THE CANADIAN ENTOMOLOaiST. 177 



between these two and the other Satmdersii. Still, aside from these 

 three, the specimens in either set are greatly uniform among themselves 

 and the contrast is great between the two sets as to this feature of 

 decoration with white. 



I am unable to find other severance than by this white decoration ; 

 indeed, the two series are in other respects such counterparts that if the 

 sex characters permitted, I should conclude ednsa the male and Saimdersii 

 the female. But very evidently such is not the case, for both series seem 

 to contain males only. 



LUNATA COMPARED WITH THE OTHERS. 



Individuals more robust in body than those of the two other series, but 

 as to expanse of wings about the same average. 



Of the 42 specimens almost all have the general appearance of females. 

 Of only one or two would there seem any doubt, and even these do not 

 look like males, but their appearance is less conclusive as to their sex. In 

 the other sets the reverse seems the case ; all but two or three decidedly 

 appear to be males, and the exceptions do not look like females, but 

 merely have less definite characters. 



In color lunata contrasts with the others by a more sombre style 

 of decoratioff 



On under side the markings of the three series agree as to essentials, 

 the difference being in shading, &c. Ednsa and Saundersii are alike, 

 with a moderate range of individual variation. Lunata has on an average 

 more distinct markings, varying to partial obsoleteness of the strong lines. 



The three forms occur coincidently, the season of chief abundance, as 

 indicated by the dates of my specimens, being from mid-July to about 

 loth August. A few bright ones show dates during September, others 

 late in October, and one lunata November 12th. Of the late flight, part 

 survive winter and are found in June (May more rarely) much worn. No 

 evident fresh examples dated earlier than July 15th. 



P. S. — On p. 136 see Mr. Grote's correction of p. 89. 



A few lunata and Saimdersii taken during the past two weeks (August 

 nth) tend to confirm the foregoing comments. 



