6 PHYTOLOGIA

An international journal to expedite plant systematic, phytogeographical
and ecological publication

Vol. 69 October 1990 No.4

CONTENTS

--G.L. NESOM, Taxonomy of the Erigeron coronarius group of
Erigeron sect. Geniculactis (Asteraceae: Astereae) ............. 237

» ~ G.L. NESOM, Two new species of Erigéron (Asteraceae: Astereae)
pay 9) ERE EAN SRR RL RE ht Se 254

| gy D.R. WINDLER, J.J. MARRON, & S.G. SKINNER, Crota/aria /otifolia
rt Linnaeus and Crotalaria purdiana Senn, a clarification
of the species (with notes on Crotalaria axillaris Aiton) ....... 258

a SUNDELL, Notes on Arizona Asclepias (Asclepiadaceae) with a
Co iech tO ihe Pe SEE is iad ee eee 265

_~ LL.B. SMITH & W.J. KRESS, New genera of Bromeliaceae ............ 271

_-+J.L. REVEAL, Minor new combinations in Toxicodendron
A EO eee ee oe Bore laaletebesteyanpersauesgeedenses 275

- SUNDBERG, S.D. & G.L. NESOM, A new species of Erigeron
Asteraceae: Astereae) from Chihuahua, México ..........c0ee 278

~ NESOM, G.L., Taxonomy of Heterotheca sect. Heterotheca

(Asteraceae: Astereae) in México, with comments on the taxa
OOP UME CPIM a RAL OSS obs ccd son low caos done deman coe ceccecreus aubsomeu does evacensel 282

Contents continued on the inside cover.

Published by Michael J. Warnock
185 Westridge Drive Huntsville, Texas 77340 U.S.A.
PHYTOLOGIA is printed on acid free paper.

- TURNER, B.L., Two new species of Verbesina (Asteraceae,
Heliantheae) from the states of Jalisco and Guerrero,
MGKIGO oc. ocdeeecscesesedepesdoechlafeachousch bas chsae qehebay icon edhe aan 295

- TURNER, B.L., Two new species of Perityle (Asteraceae, Helenieae)
From MEXICO <ciijecassesoceceiclecsceeseonceseloe lontetted euoutcn See seen 297

KARTESZ, J.T. & K.N. GANDHI, Nomenclatural notes for the North
American flOra. VV ......ccsesssccccspencsnconseassecepsananainss Siento 301

eI EO

LIBRARY

DEL 16 1990

NEW YORK
BOTANICAL GARDEN

PHYTOLOGIA (ISSN 00319430) is published monthly by Michaei J. Warnock,
185 Westridge Drive, Huntsville, TX 77340-8916. Second Class postage
at Huntsville, TX. Copyright ©1990 by PHYTOLOGIA. Domestic individual
subscription (6 issues): $18,00. Domestic institutional subscription (6
issues): $20.00. Foreign and/or airmail postage extra. Single copy sales:
Current issue and back issues volume 67 to present, $3.50; Back issues
(previous to volume 67), $3.00 (add $.50 per copy postage and handling US
[$1.00 per copy foreign]). Back issue sales by volume: $17.00 per volume
42-66 (not all available as complete volumes); $21.00 per volume 67-present,
add $2.00 per volume postage US ($4.00 per volume foreign). POSTMASTER:
rel wien on tae changes to Phytologia, 185 Westridge Drive,Huntsville, TX
40-8916.

Phytologia (October 1990) 69(4):237-253.

TAXONOMY OF THE ERIGERON CORONARIUS GROUP OF ERIGERON
SECT. GENICULACTIS (ASTERACEAE: ASTEREAE)

Guy L. Nesom
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

The Erigeron coronarius E. Greene group comprises five closely re-
lated species, ranging from south central México to the southwestern
United States, and characterized by habitats primarily in grasslands,
taproots, narrow and essentially entire leaves, erect buds, and narrow,
white, reflexing ligules. Erigeron janivultus Nesom and E. coronarius
have relatively wide ranges in México. Erigeron zacatensis sp. nov. is
a relatively narrow endemic of Zacatecas and is probably derived from
E. coronarius or E. janivultus. Erigeron arisolius sp. nov., which oc-
curs in southern Arizona and northern Sonora, is most similar to £.
coronarius. Erigeron sceptrifer sp. nov. occurs in central Chihuahua,
adjacent Sonora, and southeastern Arizona and appears to be more dis-
tantly related to the other taxa of the group. Four of the taxa are known
to exist primarily as diploids, but tetraploids also have been discovered
in three of them. In a broader context, the E. coronarius group is itself
placed within a larger group of species and the whole assemblage is rec-
ognized here as a new section, Erigeron sect. Geniculactis, typified
by E. coronarius.

KEY WORDS: Erigeron, Asteraceae, Astereae, Arizona, México

Erigeron coronarius E. Greene is a relatively common species of grassland
habitats and disturbed sites in Chihuahua and Durango, México. In studies of
Mexican Erigeron, the plants most closely similar to E. coronarius have been
recognized, including several previously undescribed taxa, two species of which
are known from both México and the United States. In the following study,
all taxa of the E. coronarius group are mapped (Maps 1 and 2) and provided
with a formal taxonomic treatment as well as an hypothesis regarding their
position within the genus.

237

238 PHY TOLOGIA volume 69(4):237-253 October 1990

E. coronarius

@ var. coronarius

© var. durangensis
4 E. janivultus

@ E. zacatensis

Map 1. Geographic distribution of Erigeron coronarius, E. janivultus, and E.
zacatensis.

Nesom: Taxonomy of the Erigeron coronarius group 239

@
e E. arisolius
® AE. sceptrifer
e e
e%
) ar i
e ce *e r
e ®e
e
\ ; “
Sits
A

Map 2. Geographic distribution of Erigeron arisolius and E. sceptrtfer.

240 PBRYT OO GIA volume 69(4):237-253 October 1990

I. TAXONOMY OF THE ERIGERON CORONARIUS GROUP

Erigeron coronarius E. Greene

Erigeron coronarius E. Greene, Pittonia 2:167. 1891. TYPE: MEXICO.
Chihuahua: Pine plains, base of the Sierra Madre, 2-3 mi S of Guerrero,
27 Sep 1887, C.G. Pringle 1275 (HOLOTYPE: US!; Isotypes: GH!,
MEXU!, NY!, PH!).

Erigeron howardit M.E. Jones, Contr. West. Bot. 12:45. 1908. TYPE:
MEXICO. Chihuahua: Colonia Judrez, Sierra Madre Mts., 12
Sep 1903, M.E. Jones s.n. (HOLOTYPE: POM!, US photo!).
Achaetogeron howardii (M.E. Jones) S.F. Blake, Contr. U.S. Natl.
Herb. 29:127. 1945.

Annuals to short lived perennials from a taproot. Stems 1-5 dm tall, single
or multiple from the base, prominently yellow ribbed, coarsely hispid with
thick based, spreading to slightly ascending, well spaced hairs, mostly along
the ribs, minutely stipitate glandular. Leaves hispid-strigose, ascending, often
strictly so, linear or narrowly oblong to linear lanceolate or linear oblanceolate,
entire or sometimes with 1-2 pairs of coarse, rounded lobes, the midcauline 1-
3 cm long, 1-2(-3) mm wide, narrowed upwards. Heads hemispheric, 5-6(-7)
mm wide, erect in bud, loosely corymboid; phyllaries with thick, raised, dark
orange midvein, abaxially and adaxially, hispid with stiffly spreading curved
hairs, minutely granular glandular, the inner 2.5-4.0 mm long. Ray flowers ca.
80-200 in 2-3 series, the corollas white, less commonly light lavender or pinkish,
6-7 mm long, with ligules 0.3-0.6 mm wide, reflexing. Disc corollas 1.8-2.5 mm
long, swollen just above the tube. Achenes 0.5-0.8 mm long, sparsely strigose
to glabrate, with 2 orange nerves; pappus of (0-)4-8 fragile bristles, the outer
a hyaline corona with a fimbriate or laciniate apex or sometimes a series of

basally fused hyaline scales 0.1-0.3 mm high.

KEY TO THE VARIETIES

1. Plants annual; stem pubescence spreading above and below; phyllaries
usually green; pappus of 4-8(-10) bristles, with a corona 0.1-0.2 mm
PAYS 5s Seara vivian piataiciatereiosdalenmea womia Suiew poss ute eee eee var. coronarius

1. Plants annual to short lived perennial; stem pubescence spreading above,
often ascending-appressed below; phyllaries usually purple tipped; pap-
pus of 2-6 bristles, with a corona 0.2-0.3 mm high .... var. durangensis

Nesom: Taxonomy of the Erigeron coronarius group 241

Erigeron coronarius E. Greene var. coronarius

Chihuahua and northern Durango; roadsides, pastures, grasslands, oak or
pine woodlands; (1450-)1850-2650 m; June-October. Chromosome numbers,
n=9, 18.

Representative collections examined: MEXICO. Chihuahua: Road to Ma-
jalco, 9 mi W of Hwy 45 turnoff, De Jong & Longpre 922 (MU) (voucher for
chromosome count of n=9, as E. divergens Torr. & A. Gray [De Jong & Long-
pre 1963]); near Balleza, 23 Sep 1898, Goldman 136 (GH, NY, US); 2.6 mi
N of jct of Hwy 16 on road to Barbidos, 4 Aug 1975, Lewis & Bierner 213
(LL) (voucher for chromosome count of n=9 by B.L. Turner [as indicated on
sheet], apparently previously unpublished); 8 mi N of Santo Tomas, 21 Sep
1939, Mueller 3888 (GH, MICH, SMU, TENN); 21 mi ESE of Cuauhtémoc
on Hwy 16, 8 Aug 1977, Nesom 626 (MEXU, TEX) (voucher for chromosome
count of n=9, reported here); 13 mi W of Cuauhtémoc on Hwy 16, 8 Aug
1977, Nesom 626 (MEXU, TEX) (voucher for chromosome count of n=9, re-
ported here); 13 mi W of Cuauhtémoc on Hwy 16, 8 Aug 1977, Nesom 627
(MEXU, TEX) voucher for chromosome count of n=18, reported here); 15 mi
W of Madera on logging road, 9 Aug 1977, Nesom R636 (MEXU, NCU, TEX)
(voucher for chromosome count of n=9, reported here); 30 mi SW of Parral
on Hwy 24 toward El Vergel, 21 Aug 1981, Nesom 4462 (MEXU, NMC, NY,
TEX); 11 mi N of Madera on Hwy 16 toward Las Varas, 24 Aug 1981, Nesom
4507 (ANSM, ASU, ARIZ, CAS, CIIDIR, COLO, ENCB, F, GH, GUADA,
IBUG, MEXU, MICH, NCU, NMC, OBI, PATZ, RSA, TEX, UC, UCR, UNM,
US, WIS); Rosario, E of La Junta, 14-15 Sep 1934, Pennell 18736 (GH, NY,
PH, US); 14 mi SW of Chihuahua City, 29 Aug 1961, Powell & Edmondson
979 (TEX) (voucher for chromosome count of n=18 [Powell & Turner 1963]);
Mapula Mts., 11 Nov 1886, Pringle 897 (MEXU, NY, US); 39 mi W of Cd.
Chihuahua on Hwy 16, 20 Aug 1967, Stuessy 1025 (DUKE, COLO, TEX, UC,
WIS) (voucher for chromosome count of n=9, as E. divergens [Keil & Stuessy
1975]); Cerro del Nido complex, 7.5 mi W of Bella Vista, on Mesa La Bo-
quilla, 15 Jul 1981, Worthington 7307 (COLO, TEX, UTEP). Durango: 7 mi
S of Las Nieves on Hwy 45, 20 Aug 1981, Nesom 4458 (ARIZ, ASU, ENCB,
GH, MEXU, NMC, NY, TEX); 11 km NW of Santiago Papasquiaro, 22 Aug
1983, Worthington 11251 (TEX); weedy playground in Santiago Papasquiaro,
25 Aug 1983, Worthington 11461 (TEX).

Erigeron coronarius E. Greene var. durangensis Nesom, var. nov. TYPE:
MEXICO. Durango: 25 mi W of Cd. Durango on Hwy 40, 7400 ft,
abundant along roadsides, 18 Aug 1981, G. Nesom 4428 (HOLOTYPE:
TEX!; Isotypes: GH!, MEXU!, US!).

Differt a E. coronario E. Greene var. coronario phyllariorum
apicibus purpureis et pappi setis paucioribus atque corona leviter
longiore.

242 PHYTOL O: GiA volume 69(4):237-253 October 1990

Durango and west central Zacatecas; grassy roadsides, disturbed sites, ar-
eas of shrublands and juniper to pine-oak woodlands; 2050-2400 m; August-
November. Chromosome numbers, n=9, 18 pairs.

Representative collections examined: MEXICO. Durango: 14 mi W of Cd.
Durango on Hwy 40, 7 Aug 1961, DeJong & Longpre 973 (TEX) (voucher
for chromosome count of n=9 [DeJong & Longpre 1963]); 38 mi W of Cd.
Durango on Hwy 40, 8 Aug 1961, DeJong & Longpre 994 (MEXU, NY, TEX)
(voucher for chromosome count of n=9 [DeJong & Longpre 1963]); ca.:31 mi
SW of Durango, 16 Aug 1960, King 3746 (MICH, TEX, UC, US) (voucher
for chromosome count of n=9 by B.L. Turner [as indicated on the sheet],
previously unpublished); Tepehuanes, 28 Jul 1944, Fisher 44261 (GH, SMU);
18 mi WSW of Durango on Hwy 40, 6 Aug 1977, Nesom 614 (GUADA, MEXU,
TEX, WIS) (voucher for chromosome count of n=9, reported here); 0.5 mi E
of Llano Grande on Hwy 40, 18 Aug 1981, Nesom 4433 (ENCB, GH, MEXU,
TEX); 3.5 mi W of Llano Grande at Las Cumbres on Hwy 40, 19 Aug 1981,
Nesom 4448 (MEXU, TEX); city of Durango and vicinity, Apr-Nov 1896,
Palmer 851 (GH, MO, UC, US); city of Durango and vicinity, Apr-Nov 1896,
Palmer 929 (GH, MO, NY, UC, US); Otinapa, 25 Jul-5 Aug 1906, Palmer
424 (CM, GH, MO, NY, UC, US); Sandia Station, 16 Oct 1905, Pringle 18647
(ARIZ, CAS, GH, LL, MICH, SMU, TEX, US); 5 mi W of Durango on Hwy
40, 28 Jun 1974, Roberts & Keil 10838 (OS) (voucher for chromosome count
of n=9 [Keil & Stuessy 1977]); 11 km NW of Santiago Papasquiaro, 22 Aug
1983, Worthington 11218 (TEX). Zacatecas: Ca. 2 mi W of Sombrerete on
Hwy 45, 18 Aug 1981, Nesom 4424a (TEX).

My conclusions here contrast with those of an earlier, unpublished study
(Nesom 1980), where I viewed the plants from the southern half of Durango
as most closely related to Erigeron janivultus Nesom and referred to them
as a variety of the latter species. Both of these taxa might be treated, with
some justification, as varieties of E. coronarius, particularly since the Du-
rango plants are intermediate in geography between typical FE. coronarius and
E. janivultus. However, although they have purplish phyllaries and appressed
pubescence (on the lower stems) as in E. janivultus, they are more similar to
E. coronarius in their minutely stipitate glandular upper stems with spread-
ing hairs and their pappus with more numerous bristles and a much shorter
corona. Further, there is a much stronger discontinuity in morphology between
var. durangensis and E£. janivultus than exists between var. coronartus and var.
durangensis. Plants with distinctly purple tipped phyllaries but 6-8 pappus
bristles and short coronas, morphologically intermediate and intergrading be-
tween var. coronarius and var. durangensis, occur from northern Durango into
southern Chihuahua (e.g., Nesom 4462, Pringle 18647, Worthington 11219).

Diploids have been reported from both varieties of Erigeron coronarius
as well as from E. janivultus. While diploids appear to be more common,
tetraploids also are known from each of these three taxa (see citations and

Nesom: Taxonomy of the Erigeron coronarius group 243

references in “Representative collections” from this paper, and De Jong &
Nesom, in prep.), but within a single taxon, the tetraploids appear to be mor-
phologically indistinguishable from diploids. Only diploids have been reported
from E. arisolius.

Erigeron arisolius Nesom

Erigeron arisolius Nesom, sp. nov. TYPE: UNITED STATES. Arizona:
Cochise Co., W side of Chiricahua Mts., Hwy 181, ca. 9.5 mi S of ject
with Hwy 186, near Turkey Creek Road, abundant along roadside and
in grassy plains, 4800 ft, 28 Aug 1981, G. Nesom 4521 (HOLOTYPE:
TEX!: Isotypes: ARIZ!, ASU!, CAS!, COLO!, ENCB!, GH!, MEXU!,
MO!, NMC!, NY!, RM!, UC!, UCR!, UNM!, US!, WIS!).

E. coronario E. Greene similis sed pappo setis (10-)12-17 atque
serie exteriore setarum discretarum vel squamellarum anguste lance-
olatarum differt.

Annuals to short lived perennials from a taproot. Stems 3-7 dm tall, single
or multiple from the base, prominently yellow ribbed, coarsely hispid with
thick based, spreading to slightly ascending, well spaced hairs mostly along
the ribs, minutely stipitate glandular. Leaves hispid-strigose, ascending, often
strictly so, linear or narrowly oblong to linear lanceolate or linear oblanceolate,
entire or sometimes with 1-2 pairs of coarse, rounded lobes, the basal and lower
cauline 2.5-5.0 cm long, 2-5 mm wide, narrowed upwards. Heads hemispheric,
5-8 mm wide, erect in bud, in loose corymbs; phyllaries with a thick, raised,
dark orange midvein, abaxially and adaxially, hispid with stiffly spreading
curved hairs, minutely granular glandular, narrowly lanceolate, nearly equal
in length, 2.5-3.5 mm long. Ray flowers 125-180 in 2-3 series, the corollas
white, less commonly light lavender or pinkish, 6-7 mm long, with ligules 0.6-
1.0 mm wide, reflexing. Disc corollas 2.0-2.5 mm long, glabrous, swollen just
above the tube. Achenes 0.7-1.0 mm long, sparsely strigose to glabrate, with
2 orange nerves; pappus of (10-)12-16 fragile bristles, with an outer series of
separate setae or lanceolate scales 0.1-0.3 mm high. Chromosome number,
n=9 pairs.

Arizona (Apache, Cochise, Pima, and Santa Cruz counties), Sonora; grass-
lands, often in moist areas, sometimes with mesquite, areas of oak woodlands
in grassy openings or roadsides; (700-)1300-1650 m; May-October(-November).

Representative collections examined. MEXICO. Sonora: Mpio. Altar,
Rancho “La Discordia,” 24 Aug 1969, Araiza Q. 62 (ENCB); Fronteras, 25
Sep 1890, Hartman 54 (GH, US); S-gogogsig village, SW of Baboquivaris,

244 PHY TO LOG LA volume 69(4):237-253 October 1990

floodwater field, 14 Aug 1980, Nabham & Bahre 312 (ARIZ); 6 mi E of Fron-
teras on Fronteras-Nacozari hwy, Angostura Dam Road, 22 Aug 1981, Roth
s.n. (TEX); Rancho Carrizo, 103 mi S of Nogales and 3 mi W of Hwy 15, 30
Aug 1967, Tomelson s.n. (ARIZ).

UNITED STATES. Arizona. Apache County: 27 mi W of Carrizo, 12
Jun 1937, Peebles 19579 (ARIZ). Cochise County: Mule Mts., 25 Sep 1931,
Harrison 8242 (ARIZ, TEX, US); Chiricahua Mts., W Turkey Creek at Rock
Creek Road, 28 Jul 1981, Luetke s.n. (TEX); Mexican boundary line, S of
Bisbee, 14 Sep 1892, Mearns 903 (US); W side of Chiricahua Mts., Pinery
Creek Road, 4 mi SE of jct with road to Chiricahua Natl. Monument, 28 Aug
1981, Nesom 4519 (ARIZ, ASU, GH, MEXU, NMC, NY, TEX, UNM, US);
Huachuca Mts., 0.6 mi W of Hwy 92 on Ash Canyon Road, 30 Aug 1984,
Sanders 5138 (ARIZ, TEX); Parker Canyon Lake, Canelo Hills, 7 Jul 1985,
Sanders 5923 (TEX); Dragoon Mts., Sep 1930, Thornber s.n. (ARIZ). Pima
County: Within city of Tucson, 2 Sep 1980, Adams 271 (LL); Robles to San
Fernando, 21 Aug 1932, Harrison & Kearney 8941 (ARIZ, TEX, US); 10 mi
N of San Miguel, Papago Indian Reserv. near international boundary, 16 Aug
1931, Harrison & Kearney 8029 (ARIZ, TEX); ca. 2 mi W of Roble Jct along
Ariz Rte 86, 18 Oct 1975, Keil 11146A (ASU); (voucher for chromosome count
of n=9 pairs, reported as E. divergens [Keil & Pinkava 1976]); ca. 2 mi W of
Roble Jct along Ariz Rte 86, 18 Oct 1975, Keil 11158 (ASU) (voucher for
chromosome count of n=9 pairs, reported as E. divergens [Keil & Pinkava
1976]); just off Hwy I-19 on Papago Rd exit, 6 Sep 1975, Keil 11089A (ASU)
(voucher for chromosome count of n=9 pairs, reported as E. divergens [Keil
& Pinkava 1976]); Santa Rita Mts., Box Canyon Road, 3.2 mi W of Hwy 83,
30 Aug 1985, McLaughlin 3035 (ARIZ); mouth of Baboquivari Canyon, 15
Aug 1926, Peebles, et al. 2788 (ARIZ, TEX); Santa Catalina, 1 May 1894,
Toumey s.n. (GH). Santa Cruz County: Between Pena Blanca and Ruby, 2.4
mi W of road to Sycamore Canyon, 9 Sep 1976, McGill & Lehto L20432(ASU,
TEX); NW of Sonoita, 3.7 on road toward Gardner Canyon from Hwy 83, 14
Aug 1982, Sundberg 1595 (TEX) (voucher for chromosome count of n=9 pairs
by S. Sundberg {as indicated on sheet], apparently previously unpublished);
Pajarito Mts., Sycamore Canyon, 17 Aug 1976, Van Devender & McCarten
s.n. (ARIZ).

