2 PROCEEDINGS OF THE NATIONAL MUSEUM tol. 73 



hifasciata Fabricius and unifasciaia, new, as well as eight from the Old 

 World. Brauer and Bergenstamm apparently indicated hifasciata as 

 type, and this species was definitely designated by Coquillett,^ Latreil- 

 lia being preoccupied, LatreiUimyia was proposed to replace it, taking 

 the same genotype. Willistonia had but one originally included species, 

 Musca esuriens Fabricius. Goniomima had but one originally included 

 species, Belvosia luteola Coquillett, which was also designated as type. 

 Triachora had but one originally included species, Latreillia unifasciata 

 Robineau-Desvoidy, which was also designated as type. Belvosiomima 

 had but one originally included species, fosteri, new, which was also 

 designated as type. Belvosiopsis was proposed for the single species 

 hrasiliensis, new, also designated as type; this I place as a synonym 

 of Belvosia leucopyga Van der Wulp. 



Thus it appears that there are seven different generic names to 

 consider in this group, each based on a different type species. I have 

 studied with care all the type species, with reference not only to the 

 generic characters originally mentioned but to others which might 

 have this value. In nearly 40 species, represented by about 700 

 specimens, there is such a complete blending of the characters that I 

 can not find one which will divide the mass in a consistent manner; 

 nor do I see how a satisfactory division can be made into half a dozen 

 or more genera. The genera have all been made on too few species, 

 and the describers attribute higher value to length of antennae, dis- 

 tance of vibrissae from oral margin, etc., than I can find. It is true 

 that certain species when compared with each other seem generically 

 distinct in the absence of others which connect them ; and I thought 

 for some time that two or three genera could be maintained. The 

 genitalia are very homogeneous and the reproductive habits are the 

 same throughout, as far as known. 



Williston; in his 1893 paper, discussed the characters in an able 

 manner, and showed that those used by Brauer and Bergenstamm 

 were not of generic value. He saw that if they were even specific 

 there must be many more species than were then supposed, but left 

 the question of specific limits for future study. 



The generic characters of Belvosia as here taken are as follows: 

 The head is uncommonly wide, but short, the anterior surface flat- 

 tened; length at vibrissae not much less than at antennae; front broad 

 in both sexes; eyes bare; ocellar bristles absent, frontals varying widely 

 in the sexes and in different species, from three irregular rows to one 

 nearly regular; proclmate orbitals present in all females, and in the 

 males of luteola, unifasciata, fosteri, ochriventris , slossonae, and equi- 

 noctialis ; vibrissae considerably above the oral margin, the distance 

 in general about the same as the length of the second antennal joint; 

 parafacials broad, bare; facial ridges always bristly to nearly one-half 



> Proc. U. S. Nat. Mus., vol. 37, 1910, p. ."iSS. 



