peregrina L; purslane speedwell; JS &TD 10692;
I LS3744
Chamaesaracha sordida (Dunal) A Gray; hairy false nightsha-
de; JS 10474
*Datura stramonium L; jimsonweed; JS 8484
Nicotiana repanda Willd. ex Lehm.; fiddle leaf tobacco; JS
8764
Nicotiana trigonophylla Dunal; desert tobacco; LS 3868
Physalis cinerascens (Dunal) Hitchc var. cinerascens; small-
flower groundcherry; JS 8503; LS 3777; 3950; 4065
Physalis mollis Nutt var. mollis; field groundcherry; KW s.n.
Solanum citrulli folium A. Braun var. citrullifolium; melon
nightshade; JS 10412
Solanum dimidiatum Raf.; western horse nettle; JS 9955
Solanum elaeagnifblium Cav.; silver leaf nightshade; JS 1 0051 ;
FG 8293; LS 3736
Solanum ptycanthum Dunal; American nightshade; KW s.n.;
LS4553
Solanum rostratum Dunal; buffalo bur; JS 9954; LS 4075
i coronopus (Dunal) A. Gray; green false i
Qhtshade; KW s.n.
Qiamaesaracha edwardsiana Averett; Plateau flase nights!
Hermannia texana A. Gray; Mexican mallow; LS 3!
Ulmaceae
Celtis laevigata Willd. var. laevigata; sugar hat
18128;KWs.n.
Celtis laevigataWiWd. var. reticulata (Torn) L.D. Bens
hackberry; JS 9994; DH s.n.; LS 4198
5625 Ig°
\merican elm; JS 9989
Ulmus crassifolia Nutt.; cedar elm; JS &TD 10735; DH s.n,- KW
s.n.; LS 4092
Ulmus rubra Muhl; slippery elm; JS 10003; KW s.n.
Boehmeria cylindrica (L) Sw.; false nettle; JS 100T9JLS4103
Parietaria pensylvanica Muhl. ex Willd; Pennsylvania pelitory;
JS &TD 10691; KW sji.;LS 3719
Urtica chamaedryoides Pursh; heart-leaf nettle; JS &TD 1 0699;
LH4760
Valerianaceae
Valerianella amarella (Lindh.ex Engelm.) Krok; hairy com salad;
JS 1 0427; KWs.n.;LH 4781
Valerianella stenocarpa (Engelm. ex A. Gray) Krok; bigflower
comsalad; JS &TD 10721 ; KW s,n.
ssima (Gillies & Hook.) Troncoso; whitebrush; JS
10002; LS 391 3
Glandularia bipinnatifida (Nutt) Nutt. var. bipinnatifida; Dakota
vervain; FG 8255; KW s.n.; LS 3747; 4061
Glandularia pumila (Rydb.) Umber; pink vervain; JS & TD 1 0754;
KWs.n^ LH4756
Phyla nodiflora (L) Greene; frogffuit; JS 1 81 29; LS 3849
* Verbena brasiliensis Veil.; Brazilian vervain; LS 3865; 4098
Verbena canescens Kunth; gray vervain; JS &TD 1 0753; FG8260;
KW s.n^ LS 3726; 3905; 396 1; 3955
Verbena halei Small; slender vervain; JS 1045
1 ; LS 391 5
Jissr
521
Hatch, S.L, K.N. Gandhi, and L.E. Brown. 1 990. Checklist of the vascular plants ofTexas.Tex. Agric Exp. Sta.MP-1 655,
Texas A&M Univ., College Station.
Heck, K.L, G. Belle, and D. Simberloff. 1975. Explicit calculation of the rarefaction diversity measurement and the
determination of sufficient sample size. Ecology 56:1459-1461.
Hqagland, B.W. 2000. The vegetation of Oklahoma: a classification of landscape mapping and conservation
planning. Southw. Naturalist. 45385-420.
Hurlbert, S.H. 1 971 .The concept of species diversity: a critique and alternative parameters. Ecology 52557-586.
Hurlbert, S.H. 1 978. The measurement of niche overlap and some relatives. Ecology 59:67-77.
Hurlbert, S.H. 1 982. Notes on the measurement of overlap. Ecology 61 :44-49.
Jaccard, P. 1 902. Gesetze der Pflanzenvertveilung in der alpinen region. Flora 90:349-377.
Jones, S.D., J.K. Wipff, and P.M. Montgomery. 1 997. Vascular plants of Texas. Univ. ofTexas Press, Austin.
• Lbca Geosystems 2008. Imagine Software, Version 8.7. Available online at http://www.leica-geosystems.com.
Heerbrugg, Switzerland.
Looman, J. and J.B. Campbell 1 960. Adaptation of Sorenson's K (1 948) for estimating unit affinities in prairie vegeta-
tion. Ecology 41:410-416.
McGuinness, K.A. 1 984. Equations and explanations in the study of species-area curves. Biol. Rev. 59:423-440.
NatureServe 2008. NatureServe Explorer Database (ttpV/www.naturese rve.org/). Arlington, Virginia.
Neill A.K. AND H.D. Wilson. 2001 . The vascular flora of Madison County, Texas. Sida 1 9:1 083-1 1 21 .
Ney-Nifle, M. and M. Mangel 1 999. Species-area curves based on geographic range and occupancy. J.Theor. Biol.
196:327-342.
NRCS, USDA. 201 0.The PLANTS Database (http://plants.usda. gov.). National Plant Data Center, Baton Rouge, LA.
Riskind, D.H. and D.D. Diamond. 1988. An introduction to environments and vegetation. In: B.B. Amos and F.R.
Gehlbach, eds. Edwards Plateau vegetation: plant ecological studies in central Texas. Baylor University Press,
Waco,TX. Pp. 1-16.
Singhurst, J.R., J.c. Cathy, D. Prochaska, H. Haucke, G.C. Kroh, and W.C. Holmes. 2003. Vascular flora of Gus Engeling
Wildlife Management Area, Anderson County, Texas. Southe. Naturalist 2(3)347-368.
Singhurst, J.R., L.L. Sanchez, D. Frels, Jr,T.W. Schwertner M. Mttchell S. Moren, and W.C. Holmes. 2007. An annotated
• list of the vascular flora of Mason Mountain Wildlife Management Area. Southeastern Naturalist 6: 683-692 +
Supplementary Appendix 1 . [species checklist], http://dx.doi.Org/1 0.1 656/S497.S1), pp. 1-19+ Supplementary
Appendix 2. [plant associations], http://dx.d0i.0rg/l 0.1 656/S497.S2). Pp. 1-5.
Sbrenson,! 1948. A method of establishing groups of equal amplitude in plant sociology based on similarity
I of species content, and its application to analyses of the vegetation on Danish commons. Det Kongelige
I Danske Videnskabernes Selskab. Biologiske Skrifter Bind V Nr 4. 1 948. 1. Kommission Hos Ejnar Munksgaard.
Kobenhavn.
■ TerletzkEy, P.A. and O.W. Van Auken. 1 996. Comparison of cedar glades and associated woodlands of the southern
Edwards Plateau. Texas J. Sci. 48:55-67.
Turner B.L, H. Nichols, G. Denny, and O. Doron. 2003. Atlas of the vascular plants ofTexas. Sida, Bot. Misc 24. (2
: vols.). Bot. Res. I nst. Texas, Fort Worth.
University of Texas Bureau of Economic Geology. 1981. Llano Sheet, 1 :250,000. The University ofTexas, Austin, TX.
NRCS, USDA. 201 0.The PLANTS Database (http://plants.usda.gov) National Plant Data Center, Baton Rouge, LA.
Ugiand, K.I., J.S. Gray, and K.E. Ellingsen. 2003.The species-accumulation curve and estimation of species richness.
J- Animal Ecol. 72:888-897.
: v*n Auken, O.W. 1988. Woody vegetation of the southeastern escarpment and plateau. In: B.B. Amos and F.R.
Gehlbach, eds. Edwards Plateau vegetation: plant ecological studies in central Texas. Baylor University Press,
Waco,TX.Pp. 43-55.
I Wallace, R.K., Jr 1 981 . An assessment of diet-overlap indices. Trans. Amer. Fisheries Soc. 1 1 0:72-76.
Wluams, j.k. AND W. Lutterschmidt. 2006. Species-area relationships indicate large-scale data gaps in herbarium
If Elections. Lundellia 9:41-50.
522
Journal of the Botanical Research Institute ofTexas4(1)
ANNOUNCEMENT
The 2010 Applications for the Delzie Demaree Travel Award
Applications for the 2010 Delzie Demaree Travel Award should include a letter from the applicant telling
major professor. Please send letters of application to: Dr. Donna M.E. Ware, P.O. Box 8795, Herbarium, Biol-
ogy Department, The College of William and Mary, Williamsburg, VA 23185-8795, U.S.A. 1-757-221-2799;
Email: ddmware@wm.edu. The period for receiving applications will end three weeks prior to the date of
the symposium if a sufficient number of applications are in hand at that time. Anyone wishing to apply
after that date should inquire whether applications are still being accepted before applying. The 57th Annual
Systematics Symposium, Missouri Botanical Garden dates for 2010 are 15-16 October 2010.
