294 



NORTH AMERICAN DIPTKRA. 



in such small insects, and will usually require the use 

 of a compound microscope; indeed such a microscope, 

 with a one inch or half inch objective, is advised for the 

 study of most of the species of the family. The absence 

 or presence of the basal cells is not an important charac- 

 ter in these and allied flies, and may not even have spe- 

 cific value. Aulacigaster has been located in various 

 families, but it seems to find its most natural place here. 

 From the Geomyzidae the distinction of some of the gen- 

 era is very difficult, if not impossible at present. One 

 would better consult that family in case of doubt. 



The larvae of Phytomyza, and probably also of Par- 

 amyia, are leaf miners; those of Ochthiphila have been 

 found in the galls of Triticum repens. The larvae of Agro- 

 myza are elliptical in shape, the hind stigmata situated 

 upon small rounded plates on the under side of the last 

 segment; the abdomen is provided with false legs, with- 

 out bristles. The larvae of Lcucopis are cylindrical, 

 thicker posteriorly; the skin roughened with short hairs; 

 hind stigmata elongate, tube-like and widely separated. 

 The larvae creep leech-like, or like geometrid larvae. 



TABLE OF GENERA. 



1. Posterior cross-vein absent (Phytomyzinae) . .... 2 

 Posterior cross-vein present. ....... 3 



2. Proboscis elongate, folding; palpi long and tbickened (18). 



Paramyia. 

 Proboscis and palpi not elongated (16). . . Phytomyza. 



3 . First posterior cell narrowed in the margin; proboscis long and 



geniculate.* Eusiphona. 



First posterior cell not or but slightly narrowed in the margin. 4 



4. Posterior cross-vein situated before the middle of the wing, the two 



cross-veins approximated 5 



The cross-veins not approximated. ...... 12 



* I do not know this genus, originally described as a tachinid; it 

 must have a curious resemblance to Slylogaster (Conopidae). 



