128 Greene Revision of the Genus Wislizenia. 



Gila Valley in southern Arizona. From another region remote 

 from this and more remote still from Texas and northern 

 Arizona were a series of specimens all with valves strongly 

 ribbed but not reticulate ; while again, and from very far to the 

 northward of what had been regarded as the range of Wislizenia 

 there appeared a member of the genus exhibiting almost the 

 valves of the original Texan species, but in foliage and habit 

 extremely different from all ; so that it became manifest that 

 the genus must needs be revised, and a fair number of excellent 

 geographic species given recognition on characters of the fruit 

 chiefly ; though each species seems to have its own particular 

 area, and is thus rather perfectly isolated, topographically and 

 climatically, from every other. 



While pursuing this line of research, Mr. J. N. Rose pleas 

 antly surprised me by bringing forth a series of specimens of his 

 own gathering in Sonora and Lower California, upon which he 

 had undertaken a critical study long since, which study had 

 been interrupted, and these, together with the manuscript on 

 them, he generously submitted to me, as an aid to this general 

 revision. His own Sonoran species, both of them well marked 

 in character, conclude the subjoined list of species mostly new. 



Out of the ten species here recognized, all seem true to the 

 original account of the group as a genus of annuals, except one 

 inhabiting the Lower California!! peninsula which seems to be 

 not only of perennial duration, but is even suffruticose, if not 

 indeed altogether shrubby. 



In the course of this study, no modification of fruit, such as 

 might tend in any degree to the weakening of the generic char 

 acter of Wislizenia has been observed ; neither does their appeal- 

 in any of the seven new species, the least habital leaning towards 

 Oxystylis ; which genus, still monotypical, might in almost any 

 event be rested on its strange habit, with lateral and densely 

 glomerate inflorescence ; in both which particulars it is most 

 unlike any and every other genus of the family. 



KEY TO THE SPECIES BY THE VALVES. 



Valves obovoid or pyriform, rounded at apex, either not tuberculate or 

 the tubercles few and low. 



Valves short pyriform, with few lines, a well defined reti 

 culation, and a few low tubercles at summit. 



1. W. refracta. 



