230 NEW YORK STATE MUSEUM 



The " subcostal cross vein " of many systematic dipterologists 

 is not included in this list because it is not a cross vein at all, 

 but the free part of Sc^ as already pointed out. The basal one 

 of these five cross veins, the humeral, is very constant in posi- 

 tion and relations, only disappearing by atrophy when the sub- 

 costa, which it braces^ atrophies. 



Of the four distal cross veins, the two innermost (r-m and m-cu) 

 lie between principal veins and effect strong and permanent unions. 

 The other two {r and m) lie in the middle of the principal forks 

 of the radius and of media respectively, usually fall outside the 

 cord and are far less constant. These two disappear by atrophy; 

 the former, only by fusion of veins together, making the joinings 

 stronger. The radial cross vein may enter strongly into the forma- 

 tion of the cord, as in Conosia [pi. 21, fig. 5] and the more typical 

 Eriopterini [pi. 23] or it may be quite eliminated as in many of 

 the Rhamphidiinae [pi. 29] and Paratropeza [pi. 21, fig. 4]. It 

 rarely disappears through the fusion of adjacent veins, as in Lechria 

 [pi. 19, fig. 5] ; more often it is eliminated by fusion from the apex 

 backward as in the Cylindrotomini [pi. 15, fig. 4, 5, 6]. Its posi- 

 tion in relation to the forking of the radial sector is very different 

 in different crane flies ; and in this there probably lie unused generic 

 characters of value. 



The radio-medial cross vein always enters into the formation of 

 the cord, being always present or accounted for. It disappears 

 only by fusion of adjacent veins upon it. That fusion may be 

 brief, as in Rhamphidia [pi. 14, fig. 4] or more extensive, as in 

 Ptilogyna [pi. 17, fig. 4], Liogma [pi. 15, fig. 5], Mongoma [pi. 21, 

 fig. 6], or Paratropeza [pi. 21, fig. 4], each representing a different 

 group, and all highly specialized. It is situated at an unusual dis- 

 tance from the base of the wing in Conosia [pi. 21, fig 5], being 

 beyond the median cross vein. 



The median cross vein usually lies without the cord (except in 

 the case of Conosia, just cited) and when far without, it appears 

 to be relieved of great responsibility and tends to disappear along 

 with the superfluous branches of the median vein. As a 

 rule it disappears in advance of the atrophy of M3. Owing 

 to the upward deflection of the base of M3, this cross vein 

 and the deflected portion of that vein equally support the 

 tip of M3 in very many cases; and either may be eliminated, 

 leaving the tip supported on the other. In fact both may go, and 

 leave the tip hanging in the membrane unsupported, as illustrated 



