ENTOMOLOGY 949 
Sterrhocera and Archiplatius being unknown to me, will not be discussed here. 
The characteristics of Tylopelma, Potisa, and Parhaematopota are obvious enough 
in the respective type species. If, however, one examines a large number of 
species, he finds within the genus Haematopota such a bewildering variety in the 
shape of the antennae, that it seems impossible to arrange all the species satis- 
factorily into the groups defined in the foregoing key. I, therefore, at present 
fully subscribe to Austen’s statement that ‘‘no valid division of the genus 
Haematopota, at any rate into categories higher than groups or subgenera, is 
possible, since, although within the limits of the genus great differences exist in 
the shape of the frontal callus, antennae, front and hind tibiae, ete., it would be 
difficult to find two species showing identical differences from the genotype, yet 
all are united by the well-known, highly characteristic, and distinctive wing- 
markings, as well as by a general facies” (1912, Bull. Ent. Res., II, p. 405, 
footnote). 
According to Enderlein’s key, 7'ylopelma is based upon the combination of 
the disciform basal division of the third antennal segment with the presence of a 
transverse furrow before the apex of the first antennal segment. His generic 
diagnosis, however, says: ‘‘This genus differs from Haematopota Meigen, 1803, 
as follows: first antennal segment on the upper side with a sharp transverse 
furrow before the apex, forming a transverse swelling at the apex, which, seen 
from above and in profile, gives almost the impression of a distinct antennal 
segment.’ Unfortunately, Enderlein overlooked that this is precisely the 
structure of the first antennal segment in the genotype of Haematopota, H. 
pluvialis (Linnaeus); but in that species the basal division of the third segment 
is not disciform. ‘The following African species have a more or less distinct 
notch before the apex of the first antennal segment: H. decora Walker, H. 
angustipalpis (Enderlein), H. confluens (Enderlein), H. coronata Austen, H. 
fasciatapex Edwards, H. mactans Austen, H. ochracea Bezzi (= Tylopelma patelli- 
corne Enderlein), H. frilleborni (Enderlein), H. nigripennis Austen, H. niveripalpis 
(Enderlein), and H. vittata Loew (= H. pulchrithorax Austen). Not all of these, 
however, have a disciform basal division of the third antennal segment, while 
some have fringed legs and might equally well be placed in Austenia. 
The following African species have a broad, disciform basal division of the 
third antennal segment and no furrow before the apex of the first: H. brunnes- 
cens Ricardo, H. cognata (Grinberg), H. corsont Carter, H. edax Austen, H. 
exiguicornuta Carter, H. harpax Austen, H. hastata Austen, H. ingluviosa Austen, 
H. pinguicornis Carter, H. rufula (Surcouf), H. sanguinaria Austen, and probably 
also H. lathami Surcouf. Some of these, however, do not have the peculiar 
shape of the first antennal segment of Parhaematopota cognata. H. harpax, for 
instance, is plainly transitional to Potisa, although the basal division of the third 
antennal segment is somewhat angular above. 
Potisa was entirely based upon the peculiar shape of the antennae, which 
Surcouf described as follows: ‘‘First antennal segment as long as the head, 
swollen from the base on, almost cylindrical; second segment very small, cres- 
cent-shaped; third segment with the basal division ovoid, swollen, slightly nar- 
