PII: SOO21-9290(96)00171-6

ISB KEYNOTE LECTURE

MUSCLE ARCHITECTURE AND FUNCTION IN HUMANS

Tetsuo Fukunaga, Yasuo Kawakami, Shinya Kuno, Kazuo Funato and Senshi Fukashiro
Department of Life Sciences. University of Tokyo, 3-8-1, Komaba, Meguro, Tokyo. Japan

Abstract-The present study focused on architectural factors which are considered to influence the linkage of
muscle fiber and joint actions. By means of real-time ultrasonography we can observe clearly and noninvasively
in oiao the movement of fascicle and aponeurosis in human muscle and measure directly the changes in pennation
angle and length of fascicle during muscle contraction. During dorsi and plantar flexion without load the
movement of tendinous tissue in human tibialis anterior muscle (TA) appeared to synchronize with the displace-
ment of the ankle joint, indicating that the muscle-tendon complexes are stiff relative to the applied force. which is
fairly small in the case of foot shaking motion. On the other hand, when the ankle joint was fixed and the TA
contracted ‘statically’ the ultrasonic echo from deep aponeurosis in the TA was observed to move proximally,
indicating the elastic component (i.e. mainly tendinous tissue) was stretched significantly by the contraction force
of muscle. In the case of the knee joint, a length of fascicle in vastus lateralis decreased by 18% with the extension of
the knee passively from a 100’~ flexed position. When the knee extensors contracted ‘statically’ the fascicle length
decreased at every joint angles and its magnitude was greater (30%) when the knee was closer to full extension than
(5%) at the flexed positions. The present results clearly show that the architecture of actively contracting muscle
tibers differ considerably than that which occurs when movement is passively induced. The use of cadaver data in
the study of architecture and modeling of muscle functions would result in inaccurate, and in some cases even
erroneous results. (1) 1997 Elsevier Science Ltd. All rights reserved.
Kr! I\YI&S: Ultrasonography: Architecture; Muscle fibre; Tendon.

