293 



Family CEClDOMYIIDAB 



I have not devoted niiich time to rearing this family, and my ma- 

 terial is limited, consisting of a few species that were picked up in the 

 course of other work, those that form the existing collection of this 

 Laboratory, and a number that were kindly supplied by Dr. E. P. Felt. 

 Judging from my examination of this material, there appear to be 

 quite a number of good distinguishing characters for both larvae and 

 pupae. Owing to the very large number of species included in the 

 family it would recjuire a great amount of work and the expenditure 

 of much time to devise a satisfactory classification of the early stages. 

 I have figured a few details of some of the larvae and pupae as in- 

 dices of the general character of these stages. KiefTer, in "Genera 

 Insectorum"*, has given, in addition to generic synopses of imagines, 

 a summary of the larval and pupal characters of the family, and keys 

 to the known larvae of the European genera. These keys will no 

 doubt prove serviceable to students of the North American species. 



Dr. E. P. Felt has been for several years studying this family, and 

 his work on the imagines has done more to clear up this obscure group 

 than that of any other worker either in America or abroad. 



CHARACTERS OF THE) FAMILY 



Larva. — The head is not so heavily chitinized as in other families 

 of the Nematocera, and on that account, and because it is generally 

 incomplete posteriorly and very small, it is rather difficult to dis- 

 tinguish details, especially in the small species ; the eyes and mandibles 

 are apparently absent, but the antennae are long and 2- or 3-jointed. 

 The presence of 13 segments, exclusive of the head and of lateral 

 abdominal spiracles, is sufficient to distinguish the larvae of this fam- 

 ily from any other in the order except Bibionidae, and here the com- 

 plete head of the latter serves to distinguish the two families. Nor- 

 mally there is present also, at least in the mature larvae, a chitinized 

 plate under the membrane on the ventral surface of the second 

 thoracic segment. This plate differs considerably in the different 

 species, and its form is of value for identification purposes. It serves 

 the larvae as a means of propulsion in making their leaps after leaving 

 their pabulum to pupate, or when they are removed therefrom and 

 placed upon a dry surface. Many species form galls upon plants, and 

 may be identified by these alone, as each species usually attacks but 



*Fasc. 152. (1913) 



