Art. XVI.—On the Origin of the Sternum.
[Read before the Otago Institute, 14th October, 1890.]
According to Wiedersheim, one of the leading modern authorities on vertebrate morphology, nothing is known of the phylogeny of the sternum, in spite of the fact that its development in many of the more important types has been worked out in detail.
Two kinds of sternum are distinguished by morphologists—the costal sternum, formed by the concrescence of the ventral ends of ribs, and found in Amniota (birds, reptiles, and mammals); and the omosternum (coracoid sternum, or clavicular sternum), formed by segmentation of the shoulder-girdle, and highly characteristic of Amphibia. No trace of a sternum has hitherto been described in fishes.
In all Amniota in which a sternum occurs, its development takes place in the same way. The ribs of each side gradually grow downwards, enclosing the chests. As they do
so, those of each side unite by their ventral ends, forming, a longitudinal band of cartilage (Pl. XIX., fig. 5, st.); the two bands thus constituted then approach one another and unite in the middle ventral line (fig. 6), forming a median structure; areas of connective tissue appear separating it from the ventral ends of the ribs, and thus the sternum is formed as a purely costal product.
In the tailed Amphibia the sternum has a double origin A pair of cartilaginous bands appear in the inscriptiones tendineæ of the mm. recti abdominis, and a pair of narrow strips are separated off from the posterior borders of the coracoids. From these four elements the sternum is produced. Ruge considers that the first-named chondrites* are to be looked upon as vestigial ribs; the others are obviously parts of the shoulder-girdle.
In the tailless Amphibia, such as the frog, the sternum has a similar origin, but in many genera there are formed in addition paired cartilages in front (cephalad) of the shoulder-girdle, and arising by the detachment of narrow bands from the procoracoids. These unite in the middle line, and form a median cartilage (fig. 3, pr. om. st.), which was distinguished by my father as the omosternum, the name sternum, being restricted to the post-coracoidean element (pt. om. st.). As, however, both structures are formed from the shoulder-girdle, Albrecht's suggestion to call the anterior cartilage the pre-omosternum, and the posterior the post-omosternum, is worthy of adoption, especially as these names suggest homologies with the similarly-formed median elements of the pelvis, the pre-pelvisternum (epipubis) and post-pelvisternum (os cloacæ).
In order to form any clear notion of the origin of this element of the skeleton, a sternum in some form lower than the Amphibia is obviously a desideratum. I believe I have discovered such a structure in the Perlon Shark (Notidanus indicus).†
The shoulder-girdle of Elasmobranchs is an inverted arch of cartilage attached to the vertebral column either by muscles only (Selachians), or by fibrous union (Rays), and having the pectoral fins attached, one on either side, to its posterior
[Footnote] * Chondrite = an independent cartilaginous element or centre of chondrification.
[Footnote] † Since writing this paper I find that the structure called in the ensuing description the pre-omosternum was discovered in 1884 by Haswell, who says, “The shoulder-girdle [of Notidanus indicus] is remarkable for the presence in the middle ventral line of a distinct foursided lozenge-shaped cartilage, let into the arch, as it were, in front… The intercepted cartilage is temptingly like a pre-sternal, but the absence of such an element in the skeleton of any group nearer than the Amphibia seems to preclude this explanation” (Proc. Linn. Soc. N.S.W., vol. ix.).—March, 1891.
border. The portion of the continuous cartilage lying above (dorsad of) the articulation of the fin is the scapular region; that below (ventrad of) it the coracoid region. In the embryo the shoulder-girdle consists of distinct paired cartilages, which afterwards unite with one another in the middle ventral line by concrescence of their coracoid portions.
In a skeleton of Notidanus indicus, prepared a few months ago for the Otago University Museum, the middle region of the shoulder-girdle (figs. 1 and 2) is produced in front into a blunt process, while it is evenly curved posteriorly. Two curved areas of fibrous tissue, with their convexities towards the median plane, extend from the anterior to the posterior border, touching one another in the centre, and thus bounding two distinct cartilaginous areas—an anterior (pr. om. st.) of a rhomboid, a posterior (pt. om. st.) of a triangular form. The two cartilages are particularly well seen when the shoulder-girdle, which has been prepared by impregnation with car-bolized glycerine jelly, is held up to the light.
I think there can be no doubt that the anterior rhomboid cartilage (pr. om. st.) is to be considered as a pre-omosternum, the posterior triangular piece (pt. om. st.) as a post-omosternum, the coracoid portion of the shoulder-girdle being related to the two median elements in much the same way as the coracoids of Anura to the sternal cartilages (compare figs. 2 and 3).
