ARTICLE IN PRESS

Journal of Biomechanics 43 (2010) 627–631

Contents lists available at ScienceDirect

Journal of Biomechanics
journal homepage: www.elsevier.com/locate/jbiomech
www.JBiomech.com

Vertical forces on the horse’s back in sitting and rising trot

Patricia de Cocq a,b,n, Anne Mariken Duncker c, Hilary M. Clayton d, Maarten F. Bobbert c, Mees Muller a,
Johan L. van Leeuwen a
a
Experimental Zoology Group, Animal Sciences Group, Wageningen UR, P.O. Box 338, 6700 AH Wageningen, The Netherlands
b
Animal Husbandry and Animal Health, HAS Den Bosch, P.O. Box 90108, 5200 MA ‘s-Hertogenbosch, The Netherlands
c
Research institute MOVE, Vrije Universiteit, van der Boechorstraat 9, 1081 BT Amsterdam, The Netherlands
d
Mary Anne McPhail Equine Performance Center, Department of Large Animal Clinical Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, MI 48824,
USA

a r t i c l e in fo abstract

Article history: In equestrian sports, it is generally assumed that rising and sitting trot load the horse’s back differently.
Accepted 19 October 2009 The objective of this study was to quantify the load on the horse’s back in these riding techniques.
Kinematic data of 13 riders were collected in rising and sitting trot. The time-history of the position of
Keywords: the rider’s centre of mass (CoM) was calculated, and differentiated twice to obtain the acceleration of
Equus caballus the CoM. The reaction force between the rider and the horse’s back was calculated from the
Locomotion acceleration. Forces were divided by the body weight of the rider to obtain dimensionless forces. As
Riding technique expected, the computed average vertical force did not differ between riding techniques and was not
Rising trot significantly different from the body weight of the riders. At trot, two force peaks were present during
Sitting trot
one stride cycle. Both peaks in rising trot were significantly lower compared to sitting trot (peak 1:
2.54 7 0.30 versus 2.92 7 0.29; po 0.001; peak 2: 1.95 7 0.34 versus 3.03 7 0.32; p o 0.001). This
supports the general assumption that rising trot is less demanding for the horse than sitting trot.
& 2009 Elsevier Ltd. All rights reserved.

1. Introduction be used to evaluate the loading of the horse by a rider. Since on

average the net vertical force on the back of the horse is equal to
In equestrian sports, it is generally assumed that riding technique the body weight of the rider, possible differences between rider
influences the loading of the horse’s back. An example of these riding techniques are to be found in the variation of force patterns
techniques are sitting and rising trot. In sitting trot, the rider remains around the average value on the horse’s back.
seated in the saddle. In rising trot, the rider sits in the saddle during The forces between rider and horse have been measured
half of the stride and rises from the saddle during the other half of with electronic force measuring devices that have been shown
the stride in coordination with the horse’s limb movements. The to provide valid and reliable measurements of normal
common opinion in the equestrian world is that rising trot is less forces between the saddle and the horse’s back in the standing
demanding for the horse than sitting trot. It is therefore used to train position (De Cocq et al., 2009b). Such devices have been used to
younger horses that have not yet developed sufficient muscle study the influence of the rider on a standing horse (De Cocq et al.,
strength for ridden exercise and to relax older horses in the 2009b) and to compare forces on the horse’s back during
warming-up and cooling down phases of a training session. sitting trot and rising trot (Peham et al., 2008, 2009). Although
Studies investigating the influence of the rider on the horse the latter studies did not find significant differences in
have evaluated the effect of weight (Clayton et al., 1999; De Cocq peak forces, the second peak force in rising trot, i.e. the peak that
et al., 2004; Sloet van Oldruitenborgh-Oosterbaan et al., 1995) and occurs at the moment the rider is standing, was lower than in
riding technique (De Cocq et al., 2009a; Roepstorff et al., 2009) by sitting trot.
comparing limb and back movements of the horse. Although these An alternative method of obtaining the forces exerted on the
studies provide useful information on the effects of loading, a horse by the rider, is to use rider kinematics to calculate the
direct method to evaluate the loading of the horse by a rider reaction force from horse to rider. This force will be equal but
would be preferable. Forces exerted on the horse by the rider can opposite in direction to the force of the rider on the horse’s back.
This method has, to our knowledge, not been applied in
equestrian sports but does give the opportunity to compare net
n
forces of the rider on the horse’s back between different riding
Corresponding author at: Experimental Zoology Group, Animal Sciences Group,
Wageningen UR, P.O. Box 338, 6700 AH Wageningen, The Netherlands. Tel.:
techniques. We expected to find the main differences between the
+ 31 317 482749; fax: +31 317 483955. riding techniques in the vertical force and expected that the forces
E-mail address: Patricia.deCocq@wur.nl (P. de Cocq). in forward–backward and sideward direction are relatively small.

