completes its growth about the month of October or November. It now prepares for its first metamorphosis by ceasing to eat, and thus completely emptying its digestive tube; then it seeks the hollow of some tree, or hole in some neighbouring wall, and having found a suitable spot, it begins its preparations.
Unlike the silkworm, this caterpillar spins no cocoon for its concealment and protection, but undergoes its metamorphosis in the open air. It now commences covering the spot it has chosen with filaments which cross each other in every direction; and this silken couch, delicate in texture, but withal of considerable strength, serves as a solid and firm support for the hinder limbs. Then bending its trunk and head posteriorly almost to the middle of the back—like an acrobat who makes a hoop of his body—it fixes a thread first on one side and then upon the other, and continues the operation till it has formed a kind of girth, composed of about fifty filaments. This done, it straightens its body, and undergoes its last moulting; the animal however which emerges from the cast-off skin is no longer a caterpillar, but a chrysalis, which is sustained horizontally by the hooklets of its tail and the girth we have described.
The Pieris in the new condition which it will maintain during the winter, bears hardly any resemblance to the caterpillar. The skin, which is dense and horny, is covered with a sort of varnish, thrown out at the moment of the metamorphosis, and rapidly dried. It has now assumed an ashy hue, picked out with black and yellow. The body has become thicker, but, as it were to compensate for this increase, has been shortened by about one-third. Instead of being made up of rings from end to end, it now exhibits two principal segments. The hinder one alone, which is short and conical, presents the annulose condition, although there is a keel-like elevation upon the dorsal portion of the anterior segment, and a kind of crest upon its undersurface. The head and feet seem at first sight to have disappeared altogether. On closer examination, however, we can detect a series of rounded elevations and projections, arranged symmetrically. Knowing what this insert mass will eventually become, we can almost fancy that we see the various organs beneath the skin, or rather beneath the cement which invests it; the proboscis, antennæ, and wings being indicated in the same manner as the form and proportions of a mummy are rudely mapped out by the bandages which enshroud it. To all intents and purposes the chrysalis is a mummy.
About the middle of spring, or beginning of summer, the Pieris undergoes its second metamorphosis. Its envelope splits along the dorsal portion, and the organs which had been inclosed by the crests and elevations come out, as if from a case; then the entire animal disengages itself, and from the chrysalis coffin there emerges a perfect butterfly. At first its feet are of too pliant a character to support the body; the wings are heavy, thrown into microscopic zigzag folds, and unfit for flight; and the proboscis with its component halves often separate, is extended in a right line. But after a while the surrounding liquids are evaporated, the limbs are strengthened, the proboscis is adjusted and coiled up, the wings are unfolded, and the insect, which in its early days was "a creeping thing," and afterwards a motionless one, flies to the nearest flower and makes its first repast.
VARIATIONS IN BRITISH PLANTS.
I have here set down a few examples of variation which have come under my notice during the last two or three years. I have observed that the species contained in the order Ranunculaceæ have a peculiar tendency to depart from their ordinary forms; and many instances of such variations occur in the earlier numbers of the Naturalist. In addition to those therein recorded, I may mention that I have observed in Brompton cemetery a curious variety of the Bulbous Crowfoot (Ranunculus bulbosus), which bears flowers of a pale yellow or cream colour; though it differs in no other respect from the typical form: this variety is apparently constant, about a dozen examples having been observed in the same spot for the last four or five years. A friend has observed a single plant of the same species, at High Wycombe, Bucks, having blossoms "as double as those of the Dahlia," the whole of the stamens being transformed into petals. Another species, the Creeping Crowfoot (R. repens) is peculiarly variable in the relative number of stamens and petals; and semi-double varieties are in some places equally common with the ordinary form. The Lesser Spearwort (R. Flammula) also occasionally exhibits a similar peculiarity. The singularly deformed petals of the Wood Crowfoot (R. auricomus) must be familiar to all who know the plant; regularly shaped
