172 HINTS ON THE ANCESTRY OF INSECTS. . 



two simple tubes with few branches, while there are no stig 

 mata, or breathing holes, to be seen in the sides of the body. 

 This fact sustains the view of Gegenbaur* that at first the tra 

 cheae formed two simple tubes in the body- cavity, and that the 

 primary office of these tubes was for lightening the body, and 

 that their function as respiratory tubes was a secondary one. 

 The aquatic Protoleptus, as we may term the ancestor of Lep- 

 tus, may have had such tubes as these, which acted like the 

 swimming bladder of fishes for lightening the body, as suggested 

 by Gegenbaur. It is known that the swimming bladder of fishes 

 becomes developed into the lungs of air-breathing vertebrates 

 and man himself. As our Leptus adopted a terrestrial life and 

 needed more air, a connection was probably formed by a minute 

 branch on each side of the body with some minute pore (for 

 such exist, whose uses are as yet unknown) through the skin, 

 which finally became specialized into a stigma, or breathing 

 pore : and from the tracheal system being closed, we now have 

 the open tracheal system of laud insects. 



The next inquiry is as to the origin of the wings. Here the 

 question arises if wingless forms are exceptional among the 

 winged insects, and the loss of wings is obviously dependent 

 on the habits (as in the lice), and environment of the species 

 (as in beetles living on islands, which are apt to lose the hinder 

 pair of wings), why may not their acquisition in the first place 

 have been due to external agencies ; and, as they are suddenly 

 discarded, why may they not have suddenly appeared in the first 

 place ? In aquatic larvae there are often external gill-like organs, 

 being simple sacs permeated by tracheae (as in Agrion, Fig. 129, 

 or the May flies). These organs are virtually aquatic wings, 

 aiding the insect in progression as well as in aerating the blood, 

 as in the true wings. They are very variable in position, some 

 being developed at the extremity of the abdomen, as in Agrion, 

 or along the sides, as in the May flies, or filiform and arranged 

 in tufts on the under side of the body, as in Perla; and the natu 

 ralist is not surprised to find them absent or present in accord 

 ance with the varying habits of the animal. For example, in the 

 larvae of the larger Dragon flies (Libellula, etc.) they are want 

 ing, while in Agrion and its allies they are present. 



*Vergleichende Anatomie, 2te Auflage, 1870, p. 437. I should, however, here add 

 that I am told by Mr. Putnam that some fishes which have no swim-bladder, are 

 surface-swimmers, and vice versa. 