The Delzie Demaree Travel Award was established in 1988 honoring Delzie Demaree who attended
35 out of a possible 36 symposia before he died in 1987. Delzie Demaree was a frontier botanist, explorer,
discoverer, and teacher. His teaching career as a botanist began in Arkansas at Hendrix College in 1922. He
also taught botany at the University of Arkansas, Navajo Indian School, Yale School of Forestry, Arkansas
A&M, and Arkansas State University at Jonesboro where he retired as professor emeritus in 1953. One of
the things he enjoyed most as a botanist was assisting students with their field botany research.
Anyone interested in making a contribution to Delzie Demaree Endowment Fund, which supports the
travel award, may make contributions by VISA or MasterCard or by a check, payable to Botanical Research
Institute of Texas, to Barney Lipscomb, 500 E 4th Street, Fort Worth, TX 76102-4025, U.S.A. 1-817-332-
7432; Email: barney@brit.org. Thank you.
I. Bot. Res. Inst. Texas 4(1): 522. 1
ANNOTATED CHECKLIST OF THE VASCULAR PLANTS OF FORT HOOD, TEXAS
Laura L. Hansen
Department of the Army
Directorate of Public Works
Environmental Division
Natural Resources Management Branch
4219 77th Street
Fort Hood, Texas 76544-5028, USA
ABSTRACT
in Bell and Coryell counties,
RESUMEN
11 y Coryell, Texas. Este
INTRODUCTION
Fort Hood Military Reservation, located in central Texas (Fig. 1), is 88,742 hectares. There is no agreement
as to what vegetation region defines this part of Texas. Diggs et al. (1999) place it within the Lampasas Cut
Plain. LBJ School of Public Affairs (1978) considers it to be in the Edwards Plateau Ecoregion. Others have
placed it in the Cross Timbers and Prairies (Gould 1962; Correll & Johnston 1979; Diggs et al. 1999) and
in the Blackland Prairie (Mahler 1988). The flora shows affinities to all of these ecoregions.
The Army acquired the lands of Fort Hood by eminent domain beginning in 1942. Prior to this, the
lands of Fort Hood were privately held and used for livestock grazing and crop production. Fort Hood is an
Army base used primarily for military training, including large-scale troop and ground vehicle maneuvers,
live fire weapons and aviation training. There are over 45,000 active duty Army personnel, as well as Army
National Guard personnel training on Fort Hood. The land is also managed for multiple use, including
recreation, fishing, hunting, and livestock grazing.
The first known checklist of Fort Hood was compiled by CEMML (1994). It was based on an inventory of
the Fort Hood Herbarium (HOOD) and additional field collections during 1992-1993. The list documented
642 taxa. However, 67 of these were deleted from the list, as follows: 29 names were synonyms, 28 were
corrected to a taxa already listed, and 10 lacked vouchers and were out of range. Thus, the number of taxa
was reduced to 575.
The purpose of this present study is to update the checklist and provide information on frequency of
occurrence and habitat for each taxon.
SITE DESCRIPTION
Fort Hood is located in Bell and Coryell counties, from 31.0105 N to 31.3929 N and 97.4676 W to 97.9121 W.
Elevation ranges from 180 m to 375 m above mean sea level. Terrain varies from riparian corridors, alluvial
bottomlands, rolling grasslands, steep slopes and canyons, and flat topped mesas. The underlying geology
is Edwards Limestone, Comanche Peak Limestone, Walnut Clay, Paluxy Sand, and Glen Rose Limestone.
Soils are classified as clay loam, silty clay, sandy loam, and limestone rock outcrop (NRCS 2009b).
Journal of the Botanical Research Institute of Texas 4(1)
524
I . Location of Fort Hood Military Reservation in Tc
The climate is characterized by long, hot summers and mild winters. Mean high temperature for July is
35°C, with a low of 23°C. January mean high and low are 15°C and 5°C, respectively. Annual precipitation
averages 69 cm, with most rain falling in the spring and fall (Weatherbase 2009).
Historic vegetation likely consisted of tall- or mixed-grass prairies, oak savannas, oak shrublands, oak-
juniper woodlands, riparian forests, and floodplain forests (Bray 1904; Palmer 1920; Tharp 1939; Gehlbach
1988; Van Auken et al. 1979, 1980, 1981). Fire plays a role in these communities and many native plants
are fire adapted (Bray 1904; Smeins 1980; Reemts & Hansen 2008).
Current vegetation is classified as forest (8%), woodland (40%), shrubland (6%), grassland (34%), and
not vegetated/developed (12%), with 35 associations described and 70 classes (Teague & Reemts 2007). Major
forest communities include Quercus buckleyi-Fraxinus texensis-Juniperus ashei Forest and Acer grandidentatum-
(Quercus muehlenbergii)/Carex edwardsiana Lampasas Cutplain Forest. Woodlands are primarily Quercus
fusiformis-Quercus buckleyi-Ulmus crassifolia/Schizachyrium scoparium Woodland and Quercus stellata- (Quercus
marilandica, Ulmus crassifolia)/Schizachyrium scoparium Woodland, Juniperus ashei-Quercus sinuata var. brevi-
loba Woodland, and Prosopis glandulosa/Bouteloua curtipendula-Nassella leucotricha Woodland. Shrublands
are dominated by Quercus sinuata var. breviloba Shrubland. Grassland areas include disturbed herbaceous
vegetation and Schizachyrium scoparium-Bouteloua curtipendula-Nassella leucotricha Herbaceous Vegetation
(Teague & Reemts 2007).
No federally listed plant species are known to occur on Fort Hood. However, three federally listed birds
are known on Fort Hood. These include, the golden-cheeked warbler (Dendroica chrysoparid), the black-
capped vireo (Vireo atricapillus ), and the bald eagle ( Haliaeetus leucocephalus ). Golden-cheeked warbler is a
Hansen, Flora of Fort Hood, Texas
525
neotropical migrant songbird that nests only in central Texas. Its preferred habitat consists of old-growth
mixed woodlands, where Ashe juniper is mature enough to have shredding bark (Pulich 1976; Kroll 1980;
Ladd & Gass 1999; Dearborn & Sanchez 2001; Baccus et ai. 2007). Black-capped vireo is also a neotropical
migrant songbird. Its preferred habitat consists of low, scrubby clumps of mainly deciduous species, with low
Ashe juniper cover and open spaces in between clumps (Graber 1961; Grzybowski et al. 1994; Grzybowski
1995). Bald eagle is a winter resident on the eastern side of the installation, along the shores of Belton Lake.
METHODS
Additional field collecting was conducted from 1995 to 2010. Voucher specimens were collected according
to standard collecting methods. Material was keyed with appropriate floras (Bailey 1949; Stanford 1976;
Correll & Johnston 1979; Diggs et al. 1999; FNA 1993+) and compared to specimens at BAYLU, HOOD,
TEX, and LL. Specimens were deposited at HOOD, with some specimens and duplicates located at BAYLU,
CMML, HABAYC, HPC, LL, MO, SWT, TAES, and TEX. Herbarium searches yielding additional taxa
were conducted at HABAYC, HPC, LL, and TEX. Nomenclature follows NRCS (2009a), with the following
exceptions: Buchnera floridana, Centaurium muehlenbergii, and Phlox villosissima follow Turner et al. (2003)
and Mirabilis latifolia follows FNA (1993+). Native is defined as native to Texas, and introduced is defined
as introduced to Texas (as list in Nesom 2009).
The checklist includes all native and introduced taxa known to occur on the lands of Fort Hood. Many
ornamental species, planted by the previous land owners, have naturalized or persisted, and are thus included
in this checklist. However, plantings of ornamental species in the garrison areas are not included.
RESULTS
This study documents an additional 439 taxa to the checklist of Fort Hood, bringing the total to 1,014 taxa
(species and below) in 896 species, 513 genera, and 120 families. This represents 43% of the 2,376 taxa
recorded for north central Texas (Diggs et al. 1999). There are 3,793 voucher specimens in the Fort Hood
Herbarium (HOOD). Seven hundred ninety-eight are presumed new county records, with 245 records from
Bell County and 553 from Coryell County.
Families with the greatest number of taxa are the Poaceae (151), Asteraceae (133), Fabaceae (80), Cyper-
aceae (42), Euphorbiaceae (37), and Lamiaceae (26). These six families comprise 46.3% of the flora. Other
large families include Apiaceae (23), Brassicaceae (21), Onagraceae (20), Rosaceae (20), Scrophulariaceae (20),
and Apocynaceae (19). Genera with the greatest number of taxa are Carex (13), Cyperus (12), Dichanthelium
(11), Eragrostis (11 \Juncus (10), Chamaesyce (9), Dalea (8), Quercus (8), Panicum (8), AUium (7), Croton (7),
Euphorbia (7), Gaura (7), and Paspalum (7).
