THE ANATOMICAL RECORD 248:456463 (1997)

Distribution of Primary Motor Nerve Branches and Terminal Nerve Entry Points to the Forearm Muscles
JIE LIU, ROBERT W.H. PHO,* BARRY P. PEREIRA, HUI-KING LAU, AND V. PREM KUMAR Department of Orthopaedic Surgery, National University of Singapore, Singapore

ABSTRACT Background: The information available on innervation pattern of the human forearm muscles in standard anatomy texts, although adequate for routine procedures, is not detailed enough for surgical reconstruction in complex injuries of the limb and for paralytic conditions of the forearm from peripheral nerve and spinal cord injuries. Methods: The innervation pattern in 10 cadaveric forearms was studied. The contributions of the main nerve trunks to each forearm muscle was examined. The location and number of the primary motor nerve branching points and of the terminal nerve entry points to each muscle were investigated. The location of both the primary nerve branching points and terminal nerve entry points was presented as a percentage of forearm length measured from the lateral humeral epicondyle to the radial styloid. Results: Seven of 19 forearm muscles were innervated from a single branch from the main nerve trunk. The remaining 12 received more than one primary branch. Two of 19 forearm muscles had only one terminal nerve entry point. The others had two or more each. In 13 of 19 forearm muscles, the statistical median location of the primary motor nerve branching points was within the proximal one-third of the forearm length and either more proximally or distally for the remainder. The statistical median location of the terminal nerve entry points was within the proximal one-third in 9 forearm muscles and within the middle one-third of the forearm in 8 forearm muscles. In two, it was located proximal to the elbow and in the distal one-third of the forearm, respectively. Conclusions: In lacerations across the forearm, where main nerve trunks are divided, mere repair of the nerve trunks would not address the denervation of muscle or segments of muscle by the division of the primary (or secondary) nerve branches traversing the wound and which took origin proximal to the laceration either from the divided nerve trunks or from other undamaged nerve trunks. Although the main nerve trunks may be intact, segmental crush injuries will defunction muscles by direct muscle damage or by damage to the terminal nerve entry points to the muscle. Knowledge of the location of the nerve branches and the terminal nerve entry points facilitates the insertion of electrodes at the motor points of forearm muscles for functional electrical stimulation in upper motor neuron lesions. The information in this study may also be usefully applied in selective denervation procedures to balance muscles in spastic upper limbs. Anat. Rec. 248:456463, 1997. r 1997 Wiley-Liss, Inc. Key words: innervation pattern; primary nerve branches; terminal nerve entry points; forearm muscles The detailed innervation pattern of the human forearm muscles, which is lacking in standard anatomy texts (Mortensen and Pettersen, 1966; Hollinshead, 1983; Williams et al., 1989), has been supplemented by the works of Brash (1955), Hollinshead and Markee (1946), Linell (1921), Reid (1920), Seddon et al. (1943), and Sunderland (1968). However, practicing surgeons
r 1997 WILEY-LISS, INC.
Contract grant sponsor: National Medical Research Council; Contract grant number: NMRC/96/0066; contract grant sponsor: Shaw Foundation; contract grant number: GR05988N; contract grant sponsor: National University of Singapore; contract grant number: RP950330. *Correspondence to: Robert W.H. Pho, Department of Orthopaedic Surgery, National University of Singapore, 5 Lower Kent Ridge Road, Singapore 119260. Received 4 November 1996; accepted 10 February 1997.

