The caterpillar has a green head, which is larger than the first ring of the body (1st thoracic), covered with short fine whitish hairs, with which are mixed a few dark hairs. The body is rather brighter green, with darker lines, edged with yellowish, along the back and sides; the skin is transversely wrinkled, the rings being subdivided, and the whole of the body is clothed with fine whitish hair and a few dark hairs arising from warts; the anal points are whitish and also hairy. It feeds on various grasses, among which are _Triticum repens_ and _Dactylis glomerata_. [Illustration: _Pl._ 80. =Speckled Wood.= _Spring Brood:_ 1, 2 _male_; 3, 5 _female_. _Summer brood:_ 4, 6 _male_; 7 _female_.] [Illustration: _Pl._ 81. =Speckled Wood.= _Egg, natural size and enlarged; caterpillar and chrysalis._] The chrysalis is pale green, tinged with yellowish or whitish; the edges of the wing covers are brown, and there are whitish dots on the body. According to Hellins the colour varies, and green chrysalids may be covered all over with very fine smoky freckles. Barrett states that they are occasionally brownish with darker brown lines. Suspended by the cremaster from a silken pad. From eggs laid in early May butterflies were reared at the end of June; and from eggs laid at the end of June butterflies resulted during middle August. Early July eggs produced perfect insects in early September, and from caterpillars fed up in October butterflies were obtained in November. These observations were not all made in the same year. Barrett writes, "In the south of Surrey in 1862, the first emergence took place in April in abundance, these specimens became worn and disappeared, and a second emergence took place at the end of May, a third at the end of July, and a fourth in September; the next year the first emergence was in the third week in March, and again four broods were observed, but this is not the case every year, three emergences being probably the rule." Mr. Joy has recorded that of caterpillars, resulting from a pairing induced in captivity, in August, eighty per cent. hibernated as pupæ, twenty per cent. as half-fed caterpillars. Butterflies from the winter pupæ emerged in May, but the caterpillars that had gone through the winter in that state did not produce butterflies until June. Possibly something of this sort occurs in the open, and we may suppose that the early and late spring butterflies are not separate broods, but early and late emergences of one brood. Butterflies seen on the wing in November may be a few individuals that, owing to favourable weather, have emerged from chrysalids which under ordinary conditions would have remained as such during the winter.