There are 166 taxa introduced to Texas in 165 species (Nesom 2009) recorded, comprising 16.4% of
the flora of Fort Hood. Twelve taxa (Table 1) have a global or state ranking (Carr 2005; Poole et al. 2007),
representing 1.2% of the flora. Opuntia engelmannii var. linguiformis, which is ranked G5T1QS1, is not in-
cluded in Table 1, because I am certain it was planted, as it is only found at a few old homesites. Two taxa
are endemic to north central Texas (Diggs et al. 1999), Yucca pallida and Croton alabamensis var. texensis.
Taxa that are endemic to the state and restricted to the Edwards Plateau (Carr 2007) include Argythamnia
simians. Astragalus wrightii, Carex edwardsiana, Chaetopappa bellidifolia, Monarda punctata ssp. punctata var.
intermedia, Muhlenbergia involuta, Parthenocissus heptaphylla, Passiflora affinis, Prunus serotina var. eximia,
%ntx platanifolius ssp. platanifolius, Vitis monticola, and Yucca rupicola.
The following taxa are endemic to Texas but not restricted to the Edwards Plateau Ecoregion (Carr
2007): Amphiachyris amoena, Arabis petiolaris, Argemone aurantiaca. Astragalus rejlexus, Castilleja purpurea
yar. lindheimeri. Clematis texensis, Cryptantha texana, Dalea hallii, Dalea tenuis, Daucosma laciniata, Dichondra
recurvata, Lechea san-sabeana, Lesquerella recurvata, Malvastrum aurantiacum, Matelea edwardsensis, Nolina
lindheimeriana, Pectis angustifolia var. fastigiata, Pediomelum cyphocalyx, P. hypogaeum var. scaposum, P. lat-
Mipulatum var. appressum. Phlox roemeriana, Ruellia drummondiana, Salvia engelmannii, Silphium albiflorum,
Tradescantia edwardsiana, T humilis, Tridens congestus, and Verbesina lindheimeri.
526
Astragalus reflexus
Dalea hallii
Pediomelum cyphocalyx
Styrax platanifolius ssp. platan ifolius
Hexalectris nltida
The following are comments on unusual or interesting plant distributions. Isoetes butleri has few reports
from Texas. Cyperus eragrostis is known in Texas only from Harris County. Cynanchum barbigerum is at the
northern limit of its distribution.
The following taxa are at the western limit of their distribution: Acmella oppositijolia var. repens, An-
dropogon virginicus, Botrychium virginianum, Carex corrugata, Carex lurida, Claytonia vi rginica, Commelina
diffusa, Cyperus erythrorhizos, Cyperus pseudovegetus, Dichanthelium depauperatum, Dichanthelium malacophyl-
lum, Diodia virginiana, Elephantopus carolinianus, Eragrostis hirsuta, Erechtites hieraciifolia, Eryngium hookeri,
Euonymus atropurpurea var. atropurpurea, Gaura longiflora, Helianthemum rosmarinifolium, Ilexvomitoriajuncus
di chotomusjuncus effusus var. solutus, Juniperus virginiana, Leersia virginica, Lespedeza procumbens, Lespedezfl
repens, Paspalumjloridanum, Phryma leptostachya, Pleopeltis polypodioides ssp. michauxiana. Polygala incamata,
Ranunculus fascicularis, Scirpuspendulus, Scleria ciliata, Scleria oligantha. Senna marilandica, Stylosanthes biflora,
Thelypteris kunthii, Trepocarpus aethusae, Tridens congestus, Viola bicolor, Vitis vulpina, and Zanthoxylum clava-
The following taxa are at the eastern limit of their distribution: Ammoselinum popei, Asclepias engelman-
niana, Astrolepis integerrima, Chamaesaracha coniodes, Dalea nana var. nana, Juniperus pinchotii, Maurandella
antirrhiniflora, and Wedelia texana.
DISCUSSION
When compared to other floristic studies within the Edwards Plateau, Fort Hood has a higher number of
taxa. For example, 583 taxa were found in central and western Bell County (Beaty 1978), 693 taxa at Mason
Mountain Wildlife Management Area, Mason County (Singhurst et al. 2007), 728 taxa at Balcones Canyon-
lands National Wildlife Refuge, Burnet, Travis, and Williamson counties (BCNWR 2008), and 725 taxa at
Kerr Wildlife Management Area, Kerr County (Singhurst et al. 2010). This high number of taxa occurring
on Fort Hood is likely a result of its large size, its location at the intersection of three ecoregions, the large
number of plant communities, and the inclusion of persisting cultivated species.
The number of taxa documented at Fort Hood is compatible to other large-sized military installations.
Fort Benning (2005) reported 1,350 taxa at Fort Benning, Georgia (73,650 ha), Anderson (2007) reported
1,132 taxa at White Sands Missile Range, New Mexico (830,000 ha). Fort Bliss (2001) reported 1,200 taxa
at Fort Bliss, Texas and New Mexico (453,250 ha), and Sorrie et al. (2006) reported 1,185 taxa at Fort Bragg,
North Carolina (60,140 ha).
Texas endemics comprise 4.1% of the flora. This is similar to the 4.0% reported for the flora of north
central Texas (Diggs et al. 1999), but lower than 5.9% reported for Travis County (Carr 2004). There is a
high percentage of introduced taxa (16.3%), but lower than 17.7% reported for north central Texas (Diggs
Hansen, Flora of Fort Hood, Texas
527
et al. 1999). This compares favorably with Fort Bragg, North Carolina, which has 16.8% introduced taxa
| (Sorrie et al. 2006). However, other military bases had much lower numbers of introduced taxa. White Sands
Missile Range, New Mexico has 11.4% introduced (Anderson 2007) and Fort Benning, Georgia has 4.4%
J (Fort Benning 2005).
ANNOTATED CHECKLIST OF VASCULAR PLANT TAXA
Taxa are arranged by divisions, with flowering plants subdivided into classes. Within these groups, taxa
are listed alphabetically by family, genus, species, and infraspecific rank. Taxa that are endemic to Texas
are listed in bold font. Common names follow NRCS (2009a). Taxa are defined as native to Texas or intro-
f duced, and duration is given. Frequency of occurrence is given as: rare, very difficult to find and limited to
I pe or a few locations; occasional, difficult to find but found in several locations; frequent, widely scattered
and not difficult to find; and common, easily found in most habitats. Brief habitat descriptions are given
and the most recently collected specimen is listed for each county with collector and number. Specimens
representing presumed new county records are listed in bold font. Specimens are located at the Fort Hood
The following abbreviations are used for collectors: BH = Bryan Hummel; BRJ = B.R. Jones; BT = Billie
L- Turner; CC = Chuck Coffey; CLGY = C.L. & G. York; CLY = C.L. York; CR = Charlotte Reemts; GG =
Greg Gust; JC = John Cornelius; JJ = Joye Johnson; JS = John Stone; JW = J.K. Wipff; KSNH = K. Schulz
& N. Hastings; KSRS = K.A. Schulz & R.M. Smith; LH = Laura Hansen (LS = Laura Sanchez); RKGL = R.
Krohn & G. Liburn; RTBS = Ross T. Bowlin Sherwood; SEW = Simon E. Wolff; SKJ = Sheila K. Jackson;
SJC = Sgt. J. Cohn; SMKL = Schulz, Morrison, Krohn, & Liburn; VM = V. Morrill; and WRC = W.R. (Bill)
I Carr.
LYCOPODIOPHYTA
Isoetaceae
f Isoetes butleri Engelm., limestone quillwort Native, perennial.
Rare in seasonally seeping limestone soil. Bell Co.: LH
491 6, Coryell Co.: CR 68
EQUISETOPHYTA
Equisetaceae
' fyuisetum hyemale L. var. affine (Engelm.) A.A. Eat., scouring-
I rush horsetail. Native, perennial. Rare in moist soil along
rivers and creeks. Bell Co.: LS 760, Coryell Co.: LH 6031
PTERIDOPHYTA
Aspleniaceae
Asplenium resiliens Kunze, black-stem spleenwort Native, pe-
rennial. Frequent on limestone in mesic wooded canyons.
Bell Co.: LS 4597, Coryell Co.: LS 2634
°ryopteridaceae
Cyrtomium falcatum (Lf.) C. Presl, Japanese net-vein holly fern.
Introduced, perennial. Occasional on mesic limestone
cliffs and sinkhole openings, naturalized. Bell Co.: LS 4651,
Coryell Co.: LS 2648
Woodsia obtusa (Spreng.) Torn ssp. occidentalis Windham,
Pleistocene terrace deposits of the Leon River. Coryell
Co: LS 3383
Marsileaceae
Morsilea vestita Hook. & Grev., hairy waterdover. Native, peren-
nial. Frequent in seasonally damp soil. Bell Co.: LH 51 77,
; Coryell Co.: LH 5163
Ophioglossaceae
Botrychium virginianum (L) Sw, rattlesnake fern. Native, p
rennial. Rare in mesic wooded canyons. Bell CozLS 424
Coryell Co.: LH 5851
mesic wooded canyon. Bell Co.: LS 2106
Pteridaceae
Adiantum capillus-veneris L., common maidenhair. Native,
perennial. Frequent on seeping limestone in riparian
areas and mesic wooded canyons. Bell Co.: LS 1974,
Coryell Co : LS 3968
Argyrochosma dealbata (Pursh) Windham, powdery false
cloak fern. Native, perennial. Occasional on dry limestone
rimrock at upper edges of wooded canyons and slopes.
