RUMICIA PHL^IAS. 387 



arise so close together, that, whilst hardly appearing to have their 

 "bases conjoined, it is difficult to be sure they have not, especially on 

 the forward segments. Tubercle ii carries a short hair O-lmm. long, 

 deflexed, arising close to, and outside, i. Halfway between this and 

 the spiracle, in middle of segment, is a large lenticle (a little larger 

 than spiracle). Slightly above this, and at the posterior border of the 

 segment, is a very minute hair, and a similar one about halfway 

 between the lenticle and spiracle ; this varies in position, usually 

 rather nearer spiracle ; on one side of one segment it is close under 

 the lenticle ; again, there is a similar minute hair behind, and a little 

 below, spiracle. In two specimens (British), on the 4th, 5th, and 6th 

 abdominal segments, in front, and outside i, where an accessory hair 

 is common in Lycaenid larvae, is a small hair of peculiar structure ; it 

 is not more that about OOlmm. long, and is fan-shaped, with several 

 points at the end. In two Riviera specimens these are absent, except 

 that, on one side, in one specimen, on the 6th abdominal segment, is 

 a similar " hair, but a little more baton-like in shape. Below the 

 spiracle (on flange) are three hairs on a level, the middle one a little 

 the longest, 0-2mm. ; on the 3rd, 4th, and 6th abdominal segments 

 there is a fourth smaller hair in front of these ; lower is a very large 

 lenticle, then a small hair (0-02mm.), and then two at the base of the 

 prolegs. On the 7th abdominal, i is present, with four lenticles 

 behind it. On the 8th abdominal, i is present, with two lenticles 

 behind it. On the 9th abdominal, i is present, and the two hairs are 

 some little way apart, and there are no lenticles. The lenticle 

 between i and the spiracle is absent on the 7th, 8th, and 9th abdominal 

 segments. The 10th abdominal has an anal plate, with two lenticles 

 on each side, and with five hairs beside and behind it, on a surface 

 with long sharp skin-points ; the 8th (with 9th) abdominal segment 

 shows four marginal hairs. The submarginal lenticles, which are very 

 large (twice the diameter of the spiracle), are present only on the 3rd, 

 4th, 5th, and 6th abdominal segments, but on the 1st and 2nd its 

 place is taken by a long hair. On the metathorax the spiracle and the 

 lenticle above it are absent ; there are four equal marginal hairs ; in 

 line of submarginal lenticle are two small hairs, with a small lenticle 

 behind them. On the mesothorax, tubercle ii carries a hair exactly 

 like i (at least there is such a hair, that it truly represents ii is another 

 matter); there are four marginal hairs, and a lenticle in front of them, 

 two hairs and a lenticle below, as in metathorax; no hairs at the leg- 

 bases. On the prothorax are seven marginal (?) hairs on either side, which 

 run round below plate, but above spiracle ; below the spiracle there is 

 one hair. The plate is square, with three long hairs at each anterior 

 angle, minute ones behind, of these the largest (very small) are near 

 the posterior angles. The general surface is faintly reticulated with a 

 largish mesh. The front of the prothorax is nodulated with rounded 

 skin-points, much closer together than the meshes of the skin-netting 

 behind. The prolegs have thirteen to fifteen hooks, more than usually 

 in a complete line (not divided into two pads) (May 5th, 1906). 

 Second instar (newly-moulted) : The colours are more blurred and the 

 white-yellow band looks straight ; there are two sets of hairs on the 

 slope between the dorsal and lateral flanges, but the slope is a little 

 rounded both in this and the first instar (fullfed). [The typical 

 Lycaenid slope is not at all marked, except in quite young larvae in the 