INTRODUCTION 1993, 1994, 1995; Fukunaga et al., 1992, 1996; Narici

et al., 1992; Rugg et al., 1990). In this article, the behavior
Muscle contraction generates movements of the tendon, of the muscle-tendon complex in humans revealed by
which produce joint motions. Because behavior of these techniques is reviewed, the changes in the penna-
a muscle during contraction in humans cannot be di- tion angle and the fascicle length during muscle contrac-
rectly observed, in vivo functions of human muscles have tion are especially formulated.
been evaluated from joint actions. In fact, however,
joint actions not only reflect the intrinsic character-
istics of muscle fibers but are affected by anatomical VISUALIZATION OF TENDINOUS MOVEMENT
factors within a muscle, and within a joint system DURING JOINT ACTION
(Bobbert et ul., 1990; Powell et al., 1984; Roy and Edger-
ton, 1992; Trotter, 1993; Wickiewicz et al., 1983). The A previous paper (Fellows and Rack, 1987) has shown
force-velocity relationship of human elbow flexor some possibilities of the estimation of the movement of
muscles has been reported to be similar to that of muscle contracting muscle in human elbow flexors using X-ray
fibers (Wilkie, 1950), but the similarity is controversial if photography or needles inserted into the muscles,
one considers the multiple factors lying between muscle indicating the biceps muscle shorten or is lengthened in
fibers and a joint: from observations at the joint, one a phase corresponding to the flexion and extension of
cannot get information on how much muscle shortening elbow joint. Some more recent papers (Henriksson-
or lengthening occurs during joint actions. Especially, in Larsen et ul., 1992; Kawakami, 1993; Kuno and
pennate muscles in which fascicles are arranged diago- Fukunaga, 1995; Rutherford and Jones, 1992) have
nally to the line of pull of the muscle (Bobbert et ul., 1990; shown that fascicles and aponeurosis can be visualized in
Roy and Edgerton, 1992; Trotter, 1993), it is almost humans by means of ultrasonography and pennation
impossible to describe the behavior of a muscle-tendon angles can be measured in vim.
unit from the joint movement. In the present study, the movement of tendinous tis-
Recently, development of imaging techniques such as sues in the human tibialis anterior muscle (TA) was
MRI and ultrasonography is remarkable, making pos- measured by the use of ultrasonic apparatus with an
sible the determination of the architecture of the muscle- electronic linear array probe of 7.5 MHz wave frequency
tendon complex in humans (Henrikkson-Larsen et al., (ALOKA, SSD-2000, Jpn). The echoes from interspaces
1992: Rutherford and Jones, 1992; Kawakami et al., of the TA were sometimes imaged more clearly when the
plane of probe was changed slightly diagonally to the line
drawn with the pen, in which case the recreated image
Address correspondence to T. Fukunaga. Department of Life was used for the measurement. The experimenter visually
Sciences, University of Tokyo, 3-R-I. Komaba, Meguro, Tokyo, Japan. confirmed the echoes reflected from aponeurosis and
457
interspaces among fascicles in TA through a comparison of
ultrasonic measurements with direct measures (Kawakami
subcutaneous
of (I/.. 1993). The cross point (P) between two echoes. one r adipose tissue
from deep aponeurosis and the other from fascicles. was
easily determined on the ultrasonogram (Fig. 1).
When an ankle joint was dorsi- and plantar flexed
without additional load, it was observed that the
P moved proximally during dorsiflexion and distally
during plantarflexion. Figure 2 shows an example of the
distance, D, moved by P and joint angle, 0, during dorsi-
and plantarflexion without an additional load. The dis-
placement of P studied here is considered to indicate the
excursion of the tendinous tissue in the TA during muscle
contraction. The D appeared to synchronize with the
displacement of the joint angle during dynamic actions,
indicating that the muscle-tendon complexes are stiff
relative to the applied force---fairly small in the case of
foot shaking motion without load. In the human elbow
flexor muscle, no detectable phase shift was observed
between intramuscular and joint displacements during Fig. 1. Ultrasonic image of the human lower leg. Subo~~t,~ncoua adipose
voluntary elbow flexion-extension movements with no tissue and muscle (tibialis anterior) and internal aponcurobis are 1 isua-
ized. The cross-point (P) was determined from the cchoe~ ~>rf deep
additional load (Fellows and Rack, 1987). aponeurosis and fascicles. A placement makct- was fixed IO \kln on the
During dorsiflexion the changes in 6, and D were highly TA at the mid-point of the tibia1 bone as a refcl-encc point I~J lhat doea
correlated in every subjects (1. = 0.93- 0.99) (Fig. 3). not move during contractions.
Because the TA is a bipennate muscle with its proximal
fibers attached to the tibia1 bone, it is reasonable to
assume that the proximal attachment does not move
during contraction (Fig. 4). Thus, displacement of the
central aponeurosis which is attached to tuberosity of
metatarsal in the foot, i.e. movement of P is considered to
be equivalent to the movement of the distal tendon. It
follows therefore that the slope of the regression line in
Fig. 3 (i.e. the distance moved by P per 1 rad of ankle
dorsiflexion) represents the moment arm of TA. Values i 20-
ranging from 44 to 48 mm over a motion range of
17 - 34 were determined for five subjects, which were in
o-
good agreement with a previous report. i.e. TA moment
arm measured by MRI changed between 51 and 42 mm , I I I
0 2 4 6 8 10
depending on ankle joint angles (Rugg et al., 1990).
The tendinous movement observed in this study was Time (s)
caused by the change in length of muscle fibers. However,
the tendinous movement might include elongation of the Fig. 2. Displacement of P (D) and angle of ankle joint (0) during
dorsi- and plantar-flexion. The D was measured at every 25ms.
aponeurosis (although not a large amount). Furthermore, it Zero degree of ankle joint (8) represents at the natural position of
has been shown that the fibers do not run from the origin foot and the D represents the distance from the natural position
to insertion, but taper within fascicles (Hijikata et al., moved by P.

