has clearly proved, from various experiments which he himself made, that these birds cannot continue long under water without being drowned. I do not mean to deny that swallows may have occasionally been found under water; for it is well known that they have and this probably has given rise to the absurd notion entertained by some, that the whole of the species winter in that element. But I should certainly attribute their being found in such situations to mere accident; and I think it might have been occasioned by some such circumstance as the following:

:

It is well known that, towards the latter end of autumn, swallows frequently roost by the sides of lakes and rivers; we will suppose, therefore, that a number of these birds had retired to roost on the banks of some shallow and muddy river at low tide, and that they had been induced by the cold to creep among the reeds or rushes which might grow in the shallow parts of the river, and that while in this situation, driven into a state of torpidity by the cold, they had been overwhelmed, and perhaps washed into the current, by the coming in of the tide.3

Having thus accounted for the manner in which swallows might chance to get into the water, it remains to be considered by what means they may have been sometimes taken out alive. Let us suppose, therefore, that some fishermen, as is very likely to be the case, had availed themselves of the coming in of the tide to catch fish, and that the swallows, which we have before supposed to have been carried into the current, coming in contact with their nets, were consequently drawn out of the water by them, and, not having been long under water, were not completely drowned.

There are several other circumstances which seem to favor the opinion, that these birds remain concealed during winter in this country: Among others, the most striking is, that swallows,

'The experiments of Mr. Pearson, related in Bewick's Birds, show, I think, that the swallow has no great propensity to become torpid in winter, unless operated on by some other circumstance than merely the time of the year.

2 This circumstance may have contributed to induce some to believe that they go into the water.

3 There have been occasional instances of other birds besides swallows having been found in a state of torpor during winter. I think I recollect a cuckoo being found in such a state.

hirundines rusticae, as well as martins, hirundines vrbicae, have sometimes appeared very late in autumn, a considerable time after they were all supposed to have taken their departure.'

Again, as I have before had occasion to observe, they have sometimes been taken out of the water, in winter, in a torpid state, not only out of rivers, but also out of lakes, and stagnant pools, and even out of bogs." They have likewise been found concealed in the crevices of rocks, in holes of old decayed trees, in old ruined towers, and under the thatch of houses.3

From the consideration of the above facts alone, without making any further inquiry into the subject, many persons have concluded that the whole tribe always winter in similar situations. It seems, however, much more probable that those birds, which may have been found in a state of torpidity, as above described, had, owing to some accident, been hatched later in the year than ordinary, and that consequently they had not acquired sufficient strength to undergo the fatigue of a long journey upon the wing, at the time when the migration of the rest of their species took place. It is very probable that many of these, in order to shelter themselves from the inclemency of the weather, may have retreated to holes of rocks, and other recesses, where, from cold and hunger, they may

'Of this we have several instances: Bewick, in his History of British Birds, Introduction, p. xvii. takes notice of having seen a straggling swallow so late as the end of October; and White, in his Natural History of Selborne, mentions having seen a house martin flying about in the month of November. Montagu, in the Supp. to Orn. Dic. mentions having seen many swallows and martins as late as November, 5, 1805. To which I may add, that, in the year 1804, I saw several, both martins and swallows, flying about in the neighbourhood of London, as late as October the 19th. I have since frequently seen them later than that time.

2 For further particulars relative to the torpidity of swallows, see Miscel lanies by the Hon. Daines Barrington, page 225 and sequel; also Buffon, Hist. Nat. des Oiseaux, 4to. Paris, 1780, Plan d'Ouvrage, p. xiii.

3 A great many sand martins' holes have been opened in winter, and nothing has been found in them but old nests.-See Phil. Trans. vol. li. p. 463.

In October, 1810, I opened several sand martins' holes near Dorking in Surry, and found in one of them a variety of very small bony substances, which might be part of large insects, mixed with dirt.

I

have sunk into a state of torpidity. Others, for the same reason, may have crept among the weeds, which grow by the sides of rivers and ponds, where they may have been overwhelmed by the increase of the water, occasioned by the heavy rains which often happen towards the end of autumn, and some, which may not have been long immersed, may probably have been restored to life, when brought into the sunshine, or before a fire.

But that the chief part of each species migrate, is so well established by a multitude of corresponding facts, that it seems almost an absurdity to doubt of it. In the first place I would observe, that if these birds lay concealed in winter, in the same countries which they inhabit in summer, they would probably make their first appearance in spring, in mild weather, and would appear sooner in early than in late seasons, which is quite contrary to experience. For several years past I have observed that chimney swallows have appeared first in cold weather. I have sometimes seen them as early as April the 2d, when the mercury in the thermometer has been below the freezing point. On the other hand, I have often taken notice, that during a continuance of mild weather for the space of a fortnight, in the month of April, not so much as one swallow has appeared.