0021-9290/$ - see front matter & 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jbiomech.2009.10.036
 ARTICLE IN PRESS
628 P. de Cocq et al. / Journal of Biomechanics 43 (2010) 627–631

The objective of this study was to quantify the force on the used (Zatsiorsky, 2002). The filtered positional data of the segmental mass centres
were differentiated twice using the five point method to obtain accelerations. The
horse’s back in sitting and rising trot by using kinematic data of
vertical force contributions of the segments, mi ðz€ i  gÞ, were determined from
the rider. these accelerations. Forces were normalised by dividing them by the body weight
of the rider and are therefore dimensionless. Strides were interpolated to 100% of
the stride cycle.
2. Materials and methods The same approach was used for the horizontal forces. The equation for the
forward–backward forces was
The study was performed with approval of the All University Committee for
X
4
Animal Use and Care and the University Committee on Research Involving Human Fx_rider ¼ mi ðx€ CM;i Þ; ð3Þ
Subjects at Michigan State University, and with full informed consent of the riders. i¼1

where Fx_rider is the forward–backward component of the reaction force vector and
2.1. Riders, horses and saddles x€ CM;i is the forward–backward acceleration of the centre of mass of the ith
segment. The equation of the sideward forces was
In this study 13 female riders (mean age7 SD 31714.3 years, height
X
4
1.6770.1 m, mass 62 7 5.6 kg) and two horses participated. One horse, 8 years Fy_rider ¼ mi ðy€ CM;i Þ ð4Þ
old, 1.53 m in height at the withers and 451 kg, was ridden by 6 riders. The other i¼1
horse, 24 years old, 1.63 m in height at the withers and 667 kg, was ridden by 7
where Fy_rider is the sideward component of the reaction force vector and y€ CM;i is
riders. Riders were weighed using a scale before the measurements. Both horses
the sideward acceleration of the centre of mass of the ith segment.
were clinically sound and were ridden with their own saddle.

