144 TEXT-BOOK OF ENTOMOLOGY 



lateral continuations of the dorsal plates of the body-segments with 

 which they are connected." 



Now, speculating on the primary origin of wings, we need not suppose that 

 they originated in any aquatic form, but in some ancestral land insect related 

 to existing cockroaches and Termes. We may imagine that the tergites (or 

 notum) of the two hinder segments of the thorax grew out laterally in some 

 leaping and running insect ; that the expansion became of use in aiding to sup- 

 port the body in its longer leaps, somewhat as the lateral expansions of the 

 body aid the flying squirrel or certain lizards in supporting the body during 

 their leaps. By natural selection these structures would be transmitted in an 

 improved condition until they became flexible, i.e. attached by a rude hinge- 

 joint to the tergal plates of the meso- and metathorax. Then by continued use 

 and attempts at flight they would grow larger, until they would become perma- 

 nent organs, though still rudimentary, as in many existing Orthoptera, such as 

 certain Blattarise and Pezotettix. By this time a fold or hinge having been 

 established, small chitinous pieces enclosed in membrane would appear, until 

 we should have a hinge flexible enough to allow the wing to be folded on the 

 back, and also to have a flapping motion. A stray tracheal twig would naturally 

 press or grow into the base of the new structure. After the trachea running 

 towards the base of the wing had begun to send off branches into the rudimen- 

 tary structure, the number and direction of the future veins would become 

 determined on simple mechanical principles. The rudimentary structures beat- 

 ing the air would need to be strengthened on the front or costal edge. Here, 

 then, would be developed the larger number of main veins, two or three close 

 together, and parallel. These would be the costal, subcostal, and median veins. 

 They would throw out branches to strengthen the costal edge, while the branches 

 sent out to the outer and hinder edges of the wings might be less numerous and 

 farther apart. The net-veined wings of Orthoptera and Pseudoneuroptera, as 

 compared with the wings of Hymenoptera, show that the wings of net-veined 

 insects were largely used for respiration as well as for flight, while in beetles 

 and bees the leading function is flight, that of respiration being quite subordi- 

 nate. The blood would then supply the parts, and thus respiration or aeration 

 of the blood would be demanded. As soon as such expansions would be of 

 even slight use to the insect as breathing organs, the question as to their perma- 

 nency would be settled. Organs so useful both for flight and aeration of the 

 blood would be still further developed, until they would become permanent 

 structures, genuine wings. They would thus be readily transmitted, and being 

 of more use in adult life during the season of reproduction, they -would be still 

 further developed, and thus those insects which could fly the best, i.e. which 

 had the strongest wings, would be most successful in the struggle for existence. 

 Thus also, not being so much needed in larval life before the reproductive 

 organs are developed, they would not be transmitted except in a very rudi- 

 mentary way, as perhaps masses of internal indifferent cells (imaginal discs), 

 to the larva, being the rather destined to develop late in larval and in pupal life. 

 Thus the development of the wings and of the generative organs would go hand 

 in hand, and become organs of adult life. 1 



The development and structure of the tracheae and veins of the wing. 

 The so-called veins ("nervures") originate from tine tracheal 



1 Reproduced from the author's remarks in Third Report U. S. Eut. Commission, 

 pp. 2G8-271, 1883. 



