50 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 
n.301).’’ The type is in the herbarium of Drake de Castillo. The name was earlier 
mentioned by Hemsley.¢ 
~ Panicum proliferum chloroticum Hack. in Repert. Nov. Sp. Fedde 7: 343. 1909. 
Based on P. chloroticum Nees. 
This species was referred by Pursh,® as it has been by most later authors, to P. 
proliferum Lam. The latter is, however, the same as P. miliare Lam.,an Old World 
species. (4a Safir Reyne of ie 
DESCRIPTION. © aut riot Rete, Ber, (-2hbe 
Rap, (G0'4\, 146g 
Plants usually freely branching, ascending or spreading from a geniculate base, or 
sometimes erect, usually smooth throughout, or, in tropical forms, more or less pubes- 
cent; culms somewhat compressed, often thick and succulent, drying furrowed, 
usually 50 to 100 cm. long, in robust specimens as much as 2 meters long, the nodes 
smooth, at least the lower swollen; sheaths often com- 
pressed, usually longer than the internodes, ciliate on the 
margin toward the summit; ligules 1 to 2mm. long; blades 
flat or in small specimens sometimes folded, glabrous or 
sparsely pilose above, 10 to 50 cm. long, 3 to 20 mm. wide, 
at base about as wide as sheath, the white midnerve usu- 
ally prominent; panicles terminal and axillary, included 
at base or tardily short-exserted, many-flowered, 10 to 40 
cm. long or more, the main branches ascending, or finally 
spreading or even reflexed, the short branchlets appressed, 
: “™- bearing short-pediceled, often rather crowded spikelets, 
From specimen of Z’. genic- the axes angled and scabrous; spikelets narrowly oblong- 
latum Muhl. in Elliott Her- 5 7 
barium. ovate, 2 to 3.2 mm., usually abeut 2.5 mm. long, about 0.9 
mm. wide, acute, often greenish purple; first glume one- 
fifth to one-fourth the length of the spikelet, truncate or broadly triangular; second 
glume and sterile lemma more or less pointed beyond the fruit, rather faintly 7-nerved, 
the palea of the sterile floret present or wanting; fruit 1.8 to2 mm. long, about 0.8 mm. 
wide, elliptic. 
This species as it occurs in the United States is usually glabrous throughout but 
varies much in the size of the blades and of the spikelets, the latter varying from 2 to 
3.2 mm. in length. Not uncommonly specimens occur with the upper surface of 
some or all of the blades sparsely or even densely pilose, such as: CONNECTICUT, 
Wilson 1248; New Yorx, Young in 1872; PENNSYLVANIA, Heller in 1900; DELAWARE, 
Commons 230; Kansas, Carleton in 1892; Fioripa, Chase 4294, Combs 94, 1251. One 
series of specimens from Florida, Nash 567,¢ is low, 20 to 30 cm. high, with narrow 
blades pubescent above, and papillose-hispid sheaths. Nash’s no. 372 from the same © 
locality is glabrous throughout, except the ciliate margin of the sheaths, but otherwise 
is the same as his no. 567. Two Cuban specimens, Hitchcock 149 and Wright 3860, are 
like Nash’s no. 567. Many of the West Indian specimens have blades pilose apove, 
_ some of which have spikelets about 2 mm. long and others about 3 mm. long. Such 
are: Brace 3742, Britton & Cowell 432, Curtiss 177, Duss 3178, Eggers 4405, 4512, Geogr. 
Soc. Baltimore 489, Hitchcock 150, Wright 3861. The South American specimens cited 
are glabrous. Those from Arechavaleta and Morong 543 have small spikelets as in the 
Fig. 31.—P. dichotomifiorum. 
Mexican Gramineae’ is not yet published; but being already printed off and M. 
Fournier having obligingly supplied me with a copy, I feel bound in so far as I am con- 
cerned, to treat it as having already taken date.’’ The Kew copy ends with page 150 
and lacks index, title-page, and plates. 
a Biol. Centr. Amer. Bot. 3: 489. 1885. 
6 Fl. Amer. Sept. 1: 68. 1814. 
¢ This number was distributed under an unpublished varietal name. 
