323 



August 14 they had almost ceased to come in, partly tipcause of the 

 lateuess of the season, but more on account of the cold and rainy 

 weather that followed. 



The Sphiugidte were comparatively few in number, being most abun- 

 dant from July 9 to 17; 21 species were taken in 122 examples. The 

 most abundant species was Paonias myops, followed by P.excoecatus and 

 Protoparee celeus, as maj' be seen from the list. 



The Bombycidae were much more numerous, 93 species being taken 

 in 7,871 examples. The three species, Halisidota tesselata, CUsiocampa 

 americana, and Hyphantria cunea were most abundant, comprising 82 

 per cent, of the whole. H. tesselata and C. americana comprise 65 per 

 cent., while H. cunea forms 48 per cent, of the remainder, leaving only 

 18 per cent, for the other 90 species. It will be seen that the number 

 of specimens captured increases rapidly to a maximum on June 24 and 

 then decreases regularly, with the exception of July 28, when but a 

 small lot were taken. The numbers represent the number of each spe- 

 cies that was taken sufficiently well preserved for identification. In a 

 few instances they are followed by an interrogation mark (?) when the 

 identification was at best doubtful. 



From a consideration of the figures I think at a fair estimate the 

 electric lights of Poughkeepsie must destroy annually 25,000 Bomby- 

 cidsB, of which fully 8,000 are Halisidota tesselata. I wish to express 

 my thanks to Mr. James H. Van Norstrand, of Poughkeepsie, who has 

 charge of trimming the lamps that I visited, for his kind cooperation 

 in the collection of these moths. 



