Paper - A contribution to the embryology of the fore-limb

From Embryology
Embryology - 18 May 2024    Facebook link Pinterest link Twitter link  Expand to Translate  
Google Translate - select your language from the list shown below (this will open a new external page)

العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Rutherford NC. A contribution to the embryology of the fore-limb. (1914) J Anat Physiol. 48: 355-377.

Online Editor  
Mark Hill.jpg
This 1914 historic paper describes upper limb development in the embryo. See links below for the modern notes.

Modern Notes:

Musculoskeletal Links: Introduction | mesoderm | somitogenesis | limb | cartilage | bone | bone timeline | bone marrow | shoulder | pelvis | axial skeleton | skull | joint | skeletal muscle | muscle timeline | tendon | diaphragm | Lecture - Musculoskeletal | Lecture Movie | musculoskeletal abnormalities | limb abnormalities | developmental hip dysplasia | cartilage histology | bone histology | Skeletal Muscle Histology | Category:Musculoskeletal
Historic Embryology - Musculoskeletal  
1853 Bone | 1885 Sphenoid | 1902 - Pubo-femoral Region | Spinal Column and Back | Body Segmentation | Cranium | Body Wall, Ribs, and Sternum | Limbs | 1901 - Limbs | 1902 - Arm Development | 1906 Human Embryo Ossification | 1906 Lower limb Nerves and Muscle | 1907 - Muscular System | Skeleton and Limbs | 1908 Vertebra | 1908 Cervical Vertebra | 1909 Mandible | 1910 - Skeleton and Connective Tissues | Muscular System | Coelom and Diaphragm | 1913 Clavicle | 1920 Clavicle | 1921 - External body form | Connective tissues and skeletal | Muscular | Diaphragm | 1929 Rat Somite | 1932 Pelvis | 1940 Synovial Joints | 1943 Human Embryonic, Fetal and Circumnatal Skeleton | 1947 Joints | 1949 Cartilage and Bone | 1957 Chondrification Hands and Feet | 1968 Knee
Historic Disclaimer - information about historic embryology pages 
Mark Hill.jpg
Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

A Contribution to the Embryology of the Fore-Limb Skeleton

By N. C. Rutherford, M.B., F.R.C.S.


THE concise account of the development of the human skeleton recently published in the Keibel and Mall Text-book of Human Embryology forms the standpoint of present knowledge of and opinion on the subject. Regarded from this standpoint, certain observations of some importance resulting from a study of the development of the shoulder girdle and forelimb skeleton in man, recently undertaken by the writer, seem worthy of publication as supplementing the account of the cardinal features in the growth of these cartilages and giving an account of the early phenomena of the ossification of the scapula. In considering the observations made on the development of the skeletal arch which overhangs and strengthens the shoulder-joint, its origin from one of a smaller perimeter will be the chief subject of discussion. The embryonic material employed consists of human embryos and foetuses of from 10 to 30 mm. in greatest length, reconstructions of the fore-quarter skeletons being made in two cases, one a 20-millimetre and the other a 30-millimetre specimen. The specimens cited in comparison are in the Museum of the Royal College of Surgeons. As it is on the reconstructions that the initial Observations were made and round them that the whole work centres, a short description of them forms the first part of this paper.

Technical Note. — In making the drawings of the cartilaginous skeleton, care was taken to indicate exactly the margins of the fully differentiated cartilage as contrasted with the transitional perichondrial layers in all cases except that of the clavicle in which the easily followed enveloping layer was shown.


Model of the Fore-quarter Skeleton of a, 20 mm Human Embryo (figs. 1 and 2).

This model includes :— Twelve vertebrae of which the upper two are cervical, nine complete and two incomplete ribs, one lateral half of the sternum, all the elements of the limb and girdle skeleton.

The vertebrae are represented by their bodies, and continuous with these are the neural arches of the right side; cartilaginous continuity around the notochord exists between the bodies. They form a markedly arched series, the dorsal convexity being so great as to bring the two upper ones each rather ventral to the succeeding member of the series than on top of it; these two cervical vertebrae have procoelous bodies of approximately tetrahedral form, apex directed caudo-ventrally. The neural arches, imbricated above and in linear series below, show small lateral projections at their dorsal extremities representing demispines; the seventh cervical vertebra shows a separate chondral element, its costal process, which is in close articulation with the body.

Rutherford1914 fig01.jpg

Fig. 1. Wax-plate reconstruction from human embryo 20 mm. greatest length. Magnification in figure about x 8.5.

The disproportionate length of the ribs and the approximately horizontal ‘position each occupies give to the whole thoracic skeleton an appearance of marked inspiration. In general the cross-section of each rib is circular, but in front there is dorso-ventral flattening, whilst behind the flattening is from above downwards; the anterior extremities of ribs I. to V. articulate with a delicate sternal bar of cartilage which merges above partly in the cellular nonchondral interclavicular mass; VI. and VII. meet and support each other caudal to the other extremity of this bar ; all turn slightly dorsally, and I. to V. then acutely caudally in contact with the sternal bar. Like VI. and VII., VIII. IX. X. and XI. bend forward and so aredisposed as in the adult.

