536 W. L. TOWER, 



3. Significance of the Types of Wing DcTelopment. 



In the preceding section 1 have shown that the wings of Coleo- 

 ptera develop during larval life in one of three ways, and that the 

 beginning and the result of the process is in every case the same; 

 the formation of the adult wings which have essentially the same struc- 

 ture. This interesting variation in the intermediate stages of the same 

 organ in the same group is probably due to two causes, variety of 

 larval habits and habitat, and rapidity of development. 



In the species studied it was evident that the prevailing type of 

 wing development is a simple one where the wings develop directly 

 beneath the cuticula without the formation of even a trace of a wing 

 sac. In all of the beetles which show this type of development the 

 larval period is long, in some cases {Scarabaeidae and Cerambycidae) 

 being two, three, or even more years in duration. The Curculionidae 

 are, however, an exception to this rule, but in this family the wings do 

 not appear until the last instar, when there is no need of a wing 

 sac. In most of the species studied, which fall into this class as 

 regards wing development, the maximum growth is attained in early 

 winter and the larva then lies dormant until warm weather comes again. 

 The last larval moult preceding the final one occurs in late autumn, 

 and soon after this the wing discs are found as slightly thickened areas 

 upon the sides of the thoracic segments. They remain in this simple 

 state until the following spring, when, with the increase of temperature, 

 development begins again. 



This type of wing development forms, I believe, a connecting hnk 

 between the simpler condition of the Heterometabola and the more 

 complicated one of the Holometabola. As I was unable to find any 

 satisfactory account of the origin and development of the wings of 

 Heterometabolic insects, I have made for purposes of comparison, a 

 careful study of the rise and development of the wings in some 

 Hemiptera and Orthoptera, but I shall present here only such data 

 as are needed for the sake of comparison, reserving the rest of my 

 observations for a later communication. 



In Anasa tristis de G. (Hemiptera) the wings are present as 

 disc-like thickenings of the pleural hypodermis of the last two thoracic 

 segments. These are present in the youngest larvae and probably 

 arise in the embryo. Each disc is supplied with a small tracheal 

 branch from the longitudinal tracheal trunk, and is in exactly the same 

 position upon the thorax of the larva as are the discs in Coleopter^ 