Erigeron arisolius in Arizona has been identified as E. divergens Torr. &
A. Gray, although a number of sheets have annotations noting that the plants
are not typical of the latter. In spite of the morphological similarity between
the two, E. divergens belongs to a different section of the genus (sect. Oly-
gotrichium (see Nesom 1989), and there is no evidence of any form of gene
flow between them. They are particularly distinct in the field, where the erect
buds and reflexing rays of E. arisolius can be contrasted with the nodding
buds and straight rays of E. divergens. The following couplet summarizes
differences between them.

Nesom: Taxonomy of the Erigeron coronarius group 245

1. Stems with coarse, thick based hairs arising mostly from the cauline ribs,
minutely but prominently stipitate glandular; buds erect; rays reflexing;
achenes with orange nerves; pappus bristles 10-16 ......... E. arisolius

1. Stems with relatively thin based hairs arising evenly from the ribs and
interstices, sometimes minutely granular glandular, rarely stipitate glan-
dular near the heads; buds nodding; rays straight or closing upwards;
achenes with whitish nerves; pappus bristles 7-10 ......... E. divergens

Erigeron arisolius is most similar to E. coronaritus. Although the two
species are similar in habit and most morphological features, they have en-
tirely separate geographic ranges and I have seen no specimens that could
be regarded as intermediate. They are distinguished by the contrasts in the
following couplet, most sharply by features of the pappus.

1. Plants 3-7 dm tall; largest leaves 2.5-5.0 cm long, 2-5 mm wide; rays with
ligules 0.6-1.0 mm wide; achenes 0.7-1.0 mm long; pappus of (10-)12-17
bristles, with an outer series of separate setae or narrowly lanceolate
Squainellacs so. oe SOE eee eae oe ee oan na eee E. arisolius

1. Plants 1-5 dm tall; largest leaves 1-3 cm long and 1-2(-3) mm wide; rays
with ligules 0.3-0.6 mm wide; achenes 0.5-0.8 mm long; pappus of (2-)
4-8(-10) bristles, with an outer, hyaline corona of broad, basally fused
sguamellae 2.5. Act Meeie sles cmtine ads Po seme sateen dt E. coronarius

Although Erigeron artsolius is a discrete taxon, not intergrading with oth-
ers of the E. coronarius group, it may be viewed as the northern extreme in
a north/south trend of variation in several morphological characters, from E.
arisohius to E. coronarius var. coronarius, to var. durangensis, and finally to
E. janivultus, the southern extreme. From north to south, the plants of this
assemblage generally become shorter in stature, longer lived, fewer headed,
and the pappus bristles are reduced in number and the corona increased in

height.

Erigeron janivultus Nesom

Erigeron janivultus Nesom (nom. nov.), Sida 9:223. 1982. Based on: Erigeron
lineartfolius S. Wats., Proc. Amer. Acad. Arts 26:139. 1891. TYPE:
MEXICO. México: Bluffs and plains near Flor de Maria, 4 Sep 1890,
C.G. Pringle 3242 (HOLOTYPE: GH!; Isotypes: CM!, ENCB!, MEXU-
2 sheets!, MSC!, NY!, PH!, UC-2 sheets!, US-2 sheets!). Not Erigeron
linearifolius Cav., 1801.

246 PH Y?ROGO GTA volume 69(4):237-253 October 1990

Achaetogeron ascendens Greenm., Proc. Amer. Acad. Arts 41:254. 1905.
TYPE: MEXICO. Hidalgo: Meadows near Buena Vista Station
(Cuyamaloya), 4 Aug 1904, C.G. Pringle 8851(HOLOTYPE: GH!;
Isotypes: CM!, ENCB!, MEXU-2 sheets!, NMC!, NY!, PH!, UC!,
US-2 sheets!). Not Erigeron adscendens Turcz., 1851.

Short lived perennial herbs from a taproot, often with a short, woody
caudex with slender branches. Stems 4-33 cm tall, sparsely but consistently
strigose, sometimes minutely stipitate glandular just below the heads. Leaves
strigose with appressed-ascending trichomes, minutely stipitate glandular, the
basal usually absent by flowering, the cauline linear to linear lanceolate, ses-
sile, mostly 5-30 mm long, 0.5-3.0 mm wide, even sized upwards, the margins
entire or with 1-2 pairs of narrow, lateral lobes. Heads hemispheric, 5-9 mm
wide, in a loose corymb on peduncles 3-28 mm long; phyllaries in 3-4 series of
nearly equal length, the longest 4-5 mm long, mostly narrowly lanceolate, usu-
ally purple tipped or purple over the whole surface, hirsute-strigose, minutely
stipitate glandular, remaining erect after maturation and release of the ach-
enes. Ray flowers ca. 80-200 in 2-3 series, the corollas drying white or lavender
tinged, 6-9 mm long, the ligules 0.7-0.9 mm wide, reflexing at maturity. Disc
corollas 2-4 mm long, the lower limb somewhat inflated and indurated; collect-
ing appendages of the style branches triangular to shallowly deltate. Achenes
(0.8-)1.0-1.5 mm long, with 2(-3) thickened ribs, sparsely strigose; pappus of
ray and disc a conspicuous hyaline corona with a fimbriate or laciniate apex
0.3-0.7 mm high, the inner series of 0-5 bristles 1/2-3/4 as long as the disc
corollas. Chromosome numbers, n=9, 18 pairs.

Durango, Zacatecas, Jalisco, Hidalgo, México, Tlaxcala; roadsides, grass-
lands, openings in oak woods; 1850-2800 m; June-October.

Representative collections examined: MEXICO. Hidalgo: 10 km W of
Alfajayucan, 16 Jul 1965, Gonzalez Q. 2702 (ENCB); toward Singuilucan, 2
mi S of jct with Hwy 130 at Cuyamaloya, 9 Aug 1981, Nesom 4382 (ENCB,
GH, MEXU, RM, TEX); Marques Station, 26 Aug 1905, Pringle 13534 (ARIZ,
MICH, MSC, NMC, SMU); 1 km S of Guajolote, near Epazoyucan, 22 Jul
1979, Rzedowski 36242 (ENCB); 12 km E of Apam, 27 Jun 1966, West FF10
(MICH, WIS). Jalisco: ca. 11 mi SE of Lagos de Moreno, 7 Sep 1952, Mc Vaugh
12815 (MICH). México: 5 km N of Atlacomulco, 28 Jun 1960, Beaman 3370
(TEX) (voucher for chromosome count of n=9, as E. coronarius [Beaman &
Turner 1962]); 9 mi E of Villa Victoria on Hwy 15, 29 Jul 1965, Kral 25156
(ENCB); 11 km NNW of Ixtlahuaca on Hwy 55, 16 Aug 1981, Nesom 4412
(ASU, ARIZ, CIIDIR, COLO, GH, GUADA, MEXU, MO, NMC, NY, RM,
TEX, US, WIS); 1.5 km N of Atlacomulco, 16 Aug 1981, Nesom 4420(MEXU,
TEX); Flor de Maria, 10 Sep 1892, Pringle 7368 (NY, POM); 20 km NE of
Texcoco, 23 Jun 1966, Rzedowski 22462 (ENCB, MICH, MSC); 6 mi S of
Acambay, 7 Nov 1970, Webster & Breckon 16246 (MICH). Tlaxcala: NW of

Nesom: Taxonomy of the Erigeron coronarius group 247

Cauhula, W of San Antonio Calpulalpan, 8 Oct 1967, Villegas 676 (ENCB).
Zacatecas: ca. 2 mi ESE of Sombrerete on Hwy 45, 5 Aug 1977, Nesom R611
(CIIDIR, GH, MEXU, MICH, NY, TEX, US); ca. 2 mi W of Sombrerete on
Hwy 45, 18 Aug 1981, Nesom 44246 (CIIDIR, MEXU, MICH, NY, TEX).

Erigeron janivultus is distinguished from EF. coronarius by its longer du-
ration, lignescent, branching caudex, basally ascending stems, strigose stem
pubescence, eglandular upper stems and phyllaries, and thicker ribbed, longer
achenes with a longer coroniform pappus and fewer bristles.

Two collections from the vicinity of Sombrerete in west central Zacatecas
(Nesom 611 and 44246) are intermediate in some respects between E. janzvul-
tus and E. coronarius. Although I have identified them as the former, empha-
sizing their eglandular stems with appressed pubescence, they are similar to
E. coronarius in their short disc corollas and pappus with a short (0.2-0.3 mm)
corona and relatively numerous (5-8) bristles. In fact, further collections in
this region might furnish evidence that the two taxa are best considered con-
specific, even though they are very different at their geographic extremities.
Earlier, Blake (1945) also recognized the close relationship between E. coro-
narius and E. janivultus and accurately described essential differences between
them. At the collection site for Nesom 4424), relatively typical E. coronarius
was also collected (Nesom 4424a) as well as the discoid taxon referred to below
as E. zacatensis (which see for additional comments).

Erigeron zacatensis Nesom, sp. nov. TYPE: MEXICO. : Zacatecas: ca. 2
mi W of Sombrerete on Hwy 45, at crest of hill overlooking town, grazed
area with scattered, low Fabaceous shrubs and Opuntia, very scattered,
8200 ft, 18 Aug 1981, G. Nesom 4423 (HOLOTYPE: TEX!; Isotypes:
IBUG!, CIIDIR!, ENCB!, GH!, MEXU!, MICH!, NY!, US!).

Differt a E. janivulto Nesom ligulis multum redactis et pappi
setis numerosioribus atque corona breviore.

Annual to short lived perennial herbs from a taproot, often with a short,
woody caudex with slender branches. Stems 10-20 cm tall, erect to basally as-
cending, few branched at about the middle, sparsely pubescent with appressed
to spreading ascending hairs, eglandular. Leaves strigose with appressed-
ascending trichomes, eglandular, the basal usually absent by flowering, the
cauline linear to linear lanceolate, sessile, mostly 5-30 mm long, 0.5-2.0 mm
wide, even sized upwards, the margins entire. Heads hemispheric, 8-11 mm
wide, in a loose corymb on peduncles mostly 2-6 cm long; phyllaries in 3-4
series of nearly equal length, the longest 2.5-4.0 mm long, mostly narrowly
lanceolate, usually purple tipped, hirsute strigose, minutely granular glandu-
lar, spreading after maturation and release of the achenes, but not reflexing.
Ray flowers ca. 150-250 in 2-3 series, fertile, the ligules 0.1-0.2 mm wide, 0.5-
1.0 mm long, not extending past the phyllaries. Disc corollas 2.2-2.8 mm long,

248 PHY TOLOGLA volume 69(4):237-253 October 1990

the lower limb somewhat inflated and indurated; collecting appendages of the
style branches triangular to shallowly deltate. Achenes 0.5-0.8 mm long, with
2(-3) thickened, whitish ribs, sparsely short strigose; pappus of ray and disc
a conspicuous hyaline corona with a fimbriate or laciniate apex 0.2-0.3 mm
high, the inner series of 5-10 bristles 3/4 as long as the disc corollas.

Zacatecas; shrublands, grasslands; 2150-2500 m; August-October.

Additional collections examined: MEXICO. Zacatecas: ca. 20 mi NW of
Fresnillo on Hwy 45, Davidse 10005 (MO); 8 mi NW of jct in Hwys 45 and
49 on Hwy 49, 29 Jun 1972, Denton 1721 (ENCB, MICH); Hwy 45, 1 km
E of turnoff to Est. Frio, 23 Aug 1979, Lane 2704 (TEX); between Fresnillo
and Sombrerete on Hwy 45, 6.0 mi W of turnoff to Santa Rosa, 28 Sep 1984,
Sundberg 2914 (MEXU, TEX).

In their relatively short disc corollas and pappus of a short corona with 5-10
bristles, these plants are similar to Erigeron coronarius, but in their eglandular
stems with appressed-ascending hairs, they are more similar to E. janivultus,
particularly those plants so identified from Zacatecas. Erigeron zacatensis
differs from both taxa in the greatly reduced ligules that barely extend past
the involucral bracts, although it may be a derivative of one or the other or
both (as a hybrid) of them. Nearly identical plants have been collected at
several separate, relatively closely situated localities in west central Zacatecas
(Map 1), apparently indicating that the taxon is stabilized and reproducing.
Plants of both E. coronarius and E. janivultus also were growing at the type
locality of E. zacatensis (see comments following EF. janivultus).

Erigeron sceptrifer Nesom, sp. nov. TYPE: MEXICO. Chihuahua: Mpio.
Cuauhtémoc, 17.6 km W of Cuauhtémoc on Hwy 16, abundant along
grassy roadsides, area of pine-oak-juniper, 23 Aug 1981, G. Nesom 4477
(HOLOTYPE: TEX!; Isotypes: ANSM!, ARIZ!, ASU!, CAS!, CHAPA!,
CIIDIR!, COLO!, ENCB!, F!, GH!, GUADA!, IBUG!, MEXU!, MICH!,
MO!, NMC!, NY!, OS!, RM!, RSA!, WIS!, UC!, UCR!, US!).

E. coronario E. Greene similis sed corollis radii brevioribus et
pappo setarum numerosioribus sine corona exteriore imprimis dif-
fert.

Plants annual from a slender taproot. Stems 3-8 dm tall, single from the
base or more commonly with numerous, ascending branches from the base,
moderately pubescent with ascending to spreading hairs arising primarily from
the prominent ridges, minutely granular glandular to essentially eglandular.
Leaves entire, linear to linear oblanceolate, 1-3 cm long, 0.5-1.5 mm wide
(the lowermost sometimes up to 4.0 mm wide and with 1-2 pairs of teeth),
minutely granular glandular above and below, moderately pubescent with as-
cending hairs, these commonly thicker based and more dense on the midrib and

Nesom: Taxonomy of the Erigeron coronarius group 249

marginal areas. Heads hemispheric, 4-6 mm wide, on short peduncles, numer-
ous in corymboid panicles; phyllaries linear lanceolate, in 3-4 weakly graduated
series, the inner 3.0-3.5 mm long, very sparsely hirsute strigose, eglandular or
the outer sparsely and minutely granular glandular; receptacles shallow con-
vex. Ray flowers (60-)85-130(-195), 3.8-5.0 mm long, the ligules 0.2-0.4 mm
wide, white, drying pink to blue or purple. Disc corollas 1.8-2.3 mm long, with
the lower limb somewhat indurate inflated; collecting appendages of the style
branches deltate, 0.1 mm long. Achenes 0.6-0.7 mm long, compressed, with
2 orange nerves; pappus of 9-14 fragile bristles slightly shorter than the disc
corollas, with a few outer setae or lanceolate scales 0.1-0.2 mm long.

Chihuahua, Sonora, southeastern Arizona; grasslands, with scattered ju-
niper, oak, or oak-pine; 1900-2100 m; July-October.

Additional collections examined: MEXICO. Chihuahua: 15.6 mi S of
Zaragoza, road from Cuauhtémoc to Buenaventura, 4 Aug 1975, Lewis &
Bierner 216 (LL); Hwy 16, 1.5 mi E of jct at La Junta with road toward
Creel, 23 Aug 1981, Nesom 4483 (ANSM, ARIZ, ASU, CAS, CIIDIR, COLO,
ENCB, GH, GUADA, MEXU, MICH, NMC, NY, RM, RSA, TEX, WIS, UCR,
US); 7.5 mi SW of La Junta (Hwy 16 jct) on road toward Tomochic, 24 Aug
1981, Nesom 4494 (ARIZ, ENCB, MEXU, TEX); 2 mi W of Matachic on Hwy
16, 24 Aug 1981, Nesom 4497(ASU, CIIDIR, MEXU, TEX); 9 mi W of Guer-
rero on Hwy 16, 24 Aug 1981, Nesom 4495 (ARIZ, ASU, CHAPA, CIIDIR,
ENCB, GH, MEXU, NMC, NY, TEX, US). Sonora: Dist. Alamos, lower part
of Estrella Canyon, 31 Oct 1933, Gentry 722 (DS).

UNITED STATES. Arizona: Cochise Co., Parker Canyon Lake, 12 Oct
1968, Pinkava, et al. 14609 (ASU, NCU).

Erigeron sceptrifer Nesom is recognized by its annual duration, spreading-
ascending stem pubescence, linear leaves, erect buds, numerous, very small
heads with short rays that become purplish upon maturity or drying, small,
orange-ribbed achenes, and its pappus of 9-14 fragile bristles, without an outer
corona. The rays reflex at maturity, although they tend to be more curved re-
flexed than distinctly bent at the tube/ligule junction, as are the other species
closely related to it. It is tentatively included here, but the ligules also have a
slight tendency to turn upwards at night, a behavior that is anomalous among
the species of the E. coronarius group. At two of its collection sites in Chi-
huahua, E. sceptrifer was growing intermixed with E. coronarius, which was
the more abundant, but there were no plants with any degree of intermediacy
between the two.

250 PH YT OU O'CLTA volume 69(4):237-253 October 1990

ARTIFICIAL KEY TO SPECIES OF THE ERIGERON CORONARIUS
GROUP

1. Ray corollas not extending past involucral bracts, the heads appearing
CBSENPIENY CISCOMD 6025 cepts = egies ood are ¢b eyeie ae Pe E. zacatensis

1’ Ray corollas prominent, extending well past the involucral bracts .... (2)

2. Ray corollas 3.8-5.0 mm long; pappus of 9-14 bristles, with an outer
series of separate bristles or scales ................... E. sceptrifer

2’ Ray corollas 6-9 mm long; pappus of 0-17 bristles, with an outer
corona or series of separate bristles or scales .................. (3)

3. Pappus of (10-)12-17 bristles, with an outer series of separate setae or
narrowly lanceolate squamellae ...............-..eeeeeeees E. arisolius

3’ Pappus of 0-8 bristles, with an outer series fused into a definite, complete
or nearly, complete; corona : .): «0.5.54. .v)0. «She bcl ee eee (4)

4. Stems with appressed to appressed-ascending hairs; stems, leaves,
and phyllaries eglandular; pappus with 0-5 bristles; disc corollas 2-4
Mt LONG. 5s, 5 oshg tae ens ae 6 Waves ache eee E. janivultus

4’ Stems with spreading to spreading ascending hairs, at least on the
upper portions of the stems; upper stems, leaves and phyllaries
glandular; pappus with (2)4-10 bristles; disc corollas 1.6-2.5 mm
longer: «jield .A ask, hati ee E. coronarius

I]. A NEW SECTION OF ERIGERON, INCLUDING THE E.
CORONARIUS GROUP

Erigeron coronarius and its close relatives were tentatively positioned near
the “E. pumilus group” of sect. Asteroidea, primarily based on their essentially
entire leaves, erect buds, white, reflexing ligules, and basally persistent pappus
bristles (Nesom 1989). The similarities among these taxa strongly suggest
that the whole group is monophyletic, and I recognize it as a separate section.
Distinctions among its taxa restricted to the United States were outlined in
an earlier paper (Nesom 1983).

Erigeron sect. Geniculactis Nesom, sect. nov. TYPE SPECIES: Erigeron
coronarius E. Greene.

Nesom: Taxonomy of the Erigeron coronarius group 251

Erigeron (sp. group) Pumili Rydb., Fl. Colorado 359. 1906, in clave.
TYPE SPECIES: Erigeron pumilus Nutt., Gen. N. Amer. Pl. 2:147.
1818.

Habitationibus gramineis, radicibus palaribus, foliis angustis
admodum integrisque, gemmis erectis, flosculis radii numerosis li-
gulis albis reflexis, pappo setis basaliter persistentibus et serie ex-
terna saepe valde evolutis dignoscenda.

Species included: Erigeron aphanactis (A. Gray) E. Greene, E. arisolius
Nesom, E. clokeyi Crong., E. concinnus (Hook. & Arn.) Torr. & A. Gray,
E. coronarius E. Greene, E. engelmanni A. Nels., E. goodrichu Welsh, E.
janivultus Nesom, E. pumilus Nutt., E. sceptrifer Nesom, and E. zacatensts
Nesom.

The plants of sect. Geniculactis are distinguished by their grassland habi-
tats, taproots, narrow and essentially entire leaves, erect buds, numerous ray
flowers with white, reflexing ligules, disc corollas with swollen, white indurated
throats, and pappus with relatively few, basally persistent bristles and a ten-
dency to produce a strongly elaborated outer series. In contrast to the taxa
occurring primarily in México (the “Erigeron coronarius group”), which tend
to produce numerous heads in paniculate cymes and achenes mostly 0.6-1.0(-
1.5) mm long, those species restricted to the United States (the “E. pumillus
group”) tend to produce solitary heads on monocephalous stems and achenes
1.6-2.2 mm long.

While it is almost certain that these two groups represent separate evolu-
tionary lineages, several parallel developments between their respective species
are notable. The pappus of Erigeron concinnus is a highly elaborated whorl of
broad, sometimes basally united scales, very similar to the outer pappus of E.
coronarius. Further, the ligules of the pistillate flowers of E. aphanactis vary
from completely absent to filiform projections ca. 1 mm long, matching those
found in E. zacatensis.

