440 BRITISH LEPIDOPTERA. 



Sphingid superfamilies. In this connection one may notice that 

 although the naked Sphingid larvae show primitive tubercles of a 

 generalised kind in the first stage, and these may lead one to assume the 

 larvae to be of a more ancestral form than those of the Lachneids, yet the 

 development of the adult Sphinx-like larva of Endromis versicolora, from 

 a Lachneid-like larva in its first instar, together with a somewhat 

 parallel development in the larva of B. mori suggest that the Sphingids 

 may be a special development of the original Lachneid stirps. With 

 regard to the supplementary prespiracular wart, already noticed as 

 being present on the anterior border of the segments, it is highly 

 improbable that it has any connection with the primary subspiracular 

 tubercle v ; it must be secondary and independently developed. One 

 may here note the prominent development of i on abdominal segment 

 8, which in some larvae forms a sort of hump ; it is, one suspects, a 

 good evolutionary character but one does not feel any confidence in 

 dealing with it owing to a similar parallel development in the Notodonts 

 and other more or less unrelated families." 



It may be well to note here the great development of the prothoracic 

 lateral tubercles in the early stadia of the Lachneid larvae. Bacot 

 terms them " ear-tubercles." Packard observes that those of the 

 newly-hatched larvae of Artace rubripalpes project outwards and consist 

 of a large, piliferous, amber-coloured tubercle, three times as large as 

 those behind it on the succeeding segments. Other marked features 

 in some Lachneid larvae are the extra size and length of the thoracic 

 segments. One may also mention that the arrangement of the dorsal 

 tubercles on the meso- and metathorax appear to be, in the species 

 examined, the same as on the abdominal segments. 



We have already referred to the larva of Eustaudingeria vandalicia, 

 which Dyar describes as a curious larva, having but little the appear- 

 ance of those of the modern Lasiocampidae, its abundant, rather stiff 

 hairs and conspicuous warts giving it the appearance of an Arctian. 

 He describes the tubercles as " i and ii large, elongate transversely, iii 

 more rounded, iv and v in line below the spiracle, vi moderate, leg- 

 plates pale ; on the thorax three warts above the stigmatal warts, the 

 two upper in line longitudinally, large, equal, elongate transversely, 

 the third wart more rounded, the stigmatal and subventral warts 

 smaller." He adds that " this arrangement is exactly that of the first 

 stage of Tobjpe velleda." The suppression of the larval tubercles in 

 this latter species is discussed by Dyar (Proc Bast. Soc. Nat. Hist., 

 xxvii., pp. 144-5) ; he observes that the arrangement of the warts in 

 stage 1 is, in this species, as in Clisiocampa (vide, Psyche, vii., pp. 

 259-260), but the confusing secondary warts on the anterior part of 

 the segments in the latter genus are not present here. On the abdomen 

 v is smaller than iv, and all except i and vi are greatly reduced. These 

 two warts, i and vi*, alone are present in the adult larva. On the 

 thorax there are three warts above the stigmatal wart, the middle one 

 posterior to the others. The two lower are rudimentary, and in the 

 adult larva only three warts persist, corresponding probably to 1 + lb, 

 iv -+- v, and vi." It appears to us that the above description of the 

 large tubercular warts i, ii, iii of Eustaudingeria agrees much more closely 

 with that of the Eutrichids in their first instar than with Malacosoma 



*Not ii and v as stated in Ann. New York Acad. Science, viii., p. 229. 



