Journal of Biomechanics 33 (2000) 659}668

The relationship between force depression following shortening

and mechanical work in skeletal muscle
W. Herzog*, T.R. Leonard, J.Z. Wu
Human Performance Laboratory, The University of Calgary, Faculty of Kinesiology, 2500 University Drive, N.W., Calgary AB T2N 1N4, Canada
Accepted 20 December 1999

Abstract

Force depression following muscle shortening was investigated in cat soleus (n"6) at 373C for a variety of contractile conditions
with the aim to test the hypotheses that force depression was independent of the speed of shortening and was directly related to the
mechanical work produced by the muscle during shortening. Force depression was similar for tests in which the mechanical work
performed by the muscle was similar, independent of the speed of shortening (range of speeds: 4}256 mm/s). On the other hand, force
depression varied signi"cantly at a given speed of shortening but di!erent amounts of mechanical work, supporting the hypothesis
that force depression was not speed * but work dependent. The variations in the mechanical work produced by the muscle during
shortening accounted for 87}96% of the variance observed in the force depression following shortening further supporting the idea
that the single scalar variable work accounts for most of the observed loss in isometric force after shortening. The results of the present
study are also in agreement with the notion that the mechanism underlying force depression might be associated with an inhibition of
cross-bridge attachments in the overlap zone formed during the shortening phase, as proposed previously (Herzog and Leonard, 1997.
Journal of Bimechanics 30 (9), 865}872; MareH chal and Plaghki, 1979. Journal of General Physiology 73, 453}467).  2000 Elsevier
Science Ltd. All rights reserved.

Keywords: Whole muscle mechanics; Mechanism of muscular force production; Concentric contraction; Cat soleus

1. Introduction scribe force depression mathematically. One of the excep-

The isometric force in muscle following shortening
against resistance is lower (depressed) compared to the
isometric force obtained at the same length without prior
shortening. This observation has "rst been described by
Abbott and Aubert (1952), and has since been con"rmed
on many occasions (e.g. Edman et al., 1993; Herzog and
Leonard, 1997; MareH chal and Plaghki, 1979; Meijer et al.,
1997; Sugi and Tsuchiya, 1988). The mechanism underly-
ing force depression following shortening is not known,
although several mechanisms have been proposed in the
past (Edman et al., 1993; Granzier and Pollack, 1989;
Herzog and Leonard, 1997; MareH chal and Plaghki, 1979).
Because of the lack of a generally accepted mechanism
for force depression following muscle shortening, re-
searchers have largely refrained from attempting to de-
* Corresponding author. Tel.: #1-403-220-8525; fax: #1-403-284-
3553.
E-mail address: walter@kin.ucalgary.ca (W. Herzog).
tions is the work by MareH chal and Plaghki (1979) who
showed that force depression is linearly related to the
amount of shortening and that the linearity constant
depended on where on the force}length relationship the
experiments were performed. They further described that
force depression was an inverse exponential function of
the speed of shortening. Similar "ndings have been re-
ported by others (Abbott and Aubert, 1952; De Ruiter
et al., 1998).
Recently, based on results in cat soleus, it was pro-
posed that force depression may be independent of the
speed of shortening, but rather may depend on the force
during shortening (Herzog and Leonard, 1997). This
"nding led to the suggestion of a mechanism of force
depression following shortening based on the mechanical
deformation of actin "laments entering the actin-myosin
overlap zone during shortening (MareH chal and Plaghki,
1979), and a corresponding change in the relative orienta-
tion between myosin cross-bridges and actin-binding
sites that might reduce the probability for cross-bridge
attachments in the newly formed overlap zone. Since

