Summary and General Conclusions 461 



this idea of many individuals varying at the same time, and 

 in the same direction, at once involves the conception that 

 evolution moves forward by some force residing in the organ- 

 ism, driving it forwards or backwards. Instability comes, 

 perhaps, nearer to expressing this idea than any other term, 

 and yet to evolve from a protozoan to a man implies the idea 

 of something more than simple unstableness. 



The idea that Weismann has touched upon in this connec- 

 tion, namely, that the survival of a given form determines the 

 future course of evolution for that form, is very plausible, and 

 also fits in well with the results of our experience in the field 

 of the inheritance of variations. We see new variations or mu- 

 tations departing in some or in many characters from the 

 original type, apparently by new combinations or perturba- 

 tions of those already present. We never expect to see a 

 bird emerge from the egg of an alligator. Thus it appears 

 that by the survival of certain forms the future course of 

 evolution is determined in so far as the new types of muta- 

 tion are thereby limited. Weismann means, however, that 

 in this way new plus or minus steps will be indefinitely deter- 

 mined amongst the new fluctuating variations, but this state- 

 ment is contradicted by our experience of the results of 

 artificial selection. The upper limit does not keep on pushing 

 out indefinitely in the direction determined by the first selec- 

 tion, but is soon brought to a standstill. So that, as far as 

 Weismann's hypothesis is concerned, the idea appears to have 

 no special value. On the other hand, this idea may be fruitful 

 if applied to mutations, but here unfortunately we have not 

 sufficient experience to guide us, and we do not know defi- 

 nitely whether a new character that appears as a mutation 

 will be more likely, in subsequent mutations, to go on increas- 

 ing in some of the descendants. Thus, while the mutation 

 theory must assume that some new characters will go on 

 heaping up, we lack the experimental evidence to show that 

 this really occurs. It would be also equally important to 