The clavicle is represented in its inner four-fifths by a very stout bar prolonged as to its outerfifth in the form of a flattened plate by bevelling at the expense of its costal aspect. The curve of the lateral part of the adult bone is practically absent. The cellular, constitution can be well seen in fig. 6; it is made up of two ossific centres on the point of coalescence within a matrix of young cartilage cells, the whole being surrounded by a very dense cellular periosteum; this last is prolonged into (1) a cellular mass surrounding the tip of the acromion, (2) a similar mass at the opposite end, the inter-clavicle, and (3) a tract or band connecting behind the coracoid with the pre-scapula in which there is an appearance of early chondrification. The wedge of cartilage described by Professor Fawcett and assumed by him to be associated with the formation of the deltoid tubercle is well shown in fig. 6. In this figure the deltoid is not associated with it.

File:Rutherford1914 fig02.jpg

Fig. 2. Wax-plate reconstruction from human embryo 20 mm. Schematic representation of scapula and limb skeleton. Magnification in figure x 7'14.

The scapula is as yet incompletely merged with the coracoid, as indicated by the Width of the scapular notch and by a line of division “across the glenoid cavity rather below its middle. The blade is irregularly diamond-shaped; its upper limit lies opposite the caudal extremity of the body of the sixth cervical vertebra, and its surfaces are parasagittal.

The coracoid is a bulbous mass of cartilage joined with the scapula by a narrow neck and having a short cornuate process projecting upwards above the greater tuberosity of the humerus and a long one directed caudoventrally in the neighbourhood of the lesser tuberosity, considerably below the clavicle.

The acromion is a delicate bar showing an approximately right angle bend just above its root and a gentle curve beyond that to its tip, which projects slightly beyond the lateral extremity of the clavicle. Its root is immediately behind the edge of the scapular part of the glenoid cavity.

File:Rutherford1914 fig03.jpg

Fig. 3. Wax-plate reconstruction from human embryo about 30 mm. greatest length. Magnification about x 8.5. Ossific plate shaded.

There is complete absence of the spine, but, on the other hand, there is present a distinct supraspinous element (which for reasons which will appear I will call the “ prescapula”), separated by a slight interval. near the middle of its length from the main part of the blade or “ postscapula ” and by a junctional zone dorsally to this. This prescapula is inclined at a slight angle to and occupies a more medial plane than the postscapula, as in fig. 5; a prominent ridge which lies on the superficial surface of the post-scapula, caudal to the interval between the two elements of the blade, indicates the future basis of support for the spine. By reason of this arrangement there exist two notches in this region, one which for my present purpose may be called coraco-prescapular, and becomes the future incisura scapularis, and the other which I will call acromio-postscapular, a temporary feature later filled up by the growth of the spine. Between these two notches is the narrow area of origin of the supraspinatus muscle which occludes on one aspect the interval between the two parts of the blade.

The roots of the acromion and coracoid, and most markedly that of the prescapula, are invested by a thick layer of cells which surrounds the origin of supraspinatus and forms the scapular extremity of the previously mentioned band or cellular condensation which reaches behind the main mass of the coracoid to be continuous" with the periosteum of the outer segment of the clavicle (fig. 6).

File:Rutherford1914 fig04.jpg

Fig. 4. Wax-plate reconstruction from human embryo about 30 mm. Schematic representation of scapula and limb skeleton. Magnification about x 6.

The humerus has a very remarkable shape. The shaft is bent outwards and forwards so as to form a salient angle opposite the deltoid insertion; its proximal half is arched outwards and its distal half backwards. Neither extremity in its relation to the axis of the shaft occupies a position corresponding to that of the adult bone; the proximal end is rotated somewhat outwards, and its articular surface is divided by a groove into two flattened condyles corresponding with the coracoid and scapular segments to which it is applied; the distal end occupies a plane such as might result from a great exaggeration of that angle between the axis of the trochlea and the long axis of the shaft which is characteristic of the adult. By reason of this the inner lip of the trochlea forms the distal extremity of the humerus, and the capitulum lies on the lateral aspect at a higher level. A delicate outgrowth from the perichondrium of the shaft near its lower end ventro-lateral to the median nerve indicates the position of the entepicondylic spine (fig. 7).

What has been said of the distal articular surfaces of the humerus conveys equally the relative position of the contiguous extremities of the radius and ulna. These two cartilages lie parallel, one above the other, but not quite in the same plane, owing to what may be ‘described as a very slight degree of pronation. There is present between their distal ends an independent cartilaginous nodule embedded in a thickly crowded mass of cells. This may represent the intermedium.

File:Rutherford1914 fig05.jpg

Fig. 5. Coronal section, human mbryo about 30 mm. Prescapa and postscapula, ossifying pine of scapula and trapezio-deltoid intersection.

