ENTOMOLOGY 863 
cism. Nevertheless, I venture to present a tentative grouping embodying my 
views of the natural relationships between the several genera. 
I recognize among the Tabanidae three main divisions, which I shall rank 
as subfamilies, although I am fully aware that one of them (Coenomyiinae) 
is farther removed from the rest of the family than the two other subfamilies 
are from each other. The proper solution of this difficulty might be to raise 
the Coenomyiinae to family rank; but this course would obscure the signifi- 
cance of this particular group as a connecting link between the remaining 
Tabanidae and the Leptidae. 
1. Subfamily Comnomyimnak. — Eyes broadly separated by the frons in 
the female, contiguous (holoptic) in the male. Ocelli present. Proboscis short, 
well-developed; palpi normal. Third antennal segment composed of eight 
divisions. Hind tibiae with two spurs. Wings with the anal and all posterior 
cells broadly open; sixth longitudinal vein (An) strongly wavy; squamae 
present, though smaller than in most other tabanids. This subfamily comprises 
Coenomyia Latreille, Pelecorhynchus Macquart, and Coenura Bigot, none of 
which are represented in Africa. 
These three genera form the transition between the Tabanidae and the 
subfamily Xylophaginae among the Leptidae. The next steps in the passage 
are provided by Arthropeas Loew and Glutops Burgess. Whether any of the 
other genera included by Kertész (1908, ‘Cat. Dipt.,’ III, pp. 141-146) in his 
family Coenomyiidae, are related to Coenomyia I am unable to state.1. Although 
I cannot enter into a further discussion of this group, I may point out that 
Bigot and Philippi placed Coenura in the Coenomyiidae (which Bigot regarded 
as a division of his Tabanidi); Schiner regarded it as allied to Arthropeas and 
Osten Sacken placed it in the Xylophagidae. On the other hand, Gerstaecker 
and Brauer were positive that it was one of the Pangoniinae. 
2. Subfamily PancontInaE. — Eyes more or less broadly separated by 
the frons in the female, separated or contiguous in the male. Ocelli either 
present or absent. Proboscis of variable length, often longer than the head, 
in some cases aborted. Hind tibiae with two spurs, which sometimes are very 
small. Wings with the sixth longitudinal vein (An) straight. 
The numerous genera of this subfamily appear to form three fairly natural 
groups which I shall call tribes. 
2a. Tribe PANGonrIINI. — Eyes hairy or bare, not spotted or streaked in 
life. Ocelli present or absent. Proboscis and palpi well-developed, the former 
of variable length. Antennae short, the second segment at most half as long 
as the first. This tribe contains most of the genera of Pangoniinae. 
2b. Tribe Curysoprni. — Eyes bare, in life with spots or zigzag streaks. 
Ocelli present. Frons of female at least as wide as long, often wider. Pro- 
boscis and palpi well-developed, the former at most slightly longer than the 
head. Antennae elongate; the first segment slender or swollen; the second 
1 T have recently learned in correspondence that Dr. J. M. Mackerras, of Sydney, has independently 
reached the conclusion that Pelecorhynchus belongs in a distinct subfamily from the remainder of the 
Tabanidae. 
