146 PROC. ENT. SOC. WASH., VOL. 22, NO. 6, JUNE, 1920 



early development of polymorphism. However, the highly 

 specialized different types of colonizing reproductive forms (with 

 the exception of the winged first form) and the "intermediates" 

 very probably have been evolved comparatively recently in geo- 

 logical history since the Tertiary period. 



The three main evolutionary factors are Lamarck's factor of 

 the origin of variations through environmental action; Darwin's 

 the survival of chance variations, which are adapted to the en- 

 vironment; the DeVries' chance variations, i. e., mutations, which 

 may or may not be useful, i. e., adapted to the environment. 



A morphological study of termites indicates to the writer that 

 certain structural peculiarities in case of the workers and soldiers, 

 which appear to indicate adaptation to environment and as the 

 result of use, are rather to be explained as chance variations \vhich 

 have survived through adaptation to environment, (Snyder, 

 1919). 1 The workers of certain species of subterranean termites 

 living in the hard dry soil of the semi-arid southwestern states 

 have the prothoracic tibiae enlarged, i. e., subfossorial, and the 

 enlarged legs might appear to be the result of use or response 

 to environment. However, other termites living under the same 

 conditions, have not the enlarged tibiae. 



The highly specialized soldier caste is in certain species often 

 of little apparent use to the colony life. There is great variation 

 in the size and shape of both the mandibulate soldiers and the 

 nasuti. In the case of species in the genus Capritermes Wasmann, 

 the peculiar twisted shape of the mandibles of the soldier caste 

 can certainly not be explained as a useful adaptation; the man- 

 dibles must be almost useless 'for the purpose of defense against 

 marauding enemies of the colony. 



It appears to the writer that the origin of the castes of termites 

 can only be explained by DeVries' mutation theory. It will be 

 remembered that mutations are chiefly of two kinds progressive 

 and regressive ; progressive mutations possess new characters not 

 present in the parents and regressive mutations are those which 

 have lost some of the parental characters. 



In termites, in addition to the worker, soldier, the three stable 

 reproductive types, and the "intermediate" reproductive types, 

 there are other forms which may be considered to be mutants. 

 I have already discussed these forms in a previous paper, (Banks 

 and Snyder, 1920) : 



"Certain abnormalities have been noted, (Snyder, 1915) in the metamor- 

 phosis of the nymphs of the first form to the winged adult. Individuals 



1 1919. Snyder, T. E. Some significant structural modifications in Nearc- 

 tic termites. Proc. Ent. Soc. Wash., Vol. 21, No. 5, May. 