DISTRIBUTION OF NERVE BRANCHES TO FOREARM MUSCLES

457

still face great variations in the locations of branching points from the nerve trunks and terminal nerve entry points, which can be frustrating when they attempt to locate motor points for insertion of electrodes for functional electrical stimulation. Denervation procedures (Garland et al., 1980; Mertens and Sindou, 1991) for spastic muscles may be confounded by the variation in the location of the branching points in different patients and even in different limbs in the same patient. Assessment of any remaining undenervated and undamaged muscles in clean lacerations of the main nerve trunks in the forearm or in segmental crush injuries of the forearm is also difficult if knowledge of the branching pattern and its variation to the various muscles is lacking. Subsequent reconstruction is also difficult. Our purpose is to present succinctly the innervation pattern of the forearm muscles with emphasis on the number of nerve branches and terminal nerve entry points to each muscle and the range of location of these anatomical and surgically important entities with respect to xed anatomical landmarks in different cadaveric specimens.
MATERIALS AND METHODS

as the primary nerve branching point. Secondary branches were dened as the subdivisions of the primary branch into two or more branches. The number of the branches into the muscle belly and the locations of their entry points into the epimysium were also recorded. These branches were referred to as the terminal nerve branches, and the entry points into the epimysium were referred to as the terminal nerve entry points (Liu et al., 1994, 1995). All measurements were done in situ by using a vernier caliper (Mitutoyu, Japan). The number of primary motor nerve branches to the various muscles and the number of terminal nerve entry points are reported as the mean with one standard deviation based on 10 samples for each muscle. The distribution of the locations is reported as a statistical median location along the length of the forearm. The location is expressed as a percentage of the forearm length, which was dened as the distance between the lateral humeral epicondyle and the styloid process of the radius (Fig. 1). Locations proximal to the lateral epicondyle were expressed as negative values.
RESULTS

Ten fresh upper limbs from adult human cadavers of Asian descent were used in this study. In all, the cause of death did not affect the muscles or nerves in the forearm. During dissection, the limbs were extended at the elbow, with the wrist in the neutral position. Midline volar and dorsal incisions were made from above the elbow to below the wrist. The main trunks of the median, ulnar, and radial nerves and the anterior interosseous and posterior interosseous nerves were exposed. Under a 332 magnication operating microscope (Carl Zeiss, Germany), the nerve trunks and their branches were individually followed distally to the terminal nerve entry points in the forearm muscles. Nineteen forearm muscles (8 exors and 11 extensors) were studied in each limb. The number of motor nerve branches arising from the main nerve trunk to the individual muscles and the locations of these branching points were recorded. These branches were referred to as the primary nerve branches and the branching point
Abbreviations AIN APL BR ECRB ECRL ECU EDC EDM EI EPB EPL FCR FCU FDP FDS FPL MN PIN PL PQ PT SUP UN anterior interosseous nerve abductor pollicis longus brachioradialis extensor carpi radialis brevis extensor carpi radialis longus extensor carpi ulnaris extensor digitorum communis extensor digiti minimi extensor indicis extensor pollicis brevis extensor pollicis longus exor carpi radialis exor carpi ulnaris exor digitorum profundus exor digitorum sublimis exor pollicis longus median nerve posterior interosseous nerve palmaris longus pronator quadratus pronator teres supinator ulnar nerve

A total of 190 forearm muscles, 50 forearm main nerve trunks, 289 primary motor nerve branches with their branching points from the main nerve trunks, and 775 terminal nerve branches with their entry points were observed.
Contributions From the Nerve Trunks to Each Muscle

The main nerve contribution to each muscle is shown in Table 1. Most muscles, with the exception of the exor digitorum sublimis (FDS), exor digitorum profundus (FDP), and extensor carpi radialis brevis (ECRB) were innervated by branches arising from one nerve trunk. In the case of FDS, innervation was from the median nerve alone in 9 specimens (90%); in 1 specimen (10%), there was an additional innervation from the anterior interosseous nerve. For the FDP muscle, 8 specimens (80%) were innervated by both the anterior interosseous nerve and the ulnar nerve, and 2 (20%) were innervated solely by the anterior interosseous nerve. Where the FDP was innervated by both nerves, the anterior interosseous nerve was noted to give more primary branches and terminal nerve entry points than the ulnar nerve (average ratio 5 3:1; Fig. 2). For the ECRB, 9 specimens (90%) were innervated by the posterior interosseous nerve and 1 (10%) by both the posterior interosseous and the supericial branch of the radial nerve.
Number of Primary Nerve Branches and Terminal Nerve Entry Points to Muscles

The mean number of primary branches and terminal nerve entry points to each muscle is also summarized in Table 1. Three exors (exor carpi radialis [FCR], palmaris longus [PL], and pronator quadratus [PQ]) had only one primary branch, and the remaining 5 exors (pronator teres [PT], FDS, FDP, exor pollicis longus [FPL], and exor carpi ulnaris [FCU]) each received a mean of two or more primary branches. The PL shared its primary nerve branch from the median nerve with the FDS (Fig.

