308 BRITISH LEPIDOPTERA. 



idea of an oblique line, formed in them by the yellow secondary 

 tubercles and hairs which do not exist here ; a darker dorsal line 

 is also now present. As the larva grows, the horn looks less pro- 

 portionally long. It is now also evident that the fine hairs exist 

 only on the horn, and that what at first could be easily mistaken 

 for the surface of the segment at its base, is really a portion of the 

 horn. Elsewhere the segments are perfectly smooth, but, in well- 

 grown specimens, are some dark points in the oblique lines that 

 look as if there were rough points or some representation of them 

 beneath the cuticle ; they may really be so, representing points to 

 occur in next stadium (Chapman). The tubercles on the meso- 

 and metathorax differ altogether in position from those of the 

 Eumorphids (e.g., Theretra porcellus), i and ii being close together 

 not only on the same subsegment, but occupying the same chitinised skin 

 area, i only bearing a seta, ii being largely atrophied,* while on the 

 abdominal segments there are two subsegments between i and ii ; 

 on the abdominal segments iii is directly above the spiracle on 

 the subsegment posterior to that bearing i ; iv is beneath and slightly 

 posterior to spiracle, whilst the prespiracular (? v) is in the same 

 line as the spiracle, but on the front part of the subsegment anterior 

 to that bearing i ; iii, on the thoracic segments, consists of two setae on 

 a raised skin area ; there appears to be only one subspiracular tubercle. 

 The primary setae are slightly knobbed at the top and there is no 

 trace of secondary hairs; on the 8th abdominal the anterior 

 trapezoidals (i) are placed on the top of the caudal horn. When 

 nearly fullfed in this skin, the oblique stripes are clear and distinct, cut 

 by the subdorsal line, which also is very clearly marked. There 

 appear to be only 7 subsegments to the thoracic, but 8 subsegments 

 to the abdominal, segments (Bacot). The larva just after emergence 

 is 5'25mm. long when extended in walking, the horn 3mm. long 

 (in some, however, the horn is nearly as long as the body). The 

 colour of the body is a pale, transparent (very slightly greenish-) yellow, 

 the head much greener than the body (see Trans. Ent. Soc. Loud., 

 1885, pi. vii., fig. 1) ; the horn black, but the upper half transparent 

 and greenish, the green internal fluid shining through the dark 

 exterior, it tapers very slightly, is straight except that the upper part 

 is bent downwards in a gentle curve, the blunt tip is bifid, and 

 each process ends in a fine and fairly long bristle, so that the 

 structure is prong-like when viewed from above. The body is 

 thinly clothed with long fine black hairs, arranged in 4 rows, two 

 dorsal f (one on each side of the middle line) and two lateral}: (a little 

 above the spiracles) ; each of these four rows is made up of two hairs § 

 on each segment from the 1st to the 7th abdominal (both inclusive). 

 Posterior to the latter the hairs are more abundant, but still thin, 



* Possibly only in the example examined ; one suspects that the seta of ii is 

 usually present (Bacot). Chapman adds that the seta; of i and ii are equally developed 

 as "well as the two setne of iii, and that Bacot's example must have been abnormal. 



t Evidently refer to the rows in which i and ii are placed longitudinally. 



% Appear to refer to iii on either side. 



I An obvious error. The setae of] and ii may be considered as being in the 

 same line on each segment, but iii is single-haired on the abdominal segments, and 

 has no other hair in line with it, so that here Poulton adds an extra tubercle and seta. 

 A later note of Poulton's suggests that he may be here including the prespiracular 

 (? v) as being in the same line as iii (see last footnote on p. 309). 



