Evolutionary Change in the Primate Wrist and Inferior Radio-ulnar Joints 0. J. LEWIS D e p a r t m e n t of A n a t o m y , St. Bartholomew's Hospital Medical College, Charterhouse Square, L o n d o n E.C.1. ABSTRACT The joints at the junction of the forearm and hand are described in a number of Primates, including Man. It is shown that the original lower articular extremity of the ulna recedes from its primitive articulation with the triquetral and pisiform to become the ulnar styloid process, while a neomorphic ulnar head is elaborated entering into a new synovial inferior radio-ulnar joint. The lower capsule of this joint becomes the triangular articular disk. In the anthropoid apes a meniscus (with a laterally-directed concave free border bounding the ulnar styloid) develops in the interval between the receding ulna and the carpus. In gibbons this meniscus presents a lunula (0s Daubentonii) which is also present in the human fetus as a transient cartilaginous nodule often erroneously homologized with the 0s intermedium. The upper articular surface of the human wrist joint is formed by the radius, the triangular articular disk and the homologue of the ape meniscus. Between the two latter components is a n aperture, similar to that of apes, leading into a pea-sized diverticulum of the wrist joint - the pre-styloid recess. The mechanism of the evolution of new synovial joints is discussed. Despite numerous observations recorded during the past two centuries concerning the comparative morphology of the joints between the primate forearm and hand, no clear or coherent understanding of these structures has as yet emerged. Even more surprisingly the modern textbook accounts of the human wrist joint are often difficult to reconcile with the findings of direct observation in the dissecting room. In Man and in other Primates a number of such findings call for intelligible and convincing explanation. Primate evolution from the Prosimii to the Hominoidea has been characterized by an increased mobility of the hand, effected by two major skeletal modifications. Firstly the ulna has become withdrawn from its primitive articulation with the pisiform and triquetral bones (Flower, 1885; G r a d , '55), its original carpal articular extremity being finally represented by the human styloid process; this change has increased the range of hand adduction (Wood Jones, '49). Secondly, and correlated therewith, pronation and supination have been enhanced by the elaboration of a new ulnar capitular articulation within a synovial inferior radio-ulnar joint. This neomorphic ulnar head is excluded from the wrist joint cavity by a triangular articular disk arisANAT. REC., 151: 275-286. ing, apparently, as a modification of the lower part of the capsule of the inferior radio-ulnar joint (Wood Jones, '49). Two questions therefore pose themselves, viz. (1) by what agency has the ulnar styloid been excluded from the wrist joint ( a process recapitulated - save very exceptionally - in human ontogeny) and ( 2 ) has there been evolution of an entirely new, diarthrodial, inferior radioulnar joint? The gibbon carpus contains an additional bony element, situate between the ulnar styloid and the triquetral and pisiform bones, apparently first described by Daubenton (1766) and hence often known as the 0 s Daubentonii; it was subsequently observed by Leboucq (1884) and by Kohlbrugge (1890) (fig. 1A). It is also established that the human embryo, during the second to fourth month period, exhibits a transient nodule of hyaline cartilage at a comparable carpal site, i.e., in the lose connective tissue immediately distal to the developing triangular articular disk (Leboucq, 1884; Corner, 1898). Even in the adult a radiographic opacity is sometimes observable in this same situation (Wood Jones, '49; Oliver, '62), while Pfitzner (1895) described an actual bone (trique275 276 0. J. LEWIS B Fig. 1 A, the carpus of Hylobates leuciscus (from Kohlbrugge, 1890). B, a tracing of a normal arthrogram of the human wrist joint, showing the pre-styloid recess (from Kessler and Silberman, '61). Lu, lunate; OsD, 0 s Daubentonii; Pi, pisiform; Ra, radius; Sc, scaphoid; Tr, Triquetral; U1, ulna. trum secundarium, triangulare) here in an osteological preparation. The gibbon carpal element and the identically located cartilaginous nodule of the human fetus have been interpreted as serially homologous with the pedal 0s trigonum (Leboucq, 1886); and just as the 0s trigonum has been said to represent the 0s intermedium tarsi so this carpal element has been interpreted as an 0s intermedium antibrachii (Poirier and Charpy, '11; Testut, '04).' Since it is obvious that an 0s Daubentonii can only appear subsequent to the retreat of the ulna from its direct articulation with the carpus, it is tempting to suppose that such appearance is