SPHINX LIGUSTRI. 307 



but, shortly afterwards, the horn becomes very nearly black. Except 

 for the ordinary setae of the primary tubercles, the larva is quite 

 smooth *, i.e., there are no secondary hairs, nor any very evident 

 division into subsegments. As an exception to the smoothness of 

 the larva, the horn (and apparently the segment about its base) 

 is thickly clothed with short hairs, these hairs are expanded at 

 their points, and many of them, especially towards the extremity 

 of the horn, have a bifurcate character, being fish-tailed like those 

 of Mimas tiliae. The two terminal hairs are similar to those on 

 the tubercles, but hardly so long, and are, doubtless, the setae of 

 tubercles i of this segment. The setae of the tubercles are rather 

 long, dark, and terminate in an enlargement as of a small funnel- 

 shaped structure, and with some traces of spiculation. In a mounted 

 specimen the subsegments become visible, and the tubercles are 

 placed thus — i on the 3rd subsegment, iii on the 4th subsegment, 

 and ii on the 6th. There is a prespiracular tubercle which cannot 

 possibly be a subspiracular (v) moved upwards, because it is very 

 definitely placed on subsegment 2, and these subsegments are all 

 dorsal divisions, altering, mixing, and rearranging themselves at 

 the flanges belonging to the spiracular and subspiracular region. 

 A subspiracular tubercle (iv) is on the lower aspect of a flange-swelling 

 passing from behind the spiracle in some degree from subsegment 

 5, and running downwards and forwards. Below this is one more 

 tubercle nearly at base of proleg. The dorsal (? i + ii) and sub- 

 dorsal (? iii) tubercles on the 2nd and 3rd thoracic segments carry 

 two setae, one in front and one behind (in one case, the dorsal on the 

 2nd thoracic, has three). There are no small or secondary hairs on the 

 general surface, those of the horn are arranged in circles round 

 the horn, not easy to count at its base, but more evident further 

 on, and about 130 in number. Each circle has some 12 or 14 

 hairs. Those in adjacent circles alternate, or the hairs in alternate 

 rows are on the same line, or even more frequently the hairs of rows 1 

 and 4 are in line rather than those of 1 and 3. The alignment in 

 this way is, indeed, a little irregular, but is sufficiently regular to 

 give a fir-cone spiral appearance to portions of the horn in many 

 aspects. The crochets to the prolegs appear to be about 9 in number. 

 As the larva gets a little older the subsegments become more 

 distinct, 8 in number, the narrow longitudinal yellow line is fairly 

 distinct and the oblique yellow lines, though pale, are quite 

 evident f ; they are very slender, they start from i of one segment 

 and pass to iii of the preceding segment and then on to the pre- 

 spiracular, these tubercles being in its actual line, and, as in the 

 Amorphids, suggesting that their yellow coloration originated the 



* Poulton erroneously says {Trans. Ent. Soc. Land., 1886, p. 141): "In the 

 newly-hatched larva of S. ligustri the body was- covered with hairs or bristles, 

 which sprang from ordinary shagreen-dots, whilst there were also two lateral and two 

 dorsal rows of longer bristles springing from larger shagreen-dots which bore a 

 special relation to the larval markings which appeared later," &c. 



t Hellins states (Buckler's Larvae, &c, ii., p. in) : " When the first moult is 

 passed .... the slanting stripes on the side now become visible, i.e., on 

 each subdivision there is one pale point in the line of the streak, the streaks thus 

 being lines of pale greenish- yellow dots." The fact is these oblique stripes, as noted 

 above, are visible in the 1st instar, i,e., before the 1st moult. 



