The Thoracic Basket

The vertebral column, as already pointed out, has to do with the tubular dorsal nerve cord. The other one of the two essential tubes that characterize the vertebrate body, namely, the digestive tube, is encircled and protected at least partially by another part of the skeleton, the thoracic basket. This consists primarily of the thoracic vertebrae, the ribs, and the breastbone, with the pectoral girdle sometimes playing a supplementary role.

Since the digestive tube in most instances is many times longer than the body, it is coiled up in a compact mass that requires a “basket” for its skeletal protection instead of a straight narrow sheath such as is adequate for the nerve cord.

The thoracic basket first comes into its own with life on land in the class Reptilia, where a true breastbone and encircling ribs are present. This combination of bones and cartilages, which is hung on the anterior part of the skeletal axis, not only encloses a considerable portion of the digestive organs, but also furnishes protection to other soft viscera, notably the heart and lungs that are in great need of skeletal protection. In higher vertebrates still other organs, originally contained within the thoracic basket, have lost this protection, either by the contraction or degeneration of the structure itself, thus exposing the parts formerly protected, or by shifting their position. Liver, spleen, pancreas, and the small intestine, much of which is exposed, receive only partial protection from the thoracic basket, while the kidneys and gonads have migrated far posteriorly from their embryonic or ancestral position within the basket.

The backbone, thoracic basket, and girdles

The thoracic basket in man is marked off at either end by more or less transverse areas, while the sides are made up of skeletal elements embedded in muscular walls. The top of the basket is a somewhat restricted region, the margin of which is determined by the first thoracic vertebra, the first pair of ribs, and the upper end of the breastbone (Fig. 448). This enclosed space marks a narrow passage-way through which there are crowded side by side various structures, providing for traffic between the head and trunk (Fig. 449). There is first the trachea, connecting the imprisoned lungs with the outside world; next, the esophagus, that passes along food and drink, placing it safely beyond normal recall; the vagus nerves, wandering far from their headquarters in the brain to supply the distant viscera below; the large carotid arteries and jugular veins, which collect and distribute the blood of the head; and finally, a loop of the thoracic duct, that brings back into the confines of the venous system escaped white blood cells from their peregrinations throughout the tissues of the body.

Diagram of the anterior opening of the thoracic basket as seen from above

The floor of the basket, which in man is larger than the top area, is bounded behind by the last thoracic vertebra together with the short twelfth pair of floating ribs attached thereto. On the sides and in front the margins of the floor are determined by the tips of the tenth and eleventh pairs of ribs and by the cartilages of other posterior ribs that attach the latter to the breastbone, as well as by the lower end of the breastbone itself. The floor is also closed crosswise by the vaulted diaphragm which separates the cavities containing the lungs and heart above from the abdominal cavity below. This partition, or floor, of the thoracic basket tends to slope backward from the sternum towards the backbone, while the slant of the margin around the space marking the smaller top of the basket slopes upward from the sternum towards the backbone, since the sternal length is shorter than the vertebral length.

In general, therefore, the contour of the whole basket is somewhat cone-shaped with the smaller, upper end towards the head.

The space within the thoracic basket is partially divided into right and left spaces bulging dorsally like bay windows from the general cavity, since the upright column formed by the centra of the vertebrae, stacked one upon the other, stands out into the cavity, thus serving to a certain degree as a longitudinal partition. Within these lateral enlargements are packed the lungs. The heart lies midway between the lungs instead of far over on the left side as melodramatic actors are wont to indicate.

Cross section through the thorax of a human embryo

When compared in cross section the shape of the thorax of the human embryo is seen to be decidedly different from that of the adult (Fig. 450). In the embryo the dorso-ventral diameter exceeds that from side to side. In the human adult, on the contrary, where the visceral weight exerts a pull parallel to the backbone instead of at right angles to it, the greater diameter of the basket is no longer dorso-ventral, but from one side to the other. The general shape of the thorax can be modified by gymnastic and military exercises, by tight lacing, and by pathological conditions such as are induced, for example, by rickets or tuberculosis.

Quadrupeds in general resemble the human embryo with respect to the relative dimensions of the thoracic basket, which are correlated with the weight of the hanging viscera that exert a mechanical pull downward from the backbone towards the sternum. Reptiles whose bellies drag upon the ground, as well as such animals as sirenians and cetaceans whose viscera are supported in part by the surrounding medium of water, have a flattened thoracic basket like that of adult bipeds.