in some fashion directly associated with this retreat. Now Lewis ('64) has shown that the comparable free "0s intermedium tarsi" of the marsupial (held to correspond to the anomalous human 0s trigonum) is nothing more than a lunula, i.e. an ossification within the thick part of an intra-articular meniscus. The likelihood suggests itself therefore that the correspondingly placed ossicle in the gibbon carpus is nothing more than a similar functional adaptation. During the last century anatomists (Henle, 1856; Morris, 1879) described the proximal articular surface (the receptive cavity) of the condyloid human wrist joint as a smoothly concave ellipse whose transverse extent was limited by attachments to the margins of the upper articular surfaces of the scaphoid and triquetral bones. But clearly the medialmost part of this surface (that part related to the triquetral facet) cannot be the triangular articular disk itself. Henle (1856) described this disk as undergoing an apical cleavage into upper and lower laminae, the former attaching to the root of the ulnar styloid, the lower continuing as part of the curved articular surface. Henle's account does not accord with the modern textbook descriptions which regard the entire apex of the triangular articular disk as attached to a pit at the base of the ulnar styloid and whose accompanying illustrations are not always reconcilable with their text. (Pictures frequently show increasing thickness of the disk on being traced medially. ) French anatomists (Testut, '04; Poirier and Charpy, '11) have described a promi1 Actually two additional ossicles have been described just proximal to the first row of carpalia in the aibbon. viz.. the aDDarentlv constant 0 s Daubentonii- and ' another (rkiatively uncommon) ossicle, more laterally adjacent to the radius, the 0 s Camperi (Kohlbrugge, 1890). Thilenius (1895, 1896) interpreted the 0 s Camperi and the human embryonic carpal nodule as homologues and representative of the 0 s intermedium, but this distinction between the two supernumerary gibbon ossicles has not been retained in later derivative accounts. 277 EVOLUTION O F WRIST JOINT nent cul-de-sac (the pre-styloid recess) establishing a relationship between the wrist joint cavity proper and the ulnar styloid process. This pea-sized diverticulum of the wrist joint cavity has recently been demonstrated constantly in arthrograms (Kessler and Silberman, '61) (fig. 1B). Yet it escapes notice in British anatomical treatises, and its precise topography is ignored by the French anatomists, although Poirier and Charpy state that occasionally the wrist joint's internal lateral ligament is of a tubular nature, occupied by the pre-styloid recess and attached to the base of the ulnar styloid, whose tip plays freely within its synovially lined cavity. It seems not unlikely that the pre-styloid recess is related in some way to the phylogenetic and ontogenetic retreat of the ulnar styloid from the carpus. It is shown below that a clear concept of wrist joint phylogeny clarifies the present unsatisfactory morphological picture and necessitates some reappraisal of traditional interpretations of human wrist joint morphology. crown-rump length and of serial coronal sections of the hand and forearm of a fetus of 13 cm crown-rump length. OBSERVATIONS Trichosurus vulpecula Arrangements herein typify the basic mammalian carpus pattern. The large and elongate pisifonn articulates with the triquetral : a concave articular surface, largely triquetral but completed anteriorly by the pisiform, receives the convex articular lower extremity of the ulna, no meniscus intervening. Immediately proximal to its spherical lower end the ulna constricts as a neck, proximal whereto it expands into an evenly fusiform enlargement traversed about mid-way by the distal epiphyseal line. This ulnar enlargement is intimately united to the radius by ligamentous tissue representing a modification of the distal portion of the interosseous membrane. No synovial cavity is present: the inferior radio-ulnar joint is a syndesmosis. Lemur fulvus (fig. 2A) In many respects the arrangements are The wrist and inferior radio-ulnar joints similar to those in Trichosurus but the were first studied in the common brush- articular surfaces of ulna on the one hand, tailed possum (Trichosurus vulpecula) and triquetral and pisiform on the other, which demonstrated the generalized mam- show a lesser degree of curvature and the malian arrangement. Against this estab- pisiform makes a greater relative contribulished basic pattern the articulations were tion to the lower joint surface. The ulna dissected in examples of the following pri- again presents a fusiform swelling, travmate species : brown lemur (Lemur ful- ersed by the lower epiphyseal line, and v u s ) , weeper capuchin monkey (Cebus united