458

J. LIU ET AL.

Fig. 1. Measurement details. The distances from the lateral epicondyle of the humerus to the primary motor nerve branching points and the terminal nerve entry points were recorded for each individual muscle. These distances were expressed as a percentage of the length of the forearm, which is the distance between the lateral epicondyle and the styloid process of the radius. Details for FPL are given.

3). This sharing was also seen in one specimen where both FDP and FPL shared the primary branch from the anterior interosseous nerve. In addition, the secondary nerve branch to PL always passed through the muscle belly of FDS before innervating PL (Fig. 3). Four extensors (extensor digiti minimi [EDM], extensor carpi ulnaris [ECU], extensor pollicis brevis [EPB], and extensor indicis [EI]) consistantly had only one primary branch, whereas the other 7 (brachioradialis [BR], extensor carpi radialis longus [ECRL], ECRB, supinator [SUP], extensor digitorum communis [EDC], aabductor pollicis longus [APL], and extensor pollicis longus [EPL]) may have had one, two, or three branches, depending on the specimen. Only the SUP displayed a mean of two or more branches. In about 50% of the specimens, EDM and EDC, ECU and EDC, and EPL and EI were found to receive the secondary branches from the same primary branch from the posterior interosseous nerve. Similarly, occurrence of two muscles being innervated from the same primary branch was seen in BR and ECRL, ECRL and ECRB, APL and EPB, and EPB and EPL. EI in one specimen, besides sharing

its primary branch with EPL, had its secondary branch rst passing through the EPL muscle belly before innervating the EI. Fourteen of 19 forearm muscles (7 exors and 7 extensors) each had two or more terminal nerve entry points. EI and EPB had only one entry point, and the number of terminal entry points on PL (Fig. 3), EPL, and EDM was between one and two.
Distribution of the Primary Nerve Branching Points and Terminal Nerve Entry Points

The distribution of the primary nerve branching points and terminal nerve entry points in the forearm is schematically represented in Figure 4. The location of each point was expressed as a percentage of the forearm length. In our series, the mean forearm length was 260.0 6 16 mm. The primary nerve branching points in all the BR and ECRL specimens were located proximal to the transverse plane of the lateral epicondyle. Three of 10 specimens of PT and 1 of 10 ECRB also had branching points in this region. All FCR, PL, FCU, SUP, and EDM

DISTRIBUTION OF NERVE BRANCHES TO FOREARM MUSCLES

459

TABLE 1. Contribution of the main nerve trunks to each muscle and the number (mean 6 S.D.) of primary motor nerve branches and terminal nerve entry points Number of primary motor nerve branches 2.0 6 0.67 1.0 1.0 4.2 6 1.23 3.7 6 1.49 2.0 6 0.82 1.0 2.0 6 0.94 1.2 6 0.42 1.2 6 0.42 1.1 6 0.32 2.1 6 1.20 1.1 6 0.32 1.0 1.0 1.1 6 0.32 1.0 1.4 6 0.52 1.0 Number of terminal nerve entry points 3.4 6 0.70 3.1 6 1.10 1.1 6 0.33 5.6 6 1.65 5.8 6 1.23 3.0 6 0.67 2.4 6 0.52 2.5 6 0.71 2.4 6 0.70 3.3 6 0.82 2.5 6 0.71 3.1 6 0.35 4.3 6 0.67 1.2 6 0.67 2.89 6 0.78 2.0 1.0 1.7 6 0.67 1.0

