— 182- 



drawings were afteiwaid made, and the word Cosmosoma wherever it oc- 

 curs should read Didasys. I had no (^ specimen of Cosmosoma before 

 me when that paper was written, hence my failure to note the peculiar 

 structure hereafter described . 



The head presents nothing noteworthy, and the figures sufficiently 

 explain its structure. The front is somewhat protuberant, nearly quad- 

 rate, suddenly receding, and with a deep concavity below the base of 

 the antennae, leaving a distinctly marked intervening carina extending to 

 the suture. The ocelli are distinct, as usual situated close to the com- 

 pound eye and behind the base of the antennae. 



The dorsum of thorax appears slightly lengthened in the figure, be- 

 cause the convexity is not shown. Of the prothorax nothing is visible 

 from above. 



The mesonotum or scutum is, as usual, the most prominent, and is 

 oblong, somewhat widening posteriorly. The upper surface presents 

 nothing of note, and does not seem to differ from its allies. Viewed from 

 the side it shows more that is characteristic, and differs essentially from 

 that oi Zygaena by the slight development of the meta-thorax. No com- 

 parisons can be made at present because of the lack of knowledge of 

 allied forms, and I shall simply let the figure speak for itself 



The abdomen is most peculiar. It consists of 8 segments, includ- 

 ing the membraneous connecting segment at base, and excluding the 

 specialized rings forming the genital armature. 



The membraneous ring at base consists of a dorsal plate a, an in- 

 flated process, covering a spiracle, b, and a minute sternal ring, not vis- 

 ible from below unless the abdomen is removed. The second dorsal 