The ribs, which have been of particular human interest ever since the days of the Garden of Eden, are the most conspicuous part of the thoracic basket. They vary in length from mere immovable tips attached to the transverse processes of the vertebrae to hooplike bands of bone entirely encircling the body. Primitively and embryologically there is a pair of ribs for every vertebra from axis to tail. Some fishes have two pairs to a single vertebra. The general tendency, however, both ontogenetically and phylogenetically, is towards a reduction in number at either end of the series.

It is known that the ribs of all vertebrates are not homologous structures, since some of them arise in the sheet of connective tissue that separates the dorsal from the ventral metameric muscles of the body wall, while in other instances they develop between the peritoneal lining and these muscles.

A typical human rib is a somewhat flattened bone so bent and twisted that it cannot be made to lie flat without rocking back and forth when placed on a table (Fig. 451). Its slender and somewhat elastic main body is pieced out at the sternal end in mammals with a flexible cartilage, not shown in the figure. At the dorsal end, which joins the thoracic vertebra in two places, are two prominences known as the head and tubercle respectively, separated from each other by the neck which is slightly narrower than the body, or the main part of the rib. Both head and tubercle bear articular facets by means of which the rib plays upon its vertebral support, maintaining its attachment with a minimum of muscular equipment.

The fifth right bony rib of man, as seen from below. The thoracic basket

Along the inner margin of each rib, for a part of its length at least, is a shallow groove, the sulcus, within the protection of which a vein, or artery, and a nerve lie parallel to each other in harmonious safety.

There are normally twelve pairs of ribs in man (Fig. 452), although much evidence from comparative anatomy, as well as from embryology, supports the conclusion that the ancestors of modern man had more. That reduction in the length of ribs is going on is indicated by their obvious degeneration in the posterior region of the thoracic basket, where at least two pairs fail to reach the sternum.

To understand the nature of this degeneration it should be stated that all the ribs articulate at one end with thoracic vertebrae, while at the other end only the first seven pairs, or true ribs, join the sternum. The remaining five pairs are known as false ribs. Of these the eighth, ninth, and tenth secure anchorage to the sternum indirectly by means of the cartilages of the seventh pair of true ribs. Ordinarily the eleventh and twelfth pairs, which are called floating ribs, have so far degenerated that they only partially encircle the body, thus failing to make even a vicarious attachment to the sternum.

The ribs of man, therefore, increase in length from the first to the seventh or eighth pairs and then successively decrease to the twelfth pair, which may be reduced to mere stubs hardly more than an inch long.

Cervical ribs on the seventh cervical vertebra of a human adult individual. Reduction of the first pair ribs

In medical literature numerous cases are cited of extra human ribs, either at the cervical or at the lumbar ends of the thoracic series, persisting in adult life. Instances are given of a pair of ribs attached to the seventh cervical vertebra, that completely encircled the upper area of the thoracic basket, joining the sternum quite after the manner of true ribs. Persistent cervical ribs, however, are more frequently incomplete, failing of direct sternal attachment (Fig. 453). On the other hand, the first pair of thoracic ribs may sometimes be incomplete (Fig. 454), resulting in a subnormal number. Additional ribs are of more frequent occurrence on the first lumbar vertebra than in the cervical region. Extra lumbar ribs when they appear are called “gorilla ribs” because they represent the normal condition in gorillas and chimpanzees. The closely related gibbons among the apes normally have fourteen pairs, while the orang-utans, like their human kin, have only twelve. The anatomist Rabl in a report upon 640 human bodies examined in the dissecting rooms of the University of Prague says that he found forty of them, or a little more than six per cent, with a gorilla rib on at least one side. Two out of the 640 had only eleven pairs of ribs.

Gorilla ribs are about three times more frequent in the human male than in the female, a fact difficult to harmonize with Adam’s historic loss. Albertus Magnus (1193-1280), who did not always base his anatomical conclusions upon direct observation, discusses very learnedly the supposed discrepancy in the number of ribs in the two sexes.

With respect to length, the eleventh pair of floating ribs varies from six to eleven inches, while the twelfth pair, ranging from mere stubs less than an inch long to approximately a foot in length, may nearly encircle the body.

Posterior part of the vertebral column, including three thoracic vertebrae and their ribs

In fetal life ribs are not only present upon the seventh cervical vertebra but also upon all the lumbar vertebrae (Fig. 455). Moreover, rudiments of ribs, which afterwards fuse with the centra to form the lateral masses of the sacrum, are attached to the sacral vertebrae.

