GRASSES OF MICHAUX's FLORA BOREALI-AMERICANA. 149 



is also on this sheet a small specimen of P. lindheimeri Nash. There are two other 

 sheets from the same source, but without locality. One is P. verrucosum Muhl. The 

 other appears to be P. gravius Hitchc. & Chase, though it may be P. dichotomum. In 

 determining which plant shall be taken as the type it is to be noted that the locality 

 given in the description is "Carolina." The only specimen having this locality upon 

 the label is the one in the herbarium of Drake de Castillo, which is P. ashei. The 

 description, however, mentions that the nodes are barbed, which applies to P. gravius, 

 the plant in the Michaux herbarium, and to none of the others concerned. The speci- 

 men in the Michaux herbarium (excluding the small plant P. lindheimeri) has there- 

 fore been taken as the type, although it does not come from Carolina. Michaux evi- 

 dently confused several species, but we must surely apply the name to a species with 

 barbed nodes. The type is not what has been called P. barbulatum in all recent botan- 

 ical works. This latter species has a smaller spikelet (1.5 mm. long), while P. gravius 

 has spikelets 2 mm. long. The plant commonly called P. barbulatum must take the 

 name P. microcarpon Muhl.; Ell. Bot. S. C. & Ga. 1816 (not Muhl. Gram. 1817, which 

 is P. polyanthes Schult.). 



Panicum ramulosum Michx. 



"In pratis, cespitosis Carolinae." A poor specimen without spikelets, but certainly 

 of the angustifolium group, apparently P. angustifolium Ell. This name antedates 

 any of those applied to P. angustifolium and its allies, but on account of the fragmentary 

 condition of the specimen it would not be wise to take it up. There is nothing 

 in the description which will identify the plant any more certainly. In the herbarium 

 of Drake de Castillo are two specimens from Michaux sent by Richard under this name. 

 One is P. dichotomum L., the other is P. lindheimeri Nash. There is also a specimen 

 of the latter species in the Berlin herbarium sent by Richard under the name of 

 P. ramulosum. 



Panicum melicarium Michx. 



"In Carolina ad ripas rivorum amuviente mari inundatus [sign for perennial].". 

 The specimen is not a Panicum at all, but Panicularia elongata (Torr.) Kuntze. The 

 species has been much misunderstood and was rendered doubtful by the character 

 mentioned in the description of a sterile rudiment of a second flower, a character not 

 found in the genus Panicum. The spikelets of the specimen are past maturity and 

 consist of empty glumes or with the lowermost florets still attached. This floret 

 bears behind it the joint of the rachilla leading to the second floret, thus explaining 

 the character mentioned by Michaux. This species becomes Panicularia melicaria 

 (Michx.) 



Panicum divaricatum Michx. 



"In cespitosis excelsarum montium Carolinae Septentrionalis [sign for perennial]. -'- 

 This is Festuca obtusa Spreng. (F. nutans Spreng.). Michaux doubted that this 

 was referable to Panicum. The spikelets are past maturity and like the preceding 

 species consist of empty glumes or with the addition of the lowermost floret, which 

 bears, of course, the joint of the rachilla. Michaux describes the spikelet as being 

 2-flowered, the second flower being a sterile pedicel. This species is of course quite 

 different from Panicum divaricatum L. 



Oryzopsis asperifolia Michx. 



"In praeruptis et saxosis per tractus montium a sinu Hudsonis ad Canadam [sign 

 for perennial]." The specimen belongs to the species described under this name in 

 Gray's Manual. 



Agrostis indica " Sw. obs.' ? 



"A Virginia maritima ad Floridam [sign for perennial]." The specimen is Spo- 

 robolus indicus (L.) R. Br. 



