J?54 ON THE EXTINCT MAMMALIA OP 



CoNCLtJDiNG Remarks. 



In a retrospect of the tertiary fiiunae of Dakota and Nebraska, exemplified by the 

 fossil remains described in the preceding pages, amid the rich evidences of mammalian 

 life one cannot but be struck with the extreme deficiency in other forms. With the 

 exception of one or two species of turtles, and a single terrestrial mollusk, no repre- 

 sentative of any other Class or Order has been discovered in immediate association 

 with the extinct mammals. 



The geological constitution of the localities concerned make it appear that the 

 formations in which the mammalian fossils occur are the deposits of ancient lakes, or 

 of the estuaries of streams communicating with the latter. In this view of the forma- 

 tions we are led to inquire why they exhibit no traces of fishes or aquatic mollusca 

 mingled with the multitude of relics of terrestrial mammals. Even remains of the 

 latter of decided aquatic habit are absent. With the exception of the marsh-loving 

 Rhinoceros and the Beaver, no amphibious mammals have been discovered, not even 

 the Hippopotamus, whose remains are frequent in cotemporaneous formations of 

 Europe and Asia. 



The constitution of the skeletons of most fresh-water fishes, thoughLcomparatively 

 unfavorable to their preservation as fossils, can hardly be admitted as a sufficient 

 reason for the total absence of their remains in the formations in question. The 

 conditions during which the formations of the Mauvaises Terres were deposited would 

 appear to have been especially favorable for the preservation of the most delicate 

 structures. The mammalian fossils, in the perfect preservation of their original sharp- 

 ness of outline without the slightest trace of erosion, and the character of their con- 

 taining matrix, indicate quiet water with a soft muddy bottom. The chemical 

 constitution, too, of the matrix and fossils, indicate a petrifying quality in the mud 

 and water favorable to the preservation of any animal skeleton. 



The absence of remains of fishes and aquatic moUusks in association with the 

 mammalian fossils, both in the Niobrara and Nebraska formations, may be accounted 

 for by supposing that the lakes in which were formed the deposits containing the 

 fossils may have been periodically subjected to admissions of salt water from the 

 ocean, thus inducing a condition unfavorable to life in the lake waters. 



The turtle sliells mingled in profusion with the mammalian fossils, though exhibit- 

 ing in some respects the constitution of land turtles, in others partake of the character 

 of the emydeans, so as to suggest the probability of amphibious habits. 



It is remarkable that among the multitude of remains of mammals and turtles, 

 there are none of Crocodiles. Where were these creatures, when the shores of the 



