Available online at www.sciencedirect.com

www.elsevier.com/locate/tvjl

The Veterinary Journal 175 (2008) 332–337

TheVeterinary Journal

Forces and pressures beneath the saddle during mounting fromthe ground and from a raised mounting platform

C.A. Geutjens, H.M. Clayton *, L.J. Kaiser

Mary Anne McPhail Equine Performance Center, Department of Large Animal Clinical Sciences, College of Veterinary Medicine,

Michigan State University, East Lansing, MI 48824, USA

Accepted 27 March 2007

Abstract

The objective was to use an electronic pressure mat to measure and compare forces and pressures of the saddle on a horse’s back whenriders mounted from the ground and with the aid of a mounting platform. Ten riders mounted a horse three times each from the groundand from a 35 cm high mounting platform in random order. Total force (summation of forces over all 256 sensors) was measured andcompared at specific points on the force–time curve. Total force was usually highest as the rider’s right leg was swinging upwards and wascorrelated with rider mass. When normalized to rider mass, total force and peak pressure were significantly higher when mounting fromthe ground than from a raised platform (P < 0.05). The area of highest pressure was on the right side of the withers in 97% of mountingefforts, confirming the importance of the withers in stabilizing the saddle during mounting.� 2007 Elsevier Ltd. All rights reserved.

Keywords: Electronic saddle mat; Total force; Peak pressure; Equestrian; Kinetics

Introduction

Regardless of sporting discipline or level of training,horses being trained under saddle must be mounted eachtime they are ridden. In some sports, the rider is usuallyassisted by receiving a ‘‘leg up’’, whereas in other sportsthe rider mounts using the stirrup, with or without the helpof a raised mounting platform. Intuitively, it might beassumed that if the horse is habitually mounted from thesame side, asymmetrical forces and pressures on the horse’sback may result in soreness or unequal muscular develop-ment as a consequence of the horse bracing against the uni-lateral force.

Pressure mapping technology, which has been used inclinical and research applications by both the medicaland veterinary professions, can be applied to measure totalforce of the saddle on the horse’s back and localized pres-sure distribution. An electronic pressure mat consists of an

1090-0233/$ - see front matter � 2007 Elsevier Ltd. All rights reserved.

doi:10.1016/j.tvjl.2007.03.025

* Corresponding author. Tel.: +1 517 432 5630; fax: +1 517 432 3442.E-mail address: claytonh@msu.edu (H.M. Clayton).

array of sensors that are calibrated individually to recordthe force applied perpendicular to the surface of the sensor.Software sums the force over all sensors (total force) anddisplays the pressure distribution over the entire surfaceof the mat, allowing temporal and spatial distribution ofhigh pressure areas to be defined. In ridden horses, previ-ous publications have reported differences in pressure dis-tribution and peak pressures between gaits (Jeffcott et al.,1999; Fruehwirth et al., 2004). A high correlation wasfound between total pressure under the saddle and massof the rider (Jeffcott et al., 1999; De Cocq et al., 2005).The effect of rider mass increases with speed (Fruehwirthet al., 2004) as a consequence of inertial effects related tothe suspension phase. An electronic pressure mat has alsobeen used to assess saddle fit, with localized areas of highpressure being taken as an indication for poor saddle fitting(De Cocq et al., 2005; Harman, 1994, 1997, 1999; Jeffcottet al., 1999; Werner et al., 2002; Meschan et al., 2007).

Every time a horse is ridden, it must be mounted. Tradi-tionally, horses are mounted from the left side. Riders maychoose to mount from the ground or with the aid of a

C.A. Geutjens et al. / The Veterinary Journal 175 (2008) 332–337 333

raised mounting platform. In mounting from the ground,the rider may step up using the stirrup, vault on withoutusing the stirrup or be boosted up by an assistant (legup). When mounting from a raised platform, the ridermay step into the stirrup or, if the platform is high enough,throw the leg over the horse’s back without using the stir-rup. It seems reasonable to speculate that the horse’s backmay be subjected to large and asymmetrical forces duringmounting with a stirrup, and the long-term effects may bemore problematic if the rider always mounts from the sameside. Although many horses show signs of resistance or dis-comfort when being mounted (Harman, 1999) forces andpressures on the horse’s back during mounting do notappear to have been investigated and the authors are notaware of any previous reports comparing different mount-ing techniques.

The purposes of this study were to measure and comparetotal force and pressure distribution beneath the saddle asriders mounted using the stirrup from the ground and froma raised mounting platform. The experimental hypothesis isthat mass-normalized maximal total force and peak pres-sure on the horse’s back are higher when mounting fromthe ground than from a mounting platform.