0021-9290/00/$ - see front matter  2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 0 2 1 - 9 2 9 0 ( 0 0 ) 0 0 0 0 8 - 7
660 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668
actin myo"laments are known to be compliant (Gold-
man and Huxley, 1994; Higuchi and Goldman, 1995;
Huxley et al., 1994; Kojima et al., 1994; Wakabayashi
et al., 1994), and since this compliance is associated with
changes in the three-dimensional orientation of actin and
myosin "laments (Daniel et al., 1998), a stress-dependent
inhibition of cross-bridges in the overlap zone formed
during shortening becomes an attractive (and possible)
mechanism (Herzog, 1998). Results on isolated frog "bres
show that sti!ness decreases in parallel with the amount of
force depression following shortening (Sugi and Tsuchiya,
1988), supporting the idea that force depression may be
associated with a decrease in the number of attached
cross-bridges compared to purely isometric contractions.
If force depression following shortening is caused by
a stress-dependent deformation of actin (and possibly
myosin) "laments, force depression should be indepen-
dent of the speed of shortening, but should depend on the
amount of shortening and the stress (force) acting on the
myo"laments during shortening. Furthermore, force de-
pression should occur continuously during the shorten-
ing phase. Therefore, it appears that force depression
following shortening might depend primarily on a single
scalar variable, the work performed by the muscle during
the shortening phase. Although it has been suggested
that work may be associated with force depression (De
Ruiter et al., 1998; Granzier and Pollack, 1989; Herzog,
1998), no systematic study has been conducted to investi-
gate the possible relationship between work and force
depression in skeletal muscle.
The purpose of this study was to test whether or not
the single scalar variable work could account for most of
the observed force depression following skeletal muscle
shortening under a variety of conditions. Speci"cally,
shortening contractions were performed by keeping two
of the three determinant variables of shortening (amount
of shortening, speed of shortening, and muscle activation
during shortening) constant, while the third variable was
varied in a systematic way. Furthermore, tests were per-
formed with variable speeds and variable activation of
the muscle during the shortening phase.
2. Methods
Force depression following shortening was determined
in cat soleus (n"6) using a setup that has been described
earlier (Herzog and Leonard, 1997; Herzog, 1998). Brie-
#y, cats were anaesthetized using a nitrous oxide, halo-
thane, oxygen mixture. The soleus, soleus tendon and
calcaneus were exposed using a single cut on the poste-
rior, lateral shank. The soleus tendon was isolated from
the rest of the Achilles tendon and was cut from the
calcaneus with a remnant piece of bone after a reference
length (corresponding to an 803 included ankle angle)
had been determined.
A second cut was made on the posterior, lateral thigh
and the tibial nerve was exposed and implemented with
a bipolar cu!-type electrode (Herzog et al., 1995) for
soleus stimulation. The cat was secured in a prone posi-
tion in a hammock and the pelvis, thigh, and shank of the
experimental hindlimb were "xed with bilateral bone
pins to a stereotaxic frame. The bone piece at the distal
end of the soleus tendon was attached with sutures to
a muscle puller (MTS, Eden Prairie, MN, natural fre-
quency '10 kHz). The soleus forces (100 N"10 V) and
excursions (20 mm"10 V) were measured continuously
by the muscle puller and were collected at a frequency of
200 Hz, except for tests exceeding a shortening speed of
64 mm/s for which data were collected at 1000 Hz. The
muscle length corresponding to the 803 included ankle
angle was taken as zero length; shortening from that
length was taken as negative, lengthening as positive. For
example, a muscle length of !4 mm corresponds to
a length 4 mm shorter than the reference length.
Six series of tests were performed. In the "rst series of
tests, an isometric contraction at a given length was
followed by a shortening contraction at a given speed
and constant activation but varying amount of shorten-
ing, followed by a second isometric contraction at the
new (shortened) length. The "nal length was always
!4 mm; the starting lengths were (typically) !2, 0,
#2, #4, and #6 mm. The speed of shortening was
4 mm/s and the stimulation of the tibialis nerve was three
times the a-motoneuron threshold (3 T), 30 Hz, 0.1 ms
pulse duration, and 8 s train duration. For the cat soleus
(a primarily slow twitch "bred muscle (Ariano et al.,
1973)), this stimulation protocol resulted in a fused teta-
nic contraction of (likely) all motor units.
In the second series of tests, isometric-shortening-
isometric contractions were performed with constant ac-
tivation (3 T, 30 Hz, 0.1 ms, 8 s) and constant amount of
shortening (8 mm) but varying speeds (typically, 4, 8, 16,
32, 64, 128 and 256 mm/s).
In the third series of tests, isometric-shortening-
isometric contractions were performed at a constant
speed (4 mm/s) over a constant shortening range (8 mm)
but varying levels of activation during the shortening
phase ranging from 0 to 3 T in about 4}8 steps. The exact
current provided to the tibial nerve for the varying levels
of activation were di!erent from muscle to muscle and
changed within a muscle for di!erent tests because of the
ever changing threshold level for a-motoneuron stimula-
tion. The activation of the muscle was constant for the
initial and "nal isometric contractions (3 T, 30 Hz,
0.1 ms).
The fourth series of tests was identical to the third,
except there was no shortening, and therefore no work
performed by the muscle, when the level of activation was
changed.
In the "fth series of tests, shortening over a given
distance (8 mm) and constant activation (3 T, 30 Hz,
 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668 661