Muscle Flexors PT FCR PL FDS FDP FPL PQ FCU Extensors BR ECRL ECRB SUP EDC EDM ECU APL EPB EPL EI

Main nerve contribution MN MN MN MN (90%); MN and AIN (10%) AIN and UN (80%); AIN (20%) AIN AIN UN Radial nerve Radial nerve PIN (90%); PIN and RN (10%) PIN PIN PIN PIN PIN PIN PIN PIN

within the middle and distal one-thirds, whereas those of EDM were found only within middle one-third and PQ only within the distal one-third. Muscle from 9 forearms had the statistical median occurrence of the terminal nerve entry points located within the proximal one-third of the forearm and 8 had theirs located in the middle one-third; BR and PQ had theirs proximal to the elbow and in the distal one-third of the forearm, respectively.
DISCUSSION

specimens had all their branching points located within the proximal one-third of the forearm, whereas FDP, FPL, EDC, ECU, APL, EBP, EPL, and EI specimens had their branching points within the proximal and middle one-thirds of the forearm. Two specimens of FDS also had their branching points located in the distal one-third. All PQ specimens had their branching points in the distal one-third. Most forearm muscles, with exception of PQ, EPL, EPB, and EI, had the statistical median occurrence of the branching points located within the proximal onethird. Although most of the muscles had their primary nerve branching points more proximally located than the corresponding terminal nerve entry points, EDC had some of its primary nerve branching points more distally located than the corresponding nerve entry points. Five BR specimens (50%) had all of their terminal entry points located proximal to the transverse plane of the lateral epicondyle and the remaining 5 specimens (50%) had entry points both proximal to the epicondyle and in the proximal third region. Seven ECRL specimens (70%) had their entry points scattered in the proximal one-third of the forearm, and the other three (30%) had their nerve entry points located proximal to the lateral epicondyle. PT and SUP had all their entry points within the proximal one-third of the length of forearm, and FCR, PL, FPL, FCU, ECRB, EDC, ECU, and APL had entry points within both the proximal and middle one-thirds. However, FDS and FDP had terminal entry points scattered in all three regions of the forearm. Entry points of EPB, EPL, and EI were found

The innervation pattern of the forearm muscles presented in this study, especially the number (Table 1) and the location of the primary motor nerve branching points, and the terminal nerve entry points (Fig. 4) will have clinical application for clinicians and perhaps will complement previous works (Brash, 1955; Hollinshead and Markee, 1946; Linell, 1921; Reid, 1920; Seddon et al., 1943; Sunderland, 1968). Sunderland (1968) discussed details of the motor nerve branching points from the main nerve trunks and the length of the branches to their entry points from a nerve regeneration point of view. Reid (1920) fastidiously located the surface marking of the terminal nerve entry points for EMG analysis. Linell (1921) reported a constant relationship between the length of forearm and the distance of motor nerve entry points from the lateral epicondyle and obtained constant gures for each muscle based on an average limb length. However, considerable variations in locations of primary nerve branching points and terminal nerve entry points can occur, as shown in our study, and even between the two sides of the same subject. Thus, the locations expressed as a range with a statistical median may be more useful. With regard to the contribution of the main nerve trunks to muscles, the data generally agreed with those of previous reports (Linell, 1921; Seddon et al., 1943; Mortensen and Pettersen, 1966; Sunderland, 1968; Hollinshead, 1983; Spinner, 1984; Tountas and Bergman, 1993; Williams et al., 1989). There were, however, interesting variations in the innervation of FDP, ECRB, and SUP. In general, FDP received nerve supply from both the ulnar nerve and the median/anterior interosseous nerve (some investigators did not differentiate these two nerves), as in 8 of our specimens (80%). Sunderland (1968) showed that this double innervation pattern to the muscle was quite constant in 19 of 20 (95%) cadaveric limbs and in 37 of 38 (97%) clinical cases. Furthermore, his clinical observation also suggested that the median nerve innervation is responsible for exing the radial two ngers and the ulnar innervation for the ulnar two ngers in 50% of cases and that different portions of the muscle power different ngers in the other 50%. Brand and Hollister (1993) documented that FDP is composed of four individual muscle bellies to the four ngers, but there is no report on the innervation to each belly. Based on their description, it is easy to understand that the innervation to FDP may be very complex and have more than one pattern. ECRB was generally understood to be innervated by the posterior interosseous nerve (Linell, 1921; Cricenti