Materials and methods

The study was performed with approval of the All University Com-mittee for Animal Use and Care and the University Committee onResearch Involving Human Subjects at Michigan State University, andwith full informed consent of the riders.

Riders

The subjects were 10 riders with a body mass of 63.2 ± 8.4 kg (range:53.9–76.5 kg) and a height of 167.7 ± 4.7 cm (range: 160–174.5 cm). Allsubjects were experienced riders who were familiar with the mountingtechnique used in the study.

Horse and saddle

The horse was a well-trained 14-year-old riding horse gelding with aheight of 145 cm. A veterinarian examined the horse and found him to beclinically sound and without signs of back pain or asymmetry.

A dressage saddle (Schleese Saddlery Service) was used that had beencustom-made for the horse. Prior to the start of the study, the fit of the saddlewas assessed using the method described by Harman (1999) and found to beacceptable. The pressure beneath the saddle was distributed in a fairly uni-form pattern over the panels without areas of localized high pressure.

Mounting technique

All riders mounted as described in the United States Pony Club Manual

of Horsemanship Basics for Beginners D Level (Harris, 1994). Whenmounting from the ground, the rider stood next to the left shoulder of thehorse, facing the tail. The left hand held both reins at the level of thewithers. The right hand was used to turn the stirrup, and then it wasplaced on the pommel or waist of the saddle but not on the cantle. Therider hopped two or three times pushing off with the right leg and, asweight was transferred to the left foot in the stirrup, the left knee wasstraightened. The right leg swung over the horse’s back, the rider’s seatwas lowered into the saddle and the right foot was inserted into the rightstirrup.

The mounting platform was 35 cm high. When mounting with theplatform, the rider stood on the platform facing the side of the horse. Theleft hand held both reins at the level of the withers. The right hand wasused to turn the stirrup, and then it was placed on the pommel. The leftfoot was placed in the left stirrup with the toe pointing towards the horse’shead and the rider mounted in one smooth movement without hopping.The right leg was swung over the horse’s back, the rider settled into thesaddle and the right foot was inserted into the right stirrup.

Data collection

The electronic pressure mat used in the study (Pliance Saddle System,Novel GmbH), consists of separate mats for the left and right sides, eachof which has 128 sensors arranged in 8 columns and 16 rows. Prior to thestart of data collections, each half of the mat was calibrated in a specialdevice consisting of a rubber membrane, housed within a secure unit thatis filled with air using a compressor. The associated software calibrateseach sensor individually.

One half of the mat was placed on each side of the horse’s back with itsmedial edge parallel to the dorsal midline. The two halves of the mat werejoined and anchored with adhesive tape leaving a separation of 5 cm over thevertebral spinous processes. The saddle was lifted above the horse’s back thenlowered gently onto the pressure mat. The girth was tightened gradually,alternating between the left and right sides. When the girth was adequatelytight, the position of the pressure mat on the horse’s back and the position ofthe edges of the saddle on the pressure mat were marked so that the positioncould be checked and, if necessary, adjusted between riders.

The pressure mat was initialized to zero after tightening the girth andwas re-zeroed between riders. For each rider-horse combination, sixmountings were recorded: three with a mounting platform and three fromthe ground, which were performed in random order. Data were recordedwith a sampling frequency of 30 Hz. A camcorder recorded each mountingand video data were synchronised with pressure data.

Data analysis

The data were processed using Excel software. Forces were evaluatedin terms of: total force, which is the summation of the forces over allsensors (Newtons – N); and force normalized to rider mass (N/kg), whichfacilitates comparisons between rider’s of different sizes. The normalizedforce was plotted during each mounting trial and peaks in the force profilewere measured and correlated with kinematic events using the videorecordings. The force was also evaluated in terms of its distributionbetween the four quadrants of the pressure mat: left front (LF), right front(RF), left rear (LR) and right rear (RR).

The pressure profile was evaluated in terms of: peak pressure – thehighest pressure recorded in any sensor (N/cm2) at the time of peak ver-tical force; and the highest pressure normalized to rider mass (N/cm2/kg)at the time of maximal vertical force.

Statistics

The data were analysed in Statistica (Statsoft). Means ± SD werecalculated for the peak value of the total force during leg swing and for thetotal force when the rider was sitting in the saddle after mounting. Peakpressure and normalized peak pressure were recorded at the moment ofmaximal vertical force. Mean values ± SD were calculated for each riderand mounting technique.

