ELSEVIER Human Movement Science 13 (1994) 557-568

Human poly-articular muscles: An anatomical comment

R.H. Rozendal

Faculty of Human Movement Sciences, Dept. of Functional Anatomy, Free University, Van der Boechorststraat 9, 1081 BTAmsterdam, The Netherlands

Abstract

Bi-articular muscles form a subclass of poly-articular muscles. Muscles in human limbs belonging to these classes are enumerated. For some of these muscles experimental evidence had been given that they direct forces on the environment and move segments of the limbs in a unique way. They perform the combination of these effects in cooperation with mono-articular antagonists. Such muscles are situated in the lower limb. Analyses of upper limb movements were less convincing. They were not complete. In the literature the kinematic studies were found from which a hypothesis about the functioning of the poly-articular muscles moving the fingers may be deduced. Expansion of an experiment of Long et al. (1970) with analysis of forces exerted may provide an interpretation of the cooperation of finger poly-articular muscles. The terms antagonist and synergist are rede- fined.

1. Introduction

In descriptive anatomy, bi-articular muscles are considered as a sub-class of poly-articular muscles. The latter differ from mono-articular muscles which span just one joint. Bi-articular CBA) muscles span two joints and poly-articular (PA) muscles span more than two joints. BA muscles have been studied for a long time (Van Ingen Schenau, 1990; Van Ingen Schenau et al., 1990). The anatomical relationship between points of attachment and joints spanned has defied anatomists to define its func- tional consequences. A classification of morphological as well as functional characteristics would be the ultimate goal of these endeavours (Andrews and Hay, 1983; Zajac, 1993). A functional classification may only emerge from research in which muscle functions are proven experimentally. The

0167.9457/94/$07.00 0 1994 Elsevier Science B.V. All rights reserved SSDI 0167-9457(94)00030-l

558 R.H. Rozendal /Human A4oc:ement Science 13 (1994) 557-568

ultimate goal of such a programme is to define the interaction between the nervous system and the so-called effector apparatus (Van Ingen Schenau, 1989; Van Ingen Schenau et al., 1992) as well as the interaction between the neuromuscular system and the environment (Hogan, 1990). Bernstein (1967) stated that in order to understand coordination (defined by him in terms of the abovementioned interactions), one had to study the anatomi- cal and physiological sources of indeterminacy. The existence of muscles like BA muscles and their antagonists spanning a certain joint is one of these indeterminacies, nowadays often called redundancy (see e.g. Gielen et al., 1990). As long as this research programme initiated by Bernstein (1967) has not been completed, a number of questions on coordination remain unanswered. Recent experimental work on BA lower limb muscles by Van Ingen Schenau and coworkers (1987, 1992) has led to the formula- tion of a theory in which so-called geometrical and anatomical constraints and coactivation of mono-articular and BA antagonists in solving conflicts between functional demands of direction of displacement and direction of force exertion on the environment could be made plausible (Van Ingen Schenau et al., 1992). Hogan (1990) stated that PA muscles couple joints by contributing to limb impedance in accordance with the task at hand. The mechanical impedance of multi-joint systems is conditioned by the geome- try and the synergetical activation of opposing muscles. Regulation of motion and manipulation of the environment may then result in minimiza- tion of energy dissipation without a complex burden on neural coordinative processes. Such theories presuppose an effective organization of the anatomical structure. Do these elegant theories linking morphological data on gross anatomy of muscles with function relate to a unique case or could they be generalized? For which kind of multi-joint systems could they be valid?

