122 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



outline a weakly developed salient angle which later becomes 

 the strongly developed salient angle of the triangle. One 

 specimen of longicaudus (No. 19976) is unusual in having the 

 second outer triangle opening into the inner triangle forming 

 with it a transverse loop which opens very narrowly into the 

 posterior loop. In a few examples of longicaudus (Nos. 21143, 

 19130, 19174, 19985, and 21154) the second outer triangle 

 opens narrowly into the posterior loop. 



In all examples of the subgenus Phenacomys at hand, except 

 in one specimen of P. orophihis (No. 101058), the outer 

 portion of the crescent or boomerang is not closed to form a 

 second outer triangle, but communicates, usually broadly, with 

 the inner portion. 



In the original description of the genus Phenacomys, Mer- 

 riam states (1889, p. 31) that sometimes the outer loop of 

 the trefoil is closed, giving the tooth two external closed 

 triangles and a postero-internal loop. Elliot's illustration 

 (1899b, p. 255) of the upper tooth-row of olympicus shows 

 the second outer triangle closed as in longicaudus. A cut of 

 the upper tooth-row of orophilus published by Merriam (1891, 

 opp. p. 130, pi. Ill, fig. 3) shows a small second outer triangle 

 closed off from the postero-internal loop, as in longicaudus. 

 Similar relations hold in another illustration in the literature 

 (Elliot, 1901, p. 167). 



It is clear from these facts that longicaudus could not be 

 certainly identified on the enamel pattern of the back upper 

 molar. But there is an average difference between the situa- 

 tion in longicaudus and that in subgenus Phenacomys, in which 

 latter, ordinarily, the crescent or boomerang looks as if it 

 had been held in the middle while still soft and its arms 

 pulled out anteriorly. In Arhorimus the outer arm of the 

 crescent becomes the second outer triangle, which is generally 

 closed in both directions, and the postero-internal loop, which 

 is rounded in outline and lies less in an antero-posterior position 

 than the inner arm of the crescent in subgenus Phenacomys. 



(3) Front lower molar 



Agrees with all the species of the genus of which I have 

 material before me in possessing an anterior tripartite trefoil, 



