33] LEPIDOPTEROUS LARVAE— FRACKER 33 



(Fig. 5), where it is about half way between the spiracle and the dorso- 

 meson, to that in most Prenatae (Pig. 21), where it is closer to the 

 dorsonieson than any other seta of the segment. Intermediate stages 

 are shown by Yponomeutidae (Fig. 35) and Acrolophidae (Fig. 7). 



The specialization of the mesothorax and metathorax of Frenatae 

 has been mentioned. No intermediate stages exist to show the relations 

 between the setal plan of these segments of Hepialus and those of Frena- 

 tae. A comparison of the thoracic with the abdominal segments of the 

 newly hatched larva of Feltia (Figs. IS, 19, 20, 29, 30), however, indi- 

 cates the direction the migration has taken. Segments 6, 7, and 8 show 

 alpha and beta in their normal positions while on segments 1 and 2 

 beta is almost directly ventrad of alpha. The large number of groups 

 in which these two setae are associated in the abdomen, either by being 

 borne on a single pinaculum, or uniting to form a single verruca, shows 

 that they may easily become adjacent. The conclusion is therefore 

 justified that the two dorsal setae of the mesothorax and metathorax 

 are alpha and beta, in spite of their changed position. Hofmann, Dyar 

 and Quail agree that this is the true condition. 



Gamma. This seems to be a comparatively new seta on all but the 

 prothorax, where it has become established. The newly hatched larva 

 bears it only on the one segment and there its position is constant. It 

 is often represented on meso- and metathoracic and abdominal segments 

 as a more or less minute subprimary seta near the cephalic border of 

 the segment and in Hepialus is associated with two other smaller ones. 

 While there is some doubt about the accuracy of considering a primary 

 seta of the prothorax homotypic with the subprimary one of other seg- 

 ments, the fact that the two bear exactly the same relations to the other 

 structures of the segment makes it inadvisable to use a different name. 

 This will be discussed under "Eta" on a later page. The probability 

 is that gamma is a more recent seta than the other primaries and be- 

 came first established on the prothorax, the order of appearance on the 

 different segments now being shown by ontogeny. 



Delta. Between beta and rho on the prothorax of all caterpillars 

 and also on the other thoracic segments of Hepialus is delta, a seta never 

 found in any other position or on any other segment. If ancestral, it 

 has since become lost on the abdomen of all lepidopterous larvae and on 

 the last two thoracic segments of all Frenatae. 



Epsilon. The prothoracic segments of Hepialus and Pseudana- 

 phora bear the third seta of the cephalic row in its typical position. In 

 connection with rho it goes through various changes which are some- 

 times hard to follow on the other body-segments but are evident on the 