0.1 ms) was performed using a variety of changing speeds.

The changes in speed were controlled by the computer
that controlled the muscle puller. In our setup, any speed
change that can be expressed as a function of time can be
implemented. The speci"c speed changes chosen for this
experiment are shown as part of Fig. 5 in the results
section. In the last series of tests, the amount of shorten-
ing (8 mm) and the speed of shortening (4 mm/s) were
kept constant whereas activation during the shortening
phase was varied.
In all six series of tests, two identical isometric-
shortening-isometric tests were preceded and followed
by an isometric reference contraction at the "nal length.
Force depression was determined as the di!erence in
force between the isometric reference contraction and the
steady-state isometric force following shortening (i.e.
3 s after the end of any shortening contraction in a given
experiment). This di!erence in force was determined
at the same instant in time for the reference and the
experimental contractions to avoid any bias caused by
fatigue-related loss of force. Work performed by the
muscle was calculated as the area under the muscle
force}length change graph. Calculating the work for each
trial of all six series of tests gave, on average, about 50}60
values per muscle that could be related to the force
depression values. These 50}60 values represent 25}30
independent measurements that were repeated once.
Since the repeat measurements were nearly identical in
all cases, they were not used for deriving the relation
between work and force depression. A least-squares "t-
ting power function that was forced through the origin
(zero force depression, zero work) was used to describe
Fig. 1. Representative force}time histories of isometric reference con-
the relationship between work and force depression. tractions at a muscle length of !4 mm, and force}time histories for
Also, force depression values were plotted as a function of isometric-shortening-isometric contractions (a) in which the shortening
the speed of shortening. All experimental procedures distance was varied from 2 to 10 mm as shown (b). Shortening speed
were approved by the animal ethics committee of the (4 mm/s) and activation (maximal) were kept constant. The correspond-
University of Calgary. ing work}time histories for the above contractions (c).

