LACHNEIDES. 437 



wise specialised larva retaining a useful ancestral character in the posi- 

 tion of the warts, which parts have, however, been specialised in another 

 direction by the development of the lateral hairs arising therefrom, in 

 order to meet new protective needs. These hairs bend downwards to 

 the resting-place, and thus produce a continuity of surface, which is 

 an aid to the perfecting of the resemblance which the larva bears to its 

 surroundings, by minimising the effect of light and shade. It is the 

 lateral hairs (from iv and v) that most largely produce this result. 

 The dorsal secondary hairs of some Lachneid groups develop into the 

 well-known, offensive, urfcicating hairs. With regard to the ordinary 

 primary hairs of the larvae of this group, it would appear that those of 

 many species are finely serrated in the first instar, but the character 

 is soon lost, and no larva of the British species seems to have such 

 hairs in the adult stage. One would suspect, however, from the 

 frequency with which they occur in the early stages that the ancestral 

 Lachneid larva had serrated hairs — those of Poecilocampa populi, 

 Trichiura crataegi, Malacosoma neustria, M. franconica, M. castrensis, 

 Pachi/t/astria trifolii, and Cosmotriehe Rotatoria have them when young 

 — but the character has been gradually lost, and is now, as we have 

 just noted, only present in the early instars of certain larvae. The 

 development of secondary hairs is, however, very remarkable. These, 

 in certain specialised larvae, present very striking developments — scale- 

 like, spear-like, grass-blade-like, &c, in Eutricha quercifolia, Dendro- 

 limus pini, &c, besides the long, fine, spindle-shaped urfcicating hairs, 

 in Macrothylacia rubi, Lasiocampa quercits, &c. 



A most interesting note on the structure of certain scales (or 

 modified hairs) found in the larva of E. quercifolia is written by 

 Packard, who says that, in examining the median dorsal tufts on the 

 2nd and 3rd thoracic segments of this species, he found that they 

 were composed of broad, lanceolate, oval scales, which were opaque 

 and dark steel-purple in colour, with the surface quite regularly 

 striated, though not invariably so ; the striae not appearing to extend 

 to either end. These scales vary in shape and size, some being narrow 

 and with a simple point at the distal end, while the majority are 

 variously notched or toothed. They thus appear to be true scales, 

 like those on the wings of lepidoptera. In the same species, the 

 lateral tufts along the body, each contain a few long hairs Avith 

 flattened ends, the latter varying in shape from oval to triangular, 

 with the ends often very broad and ragged, and with from one to four 

 very irregular teeth ; no striae are perceptible on these, and the hairs 

 throughout are colourless and transparent. He then adds that, on 

 examining the lateral tufts of Gastropacha americana, he found " some 

 very long similar hairs flattened at the end, and of extraordinary form, 

 usually projecting beyond the simple hairs ; some ending in regular 

 lanceolate oval shapes, with the point much attenuated, others broader, 

 while some are oval and broad at the end which terminates in a fine 

 attenuated point, with usually three minute teeth at the base. They 

 are similar in shape to those of Gastropacha quercifolia. On turning 

 over the beautiful plates of Uurmeister's Atlas of the Lepidoptera of the 

 Argentine Republic, one finds that the author represents on pi. 22, fig. 

 9, similar long hairs, much flattened and expanded at the ends, with 

 three, four, or five long slender teeth, in the larva of his Olisiocampa 

 proximo,, which, however, seems to differ from Clisiocampa proper. The 