Bell Co.: LS 3979, Coryell Co.: LH 5066
Astrolepis integerrima (Hook.) Bentham & Windham, hybrid
cloak fern. Native, perennial. Rare on limestone in mesic
wooded canyons. Bell Co.: LH 5400, Coryell Co.: LH 5259
Cheilanthesalabamensis (Buckley) Kunze, Alabama lip fern. Na-
tive, perennial. Frequent on limestone in mesic wooded
canyons. Bell Co.: LS 2107, Coryell Co.: LS 2067
Cheilantheshorridula Maxon, rough lip fern. Native, perennial.
Rare on limestone in mesic wooded canyons. Bell Coz LH
5538, Coryell Co.: LH 6564
Pellaea atropurpurea (L) Link, purple diff-brake. Native, peren-
nial. Frequent on limestone in mesic wooded canyons.
Bell Co.: JS 4228, Coryell Co.: LS 2749
528
Journal of the Botanical Research Institute of Texas 4(1)
perennial. Frequent in shade along banks of creeks and
seeping limestone of mesic canyons. Bell Co.: LS 2720,
Coryell Co.: LS 3967
CONIFEROPHYTA
Juniperusashei Buchh., Ashe's juniper. Native, perennial. Com-
mon in most habitats, especially rocky limestone. Bell Co.:
GG 345, Coryell Co.: LS 2083
Juniperus pinchotii Sudworth, Pinchot's juniper. Native,
LH 6241
Juniperus virginiana L, eastern red-cedar. Native, perennial.
Occasional along wooded riparian corridors. Bell Co.: LH
4918, Coryell Co.: LH6303
Platycladus orientalis (L.) Franco, Oriental arbor-vitae. Intro-
duced, perennial. Rare at old homesites and cemeteries,
cultivated, persisting. Coryell Co.: LS 2597
Taxodium distichum (L.) Rich., bald cypress. Native, perennial.
Rare along banks of a fishing pond, possibly a waif. Coryell
Co: LH 6736
MAGNOLIOPHYTA: Dicotyledonae
Acanthaceae
Didiptera brachiata (Pursh) Spreng., branched foldwing. Na-
tive, perennial. Occasional in moist soil along wooded
rivers and creeks. Bell Co.: LH 5130, Coryell Co.: LH 5450
Dyschoriste linearis (Torr. & A. Gray) Kuntze var. linearis, snake-
herb. Native, perennial. Common In grasslands and open
woodlands. Bell Co.: LH 5907, Coryell Co.: LS 2421
Justida americana (L) Vahl, American water-willow. Native,
perennial. Common along banks of rivers and creeks. Bell
Co.: LS 2499, Coryell Co.: LS 2553
Alternanthera caracasana Kunth, mat chaff-flower. Native,
perennial. Occasional in sandy soil and disturbed areas.
Bell Co.: LH 6059, Coryell Co.: LH 5464
Amaranthus albus L., tumbleweed. Introduced, annual. Oc-
casional on banks of creeks and other disturbed areas.
Bell Co: LS 2920, Coryell Co.: LS 4473
Amaranthus blitoides S. Watson, prostrate pigweed. Intro-
duced, annual. Occasional in disturbed areas. Bell Co.:
CLGY 54437 (TEX), Coryell Co.: LH 6159
Amaranthus palmeri S. Watson, carelessweed. Native, annual.
Occasional along roadsides and in disturbed areas. Coryell
Co LS 3898
Amaranthus polygonoides L, tropical amaranth. Native, annual.
Rare on sandy Pleistocene t
Amaranthus retroflexus L, red-root amaranth. Introduced,
annual. Rare in disturbed areas. Bell Co.: CLGY 55269
(HABAYC), Coryell Co.: LS s.n.
Amaranthus tuberculatus (Moq.) Sauer, rough-fruit amaranth.
Native, annual. Rare in disturbed areas. Bell Co.: SJCT41
(LL)
Froelichia gracilis (Hook.) Moq., slender snake-cotton. Native,
annual. Occasional on sandy Pleistocene terrace deposits.
Bell Co.: CLGY 55109 (HABAYC), Coryell Co.: LH 6154
Gossypianthus lanuginosus (Poir.) Moq. var. lanuginosus,
cotton-flower. Native, perennial. Occasional in sandy soil
and disturbed areas. Bell Co.: CLGY 54340 (HABAYC),
Coryell Co.: LS 4474
corridors. Bell Co.: LS 1 979, Coryell Co.: LS 4467
Ruelliahumilis Nutt, fringe-leaf wild petunia. Native, perennial.
Frequent in partial shade of woodlands. Bell Co.: LS 3962
Coryell Co.: LS 2872
Ruellia metziae Tharp, Metz's wild petunia. Native, perennial.
Frequent in grasslands and limestone outcrops. Bell Co.:
LH 5067, Coryell Co.: LH 5041
Ruellia nudiflora (Engelm. & A Gray) Urban var. nudiflora, violet
wild petunia. Native, perennial. Common in grasslands
and open woodlands. Bell Co.: LS 2932, Coryell Co.: LS
Ruellia nudiflora (Engelm. & A Gray) Urban var. runyonii (Tharp
& F.A Barkley) B.L Turner, Runyon's wild petunia. Native,
perennial. Frequent in grasslands and open woodlands!
Bell Co.: LH 5052, Coryell Co.: LH 5298
ar. sinuosum (Rehder) Little, can-
yon maple. Native, perennial. Occasional in mesic lime-
stone canyons. Bell Co.: LS 4482, Coryell Co.- LS 4624
Acer negundo L var. negundo, box-elder. Native, perennial.
Rhus lanceolata (A. Gray) Britton, flame-leaf sumac. Native,
perennial. Common in most habitats, especially in sunny
disturbed areas. Bell Co.: LS 2919, Coryell Co.: LS 261 5
Rhus trilobata Nutt., skunkbush sumac. Native, perennial.
Common on rocky limestone slopes and mesatops. Bell
Co.: LS 2202, Coryell Co.: LS 2637
Rhus virens Lindh. ex A. Gray var. virens, evergreen sumac Na-
tive, perennial. Common on rocky limestone slopes and
mesatops. Bell Coj GG 344, Coryell Co.: LH 5330
Toxicodendron radicans (L) Kuntze ssp. verrucosum (Scheele)
woodlands. Bell Co.: LS 2221, Coryell Co.: LS 2663
Apiaceae
Ammoselinum popei Torr. & A Gray, plains sand-parsley. Na-
tive, annual. Rare in shade along banks of creeks. Bell
Co.: LH 5838
Bifora americana Benth. & Hook. f. ex S. Watson, prairie-bishop.
Native, annual. Common in grasslands. Bell Co.: LS 2323,
Coryell Co.: LS 2533
532
Native, annua! or biennial. Common in open woodlands,
roadsides, and disturbed areas. Bell Co.: LH 5462, Coryell
Co: LS 4510
Hymenopappus scabiosaeus L'Her. var. corymbosus (Torr. & A.
Gray) B.L Turner, Carolina woolly-white. Native, biennial.
Common in grasslands and open woodlands. Bell Co.: LS
1707, Coryell Co.: LS 2432
Hymenopappus tenuifolius Pursh, chalkhill. Native, biennial.
Common on dry rocky limestone. Bell Co.: LH 5043, Coryell
Co.: LS 2523
Hypochaeris brasiliensis (Less.) Benth. & Hook, ex Griseb., Brazil-
ian cat's ear. Introduced, biennial. Rare in disturbed areas.
Coryell Co.: LS 2451
Iva angustifolia Nutt ex DC., narrow-leaf sumpweed. Native,
annual. Common in disturbed grasslands and in moist soil
along banks of creeks. Bell Co.: LS 2952, Coryell Co.: LS 3444
Iva annua L, annual marsh-elder. Native, annual. Common
along banks of creeks and other damp areas. Bell Co.: LS
2988, Coryell Co.: LS4503
Krigia caespitosa (Raf.) Chambers, dwarf dandelion. Native,
annual. Frequent in damp soil of disturbed areas. Bell Co.:
LH 5642, Coryell Co.: LH 6375
Krigia wrightii (A. Gray) K.L. Chambers ex KJ. Kim, Wright's
II Co.: LSI 983,
Coryell Co.: LH 5191
Lactuca ludoviciana (Nutt.) Riddell, western wild lettuce. Na-
tive, biennial. Common in woodlands and along riparian
corridors. Bell Co.: LS 3264, Coryell Co.: LS 2744
Lactuca serrioia L, prickly lettuce. Introduced, annual. Com-
mon in disturbed areas. Bell Co.: LH 4993, Coryell Co:
LS4507
Laenneciacouiteri (A. Gray) G.L Nesom, Coulter's horseweed.
