SPHINGIDES. 377 



characters, and it is difficult to believe that such forms could 

 arise unless there were some genetic relationship. Chapman 

 disagrees entirely with Bodine, and writes (in litt.) : " The 

 antennae of the Sphingids present certain very distinctive 

 characters. The female antenna of an Amorphid looks very like 

 that of any other moth with a simple antenna — scaled dorsally, with 

 fine hairs ventrally and sundry bristles. A closer examination, how- 

 ever, shows that its true axis is not its apparent one. Each joint 

 is articulated to its neighbour, not by its whole surface, but by a 

 small circle close to the dorsal surface, and we might call the lowest 

 portion beneath the articulating axis a pectination, or a pair of 

 pectinations coalesced. It is very possible that this is its real 

 nature. The dorsal scaled area has the scales arranged in some- 

 what irregular transverse rows, never quite regularly, often very 

 irregularly, in the manner usually associated with pectinations past 

 or present. A second character of the Sphingid antenna is the 

 terminal segment or joint, which is somewhat lengthened and of 

 conical form in Amorphids, and becomes, in the other families, much 

 lengthened and armed with various bristles, so as to have quite a 

 special facies, often amounting to a sort of tassel. Another peculiarity 

 of the Sphingid antenna is the peculiar arrangement of the long 

 hairs of the $ antenna. The lower surface is divided into two 

 lateral portions by a central longitudinal keel or ridge, often very 

 marked, almost always observable. If an ellipse be drawn on the 

 ventral surface of the antenna, with this keel as the shorter axis, 

 and the extremities just touching the middle of the margin of the 

 scaled dorsum, it would mark exactly the line occupied by the 

 origin of the hairs, excepting that a central line by the keel is 

 usually vacant. The longest of these hairs is nearest the central 

 line, and they are so disposed that, looking at the antenna on its 

 ventral aspect, those on either side of the ventral line form a pocket 

 like the toe of a shoe, by arching across the half ellipse between 

 their bases, those on the proximal meeting those on the distal side 

 of this area, with the opening into the pocket directly facing the 

 observer, the individual hairs not radiating in all directions from 

 the ventral line of the antenna, but directly outwards from its central 

 plane. The central line or keel and the area within these pockets 

 are covered with very fine sense-hairs, the angles outside the lines of 

 hairs are unclothed. In the female antenna the whole ventral 

 surface carries these fine appressed sense-hairs. The description 

 given above of the ellipse of hairs is not quite accurate, at least 

 for many species ; as seen from the underside, the line looks single, 

 but, at the further end, next the scaling, it is not so ; there, there 

 are several hairs in short rows parallel with the length of the 

 antenna, and very like the rows of hairs on the plumules ot Lach- 

 neids. Still, though the Sphingid antenna is probably derived 

 from some forms of pectinated antenna, it is difficult to understand 

 any mutation by which these hairs could be derived from those of the 

 Lachneids. On the line of the keel at the distal margin of each 

 segment is a hair, not very obvious basally, but, beyond the middle 

 of the antenna, becoming short and thick as a little baton, and pro- 

 jecting markedly on the convexity of the antenna, where it curves 

 round towards the end. The characters of the Sphingid antenna 



