HITCHCOCK AND CHASE — NORTH AMERICAN PANICUM. 



87 



DESCRIPTION. 



Plants like P. repensAn habit; culms on the average lower, rarely over 30 cm. high; 

 sheaths and blades usually glabrous, more crowded than common in P. re-pens; panicle 

 smaller, narrower, more densely flowered, commonly purple; spikelets 2 to 2.4 mm. 

 long, about 1 mm. wide; first glume broadly triangular, one-third to nearly half the 

 length of the spikelet; second glume slightly shorter than the sterile lemma. 



This species is closely allied to P. repens and approached by occasional specimens 

 of that species, which varies more than does this. 



DISTRIBUTION. 



Sea beaches, along the Gulf Coast from Alabama to Louisiana and south to Vera 

 Cruz; also on the coast of Uruguay. 

 We have seen no specimens of this 

 species from the Old World. 



Alabama: Mobile, Mohr in 1881. 

 Mississippi: Mississippi Sound, 

 Smith in 1885; Horn Island, 

 Tracy 7753; Deer Island, Sey- 

 mour in Seymour & Earle 

 Mex. Gulf Fl. 91825; Ship Is- 

 land, Tracy in 1898; Petit Bois 

 Island, Tracy 4566; Bay St. 

 Louis, Langlois in 1883. 

 Mexico: Vera Cruz, Milller 2177, 



Pringle 5569 (Hitchcock Herb.); Antigua, Liebmann 450; Coatzacoalcos, C. L. 

 Smith 913. 

 Uruguay: Maldonado, Baldwin (tlitchcock Herb.). 



Fig. 78.— Distribution of P. gouini. 



44. Panicum virgatum L. 



Panicum virgatum L. Sp. PI. 59. 1753. Linnaeus gives a short diagnosis of his 

 own, also quotes a diagnosis from Gronovius, and gives as habitat "Virginia." The 

 type specimen, in the Linnsean Herbarium, appears to have been received from 

 Gronovius as the sheet bears Gronovius's phrase name and the number 578, which is 

 the Clayton number referred to by Gronovius. a Gronovius's specimen of Clayton's 

 no. 578, in the herbarium of the British Museum agrees with the Linnaean specimen. 

 These represent the medium form of the species with open panicle and spikelets 4.1 

 to 4.2 mm. long. Clayton's no. 606 is the same. 



Panicum coloratumWalt. Fl. Carol. 73. 1788, not L. 1767. There is no specimen 

 of this in Walter's herbarium. The brief description applies well to P. virgatum, to 

 which Pursh& referred Walter's species. 



Eatonia purpurascens Raf. Journ. de Phys. 89 : 104. 1819. For locality the author 

 gives' ' ' Dans les marais maritimes de New- York, etc . " In the De Candolle Herbarium 

 is a specimen from Rafinesque so named by him. It bears the data "Long Island, 

 Rafinesque 1819." This consists of a leaf and a narrow panicle of P. virgatum, with 

 spikelets 3.5 mm. long. Rafinesque's description well agrees with this species but 

 his comparisons of it with other genera and especially the names he gives as synony- 

 mous are misleading. c 



Panicum pruinosum Bernh.; Trin. Gram. Pan. 191. 1826. This is given as a 

 synonym of P. virgatum (1 glaucum. Trinius states " V. spp. Am. bor. (Bernhardi sub 



«F1. Virg. 2: 133. 1743. 



& Fl. Amer. Sept. 1 : 67. 1814. 



cSee Scribner, Rhodora 8 : 137-138. 1906. 



