THE METHOD OF CREATION OF ORGANIC FORMS. 175 



jects are furnished — in other words, for the causes of the origin of 

 the fittest. 



It has seemed to the author so clear from the first as to re- 

 quire no demonstration, that natural selection includes no ac- 

 tively progressive principle whatever ; that it must first wait for 

 the development of variation, and then, after securing the survival 

 of the best, wait again for the best to project its owm variations 

 for selection. In the question as to whether the latter are any 

 better or worse than the characters of the parent, natural selec- 

 tion m no wise concerns itself. 



I. ON THE LAW OF ACCELEKATION" A:N"D RETAEDATIO]!^-. 



There are two modes of demonstration of evolution, both de- 

 pending on direct observation. One of these has been success- 

 fully presented by Darwin. He has observed the origin of 

 varieties in animals and plants, either in the domesticated or 

 wild states, and has shown, what had been known to many, the 

 lack of distinction in the grades of difference which separate 

 varieties and species. But he has also pointed out that species 

 (such, so far, as distinctness goes) have been derived from other 

 species among domesticated animals, and he infers by induction 

 that other species, wiiose origin has not been observed, have also 

 descended from common parents. So far I believe his induction 

 to be justified ; but when from this basis evolution of divisions 

 defined by important structural characters, as genera, orders, 

 classes, etc., is inferred, I believe that we do not know enough of 

 the uniformity of Nature's processes in the premises to enable us 

 to regard this kind of proof as conclusive. 



I therefore appeal to another mode of proving it, and one 

 which covers the case of all the more really structural features of 

 animals and plants. 



It is well known that in both kingdoms, in a general way, 

 the young stages of the more perfect types are represented or imi- 

 tated with more or less exactitude by the adults of inferior ones. 

 But a true identity of these adults with the various stages of the 

 higher has, comparatively, rarely been observed. Let such a case 

 be supposed. 



In A we have four species whose growth attains a given 

 point, a certain number of stages having been passed prior to its 

 termination or maturity. In B we have another series of four 

 (the number a matter of no importance), which, during the period 



