28 ILLINOIS BIOLOGICAL MONOGRAPHS [28 



and they are in the same position here as the Tan group on the thorax, 

 so the entire homology is clear, as labeled on the plate. 



On the proleg-bearing segments, omega varies between a mesocephalic 

 position at the base of the leg in some species to a laterocephalic location 

 in others. In the latter position it is quite closely associated with the 

 Pi group and seems to belong to it rather than to the group from which 

 it was derived. On segments 7, 8, and 9, it is, however, wanting, so that 

 these segments bear four subventral setae, nu and pi representing the 

 still bisetose Pi group, tau alone remaining of the group which bears 

 its name, and, as usual, sigma near the medioventral line. 



In many caterpillars the ninth abdominal segment shows a very 

 puzzling condition, analogous to that on the mesothorax. There is a 

 tendency for the setae to arrange themselves in a single transverse line, 

 challenging the investigator to say which is which. The larvae of Hepialus 

 (Fig. 14), however, have not undergone so much change, and homologiz- 

 ing the setae on this body-segment is not so difficult. As usual alpha and 

 beta are found near the mediodorsal line with rho and theta below them. 

 Epsilon, altho present on the preceding segments, is wanting here, but 

 it is such an evanescent seta that its absence in this place is not sur- 

 prising. Kappa and eta are somewhat closer together and farther caudad 

 and dorsad than before, but the homology is clear. This leaves only the 

 subventral setae, which, as was stated in the last paragraph, are identical 

 with those on segments 7 and 8. 



The most difficult problem of all still remains and we can give it 

 only a partial answer. Had we all the ancestors of Hepialus there 

 might still be a doubt as to the meaning of the anal segment. Is there 

 a tenth segment and then a telson representing the eleventh? Does 

 the proleg belong to the tenth or to the eleventh segment and does it 

 bear the setae of both segments or of only one of them ? What is the ori- 

 gin of the setae on the caudal aspect ? The condition in Hepialus larvae 

 is as follows : 



On the dorsal half of the last abdominal segment (Pig. 14) is a 

 semicircular plate whose diameter is the cephalic border of the segment. 

 On each side of the meson this plate bears three setae, two in a longi- 

 tudinal line comparable to the position of alpha and beta on the other 

 segments, and one farther laterad and cephalad. Between the plate 

 and the anus is a pair of fleshy projections, the suranal lobes, each bear- 

 ing one seta on its caudal aspect. Below the anus another pair of lobes 

 and the prolegs bear a total of eight setae. Two of these are on the cau- 

 dal aspect of the ventral lobe, four on the lateral aspect and one on the 

 caudal aspect of the proleg, and the other cephalad of the proleg. 