20
20? 1

I5 -
Subj. A
I Subj. B

b io 2b 0 10 20 0 10 20

8 (de@ 61 (deg) 0 (de@

Fig. 3. The relationships between D and H for three subjects. Significant positive correlation was observed
in all subjects.
 Muscle
Fig. 4. Schematic illustration for measuring the movement
the P and the moment arm of TA during ankle dorsiflexion.
dorsiflexion the point P on deep aponeurosis in TA moved proximally.
The moment arm is approximately equivalent to the distance D during
1 rad of dorstflexion.
1993). Elasticity of the attachments between tapering
fibers might affect the tendinous movement. The present
study cannot answer what amount in the tendinous
movement is muscle and what amount is tendon. In
addition, the present results are based on the measure-
ments of the movement of one certain point on the
aponeurosis. It has been shown in rat muscles that there
is some variability in the elasticity over the aponeurosis
(Zuurbier et nl., 1994). Such heterogeneity may exist in
human muscles, in which case the movement of separate
areas on the aponeurosis might not be the same. This
uncertainty would be the subjects, who can be investi-
gated with the present method in future studies. This
method would also have other applications in the field of
muscle physiology and biomechanics in humans.
MUSCLE SHORTENING DURING ‘ISOMETRIC/STATIC’
ACTION
According to Hill’s classical model, the muscle-tendon
complex consists of contractile and series elastic compo-
nents (CC and SEC, respectively). The major part of SEC
is located in the tendinous tissues (Huijing, 1992). The
SEC demonstrates elastic behavior in most movements
and especially when the muscle is activated and simulta-
neously stretched prior to concentric action (Komi,
1984). The characteristics of SEC have been indirectly
estimated by special in vivo and in vitro methods (Huijing,
1992; Poussen et nl., 1990).
We can observe in civo the movement of tendinous
tissue in tibialis anterior muscle (TA) during ‘static’ con-
traction by means of ultrasonography. The subject lay
supine on a bed and was requested to perform ‘static’
architecture and function in humans
q 0.00 rad
l 0.26 rad
0 5 10 15 20
Fig. 5. Relationship between the distance (AI) moved by the P on
the aponeurosis and the static force (AF) exerted by ‘static‘ contrac-
tion of dorsiflexor at different angles of the ankle joint. 0.00 rad repres-
ents the natural position of the foot, and 0.26 and 0.52 rad represent the
positions dorsiflexed from the natural position, respectively. AF was
obtained from the ‘static’ measured torque of dorsiflexion divided by
the moment arm.
distance of
During dorsiflexion with gradually increasing force in each trial
at three ankle joint angles. The ultrasonic echo from deep
aponeurosis in TA (same as the point P in Fig. 1) was
observed to move proximally during ‘static’ dorsiflexion,
which is in good agreement with our previous study
(Fukashiro et al., 1995). The distance moved by P(A I) was
considered as the changes in tendon length. The tension
exerted by muscle contraction (AF) was calculated from
the joint torque measured by a dynamometer and the
moment arm reported previously (Rugg et (ll., 1990).
The tension (A.F))length (AI) curve during ‘static’ con-
traction (Fig. 5) was well-fitted to quadratic regression,
which was quite similar to that of the isolated tendon
(Woo, 1981).
The force-length curve of the human tendon can be
directly but noninvasively obtained by the ultrasonic
method, and it is considered that during static muscle
contraction an elastic component (i.e. mainly tendinous
tissue) of the muscle-tendon complex is stretched signifi-
cantly by the contraction force of the muscle.
FASCICLE LENGTH RELATED TO JOINT ANGLE
Most skeletal muscles run across the joints; thus, the
shortening of muscle fibers induces changes in joint
angles. The muscle architecture (geometric character-
istics of fiber or tendon), therefore, has a considerable
effect on the force-producing capacity of the muscle (Roy
and Edgerton, 1992; Huijing, 1992).
We measured the length and angle of fascicle (fiber) of
the vastus lateralis muscle and observed changes in the
fascicle length and pennation angle during knee exten-
sion. In the ultrasonic image obtained from the vastus
lateralis muscle at 50% of the distance from the greater
trochanter to the lateral epicondyle of the femur (Fig. 6)
the echoes reflected from the fascicles and deep aponeur-
osis can be observed. One of the fascicles was carefully
traced from the proximal and distal ends. and the
length of fascicle was measured by the use of a digital
 Knee angle 15” Knee angle 50”

