NOTES ON COSSIDaS. 95 



iv and v approximate on one plate common to both, and basal 

 setae. Most of the setae are typical, long, hollow, bristle-like, 

 and pointed ; but on the anterior trapezoidal and anterior sub- 

 spiracular tubercles are short setae with curious tips, which we 

 will call trumpet-like. Immediately on noticing these curious 

 setae, I observed also that the thoracic segments had two exactly 

 similar setae (fig. 2). These were conspicuous on the specimens 

 which I first examined. Mr. Bacot questioned the shape of the 

 setae, the specimens he had examined not apparently exhibiting 

 the trumpet-like form of seta. I therefore gave further attention 

 to the matter, and am satisfied that the two setae on the thoracic 

 segments are exactly similar to the two on the abdominals. The 

 figure (4) of an abdominal anterior trapezoidal seta was drawn 

 from one of Mr, Bacot's slides; on that specimen are other more 

 or less trumpet-like setae, the curve of the seta is probably due to 

 pressure of the cover-glass. Although not absolutely certain, I 

 believe we have here a functional seta of a remarkable kind 

 lohich can open and shut; when open at the tip the seta is 

 trumpet-like, closed it loses this appearance. 



The position of all the tubercle setae on the thoracic segments 

 differs from that of the abdominals. On C. cossus, in addition 

 to the displacement of thoracic setae from spiracular reasons 

 above referred to, one notes iii b has been eliminated, resulting 

 in the approximation of iii and the anterior subspiracular tubercle, 

 and, moreover, the movement of this to a position above iii. 

 The homologues appear to be in lateral descending order on the 

 thoracic segments : i, ii approximate, subspiracular seta moved 

 up ; then iii ; single subspiracular below and basal. 



It is unfortunate that I cannot carry these observations 

 further as regards C. cossus. The nearest material I have being 

 adult Zeuzera pyrini, which is not really of value to compare 

 with C. cossus newly-hatched. On the thoracic segments of Z. 

 pyrini adult larva (fig. 5) : i, ii are separate ; iii a, b approxi- 

 mate — I do not think the subspiracular seta is here moved 

 up, iii a* b exists definitely on the abdominal segments ; iv, v 

 are curiously above what is perhaps a scar of the eliminated 

 spiracle, and, moreover, the setae are on separate tubercles — such 

 spiracle scars exist in all Zeuzerinae I have seen ; vi is below, 

 and on the base of legs another single seta. 



The abdominal tubercle setae of Z. pyrini (fig. 6) : i, ii remote 

 normal; iii a,* 6 ; iv, v approximate; vi posterior; and three 

 basal setae. A minute, subdorsal, anterior seta may also be 

 noted ; this, like the spiracular scar, persists on all species allied 

 to Zeuzera. 



As the result of criticism, Dr. Dyar admits that the tubercle 

 setae of the thorax are homologous with those of the abdominal 



"- See my note, foot of p. 94. — A. B. 



k2 



