THE CANADIAN ENTOMOLOGIST. 197 



What has been thus proven in Crustacea will certainly be observed 

 also with other Articulates. Since insects do not possess a post-abdomen, 

 there cannot occur the same differences as in the case cited, but analogous 

 ones will not be wanting. It is obvious that so-called " salt insects " are 

 the first ones which will need new and careful study. Those known are 

 Coleoptera, Diptera, Hemiptera and Orthoptera, and the species are often 

 nearly related to other ones which do not live in salt regions. Further, it 

 is evident that similar changes will be the result of different conditions of 

 life. So-called " local varieties " are certainly nothing else, and a vast 

 field of observation and study is opened by the remarkable discoveries 

 of Mr. Schmaukevitch. I believe that we are now justified when we 

 exclude from generic characters all the following ones : 



i. Every character based on the number of parts, when the number 

 ceases to be a small one ; the more so when it varies in related species* 

 If a number is larger than about a dozen, we can never rely upon the 

 constancy of the number in antennal joints or anal appendages. In 

 spines, bristles, spurs, a much smaller number is constant ; transversal 

 veins of the wings belong to the same category. 



2. The external coating.. ^ the body, consisting in hairs, scales and 

 other appendages, is not a generic character. The hairs, tufts, brushes, 

 spines, spurs, are often only sexual and can not be considered generic char- 

 acters ; also, hairy eyes, since we find this character changing in the 

 most related species and probably in the same species in Diptera. 



3. The presence or want of the ocelli or eyes is not a generic char- 

 acter. 



4. The veins of the wings give only to a certain degree generic 

 characters, viz : the principal branches, but certainly not after their bifur- 

 cation. 



Having arrived so far by exclusion, it is important to state what is 

 left for generic characters. 



So *?r as I am advanced in the study of generic characters, I think 

 t 1 """ /wing should be used : 



1. The form and relation of the three principal parts of the body. 



2. The organs providing nutrition (mouth parts). 



3. The organs making possible the working of the mouth parts, i. e.> 

 the organs of locomotion. 



The anatomical characters may be of prominent help. At present our 

 knowledge as to their details is too limited to admit our using them to a 