by a ligament to the radius. In this nigrivittatus), olive colobus monkey (Proco- case, however, a small synovial cavity ocZobus v e r u s ) , white-thighed colobus mon- cupies the interior of the ligamentous key ( Colobus polykomos vellerosus), white- bond, and each bone presents a tiny cartinosed monkey (Cercopithecus nictitans), lage-clothed articular facet on its epiphysilvery gibbon (Hylobates lar leuciscus), sis. Two prominent ligaments, converging lar gibbon (Hylobates lar) and chimpanzee on the lunate from the neck of the ulna and the front of the radial styloid process, ( P a n satyrus). The corresponding joints were carefully are seen on the inner aspect of the anteexamined in 19 adult human arms; in five rior capsule of the wrist joint; these are the of these specimens the radiocarpal joint palmar radiocarpal and ulnocarpal ligacavity was injected with neoprene latex ments. prior to dissection. The joints were also Cebus nigrivittatus examined, with the aid of the dissecting Again a large pisiform forms, with the microscope, in human fetuses of 15 cm and 16 cm crown-rump length; the find- triquetral, a concave facet receiving the ings here were confirmed by the examina- convex lower extremity of the ulna, withtion of serial transverse sections of the out the intervention of any meniscus. forearm and hand of a fetus of 24 cm Again the ulna has a constricted neck just MATERIAL AND METHODS 2 78 0. J. LEWIS A B Fig. 2 A, the right wrist of Lemur fulvus. B, the right wrist of Cercopithecus nictitans. In each case the hand has been flexed on the forearm after opening the wrist joint from behind. In each case the palmar ulnocarpal and radiocarpal ligaments are shown converging on the lunate and in B is shown the neomorphic ulnar head and the emergent triangular articular disk. In each case the lower ulnar epiphyseal line is represented by the series of short vertical lines. The lettering is as in figure 1. above its distal extremity (the future styloid process) but, instead of a fusiform swelling related to the radius, manifests an asymmetrical enlargement - a laterally placed prominence which clearly heralds the appearance of the “head” of human anatomy. (To avoid confusion, the terms of human anatomy - styloid process and head, respectively - will be used in the following descriptions for the original carpal articular extremity and for this newlyevolved enlargement.) A tough ligament unites this ulnar head to the radius without the intervention of any synovial cavity. The lower aspect of this ligament is smooth, attaches medially to the neck of the ulna, and forms part of the upper articular surface of the wrist joint, just as does the triangular articular disk of the human inferior radio-ulnar joint. A separate ligamentous band (the palmar ulnocarpal ligament) joining the neck of the ulna to the lunate, with attachment to the pisifonn between, is related to the anterior margin of this homologue of the human articular disk. There is also a prominent palmar radiocarpal ligament joining the front of the radial styloid to the lunate. Procolobus verus In all essentials the arrangements reflect those described for Cebus. Colobus polykomos vellerosus The articulations under consideration closely resemble those of the preceding species with one further modification, namely the appearance of a small synovial cavity (essentially a bursa) within the interosseous ligament between the ulnar head and the radius. The cavity of this bursa is related particularly to the ulnar head, and the radius here bears no true articular facet, hence the joint could not justifiably be termed a true diarthrosis. Cercopithecus nictitans (fig. 2B) The joints resemble those of Colobus polykomos vellerosus save that the elaboration of a synovial inferior radio-ulnar joint has progressed further. There is a considerable radio-ulnar synovial cavity and the ulnar head now has a convex, carti- -.-l,..l JLVULU l l U l Y 279 ur WRIST JOINT lage-clothed, articular surface; the articular facet on the radius is more rudimentary. Hylobates Zar leuciscus (fig. 3) There is a fully elaborated synovial inferior radio-ulnar joint, separated from the wrist joint by a triangular articular disk. The original ulnar articular extremity, now transformed into a recognizable styloid process, has retreated somewhat from the carpus. Correlated with this is the appearance in this interval of a very large, obtrusive intra-articular meniscus, which thus intervenes between the cartilage-clothed ulnar styloid process and the triquetral, where the latter bears the pisiform articulation. The meniscus has a concave free border facing laterally and a convexity attached to the wrist joint capsule. Its two horns have important attachments: the posterior is inserted on the radius behind the inferior radio-ulnar joint; the