2.2. Data collection 2.4. Statistics

Kinematic data were collected using a Motion Analysis System (Santa Rosa, CA, Means 7 SD were calculated from 4 strides of each rider in each condition.
USA) with eight infrared cameras operating at 120 Hz. The cameras were positioned Data were checked for normality of distribution using a Kolmogorov-Smirnov test.
in a riding arena around a calibrated volume measuring 8 m long  2 m wide  2.5 Data were analysed statistically in a GLM-repeated measures test followed by a
m high. A dynamic calibration procedure was performed using a wand with a width post hoc Bonferroni test using SPSS software (SPSS Inc., Chicago, Illinois, USA).
of 0.5 m between two markers; the maximal standard deviation of the measured Riding technique (sitting or rising) was the within-subject factor. The horse was
width was 0.0015 m. Horse and rider were prepared by placing spherical (20 mm included as a between-subject factor. For the average force, the rider’s weight
diameter) and cubic (6 mm length) infrared light retroreflective markers. The measured using a scale was also included in the analysis. A p-value of o0.05 was
markers of the rider were placed on the skin above the approximate joint centres of considered statistically significant.
the shoulder, elbow, wrist, hip and knee, as well as on the head and back (chin,
spinous processes of vertebrae C7 and T12). The markers over the joint centres of the
ankle and on the toe were attached to the shoe of the rider. The riders wore special 3. Results
clothes to enable placement of the markers directly on the skin. On the back of the
rider the spherical markers were used, to ensure that they were visible by the
cameras. On the horse, markers were placed on the hoofs. The markers were Data of one rider of horse 1 were not used because markers
attached to the skin with double-sided adhesive tape and glued to the hooves. were frequently lost from view. Normalised force values are
Rider and horse warmed up for 5–10 min prior to data collection. Data were presented in Tables 1 and 2. Normalised force patterns are shown
collected while the horse trotted in a straight line at its preferred speed (average
in Fig. 1.
3.11 m/s) with a rider performing sitting trot or rising trot in random order. The
trials within one rider had a maximal speed range of 7 0.05 m/s. Six trials at both In rising trot, the average vertical force was 0.9670.09. In
sitting and rising trot were collected within the allowed speed range. During each sitting trot, the average force was 0.96 70.09. Rising trot and
trial, one full stride was collected. Four full strides at both sitting and rising trot sitting trot did not differ significantly. There was no significant
were analysed for each horse/rider combination.
difference between the average vertical force of either riding
technique and the body weight of the rider. The two horses did
2.3. Data analysis
not differ in average vertical force.
At trot, two force peaks were present during one stride cycle.
The reconstruction of the 3D position of each marker was based on a direct
linear transformation algorithm. The raw coordinates were imported into Matlab
For the first force peak, values for rising trot were lower than
(MathWorks Inc., Natick, MA, USA). The raw kinematic data were filtered using a those for sitting trot (2.5470.30 versus 2.92 70.29). The second
zero-lag 8 Hz second order (effectively fourth order) low pass Butterworth filter force peak was also lower at rising trot (1.95 70.34 versus
before calculating derivatives. During rising trot, not all riders rose on the same 3.1170.39). Both peaks of horse 2 were significantly higher than
diagonal. Individual stride cycles were therefore extracted, with the beginning of
the peaks of horse 1.
each stride cycle defined as the moment of hoof contact of the hindlimb that was
grounded when the rider was sitting in the saddle during rising trot. Consequently,
all riders sat in the saddle during the first half of the stride cycle and rose from the Table 1
saddle during the second half of the stride cycle in rising trot. Data of riders that Normalised forces (mean 7SD) in rising trot and sitting trot.
rose during ground contact of the left hindlimb were mirrored in order to be able
to compare the sideward forces. The same hoof sequence was used to define the Variables Rising trot Sitting trot pvalue
stride cycle in sitting trot. Detection of the moment of hoof contact was based on
the horizontal velocity profile of the marker on the hoof (Peham et al., 1999). Average vertical force 0.967 0.09 0.96 7 0.09 40.999
The amount of vertical force on the rider depends on the mass and acceleration Peak vertical force 1 2.54 7 0.30 2.92 70.29 o 0.001nn
of the rider, in accordance with Newton’s second law Peak vertical force 2 1.95 7 0.34 3.11 70.39 o 0.001nn
Average forward–backward force  0.027 0.01 0.01 7 0.02 o 0.001nn
Fz_rider ¼ mB ðz€ MCB  gÞ ð1Þ
Peak forward force 1 0.48 7 0.17 0.50 70.25 0.068
where Fz_rider is the vertical component of the reaction force vector, mB is body Peak forward force 2 0.46 7 0.16 0.67 7 0.20 o 0.001nn
mass, g is the acceleration due to gravity (9.81 m/s2), and z€ MCB is the vertical Peak backward force 1  0.24 7 0.13  0.54 7 0.20 o 0.001nn
component of the acceleration of the body’s mass centre. If the body is subdivided Peak backward force 2  0.687 0.20  0.49 70.22 0.633
into 4 rigid segments, Eq. (1)) can be written as Average sideward force  0.00 7 0.01  0.00 7 0.01 0.301
Max sideward force 0.217 0.07 0.27 7 0.09 0.007nn
X
4
Fz_rider ¼ mi ðz€ CM;i  gÞ ð2Þ Min sideward force  0.247 0.07  0.30 7 0.08 0.009nn
i¼1
Forces are expressed in dimensionless units (force/body weight of the rider). A
where mi is the mass of the ith segment and z€ CM;i is the vertical acceleration of the
positive sideward force is directed towards the side of the hindlimb which lands at
centre of mass of the ith segment. Four body segments were defined; foot, lower
the beginning of the stride cycle; a negative sideward force is directed to the
legs, upper legs, and the upper body including the arms, the hands and the head.
opposite site.
Data on the segmental masses (percentages of body mass) and positions of
nn
segmental mass centres (percentages of segment lengths) in female athletes were Values are significantly different (po 0.05, Bonferroni correction).
 ARTICLE IN PRESS
P. de Cocq et al. / Journal of Biomechanics 43 (2010) 627–631 629