----a
--x-
Fig. 6. Ultrasonic image obtained from human vastus lateralis muscle at two different joint angles during
relaxed and tensed conditions with 10% of maximal static strength. The ultrasonic transducer was placed
on the skin over the vastus lateralis muscle at 50% of the distance from the great trochanter to lateral
epicondyle of femur. The fascicle length (fl) was determined as the length of a line drawn along the
ultrasonic echo parallel to fascicles. The fascicle angle (0) was determined as angles between the echoes
obtained from fascicles and deep aponeurosis in the ultrasonic image.

curvimeter. The fascicle angles, defined as the angles 1501

I *
between the echoes of the deep aponeurosis and the
echoes from interspaces among fascicles were also meas-
ured. When the knee was extended passively from 100’
flexion (0” = full extension) without muscle contraction,
the fascicle length decreased by 18%, from 129 + 9 mm
(mean + SE.) at 100” to 107 f 8 mm at 20” (Fig. 7).
When the knee extensors contracted ‘statically’ with 10%
of their maximal voluntary strength (MVC) at different
joint angles, in which case the MVC changed due to the
----.-.- Relaxed
joint angle, the fascicle length decreased and its magni- - Tensed
tude was greater (30%) when the knee was closer to full
extension than that (5%) at the flexed positions (Fig. 7).
During ‘static’ contractions of the quadriceps muscle so!
100 80 60 40 20
a patellar tendon was oriented anteriorly at extended
knee positions (Herzog and Read, 1993) suggesting lax- Knee joint angle (de@
ity of the joint in this condition. Because the
Fig. 7. Changes in fascicle lengths at various angles of the knee joint
muscle-tendon complex consists of muscle fibers and (n = 6. mean and SE.). After measurements of the fascicle length with-
tendinous tissues, or of CC and SEC (Bobbert et al., out muscle contraction at each joint angle the subject was requested to
1990) the differences in changes in fascicle length be- perform isometric contractions of knee extensor muscles at 10% MYC
tween the relaxed and tensed conditions would be caused by using a specially designed dynamometer (MYORET, Asics, JPN) in
by the laxity of the relaxed muscle and the lengthening of which the gravitational effect was controlled and the fascicle length was
measured during contraction as well.
the SEC during contraction. Because lengthening of the
SEC is a function of tension applied to it (Bobbert et al.,
1990) the lengthening of SEC would be expected at each the slackness of the muscle-joint system when the knee is
joint angle. Larger shortening of the fascicles at the extended. The fascicles are connected to the deep apo-
extended that at flexed positions would be due to neurosis with an angle, which was defined as a fascicle
 Muscle architecture and function in humans 461

