TRANSFORMISM 265 



becomes a hereditary one, but so far from being inconceivable 

 was such a conclusion that, until the time of Weismann, it was 

 commonly held. In pure natural history the transmission of 

 acquirements seems to be the most obvious way of explaining 

 many cases of transformism — the adoption of a gasteropod shell 

 as a shelter by the Hermit-crab, for instance. Of course so great 

 is the generality of the logic of natural selection that such cases 

 are equally well explained (logically) by the latter hypothesis. 



If we admit that acquirements may be slowly transmitted by 

 heredity it becomes easy to formulate the corresponding hypothesis 

 of transformism. The increased use of some bodily part reacts on 

 structure so that bodily proportions become changed — in such a 

 way we might explain (logically) the gradual increase in size and 

 efficiency of the 3rd digit in the feet of the Tertiary horses ; the 

 practical elimination of the other digits and so the evolution of 

 the Equidae. Or the occasional swimming of some terrestrial 

 mammals may be thought about as becoming a habit, reacquired 

 by imitation, generation after generation and reacting on structure 

 until the forms of the limbs became so changed as to lead to the 

 evolution of flippers adapted for locomotion in water. Obviously 

 the logical argument is a very plausible one. 



87^. The Evidence for Lamarckism. Some things are 

 demonstrable, (i) The general validity of the method of trial 

 and error : observation of organisms living in the wild and 

 laboratory experiments prove this. (2) The passage of an 

 individually acquired activity into a habitual one. (3) The auto- 

 matism of habit and the performance of learned, skilled activities 

 without consciousness. (4) The effects of use and disuse of organs 

 and bodily parts in producing structural and functional changes. 

 There remains the all-important demonstration — that such indivi- 

 dual acquirements may reappear, not as renewed acquirements 

 but as hereditary qualities, in the progeny. 



Here we must make appeal to facts. And it has to be admitted 

 at once that the appeal is inconclusive. Logically the hypothesis 

 of natural selection is sound, but we find it impossible now to show 

 that the naturally occurring variations by fluctuation on which 

 Darwin and Wallace built their case are changes that reappear in 

 the progeny — or are transmitted by heredity. Nor can we show 

 that mutations occur so frequently as to constitute such an 

 abundant material for selection that transformism must result. 



