82 
Psyche 
[June-Sept. 
the late Major Austen, I was able to examine the holotype 
and paratype at the British Museum. This species, which 
has not been figured, is close to H. longipennis , the most im- 
portant differences being given in the key. In addition, the 
vertex is somewhat narrower than in longipennis, with the 
inner margins slightly converging toward the occiput; the 
postvertex is shorter; the inner orbits (or parafrontalia) 
are narrower and of more uniform width throughout; the 
fronto-clypeus also narrower. The insect is mostly reddish- 
yellow ; but scutellum, postvertex and fronto-clypeus are al- 
most wholly pale ivory-yellow. I have seen also a male from 
Tanganyika Territory (West shore of Lake Manyara). 
This sex is almost exactly like the female. It has eight 
strong pale-colored preapical bristles on the scutellum (also 
present in both types), placed in a single row and widely 
divided into two groups of four each. The structure of the 
abdomen is not known, but is probably similar to that of 
equina and longipennis, with minor differences in the rela- 
tive size of the median tergal plates. 
In the male of fulva examined, the parameres of the 
genitalia are similar to those of equina and longipennis, but 
the terminal point is blunter. H. fulva and H. martinaglia 
are the smallest members of the genus. 
4. Hippobosca variegata Megerle von Miihlfeld, 1803, 
(actual date of publication!), Appendix ad Catal. Insect. 
Nov. 1802 Viennae Austriae Vendita, p. [14] (unnumbered) 
(Bengal) . 
Synonyms: H. maculata Leach, 1817; H. bipartita Mac- 
quart, 1843; H. aegyptiaca Macquart, 1843; H. fossulata 
Macquart, 1843 ;* H. sudanica Bigot, 1884; H. sivae Bigot, 
1885; H. calopsis Bigot, 1885; H. aegyptiaca var. benga- 
lensis Ormerod, 1895. 
H. variegata appears to be the oldest valid name for this 
species, the date 1823 given by Wiedemann being erroneous 
(see Schenlding 1935, Arch. Morph. Taxon. Entom., Ber- 
lin-Dahlem, II, p. 156). The original description is of the 
briefest: “ex Beng. Aff. equin. sed maj. magisque varieg. 
(1 Exemplar).” Yet it is sufficient to validate the name, 
especially in view of the fact that Wiedemann (1830, 
iMacquart’s three names should be dated 1843, when the Mem. Soc. 
Sci. Lille for 1842 were actually published. 
