190 OUTLINES OF EVOLUTIONARY BIOLOGY 



selection, the welfare of the insect community as a whole, rather 

 than that of its constituent individuals, being in this case the 

 determining condition through which natural selection operates. 

 The case of such communities, however, is exactly analogous to 

 that of individuals, except that the single cells are replaced by 

 complete and separate multicellular units. The whole community 

 may be looked upon as one individual of a higher order, and the 

 problem of the transference of stimuli from the bodies of the 

 neuters to the germ cells of the perfect insects differs in no 

 essential respect from that of the transference of similar stimuli 

 from somatic cells to germ cells in an ordinary individual. The 

 close association in which members of such communities live 

 may be supposed to facilitate the transfer of such stimuli (arising 

 from modification of the somatic cells of the neuters in response to 

 the environment) to the germ cells of the males and perfect 

 females, without the aid of material conductors. 



In the present state of our knowledge, of course, any suggestions 

 which may be put forward on this subject must be regarded as mere 

 hypotheses, incapable of demonstration, and as such they will 

 doubtless appear to many people to be unwarrantable. If, however, 

 they serve to show that there is no a priori reason for denying the 

 possibility of the transmission of somatogenic characters to the 

 germ cells, and thence to future generations, they will serve a 

 useful purpose. 



We must, however, again point out that, under normal cir- 

 cumstances, it probably takes many generations before any 

 impression produced by modification of the soma upon the germ 

 cells becomes so deeply ingrained as to find full expression in the 

 offspring produced by their development. The impression, once 

 made, appears to be equally difficult to remove, and hence has 

 probably arisen that conservatism or inertia of the germ cells to 

 which we have already alluded. Occasionally, however, it appears 

 that a suddenly acquired somatogenic character at once makes a 

 deep impression upon the germ cells. Why this should be so we do 

 not know, but it is perhaps no more remarkable than the well 

 known fact that the germ cells themselves may occasionally throw 

 aside their conservatism and give rise to sports or mutations. 1 



It is well known that, apart altogether from the highly 

 specialized brain cells, living protoplasm frequently has the power 



1 It is possible, however, that the suddenness of a mutation is apparent rather 

 than real, at any rate in some cases ; compare p. 106. 



