OF THE TABANID^ OF THE UNITED STATES. 425 



7. The T.abani with imicolorous eyes belong to several distinct groups : — 



a. T. melanocerus and nigricornis. 



h. T. Orion and Actceon seem also to have imicolorous eyes, at least I have not 

 been able to revive the colors by moisture. 



c. T. americanus has uniformly green eyes ; those of T. giganteus are also uni- 

 colorous. 



8. T. Reinioardtu has two bluish green stripes, with a rather broad interval, the lower 

 one bent towards the upper one at tlie outer end, Avithout reaching it ; thus, its pubescent 

 eyes notwithstanding, it resembles in this respect tlie Taljani of the first group. 



9. T. cerastes has a single blue crossband across the middle of the eye. 



10. The whole subgenus Therioplectes has three or four green or bluish green cross- 

 bands on purple ground ( T. zoncdls and cinctus have four dull green bands, T. trlsjnhis 

 four bright ones, etc.). 



These instances plainly show the connection which exists between the coloring of the 

 eye and the relationship of the species ; but they show at the same time the great variety 

 of types which exist within the genus Tabanus. 



The subdivision of the genus Tabanus into smaller genera, natural groups or sub- 

 genera, lies beyond the scope of the present paper. In order to be carried out success- 

 fully, such an attempt would require the study of the species from all parts of the world. 

 Tabanus, on the whole, is a remarkably compact genus, but with a great variety of specific 

 types. The wiser course, perhaps, will be to treat the subdivisions as subgenera only, pro- 

 viding them, however, with separate names, as has been done in tlie genus Trypeta. I will 

 confine myself here to a few remarks about the subgenus Therioplectes, proposed by Dr. 

 Zeller in 1842 for the Tabani with pubescent eyes. The subgenus thus defined contains 

 some heterogeneous elements in Europe {T. I'tistiais niidfitlvics), and still more so in North 

 America. We imj^rove the definition of this group, however, through the use of a charac- 

 ter which has been hitherto neglected, but which seems to be a character of higher order, 

 that is, more indicative of a natural relationship than the pubescence of the eyes ; I mean 

 the presence of a more or less distinct ocelligerous tubercle on the vertex. Often a rudi- 

 mentary ocellus is perceptible upon the tuljerde ; sometimes quite distinctly (for instance, 

 in T. cuictus, as was already noticed by Meigen). This character, more distinct in the 

 female, is often also visible in the male. The ocelli seem to be rudhnentary, and should 

 not prevent the absence of ocelli from being quoted among the characters of the genus. 

 The new definition, as far as applicable to the American species, would run thus : 



Therioplectes. Eyes pubescent ; ocelligerous tubercle more or less distinct; eyes ( ? ) 

 with three or four bright green or bluish crossbands ; head of male not differing much in 

 size and shape from that of the female ; the difference in size between the large and small 

 facets upon it is but very moderate, and hence the line dividing them rather indistinct 

 (especially in dry specimens). 



Thus limited, Therioplectes forms a rather homogeneous group of twelve American sj^e- 

 cies (Nos. 43-54). The two species T. hicolor and fithescens, as well as their relatives, 

 the European T.fulvus and ricsticus, would be excluded from it and form a separate group, 

 which I propose to call Atylotus, and which may be characterized as follows : — 



MEMOIRS B03T. SOO. NAT. HIST. VOL. n. 107 