Native, annual. Rare in disturbed alluvial floodplain. Bell
Co: LH 6246
Liatris aestivalis G.L Nesom & R. O'Kennon, summer gayfeather.
Native, perennial. Frequent in grasslands on dry rocky
limestone. Bell Co.: LH 5373, Coryell Co.: LS 2003
Uatris mucronata DC., narrow-leaf gayfeather. Native, peren-
nial. Common in grasslands on rocky limestone. Bell Co.:
LS20H Coryell Co.: LS 4493
Journal of the Botanical Research Institute of Texas 4(1)
Packera obovata (Muhl. ex Willd.) W.A. Weber & A. Love,
round-leaf groundsel. Native, perennial. Frequent in mesic
woodlands and canyons. Bell Co.: LH 5588, Coryell Co.:
KSNH 731
Packera tampicana (DC.) C. Jeffrey, Great Plains ragwort Na-
tive, annual. Frequent in damp open areas and roadside
ditches. Bell Co.: LH 5688, Coryell Co.: LH 5853
Palafoxia callosa (Nutt.) Torr. & A. Gray, small palafoxia. Native,
annual. Common in disturbed grasslands. Bell Co: LS 2046,
Coryell Co.: LS 4430
Parthenium hysterophorus L, false ragweed. Native, annual.
Common in disturbed grasslands. Bell Co.: LH 5056, Coryell
Co: LS 4476
Pectis angustifolia Torr. var. fastigiata (A. Gray) Keil, lem
onscent. Native, annual. Occasional in shallow pockets
of soil on limestone outcrops and mesatops. Bell Co.: LH
5326, Coryell Co.: LH 5064
Coryell Co.: LH 5871
Pluchea camphorata (L) DC, camphorweed. Native, annual
or perennial. Common along banks of creeks and other
damp areas. Bell Co.: LH 5267, Coryell Co.: LH 5306
Pluchea odorata (L) Cass., sweetscent. Native, annual. Com-
mon along banks of creeks and other damp areas. Bell
Co.: LS 2989, Coryell Co.: LS 4437
Pseudognaphalium obtusifolium (L) Hilliard & B.L. Burtt fra-
grant cud-weed. Native, annual. Frequent in woodlands
of limestone slopes and adjacent mesatops. Bell Co.: LH
5399, Coryell Co.: LH 5258
Pyrrhopappus grandiflorus (Nutt) Nutt, tuber false dandelion.
Native, perennial. Occasional on sandy Pleistocene terrace
deposits. Coryell Co.: LH 5673
Pyrrhopappus pauciflorus (D. Don) DC, Texas dandelion. Native,
annual. Common in grasslands and mesic woodlands. Bell
Co: LS 2215, Coryell Co.: IS 2251
Ratibida columnifera (Nutt.) Woot. & Standi, Mexican hat.
Native, perennial. Common in grasslands and open
woodlands. Bell Co.: LS 2496, Coryell Co: LS 2424
Rudbeckia hirta L var. pulcherrima Farw, black-eyed Susan.
Native, annual or perennial. Common in grasslands and
open woodlands. Bell Co.: LS 2577, Coryell Co.: LS 3373
Silphium albiflorum A. Gray, white rosinweed. Native, peren-
Lygodesmia texana (Torr. & A. Gray) Greene, Texas skeleton-
plant. Native, perennial. Common in grasslands. Bell Co:
LS 2693, Coryell Co: LS 2673
Marshallia caespitosa Nutt, ex DC var. signata Beadle & F£
Boynt, Barbara's buttons. Native, perennial. Common in
grasslands on rocky limestone, especially in seepy areas.
Bell Co: LS 1 777, Coryell Co: LS 2428
Melampodium leucanthum Torr. & A. Gray, rock daisy. Native,
perennial. Common in grasslands on dry rocky limestone!
BeH Co: LS 1852, Coryell Ca: LS 2286
Onopordum acanthium L, cotton-thistle. Introduced, biennial.
Rare on roadside shoulder. Bell Co: LS 3562
LS 3573,
Bell Co: LS 1881, Coryell Coryell Co: LS 4424 '
Silphium laciniatum L, compassplant. Native, perennial. Rare
on dry rocky limestone. Coryell Co: LS 1951
Silphium radula Nutt, rough-stem rosinweed. Native, peren-
nial. Common along riparian corridors. Bell Co: LH 5045,
Coryell Co: LS 3977
Silybum marianum (L.) Gaertn., blessed milk-thistle. Intro-
Bell Co: LS 2809, Coryell Co: LH 6475
Simsia calva (Engelm. & A. Gray) A. Gray, awnless bush-sun-
flower. Native, perennial. Common in open woodlands.
Bell Co: LS 1489, Coryell Co: LS 2407
Smallanthus uvedalius (L) Mack, ex Small, bear's foot Native,
Hansen, Flora of Fort Hood, Texas
perennial. Occasional in shade along riparian corridors.
Bell Co.: LH6097
Solidago altissima L., late goldenrod. Native, perennial. Com-
mon in damp soil along banks of creeks and in post oak
woodlands. Bell Co: LS 2995, Coryell Co.: LH 5468
Solidago gigantea Aiton, giant goldenrod. Native, perennial.
Common in moist soil along banks of creeks and rivers.
Bell Co.: LS 2059, Coryell Co.: LS 4504
Solidago nemoralis Aiton var. longipetiolata (Mack. & Bush)
Palmer & Steyerm., gray goldenrod. Native, perennial.
Common in grasslands on dry rocky limestone. Bell Co:
LS 2050, Coryell Co: LS 2094
Solidago radula Nutt., western rough goldenrod. Native,
perennial. Common on dry rocky limestone. Bell Co: LS
2049, Coryell Co: LS 2008
Sonchus asper (L) Hill, spiny sow-thistle. Introduced, annual.
Frequent in disturbed areas. Bell Co: LS 1856, Coryell
I Co: LS 2236
Sonchus oleraceus L, common sow-thistle. Introduced, an-
nual. Rare in disturbed areas. Bell Co: LH 6339, Coryell
Co: LH 5001
Native, annual. Common along banks of creeks, ponds,
| and other damp areas. Bell Co: LS 4587, Coryell Co: LS
4506
■ Symphyotrichum drummondii (Lindl.) G.L Nesom var. texanum
, (Burgess) G.L. Nesom, Texas aster. Native, perennial. Com-
■t mon in shade along riparian corridors and mesic canyons.
Bell Co: LS 2516, Coryell Co: LH 5476
■ Symphyotrichum ericoides (L) G.L Nesom var. ericoides, white
heath aster. Native, perennial. Common in grasslands. Bell
P Co: LS 4589, Coryell Co: LH 5469
I Symphyotrichum oblongifolium (Nutt.) G.L. Nesom, aromatic
aster. Native, perennial. Rare in sandy soil. Coryell Co:
| LS 2084
I Symphyotrichum praealtum (Poir.) G.L Nesom, willow-leaf
I aster. Native, perennial. Frequent in moist soils and
along unshaded creek banks. Bell Co: LS 4588, Coryell
I Co: LH 6305
| Symphyotrichum sericeum (Vent.) G.L. Nesom, silky aster,
y Native, perennial. Common in grasslands on dry rocky
limestone. Bell Co: LS 4593, Coryell Co: LS 3634
Taraxacum officinale F.H. Wtgg„ common dandelion. Intro-
I duced, biennial or perennial. Common in disturbed areas
and lawns. Bell Co: LS 1695, Coryell Co: LS4599
Tetraneuris linearifolia (Hook.) Greene, slender-leaf bitterweed.
Native, annual. Common in grasslands and on limestone
mesatops. Bell Co: LH 6414, Coryell Co: LS 2903
Tetraneuris scaposa (DC.) Greene var. scaposa, slender-stern
I bitterweed. Native, perennial. Common in grasslands
i on dry rocky limestone. Bell Co: LS 2129, Coryell Co:
1 152252
I Thelesperma filifolium (Hook.) A. Gray var. filifolium, stiff
S greenthread. Native, annual or perennial. Common in
grasslands and open woodlands. Bell Co: LS 3250, Coryell
II Go: LS 2413
thelesperma simplicifolium A. Gray, slender greenthread.
Native, perennial. Common in grasslands and open
woodlands. Bell Co: LS 2498, Coryell Co: 2498
Thymophylla pentachaeta (DC.) Small, common dogweed.
Native, perennial. Frequent in grasslands on rocky lime-
. Bell Co: LS 1672, Coryell Co:
Townsendia exscapa (Richardson) Porter, stemless Townsend
daisy. Native, perennial. Rare in grasslands on rocky lime-
stone. Bell Co: LH 5574, Coryell Co: LH 5566
Tragopogon dubius Scop., yellow salsify. Introduced, biennial.