Of the carpal cartilages those on the postaxial side are far in advance of those on the pre-axial side of the wrist; the same is true of the metacarpals and phalanges, the three digits of the ulnar side being almost equal in length and slightly longer than the index. End plates to the terminal phalanges are just recognisable by their dark staining and plexiform character. A Contribution to the Embryology of the Fore-Limb Skeleton 361

Model of Fore-quarter Skeleton of 30 mm. Human Fcetus (figs. 3 and 4;).

This reconstruction shows ten half-vertebrae, of which three are cervical, four complete, and three incomplete costal arches, the manubrium sterni, and representatives of three sternebrae and all the elements of the limbgirdle and free limb.

The position of the vertebrae has altered owing to the decrease of the spinal curve, and the demispines on the extremities of the neural arches are now more marked. The ribs are now much more slender, except dorsally, where they are very massive and show commencing angulation.

File:Rutherford1914 fig06.jpg

fiG. 6. Horizontal section through shoulder of 20 mm embryo.

sc., scapula; Acr., acromion; H., humerus; Co., coracoid; CL, clavicle; M., supraspinatus; and P., prescapulo-clavicular band shown attached dorsally where scapula is covered by thick cellular layer.

The manubrial cartilage is pierced by a small hole, and there is a median cleft in the third sternebra.

The clavicle has developed rapidly, and has now almost the adult form with two curvatures; the outer extremity is, however, not so flattened as in the adult.

The scapula reaches headwards to the level of the caudal border of the seventh cervical vertebra; its diamond shape is still preserved, but the length of that border which runs from the coracoid to the upper angle has increased disproportionately; it follows that T the “ prescapular ” bar is greatly increased‘ in size, the slit separating it from the postscapula has lengthened, and there is- quite a marked groove indicating the boundary between the two elements and extending from the slit to the dorsal border, which it notches. The growth of the prescapula’ has greatlyexceeded that of the postscapula in the 20-30 mm. period, and it now bears a greater proportion to the whole blade than does the resulting supraspinous part of the blade in the adult.

File:Rutherford1914 fig07.jpg

Fig. 7. Horisontal section, human embryo 20 mm., to show : H., humerus ; 1%., radius in cartilaginous continuity; Med., median nerve with entepicondylic process of perichondrium on its lateral "aspect (*).

The coracoid is little changed, being only rather smaller relatively. The glcnoid surface is now concave and undivided; it seems to have increased relatively in its lower blade part. The acromion has grown considerably and become flattened and spatula-like, so that the acromial angle of the. adult. is now recognisable. The tip of the process projects considerably ventral to the outer end of the clavicle. The right-angle bend near the root described in the previous stage is now a sharp projecting elbow or spur of cartilage connecting with a tract of young, actively growing cells which passes backwards over the dorsum scapulae, under cover of the small-celled fibrous intersection between trapezius and deltoid, nearly to the vertebral border. This tract represents the developing spine, and in it, especially . near. the L acromial elbow, ossification is well advanced (fig._.8). The ossifying region is shaded in the figures of the model, and it can be seen to extend into a projection at the scapular notch, over the edge of which it is continued on to the medial surface of the blade. The cells in which the ossification takes place are derived from the mass which in the 20-mm. stage clothed the roots of the postscapula, coracoid, and acromion.

File:Rutherford1914 fig08.jpg

Fig. 8. Coronal section, human embryo 30 mm. Shows elbow of acromion and ossification of spine (*).

The humerus is constricted in the region of spreading ossification, and is gently curved convex backwards. The two curvatures seen in the earlier stage are much less prominent in this later one owing to the great increase in length of the humerus, but both are easily identifiable still.

The upper end is now only slightly out-turned as compared with that of the adult bone, and the lower end has assumed, in the inclination of its axis to that of the shaft, a position intermediate between that in the earlier model and that representing the average of the two sexes in adult bones.

The radius and ulna likewise show constrictions in the ossifying zones. They are now very decidedly crossed. The ulnar styloid seems to be prolonged into the chondrifying discus articularis, and the same is true of the ulnar side of the lower extremity of the radius: between the two prolongations is a dense, unchondrified cell-mass. The contrast_ between this condition and that described for the 20-mm. stage is remarkable.

File:Rutherford1914 fig09.jpg

Fig. 9. Corona1 section, human embryo 30 mm. 0., centrale continuous through faintly staining zone with N., navicular.

The carpal elements, especially those of the distal row, are well forward in development, and there is visible, both on the palmar and dprsal aspects of the wrist, a distinct centrale; on that side it lies between navicular, lunar, and trapezoid, and on this between navicular, capitatum, and trapezoid; its surface isolation does not correspond with the conditions in the depth of the articulation, where it is continuous, through a faintly staining transitional zone, with the navicular (fig. 9). The navicular is in only narrow contact with the trapezium, which seems isolated at the radial end of-the distal row and is slightly smaller than the trapezoid still.

The metacarpals and phalanges of the free digits are so disposed as to suggest that the hand is grasping a globe of small diameter. The end plates of the terminal phalanges are very broad.