3. Results shortening phase, and since the amount and speed of

Force depression following muscle shortening was dir-
ectly related to the amount of shortening (Fig. 1a); sim-
ilarly, the work performed by the muscle was increased
with increasing amounts of shortening (Fig. 1c).
Force depression was inversely related to the speed of
shortening (Fig. 2a), and, for a given amount of shorten-
ing and constant activation, work of the muscle was
decreased with increasing speeds of shortening (Fig. 2c).
This latter result is caused by a decrease in the force
during the shortening phase as the speed of shortening
increases. It is merely a re#ection of the force}velocity
property of skeletal muscle (Hill, 1938).
Force depression was increased for increasing levels
of tibial nerve stimulation (Fig. 3a). Increased
stimulation caused an increase in the force during the
shortening were kept constant, the work performed by
the muscle during shortening was also increased for in-
creased tibial nerve stimulations (Fig. 3c). When activa-
tion was decreased for a 2 s period without any
shortening of the muscle, no work was performed in the
process and there was no force depression upon full
reactivation (Fig. 4).
Performing isometric-shortening-isometric contrac-
tions at variable speeds or variable stimulation of the
tibial nerve during the shortening phase resulted in force
depressions that were directly related to the di!erent
amounts of work performed by the shortening muscle
(Figs. 5 and 6, respectively).
The work performed by the muscle during shortening
accounted, on average, for 92% ($3% S.D.) of the force
depression for all six muscles with a range from 87 to
662 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668
Fig. 2. Representative force}time histories of isometric reference con-
tractions at a muscle length of !4 mm, and force}time histories for
isometric-shortening-isometric contractions (a) at di!erent speeds of
shortening varying from 4 to 128 mm/s (b). Shortening distance (8 mm)
and activation (maximal) were kept constant. The corresponding
work}time histories for the above contractions (c).
96% Figs. 7(a}f). Although, a casual glance at Figs. 7(a}f)
might give the impression that the force depression-work
relationships vary much across muscles, it should be
noted that the raw data for all muscles are remarkably
similar. When combining all force depression and work
values shown in Figs. 7(a}f) in a single graph, the result-
ing power relationship is still strong (Fig. 8; r"0.85), i.e.
85% of the variance in force depression across all six
muscles and all six tests was explained by the variance in
the mechanical work of the muscle.
When plotting force depression as a function of the
shortening speed, two observations were made: "rst,
force depression was similar for experiments performed
at vastly di!ering speeds provided that the work during
shortening was similar, and second, force depression var-
ied greatly for a given speed of shortening, provided that
the work varied as well (Fig. 9).
Fig. 3. Representative force}time histories of isometric reference con-
tractions at a muscle length of !4 mm, and force}time histories of
isometric-shortening-isometric contractions at four levels of activation
varying from zero to maximal during the shortening phase (a) Note,
that the magnitude of force depression increased with increasing work
performed by the muscle; i.e. the amount of force depression increased
from trace b to c, to d, to e, as expected. The shortening distance (8 mm)
and shortening speed (4 mm/s) were kept constant (b). The correspond-
ing worktime histories for the above conditions (c).
4. Discussion
Force depression following shortening in skeletal
muscles has been associated primarily with the amount
of shortening, and the shortening speed (Abbott and
Aubert, 1952; MareH chal and Plaghki, 1979). The most
common mechanism related to force depression has been
sarcomere length non-uniformity (Edman et al., 1993).
For a variety of reasons, shortening speed and sarcomere
length non-uniformity appear less attractive as candidate
mechanisms at present than they might have in the past.
For example, it has been demonstrated earlier (Herzog
and Leonard, 1997), and in this study, that force depress-
ion may vary substantially in tests in which the shorten-
ing distance and speed are kept constant and force during
 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668
Fig. 4. Representative force}time histories of isometric contractions in
which activation was changed for a 2 s period following full force
development. Note, no work was performed during this process as there
was no muscle shortening. There was no force depression associated
with the change in activation levels.
Fig. 5. Representative force}time histories of isometric reference con-
tractions at a muscle length of !4 mm, and force-time histories of
isometric-shortening-isometric contractions (a). The speed of shorten-
ing was varied continuously in these contractions, whereas the shorten-
ing distance (8 mm) and activation (maximal) were kept constant (b).
The corresponding work}time histories for the above conditions (c).
663
Fig. 6. Representative force}time histories of isometric reference
contractions at a muscle length of !4 mm, and force}time histories
of isometric-shortening-isometric contractions (a). The activation
during the shortening phase was varied systematically, here, for
example, by deactivating the muscle for the "rst 500, 1000, 1500, and
2000 ms of the shortening phase and then reactivating it maximally.
The distance (8 mm) and speed of shortening (4 mm/s) were kept con-
stant (b). The corresponding work}time histories for the above condi-
tions (c).
shortening is varied (Fig. 3). Furthermore, we found
consistently that force depression was similar when the
mechanical work performed by the muscle during
shortening was similar, even if the speed of shortening
was changed from the low (4 mm/s) to the high extreme
(256 mm/s) tested in this study (Fig. 9). These results
suggest that it is the work rather than the speed of
shortening that is related to the force depression.
Similarly, the sarcomere length non-uniformity mecha-
nism cannot explain the observation by Granzier and
Pollack (1989) that force depression is virtually identical
for "xed end and sarcomere length controlled contrac-
tions in isolated frog skeletal muscle "bres. Also, sar-
comere length non-uniformity has been suggested as
664 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668

