38 GENERAL PRBLIMINAKT CONSIDERATIONS. 



forming one species into another ; for every subsequent change 

 in the conditions of existence would give rise to fresh changes, 

 of advance or retrogression, so that it would become impossible for 

 them to develope any further in one particular direction. Under 

 such a theory as this it would evidently be quite superfluous to 

 investigate the influence of outer circumstances on animals and 

 on their organs and mode of life. 



But this objection rests, as it seems to me, on the false 

 assumption that the external conditions are constantly and 

 rapidly altering, so that each variation caused by them is coun- 

 teracted at once by some antagonistic external influence. This 

 assumption is, as we know, in direct contradiction to the fact 

 that the external conditions in reality remain constant through 

 extraordinarily long secular periods. Thus the assumption would 

 seem well-founded, that animals might be acted upon merely 

 by the constant, uniform repetition of certain influences,' and 

 strongly enough affected to become capable of maintaining 

 the characters thus acquired, even when the external causes 

 which gave rise to them were removed by some fresh change. 

 Hence it is impossible in our enquiry to ignore the transform- 

 ing influence of the conditions of existence, merely in order to fall 

 in with a somewhat commonly accepted dogma : for it is only 

 by assuming that such effects are possible, and directing our 

 enquiries and experiments accordingly, that we shall be able to 

 arrive at any decision on the question whether such trans- 

 forming influences have played any part in the development of 

 animal types or not. 



After these somewhat long, but indispensable, general con- 

 siderations, I must briefly indicate the classification and arrange- 

 ment of the material which was best fitted to elucidate the 

 study of the action of external conditions on animal life. At 

 first sight it might seem that it would be well to distinguish 

 the Transforming from the Selective influences. Such a classifi- 

 cation would, however, involve us in many inconveniences. In 

 the first place the two divisions would be widely different in 

 extent ; for while we have no particularly rich store of experi- 

 mentally grounded facts, even with regai-d to the selective 

 influences of external conditions, with regard to direct trans- 



