15 



27 



26 



8 



26 INTRODUCTORY REMARKS. 



relations of height to length aud breadth, and also varying in anatomical details, so that the author 

 was at first led to suppose they indicated five different species. Since describing these in the "Ancient 

 Fauna of Nebraska," the examination of many other specimens has led him to view them all as 

 pertaining to a single species. This appears to have been a land turtle, for which the name of Testudo, 

 or Stylemys Nebrascensis, is retained. 



The almost perfect shell of an adult specimen, in the Museum of the Academy, obtained by Dr. 

 Hayden in the expedition of 1866, has the following measurements : 



Length of sternum, . . . . . - . . .20 inches. 



Breadth " ...... 



Length of antero-posterior curve of carapace, 



Length of transverse " "... 



Height of carapace above level of sternum, 



The collection of fossils from the Niobrara River in quantity forms but a comparatively small 

 portion of those forming the material of the succeeding pages. It was exclusively obtained by Dr. 

 Hayden in the only known visit made to the locality, in 1857. The fossils were for the most part 

 picked up during the expedition, from the loose sands of the locality. They are remarkably well 

 preserved, and are devoid of adherent matrix. The bones are not fissured, nor are they generally 

 water-rolled, but usually retain their original freshness of shape. A few, however, are water-rolled, 

 and many small fragments of teeth also present this appearance. 



Most of the Niobrara fossils are like those of the Mauvaises Terres, completely petrified, but this 

 concerns the structure of the bone and dental tissues alone. The interstices of the spongy substance of 

 the bones, the medullary cavities, and the cavities aud depressions of the skulls and teeth, are devoid 

 of adherent matrix or infiltrated mineral matter. Other of the fossils appear only partially petrified, 

 and some appear to have undergone but little change except in losing some of the bone cartilage. It 

 is even uncertain that a few of the bones viewed as fossils may not have belonged to recent animals. 

 Many of the fossils contained within their cavities loose quartzose sand, like the ordinary white quartz- 

 ose sand of certain river and sea shores. 



The collection of Niobrara fossils is almost equally devoid of any other than mammalian remains, 

 and those of a species of turtle. A couple of nearly uncharacteristic fragments of bird bones are 

 suspected to have belonged to the recent Sand-hill Crane. The remains of turtles, from a want of 

 adherent matrix to hold the parts together, consist of a multitude of fragments of the shells, of all ages. 

 They perhaps indicate a species distinct from that of the Mauvaises Terres, and the author in this view 

 has named it Testudo or Stylemys Niohrarensis. It is certainly very like the Mauvaises Terres Testudo, 

 and probably was the unchanged successor of the latter, and therefore not a distinct species. 



The few remaining fossils described in the body of the work, from Texas and California, in general 

 character bear a close resemblance to those of the Mauvaises Terres. 



Of fossil invertebrata only a single species of moUusk was found in association with the fossil bones 

 in the several localities mentioned. This consists of the casts of a land snail, occurring abundantly in 

 the Mauvaises Terres, and described by Messrs. Meek and Hayden under the name of Helix Leidyi. 



Among the fossil mammalian remains of the tertiary formations of Dakota and Nebraska thus far 

 brought to our notice, there are no traces of Primates, neither of Man nor Monkeys ; none of Bats, 

 Edentates, Marsupials, Monotremes, Seals, Sirenians nor Cetaceans. The other orders are well 

 represented, especially the Carnivoi-a, Pachyderms and Ruminants. The Solidungulates are better 

 represented than in the recent fiiunse of any part of the world, and indeed the pliocene deposit of the 