Elsewhere in Erigeron, outside of sect. Geniculactis, reflexing ligules are
known with certainty only in sect. Scopulincola (Nesom 1989b, 1990), which is -
distinguished by its cliffside habitats, obovate leaves, erect buds, and relatively
few ray flowers with broad ligules, and in sect. Polyactis (Nesom 1989a), which
is distinguished by its pinnatifid or coarsely toothed leaves, arching- pendant
buds, and basally caducous pappus bristles or lack of bristles. As noted earlier
(Nesom 1989b), two monocephalous species of the EF. caespitosus Nutt. group
of sect. Asteroidea also appear to have reflexing rays and need to be inves-
tigated in more detail for the possibility of a close relationship with the E.
pumilus group of sect. Geniculactis.

252 PH Y-T.OL.0:G 1A volume 69(4):237-253 October 1990
ACKNOWLEDGMENTS

I thank Dr. B.L. Turner and Dr. P.O. Karis for their review and comments.
Loans of specimens were obtained from ARIZ, ASU, CAS, CM, COLO, ENCB,
GH, MEXU, MICH, MSC, NMC, NY, OS, PH, POM, RM, SMU, TENN, UC,
UCR, UNM, US, UTEP, and WIS.

LITERATURE CITED

Beaman, J.H. & B.L. Turner. 1962. Chromosome numbers in Mexican and
Guatemalan Compositae. Rhodora 64:271-276.

Blake, S.F. 1945. Asteraceae described from Mexico and the southwestern
United States by M.E. Jones, 1908-1935. Contr. U.S. Natl. Herb. 29:117-
137.

De Jong, D.C.D. & E.K. Longpre. 1963. Chromosome studies in Mexican
Compositae. Rhodora 65:225-240.

Keil, D.J. & D.J. Pinkava. 1976. Chromosome counts and taxonomic notes
for Compositae from the United States and Mexico. Amer. J. Bot.
63:1393-1403.

Keil, D.J. & T.F. Stuessy. 1975. Chromosome counts of Compositae from
the United States, Mexico, and Guatemala. Rhodora 77:171-195.

. 1977. Chromosome counts of Compositae from Mexico and the
United States. Amer. J. Bot. 64:791-798.

Nesom, G.L. 1980. A revision of the epappose species of Erigeron (Asteraceae—
Astereae). Ph.D. dissertation, University of North Carolina, Chapel Hill.

. 1982. Nomenclatural changes and clarifications in Mexican Erigeron

(Asteraceae). Sida 9:223-229.

. 1983. Taxonomy of Erigeron concinnus (Asteraceae) and its sepa-
ration from E. pumilus. Sida 10:159-166.

. 1989a. Taxonomy of Erigeron sect. Polyactis (Compositae: Aster-
eae). Phytologia 66:415-455.

. 1989b. Infrageneric taxonomy of New World Erigeron (Asteraceae:
Astereae). Phytologia 67:67-93.

Nesom: Taxonomy of the Erigeron coronarius group 253

. 1990. Taxonomy of the Erigeron pringlet group (Asteraceae: Aster-
eae). Phytologia 69:226-233.

Powell, A.M. & B.L. Turner. 1963. Chromosome numbers in the Compositae.
VII. Additional species from the southwestern United States and Mexico.

Madrono 7:128-140.

Phytologia (October 1990) 69(4):254-257.

TWO NEW SPECIES OF ERIGERON (ASTERACEAE: ASTEREAE)
FROM MEXICO

Guy L. Nesom
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

Erigeron macdonaldii sp. nov. (sect. Jmbarba) is described from
the peak of Cerro Mohinora in southern Chihuahua, where apparently it
is narrowly endemic. Erigeron vicinus sp. nov. (sect. Olygotrichium)
is described from the Sierra Madera del Carmen in northwestern Coahuila.

KEY WORDS: Erigeron, Asteraceae, Astereae, México

Two previously undescribed species have come to light in studies of Mex-
ican Erigeron. One is from an area of high elevation in southern Chihuahua,
the other from a montane area of northwestern Coahuila.

Erigeron macdonaldii Nesom, sp. nov. TYPE: MEXICO. Chihuahua:
Mpio. Guadalupe y Calvo, ca. 13 mi SW of Guadalupe y Calvo; rock
faces on NW side of uppermost summit of Cerro Mohinora; plants in
crevices at lip of precipice, with scattered, “krumholz” pines, 3250 m,
21 Aug 1988, McDonald & Nesom 2472 (HOLOTYPE: TEX!; Isotypes:
GUADA!, GH!, MEXU!, NY!, US!).

E. fraternus E. Greene similis sed plantis parvioribus, capitulis
solitar.is, floribus radii parvioribus, et acheniis setas pappi efferen-
tibus differt.

Perennial herbs from a system of short, slender, lignescent rhizomes with
fibrous roots. Stems basally ascending, 9-13 cm tall, sparsely pubescent with
spreading to ascending appressed hairs, eglandular. Leaves eglandular, strigose
adaxially with closely appressed hairs, nearly glabrous abaxially, the basal
leaves persistent, spatulate, 15-50 mm long with petioles 1/2-2/3 the length of
the leaf, the blades 4-10 mm wide, entire or usually with 1-2(-3) pairs of teeth
or shallow lobes, the cauline leaves narrowly oblanceolate, not clasping, entire,
5-12 mm long, widely and evenly spaced, little reduced upward. Heads solitary,

254

Nesom: Two new species of Erigeron from México 255

5-9 mm wide; phyllaries purple tipped, sparsely strigose hirsute, eglandular, in
2-3 series of nearly equal length, the longest 3-4 mm long. Ray flowers (9-)13-24
in a single series, the corollas 8-9 mm long, the ligules 1.5-2.0 mm wide, white,
drying white to purplish, remaining essentially straight at maturity. Disc
corollas 2.5-2.8 mm long, funnelform, not swollen or indurated; style branches
with triangular collecting appendages 0.2 mm long. Achenes ca. 1.5 mm long,
2 nerved, sparsely strigose; pappus of 5-7 basally persistent bristles 1/2-2/3
the length of the disc corollas, with an outer series of minute squamellae ca.
0.1 mm long.

The epithet commemorates Dr. Andrew McDonald, who will be remem-
bered for his comprehensive systematic studies of the Convolvulaceae as well
as his surveys of the alpine vegetation of México.

In its geographic location and its rhizomatous habit, toothed leaves, erect
buds, and basally 3 ridged phyllaries, Erigeron macdonaldii clearly is most
similar to the seven species of sect. Imbarba Nesom’(Nesom 1989b, 1989c),
which are restricted to México and Guatemala, and primarily to the Sierra
Madre Occidental of México. Particularly in their sparse vestiture, small heads
with few, short rays, and thin, obovate-spatulate, strongly toothed leaves, the
plants of E. macdonaldi are most similar to E. fraternus E. Greene. The latter
ranges from central Chihuahua to central Durango and grows in the immediate
area of Cerro Mohinora, although it is not found on the highest part of the
peak itself.

Erigeron macdonaldi differs from E. fraternus in features summarized in
the following couplet.

1. Plants from a relatively diffuse system of slender rhizomes; stems 9-13 cm
tall; basal leaves 4-10 mm wide; heads solitary; ray flowers 9-24; pappus
of 5-7 short, persistent bristles ....... 0... -:<200) Aee.L et E. macdonaldit

1. Plants usually with a definite caudex, producing basal offsets; stems mostly
15-60 cm tall; basal leaves mostly 5-17(-23) mm wide; heads few, but in
a loose corymb; ray flowers 18-80; pappus without
TAPAS EGR i ite ta ale eth oc ical cee Btw a ent a eeiahals E. fraternus

The new species is the only member of sect. Jmbarba to consistently pro-
duce a pappus with bristles, although the bristles are relatively few and short.
Erigeron fraternus rarely produces one or two, apparently vestigial bristles,
0.3-1.0 mm long, but like the other species of the section, a short corona may
be produced but the pappus normally is completely without bristles. In this
respect, E. macdonaldii may be regarded as the most primitive among its
immediate relatives.

Erigeron macdonaldii is known only from a single site and one collection
from the uppermost reaches of the peak of Cerro Mohinora, where the plants
grow on the steep rock walls that shear off toward the northwest. This peak and

256 PHYTOL OGIra volume 69(4):254-257 October 1990

the surrounding area in southern Chihuahua is particularly rich in species of
Erigeron. Four species of sect. Polyactis (Nesom 1989a) are primarily restricted
in geographic range to that area, and E. mohinorensis Nesom of sect. Jmbarba
is endemic to the open pine-fir woods in the area immediately around the peak.

Erigeron vicinus Nesom, sp. nov. TYPE: MEXICO. Coahuila: Mpio.
Ocampo, Sierra Madera del Carmen, logging road ca. 1 km past Campo
4 in mesic forest, ca. 2500 m, 15 Apr 1976, D. Riskind & T. Patterson
1988 (HOLOTYPE: TEX!; Isotypes: MEXU!, TEX!).

E. pubescentt Kunth similis sed plantis rhizomata breves effer-
entibus et caulibus prope basim pilosis differt.

Fibrous rooted perennials producing slender, ligneous, rhizomelike basal
offsets 1-7 cm long. Stems basally ascending, 10-30 cm tall, monocephalous
or sometimes with a single branch, reddish at base, strigose above, pilose with
spreading-deflexed hairs on the basal third or the very base. Basal leaves
persistent, spatulate, blades obovate, entire or with 1(-2) pairs of shallow
teeth, 2-4 cm long, the blades 5-11 mm wide, stiffly pilose with long-ciliate
petioles, the cauline oblanceolate, not clasping, strigose. Heads 7-10 mm wide,
solitary on nearly naked peduncles 4-12 cm long; phyllaries in 2-3 series of
nearly equal length, the longest 4-5 mm long, sparsely hirsute, sometimes
sparsely and minutely granular glandular. Ray flowers 65-95 in 1-2 series, the
corollas 7-10 mm long, the ligules 0.4-0.8 mm wide, white above with a lilac
midstripe beneath. Disc corollas 1.8-2.2 mm long, slightly swollen above the
tube. Achenes sparsely strigose, 2 nerved, 0.6-0.8 mm long; pappus of 8-11
fragile bristles slightly shorter than the disc corollas, with an outer series of
squamellae ca. 0.1 mm long.

Coahuila (Sierra Madera del Carmen); pine to pine-fir woods, 2200-2500
m; flowering April-August.

Additional collections examined: MEXICO. Coahuila. Mpio. Ocampo,
Sierra Madera del Carmen: upper end of Dos Canyon at road fork to Campo
Uno, 23 Jun 1976, Fryzell 2724 (LL); logging road from Campo 4 at head of
mesic draw, 28 May 1975, Riskind & Patterson 1817b (LL); Campo El Tres,
5 Aug 1974, Wendt & Adamcewicz 492 (LL); Canon El Dos, ca. 3/4 mi SE of
Campo Dos, above moist mesic central part of canyon, 3 Apr 1974, Wendt, et
al. 129c (LL).

Erigeron vicinus is clearly a member of sect. Olygotrichium (Nesom 1989c)
and in its perennial duration, fibrous roots, strigose upper stems, and ray
corollas with a lilac midstripe, it is most similar to E. pubescens Kunth and the
E. modestus A. Gray-E. metrius S.F. Blake complex. Particularly, in its stems
that are reddish and pilose near the base, it is similar to E. modestus and E.
metrius, but I have seen no other plants among hundreds of collections of these,

Nesom: Two new species of Erigeron from México 257

with the distinctive rhizomatous habit of FE. vicinus. The new species appears
to be highly localized geographically, and the five known collections are all
very similar among themselves. The epithet is intended to reflect the locality
of the species, which is near the international boundary between México and
the United States.

ACKNOWLEDGMENTS

I thank Dr. B.L. Turner and Dr. A. McDonald for their review and com-

ments on the manuscript.

LITERATURE CITED

Nesom, G.L. 1989a. Taxonomy of Erigeron sect. Polyactis (Compositae:
Astereae). Phytologia 66:415-455.

. 1989b. Four new species of Erigeron (Compositae: Astereae) from

northern México. Phytologia 66:488-495.

. 1989c. Infrageneric taxonomy of New World Erigeron (Asteraceae:

Astereae). Phytologia 67:67-93.

Phytologia (October 1990) 69(4):258-264.

CROTALARIA LOTIFOLIA LINNAEUS AND CROTALARIA PURDIANA
SENN, A CLARIFICATION OF THE SPECIES (WITH NOTES ON
CROTALARIA AXILLARIS AITON)

D.R. Windler, J.J. Marron, & S.G. Skinner

Biology Department, Towson State University, Towson, Maryland 21204
WSeA:

ABSTRACT

,

A study of the closely similar Central and South American taxa
Crotalaria lotifolia L. and C. purdiana Senn, shows that they are in-
deed distinct species. A scatter diagram, a series of graphs and a new
key summarize their differences. The African species C. azillaris Aiton
appears to be related.

KEY WORDS: Taxonomy, Leguminosae, Crotalaria, West Indies,
Central America, South America, Africa

Crotalaria lotifolia L. is a shrub which can be readily differentiated from
most other native American species of the genus by its axillary inflorescences.
It was first reported (in polynomial form) by Sloane (1725) from Jamaica and
subsequently by Dillen (1732) in Hortus Elthamensis. When Linnaeus treated
the species in Species Plantarum, he cited both polynomials and chose Dillen’s
words “lott folio” as the basis for his trivial name. An orthographic error led
to the publication of the name in Species Plantarum as “latifolia,” the correct
spelling beirg lotifolia.

In 1920, Fawcett & Rendle selected the Sloane collection (Sloane Herbar-
ium, Vol. 6, #5, BM!) as the lectotype for Crotalaria lotifolia. Previously,
the species had been chosen as the lectotype species for the genus (Britton &
Brown 1913; see also Farr, et al. 1979).

In 1938, H.A. Senn described a new species from Colombia which is mor-
phologically related to Crotalaria lotifolia and named it C. purdiana Senn,
after the collector, William Purdie. In his revision of the North American
crotalarias, Senn (1939) separated the two species in the following way:

258

Windler, et al.: Crotalaria lotifolia and C. purdiana 259

Q. Inflorescence few-flowered (1-3 flowers), leaves glabrous above (West In-

BCR) 50s cine 5 5'n 5 erate IME. 2 thos Scioto cle ate Seine ysis C. lotifolia

Q. Inflorescence several-flowered (4-8 flowers), leaves hirtellous above (Colom-

Biarand Cuba) ce. ocsdacct ese dsvuencdetdboomaecapwicee ce C. purdiana

During preparation of a revision of the American species of Crotalaria, the
senior author questioned whether C. lotifolia and C. purdiana might better
be treated as varieties of a single species or as a single variable species. Both
are trifoliolate shrubs with axillary inflorescences and similarly shaped fruits.
Senn’s use of flower number per inflorescence as his main character was brought
into question by the difficulty of counting flower pedicel scars on the short
inflorescences. In addition, the leaf pubescence was found to be more variable
than Senn had indicated in his key. All specimens of both species are pubescent
beneath, usually with appressed trichomes, 0.2-0.3(0.5) mm long. The upper
leaf surfaces of all C. lotifolia specimens and of C. purdiana from Barbados,
Jamaica, and Martinique are glabrous. The upper leaf surfaces of C. purdiana
specimens from Cuba, Colombia, and Venezuela have very short trichomes
(0.1-0.2 mm) on the upper surface, either generally distributed or distributed
in patches between glabrous areas.

With these factors in mind, the authors further examined Senn’s delin-
eation of the two species. During a preliminary survey of the material, the
senior author observed that flowers of Crotalaria lotifolia had twisted carinal
beaks, that those of C. purdiana lacked the twist, and that in general the
leaves of C. lotifolia were smaller than those of C. purdiana. An analysis of
22 specimens of C. lotifolia and 30 specimens of C. purdiana for leaf measure-
ments, inflorescence length, and number of flowers per inflorescence resulted
in the data plotted in Figure 1. A more graphic comparison of the two species
is shown in the scatter diagram (Figure 2).

Based on our observations, we accept Crotalaria lotifolia and C. purdiana
as distinct and offer the following couplet to separate the species.

A. Terminal leaflets 35-82 mm long, petioles 20-90 mm long; inflorescence
2-12 mm long, bearing 4-8 flowers; carinal beak not
EE ot ee ee eee ne eae en RMT ES SEM Soot, SE Pn RO C. purdiana

A. Terminal leaflets 13-30 mm long, petioles 14-30 mm long; inflorescence
1-2 mm long, bearing 1-3 flowers; carinal beak twisted ..... C. lotifolia

Senn (1939) recognized a variety of Crotalaria lotifolia and named it after
the collector of the type, H.F. Eggers. The plants are small and in our view
are facultative dwarf responses to dry conditions. We do not view his variety
as distinct from the typical variety. Our concepts of the taxa are documented
by the following citations.

260 PHYTOLOGIA volume 69(4):258-264 October 1990

TERMINAL LEAFLET LENGTH
|} C. purdiana
Hoje C.lotifolia

i?) 10 30 50 70 90 [mm]

TERMINAL LEAFLET WIDTH

ebay -————| C. purdiana
(Seep ———4 C. lotifolia

te) 10 20 30 40 50 Imm)

PETIOLE LENGTH

| $1 C. purdiana
ae el C. lotifolia
10 30 50 70 90 {mm}

# PEDICELS/ INFLORESCENCE

} oer C.purdiana
——— C. lotifolia

INFLORESCENCE LENGTH

EERE — 4 C_purdiana
nh C. lotifolia

2 4 6 8 10

°o

12 Imm]

Figure 1. A comparison of diagnostic characters
which separate two species of Crotalaria. Mean,
range, and standard deviation are indicated for
each charocter.

261

Crotalaria lotifolia and C. purdiana

Windler, et al.:

‘DUDIPINA D[DIDJOID PUD Hj/Oj1}O] DUD|DJOID JO UOYOIDdes Oy Buyr}dep wWoGpIp seyDIS “7 euNB}4

[ww] HLONAT L31dV37

¥ WXv/SS31 YO $130I03d 2 V'e@
WXv/S1390IGAad Z NVHL SHOW +

G3LSIML VWNINVD @

G31SIML LON WNIHVD ¥

Aa

{ww} HLON3AT SJONSOSSYOTANI

262 Pony TOOL O Gel A volume 69(4):258-264 October 1990

SPECIMENS STUDIED
CROTALARIA LOTIFOLIA

GUATEMALA. Santo Tomas, Friedrichsthal 272 (W).

BAHAMA ARCHIPELAGO. Great Exuma, along Flamingo Drive, near
Palm Hill Road, S of Georgetown, Correll & Correll 47929 (BM, NY).

CUBA. Cayo Paloma, Camaguey, Shafer 2572 (BM, NY); Vic. Puebla
Romano, Camaguey, Shafer 2478 (BM, NY, PH).

PUERTO RICO. Coamo, Hess & Stevens 3975 (NY); Vic. Coamo Springs,
Britton, et al. 6364 (NY); Guanica, Sintensts 3650 (MO, NY, P); Guayanilla,
Britton & Britton 9342 (NY); Ponce to Penuelas, Britton & Cowell 1281(NY);
8 mi W of Ponce, Heller 6273 (G, MO, P, PH); Vieques Island, Clabaza to
Ensenada Honda, Shafer 2944 (NY); Ensenada Honda to Puerto Medio, Shafer
3016 (NY).

VIRGIN ISLANDS. Angada, Fishlock 001 (NY, PH). Saint Croix, Chris-
tiansted, J.N. Rose, et al. 3620 (NY). Saint Jan, Coral Bay, Raunkiaer s.n.
II-1906 (P); Lameshur, Britton & Shafer 512 (NY). Virgin Gorda, N Sirma,
Fishlock 23 (NY). Saint Thomas, Soldier Bay Dalen, Eggers s.n. 1880 (MO,
PH). Water Island, Eggers 531 (G, P).

WINDWARD ISLANDS. Guadeloupe, St. Barthelemy, Grande Saline, Le
Gallo (NY).

CROTALARIA PURDIANA

MEXICO. Yucatan: Hda. San Francisco, Enriquez 811 (MEXU)); No spe-
cific locality, Gaumer 24264 (F).

BAHAMA ARCHIPELAGO. No specific locality, Brace 411 (NY).

CUBA. Camaguey: Tiffen, R.L. Shafer 2891 (NY). Oriente: Antella,
Britton, et al. 12441 (NY); Puerto Padre, Montenegro 17116 (NY); Santiago
harbor, Britton 1882 (NY); Santiago, Britton & Britton 12915 (NY), Ekman
7761 (NY), Havard 80 (NY), 85(NY), C.L. Pollard, et al. 268 (MO, PH).

JAMAICA. Santa Catherine: Hellshire Hills, C.D. Adams 10771 (MO).
Clarendon: Harris Savanna, Proctor 34308 (BM, NY); Inverness, Harris 12723
(BM, MO). Manchester: Between Cut River and Canoe Valley, Proctor 35474
(NY). Trelawny: 1.5 mi W of Rio Bueno, Proctor 31562 (NY). Parish not
given: Great Goat Island, Harris 9323 (BM, NY), 12520(BM, MO, NY, PH).

HAITI. Near Jean Rabel, Leonard & Leonard 13024 (NY).

DOMINICAN REPUBLIC. Bellomar to Chedo, 10 mi NE of Cabo Rajo,
Liogier 16937(NY, P); Near Boca de Yuma, Liogier 12279 (NY, P).

WINDWARD ISLANDS. Martinique: Presquile de la Caravelle, commune
Trinite, Egler 39-236 (NY); also C. lotifolia (Herbarium Lamarck, P!); Saint

Windler, et al.: Crotalaria lotifolia and C. purdiana 263

Vincent: Tobago Cays, Howard 11030 (BM, NY). Beadimouns, Eggers 7354
(P):
COLOMBIA. Magdalena: Cerrejon, Haught 6718 (P).