460

J. LIU ET AL.

Fig. 2. FDP innervated by the anterior interosseous nerve, which gives off three primary motor nerve branches (white arrows) to the FDP; the ulnar nerve gives off one (black arrow). The anterior interosseous nerve gives off ve terminal nerve entry points (white arrowheads); the ulnar nerve gives off only two (black arrowheads).

et al., 1994). However, the main radial nerve or the supercial branch of the radial nerve (Brand and Hollister, 1993) may also innervate it in different specimens. One of 10 specimens in this study received a branch from the supercial radial nerve. Although Sunderland (1968) reported that 20% of SUP in his series had their nerve branches from the radial nerve, this was not observed in the present study (all of these were innervated by the posterior interosseous nerve) nor in other reports (Mortensen and Pettersen, 1966; Hollinshead, 1983; Spinner, 1984; Tountas and Bergman, 1993; Williams et al., 1989). Flexors in general tended to have more primary motor nerve branches and terminal nerve entry points than the extensors. This nding may be related to the need to generate more power in the exors and, in the case of FDS and FDP, the need for complex function. Recent reports (English and Letbetter, 1982; Segal et al., 1991; Segal, 1992; Thomson et al., 1991) have suggested the relationship of terminal nerve entry points to subdivisions of muscles and neuromuscular compartments. Each entry point may correspond to a neuromuscular compartment that can function indepen-

dently from other compartments (English and Letbetter, 1982). Thus, muscles with complex functions such as FDS, FDP, and EDC have a greater number of terminal nerve entry points than do other forearm muscles (Schieber, 1991). Our study, like previous reports (Mortensen and Pettersen, 1966; Hollinshead, 1983; Tountas and Bergman, 1993), showed that PL always shares one primary motor nerve branch with FDS from the median nerve and that the secondary or terminal branch to PL passes through the proximal part of the muscle belly of FDS before innervating it. This nding may be related to the evolutionary development of PL and FDS because PL may substitute for the ring-nger slip of FDS (Tountas and Bergman, 1993). The same reasoning can be applied to EDM because it shares a nerve branch with EDC from the posterior interosseous nerve and other nerve sharing instances. In one of our cases, the terminal nerve branch to EI also passed through the belly of EPL rst before innervating EI. This phenomenon has not been observed in previous reports (Mortensen and Pettersen, 1966; Hollinshead, 1983; Sunderland, 1968) and may be a less common variation.

DISTRIBUTION OF NERVE BRANCHES TO FOREARM MUSCLES

461

Fig. 3. Primary motor nerve branch (white arrow) shared by PL and FDS. Secondary nerve branch to PL (black arrow) passing through the belly of FDS before innervating it. In this specimen, PL has two terminal nerve entry points (white arrowheads). The black arrowhead indicates one of the terminal nerve entry points on FDS.

Segmental crush injuries of the forearm over the proximal one-third will effectively defunction most of the exor compartment (Fig. 4), whereas some of the extensor muscles (BR, ECRL, EPB, EPL, EI) and PQ will survive with intact innervation. Middle one-third forearm crush injuries may defunction FDS, FDP, FPL, and some extensors (EDM, APL, EPB, EPL or EI),

whereas the rest of the exor muscles will survive only marginally harmed. Distal one-third injuries essentially leave the forearm musculature intact. In the clinical situation, however, the actual damage to soft tissue is usually not conned to the zone of injury but extends proximally and distally as the energy of the force is dissipated.

