The Veterinary Journal 2000, 160, 225234

doi: 10.1053/tvjl.2000.0507, available online at http://www.idealibrary.com on

Body Centre of Mass Movement in the Sound Horse

H. H. F. BUCHNER, S. OBERMLLER and M. SCHEIDL

Clinic of Orthopaedics in Ungulates, University of Veterinary Medicine Vienna, Veterinrplatz 1, A-1210 Wien, Austria

SUMMARY
The body centre of mass (BCM) is a key factor in the analysis of equine locomotion, as its position and move-
ment determines the distribution and magnitude of loads on the limbs. In this study, the three-dimensional
(3D) movement of the BCM in walking and trotting horses was assessed using a kinematic, segmental
method. Thirty markers representing 20 body segments were recorded in 12 sound horses while standing,
walking and trotting on a treadmill using a high-speed video system. Based on segmental inertial data, 3D
positions of the segmental centres of mass as well as the total BCM were calculated. The position within the
trunk during square standing and the movements of the BCM were determined for the three planes.
The position of the BCM in the standing horse is presented relative to external reference points. At the
trot, vertical displacement amplitude of the BCM amounted to 53 (6) mm as mean (sd), which was 27%
smaller than external trunk movement. Medio-lateral displacement amplitude of the BCM was 19 (4) mm,
34% less than trunk amplitude. Sagittal forward-backward oscillations of the BCM independent from
general forward movement were 13 (3) mm, being 24% less than trunk movements. At the walk, vertical,
medio-lateral and sagittal BCM movements were smaller than trunk movements by 43, 65 and 65%
respectively.
The results show reduced and efficient BCM movements compared to the trunk and form a basis for the
assessment of various clinical conditions such as lameness, the influence of a rider and various dressage
performances.
2000 Harcourt Publishers Ltd
KEYWORDS: Horse; centre of mass; kinematics; segment model; locomotion.

INTRODUCTION Wanless, 1999) and estimations of the BCM posi-

tion in the horse are used to guide riders to find
The determination and analysis of the body centre
the best position with minimal disturbance. Early
of mass (BCM) is a key part of biomechanic analysis
scientific studies about the position of the BCM in
dealing with the kinetics of a body. BCM is defined
standing horses were published in 1934 by Benke.
as the centroid of all mass elements of an object
Using a Borelli moment table, he was able to deter-
(Nigg, 1994) and its position and movement pat-
mine the position of the BCM in the sagittal axis.
tern determines, for example, the load distribution
Various head positions of the horse and the influ-
on the limbs and the external work of locomotion.
ence of rider posture on the overall centre of mass
In humans, the BCM movement is widely used both
were studied. Using this method, however, the posi-
as tool for basic biomechanic research (Cavagna
tion of the BCM only in the craniocaudal axis in
et al., 1963; Winter, 1979) as well as in clinical appli-
the standing horse could be measured. Knoll
cations (Saunders et al., 1953; Simon et al., 1977;
(1934) was the first to measure the different posi-
Iida & Yamamuro, 1987).
tions of the BCM in the moving horse. He meas-
In horses, the position of the BCM is frequently
ured the segmental masses of equine cadavers and
mentioned in equitation books (Romaszkan, 1957;
used plastillin models of the segments to deter-
mine segmental centres of mass assuming equally
Correspondence to: Dr H.H.F. Buchner, Clinic of Orthopaedics distributed density. Combining these segmental
in Ungulates, University of Veterinary Medicine Vienna, data and pictures from kinematic recordings, Knoll
Veterinrplatz 1, A-1210 Wien, Austria. Tel.: +43 1 25077 5508; fax.:
+43 1 25077 5590; e-mail: Florian.Buchner@vu-wien.ac.at could determine geometrically the BCM position

