480 BRITISH LEPIDOPTERA. 



continued forwards to the anterior limits of the 6th abdominal 

 segment. There are oblique stripes (chiefly made up of shagreen 

 dots) just above the claspers on the 3rd, 4th, and 5th abdominal 

 segments. These lines are more nearly horizontal than the ordinary 

 oblique stripes, and their relation to the latter is doubtful. They may 

 represent the forward extension of the 4th, 5th, and 6th oblique 

 stripes respectively, but they have also the appearance of a sub- 

 spiracular line twisted into partial parallelism with the oblique 

 stripes. An examination of the larva in this stadium, therefore, adds 

 nothing to our knowledge of these lines, which are found during the 

 whole subsequent life of the larva, and which have previously been 

 described in the last stage (see, Trans. Ent. Soc. Lond., 1885, p. 297). 

 The extension of the first stripe and the semicircular crown are not 

 equally distinct in later stages, but the latter is easily recognisable. 

 The 8th stripe disappears. The whole comparison strongly confirms 

 Weismann's conclusion as to the extreme uniformity, and, therefore, 

 the ancestral character, of the appearances witnessed in the ontogeny 

 of this species (Poulton, Tratis. Ent. Soc. London, 1887, pp. 281 — 282). 

 For notes on the dichotomous hairs of this larva see Trans. Ent. 

 Soc. Lond., 1885, p. 296. Notes on moulting have already been 

 given (anted, ii., pp. 17 and 59). 



Variation of larva. — The larvae of A. populi, like those of ' S. 

 ocellata, vary considerably in colour in their later stages, ranging from 

 dull green or sage-green to quite a bright yellow. The spiracles 

 are often surrounded by a red spot, and it is not uncommon to 

 get larvae with an entire or partial subdorsal row of similar spots. 

 These spots are said to simulate the small red galls so often 

 found on willow and poplar leaves. In one larva that I bred this 

 year, the spots of the lower row were absent from the thoracic 

 segments, whilst those of the upper row were present on those 

 segments. I used to have a notion that the different forms were 

 from different broods, chiefly because I had usually found the dull 

 green forms on black poplar and the bright ones on Lombardy 

 poplar, sallow or willow. It is true that I have occasionally found 

 the latter on black poplar, but I do not remember ever taking the 

 dull forms on Lombardy poplar. This year, however, I bred both 

 forms from eggs laid by a single female, and Buckell has had the 

 same experience (Bacot). Poulton has made many observations on the 

 red-spotted form of the larva of this species (Trans. Ent. Soc. London, 

 1885, p. 297 ; 1886, pp. 284 et sea.), and observes that the spots may 

 be present in the early stadia, and that the peculiarity may be increased, 

 or, on the other hand, the spots may decrease or altogether disappear 

 in the later stadia, and he considers it probable that the spots 

 are present upon a much larger proportion of young larvae than 

 upon those in the later stages, although, in a certain proportion of the 

 latter, the character reaches a pitch of perfection not hitherto described 

 in the earlier stages. Based upon the examination of 22 specimens, 

 White observes (loc. cit., p. xxv) that the spots in the dorsal and spiracu- 

 lar rows usually develop on the segments in the following order : 



Dorsal row 3, 7, 4, 2t, 5, 1, 6, 2, 3t, it, 8. 



Spiracular row .. .. .. 4, 3, 5, 2t, 7, 6, 2, 8, 1, 3t. it. 



He shows that the dorsal is the more important series, and that 

 this row commences most frequently on the 3rd abdominal segment, 



