lOO EVOLUTION, OLD AND NEW. 



the immutability of species, and indicates only the 

 doctrine of the permanence of essential features and the 

 variability of details (toutes les touches accessoires) ; 

 he repeats this eleven years later in his ' Epoques de la 

 Nature'" (published 1778).* 



But I think I can show that the passages which M. 

 Geoffrey brings forward, to prove that Buffon was in the 

 first instance a supporter of invariability, do not bear 

 him out in the deduction he has endeavoured to draw 

 from them. 



" What author," he asks, " has ever pronounced more 

 decidedly than Buffon in favour of the invariability of 

 species ? Where can we find a more decided expression 

 of opinion than the following ? 



" ' The different species of animals are separated from 

 one another by a space which Nature cannot overstep.' " 



On turning, however, to Buffon himself, I find the 

 passage to stand as follows : — 



" Although the different species of animals are sepa- 

 rated from one another by a space which Nature cannot 

 overstep — yet some of them a;p]proach so nearly to one 

 another in so many respects that there is only room enough 

 left for the getting in of a line of separation between 

 them"\ and on the following page he distinctly en- 

 courages the idea of the mutability of species in the 

 following passage : — 



" In place of regarding the ass as a degenerate horse, 

 there would be more reason in calling the horse a more 

 perfect kind of ass (un ane perfectionne), and the 



• ' Hist. Nat. G^a.,' torn. ii. p. 391, 1859. 

 t Tom. V. p. .'59, 1755. 



