Eur J Appl Physiol (1995) 70:379 386

© Springer-Verlag 1995

C. Bosco • A. Belli • M. Astrua • J. Tihanyi R. Pozzo - S. Kellis • O. Tsarpela • C. Foti R. Manno • C. Tranquilli

A dynamometerfor evaluation of dynamic muscle work

Accepted: 14 October 1994

Abstract The validation of a new dynamometer for evaluation of dynamic muscle work is presented. The device was based on a precise measurement of load displacements of any machine using gravitational loads as external resistance. It allowed, through a sensor consisting of an infrared photo interrupter, the calculation of velocity, force and power during concentric, eccentric and stretch-shortening cycle activity. To validate the dynamometer 33 male and female track and field athletes (12 throwers and 21 jumpers) participated in the study. The throwers (4 women and 8 men) were asked to perform half-squat exercises on a slide machine with a load of 100% of the subject's body mass. The day-to-day reproducibility of half-squat exercises gave a correlation coefficient of r = 0.88, 0.97 and 0.95 for average push-off force (AF), average push-off velocity (AV), and average push-off power (AP) respectively. Comparison of half-squat measurements was performed against jumping and running test evaluation by

C. Bosco - C. Foti - R. Manno Department of Physical Medicine and Rehabilitation, Faculty of Medicine and Surgery, University of Roma Tor Vergata, Rome, Italy A. Belli Laboratoire de Physiologie GIP Exercice, Facult6 de M~decine de Lyon-Sud, Lyon, France M. Astrua • C. Bosco Italian Track and Field Association - F.I.D.A.L., Rome, Italy C. Bosco ( ~ ) . J. Tihanyi Department of Exercise Physiology, Hungarian University of Physical Education, Alkotasu, 44-1123 Budapest, Hungary R. Pozzo Institute for Biomechanics DSHS, K6hn, Germany S. Kellis. O. Tsarpela Department of Physical Education and Sport Science, Aristotle University of Thessaloniki, Greece C. Tranquilli Institute of Sport Science

C.O.N.I., Rome, Italy

the jumpers (7 women and 14 men). The interrelationships among the different variables studied demonstrated a strong correlation between AF, AV and AP and sprinting and jumping parameters (r = 0.53-0.97; P < 0.05-0.001). Using values of AF, AV and AP developed in half-squat exercises executed with different loads, ranging from 35% to 210% of the subject's body mass, it was also possible to establish the force-velocity and power-velocity relationships for both male and female jumpers. In any individual case, the maximal error due to the measurement system was calculated to be less than 0.3%, 0.9% and 1.2% for AF, AV, and AP respectively. Given the accuracy of the ergometer, the high reliability found between 2 days of measurements, and the specificity of the results it is suggested that the dynamic dynamometer would be suitable for evaluation of athletes performing specific skills. In addition, because single and multiple joint movements involving appropriate muscle groups can be easily performed, physiological characteristics could be evaluated for both athletic and rehabilitation purposes. Therefore, because of its simplicity of use and application, and its low cost the dynamometer would be suitable for both laboratory and field conditions. Key words Dynamic d y n a m o m e t e r . Gravitational loads • Force-velocity relationship • Muscle power • Muscle sex difference

Introduction Evaluation of muscle behaviour has been extensively performed using in the main isometric or isokinetic dynamometers. A disadvantage of the isometric dynamometer is that neither work nor power can be measured. On the other hand, even if isokinetic measurements offer safety when testing for clinical purposes (e.g. rehabilitation, post trauma post surgery, etc.), the diagnosis reflects the nature of motion imposed. In

380

contrast, only few devices have been described for measurements of muscle strength, work and power under isotonic or ballistic conditions. These apparatus have used, as external resistance, inertia wheels (e.g. Kaneko 1970), inertia loads (e.g. Tihanyi et al. 1982; Avis et al. 1985), wheels and loads (De Koning et al. 1982). In addition, exercises executed with own body mass (e.g. Cavagna et al. 1968) or with extra loads, have been performed on force platforms (e.g. Bosco and Komi 1979a). However, although these devices and methods have been very practical offering good opportunities for studying natural muscle activity they have not been extensively used. With these objections in mind, a dynamometer was constructed for measuring muscle work during both concentric and eccentric activity which can be used with any weight lifting equipment. This paper presents a preliminary investigation on athletes, which has been conducted to study the accuracy and the reproducibility of measurements during strength training exercises, as well as the specificity of the measurements against basic muscle evaluations as in jumping and running.