1090-0233/00/030225 + 10 $35.00/0 2000 Harcourt Publishers Ltd

226 THE VETERINARY JOURNAL, 160, 3

Markers

Segment
boundaries
z

Fig. 1. The 20-segment model of the horse. Straight lines indicate the segment boundaries, grey circles the markers for
the recording of the reference points and back movement. The cross indicates the position of the body centre of mass
(BCM) in the square standing horse. The movement axes (x, y, z) of the global co-ordinate system are indicated.

in the sagittal plane at various instancets during all
gaits and during jumping. However, the database
was still limited (there were no cadaver studies on
the position of the segmental centres of mass) and,
probably owing to the huge amount of work
involved in the data analysis, only exemplary posi-
tions and no total pattern of the motion were
reported. Basing on the work of Knoll (1934),
Krger (1941) discussed in more detail the BCM
displacement. He first mentioned the possible
significance of craniocaudal BCM movement in
lame horses as means of relief of a lame limb.
In a survey reported by Leach and Crawford
(1983) on future research in equine locomotion,
the determination of the BCM and the influences
of various segments on BCM movement were iden-
tified as major areas for future study. Since then,
two-dimensional (2D) segmental centre of mass
data for Thoroughbred horses (Sprigings & Leach,
1987; Kubo et al., 1992), as well as the complete
three-dimensional (3D) inertial data set of Dutch
Warmblood horses (Buchner et al., 1997) have
been published. Combining these accurate data
with modern gait analysis and computation tech-
niques allows detailed and reliable determination
of BCM movement patterns in moving horses to be
determined.
The purpose of this study was to determine the
3D position of the BCM within the horse as well as
the locomotion pattern of the BCM at the walk and
the trot for all planes of motion using a kinematic,
segmental method.
MATERIAL AND METHODS
Horses
Twelve horses, seven geldings and five mares, aged
between four and 22 years were used. All were
Warmblood horses of different size with a body
mass between 450 and 670 kg. The horses
belonged to the University of Veterinary Medicine
of Vienna and were clinically examined for absence
of any lameness. All horses were well trained to
treadmill locomotion (Buchner et al., 1994), shod
with standard open shoes and accustomed to the
recording environment.
 CENTRE OF MASS IN THE HORSE
Segment model
For the assessment of the BCM, a 20-segment
model was chosen (Fig. 1). In accordance with the
inertial segmental data set of Buchner et al. (1997)
the following segments were defined: head, neck,
trunk, tail, digit, metacarpus, anterbrachium and
shoulder for each fore limb, digit, metatarsus, crus
and thigh for each hind limb. Relative segmental
masses and position vectors for all segmental cen-
tres of mass were taken from Buchner et al. (1997).
Owing to significant skin displacement of the
marker at the neck/trunk junction (C6), its posi-
tion during locomotion was calculated based on
the triangular position of the marker at the withers
(Th5), the first coccygeal vertebra and C6 during
quiet standing.
Kinematic recording
Spherical passive markers, 4 cm in diameter, were
attached to the skin of the horses at specified refer-
ence points (Buchner et al., 1997) (Fig. 1). An addi-
tional five back markers were used at the level of
the dorsal spinal processes of thoracal vertebrae
Th5, Th10, Th16, the lumbar vertebra L3 and the
cranial edge of the sacral bone to analyse the BCM
position relative to trunk movement.
The locomotion pattern of the horses was
recorded while standing, walking (1.6 ms1), and
trotting (3.9 ms1) at a constant speed on a tread-
mill (Mustang 2200, Kagra AG). Kinematic data
were collected using the ExpertVision analysis
high-speed video system (Motion Analysis
Corporation) at a frame rate of 120 Hz during 10 s.
Six cameras in total with a resolution of 240 833
points were used on both sides of the treadmill.
Kinematic analysis
Following calibration of the cameras the markers
were tracked automatically under manual control.
The raw data were smoothed and stored. General
stride variables (stride duration, stride length,
stance duration) were calculated and all strides
normalised to 100% stride duration. The stance
phase was defined by a statistical method determin-
ing the horizontal speed distribution of the hoof
(Peham & Scheidl, 1998). The kinematic variables
were calculated for each stride, controlled for
irregular strides and averaged for each recording.
No correction for skin movement (except C6) was
performed.
227
BCM calculation
The calculation of the BCM was a two-step proce-
dure. Firstly, for each frame and each segment, the
3D position of the segmental centres of mass within
the global co-ordinate system were calculated from
the reference marker positions and using the iner-
tial segmental data from Buchner et al. (1997).
Using only two markers for each segment, no rota-
tional movement along the longitudinal segmental
axis could be assessed, but this movement is
assumed to be very small and negligible. Shoe mass
and position of the CM were considered with the
digit data. In the second step, a weighted sum of all
segmental centres of mass was calculated to give
the x, y and z co-ordinates of the total body centre
of mass for each frame.
The equation for the x- co-ordinate is given by:
rm1x1 + rm2x2 + rm3x3 + + rmnxn
xBCM =
M
where rmn equals relative mass of segment n; xn
equals x-co-ordinate of CM of segment n and M
equals body mass.
In this global co-ordinate system, the x-axis is
positive in the forward moving direction of the
horse, the y-axis is positive in the left lateral direc-
tion and the z-axis is positive in the vertical upward
direction (Fig. 1). The position of the BCM in the
standing horse is related to the position of external
trunk or limb markers (Th5, Th16, Tuber spinae
scapulae and Trochanter major). To determine the
position of the BCM from external markers, regres-
sion equations for x and z co-ordinates of the BCM
were tested for their significance. As independent
variables, the sagittal distance between Th16 and
Th5 as well as withers height were used.
BCM movement at the walk and the trot were
then analysed for its displacement in
forwardbackward, mediolateral and vertical direc-
tion. For visualization, 2D movement is shown in
the three movement planes: xz, yz and xy plane
(Figs 2ac). To allow the comparison of BCM and
Th16 in one graph, the positions of both points
were standardized to their position in the standing
animal (stand = 0). The movement of the BCM
within the body was calculated as the difference of
Th16 and BCM, and is shown as relative BCM
movement in Figures 3ac. The displacement
amplitudes of the BCM were calculated and com-
pared to those of the trunk marker Th16.
228 THE VETERINARY JOURNAL, 160, 3

