THE CANADIAN ENTOMOLOGIST. 197 



CECIDOMYIID.E: A STATEMENT. 



BY E. P. FELT, ALBANY, N. Y. 



The Gall Midges, though extremely small in size and frequently of 

 very uniform colouring, are, nevertheless, easily referable, for the most 

 part, to family, genera and species, by characters found on the insects 

 themselves. Inasmuch as two and sometimes three or four different 

 species may inhabit the same gall or very similar galls on the same plants, 

 it must be admitted that a system dependent largely upon plant deforma- 

 tions is not entirely satisfactory. Moreover, a number of species, 

 including in this category practically all the members of two important 

 subfamilies, produce no vegetable deformations. Obviously these latter 

 species, if identified at all, must be separated by characters presented by 

 the insects themselves. The minute we allow this, there must be some 

 method of distinguishing them from the host of other species with which 

 they might be confused if taken in the field away from the galls in which 

 many forms breed. Our correspondents may be interested to know that 

 the preliminary descriptions issued from this office have, in every instance, 

 been condensed from more detailed characterizations (not to mention 

 numerous photomicrographs and other illustrations), all bearing the same 

 number and easily associated with specimens similarly marked, 

 consequently there can be no question later on as to the identity of the 

 forms described. Furthermore, we are now engaged in a serious 

 systematic and biologic study of this group, and have already well in hand 

 a series of tables for the separation of not only subfamilies and genera, 

 but also species, in such a manner that all may be recognized irrespective 

 of the plant or material in which they breed. This, it seems to us, is the 

 only logical basis for a classification, and something that is imperatively 

 needed, particularly when it is remembered that very few of the published 

 descriptions are sufficient for the identification of adults, unless they are 

 taken in connection w'th the galls. Obviously, progress must be seriously 

 hindered if this condition is allowed to persist, since many of the earlier 

 described species can be identified only by securing the galls and breeding 

 the insects therefrom. We would not imply by the above that the 

 biological study of this group should be neglected, far from it ; this phase 

 should be pushed with all possible vigour. We do urge the necessity of a 

 thorough study of the adults and the replacing of the present more or less 

 insufficient descriptions by characterizations that will bring out the specific 

 differences most clearly. 



The above statements are made at this time owing to the fact that 



June. 1907 