In discussing the features of the models described, the opportunity of reference to other human embryos in the collection at the London Hospital will be made use of. These embryos comprise 8 mm, 10 mm, l5.5 mm, 24.5 mm, and 25 mm stages, the measurements being in all cases those of greatest length. i _

The Vertebrae

The position and the curious form of the cervical vertebrae in Model I. led me to seek in earlier stages for an explanation of the pointed caudal projection of their bodies. It is only in the cervical vertebrae that this condition is very ma_rked,.and at first it was thought to be purely a positional moulding. Naturally shape and position are related features in ontogeny, a fact which in this instance is confirmed by the further observation that a much less and rapidly diminishing degree of- the same moulding is visible in the upper thoracic bodies. Still, it seemed only reasonable to expect that the shape of the cervical and upper thoracic bodies in the adult, with their prominent ventro-caudal lips, would require for its production something more than a process of purely adaptive growth of the scleroblastema of what may be called the typical centrum. It remains to explain the two correlated features by means of some material organic addition to these vertebrae during their development; such an organic addition is provided for in the 10-mm. stage when there is present connecting the costal elements of all the upper vertebrae a hypochordal bar of condensed cellular tissue (fig. 10) into which the chondrification of the body can extend in a caudo-ventral direction. This extension will be facilitated by the positional relations of the upper vertebrae, which here form a more sharply curved arch than those lower in the series, where the hypochordal element is less easily identifiable. Lewis, in touching upon the hypochordal element, limits it to the upper three cervical vertebrae and connects it, not with the costal, but with the neural arches. Bardeen notes its presence as low as the thoracic region.

File:Rutherford1914 fig10.jpg

Fig. 10. Human embryo 10 mm Trans. section costo-hypochordal part of vertebral blastema shown stretching between the ventro-lateral myotome buds in front of chords.

The Ribs

Apart from the excessive length of the ribs in relation to their average cross-section, the curve of the costal arches and the shape of the thoracic cavity resulting are noteworthy, since, in both models, the proportion between the depth and breadth of this is reversed as compared with the adult form; further, the cylindrical form of the earlier stage is seen to be transformed into a truncated cone in the later, the profile view of which shows it to be much blunter than after completion of its. full growth and expansion. In Model I. the ventral ends of the ribs are free, with the exception of the first, which is fused with the sternal bar, whilst the second, third, fourth, and fifth turn caudally alongside it, being accommodated in fossae on its lateral aspect; in Model II. these have fused with the side of the sternum. The cartilaginous sternal bar ends, in the 20-mm. stage, opposite thelfifth rib, and the ventral extremities of the lower ribs turn headwards, the sixth and seventh ending in a condensed blastema continuous with the cartilaginous sternum. It would thus appear that the thoracic skeleton is furthest advanced in development at its inlet, and that the fusion between rib and sternal cartilages is a secondary process, like the fusion of the sternebral halves, as seen occurring in the third body sternebra of the 30-mm. stage. The use of the term “sternebra” in blastemal and cartilaginous stages is of course unjustifiable, but it has been employed for convenience sake to designate those parts of the continuous bar or bars which correspond to the bony segments of later stages.

The Clavicle

The form of the clavicle in Model I. is very striking by reason of the absence of that curvature which characterises its outer segment in Model II. and in the adult. In the 10-mm. stage, which may be taken as the equivalent of that, '£.e. 106 mm., described by Lewis, the blastematous clavicle extends inwards only one-third of the distance to the ventral end of the first rib, which it reaches in the 15°5-mm. stage. Arrived at this stage in its growth, the clavicle is therefore represented only by that part which, if form be any criterion, is the equivalent, in the adult, of its inner two-thirds at most; at this stage it seems to halt until a short time before that represented in Model I. ; then begins a further and distinct stage in growth coincident with the onset of ossification, one in which the lateral third of the clavicle of the adult is evolved from the cellular mass at the distal end (fig. 11). In this way the clavicle achieves its full development, arriving at the condition seen in Model II. The striking difference in the two stages of the development of a bone whose two parts are just as strikingly different in the adult, separated as those stages are by a considerable period of time, seems to warrant that any theory of the growth of the clavicle should take full cognisance of the causal facts that in the first stage the growth is axipetal, in the second, axifugal. It may be contended that the axifugal growth of the outer segment here emphasised is undeserving of emphasis, since it is equalled by contemporary axipetal growth of the inner segment; but it must be remembered, on the contrary, that the proportionate increase of the two growing ends differs widely and remains in favour of the outer, otherwise no such change in shape as occurs between that in Model I. and that in Model II. would take place.

File:Rutherford1914 fig11.jpg

Fig. 11. Horizontal section, human embryo 15.5 mm.

01. clavicle continuous laterally with suprahumeral cell mass (S.H.M.) ; 15r., prescapula; Sc., postscapula; Acr., acromion; 0.H., omohyoid.

