1908.] 165 



not be a true Pftora after all. However, in the number of bristles oil 

 the scutellum we have a thoroughly satisfactory character, and I have 

 taken advantage of it to separate the species into those with four or 

 more bristles (Section B), and those with only two. The former run 

 to nearly twenty forms, leaving the large number of seventy, or there- 

 abouts, to be dealt with in the latter. These T further break up 

 according as the wing cost a is long (Section C) or short (Section D). 

 This character, however, has the disadvantage, common to all com- 

 parative qualities, of leavinj a doubt sometimes into which division 

 an insect should go ; but at the same time it has the advantage of 

 being an easily come-at-able one, for in whatever way an entomologist 

 may pin or set his insect, the wings are always in view. 



On the whole the arrangement leads, I think, to a fairly natural 

 grouping of the species. Section B contains most of the species 

 which in size and structure come nearest to those in Group I. 

 Section C also consists chiefly of large species, and only includes a 

 very few which can be called tiny ; whereas the reverse occurs in 

 Section D, nearly all the tiny forms being found here, and whilst 

 some may be of fair size, scarcely one is actually large. 



For the further grouping within the Sections much use has been 

 made of the colour of the halteres, the length of the costal fringe, 

 and of the condition of the meso-pleurae, whether bare or bristly. 

 I am indebted to Mr. Collin for painting out to me the last-mentioned 

 characteristic, and a very important one it is The position of these 

 bristles is in the upper hind corner, below the edjje of the dorsum. Two 

 types may be recognised : one in which all the bristles or bristly hairs 

 are of equal or nearly equal size ; the other in which one, very rarely 

 more, is much longer than the rest and lies far back, up against the 

 posterior suture. The costal fringe can scarcely be considered a very 

 important structure, yet perhaps for this very reason, and because it 

 is apparently uninfluenced b} sex, its length, whilst varying greatly 

 among the different species, is remarkably constant so far as each 

 individual one is concerned. In this respect it has the advantage 

 even of the scutellar bristles. For in the males of at least three of 

 the species in Section B the anterior bristle is considerably smaller 

 than the posterior one ; and accompanying this reduction, there is 

 more or less individual variability, which in spinigera may be so 

 extreme that a specimen here and there might well be supposed to 

 be one of the two-bristled species, the bristle being little more than 

 a fine hair, such as is found among some of these. As regards the 

 reliability of the colour of the halteres, I need add nothing to what 