The variables were tested for normality of distribution using the Kol-mogorov–Smirnov statistic. If a normal distribution was present, com-parisons between mounting with and without a platform were made usingStudents t test for dependant samples. For variables that were not normallydistributed, the non-parametric Mann U Whitney test was used. Between-rider comparisons were made using mass normalized values. Correlationcoefficients were determined between rider mass and maximal total forcefor mounting with and without a mounting platform. When significantcorrelations were identified, regression equations were developed. Absolute

334 C.A. Geutjens et al. / The Veterinary Journal 175 (2008) 332–337

values of the variables were used for tests of correlation and regression. Forall statistical tests, a probability of P < 0.05 alpha was used.

Results

The force and video data were used to divide the mount-ing effort into phases. Mounting began when the force onthe saddle mat increased above baseline and ended whena steady force was recorded on completion of the mountingeffort (Figs. 1 and 2). When mounting from the ground,small peaks in the force curve occurred during each hopand the force did not return to baseline after hoppingstarted. Instead, there was a progressive increase in totalforce beneath the saddle after each hop, followed by a stee-per increase when the right foot left the ground. Whenusing a mounting platform, there was a smooth increasein force after the right leg was raised.

A large and consistent force peak coincided with weightacceptance on the left stirrup as the right leg was swingingupwards for riders mounting from the ground and from aplatform. During this stage, mass normalized total forcewas significantly higher when mounting from the groundthan when using a platform (Table 1). A second force peaksometimes occurred as the rider’s buttocks contacted thesaddle (e.g. Fig. 2) and, occasionally, this force was higherthan the first peak. In several mounting efforts, there was

Fig. 1. Graph showing total force during a typical trial for mounting fromdistribution under the saddle are shown for the points marked 1–5 on the graphthe right leg swings upwards. In the pressure maps, the cranial part of the ma

also a third force peak just after the right foot was placedin the stirrup (e.g. Fig. 2). This appeared to be a habitualmotion in which the rider weighted the right stirrup toadjust the position of the saddle.

Since the peaks associated with the rider contacting thesaddle and adjusting the position of the saddle were notconsistently present, total force for these peaks was notused in the statistical comparison. Total force at the endof the mounting effort when the force trace had stabilised,did not differ significantly between the two methods ofmounting (Table 1). Partitioning of the total force byquadrant of the saddle indicated that the highest totalforce was in the RF quadrant in 85% of the mountingswith a platform and in 87% of the mountings from theground. In the other trials, the total force was highestin the LF quadrant.

The pressure distribution showed localized areas of highpressure. The initial weighting of the stirrup was associatedwith pressure increases in the LF lateral, LR medial andRF medial quadrants (Fig. 1: 1–3). As more weight wastransferred to the stirrup, the greatest increases in pressurewere in the sensors in the RF quadrant adjacent to thewithers and in the lateral part of the LF quadrant (Figs.1: 4 and 2: 2). Sensors in the medial part of the LF quad-rant were unweighted at this time. As the rider became

the ground. Still photographs and the corresponding maps of pressure. This sequence shows the most common pattern with a single force peak ast is up and the left side is to the left.

Fig. 2. Graph showing total force during a trial for mounting from a platform. Still photographs and the corresponding maps of pressure distributionunder the saddle are shown for the points marked 1–5 on the graph. This sequence shows force peaks as the right leg swings upwards (2), as the rider’sbuttocks contact the saddle (3) and as the rider steps into the right stirrup to adjust the saddle position (4). After this adjustment, there is a redistributionof pressure from the left to right side. In the pressure maps, the cranial part of the mat is up and the left side is to the left.

Table 1Mean (± SD) for peak value of the total force (TF), mass-normalized TF,maximal pressure and mass-normalised maximal pressure at mid-swingand when the rider was sitting in the saddle for riders mounting from a35 cm high mounting platform and from the ground (N = 10)

Mounting fromplatform

Mounting fromground

TF at midswing (N) 585.67 (85.11) 644.69 (67.74)Mass-normalised TF at

midswing (N/kg)9.29 (0.91)* 10.24 (0.56)*

TF sitting (N) 589.47 (92.84) 592.16 (80.51)Mass-normalised TF sitting (N/

kg)9.31 (0.39) 9.38 (0.55)

Peak pressure (N/cm2) 2.68 (0.43) 3.35 (1.07)Mass-normalised peak pressure

(N/cm2/kg)0.041 (0.003)* 0.052 (0.013)*

Significant differences between mass normalized values are indicated byasterisks. Fig. 3. Rider mass plotted against maximal total force during mounting

from the ground and from a raised platform (N = 10).

