1 



THE IDENTIFICATION OF WALTER^ GRASSES. 39 



rated by Professor Scribner as A. intermedia. In 1789, 

 the year following the publication of Walter's work, 

 Eraser, who was familiar with Walter's plants, wrote an 

 article in the " Gentleman's Magazine," entitled M Fra- 

 ser's Carolina Grass," in which he gives a good descrip- 

 tion accompanied by a plate, and cites Comucopiae per- 

 ennans Walt, as a synonym. A specimen of this is in 

 the herbarium of M. de Candolle at Geneva. 



altissima 3. culmo erecto, duro; panicula coarctata ; flor 



magnis. 



No specimen. Pursh refers this to his Trichodium 

 elatum along with Agrostis dispar Michx. The latter is 

 Agrostis alba L., and I think it much more likely that 

 this is also the basis of Walter's C altissima. It is true 

 that his generic description calls for a " 1-valved corolla," 

 but the palet in Agrostis alba is not easily seen and might 

 well have been overlooked. Trichodium elatum is a 

 comparatively rare plant, especially in Walter's region, 

 and if seen he is not likely to have separated it from his 

 C. perennans, judging from his work in general. 



31. Alopecurus. CaL 2-valvis. Cor. 1-valvis, apice 



simplich 

 pratensis 1. culmo spicato erecto, glumis villosis. 



There is a specimen in the herbarium labeled "Alope- 

 curus 470," which consists of a spike and portion of 

 culm but no leaves and only a few spikelets remaining 

 at the base of the spike. The awn is about twice the 

 length of the spikelet. It appears to be A. geniculatus. 

 Whether this represents his A. pratensis or A. carolini- 

 anus it is impossible to say. Elliott refers A. pratensis 

 Walt, doubtfully to A. geniculatus. The brief descrip- 

 tion which is copied from Linnaeus applies well enough. 