The forward–backward force had two forward peaks and two

Table 2 backward peaks. The forward peak in the second half of the stride
Normalised forces (mean 7 SD) of horse 1 and horse 2. cycle was significantly lower in rising trot compared to sitting trot
(0.4670.16 versus 0.6770.20). The magnitude of the backward
Variables Horse 1 Horse 2 p value
peak in the first half of the stride cycle was significantly smaller in
o
Peak vertical force 1 rising trot 2.327 0.16 2.69 7 0.28 0.008nn
rising trot (0.2470.13 versus 0.5470.20). The peak forward–
Peak vertical force 1 sitting trot 2.697 0.13 3.09 7 0.26 backward forces were higher in horse 2 compared to horse 1. The
o
Peak vertical force 2 rising trot 1.627 0.24 2.19 7 0.13 o 0.001nn pattern of the sideward force was more variable between riders.
Peak vertical force 2 sitting trot 2.737 0.06 3.38 7 0.26
o The maximum forces to the left and to the right were higher
Peak forward force 1 rising trot 0.357 0.13 0.58 7 0.12 0.080
Peak forward force 1 sitting trot 0.277 0.12 0.68 7 0.16 during sitting trot compared to rising trot. There was no difference
o
Peak forward force 2 rising trot 0.317 0.05 0.57 7 0.11 0.001nn in sideward force between the horses.
Peak forward force 2 sitting trot 0.517 0.13 0.79 7 0.14
o
Peak backward force 1 rising trot  0.13 7 0.09  0.32 7 0.09 0.009nn
Peak backward force 1 sitting trot  0.39 7 0.14  0.65 7 0.16 4. Discussion
o
Peak backward force 2 rising trot  0.55 7 0.04  0.78 7 0.21 0.001nn
Peak backward force 2 sitting trot  0.49 7 0.22  0.67 7 0.24
Peak vertical forces on the horse’s back were smaller in rising
Forces are expressed in dimensionless units (force/body weight of the rider). trot than in sitting trot, especially in the part of the stride where the
nn
Values are signi.cantly different (p o 0.05, Bonferroni correction). rider is standing in the stirrups, which agrees with Peham

Fig. 1. Normalised forces of the horses on the riders in rising and sitting trot. (a) Individual patterns of vertical reaction force of the horses on the riders in rising trot. (b)
Individual patterns of vertical reaction force of the horses on the riders in sitting trot. (c) Mean 7sd patterns of vertical reaction force of the horses on the riders in rising
and sitting trot. (d) Individual patterns of forward-backward force of the horses on the riders in rising trot. (e) Individual patterns of forward-backward force of the horses
on the riders in sitting trot. (f) Mean7 sd patterns of forward-backward force of the horses on the riders in rising and sitting trot. (g) Individual patterns of sideward force of
the horses on the riders in rising trot. (h) Individual patterns of sideward force of the horses on the riders in sitting trot. (i) Mean 7 sd patterns of sideward force of the
horses on the riders in rising and sitting trot. Force patterns for each rider–horse combination were calculated using four strides. Mean 7 SD patterns were calculated using

data of four strides of 12 rider–horse combinations: : rising trot; : sitting trot; oooo: riders on horse 1 (n= 5); +  +  : riders on horse 2 (n= 7). Forces are

expressed in dimensionless units (force/body weight of the rider). Note that the horizontal forces have been plotted on a different scale than the vertical forces.
 ARTICLE IN PRESS
630 P. de Cocq et al. / Journal of Biomechanics 43 (2010) 627–631