The microprocessor worked internally with a 10-bts time resolution. When the loads were moved by the subjects the signal from the optical transducer interrupted the microprocessor every 3 mm of displacement. Thus, it was possible to calculate velocity, acceleration, force, power, and work corresponding to the load displacements. The velocity (v, in metres per second) during an elementary measurement of displacement (Ad = 3 mm) was calculated using the following equation: v = kd x At-1

(1)

where Ar is the time (in seconds) to perform the elementary displacement with a resolution of 10 ps. The acceleration (a, in metres per second squared) was derived as follows: a - Av x At- 1

(2)

where Av is the difference between the velocity of the considered elementary displacement and the velocity of the preceding elementary displacement. The corresponding force (F, in Newton) was calculated as follows: F = (mxg) + (mx a)

(3)

where m is the mass of the load and g is the acceleration due to gravity (9.81 m s-2). Finally, the corresponding mechanical power (P, in watts) was computed as follows: P = F x v.

Methods Apparatus The device, called Ergopower (Ergotest Technology A.S., Langensund, Norway), was based on a precise measurement of the load displacements of any machine using gravitational loads as external resistance (e.g. leg press, lat machine, barbell, etc.). It is illustrated in diagramatic form in Fig. 1. The vertical displacements of the loads were monitored with simple mechanics and a sensor arrangement. The loads were mechanically linked to a shuttle which glided on a track bar. The sensor consisted of two infrared photo interrupters, locked in the shuttle, facing an optical code strip stuck to the track bar. The two outputs from the sensor were phase-shifted by 90 °, allowing the detection of the movement direction (up or down). The sensor was interfaced to an electronic device which included a microprocessor and adequate software stored in PROM.

(4)

For each exercise repetition, the average velocity (V), the average force (F) and the average power (P) were the means of all v, F and P elementary values, respectively, measured during all the time necessary to perform a complete repetition (Fig. 2).

System accuracy The relative errors (in percentages) of measured displacement (Ad%) and of measured time (At%) are respectively: Ad Ad% = - - x 100 d

(5)

At At = - - x 100 t

(6)

where Ad is the absolute error (3 mm) of the displacement measurement, At is the absolute error (0.01 ms) of the time measurement, d is the measured displacement (millimetres) and t is the measured time

Force (N),Velocity (mm.s-~),Power(w) 3000 2

0

0

0

0

~

.5

1

1:5

Time (s) Fig. 1 Diagram of the apparatus. A shuttle including infrared photo interrupter, B track bar including code strip, C Ergopower/microcomputer, D moving loads

Fig. 2 Typical example of instantaneous force, velocity and power produced by a female jumper during half squat and with a load of 140 kg

381 (milliseconds). In fact, At% is also dependent on the relative accuracy of the electronic clock, which is given by the stability of the oscillating quartz. In the present system, the quartz used (type A161A, 16MHz, IQD Ltd, Crewkerne, England) has a stability of 0.003%, the exact relative error is then: At

At% = - - x 100 + 0.003% t

(6b)

According to Eqs. 1 and 2 it is possible to calculate the relative error (as in percentage) of velocity (Av%) and of acceleration (aa%): Av%=--xl00=

+--

/)

Aa%

Aa (~ - - x 100 = a

+

A~)

xl00=kd%+kt%

(7)

x 100 = Ad% + 2At%

(8)

Assuming that the error on the gravity acceleration estimation is negligible, the relative error of force (AF%) is:

(

AF% = A m + \m

100=Am%+ 9+a]

(

Aa% x g

1)