It is a well-known fact, that the swallow, like most other birds of passage, appears earlier, and departs later, in the southern than in the northern parts of Great Britain; and it must have been observed, by every one who is attentive to natural history, that towards the latter end of September, swallows, hirundines rusticae, as well as martins, hirundines vrbicae, congregate in great numbers, and are frequently seen sitting on the tops of houses, and on rocks near the sea. These meetings usually continue for several days, after which they suddenly disappear."

1 See Montagu's Ornithological Dictionary, under the word martin.

It is by no means improbable that very cold and frosty weather in spring may sometimes drive the swallow, just arrived, into some snug retreat, where it may remain until the warm weather returns.-See Phil. Trans. vol. lxv. p. 259.

2 Swallows seldom perch on trees, except in autumn, shortly previous to their disappearance, and they then choose dead trees in preference. I have known them sit on trees earlier in summer, when the weather has been very cold. No. VIII.

Pam.

VOL. IV.

2 G

Swifts, hirundines apodes, also begin to assemble in large bodies previous to their departure, early in July: their numbers daily increase, and they soar higher in the air, with shriller cries, and fly differently from their usual mode. Such meetings continue till towards the middle of August, after which they are seldom seen.

Sand martins, hirundines ripariae, likewise flock together in autumn. Some years ago they are said to have appeared in great numbers in London and its neighbourhood.

From all the abovementioned circumstances, as well as from the great length of the wings, in proportion to the bulk of the body, of all this genus,' it must appear evident that swallows are birds of passage for it is hardly to be supposed that they would assemble together merely to hide themselves; on the contrary, it is most probable that, were this the case, each individual bird would seek a hiding place for itself.2

It will be proper now to examine the accounts of mariners and others, who have seen these birds on their passage, many hundred miles out at sea, and on whose ships they have alighted to rest, almost exhausted with fatigue and hunger; by which means we may be enabled, in some measure, to determine to what quarter of the globe they retire, when they leave Europe in autumn.

Adanson, in his Voyage to Senegal, relates, that on the sixth of October, being about fifty leagues from the coast, between the island of Goree and Senegal, four swallows alighted on the shrowds of his ship, which he easily caught, and knew to be European swallows. He adds, that these birds never appear at Senegal, but in the winter season, and that they do not build nests as in Europe, but roost every night on the sand by the sea shore. It is much to be lamented that Adanson, who was a naturalist, did not mention of what species these birds were. It is, however, most probable,

"If we calculate the velocity of this bird on the wing, and that it can and does suspend itself in the air for fourteen or fifteen hours together in search of food, it cannot fly over a space of less than two or three hundred miles in that time."-Montagu Ornith. Dict.

2 If the swift hirundo apus lay torpid during the whole of its absence, it inust sleep for a continuance of above eight months, and would afford an instance of such a preponderance of torpidity over animation, as occurs in no other known animal.

as they were seen at Senegal on the sixth of October, that they were chimney swallows, hirundines rusticae, as martins, hirundines vrbicae, seldom leave their summer haunts till after that time; and swifts, hirundines apodes, usually depart before the twenty fifth of August. With respect to bank martins, hirundines ripariae, it is very unlikely that Adanson should have mistaken them for chimney swallows, being distinguishable at first sight from the three abovementioned species, by their inferior size.

Latham, who quotes Adanson's account, evidently understood the birds in question to be chimney swallows, and supposes Seuegal and the adjacent parts of Africa to be the winter residence of this species. The writers of Le Nouveau Dictionnaire d'Histoire Naturelle, published at Paris in 1803, seem to have been of the same opinion. Celles (les hirondelles) de cheminée, say they (vol. xi. p. 18), vont jusqu'au Sénégal, ou elles arrivent vers le 9 Octobre, et en repartent au printems. Il n'est pas rare dans les migrations d'en voir en mer, qui lorsqu'elles sont trop fatiguées se reposent sur les vergues des navires; et parmi elles, on a reconnu celles qui habitent parmi nous.

Another account, which affords additional proof that swallows are birds of passage, though it will not assist us in discovering to what part of the world they go, is that of Sir Charles Wager, first lord of the admiralty; who relates, that in one of his voyages home, as he came into soundings of our channel, a great flock of swallows settled on his rigging: every rope was covered with them they hung on one another like a swarm of bees: the decks and carvings were filled with them: they seemed spent and famished, and, to use his own expression, were only feathers and bones; but, recruited with a night's rest, they resumed their flight in the morning.

Peter Collison, F. R. S., in a letter to the Hon. 1. T. Klein, mentions, that a similar circumstance happened to Captain Wright, in a voyage from Philadelphia to London; the particulars of which, it appears, the captain neglected to relate.'

• Collison adds: "I have for many years been very watchful in taking notice of the times when the swallows leave us, and I think I have twice actually seen them taking their flight. At two different years (on the 27th and 29th of September) walking in my garden at noon, on very sun-shiny days, and looking up into the sky, I distinctly saw an innumerable number