“M z
a”
- 18.
z 8- Angular velocity lOdeg/sec
aJ
"M 2v 6-
9b x
16.
9
Y
cd 14-
14

12/
100 80 60 40 20

Knee joint angle (deg) 01

Fig. 8. Changes in fascicle angles as a function of knee angles (n = 6,
mean and S.E.). Significantly higher fascicle angles in tensed than in
relaxed conditions were observed at the extended positions (i.e. knee
angle = 20 ).
angle (or pennation angle). The fascicle angle at the
relaxed condition increased from 14 to 17” with knee
extension (Fig. 8). Larger fascicle angles were observed in
tensed than in relaxed conditions at each joint angle and
greater differences between the two conditions were ob-
served when the knee was in extended positions. The
pennation angle of vastus lateralis muscle has been re-
ported to be 5520” obtained from cadavers (Yamaguchi
et al., 1990) and 6 - 27” in living humans (Henriksson-
Larsen et ul., 1992; Rutherford and Jones, 1992).
These are in good agreement with our measurements. As
the force applied to the tendon by the contraction of
muscle fibers is less than the sum of the forces of the
individual fibers, i.e. the tendon force is reduced by a fac-
tor of cos0 (0 = pennation angle), the larger the penna-
tion or fascicle angles, the more disadvantageous the
force development would be expected (Gans and Gaunt,
1991; Roy and Edgerton, 1992; Kawakami et al., 1993).
Increase of the fascicle angle at the extended positions,
therefore, implies a mechanical disadvantage in the trans-
mission of force.
In previous studies the ‘isokinetic’ contraction has
been defined as joint action at a constant angular velocity
throughout the range of motion (Wickiewicz et al., 1984).
When the knee was extended with a given angular velo-
city (i.e. ‘isokinetic’ action), the shortening velocity of
fascicle increased when the knee was in extended posi-
tions (Fig. 9) i.e. with ‘isokinetic’ extension of the knee
with 10’~si during the whole range of motion the fas-
cicle velocity was 4 mm s- ’ at flexed positions (100-60”)
and increased linearly up to double this (8 mm s-i) near
the knee extension. This would suggest that when study-
ing the functional characteristics of muscle fiber from
knee joint actions, care has to be taken as to the hetero-
geneity of the shortening velocity of muscle fibers during
isokinetic muscle actions, especially when the knee is
close to the extended positions.
Considering the smaller force production at shorter
muscle lengths and at higher contraction velocities esti-
mated from the force-length and forceevelocity relation-
ships, and the lower efficiency of force transmission at
larger angles of pennation, the work performance at the
tendon of vastus lateralis muscle is sequentially reduced
100 80 60 40 20
Knee joint angle (deg)
Fig. 9. Velocity of shortening of the fascicle in relation to knee joint
angles (n = 6. mean and SE.). The fascicle velocity was calculated from
changes of the fascicle length traveled in 10 of the joint when the knee
was extended ‘isokinetically’ with an angular velocity of 10 s-’ and
with a tension of about 10% MVC by an isokinetic dynamometer.
with knee extension, e.g. much lower MVC (81 N m) at
knee extended position (20”), was observed, compared to
knee flexed positions (244 N m at 60” and 213 N m at
100”).
PENNATION ANGLES IN HYPERTROPHIED MUSCLES
Morphologically, skeletal muscles are roughly divided
into two types, parallel fiber muscles in which fibers line
almost parallel to the line of action of the muscle and
pennate muscles in which fibers insert into the tendon at
an angle to the line of action of the muscle (pennation
angle) (Jones et al., 1989; Hijikata et al., 1993). The
pennation angle (fascicle angle) in the human triceps
brachii muscle was measured as 10 - 60” for many differ-
ent age groups (aged 10-60yr) by means of the ultrasonic
method. Figure 10 shows the relationship between the
fascicle angle and the muscle thickness for the triceps
brachii muscle with significant correlation coefficients.
which agrees with a previous study (Kawakami et al..
1993). These would imply that muscle hypertrophy might
accompany an increase in fiber angles. Maxwell et al.
(1974) predicted that for the muscle having constant
muscle length, fiber length, and fiber numbers, hyper-
trophy (i.e. increase in the diameter of fibers) must accom-
pany increases in pennation angles and muscle thickness.
The physiological cross-sectional area (PCSA) of
muscle, which is defined as the total area of fiber in
parallel and then represents the force potential produced
by muscle, is calculated by
PCSA = (MV/FL)cos 8.
where MV is muscle volume and FL is muscle-fiber
length. The cosine of the pennation angle 8 normalizes
fiber angulation to the line of action of the muscle. If
muscle hypertrophy accompanied the increase in
fiber pennation angles, then this change would be disad-
vantageous for the force development. However, an in-
crease in pennation angles would cause more contractile
material to be attached to the tendon. which would
 Friedrich, J. A. and Brand. R. A. (1990) ivlusclc ar~hllccturc 11, iltc
human lower limb. J. Biur?rrc,hrrnirs 23, Y I 95.
Fukashiro, S.. Ito. M., Ichinosc. Y., Kawahaml. Y. and I-ukun+a. t.
(1995) Ultrasonography gives directly bui noninvasivelp elastic
characteristic of human tendon i,: ri~o. El/r. J. trppi. Pllysiol. 71,
555-551.
Fukunaga. T., Roy, R. R.. Shellock, F. G.. Hodgson. J. A. and Edgerton.
V. R. (1996) Specific tension of human plantarflexorh and dorsi-
00 flexors. J. appl. Physiol. 80, I58 165.
0.00 0.05 0.10 0.15 0.20 0.25 Fukunaga. T.. Roy, R. R., Shellock. F. G., Hodgson, J. A., Lee. P. L.,
Kwong-Fu. H. and Edgerton, V. R. (1992) Physiological cross-sec-