Fig. 7. Relationship between the amount of depressed force as a function of the work performed by the muscle during the shortening phase for all six
muscles (a}f) and for all six experimental conditions (the di!erent symbols are associated with the di!erent experimental conditions). Also, shown is the
best-"tting power function that goes through the origin (0, 0) of the force}work plot and the corresponding coe$cient of variation, R.

a mechanism for force enhancement following eccentric
contractions (Edman and Tsuchiya, 1996), and it is hard
to reconcile how a single mechanism may explain two
opposite phenomena (force depression and force en-
hancement) simultaneously.
Recently, it has been found that actin myo"laments are
compliant (Goldman and Huxley, 1994; Higuchi and
Goldman, 1995; Huxley et al., 1994; Kojima et al., 1994;
Wakabayashi et al., 1994) and may account for as much
as 50}70% of the total sarcomere compliance (Daniel
 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668
Fig. 8. Relationship between the amount of depressed force as a func-
tion of the work performed. Data from all six muscles and all six
experimental tests were combined, illustrating the similarity of the
relationship across muscles and tests. Note, that neither the force nor
the work data were normalized in any way, suggesting that the relation-
ship holds independent of the size or maximal force of the muscle.
et al., 1998). Furthermore, it has been demonstrated that
actin "laments undergo the largest deformation in the
non-overlap region (Forcinito et al., 1997) and that
these deformations cause an altered orientation between
the actin-binding sites and the myosin cross-bridges
which may alter the probability of cross-bridge attach-
ment (Daniel et al., 1998). Therefore, the idea that
force depression is related to a stress-induced cross-
bridge inhibition in the actin}myosin overlap zone that
is formed during shortening (Herzog and Leonard,
1997; Herzog, 1998; MareH chal and Plaghki, 1979)
becomes a distinct possibility. This mechanism requires
that force depression following muscle shortening is inde-
pendent of the speed of shortening, but depends both on
the amount of shortening and the force during shorten-
ing. Furthermore, this mechanism requires that force
depression is a process that starts immediately upon
shortening and is cumulative during the shortening
phase. All these requirements suggest that there must be
a direct and strong link of force depression with muscular
work.
The way force depression may be associated with the
amount of work performed by the muscle through actin
"lament deformation may be explained as follows. Work,
by de"nition, is the area under the force}displacement
curve (Fig. 10). It is a function of the amount of force and
shortening the muscle undergoes. The higher the force for
a given amount of shortening, or the larger the amount of
shortening for a given force, the larger the mechanical
work produced. If the amount of shortening is constant,
one would expect the newly formed overlap zone be-
tween actin and myosin "laments to be constant, on
average (Fig. 11). For this situation, force depression just
665
depends on the amount of force during shortening. The
higher the force, the greater the elongation of the actin
"lament, and the larger the force depression, presumably
because of the increased actin "lament deformation. If
the amount of force is constant, the work performed
depends exclusively on the amount of shortening. For
this situation, the larger the amount of shortening, the
greater the newly formed overlap zone (Fig. 11, n) and the
larger the force depression, presumably because there is
an increased area of overlap between actin and myosin
"laments containing deformed actin "laments with a re-
duced probability for cross-bridge attachments com-
pared to the same zone of overlap in an isometric
contraction.
Although it has been mentioned anecdotally that
work might be related to force depression (De Ruiter
et al., 1998; Granzier and Pollack, 1989; MareH chal
and Plaghki, 1979), no systematic study has ever been
performed to explore the possible relationship between
these two variables. Here, we show that the work per-
formed by the muscle explains virtually the entire vari-
ation in force depression for a variety of di!erent tests
(Figs. 7 and 8). In addition, the relationship between the
absolute force depression and the absolute work across
all six muscles was similar (Fig. 8), resulting in a strong
relationship across all muscles and all tests (r"0.85).
We conclude from these results that work is a good
descriptor of force depression, that force depression is
likely not related to the speed of shortening but rather to
the force or work that is changed with changing speeds
(Fig. 9), and that work might not only be a descriptor of
force depression but may point directly to an underlying
mechanism.
If force depression following muscle shortening is
caused by stress-induced inhibition of cross-bridge at-
tachments in the newly formed actin}myosin overlap
zone, two so far untested hypotheses should hold: (1)
muscle in the force-depressed state should be less sti!
than muscle in the normal isometric state. Although,
such a correlation between sti!ness and depressed force
has been demonstrated in an earlier study on frog "bre
preparations (Sugi and Tsuchiya, 1988), single "bres
might behave distinctly di!erent than whole mammalian
skeletal muscle. (2) Force depression should be abolished
instantaneously if the stress in the muscle is completely
released. Although such stress release experiments have
been performed in the past (Abbott and Aubert, 1952; De
Ruiter et al., 1998; Herzog and Leonard, 1997), they were
accomplished by interrupting the stimulation during the
force-depressed phase. Therefore, one might argue that
the recovery of force was related to the interruption of
stimulation and the muscle's reactivation after a su$-
cient period of time, rather than the release of force
accompanying the interruption of stimulation. In order
to test whether a stress release without a change in
activation will abolish force depression, a muscle should
666 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668