Ribless vertebrates include amphioxus, the cyclostomes, holocephalans, skates, and such lophobranchs as Syngnathus and Hippocampus among the teleost fishes. In the ganoid Polyodon, the ribs are quite vestigial.

Among other fishes there are two sorts of ribs of diverse origin, namely, ventral and dorsal ribs, both of which are connected at one end with the vertebral column, although entirely unattached at the other end in the absence of a sternum (Fig. 456).

Diagrams of the three patterns of rib arrangement

Ventral ribs simply represent vertebral haemal arches, anterior to the anus, that have spread apart enough to incompletely encircle the body cavity (Fig. 456c). They are therefore sometimes known as haemal ribs. They lie entirely inside the muscles of the body wall, next to the peritoneal lining of the body cavity at the points where the myocommata join this layer. The ribs of the dipnoi, as well as of most teleosts and ganoids, are of this character.

Dorsal ribs, on the other hand, grow out from the transverse processes of the vertebrae at the points where the myocommata cross the horizontal septum. They therefore extend between the dorsal and ventral groups of muscles that form the body wall (Fig. 456a). Most elasmobranchs, together with all vertebrates above fishes, have dorsal ribs. In elasmobranchs and modem amphibians these are short and insignificant; but, beginning with reptiles, they become longer, in many instances encircling the body cavity and joining at the ventral ends with the breastbone to form a complete thoracic basket.

There are a few fishes, notably the ganoid Polypterus, and certain clupeoid and salmonoid teleosts, which have both dorsal and ventral ribs, one outside the other (Fig. 456b) , making two pairs of ribs to each vertebra.

Amphibian ribs like those of all tetrapods are of the dorsal type, although they are never more than mere stubs. In urodeles they are present even on anterior caudal vertebrae which also have haemal arches and therefore are associated with the homologues of both ventral and dorsal ribs. In no modern amphibian do the ribs encircle the body cavity, but in the fossil stegocephals strong well-developed ribs occurred. The reduced ribs of modern amphibians are therefore to be regarded as vestigial rather than as primitive structures.

A vertebra with a pair of ribs of a young salamander

The short ribs of urodeles are forked at their proximal ends like a letter Y, thus having two points of attachment to the vertebra (Fig. 457), in that way forming a passage-way between the fork and the vertebra, called the vertebrarterial canal. A similar canal persists in mammals and even in man, where one branch of the fork becomes the “head” and the other the “tubercle” of the rib (Figs. 451 and 458), thus preventing the large vertebral artery that lies therein from being disturbed when the neck is stretched or twisted about.

Diagram to show articulation of rib with both centrum and transverse process of vertebra

The ribs of the Anurans are not forked, and become small immovable tips attached to the transverse processes of the vertebrae.

Among reptiles the ribs found in lizards and crocodiles are most typical, becoming differentiated into a dorsal bony vertebral division, homologous to the rib of the urodele and a ventral cartilaginous sternal division that attaches to the sternum, which is something entirely new in costal devices. Some lizards and crocodiles even have a third bony segment intercalated between these two parts.

As there is no sternum in snakes all the ribs are “floating,” which makes it possible for them to aid materially in locomotion, because the snake grips the ground with its ventral scales and with the ends of its movable floating ribs.

Turtles do not have cartilaginous sternal elements in their ribs. Instead, the bony vertebral parts, which flatten out and join together, are overlaid by dermal costal plates to form part of the dorsal shell, or carapace.

In Sphenodon, that “living fossil” of New Zealand, there are several unusual ribs in the tail region, which suggests an ancestral prodigality of ribs not characteristic of more modern vertebrates.

As might be expected, birds present extreme modifications in their ribs. Both the sternal and vertebral parts are entirely ossified, making a firm but withal expansible thoracic basket for the attachment of the powerful muscles of flight. This necessary firmness or solidarity is further enhanced by the fact that most of the ribs are “true,” that is, connected with the sternum, while in the sacral region the ribs fuse solidly with the vertebrae to form the large strong characteristic “backpiece,” or synsacral complex, that is so effective in the support of the body weight. The ribs of birds are thin and flat, affording ample surface for muscle attachment, and in addition they are usually provided with light flat supplementary uncinate processes (Fig. 459), which serve to splice the ribs of the thoracic basket together into a firm resistant unit for the attachment of flight muscles.