Muscle thickness/upper arm length tional area of human leg muscles based on magnetic resonance
imagining. J. orrhop. Rus. 10, 9%934.
Fig. 10. Relationship between fascicle angles of triceps brachii (long Cans, C. and Gaunt. A. S. (1991) Muscle architecture in relation to
head) and ratio of muscle thickness to upper arm length for 537 male function. J. Biomechrmi~~s 24, 53 -65.
and female subjects (aged IO-60 yr); r = 0.81, p < 0.001. Hather. B. M.. Adams. G. R.. Tesch. P. A. and Dudley, G. A. (1992)
Skeletal muscle responses to lower limb suspension in humans. J.
c~ppl. Physiol. 72, 1493 1498.
increase the muscle volume and consequently increase Henriksson-Larsen. K., Wretling. M.. Lorentzon, R. and Oberg. L.
PCSA. (1992) Do muscle fiber size and fiber angulation correlate in pennated
It must be noted that the pennation angles measured human muscles’? Etrr. J. uppI. Physiol. 64, 6% 72.
by ultrasonography were those between the aponeurosis Herzog, W. and Read. J. (1993) Lines of action and moment arms of the
major force-carrying structures crossing the human knee joint. J.
and the fascicles. The pennation angles may not be accu- Anat. 182, 213.-230.
rate vectors for predicting the effect of angulation of Hijikata, T., Wakisaka. H. and Nlida, S. (1993) Functional combmation
fibers on force transfer to the tendon if the aponeurosis of tapering profiles and overlapping arrangements in nonspanning
lies at an angle to the line of action of the muscle. But, in skeletal muscle fibers terminating intrafascicularly. A/l(lt. Rec. 236,
602-6 IO.
the case of elbow extensor muscles, according to visual Huijing. P. A. (1992) Elastic potential of muscle. In Strelqrh UM~ Potvet
observation on human cadavers, the angle of the apo- ill Sport (Edited by Komi. P. V.). pp. 151-168. Blackwell Scientific
neurosis on the deep border to the long axis of the long Publications. Oxford.
head was small compared with that of the opposite Jones, D. A. and Rutherford, 0. (1987) Human muscle strength train-
aponeurosis; thus, the error in predicting the impact of ing; the effects of three different regimes and the nature of the
resultant changes. J. Physio/. 391, I -1 I.
fiber angulation on the force transfer from the pennation Kawakami, Y.. Abe. T. and Fukunaga, T. (1993) Muscle-fiber penna-
angle would be small (Kawakami et al., 1993). tion angles are greater in hypertrophied than in normal muscles.
J. appl. Phgsiol. 74, 2740-2744.
Kawakami. Y.. Abe, T.. Kuno. S. and Fukunaga, T. (1995) Training-
CONCLUSION
induced changes in muscle architecture and specific tension Eur. J.
Appl. Physiol. 72, 37113.
As described in the Introduction there are some archi- Kawakami. Y., Nakazawa, K., Fujimoto, T., Nozaki, D., Miyashita. M.
tectural factors that lie between muscle fibers and joints. and Fukunaga. T. (1994) Specific tension of elbow flexor and extensor
We focused on the architectural factors which are con- muscles based on magnetic resonance imaging. Eur. J. crppl. Phyioi.
sidered to influence the linkage of muscle fiber and joint 68.139-147.
Komi, P. V. (1984) Physiological and biomechanical correlates of
actions. By means of real-time ultrasonography we can muscle function: effects of muscle structure and stretch-shortening
observe clearly and noninvasively in vivo the movement cycle on force and speed. Euerc. Sport SC;. Ret).. Vol. 12 (Edited by
of fascicle and aponeurosis in human muscle and Terjung. R. L.), pp. 81-121. Franklin Institute Press, Philadelphia.
measure directly the changes in the pennation angle and Kuno. S. and Fukunaga, T. (1995) Measurement of muscle fiber dis-
length of fascicle during muscle contraction. placement during contraction by real-time uitrasonography in hu-
mans. Eur. J. appl. Physiol 70, 45-48.
To date, architectural data of skeletal muscles have been Maxwell, L. C.. Faulner. J. A. and Hyatt, T. (1974) Estimation of tibres
derived mainly from cadaver specimens (Arnis et al., 1979; in guinea pig skeletal muscles. J. appl. Physiol. 37, X9-264.
An et al., 1981; Friedrich and Brand, 1990; Wickiewicz et al., Narici. M. V.. Roi, G. S. and Landoni. L. (1992) Assessment of human
1984). The present results clearly show that the architecture knee extensor muscles stress from ip~ tk~ physiological cross-sec-
of actively contracting muscle fibers differs considerably tional area and strength measurements. Eur. J. appl. Phwiol. 65,
438444.
than that which occurs when movement is passively in- Poussen, M.. Hoecke, J. V. and Goubel, F. (1990) Changes in elastic
duced. The use of cadaver data in the study of architecture characteristics of human muscle induced by eccentric exercise. J.
and modeling of muscle functions would result in inaccur- Biomecharlics 23, 343-348.
ate. and in some cases even erroneous results. Powell, P. L., Roy, R. R., Kanim, P. M., Belle. A. and Edgerton. V. R.
(1984) Predictability of skeletal muscle tension from architectural
determinations in guinea pig hindlimbs. J. appl. Physiol. 57,
REFERENCES 1715~1721.
Roy, R. R. and Edgerton, V. R. (1992) Skeletal muscle architecture and
Amis. A. A., Dowson, D. and Wright, V. (1979) Muscle strength and performance. In&en& and Power in Sport (Edited by Komi. P. V.).
musculo-skeletal geometry of the upper limb. Engny Med. 8, 4148. ~JD. 115 -129. Blackwell Scientific Publications, Oxford.
An, K. N.. Fui, F. C., Morrey, R. L.. Lincheid, R. L. and Chao, E. Y. R&g, S. G.. Gregor. R. J.. Mandelbaum. B. R. and Chui. L. (1990)
(1981) Muscles across the elbow joint: a biomechanical analysis. In rCw moment arm calculations at the ankle using magnetic rexon-
.I. Biomrchnnics 14, 659 -669. ante imaging (MRI). J. Biontrclttrr~i~~.r 23, 495- 501.
 Muscle architecture and function in humans 463