462

J. LIU ET AL.

Fig. 4. Distribution (vertical bars) of location of the primary motor nerve branching points and distribution of location of the terminal nerve entry points expressed as a percentage of the forearm length (lateral epicondyle of humerus to the styloid process of radius). The shorter horizontal bars represent the sites where the most distal and

most proximal primary motor nerve branching points and terminal nerve entry points in the different muscles. The longer horizontal bars represent the statistical median location of the primary motor nerve branching points and the terminal nerve entry points.

Figure 4 also enables assessment of muscle denervation with lacerations involving the major nerve trunks. Laceration over the proximal one-third of the forearm will effectively denervate all the muscles innervated by the major nerve trunk. Mid one-third forearm lacerations of the median and ulna require repair essentially for hand intrinsic muscle function and sensation because almost no forearm muscles will be completely denervated with the exception of PQ. However, merely repairing a major nerve divided in a transverse laceration of the forearm will not address the division of the primary and secondary nerve branches traversing the laceration having arisen from the divided major nerve and other major nerves proximal to the laceration. Recovery in such a situation will be expected to be incomplete. The information on the number and the locations of motor nerve branching points and terminal nerve entry points is also useful in selective denervation of spastic muscles to equalize tone between agonist and antagonist muscles (Garland et al., 1980; Mertens and Sindou, 1991). This surgical procedure enables restoration of function, improves cosmesis, and may facilitate nursing care with better positioning of the spastic limbs. Information on the neuromuscular compartment supplied

by each terminal nerve entry point will probably be required before deciding which terminal nerve branch should be sectioned to restore optimal muscle function. The location of the terminal nerve entry points to the different muscles is useful for electrophysiological studies (Reid, 1920) and facilitates the application of electrodes to activate muscles paralyzed from upper motor neuron lesions (Hoshimiya et al., 1989; Kilgore et al., 1989; Lau et al., 1995; Liu et al., 1995; Nathan, 1989), which applies to patients with traumatic spinal cord injury, cerebravascular accidents, and even cerebral palsy. Electrodes may be surface (percutaneous), being introduced into the terminal nerve entry points through the skin, or may even be applied by open surgery by direct suture to terminal nerve entry points. By stimulating the points electrically with computer control, useful hand function can be achieved. Although the branches to the different muscles have been outlined in detail in this study, additional physiological studies are indicated to determine the extent of each muscles neuromuscular compartment innervated by a particular branch and its terminal nerve entry point (English and Letbetter, 1982; Thomson et al., 1991). Only then can the ideal branches or motor nerve entry points be selected for neuromuscular stimulation.

DISTRIBUTION OF NERVE BRANCHES TO FOREARM MUSCLES

463

The present study, unlike previous work (Linell, 1921; Sunderland, 1968), has illustrated the great variation in the innervation pattern of the forearm musculature. Such information may prove useful to the practicing surgeon who may nd information from standard texts (Mortensen and Pettersen, 1966; Williams et al., 1989) and even those designated for surgeons (Hollinshead, 1983) incomplete. Further studies extended to other sections of the human limb may be indicated for a complete picture.
ACKNOWLEDGMENTS

We are grateful to Mr. E.L. Leow for the illustration, Associate Professor E.H. Yap for the use of the cadaver dissection room, and Mr. Tan Boon Kiat for the photography.
LITERATURE CITED
Brand, P.W., and A. Hollister 1993 Clinical Mechanics of the Hand, 2nd ed. Mosby Year Book, St. Louis. Brash, J.C. 1955 Neuro-Vascular Hila of Limb Muscles. E&S Livingstone, Edinburgh. Cricenti, S.V., M.A. DeAngelis, L.J.A. DiDio, N.A. Ebraheim, R.E. Rupp, and A.S. DiDio 1994 Innervation of the extensor carpi radialis brevis and supinator muscles: Levels of origin and penetration of these muscular branches from the posterior interosseous nerve. J. Shoulder Elbow Surg., 3:390394. English, A.W., and W.D. Letbetter 1982 Anatomy and innervation patterns of cat lateral gastrocnemius and plantaris muscles. Am. J. Anat., 164:6777. Garland, D.E., R. Thompson, and R.L. Waters 1980 Musculocutaneous neurectomy for spastic elbow exion in non-functional upper extremities in adults. J. Bone Joint Surg., 62A:108112. Hollinshead, W.H., and J. E. Markee 1946 The multiple innervation of limb muscle in human. J. Bone Joint Surg., 28:721731. Hollinshead, W. 1983 Anatomy for Surgeons. The Back and Limbs, 3rd ed. Harper & Row, Philadelphia. Hoshimiya, N., A. Naito, M. Yajima, and Y. Handa 1989 A multichannel FES system for the restoration of motor functions in high spinal cord injury patients: A respiration-controlled system for multijoint upper extremity. IEEE Trans. Biomed. Eng., 36:754 762. Kilgore, K.L., P.H. Peckham, G.B. Thrope, M.W. Keith, and K.A. Gallaher-Stone 1989 Synthesis of hand grasp using functional

