STAPHYLINID^. 53 



the front angles well rounded and the sides slightly and roundly angulate 

 just in front of the middle; it is of the same width as the head, or perhaps 

 slightly narrower, and shows the faintest sign possible of a longitudinal 

 median angulation. The elytra are considerably broader than the thorax, 

 narrowed and rounded in passing forward toward the humeral angle, have 

 apparently the same surface structure as the thorax, and are about as long 

 as broad. The abdomen is very regularly elongate obovate, the sides 

 being nowhere quite parallel but slightly rounded and the narrowed tip 

 with a regular ovate outline. The remanis of the legs show them to have 

 been similar in length and stoutness to those of the species mentioned. 



Length of body, 19.5 mm.; width of thorax, 3.85 mm.; of elytra, 

 6.5 mm. 



Florissant, Colorado; one specimen. No. 16410. 



Staphylinus sp. 



Another species occurs at Florissant, apparently belonging here or to 

 Ocypus, and of about the size of 0. ater Grav. It has a somewhat similarly 

 shaped head and thorax, which are smooth and glabrous, or nearly so, but 

 the only specimen obtained (No. 11751) is so badly broken that it can not 

 be further described. 



PHILONTHUS Curtis. 



A dominant cosmopolitan genus, of which eighty or ninety species are 

 ah-eady known in the United States and Canada. Including those here 

 given, seven fossil species have been described from the earlier Tertiary 

 deposits of Colorado, France, and the Rhine, and the genus has also been 

 recognized in amber. A single species has been found in the Pleistocene 

 of Bavaria. 



The species which we have here grouped under this generic name agree 

 in certain characteristics by which they differ from modern species of this 

 genus. This is particularly the case in the nearly equal width of the head 

 and prothorax, the brevity and rather uniform breadth of the latter, and the 

 shortness of the antennae. It is probable that if we could become better 

 acquainted with their entire structure we should be forced to separate them 

 as a distinct generic type. 



