J. Anat. (1977), 124, 3, pp. 627-632 627 With 2 figures Printed in Great Britain The coraco-acromial ligament and projection index in man and other anthropoid primates

RUSSELL L. CIOCHON AND ROBERT S. CORRUCCINI Department ofAnthropology, University of California, Berkeley, CA 94720 U.S.A. and Division ofPhysical Anthropology, Smithsonian Institution, Washington, D.C. 20560 U.S.A. (Accepted 11 October 1976)

INTRODUCTION Studies on the functional anatomy of the primate shoulder (Keith, 1894, 1923; Miller, 1932; Ashton & Oxnard, 1963; Washburn, 1968; Morbeck, 1972; Corruccini & Ciochon, 1976) have shown a number of specialized morphological features in Hominoidea which differ markedly from the presumed ancestral pattern typified by the condition in cercopithecoid monkeys. These differences have been correlated with a radical shift in the locomotor/feeding adaptations of hominoid primates with which the term 'brachiation' has been associated by many authors in the past. One particular, heretofore relatively neglected, anatomical feature of the shoulder of hominoids is the presence of a strong triangular ligament extending from the tip of the costal surface of the (lateral to the clavicular facet) to the whole postero-superior border of the . The coraco-acromial ligament acts with the acromion and coracoid process to limit superior movement of, and to serve as a protective cuff for, the head of the ; it also separates the deltoid and supraspinatus muscles which abduct the humerus. A homologue of this ligament has never been detected in lower anthropoid primates (i.e. cercopithecoid and ceboid monkeys), raising the possibility that the ligament is unique to the Hominoidea (including, of course, man). The objects of the present study are to present a few new observations on the comparative presence or absence of the coraco-acromial ligament, and to evaluate the functional significance and variance of an associated mensural trait, the coraco- acromical projection index.

MATERIALS AND METHODS Visual observation of a large series of prepared hominoid, cercopithecoid and ceboid scapulae revealed that the coracoid and acromion processes of the Homi- noidea are relatively the largest and project far laterally from the plane of the glenoid cavity. The enlarged acromion in hominoids is correlated with a much thickened deltoid muscle which originates from it and is inserted far distally on the humerus. Pectoralis minor is inserted on the enlarged laterally projecting coracoid process rather than on the humerus, as is the case in the lower anthropoid primates. Together these processes provide both an advantageous biomechanical arrangement for deltoid and trapezius and an anchorage for the strong coraco-acromial ligament. 628 R. L. CIOCHON AND R. S. CORRUCCINI In particular, deltoid is afforded a powerful angle of pull during abduction without the possibility of its dislocating the humerus in a vertical direction. Along with the modification of these scapular processes in the Hominoidea, changes in the glenoid cavity have also occurred: thus hominoids have a relatively wide and flat glenoid cavity, with a non-projecting . In hominoids also the humeral head is well-rounded and expanded, relative to the slender shaft, in the antero-posterior dimension. These modifications, when coupled with the above-mentioned muscular rearrangements, allow a much greater range of

0,0

/ ~~~~~~~~~~~4,

Fig. 1. Pictorial representation of the coraco-acromial projection viewed from the superior aspect of the . The scapula of Homo sapiens is utilized here to depict the hominoid condition. circumduction, with the ability to move and reach in all directions greatly enhanced. Conversely, the Cercopithecoidea and most of the Ceboidea are characterized by non-projecting coracoid and acromion processes and absence of a coraco-acromial ligament; a narrow glenoid cavity which is concave, with a projecting supraglenoid tubercle; and a humeral head which is much elongated in the medio-lateral dimen- Coraco-acromial projection index 629 sion.* This morphology reflects a much more fixed and stable gleno-humeral joint than in hominoids, with movement restricted to a partial arc of flexion and extension. In order to quantify the observed qualitative differences in the shoulders of hominoids, cercopithecoids and ceboids we have devised the following interval scale measurement, termed the coraco-acromial projection, which is defined as the minimum perpendicular distance from a line connecting the most lateral points on the coracoid process and the acromion to the most lateral projection of the supra- glenoid tubercle. This is recorded as a negative value in cases where the supraglenoid tubercle projects further laterally than either the coracoid or the acromion. Figure 1 illustrates this measurement. In order to cancel out some of the effects of body size on the measurement's gross magnitude we found it necessary to employ a ratio of the value. The height of the glenoid cavity is used as a divisor in this ratio and is defined as the distance between the most superior extension of the glenoid articular surface at the supraglenoid tubercle and the most inferior extent of the articular surface immediately superior to the infraglenoid tubercle taken at the glenoid cavity midline. These measurements were made on 433 scapulae representing 20 species of Anthropoidea. The specimens were studied at 15 museums in Great Britain, Europe and North America. There were five species of Hominoidea, eight species of Cercopithecoidea and seven species of Ceboidea (see Table 1). We selected these species because they exhibit different locomotor-behavioural repertoires and because they represent the major taxonomic groupings within the Anthropoidea. Approxi- mately equal numbers of each sex were obtained for each species in order to balance any possible sample skewing effects due to sexual dimorphism. Approximately 90 % of the measured specimens were caught in the wild. The remaining 10 % were from zoological gardens and laboratories because of the lack of wild-caught representa- tives of certain species. All specimens in the study were examined for disease, healed fractures, gunshot wounds, arthritic joint surfaces and other abnormalities. Only adult specimens (as judged by fully fused epiphyses) lacking obvious pathology were selected.

