270 BRITISH BUTTERFLIES. 



describes it, it is really a line of very pale pink down the middle of 

 the darker pink lateral line, and having all the appearance of, and 

 probably actually being, a thread of white tissue imbedded in the pink 

 material, just as the latter is sunk some way beneath the actual skin 

 surface. The same specimen shows distinctly yellow oblique lines 

 down the slope. These also are not skin marking, but are slightly 

 irregular threads of yellow matter, taking a greenish hue from over- 

 lying fluids, because sunk to some depth beneath the skin. Each 

 oblique line beginning at a dorsal prominence passes obliquely back- 

 wards and downwards, crossing the next two segments, the first high 

 up, the second above the spiracle, and ending on the fourth. by being 

 lost in a yellowish upper border of the lateral pink mass. Owing to 

 their being rather obscure at either end of the larva, due to the colour- 

 ing matter being buried, it is difficult to say on what segments these 

 oblique lines begin and end, they are, however, tolerably evident on the 

 2nd and 3rd thoracic and lst-6th abdominal segments. June 16th, 

 1908. — Larva of wed on [salmacis ?) from Middlesbrough. Has no 

 dorsal red line and only the faintest indication of a dorsal difference of 

 colour from that of the larva. The lateral line is less marked than in 

 the larvae noted earlier (from South of England), so far as memory is 

 reliable, the colour also is less brick-red, more pink, but this is pro- 

 bably due to larva being quite mature and beginning to think of pupation, 

 though it is still active and unchanged in form. On mounting this 

 specimen it is found to contain nine hymenopterous (chalcid ?) larvae 

 (Chapman). The full-grown larva has a length of 6^-7 lines. Its 

 body is much arched, and so contractile that the creature can appear 1-| 

 lines shorter, whereby it naturally becomes much more deeply arched. 

 The much concealed black head has a whitish transverse streak above 

 the mouth ; the dark palpi are whitish at the base. The ground-colour of 

 the body is an agreeable pale green ; the deeply-seated, brownish-purple 

 coloured dorsal lino reaches from the beginning of the mesothorax to 

 the beginning of the penultimate segment; the rather flat anal plate is 

 semi-oval, and in the middle of each side slightly concave. On each 

 side of the body, from above obliquely down wards and posteriorly, go faint 

 pale stripes, only just perceptible, and in many points of view quite 

 invisible. The incisions of the segments are deep above, whereby on 

 each segment near the dorsal line an eminence arises, which bears a 

 multitude of white bristles of unequal length, almost radiating. Below 

 this warlike eminence is a second less conspicuous, with similiar 

 bristles. Both eminences have hollows in the middle, which the larva 

 can raise or depress at will. The lateral wart, clothed with longer 

 projecting bristles, in which the spiracle is not perceptible, is purple- 

 red, and forms the rather broad lateral stripe, which, however, does not 

 reach the head, since the prothorax is either altogether green at the 

 sides, or is only pale reddish posteriorly. The anal plate is purple- 

 coloured only for a narrow space anteriorly at the sides. The belly is 

 pale green with many whitish bristles. The ventral legs are short, 

 pale yellowish, rather transparent, with short cylindrical feet, with a 

 darker yellowish circlet of hooks ; the pectoral legs are spotted with 

 black anteriorly (Zeller). 



Symbiosis of Ants with the Larva of A. medon. — The following 

 note on this subject is interesting, though the author was unaware of 

 of the existence of the honey-gland and its use : — My next observation 



