may be taken as (very conveniently from its large size) showing the 

 Tortrix character. The larva, with its circles of hooks of alternate 

 lengths on the prolegs, is of strictly Tortrix pattern. The imago 

 agrees with Tortrix in every detail, the venation is almost identical 

 with that of C. pomonana" etc. Pomonana here mentioned is a 

 member of the genus Carpocnspa, and it should not be surprising 

 to find a species shown to be so closely allied, having similar habits 

 to the members of that genus. The species of Carpocaspa feed in 

 various fruits, such as nuts, acorns, apples, etc., and are full-fed in 

 the autumn, when they quit the fruit in which they have fed. So 

 far as I am aware they do not really enter the earth, but spin up in 

 any rubbish that they happen to meet with, or under loose bark or 

 moss, where they remain until the spring before assuming the pupal 

 stage. 



This appears to me to be just what such evidence as we have 

 regarding Cossiis points to. We should expect it to leave the mines 

 in which it has fed in the autumn of the year in which it attains full 

 growth. If the tree in which it has passed its earlier life provided 

 suitable dry, soft, rotten wood for its hybernation and ultimate pupa- 

 tion, the necessary conditions for shelter would be fulfilled. Experi- 

 ment has shown us that the larvae pupate readily in sapless timber. 

 Our forefathers knew that the larvte fed in trees ; they found pupae 

 in trees, though possibly not in the burrows in which the larvae had 

 fed, and therefore did not search for them elsewhere. Coming now 

 to the evidence as to subterranean pupation, the general drift appears 

 to me, where definite particulars are given, to show that the sites 

 selected are in earth of a light fibrous nature. The centre of a 

 tennis lawn may certainly suggest something of a more solid charac- 

 ter, but we are given no notion of what the lawn was laid on, or 

 whether the cocoon was formed any deeper than the roots of the 

 turf. But in the case of the pupje found annually in the " sand- 

 cops," we are clearly told that the cocoons were detected by feeling 

 along the top of the "cop," with the fingers, for a soft place, showing 

 that the cocoon was made in the turfy covering, and not in the harder 

 soil beneath. Another case where larvas were found some 8 or 9 

 inches below the surface of " soil that had been used for dahlias and 

 sunflowers " is sufficiently suggestive of something akin to a loose 

 rubbish heap; and the light, fibrous nature of the "earth" from 

 which the pupae skins were found last summer has already been de- 

 monstrated. 



The conclusion, therefore, that I arrive at is that Cossus is not a 

 purely "internal" species like Zeuzera pyrina or Sesia sphegifortnis, 

 for instance, nor is it truly subterranean as some of the Sphingidce, 

 but that in its habits it more closely follows its near allies. The 

 larva, on leaving its food on becoming full-fed in the autumn, taking 

 advantage of the first dry friable substance that comes in its way, 

 enters it to make its final hybernaculum, in which it also pupates in 

 the following spring. 



