REVISION OF ELEODIINI BLATSDELL. 411 



Lo7ig{collis has a wide range of distribution. In Mexico it occurs 

 in the States of Coahuila, Chihuahua, and Sonora. 



In the United »States it is abundant in Texas, Arizona, New 

 Mexico, Kansas, and Colorado. It probably extends into Nebraska, 

 Wyoming, and Utah. A specimen in the National Museum collec- 

 tion bears a label "Oregon.'' I believe this is erroneous; at any 

 rate it remains to be verified. Longicollis occurs in Colorado at an 

 elevation of 7,000 feet (Wickham). Innoceths thus far is Only known 

 to occur in Lower California. I have no knowledge of it occurring 

 on the mainland in Mexico. 



Genealogy. — It would appear after a careful survey of the large 

 series before me representative of the subgenus Steneleodes that the 

 species have descended from a common ancestral stock along two 

 distinct and divergent lines. Of course this only applies to those 

 species included in our fauna. Similar study of the Mexican species 

 would be the oidy way to definitely settle this question. 



The Mexican forms related to innoeeyis I 

 know nothing about. Longicollis and gigan- cigaidea 



tea are closely related, and I believe the »mata^ ^ ^oenb-Es 

 former had its birth in Mexico, diffusion 

 taking place to the nortlnvard, a western 

 modification resulting in the latter. 



Unfortunately I have not seen any marked 

 races of longicollis; there are undoubtedh' 

 some to be foiuid in Mexico. Gigantea has 

 apparently thrown off two divergent forms — 

 gentilis, the older, and estriata., the more 

 recent. I might say that innacens is the more pj^ t.-genealogical dia- 

 heteromorphic; longicollis.^ gentilis^ and graai of the sdbgends 



, , ', \ . r-t , • , STENELEODES. 



gigantea as the more homomorphic. t striata 

 is slightly intermediate. 



Innocens has the abdomen oblique in the male ; longicollis., gentilis, 

 and gigantea have the al)domen horizontal in both sexes; estriata has 

 the abdomen slightly, but undoubtedly, oblique. This may be an 

 atavistic reversion to the primitive form, which may have had an 

 oblique abdomen. 



Our species and their races may be separated as follows : 



Elytra estriate: surface luster shining and bright. 



Prouotum feebly convex antero-posteriorly, normally strongly convex from 

 side to side; sides rather evenly arcuate when viewed from above. 

 Bast of the Colorado Kiver longicollis. 



Pronotum rather strongly convex antero-posteriorly ; sides strongly arcuate 

 anteriorly. West of the Colox-ado River and Sierra Nevada Moun- 

 tains gigantea. 



Pronotum moderately and evenly convex, subquadrate. 



