42 BRITISH FLIES 



North American Diptera (1905) excepting in the exchange of places 

 between the Leptidfn and Tahanidm, and in the fact that Ahhich uses 

 no snperfamiUes or subfamilies ; Aldrich, however, reversed the sequence 

 of the genera in the Slraiiomyidm. The main points upon which I have 

 differed from Osten Sacken are as follows : — 



Firstly ; I include the Nemcstrinidm and Cyrtid.oi in the Erpmoch/eta 

 instead of doubtfully leaving them in th(j TiiOMOPTEKA, and my reason 

 for this is that those two families agree with the Ere.mocili;ta of Osten 

 Sacken in being absolutely eremochtetous and in having three pulvilli. 

 1 consider tlio combination of these two impftrtant characters separates 

 them naturally and scientifically from the other TiiOMOPTKUA, while it unites 

 them with the Vaivmocmmyx. My interpretation of the EnEMOCHynw 

 exactly coincides with Brauer's HoMfEODACTYLA, and, as Jirauer's system 

 was mainly founded on embryonic characters, the coincidence becomes a 

 very strong confirmation of its accuracy, i prefer Osten Sacken's title 

 for the group because it is founded upon the perfect insects, besides being 

 a better word. The Tkomopteka (as interpreted Ijy me) are composed of 

 (jnly the Bomhylvhn and Therevidm, as I remove the Hixnopinido: to the 

 next superfamily even though I believe in their affinity t(j the Thercvidm ; 

 in this ari'angement there is great divergence from Brauer's views. 



Secrmdly ; I have differed from Osten Sacken in removing tlii! 

 Hcenop'midw from the Tjiomoi'TEJia altogether, and in uniting them with 

 the Mydaidm in a common superfamily, for whicii I jiroposo the name 

 of Dermatina. Oston Sacken had already treated the ]\fydaid(n as a 

 superfamily in themselves, and although Brauer widely separated the 

 Hnenopinidji: and Myddidc. in the arrangement 1 have (j noted above, he 

 had in the ])reviou8 year most elaborately worked out the rolationshii* 

 of Sciincidnus and proved over and over again its affinity to the ^fyd((id(l', 

 in fjict so much so that he wrote "Jene kur/fiihlerigen Mydaiden 

 " scheinen den unzweifelhaftosten Ubergang ym ^cenopiv/itH zu zeig(;n " ; it 

 is again remarkable that when working on independent lines we have 

 arrived at the same result. The term Dermatina (oep/uLUTivo^) refers to 

 the leathery appearance of most of the species, which are however not 

 leathery like the Pupirara but in the sense of a dull dark smooth br^ly- 

 surface entirely devoid of bristle and pubescence. 



Thirdly; I have acknowledged the family rank of the Apiocerido:, as I 

 cannot associate them with the Dermatina ])ecause they arc distinctly 

 chajtophorous, nor is it possible to associate them with the Thcremdm 

 because of tlu; v(;)y distinct venation, in which the subc(jstal vein is long 

 and the veins on the apical part of the wing are curved up so much that 

 tli(; upper veinlet from the discal cell ends before the wing-tip. Osten 

 Sacken at one time contended that they were true A&ilinm (allied to some 

 species of Erax), but he subsequently considered that they were entitled 

 to be considered a subfamily of equal rank with the Asilince. I am 