Landsmeer (1961) characterized multi-joint systems moved by the combi- nation of mono-articular and PA muscles. It appears that neither the number of segments nor the relative direction of rotation is of importance. Flexing systems as well as zig-zagging systems could be driven by those combinations of muscles, cooperating in moving the segments of the system relative to each other. He did not study the forces exerted on the environ- ment by such systems. From examples given by Van Ingen Schenau (1990) it is also clear that both types of systems are to be included: jumping vertically is performed by extension of a zig-zag system, moving objects with the hand over a table by a flexing system. Both systems will have joints characterized by their range of motion indicated as an anatomical con-

R.N. Rozendal /Human Movement Science 13 (1994) 557-568 559

straint by Van Ingen Schenau (1990). Both systems shorten and lengthen by rotating their segments. And in both systems muscles spanning the various joints must be able to solve conflicts between direction of force application on the environment and of movement of the segment. Or, in both systems force generated by muscles as well as gravitational and inertial effects will interact in a complex way such that PA muscles apparently “accelerate joints opposite to the torques they produce” (Zajac, 1993). He does not believe that explanations of neuromuscular function may be generalized without experimentation and inverse and direct dynamic modelling. One might be more optimistic about the possibilities of generalization on a more abstract level and phrase the questions to be answered.

One question pertains to the inclusion or exclusion of certain muscles in the class of BA muscles for which hypotheses could be tested following from the cited theories. Another question is if theories on BA muscles could also be extended to the larger anatomical class of poly-articular muscles. A third question is if the allocation of muscles to classes of antagonists and synergists remains feasible with a view to these developing theories. The first question is a problem of morphology or gross human anatomy. It will appear that the second question belongs to the same domain. However, an example of an experimental approach based on the literature will be given in order to judge the proof of the classification and its functional consequences. The third question is a problem of functional classification relating to the conditions in which activation of muscles is realized (Sherrington, 1905).

Differences in form and function between upper and lower limbs have been evolved related to the erect posture in the human species. Thus the first and second question have to be answered separately for arm and leg.

2. Anatomical classification problems, lower limb

In the leg the long flexors and extensors of the toes are undoubtedly PA muscles spanning more than two joints, which cooperate with short muscles in the foot, move the toes and help to apply force on the environment. The flexor and extensor system in the lower extremity resembles that in the upper of which the relation between form and function is more extensively analyzed. It will be noted here that the short muscles are by no means simple mono-articulars; they insert in aponeurotic ensembles, or their tendons run along more than one joint. Thus these short muscles are PA.

In the classification of BA muscles problems may be posed with respect

560 R.H. Rozendal /Human Movement Science 13 (1994) 557-568

to muscles like the human soleus muscle. It passes over a joint proximal and one distal of the talus. Talus is a bone without any attachment of muscles. Landsmeer (1961) studied the kinematics of such systems and concluded that the movement of the intercalated bone will be governed completely by the geometric relation of the joints and tendons. The unique terminal position of the bone is reached by equilibrium between muscular and ligamentous moments. Only a (hypothetical) muscle attached to the intercalated bone renders muscular governing of the position of that bone possible (Landsmeer, 1961). Talus cannot be positioned independently from the position of the bones of the leg and foot. Therefore the soleus muscle should be excluded from the class of BA muscles. It is a mono-artic- ular muscle.

The circumstances for which the theory was originally formulated by Van Ingen Shenau et al. (1987) could be characterized as a sagittal planar motion in which hamstring muscles and rectus femoris could be seen as antagonists of mono-articular hip flexors and knee extensors, respectively hip extensors and knee flexors. Some of the hamstring muscles are also hip abductors and exorotators, others are adductors and endorotators. Mono- articular adductors and endorotators, abductors and exorotators are pre- sent in the thigh. Provided the statements about the function of muscles are limited explicitly to well-analyzed planar movements, one could not find any a priori reason why these muscles in movements in such planes would be exempt from such a theory. Maybe the sartorius muscle is a case apart, needing hardware solutions in the nervous system completing the solutions in the hardware of the effector apparatus (Loeb and Richmond, 1992): its runs in a fascial envelope and has in the cat and possibly in man an internal structure resulting in forces transversal to the longitudinal axis of the limb. It is undoubtedly a BA muscle.

A rotatory function with respect to the leg when the knee is flexed is ascribed to hamstring muscles. As mono-articular antagonists are available the popliteus muscle and the short head of the biceps femoris muscle.

Muscles which are difficult to classify morphologically are the tensor fasciae latae muscle and that part of the gluteus maximus that inserts in the fascia lata, a reinforced strand of the fascial envelope of the thigh.

