87 



sociated are all Incompletse is a problem to be worked out ; 

 there is little in the appearance of the imagines of our species 

 to suggest the profound difference which exists between them 

 and the Bombycidse, but the larvae and pupae show how 

 widely apart these two families should be placed. The legs of 

 the larvae in the Limacodidse are evanescent, and, on the other 

 hand, they are perfectly developed in the Zygaenidae, showing 

 how very little affinity some of the families of Incompletae 

 have among themselves. 



The very name of the sub-order Lepidoptera-Heterocera 

 is suggestive of the heterogeneous character of the insects 

 included therein. It is a step in the right direction to divide 

 it into Obtectae and Incomplete, making with the Rhopalocera 

 three sub-orders ; for many years past it has appeared to 

 me that the oral organs of the imagines have been far too 

 little considered in classification. 



All field naturalists will bear me out in the statement, that 

 there are many large families in which the structure of the 

 parts of the mouth in the imagines is such that they are unable 

 to feed ; for instance, the Noctuidae, the spiral tongue or 

 maxillae of which is greatly elongated, often swarm at sugar 

 baits, and on flowering trees and shrubs, such as the lime, 

 willow, ivy, and others ; but the Geometridae, in which the 

 maxillae are short and weak, being nearly membraneous, are 

 very rarely if ever so attracted, neither do they visit flowers. 

 The same may be said of the Bombycidae, the maxillae of which, 

 when present, are so short and weak as to be useless as a 

 tongue ; yet the latter family is placed by systematists next 

 to the Sphingidae, which have the most highly developed 

 suctorial mouth, the tongue being often longer than the body, 

 and sometimes in certain exotic species it is several inches 

 in length. 



In my judgment, when these deeply-seated differences of 

 the imagines in structure and habits are considered, together 

 with those pointed out by Dr. Chapman in the pupae, it is 

 a very unnatural combination to associate under the ob- 

 jectionable name of Sphinges and Bombyces, among others, 

 such widely different families, as the Sphingidae, Bombycidae, 

 Zeuzeridae, and Hepialidae. The subject of classification is 

 one that time will not permit me to pursue further, and I 

 shall therefore add no more on that point. 



It is a well-known fact to those who have entomologized 

 in the higher Alps that an unusual proportion of nielanic and 

 phaeic varieties of Rhopalocera are found, and the question 

 arises as to the cause of such aberrations. In these altitudes 



