276 



Illinois River at Havana. A detailed description of all stages of this 

 species is given herein. The Chironomidcv, though resembling the 

 CiiUcidce in many respects, including the form of the adult and the 

 aquatic habits of the larva:, do not present in the larval stage such 

 characters as those which are so readily appreciable and so easily ac- 

 cessible in the larvcC of the latter family. It is only in the terrestrial 

 species of Ceratopogonince that one meets with conspicuously spinose 

 forms, and in the great majority of cases a clue to the specific, and 

 even generic, identity of a larva must be sought in the structure of the 

 head and its parts, and also in the form of the anal appendages. 



The members of this family are among the commonest and most 

 widely distributed of the two-winged flies, occurring on every conti- 

 nent and in all the faunal areas into which these continents have been 

 divided. Although considerably over one thousand species have been 

 described, it is certain that this is but a small fraction of the total num- 

 ber of species which must occur throughout the world. The reasons 

 for this paucity of described species are obvious. The fragile struc- 

 ture of most of the species prevents their being readily preserved, and 

 their frequent close similarity deters all but a few specialists from 

 working on the family. Despite the extremely fragile structure of the 

 members of this family many fossil species have been described by 

 various authors. Conjectures as to the possible sources from which 

 the family sprang must remain, as in the case of other families, mere 

 conjectures, serving only to create purely academic discussion, which 

 has no proper place in the present paper. It may, however, be useful 

 to students of the Diptcra to indicate, as clearly as possible from the 

 available data, the characteristics of the family. 



In this paper I have introduced in descriptions of genera and spe- 

 cies a few characters which have not previously been used by writers 

 in dealing with the family. I have in every case endeavored to find 

 coordinated characters in the sexes, and in dealing with the genera I 

 have, wherever possible, associated the characters found in the adults 

 with a certain set of characters found in larva: and pupae. I have, I 

 believe, met with a considerable degree of success in deciding some of 

 the more difficult points, particularly in the Ceratopogonince. The 

 presence of the thoracic cavities in Cidicoides and their absence from 

 the members of closely allied genera serves as a much more satisfac- 

 tory and more readily appreciable character for the separation of the 

 genera than that previously in use, namely, the size of the empodia in 

 comparison with the size of the claws. I have of course followed pre- 

 vious authors in the main lines of separation, only occasionally deviat- 

 ing when convinced that such course is expedient and conducive to a 

 better understanding of the family; but in descriptions of species I 