anterior is affixed partly to the radius, but mainly to the lunate which it reaches by merging with the palmar ulnocarpal ligament observed in monkeys spanning the interval between ulnar neck and lunate at the anterior margin of the triangular articular disk. Within its very thick medial portion the meniscus is ossified, i.e., it contains a lunula. The size and position of this ossicle proclaim it as the 0s Daubentonii. Nothing corresponding to the 0s Camperi was observed. A strong palmar radiocarpal ligament is present. Hylobates lar (This is a young specimen and at the very start of the Juvenile I1 period, according to the criteria of Schultz, '44: the permanent first molars and central incisors are all erupted and the permanent lower lateral incisors are just erupting; the coracoid process is not yet united to the remainder of the scapula and ilium and pubis are ununited.) TrD . RcL Fig. 3 The right wrist of HyZobates lar Zeuciscus with the posterior joint capsules removed, and the hand flexed on the forearm. Lun, lunula; Men, intra-articular meniscus; RcL, palmar radiocarpal ligament; TrD, triangular articular disk; other lettering as in figure 1. 280 0. J. LEWIS The joints of this specimen resemble those of the previous species, but the unossified intra-meniscal lunula consists of a well-dehed, circumscribed, nodule of hyaline cartilage. Pan satyrus (fig. 4 ) The arrangements here present a clear transition between those in the gibbon and man. A meniscus is present having similar attachments to that of the gibbon, but containing no lunula. The wrist joint characteristics of this species, and their correlation with the human condition, can be best demonstrated by opening the wrist joint from behind so that the meniscus (whose posterior horn, in any case, attaches to the radius) is folded forward with the forearm bones, and thus appears as part of the receiving cavity for the carpal condyle. A deficiency is then apparent in this upper articular surface, located near the apex of the triangular articular disk: it is in fact the gap within the concavity of the meniscus, and here is exposed the primitive ulnar articular extremity (the styloid process). This gap leads into what is virtually an upper compartment of the wrist joint wherein lies the still articular cartilage-covered ulnar styloid process. This upper synovial cavity has its greatest proximal extent on the anterior aspect of the styloid process, and its lining membrane presents prominent synovial fringes, some of which protrude into the lower compartment through the connecting aperture. The constitution of the human upper articular surface is thus clearly foreshadowed, the upper synovial compartment being the precursor of the human pre-styloid recess. Homo (fig. 5 ) The adult joints. Understanding of the chimpanzee condition elucidates the human pattern. The ulna has withdrawn further from its primitive carpal articulation and no longer presents itself, as in Pan, through the orifice of a meniscus. But even in Homo the continuous curve of the upper articular surface presents an opening - the aperture leading into the pre-styloid recess. This opening is found close to the apex of the triangular articular disk and is easily overlooked since its margin is crenated and not sharp and TrD I I - -RcL Fig. 4 The right wrist of Pan satyrus with the posterior joint capsules removed and the hand flexed on the forearm. Men, intra-articular meniscus; TrD, triangular articular disk; UlSt, ulnar styloid; other lettering as in figure 1. 281 EVOLUTION OF WRIST JOINT UI S t I I I I TrD I I - - - -RcL Fig. 5 The right human wrist joint opened from behind; the dorsal radiocarpal and the palmar ulnocarpal ligaments have been cut and the hand flexed on the forearm. The broken line indicates the position of the pre-styloid recess whose opening into the wrist joint, with its crenated margin, is shown in solid black. RcL, palmar radiocarpal ligament; TrD, triangular articular disk; UlSt, ulnar styloid; other lettering as in figure 1. prominent as in Pan. It readily admits a blunt probe into the pre-styloid recess, the homologue of the upper compartment of the chimpanzee wrist joint cavity, and similarly situated in relationship to the ulnar styloid; the major prolongation of this diverticulum is anterior to the ulnar styloid. The upper articular surface of the human wrist joint is thus a composite structure formed by the amalgamation of three structures - articular surface of the radius, triangular articular disk and the anthropoid ape meniscus. The last component has become that portion of the surface which lies in contact with the triquetral articular facet. Variations in size of the human ulnar styloid are associated with corresponding variations in the prestyloid recess, which may project more or less horizontally from the wrist