HITCHCOCK AND CHASE—-NORTH AMERICAN PANICUM. 49 
aagreste, the equivalent of the species, “Habitat in graminosis cultis prope Soteropo- 
lin et Oeiras provinciae Bahiensis et Piauhiensis;”’ £ sylvesire, ‘Habitat in sylvis ad 
Almada, Ferradas et in via Felisbertia districtus Insulanorum prov. Bahiensis, 
(Martius et Maximil. Princ. Neovid.);”’ 7 pingue, ‘Habitat in cultis ad Soteropolin, 
provinciae Bahiensis.’’ Specimens of none of these could be found in the Trinius 
Herbarium, and therefore the specimens described by Nees, which are in the Munich 
Herbarium, may be considered the types. These are all labeled with the published 
data as given above. The types of agreste and pingue are similar, having broad blades 
and rather dense panicles of small spikelets 2.2 to 2.3 mm. long as in Chase 4234 from 
Florida and Morong 543 from Paraguay. The type of variety sylvestre differs in having 
narrow and shorter blades, smaller, more open panicle, and larger spikelets about 3 mm, 
long as in Riedel 959 from Brazil. This form may prove to be a distinct species, None 
of the specimens shows the base of the plant. Kunth erroneously refers variety 
sylvestre to Panicum brachiatum Poir., which is a species of Chaetochloa, 
Panicum elliottti Trin.; Nees, Agrost. Bras. 170, 1829. This is mentioned as a syn- 
onym under P. proliferum Lam. which latter name Nees applies to P. dichotomiflorum 
Michx. The type was not found in the Trinius Herbarium nor at Munich, 
Panicum retrofractum Delile; Desv. Opusc. 96. 1831. Desvaux gives no locality 
other than “America borealis.” The type, in the Jussieu Herbarium, is from 
*‘Caroline,’’ and is the typical form, 
Panicum hygrophilum Salzm.; Steud. Syn. Pl. Glum, 1:71. 1854. “Bahia.’’? In 
the National Herbarium is a specimen labeled P. hygrophilum Salzm. from Bahia, 
which agrees with Nees’s variety sylvestre. There is a specimen of the same in Van 
Heurck’s herbarium, where is located the original set of Salzmann, and duplicates in 
Hackel’s and other European herbaria, but we do not know which specimen was seen 
by Steudel. 
Panicum proliferum pilosum Griseb. Cat, Pl, Cub. 232. 1866. “Wrfight] a. 1865, 
ad lagunas.”” The type, in the Grisebach Herbarium, is labeled, ‘Around lagunas 
in wet or damp ground, Hanabana,’’ no. 186. This is a small plant with spreading or 
decumbent culms, papillose-hispid sheaths and blades villous above. Nash’s no. 567 
from Eustis, Florida, is similar to this but has somewhat larger spikelets. 
Panicum proliferum strictum Griseb. Cat. Pl. Cub. 232. 1866. “Wrlight] 3456.” 
The type, Wright’s no. 3456 from Cuba, is in the Grisebach Herbarium. The spikelets 
are about 3 mm. long, the sheaths smooth, the blades villous above. 
Panicum proliferum geniculatum Wood, Bot. & Flor. 392. 1874. This is probably 
based on P. geniculatum Ell., though that name is not mentioned; no locality nor 
specimen is cited. Vasey makes the same combination, basing it upon P. genicu- 
latum Ell. 
Panicum amplectans Chapm. Bot. Gaz. 3:20. 1878. ‘South Florida.’? The type, 
in the Chapman Herbarium at Biltmore, was collected by Blodgett. 
Panicum francavillanum Fourn. Mex. Pl, 2:25, 1886.¢ ‘Tacabaya (SCHAFFN[ER] 
@Enum. Pl, 1: 155. 1833. 
6 Grasses U. 8. 12. 1883. 
¢ The date given on the title-page of this work is 1886. A set of proof sheets was 
supplied to Bentham in 1880 and is referred to by the latter author in his paper, 
Notes on Gramineae, read November 3, 1881, and published in the Journal of the 
Linnaean Society (Botany 19: 14-134. 1881). Fournier’s names are also cited by 
Hemsley (Biol. Centr. Amer, 3: 1885), to which work they are referred by the Index 
Kewensis, but the names are there usually nomina nuda. The proof sheets mentioned 
above are in the library at Kew, marked, ‘‘Proof sheets of Mr. Fournier Gramineae, 
1881. From Mr. Bentham.’’ They are stamped, ‘‘lre Epreuve 18 Mai 1880.”’ Ben- 
tham says of these (Notes on Gramineae, p. 20), ‘‘Eugéne Fournier’s ‘Enumeration of 
41616°—vor 15—10——4 
