374 BRITISH LEPIDOPTERA. 



elpenor (loc. cit., pp. 146 — 147). Poulton observes ( Morph. Lep. Pupa, 

 p. 200) that, in the external genital organs of the pupa of Sesia 

 stellatarum, there is a median line which traverses the prolongation 

 of the 10th abdominal. Poulton .gives excellent descriptive figures 

 of the terminal abdominal organs of the pupae of many Sphingid species 

 (vide, loc. cit., pp. 202 — 206, pi. xx., figs. 1 — 26). [Jackson finds no 

 external indications of sex in the larvae of Sphinx lignstri, etc.] 



We have already given (antea, pp. 357 et sea.) the diagnoses ot 

 Herrich-Schaffer, Meyrick and others, of the imagines of this group. 

 Fernald further notes (Sphingidae New England, pp. 96—97) the 

 following characters : 



The thorax is well -developed in all the species and has the scales appressed 

 over the surface, or there is a short stout erect tuft from each side of the 

 metathorax, while some species have a central ridge of scales along the middle. 

 The abdomen is stout and tapering to the apex, aud there are fine spines along 

 the edge of the segments in some of the species, which are concealed by the 

 scales. Some species have anal tufts more or less fandike and small tufts along 

 each side. The ? ovipositor consists of two short pieces, one on each side, 

 rounded at the outer end, and clothed with short hairs. The external organs 

 of generation in the male are quite complex, and, besides the intromittent organ, 

 consist of a clasper on each side, and a central plate above, to which is attached a 

 hook curving downward, and, beneath this, is a projection, which is generally 

 shorter than the hook, and curves up somewhat at the end. These two may be 

 represented by the thumb and finger separated from each other by a little space and 

 slightly bent. They vary much in form and in their relative length in different 

 species. The side pieces or claspers also vary very much in form and size, and often 

 have a variously-shaped spine, hook, or clasping-organ connected with the lower and 

 inner side. These appear to be only a modification of one part of the side-piece and 

 not a distinct body. The leg's are well-developed and of moderate and nearly equal 

 length. The fore tibiae have a stout spur called the tibial epiphysis on the inside, 

 and, in some species, there are, more or less, spines all over their surface. The 

 middle tibiae have a pair of unequal spines at the outer end, and are also spinose in 

 some species. The hind tibiae are rarely spinose, but have a pair of spurs at the 

 end, and, in most species, a second pair near the middle. The tarsi of all the legs 

 are five -jointed, spinose, and are armed with a pair of simple claws at the end. 



The specialisation of the imagines is carried to great length 

 in some particulars, the proboscis, antennae, wing-shape, scales, 

 &c, being specially developed in various groups. It is remarkable 

 that the mandibles, which can hardly be said to exist at all in the 

 lepidopterous imago, appear to be present as obtuse projections on 

 each side of the labrum in AmorpJia populi, PJilcgctliontius Carolina, 

 and possibly several other species. The forewings are usually pointed 

 at the apex, long in proportion to their width, the nervures exceedingly 

 strong towards the base, and the hindwings small. The Amorphids 

 again form an exception. The Sphingid abdomen is usually pointed, 

 but in the Sesiids (Macroglossids) tufted apically, with one or two 

 tufts of long scales capable of considerable expansion. The Sphingid 

 proboscis is frequently developed to great length. Even in Agrius 

 convolvuli, the proboscis is some 5 inches in length, and those of 

 some South American species are double this, the specialisation 

 enabling them to stick the nectar from long tubular flowers, and no 

 doubt are specially modified to aid in giving effect to cross-fertilisation. 

 On the other hand, specialisation has, in the Amorphids, resulted 

 in reducing the tongue to a very small and almost functionless 

 structure. Westwood has noticed that there is an interesting con- 

 nection between the variation in the length of the tongue and the 

 rapidity of flight, both depending on the habit of sucking nectar 



