50 BULLETIN NO. 3, DIVISION OF ENTOMOLOGY. 



in front than in Xystus, not so elongate or so elevated dorsally. The 

 male antennae are bipectinate ; the lamella? rather short and ciliate. 

 The female antenna? are serrated. It is allied to the European Cossus 

 terebra F., but is a larger insect. It differs from C. querciperda Fitch 

 by the absence of any yellow on the male hind wing, and by its darker 

 color and closer reticulations. 



In color this species is black and gray. The edges of the thorax and 

 collar are shaded with gray, more noticeable on some specimens than 

 others. The primaries are covered with black reticulations, which are 

 not always identical in their minor details in different specimens, nor 

 sometimes on both wings in the same specimen. Beyond the cell there 

 is a transverse continuous line, broader than the rest, and outwardly 

 bent over median nervules. The brown color is blackish over nearly 

 two-thirds of the primaries from the base, and outwardly gray; hind 

 wiDgs rounded in both sexes, with blackish hairs at base, pale and sub- 

 pellucid, with short gray fringe, before which there is a narrow black- 

 ish edging. The abdomen is blackish. The males are smaller than the 

 females. The smallest male expands about 40 mm , the largest female 

 over 60 mm (see Plate I, Figs. 10, 11, and 12). While thus far the Centre 

 (N. Y.) locality has proved to be the chief home of this Cossus, it will 

 undoubtedly be found elsewhere wherever the Populus tremuloide* 

 is found. Several pupa-cases of this species have been found in the 

 corporate limits of Albany. Usually trees of less than 1 foot in diameter 

 are attacked, although in one instance a pupa-case was found in a tree 

 measuring 16 inches in diameter. 



It is a very different matter to observe the changes of insect life from 

 the eggs to the imago when feeding upon the foliage of vegetation than 

 where the larva? have bored deep into a tree trunk and feed upon the 

 ligneous fiber and its circulating fluids. To obtain this information it 

 has been necessary several times each year to cut down trees bearing in- 

 dications of its ravages, and to dissect them into fragments the size of 

 kindling-wood. The mouths of October, April, and June were selected 

 as suitable times for such investigations. October 14 we visited a tree 

 for the purpose of obtaining caterpillars, and from a limb 4 feet in length 

 six caterpillars were taken, two of which were occupying cells as seen in 

 the engraving. 



April 2 we cut from a tree a limb 3 feet in length, and in it we found 

 seventeen caterpillars of three distinct sizes, indicating a growth for each 

 year. The larger ones were not fully grown. All of them were actively 

 passing through their tunnels in the wet wood, through which the sap 

 was freely flowing. Not any of the caterpillars were occupying pupa- 

 cells at this time. June 12, 1881, we again visited a tree when the insects 

 were emerging. The tree selected was far advanced in decay, from the 

 effects of the tunneling of the larva? ; only about 4 feet of the trunk was 

 alive, with a few lateral branches in foliage, scarcely enough to support 

 its respiration. In the trunk were found fresh pupa-cases, pupa?, and 