joint cavity or may be more vertically aligned. The ulnar styloid, where related to the recess, may present no special modification, but as a cartilage-covered projection may sometimes actually invaginate the synovial cavity. The recess was present in all the hands examined and was clearly demonstrated in all specimens injected with neoprene latex: its extent varied from 6 mm to 10 mm. Not to be confused with the opening into the pre-styloid recess is a common deficiency (apparently pathological) in the triangular articular disk and an additional frequent communication with pisotriquetral articular cavity; the latter is clearly a retention of the normal communication found here in anthropoid apes. In Man, as in the other Primates described, a strong ligament passes from the 282 0 . J. LEWIS radius adjacent to its styloid process, to attach to the front of the carpus and particularly to the lunate. This ligament is clearly apparent only when viewed from the joint interior and few of the authoritative descriptions give it adequate emphasis. The human fetus. In both specimens ( 15 cm and 16 cm crown-rump length) examined by dissection the arrangements provided a connecting link between the features observed in Pan and those in Homo and confirmed the morphological interpretation given above : thus the ulnar styloid approached the triquetral quite closely and the intervening homologue of the ape meniscus was more clearly apparent as a separate entity, with a more clearly defined margin than obtains in the adult condition. This thin free margin was closely applied to the under surface of the triangular articular disk, and formed a valve-like entrance, which freely admitted a probe, to the pre-styloid recess and the adjacent styloid process. Examination of transverse serial sections of the hand and forearm of a fetus of 24 cm crown-rump length confirmed these findings. The coronal serial sections from the 13 cm fetus also demonstrated the pre-styloid recess with well-defined margin of entrance and the flap-like homologue of the ape meniscus, with attachment in the most anterior sections to the lunate. These sections revealed the great thickness of the triangular wedge of tissue intervening between the ulna and triquetral and containing the ulnar styloid and the prestyloid recess. The synovial membrane walling the recess was extremely vascular. DISCUSSION Evolution of the wrist joint. An important feature of primate evolution, viz. increasing mobility of hand on forearm, centers around the retreat of the ulna from its primitive carpal articulation. Pari passu with this, a thick meniscus develops in the interval so created, which is not surprising in the light of Whillis's ('40) work on joint development. At all sites of such development a mesenchymal joint disk forms, and where not subjected to pressure may remain as an intra-articular structure; at an early stage therefore a meniscus is present at the periphery of most joints. Retention of a meniscus at the wrist of adult anthropoid apes thus follows as a logical consequence upon the withdrawal of the ulna. The general failure to recognize this meniscus, or to distinguish it from the morphologically different triangular articular disk, has occasioned a great deal of confusion. In gibbons the thickest part of this meniscus exhibits an ossification, i.e., a lunula. (Grassk's ('55) statement that the supernumerary gibbon carpal ossicle is a sesamoid in the tendon of M. extensor carpi ulnaris is without justification). The work of Pearson and Davin ( Y l ) , Pedersen ('49) and Barnett ('54) has revealed that such lunulae, found in many intra-articular menisci, are functional modifications apparently preventing compression of the thickest portions. There can be little doubt that the transient cartilaginous nodule of the human fetal wrist represents an ontogenetic recapitulation of such a lunula. Its appearance in the connective tissue distal to the triangular articular disk and its early attachment to the lunate (Leboucq, 1884; Henckel, '31) support this view. The recognition of these primate skeletal elements as intra-meniscal structural modifications seems effectively to dispose of the commonly held view that they represent an 0 s intermedium. (Discussion as to whether the 0 s Daubentonii or the 0s Camperi' represent the intermedium is likewise redundant: more than one lunula may be found in a meniscus and there is little doubt that the rare 0s Camperi is such a supernumerary lunula.) The view of Leboucq (1884) and of Corner (1898) that these structures represent the carpal element of a lost post-minimal digit, of which the pisiform is the sole remaining part, is no longer tenable. Both Henckel ('31) and Olivier ('62) have earlier rejected interpretation of the elements in question as part of the primordial skeleton and have