Rutherford, 0. M. and Jones. D. A. (1992) The relationship of muscle muscle. J. PhJrsiol. 110, 249-280.
and bone loss and activity levels with age in women. Eur. J. nppl. Woo, S. L.-Y. (1981) The effect of exercise on the biomechanical and
Ph&~l. 65, 433437. biochemical properties of swine digital flexor tendons. J. hionleclr.
Trotter, J. A. (1993) Functional morphology of force transmission in Enqng 103, 51-56.
skeletal muscle. A&I Anat. 146, 205-222. Yamaguchi, G. A., Sawa, G. U., Moran. D. G.. Fcssler.
Wickizwicz. T. L., Roy. R. R., Powell, P. L. and Edgerton. V. R. (1983) M. J. and Winters. .I. M. (1990) Mdtiplv Muscle Sysren~s
Muscle architecture of the human lower limb. C/in. Orthop. Rrlat. (Edited by Winters. J. and Woo. S.). pp. 717-773. Springer. New
Rrs. 179, 275.-283. York.
Wickiewicz. T. L., Roy, R. R., Powell. P. L.. Perrine. J. J. and Edgerton, Zuurbier, C. J., Everard, A. J., van der Wees, P. and Huijing, 1’. A. ( 1994)
V. R. (1984) Muscle architecture and force-velocity relationships in Length-force characteristics of the aponeurosis in the passive and
humans. J. rrppl. Ph~siol. 57, 4355443. active muscle condition and in the isolated condition. J. Bio-
Wilkle, D R. (1950) The relation between force and velocity in human tnrr~/~unic~s 27, 445453.