et al. (2008, 2009). In our study we also found a significantly lower saddle in the second part of the stride cycle is of practical
first peak in rising trot compared to sitting trot, which Peham et al. relevance regarding the loading of the back of the horse, while the
(2008) indicated might not be significant in their study due to the reduction in the first part of the stride cycle when the rider sits in
large variance in the saddle force data. This large variance in the saddle is of less importance. Another study by Roepstorff et al.
Peham’s study may be a result of transferring the saddle pad (2009) demonstrates that the vertical ground reaction force of the
between horses, which tends to increase the variability (De Cocq et horse is increased during the sitting phase of the rising trot
al., 2009b). Furthermore the variance in back shape of the horses compared to the rising phase. But because the difference was
will also influence the results of saddle force measurements as the minor in absolute terms, they think it is unlikely to have a direct
saddle force devices measure normal forces and the pressure mat is impact on the occurrence of locomotor injuries.
arched over the back of the horse in the frontal plane. In addition to reducing peak forces, standing up by the rider
Another difference between vertical force calculations from rider might also reduce the energy expenditure of the horse. Standing
kinematics and saddle force measurements is that in the former the up is also used by jockeys during horse racing (Pfau et al., 2009).
saddle was not included while it is included in the saddle force Jockeys adopt a posture in which the quasi-elastic capacities of
measurements. Fruehwirth et al. (2004) found a maximal vertical their legs can be used. With this posture they are able to reduce
force of 302.4733.9 N underneath the saddle of an unridden horse the peak loading on the horse’s back enabling the horse to go
at trot. But because the saddle moves with the horse and the saddle’s faster. The same method for reducing peak forces can be used by
mass is relatively low compared to a rider, this is unlikely to affect making an elastic interface between the backpack and the body
the difference in forces between riding techniques. (Rome et al., 2005; Foissac et al., 2009) and by using springy poles
The peak forward–backward and sideward forces were also in carrying loads (Kram, 1991). As load carrying also has energetic
lower in rising trot compared to sitting trot. Peham et al. (2008, costs, the question remains whether the use of elastic coupling
2009) found decreased movement of the centre of pressure in the also influences the energy expenditure of the carrier. If this is the
forward–backward direction in rising trot compared to sitting trot. case, elastic coupling can be used to reduce the energy
In the sideward direction, they did not find a difference in the expenditure of load carrying. Taylor et al. (1980) hypothesized
movement of the centre of force. Less movement of the centre of that it is the cost of generating force to support body mass that
mass of the rider explains the lower forces in rising trot found in determines the cost of running, and not the mechanical work that
this study. has to be done. If this is true, elastic coupling does not influence
There was a significant difference in peak vertical and forward– the energetics. Foissac et al. (2009), however, found that the
backward force between the two horses, which had been trained mechanical properties of a backpack (stiffness and damping
for different types of riding and were of a different age. Vertical coefficient) did indeed affect the energetics of walking in
ground reaction forces during trotting vary between breeds of humans carrying backpacks. With respect to equestrian sports,
horses that are used for different occupations (Back et al., 2007) as by using the legs as an elastic coupling during rising trot, the rider
a consequence of differences in vertical displacement of the body may not only reduce peak forces but may also reduce the
during the stride. In this study there was indeed a significant energetic cost.
difference in vertical displacement between the two horses (horse In conclusion, the peak forces in rising trot were lower than the
1: 0.05570.008 m; horse 2: 0.088 70.004 m; p o0.001) while peak forces in sitting trot. This supports the assumption that
the horses were moving at approximately the same stride rising trot is less demanding to the horse than sitting trot. Rising
frequency. This could also explain the observed differences in trot can therefore be used to prevent injuries in the horse. The legs
the peak vertical forces in our study as a consequence of larger of the rider act as a quasi-elastic coupling between rider and
vertical oscillations of the rider. Furthermore age influences back horse. Whether this coupling also reduces the energetic costs of
kinematics. Johnston et al. (2004) found that age was negatively load carriage could be tested by comparing the energetics of
correlated to extension and flexion of the thoracolumbar junction. horses ridden at the same speed in rising trot and sitting trot.
This could also influence the rider’s kinematics.
When calculating forces from kinematics of the rider, several
causes for errors exist. One important cause for errors is skin Conflict of interest
displacement. The markers are attached to the skin and can move
relative to the segmental mass centres. We used 4 segments and All authors deny having any financial and personal relation-
assumed that these segments were rigid. However, there was a ships with other people or organizations that could inappropri-
variance in segment length. The standard deviation of the upper ately influence our work.
body length was 0.008 m, that of the upper leg was 0.004 m, that of
the lower leg was 0.002 m and that of the foot was 0.0003 m. This
variance is a result of skin displacement and the measurement Acknowledgements
error of the cameras (0.0015 m for 0.5 m wand). Filtering and
differentiation may also be a source of error. We estimated this The authors thank the riders who participated in the experi-
error using different cutoff frequencies. Cutoff frequencies of 16, 30 ment. We give special thanks to LeeAnn Kaiser for her invaluable
and 50 Hz produced similar results to those presented here. Using a technical support.
cutoff frequency of 4 Hz resulted in an underestimation of the
accelerations. Therefore, it seems that these errors are small and do
not affect the differences between the conditions. References
The question remains whether the decrease of peak forces seen
in rising trot is indeed more comfortable for the horse. A study on Back, W., MacAllister, C.G., van Heel, M.C.V., Pollmeier, M., Hanson, P.D., 2007.
Vertical frontlimb ground reaction forces of sound and lame Warmbloods
back kinematics of the horse showed no difference in peak
differ from those in Quarter horses. Journal of Equine Veterinary Science 27,
extension of the back between rising and sitting trot (De Cocq et 123–129.
al., 2009a): in rising trot peak extension occurred during the first Clayton, H.M., Lanovaz, J.L., Schamhardt, H.C., van Wessum, R., 1999. The effect of a
(higher) force peak; during the second (lower) force peak rider’s mass on ground reaction forces and fetlock kinematics at the trot.
Equine Veterinary Journal 30 (Suppl.), 218–221.
extension of the back was less than in sitting trot. This indicates De Cocq, P., van Weeren, P.R., Back, W., 2004. Effects of girth, saddle and weight on
that the reduction in peak force when the rider rises out of the movements of the horse. Equine Veterinary Journal 36, 758–763.
 ARTICLE IN PRESS
P. de Cocq et al. / Journal of Biomechanics 43 (2010) 627–631 631