C.A. Geutjens et al. / The Veterinary Journal 175 (2008) 332–337 335

seated, pressure on the front half of the saddle decreased(Figs. 1: 5; 2: 3–5). If the rider stepped heavily into the rightstirrup to adjust the saddle, pressure increased in the rearhalf of the saddle, especially on the right side but pressurestill remained somewhat higher on the left than the rightside (compare Fig. 2: 3 with 4).

The highest pressure recordings were in the RF quad-rant as the left stirrup was accepting the rider’s weightand the rider’s right leg was swinging upwards (Figs. 1: 4and 2: 2). Peak pressure was located in the RF quadrantadjacent to the withers in 97% of the trials when mounting

336 C.A. Geutjens et al. / The Veterinary Journal 175 (2008) 332–337

from a platform and in 93% of the trials when mountingfrom the ground. In the remaining trials, peak pressurewas located in the LF quadrant. Mass normalized pressureat the time of peak vertical force was significantly higherwhen mounting from the ground than when using a plat-form (Table 1).

Maximal total force was significantly correlated withrider mass for both mounting techniques (Fig. 3). The fol-lowing regression equations were developed: for total forcewhen mounting from the ground: y = 177 + 7.4 * mass(r = 0.96; r2 = 0.84; P < 0.05); for total force when mount-ing from a platform: y = 121 + 7.4 * mass (r = 0.72;r2 = 0.52; P < 0.05).

Discussion

The mounting procedure used in this study is taughtthrough the Pony Club and other equestrian programs.Riders are discouraged from grasping the cantle of the sad-dle since this may, over a period of time, twist the tree orexert an exaggerated twisting force on the horse’s back(Harris, 1994). Even without pulling on the cantle, pressureon the horse’s back and, by inference, pressure on the sad-dle were shown to be highly asymmetrical and sufficient tocause stretching of the stirrup leather or twisting of the sad-dle tree that could have a further detrimental effect on thehorse’s well-being and performance (Harman, 1999).

During mounting, the horse’s body is stabilized by mus-cular contractions to avoid loss of balance in the face of alarge force applied rapidly on one side of the body. Whenthe rider habitually mounts from the same side, usually theleft, this bracing may lead to asymmetrical muscular devel-opment, especially in the horse’s shoulders. Visible differ-ences in muscular prominence between the left and rightshoulders are well recognized (Harman, 1999) with the leftshoulder being more prominent in the majority of horses. Itis conceivable that bracing of the forequarters duringmounting contributes to this asymmetry.

Riders selected for this study had a wide range of body-weights, which allowed the effects of bodyweight on abso-lute forces and pressures to be evaluated but introducedconsiderable between-rider variability in the absolute val-ues for total force and peak pressure. Consequently, massnormalized values were used to make between-rider com-parisons, which clearly showed that heavier riders exert lar-ger forces and pressures on the horse’s back, thussupporting the experimental hypothesis. Ease of mountingis likely to be related to height of rider relative to height ofhorse. Taller riders find it easier to reach the stirrup withthe left foot, and require a smaller elevation of their centreof mass from the ground. A raised mounting platform isoften used to facilitate mounting when there is a large dis-crepancy in height between a short rider and a tall horse.The rider’s agility may also play a role. The subjects of thisstudy were young and athletic; less agile riders may havemore difficulty in mounting leading to higher forces thanare presented here.

The results presented here indicate that the likelihood ofdamaging the saddle or injuring the horse during mountingis greater for a heavier rider. For example, the regressionequations developed in this study indicate that whenmounting from the ground, a rider with mass 50 kg wouldexert a maximal total force of 547 N, whereas a rider withmass 100 kg would exert a maximal total force of 914 N.These findings suggest that mass of the rider is an impor-tant consideration with regard to the possible detrimentaleffects of mounting from the ground, and that heavier rid-ers should be encouraged to use a mounting platformregardless of their height or level of agility.

The rider’s weight (N) is calculated as mass multipliedby the gravitational force (9.81). When the rider was sittingin the saddle after completion of the mounting effort, themass normalized total force was slightly less than therider’s weight because some of the weight was distributedoutside of the area covered by the pressure mat, such asbeneath the thighs or knees. However, total force whenthe riders were seated in the saddle was almost identicalfor the two mounting methods, as would be expected.

