384 BRITISH LEPIDOPTERA. 



have specialised from a moderate-tongued ancestor, the develop- 

 ment has left its mark on the pupa ; on the other hand, the 

 Amorphids have specialised in the direction of loss of tongue, 

 and, therefore, in general appearance, the pupa is much more like 

 those of its tongueless relatives (sens. Int.). In the imaginal stage, 

 also, the habits of the Amorphids and Sphingids are so different 

 that specialisation is again in opposite directions and considerable 

 difference in appearance results. 



Comparing the British Amorphid with the Sphingid species, Bacot 

 gives (Ent. Rec, vi., pp. 178 et seq.) the following interesting details : 

 " In their earlier stadia, the larvae of the Amorphids and of Sphingids 

 (Sphinx ligustri) have many characters in common, e.g., S. ligustri, 

 in its first skin, has forked hairs, though these are black, and are 

 thinly scattered compared with the ' door-mat ' appearance which is 

 characteristic of the Amorphids. The shape of the head is similar 

 to that of Mimas tiliae in its earlier stages, and there are other minor 

 points of resemblance. In the imagines, however, the only points of 

 resemblance that I can see are that the head and eyes of S. ligustri 

 are small, and that the antennae are somewhat similar in stucture. 

 On the other hand, the resting-position is quite different, S. ligustri, 

 like Manduca atropos, Agrins convolvuli, &c, resting with the fore- 

 wings sloped over the back and the hindwings folded underneath them. 

 The resting-position of Eumo?pha appears, at first sight, to be some- 

 what similar to that of the Amorphids, but the pupae are very different. 

 Whatever may be the right place for the Amorphids, I can see no 

 reason for putting Manduca atropos next to them, for I think that, 

 without question, S. ligustri is a nearer relation, and it is probably a 

 link between the two. Turning to individual differences, the hind tibiae 

 of M. tiliae have four spurs, while those of the other Amorphid 

 species, A. populi and S. ocellata, have only two. I find that 

 Manduca atropos, Sphinx ligustri, Eunwrpha elpenor and Heniaris 

 fuciformis also have four spurs to their tibiae. The males of M. 

 tiliae have a frenulum*, but in the females this is only rudimentary, 

 the loop being entirely absent, and the bristle being replaced by a 

 number of short slender ones which are of no apparent use. The 

 males of Smerinthus ocellata possess a small bristle, the females a 

 group of small ones, but there is no trace of the loop in either 

 sex. I examined a number of specimens of A/norpha populi, but 

 could find no trace of loop or bristle in either sex, except in one 

 female, which had a group of small bristles on one hindwing only. No 

 doubt this was simply an instance of reversion, the frenulum apparently 

 being, as a rule, altogether wanting in this species. A consideration 

 of all the different characters present in the several stages leads 

 one to think that M. tiliae is the oldest form, and that .S. ocellata, 

 though its habits and food are very different, as is also the colora- 

 tion of the imago, is not really so widely removed from M. tiliae 



* Griffiths notes that the frenulum probably reaches its highest development 

 among the typical Sphingidac, but the case is far different with the Amorphine 

 group, many species of which, including the giant Coequosa triangularis from 

 Australia, possess it either weakly developed or quite abortive. In tins, as in 

 several other respects, they approximate to the broad-winged Bombycids — 

 Lasiocampa, Attacus, Satumia, and their allies — to which they are probably nearly 

 related. 



