Form and Function: A Contribution to the History of Animal Morphology

by E. S. Russell

There was a close spiritual affinity between the speculative evolutionists and the transcendentalists. Both showed the same subconscious craving for simplicist conceptions—the transcendentalists clung fast to the notion of the absolute unity of type, of the ideal existence of the "one animal," and the evolutionists did precisely the same thing when they blindly and instinctively accepted the doctrine of the monophyletic descent of all animals from one primeval form. Geoffroy persisted in regarding Arthropods as being built on the same plan as Vertebrates: Dohrn and Semper did nothing different when they derived both groups from an ancestor combining the main characters of both. The determination to link together all the main phyla of the animal kingdom and to force them all into a single mould was common to evolutionary and pre-evolutionary transcendentalists alike.

From the fact that all Metazoa develop from an ovum which is a simple cell, the evolutionists inferred that all must have arisen from one primordial cell. From the fact that the next step in development is the segmentation of the ovum, they argued that the ancestral Metazoa came into being through the division of the primal Protozoon with aggregation of the division-products. From the fact that a gastrula stage is very commonly formed when segmentation has been completed, they assumed that all germ-layered animals were descended from an ancestral Gastræa.

They quite ignored the possibility that a different explanation of the facts might be given; they seized upon the simplest and most obvious solution because it satisfied their overwhelming desire for simplification. But is the simplest explanation always the truest—especially when dealing with living things? One may be permitted to doubt it. It is easy to account for the structural resemblance of the members of a classificatory group, by the assumption that they are all descended from a common ancestral form; it is easy to postulate any number of hypothetical generalised types; but in the absence of positive evidence, such simplicist explanations must always remain doubtful. The evolutionists, however, had no such scruples.

Phylogenetic method differed in no way from transcendental—except perhaps that it had learnt from von Baer and from Darwin to give more weight to embryology. The criticisms passed by Cuvier and von Baer upon the transcendentalists and their recapitulation theory might with equal justice be applied to the phylogenetic speculations which were based on the biogenetic law. There was the same tendency to fix upon isolated points of resemblance and disregard the rest of the organisation. Thus, on the ground of a presumed analogy of certain structures to the vertebrate notochord, several invertebrate groups, as the Enteropneusta, the Rhabdopleura, the Nemertea, were supposed to be, if not ancestral, at least offshoots from the direct line of vertebrate descent. And if other points of resemblance could in some of these cases be discovered, yet no successful attempt was made to show that the total organisation of any of these forms corresponded with that of the Vertebrate type. With the possible exception of the Ascidian theory, all the numerous theories of vertebrate descent suffered from this irremediable defect, and none carried complete conviction.

In spite of the efforts of the evolutionists, as of those of the transcendentalists, the phyla or "types" remained distinct, or at best connected by the most general of bonds.

The close affinity of transcendentalists and evolutionists is shown very clearly in their common contrast in habits of thought with the Cuvierian school. It is the cardinal principle of pure morphology that function must be excluded from consideration. This is a necessary and unavoidable simplification which must be carried out if there is to be a science of pure form at all. But this limitation of outlook, if carried over from morphology to general biology becomes harmful, since it wilfully ignores one whole side of life—and that the most important. The functional point of view is clearly indispensable for any general understanding of living things, and this is where the Cuvierian school has the advantage over the transcendental—its principles are applicable to biology in general.

Geoffroy and Cuvier in pre-evolutionary times well typified the contrast between the formal and the functional standpoints. For Geoffroy form determined function, while for Cuvier function determined form. Geoffroy held that Nature formed nothing new, but adapted existing "materials of organisation" to meet new needs. Cuvier, on the other hand, was always ready to admit Nature's power to form entirely new organs in response to new functional requirements.

The evolutionists followed Geoffroy rather than Cuvier. They laid great store by homological resemblances, and dismissed analogies of structure as of little interest. They were singularly unwilling to admit the existence of convergence or of parallel evolution, and they held very firmly the distinctively Geoffroyan view that Nature is so limited by the unity of composition that she can and does form no new organs.

By no one has this underlying principle of evolutionary morphology been more explicitly recognised than by Hubrecht, who in his paper of 1887, after summarising the points of resemblance between Nemertines and Vertebrates which led him to assume a genetic connection between them, writes as follows:—"At the base of all the speculations contained in this chapter lies the conviction, so strongly insisted upon by Darwin, that new combinations or organs do not appear by the action of natural selection unless others have preceded, from which they are gradually derived by a slow change and differentiation.

"That a notochord should develop out of the archenteric wall because a supporting axis would be beneficial to the animal may be a teleological assumption, but it is at the same time an evolutional heresy. It would never be fruitful to try to connect the different variations offered, e.g., by the nervous system throughout the animal kingdom, if similar assumptions were admitted, for there would be then quite as much to say for a repeated and independent origin of central nervous systems out of indifferent epiblast just as required in each special case. These would be steps that might bring us back a good way towards the doctrine of independent creations. The remembrance of Darwin's, Huxley's, and Gegenbaur's classical foundations, and of Balfour's and Weismann's brilliant superstructures, ought to warn us away from these dangerous regions" (p. 644).