NOTES TOWARDS A LIFE-HISTORY OF THESXOR BALLUS. 257 



in three sets ; assuming a complete circle, these three sets are at the 

 outer margin, and, at the anterior and posterior points of the inner 

 margin, about 120° apart. Each of the two inner sets consists of two 

 hooks with an abortive chitinous point between them ; the outer one 

 of only two chitinous points ; that this represents hooks at all is no 

 doubt quite open to dispute (pi. xiii.,fig. 1). The anal proleg is identical, 

 except that the outer hooks (?) are not represented. The chitinous 

 portions of true legs are black, and the hooks are nearly straight after 

 being sharply bent at their bases (pi. xiii.,fig. 2). In afull-grown specimen 

 in this skin, the reddish spiracular area is more pronounced forwards 

 and dorsally, slightly outlming the oblique bar of later skins. 



.'Third imtar: — In this stage the larva is 3mm.-3-5mm. 

 long, and has now the characteristic Lycfenid form, viz., a marked 

 marginal flange, from which the sides slope up to a narrow, but 

 rather flat, dorsal plane, this dorsal plane includes the 1st thoracic 

 segment, or almost does so, but stops abruptly with the 6th abdominal 

 segment ; then the following segments form a slope to the latter 

 part of the marginal flange. The marginal flange is yellow, as is also 

 the dorsal plane, the slopes being reddish. This broad description, 

 however, requires certain prominent markings to be noted in modifica- 

 tion of it. Dorsally, there is a deep purple-brown, almost black, line 

 spreading out on the mesothorax, with rather bright red wings, on a 

 transverse line to meet the red lateral markings. The prothorax has a 

 marked black dorsal plate. The reddish slopes are a dull purplish-red. 

 On the first six abdominal segments, and nearly as marked on the last 

 two thoracic ones, they present marked oblique red lines sloping back- 

 wards and ventral (reverse to Sphinx larval lines). Below these, the red 

 *is modified by two paler (yellowish) clouds on each segment, being 

 close below the oblique lines ; the forward is more dorsal. In the 

 centre of each of these oblique bands on the six abdominal segments 

 (but not on the thoracic) is a hollow or depression, its margins a 

 little darker, the bottom paler, longest in an antero -posterior direction, 

 and slightly expanded at the ends in some cases. These margms of 

 the hollow appear to overhang a little. In looking for the lenticles of the 

 previous skin, it seems at first that they are wanting. A small one, with 

 veiy dark margin, exists on quite the dorsum of the 2nd thoracic. On 

 the 2nd, 8rd, and 4th abdominal segments, minute rings exist towards 

 the front margin of segments, in line with the hollows above noted. 

 These might, so far as their appearance goes, be hair-spots that had 

 lost their hairs. On the abdominal segments also, near the middle 

 line towards the front of segment, is a minute, dark, chitinous tubercle, 

 with, perhaps, an indication of a paler summit. A small raised tubercle 

 also exists above the spiracle where the lenticle occurred in previous 

 skin. This is a black ring with plain centre, smaller than lenticle of 

 previous skin. Two exist on each side of the 7th abdominal segment. 

 On the 7th, the spiracle is much higher than on the 6th, and, on the 8th, 

 it is very large and quite dorsal (nearer middle line than to flange). The 

 long (•4mm.) spiculate hairs are much as in previous skin, bat are (com- 

 paratively) smaller, the dorsal group (on yellow back) are about ten in 

 number, with four or five of them larger, and two or three of these 

 black, the rest white ; two or three more are on the yellow dorsal 

 colour where it stretches down above the oblique band, at the posterior 

 margin of the segment. Round the spiracles is another group, of which 