The Scapula

The most striking features of the growth of the scapula, as brought out by the two models, are the development of the supraspinous portion, of the blade and the late appearance and mode of origin of the spine. The partial separation of the growing cartilage which represents the supraspinous part of the blade, to which the name prescapula has been given, especially excites curiosity. This separation is first fully recognisable in the 15°5-mm. embryo (fig. 11), and reaches its greatest differentiation in that represented by Model II. ; it is the more remarkable in that the slit seen in the two models between pre- and postscapula is occupied only by cells of the limiting membranes or perichondria of the two parts (fig. 5), and is not a passage for vessels or nerve filaments. For this reason it is to be considered a truly morphological separation, and not merely a casual fenestration, such as is a noticeable feature of the cartilages of certain amphibian and reptilian shoulder girdles. It is further especially to be remarked that the prescapular segment is fully continuous with the rest of the blade laterally, where it bounds the scapular notch, and that, at its other extremity, it is marked off from the blade in both models by a distinct groove and projects somewhat from the vertebral border, where it forms a part of this; in the former situation also it has a decided ventral projection clothed by a distinct cell-mass, among the surface layers of which the fibres of origin of supraspinatus are interspersed. A comparable condition. is found in the scapula of the dolphin, Delphinus delphis (fig. 12 (III.)), a condition made the more noteworthy by the absence of anything (resembling a true acromion process. In the toothed whale, on the other hand, an acromion process is seemingly present, but the prescapular equivalent is absent to all appearances (fig. 12 (I.)), whilst in Mesoplodon Grayii a condition is present which can be regarded as an intermediate stage between the two; in it, to adopt the accepted terminology, the acromium is bent backwards into a prescapular position, without, however, becoming continuous with the blade at its vertebral extremity (fig. 12 (II.)). Here, seemingly, are three stages in the formation of a prescapular element out of what has been previously considered an acromion process—-an assumption which, in the absence of the clavicle and the scapular spine among Cetacea must, for the moment, remain a doubtful quantity. There is, then, in the development of the human scapula an element, the true morphological importance of which has hitherto been disregarded, and which I have called the prescapula, whilst in the scapula of the adult dolphin a similar element is separable which, in its turn, is to be compared with the so—called acromion of Mesoplodon Grayii and of the toothed whale. Turning to the development of the spine of the scapula, we find that appearances indicate an origin for it in a cellular proliferation connecting with the elbow or spur of the developing acromion process (as distinct from the adult acromial angle); this undergoes a precocious ossification (fig. 8) as it spreads dorsally under the trapezius-deltoid intersection from the region of the scapular notch. The nature of the cells is, from their appearance alone, not easy to discern, but such characteristics as they possess, coupled with their secondary continuity with the fully differentiated cartilage of the acromial elbow, places them as derivatives of cartilage cells. Perhaps the most striking fact brought out by the second model is the formation of a bony dorsal boundary to the scapular notch through the agency of cells which in the 30-mm. stage are continuous with those ossifying under the acromion, and which are only represented in the 20-mm. stage by the mass lying most thickly at the root of the prescapula. The position and relations of this cellular proliferation are very suggestive, especially in view of the ossific process which takes place in it so early after its assumption of rapid growth. firstly, it overlies the junctional zone between coracoid and scapula; secondly, it lies at the dorsal extremity of a band of thickly crowded cells which passes medial to the coracoid (figs. 6 and 11) to divide into three well-marked strands merging with the periosteal layer of the outer segment of the clavicle; thirdly, this connexion is identifiable in the 15°5-mm. embryo as a dense strand forming the mesial boundary of a cell-mass which includes more laterally the prechondral representatives of acromion and clavicle, the middle of which is less condensed than its mesial and lateral margins. The mesial margin is well seen in fig. 11, and can be traced dorsally to the nucleus of chondrification representing the prescapular element and ventrally to the clavicle, whilst the middle third of it gives origin to the omohyoid, just lateral to the subclavian vein. The ossification of the scapula commencing at a junctional zone is remarkable as a parallel phenomenon to that described by Professor Fawcett in the ossification of the clavicle ; closer examination may show even greater resemblance in the details of the process; whilst further, in view of the importance of junctional zones as centres of growth, brought out by the work of Professor Geddes on the Vertebrate Limb, an even more widely generalised significance may be attributable to this fact in the growth of the scapula. The cell strand connecting prescapula and clavicle is so far differentiated in fig. 6 that it represents the typical position of those ligamentous fibres which in many adults pass along the upper part of the conoid ligament into the transverse ligament; in this figure the omohyoid origin is not connected with these fibres, a factwhich is referable to the dorsal migration of the origin towards the scapula.

File:Rutherford1914 fig12.jpg

Fig. 12. I., Scapula of Physeter Macrocephalus ; II., Scapula. of Illesoplodon Grayii; III., Scapula of Delphinus delphis.