A part of the skeleton of a goose, showing uncinate processes

In Archaeopteryx, the oldest known fossil bird, the ribs were rounded instead of flat, like those of lizards, and lacked uncinate processes, although Sphenodon, crocodiles, and some fossil reptiles, such as Eryops, show these supplementary costal inventions.

In mammals the sternal part of the ribs remains cartilaginous in adult life, thus allowing greater freedom in respiratory movements. The total number of ribs in mammals varies from nine pairs in the bottle-nosed whale, Hyperoodon, to twenty-four in the primitive two-toed sloth, Choloepus, of South America, which finds an elongated thoracic basket useful since it spends most of its time suspended upside down from the limbs of trees.

True ribs, that reach the sternum directly by means of their own cartilage segments, vary from two pairs in the sea-cow Trichechus, to ten pairs in the agile spider monkey Ateles.

The Sternum, or Breastbone

As contrasted with the backbone the frontbone, or sternum, is the terrestrial part of the thoracic basket, that is it appears first in the evolutionary history of vertebrates that locomote on land.

The need of such a strengthening structure to knit together the whole thoracic basket into a firm skeletal unit on which the muscles of the anterior legs or arms may find suitable anchorage, is not apparent for the more primitive water-dwellers like fishes that go forward by lateral tail motion rather than by the leverage of bilateral appendages.

Not only fishes are devoid of any kind of sternal apparatus, but also some of the lowest urodeles, for instance, Proteus and Amphiuma, as well as the footless caecilians, snakes, and turtles.

Ventral view of Necturus showing sternal cartilages, extending into myocommata

The sternum develops in the ventral median septum of the anterior trunk or thoracic region of tetrapods. In amniotes it forms the ventral element which, with ribs and vertebrae, completes the thorax. The earliest structure recognized as a sternum is found in urodeles. In Necturus (Fig. 460) it is a rather insignificant appearing, irregular mass of cartilage extending over several segments in the ventral septum and with branches running into adjacent myosepta. Here it seems to arise independent of other parts of the skeleton such as ribs and pectoral girdle. In other urodeles and lower anurans it is a more compact plate of cartilage near the posterior part of the girdle (Fig. 461).

Pectoral girdle and sternum of Amblystoma, ventral view

In Rana and other higher anurans the sternum consists of several midventral elements unpaired in the adult (Fig. 462). In the mid-line between the two halves of the pectoral girdle is an epicoracoid cartilage, ordinarily considered a part of the girdle but also actually in line with the structures usually recognized as belonging to the sternum. At the anterior end of the epicoracoid is an omosternum the basal part of which is ossified. A similar cartilaginous and bony xiphisternum is connected to the posterior end of the epicoracoid. It is debatable whether or not this posterior part corresponds to the cartilaginous plate of lower Amphibia, although many authors call its bony part the archisternum. These parts all arise as paired structures which subsequently fuse.

Sternum and pectoral girdle of a frog, Rana, and Iguana ventral view

In Lacertilia (Fig. 463) and Grocodilia (Fig. 464) there is an oblong, cartilaginous plate continuous with which are usually two hornlike, posterior processes. To the anterior end of the sternum the pectoral girdle is attached. Ribs connect with the sides and with the posterior horns. These animals may have been the first to have the ribs attached to the sternum, although some palaeontologists have suggested the possibility that the ribs of Stegocephalia reached such a midventral structure. Snakes and turtles are without sterna.

Sternum, gastralia and parts of the girdles of the alligator

Birds have a large thin sternum of replacing bone that affords attachment for the pectoral muscles of flight. It is spread out flat in running birds (ratites), but in all flying birds (carinates), a carina, or keel, projects at right angles to the breastbone proper, furnishing considerable additional surface for muscle attachment. The pectoral girdle and bony sternal divisions of the ribs attach to the sternum. A comparison of the keels on the breastbones of an albatross, a pigeon, and a hummingbird (drawn to the same scale in Figure 465) shows how large a relative surface for muscle attachment is necessary to enable the tiny hummingbird to maintain its marvelous vibrating wing movements, which are so rapid that the wings themselves, like the spokes of a swiftly rotating wheel, blend out of sight.

Comparsion of the keel of the albatross

A carina is also a characteristic of extinct flying reptiles (pterodactyls) as well as of flying bats, and in burrowing moles among mammals, although in this latter case it is probably not homologous with the carina of birds.