3. Classification problems, upper limb

The morphology of the human arm is characterized by the presence of more degrees of freedom than the leg. The shoulder girdle adds to this

R.H. Rozendal /Human Mouement Science 13 (1994) 557-568 561

(Prank, 1988, 1991). From a morphological viewpoint, muscles originating from the trunk and inserting in the arm pass the sternoclavicular, the acromioclavicular and shoulder joints. More functionally seen, muscles like the pectoralis major and latissimus dorsi pass the shoulder girdle and the shoulder joint. The girdle moves with respect to the trunk concomitantly in the sternoclavicular joint and the scapulothoracic gliding plane (Prank, 1991; Van der Helm, 1991), rendering both named muscles BA muscles (Veeger et al., 1991 a,b; Van der Helm et al., 1992). Consequently the muscles between trunk and girdle and those between girdle and arm are mono-articulars, some of them situated so that they could act antagonisti- cally to the BA muscles.

The long head of the triceps brachii muscle and both heads of the biceps brachii are BA muscles, originating from the scapula and inserting in the forearm. Mono-articular antagonists in the elbow joint are well known: brachialis, brachioradialis muscles may flex, the short heads of triceps muscle may extend this joint. The pronator teres muscle could act as an antagonist against the supinator muscle synergistic with the biceps muscle in supination of the forearm, while the triceps acts as an antagonist blocking flexion (Zuylen et al., 1988). In the forearm, muscles passing the wrist joint only have to be considered as mono-articulars on the same ground as the soleus muscle. A number of muscles named as prime movers of the wrist originate from the humerus and pass the elbow as well as the wrist joint; flexors and extensors of the wrist at the ulnar and at the radial side. Morphologically these are BA muscles.

The long flexors and extensors of the fingers are like those of the toes undoubtedly PA, spanning more than two joints. The morphology of these muscles as well as the kinematics have been studied by Landsmeer and various co-workers (Landsmeer, 1955, 1958; Spoor and Landsmeer, 1976). Most of these muscles are attached to the bones of the forearm. Their tendons pass the wrist and metacarpophalangeal (MP) as well as the interphalangeal (IP) joints before inserting in either the middle or the distal phalanx of the digit in question. The presence of a dorsally located aponeurotic extensor assembly (Landsmeer, 1955) complicates the formula- tion of hypotheses about the function of these muscles on the basis of morphology (see Fig. 1). They must cooperate with interosseus muscles which also attach to the fingers through this assembly. An essential property of the assembly is that extensor digitorum and interosseus muscles attach to it by medial bands which always remain at a constant distance from the centre of the proximal IP joint, passing dorsally to it, and lateral

562 R.H. Rozertdal/ Human Mocement Science I3 (1994) 557-568

Fig. 1. Schematic drawing of the main structures of a human finger. W = wrist joint; MP =

metacarpophalangeal joint; PIP, DIP = proximal and distal interphalangeal joint resp.; med = medial,

lat = lateral bands of extensor ( = ext) interosseus ( = int), and lumbrical ( = lumb) muscles; fl.p = deep

flexor muscle; fl.s = superficial flexor muscle. Note that the deep flexor perforates the superficial at the

level of the proximal phalanx. Flexor muscles are prevented from bowstringing by connective tissue

tunnels which are not illustrated.

bands which pass this joint laterally. Lumbrical muscles run between the tendons of the deep flexor and the extensor assembly in more or less the same way as interosseus muscles do. Thus interosseus and lumbrical muscles, taken together as intrinsic muscles of the finger, pass the MP joint ventrally and the IP joints dorsally as well as laterally. Moment arms for flexors and the dorsal part of the extensor assembly differ systematically, but the moment arm of the lateral bands may vary with the pull of the various muscles connected to the extensor assembly. Landsmeer (1955) already noted that finger systems like this are shortened by flexion due to the tension of the interosseus muscles. In patients with paralysis of the interossei the fingers shorten by zig-zagging, due to the unopposed action of the long flexors and extensors. Landsmeer (1961) stated that a third muscle is necessary for stabilizing biarticular systems with two BA muscles opposing each other. The interosseus muscles oppose the effects of the extensor muscles in the MP joints and diminish the effects of the lateral bands of the extensors. These bands couple the orderly flexion of the IP joints. With the help of a Lumbrical-minus model of a single finger the conditions of movement of this complicated system have recently been analyzed by Leijnse et al. (1992) on the basis of a displacement model in which the input is essentially morphological.