VENEZUELA. Aragua: Costa de La Costa, Arzsteguieta 5142 (G, P).

In 1968, Polhill commented on the similarity between the African species
Crotalaria azillaris Aiton and C. lotifolia, and stated that the latter lacks a
twisted carinal beak. This statement is in conflict with our understanding of
C. lotifolia. Since we do not know the nature of the material on which Polhill
based his comment, we can only speculate that he was examining specimens
of C. purdiana annotated by Senn as C. lotifolia. We have examined nine
specimens of C. azillaris from the Belgian Congo at the Kew Herbarium.
Leaflet size and petiole length are similar to C. purdiana. Characters of the
inflorescence, including the twisted carina are more similar to C. lotifolia. The
fruits of C. azillaris measured ranged from 3.5-5.0 cm long with a stipe of
0.6-1.0 cm long. Crotalaria lotifolia and C. purdiana have legumes measuring
less than 2.5 cm and stipes measuring less than 0.3 cm. Seeds of C. azillaris
range from 4 to 5 mm in length. The largest of the C. lotifolia and C. purdiana
seeds is 3.8 mm long. In general appearance the three species are very similar,
but the nature of their relationship is difficult to determine.

Polhill (1968; 1982) reorganized the subgeneric classification of the genus
Crotalaria based on floral characteristics, including the twist of the carina. It
is interesting that two seemingly closely related species such as C. lotifolia and
C. purdiana would differ in this putatively conservative character.

ACKNOWLEDGMENTS

Appreciation is extended to the curators of the various herbaria from which
specimens were borrowed (F, GH, K, MEXU, MO, NY, P, PH). Graphs and
scatter diagrams were prepared by Ms. Vicki Dodson, Towson State Univer-
sity Graphic Arts Department. Dr. Robert DeFilipps, Botany Department,
Smithsonian Institution and Dr. Guy Nesom, Herbarium, University of Texas
are thanked for their reviews of this paper.

LITERATURE CITED

Britton, N.L. & A. Brown. 1913. Illustrated Flora of the Northeastern United
States ...edition 2. Volume 2. New York.

Dillen, J.J. 1732. Hortus Elthamensis. 2 Volumes. London.

264 PHY TROLOGIA volume 69(4):258-264 October 1990
Farr, Ellen R., et al. (eds.). 1979. Index Nominum Genericorum (Plantarum).
Reg. Veg. 100. Bohn, Scheltema & Holkema, Utrecht.
Fawcett, W. & A.B. Rendle. 1920. Flora of Jamaica. London.
Linnaeus, C. 1753. Species Plantarum. 2 Volumes. Stockholm.

Polhill, R.M. 1968. Miscellaneous notes on African species of Crotalaria II.
Kew Bulletin 22:196-348.

. 1982. Crotalaria in Africa and Madagascar. 389 pp. A.A. Balkema.
Rotterdam.

Senn, H.A. 1938. Journal of Botany 76:298. |

. 1939. North American species of Crotalarza. Rhodora 41:317-367.

Sloane, Sir H. 1725. A Voyage to the Islands Madera, Barbados, Nieves, S.

Christophers and Jamaica .... 2 Volumes. London.

Phytologia (October 1990) 69(4):265-270.

NOTES ON ARIZONA ASCLEPIAS (ASCLEPIADACEAE) WITH
A NEW COMBINATION

Eric Sundell

Department of Natural Sciences, University of Arkansas at Monticello,

Monticello, Arkansas 71655 U.S.A.

ABSTRACT

Relationships between several pairs of closely related Asclepias taxa
from the western United States are discussed. A new combination, A.
uncialis E. Greene var. ruthiae comb. nov., is proposed.

KEY WORDS: Asclepias, Asclepiadaceae, Arizona

Introduction

In the “Contributors’ Guide for the New Vascular Plants of Arizona,”
the editorial committee recommends a conservative view in the recognition of
species. “A describable morphological discontinuity between species is essen-
tial. Members of complexes that intergrade considerably in morphology are
considered in infraspecific rank, and those that have no consistently diagnostic
morphological characters are not treated formally.” This is sensible editorial
policy for a manual that will be broadly used; it is also sound taxonomy in the
orthodox tradition, the lineal descendent of Charles Darwin’s (1859) observa-
tion that “wide-ranging, much diffused, and common species vary most.” At
the inception of phylogenetic taxonomy, Darwin was able to clarify the role of
the taxonomist who tries to make sense of this variability: “We have seen that
there is no infallible criterion by which to distinguish species and well-marked
varieties; and when intermediate links have not been found between doubtful
forms, naturalists are compelled to come to a determination by the amount of
difference between them, judging by analogy whether or not the amount suf-
fices to raise one or both to the rank of species.” Recognizable and describable
discontinuities between groups do not automatically indicate species status or
even formal taxonomic recognition for those groups; the discontinuities must
also be significant.

265

266 Pea YTOLO GEA volume 69(4):265-270 October 1990

During preparation of a treatment of the Asclepiadaceae for the Arizona
flora project, I have puzzled over specimens representing several pairs of closely
related Asclepias taxa. In his monograph of the North American species of
Asclepias, Woodson (1954) recognized all of them at the rank of species: A.
engelmanniana Woods. and A. rusbyi (Vail) Woods.; A. involucrata Engelm. ez
Torr. and A. macrosperma Eastwood, and A. uncialis E. Greene and A. ruthiae
Maguire. Yet the determinations supplied by Arizona botanists for their own,
local collections of those taxa, in particular the first two pairs, suggest that
Woodson’s treatment is unsatisfactory. Annotations and re-annotations, some-
times by the same individual, pepper many of the specimens, jumping back
and forth between the taxa in question. Woodson rightly described the species
of Asclepias in the United States as “unusually well defined and properly ap-
preciated by the botanical public.” In order to more satisfactorily reflect the
magnitude of their affinities, that is, “judging by analogy,” it seems practical,
in light of available taxonomic information, to restrict recognition of those
taxa to infraspecific rank.

Asclepias uncialis E. Greene

Asclepias uncialis comprises a poorly understood complex of intergrading
morphological phases from widely scattered localities on the high deserts and
dry plains of Nevada, Utah, Colorado, Arizona, and New Mexico. Included
in this complex are the more robust, broader leaved plants hitherto sepa-
rated from A. uncialis as A. ruthiae (and including A. eastwoodiana Barneby).
Whether those plants deserve formal recognition at even the infraspecific level
is a question that future taxonomic and biosystematic studies must answer.
My own morphological examination indicates that maintaining distinct enti-
ties at the rank of species is not warranted: I can find no clear, consistent, and
significant discontinuity. For purposes of a floristic treatment of the Milkweed
Family for .he state of Arizona, where A. uncialis is not only rare, but rep-
resented by its extreme forms, it seems best to formally retain those forms at
infraspecific rank. Therefore, the following new combination is proposed:

Asclepias uncialis E. Greene var. ruthiae (Maguire in Maguire & Wood-
son) Sundell, comb. nov. BASIONYM: Asclepias ruthiae Maguire in
Maguire & Woodson, Ann. Missouri Bot. Gard. 28:245. 1941.

Asclepias eastwoodiana Barneby, Leafl. W. Bot. 4:210. 1945.
Asclepias sanjuanensis Heil, Porter & Welsh, Great Basin Naturalist
49:100. 1989.

Extreme forms of the two varieties can be distinguished according to the
following key:

Sundell: Notes on Arizona Asclepias 267

1. Plants relatively small, stems to 0.5-1 dm long; lower leaf blades lanceo-
late, grading to linear lanceolate or linear above, pubescent only on the
margins and veins; flowers relatively small, corolla lobes 3-4 mm long,
anther wings 1.1-1.4 mm long; crown hoods with marginal lobes wel!

developed, the horns tangentially flat .................... var. uncvalis

1. Plants relatively large, stems mostly 1-2 dm long; lower leaf blades mostly
ovate, broadly elliptic or circular, grading to narrowly ovate or broadly
lanceolate above, margins pubescent, upper and lower surfaces pubescent
to glabrous; flowers relatively large, corolla lobes 4-6 mm long, anther
wings 1.3-1.7 mm long; crown hoods with marginal lobes indistinct to
well developed, the horns subdigitate ..................4.. var. ruthiae

Asclepias uncialis var. uncialis occurs in the southern and eastern portion
of the species’ range, from western Arizona and southwestern New Mexico to
Colorado and western Oklahoma; var. ruthiae is found farther to the west,
where it appears to be more common, or at least more frequently collected,
from central Nevada to southeastern Utah, northern Arizona, and northwest-
ern New Mexico. Plants recently described as A. sanjuanensis (Heil, et al.
1989) from northwestern New Mexico are somewhat intermediate and, in
stature, leaf outline, and crown form, bridge an already indistinct morpho-
logical gap.

Collections of the typical variety of Asclepias unczalis are apparently un-
common, a situation that tends to promote species inflation based on minor
discontinuities. Of particular interest is an Alice Eastwood collection (8259,
GH [photocopy seen], US!) from Silver City, New Mexico, the type locality,
exhibiting a more robust habit and, as far as I am able to judge, indistinguish-
able from type material of A. sanjuanensis (Heil 4338, NMC!). The authors
of A. sanjuanensis recognize a distant affinity to A. unczalis, but tentatively
derive their new species, as well as A. eastwoodzana, from the geographically
intermediate A. ruthiae. Collection of additional flowering and fruiting mate-
rial from Arizona and western New Mexico will help to resolve the problem,
most likely by further blurring the distinctions between these various entities.

Asclepias involucrata Engelm. ez Torr.

Asclepias involucrata of the southwestern United States and northern Mex-
ico is so variable in pubescence, leaf outline, and crown form, that its distinc-
tiveness has been insufficiently appreciated. At least in Arizona, where it is
sufficiently common to be rather well represented in state herbaria, the dwarf,
spreading habit, more or less erect leaves with short-wooly margins, and ter-
minal, bracteate umbels clearly mark it as a cohesive genetic and taxonomic

268 PHY TO bO:GBa volume 69(4):265-270 October 1990

entity. In the northern part of the state, plants in some populations exhibit
one or more of the following character extremes: ovate leaves (especially the
lower ones), persistent wooly pubescence on leaf undersurfaces, oblong and
upright crown hoods, and larger seeds. These have been segregated as Ascle-
pias macrosperma or A. involucrata var. tomentosa Eastwood. However, these
character states do not consistently correlate, nor are there clear morphologi-
cal discontinuities discernible between contrasting conditions. Like McDougall
(1973), I am unable to treat such plants formally.

Asclepias engelmanniana Woods.

This tall and distinctive milkweed is a common element of prairies and
floodplains on the high plains of Nebraska, Colorado, and Kansas. To the
southwest, it ranges as far as Arizona, where it occurs sporadically along creeks
and in open woodlands in the northeastern half of the state. Plants with more
or less well developed crown horns have been recognized by Kearney & Peebles
(1960) at the infraspecific level as A. engelmanniana var. rusbyi (Vail) Kearney
(A. rusbyz Vail). Most Arizona plants possess at least a nub of a horn and, with
some misgiving, are assigned here. However, even within the same collection
(Lehto L23157 from Gila Co.), the horn can vary from well developed (ca.
1 mm long and scarcely exserted beyond the hood rim [ARIZ!]) to entirely
suppressed and represented by a ridge at the base of the back wall of the hood
(ASU!). Typical A. engelmanniana of the Great Plains, lacks both the horn
and the ridge.

Additional Problems

Two additional pairs of species, each with one representative in the Arizona
flora, pose similar, unresolved taxonomic problems.

Asclepias angustifolia Schweig. of México and southern Arizona has never
been circumscribed broadly enough to include the relatively localized A. terana
Heller of south Texas and northeast México. In the United States, the species
are easily separated morphologically by leaf outline, mostly linear to narrowly
lanceolate, to 12 cm long, in A. angustifolia, mostly oval to oblong elliptic, to 7
cm long, in A. terana. The difference is distinct, but certainly no greater than
that between, e.g., the subspecies of Sarcostemma cynanchoides Decne., also
of the American Southwest. Plants with somewhat intermediate leaf outlines
occur in northern México and, based on specimens at ARIZ, have no trouble
passing as A. angustifolia to Arizona botanists and as A. tezana to their Texas
colleagues! Ecologically, the species differ very little, occuring along creeks and
canyons in rocky and otherwise arid country.

Sundell: Notes on Arizona Asclepias 269

Asclepias nyctaginifolia A. Gray of California, Arizona, and New Mexico
is clearly a segregate of A. oenotherioides Cham. & Schlecht., which ranges
from Texas southward into much of Central America. The species are main-
tained based on differences in the length and basal taper of the crown hoods.
Invoking the Darwinian method of analogy to other members of the genus,
I would suggest that the distinction, if it were to survive the examination
of currently available specimens, is weak and the taxa good candidates for
taxonomic reappraisal.

Good Field Notes for Milkweed Collections

Good morphological information accompanying milkweed collections is rare.
In the arid Southwest, most Asclepias species are relatively undercollected.
For example, of the 26 species of Asclepias recognized in Arizona, only five
were sufficiently represented, among the specimens borrowed from the major
Arizona herbaria, to occupy more than a single shelf in a herbarium case.
Few botanists could be familiar with those milkweeds in the field. Thus the
recording of appropriate detailed morphological data is not at all a formality.

Useful information that is more often than not lacking on milkweed collec-
tions includes the following:

-root morphology

-habit; in particular, stem number, disposition, and length

-corolla form and disposition, particularly of the lobes

~crown form: standard botanical terms are not always available to describe
the three dimensional form and disposition of the hoods

-flower color: corolla, crown, and anther head should be noted.

ACKNOWLEDGMENTS

I wish to thank the curators of the following herbaria for making their
collections of Asclepias available to me: ARIZ, ASC, ASU, CM, COLO, CS,
MNA, NMC, NY, UNM, and US. I am grateful to Jim Locklear, Director of
the Dyck Arboretum of the Plains, Hesston, Kansas, for keeping me informed
on the whereabouts of A. uncialis specimens and to Steve Lynch, Louisiana
State University, Shreveport, for helpful conversations and communications
concerning several of the species discussed in this paper.

270 PHY TOLDOG TA volume 69(4):265-270 October 1990

LITERATURE CITED

Darwin, Charles. 1859. The Origin of Species by Means of Natural Selection.

Heil, K.D., J.M. Porter, & S.L. Welsh. 1989. A new species of Asclepias (As-
clepiadaceae) from northwestern New Mexico. Great Basin Naturalist

49:100-103.

Kearney, T.H., R.H. Peebles, & Collaborators. 1960. Arizona Flora, ed.
2, with Supplement by Howell, J.T., E. McClintock, & Collaborators.
Univ. of California Press, Berkeley.

McDougall, W.B. 1973. Seed Plants of Northern Arizona. The Museum of
Northern Arizona, Flagstaff.

Woodson, R.E., Jr. 1954. The North American species of Asclepias L. Ann.
Missouri Bot. Gard. 41:1-211.

Phytologia (October 1990) 69(4):271-274.

NEW GENERA OF BROMELIACEAE
Lyman B. Smith & W. John Kress

Department of Botany, Smithsonian Institution, Washington, D.C. 20560
USA:

ABSTRACT

New combinations are enumerated in two new genera of Bromeli-

aceae.
KEY WORDS: Bromeliaceae, Neotropics, nomenclature

The generic names Podaechmea stat. nov. and Platyaechmea stat. nov.
(Smith & Kress 1989) were invalidly published under Art. 33.2 because their
basionyms were omitted by error. The new combinations within these genera
were also invalidly published under Art. 43. All are herewith validated.

Podaechmea (Mez) Smith & Kress, stat. nov. BASIONYM: Aechmea sub-
gen. 1. Podaechmea Mez, DC. Monogr. Phan. 9:191. 1896. TYPE
SPECIES: Podaechmea lueddemanniana (K. Koch) Smith & Kress.

Podaechmea ferruginea (L.B. Smith) Smith & Kress, comb. nov.
BASIONYM: Aechmea ferruginea L.B. Smith, Contr. Gray Herb.
98:5, pl. 1, figs. 1-3. 1932.

Podaechmea galeottii (Baker) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea galeottii Baker, Handb. Bromel. 51. 1889.

Podaechmea lueddemanniana (K. Koch) Smith & Kress, comb. nov.
BASIONYM: Pironneava lueddemanniana K. Koch, Wochenschr.
Gartnerei Pflanzenk. 9:182. 1866.

Podaechmea macvaughii (L.B. Smith) Smith & Kress, comb. nov.
BASIONYM: Aechmea macvaughii L.B. Smith, Phytologia 10: pl.
1, figs.

Podaechmea mexicana (Baker) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea mezicana Baker, J. Bot. 17:165. 1879.

271

272 PPHYsTsOLLiGCeG'T A volume 69(4):271-274 October 1990

Platyaechmea (Baker) Smith & Kress, stat. nov. BASIONYM: Aechmea
section Platyaechmea Baker, J. Bot. 17:130. 1879. TYPE SPECIES:
Platyaechmea disticantha (Lemaire) Smith & Kress.

Platyaechmea anomala (L.B. Smith) Smith & Kress, comb. nov.
BASIONYM: Aechmea anomala L.B. Smith, Caldasia 3:237, fig.
1945.

Platyaechmea caesia (Baker) Smith & Kress, comb. nov. BASIONYM:
Aechmea caesia E. Morren ez Baker, Handb. Bromel. 43. 1889.

Platyaechmea chantinii (Carriere) Smith & Kress, comb. nov. BA-
SIONYM: Billbergia chantinii Carriere, Rev. Hort. 50:112, fig. 22.
1878; 52:272, figs. 54-56. 1880.

Platyaechmea contracta (Schultes filius) Smith & Kress, comb. nov.
BASIONYM: Billbergia contracta Martius ez Schultes filius in Roe-
mer & Schultes, Syst. Veg. 7(2):1263. 1830.

Platyaechmea dealbata (Baker) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea dealbata E. Morren ez Baker, Handb. Bromel.
58. 1889.

Platyaechmea dichlamidea (Baker) Smith & Kress, comb. nov. Pla-
tyaechmea dichlamidea (Baker) Smith & Kress var. dichlamidea.
BASIONYM: Aechmea dichlamidea Baker, J. Bot. 17:133. 1879.

Platyaechmea dichlamidea (Baker) Smith & Kress var. pariaensis
(Pittendrigh) Smith & Kress, comb. nov. BASIONYM: Aechmea
dichlamidea Baker var. pariaensis Pittendrigh zn L.B. Smith, Phy-
tologia 18:137, pl. 1, fig. 1. 1969.

Platyaechmea dichlamidea (Baker) Smith & Kress var. trinitensis
(L.B. Smith) Smith & Kress, comb. nov. BASIONYM: Aechmea
dichlamidea Baker var. trinitensis L.B. Smith, Proc. Amer. Acad.
Arts (Contr. Gray Herb. #102) 68:145, pl. 1, fig. 3. 1933.

Platyaechmea disticantha (Lemaire) Smith & Kress, comb. nov.
Platyaechmea disticantha (Lemaire) Smith & Kress ‘var. distican-
tha forma disticantha. BASIONYM: Aechmea disticantha Lemaire,
Jard. Fleur. 3:pl. 269. 1853.

Platyaechmea disticantha (Lemaire) Smith & Kress forma albiflora
(L.B. Smith) Smith & Kress, comb. nov. BASIONYM: Aechmea
disticantha Lemaire forma albiflora L.B. Smith, Arq. Bot. Estado
Sao Paulo 1:102. 1943.

Platyaechmea disticantha (Lemaire) Smith & Kress var. glaziovii

(Baker) Smith & Kress, comb. nov. BASIONYM: Aechmea glaziovi
Baker, J. Bot. 17:133. 1879.

Smith & Kress: New genera of Bromeliaceae 273

Platyaechmea disticantha (Lemaire) Smith & Kress var. schlum-
bergeri (Mez) Smith & Kress, comb. nov. BASIONYM: Aechmec
disticantha Lemaire var. schlumbergeri E. Morren ez Mez, Martius

Flora Brasiliensis 3(3):343. 1892.

Platyaechmea disticantha (Lemaire) Smith & Kress var. vernicosa
(E. Pereira) Smith & Kress, comb. nov. BASIONYM: Aechmea
disticantha Lemaire var. vernicosa E. Pereira, Bradea 2(947):308.
1979.

Platyaechmea fasciata (Lindley) Smith & Kress, comb. nov. Platy-
aechmea fasciata (Lindley) Smith & Kress var. fasciata. BASIO-
NYM: Billbergia fasciata Lindley, Bot. Reg. 13: pl. 1130. 1828.

Platyaechmea fasciata (Lindley) Smith & Kress var. flavivittata
(Reitz) Smith & Kress, comb. nov. BASIONYM: Aechmea fasciata
(Lindley) Baker var. flavi-vittata Reitz, Sellowia 33:55. 1981.

Platyaechmea fasciata (Lindley) Smith & Kress var. pruinosa (Re-
itz) Smith & Kress, comb. nov. BASIONYM: Aechmea fasciata
(Lindley) Baker var. pruinosa Reitz, Sellowia 33:55. 1981.

Platyaechmea fasciata (Lindley) Smith & Kress var. purpurea (Guil-
lon) Smith & Kress, comb. nov. BASIONYM: Billbergza rhodocyanea
Lemaire var. purpurea Guillon, Rev. Hort. 55:453. 1883.

Platyaechmea flavorosea (E. Pereira) Smith & Kress, comb. nov.
BASIONYM: Aechmea flavo-rosea E. Pereira, Bradea 2(49):321,
fig. (p. 323). 1979.

