AMERICAN DIPTERA. 



197 



teron lies in the structure and neuration of the wings. We are 

 mainly assisted by the venation in assigning a fly its systematic posi- 

 tion, and this is because the veins in an insect's wing have become 

 fixed through long periods of evolutionary change. If we take 

 venation as our chief criterion, the family Empididse consists of an 

 incongruous assortment of genera. This is much more marked than 

 in most other families of Diptera.so much so that what is now given 

 as one family was formerly divided into three. In groups like the 

 Dolichopodidc^ or Syrphidse venation is of paramount importance, 

 and exhibits relatively little variation for all the members of the 

 family. But in the Empidid^e the number and the arrangement of 

 the veins may vary considerably. The larger genera have the full 

 numl)er of veins and cells, but in the smaller forms there is a basal 

 coalescence of the posterior veins, until the anal cell vanishes and 

 the posterior cells become reduced to two. The next character in 

 which Empidid^ are variable is the structure of the antennae. Here, 

 again, the larger genera have normal three-jointed antennse, of which 

 the third joint is typically longer than the first or second, and bears 

 a variable terminal style or arista. In many of the smaller genera 

 the antennse are reduced in length, the first and second joints may 

 become fused into one, and the arista may become even sub-dorsal, 

 in which case these insects, superficially at least, bear a close resem- 

 blance to the smaller Muscidse, a resemblance which is heightened 

 by the similarity of the wings. 



Work on the homology of the parts of the male hypopygium in 

 the different genera and families of flies has made but little advance 

 as yet. That much is to be accomplished in this line may be judged 

 from the similar work in other groups, as for example, the happy 

 results obtained by Prof. J. B. Smith from the study of Lachnosterna, 

 and especially since the Erapididse present so large a number of 

 interesting and anomalous forms. Even in a single genus, we may 

 cite here Platypalpus or Empis, there is a wide range in the plasti- 

 city of similar parts. For this reason we have laid greater stress 

 than usual upon the conformation of the male genitalia, relying not 

 so much, however, on verbal description as upon the series of figures 

 of the various forms. 



With these words of introduction we may come to a definition of 

 the family. EmpididiB are distinguished as such from all other flies 

 by the following conjbination of characters : Body chitinous ; an- 



TRANS. AM. ENT. SOC, XXVIII. •'^^^' ^^^^ 



