It is with much interest, then, that we turn to Cuvier’s great work, which brought him such immediate and widespread fame, in order to see how he treated his subject. His general views are contained in the preliminary remarks in his well-known “Essay on the Theory of the Earth” (1812), which was followed in 1821 by his Discours sur les Révolutions de la Surface du Globe.
It was written in a more attractive and vigorous style than the writings of Lamarck, more elegant, concise, and with less repetition, but it is destitute of the philosophic grasp, and is not the work of a profound thinker, but rather of a man of talent who was an industrious collector and accurate describer of fossil bones, of a high order to be sure, but analytical rather than synthetical, of one knowing well the value of carefully ascertained and demonstrated facts, but too cautious, if he was by nature able to do so, to speculate on what may have seemed to him too few facts. It is also the work of one who fell in with the current views of the time as to the general bearing of his discoveries on philosophy and theology, believing as he did in the universality of the Noachian deluge.
Like Lamarck, Cuvier independently made use of the comparative method, the foundation method in palæontology; and Cuvier’s well-known “law of correlation of structures,” so well exemplified in the vertebrates, was a fresh, new contribution to philosophical biology.
In his Discours, speaking of the difficulty of determining the bones of fossil quadrupeds, as compared with fossil shells or the remains of fishes, he remarks:[96]
“Happily comparative anatomy possessed a principle which, well developed, was capable of overcoming every difficulty; it was that of the correlation of forms in organic beings, by means of which each kind of organism can with exactitude be recognized by every fragment of each of its parts.—Every organized being,” he adds, “forms an entire system, unique and closed, whose organs mutually correspond, and concur in the same definite action by a reciprocal reaction. Hence none of these parts can change without the other being also modified, and consequently each of them, taken separately, indicates and produces (donne) all the others.
“A claw, a shoulder-blade, a condyle, a leg or arm-bone, or any other bone separately considered, enables us to discover the kind of teeth to which they have belonged; so also reciprocally we may determine the form of the other bones from the teeth. Thus, commencing our investigation by a careful survey of any one bone by itself, a person who is sufficiently master of the laws of organic structure can reconstruct the entire animal. The smallest facet of bone, the smallest apophysis, has a determinate character, relative to the class, the order, the genus, and the species to which it belongs, so that even when one has only the extremity of a well-preserved bone, he can, with careful examination, assisted by analogy and exact comparison, determine all these things as surely as if he had before him the entire animal.”
Cuvier adds that he has enjoyed every kind of advantage for such investigations owing to his fortunate situation in the Museum of Natural History, and that by assiduous researches for nearly thirty years[97] he has collected skeletons of all the genera and sub-genera of quadrupeds, with those of many species in certain genera, and several individuals of certain species. With such means it was easy for him to multiply his comparisons, and to verify in all their details the applications of his laws.
Such is the famous law of correlation of parts, of Cuvier. It could be easily understood by the layman, and its enunciation added vastly to the popular reputation and prestige of the young science of comparative anatomy.[98] In his time, and applied to the forms occurring in the Paris Basin, it was a most valuable, ingenious, and yet obvious method, and even now is the principal rule the palæontologist follows in identifying fragments of fossils of any class. But it has its limitations, and it goes without saying that the more complete the fossil skeleton of a vertebrate, or the remains of an arthropod, the more complete will be our conception of the form of the extinct organism. It may be misleading in the numerous cases of convergence and of generalized forms which now abound in our palæontological collections. We can well understand how guarded one must be in working out the restorations of dinosaurs and fossil birds, of the Permian and Triassic theromorphs, and the Tertiary creodonts as compared with existing carnivora.
As the late O. C. Marsh[99] observed:
“We know to-day that unknown extinct animals cannot be restored from a single tooth or claw unless they are very similar to forms already known. Had Cuvier himself applied his methods to many forms from the early tertiary or older formations he would have failed. If, for instance, he had had before him the disconnected fragments of an eocene tillodont he would undoubtedly have referred a molar tooth to one of his pachyderms, an incisor tooth to a rodent, and a claw bone to a carnivore. The tooth of a Hesperornis would have given him no possible hint of the rest of the skeleton, nor its swimming feet the slightest clue to the ostrich-like sternum or skull. And yet the earnest belief in his own methods led Cuvier to some of his most important discoveries.”
Let us now examine from Cuvier’s own words in his Discours, not relying on the statements of his expositors or followers, just what he taught notwithstanding the clear utterances of his older colleague, Lamarck, whose views he set aside and either ignored or ridiculed.[100]