the application of the principle: but the variation never proceeds beyond certain limits; and therefore no more interferes, eventually, with the uniformity of the specific character of animals, than the periodical oscillations of the celestial bodies counteract the general regularity of their motions.

We are now therefore to consider the nature of the disturbing cause, if I may borrow that expression for a moment, which occasionally interferes with the uniformity of specific character. And, with respect to specific forms, it may be remarked; that, although it is to a certain extent true that all organized bodies have the power of producing beings resembling themselves, yet circumstances of temperature, and of quantity or quality of food, and other causes, have usually some influence in the developement of the body of each individual; thereby producing some corresponding variation in the form: and, consequently, the resemblance between the parent and offspring is never perfect. But-and this is a fact of the highest importance there is no ground for believing that such variations proceed beyond certain limits; no ground therefore for believing that any of the above mentioned circumstances could have produced all the differences perceptible in organized bodies; could have advanced for instance, by a gradual alteration of structure, a lower to a higher species. Experience, on the contrary, founded on an examination of the records of remote antiquity, seems to show that the limits of variation were ever the

same that they are now. It appears for instance rom the mummies of Egypt,* that the general form, and size, and proportions were the same three thousand years since, that they are at present; as well in various other animals as in man; and in all physiological probability therefore were the same three thousand years before that period: so that we cannot refuse to admit, that certain forms have, without

* Vid. Cuvier, Oss. Foss, i. Disc. Prelim. p. 75, 80.

exceeding the limits above discribed, been perpetuated from the creation.

From various circumstances, however, as has been already stated, the offspring never exactly resembles the parent; and by the extension of those causes which occasion a difference of character, the variation from the common parent may possibly become so great, and so permanent in individuals of the same species, as to exceed in some respects the difference observable in individuals of different species. Such appears to be the fact, when, in the dog species, we compare the grayhound with the turnspit; or the Newfoundland-dog with the Blenheim spaniel: and yet, even in such instances, which perhaps may be considered as comprising the extreme limits of variation, the specific character is never so far obscured, but that a child who had been accustomed to see a variety of dogs, and also of other animals, would recognise the character of the dog in each individual of that species.

It is true, indeed, that it would be difficult not only for a child, but even for the most experienced observer, to define those characters by which the specific resemblance is recognised upon a transient view of the animal. Yet, although not obvious on a superficial examination, nature has not left this point undefinable: for, in almost every instance, the form and number of the bones are so accurately preserved, that, however the colour, or the size and the general form of the body may be altered, we have satisfactory criteria of the species in the points just mentioned. But, of all the constituent parts of the body, this observation holds most eminently with respect to the teeth: and in the case of quadrupeds, which principally constitute the highest class of the animal kingdom, and in which class alone any considerable degree of variation is likely to be observed, we have almost always a ready mode of judging of the identity of specific character by an examination of the teeth; for they in almost every instance have teeth, which

are entirely wanting throughout the whole class of birds, and often in reptiles and in fish.

In investigating the remote causes of specific variation, we find that domestication is the most general and extensive; and that the effects are produced principally by the joint operation of the following means, namely, diet, general regimen, and the due selection of individuals for the purpose of breeding.* While animals exist in a state of nature, it does not appear that the circumstances in which they are placed give rise to much variation, even in their external and fugitive characters. A uniformity of size and colour is usually observable in the several individuals of the same species; as in the instances of the wild cat and rabbit. Nor is the character liable to be changed by intercourse among individuals of different species. Although, for instance, the hare and rabbit are so nearly allied in form and size and colour, we never meet with a hybrid or mule of those species.

In domesticated species a variation first in colour and then in size, usually takes place, to an extent proportional to the degree of domestication. Cats, which are less subjugated to man than horses or dogs, vary little more than in colour; scarcely at all in size. And in horses, on the same principle, there is a less degree of variation that in dogs. In the dog, which is of all species the most domesticated, the variation extends to the production of an additional toe, and corresponding metatarsal bone in the

*Burkhardt observes, in his notes on the Bedouins, p. 111, and 139, that in barren parts of the desert of Arabia, or in seasons of scarcity, camels and sheep do not multiply so extensively as in fertile plains and seasons. A similar observation would probably hold good with respect to the ratio of increase among the Tchutzki and other tribes of north-eastern Russia, and the inhabitants of New Holland or any other part of the world where the supply of food is scanty.

