in Buiierfies of the Ce?;?/* Precis. 39 



hvana^ and all Lis experiments clearly show that the climatic 

 is the primary, and the selective the secondary cause ; that 

 is to say, the tendency to variation in certain directions was 

 initiated by climatic agencies, and subsequently utilized and 

 developed to a more or less high degree by natural selection 

 for protective purposes. In cases of adaptive seasonal dimor- 

 phism selection can have only come into play in order to 

 enable a given species to adapt itself to the varying conditions 

 at two diiferent seasons ; but if these climatic changes were 

 eliminated, it seems evident that natural selection alone 

 would be incapable of producing seasonal dimorphism as we 

 now see it. It is true that in certain cases Prof. Weismann 

 found that certain pupfe out of a number experimented on 

 produced the form that would have appeared in nature in 

 spite of artificial temperatures, and it is on these exceptions 

 that he founded the above ]n-opositions. But the more we 

 investigate the matter the more clear it becomes that the 

 actual cause which induces the change in a well-marked case 

 of dimorphism is of a highly complex character: the vastly 

 preponderating cause is either temperature or humidity, or 

 both, as the case may be ; but there is something beyond 

 this of which we as yet know nothing. This is evidenced by 

 such cases as that I have already quoted, where two larvae 

 reared from the egg under precisely similar conditions pro- 

 duced the full wet and full dry forms respectively ; and not 

 a few similar instances have come under my notice. 1 am 

 not therefore inclined to attach very great importance to the 

 exceptions found by Prof. Weismann in his temperature- 

 experiments, for I doubt the possibility of artificially pro- 

 ducing all the factors necessary to induce the transition from 

 one seasonal form to another in every case j and, moreover, I 

 am of opinion that the personal equation, if one may apply 

 such a term to insects, must be taken into account. 



We can, as a general rule, obtain a very fair idea as to 

 how far natural selection has taken part in the production of 

 any given case of seasonal dimorphism by taking into con- 

 sideration the amount of difference between the two extreme 

 ibrms and the rapidity of transition trom one to anoilier, for 

 the more highly differentiated they are and the more abrupt 

 the transition the greater has been the influence of selection. 

 Indeed one could almost trace a scale of development from a 

 species like Acraa unemosa, Hew., where the slight change 

 in the black markings is probably due entirely to climatic 

 causes; through the Teracoli, whose seasonal differences are 

 part climatic, part selective; and culminating in Precis sesamus, 



