Septembeb 10, 1909] 



SCIENCE 



329 



better than our predecessors in unraveling 

 real genealogies. At least Professor De- 

 peret's genealogical table of the Lower Ter- 

 tiary pig-like Anthracotheriida;, which he 

 publishes as an illustration of "evolution 

 reelle," seems to me to be no more exact 

 than several tables of other groups by 

 previous authors which he criticizes. His 

 materials are all fragmentary, chiefly jaws 

 and portions of skulls; they were obtained 

 from several isolated lake-deposits, of 

 which the relative age can not be deter- 

 mined by observing the geological super- 

 position ; and they represent a group which 

 is known to have lived over a large part of 

 Europe, Asia, northern Africa and North 

 America. There is therefore no certainty 

 that the genera and species enumerated by 

 Professor Deperet actually originated one 

 from the other in the region where he hap.- 

 pened to find them; he has demonstrated 

 the general trend of certain changes in the 

 Anthracotheriidffl during geological time, 

 but really nothing more. 



Even when a group of animals seems to 

 to have been confined to one comparatively 

 small region, where the series is not com- 

 plicated by migration to and from other 

 parts of the world, modern research still 

 emphasizes the difficulty of tracing real 

 lines of descent. The primitive homed 

 hoofed animals of the family Titano- 

 theriidfe, for example, are only known 

 from pai't of North America, and they seem 

 to have originated and remained there 

 until the end. As their fossil skeletons are 

 abundant and well preserved, it ought to be 

 easy to discover the exact connections of 

 the several genera and species. Professor 

 Osborn has now proved, however, that the 

 Titanotheres must have evolved in at least 

 four distinct lines, adapted "for different 

 local habitat, different modes of feeding, 

 fighting, locomotion, etc., which took origin, 

 in part at least, in the Middle or Upper 

 Eocene." They exhibit "four distinct 



types in the shape and position of the 

 horns, correlated with the structure of the 

 nasals and frontals, and indicative of dif- 

 ferent modes of combat among the males." 

 The ramifications of the gi'oup are indeed 

 so numerous that the possibility of follow- 

 ing chains of ancestors begins to appear 

 nearly hopeless. 



Among early reptiles the same difficul- 

 ties are continually multiplied by the 

 progress of discovery. About twenty years 

 ago it began to appear likely that "we 

 should soon find the terrestrial ancestors of 

 the Ichthyosauria in the Trias; and some- 

 what later a specimen from California 

 raised hopes of obtaining them by system- 

 atic explorations in that region. During 

 more recent years Professor J. C. Merriam 

 and his colleagues have actually made these 

 explorations, and the result is that we now 

 know from the Californian Trias a multi- 

 tude of reptiles, which need more explana- 

 tion than the Ichthyosauria themselves. 

 Professor Merriam has found some of the 

 links predicted between Ichthyosaurs and 

 primitive land reptiles, but he has by no 

 means reached the beginning of the marine 

 group ; and while making these discoveries 

 he has added greatly to the complication 

 of the problem which he set out to solve. 



Serious difficulties have also become ap- 

 pai'ent during recent years in determining 

 exactly the origin of the mammals. For a 

 long time after the discovery of the Ano- 

 modont or Theromorph reptiles in the 

 Permian-Trias of South Africa, it seemed 

 more and more probable that the mammals 

 arose in that region. Even yet new reptiles 

 from the Karoo formation are continually 

 being described as making an astonish- 

 ingly near approach to mammals; and, so 

 far as the skeleton is concerned, the links 

 between the two grades are now very nu- 

 merous among South African fossils. 

 Since these reptiles first attracted atten- 

 tion, however, they have gradually been 