(9)

a+l

where m is the mass, Am and Am% are respectively the absolute and the relative error of mass, a is the acceleration measured by the Biorobot and g is the acceleration of the gravity (9.81 m. s 2). In the present study, the mass used (barbell and weights) were always in conformity with the regulation of the International Weightlifting Federation which requires a Am% of 0.04%. Finally, the relative error of power (AP%) is: AP% = AF% + kv%

(10)

Calibration procedure Since the instrument was designed to be applicable to different types of apparatus using moving loads which are mechanically guided, the friction losses of these external systems must be assessed. This was possible by allowing the system to "free fall" with a known mass and then by calculating the friction force (Ff) as follows:

Ff=mf(g-a)

(11)

where m is the falling mass, g is the acceleration of gravity (9.81m's 2) and a is the measured acceleration. In the present experiment, the Ff calculated in five different trials was 8.4 (SD 1.4) N. This Ff was added during the concentric part (raising of the loads) and subtracted during the eccentric part (falling of the loads) of the movement. Finally, the mass of the shuttle was measured (0.19 kg) and added to the mass of the moving load for the computations.

Table 1 Anthropometric characteristics of the subject groups

Subjects A group of 33 track and field athletes, divided into two groups (12 throwers and 21 jumpers), volunteered to participate in the study. The physical characteristics of the subjects are given in Table 1. All the athletes were fully motivated and well accustomed to the development of maximal effort during the test.

Measurements The throwers reported twice to the laboratory. On the first visit several sub-maximal and maximal dynamic half-squat exercises were performed on a slide machine (guided horizontal barbell) with an extra load similar to the subject's body mass. The body position was standardised and the number of trials needed for stable measurements was studied. Two or three trials were sufficient for reaching a plateau in performance. The next two consecutive trials were used for the study of the trial reproducibility. On the 2nd day the procedure was the same procedure as the st day, and at almost the same time of day. All the performances were recorded with the dynamometer, and the best trial of each day was used for determining the reproducibility during 2 different days (Tornvall 1963). The jumpers were tested in two different periods at 3-day intervals. In the first evaluation the athletes performed the following jump tests on a resistive platform (Bosco et al. 1983a) connected to a digital timer (accuracy _+0.001 s) (Ergojump Psion CM, MA.GI.CA., Rome, Italy): a squat jump (SJ), a counter movement jump (CMJ) and a squat jump executed with an extra load (barbell on the shoulder) similar to the subject's body mass (SJbm)- The SJ performed with and without extra loads were executed starting from knee joint pre-set at 90 °. No allowance for preparatory counter movement was accepted. In CMJ, the subject started from an erect standing position on the resistive platform and the end of the eccentric phase corresponded to the starting position of SJ. The rise of the center of gravity above the ground (h in metres) in SJ, SJbm and CMJ were measured from flight time (tf in seconds), applying ballistic laws: h = t 2 x g x 8 -1

(12)

where 9 is the acceleration of gravity (9.81 m" s-Z). To avoid unmeasurable work, horizontal and lateral displacements were minimised, and the hands were kept on the hips. In addition to the jumping tests, the time to run at maximal speed a 30-m dash was recorded with an electronic timer ( _+ 0.001 s). Then 3 days after the first measurement, the jumpers reported to the laboratory for measurement during half-squat exercises performed with loads ranging from 35% to 210% of the subject's body mass. The best trial of three measurements of each load was used for statistical analysis.

Subject groups

Age (year)

Body mass (kg)

Height (cm)

Mean Throwers

Men (n = 8) Women (n = 4)

20.6 20.5

SD

Mean

SD

Mean

SD

2.4 1.1

82.0 59.0

9.7 3.1

185.3 170.7

6.9 4.1

Jumpers

Men (n = 14) Women (n = 7)

21.6 22.1

4.6 2.6

70.7 60.3

5.7 3.9

183.5 172.6

5.1 5.4

382 StatisticaI analysis Ordinary statistical methods were employed, including means (x), standard deviation (SD) and standard error (SEM). The Pearson product moment correlation coefficient (r) was used for test re-test measurements reproducibility and to show the relationship among selected variables. The coefficient of variation (CV) of test re-test measurements was calculated using the following equation (Torstenson 1976):

CV =

SD 2 0 0 x m - ~ x ( x l + x 2 ) -*

(13)

,/2/

where xl and x2 are the mean average values of the two successive measurements and SD is the standard deviation of the mean difference between two measurements. To determine the possible differences between test re-test measurements Student's t-test for paired observations was used, while unpaired observations were used between the parameters for men and women. The probability level was set a priori at 0.05 to determine statistically significant differences.