See, on this subject a letter, published by sir John Sebright in 1809, on the art of improving the breeds of domestic animals.

hind foot.* And in the human species, in the individuals of which, from their varied intercourse and modes of living, the limits of variation may antecedently be expected to have the widest range, there are families having six fingers.

In concluding this part of the subject, I would observe that the principle, which we have just now been examining, is of very great importance as the basis of a physiological argument with reference to the identity of the human species throughout the world. For, inasmuch as all the variations in colour, form, and size, of the different nations of mankind, come within the acknowledged limits of specific variation in the animal kingdom, we have hence satisfactory physiological proof that all the varieties. of the human race may have proceeded from one common parent. Of the truth of the general position indeed, of which the human species is a particular instance, the work of Aristotle now under consideration is in itself a strong argument: for, notwithstanding the lapse of ages which has taken place since it was written, the description of many species is so accurate, as to leave no doubt of the identity of those described by Aristotle with those to which the description is applicable at the present day.t

* Ann. du. Mus. tom. xviii. p. 342. pl. 19.

It can hardly escape observation, or fail to excite surprise, 'that in the work now under consideration, Aristotle usually contents himself with stating facts: he very rarely reasons on their final causes; thus omitting what Cuvier calls one of the most beautiful and useful points in natural history. The following are, I believe, the only instances in which he deviates from mere description. He observes, when speaking of fish, that a great proportion of the spawn of those animals is destroyed in various ways, and that if this were not the case the species would become too

numerous.

(Τὰ μὲν πολλὰ ὠδ ιἱ ἄρρενες ανακάπτουσι, τὰ δ ̓ ἀπόλλυται ἐν τῷ ὑγρῶ ὅσα δ' ἂν ἐκτέκωσιν εἰς τούς τόπους εἰς οὓς ἐκτίκτουσι, ταῦτα σώζεται· εἰ γὰρ πάντα ἐσώζετο, παμπληθὲς ἄν τὸ γένος ἣν ἑκάστων. Ρ. 169.)

SECTION IV.

On those Animal Forms called Monsters, or Lusus Naturæ.

THE subject of the present section is naturally connected with that of the latter part of the preceding: and, although the occasion neither requires nor would justify even a brief examination of the laws which regulate the formation of monsters, or lusus naturæ, as they are often called, especially as they have been lately illustrated by that ardent French physiologist Geoffroy St. Hilaire; it will not be perhaps considered impertinent to make a few observations on those remarkable productions, considered with respect to one of the probable final causes of their existence.

The term lusus naturæ is applied to those natural productions, which vary in any remarkable degree, with respect to form, colour, structure, size, &c. from the general character of the individuals of the same species. The term literally taken, implies a sportive effort of the creative power of Nature; and for the purpose of general description there is no objection to this term, being, as it now is, familiarized by long continued use. But as we have no ground for supposing that nature, or, to use the more proper expression, that the providence of the Creator ever acts without some wise and beneficent purpose, we must consider the term in a philosophical point

On another occasion he observes, that though the spring is the general season for propagation, yet occasionally the rule is set aside; where, for instance, the preservation of the offspring is the result:

(Ορμητικώτατα μὲν οὖν ὡς ἐπὶ τὸ πᾶν εἰπεῖν πρὸς τὴν ὀχείαν τὴν ἐαρινὴν ὥραν ἐστίν· οὐ μὴν τὰ πάντα γε ποιεῖται τὸν αὐτὸν καιρὸν τῆς ὀχείας, ἀλλὰ κ πρὸς τὴν ἐκτροφὴν τῶν τέκνων ἐν τοίς καθήκουσι καιροῖς. Ρ. 181.)