RESULTS Table 1 summarizes the metric results of the coraco-acromial projection, glenoid height and coraco-acromial projection index measurements for all the species of Anthropoidea in our comparative sample. These basic statistical results are fully compatible with additional observations on about 600 prepared scapulae and some wet dissections of new taxa. Figure 2 displays the metric results presented in Table 1. A complete separation between hominoids and cercopithecoids is evident. There is almost as much variance between these two groups as there is within each of them. The ceboid monkeys, however, appear to be much more generalized. The outer limit of the ranges of their coraco-acromial indices overlap with both the hominoids and the cercopithecoids. Except for the howler monkey, Alouatta, ceboids uniformly display much less intragroup variance than either the Hominoidea or the Cercopithecoidea. Alouatta's

* All topographical references used in this study refer to the standard human anatomical position. All non-human specimens were placed in this position before measurements were made. In addition, all descriptive anatomical nomenclature used here conforms with the terminology adopted in Nomina Anatomica 3rd edition (1966). 630 R. L. CIOCHON AND R. S. CORRUCCINI

Table 1. Coraco-acromial projection and index in selected Anthropoidea Coraco-acromial projection Glenoid height Index Taxon (N) Mean S.D. Mean S.D. Mean S.D. S.E. Hominoidea Homo sapiens (26) 19-42 2-72 35-43 3-45 0 55 0-07 0-027 Pan troglodytes (23) 20-88 1-90 32-91 2-62 0-64 0 09 0-038 Gorilla gorilla (22) 31-32 6-73 50-19 6-84 0-62 0-08 0 034 Pongo pygmaeus (24) 17-22 2-62 37-11 5 09 0 47 0-08 0 033 Hylobates lar (22) 12-60 0 97 14-16 0-98 0-89 0 09 0-038 Cercopithecoidea Colobus guereza (22) 3-47 0-66 15-93 1-17 0-22 0-04 0-017 Nasalis larvatus (21) 5-14 1-07 20-24 2-78 0-25 0-05 0-022 Presbytis rubicunda (22) 3-11 0 35 13-55 0-66 0-23 0-02 0 009 Cercopithecus mitis (22) 2-43 0-75 14-95 1 54 0-16 0 05 0-021 Erythrocebus patas (14) 1-71 0-54 17-04 2-06 0-10 0-04 0-021 Macaca fascicularis (21) 1-87 0 59 12-84 1-45 0-14 0-04 0-017 Papio cynocephalus (22) -0-18 1-23 24-45 3-54 -0-01 0-05 0-021 Theropithecus gelada (24) -1-30 0-62 21-25 2-26 -0-06 0 03 0-012 Ceboidea Ateles paniscus (23) 9-62 1 19 15-96 1-26 060 006 0025 Lagothrix lagothrica (21) 6-51 0-81 16-22 1 22 0 40 0 05 0-022 Alouatta villosa (24) 4-42 0-94 16-06 1-39 0-28 0 05 0-020 Cacajao calvus (15) 6-67 1 31 13 34 1-34 0 50 0-07 0-036 Pithecia pithecia (18) 4-66 0-76 10 09 0.99 0-46 0 07 0 033 Cebus apella (24) 4-78 1-05 11-72 1-12 0 41 0 07 0-029 Saimiri sciureus (23) 2-88 0-41 7-47 0-42 0-39 0-06 0-025

Hor7o sapiens a Pan troglodytes m GorY/Ja gorilla - Pongo pygmaeus Hylobates lar Colobtus guiereza a Nasalis larvatus a Presbytis rujbicujnda a Cercopithecus tnitis Erythrocebus patas a Macaca fascictularis a Papio cynocephalus n TheropithecLus gelada Ate/es paniscuis ± Lagothrix lagothrica w Alonatta vil/osa Cacajao calvuJs Pithecia pithecia Cebus ape/la Saimiri sciureus I I I I I 1 --0 40 -020 0 0 20 0 40 0 60 0 80 1 00 1 20 Coraco-acromial projection indcx

