110 
Psyche 
[Sept -Dec. 
wings, with a complete venation, and are as good fliers as any 
of the Hippoboscidse. Some species, however, such as the In- 
dian O. comosa Austen, show a decided shortening of the 
wing, though the venation remains normal. The further reduc- 
tion of the wing to a functionless condition apparently pro- 
ceeded along two different lines. In one, leading to Stenepteryx , 
the wing retained its length, but reduced its area by losing most 
of the membrane, while some of the veins disappeared and the 
remainder were crowded near the costa. In the other, leading 
through Crataerina and Myophthiria to Brachypteromyia , the 
whole wing was shortened, with concomitant loss of most of the 
venation. All known species of these four subapterous genera 
appear to be incapable of flight. Moreover, the reduction of 
the wings was accompanied by other changes which affected 
nearly all parts of the body. The most striking of these are the 
lengthening of the head, the reduction of the eyes, the loss of 
the ocelli, the obsolescence of the dorsal thoracic sutures, the 
gradual disappearance of most of the tergal plates of the ab- 
domen, and the unusual development of the legs and particu- 
larly of the claws. 
The restricted choice of a host and even more so the type of 
host selected were no doubt the main factors which determined 
this orthogenetic evolution. Most species of Ornithomyia show 
very little host specificity, being found mainly on a variety of 
passerine and some game birds. Some species, however, are 
restricted to swallows and martins (Hirundinidse) and O. 
comosa is one of these. It may be significant that typical 
Stenepteryx hirundinis (Linne) occurs on swallows; while a 
form regarded either as a race of hirundinis or as a distinct 
species ( cypseli Rondani) lives on swifts (Apodidae or Micro- 
podidae). All known species of Cratcerina , Myophthiria and 
Brachypteromyia , on the other hand, are restricted to the 
Apodidae. It seems reasonable to assume that the ancestors of 
these swift-flies were swallow-flies and that the change of hosts 
— from swallows to swifts — was in some way responsible for 
the series of morphological changes eventually culminating in 
Brachypteromyia. As in most such cases, the mechanism of these 
changes is a matter of speculation; but the end result is readily 
recognized as an efficient adaptation to the peculiar morphology 
and habits of the host. 
Swallows and swifts are amongst the most aerial of birds, 
