THE CANADIAN ENTOMOLOGIST. 299 



( Corylus rostrata) and the red raspberry. It occurs in both dry and 

 fairly moist situations. The specimens from Tobermory were taken 

 under similar circumstances, while at De Grassi Pt. they seem to be 

 confined to swampy ground where the vegetation is of a boreal character. 

 In such places I have taken them on bushes, chiefly raspberry, but have 

 often found them on the branches and trunk of the Arbor-vitae, sometimes 

 8 or lo ft. from the ground. I have never observed this habit in the 

 north, although the species is far more abundant there, but Mr. J. A. G. 

 Rehn says, in an interesting article on " The Habits and Distribution 

 of Podisma variegata " (Ent. News, XT., 630), that in Pennsylvania they 

 occur on the branches of hemlock, and that when removed they will 

 quickly return. 



From these various facts it may be inferred that P. glacialis is the 

 more primitive form, especially as the genus is typically an alpine one, 

 and that it once inhabited a much larger area, but after the retreat of the 

 ice-sheet it disappeared from this area, except in the northern part and on 

 the mountains farther south. Variegata, on the other hand, may be 

 regarded as an incipient species, the product of an effort on the part of 

 the parent species to survive amid the altered conditions of its environ- 

 ment. These conditions, as we go southward, diverge more and more 

 from those to which the insect was originally adapted, and hence it is not 

 surprising to find slight modifications of structure and colour-pattern 

 corresponding in degree with these changes. 



Its occurrence in swampy stations southward is what would be 

 expected from the fact that wet soil is a poor conductor of heat, and such 

 places are cooler than the more open, dry country, but its fondness for 

 hemlock in Pennsylvania seems to indicate a distinct specialization in the 

 insect's habits in this locality. Further observation, however, is desirable 

 on this point. 



As many of my Canadian specimens can be classed equally well with 

 glacialis or variegata, it will be necessary to give a new racial name to 

 these forms, and I have accordingly subdivided the species as follows, 

 though it will be understood that these different geographical races cannot 

 be sharply separated from one another : 



A. Antennae distinctly shorter than hind femora ( J ), nearly three-fourths 

 as long ( 9 ). Eyes not very prominent. Hind femora nearly 

 uniform green externally, obscurely bifasciate with darker green. 

 Furcula crossing basal fourth or fifth of supra-anal plate. Cerci of 



