ANTHROCERA FILIPENDUL^. 525 



segment (widest part) nearly 5 mm. It tapers much towards anus 

 (the abdomen being rather long and pointed) slightly towards head. 

 The eye-covers, bases of antennae and face piece (a ventral plate) 

 project noticeably. The sexual organs well-marked. The abdominal 

 segments 3-7 are free (probably also 1 and 2) ; the wing-cases are 

 soldered (but not firmly) together and to appendages ; the ends of the 

 3rd pair of legs and maxillas project unattached from beneath the wings 

 and reach the 6th abdominal segment ; the antennae are long, reach- 

 ing with the wings just beyond the 4th abdominal ; a small portion of 

 the hind-wings extend beyond the fore-wings over the 1st and 2nd 

 abdominal segments. The abdominal segments are dull or dead 

 black in colour, the wings, thoracic segments and anus, shiny, the 

 intersegmental areas between abdominal segments 2-8 are paler, but 

 show best between 4-5, 5-6, 6-7, 7-8. Above and behind the spiracles, 

 on abdominal segments 1-7, is a hollow, with the chitin in radiated 

 folds (as though the end of a finger had been plunged, with a twisting 

 movement, into putty or dough), and below the spiracle there is a 

 hollow, less organically marked. These place the spiracles on a slightly 

 marked and rounded lateral flange (the same structure, no doubt, that 

 is so marked in Adscitids, and give these pupae so flattened an appear- 

 ance). On the dorsal anterior edge of abdominal segments 3-9 there 

 is a ridge or row of strong curved hooks pointing backwards, and on 

 the posterior edge of abdominal segments 6-7, a few minute spicules, 

 or bristles, can be traced (they could hardly be called a row, as in the 

 Psychids). The 9th segment has the usual row of hooks, somewhat 

 modified, and a few nodules along its hind margin ; the 10th bears 

 somewhat dorsally some blunt points, about a dozen on each side, 

 rather irregularly grouped, but capable of being regarded as a double 

 or treble row. There is a slight tendency to a dorsal constriction or 

 waist between the thoracic and abdominal segments. Of the head-parts 

 the eye-covers and labrum are prominent, the base of the maxillae has 

 a projection that probably represents the maxillary palp, and there is 

 an internal chitinous piece that may represent a further portion ; 

 other small and obscure pieces probably represent the mandibles. 

 Bacot remarks that, having examined a number of pupaa collected 

 within a few square yards, at Sandown, in August, 1898, he found the 

 same difference with regard to the development of the anal spines as 

 exists in the pupae of A. trifolii, some of the pupae of A. filipenclulae 

 having then quite as well developed as those of the latter species. 

 Hellins describes the pupa as being "about 16 mm. long, cylindrical, 

 of even bulk, except that the headpiece slopes rapidly from the back 

 and stands out distinctly, and the last three segments of the abdomen 

 taper. The antennaa-cases are strongly formed and well-developed, 

 free at their tips, as is also the tongue-case, for some distance, and the 

 edges of the wing-cases ; the anal segment ends in a somewhat rounded 

 boss, without a spike, but bearing some short sharp points, reaching 

 to the spiracles and pointing backwards. The colour is glossy black 

 on the head, thorax, wing-cases and tail ; the abdomen is more dingy 

 black, with the segmental folds showing dingy greenish." Moore 

 notes (Ent. Eec, iii., p. 37) a pupa found at Deal, in August, 1890, 

 which was alive after a lapse of 17 months. 



Dehiscence.— On dehiscence, the head, eyepieces and antennae 

 form one piece. The dorsal headpiece does not carry the eyepiece, 