Platyaechmea moorei (Luther) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea moore: Luther, J. Brom. Soc. 30(4):176, fig.
1980.

Platyaechmea retusa (L.B. Smith) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea retusa L.B. Smith, Phytologia 10:484, pl. 4,
figs. 3-5. 1964.

Platyaechmea romeroi (L.B. Smith) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea romero: L.B. Smith, Phytologia 5:282, pl. 1,
figs. 4-6. 1955.

Platyaechmea serrata (L.) Smith & Kress, comb. nov. BASIONYM:
Tillandsia serrata L., Sp. Pl. 286. 1753.

Platyaechmea smithiorum (Mez) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea smithiorum Mez, DC. Monogr. Phan. 9:246.
1896.

Platyaechmea tessmannii (Harms) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea tessmanni Harms, Bot. Gart. Berlin 9:1153.
1927.

274

PHY TPOTOCGPA volume 69(4):271-274 October 1990

Platyaechmea tillandsioides (Schultes filius) Smith & Kress, comb.
nov. Platyaechmea tillandsioides (Schultes filius) Smith & Kress
var. tillandsiordes. BASIONYM: Billbergia tillandsioides Schultes
filius in Roemer & Schultes, Schultes, Syst. Veg. 7(2):1269. 1830.

Platyaechmea tillandsioides (Schultes filius) Smith & Kress var.
kienastii (Mez) Smith & Kress, comb. nov. BASIONYM: Aechmea
kienastzz E. Morren ez Mez, DC. Monogr. Phan. 9:243. 1896.

Platyaechmea wittmackiana (Regel) Smith & Kress, comb. nov.
BASIONYM: Quesnelia wittmackiana Regel, Gartenflora 37:497,
pl. 1281, fig. 2. 1888.

Platyaechmea zebrina (L.B. Smith) Smith & Kress, comb. nov. BA-
SIONYM: Aechmea zebrina L.B. Smith, Phytologia 4:358, pl. 2,
figs. 1, 2. 1953.

LITERATURE CITED

Smith, L.B. & W.J. Kress. 1989. New or restored genera of Bromeliaceae.

Phytologia 66(1):70-79.

Phytologia (October 1990) 69(4):275.

MINOR NEW COMBINATIONS IN TOXICODENDRON (ANACARDIACEAE)
James L. Reveal

Department of Botany, University of Maryland, College Park, Maryland
20742-5815 U.S.A.

ABSTRACT

Three new combinations are proposed within Tozicodendron radi-
cans (L.) Kuntze: var. pubens (Engelm. ez S. Watson) comb. nov., var.
barkleyi (Gillis) comb. nov., and var. negundo (E. Greene) comb. nov.

KEY WORDS: Tozicodendron, Anacardiaceae, poison ivy, North

America

The variation within the Tozicodendron radicans (L.) Kuntze (Anacar-
diaceae) complex is better treated at the varietal rank than the subspecific;
accordingly, the following new combinations are proposed:

Toxicodendron radicans (L.) Kuntze var. pubens (Engelm. ez S. Watson)
Reveal, comb. nov. BASIONYM: Rhus tozicodendron L. var. pubens
Engelm. ez S. Watson, Bibliogr. Inder N. Amer. Bot. 185. 1878.

Toxicodendron radicans (L.) Kuntze var. barkleyi (Gillis) Reveal, comb.
nov. BASIONYM: Tozicodendron radicans (L.) Kuntze subsp. barkley:
Gillis, Rhodora 73:224. 1971.

Toxicodendron radicans (L.) Kuntze var. negundo (E. Greene) Reveal,
comb. nov. BASIONYM: Tozicodendron negundo KE. Greene, Leafl. Bot.
Observ. Crit. 1:117. 1905.

ACKNOWLEDGMENTS

The study of early temperate North American types of vascular plants
is part of the Linnaean Plant Name Typification Project and is supported by
National Science Foundation Grant BSR-8812816. The paper was reviewed by
F.R. Barrie, C.E. Jarvis, and D.H. Nicolson. This is Scientific Article A-6058,
Contribution No. 8220, of the Maryland Agricultural Experiment Station.

275

Phytologia (October 1990) 69(4):276-277.

COMBINATIONS MADE IN THE CHECKLIST OF THE VASCULAR
PLANTS OF TEXAS

Kancheepuram N. Gandhi

The North Carolina Botanical Garden, Department of Biology, Coker Hall,
University of North Carolina, Chapel Hill, North Carolina 27599-3280 U.S.A.

ABSTRACT

Four new combinations, proposed by Gandhi in the Checklist of the
‘ascular Plants of Tezas (Hatch, et al. 1990), are validated here.

K2Y WORDS: Checklist, Texas, Asteraceae, Commelinaceae

In the Checklist of the Vascular Plants of Tezas (Hatch, et al. 1990), Gandhi
proposed four new combinations. Unfortunately, the requirements of Art. 33.2
of the International Code of Botanical Nomenclature (Greuter 1988) were not
met and the combinations remain invalid. Hence, those combinations are
proposed here for validation and the relevant page numbers of the Checklist
are indicated.

Asteraceae

p. 119. Haplopappus spinulosus (Pursh) DC. var. chihuahuanus (B.
Turner & Hartman) Gandhi, comb. nov. BASIONYM: Machaeranthera
pinnatifida (Hook.) Shinners var. chthuahuana B. Turner & Hartman,
Wrightia 5:311. 1976.

p. 120. Heterotheca subaxillaris (Lam.) Britt. & Rusby var. psam-
mophila (Wagenkn.) Gandhi, comb. nov. BASIONYM: Heterotheca
psammophila Wagenkn., Rhodora 62:76. 1960. ;

p. 120. Hymenopappus scabiosaeus L’Her. var. riograndensis (B. Turner)
Gandhi, comb. nov. BASIONYM: Hymenopappus artemisifolius DC.
var. riograndensis B. Turner, Rhodora 58:305. 1956.

276

Gandhi: New combinations in Checklist of Vascular Plants of Tezas 277

Commelinaceae

p. 38. Tradescantia leiandra Torr. var. glandulosa (Correll) Gandhi,
comb. nov. BASIONYM: Setcreasea letandra (Torr.) Pilg. var. glandu-
losa Correll, Madrono 19:187. 1968.

ACKNOWLEDGMENT

The author thanks Dr. John T. Kartesz for helpful comments.

REFERENCES

Greuter, W. (Chairman, Editorial Committee). 1988. /nternational Code of
Botanical Nomenclature. Regnum Veg. Vol. 118.

Hatch, S.L, K.N. Gandhi, & L.E. Brown. 1990. Checklist of the Vascular
Plants of Teras. MP-1655. The Texas Agricultural Experiment Station,
The Texas A&M University System, College Station, TX.

Phytologia (October 1990) 69(4):278-281.

A NEW SPECIES OF ERIGERON (ASTERACEAE: ASTEREAE) FROM
CHIHUAHUA, MEXICO

Scott D. Sundberg & Guy L. Nesom

Department of Botany, University of Washington, Seattle, Washington 98195
“USS.A:
&
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

A new species, Erigeron byei spec. nov., is described from south-
central Chihuahua, México. Based on similarities in habit, vestiture,
and capitular and floral morphology, it appears to be most closely re-
lated to E. ortegae S.F. Blake and E. oryphyllus E. Greene, and is placed
as the third species in Erigeron sect. Spinost. The new species has nine
pairs of chromosomes.

KEY WORDS: Erigeron, Asteraceae, Astereae, México

The following previously undescribed species is recognized from the moun-
tains of west-central Chihuahua, México.

Erigeron byei Sundberg & Nesom, spec. nov. (Figure 1) TYPE: MEXICO.
Chihuahua: Mpio. Batopilas, 2.5 mi S of Creel on main road to Batopi-
las, steep, terraced slopes, pine-oak woodlands, 19 May 1985, R. Scott
471 with T. Ayers, M. Lavin, & A. Whittemore (HOLOTYPE: TEX!;
Isotypes: ARIZ!, ASU!, COLO!, GH!, ILL!, MEXU!, NY!, OBI!, RM!,
UNM!, US!). ;

E. ortegae S.F. Blake similis vestimento fere glabro, foliis cauli-
nis bracteatis, gemmis erectis, phyllariis trinervibus, et ligulis cir-
cinatis sed differt statura multo parviore, rhizomatibus tenuibus,
foliis basalibus persistentibus, et capitulis parvioribus flosculos par-
viores efferentibus.

278

Sundberg & Nesom: New species of Erigeron from Chihuahua 279

2 7 7
A
AY ASS S
SPSS
= “ CJ

Figure 1. Erigeron byei. A. Habit. B. Capitulum and detail of phyllary. C.
Details of ray and disc flowers. D. Achene.

280 PHY. TO.LOGIA volume 69(4):278-281 October 1990

Perennial, caespitose, glabrous herbs from a system of thin, lignescent,
thizomelike caudex branches. Stems 7-20 cm tall, usually with 1-2 ascending
branches near midstem. Basal leaves persistent in rosettes, sessile, narrowly
elliptic-oblanceolate, 8-26 mm long, 1-2 mm wide, entire, 1 nerved or faintly 3
nerved, with a mucronulate apex, the cauline leaves few and sharply reduced
in size to linear bracts 1-4 mm long. Heads 5-8 mm wide, erect in bud, solitary
on nearly naked peduncles 5-10 cm long; phyllaries in 3-4 strongly graduated
series, elliptic-lanceolate with obtuse to acute apices, thin herbaceous with
scarious margins and 1-3, filiform, orange resinous nerves, sometimes purple
tinged, at least on the upper half, the inner series 4-5 mm long, 0.5-0.8 mm
wide. Ray flowers 9-18, the corollas white, drying white or purplish, 5-7 mm
long, the ligules 1.4-2.0 mm wide, coiling at the tips. Disc flowers 24-29, the
corollas 3.5-4.2 mm long, not inflated or indurated; collecting appendages of
style branches deltate, 0.1-0.2 mm long. Achenes sparsely strigose, cylindrical
to slightly flattened, 1.8-2.5 mm long, 0.4-0.5 mm wide, with (2-3)4(5) thin,
orange nerves; pappus of 15-27 barbellate bristles 2.9-3.8 mm long and a few
outer setae 0.1-0.5 mm long. Chromosome number, n=9 pairs (Fig. 1), as
determined from meiotic counts from buds collected with the type specimens,
Scott 471.

Additional collections examined: MEXICO. Chihuahua: Mpio. Batopilas:
N of Quirire, pine-oak forest, along arroyo in crevice of bedrock, 3 Jun 1973,
Bye 3951 (UCR); 4.5 mi N of Quirire, in arroyo (Rio Batopilas drainage), pine-
oak forest with Arbutus, Buddleia, from rock ledge in arroyo, 2150 m, 31 May
1984, Bye, et al. 12841 (COLO, TEX); 15.1 km SW of the Creel-Guachochi
road on the road to Batopilas; on wet, vertical rock in small canyon, SE-facing,
mostly shaded, 2390 m, with Quercus hypoleucoides, Q. coccolobifolia, Q. cf.
fulva, Pinus latifolia, and Alnus, 18 Oct 1986 [past fir], Spellenberg & Zucker
8932 (TEX).

Erigeron byei is named for Dr. Robert Bye, of the Jardin Botanico del
Instituto de Biologia, UNAM (México), who first collected the species. He
continues to be an active collector and student of the flora of the Sierra Madre
Occidental, particularly that of central Chihuahua. The new species appears
to be narrowly endemic to the area just south of Creel, Chihuahua.

The new species is most similar to Erigeron ortegae S.F. Blake (= Aster
spinosus Benth., see Sundberg 1986) and its putative sister species, E. ozy-
phyllus E. Greene (Nesom 1989), in its nearly glabrous vestiture, cauline leaves
strongly reduced in size to linear bracts, erect buds, thin herbaceous phyllaries
with three, prominent, orange resinous nerves, ray flowers with short, white,
coiling ligules, deltate collecting appendages of the disc style branches, and 4-5
nerved achenes. It differs from both species primarily in its habit and much
smaller stature. Both FE. ortegae and EF. ozyphyllus produce stems 0.5-2.5 m
tall from a system of thick rhizomes, and the plants are essentially leafless by
flowering. Erigeron byez is distinctly caespitose in habit, as the plants produce

Sundberg & Nesom: New species of Erigeron from Chihuahua 281

a system of slender rhizomes or rhizomelike caudex branches with clustered
rosettes of persistent basal leaves at the rhizome tips. Further, the heads of
E. byet are smaller and have fewer rays than either of the other species.

Despite the striking difference in habit, these three species appear to be
most closely related among themselves. Erigeron ortegae and E. ozyphyllus
have been segregated as Erigeron sect. Spinost (Alexander) Nesom & Sundberg
(Nesom 1989), and LZ. byez joins them as the third species of the section. A full
taxonomic treatment of E. ortegae, with a detailed discussion of relationships,
will be presented separately (Sundberg in prep.).

ACKNOWLEDGMENTS

This research was supported in part by National Science Foundation Grant
BSR-8313002. We thank Dr. B.L. Turner and Dr. D. Morgan for their review
and comments on the manuscript and Dr. L.A. Vorobik for drawing Fig. 1.

LITERATURE CITED

Nesom, G.L. 1989. Infrageneric taxonomy of New World Erigeron (Com-
positae: Astereae). Phytologia 67:67-93.

Sundberg, S.D. 1986. The systematics of Aster subg. Ozytripolium (Com-
positae) and historically allied species. Ph.D. dissertation, Univ. Texas,
Austin.

Phytologia (October 1990) 69(4):282-294.

TAXONOMY OF HETEROTHECA SECT. HETEROTHECA (ASTERACEAE:
ASTEREAE) IN MEXICO, WITH COMMENTS ON THE TAXA OF THE
UNITED STATES

Guy L. Nesom
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

Four species are recognized in Heterotheca sect. Heterotheca, all of
which occur in México: H. grandiflora Nutt., H. subazillaris (Lam.)
Britt. & Rusby, H. inuloides Cass. (with three varieties), and H. lep-
toglossa DC. Heterotheca inuloides and H. leptoglossa are endemic to
Mexico, and a new variety restricted to Oaxaca and Puebla is described:
H. inuloides var. viridis var. nov. Significant variation also occurs
within H. subazillaris, but in a study of the species over its entire range,
it has not been possible to recognize meaningful patterns so that even
varietal names could be consistently applied. A key to the species is
provided as well as discussions of variation and a distribution map for
the Mexican taxa.

KEY WORDS: Heterotheca, Asteraceae, Astereae, México

Heterotheca sect. Heterotheca was monographed by Wagenknecht (1960),
who recognized seven species, and it was the subject of further biosystematic
study and comments by Harms (1965a; 1968). The generalized distribution
map presented by Harms (1965a) for the members of the H. subazillaris (Lam.)
Britt. & Rusby complex was based primarily on the discussion and cited spec-
imens of Wagenknecht.

Within the genus, plants of sect. Heterotheca are characterized primarily
by their dimorphic achenes. The disc achenes are compressed, strigose to
sericeous, and with a well developed pappus, typically including 1-3 series of
long barbellate bristles and a much shorter, outer series of bristles or scales;
the ray achenes are 3 angled, smaller, glabrous or much less pubescent, and
completely epappose or rarely with a few short bristles or a minute corona.
Further, all species are taprooted and have stems, leaves, and phyllaries with
a vestiture of stipitate glandular hairs and spreading, nonglandular hairs with

282

Nesom: Taxonomy of Heterotheca sect. Heterotheca 283

osteolate bases, midcauline leaves sessile and clasping, heads in a corymboid
capitulescence, and yellow, coiling rays.

In the view of Semple, et al. (1980), Heterotheca also includes another
group of approximately 25 species, sect. Phyllotheca (Nutt.) V. Harms. They
are separated from sect. Heterotheca by their monomorphic achenes, although
the ray achenes in sect. Heterotheca occasionally produce few to numerous
pappus bristles, narrowing the morphological gap between the two sections.
Two species are rayless, each of them apparently having been derived indepen-
dently from taxa of sect. Phyllotheca, according to Semple, et al. (1988), who
nevertheless retained one of the rayless species as the sect. Ammodia (Nutt.)
V. Harms. In more conservative views (see Semple, et al. 1980, for a review),
Heterotheca has been considered to include two groups primarily endemic to
the southeastern United States, both now regarded with good evidence by
Semple as separate genera: Chrysopsis (no Mexican taxa) and Pityopsis (one
species ranging into México).

Even though the species of sect. Heterotheca have relatively recently been
studied, identifications and annotations since then have been highly incon-
sistent. In contrast to the seven species previously recognized, the present
report recognizes only four in the section, two of which encompass significant
infraspecific variation. At the most general level, they are divided into a north-
ern group of small headed taxa and a southern group of larger headed taxa.
All four of the species occur in México.

Although this study was primarily intended to address the patterns of vari-
ation in México, it is based on a survey of plants over the entire range of the
section, covering the United States as well, in an attempt to meaningfully
apply varietal nomenclature to the Mexican representatives. The ecological
summaries below, however, refer only to the Mexican plants. Additional syn-
onyms and information on typology are provided by Wagenknecht (1960).
Meiotic chromosome numbers of n=9 pairs are known for most of the taxa (as
summarized by Harms 1965b, Semple 1977, and as noted below).

KEY TO THE SPECIES

1. Innermost phyllaries (8-)9-12 mm long; plants of south central Mexico (3)

1’ Innermost phyllaries 5-9 mm long; plants of northern Mexico and the
Wantedtstates: SU jee Shot rh re. RR, Be Ne fet. eee (2)

2. Upper cauline leaves usually at least slightly subclasping to not at
all clasping; innermost phyllaries 5-8 mm long; ray achenes usually
completely glabrous, disc achenes usually with 1-3 resinous veins;
disc corollas glabrous or nearly so; outer pappus strongly developed;
herbage Crying green oo. c.9.<. hee te hoe aloe oe tee H. subazillaris

284 PHY BOLO GiA volume 69(4):282-294 October 1990

2’ Upper cauline leaves not clasping; innermost phyllaries 7-9 mm long;
the ray achenes minutely strigose at least on the angles, commonly
on the faces as well, the disc achenes usually without resinous veins;
the disc corollas prominently hairy on the throat; the outer pap-
pus weakly developed; herbage drying dark green to blackish green
H. grandiflora

3. Perennials or biennials; phyllaries purple at the very apices, the outer 2-3
series with vestiture mostly restricted to the distal half, sparsely glandu-
lar to nearly eglandular, with appressed to ascending apressed, relatively
thin based, nonglandular hairs along the midregion ..... H. leptoglossa

3’ Annuals or biennials; phyllaries either strongly purplish on the tips, mar-
gins, and midregions or not at all purplish, the outer 2-3 series with
prominent vestiture from base to apex, with thick based, stipitate glands
and spreading, thick based nonglandular hairs, or nonglandular hairs
commonly absent, in var. rose? .\. . 2-48 .<c 001 seiee eee eee H. inuloides

1. HETEROTHECA GRANDIFLORA Nutt.

Heterotheca grandiflora Nutt., Trans. Amer. Philos. Soc. 2 7:315. 1840.
TYPE: UNITED STATES. California: near Santa Barbara, [1836], Nut-
tall s.n. (PH, not seen).

Annuals to biennials or short lived perennials; stems and leaves sparsely
to densely hispid pilose, the herbage with a strong tendency to dry with a
pronounced dark cast. Cauline leaves narrowed to a sessile or subsessile, non-
clasping, attenuate base, at least on the upper stems. Phyllaries with only
glandular hairs, usually purple tipped, the innermost 7-9 mm long. Ray ach-
enes minutely strigose at least on the angles and commonly on the faces as well;
pappus with a weakly developed outer series of scales or bristles. Chromosome
number, n=9 pairs.

Baja California Norte, Sonora, California, southern Arizona; commonly in
disturbed sites, sandy soil, dunes, grasslands, chaparral; 3-50(-1200 in the US)
m; (May-)June-October(-January).

Heterotheca grandiflora appears to be strongly genetically isolated from H.
subazillaris, which is distributed from its native range in the eastern United
States westward to California, where it co-occurs with H. grandiflora. It is
possible that gene flow has occurred from H. grandiflora to H. subazillaris (see
comments following H. subazillaris), but there is no evidence, however, that
the reverse might be true. The strongly darkening pigments do not occur in
any other taxa of the section.

Nesom: Taxonomy of Heterotheca sect. Heterotheca 285

2. HETEROTHECA INULOIDES Cass.

Heterotheca inuloides Cass., Dict. Sci. Nat. 51:460. 1827. TYPE: MEXICO.
without other data (Probable holotype: Fl-Herb. Webbianum ez Herb.
Desfontaines, GH photo!). According to Cassini, this plant was said by
Desfontaines to have originated in México, sent by de Candolle from
Geneva to the Jardin du Roi (presumably in Paris). The sheet at FI was
annotated in Cassini’s hand as “Heterotheca inuloides, H. Cass.” There
is in G-DC (fiche!) a very similar specimen noted as from “J. de Paris
1828” that is possibly a duplicate of the type.

Annuals, biennials, or perennials; stems sparsely to moderately hispid-
pilose. Lower leaves petiolate or epetiolate, coarsely serrate, the upper be-
coming entire, sessile, usually at least subclasping. Phyllaries with a dense
vestiture of stipitate glands, on the outer phyllaries these distributed from the
base to apex, also densely hirsute-pilose in two varieties with long, nonglandu-
lar hairs, but commonly lacking nonglandular hairs in var. rosei, the innermost
phyllaries (8-)9-12 mm long. Ray achenes usually minutely but prominently
strigose on the faces and angles, sometimes glabrous; pappus with a prominent
outer series of scales or bristles, these sometimes weakly developed.

