Indeed, Professor Owen himself remarks that “the neurapophyses are more constant as osseous or cartilaginous elements of the vertebræ than the centrums” (p. 97). Thus, then, it appears that the several elements included in the “ideal typical vertebra” have various degrees of constancy, and that no one of them is essential. There is no one part of a vertebra which invariably answers to its exemplar in the pattern-group. How does this fact consist with the hypothesis? If the Creator saw fit to make the vertebrate skeleton out of a series of segments, all formed on essentially the same model—if, for the maintenance of the type, one of these bony segments is in many cases formed out of a coalesced group of pieces, where, as Professor Owen argues, a single piece would have served as well or better; then we ought to find this typical repetition of parts uniformly manifested. Without any change of shape, it would obviously have been quite possible for every actual vertebra to have contained all the parts of the ideal one—rudimentally where they were not wanted. Even one of the terminal bones of a mammal’s tail might have been formed out of the nine autogenous pieces, united by suture but admitting of identification. As, however, there is no such uniform typical repetition of parts, it seems to us that to account for the typical repetition which does occur, by supposing the Creator to have fixed on a pattern-vertebra, is to ascribe to him the inconsistency of forming a plan and then abandoning it.
If, on the other hand, Professor Owen means that the “ideal typical vertebra” is a crystalline form in antagonism with “the idea or organizing principle;” then we might fairly expect to find it most clearly displaying its crystalline character, and its full complement of parts, in those places where the organizing principle may be presumed to have “subdued” it to the smallest extent. Yet in the Vertebrata generally, and even in Professor Owen’s Archetypus, the vertebræ of the tail, which must be considered as, if anything, less under the influence of the organizing principle than those of the trunk, do not manifest the ideal form more completely. On the contrary, as we approach the end of the tail, the successive segments not only lose their remaining typical elements, but become as uncrystalline-looking as can be conceived.
Supposing, however, that the assumption of suppressed or undeveloped elements be granted—supposing it to be consistent with the hypothesis of an “ideal typical vertebra,” that the constituent parts may severally be absent in greater or less number, sometimes leaving only a single bone to represent them all; may it not be that such parts as are present, show their respective typical natures by some constant character: say their mode of ossification?
To this question some parts of the Archetype and Homologies seem to reply, “Yes;” while others clearly answer, “No.” Criticising the opinions of Geoffrey St. Hilaire and Cuvier, who agreed in thinking that ossification from a separate centre was the test of a separate bone, and that thus there were as many elementary bones in the skeleton as there were centres of ossification, Professor Owen points out that, according to this test, the human femur, which is ossified from four centres, must be regarded as four bones; while the femur in birds and reptiles, which is ossified from a single centre, must be regarded as a single bone. Yet, on the other hand, he attaches weight to the fact that the skull of the human fœtus presents “the same ossific centres” as do those of the embryo kangaroo and the young bird. (Nature of Limbs, p. 40.) And at p. 104 of the Homologies, after giving a number of instances, he says—
“These and the like correspondences between the points of ossification of the human fœtal skeleton, and the separate bones of the adult skeletons of inferior animals, are pregnant with interest, and rank among the most striking illustrations of unity of plan in the vertebrate organization.”
It is true that on the following page he seeks to explain this seeming contradiction by distinguishing
“between those centres of ossification that have homological relations, and those that have teleological ones—i.e., between the separate points of ossification of a human bone which typify vertebral elements, often permanently distinct bones in the lower animals; and the separate points which, without such signification, facilitate the progress of osteogeny, and have for their obvious final cause the well-being of the growing animal.”
But if there are thus centres of ossification which have homological meanings, and others which have not, there arises the question—How are they always to be distinguished? Evidently independent ossification ceases to be a homological test, if there are independent ossifications that have nothing to do with the homologies. And this becomes the more evident when we learn that there are cases where neither a homological nor a teleological meaning can be given. Among various modes of ossification of the centrum, Professor Owen points out that “the body of the human atlas is sometimes ossified from two, rarely from three, distinct centres placed side by side” (p. 89); while at p. 87 he says:—“In osseous fishes I find that the centrum is usually ossified from six points.” It is clear that this mode of ossification has here no homological signification; and it would be difficult to give any teleological reason why the small centrum of a fish should have more centres of ossification than the large centrum of a mammal. The truth is, that as a criterion of the identity or individuality of a bone, mode of ossification is quite untrustworthy. Though, in his “ideal typical vertebra,” Professor Owen delineates and classifies as separate “autogenous” elements, those parts which are “usually developed from distinct and independent centres;” and though by doing so he erects this characteristic into some sort of criterion; yet his own facts show it to be no criterion. The parapophyses are classed among the autogenous elements; yet they are autogenous in fishes alone, and in these only in the trunk vertebræ, while in all air-breathing vertebrates they are, when present at all, exogenous. The neurapophyses, again, “lose their primitive individuality by various kinds and degrees of confluence:” in the tails of the higher Vertebrata they, in common with the neural spine, become exogenous. Nay, even the centrum may lose its autogenous character. Describing how, in some batrachians, “the ossification of the centrum is completed by an extension of bone from the bases of the neurapophyses, which effects also the coalescence of these with the centrum,” Professor Owen adds:—“In Pelobates fuscus and Pelobates cultripes, Müller found the entire centrum ossified from this source, without any independent points of ossification” (p. 88). That is to say, the centrum is in these cases an exogenous process of the neurapophyses. We see, then, that these so-called typical elements of vertebræ have no constant developmental character by which they can be identified. Not only are they undistinguishable by any specific test from other bones not included as vertebral elements; not only do they fail to show their typical characters by their constant presence; but, when present, they exhibit no persistent marks of individuality. The central element may be ossified from six, four, three, or two points; or it may have no separate point of ossification at all: and similarly with various of the peripheral elements. The whole group of bones forming the “ideal typical vertebra” may severally have their one or more ossific centres; or they may, as in a mammal’s tail, lose their individualities in a single bone ossified from one or two points.