neuromuscular stimulation. IEEE Trans. Biomed. Eng., 36:761 770. Lau, H.K., J. Liu, B.P. Pereira, V.P. Kumar, and R.W.H. Pho 1995 Fatigue reduction by sequential stimulation of multiple motor points in a muscle. Clin. Orthop., 321:251258. Linell, E.A. 1921 The distribution of nerves in the upper limb, with reference to variablities and their clinical signicance. J. Anat., 55:79112. Liu, J., L. Retnam, H.K Lau, B.P. Pereira, V.P. Kumar, and R.W.H. Pho 1994 A rabbit muscle model for studying contraction characteristics of muscle with multiple motor points. Muscle Nerve, 17:1477 1479. Liu, J., H.K. Lau, W.X. Min, B.P. Pereira, V.P. Kumar, and R.W.H. Pho 1995 Contractile characteristics on electrical stimulation of muscle with multiple motor points: an in vivo study in rabbits. Clin. Orthop., 313:231238. Mertens, P., and M. Sindou 1991 Selective peripheral neurotomies for the treatment of spasticity. In: Neurosurgery for Spasticity. M. Sindou, R. Abbott, and Y. Keravel, eds. Springer-Verlag, New York, pp. 119132. Mortensen, O.A., and J.C. Pettersen 1966 The musculature. In: Morris Human Anatomy, 12th ed. B.J. Anson, ed. Mcgraw-Hill, New York, pp. 421612. Nathan, R.H. 1989 An FES-based system for generating upper limb function in C4 quadriplegic. Med. Biol. Eng. Comput., 27:549556. Reid, R.W. 1920 Motor points in relation to the surface of the body. J. Anat., 18:271275. Schieber, M.H. 1991 Individuated nger movements of rhesus monkey: a means of quantifying the independence of the digits. J. Neurophysiol., l65:13811391. Seddon, H.J., P.B. Medawar, and H. Smith 1943 Rate of regeneration of peripheral nerves in man. J. Physiol., 102:191215. Segal, R.L. 1992 Neuromuscular compartments in the human biceps brachii muscle. Neurosci. Lett., 140:98102. Segal, R.L., S.L. Wolf, M.J. DeCamp, M.T. Chopp, and A.W. English 1991 Anatomical partitioning of three multiarticular human muscles. Acta. Anat., 42:261266. Spinner, M. 1984 Kaplans Functional and Surgical Anatomy of the Hand, 3rd ed. J.B. Lippincott, Philadelphia. Sunderland, S. 1968 Nerves and Nerve Injuries. E&S Livingstone, Edinburgh. Thomson, D.B., S.H. Scott, and F.J.R. Richmond 1991 Neuromuscular organization of feline anterior sartorius: I. Asymmetric distribution of motor units. J. Morphol., 210:147162. Tountas, C.P., and R.A. Bergman. 1993 Anatomic Variations of the Upper Extremity. Churchill Livingstone, New York. Williams, P., R. Warwick, M. Dyson, and L. Banister 1989 Grays Anatomy, 37th ed. Churchill Livingstone, Edinburgh.