232 A CloKsiJiealion of Lepidopte?'Ous Larvae. 



Superfamily SATURNINA. 



We have an unpaired 'lorsal tubercle on both the 8th and 9th 

 abdominal segments, the first corresponding to i, the second to ii, 

 as tubercle i is not present on the 9th segment. The Citheroniidse 

 is the most generalized family, tubercles i (on 8th abdominal seg- 

 ment), ii, iii, V, and vii being present in Citheronia splendens. In 

 the other genera the number of tubercles on the 8th segment is 

 reduced, usually by the removal of ii ; but in Anisota, i has disap- 

 peared. It should be understood that I exclude the Lacosomid^e 

 from the Saturnina ; otherwise I accept Prof. Comstock's arrange- 

 ment. 



Family Sphingidae. 



The position which I have assigned to this family is only tenta- 

 tive. The larvae are so highly specialized that it is very difficult to 

 get any clue to their relationship. The only trace of the tubercles 

 that is left, is the "caudal horn," an unpaired dorsal process on the 

 8th abdominal segment. The view which I have adopted here, that 

 this horn represents the consolidated tubercle i of the Saturnina 

 has no more to support it than the other view that the horn repre- 

 sents the base of the unconsolidated pair of tubercles i, the tubercles 

 themselves having disappeared. The tendency to the formation of 

 this structure has been referred to in the case of the Notodontida? 

 and Lasiocampidse. This view would place the Sphingidse at the 

 end of the Frenulum Conservers, next to the Lasiocampidae in my 

 table ; and I wish to state that, as far as I am able to judge from 

 the mature larvae, either position is equally supported.' 



1 The larva in stage I, wliicli I have excluded from consideration, is figured 

 by Weismann in the case of the European Deilephila euphorbia (Studies in the 

 Theory of Descent, PI. V, fig. 38), and setse are present. Unfortunately, the 

 figure appears to kave been made without sufficient care as regards the arrange- 

 ment of the tubercles. There is no description except " on each segment there 

 are a number (mostly ten) of small warts, each of which emits a single bris- 

 tle." The figure represents all the tubercles present except iv, though v and 

 vi seem to be situated too high. On abdominal segment 8, the two setse of 

 tubercle i are borne on the apex of the caudal horn. If the structure of this 

 embryonic larva is to be trusted, the conclusion seems irresistable that the 

 alternative position which I suggest for the Sphingidgs is the proper one, and 

 not the one which I have adopted above, following the conclusions of Mr. A. R. 

 (irote (Canad. Entom., vol. XXII, p. 15), and of Prof. E. B. Poulton (Trans. 

 Ent. rioc. London, 1888, pp. 568-574). 



