444 BRITISH LEPIDOPTERA. 



cremaster is very characteristic, it occupies the dorsal half of the 10th 

 abdominal segment, and the end of the pupa being large and rounded, 

 this is usually a considerable area. It is covered with short stiff hairs, 

 almost like a cocoanut mat in some cases, even when there are hooks ; 

 where there are hooks, they are these same hairs, very numerous, of 

 equal size, and scattered over a considerable area. In a few species, 

 the cremaster seems to have disappeared in so far as there are no hairs 

 or hooks. There are never any spinous hooks or processes such as we 

 meet with in Sphingids, Noctuids, &c. 



The Lachneid pupa has frequently been shown to be not unlike 

 the Notodont, and in some cases the resemblance is somewhat marked. 

 The points of distinction, according to Chapman, are : — 



1. Labial palpi. — Present in Lachneid, not in Notodont pupa. 



2. Dorsal head-jyiece. — Present in Lachneids (usually a very decided quad- 

 rangular piece). In Notodonts, this is usually a very narrow strip, slightly broader 

 at one end, often evanescent, especially in the living pupa, in some of these a trace 

 is visible after dehiscence. 



3. Hairs. — Distributed over general surface in Lachneids. Antenna-basal 

 hairs, and sometimes those of tubercles proper, only present in Notodonts. 



4. Texture. — In Lachneids the texture is more soft, flimsy, and flexible, and 

 where the surface is smooth, is often transparent. In Notodonts the texture is hard, 

 brittle, and rarely or never transparent. 



5. Cremaster. — In Lachneids this consists of fine hairs or hooked bristles 

 spread over a considerable surface. In Notodonts it bears a definite spine or spines 

 with hooks as processes of pupa. Where there is no cremaster the surface is much 

 more polished in Notodonts. 



6. Outline. — In Lachneids the general outline is curved as in Tortricids ; in 

 Notodonts it is straight. 



In a great majority of cases these characters, or some of them, are 

 sufficiently pronounced to enable a pupa to be referred to its proper 

 family. 



The coloration of the imagines is generally brown of some shade. 

 This has probably been developed from a dull buff as exhibited by the 

 undersides of such species as T. crataegi and C. Rotatoria ? . The 

 bright yellow and rich chocolate colorations met with are certainly 

 specialised conditions of this colour. In the markings it will be 

 noticed that there are four transverse lines crossing the forewings : (1) 

 Very short, quite at the base. (2) Two, crossing the centre of the 

 wing, one each side of the discoidal spot. (3) A submarginal wavy 

 line. All the variations in the markings are shown to be modifications 

 of these lines, and we are inclined to look on the fasciated species, as 

 represented by L. quercus, and the unicolorous aberrations of P. trifolii, 

 M. castrensis, &c, as specialised rather than generalised forms. In the 

 hindwings the short basal line is absent, but the three other transverse 

 lines of the forewing are continued on the hindwing in some of the 

 species. These transverse lines — four on forewing and three on hind- 

 wing — are, in some modified form or other, traceable throughout the 

 Saturniids, Endromids, and Sphingids, and also in the Geometrids, 

 Noctuids, Pyralids, &c, and, there can be but little doubt, form one of 

 the most generalised types of marking on lepidopterous insects. 



With regard to the structural peculiarities of the imago we may 

 note that in this superfamily, Kellogg states (Taxonomic value of scales 

 of Lepulojitera, p. 89) that he has examined scales of the North Ameri- 

 can genera — Clisiocavijia*, Artacc, Tolype, HctervpacJia, Gaatropacha/'' , 



* Clisiocamim equals Malacosoma and Gastrojuicha equals Eutricha. 



