Each lobe of the second and third pair composed of two lobules of 

 which the inner one is larger than the outer. Second pair always 

 smaller than the first ; third pair always smaller than the second. 

 Gland-spines usually simple though sometimes forked at their tips 

 and usually shortest near the median lobes, gradually increasing in 

 size toward the anterior part of the pygidium. 



The spines are usually plainly visible ; those on the dorsal larger 

 than those on the ventral surface. First row of dorsal gland-orifices 

 absent : second row usually represented by the anterior group only, 

 though in one species (dysoxylt) the posterior group only is present ; 

 third and fourth rows always present. 



Circumgenital gland-orifices always present and arranged in five 

 groups. 



SCALE OF MALE. Elongated ; tricarinate, unicarinate or without 

 carinae. Ventral scale complete, forming with the upper part a 

 complete tube. Plate 2, Fig. 8 A represents the male scale of salicis, 

 the type of this genus. Plate 5, Fig. i shows it in cross sections. 



REMARKS. 



Of the fifteen species and one variety now known in the genus 

 Chionaspis, eleven are from America and one each from Europe, 

 Japan, New Zealand, Ceylon and South Africa. Most of the species 

 are rather northern in their range of distribution, and as a rule 

 occur on the bark of the 'host plant. 



Some of the species are very closely related and it has been only 

 by careful study of a very large amount of material that I have 

 been able to find trustworthy characters by means of which they can 

 be separated. It is my opinion that the true European sahcis has 

 not been discovered in America and that what Prof. Comstock 

 believed to be this species is distinct, though closely related to it. 

 After a careful study of a mass of material from all parts of this 

 country. I have come to the conclusion that Prof. Comstock's salicis 

 is identical with the species described by Walsh as salicis-nigrae and 

 it seemed proper to adopt his name, even though his type is lost and 

 no comparison with it can now be made. 



I have adopted salicis L. as the type of the genus Chionaspis for 

 the reason that in his original description of the genus, Signoret 

 mentioned this species as an example, though he did not state defi- 