The explanation of the observed facts bases itself mainly upon the presence of the large cell-mass in the upper shoulder region in the 15.5-mm. stage. This mass is continuous with the acromion laterally, with the prescapula dorsally, and with the outer extremity of the chondrifying clavicular bar ventrally. It is in this mass that the specialisation of the clavicle-acromion arch so closely related to the evolution of a freely movable shoulder-joint must occur, and, since this specialisation is especially important in man, the steps of its development may be expected to abound in matter for comparison with the vertebrate girdles down even to those of thereptilia. Taking into consideration the variety of forms among the mammalia assumed by these portions of the skeleton, the cell-masses which are their precursors in ontogeny must be regarded as possessing a remarkable degree of lability, and, for this very reason, as being likely to preserve in the stages of their early development certain main features of their common and very early precursors in phylogeny. The earliest form of the human shoulder girdle, as figured by Lewis for a 10'5-mm. embryo, is a rectangular mass of crowded cells representing a scapula and situated high in the cervical region, to all appearance a plate-like headward prolongation of the axial condensation of the free limb.‘ In it, again, according to Lewis’s figure, there is a shallow furrow which may or may not be the first indication of the slit found in the cartilaginous scapula of later stages; if it be not, I am inclined strongly to the view that this rectangular mass of cells represents rather the prescapula of the cartilage stages than the postscapula, and from its caudo-lateral angle the postscapula may grow out; this View would be little more than speculation if it were not that the prescapula of a 15'5-mm. embryo shows a more advanced stage of differentiation than does the rest of the scapula. This is, however, a point on which the material at my disposal does not provide any further evidence of conclusive value, and which is of minor importance for the further development of the deductions made from the other facts observed in order to elucidate the connexion of prescapula and clavicle. This connexion has been fully described and figured, as has also the evolution of the so-called acromion in the Cetacea, through which it becomes converted in Delphvlmos delphis into a remarkable semblance of the human cartilaginous prescapula; and it only remains, in the first place, to emphasise that the connexion of the human prescapula with the clavicle is made at the mesial extremity of the lateral segment of the clavicle, and, in the second place, to postulate that the Cetacean so-called acromion connected with the clavicle of Cetacean progenitors, in order to reach the conclusion that the labile substance of the shoulder region can produce under evolution a mesial and a lateral arch ; that one prescapulo-palaeoclavicular — possibly identifiable in Cetacean ontogeny—this one acromioneoclavicular in nature.

In support of this conclusion and of the one postulate furthering it are a number of indisputable facts :—

1. That the ingrowth of the clavicle in stages from 10 mm. onwards occurs, not from the region of the tip of the acromion (sensu stricto as here applied), but from the ventral extremity of the strand of cells which forms the mesial limit of the suprahumeral arch mass, the outer periphery of which will later give rise to the outer segment of the clavicle ventrally and to the acromion dorsally. This is beautifully demonstrated in fig. 11, 15.5, which, it will be remembered, comes from a stage 155 mm., in which the clavicle has reached to the ventral extremity of the first rib.

2. That the variations in origin of the posterior belly of the omohyoid muscle almost never follow the lateral or acromio-clavicular, but nearly always the mesial or prescapulo-precoracoid arch.

3. That in the two-toed sloth, Uholwpus H0fi"ma'n/n/£7}, the prescapula and acromion are synostosed, and with the complete arch thus constituted from the periphery of the suprahumeral mass the clavicle articulates, which clavicle, to judge by its sole curvature, a ventral convexity, represents only the precoracoid portion of that bone (fig. 13). By reason of the primitive scale of general mammalian organisation found in the Edentata, and also by reason of the habits of the sloth, this fact supports both the conclusion I have come to and the postulate regarding the Cetacean clavicle necessary to its validity in the comparative sense.

File:Rutherford1914 fig13.jpg

Fig. 13. Scapula. Clavicle of two-toed sloth, Cholaepus Hofimannii, showing acromio-prescapular arch.

To summarise, the shoulder arch in man shows in its development a fair epitome of the phylogenetic steps in its specialisation from a type which would be fairly represented by the shoulder girdle of the toothed whale, did this possess a clavicle, a type extant in both Ornithorhynchus and Echidna. With advancing specialisation the original arch becomes converted to other uses; the primitive acromion becomes the supraspinous plate, and a new acromion is evolved, whose earliest trace is possibly identifiable in Echidna and whose intermediate form is exemplified in the sloth, this new acromion having a wider sweep and necessitating an addition to the primitive clavicle. Meanwhile the intermediate portion of the original arch merges in, or specialises as the transverse and part of the conoid ligaments of man.

This theory of the development of two arches in the mammalian shoulder girdle deserves a reference to the conditions in lower vertebrates. In the reptiles the so-called clavicle often reaches far on to the dorsum of the scapula, whilst as yet no acromion can be said to exist, indeed the dorsal portion of the clavicle occupies a position on the reptilian scapula which is only comparable with that of the scapular spine in man, or in other mammals in which a truly homologous spine is developed. Such an extension of the clavicle may reasonably be supposed, I think, to follow the line of our prescapulo-precoracoid arch, and thus leads on to the hypothesis that that cellular covering of the root of the prescapula (fig. 6) which ossifies to form a bony spicule, the definitive dorsal edge of the scapular notch, and which further extends to the root of the acromion and under the trapeziusdeltoid intersection to form the spine of the scapula, ossifying as it goes, is the homologue of part’ of the original reptilian clavicle. This is only,‘ however, a hypothesis, but it has so many favourable features that it seems worthy of the more exhaustive examination of embryonic and comparative material necessary for its proof and acceptance as a well-grounded theory.

