HELL CREEK BEDS, CERATOPS BEDS AND EQUIVALENTS 235 



made plain, that, so far as present knowledge goes, the relationships 

 of the mammals of the Ceratops beds are, with the exception of the 

 two doubtfully placed Judith River forms, exclusively Tertiary 

 and not Cretaceous. That they had an anterior period of develop- 

 ment is, of course, probable, but we do not know at present what that 

 was, or how far back we must go for their starting point. 



From what Mr. Gidley has stated it appears that the mammals of the 

 "Ceratops beds" cannot be considered as affording evidence against 

 the reference of these beds to the Tertiary. The Ceratops mammals 

 cannot be regarded as the direct ancestors of those in the Puerco — 

 for this would imply a considerable time interval — but rather as a 

 different, possibly slightly older, phase of the same evolutionary 

 stage. Recently Mr. Gidley has been studying a very interesting 

 mammalian fauna secured in the vicinity of Melville, Montana, in 

 beds of the Fort Union, which are of undoubted Torrejon affinity, 

 while in beds of the upper member about 1200 feet above its base 

 is another fauna which is even more typical of the Torrejon. Mr. 

 Gidley permits me to say that the mammals of the " Ceratops beds" 

 are apparently more closely related to those of the Fort Union 

 (Torrejon) than they are to those of the Puerco, there being among 

 the multituberculates as already stated three genera in common with 

 the former and none with the latter. The evidence afforded by the 

 trituberculates in general is similar, that is, they show more aflSnities 

 with those of the Torrejon than with the Puerco. 



c Chelonians. 



Remains of turtles appear to be of rather common and widespread 

 occurrence in the lower Fort Union, but they are usually so frag- 

 mentary as to be difficult or impossible of satisfactory identification; 

 in any event, among the numerous specimens brought in in recent 

 years by the writer and others, it has not been found possible to make 

 even generic determinations by the paleontologists to whom they 

 have been submitted. 



It has recently been confidently asserted on eminent authority 

 that the turtles of the "Ceratops beds" are very closely related to 

 those of the Judith River formation; in fact that several species are 

 identical, and others so close as to be separated with difficulty. 