It may be objected that the cartilages in question are unpaired in Notidanus, while in Amphibia they arise from the union of paired chondrites. But, in the first place, we know nothing of the development of the shoulder-girdle in Notidanus, and, in the second place, an actual unpaired origin would only mean that the sternal elements were detached after the union of the coracoids with one another, each of them being morphologically paired since each is derived in an equal degree from the two originally separate halves of the pectoral arch.
It will be seen that the omosternum of Notidanus is related to the shoulder-girdle in much the same way as the copulæ (basi-hyal; basi-branchials) to the visceral arches.
The question then arises as to whether there is any genetic connection between the omosternum of Notidanus and Amphibia and the costal sternum of Amniota. I am disposed to think that the latter is derivable from the former, its present mode of origin being a case of retarded development.
In the early Amniota we may suppose there to have been a post-omosternum, developed as in Amphibia, but separated from the coracoids before chondrification, and joined at a late period of development by the first pair of ribs. In a subsequent stage of evolution we should have the second and following ribs becoming successively united in the same manner.
A simultaneous retardation in the chondrification of the sternum would result in the development of a sternum formed of indifferent tissue and subsequently chondrified from the ribs, and from this condition of things it is but a step to the earliest stage in existing Amniota, in which the first indication of the breastbone consists of paired patches of cartilage formed by the union of the anterior thoracic ribs.
Any facts tending to show that any portion of the sternum originates independently of ribs will support this view, and in this connection Goette's observations on the development of lizards are significant. In an early stage of Cnemidophorus each half of the sternum is a triangular patch of tissue (fig. 4, st.) extending beyond the level of the third thoracic rib, but in connection only with the first (Th. Rb. 1): in other words, that part of the sternum which corresponds with the second and third ribs is formed independently of them, and as a backward growth from the anterior portion.
My own observations on Apteryx tell in the same direction. In the earliest stage in which the sternum is present it extends backwards to the level of the third thoracic rib; the first two ribs are united to it by joints, the third is loosely attached by connective tissue. In the next stage, the first three ribs are attached by joints, and the fourth by fibrous tissue. That is, as it appears to me, the portion of the sternum corresponding to the third and fourth ribs is formed by a backward growth of the anterior region and quite independently of the last two ribs, the union of which with it is a secondary process.
I am disposed to consider the stages in the phylogeny of the sternum to have been somewhat as follows:—
|1.||Segmentation of anterior and posterior elements (pre- and post-omosterna) from ventral ends of coracoids: there is no evidence to show whether these were originally paired or unpaired, but the former seems more likely.||Notidanus.|
|2.||Pre-and post-omosterna arise each from paired elements, which after wards unite, segmented from the coracoids: the pre-omosternum may be absent, and the post-omosternum may be formed in part from chondrites (? vestigial ribs) formed in the inscriptiones stendineæ (Urodela).||Amphibia.|
|3.||Disappearance of pre-omosternum: late union of first pair of thoracic ribs with post-omosternum.||Hypothetical.|
|4.||Post-omosternum separated from coracoids while still in the state of indifferent tissue, and chondrification retarded to a later period than that of the ribs.||Hypothetical.|
|5.||Sternum (post-omosternum) unites very early with the first pair of ribs, and chondrification extends into it from them; after chondrification it grows backwards, the remaining sternal ribs uniting successively with it.||Cnemidophorus (? Apteryx)|
|6.||Development of sternum further retarded until it first appears in the form of paired longitudinal bars, formed by the concrescence of the thoracic ribs.||Mammalia.|
Description Of Plate XIX.
Fig. 1. Shoulder-girdle of Notidanus indicus, viewed obliquely from the left side; one-half natural size: Cor., coracoid; scap., scapula; pr. om. st., pre-omosternum; pt. om. st., post-omosternum.
Fig. 2. Mid-ventral portion of the same, from beneath; natural size: Letters as before.
Fig. 3. Part of shoulder-girdle and sternum of a young frog (after W.K. Parker): Pr. Cor., pro-coracoid; cl., clavicle; the other letters as before.
Fig. 4. Early stage in the development of the shoulder-girdle and sternum of a lizard (Cnemidophorus) (after Goette): Th. Rb. 1., first thoracic rib; other letters as before.
Fig. 5. Early stage in the development of the sternum of man (after Ruge): Letters as before.
Fig. 6. Later stage of the same (after Ruge): x. pr., xiphoid process; other letters as before.