Fig. 2. Graphic representation of the mean, two standard errors of the mean, one standard deviation, and total range, for the coraco-acromial projection index in selected Anthropoidea. Coraco-acromial projection index 631 somewhat unique position can be readily explained by the fact that it possesses a long but downward projecting coracoid process, which makes the coraco-acromial projection and therefore the index significantly lower. The position of the spider monkey, Ateles, is also of interest since it converges significantly on the hominoid condition. The ranges overlap in all cases except with that of the gibbon, Hylobates (Fig. 2). The functional anatomical interpretation of the results presented in Figure 2 correlate with the general fact that hominoid shoulder joints are much more mobile than cercopithecoid shoulder joints, and most ceboid shoulder joints are signi- ficantly more mobile than Old World monkey shoulders.

DISCUSSION High values of the coraco-acromial projection index are generally associated with the presence of a coraco-acromial ligament. The locomotor and anatomical con- vergence of Ateles on the Hominoidea raised the possibility of their possessing a coraco-acromial ligament. From observations made on five embalmed ceboid monkeys (M. Schon & L. Ziemer, personal communication), as well as on many semi-macerated scapulae retaining ligamentous tissue, it was clear that (a) the convergent New World monkeys (Ateles, Lagothrix, Cacajao) do not have coraco- acromial ligaments and, (b) Hylobates, the most primitive hominoid in many respects, does not retain the primitive condition in the shoulder joint but shares with hominoids a well-developed coraco-acromial ligament. It is thus confirmed that this ligament is indeed a uniquely specialized, derived character in hominoids, as had previously been suspected. The occurrence of high coraco-acromial projection indices in some ceboid primates without the existence of a corresponding coraco-acromial ligament might be explained by certain basic difference in the locomotor-behavioural repertoire of hominoids and ceboids. In particular, Ateles has a prehensile tail which, as is well known, serves as a fifth limb. This animal spends much of its time suspended by its tail and one or two limbs. This form of support behaviour in Ateles might well preclude the need for a mechanism limiting the superior range of movement of the humerus. The morphological peculiarities of Ateles may also reflect the fact that the force pattern impressed upon the shoulder in this species is predominantly tensile (Ashton, personal communication). Man corresponds closely to the anthropoid apes, and differs sharply from the non-hominoids, in his possession of a strong coraco-acromial ligament and in his shoulder morphology generally. This adds one more derived trait to the long list of traits which indicate a unified ancestry for, and close relationship between, the living hominoids.

SUMMARY The coraco-acromial ligament in man is a trait shared only with other hominoids (apes) among anthropoid primates. The associated form of the coracoid process and acromion, their lateral projections and the shape of the glenoid cavity likewise distinguish the Hominoidea. These anatomical features add credence to the view that the living hominoids share an independent ancestry within the Anthropoidea associated with the development of several unique locomotor/feeding adaptations. 632 R. L. CIOCHON AND R. S. CORRUCCINI This research was supported by the National Science Foundation (Grant GS- 31943X) and the National Institute of General Medical Sciences (Training Grant GM-1224). We owe a special debt of gratitude to S. L. Washburn for providing the initial impetus for the project. We also wish to thank N. T. Boaz, W. E. Meikle and H. H. Srebnik for commenting on the manuscript. The authors, however, assume full responsibility for its content. Curatorial assistance and access to specimens has been provided by the following: S. Anderson, F. Baud, L. de la Torre, A. M. Husson, D. C. Johanson, S. Kortlucke, B. Lawrence, X. Misonne, P. Napier, R. Orr, F. Petter, H. Schaefer, A. H. Schultz, P. Starck and R. Thorington. We thank J. K. Andreasen and M. B. Schwartz for help with the Calcomp Plot routine.

REFERENCES ASHTON, E. H. & OXNARD, C. E. (1963). The musculature of the primate shoulder. Transactions of the Zoological Society ofLondon 29, 553-650. CORRUCCINI, R. S. & CIOCHON, R. L. (1976). Morphometric affinities of the human shoulder. American Journal ofPhysical Anthropology 45, 19-38. KEITH, A. (1894). The myology of the Catarrhini - a study in evolution. Ph.D. Dissertation, University of Aberdeen. KEITH, A. (1923). Man's posture. British Medical Journal 1, 451-454. MILLER, R. A. (1932). Evolution of the pectoral girdle and forelimb in the primates. American Journal ofPhysical Anthropology 17, 1-56. MORBECK, M. E. (1972). A re-examination of the forelimb of the Miocene Hominoidea. Ph.D. Dis- sertation, University of California, Berkeley. WASHBURN, S. L. (1968). The Study of Human Evolution. Eugene: University of Oregon Press.