Rare in disturbed area. Bell Co: LH 6635
Verbesina encelioides (Caw.) Benth. & Hook. f. ex A. Gray, cowpen
daisy. Native, annual. Frequent in grasslands and disturbed
areas. Bell Co: LH 5394, Coryell Co: LS 2023
Verbesina lindheimeri B.L. Rob. & Greenm., Lindheimer's
crownbeard. Native, perennial. Rare in wooded canyons
and slopes. Bell Co: LH 51 61 , Coryell Co: LS s.n.
Verbesina virginica L, frostweed. Native, perennial. Common
in shade of woodlands and riparian corridors. Bell Co: LS
1976, Coryell Co: LH 5275
Vernonia baldwinii Torr. ssp. interior (Small) Faust, western
ironweed. Native, perennial. Common in moist soil along
banks of creeks and rivers. Bell Co: LS 2947, Coryell Co:
LS 4495
Vernonia x guadalupensis A. Heller (pro sp.) [baldwinii x lind-
heimeril hybrid ironweed. Native, perennial. Frequent in
grasslands and along banks of creeks. Bell Co: LH 5144,
Coryell Co: LH 5274
Vernonia lindheimeri A. Gray & Engelm., woolly ironweed.
Native, perennial. Common on dry rocky limestone. Bell
Co: LS 3575, Coryell Co: LH 5482
Viguieradentata (Cav.) Spreng, golden-eye. Native, perennial.
Common at woodland edges on limestone. Bell Co: LH
5397, Coryell Co: LH 4997
Wedelia texana (A. Gray) B.L Turner, hairy wedelia. Native,
perennial. Rare in grasslands and open woodlands. Bell
Co: LH 5942
Xanthisma texanum DC ssp. drummondii (Torr. & A. Gray)
Semple, Texas sleepy daisy. Native, annual. Rare among
construction debris, possibly a waif. Bell Co: LH 5055
Xanthium strumarium L. var. canadense (Mill.) Torr. & Gray,
Canada cocklebur. Native, annual. Common along banks
of creeks, ponds, and other damp areas. Bell Co: LS 2991 ,
Coryell Co: LS 201 7
Mahonia trifoliolata (Moric.) Fedde, agarita. Native, perennial.
Common in grasslands and woodlands on dry rocky
limestone. Bell Co: LS 4629, Coryell Co: LS 2635
Nandina domestica Thunb, sacred bamboo. Introduced,
perennial. Occasional near old homesites, cultivated,
naturalized. Bell Co: LS 2104, Coryell Co: LS 4191
Bignoniaceae
Campsis radicans (L) Seem, ex Bureau, common trumpet-
creeper. Native, perennial. Occasional near old homesites,
cultivated, naturalized. Bell Co: LS 759, Coryell Co: LH
6638
Catalpa speciosa (Warder) Warder ex Engelm, northern ca-
540
Journal of the Botanical Research Institute of Texas 4(1)
ruial. Frequent in sandy soil and post oak woodlands. Bell
Co.: LS 2061 , Coryell Co.: LS 3428
Strophostyles leiosperma (Torr. & A. Gray) Piper, slickseed
Coryell Co.: WRC 17591 Y
Styphnolobium affine (Torr. & A. Gray) Walp, Eve's necklace.
Native, perennial. Frequent in most woodlands. Bell Co.:
GG 372, Coryell Co: LSI 890
Trifolium repens L„ white clover. Introduced, perennial. Rare
in disturbed areas, along creek banks, and in lawns. Bell
Co.: LH 6073, Coryell Co.: LS 3288
Vicia ludoviciana Nutt, deer pea vetch. Native, annual. Com-
mon in grasslands and open woodlands. Bell Co.: LS 2322,
Coryell Co.: LS 2249
Vicia sativa L. ssp. nigra (L.) Ehrh, common vetch. Introduced,
annual. Frequent in disturbed grasslands and roadsides.
Bell Co.: LH 5618, Coryell Co.: LH 4661
Vicia vilbsa Roth ssp. varia (Host) Corb, winter vetch. Intro-
duced, annual. Rare in grasslands and disturbed areas.
Coryell Co.: VM s.n.
Garrya ovata Benth. ssp. lindheimeri (Torr.) Dahling, Lindhei-
mer's silktassel. Native, perennial. Rare in a mesic wooded
canyon. Bell Co.: LS 2324
Muhlenberg's centaury. Introduced, annual. Frequent
in seasonally seeping limestone soil. Bell Co.: LS 3301,
Coryell Co.: LH 6614
Centaurium texense (Griseb.) Fernald, Texas centaury. Native,
annual. Common in grasslands on dry rocky limestone.
Bell Co.: LS 2488, Coryell Co.: LS 2538
Eustoma exaltatum (L.) Salisb. ex G. Don ssp. russellianum
(Hook.) Kartesz, bluebell gentian. Native, annual or peren-
nial. Common in seasonally seeping limestone soil and
along creek banks, especially in sun. Bell Co.: LS 4590,
Coryell Co.: LS 3567
Sabatia campestris Nutt, prairie rose gentian. Native, annual.
Occasional on sandy soils and post oak woodlands. Bell
Co.: CLGY 54441 (HABAYC), Coryell Co.: LS 3370
Quercus buckleyi Nixon & Dorr, Buckley’s oak. Native, perennial.
Common on dry limestone slopes, canyons, and mesas.
Bell Co.: LS 21 33, Coryell Co,- LS 2127
Quercus fusiformis Small, Plateau live oak. Native, perennial.
Common in most habitats. Bell Co.: LS 2205, Coryell Co :
LS2636
Quercus macrocarpa Michx, bur oak. Native, perennial. Com-
mon along riparian corridors. Bell Co.: LS 1697, Coryell
Co.: LH 4749
>, blackjack oak. Native, peren-
nial. Common on sandy soils and post oak woodlands.
Bell Co.: LS 4630, Coryell Co.: LH 4665
Quercus muehlenbergii Engelm, chinkapin oak. Native, peren-
nial. Common in mesic wooded canyons and slopes. Bell
Co.: LSI 994, Coryell Co.: LS4635
Quercus shumardii Buckley, Shumard red oak. Native, peren-
nial. Common along riparian corridors. Bell Co.: LH 5643,
Coryell Co ,:LH 6304
Quercus sinuata Walter var. breviloba (Torr.) C.H. Mull, Bigelow's
Co: LS 2284
Quercus stellata Wangenh, post oak. Native, perennial. Com-
mon on sandy soils and post oak woodlands. Bell Co.: LS
2321, Coryell Co.: LS 2274
Corydalis micrantha (Engelm. ex A. Gray) A. Gray ssp. australis
(Chapm.) GB. Ownbey, southern corydalis. Native, annual.
Occasional on sandy Pleistocene terrace deposits. Bell Co.:
CLGY 54023 (TEX), Coryell Co.: LH 5596
Corydalis micrantha (Engelm. ex A. Gray) A. Gray ssp. micran-
tha, small-flower corydalis. Native, annual. Occasional on
sandy Pleistocene terrace deposits. Bell Co.: LH 5586
Erodium cicutarium (L) L'Her. ex Aiton, filaree. Introduced,
annual. Common in grasslands and disturbed areas. Bell
Co.: LS 1701, Coryell Co.: LS 2225
Common in grasslands and disturbed areas. Bell Co; GG
347, Coryell Co.: LS 2226
Geranium carolinianum L, crane's-bill. Native, annual. Com-
mon in grasslands and disturbed areas. Bell Co.: LS 17G3,
Coryell Co.: LS 2231
Geranium texanum (Trel.) A. Heller, Texas geranium. Native,
annual. Common in grasslands and disturbed areas. Bell
Co- LS 1 836, Coryell Co.: LS 2643
Haloragaceae
Myriophyllum heterophyllum Michx, two-leaf water-milfoil. Na-
tive, perennial. Occasional in ponds and lakes, submersed.
Bell Co.: LH 5265, Coryell Co.: LH 5484
Hippocastanaceae
Aesculus glabra Willd. var. arguta (Buckley) B.L Rob, Texas
buckeye. Native, perennial. Frequent along riparian cor-
ridors and in mesic wooded canyons. Bell Co, LS 4638,
Coryell Co.: LS4625
Hydrophyllaceae
Nama hispidum A. Gray, sandbells. Native, annual. Rare on san-
dy Pleistocene terrace deposits and post oak woodlands.
Bell Co.: CLGY 54438 (HABAYC), Coryell Co,- LH 6541
Namajamaicense L, Jamaican weed. Native, annual. Occa-
sional on sandy Pleistocene terrace deposits. Bell Co; LH
5687, Coryell Co.: LH 5752
Nemophilaphacelioides Nutt, baby blue-eyes. Native, annual.
