10 Journal New York Entomological Society. [Vol. xm. 



presupposes that the ancestors of the winged insects were aquatic 

 forms and that still farther back in the line of descent the pregenitors 

 of these aquatic forms must have been terrestrial, in order to have 

 evolved a tracheal system in the body, from which later, tracheal gills 

 could be developed ; as otherwise, if gills had been developed at all 

 by these primitive insects they would certainly have been blood gills. 

 But all the evidence points strongly to the supposition that the 

 ancestors of the winged insects were terrestrial. It has been shown 

 by Packard (1898) that tracheal gills are adaptive, secondary, tem- 

 porary larval structures that do not persist in the adult and "are 

 not ancestral, primitive structures." Tower also rightly objects that 

 there is no resemblance between the fundamental type of wing vena- 

 tion, as established by Comstock and Needham, and the tracheation 

 of any known tracheal gill, which should certainly be the case if the 

 wings had been developed from any such structure. 



The second theory and the one to which it seems to me that all 

 the present evidence points strongly is the one put forward by Graber 

 (1867) and Muller (1875) and strongly supported by Pancritius 

 (1884) and Packard (1898), that the wings have arisen as simple 

 outgrowths or evaginations of the integument at the suture between the 

 dorsum and the pleurum. After working on. the development of the 

 wings in the Termites, Muller declared that the wings of insects have 

 been derived from lateral continuatuons of the dorsal plates of their 

 respective segments. 



I have shown that in the Coleoptera the wing arises on the 

 pleurum, at or near the future position of the dorsb-lateral suture, as 

 a thickening of the body wall, which, in the simplest type, begins as 

 a simple pushing outward and downward of this thickening (Fig. 4). 

 In another paper I shall show exactly the same process in the begin- 

 ning wing of the Neuropteron, Raphidia sp. the wing developing as an 

 outpushing of the thickened hypoderm at the lateral fold. In certain 

 Hemiptera, sections of the early stages of which I have examined, it 

 was evident that the wing developed as a simple pushing out and fold- 

 ing of the body wall near the lateral suture. 



All this is in line with the conclusion of Packard (1898) that 

 " the wings are essentially simple dcrsal outgrowths of the integument, 

 being evaginations of the hypodermis," and of Comstock and Need- 

 ham (1899) that the " wings arise as sack-like folds of the body wall 

 at the point where the suture between the tergum and the pleurum 