Key to the varieties

1. Outer phyllaries densely hispid-pilose from base to tip with spreading to
spreading ascending, nonglandular hairs ..............0.0eeeeeeeees (2)

1’ Outer phyllaries without nonglandular hairs or the nonglandular hairs
present mostly on the ‘distal third ...:....02.-0--e4 + senaee > see var. Toset

2. Phyllaries distinctly purplish at least at the apices and commonly
along the distal margins and midregion as well, with thick based,
nonglandular trichomes: < 2/5 2.5...) sfe.so0e oe tectemaae var. inuloides

2’ Phyllaries completely without purple coloration, with relatively thin
based nonglandular trichomes ..............---+.+--+- var. viridis

2a. HETEROTHECA INULOIDES Cass. var. INULOIDES

Chromosome number, n=9 pairs.

Nuevo Leon, Durango, Zacatecas, San Luis Potosi, Hidalgo, Veracruz, Col-
ima, Michoacan, México, Tlaxcala, Puebla; roadsides, fields, openings in pine-
oak or pine woodlands; 2000-3000 m; (May-)August-November(-December).

286 PUY T0L,0 GLA volume 69(4):282-294 October 1990

Plants of several scattered collections from far outside its main range must
be identified as var. inuloides: Nuevo Leon: Dulces Nombres, 12 Aug 1948,
Meyer & Rogers 2951 (GH). San Luis Potosi: region of S.L.P., 1878, Parry
& Palmer 372 (GH). And as noted by McVaugh (1984), other populations
with var. inulozdes-like phyllaries occur well within the range of var. rosez, in
the areas of northeastern Nayarit, southeastern Durango, and adjacent Za-
catecas. To show their location, these are mapped (Map 1) as var. inulozdes,
although field study may prove these to be variants within more rosez-like pop-
ulations. Further, plants that might be interpreted as intermediate between
var. inuloides and var. rosez occur in the vicinity of Uruapan and Tancitaro,
Michoacan, but these are mapped as var. roset.

2b. HETEROTHECA INULOIDES Cass. var: ROSEI Wagenk.

Heterotheca tnuloides Cass. var. roset Wagenknecht, Rhodora 62:69. 1960.
TYPE: MEXICO. Jalisco: sandy soil, grassland and roadsides, 30 mi
E of Guadalajara, 25 Jul 1956, B.L. Wagenknecht 2846 (HOLOTYPE:
KANU!).

In addition to the paucity or complete lack of nonglandular hairs, the
phyllaries of var. rosei usually are completely lacking in purple coloration,
they are more densely stipitate glandular than in var. znuloides, and the glands
usually have thicker stipes. Judging from their more slender taproots, these
plants also tend to be more short lived than those of the two eastern varieties
of Heterotheca inuloides. Chromosome number, n=9 pairs.

Durango, Zacatecas, Nayarit, Jalisco, Michoacan, Colima; matorral, shrub-
lands, oak-juniper to pine-oak and pine woodlands, roadsides and other dis-
turbed sites, pond edges; 1200-2400 m; (March-)June-November.

2c. HETEROTHECA INULOIDES Cass. var. VIRIDIS Nesom, var. nov.

Heterotheca inuloides Cass. var. viridis Nesom, var. nov. TYPE: MEX-
ICO. Oaxaca: Distrito del Centro, Monte Alban, 1850 m, 14 Oct 1932,
C. Conzatti 4794 (HOLOTYPE: LL!; Isotype: MEXU).

A H. inuloides Cass. var. inuloides differt phyllariis absque
pigmento purpurato ac trichomatibus tenuioribus ad bases lon-
gioribusque.

These plants are similar to those of var. inuloides in their phyllaries with
a densely hirsute-pilose vestiture but consistently different in their complete
lack of purple coloration and the longer nonglandular trichomes with thinner
bases. Intermediates between the two taxa occur in northern Puebla and along
the adjacent border with Veracruz. Chromosome number, n=9 pairs.

Taxonomy of Heterotheca sect. Heterotheca 287

Nesom:

SURIIIxeqns “H g
H ®@

SIPUIA ‘2A @

wo ipuess

12sQ1 “18A ©
soplojnur “1eA =
soprojnur “H

wsso[301d3] “-H W

Map 1. Geographic distribution of the taxa of Heterotheca sect. Heterotheca

in México.

288 PHY TOLOGIA volume 69(4):282-294 October 1990

Veracruz, Puebla, Oaxaca; openings in oak-juniper and pine-oak wood-
lands, roadsides and other disturbed sites; 1550-2300 m; June-November
(-February).

Representative collections examined: MEXICO. Oaxaca: Valley of Oax-
aca, 19 Apr 1896, Conzatt: 125 (GH); Km 58 S of Oaxaca on road to Puerto
Escondido, 14 Aug 1975, Davidse 9623 (LL); between Oaxaca and Puerto Es-
condido, 8 km S of Sola de Vega, 25 Jun 1986, Diggs, et al. 4005 (NY); 20
mi S of Nochixtlan along rte 190, 1 Feb 1960, King 2515 (TEX) (voucher for
chromosome count of n=9, as annotated by Ellison, apparently previously un-
reported); ca. 40 mi SE of Oaxaca, 15 Jun 1960, King 2897 (TEX) (voucher
for chromosome count of n=9 pairs [Turner, et al. 1961]); 15 km SE of Miahu-
atlan, 6 Jul 1969, Marcks & Marcks 1048 (LL); 19 mi W of Oaxaca along rte
190, 15 Aug 1961, Powell & Edmondson 703 (TEX); 11 km N of Ixtlan, 17 Sep
1965, Roe & Roe 1989 (LL); 33 km NE of Cuajimoloyas, 14 May 1983, Torres
2891 (TEX). Puebla: Supercarretera México-Puebla, 24 Apr 1966, Boege 66
(GH); Orizaba, 1829, Botter: 799 (GH); 1 mi E of Hwy 125, N of San Salvador
on road to Tlachichuca, 7 Jun 1983, Dorr 2673 (TEX); 5 mi E of Tlahuapan
along rte 190, 24 Jan 1960, King 2236 (TEX); W of San Martin Texmelucan, 1
Aug 1953, Manning & Manning 53689 (GH); 3 km E of center of San Nicolas
Buenos Aires, 7 Sep 1986, Nee & Soule 33040 (TEX). Veracruz: S of Totalco,
27 Aug 1968, Ramos 269 (GH).

3. HETEROTHECA LEPTOGLOSSA DC.

Heterotheca leptoglossa DC., Prodr. 5:317. 1836. TYPE: MEXICO. [Edo.}
Guanajuato: Leon to Guanajuato, 1827, Mendez s.n. (HOLOTYPE: G-
DC fiche!; Isotype: GH).

Perennials. Phyllaries linear-lanceolate, 0.5-0.7(-1.0) mm wide, the inner-
most 9-11 mm long, without scarious margins, purple at the very apices, nearly
eglandular to sparsely sessile glandular or the glands short stipitate near the
phyllary apex, sparsely strigose in the midregion with appressed to appressed
ascending nonglandular hairs. Ray achenes commonly minutely strigose on the
angles and faces, sometimes hairy only on the angles or still less commonly
completely glabrous; outer pappus of strongly developed, lanceolate scales.
Chromosome number, n=9 pairs (reported as Heterotheca inuloides by Turner
& Johnston 1961).

Zacatecas, Aguascalientes, San Luis Potosi, Jalisco, Guanajuato, Querétaro;
roadsides and other disturbed sites, brushlands; 1750-2350 m; July-December
(-January).

Heterotheca leptoglossa was recognized by Wagenknecht (1960) as a dis-
tinct species, but he cited only two collections (other than the type) from
the geographical area of the species as recognized here. He added, however,

Nesom: Taxonomy of Heterotheca sect. Heterotheca 289

citations of other collections from Chihuahua, Sinaloa, and Sonora, all recog-
nized here as H. subazillaris (Map 1, region “a”). McVaugh (1984) included
H. leptoglossa within his concept of H. inuloides var. roset.

This taxon might reasonably be considered as a fourth variety of Het-
erotheca inuloides, but it appears to be genetically isolated from both var. in-
uloides and var. roset. The former is mostly contiguous with it in geographic
range and the latter is partially sympatric (Map 1). Heterotheca leptoglossa is
more similar to var. rosei in its very sparsely hairy phyllaries, but the former
differs in its more strongly woody taproots (and presumably longer duration)
and phyllaries that are purple tipped and less densely glandular with relatively
thin stiped glands. Further, the nonglandular hairs are usually appressed in
orientation, in contrast to the spreading-erect ones of var. roset.

4. HETEROTHECA SUBAXILLARIS (Lam.) Britt. & Rusby

Heterotheca subazillaris (Lam.) Britt. & Rusby, Trans. New York Acad. 7:10.
1887. BASIONYM: Inula subazillaris Lam., Encycl. Meth. Bot. 3:259.
1789. TYPE: UNITED STATES. “Carolina,” D. Walter s.n. (HOLO-
TYPE: P, not seen).

Heterotheca chrysopsidis DC., Prodr. 5:317. 1836. TYPE: MEXICO.
[Coahuila]: Saltillo, Jan 1828, Berlandier 1830 [109] (HOLOTYPE:
G-DC fiche!; Isotypes: F!, GH, NY, PH).

Heterotheca latifolia Buckl., Proc. Acad. Nat. Sci. Philadelphia 13:459.
1862. TYPE: UNITED STATES. Texas: Llano Co., S.B. Buckley
s.n. (PH, not seen). Heterotheca subazillaris (Lam.) Britt. & Rusby
var. latifolia (Buckl.) Gandhi & Thomas, Sida, Bot. Misc. 4:110.
1989.

Heterotheca latifolia Buckl. var. arkansana Wagenk., Rhodora 62:105.
1960. TYPE: UNITED STATES. Arkansas: Logan Co., Magazine
Mountain, 10 Sep 1930, D.M. Moore 30142 (HOLOTYPE: TEX!).

Heterotheca latifolia Buck]. var. macgregoris Wagenk., Rhodora 62:103.
1960. TYPE: UNITED STATES. Kansas: Morton Co., 9 mi N of
Elkhart, bluffs along Cimmaron River, dry sandy prairie, 27 Aug
1951, R.L. McGregor 5163 (HOLOTYPE: KANU!).

Heterotheca psammophila Wagenk., Rhodora 62:76. 1960. TYPE:
UNITED STATES. Arizona: Yavapai Co., 1 mi S of Sedona, 23 Aug
1957, B.L. Wagenknecht 4824 (HOLOTYPE: KANU not seen).

Heterotheca subazillaris (Lam.) Britt. & Rusby var. petiolaris Benke,
Rhodora 30:201. 1928. TYPE: UNITED STATES. Texas: [Galve-
ston Co.], Galveston, 12 Mar 1928, H.C. Benke 4585 (HOLOTYPE:
F!).

290 PHYTOLOGTA volume 69(4):282-294 October 1990

Heterotheca subazillaris (Lam.) Britt. & Rusby var. procumbens Wa-
genk., Rhodora 62:75. 1960. TYPE. UNITED STATES. Alabama:
Mobile Co., flattish dunes, ca. 1 mi SW of Dauphin Island Post Of-
fice, R.M. Harper 3801 (HOLOTYPE: GH; Isotypes: F!, NY, PH,
US).

Annuals to short lived perennials. Upper cauline leaves ovate lanceolate
to oblong, with a sessile or subsessile, commonly cordate, subclasping base,
but sometimes not at all clasping. Phyllaries lanceolate, the innermost 5-
8(-9) mm long, purple at the apices or without purple pigments, sparsely to
densely glandular, usually sparsely to moderately invested with nonglandu-
lar trichomes but sometimes only glandular. Ray achenes usually completely
glabrous, rarely slightly and minutely strigose on the angles; outer pappus of
strongly developed, lanceolate scales. Chromosome number, n=9 pairs.

Sonora, Chihuahua, Sinaloa, Coahuila, Nuevo Leon, Tamaulipas, San Luis
Potosi, Veracruz, widely distributed from coast to coast in the United States,
primarily in the southern half of the country; in a variety of habitats, often
disturbed, from beach dunes to Larrea desert to pine woodlands; (0-)1200-2500
m; July-December(-February).

Within what is considered here as a single, variable species, Wagenknecht
(1960) treated Heterotheca subazillaris (with 2 varieties), H. latifolia Buckl.
(with 3 varieties), H. psammophila Wagenk., and H. chrysopsidis DC. Be-
fore that, Shinners (1951) had accepted H. latifolia as a good taxon with the
caveat that it was probably only varietally distinct from H. subazillaris, “the
southeastern representative of this species.” The status of the Mexican H.
chrysopsidis had never been evaluated, and H. psammophila had not previ-
ously been recognized. Characters used to delimit these taxa have included
duration (as surmised from the taproot thickness), plant height, stem width,
length of trichomes on the stems and leaves, petiole length of lower cauline
leaves, size, shape, and insertion of upper cauline leaves, toothing of leaf mar-
gins, shape of the capitulescence, head size (including phyllary, disc corolla,
and pappus length), degree of investment of phyllaries with glandular and
nonglandular trichomes, and the pattern of pigmentation in the phyllaries.

Attempts to characterize taxa at any level within the Heterotheca subazil-
laris complex must be based on the supposition that geographic patterns can
be identified, but among what seem to be shifting and variable constellations
of morphological features, I have not been able to discover patterns that would
allow a consistent application of names at even varietal rank. Harms (1968,
p. 9) ventured that perhaps the “entire complex should still be accepted as a
single, polymorphic, polytypic species,” and Cronquist (1980) apparently took
the same conservative view as presented in the present study, although he did
not list synonyms of what he considered to be the widespread H. subazillaris.
Geographic trends can be observed in this species, and plants can often be

Nesom: Taxonomy of Heterotheca sect. Heterotheca 291

identified as having originated in certain geographic areas, but in my opin-
ion, plants occurring in broad areas of intergradation are significantly more
numerous than those that have been given names and that are said to have
a distinctive combination of traits. In this, my view seems to be very near
that of Semple (1990, p. 226) with regard to plants of certain species of sect.
Phyllotheca, who regards them as “morphotypes,” “ ...sometimes semidistinct
regional ’races’ that grade into each other to such an extent that continued
recognition cannot be justified ....”. Whoever may choose to recognize segre-
gate taxa within the H. subazillaris complex will be challenged to present more
than simple diagnoses and taxonomic combinations to justify their decisions.

What has been considered typical Heterotheca subazillaris (var. subazillaris
or var. procumbens) occurs in sandy habitats along the coast of the eastern
United States to central México. These plants are said to be distinguished
from H. latifolia by their procumbent or at least less than erect habit, merely
scabrous stems and leaves, thicker leaves, and more densely hirsute-villous
hairs on the phyllaries. Nearly identical plants, however, are found in sandy
habitats far inland within the range of H. latifolia, and plants with none of the
putatively diagnostic features of typical H. subazillaris are commonly collected
in coastal and near coastal sites. Further, Burk (1961), in a study apparently
little appreciated by some systematists, demonstrated that plants of H. sub-
azillaris are phenotypically highly plastic in features of habit, vestiture, and
leaf morphology. He transplanted sets of North Carolina plants from a barrier
island (var. procumbens) and from an inland habitat (H. latzfolia) to a common
garden on the piedmont, where they “developed into virtually identical sets of
plants.” Using cloned individuals, he also observed that vestiture was signifi-
cantly modified depending on the season and whether the plants were grown in
a greenhouse or outdoors. Burk concluded that it was preferable to “retain the
former concept of H. subazillaris as a single, highly variable species.” Harms
(1965a) also documented the ease of environmental modification among plants
of H. subazillaris.

Harms (1965a, 1968) noted that he was able to distinguish Heterotheca
latifolia var. macgregoris Wagenk. but that he could not separate it from var.
arkansana Wagenk. Wagenknecht (1960) separated both of the latter two va-
rieties from typical H. latifolia on the basis of their putatively more pilose
or velutinous upper leaf surfaces, but I cannot confirm the existence of any
morphological features or geographic localization consistent enough to war-
rant the formal recognition of either taxon. Hatch, et al. (1990) have listed
(without comment) both var. arkansana and var. macgregoris as synonyms of
var. latifolia.

Heterotheca subazillaris var. petiolaris Benke has been considered a syn-
onym of typical H. subazillaris (Burk 1961) and of H. latifolia (Wagenknecht
1960). But plants similar to the type of var. petiolaris with particularly long
petioles and narrowly oblong-lanceolate blades (basal and lower cauline leaves)

292 PHYPODOCG PA volume 69(4):282-294 October 1990

are common from Galveston County, Texas, into Tamaulipas of northeastern
México and appear to represent a morphotype outside of both H. subazillaris
and H. latifolia, as conceived of by Wagenknecht. In any case, “var. petio-
laris” intergrades with other morphotypes of the same region and cannot be
consistently distinguished from them.

As noted by Wagenknecht, plants of Heterotheca chrysopsidis (from south-
ern Coahuila and adjacent Nuevo Leon) are slightly longer lived, shorter in
stature, have petiolate, prominently thickened (much like coastal variants)
leaves with broadly ovate blades, and the pappus tends to be deep reddish
brown. These features are most pronounced in plants from the gypseous areas
of Nuevo Leon. Nevertheless, I have found it impossible to regard these traits
as anything but tendencies, since many plants from the same area are iden-
tical with those that are common in central Texas, and the distinctive traits
disappear gradually northward through Coahuila.

Heterotheca psammophila was recognized by Wagenknecht from Arizona,
southwestern New Mexico, trans-Pecos Texas, and northern México as a taxon
with particularly large leaves, large heads, and highly glandular phyllaries.
The plants do seem to have consistently larger leaves, but plants with similar
leaves grade eastward across a wide area and sometimes appear in populations
of central Texas, Oklahoma, and Kansas. Further, in Arizona and western New
Mexico, while some individuals produce heads that are among the largest in
the species (innermost phyllaries 7-9 mm long), other large leaved plants in
the same area commonly produce small heads (innermost phyllaries 5-7 mm
long). There is no consistent distinction in the degree of glandularity, and in
the United States, no consistency in the presence or absence of nonglandular
hairs.

In contrast, there does appear to be a degree of isolation between two
weakly differentiated “morphotypes” of Heterotheca subazillaris in the moun-
tains of northwestern México. Indeed, this is the only “real” boundary I have
been able to discover in the H. subazillaris complex. Plants of region “a” (Map
1) have phyllaries with purple tips, without non-glandular hairs, and the in-
nermost are 6-8(-9) mm long. To the east in central Chihuahua (region “b”),
the phyllaries usually lack purple tips, almost always produce nonglandular
hairs, and the innermost are 5-6 (-7) mm long. The plants of region “a” might
justifiably be regarded as a separate variety, presumably corresponding to H.
psammophila, were it not for the apparent complete intergradation in Arizona
of these “psammophila-like” plants with more typical H. subazillaris.

The occurrence of dense glandularity and relatively large heads in the west-
ern part of the range of Heterotheca subazillaris prompts the speculation that
these features may have appeared in response to the input of genes from H.
grandiflora.

Nesom: Taxonomy of Heterotheca sect. Heterotheca 293
ACKNOWLEDGMENTS

I thank Dr. B.L. Turner and Dr. P.O. Karis for their review and comments
on the manuscript and the staffs of F, GH, KANU for loans of specimens.

LITERATURE CITED

Burk, C.J. 1961. Environmental variation in Heterotheca subazillaris. Rho-

dora 63:243-246.

Cronquist, A. 1980. Asteraceae. Vascular Flora of the Southeastern United
States, vol. 1. Univ. North Carolina Press, Chapel Hill.

Harms, V.L. 1965a. Biosystematic studies in the Heterotheca subazillaris
complex (Compositae: Astereae). Trans. Kansas Acad. Sci. 68:244-257.

. 1965b. Cytological evidence supporting the merger of Heterotheca
and Chrysopsis. Brittonia 17:11-16.

. 1968. Nomenclatural changes and taxonomic notes on Heterotheca,
including Chrysopsis, in Texas and adjacent states. Wrightia 4:8-20.

Hatch, $.L., K.N. Gandhi, & L.E. Brown. 1990. Checklist of the Vascular
Plants of Teras. Texas A & M Agric. Expmt. St., College Station.

McVaugh, R. 1984. Heterotheca. Flora Novo-Galiciana 12:505-508. Univer-
sity of Michigan, Ann Arbor.

Semple, J.C. 1977. Cytotaxonomy of Chrysopsis and Heterotheca (Com-
positae: Astereae): a new interpretation of phylogeny. Canad. J. Bot.

55:2503-2513.

. 1990. Neotypification of Amellus villosus, the identity of a Brad-
bury collection, and typification of some other goldenasters (Compositae:
Astereae). Brittonia 42:221-228.

—______., V. Blok, & P. Heiman. 1980. Morphological, anatomical, habit,
and habitat differences among the goldenaster genera Chrysopsis, Het-
erotheca and Pityopsis (Compositae - Astereae). Canad. J. Bot. 58:164-
ae

Semple, J.C., C. Leeder, C. Leuty, & L. Gray. 1988. Heterotheca sect. Am-
modia (Compositae: Astereae): A multivariate study of H. oregonu and
specimens of Brewer’s (Golden)aster. Syst. Bot. 13:547-558.

294 PHYTOLOGIA volume 69(4):282-294 October 1990

Shinners, L.H. 1951. The North Texas species of Heterotheca, including
Chrysopsis (Compositae). Field & Lab. 19:66-71.

Turner, B.L., W.L. Ellison, & R.M. King. 1961. Chromosome numbers in the
Compositae. IV. North American species with phyletic interpretations.
Amer. J. Bot. 48:216-223.

