988 REPORT OF THE HARVARD AFRICAN EXPEDITION 
the abdominal spiracles are clearly placed in the membrane between the ter- 
gites and sternites, while in the other calyptrate Muscoidea (with the excep- 
tion of the Gasterophilinae), they are placed in the second to fifth tergites. 
Furthermore, the chitinized prosternal plate is lacking, and the metasternum 
has a unique development, being on a level with the mesosternum and pro- 
jecting forward so as to separate the mid-coxae. Malloch suggests that these 
peculiarities are perhaps correlated with the highly specialized and strictly 
blood-sucking habits of tsetse-flies, while the development of strong bristles on 
the bases of the fore coxae may help protect the swollen base of the proboscis. 
The tsetse-flies show a number of additional, important characters that 
set them off from the other Muscoidea Calyptrata. The feathered or plumose 
hairs of the antennal arista are unique among the Muscoidea. The unusual 
course of the fourth longitudinal vein of the wing, with its strong downward 
bend before the strongly oblique anterior transverse vein, is also unparalleled 
in the group. 
The proboscis is very highly specialized, consisting of a swollen basal bulb 
and an extremely slender needle-like rod (which includes the labium, labrum, 
and hypopharynx). At rest it is completely ensheathed by the long and slender 
palpi. While the proboscis is built essentially like that of the other blood- 
sucking Muscoidea, its finer structure is separated by a broad gap from that 
of the nearest relative, Stomoxys. The females are pupiparous? and as a result 
their genital organs are profoundly modified. The male genital armature is 
completely hidden within the hypopygium or peculiarly modified eighth abdomi- 
nal segment. 
The tsetse-flies have generally been placed in the subfamily Stomoxydinae 
of the Muscidae. While it is true that they show more similarity to Stomoxys 
than to any other living genus of flies, it is doubtful whether this is indicative 
of any true relationship. Considering the high antiquity of the genus Glossina, 
which during Miocene times was represented by several species seemingly as 
highly specialized as any living at present, there is every reason to believe 
that it forms a distinct off-shoot in the evolution of the Calyptrata. Stomoxys 
and its relatives, Haematobia, Stygeromyia, Bdellolarynx, and Haematobosca, 
are most probably a much more recent branch of the Musca stock, having no 
genetic relation with Glossina. 
Attention may be called here to a hatched puparium of a muscid fly which 
was found under dry leaves in a forest clearing at Camp No. 3, on the Du River, 
Liberia. This puparium agrees with that of a tsetse-fly in size and in bearing 
at the posterior extremity a pair of almost spheroidal lobes separated by a 
narrow, but deep, notch. That this puparium is, however, not of a species of 
Glossina is shown by several peculiarities. The several segments are clearly 
set off by slight constrictions. ‘The external surface is not uniformly dull and 
shagreened, but rather smooth and shiny (at any rate under a handlens) and 
1 While the female tsetse-flies actually deposit full-grown larvae and not pupae, it seems neverthe- 
less more appropriate to call them pupiparous rather than larviparous, so as to distinguish their case 
from that of the many Diptera which deposit young larvae, freshly hatched from the eggs. 