considered them to be "secondary structures." This vague alternative is given substance by the present demonstration that they are actually intra-meniscal lunulae . The gibbon wrist meniscus, with its contained lunula, corresponds in position to a ~~ 2 ~ See footnote 1. EVOLUTION OF WRIST JOINT similar meniscus, also containing an ossicle, found in the marsupial foot (Lewis, ’64). Leboucq (1886) had some justification for his comparison of the pedal 0 s trigonum with the additional skeletal element of the hominoid wrist. Unfortunately he was unaware that both represent functional adaptations to similar changes in upper and lower limb, readily appreciable on the basis of the general principles of joint development : thus the initiating factor in the appearance of the forelimb meniscus is retreat of the ulna from the triquetral (ulnare) ; the analogous change in the foot is withdrawal of the fibula from articulation with the calcaneus (fibulare). Since mere regression of the ulna automatically initiates the changes leading to the appearance of a meniscus and lunula, it would not be surprising if these structures had appeared more than once in mammalian forms. Indeed, Daubenton (1766) noted the ossicle bearing his name to be present in the magot (Macacus inuus ) . T h e h u m a n wrist joint. The human joint bears the clear imprint of its evolutionary history, in the light of which a number of its features gain emphasis and added significance. The upper articular surface is formed from three components - the radius, the triangular articular disk and the homologue of the ape meniscus. The latter two components are fairly well blended, thus obscuring somewhat the identity of the most medial part of the receptive cavity (that part in contact with the triquetral) as the homologue of the ape meniscus. The opening bounded by the meniscus in Pan remains in Man as the entrance to the pre-styloid recess. This synovial diverticulum enters into relationship with the ulnar styloid process and perhaps invests its lower part, in which event the source is clear of the description by Poirier and Charpy (’11) of the occasional synovial-lined tubular form which may be assumed by the internal lateral ligament of the wrist joint. It was noted above that the wall of the pre-styloid recess is very vascular: from this feature, presumably, is derived Henle’s (1856) description of an aggregation of blood vessels separating the apex of the triangular articular disk into 283 two ligamentous laminae - the “ligamentum subcruentum.” The human palmar ulnocarpal ligament and the palmar radiocarpal ligament are both significant structures throughout primate evolution. Some of these features would seem to merit clinical consideration. Thus, the relationship of the pre-styloid recess may be of significance in those cases of Colles’ fracture in which the ulnar styloid is avulsed. There is also little doubt that this blind cul-de-sac is a common site for pathological change. The two palmar ligaments converging on the lunate are worthy of note when considering the mechanism of dislocation of the lunate bone. Evolution of the inferior radio-ulnar joint. The Primates exhibit the transition in this situation from a mere bursa to a typical diarthrosis. Elaboration of this diarthrosis can only occur as a sequel to prior changes in the wrist joint, for full pronation and supination are possible only after the carpus is freed from its direct ulnar articulation. The lower capsule of the new diarthrosis, broadened by development of the neomorphic ulnar head, becomes the triangular articular disk. The appearance of new diarthroses has considerable interest. Other examples are known: Bock (’59) has described a secondary jaw articulation found in certain birds, and Lewis (’59) has described a coraco-clavicular diarthrosis (probably present at birth) occurring as a variant in Man, It is of course well known that pseudoarthroses, with all the attributes of normal diarthroses, may arise at fracture sites; clearly, the capacity to form a new synovial joint is a normal bodily response which may be realized during an individual’s lifetime in response to use. But the joint under consideration is inherited. It fits into the category of adaptations known as pseudoexogenous, i.e., those which could arise as an adaptive response during an individual’s life time but which, in fact, are of developmental origin. This type, at first sight, appears to require a Lamarckian interpretation, but, in fact, such contingencies are adequately covered in the modern synthetic theory of evolution; it is clear that developmental preemption of such a character is of selective 284 0. J. LEWIS advantage (Medawar, ’57). The underlying genetical and natural selection basis of such adaptations, which has been called “genetic assimilation,” has been