312 
Psyche 
[December 
races have been derived. According to C. Keller (1902), 
the ancestor of the heavy races of horses was a wild equine 
which lived in Europe during the Pleistocene ( Equus cabal- 
lus Linnaeus) and even survived until historic times. The 
lighter races, on the other hand, had an Asiatic origin and 
were most probably derived from Equus przewalskii Poli- 
akoff, a wild equine of Central Asia, of which a few speci- 
mens may be living yet in a feral state in Dzungaria. I 
am inclined to the view that both these ancestral species 
were infested with H. equina, although, of course, this 
is a mere hypothesis. It may yet be possible to find the 
insect on the living E. przewalskii. 
Affinities. — H. equina and H. capensis appear to be 
very closely allied, although there can be no doubt that 
they are specifically distinct. After a very careful study, 
I have found that the only structural difference, which 
seems to be entirely reliable, is found in the shape of the 
fronto-clypeus. I had thought at first that the wing vena- 
tion afforded some additional distinctive features. The 
relative distance from the tip of the second longitudinal 
vein to that of the first and third longitudinals appears 
to be highly variable. The length of the anterior basal 
cross-vein (or second basal cross-vein; m-cu or M 3 ) is per- 
haps more reliable: in H. capensis this cross-vein is fairly 
straight and, as a rule, about as long as the distance from 
its tip to the anterior cross-vein (r-m) ; in H. equina it 
is more curved and generally much longer than the dis- 
tance from its tip to the anterior cross-vein. 
In one male of H. equina, from Sardinia, the left wing 
has a double anterior cross-vein, enclosing a small super- 
numerary cell ; the right wing is normal. 
A further specific character may perhaps be found in 
the size and shape of the sclerotized upper plate of the 
vertex (vertical triangle). In all the 29 specimens of 
H. equina examined, this plate occupies about the upper 
third of the vertex, being at most half the length of the 
medio-vertex (or frontalia) ; it is almost semi-circular in 
outline and much wider at the occiput than long on the 
middle line. In most of the specimens of H. capensis seen, 