De Cocq, P., Prinsen, H., Springer, N.C.N., van Weeren, P.R., Schreuder, M., Muller., Peham, C., Hofmann, A., Molsner, J., Borkenhagen, B., Kuhnke, S., Baltacis, A., 2008.
M., van Leeuwen, J.L., 2009. The effect of rising trot and sitting trot on back Forces acting on the horses back and the stability of the rider in sitting and
movements and head–neck position of the horse. Equine Veterinary Journal 41, rising trot—a comparison. Pferdeheilkunde 24, 337–342.
423–427. Peham, C., Kotschwar, A.B., Borkenhagen, B., Kuhnke, S., Molsner, J., Baltacis, A.,
De Cocq, P., Clayton, H.M., Terada, K., Muller, M., van Leeuwen, J.L., 2009. Usability 2009. A comparison of forces acting on the horse’s back and the stability of the
of normal force distribution measurements to evaluate asymmetrical loading rider’s seat in different positions at trot. The Veterinary Journal doi:10.1016/
of the back of the horse and different rider positions on a standing horse. The j.tvjl.2009.04.007.
Veterinary Journal 181, 266–273. Pfau, T., Spence, A., Starke, S., Ferrari, M., Wilson, A., 2009. Modern riding style
Foissac, M., Millet, G.Y., Geyssant, A., Freychat, P., Belli, A., 2009. Characterization of improves horse racing times. Science 325, 289.
the mechanical properties of backpacks and their influence on the energetics Roepstorff, L., Egenvall, A., Rhodin, M., Byström, A., Johnston, C., van Weeren, P.R.,
of walking. Journal of Biomechanics 42, 125–130. Weishaupt, M., 2009. Kinetics and kinematics of the horse comparing left and
Freuhwirth, B., Peham, C., Scheidl, M., Schobeberger, H., 2004. Evaluation of right rising trot. Equine Veterinary Journal 41, 292–296.
pressure distribution under an English saddle at walk, trot and canter. Equine Rome, L.C., Flynn, L., Goldman, E.M., Yoo, T.D., 2005. Generating electricity while
Veterinary Journal 36, 754–757. walking with loads. Science 309, 1725–1728.
Johnston, C., Roethlisberger Holm, K., Erichsen, C., Eksell, P., Drevemo, S., 2004. Sloet van Oldruitenborgh-Oosterbaan, M.M., Barneveld, A., Schamhardt, H.C., 1995.
Kinematic evaluation of the back in fully functioning riding horses. Equine Effects of weight and riding on workload and locomotion during treadmill
Veterinary Journal 36, 495–498. exercise. Equine Veterinary Journal 18 (Supplement), 413–417.
Kram, R., 1991. Carrying loads with springy poles. Journal of Applied Physiology 71, Taylor, C.R., Heglund, N.C., McMahon, T.A., Looney, T.R., 1980. Energetic cost of
1119–1122. generating muscular force during running—a comparison of large and small
Peham, C., Scheidl, M., Licka, T., 1999. Limb locomotion—speed distribution animals. Journal of Experimental Biology 86, 9–18.
analysis as a new method for stance phase detection. Journal of Biomechanics Zatsiorsky, V.M., 2002. In: Kinetics of Human Motion. The Pennsylvania State
32, 1119–1124. University: Human Kinetics, p. 590.