168 Hoy a I Soc iety : — 



By suc-h division, we shall gain the advantage of having the Pari- 

 digitata arranged into two distinct Hnes of descent; every new 

 discovered form will at once have its place along one of the lines, 

 and the true pedigree of both v.dll be ascertained much sooner and 

 with <^reater accuracy. Whilst now making no such clear division, 

 pahcontologists, in projecting their genealogical tables, mix both 

 groups together ; and, according to the need of the moment, they 

 place forms belonging to one line of descent in the other, and vice 

 versa. Thus, for instance, all the Hyopotamoids and Anihraco- 

 therium are constantly moved about from one hne to the other *, 

 while their true place is along the line of Selenodont Paridigitata ; 

 and they have nothing to do with the Bunodont Suina, although 

 groups quite parallel with them may be found on the descending 

 line of Bunodont Paridigitata. 8uch parallelism, however, does 

 not imply direct links along parallels drawn across both diverging 



Divergence of Selenodont 

 and Bunodont Paridigitata. 



and descending lines ; the links are to be found only by climbing 

 along the ascending lines to near the point of separation. For 

 instance, Dicotyles and Hyomoschus occupy analogous positions ; 

 but there is no link between them along the dotted parallel. 

 Links will be found only by going up to the point near their 

 separation. 



There are, no doubt, to be found around the points of divergence 

 many forms of which it is difficult to say whether their teeth 

 are tubercular or crescentic, so thick are the lobes ; but once 

 this uncertain stage is passed, both groups keep unmistakably 

 distinct. 



Having once become con^inced that these two groups of crescent- 

 toothed and tubercular-toothed Paridigitata, after branching off 

 from a common progenitor in the early Eocene (perhaps the Creta- 

 ceous) period, followed diverging Hnes of descent, never mixing to- 

 gether, I tried to ascertain accurately, by such data as were fur- 



* In fact described constantly as Suina. See Gervais, ' Paleontologie de 

 France.' 