Thus, from a purely morphological view, the fingers are moved by a system of muscles that consists of at least four PA (flexor, extensor, lumbrical muscles and lateral band of interosseus muscle) and one neces-

R.N. Rozendal /Human Mol;ement Science 13 (1994) 557-568 563

sarily co-contracting BA-muscle (medial band of interosseus muscle). The interosseus may oppose the others and thereby stiffen the finger in any position.

4. Classification of muscles of the trunk

Anatomically most tracts of the erector spinae muscle and all abdominal muscles are PA muscles. The organization of the bones and joints of the trunk is characterized by quite different spatial proportions to those in the extremities. To consider the functional consequences would require analy- ses far beyond the scope of this paper.

5. Functional considerations

Analyses applying inverse dynamics of leg functions were published with regard to BA muscle function in vertical jumping (Bobbert and Van Ingen Schenau, 1988>, speed skating (De Koning et al., 19911, bicycling (Van Ingen Schenau et al., 1992) and sprint push-off (Jacobs and Van Ingen Schenau, 1992). A direct dynamics study of vertical jumping has been undertaken by Van Soest et al. (1993). Zajac and various co-workers also combined inverse and direct dynamic approaches in studying vertical jump- ing (Zajac, 1993). All movements studied are characterized such that they are in the plane of the zig-zagging limb system.

In the upper extremity matters are seemingly more complex than in the lower. Arm movements are frequently not conducted in one of the planes anatomists and medical people discern. In most analyses of function hardly anything is known about how the bones of the shoulder girdle move, and thus it is not certain if particular muscles act for example as shortening antagonists of BA muscles spanning either the shoulder girdle and shoul- der joint or the elbow and shoulder joint.

Veeger et al. (1991b) and Veeger et al. (1992) tried to explain the activity of biceps and triceps brachii muscles in pushing the handrim of a wheelchair ergometer. In this exercise a conflict between the direction of force exertion along the rim and the attempted motion in the elbow joint (extension) exists. They could not prove definitively that they had encoun- tered a case in which the theory of Van Ingen Schenau (1990) could be applied for the upper extremity as they did not register activity in mono-

564 R.H. Rozendal /Human Movement Science 13 (1994) 557-568

articular elbow muscles. In wheelchair propulsion large moments are developed by pectoralis major and anterior deltoid muscles. As long as the hand does grasp the handrim in front of the top dead centre, the elbow is forced to extend.

Also Gielen et al.‘s (1990) analysis of a conflict between extension and flexion moments of the elbow with a concomitant adduction of the elevated arm in the shoulder complex was only directed to BA and mono-articular elbow flexors. The possibility of participation of elbow extensors was assumed. The possibility that moment was developed by shoulder muscles other than biceps brachii had not been taken into consideration.

The long flexors and extensors of the fingers (taken together as extrin- sits) have been subjected to functional analysis by Landsmeer and various co+vorkers some decades ago. More proximally some of the extrinsic muscles of the fingers pass the elbow and wrist joints. Their function with respect to these joints is uncertain. It might be postural. Activation pat- terns will have to vary with the positions of these joints in order to keep the muscle lengths in rapport with the function with respect to the fingers. In the leg the position of the ankle and subtalar joint complexes will demand adaptation from flexors and extensors of the big and smaller toes in order to function adequately with respect to the toes. Up to now the analyses had been restricted to the functioning of extrinsics of the fingers in finger motion, while wrist and elbow were kept or imagined static.

As shown by Landsmeer (1955) on a theoretical basis and by Long (1968) experimentally, in clawing of the fingers (MP extension and IP flexion) as well as in full flexion the extrinsic muscles are all active. MP flexion with the fingers extended asks for activity of the intrinsic muscles, full extension of all joints is reached and held by activity of all muscles.