In mammals the sternum typically consists of a series of six to nine sternebrae, composed of replacing bone, followed by a small cartilaginous process (Fig. 466). To the anterior most of these elements, called the manubrium, is attached the first pair of ribs. The other pairs of true ribs attach at the junctions between sternebrae. The posterior part of the sternum is the xiphoid cartilage. In Sirenia and some Cetacea the sternum is a single bony plate.

Sternum and rib cartilages of a cat, showing separate sternebrae

In the human adult the sternum consists of three parts, namely, the manubrium, the gladiolus formed by the fusion of four sternebrae, and the xiphoidor ensiform, cartilage (Fig. 467). Of these parts the first two are formed of bone, and the last, as its name implies, of cartilage. Its superficial resemblance to a Roman sword explains why the names “manubrium,” or hand-grip, and “gladiolus” and “ensiform,” both of which signify a sword, were applied to its component parts by the early anatomists. The clavicles and the first seven pairs of ribs are attached to the sternum.

The adult human sternum

The shortening of the human sternum, which results in the conspicuous notch on the front side of the thoracic basket, follows the fusion of the sternebrae and the consequent disappearance of the intersternebral cartilages. It goes further in the female than in the male, thus allowing a relatively larger unhampered space between the sternum and the pelvis for the accommodation of a possible fetus during pregnancy.

It is now generally agreed that in mammalian embryos the earliest recognizable evidence of the sternum is a pair of ventral, longitudinal procartilaginous bars in front of which lies a single, oval mass of procartilage (Fig. 468). Each bar soon fuses with the oval mass and is in turn joined by the ventral ends of the true ribs of its side of the body. Beginning at their anterior ends the two bars gradually fuse to form a single, median cartilaginous rod. Then centers of ossification appear, usually a single one for the manubrium and a pair for each of the other sternebrae. Perhaps the manubrium is derived from the originally oval, unpaired, anterior element. The mammalian sternum, therefore, arises independent of the ventral ends of the ribs but near these structures. In reptiles and birds the sternum also arises from two longitudinal bars which, even in the earliest observed procartilaginous stage, are continuous with the ribs. The sternal elements of modern amphibians are associated with the median part of the pectoral girdle but never with ribs. There has been considerable conjecture as to whether or not the extinct primitive Amphibia (Stegocephalia) had long cartilaginous ventral continuations of their ribs which reached the mid-ventral region to join a cartilaginous sternum. Unfortunately cartilaginous elements do not lend themselves to fossilization.

Development of human sternum

Three main theories have developed concerning the origin and homologies of the sternum. According to the Girdle Theory a mid-ventral part of the pectoral girdle has separated from the rest of that structure, elongated, and become the sternum. Neither embryology nor comparative anatomy can offer much to support this concept. The Rib Theory derives the sternum from the ventral ends of the ribs. As noted above, there is some support for this from studies of the development of the amniotes. The fact that the origin of the mammalian sternum can be traced back to procartilaginous structures independent of the ribs casts a shadow of doubt on this theory. Also the situation in the Amphibia offers something of a problem. Some have tried to meet this difficulty by suggesting that primitive amphibians had long ribs which reached the ventral side and gave rise to the sternum, although there is no really good evidence of this from paleontology. Furthermore there can be no doubt that the amphibian sternum arises in ontogeny completely independent of definitive ribs. Some advocates of this theory apply it only to amniotes and consider the amphibian sternum derived from the girdle region. According to this concept the amniote sternum (neosternum) is a new (neo, new) structure not homologous with the primitive one (archisternum) of amphibians.

The third, and possibly the most plausible theory, assigns an Independent Origin to the sternum. With the development of legs and the migration onto land, new stresses develop and the need arises for both additional protection of the ventral body wall and greater surface for the attachment of appendicular muscles. Under these conditions the sternum appears as a modification of the general connective tissue of the mid-ventral region. Both in amphibians and in mammals it has been observed to develop independent of both ribs and girdle. Careful study using new techniques may enable us to recognize sternal elements of reptiles and birds in earlier stages than has been possible to date. All we know now is that the ribs and sternum of these animals are in continuity with one another when first detectable. There is no evidence that either ribs give rise to sternum or sternum to ribs. According to this theory of the independent origin of sternal elements, their connection with the pectoral girdle or ribs or both should be considered as secondary and due to proximity rather than the derivation of one part from the other.