Turner, B.L. & M. C. Johnston. 1961. Chromosome numbers in the Composi-
tae III. Certain Mexican species. Brittonia 13:64-69.

Wagenknecht, B.L. 1960. Revision of Heterotheca section Heterotheca (Com-
positae). Rhodora 62:61-76.

. 1960. Revision of Heterotheca section Heterotheca (Compositae).
Rhodora 62:97-107.

Phytologia (October 1990) 69(4):295-296.

TWO NEW SPECIES OF VERBESINA (ASTERACEAE, HELIANTHEAE)
FROM THE STATES OF JALISCO AND GUERRERO, MEXICO

B.L. Turner
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

Verbesina platanara spec. nov. and Verbesina teotepecana
spec. nov., new species from the states of Jalisco and Guerrero, respec-
tively, are described. The former is seemingly related to V. culminicola
McVaugh, but the latter apparently has no close relatives.

KEY WORDS: Asteraceae, Verbesina, México

Routine identification of Mexican Asteraceae has revealed the following
novelties in the genus Verbeszna.

Verbesina platanara B. Turner, spec. nov. TYPE: MEXICO. Jalisco:
Mpio. de San Martin Bolanos, Las Treinta Vueltas, 10 km al NW de
El Platanar, 2200 m, J. Rzedowski 26229 (HOLOTYPE: MICH!).

Verbesinae culminicolae McVaugh similis sed differt caulibus
sine alis, capitulis majoribus ligulis longioribus, et acheniis sine
pappo.

Shrubs to 1.5 m high. Stems strigulose, wingless, glabrate with age. Leaves
alternate, simple, crowded along the upper nodes, lanceolate to oblanceolate,
10-14 cm long, 1.5-3.0 cm wide, about equally tapering at both ends, sparsely
hispidulous to nearly glabrate, the margins remotely serratulate to nearly en-
tire. Head about 25 mm across the extended rays, single and axillary, the
peduncle ca. 6 cm long. Involucres ca. 4 mm high, the bracts 2-3 seriate,
grading into the obovate receptacular bracts. Ray florets ca. 11, pistillate, the
ligules yellow, ca. 1 cm long, ca. 4 mm wide. Disk florets numerous (50+),
the corollas yellow, ca. 3.5 mm long, the tube densely pubescent, ca. 1 mm
long. Achenes (immature) ca. 3 mm long, 1.5 mm wide, sparsely pubescent,
epappose.

The species is apparently most closely related to Verbesina culminicola
McVaugh on characters of the head and corolla, but differs in having wingless
stems, larger heads borne singly in the leaf axils, and epappose achenes.

295

296 PHYTOLOGIA volume 69(4):295-296 October 1990

Verbesina teotepecana B. Turner, spec. nov. TYPE: MEXICO. Guerrero:
Mpio. Tlacotepec, Cerro Teotepec, pine-fir forests, 3300 m, 11 Apr 1963,
J. Rzedowski 16500 (HOLOTYPE: MICH!).

Frutices usque ad 2.5 alti. Folia opposita sessilia lanceolata ac
auriculata. Caules teretes dense glandulosi-pubescentes sine alis.
Capitula radiata ac hemisphaerica, discus ca. 2 mm lati; involucra
10-12 mm alta, bracteae 2-4 seriatae lineari-lanceolatae apicibus
acutis. Flosculi disci numerosi (100+), corollis luteis ca. 6 mm
longis. Achenia ca. 4 mm longa, pappus aristis 2 cito deciduis 3-4
longis contans.

Shrub to 1.5 m high. Stems terete, densely glandular pubescent. Leaves 12-
15 cm long, 3-4 cm wide, sessile, auriculate at the base, the apices gradually
and narrowly acute, densely glandular pubescent on both surfaces. Heads
hemispheric, radiate, ca. 18, borne terminal in a rounded capitulescence, not
much exceeding the leaves, if at all. Involucres 10-12 mm high, ca. 2 cm
across, the bracts black, linear-lanceolate, glandular pubescent. Ray florets
15-21, pistillate, the ligules 6-8 mm long, 3-5 mm wide, yellow. Disk florets
numerous, the corollas ca. 6 mm long, yellow, the tube ca. 1.5 mm long, the
throat tubular, ca. 3.5 mm long. Achenes about 4 mm long, narrowly alate,
the pappus of 2 awns 2-3 mm long.

The opposite sessile auriculate leaves, large heads and densely glandular
pubescent foliage readily distinguish this taxon. It has no close relative among
the species of Verbeszna known to me.

ACKNOWLEDGMENTS

I am grateful to Guy Nesom for the Latin diagnosis and description, and
to him and Andrew McDonald for reviewing the manuscript.

Phytologia (October 1990) 69(4):297-300.

TWO NEW SPECIES OF PERITYLE (ASTERACEAE, HELENIEAE)
FROM MEXICO

B.L. Turner
Department of Botany, University of Texas, Austin, Texas 78713 U.S.A.

ABSTRACT

Two new species of Perityle belonging to the section Perityle are
described: P. pennellii spec. nov., from Cerro Mercado, near Cd. Du-
rango, Durango and P. stevensii spec. nov. from near Emcosa, Sinaloa.
Both species have white rays and yellow disks and are believed to re-
late to P. hofmetsteria Rydb. and several other taxa of northwestern
México with similar characteristics. A brief key and map to these taxa
are provided.

KEY WORDS: Perityle, Asteraceae, Helenieae, México

Preparation of a treatment of Perityle for the Asteraceae of Mexico has
revealed the following novelties belonging to the section Perityle (sensu Powell,

1974; Rhodora 76:229-306).

Perityle pennellii B. Turner, spec. nov. TYPE: MEXICO. Durango: Cerro
de Mercado, N of Cd. Durango, crevices of ferruginous rocks, 2000-2100
m, 24-25 Aug 1934, Frances W. Pennell 18160 (HOLOTYPE: NY!).

Perityle canescenti Everly similis sed differt foliis simplicibus
(vs. tripartitis) peranguste spathulatis differt.

Small suffruticose perennials 3-10 cm high; stems hispid-puberulous; leaves
mostly 8-10 mm long, 1-3 mm wide; petioles mostly 4-6 mm long; blades ovate
to elliptic, 2-4 mm long, 2-3 mm wide, sparsely hispid, glandular punctate, the
margins entire, or rarely faintly 3 lobed; involucral bracts ca. 15, 4-5 mm high,
sparsely puberulent, the apices acute; heads single on hispidulous peduncles
10-20 mm long; ray florets 5-8, the ligules 2-4 mm long, white; disk florets ca.
25, yellow, the corollas tubular, ca. 3 mm long, the lower portion glandular
pubescent; achenes ca. 2.5 mm long with margins callous thickened, ciliate with

297

298 PHEV TIO LIOGIA volume 69(4):297-300 October 1990

long hairs, the pappus of 2-3 rather persistent bristles 2-3 mm long, between
these a crown of slender scales ca. 0.3 mm high.

Powell (1974) recognized four species of the sect. Perityle with white rays,
yellow disks, and having large achenes (1.8-3.5 mm long) with prominent pap-
pus bristles (1.5-3.0 mm long). These are: 1) P. ciliata (Dewey) Rydb. (known
only from Arizona, U.S.A.); 2) P. canescens Everly, a taxon with 3 partite
leaves with linear segments (known only from Sinaloa, Sierra Tacuichamona);
3) P. coronopifolia A. Gray, also with 3 partite leaves, but the achenes having
margins with long-ciliate hairs as opposed to the merely puberulent margins in
P. ciliata (known only from Arizona, U.S.A.); and 4) P. hofmeisteria Rydb.,
with leaves 10-60 mm long, 2-20 mm wide, the blades broadly ovate to flabel-
late in outline, with margins deeply lobed to serrate, the heads ca. 8 mm high
(known only by a few collections from near Cd. Durango and in the vicinity
of Rodeo, Durango).

Among the Mexican species described to date, Perityle pennellii is closest
to P. canescens; it differs from the latter in possessing much smaller leaves
which are very narrowly spatulate with essentially entire margins.

The distribution of the Mexican taxa which make up the above mentioned
complex is shown in Figure 1. It is likely that additional taxa will come to
the fore as the bluffs and ledges of northwestern México are more thoroughly
explored.

Perityle stevensii B. Turner, spec. nov. TYPE: MEXICO. Sinaloa: Mpio.
San Ignacio, ca. 2.6 mi from Hwy 15 along road to microwave tower, ca.
17 mi SE of Emcosa, 10 Sep 1973, Warren Douglas Stevens 2040 (and
Mary Fairhurst) (HOLOTYPE: TEX!; Isotype: MSU).

Peritylae hofmeisteriae Rydb. similis sed differt foliis parvioribus
magis conspicue glandulosi-punctatis petiolis brevioribus et acheniis

setis pappi magis facile deciduis brevioribus (1.0-1.5 mm longis vs.
2-3 mm).

Suffruticose perennials 15-30 cm high; stems densely puberulent; leaves
mostly 15-25 mm long, 8-15 mm wide; petioles 7-17 mm long; blades del-
toid, irregularly crenulodentate, sparsely puberulent, glandular punctate on
both surfaces; heads single on puberulent peduncles 10-25 mm long; involu-
cral bracts 6-7 mm long, puberulent, the apices acute; ray florets ca. 11, the
rays white, 2.5-3.5 mm long; disk florets ca. 40, the corollas yellow, ca. 3 mm
long, atomiferous glandular; achenes (immature) ca. 2.5 mm long, sparsely
setiferous, the pappus of 1 or 2 readily deciduous bristles 1.0-1.5 mm long,
between these a short crown of scales ca. 0.2 mm high.

According to label data, the species is “common on steep rockface of road-
cut, stems woody but extremely brittle, rays white, disc yellow.”

299

Two new species of Perityle from México

Turner:

Perityle
A canescens
@ coronopifolia
© hofmeisteria
A pennellii

@ stevensii |

Fig. 1. Distribution of Perityle spp. discussed in text.

300 PR YT. On O- GTA volume 69(4):297-300 October 1990

Dr. Stevens (pers. comm.) notes that the type was collected on a side road
leading to Baila and Cerro de Culagua, which is about 79.5 road miles N of
Mazatlan; it was identified by John Strother as Perityle gentry: A.M. Powell,
but the latter species belongs to the section Laphamiaand is markedly different
from P. stevenszi.

Perityle stevensi is apparently most closely related to P. hofmeisteria,
which is discussed under the above species. It differs from that taxon in having
smaller, less deeply lobed, leaves with mostly shorter petioles. In addition, the
pappus bristles are shorter (1.0-1.5 mm long vs. 2-3 mm) and readily deciduous.
It is possible that P. stevensz will be found to have achenes without callous-
thickened margins; if so, the species will not readily relate to yet other Mexican
taxa. The following key should help the interested worker in distinguishing
among these taxa.

Key to the Mexican species of sect. Perityle having white rays, yellow disks
and large achenes with callous-thickened margins:

1. Achenes mostly 2.0-3.5 mm long; leaves simple; pappus bristles mostly
1,5-3.0, mm. longj-rarely shorter. o/s 64 «02%. « j-th aoe ee eee (2)

1’ Achenes mostly 1.3-2.0 mm long; leaves tripartite, pappus bristles mostly
O:5-1-5 mim long 2)... 1.592. sot ose oe oe ae (4)

2. Leaves narrowly spatulate, 2-4 mm wide, 8-10 mm long P. pennelli

2’ Leaves not spatulate,.much larger...:........0 sess e eee (3)

3. Leaves densely glandular punctate; pappus bristles 1-2 mm long;
Sinaloa’ 21nd siete nes ae bats yanks Poshele c 651s Da ena eee ee P. stevens

3’ Leaves sparsely glandular punctate, if at all; pappus bristles 2-3 mm long;

Dien t go. os, ct ost oid note misread pln aio oe tee oh P. hofmeisteria
4. Heads 2-5, borne in clusters; achenes with margins long-
CHARS od sven bid sis 4 o wie ow 3 5 is oo lane ee P. coronopifolia
4' Heads solitary; achenes merely puberulent ............ P. canescens
ACKNOWLEDGMENTS

I am grateful to Dr. Guy Nesom for the Latin diagnoses and to him and
Dr. A.M. Powell for reviewing the manuscript.

Phytologia (October 1990) 69(4):301-312.

NOMENCLATURAL NOTES FOR THE NORTH AMERICAN FLORA. IV.
John T. Kartesz & Kancheepuram N. Gandhi

North Carolina Botanical Garden, Department of Biology, Coker Hall,
University of North Carolina, Chapel Hill, North Carolina 27599-3280 U.S.A.

ABSTRACT

The authorship of the following names is discussed: Bouteloua gra-
cilis (Willd. ex Kunth) Lag. ez Griffiths, Brachiaria platyphylla (Munro
ez Wright) Nash in Small, Bromus unioloides Kunth, Dichanthelium lin-
earifolium (Scribn. er Nash) Gould, Echinochloa frumentacea Link, Se-
taria sphacelata (Schumach.) Stapf & Hubbard, and Sporobolus vagint-
florus (Torr. ez A. Gray) Wood. Aira elegantissima Schur and Hymen-
achne amplezicaulis (Rudge) Nees are accepted to be correct names.

KEY WORDS: Floristics, Nomenclature, Poaceae, Aira, Bouteloua,
Brachiaria, Bromus, Dichanthelium, Echinochloa, Hymenachne,
Setaria, and Sporobolus.

Introduction

Continuing with the “Nomenclatural notes for the North American flora”
(Kartesz & Gandhi 1989; 1990a; 1990b), a fourth note in the series is presented
here, toward advancing our understanding of North American plant names.

POACEAE

Aira elegantissima Schur

The European weed, commonly known as annual silver hair grass, has
been treated by many workers as Aira elegans Willd. ez Gaudin. Tutin (1980)
accepted the name A. elegantissima Schur for this grass and in synonymy
stated that A. capillaris Host (non Savi) and A. elegans Willd. ez Gaudin

301

302 Pe YAO OGin volume 69(4):301-312 October 1990

were illegitimate names. However, he did not elaborate on the illegitimacy of
the latter name.

While reviewing Gaudin’s (1811) work, we found that he numbered each
of his accepted taxa, which has historically been a common practice by var-
ious workers. In the last paragraph of the protologue of Aira caryphyllea L.
(number 7), he included the name “A. elegans Willd. ined.” and provided a
brief description. Since the name A. elegans was not separately numbered, its
inclusion within the protologue of A. caryphyllea could be interpreted as being
a described name in synonymy, a provisional name, or both (Greuter 1988,
Art. 34.1). Hence, the status of none of these can be considered legitimate.
Therefore, the name A. elegantissima is the oldest legitimate name for this
species complex.

Aira elegantissima Schur, Verh. Mitt. Siebenb. Ver. Naturw. 4(Sert. Fl.
Transs.):85. 1853.

Aira capillaris Host, Icon. Gram. Austr. 4:20, t. 35. 1809, non Savi,
1798.

Aira elegans Willd. ez Gaudin, Agrost. Helv. 1:130. 355. 1811 (pro syn.

and/or provisional name), nom. invalid.

Bouteloua gracilis (Willd. er Kunth) Lag. ez Griffiths

Some authors, such as Gould (1975) and McVaugh (1983), attribute the
name Bouteloua gracilis to (Willd. er Kunth) Lag. ez Steud. In Steudel (1840),
synonyms are provided in italics. On p. 219, 12 species of Bouteloua are listed
in italics (including “B. gracilis Lag.”). For the preceding name, Steudel cited
“Chondrosium gracile H.B.” as the accepted name. [The correct authorship
for the name C. gracile is: Willd. er Kunth, which was based on Willdenow’s
manuscript name Actinochloa gracilis]. On p. 355, Steudel listed the names of
Chondrosium species that he accepted and their synonyms, but for C. gracile,
he did not cite B. gracilis. Nevertheless, since Steudel did not accept the name
B. gracilis, he did not validate the new combination (Greuter 1988, Art. 34.1a).
Griffiths, who also attributed the name to Lagasca, inadvertently validated the
combination.

Bouteloua gracilis (Willd. er Kunth) Lag. ez Griffiths, Contr. U.S. Natl. Herb.
14:375. 1912. BASIONYM: Chondrosium gracile Willd. er Kunth, Nov.
Gen. et Sp. 1:176. 1814. Bouteloua gracilis (Willd. ex Kunth) Lag. ez
Steud., Nom., ed. 2. 1:219. 1840, pro syn.

Kartesz & Gandhi: Nomenclatural notes of North American flora-IV. 303
Brachiaria platyphylla (Munro ez Wright) Nash in Small

Webster (1988, p. 606) concluded that “combining Urochloaand Brachiaria,
excluding the type {of Brachiaria], is morphologically justified and taxonom-
ically necessary.” He made four new combinations in Urochloa. However, we
consider Brachiaria as distinct from Urochloa.

The combinations Panicum platyphyllum (Griseb.) Munro ez Wright (1871),
Brachiaria platyphylla (Griseb.) Nash, and Urochloa platyphylla (Griseb.) Web-
ster were based on Paspalum platyphyllum Griseb. Unfortunately, this ba-
sionym is a later homonym of Panicum platyphyllum J.A. Schultes (1827).
Hence, Grisebach must not be cited as the parenthetical author for these
names. Wright’s new combination should be treated as a nomen novum, with-
out a parenthetical author (Greuter 1988, Art. 72.2,.Note 1), and with its
priority starting from 1871 (not from 1866). This name thus can serve as
a basionym for B. platyphylla (Munro ez Wright) Nash and U. platyphylla
(Munro ez Wright) Webster. The correct nomenclature is given below.

Brachiaria platyphylla (Munro ez Wright) Nash in Small, Fl. Southeast. U.S.
81. 1903. BASIONYM: Panicum platyphyllum Munro ez Wright, Anal.
Acad. Cienc. Habana 8:206. 1871. Urochloa platyphylla (Munro ez
Wright) R. Webster, Syst. Bot. 13:606. 1988. Paspalum platyphyllum
Griseb., Cat. Pl. Cubensium 230. 1866, non J.A. Schultes 1827.

Bromus

Bromus catharticus Vahl

Although the names Bromus unioloides Kunth (1815) and Festuca unioloz-
des Willd. (1803) share specific epithets, they are based on different types,
the former from Ecuador, the latter from Carolina. In conjunction with his
description of B. unioloides, Kunth did not cite Willdenow either directly or
indirectly. For his B. unioloides, Kunth (1829) cited Schenodorus unzoloides
Roem. & Schult. (1817) in synonymy. Moreover, Kunth considered Willde-
now’s F. unioloides to be a different species, and in transferring it to Bromus,
named it B. willdenowii Kunth (1829), citing F. uniolordes and Ceratochloa
unioloides (Willd.) P. Beauv. (1812) in synonymy. Some authors, such as
Gould (1975), include the parenthetical author for B. unzoloides, which is in-
correct, whereas the author citation used by Hitchcock & Chase (1951, p.
833) and McVaugh (1983) is correct. We provide the following nomenclatural
summary.

Bromus catharticus Vahl, Symb. Bot. 2:22. 1791.

304 Pay TiOKh OO Gl A volume 69(4):301-312 October 1990

Festuca untoloides Willd., Hort. Berol. 3, t. 3. 1803. Ceratochloa
unioloides ( Willd.) P. Beauv., Ess. Agrost. 75, t. 15, f. 7. 1812. Bro-
mus untoloides (Willd.) Raspail, Ann. Sci. Nat. Bot. 5:439. 1825,
non Kunth 1815. Bromus willdenoww Kunth, Rev. Gramin. 1:134.
1829.

Bromus unioloides Kunth in H.B.K., Nov. Gen. Sp. 1:151. 1815. Schen-
odorus unioloides (Kunth in H.B.K.) Roem. & Schult., Syst. Veg.
2:708. 1817. Zerna unioloides (Kunth in H.B.K.) Lindm., Svensk
Fanerogamfl. 101. 1918.

Dichanthelium linearifolium (Scribn. ez Nash) Gould

Gould (1974) based his new combination Dichanthelium linearifolium on
Panicum linearifolium “Scribn.,” which was proposed in an appendix given in
volume 3 of Britton & Brown (1898). The basionym authorship is discussed
below.

The first edition of Britton and Brown’s /Il. Fl. N. U.S. was issued in three
volumes, with volume 1 in 1896 and volume 3 in 1898. The grass family was
treated in vol. 1, in which (p. 94) Britton indicated in a footnote that the grass
text was elaborated with the assistance of G.V. Nash. In his introduction to
the second edition, Britton (in Britton & Brown 1913, p. XIII) explicitly stated
that the “text of the grass family has been written by Mr. George V. Nash for
both editions.” Hence, all new names of grasses found in those editions must
be attributed to Nash.

Britton (Britton & Brown 1898) stated that the appendix in volume 3
included “new discoveries or new determinations, mostly from the west, made
while the work has been in press.” Although the name Panicum linearifolium
was attributed to Scribner, there was no indication that it was a new species
nor that the text was from Scribner. The distribution was given as “New York,
and New Jersey to Missouri.” Thus, this was not one of the new species from
the west, but a new determination. Had Nash mentioned this new species as
“P. lineartfolium Scribner, sp. nov.” (instead of “P. lineartfolium Scribner”),
we would accept Scribner to be the author of this species, but that was not
the case. It was most likely that Scribner provided the material and epithet
for this name. We conclude that the authorship of P. lnearifolium should be:
Scribn. ex Nash, and provide the following nomenclatural summary.

Dichanthelium linearifolium (Scribn. ex Nash) Gould, Brittonia 26:60. 1974.
BASIONYM: Panicum linearifolium Scribn. ex Nash in Britton & Brown,
Ill. Fl. N. U.S. 3:500. 1898.

