238 THE entomologist's record. 



looking ; traces of seven oblique stripes are present and have a head to 

 anus slope ; they start at dorsal tubercle i and end at level of spiracle (just 

 above lateral flange) near the hinder edge of the next following 

 segment ; divisions of segments very deeply marked ; subdivisions are 

 rather obscure, but I think there are four on each abdominal segment. 

 The noticeable parti-coloured forked horns are apparently develop- 

 ments of the fleshy bases of the tubercles. That on the anal segment 

 is apparently a development of the anal plate itself, and not forked. 

 A pair are situated on the prothorax, a very tall, large, and widely- 

 forked pair on the metathorax, and a large single central unpaired 

 one on the 8th abdominal. The small unforked one on the anal seg- 

 ment is also central. These horns are very formidable-looking, being 

 widely forked, and covered with smaller lateral branches, having, as 

 have the upper forks, hairs at their extremities, the branches on lower 

 half being almost equal to the forks at top in size. The horns are 

 rendered more prominent on account of their divergent colouring, the 

 lower half of the thoracic and the 8th abdominal horns and the whole of 

 the anal one being coral red. A band equal to about ^ to ^ is white 

 and the upper portion and forks are coral red. When the larva is in 

 the resting-position, they would most certainly protect it from the 

 attacks of parasites that wanted to oviposit on it, as owing to the 

 " Sphinx " attitude the thoracic horns protect the whole dorsal area, but 

 this probably is quite unnecessary for a small newly-hatched larva, 

 at any rate m its first stadium, but possibly they were an important 

 defence to the adult larva in past ages, and have now been forced 

 back into an earlier stadium by the acquisition of newer and better 

 protective characters. The four thoracic and 8th abdominal horns 

 are all developments of the Heshy base of i. This tubercle, on all 

 other segments than those producing these enormous horns (they are 

 quite half the length of the larva when it first emerges from the egg), is 

 developed into a sufficiently remarkable, fleshy, branching horn, not, 

 however, distinctively coloured ; each branch gives rise to a single hair 

 (so that we must consider i as bearing several hairs) ; there are about 

 three large branches to each horn (tubercle i), and from three to six 

 smaller ones varying from short branches to highly-developed hair- 

 bases. The hairs themselves are covered with minute, closely-set thorns, 

 and are, in some instances, bihd at tip ; ii is a small, single-haired, 

 Heshy-based tubercle, on all the thoracic, as well as abdominal, segments, 

 as far as 8th. On the segments that bear horns, however, ii is really 

 on base of horns, nearly the whole of the dorsal area of segment 

 being utilised in the base of these projections. I think I should omit 

 the prothorax as regards ii, as I doubt the hairs on posterior bases of 

 horns bearing any close analogy to ii on other segments. Abdominal 

 segments 9 and 10 both have a pair of the small tubercle i horns, 

 showing clearly that the central anal horn differs from all, save possibly 

 the prothoracic pair, in being developed independently of i. The horn 

 on the 8th abdominal is a development of both anterior trapezoidals 

 (i) combined, and is probably in some form or other a much earlier 

 development than the thoracic horns, which are evolved one from each 

 anterior trapezoidal (i). Another difference between that on the 8th 

 abdominal and thoracic horns is that the forks of the thoracic are front 

 and back on prothoracic, and nearly soon metathoracic horns, although 

 here thev fork somewhat at an angle to the median line of larva ; on bth 



