British Bracoriidse. 33 



areolet in the hind wings. The 5 of J., ephippium is apterous. 

 Abdomen of the 9 subpetiolate, lanceolate, longer than the head and 

 thorax ;. rounded at the apex, and sometimes a little spathulate in 

 the jj ; 1st segment linear, occupying at most j or | of the total 

 length of the abdomen, rarely constricted in the middle by reason of 

 the prominence of the spiracular tubercles ; 2nd suture distinct, 

 covered with a loose and translucent membrane which allows free- 

 dom of movement to the abdomen, especially in the act so familiar 

 to these insects of doubling back the abdomen under the thorax ; 

 this suture always forms a conspicuous pale mark on the abdomen ; 

 ventral valve of the $ without appendages. 



This genus, after the dismemberments above described, 

 still comprises the vast majority of the tribe. Some further 

 attempts to break up the genus have been made by Forster 

 (Synopsis, p. 247), but they are too partial and vague to be 

 really available, and are rendered useless besides by the 

 existence of transitional forms of doubtful position. The 

 number of species cannot be estimated with any approach 

 to correctness, but it is probably not much less than that 

 of the Aphidse, comprising therefore many different forms 

 respecting which we have at present no information, 

 Haliday's work, which is the most comprehensive, describes 

 twenty-four species, and the author declares that his only 

 object was to establish a few sections, by the help of which 

 future naturalists might make a more considerably advance 

 in the subject. Most of these sections, converted into 

 genera,have been detailed above; they constitute the princi- 

 pal and most obvious divisions into which the tribe can be 

 separated, and have every appearance of permanence. If 

 we seek to extend further the system of dismemberment, 

 we find, so far as I have been able to make out, only more 

 trifling characters, confined to a single species, and there- 

 fore improper for the establishment of good genera. The 

 males of Aphidiics differ much from their respective 

 females, while at the same time they resemble each other 

 so closely, that their separation by inspection of individuals 

 captured at random is almost certain to lead to error ; it 

 is necessary therefore to breed both sexes together, which 

 can generally be done without difficulty. The first table 

 therefore contains males and females of the correctness of 

 whose association there can be no doubt, and this category 

 fortunately comprises the majority of the species. A 

 second dichotomy will be devoted to a few males not 



TRANS, ENT. SOC. LOND. 1899. — PART I. (ApRIL) 3 



