of man, but in the present instance we must speak of him last, because he requires most study. We must begin then with those animals which have shells; we must go on to those which have softer coverings, as crustacea, soft animals, and insects; after these, fishes, both viviparous and oviparous; then birds; then land animals, both viviparous and oviparous."

It is clear from this passage that Aristotle had certain wide and indefinite views of classification, which though not very exact, are still highly creditable to him; but it is equally clear that he was quite unconscious of the classification that has been ascribed to him. If he had adopted that or any other system, this was precisely the place in which he must have referred to and employed it.

The honor due to the stupendous accumulation of zoological knowledge which Aristotle's works contain, cannot be tarnished by our denying him the credit of a system which he never dreamt of, and which, from the nature of the progress of science, could not possibly be constructed at that period. But, in reality, we may exchange the mistaken claims which we have been contesting for a better, because a truer praise. Aristotle does show, as far as could be done at his time, a perception of the need of groups, and of names of groups, in the study of the animal kingdom; and thus may justly be held up as the great figure in the Prelude to the Formation of Systems which took place in more advanced scientific times.

This appears, in some measure, from the passage last quoted. For not only is there, in that, a clear recognition of the value and object of a method in natural history; but the general arrangement of the animal kingdom there proposed has considerable scientific merit, and is, for the time, very philosophical. But there are passages in his work in which he shows a wish to carry the principle of arrangement more into detail. Thus, in the first Book, before proceeding to his survey of the differences of animals," after speaking of such classes as Quadrupeds, Birds, Fishes, Cetaceous, Testaceous, Crustaceous Animals, Mollusks, Insects, he says, (chap. vii.)

"Animals cannot be divided into large genera, in which one kind includes many kinds. For some kinds are unique, and have no difference of species, as man. Some have such kinds, but have no names for them. Thus all quadrupeds which have not wings, have blood. But of these, some are viviparous, some oviparous. Those which are

10 Γένη.

viviparous have not all hair; those which are oviparous have scales." We have here a manifestly intentional subordination of characters: and a kind of regret that we have not names for the classes here indicated; such, for instance, as viviparous quadrupeds having hair. But he follows the subject into further detail. "Of the class of viviparous quadrupeds," he continues, "there are many genera," but these again are without names, except specific names, such as man, lion, stag, horse, dog, and the like. Yet there is a genus of animals that have names, as the horse, the ass, the oreus, the ginnus, the innus, and the animal which in Syria is called heminus (mule); for these are called mules, from their resemblance only; not being mules, for they breed of their own kind. Wherefore," he adds, that is, because we do not possess recognized genera and generic names of this kind," we must take the species separately, and study the nature of each."

These passages afford us sufficient ground for placing Aristotle at the head of those naturalists to whom the first views of the necessity of a zoological system are due. It was, however, very long before any worthy successor appeared, for no additional step was made till modern times. When Natural History again came to be studied in Nature, the business of Classification, as we have seen, forced itself upon men's attention, and was pursued with interest in animals, as in plants. The steps of its advance were similar in the two cases;-by successive naturalists, various systems of artificial marks were selected with a view to precision and convenience;-and these artificial systems assumed the existence of certain natural groups, and of a natural system to which they gradually tended. But there was this difference between botany and zoology:-the reference to physiological principles, which, as we have remarked, influenced the natural systems of vegetables in a latent and obscure manner, botanists being guided by its light, but hardly aware that they were so, affected the study of systematic zoology more directly and evidently. For men can neither overlook the general physiological features of animals, nor avoid being swayed by them in their judgments of the affinities of different species. Thus the classifications of zoology tended more and more to a union with comparative anatomy, as the science was more and more improved." But comparative anatomy belongs to the subject of the next Book; and anything it may be proper to say respecting its influence upon zoological arrangements, will properly find a place there.

11 Είδην.

VOL. IL-27.

Cuvier, Leg. d'Anat. Comp. vol i. p. 17.

It will appear, and indeed it hardly requires to be proved, that those steps in systematic zoology which are due to the light thrown upon the subject by physiology, are the result of a long series of labors by various naturalists, and have been, like other advances in science, led to and produced by the general progress of such knowledge. We can hardly expect that the classificatory sciences can undergo any material improvement which is not of this kind. Very recently, however, some authors have attempted to introduce into these sciences certain principles which do not, at first sight, appear as a continuation and extension of the previous researches of comparative anatomists. I speak, in particular, of the doctrines of a Circular Progression in the series of affinity; of a Quinary Division of such circular groups; and of a relation of Analogy between the members of such groups, entirely distinct from the relation of Affinity.