Regarding the extrinsic muscles of the fingers and their coordianation with the intrinsic muscles in applying force while the finger is moving an experiment has been reported by Long et al. (1970). EMG was registered. He asked his subjects to translate a nut, pinched between thumb and fingertips, along a rod. They had to move the nut against the pressure of a spring towards or away from the palm of the hand. Moving back they had to resist the spring. In purchasing against the spring the MP joint was flexed while the IP joints were extended. In pulling the spring the motion was reversed. They found that in both movements all extrinsic muscles were active. In moving the nut away from the palm interosseus and lumbrical muscles were active too, in moving towards the palm only interossei were active, notwithstanding the apparent dorsal flexion of the

R.H. Rozendal /Human Movement Science 13 (1994) 557-568 565

MP joint. Although the authors did not register the force exerted, it must have been deviated in direction from the direction of the movements. At least this must be a case were PA muscles act together in applying force and causing motions in different directions.

6. Antagonists or synergists?

From the cited examples on muscles of the fingers as well as from the analysis of bicycling by Van Ingen Schenau (1989, 1990) it follows that muscles classified as antagonists on the basis of morphology may be coactivated in certain movements. Lombard (1903) had already postulated that coactivation of BA muscles could extend a frog leg, given the geometry of their moment arms. The poly-articular muscles of the human fingers have a similar relationship. They give also rise to an orderly sequence in flexion which is indispensable for prehension (Landsmeer, 1955).

The classification as antagonists and synergists which since Sherrington (1905) proposed the terms, suggests specific conditions of activation in the relevant muscles, has no general validity over all movements caused by these muscles. It had been customary in anatomical textbooks to note that in securing certain postures of body parts concerted static actions of antagonists might be needed (Williams and Warwick, 1980). It is now clear that dynamic as well as static actions occur in which so-called antagonists cooperate closely. The terms antagonist and synergist appear to be useful only to indicate (groups of) muscles of specifically different positions with respect to mechanical axes of joints.

7. A hypothesis proposed

In the papers analyzing motion and concluding to these modes of cooperation the concept of constraint is frequently used (e.g. Van Ingen Schenau et al., 1987; 1992). Such use of words is not meant to indicate that for instance anatomical structures curb functional possibilities. Anatomical structures are a priori evolved to serve optimalized motion. Structures that have to shorten and lengthen may be constructed in a zig-zag fashion. The joints in such a construction will have a range of motion which need not pass the fully extended position. In this sense reaching the end of the range of motion will not be a constraint nor is the presence of a range of motion

566 R.H. Rozendal /Human Mouement Science I3 (1994) 557-568

in itself a constraint. The active and passive structures in such a limb will be designed to stop motion at that range that will not endanger the organism, because the species would not have survived otherwise.

This may not be valid for speed skating, bicycling or even vertical jumping for sport purposes, as these functions are culturally induced and only recently (in terms of evolution) developed. But the thesis that the structure of the arm and hand are fully adapted to prehension and do not constrain but favour this function is well accepted (Landsmeer, 1955; Long 1968; Williams and Warwick, 1980). It follows that muscles in such a system cannot be considered as redundant; they have evolved to position the hand that it executes prehensive tasks and complete these tasks. It is on this basis that the following hypothesis is put forward.

The hypothesis bears on the arm-hand system, designed for grasping. Such a system is multi-jointed and will consist of more segments than the leg, both for freedom of positioning the prehensile hand in space and for enclosing the objects to handle. For that latter purpose a flexing system like the fingers is better than a zig-zagging system. It consists of four segments and is served by PA muscles. The system works along the same principles as that in the leg for controlling force application and movement.

Experimental proof must be relatively easy to give of a cooperation of extrinsic flexors and extensors of the fingers with the various intrinsic muscles. The experiment of Long et al. (1970) has to be expanded with force registration. The hypothesis is that the intrinsic muscles redistribute the moments about the joints, produced in an orderly way by the extrinsic muscles. The order is given by the lengths of the moment arms and a constant activation proportion at any length related to either elbow and wrist movement or just one of these movements. The wrist or elbow will be fixed in the latter case. At the output side, the BA muscles spanning elbow and wrist will redistribute continuously the moments about these joints according to the demands of the task. At the input side, a parallel coactivation of flexors and extensors of elbow, wrist and fingers in the forearm is postulated. Depending on the direction of the force vector the intrinsic muscles will be activated in various patterns.

