INTRODUCTION. 



The many and striking" difif'ereiices presented by the specimens which 

 have been referred to Sitanion hystrix [Ely m us si tan ion) have long been 

 recognized, bnt no one has heretofore attempted to define or classify 

 them. From the material in the herbarium of the Academy of Natural 

 Sciences of Philadelphia, it is evident that Nnttall distinguished at 

 least two species. These are shown in Plate I, the tickets attached 

 to the specimens being in NuttalTs handwriting-. The large amount 

 of material in the National Herbarium, gathcre 1 from numerous and 

 widely separated localities by many collectors, has attbrded an excellent 

 opportunity for a study of the variations which with the increase of the 

 collection became more and more apparent, and the necessity of their, 

 classification more and more evident. The present paper, pre[)ared by 

 my direction, was undertaken to meet this necessity, autl while the 

 species here defined may require some modification after further studies 

 in the field, and while some classed as species may eventually be reduced 

 to varieties, the subject as i)resented can hardly fail to be of interest 

 to the student of grasses and helpful in the close discrimination of the 

 species of a critical group of plants, 



Nuttall,' who first described the species of this genus, referred it to 

 the European ^Er/Uops and named his plant J^yilops hystrix. His 

 description was carefully drawn up and his species can be readily 

 recognized. A year later, Rafinesque- published his genus Sitanion, 

 based upon a single species, which he named iSitanion etymoides. It has 

 been found impossible to determine with certainty which of the species 

 enumerated in the present paper was the one named by IJafinesque; it 

 certainly was not, however, the grass described by Nuttall. 



Our leading authorities, Bentham and Hooker,^ Hackel,^ and Baillon,^ 

 have all reduced Sitanion to a section of Elym us. The articulate rachis, 

 readily breaking up at maturity, and the usually bifid or many parted 

 and awned empty glumes are well-defined characters, distinguishing 

 the si)ecies from Ulymus, and justifying their separation as a distinct 



1 Genera North American Plants, 1: 86. 1818. 



Mourn. Phys., 89: 103. 1819. 



^Genera Plautarnm 3: p. 1207. 



^Die Natiirlichen Pflanzenfamilien 2: part 2, p. 88. 



*Histoire des Plantes, Monographie des Graminees, 2.58. 



