CLASSIFICATION OF THE DIPTERA 155 



are constantly being made to it, classification and identification are 



somewhat difficult and unsatisfactory. The field worker 



... . Classification 



or parasitologist would do well, unless he has 



exceptional opportunities, to leave such matters in the hands of the 

 specialist, and to content himself with a knowledge of those groups 

 which are of practical importance. It is sufficient for most purposes 

 to be able to ascertain with certainty, from a study of the external 

 characters, which provide almost all the diagnostic features, that one 

 is dealing with one species of fly and not two or more, and to be able to 

 place the species in its proper genus. To do more than this requires 

 in many cases an access to a voluminous and scattered literature, and 

 also to the type specimens from which the species was originally de- 

 scribed, or at least to a specimen named by a competent authority. The 

 difficulty of identification has been greatly diminished by the keys which 

 have been drawn up by specialists for such genera as contain species 

 of practical importance. These will be found in this and the succeed- 

 ing chapter, and it is hoped that with their assistance, and with some 

 knowledge of external anatomy, the worker will be able to identify most 

 of the important forms. A few named specimens are of great value 

 for purposes of comparison. They may often be obtained in exchange 

 for other material, or the worker may make a representative collec- 

 tion from his own district, and have the specimens identified. 



The following arrangement of the groups is mainly taken from 

 Williston, who has adopted the classification of Brauer. 



The Diptera fall into three suborders, with secondary divisions, as 

 follows : 



ORTHORRAPHA. Larva with a jaw-capsule, or with a differentiated 

 head. Pupa free or enclosed in the larval skin ; 

 in either case escaping through a T-shaped 

 slit on the dorsal side of the anterior end, 

 or rarely through a transverse rent between 

 the eighth and ninth abdominal rings. The 

 imago has no frontal lunule or ptilinum. 



Nematocera. Antennae usually composed of more than six dis- 

 tinct segments, the two basal ones frequently 

 differentiated from the rest; the flagellum 

 always consisting of six to twenty or more 

 segments, similar to one another. Palpi four 

 or five jointed, and nearly always filiform. 

 Wing venation simple or complicated. Lar-vae 