0.5 s and the mean acceleration was always smaller than 4 m' s -2. Thus, applying Eqs. 5 10, the highest errors were calculated to be: Ad% =0.857%, At% = 0.005%, Av% = 0.863%, Aa% = 0.867%, AF = 0.291% and AP% = 1.158%.

Reproducibility Table 2 gives the x, SD, r, CV, of the same day and day-to-day reproducibility of F, v, P and range of motion measured in 12 throwers. These athletes were tested during half-squat exercises performed with a load similar to the subject's body mass. The CV ranged from 1.6% to 8.0% for same day and from 1.4% to 5.2% for separate day measurements. The r ranged from 0.57 to 0.84 for same day and from 0.85 to 0.97 for separate day measurements. No significant t-test values were observed between 2-day performances or between the same day evaluations.

Results Specificity System accuracy In the whole of the present study, during a single repetition, the mean displacement was always greater than 0.35 m, the working time was always greater than

The x values and SD of selected parameters of half squat (performed with a load similar to the body mass (rob) of the subjects), running and jumping tests performed by 21 jumpers are presented in Table 3. An

Table 2 Same day and day-to-day reproducibility of average force (F), average velocity (V), average power (P) and range of motion (AZ) in half-squat exercise performed by 12 throwers with a load similar to the subject's body mass. r Pearson product moment correlation coefficient, ns not significant, CV coefficient of variation for repeated measurements (see Eq. 13) Variable

Different trials Trial 1

F (N.kg t) V (m's 1) P (W-kg -1) zXZ (cm)

Different days Trial 2

Mean

SD

Mean

SD

24.8 0.96 24.0 36.8

0.7 0.1 2.5 4.7

24.9 0.97 25.3 36.1

0.9 0.1 2.1 4.3

r

CV

0.76*** 0.82*** 0.84*** 0.57ns

1.6 4.4 5.0 8.0

Day 2

Day 1 Mean

SD

Mean

SD

24.7 0.96 24.0 36.3

0.9 0.1 3.1 4.1

25.0 0.96 24.2 35.4

0.8 0.1 3.0 4.1

r

CV

0.88*** 0.97*** 0.85*** 0.85**

1.4 2.3 3.7 5.2

**P < 0.01, ***P < 0.001

Table 3 Values and corresponding correlation matrix of selected parameters from jumping test (SJ squat jump, SJbm squat jump with extra load similar to subject's body mass, CMJ counter movement jump), half squat performed on slide machine (F average force, v average velocity, P average power) and velocity in 30-m spring (30 m) studied in jump (athletes (n = 21) Variables

Mean

SD

Correlation

SJ SJ (cm) SJbm (cm) CMJ (cm) F (N-kg -1) v ( m . s -1) P (W.kg -I) 30m(s)

42.7 16.2 48.4 24.4 1.03 25.5 3.96

*P < 0.05, **P < 0.01, ***P < 0.001

4.8 2.8 4.4 1.5 12 4.5 22

0.90*** 0.93*** 0.78*** 0.76*** 0.80*** - 0.97***

Matrix

SJbm

0.89*** 0.74*** 0.53* 0.63** -0.91"**

CMJ

0.83*** 0.74*** 0.80*** 0.93***

F

0.79*** 0.90*** -0.83***

v

0.97*** --0.77***

P

0.83***

383

analysis of the interrelationship among the different variables measured for the jumpers, demonstrated a strong correlation between power output, developed in half-squat exercise, and the results of both sprinting time (r = - 0 . 8 3 , P < 0.001) and explosive jumping tests represented by SJ and CMJ (r = 0.80, P < 0.001) (see Table 3).