During the study it became clear that the reaction forceon the left stirrup due to the inertial effect of the rider’sright leg swinging upwards gave a consistent image onthe force graph and pressure profile. At this point, therewas high pressure on sensors in the RF medial and LF lat-eral quadrants, which is indicative of the effects of thehorse’s withers in stabilizing the saddle as it was pulledto the left side by the force on the left stirrup. Stability ofthe saddle during mounting is affected by the shape andprominence of the withers and by the correspondencebetween the shape of the saddle’s panels and the withers.The entire saddle is more likely to slip towards the mount-ing side on a horse with wide, flat withers, especially whenthis conformation is combined with a narrow tree that sitstoo high on the already flat withers. In an evaluation of thesaddles used regularly by 30 horses with a presenting com-plaint of back pain, 36% had a tree that was too narrow(Harman, 1997) indicating that this is a relatively commonproblem.

Fruehwirth et al. (2004) noticed that pressure was con-sistently higher beneath the LF quadrant than the RFquadrant of the saddle though the difference was not statis-tically significant. This same pattern was observed in ourstudy and it persisted even after the rider attempted to neu-tralize the saddle position by stepping into the right stirrup.It could not be determined whether this was due to asym-metrical development of the horse’s musculature or dis-placement of the saddle during mounting. Furtherinvestigation is warranted.

Maximal total forces recorded in this study (mountingfrom the ground: 10.24 ± 0.56 N/kg; mounting from aplatform: 9.29 ± 0.91 N/kg) were slightly lower than thoserecorded during walking (12.1 ± 1.2 N/kg) and consider-ably lower than those recorded during trotting(24.3 ± 4.6 N/kg) or cantering (27.2 ± 4.4 N/kg) (Fru-ehwirth et al., 2004). The inertial effect of swinging the right

C.A. Geutjens et al. / The Veterinary Journal 175 (2008) 332–337 337

leg upwards is less than the inertial effect of oscillating theentire bodyweight of the rider during the suspension phaseof trot and canter. More important in relation to equinewelfare, however, is the fact that, during mounting, thehigh forces are asymmetrically distributed, resulting in highpressures being localized in a small area on the right side ofthe withers. During ridden exercise, the force is moreevenly distributed over the four quadrants and pressuresare lower unless the saddle fits poorly.

Conclusions

It has been shown that total force and peak pressureduring mounting increase with rider mass and that totalforces and peak pressures on the horse’s back are signifi-cantly higher when mounting from the ground than whenusing a raised mounting platform. Localised areas of highpressure are present on the right side adjacent to the with-ers, which stabilises the saddle, and on the left side towardsthe lateral edge of the panel.

Acknowledgement

Red Shield Equestrian, LLC (California) and the McP-hail Endowment provided financial support for the study.

References

De Cocq, P., van Weeren, P.R., Back, W., 2005. Saddle pressuremeasuring: validity, reliability and power to discriminate betweendifferent saddle-fits. The Veterinary Journal 172, 265–273.

Fruehwirth, B., Peham, C., Scheidl, M., Schobesberger, H., 2004.Evaluation of pressure distribution under an English saddle at walk,trot and canter. Equine Veterinary Journal 36, 754–757.

Harman, J.C., 1994. Practical use of a computerized saddle pressuremeasuring device to determine the effects of saddle pads on the horse’sback. Journal of Equine Veterinary Science 14, 606–611.

Harman, J.C., 1997. Measurement of the pressure extended by saddles onthe horse’s back using a computerized pressure reading device.Pferdeheilkunde 13, 129–134.

Harman, J.C., 1999. Tack and saddle fit. Veterinary Clinics of NorthAmerican Equine Practice: Back Problems 15, 247–261.

Harris, S.E., 1994. The United States Pony Club Manual of Horseman-ship Basics for Beginners D Level. Howell Book House, New York,NY, pp. 25–26.

Jeffcott, J.B., Holmes, M.A., Townsend, H.G.G., 1999. Validity of saddlemeasurements using force-sensing array technology-preliminary stud-ies. The Veterinary Journal 158, 113–119.

Meschan, E.M., Peham, C., Schobesberger, H., Licka, T.F., 2007.The influence of the width of the saddle tree on the forces andthe pressure distribution under the saddle. The Veterinary Journal173, 578–584.

Werner, D., Nyikos, S., Kalpen, A., Geuden, M., Haas, C., Vontobel,H.D., Auer, J.A., von Rechenberg, B., 2002. Druckmessungen unterdem sattel: einem elektronischen Sattel-Messsystem (Novel GmbH).Pferdeheilkunde 18, 125–140.