Kartesz & Gandhi: Nomenclatural notes of North American flora-IV. 305
Echinochloa frumentacea Link

The combinations Echinochloa frumentacea Link (1827), Oplismenus fru-
mentaceus Kunth (1829), Panicum crusgalli L. var. frumentaceum Trimen
(1885), and E. crusgalh (L.) P. Beauv. var. frumentacea Wight (1909) are
presumed to have been based on P. frumentaceum Roxb. (1820), which is a
later homonym of P. frumentaceum Salisbury (1796). Although the latter
name is superfluous, and thus illegitimate (Salisbury cited Holcus sorghum L.
in synonymy), P. frumentaceum Roxb. must be rejected as a later homonym
(Greuter 1988, Art. 64.1, Note 1) and cannot serve as a basionym. Link’s new
combination must be considered as a nomen novum (Greuter 1988, Art. 72.2,
Note 1), with priority from 1827 (not from 1820). The combination E. frumen-
tacea Link can serve as a basionym for Kunth’s and Trimen’s combinations.
We provide the following nomenclatural summary.

Echinochloa frumentacea Link, Hort. Berol. 1:204. 1827. Oplismenus fru-
mentaceus (Link) Kunth, Rev. Gram. 1:445. 1829. Panicum crusgalli L.
var. frumentaceum (Link) Trimen, Syst. Cat. Fl. Pl. Ceylon 104. 1885.
Echinochloa crusgalli (L.) P. Beauv. var. frumentacea (Link) W. Wight,
Cent. Dict. Sup. 810. 1909. Panicum frumentaceum Roxb., Fl. Ind.
1:307. 1820, non Salisb. 1796.

Elymus alaskanus (Scribn. & Merr.) Love ssp. latiglumis (Scribn. & J.G.
Smith) Love

The subspecific new combination Elymus trachycaulus (Link) Gould ez
Shinners ssp. latiglumis (Scribn. & J.G. Smith) Barkworth & D.R. Dewey
(Great Basin Naturalist 43:562. 1983) was not validly made, because of the
lack of citation of the basionym and a full and direct reference to its author
and place of valid publication with page and date (Greuter 1988, Art. 33.2).
Hence, to date, the above new combination remains invalid.

Barkworth (pers. comm.) informed us that although she was aware of the
nomenclatural problem, she does not intend to correct it, since she presently
prefers to follow Léve’s treatment of this taxon (given below).

Elymus alaskanus (Scribn. & Merr.) Love ssp. latiglumis (Scribn. & J.G.
Smith) Love, Taxon 29:166. 1980. BASIONYM: Agropyron violaceum
(Hornem.) Lange var. latiglume Scribn. & J.G. Smith, U.S.D.A., Div.
Agrost. Bull. 4:30. 1897. Elymus trachycaulus (Link) Gould ez Shinners
ssp. latiglumis (Scribn. & J.G. Smith) Barkworth & D.R. Dewey, nom.

invalid.

306 PHA OL O'S lA volume 69(4):301-312 October 1990
Hymenachne amplezicaulis (Rudge) Nees

The genus Hymenachne consists of eight species (fide Webster 1987). The
species H. amplericaulis (Rudge) Nees (1829), a native of the West Indies,
is reported from Florida. This species was based on Panicum amplezicaule
Rudge (1805). Hsu (1978) treated H. amplezicaulis as a synonym of H. pseu-
dointerrupta C. Muell. (Bot. Zeit. 19:333. 1861), giving the range of this
grass as “Assam, Burma, Malaya to Indo-China, China and Polynesia.” Web-
ster treated both H. amplezicaulis and H. pseudointerrupta as synonyms of H.
acutigluma (Steud.) Gilliland (Gard. Bull. Singapore 20:314. 1964), with the
latter name being based on P. acutigluma Steud. (1854). Webster commented
that H. acutigluma is a native of Australia. In researching this problem, we
found that the oldest name in this complex is P. amplezicaule. Hsu and Web-
ster gave no reasons for rejecting this name and inexplicably accepting a later
name.

Bor (1960) remarked that the grass in Assam (India), known as Hymen-
achne amplericaulis, is distinct from the American H. amplezicaulis. He as-
signed the Indian element to H. pseudointerrupta. Koyama (1987) essentially
followed Bor in accepting the name H. pseudointerrupta for a Japanese species
and considered (p. 510) the name H. amplezicaulis to be misapplied to a grass
in SE Asia.

We accept Bor’s and Koyama’s interpretation that the American element
Hymenachne amplezicaulis is distinct from the SE Asian Hymenachne. Ac-
cordingly, we conclude that H. amplezicaulis is the correct name for the New
World Hymenachne. If H. acutigluma and H. pseudointerrupta are conspecific
(as indicated by Webster), then H. acutigluma has priority over H. pseudoint-
errupta. Both Hsu and Webster were correct at the time of their publications;
the former choosing the name H. pseudointerrupta; the latter choosing the
name H. acutigluma. However, both authors erred in citing H. amplezicaulis
as a synonym of their respective combinations. For the American element, the
correct nomenclature is given below:

Hymenachne amplezicaulis (Rudge) Nees, Agros. Bras. 276. 1829. BA-
SIONYM: Panicum amplexicaule Rudge, Pl. Guzan. vol. 1:21. 1805.

Non Hymenachne amplezicaule sensu Hsu or Webster.

Setaria sphacelata (Schumach.) Stapf & Hubbard

The tropical African grass Setaria sphacelata has become naturalized in
California and elsewhere in the U.S.A. The name has been attributed to “Stapf
& Hubbard” (Chipp 1929), to “Stapf & Hubbard ez Moss” (Stapf & Hubbard

Kartesz & Gandhi: Nomenclatural notes of North American flora-IV. 307

1930), to “(Schumach.) Stapf & Hubb. ez Moss” (Clayton 1966), to “(Schu-
mach.) Moss” (Clayton 1972, 1979; Clayton & Renvoize 1982; Webster 1987),
or to “(Schumach.) Stapf & Hubb.” (Hitchcock & Chase 1951; Munz 1968, p.
197 in supplement; Kartesz & Kartesz 1980).

The name first appeared in T.F. Chipp’s (1929, pp. 184, 195) article on
the Imatong Mountains of Sudan, where he stated (p. 195) “Setaria sphacelata
Stapf & Hubbard. E. of Kippia, 9000 ft. A grass 3 ft. high, one of the
commonest constituents of the mountain meadows. No. 98.” The serial no.
98 refers to the collector’s field number of the collections made in the Imatong
Mountains (as per * mark note on p. 179). A section of Chipp’s article (pp.
195-197) includes new species that were attributed to M.B. Moss, and with
all newly described species provided with detailed descriptions. However, the
species in question is not one of the new species. Hence, we speculate that
Chipp did not consider this grass to be new. Attribution of the new species to
Moss might have misled some authors, such as Clayton, to believe that Moss
was the author of the combination S. sphacelata.

On p. 184 (Chipp 1929), the mention of this species name is a nomen
nudum and does not constitute a valid publication (Greuter 1988, Art. 32.1;
Rec. 50B). In the protologue given on p. 195, no basionym is mentioned and
no reference is made to Schumacher, either directly or indirectly. If Chipp
intended the name Setaria sphacelata to be a new combination authored by
(Schumach.) Stapf & Hubb., the protologue does not meet the minimum
requirement, t.e., an indirect reference to Schumacher’s Panicum sphacelatum,
and therefore, Chipp cannot be credited with the new combination. Because
of the ambiguity of the description given by Chipp (a 5 word description: “A
grass 3 ft. high”), the alternative possibility of considering the name as sp.
nov. is also eliminated.

In light of the above analysis, we believe that the name Setaria sphacelata
does not have valid taxonomic and nomenclatural standing in Chipp’s article.
To our best knowledge, Stapf & Hubbard (1930) used this name, attributing
it to Stapf & Hubb. ez Moss, and cited the basionym (p. 796). Thus, the
new combination (S. sphacelata) was not made in Chipp’s article, but made
by Stapf & Hubbard. Accordingly, the correct authority for S. sphacelata is:
(Schumach.) Stapf & Hubbard (as given by Hitchcock & Chase, Munz, and
Kartesz & Kartesz).

Sporobolus vaginiflorus (Torr. ex A. Gray) Wood

In 1861, Wood proposed the combination Sporobolus vaginiflorus and at-
tributed the species epithet to Torrey. He also cited the following references:
“Agrostis Muhl., Crypsis Nutt.” Wood’s combination has historically been pre-
sumed to be based on Vilfa vaginiflora Torrey. We came across discrepancies

308 PLY T0.LO GTA volume 69(4):301-312 October 1990

on the authorship of both the basionym and new combination. The following
names are involved in the taxonomy and nomenclature of S. vaginiflorus:

Agrostis virginica L. 1753; Muhl. 1817; Torr. 1824.
Crypsis virginica (L.) Nutt. 1818.
Vilfa vaginiflora Torr. “in” A. Gray 1834; Torr. 1843.

Sporobolus vaginiflorus (Torr.) Wood.

In 1824, Torrey used the name Agrostis virginica L. for a northern U.S.
plant; however, he later realized that true A. virginica was taxonomically
different and more widespread than the northern plant. For the northern
grass, which he called “A. virginica L.,” Torrey excluded the Linnean type
and provided a new name: Vilfa vaginiflora. This is evident from his 1843
treatment of V. vaginiflora. He also asserted that even Muhlenberg’s “A.
virginica” belongs in Torrey’s V. vaginiflora. Torrey’s description of the plant,
along with Muhlenberg’s, although given under a misapplied name (2.e., A.
virginica), clearly apply to V. vaginiflora. Since V. vaginiflora was not a
new species, but rather a new name, no new description was necessary; a
mere reference to either Muhlenberg’s or Torrey’s description was all that was
required to validate the new name.

Between his 1824 and 1843 publications, Torrey had the name Vilfa vagznt-
flora in manuscript form. Prior to Torrey’s 1843 publication of the name V.
vaginiflora, Gray (1834) used the name and attributed it to Torrey. Gray in-
dicated that it was Torrey’s manuscript name. With reference to authorship
of the name, there has been difference of opinion, with some workers (e.g.,
Voss 1966) attributing it to Torrey and others (e.g., Kartesz & Kartesz 1980)
to Torr. ez A. Gray. Even Torrey cited the authorship as “Torr. in A. Gray”
in his 1843 work. Voss argued that Gray published Torrey’s label information
of this name (Vilfa vaginiflora Torrey, Synop. Flora, ined. Agrostis virginica
Muhl., Gram. p. 74, Torrey, Flora, v. 1, p. 89, non Elliott et Auct.). Voss
further emphasized that Torrey’s label information was sufficient to validate
the name and that Gray’s description was nonessential; hence, Torrey alone
should be credited with authorship.

We assessed the problem to determine correct authorship. Although it is
evident that Gray’s description was nonessential to validate the name, it was
Gray who actually published it. Whenever Gray used verbatim descriptions
of others, he usually provided such descriptions in quotation. However, in this
case, the description is not a critical issue. We believe that in order for Torrey
to receive credit of authorship, Gray should have specifically stated that he
published the name for Torrey; however this was not done. Mere attribution
to Torrey alone does not justify authorship to Torrey. Such attributions to

Kartesz & Gandhi: Nomenclatural notes of North American flora-IV. 309

manuscript authors were customary in the Torrey-Gray era. According to the
present practice of the ICBN committee for Spermatophyta, nomenclatural
decisions must be made on the published text (Taxon 33:300. 1984; on the
authorship of the genus Burtonia: R. Brown vs. R. Brown ez Ait. f.). Hence
we disagree with Voss and concur with Kartesz & Kartesz on the correct
authorship being: Torr. ez A. Gray.

With reference to the authorship of the new combination (Sporobolus vagini-
florus), Soil Conservation Service (1982) gave the transfer authors as: Torr.
ex Wood. As we indicated earlier, Wood attributed the name to Torrey and
cited the references to Agrostis Muhl. and Crypsis Nutt. These two references
must be construed as Agrostis virginica sensu Muhl. non L. and Crypsis vir-
ginica sensu Nutt. (non A. virginica L.). According to Greuter (1988), Art.
48.1, Wood excluded the Linnean material. Certainly, Wood was aware of
Gray’s publication, which excluded the Linnean material. Wood’s reference
to Torrey must be considered as an indirect reference to the basionym. This
interpretation is based on Greuter (1988), Art. 32, Ex. 5. Hence, Wood is the
sole transfer author for the combination S. vaginiflorus as indicated below:

Sporobolus vaginiflorus (Torr. ex A. Gray) Wood, Class-book Bot. 775. 1861.
BASIONYM: Vilfa vaginiflora Torr. ex A. Gray, Gram. & Cyp. 1:3.
1834.

ACKNOWLEDGMENTS

We thank Dr. John McNeill (Royal Ontario Museum), Dr. Mary E. Bark-
worth (Utah State University), Dr. Paul A. Fryxell (U.S.D.A. Scientist in col-
laboration with Texas A&M University), and Dr. Larry E. Brown (Houston
Community College) for their valuable suggestions for the improvement of this
manuscript. We also thank Dr. Guy Nesom (University of Texas at Austin),
Dr. Robert Webster (U.S.D.A., Beltsville, MD), and Ms. Ruth F. Schallert
(Botany Librarian, Smithsonian Institution) for providing references used in
this study.

REFERENCES

Beauvois, P. 1812. Ess. Agrost. Paris.

Bor, N.L. 1960. Grasses Burm. Ceyl. Ind. Pakist. Pergamon Press Inc., New
York, NY.

310 PAYS f. OuL 1O1G A volume 69(4):301-312 October 1990

Britton, N.L. & A. Brown. 1896. Ill. Fl. N. U.S., vol. 1. Charles Scribner’s
Sons, New York, NY.

. 1898. Ill. Fl. N. U.S., vol. 3. Charles Scribner’s Sons, New
York, NY.

. 1913. JIL Fl. N. U.S., vol. 1. Dover Publications, Inc., New
York, NY.

Chipp, T.F. 1929. The Imatong Mountains, Sudan. Bull. Misc. Inform.
177-197. [This journal was superseded by Kew Bull.].

Clayton, W.D. 1966. Studies in Gramineae: IX. Kew Bull. 20:262-264.

. 1972. Fl. W. Trop. Afr., ed. 2. vol. 3. Crown Agents Overseas
Government & Administration, London.

. 1979. Notes on Setaria (Gramineae). Kew Bull. 33:503-505.

& S.A. Renvoize. 1982. Fl. Trop. E. Afr., Gramineae, part 3. A.
Balkema, Rotterdam.

Gaudin, J. 1811. Agrost. Helv., vol. 1. J.J. Paschoud, Geneva.

Gould, F.W. 1974. Nomenclatural changes in the Poaceae. Brittonia 26:59-
60.

. 1975. Grasses Tezas. Texas A&M University Press, College Sta-
tion, TX.

Gray, A. 1834. N. Amer. Gram. Cyp., part I. (Ezsiccata).

Greuter, W. (Chairman, Editorial Committee). 1988. International Code
of Botanical Nomenclature. Regnum Veg. #118. Bohn, Scheltema &
Holkema, Utrecht.

Hitchcock, A.S. & A. Chase. 1951. Man. Grasses U.S. U.S. Dept. of Agr.,
Misc. Publ. no. 200. Govt. Printing Office, Washington, DC.

Hsu, Chien-Chang. 1978. Gramineae (Poaceae). Pp. 373-783. In: Fl. Tat-
wan, vol. 5. Epoch Publishing Co., Ltd., Taipei, Taiwan.

Kartesz, J.T. 1991. Syn. Checkl. Vasc. Fl. U.S. Canad. Greenl., ed. 2 in
press. Timber Press, Portland, OR.

& K.N. Gandhi. 1989. Nomenclatural notes for the North American
flora - I. Phytologia 67:461-468.

Kartesz & Gandhi: Nomenclatural notes of North American flora-IV. 311

. 1990a. Nomenclatural notes for the North American flora - II.
Phytologia 68:421-427.

. 1990b. Nomenclatural notes for the North American flora - III.
Phytologia 69:129-137.

Kartesz, J.T. & R. Kartesz. 1980. Syn. Checkl. Vasc. Fl. U.S. Canad. Greenl.
The University of North Carolina Press, Chapel Hill, NC.

Koyama, T. 1987. Grasses Japan. Kodansha Ltd., Tokyo.
Kunth, C.S. 1815. in H.B.K. Nov. Gen. Sp., vol. 1. Paris.

. 1829. Rev. Gramin., vol. 1. Paris.
Link, J.H.F. 1827. Hort. Berol., vol. 1. Berlin.

McVaugh, R. 1983. Fl. Novo-Galiciana, vol. 14. The University of Michigan
Press, Ann Arbor, MI.

Muhlenberg, G.H.E. 1817. Descr. Gram. Impensis Solomon. W. Conrad,
Philadelphia, PA.

Munz, P.A. 1968. Calif. Fl., Suppl. University of California Press, Berkeley,
CA.

Nees, C.G.D. 1829. Agrost. Bras. in C.F.P. Martius Fl. Bras. Enum. Pl.,
vol. 2(1)-Gramineae. Stuttgartiae.

Roemer, J.J. & J.A. Schultes. 1817. Syst. Veg., vol. 2. Stuttgartiae.
Roxburgh, W. 1820. Fl. Ind., vol. 1. Serampore.

Rudge, E. 1805. Pl. Guian., vol. 1. London.

Salisbury, R.A. 1796. Prodr. Stirp. Chat. Allerton. London.
Schultes, J.A. 1827. Mant. vol. 2. Stuttgartiae.

Soil Conservation Service. 1982. Nat. List Sci. Pl. Nam., vols. 1 & 2.
U.S.D.A., S.C.S.-T.P. #159. Government Printing Office, Washington,
DC.

Stapf, O. & C.E. Hubbard. 1930. Gramineae. Pp. 769-944. Jn: D. Prain
(ed.), Fl. Trop. Afr., vol. 9(5). L. Reeve & Co. Limited, Ashford, Kent.

Steudel, E.G. 1840. Nomencl. Bot., ed. 1, vol. 1. Typis et Sunptibus, J.G.
Cottae, Stuttgartiae et Tubingae.

312 Pay PO L OG TA volume 69(4):301-312 October 1990

. 1854. Syn. Pl. Glumac. Stuttgartiae.

Torrey, J. 1824. Fl. N. Middle U.S., vol. 1. T. & J. Swords, New York, NY.

1843. Fl. New York, vol. 2. Carroll & Cook, Albany, NY.
Trimen, H. 1885. Syst. Cat. Fl. Pl. Ceylon. Colombo.

Tutin, T.G. 1980. Gramineae. Pp. 118-267. Jn: Tutin, et al., eds., FI.
Europaeae, vol. 5. Cambridge University Press, London.

Voss, E.G. 1966. Nomenclatural notes on monocots. Rhodora 68:435-463.

Webster, R. 1987. Austral. Paniceae (Poaceae). J. Cramer, Gebruder Born-
traeger, D-1000 Berlin, D-7000 Stuttgart.

. 1988. Genera of the North American Paniceae (Poaceae: Pani-

coideae). Syst. Bot. 13:576-609.
Wight, W. 1909. Cent. Dict. Sup.
Willdenow, C.L. 1803. Hort. Berol., vol. 1. Berlin.

Wood, A. 1861. Class-book Bot., ed. s.n. A.S. Barnes & Burr, New York,
NY.

Wright, C. 1871. Anal. Acad. Cienc. Habana 8:.

i

5 00288 2304

Information for Authors

Articles from botanical systematics and ecology, including
biographical sketches, critical reviews and summaries of literature
will be considered for publication in PHYTOLOGIA. Manuscripts
may be submitted either on computer diskette, or as typescript.
Diskettes will be returned to authors after action has been taken on
the manuscript. Diskettes may be 5.25 inches or 3.5 inches but
must be written in DOS format as flat ASCII files. Typescript
manuscripts should be single spaced and will be read into the
computer using a page scanner. The scanner will read standard
typewriter fonts but will not read dot matrix print. Manuscripts
submitted in dot matrix print cannot be accepted. Use underscore
(not italics) for scientific names. Corrections made on typescript
manuscripts must be complete and neat as the scanner will not
read them otherwise. Language of manuscripts may be either
English or Spanish. Figures will be reduced to fit within limits of
text pages and therefore, should be submitted with an internal
scale and have dimensions proportional to those for text pages.
Legends for figures should be included in figures whenever
possible. Each manuscript should have an abstract and key word
list. Specimen citations should be consistent throughout the
manuscript. Serial titles should be cited with abbreviations used in
Botanico Periodicum Huntianum. References cited only as part of
nomenclatural summaries should not appear in Literature Cited.
Nomenclatural work should include one paragraph per basionym
and must provide proper (as defined by the current /nternational
Code of Botanical Nomenclature) citation of sources of epithets
and combinations.

Authors should arrange for two workers in the appropriate field to
review the manuscript before submission. Copies of reviews should
be forwarded to the editor with the manuscript. Manuscripts will
not be published without review.

Cost of publication is currently $12.00 US per page for publica-
tion without reprints. Publication with 100 reprints is provided for
$16.50 US per page, 200 reprints for $20.00 US per page. Page
charges are due with manuscript and no paper will be published
before payment is received in full. Reprints must be ordered and
paid for in advance. Page charges will be determined on the basis
of a typescript page (single spaced, 10 points, blank line between
paragraphs) with all type inside a rectangle 143 mm (horizonal) by
219 mm(vertical), not including running head and page number.
Title page should include title, author(s) name(s) and address(es).
Two blank lines should appear above and below section headings
(Abstract, Discussion, Literature Cited, etc.) in the manuscript. No
extra charge is made for line drawings provided they conform to
limitations of size and proportion for normal text. Halftones require
an extra charge of $5.00 US per page..