858 REPORT OF THE HARVARD AFRICAN EXPEDITION 
and is enclosed in a semi-transparent, brown, pocket-shaped cocoon, about 3 mm. 
in length. Projecting above the edge of the cocoon is a pair of white to greyish 
white respiratory appendages. Each appendage consists of three leaf-like proc- 
esses, two of which are united at the base.’ This brief account and the rather 
crude figure are sufficient to conclude that this supposed pupa of H. damnosum 
is totally different from the one I have described above from Liberia. 
Pomeroy (1920) has given a much more complete description of a pupa sup- 
posedly of HL. damnosum, from specimens collected in Morogoro, Tanganyika 
Territory. He states that some of the pupae contained male imagos which were 
dissected out and compared‘ with emerged adults, bred from the same locality 
and at the same time. Alsoysome females bred from the same lot of pupae were 
compared with the type of #. damnosum at the British Museum. His descrip- 
tion is as follows: ‘‘The filaments are rather pale and translucent in structure. 
They are composed of eight main lobes, bulbous and finger-like. The cephalic 
and caudal lobes very broad in the middle, pointed toward the apex. These two 
lobes are very often found split up the centre. The remaining six arise from the 
base of the main stem in pairs, and in some specimens a short broad secondary 
finger-like filament is present attached to one of the middle filaments, usually 
the first cephalic pair, about halfway up.’’ These statements as well as Pom- 
eroy’s figure cannot possibly be applied to my Liberian pupae nor to those 
described by Roubaud. They seem to fit much better the pupae studied by 
King. 
It will remain for future investigators to decide which are the true early 
stages of Husimulium damnosum. 
TABANIDAE 
The present study of African Tabanidae is not merely an enumeration of the 
species collected during the Harvard African Expedition. As I have had an 
opportunity to examine much material of this family from the Belgian Congo,’ 
I have included a synopsis of the species known from that territory, with keys 
to enable their identification. I have added some critical remarks on the classi- 
fication, on the limits of the genera and subgenera, as well as a key to the genera 
found in the Ethiopian Region (including the Malagasy Subregion). 
In attempting to name tabanids with the subjoined keys, only fairly well 
preserved specimens should be used, preferably such as have been pinned in the 
field. The relaxing jar will often spoil these flies beyond recognition and they 
can but seldom be correctly identified after having been preserved in a fluid. 
1 The caption “Simuliwm damnosum” in Faust’s ‘Human helminthology’ (1929, p. 551, fig. 271) 
is evidently due to an oversight. The figure is a copy of a drawing by Sikora in Martini’s ‘Lehrbuch der 
medizinischen Entomologie’ (1925, p. 186, fig. 120), which, however, is merely labelled ‘“Simuliwm.” 
Most probably the drawing represents the early stages of some European species. 
2 I wish to thank my friend, Dr. H. Schouteden, Director of the Congo Museum at Tervueren 
Belgium, for kindly communicating to me much valuable material. I have also been able to study one 
tabanids obtained in the Belgian Congo by the Lang and Chapin Expedition of the American Museum 
of Natural History, as well as some African specimens of the United States National Museum and the 
Museum of Comparative Zoélogy of Harvard University. 