The doctrine of Circular Progression has been propounded principally by Mr. Macleay; although, as he has shown," there are sugges tions of the same kind to be found in other writers. So far as this view negatives the doctrine of a mere linear progression in nature, which would place each genus in contact only with the preceding and succeeding ones, and so far as it requires us to attend to more varied and ramified resemblances, there can be no doubt that it is supported by the result of all the attempts to form natural systems. But whether that assemblage of circles of arrangement which is now offered to naturalists, be the true and only way of exhibiting the natural relations of organized bodies, is a much more difficult question, and one which I shall not here attempt to examine; although it will be found, I think, that those analogies of science which we have had to study, would not fail to throw some light upon such an inquiry. The prevalence of an invariable numerical law in the divisions of natural groups, (as the number five is asserted to prevail by Mr. Macleay, the number ten by Fries, and other numbers by other writers), would be a curious fact, if established; but it is easy to see that nothing short of the most consummate knowledge of natural history, joined with extreme clearness of view and calmness of judgment, could enable any one to pronounce on the attempts which have been made to establish such a principle. But the doctrine of a relation of Analogy distinct from Affinity, in the manner which has recently been taught, seems to be obviously at variance with that gradual approximation of the classificatory to the phy

13 Linn. Trans. vol. xvi. p. 9.

siological sciences, which has appeared to us to be the general tendency of real knowledge. It seems difficult to understand how a reference to such relations as those which are offered as examples of analogy can be otherwise than a retrograde step in science.

Without, however, now dwelling upon these points, I will treat a little more in detail of one of the branches of Zoology.

[2nd Ed.] [For the more recent progress of Systematic Zoology, see in the Reports of the British Association, in 1834, Mr. L. Jenyns's Report on the Recent Progress and Present State of Zoology, and in 1844, Mr. Strickland's Report on the Recent Progress and Present State of Ornithology. In these Reports, the questions of the Circular Arrangement, the Quinary System, and the relation of Analogy and Affinity are discussed.]

I

CHAPTER VII.

THE PROGRESS OF ICHTHYOLOGY.

F it had been already observed and admitted that sciences of the

same kind follow, and must follow, the same course in the order of their development, it would be unnecessary to give a history of any special branch of Systematic Zoology; since botany has already afforded us a sufficient example of the progress of the classificatory sciences. But we may be excused for introducing a sketch of the advance of one department of zoology, since we are led to the attempt by the peculiar advantage we possess in having a complete history of the subject written with great care, and brought up to the present time, by a naturalist of unequalled talents and knowledge. I speak of Cuvier's Historical View of Ichthyology, which forms the first chapter of his great work on that part of natural history. The place and office in the progress of this science, which is assigned to each person by Cuvier, will probably not be lightly contested. It will, therefore, be no small confirmation of the justice of the views on which the

"For example, the goatsucker has an affinity with the swallow; but it has an analogy with the bat, because both fly at the same hour of the day, and feed in the same manner.-Swainson, Geography and Classification of Animals, p. 129.

distribution of the events in the history of botany was founded, if Cuvier's representation of the history of ichthyology offers to us obviously a distribution almost identical.

We shall find that this is so;-that we have, in zoology as in botany, a period of unsystematic knowledge; a period of misapplied erudition; an epoch of the discovery of fixed characters; a period in which many systems were put forward; a struggle of an artificial and a natural method; and a gradual tendency of the natural method to a manifestly physiological character. A few references to Cuvier's history will enable us to illustrate these and other analogies.

Period of Unsystematic Knowledge.—It would be easy to collect a number of the fabulous stories of early times, which formed a portion of the imaginary knowledge of men concerning animals as well as plants. But passing over these, we come to a long period and a great collection of writers, who, in various ways, and with various degrees of merit, contributed to augment the knowledge which existed concerning fish, while as yet there was hardly ever any attempt at a classification of that province of the animal kingdom. Among these writers, Aristotle is by far the most important. Indeed he carried on his zoological researches under advantages which rarely fall to the lot of the naturalist; if it be true, as Athenæus and Pliny state,' that Alexander gave him sums which amounted to nine hundred talents, to enable him to collect materials for his history of animals, and put at his disposal several thousands of men to be employed in hunting, fishing, and procuring information for him. The works of his on Natural History which remain to us are, nine Books Of the History of Animals; four, On the Parts of Animals; five, On the Generation of Animals; one, On the Going of Animals; one, Of the Sensations, and the Organs of them; one, On Sleeping and Waking; one, On the Motion of Animals; one, On the Length and Shortness of Life; one, On Youth and Old Age; one, On Life and Death; one, On Respiration. The knowledge of the external and internal conformation of animals, their habits, instincts, and uses, which Aristotle displays in these works, is spoken of as something wonderful even to the naturalists of our own time. And he may be taken as a sufficient representative of the whole of the period of which we speak; for he is, says Cuvier, not only the first, but the only one of the ancients who has treated of the natural history of fishes (the province to which

1 Cuv. Hist. Nat. des Poissons, i. 13.

2 Cuv. p. 18.