i894- NEUTER INSECTS AND DARWINISM. 283 



certain number of young males in a herd of deer were regularly 

 castrated, it could not be said that a caste of stags without antlers 

 had been evolved by natural selection or by phyletic metamorphosis. 



Other cases of retrogression mentioned by Weismann are the 

 reduction in the number of facets in the compound eyes and the 

 absence of wings. It is obvious that here again we must hold, until 

 the contrary is proved, that the organs fail to develop in consequence 

 of the insufficient nutrition of the larva, and that the eggs in respect 

 of their potentialities or " determinants " are all alike. It is well- 

 known that in true cases of phyletic degeneration there is in the 

 majority of instances actual retrogression in the course of individual 

 development, as, for example, in the cases of the teeth and hind 

 limbs of whales. No evidence has been produced that anything 

 of the kind occurs in the development of the neuters of the social 

 insects. 



A very important factor in the peculiar polymorphism of these 

 insects is the phenomenon of insect metamorphosis. The wings and 

 other important organs of the perfect insect are developed not con- 

 tinuously, but within a certain brief period, i.e., during the pupa 

 stage. There is good reason for the conclusion that the conditions to 

 which the larvae are exposed effect definite differences in the meta- 

 morphosis. It is known that temperature alone produces definite 

 •results in the colours of certain lepidoptera, that is to say, the 

 temperature to which the larvae are artificially exposed. Weismann 

 says that the changes in the thorax of the worker ant are just those 

 which would arise through transmission of the deteriorating effect of 

 disuse, but the workers are sterile. He ignores, what is much more 

 important, that the changes are just those which would arise from 

 incomplete development in the metamorphosis, and the worker 

 larvae are reared in conditions which are known to lead to such 

 incompleteness in the metamorphosis. 



Are we to be told, then, it will be asked, that the workers have 

 always been exactly as they are now ; have had their peculiar 

 differences fully expressed since the first moment when the solitary 

 ancestor began to be social ? To which I reply that the conditions 

 under which the worker larvae are reared may have been gradually 

 diverging from the conditions required for the production of the 

 perfect insect, and the constitutional peculiarities of the perfect insect 

 have probably undergone gradual modification, but the differences 

 between the worker and the perfect female have always been the 

 direct result of the differences in the conditions under which the 

 larvae were reared. There is no reason, so far as minus differences in 

 structure are concerned, to suppose that the females have acquired by 

 variation and selection the constitutional property of producing 

 workers of a certain kind. There is no reason to suppose that in 

 the beginning the eggs of the female ant could not be developed into 

 .wingless neuter ants as they are now. 



