30 ILLINOIS BIOLOGICAL MONOGRAPHS [30 



first instar.* The seta] plan of Feltia gladiaria in this stage shows a 

 close correspondence to that of the newly hatched Hepialus larva. On 

 the prothorax (Pig. 17) eta of the Kappa group, all the Tau group, and 

 sigma are wanting. Otherwise the homology is clear. The mesothorax 

 and metathorax (Fig. 18) show a condition which has probably been 

 developed on account of the great mobility of these segments. The setae 

 are in a single transverse row. Above kappa, located caudad of the 

 prothoracic spiracle, are four setae instead of the five borne by Hepialus. 

 The missing one proves to be delta, for the close correspondence with 

 the abdomen shows that the dorsal two are alpha and beta; aud the 

 others are undoubtedy epsilon and rho, which are usually associated 

 together on all segments. Below kappa is pi, near the proleg. As on 

 the prothorax, the Tau group and sigma are wanting. 



The abdomen (Figs. 19, 20, 29) shows alpha and beta in their 

 usual positions near the dorsomeson except on the first few segments, 

 where they are more nearly in a transverse row, similar to their arrange- 

 ment on the metathorax. Just dorsad of the spiracle is rho, a well- 

 developed seta, and in most cases a minute point representing epsilon. 

 The latter, known as iii a in the literature, is often considered sub- 

 primary, but the presence of this rudiment, which Bacot (Quail, 1904) 

 says is of common, if not universal, occurrence in all stages of the larvae 

 of Frenatae, proves it to be primary. Its small size is the result of 

 reduction, rho having migrated caudad to the region it once occupied. 

 The other abdominal setae are just as in the first-stage Hepialus. In 

 Feltia, tau is not present, but in most Frenatae it is said to be associated 

 with nu and pi in all instars and is usually considered a member of the 

 Pi group. Its absence from the first, seventh, and eighth abdominal 

 segments is a common occurrence. Sigma is located near the medioven- 

 tral line as usual. 



The homology of the setae of segment 9 (Fig. 30) may be solved by 

 a comparison with the same segment of the mature Hepialus larva and 

 a study of other species which form connecting links between the two 

 conditions. The evidence for considering the most cephalic of the sub- 

 dorsal setae as alpha, will be given in the next section. Beta is much 

 closer to the dorsomeson, as on the prothorax. The presence in some 

 species of a minute seta (epsilon) close to the third seta shows the latter 



*Careful descriptions of first-stage larvae of many other groups have been 

 published, the prothorax usually being omitted. Since sending this paper to the 

 printer, the first instar of Prionoxystus robiniae (Cossidae) has been examined 

 by the writer. Its prothorax is identical with that shown in figures of Hepialus 

 in this stage, differing from noctuid larvae in the presence of two setae, instead 

 of only one, in the Kappa group. 