-2B

~5• i

30.

-24

Z v

£12

C~ tad

.

25.

o

?¢

-20

i,i

< :>

<

Force velocity relationship The mean values of F, P, and v reached in half-squat exercises executed with different loads (from 35% to 210% of the subject's rob) are presented for both male (n = 14) and female (n = 7)jumpers in Table 4. The F in the male group demonstrated significant differences (P < 0.01 for 210% of subject's rob, P < 0.001 for the other loads) than the female counterpart• Similar differences (P < 0.001) were observed also for v and P developed with light and submaximal load (30%-140% of subject's rob) , while no significance was noted with heavy loads (210% of the subject's rob). Using the same

Table 4 Values of average force (F35 to F21o), average velocity (v3s to Vzlo) and average power (P35 to P21o) developed by jump athletes during half-squat exercises on a slide machine, with extra loads ranging from 35% to 210% of the subject's body mass respectively. Results of unpaired students t-test comparison between men and women are given with corresponding significance level; ns not significant Variables

Men (n = 14)

Women (n = 7)

Mean

Mean

SD

16.7 20.3 23.1 23.9 27.1 30.2 33.6

0.9 1.0 0.7 1.0 1.1 1.0 1.4

4.78*** 4.33*** 4.86*** 3.86*** 4.19"** 4.18'** 3.47**

1.12 1.00 0.92 0.90 0.78 0.70 0.61

0.08 0.07 0.05 0.04 0.06 0.05 0.05

3.81"** 4.96*** 4.79*** 4.40*** 4.15"** 2.32* 0.32 ns

18.7 20.4 21.4 21.6 21.5 21.1 20.5

1.6 1.9 1.3 1.3 2.4 1.5 1.7

4.84*** 4.76*** 5.51"** 4.33*** 4.28*** 3.04** 2.06ns

SD

t-test

Average force (N" k g - 1) F35 FTo Floo Flos Fl~o F175 F21o

19.0 22.6 25.2 26.2 29.5 32.8 36.1

1.2 1.4 1.3 1.5 1.5 1.8 1.8

i

z2~

LIA tD

20

16 ]5

O

o16

"

o18

'

110

'

1'.2

'

< ¢¢ w >

1.4

AVERAGE VELOCITY ( m . s - I )

Fig. 3 Average force (F), (squares) and average powr (P) (dots), developed during half-squat exercises performed with various loads (from 35% to 210% of the subject's body mass), are shown according to the average vertical velocity (V) for male (filled symbols) and female (open symbols) jumpers. The statistical sex differences for F, P and V values together with the different loads used are shown in Table 4

variables it was possible to establish the F-v and v - P relationships for both male and female jumper groups as shown in Fig. 3.

Discussion

System accuracy In any single case the maximal error due to the measurement system was less than 0.3%, 0.9% and 1.2% for F, v and P, respectively. These maximal errors were equivalent or lower than errors generally encountered in kinetic measurements (Belli et al. 1992). Furthermore, these errors were lower than the mean differences observed among subjects and among male and female groups.

Reproducibility

Average velocity (m-s 1) V3s VTo Vloo vlo5 v14o

vl~s Vzlo

1.31 1.19 1.09 1.05 0.93 0.77 0.62

0.16 0.13 0.11 0.11 0.10 0.10 0.13

Average power (W" k g - 1) P35 PTo Ploo P~o5 Pl~o Plus P21o

25.0 26.7 27.6 27.8 27.3 25.0 22.6

3.9 4.2 3.8 5.1 3.6 4.3 5.4

*P < 0.05, **P < 0.01, ***P < 0.001

A basic requirement of any test is that repeated measurements yield consistent results. Compared to the reproducibility obtained in other tests (Thorstensson 1976; Avis et al. 1985) the reproducibility of two consecutive half-squat trials, performed on the same day, was good but not very high (Table 2). This would suggest that at least two or three trials, after reaching a plateau in performance, should be recommended for consistent evaluation (e.g. Haffajee et al. 1972). This notion is supported by the high reproducibility found between day-to-day measurements in which the best trial of each day was analysed. It should be noted that, although the size of the population examined was small (12 throwers), the r of the variables studied (F, v, P) were relatively high (r = 0.85-0.97). On the other hand,

