xin EVOLUTION 223 



necessitate the placing of the ancestral forms in a 

 different species from their descendants at the present 

 day. And in the same way, if we could trace back the 

 species of any one genus, we should find them gradually 

 approach one another in structure until they finally 

 converged in a single species, differing from those now 

 existing, but standing to ail in a true parental relation. 



It will be seen that, on this hypothesis, the relative 

 likeness and unlikeness of the various species of frogs are 

 explained as the result of their descent with greater 

 or less modification or divergence of character from the 

 ancestral form : and that we get an arrangement or 

 classification in* the form of a genealogical tree, which, 

 on this hypothesis, is a strictly natural one, since it 

 shows accurately the relationship of the various species 

 to one another and to the parent stock. So that on the 

 theory of evolution, a natural classification of any given 

 group of allied organisms is simply a genealogical tree. 



Now it is evident that the only way in which we 

 could be perfectly sure of an absolutely natural classi- 

 fication of the species of any kind of animal the frog, 

 for example would be by obtaining specimens as far 

 back as the distant period when the genus first came 

 into existence. 



Forming part of the solid crust of the earth are a 

 series of sedimentary or stratfad rocks, and the researches 

 of geologists have shown that these present a general 

 order of succession, the lowest, when undisturbed, 

 being in every case older than the more superficial 

 layers. Imbedded in these rocks are found the remains 

 of various extinct animals in the form of what are called 

 fossils ; and it might perhaps, on first considering the 

 subject, be supposed that, had a process of evolution 

 taken place, we ought to be able to find in the rocks 