experimentally supported by the work of Waddington (’57, ’60). There is a great contrast between the joints at the junction of the forearm and hand in the more primitive of the Primates and those in Man. This might suggest a change of macroevolutionary proportions, demanding a mutationist explanation. But, as shown here, the changes have been produced by gradual phylogenetic modification of existing structures, following the usual pattern in the elaboration of so-called evolutionary novelties (Mayr, ’60). Observations presented above demonstrate that some progress, at least, towards a diarthrodial inferior radio-ulnar joint has occurred more than once among the Primates, yielding yet another example of the frequency of parallel evolution. Similarly, the additional basi-temporal articulation of some birds has evolved independently several times (Bock, ’59) as has the mammalian jaw articulation itself (Simpson, ’59). The mechanism of genetic assimilation furnishes excellent opportunities for parallel evolution whereas a mutationist or saltationist explanation carries with it an inherent improbability for frequent repetition. The temporary, and functionally useless, recurrence in Man of the ape lunula provides another example of genetic assimilation of a functional modification. ACKNOWLEDGMENTS I should like to thank Professor A. J. E. Cave for providing the material used in this study and for valuable advice in the preparation of the manuscript. LITERATURE CITED Barnett, C. H. 1954 The structure and functions of fibrocartilages within vertebrate joints. J. Anat. Lond., 88: 363-368. Bock, W. J. 1959 Preadaptation and multiple evolutionary pathways. Evolution, 13: 194-211. Corner, E. M. 1898 The morphology of the triangular cartilage of the wrist. J. Anat. Lond., 32: 272-277. Daubenton, L. 1766 Histoire Naturelle G6n6rale et Particulibre avec l a Description du Cabinet du Roi. Tome XIV. Imprimerie Royale, Paris. Flower, W. H. 1885 A n Introduction to the Osteology of the Mammalia, third edition. Macmillan & Co., London. Grass6, P. P. 1955 Trait6 de Zoologie. Tome XVII (Deuxibme Fascicule). Masson et Cie, Paris. Henckel, K. 0. 1931 Beitrage zur Entwicklung der Primatenhand. 111. Uber die Entwicklung der Discus articularis des distalen RadioUlnargelenks beim Menschen. Morph. Jb., 68: 293-300. Henle, J. 1856 Handbuch der Systematischen Anatomie des Menschen. Erster Band (Zweite Abtheilung). F. Vieweg und Sohn, Braunschweig. Kessler, I., and 2. Silberman 1961 An experimental study of the radiocarpal joint by arthrography. Surg. Gynec. Obstet., 112: 33-40. Kohlbrugge, J. H. F. 1890 Versuch einer Anatomie des Genus Hylobates. Erster Teil. In: Zoologische Ergebnisse einer Reise in Niederlandisch Ost-Indien. Ed. by M. Weber. Bd. 1, 211-354. Leboucq, H. 1884 Recherches sur la morphologie du carpe chez les mammifbres. Arch. Biol., Paris, 5: 35-102. 1886 Sur la morphologie du carpe et du tarse. Anat. Anz., I: 17-21. Lewis, 0.J. 1959 The coraco-clavicular joint. J. Anat. Lond., 93: 296-303. 1964 The homologies of the mammali a n tarsal bones. J. Anat. Lond., 98: 195-208. Mayr, E. 1960 The emergence of evolutionary novelties. In: Evolution after Darwin. Ed. by Sol Tax. Vol. I: The Evolution of Life. University of Chicago Press, Chicago. Medawar, P. B. 1957 The Uniqueness of the Individual. Methuen and Co., London. Chap. IV, 79-107. Morris, H. 1879 The Anatomy of the Joints of Man. J. and A. Churchill, London. Olivier, G. 1962 Formation du Squelette des Membres chez l’Homme. Vigot Frbres, Paris. Pearson, K., and A. G. Davin 1921 On the sesamoids of the knee joint. Biometrika, 13: 133-1 75. Pedersen, H. E. 1949 The ossicles of the semilunar cartilages of rodents. Anat. Rec., 105: 1-9. Pfitzner, W. 1895 Beitrage zur Kenntniss des menschlichen Extremitatenskelets. Morph. Arb., 4: 347-570. Poirier, P., and A. Charpy 1911 Trait6 d’Anatomie Humaine. Tome I. Masson et Cie, Paris. Schultz, A. H. 1944 Age changes and variability in gibbons. Am. J. Phys. Anthrop., 2: 1129. Simpson, G. G. 1959 Mesozoic mammals and the polyphyletic origin of mammals. Evolution, 13: 405414. Testut, L. 1904 Trait6 d‘Anatomie Humaine. Tome I. Octave Doin, Paris. Thilenius, G. 1895 Das 0 s intermedium antebrachii des Menschen. Morph. Arb., 5: 1-16. EVOLUTION O F WRIST JOINT - 1896 Untersuchungen uber die morphologische Bedeutung accessorischer Elemente am menschlichen Carpus (und Tarsus). Morph. Arb., 5: 462-559. Waddington, C. H. 1957 The Strategy of the Genes. Allen and Unwin, London. 1960 Evolutionary adaptation. In: Evolution after Darwin. Ed. by Sol Tax. Vol. I: The 285 Evolution of Life. University of Chicago Press, Chicago. Whillis, J. 1940 The development of synovial joints. J. Anat. Lond., 74: 277-283. Wood Jones, F. 1949 The Principles of Anatomy as seen in the Hand. Baillibre, Tindall and Cox, London.