FUNCTION. 157 



geneous organs; since both singly and by their combinations, 

 do modified parts generate modified changes. Up to 



the highest organic types, this dependence continues mani- 

 fest ; and it may be traced not only under this most general 

 form, but also under the more special form, that in animals 

 haying one set of functions developed to more than usual 

 heterogeneity, there is a correspondingly heterogeneous ap- 

 paratus devoted to them. Thus among birds, which have 

 more varied locomotive powers than mammals, the limbs are 

 more widely differentiated ; while mammals, which rise to 

 more numerous and more involved adjustments of inner to 

 outer relations than birds, have more complex nervous 

 systems. 



§ 58. It is a generalization almost equally obvious with 

 the last, that functions, like structures, arise by progressive 

 differentiations. Just as an organ is first an indefinite rudi- 

 ment, having nothing but some most general characteristic 

 in common with the form it is ultimately to take ; so a 

 function begins as a kind of action that is like the kind of 

 action it will eventually become, only in a very vague way. 

 And in functional development, as in structural development, 

 the leading trait thus early manifested, is followed success- 

 ively by traits of less and less importance This holds 

 equally throughout the ascending grades of organisms, and 

 throughout the stages of each organism. Let us look at 

 cases : confining our attention to animals, in which func- 

 tional development is better displayed than in plants 



The first differentiation established, separates the two 

 ftmdamentally- opposed functions above named — the accumu- 

 lation of force and the expenditure of force. Passing over 

 the, [Protozoa among which, however, such tribes as present 

 fixed distributions of parts show us substantially the same 

 thing), and commencing with the lowest Coelenterata^ where 

 definite tissues make their first appearance, we observe that 

 the only marked functional distinction is between the endo- 



