in the felted body, while the extremities are free. It is from this tendency to felt that woollen cloth and stockings increase in density, and contract in dimensions, by being washed. In many places woollen stuffs are felted on a small scale by placing them in running water and under cascades: the Zetlanders expose them to the motions of the tides in the narrow inlets of the sea*.

The goldfinch (Carduelis communis, CUVIER) is more neat in the execution of its felting than the chaffinch, though its nest is not quite so tasteful; for the goldfinch's is rendered more formal and less richly varied in its colouring, by the anxiety which the bird displays not to leave a single leaf of moss or lichen projecting, all being smoothly felted with wool, which in some measure conceals the moss; whereas, in the chaffinch's nest, the lichens usually conceal the wool. In other respects the two nests are much the same, as well as the eggs; those of the goldfinch having their white ground more commonly tinged with blue, and having fewer and rather brighter spots, which are dark in the centre, and shade off into a thinly-spread purple colour. Bolton describes the goldfinch's nest as bound withi blades of dried grass and a few small roots; a circumstance which has not fallen under our observation, though this may be sometimes resorted to, for it may be seen in the nests of some chaffinches and not in others. Bolton found his nest in the bough of a plane tree (Acer Pseudo-platanus); but we have usually met with them in orchards, on elms, and more rarely in hawthorn hedges. The lining of thistle-down, ascribed to the nest of the goldfinch in most books of natural history, must be a mistake, at least with respect to the nests built in May and early in June; for none of our native thistles flower *Fleming, Philos. of Zoology, i. 89.

before the end of June, and none have down, we believe, before July. The bottom of the nest now before us is bedded with small tufts of fine wool, not much spread, and the sides with the down of coltsfoot (Tusilago farfara), and only one or two hairs or feathers; whereas the chaffinch uses little down, and seems partial to a lining either of cow's hair alone, or intermixed with a few soft feathers, neatly woven into the cup of the nest. The goldfinch frequently uses nothing besides cotton-wool for a lining.

Nest of the Goldfinch (Carduelis communis),

The truth is, that birds will in general take the materials for building which they can most easily procure. "On the 10th of May, 1792," says Bolton, "I observed a pair of goldfinches beginning

circumstance, that birds which continue in song nearly the whole year, such as the red-breast, the siskin, and the goldfinch, are obliged, after their moulting is over, to record, as if they had forgot their song. I am convinced, however, that this exercise is less a study than an endeavour to bring the organs of voice into proper flexibility, what they utter being properly only a sort of warble, of which the notes have almost no resemblance to the perfect song; and, by a little attention, we may perceive how the throat is gradually brought to emit the notes of the usual song. This view, then, leads us to ascribe the circumstance not to defect of memory, but rather to a roughness in the vocal organs, arising from disuse. It is in this way that the chaffinch (Fringilla spiza) makes endeavours during several successive weeks before attaining to its former perfection, and that the nightingale (Sylvia luscinia) tries, for a long time, to model the strophes of its superb song before it can produce the full extent of compass and brilliance *."

It might be alleged, indeed, that the old birds who sing in autumn, are influenced by association, inasmuch as this season resembles the spring; for though spring is all youth and verdure, while autumn wears the aspect of decline, and woods and fields, instead of lively green, display nothing but sombre tints of yellow and brown, yet the temperature of the air has much the same mildness, and food is equally if not more abundant. This, however, is a very partial view, with which the continuance of these autumnal songs during winter is altogether inconsistent. Colonel Montagu, however, endeavours to obviate the objection by some ingenious arguments; and, in support of the general theory, he brings forward experiments tried by himself for that express purpose. The continuation of song," he says, "in caged

* Ornithologisches Taschenbuch, Vor.

birds, by no means proves it is not occasioned by a stimulus to love; indeed, it is likely the redundancy of animal matter, from plenty of food and artificial heat, may produce it; and this is sufficient for continuing their song longer than birds in their natural wild state, because they have a constant stimulus ; whereas wild birds have it abated by a commerce with the other sex, by which, and other causes, it is prevented. It is true wild birds are heard to sing sometimes in the middle of winter, when the air is mild, animated by the genial warmth of the sun, which acts as a stimulus. But we shall now proceed to show, by experiments, that birds in their natural state may be forced to continue their song much longer than usual. A male red-start made its appearance near my house early in the spring, and soon commenced his love-tuned song. In two days after, a female arrived, which for several days the male was continually chasing, emitting soft interrupted notes, accompanied by a chattering noise. This sort of courting lasted for several days, soon after which the female took possession of a hole in a wall close to my house, where it prepared a nest and deposited six eggs. The male kept at a distance from the nest, and sometimes sang, but not so loud or so frequently as at first, and never when he approached nearer his mate. When the eggs had been sat on a few days, I endeavoured to catch the female on the nest, but she escaped through my hand. However, she soon returned, and I caught her. The male did not immediately miss his mate; but on the next day he renewed his vociferous calls, and his song became incessant for a week, when I discovered a second female; his note immediately changed, and all his actions as before returned. This experiment has been repeated with the nightingale with the same success; and a golden-crested

wren, who never found another mate, continued his song from the month of May till the latter end of August. On the contrary, another of the same species, who took possession of a fir-tree in my garden, ceased its notes as soon as the young were hatched *"

To us, however, this explanation of the facts appears too partial and contracted, the song of the birds being more naturally accounted for, as we think, from the state of their spirits than by the supposition of its having been meant as the language of courtship; and, accordingly, it does not consist with our observation that the state of the weather has much influence upon them, except in so far as it may affect their supply of food; and hence it is that caged birds remain much longer in song than if they were at large in the fields. We have further remarked, and it agrees with the experience of Mr. Sweet and others who keep tame birds, that the male will sing better, and for a longer period, when there is a female of his own species in the same cage than when he is alone; whereas, according to Colonel Montagu's explanation of his experiments, namely, that the song is uttered chiefly to attract the female to the vicinity, this circumstance ought not to take place.

The theory in question has been opposed by another which maintains the peculiar notes of various song-birds to be derived from imitation. The Hon. Daines Barrington tried a number of experiments for the purpose of supporting this latter notion, which it may prove interesting to give in his own words :

I have educated nestling linnets," says he, "under the three best singing larks, the skylark, woodlark, and titlark, every one of which, instead of

*Ornith. Dict. Intr. first ed.; p. 477, second ed.

† J. R.