The secondary coalescence of spine and acromion would well account for the proliferation which forms a spur on the acromial elbow in Model II., and a similar but more marked outgrowth in this junctional zone doubtless results in the formation of a metacromion as found in certain rodents, and markedly in the elephant. For particularly directing my attention to this structure, I am indebted to Professor Wright.

The humerus in the forms in which it appears in Models I. and II. is a very interesting object. Its curvatures are striking, and the torsion and angulation of its upper and lower extremities are brought out in a way which previously figured reconstructions have failed to indicate. This is to be attributed to the method I have adopted of drawing only the limits of the cartilage and avoiding the perichondrium, at stages in development when the differentiation of that tissue has reached or slightly overstepped its “ optimum.” The seeming contrast between the two forms is dispelled on close examination of the second model, and the change is attributable only to the enormously increased length of the shaft and its relatively decreased girth in the region of the primary ossification. Distortions of the shafts of long bones have been noted in embryo by Holl, Schomberg, and others, and attributed to disproportionately rapid growth in length; these are especially marked in hardened specimens and are therefore probably artificial. Of such artificiality I see no trace in the developing humerus, either in Lewis’s figures or my own material, and I consider that both the described curvatures of the shaft are a phenomenon of correlated growth during development of humerus and musculo-spiral nerve in the first place, somewhat exaggerated by the shortness of the shaft and the form of the unspecialised articular ends, and, possibly, later, by the related deltoid insertion in the second place. That the spiral groove of the adult humerus is so much broader than the structures lying in it is merely an indication of the early date of its production, when the girth of the nerve relative to the length of the bone was much greater and separated the origin of the two humeral heads of the triceps at the time when they were acquiring a firm attachment to the ossifying tissue‘ of the shaft; for muscles never attach themselves to cartilage, but only to the perichondrial membrane, which is later incorporated with the periphery of the bone. The causation of the spiral groove of the humerus has no relation to the torsions about to be considered.

The changes of torsion and angulation of the articular ends which take place between the 20-mm. and the 30-mm. stages are attributable in the case of the upper extremity to the pre- and postnatal alteration of position of the scapula, relative both to the axis of the humerus and to the curve of the chest wall; in the one case the change is an approach of the axillary border of the scapula to the humerus, and in the other it is a change due to the increase in length and curvature of the ribs, causing the scapula to migrate towards the vertebrae and the head of the humerus to rotate inwards to accommodate itself to the altered aspect of the glenoid. The alteration in the inclination of the distal articular surfaces to the axis of the shaft is productive of the specialisation of the elbow-joint for pronation and supination of the forearm bones. Lewis, in Keibel and Mallis T eastbook, p. 380, states that in an 11-mm. embryo, the forearm is midway between pronation and supination. This statement is incorrect, in principle and in detail, for, almost up to the 20-mm. stage, pronation and supination as movements or as postures are out of the question, owing to the form of the distal extremity of the humerus, the details of the skeleton showing that the forearm bones are parallel T and the hand apparently fully supine if the plane of the hand be projected upon the plane of the lower end of the humerus; the right-angled elbow, parallel forearm bones, and paddlepostured hand make an exact reproduction of the palaeo-reptilian fore-limb. The sex difference which is characteristic of the limb in the adult would seem to rest upon a less degree of specialisation of the upper extremity in the female than in the male, resulting in the greater “ carrying angle ” of the female. The correlation (if it be one) of this inferior grade of specialisation in the female upper limb With a higher grade of specialisation in the female pelvis is an interesting paradox.

The radius and ulna have been almost sufficiently discussed in connexion with the distal extremity of the humerus; it only remains to add that, if the inclination of forearm to arm be a true sex characteristic, the differential torsion of the two extremities of the ulna and the relative aspects of the articular surfaces of the radius should, if made capable of convenient measurement, prove a valuable means for determination of sex in osteology and anthropology.

The Carpus and Hand

The presence of a centrale in Model II. and the mode of its inclusion are not in agreement with the observations of Graefenberg. Graefenberg’s material seemingly did not include a stage as early as that of Model I., in which no centrale is to be found, for he states that it is among those carpal elements earliest differentiated, and, further, that it is probably broken up by the appearance of the lunar, and finally disappears, leaving no trace. As against these statements I have to place the facts that the centrale is unrepresented in my 20-mm. stage, yet in my 30-mm. stage it is visible on both Ventral and dorsal aspects of the carpus, where it is entirely separated from all other elements by a distinct perichondrium, and that in the depth of the massed elements it is structurally continuous with the naviculare through a faintly staining cartilaginous layer. (fig. 9). This further confirms Owen’s and MiVart’s views on the fate of the centrale, whilst opposing those of Rosenberg and Thilenius, and partially also those of Leboucq. The isolation of the trapezium on the radial side of the carpus has been noted in description of Model I.; its position confirms the observation of Hagen, one which Graefenberg professes to be unable to understand, in that it is practically confined to the ventral aspect of the wrist.

