82 



Rhodora. 



[May 



Specific characters, or those used to define s])eeies, are no less vari- 

 able than the generic, and it is the purpose of this pai)er to call atten- 

 tion to some of these in a few of our native (grasses concerning whose 

 classification there has been some <liversit\' of opinion. A l)rief dis- 

 cussion of the genera included is necessarv to the presentation of the 

 subject. 



Tkiseti'M. — Persoon, in ISO'), segregated from Airtm a number 

 of species upon wiiich he founded the genus Trisefum with the follow- 

 ing characters: 'Ta/. 2 3-florus, acuminatus, carinatus. Cor. aristis 

 2 terminalibus .subdentiformibus (glum, apice setoso-bifidis). 1 dorsali 

 recta nee contorta, flosc. ut ])lurimum glabris. {Spirillar compressae, 

 palle.scentes.)" ' The bidentate flowering glumes straight, not twisted, 

 dorsal awns and compressed spikelets were apparently the ditferential 

 characters. Persoon evidently had some doubts as to the limitations of 

 his genus, for he concludes his enumeration of the species by the obser- 

 vation: "Nonnullae generis .1 /r//r/r species ex. gr. Av. pensylvanica, 

 lupulina et purpurea hue (juofjue pertinere videntur, {juae ulterius 

 examen hinc merentur." Trisrfum pahi.sfrc he leaves in Arena. 

 Eleven species make up the genus, six of which are now referred to 

 other genera; one, the first described, to I'ruiniafa; two to Danthonia; 

 and three to Avena. The fourth si)ecies described, T. nituium Pers., 

 is still retained in the genus, and being the first species for which a fig- 

 ure is cited, must be accepted, according to the opinion of some tax- 

 onomists, as the type of the genus. 



Seven years later, in 1S12, Heauvois- takes up Triwtum, increasing 

 the number of species to eighteen, only half of which, according to the 

 Kew Index, are now retained in the genus; Vcntenata, Danthonm, 

 Aira, and Avena claiuiing seven while two are merely synonyms. 

 It may be of interest to note here that the Tri.settim .iiih.tpi'rafum 

 is the Aira suhspimta of Linnaeus^ but was jniblished six years 

 earlier by the same author under Aim .tpirafa (Linn. Sp. PI. 1: 

 64. 1753),* the name taken up by Richter in 1890 (Richt. PI. Eu. 1: 

 59.) Beauvois apparently had a very clear conception of the genus as 

 it is understood today, and his failure to apply [)roperly his diagnosis, 

 shown by his rather heterogeneous assemblage of species, only seems to 



« Pers. Syn. 1 : 97. 1805. 

 2 Agros. 88, PI. 18. fig. 1. 

 3Syst. Niit. ed. 10. 2: 87:J. 1759. 



■• Aira spicata on page 63 of Species Plantarum was changed l>y Linnaeus in the 

 errata to Aira indica. 



