XLI. FAMILY AGROMYZID^B. 



Front broad, with or without bristles. Antennae short, 

 the third joint usually rounded, sometimes a little elon- 

 gate or subquadrate; oral vibrissae usually present. 

 Arista bare or pubescent, never distinctly plumose; rare- 

 ly wanting. Genitalia rarely prominent. Wings broad; 

 auxiliary vein vestigial or indistinct, never clearly sep- 

 arated, save sometimes in its proximal part, from the 

 first vein. Second basal and anal cells always small, 

 oftentimes indistinct, or the second basal united with the 

 discal ; cross-veins often much approximated, never very 

 remote from each other. 



This family of small or minute flies, as here defined, 

 includes four or five groups which various authors have 

 either given independent rank or united with other 

 groups. Czerny would unite the Ochthiphilinae with 

 the Sapromyzidae, in which view I do not concur. The 

 limits of the Agromyzinse and Milichinse, if there be any, 

 will only be determined for our American genera by a 

 more exhaustive stud}' than I can give to them; several 

 of the genera I do not know: Eusiphona^ Hemeromyia, 

 Arctobiclla, Patodinia. Czerny and Hendel make different 

 combinations than do most other authors. The Agro- 

 myzinse, according to these authors, have divergent post- 

 vertical bristles, the Milichinse and Ochthiphilinae con- 

 vergent. 



From the Drosophilidae the members of this family 

 will be distinguished easily (in the American forms) by 

 the nonplumose or nonpectinate arista. Nearly all the 

 genera are at once distinguished from the Oscinidae and 

 Ephydridae by the distinct, though small, posterior basal 

 cells; but this character is sometimes difficult to detect 



291 