Common in shade along riparian corridors. Bell Co; LH
5620, Coryell Co.: LH 5755
Phaceliacongesta Hook, blue-curls. Native, annual. Common
. Bell Co- LH
542
Rare on roadside shoulders and seeded areas, cultivated,
not persisting. Coryell Co.: KSNH 787 (CMML)
Linum hudsonioides Planch, Texas flax. Native, annual. Rare
in sandy soil and post oak woodlands. Bell Co.: LH 6801,
Coryell Co.: LH 5671
Linum rupestre (A. Gray) Engelm. ex A. Gray, rock flax. Native,
mesas. Bell Co.: LS 2660, Corydl Co.: LS 2406 P
Mentzelia oligosperma Nutt, ex Sims, stick-leaf. Native, peren-
nial. Frequent on dry limestone slopes and mesas. Bell
Co: LH 5904, Coryell Co.: LS 2745a
partial shade. Bell Co.: LS 2933, Coryell Co.: LH 6539
Sidaabutifolia Mill, spreading sida. Native, perennial. Common
in grasslands and open woodlands. Bell Co: LH 4959,
Coryell Co.: LS 2702
Sidaspinosa L, prickly fanpetals. Native, annual. Occasional in
damp soil, disturbed areas. Bell Co.: LH 61 05
Meliaceae
Melia azedarach L, China-berry. Introduced, perennial. Corn-
2680, Coryell Co.: LS2267
Menispermaceae
Loganiaceae
Mitreola petiolata (J.F. Gmel.)Torr. & A. Gray, lax hornpod. Na-
tive, annual. Common along banks of creeks and other
damp areas. Bell Co.: LH 5268, Coryell Co.: LH 5165
Spigelia hedyotidea A. DC, prairie pinkroot. Native, perennial.
Rare in shade of woodlands. Bell Co.: LH 5903, Coryell
Go: LH 6563
Lythraceae
Ammanniacoccinea Rottb, purp
Bell Co.: LH 5180, Coryell Co.: LS 4439
Lagerstroemia indica L, common crape-myrtle. Introduced,
i homesites, cultivated, persisting.
Bell Co.: LH 6048, Coryell Co.: LS 2914
Lythrum californicumlorr. & A. Gray, California loosestrife. Na-
tive, perennial. Frequent along banks of creeks and other
damp areas. Bell Co.: LS 2843, Coryell Co.: LS 2710
Abutilon fruticosum Guill. & Perr, Indian-mallow. Native, peren-
nial. Common in grasslands and open woodlands. Bell Co.:
LS 2929, Coryell Co.: LS 2522
Callirhoe involucrata (Torr. & A. Gray) A. Gray var. involucrata,
low poppy-mallow. Native, perennial. Frequent on sandy
soils and post oak woodlands. Bell Co.: CLGY 55111 (HA-
BAYC), Coryell Co.: LH 6536
Callirhoe pedata (Nutt ex Hook.) A. Gray, finger poppy-mallow.
Native, perennial. Common in grasslands and open wood-
lands. Bell Co.: LS 3274, Coryell Co.: LS 2435
Malva neglecta Wallr, common mallow. Introduced, annual
or biennial. Occasional in disturbed areas and near old
homesites. Bell Co.: LS 2781
cum (Scheele) Walp., golden
false-mallow.
Pleistocene t<
n River. Coryell
Co: LH 5451
Malvaviscus arboreus Dill, ex Cav. var. drummondii (Torr. & A.
Gray) Schery, Drummond wax-mallow. Native, perennial.
Occasional in mesic wooded canyons and along riparian
corridors. Bell Co.: LS 2626, Coryell Co.: LH 6675
Modiola caroliniana (L) G. Don, Carolina modiola. Native,
perennial. Occasional in disturbed areas and in lawns
Bell Co.: LH 5837, Coryell Co.: LS 2639
Rhynchosidaphysocalyx (A. Gray) Fryxell, spear-leaf sida. Na-
C, Carolina snailseed. Native, pen
nial. Common in most woodlands, climbing onto shru
Bell Co.: LS 2861 , Coryell Co.: LS 3414
Mollugo verticillata L, green carpetweed. Native, annual. Rare
on sandy Pleistocene terrace deposits. Bell Co.: CLGY
55142 (HABAYC), Coryell Co.: LH 61 57
Ficus carica L, common fig. Introduced, perennial. Occasional
at old homesites, cultivated, persisting. Bell Co.: LS 1964,
Coryell Co.: LH 6639
Madura pomifera (Raf.) C.K. Schneid., Osage orange. Native,
perennial. Common along riparian corridors and fence-
rows. Bell Co.: LS 2679, Coryell Co.: LS 1907
Morus alba L„ white mulberry. Introduced, perennial. Rare
at old homesites, cultivated, persisting. Bell Co: LS 1699,
Coryell Co.: LH 5296
Morus microphylla Buckley, Texas mulberry. Native, perennial.
Common in mesic wooded canyons. Bell Co.: JS 4225,
Coryell Co.: LH 4907
Morus rubra L, red mulberry. Native, perennial. Common along
riparian corridors and in mesic wooded canyons. Bell Co:
LS 16%, Coryell Co.: LS 2758
Nyctaginaceae
Boerhavia diffusa L„ scarlet spiderling. Native, perennial. Oc-
casional in sandy soil and disturbed areas. Bell Co: LH
6670, Coryell Co: LH 5455
Mirabilis albida (Walter) Heimerl, white four-o’clock. Native,
perennial. Common in open woodlands. Bell Co: LH 5 1 75,
Coryell Co: LH 5472
Mirabilis jalapa L, marvel of Peru. Introduced, perennial. Oc-
casional at old homesites, cultivated, naturalized. Coryell
Co: LH 6061
Mirabilis latifolia (A. Gray) Diggs, Lipscomb, & O'Kennon,
broad-leaved four-o'clock. Native, perennial. Common
in open woodlands. Bell Co.: LS 2983, Coryell Co.: LH
Mirabilis linearis (Pursh) Heimerl, linear-leaf four-o'clock. Native,
perennial. Occasional in grasslands and open woodlands
Bell Co: LH 5039, Coryell Co: LH 4912
Mirabilis nyaaginea (Michx.) MacMill., wild four-o'clock. Native,
perennial. Occasional in open woodlands and mesatops
Bell Co: LH 5048, Coryell Co.: LS 4243
. Bell Co.: LS
Hansen, Flora of Fort Hood, Texas
551
Iris germanica L, German iris. Introduced, perennial. Occa-
sional at old homesites, cultivated, persisting. Bell Co.: LH
6337, Coryell Co.: LH 6345
Irispseudacorus L, pale yellow iris. Introduced, perennial. Rare
along banks of a fishing pond, possibly a waif. Coryell
Co: LH 6735
Nemastylis geminiflora Nutt., prairie celestial. Native, perennial.
Frequent in open areas of woodlands. Bell Co.: LH 5644,
Coryell Co.: LS 2193
Sisyrinchium chilense Hook., sword-leaf blue-eyed grass. Na-
tive, perennial. Common in open woodlands, grasslands,
and riparian areas. Bell Co.: LH 4754, Cory e
Sisyrinchium minus Engelm. & A. Gray, dwarf blue-eyed grass.
Native, annual. Frequent in sandy soil and post oak wood-
lands. Bell Co.: LH 6727, Coryell Co.: LH 6734
Juncus brachycarpus Engelm., whiteroot rush. Native, peren-
nial. Occasional in sandy soil. Coryell Co.: LH 5672
Juncus bufonius L„ toad rush. Native, annual. Occasional in
damp, sandy soil. Bell Co.: LH 5858
Juncus dichotomus Elliot, forked rush. Native, perennial. Com-
mon along banks of creeks and ponds, and other damp
areas. Bell Co.: LH 5892, Coryell Co.: LH 5854
Juncus dudleyi Wiegand, Dudley's rush. Native, perennial.
Occasional in damp soil. Bell Co.: KSRS 869, Coryell Co:
LH 4958
Juncus effusus L. var. solutus Fernald & Wiegand, common
rush. Native, perennial. Occasional in damp soil. Bell Co:
LH 6480
Juncus filipendulus Buckley, ring-seed rush. N
il. Bell Co.: LH 6018
Frequent in d
r Wiegand var
Allium ampeloprasum L, broadleaf wild leek. Introduced, pe-
rennial. Occasional at old homesites, cultivated, persisting.
Bell Co.: Bell LH 6052, Coryell Co.: LS 2571
Allium canadense L var. canadense, Canada garlic. Native, pe-
rennial. Common along banks of creeks and other damp
areas. Bell Co.: LH 4828, Coryell Co.: LH 4750
Allium canadense L var. fraseri Ownbey, Fraser meadow garlic
Native, perennial. Frequent in post oak woodlands and
sandy alluvium. Bell Co.: LH 6408, Coryell Co.: LS 1824
Allium drummondii Regel, Drummond's wild onion. Native,
1 5173,
5957Si°nal 'n damp areas of 9rasslands- Coryell Co.: LH
Juncus texanus (Engelm.) Coville, Texas rush. Native, perennial.
Common along banks of creeks and ponds, and other
damp areas. Bell Co.: LH 5937, Coryell Co.: LH 4956
Juncus torreyi Coville, Torrey's rush. Native, perennial. Occa-
sional in damp soil. Coryell Co.: LS 2459
Lemnaceae
lemna aequinoctialis Welw., lesser duckweed. Native, annual.