8. Conclusion

In human upper and lower limbs a class of BA muscles exist. They form a subclass of the larger class of PA muscles. For some BA muscles in the

R.H. Rozendal /Human Movement Science 13 (1994) 557-568 561

lower limb a cooperation has been proven in extending the zig-zag configu- ration of the limb. They work together with mono-articular muscles. Such systems have no redundant actuators.

The experimental proof for the existence of a similar cooperation of mono-articular and PA muscles in the arm and hand is far less convincing.

The hypothesis is put forward that PA muscles, especially those in the fingers and toes, cooperate in a specific way. A suggestion of an experiment to proof this is put forward.

The terms antagonist and synergist ought to refer only to the anatomical relationship between muscles and joints and not to the function of muscles.

References

Andrews, J.G. and J.G. Hay, 1983. Biomechanical considerations in the modelling of muscle functions.

Acta Morphologica Neerlando-Scandinavica 21, 199-223. Bernstein, N., 1967. The co-ordination and regulation of movements. Oxford: Pergamon. See also

Whiting (1984).

Bobbert, M.F. and G.J. van Ingen Schenau, 1988. Coordination of vertical jumping. Jounal of

Biomechanics 21, 241-262. De. Koning, J.J., G. de Groot and G.J. van Ingen Schenau, 1991. Coordination of leg muscles during

speed skating. Journal of Biomechanics 24, 137-146.

Gielen, S., G.J. van Ingen Schenau, T. Tax and M. Theeuwen, 1990. ‘The activation of mono- and

bi-articular muscles in multi-joint movements’. In: J.M. Winters and S.L.-Y. Woo (Eds.), Multiple

muscle systems: Biomechanics and movement organization pp. 302-311). New York: Springer

Verlag.

Hogan, N., 1990. ‘Mechanical impedance of single- and multi-articular systems’. In: J.M. Winters and

S.L.-Y. Woo (Eds.), Multiple muscle systems: Biomechanics and movement organization (pp.

149-164). New York: Springer Verlag.

Jacobs, R. and G.J. van Ingen Schenau, 1992. Intermuscular coordination in a sprint push-off. Journal

of Biomechanics 25, 953-965.

Landsmeer, J.M.F., 1955. Anatomical and functional investigations on the articulation of the human

fingers. Acta Anatomica (Supplementurn) 24, 2511-2569.

Landsmeer, J.M.F., 1958. A report on the coordination of the interphalangeal joints of the human finger and its disturbances. Acta Morphologica Neerlando-Scandinavica 2, 59-84.

Landsmeer, J.M.F., 1961. Studies in the anatomy of articulation. 1. The equilibrium of the intercalated

bone. II. Patterns of movement of bi-muscular, bi-articular systems. Acta Morphologica Neerlando-

Scandinavica 3, 287-303.

Leijnse, J.N.A.L., J.E. Bonte, J.M.F. Landsmeer, J.J. Kalker, J.C. van der Meulen and C.J. Snijders,

1992. Biomechanics of the finger with anatomical restrictions - The significance for the exercising hand of the musician. Journal of Biomechanics 25, 1253-1264.

Loeb, G.E. and F.J.R. Richmond, 1992. Archictural features of muscles with more than one action.

Eighth Meeting of the European Society of Biomechanics, Rome pp. 245-247. Lombard, W.P., 1903. The action of two-joint muscles. American Physical Education Reviews 9,

141-145.

568 R.H. Rozendal /Human Movement Science 13 (1994) 557-568

Long II, C., 1968. Intrinsic-extrinsic muscle control of the fingers. Journal of Bone & Joint Surgery

50A, 973-984.

Long II, C., P.W. Conrad, E.A. Hall and S.L. Furler, 1970. Intrinsic-extrinsic muscle control in power

grip and precision handling. An electromyographic study. Journal of Bone & Joint Surgery 52A, 853-867.

