376 J. D. Tothill — The Ancestry of Insects. 



Examples could be multiplied but sufficient lias been said to 

 indicate that the Apterygogenea occasionally preserve even 

 in tbe adult condition primitive ' ancestral ' characters, such 

 as the tritocerebral and abdominal appendages, structures 

 that have almost or completely disappeared in the Ptery- 

 gogenea. In this sense they are more generalized than any 

 other living insects. Enough has also been said to indicate on 

 the other hand that the Apterygogenea are highly specialized 

 animals as indicated by the frequent reduction of mouthparts, 

 visual organs, tracheae, etc. ; and by the development of pecu- 

 liar structures such as the caudal "spring" and the collophore. 

 Handlirsch points out that the absence of wings does not nec- 

 essarily represent a primitive condition, but that these insects 

 may have once possessed wings and lost them as a result of 

 taking up their abode in the peculiar environment in which 

 they now live. In cave insects, such as ants, and in parasitic 

 insects (especially of fur- or hair-bearing vertebrates), such as 

 the fleas and lice, reduction or loss of wings is extremely com- 

 mon. Be this as it may, the various specializations found in 

 the group seem to make it certain that it is not in the direct 

 line of descent of the great Pterygogenea complex. 



Turning to the Pterygogenea, the remarkable findings of 

 Handlirsch from the Pennsylvanian rocks engendered a new 

 point of view concerning the most generalized insect. Steno- 

 dictya (fig. 2) is typical of these ancient insects and I will con- 

 fine my remarks to it. It is a large insect with well-differ- 

 entiated head, thorax and abdomen ; with all abdominal 

 segments and with well-developed cerci. The wings are of 

 peculiar interest in that the venation is practically that of the 

 "hypothetical wing" long since suggested by Comstock and 

 Iseedham. The organs of flight are also of interest in that 

 a small pair is developed on the prothorax and because all 

 the abdominal segments show lateral wing-like outpushings. 

 This seems to indicate the method of origin of wings and does 

 away with the necessity of deriving the winged insects from 

 the Apterygogenea. 



In a word the Palasodictyoptera as illustrated by Stenodictya 

 appear to represent the ancient stock from which the present 

 Pterygogenea complex has been derived. 



If this inference is correct then the problem resolves itself 

 into discovering an ancestor for the Palseodictyoptera. Before 

 developing the problem, however, it may be pointed out that 

 the abdominal appendages on the Stenodictya larva (tig. 2, B) 

 figured seductively by Handlirsch opposite a figm-e of a trilo- 

 bite with the same sort of appendages are of doubtful phylo- 

 genetic significance. The insect was aquatic in its early stages 

 and yet most of the Pennsylvanian insects were terrestrial. In 



