378 



J. D. Tothill — The Ancestry of Insects. 



In insect embryos one of the most universal and striking- 

 features is the early development and iinal disappearance of 

 paired appendages on the abdominal segments. Graber, 

 Wheeler, Ileymons and others have figured these structures in 

 numerous insects ; I find they occur also in Paratenodera 

 (fig. 4), Chauliodes, Ranatra and Polistes. The only possible 

 interpretation seems to be that these rudiments represent a 

 condition of serial polypody in ancestral forms. This condi- 

 tion is therefore represented in the diagram. 



Fig. 3. 



Fig. 3. Generalized hypothetical ancestor of Pterygogenea. (Original.) 



In figure 4 it can be seen that the tracheae have already 

 invaginated — i. e. the spiracles are plainly visible. These 

 invaginations occur generally in insects at an extremely early 

 stage, in fact immediately after the appearance of the append- 

 ages. This seems to indicate that they are ancient structures 

 and that the immediate wingless forbears of insects were 

 tracheate and therefore terrestrial. The hypothetical ancestor 

 may, therefore, be supplied with tracheal invaginations — at 

 least ten pairs as in insects, and probably fourteen. 



There remains the head. I cannot discuss the question of 

 head segmentation at this time, interesting though it is. Vial- 

 lanes, Wheeler and Ileymons are agreed that the insect head is 

 composed of six segments and these investigators have paid 

 particular attention to the development of the brain. Folsom 