Pronk, G.M., 1988. ‘Three-dimensional determination of the position of the shoulder girdle during

humerus elevation’. In: G. de Groot, A.P. Hollander, P.A. Huijing and G.J. van Ingen Schenau (Eds.), Biomechanics XI. Amsterdam: Free University Press.

Pronk, G.M., 1991. The shoulder girdle: Analyzed and modelled kinematically. Delft: Delft University

of Technology; Faculty of Mechanical Engineering and Marine Technology.

Sherrington, C.S., 1905. On reciprocal innervation of antagonistic muscles. Proceedings of the Royal

Society B76, 160-163.

Spoor, C.W. and J.M.F. Landsmeer, 1976. Analysis of the zig-zag movement of the human finger under

the influence of the extensor digitorum tendon and the deep flexor. Journal of Biomechanics 9, 561-566.

Van der Helm, F.C.T., 1991. The shoulder mechanism: A dynamic approach. Delft: Delft University of

Technology, Faculty of Mechanical Engineering and Marine Technology.

Van der Helm, F.C.T., H.E.J. Veeger, G.M. Pronk, L.H.V. van der Woude and R.J. Rozendal, 1992.

Geometry parameters for musculoskeletal modelling of the shoulder system. Journal of Biomechan-

its 25, 129-144.

Van Ingen Schenau, G.J., 1989. From translation to rotation: Implications for theory of motor control.

Human Movement Science 8, 423-442.

Van Ingen Schenau, G.J., 1990. On the action of bi-articular muscles, a review. Netherlands Journal of

Zoology 40, 521-540.

Van Ingen Schenau, G.J., M.F. Bobbert and R.H. Rozendal, 1987. The unique action of bi-articular

muscles in complex movements. Journal of Anatomy (London) 155, l-5.

Van Ingen Schenau, G.J., M.F. Bobbert and A.J. van Soest, 1990. ‘The unique action of bi-articular

muscles in leg extensions’. In: J.M. Winters and S.L.-Y. Woo (Eds.1, Multiple muscle systems:

Biomechanics and movement organization (pp. 639-652). New York: Springer Verlag.

Van Ingen Schenau, G.J., P.J.M. Boots, G. de Groot, R.J. Snackers and W.W.L.M. van Woensel, 1992. The constrained control of force and position in multi-joint movements. Neuroscience 46, 197-207.

Van Soest, A.J., A.L. Schwab, M.F. Bobbert and G.J. van Ingen Schenau, 1993. The influence of biarticularity of the gastrocnemius muscle on vertical jumping performance; A simulation study.

Journal of Biomechanics 26, l-8. Van Zuylen, E.J., C.C.A.M. Gielen and J.J. Denier van der Gon, 1988. Coordination and inhomoge-

neous activation of human arm muscles during isometric torques. Journal of Neurophysiology 60,

1523-1548. Veeger, H.E.J., F.C.T. van der Helm, L.H.V. van der Woude, G.M. Pronk and R.H. Rozendal, 1991a.

Inertia and muscle contraction parameters for musculoskeletal modelling of the shoulder mecha-

nism. Journal of Biomechanics 24, 615-629.

Veeger, H.E.J., L.H.V. van der Woude and R.H. Rozendal, 1991b. Within cycle characteristics of the wheelchair push in sprinting on a wheelchair ergometer. Medical Science in Sports and Exercise 23,

264-271. Veeger, H.E.J., L.H.V. van der Woude and R.H. Rozendal, 1992. Load on the upper extremity in

manual wheelchair propulsion. Journal of Electromyorgraphy and Kinesiology 1, 270-280. Whiting, H.T.A. (Ed.), 1984. Human motor actions, Bernstein reassessed. (Includes text of Bernstein,

1967.) Amsterdam: North-Holland.

Williams, P.L. and R. Warwick, 1980. Gray’s anatomy (36th ed.) Edinburgh: Churchill-Livingstone. Zajac, F.E., 1993. Muscle coordination of movement: A perspective. Journal Biomechanics 26, Suppl 1,

109-124.