72 THE CANADIAN ENTOMOLOGIST. 



the larva? now are apt to be caught by cold weather and destroyed, or their 

 food plant is cut oft", so that few can reach chrysalis. Once in the chry- 

 salis stage they are safe, and sooner or later, as the weather may permit, 

 the butterflies will emerge. I am inclined to think that the butterflies of 

 the third brood do not hybernate, and that the continuance of the species 

 here depends on the few individuals which survive from this fourth brood. 

 In no other way can I account for the scarcity of this species as compared 

 with comma. Both these species feed on same plants, hop, early in the 

 season, then nettle and Bochmoia, then Celtis and elm, and neither suffer 

 to any extent from parasites. But comma is fifty times more abundant than 

 inierrogatio/iis, and in the spring while many of the former are seen, I 

 rarely see an interrogationis. In midsummer and early fall this last 

 becomes common, and if the individuals of the third brood generally 

 hybernated, surely the species ought to be common in the spring. If 

 umbrosa ever passes the winter here I have failed to discover it. All the 

 spring examples noticed by me have been Fabricii. 



First Brood — The eggs obtained from % Fabricii in April gave 



in May 2 1 umbrosa, no Fabricii. 



The results of the next succeeding, or second, brood have been vari- 

 able, just as in second brood of comma, and both forms have appeared 

 from eggs laid by one female. 



The result of the third brood has also been variable, both forms 

 resulting. This, therefore, differs from the corresponding brood of comma, 

 in which all the butterflies were of the one form, Harrisii. 



The only examples of the fourth brood raised by me to imago came 

 from larvae found on elm 10th October, and when found were past third 

 moult. They must then have proceeded from eggs laid about the middle 

 of September. The chrysalis period was much protracted, but in Decem- 

 ber there resulted 4 Fabricii, no umbrosa. I have, however, in several 

 years seen the larvae of this brood late in the fall. As some individuals 

 of every brood of any species of butterfly appear earlier than the average 

 time and others later, so individuals of this fourth brood of interrogationis 

 doubtless appear early enough in the fall to ensure early hybernation. 

 And if the chrysalis stage is reached the butterfly is sure to emerge at last 

 unless destroyed by a parasite or an enemy. In the case of comma, when 

 compared with the behavior of that species to the northward, where there 

 are but two annual broods, it is the second brood which is interpolated in 