384

the homogeneity of the population favours the possibility of obtaining a low CV. Specificity The most common methods of testing strength and power are weight-lifting, isometric, isokinetic and jumping tests. The most important criterion in selecting a test is specificity (Sale and MacDougall 1981). Since half squat represents a basic exercise used in leg strength training programme, it was chosen to study the mechanical behaviour of the muscles involved. Furthermore, it was largely used in the training programme of the subjects engaged in the present experiment. Therefore, even if weight lifting exercises, like half squat, are considered to have some limitations because the maximal resistance that can be set is restricted to F at the weakest point (e.g. Sale 1991), the muscle activation which can be realised with this exercise seems to be connected to explosive strength characteristics. The high resolution of the dynamometer used in the present experiment gave the possibility of obtaining instantaneous values of F, P and v (Fig. 2). However, average values per push-off were calculated and used for analysis, because they represent the most realistic expression of mechanical parameters since they respect the contraction cycle of each muscle group (Williams et al. 1988). Moreover, P calculated in this way has provided the best correlation with muscle effort and performance in models of cycling and running (Andrew 1983). In the present study, the averaged mechanical variables studied in half squat (F, v, P) demonstrated a remarkable relationship with sprinting and jumping performances (Table 3). Thus, since it has been previously shown that there is a positive relationship between sprinting (Mero et al. 1981) and jumping (Bosco and Komi 1979b) performances with the percentage of fast twitch (FT) fibres in the vastus lateralis muscle, it is likely that phasic motor units might be largely recruited also in half-squat exercise. This hypothesis is supported also by the fact that the linear extension v during half squat, performed with a load similar to the subject's rob, reached maximal values of 2 m' s-1 corresponding to a knee angular velocity of about 4 rad-s 1, no different from that observed by Bosco et al. (1982) in ballistic activity like SJ. A large involvement of FT fibers has been shown to occur also during isokinetic work performed at a knee angular velocity of ~zr a d ' s - 1 (e.g. Thorstensson et al. 1977) or higher (Coyle et al. 1979). Thus, there are strong indications to suggest that the pattern of motor recruitment, at high knee angular velocity, is v dependent regardless of the muscle activity involved (isokinetic vs ballistic motion) (Bosco et al. 1983b). A specific study is in progress to validate this.

Force-velocity relationship and sex difference Dynamic muscle characteristics of leg extensors recorded during natural conditions like half-squat exercise allowed us to study the force-velocity (F-v) and power-velocity (P-v) relationship in intact human skeletal muscles. The F-v and P-v relationships obtained in the present study were similar to that obtained with isokinetic leg extensor dynamometers (e.g. Perrine and Edgerton 1978). In addition to its simplicity, the advantage of the present ergometer consisted in the possibility of obtaining measurements with high contraction velocities in ballistic and in stretch-shortening cycle conditions. The F-v and P-v relationships of male jumpers were shifted to the higher values of F and P respectively compared to their female counterparts (Fig. 3). These sex differences, in general, are in agreement with those previously reported, for instance, by Bosco et al. (1986) for knee extensor muscles of sprinters, and by Kaneko (1970) for arm flexors studied in the Japanese. However, Kaneko (1970) found that the difference between sexes appeared to be more in F than in v, while the present results showed a significant difference (P < 0.01) in F and no significant differences in v and P when heavy loads (210% subject's rob) were analysed. Furthermore, the greatest difference between men and women was observed for F, v and P when light loads were used (Table4) and the men:women P ratio decreased linearly as a function of the loads (P = 0.001, see Fig. 4). The discrepancy between the present results and the findings of Kaneko (1970) seems apparent only because in the present experiments the results are presented according to the subject's mb and in Kaneko's studies in absolute values. It is worth noting, that the results of the present experiment are supported by Anderson et al. (1979). In fact, it was shown that in isometric and low speed isokinetic contractions, the values were similar in both men and women, when expressed for lean mb. In contrast, at a constant knee angular velocity of rc radian per second, men were significantly stronger than their female