Fig. 9. Relationship between the amount of depressed force as a function of the speed of shortening for all six muscles (a}f) and for the tests shown in
Figs. 1 and 2 (diamonds and squares, respectively), plus a series of shortening contractions performed at di!erent constant speeds and a shortening
distance of 4 mm (crosses). Note, that the amount of depressed force varies greatly for a given speed of shortening when the work produced during
shortening is di!erent (vertical dashed line), and further, how di!erent speeds of shortening (4}256 mm/s) result in similar force depression when the
work performed during shortening is similar (horizontal dashed line).

be shortened "rst to produce force depression. Then, the at zero external force, and so, no stress on the contractile
muscle should be allowed to shorten a second time at myo"laments. Based on the stress-dependent cross-
a speed exceeding its maximal speed of shortening, there- bridge inhibition mechanism, the force depression that
fore, most of the second shortening contraction is made was present following the "rst shortening contraction
 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668
Fig. 10. De"nition and calculation of work. Work was calculated as the
area under the force}displacement curve.
should be abolished completely following the second
shortening contraction.
Summarizing, this study provides "rst systematic evid-
ence that force depression is largely accounted for by the
mechanical work performed by the muscle during
shortening (Figs. 7 and 8). Furthermore, the results of this
study indicate that force depression following shortening
at a wide variety of speeds is the same provided the work
performed during the shortening phase is about the same
(Fig. 9). Finally, the muscle work during the shortening
phase may directly point to a mechanism of force de-
pression based on cross-bridge inhibition caused by
mechanical deformation of myo"laments. Although the
present study cannot rule out the most popular mecha-
nism associated with force depression: sarcomere length
non-uniformity; it provides support for a mechanism
based on myo"lament deformation (Herzog and Leo-
nard, 1997; Herzog, 1998; MareH chal and Plaghki, 1979)
and the associated change in three-dimensional geometry
of cross-bridge attachment (Daniel et al., 1998).
Acknowledgements
This study was supported by an operating grant from
the Natural Sciences and Engineering Research Council
of Canada (NSERC), and a postdoctoral fellowship grant
for J.Z. Wu from the Alberta Heritage Foundation for
Medical Research (AHFMR).
References
Abbott, B.C., Aubert, X.M., 1952. The force exerted by active striated
muscle during and after change of length. Journal of Physiology
117, 77}86.
667
Fig. 11. Schematic illustration of the increase in thick/thin myo"lament
overlap during shortening. The unlabelled arrow represents the com-
mon overlap zone at the long and short sarcomere length. The arrow
labelled `na represents the new overlap zone between a thick/thin
myo"lament pair which was formed because of sarcomere shortening.
Ariano, M.A., Armstrong, R.B., Edgerton, V.R., 1973. Hindlimb muscle
"ber populations of "ve mammals. Journal of Histochemistry and