The question of the relation of intermedium and discus articularis cannot be fully entered into here because of the limited amount of material under discussion, but the appearance of a separate cartilaginous nodule in the 20-mm. stage, embedded in a mass of thickly crowded cells, and the absence of such an element in the later stage, which shows chondrification of the cell-mass in continuity, on the one side, with the ulnar styloid, and, on the other, with the inner edge of the extremity of the radius, both seem to indicate early differentiation of an intermedium, followed by its disappearance and replacement by the discus.

Summary Or Conclusions


1. The first and perhaps the most critical matter, in its bearing upon the technical results of this work, is the imperative necessity, in preparing a reconstruction of the cartilaginous skeleton, to adhere slavishly, in the drawings made, to the periphery of the fully differentiated cartilage and to exclude all the surrounding transitional perichondrial tissue.

2. The second conclusion of a technical kind follows from the first, and is, that reconstructions of the developing skeleton, at stages earlier than the 376 Mr N. C. Rutherford

“optimum” of cartilaginous development, must necessarily be ambiguous since the zones of transition are much wider and more difficult of limitation, whilst the shape of the growing chondral nuclei signifies nothing of a more than ontogenetic importance. It is not desired to convey by this that the form of the condensed blastemal skeleton suffers under the same disability for comparative purposes, quite the contrary; the blastematous stage, also, has an “ optimum” development, only one which is more diflicult to assess than is that of the cartilage; it probably lies in or around the immediately prechondral period. It is further desirable to emphasise that nothing in these conclusions can apply to elements which manifest anything in the nature of direct ossification, that is to say, which, among vertebrates, may ossify otherwise than through the intermediation of cartilage; for example, a drawing and reconstruction of such an element as the chondral clavicle will convey nothing of value, because the chondrification of the clavicle never reaches an “ optimum ” in the accepted sense.

3. The blastema stage of the shoulder skeleton represents in its massive cell agglomeration two arches, one mesial, consisting of the precursors of the supraspinous portion of the scapula and the cell-strand connecting this with the clavicle at the junction of its sternal and acromial segments, and one lateral, representing the acromion and acromial segment of the clavicle, both having comparable though simpler bony representatives in the Edentate Oholcvpus Hofimannii. In the absence of the clavicle of their progenitors the Cetacea show a conversion of the dorsal pillar of the medial arch into prescapula.

4. The supraspinous portion of the scapula may possibly be the original scapula of phylogeny; it certainly is a separate morphological constituent of the scapula, as evidenced by the course of its development in man and by the various forms of its equivalent in the Cetaceans cited.

5. The formation and precocious ossification of the spine of the scapula begins in cells which, appearing at the root of the prescapula, grow on to the dorsum along the coraco-scapular junction and under the trapeziusdeltoid intersection, cells which are therefore intimately related to the mesial arch, and may be, like this, representative of the reptilian clavicle. Spine and acromion are joined up through the intermediation of these cells, and the junctional proliferation may be so marked as to result, under ossification, in the production of a metacromion, as in certain rodents, and markedly in the elephant.

6. The spiral groove of the humerus is due to the impress of the musculo-spiral nerve and is not due to torsion, which, however, is not absent during the growth of the cartilage, but affects only the proximal end.

7. The specialisation of the distal articular surfaces of the humerus takes place Within the stages examined, and its degree is different in the two sexes.

8. The carpal centrale is most fully developed in cartilage about the “optimum” period for the other carpal cartilages. It fuses with the navicular.

9. The intermedium is present in the 20-mm. stage, and disappears before the 30-mm. stage to make way for the chondrification of the articular disc from its ulnar and radial extremities.


I desire to record my thanks to Professor W. Wright for the kindly interest With Which he has followed the completion of this Work; to Professor A. C. Geddes, firstly, for the use of apparatus for reconstruction, Which was his private property, and, secondly, for the advice Which, when in any difficulty, I could always fall back upon with benefit.

The publication has been carried out with the help of a grant from the Research Fund of the London Hospital Medical College; the drawings are

by Mr W. Thornton Shiells; in the photography I have been much assisted by Mr T. Collyer Summers.

Bibliography of References

Most of the papers referred to are quoted in the Bibliography on pp. 391-397, vol. i., of the Manual of Human Embryology, by Keibel and Mall. In addition are the following :—

FAWCETT, E., “The Development and Ossification of the Human Clavicle,” Journ. of Anat. and Phys., vol. xlvii. p. 225.

GEDDES, A. (3., “The Origin of the Vertebrate Limb,” Journ. of Anal. and Phys., vol. xlvi. p. 350.

Cite this page: Hill, M.A. (2024, May 18) Embryology Paper - A contribution to the embryology of the fore-limb. Retrieved from

What Links Here?
© Dr Mark Hill 2024, UNSW Embryology ISBN: 978 0 7334 2609 4 - UNSW CRICOS Provider Code No. 00098G