PowerRatio (Men:womenxlO0) 140 120

i

~

p = 0,001 r = 0.95

1000

•

1()0 2()0 Load(%of BodyMass)

Fig. 4 Powerratio (men:womenin percentages)foundin half-squat exercise according to the loads used (from 35% to 210% of the subject's body mass, n = 7)

385

counterparts. In the light of the above observations and according to the present findings, it would appear that the maximal knee extension v is more important than maximal F in characterising sex differences. An explanation for these results can be sought in different biological effects caused by heavy resistance training in men and women. During this kind of training, an increase of serum testosterone has been shown in men and not in women (e.g. Weiss et al. 1983). Therefore, considering that all the subjects engaged in the present study, had been participating for more than 5 years in explosive and strength training programmes, it is likely that adaptive changes to training of sex hormones were more pronounced in male compared to female athletes. It has been found that the human maximal strength is determined by the cross-sectional area of the muscle (Rodhal and Howarth 1962) and by neural factors (Milner-Brown et al. 1975). More than an anabolic agent, it has been suggested that testosterone may cause a dramatic potentiation effect on neural factors and may favour the transition of type II fibres to more glycolitic profiles (for a review see Kraemer 1992). In this connection, a positive relationship has been observed, in professional football players, between the serum testosterone concentrations and maximal knee extension velocities reached in CMJ (Bosco 1993). Therefore, it can be speculated that the sex difference observed for maximal knee extension v might be strongly influenced by hormone differences. However, hormone, neurone and mechanical measurements should be combined in future experiments to validate these hypotheses.

Conclusion Given the accuracy of the ergometer, the high reliability found between 2 days of measurements (r = 0.85 0.97) and the specificity of the results, it is suggested that the dynamic dynamometer is suitable for evaluating athletes performing specific skills. These can be performed using any muscle machine which uses gravitational force as the external resistance and during the stretch-shortening muscle activation which represents the most natural pattern of muscle work. As pointed out by Sale and MacDougall (1981) the most unbiased monitoring of training occurs when the same regimen (same equipment and same movement patterns) is used for both training and testing. The present dynamometer fits well with these requirements. In addition, it is possible to perform specific tests allowing the construction of F-v and P-v relationships and the athletes' profile characteristics. A training programme is thus easy to monitor using the dynamic dynamometer and appropriate alterations can be made to a training plan on the basis of test results. Finally, because single or multiple-joint movements involving appropriate muscle groups can be easily performed, evaluation

of physiological characteristics can be realised for both athletic profiles and rehabilitation diagnoses. If preinjury F and P data are available for an athlete, the extent of the decrease in P due to the injuries can it has been shown, be quantified (Nicholas 1984). Therefore, because of its simplicity in use and application, and its low cost the dynamometer would seem suitable for use in both the laboratory and the field. Acknowledgement The authors wish to acknowledge J. Carew for reviewing the English manuscript.

