8zo THE POPULAR SCIENCE MONTHLY. 



coveries is that lie has finally succeeded in finding among this 

 mass of bones one bone, at least, which bears unmistakable traces 

 of having been connected with a humerus, the head of which 

 must have been as substantial as in cassowaries. He thus con- 

 siders it proved that the DinomithidcB, like the kiwis, descended 

 from birds which could fly.* The last missing link is thus dis- 

 covered, and the chief points in the genealogy of birds are thus 

 already settled, while many a gap which still remains will cer- 

 tainly be filled up when the rich materials recently excavated in 

 both Americas have been carefully examined by anatomists. 



The same may be also said in regard to mammals, if the recent 

 discoveries in North and South America are taken into account. 

 The earliest traces of mammals have been found, as is known, in 

 the Triassic deposits of Germany, Basutoland, the Cape Colony, 

 and North Carolina ; and it is also known, through the previous 

 remarkable works of Professors O. C. Marsh and H. F. Osborn, 

 that the Jurassic deposits of Wyoming have yielded a rich fauna, 

 among which we find the remote ancestors of various orders of 

 the present mammalia. f But the most important finds, which 

 throw a new light both on the earlier and the subsequent forms, 

 have been made in that immense area of lacustrine beds which 

 have been deposited in the region of the great salt lakes of Utah, 

 Wyoming, and Colorado, from the end of the Cretaceous period 

 down to the middle parts of the Tertiary epoch. There, and espe- 

 cially in the Eocene " Puerco " and " Wahsatch " beds, as well as 

 in the Eocene " Uinta " formation, a rich fauna of mammals has 

 been unearthed. J All those Eocene mammals had something in 

 common in their leading features, and yet they offered a sufficient 

 diversity for being considered as the probable ancestors of nearly 

 all orders of placental mammals. To mention their feet only, 

 they were adapted, in all of them, for walking upon the sole, and 

 were provided with five toes ; but it is easy to recognize in the 

 structure of the feet of the different genuses such divergences as 

 necessarily ought to evolve, under certain conditions on the one 

 side, the plantigrade foot of the bears, and, on the other side, the 

 digitigrade foot of the Ungulata (horses, camels, elephants, and 

 so on), who walk upon the points of their toes ; and, again, among 



* Nature, 1892, vol. xlv, p. 257. 



f 0. C. Marsh, in American Journal of Science, 1888 to 1S91 ; H. F. Osborn, The 

 Structure and Classification of Mesozoic Mammalia, in Journal of the Academy of Natural 

 Science of Philadelphia, vol. ix ; R. Lydekker, Catalogue of the Fossil Mammalia in the 

 British Museum, London, 1891. 



\ Cope's Synopsis of the Vertebrate Fauna of the Puerco Series, and W. Scott and H. F. 

 Osborn, The Mammalia of the Uinta Formation, in Transactions of the American Philo- 

 sophical Society, new series, vol. xvi, Parts II and III, Philadelphia, 1889. Also R. Lydek- 

 ker's paper in Nature, vol. xliii, p. Ill ; and Phases of Animal Life. 



