ANTHROCERA. 429 



of some species of this group. (3) The " minos group," in which the 

 scales are long ovals, more or less sparsely planted, and rarely bifid. 

 Whenever these scales suggest a triangular shape, the base of the 

 triangle is towards the base of the scale, the reverse being the case in the 

 other groups. A. viciae (meliloti) is said to form in the structure of its 

 scales a connecting link between the " filipendulae " and "minos" 

 groups, with more defined affinities with the latter. 



The simple scales of the "minos group " are supposed to represent 

 the most generalised form. The brilliant red scales are, in all cases, 

 rounded at the tips, and the amount of rotundity seems to determine 

 the brightness of the general appearance. The duller red or orange 

 scales are bifid or trifid. Bowell further remarks that A. filipendulae, 

 A. lonicerae and A. trifolii are, in scale structure, probably the most 

 fixed and typical of the species of the " second " group, and that the 

 external resemblance of the two latter species is not maintained in the 

 scales. 



We have already suggested the tendency of the fore-wings of the 

 males of most of our British species to be purple-blue in colour, and 

 the females green. The dark border of the hind-wings is, as a rule, 

 wider in the males than in the females, in the latter sex of A. fdi- 

 }>endidae, the border is often confined to the fringe alone. The Anthro- 

 cerid male is also (in some species, at least) specialised in having two 

 eversible feathery tufts or scent glands, immediately in front of the anal 

 segment, and placed just ventrally thereto. 



The antennas of Anthrocera are cylindrical in so far that they have 

 everywhere a smooth, circular, transverse section. The segments 

 become larger and larger as the base is left, so that the well-known 

 club-shape results ; there is some variation, in different species, in the 

 sharpness of the decline in size at the tip. Each segment is rather short 

 for its width, but the articulating surface does not occupy the whole of 

 the ends but only a central portion ; the opposed faces have neither 

 scales nor sensory hairs. The dorsal surface, from end to end, is covered 

 with scales, and the ventral with sense hairs. The only exception to this 

 that I have noted is that in some species the terminal segment is reddish 

 (free from pigment) , and without dorsal scaling. The sensory hairs form 

 a dense velvety coating to the under surface, and have amongst them 

 some larger ones on each segment. The chitinous surface itself, of the 

 ventral aspect, is covered with conical prominences. There are certainly 

 no cones (Bodine's terminology), and I think no pits. The conical 

 prominences noted look like short, thick hairs, about half the length of 

 the others, but I cannot determine that they have any articulation at 

 their base. They are larger and more regular than the serrated points 

 that cover the chitin of the pectinations of Harrisonia, but the density 

 of the pigmentation is greater than in that form, and makes observa- 

 tion even more difficult. The dorsal chitin is free from any projection, 

 except the scale-cups, and has no hairs, except one (or more ?) close to 

 the distal margin of the segment, and about the middle of the dorsum. 

 In Harrisonia, dorsal hairs amongst the scales are more numerous 

 (Chapman). 



The species included in this genus are day-flying insects, generally 

 very local, and usually abundant where they occur. They boom along 

 heavily in the hottest sunshine, or sit, sometimes several on a single 

 flower-head, sucking the nectar from the* capitula of Centaurea or. 