References Anderson M, Cote R, Coyle E, Roby F (1979) Leg power, muscle strength, and peak EMG activity in physically active college men and women. Med Sci Sports 11:81 82 Andrew JG (1983) Biomechanical measures of muscular effort. Med Sci Sports Exerc 15:199-207 Avis FJ, Hoving A, Toussaint HM (1985) A dynamometer for the measurement of force, velocity, work and power during an explosive leg extension. Eur J Appl Physiol 54:210-215 Belli A, Rey S, Bonnefoy R, Lacour J-R (1992) A simple device for kinematic measurements of human movement. Ergonomics 35:177-186 Bosco C (1993) Test di valutazione della donna nella pratica del giuoco del calcino. In: Cambi R, Paterni M (eds) II calcio femminile, a spetti medici e tecnici. Atti del Convegno Nazionale, Figc Publisher, Rome, pp 219-230 Bosco C, Komi PV (1979a) Potentiation of mechanical behaviour of the human skeletal muscle through prestretching. Acta Physiol Scand 106:467-472 Bosco C, Komi PV (1979b) Mechanical characteristics and fiber composition of human leg extensor muscles. Eur J Appl Physiol 41:275-284 Bosco C, Viitasalo J, Komi PV, Luhtanen P (1982) Combined effect f elastic energy and myoelectrical potentiation during stretchshortening cycle exercise. Acta Physiol Scand 114:543 550 Bosco C, Luhtanen P, Komi PV (1983a) A simple method for measurement of mechanical power in jumping. Eur J Appl Physiol 50:273 282 Bosco C, Mognoni P, Luhtanen P (1983b) Relationship between isokinetic performance and ballistic movement. Eur J Appl Physiol 51:357 364 Bosco C, Rusko H, Hirvonen J (1986) The effects of extra loads conditioning on muscle performance in athletes. Med Sci Sports Exerc 18:415 419 Cavagna GA, Dusman B, Margaria R (1968) Positive work done by a previously stretched muscle. J Appl Physiol 24:21 32 Coyle EF, Costill DL, Lesmes GR (1979) Leg extension power and muscle fiber composition. Med Sci Sports 11 : 12-15 De Koning FL, Binkhorst RA, Visses ACA, Vos JA (1982) Influence of static strength training on force-velocity relationship of arm flexors. Int J Sports Med 3:25 28 Haffajee D, Mortiz U, Svantesson G (1972) Isometric knee extension strength as a function of joint angle, muscle length and motor unit activity. Acta Orthop Scand 43:138-147 Kaneko M (1970) The relation between force, velocity and mechanical power in human muscle. Res J Phys Educ 14:143-147 Kraemer WJ (1992) Hormonal mechanisms related to the expression of muscular strength and power. In: Komi PV (ed) Strength and power in sport, Scientific Publications, Oxford, pp 64 67 Mero A, Luhtanen P, Viitasalo JT, Komi PV (1981) Relationship between the maximal running velocity, muscle fiber characteristics, for production of force relaxation of sprinters. Scand J Sports Sci 3:16-22

386 Milner-Brown HS, Stein RB, Lee RG (1975) Synchronistion of human motor units: possible roles of exercise and supra spinal reflex. Electroencephalogr Clin Neurophysiol 38: 245-254 Nicholas J (1984) The value of sports profiling. Clin Sport Med 3:3-10 Perrine JJ, Edgerton VR (1978) Muscle force-velocity and powervelocity relationships under isokinetic loading. Med Sci Sports 10:159 166 Rodahl K, Horwarth SM (1962) Muscle as a tissue. McGraw-Hill, New York Sale DG (1991) Testing strength and power. In: MacDougall D, Wenger HA, Green HJ (Eds) Physiological testing of high performance athlete. Human Kinetics, Champaign, Ill., pp 21-106 Sale DG, MacDougall D (1981) Specificity in strength training: a review for coach and athlete. Can J Appl Sport Sci 6:87-92

Thorstensson A (1976) Muscle strength, fiber types and enzyme activities in man. Acta Physiol Scand 98 [Suppl 443]:1-45 Thorstensson A, Larson L, Tesch P, Karlsson J (1977) Muscle strength and fiber composition in athletes and sedentary men. Med Sci Sports 9:26 30 Tihanyi J, Apor P, Fekete G (1982) Force-velocity power characteristics and fiber composition in human knee extensor muscles. Eur J Appl Physiol 48:331 343 Tornvall G (1963) Assessment of physical capabilities with special reference to the evaluation of maximal working capacity. Acta PhysioI Scand 58 [Suppl. 201]:1-101 Weiss LW, Cureton K J, Thompson FN (1983) Comparison of serum testosterone and androstenedione responses to weight lifting in men and women. Eur J Appl Physiol 50:413-419 Williams JH, Barnes WF, Signorile JF (1988) A constant load ergometer for measuring peak power output and fatigue. J Appl Physiol 65:2343-2348