Rare in still water, floating. Bell Co.: CLGY 543 1 5 (TEX)
umna minuta Kunth, least duckweed. Native, annual. Rare in
still water, floating. Bell Co.: LH 6573
Lemna obscura (Austin) Daubs, little duckweed. Native, an-
nual. Rare in still water, floating. Bell Co.: LH 6650
Co.: LH 5453
tyro delapolyrrhiza (L.) Schleid., common duckmeat Native,
annual. Rare in still water, floating. Bell Co.: CLGY 54316
(HABAYC), Coryell Co.: LH 5454
Bell Co.: JS 4231, Coryell Co.: LS 2246
Allium porrum L„ garden leek. Introduced, perennial. Rare
at old homesites, cultivated, persisting. Coryell Co.: LH
5970
Allium sativum L„ cultivated garlic. Introduced, perennial.
Rare at old homesites, cultivated, persisting. Bell Co.:
LH 6649
Allium stellatum Fraser ex Ker Gawl, prairie onion. Native,
perennial. Occasional in grasslands and open woodlands.
Bell Co.: LS 2507
Androstephium caeruleum (Scheele) Greene, blue funnel-lily.
Native, perennial. Rare in grasslands. Bell Co.: LH 6695,
Coryell Co.: KSNH 793
Camassiasciltoides (Raf) Cory, wild-hyacinth. Native, perennial.
Occasional on limestone mesatops and slopes. Bell Co:
WRC 1 7305, Coryell Co: LS 21 95
Cooperia drummondii Herbert, cebolleta. Native, perennial.
Common in grasslands. Bell Co.: LH 5058, Coryell Co.:
LS 2881
Cooperia pedunculata Herbert, giant rain-lily. Native, peren-
nial. Common in grasslands. Bell Co: GG 343, Coryell
Co: LH 5677
Erythronium albidum Nutt, white dog-tooth-violet. Native,
perennial. Common in mesic wooded canyons. Bell Co:
LS 4598, Coryell Co.: LS 461 3
Erythronium mesochoreum Knerr, dog-tooth-violet Native,
perennial. Common on mesa tops, growing in shade of
shrubs. Bell Co: LS 2259, Coryell Co: LS 2647
Muscari neglectum Guss. ex Ten, starch grape-hyacinth. Intro-
duced, perennial. Occasional in disturbed areas, natural-
ized. Bell Co.: LH 6681, Coryell Co.: LH 5603
Nothoscordum bivalve (U Britton, crow-poison. Native, peren-
nial. Common grasslands and open woodlands. Bell Co:
LS 4631, Coryell Co: LS 3435
Zigadenus nuttallii (A. Gray) S. Watson, Nuttall's death-camas.
Native, perennial. Frequent on limestone mesatops and
along riparian corridors. Bell Co: LH 6413, Coryell Co:
LH 5602
Naps guadalupensis (Spreng.) Magn us ssp. guadalupensis, com-
mon water-nymph. Native, annual. Frequent in ponds and
lakes, submersed. Bell Co.: LH 4995, Coryell Co LH 4949
JS i II «
FIRST REPORT OF PERSICARIA HISPIDA (POLYGONACEAE)
FROM NORTH AMERICA NORTH OF MEXICO (TEXAS)
Daniel E. Atha
William Carr
The New York Botanical Garden
2900 Southern Blvd
Bronx, New York 10458-5126, USA
datha@nybg.org
The Nature Consen/ancy of Texas
RO. Box 1440
San Antonio, Texas 78295-1440, USA
bcarr@tncorg
ABSTRACT
INTRODUCTION
The purpose of this paper is to report the discovery of a previously undocumented Persic
cies for the flora of North America north of Mexico and to provide a description and e;
monographic, floristic, ecological and conservation efforts.
This striking species was first encountered in Texas during a botanical inventory of a rai
County, an area rich in globally-rare species of the Tamaulipan Thomscrub Ecoregion (Fig. la). In April 2002, the
junior author made a brief stop at a small, steep-sided stock tank and expected to tally the suite of low-growing
species that characterize the flora of such sites during drought periods. These species include Cynodon dactylon
(L.) Pers., Phyla nodiflora (L.) Greene, Eryngium nasturtiifolium Juss. ex Delarbre, Glinus radiatus Ruiz & Pavon)
Rohrb., and Calibrachoa parviflora (Juss.) DArcy. The aforementioned species were present, but a taller plant
caught our attention first and resulted in an excited conversation full of joyful expletives not usually associated
with field work in South Texas. Although the species was immediately recognized as a member of the genus
ersicaria, its robust, hispid stems, large ocreae and glutinous leaves excluded it from any Persicaria known from
exas. In December 2009, a second Texas population of Persicaria hispida was located during an inventory of
a i erent Starr County ranch located about 6-7 air miles west-northwest of the first tank. The second tank
a a c ay bottom and was nearly full of water at the time. Persicaria hispida (H.B.K.) M. Gomez, occupied the
eeper parts of the tank, occurring as a lone emergent species while other species, such as Chloracantha spinosa
shoreline ^tsom' EJeocharis pdustris (L.) Roem. & Schult and Cynodon dactylon were visible along the
,■ °rC 7e B0t tanlcs were on ranches that were actively grazed by cattle and also supported small popula-
tions of white-tailed deer.
Mexic^^d™ ’S 3 distinctive species that ranges from southern South America, north to northern
Am C° ai>, n°T documente£* in Texas- Persicaria orientals (L.) Spach is the only other Persicaria in North
broIcT 15 !fPid throughout’ has leaf blades ovate, more than 3 cm wide and ocreae margins (apices)
y rotate an photosynthetic. The two species may be distinguished by the following key.
1.8-3.2 cm long, i
(2.5-)3-5 x 1.5— 2.5 n
(2.7-)3.0-3.3 x 2.6-2.9 t
Athar and Carr, Persicaria hispida in North America (Texas)
561
A
AVAILABLE TITLES FROM THE
SBM 01 1 A Quantitative Analysis of the Vegetation on the Dallas County White Rock Escarpment
SBM 02 | The Vascular Flora of St. Francis County, Arkansas
SBM 04 | Asteraceae of Louisiana
SBM 05 | The Genus Mikania (Compositae: Eupatorieae) in Mexico
SBM 06 I Frontier Botanist William Starling Sullivant’s Flowering-Plant Botany of Ohio
SBM 07 | A Taxonomic Revision of the Acaulescent Blue Violets (Viola) of North America
SBM 08 | Aster & Brachyactis (Asteraceae) in Oklahoma
SBM 09 | The Genus Mikania (Compositae: Eupatorieae) in the Greater Antilles
SBM 10 | Checklist of the Vascular Plants of Tennessee
SBM 11 1 Text Annotations & Identification Notes for Manual of the Vascular Flora of the Carolinas
SBM 12 | The “El Cielo” Biosphere Reserve, Tamaulipas, Mexico
SBM 13 | Flora de Manantlan
SBM 14 | Niebla & Vermilacinia (Ramalinaceae) from California & Baja California
SBM 15 | Monograph of Northern Mexican Crataegus (Rosaceae, subfam. Maloideae)
SBM 16 I Shinners & Mahler’s Illustrated Flora of North Central Texas
SBM 17 | The Grasses of Barbados (Poaceae)
SBM 18 | Floristics in the New Millenium: Proceedings of the Flora of the Southeast U.S. Symposium
SBM 19 | Emanuel D. Rudolph’s Studies in the History of North American Botany
SBM 20 | Generic Conspectus of the Tribe Asterae (Asteraceae) in North & Central America
SBM 21 1 A Comparative Checklist of the Plant Diversity of the Iwokrama Forest, Guyana
SBM 22 | Lloyd Herbet Shinners: By Himself
SBM 23 | Taxonomy, Distribution, & Ecology of the Genus Phaseolus (Leguminosae)
SBM 24 | Atlas of the Vascular Plants of Texas, Volume 1 & 2
SBM 25 | Los Generos de Leguminosas del Norte de Mexico
SBM 26 | Illustrated Flora of East Texas, Volume 1
SBM 27 | The Genus Psychotria (Rubiaceae) in the Philippine Archipelago
SBM 28 | Wild Flowers of Mombacho (Nicaragua)
SBM 29 | Muhlenbergia (Poaceae) de Chihuahua, Mexico
SBM 30 | Tundra to Tropics: The Floristic Plant Geography of North America
SBM 31 1 A Fifth Checklist of Tennessee Vascular Plants
OTHER PUBLICATIONS
Keys to the Vascular Plants of the Black Gap Wildlife Management Area, Brewster County
Mosses of Texas: A Manual of the Moss Flora
Eriocaulaceae of Continental North America, North of Mexico
Trees in the Life of the Maya World
Violets (Vi old) of Central & Eastern United States: An Introductory Survey
The Ecology of our Landscape: The Botany of Where we Live
For more information regarding details & pricing for these publics
please visit our website at www.britpress.org
limillM!: i l ll 1 1 I
3 1753 00367 1358