262 president's address. — section d. 



always, small and scarcely noticeable variations, the fluctuating or 

 individual variations with which we have been long familiar. If the 

 inheritance of characteristics due to mutilations becomes conclusively 

 proved, or if the cyclopean larva of Fundulus, for example, is brought 

 to maturity and is found to transmit its peculiarities, then we shall 

 have to allow more than small variations as the result of somatogenic 

 modification. 



With regard to modification, whether of the soma or of the germ 

 plasm, by direct extra-somatic influence, we notice that it is by no means 

 always adaptive. When the germ cells are modified directly, quite 

 different modifications may arise as the result of the same stimulus. 

 Thus even when both parents are subjected to the same abnormal 

 •nvironment, some of the offspring may present one modification, 

 some another, while many may present no modification that we can 

 detect. Further, quite different stimuli may be followed by the sam» 

 modification. The response is determined by the cells, not bj the 

 stimulus. It is in each case the response that that particular cell 

 has learnt to make whether to that stimulus or to another, and the 

 learning can be only by the system of trial and error. In like manner 

 when there is environmental influence that leads to the modification, 

 of some individuals, others may remain unmodified. It is true that 

 direct modifications of the soma are more often adaptive than are 

 modifications of the soma due to inheritance. But we have many 

 instances of non-adaptive modifications, as for example, the one-eyed 

 fishes or the modified Echinoderm larvae developing under the influence 

 of solutions of mineral salts. It is evident that there are subsidiary 

 problems that may prove very difficult to solve. 



On Theories of Heredity. 



Next to the work of Darwin himself that of Weismann is probably 

 the most stimulating and fruitful of biological work in the great century 

 in which Science awoke. We may not accept the view that a continu- 

 ing germ plasm, subject to modification by its environment, is proof 

 against all modification from that particular part of its environment 

 constituted by the soma. We must, none the less, admit that the 

 enunciation of that view has led to an immense amount of research 

 of a most valuable kind. We may not find ourselves able to accept 

 without question Weismann's description of the architecture of the 

 germ plasm, may not even agree with him in regarding the germ plasm 

 as limited to the chromatin matter. None the less we must admit that 

 that description has given us clearer ideas than would have been 

 possible otherwise, and a terminology most useful for describing 

 hereditary processes. It has, moreover, stimulated research to an 

 extraordinary degree. We owe much of our knowledge of minute cell 



