swift (Tachornis phænicobea) inhabited exclusively the palm trees in a few districts in the island. A colony then established themselves in two cocoa-nut palms in Spanish Town, and remained there till 1857, when one tree was blown down, and the other stripped of its foliage. Instead of now seeking out other palm trees, the swifts drove out the swallows who built in the Piazza of the House of Assembly, and took possession of it, building their nests on the tops of the end walls and at the angles formed by the beams and joists, a place which they continue to occupy in considerable numbers. It is remarked that here they form their nest with much less elaboration than when built in the palms, probably from being less exposed.

A still more curious example of change and improvement in nest building was published by Mr. F. A. Pouchet, in the tenth number of the Comptes Rendus for 1870, just as the first edition of this work appeared. Forty years ago M. Pouchet had himself collected nests of the House-Martin or WindowSwallow (Hirundo urbica) from old buildings at Rouen, and deposited them in the museum of that city. On recently obtaining some more nests he was surprised, on comparing them with the old ones, to find that they exhibited a decided change of form and structure. This led him to investigate the matter more closely. The changed nests had been obtained from houses in a newly erected quarter of the city, and he found that all the nests in the newly-built streets were of the new form. But on visiting the churches and older

buildings, and some rocks where these birds build, he found many nests of the old type along with some of the new pattern. He then examined all the figures and descriptions of the older naturalists, and found that they invariably represented the older form only.. The difference between the two forms he states to be as follows. In the old form the nest is a portion of a globe-when situated in the upper angle of a window one-fourth of a hemisphere-and the opening is very small and circular, being of a size just sufficient to allow the body of the bird to pass. In the new form the nest is much wider in proportion to its height, being a segment of a depressed spheroid, and the aperture very wide and shallow, and close to the horizontal surface to which the nest is attached above.

is

M. Pouchet thinks that the new form is an undoubted improvement on the old. The nest has a wider bottom and must allow the young ones to have more freedom of motion than in the old narrower, and deeper nests, and its wide aperture allows the young birds to peep out and breathe the fresh air. This is so wide as to serve as a sort of balcony for them, and two young ones can often be seen on it without interfering with the passage in and out of the old birds. At the same time, by being so close to the roof, it is a better protection against rain, against cold, and against enemies, than the small round hole of the old nests. Here, then, we have an improvement in nest building, as well marked as any improvement that takes place in human dwellings in so short a time.

But perfection of structure and adaptation to purpose, are not universal characteristics of birds' nests, since there are decided imperfections in the nesting of many birds which are quite compatible with our present theory, but are hardly so with that of instinct, which is supposed to be infallible. The Passenger pigeon of America often crowds the branches with its nests till they break, and the ground is strewn with shattered nests, eggs, and young birds. Rooks' nests are often so imperfect that during high winds the eggs fall out; but the Window-Swallow is the most unfortunate in this respect, for White, of Selborne, informs us that he has seen them build, year after year, in places where their nests are liable to be washed away by a heavy rain and their young ones destroyed,

Conclusion.

A fair consideration of all these facts will, I think, fully support the statement with which I commenced, and show, that the mental faculties exhibited by birds in the construction of their nests, are the same in kind as those manifested by mankind in the formation of their dwellings. These are, essentially, imitation, and a slow and partial adaptation to new conditions. To compare the work of birds with the highest manifestations of human art and science, is totally beside the question. I do not maintain that birds are gifted with reasoning faculties at all approaching in variety and extent to those of man. I simply hold that the

phenomena presented by their mode of buil ling their nests, when fairly compared with those exhibited by the great mass of mankind in building their houses, indicate no essential difference in the kind or nature of the mental faculties employed. If instinct means anything, it means the capacity to perform some complex act without teaching or experience. It implies innate ideas of a very definite kind, and, if established, would overthrow Mr. Mill's sensationalism and all the modern philosophy of experience. That the existence of true instinct may be established in other cases is not impossible, but in the particular instance of birds' nests, which is usually considered one of its strongholds, I cannot find a particle of evidence to show the existence of any thing beyond those lower reasoning and imitative powers, which animals are universally admitted to dossess.

231

VII.

A THEORY OF BIRDS' NESTS;

SHOWING THE RELATION OF CERTAIN DIFFERENCES OF COLOUR IN FEMALE BIRDS, TO THEIR MODE OF NIDIFICATION.

THE habit of forming a more or less elaborate structure for the reception of their eggs and young, must undoubtedly be looked upon as one of the most remarkable and interesting characteristics of the class of birds. In other classes of vertebrate animals, such structures are few and exceptional, and never attain to the same degree of completeness and beauty. Birds' nests have, accordingly, attracted much attention, and have furnished one of the stock arguments to prove the existence of a blind but unerring instinct in the lower animals. The very general belief that every bird is enabled to build its nest, not by the ordinary faculties of observation, memory, and imitation, but by means of some innate and mysterious impulse, has had the bad effect of withdrawing attention from the very evident relation that exists between the structure, habits, and intelligence of birds, and the kind of nests they construct.

In the preceding essay I have detailed several of these relations, and they teach us, that a consideration of the structure, the food, and other specialities of a