Cytochemistry 21 (1), 51}55.
Daniel, T.L., Trimble, A.C., Chase, P.B., 1998. Compliant realignment
of binding sites in muscle: transient behaviour and mechanical
tuning. Biophysical Journal 74, 1611}1621.
De Ruiter, C.J., de Haan, A., Jones, D.A., Sargeant, A.J., 1998. Shorten-
ing-induced force depression in human adductor pollicis muscle.
Journal of Physiology 507 (2), 583}591.
Edman, K.A.P., Caputo, C., Lou, F., 1993. Depression of tetanic force
induced by loaded shortening of frog muscle "bres. Journal of
Physiology 466, 535}552.
Edman, K.A.P., Tsuchiya, T., 1996. Strain of passive elements during
force enhancement by stretch in frog muscle "bres. Journal of
Physiology 490 (1), 191}205.
Forcinito, M., Epstein, M., Herzog, W., 1997. Theoretical consider-
ations on myo"bril sti!ness. Biophysical Journal 72, 1278}1286.
Goldman, Y.E., Huxley, A.F., 1994. Actin Compliance: are you pulling
my chain?. Biophysical Journal 67, 2131}2136.
Granzier, H.L.M., Pollack, G.H., 1989. E!ect of active pre-shortening
on isometric and isotonic performance of single frog muscle "bres.
Journal of Physiology 415, 299}327.
Herzog, W., 1998. Muscle modelling. Journal of Electromyography
Kinesiology 8, 59}60.
Herzog, W., Leonard, T.R., 1997. Depression of cat soleus forces follow-
ing isokinetic shortening. Journal of Biomechanics 30 (9), 865}872.
Herzog, W., Leonard, T.R., Stano, A., 1995. A system for studying the
mechanical properties of muscles and the sensorimotor control of
muscle forces during unrestrained locomotion in the cat. Journal of
Biomechanics 28, 211}218.
Higuchi, H., Goldman, Y.E., 1995. Sliding distance per ATP molecule
hydrolyzed by myosin heads during isotonic shortening of skinned
muscle "bers. Biophysical Journal 69, 1491}1507.
Hill, A.V., 1938. The heat of shortening and the dynamic constants
of muscle. Proceedings of the Royal Society of London, pp.
136}195.
Huxley, H.E., Stewart, A., Sosa, H., Irving, T., 1994. X-ray di!raction
measurements of the extensibility of actin and myosin "laments in
contracting muscles. Biophysical Journal 67, 2411}2421.
Kojima, H., Ishijima, A., Yanagida, T., 1994. Direct measurement of
sti!ness of single actin "laments with and without tropomyosin by
in vitro nanomanipulation. Proceedings of the National Academy
of Sciences USA 91, 12962}12966.
MareH chal, G., Plaghki, L., 1979. The de"cit of the isometric tetanic
tension redeveloped after a release of frog muscle at a constant
velocity. Journal of General Physiology 73, 453}467.
668 W. Herzog et al. / Journal of Biomechanics 33 (2000) 659}668

Meijer, K., Grootenboer, H.J., Koopman, B.F.J.M., Huijing, P., 1997.
Fully isometric length}force curves of rat muscle di!er from those
during and after concentric contractions. Journal of Applied Bio-
mechanics 13, 164}181.
Sugi, H., Tsuchiya, T., 1988. Sti!ness changes during enhance-
ment and de"cit of isometric force by slow length changes in
frog skeletal muscle "bres. Journal of Physiology 407,
215}229.
Wakabayashi, K., Sugimoto, Y., Tanaka, H., Ueno, Y., Takezawa, Y.,
Amemiya, Y., 1994. X-ray di!raction evidence for the extensibility of
actin and myosin "laments during muscle contraction. Biophysical
Journal 67, 2422}2435.