274 PHV>10LOGICAL DKVEi.OPMEXT. 



have taken place. These must be passed over as arising 

 in wavs too involved to admit of rpecitic interpreta- 

 tions ; even supposing them to have been produced by 

 causes of the hind assigned. But the probability, or 

 rather indeed the certainty, is, that some of them have not 

 been so produced. Here, as in nearly all other cases, in- 

 diect requilibration has worked in aid of direct equilibration ; 

 and in many cases indirect equilibration has been the sole 

 agency. Besides ascribing to natural selection the rise of 

 various internal modifications of other classes than tliose 

 above treated, we must ascribe some even of these to natural 

 selection. It is so with the dense deposits which form 

 thorns and the shells of nuts: these cannot have resulted 

 from any inner reactions immediately called forth by outer 

 actions ; but must have resulted mediately through the effects 

 of such outer actions on the species. Let it be understood, 

 therefore, that the differentiations to which the foregoing 

 interpretation applies, are only those most conspicuous ones 

 which are directly related to the most conspicuous in- 

 cident forces. They must be taken as instances on the 

 strength of which we may conclude that other internal 

 differentiations have had a natural genesis, though in ways 

 that we cannot trace. 