Vertical movement (m) Forwardbackward movement

0.01 0.07
a c
Th 16 Th 16
0.01
0.06

0.03
BCM 0.05 BCM
0.05

0.04
0.07

caudal cranial right left

0.09 0.03
0.02 0.00 0.04 0.06 0.08 0.10 0.12 0.02 0.03 0.04 0.05 0.06 0.07
Forward -backward movement (m) Medio- lateral movement (m)

Vertical movement (m) Fig. 2. Mean displacement pattern of body centre of

0.01 mass (BCM) and Th16 marker of 12 horses in the xz
b plane (a, seen from the right side), the yz plane (b, seen
from the front side) and the xy plane (c, seen from the
0.01 ground) during one complete stride. The arrow indi-
cates start at suspension before landing of the left fore
BCM limb and direction of the movement during the stride.
0.03

0.05
were compared using paired students t-tests. For
Th 16 all tests a significance level of P < 0.05 was chosen.
Data are presented as mean values and standard
0.07
deviations of all horses.
right left
0.09
0.02 0.00 0.04 0.06 0.08 0.10 0.12 RESULTS
Medio-lateral movement (m)
The mean position of the BCM in the standing
horses relative to the positions of the external
Statistical analysis markers is listed in Table I. For an individual local-
The regression equations were tested for statistical ization of the BCM, the position in craniocaudal
significance using Pearsons correlation coeffi- direction can be calculated from the regression
cient. Displacement amplitudes of BCM and Th16 equation using the distance between Th5 and Th16

Table I
Position of the body centre of mass (BCM) in the standing horse relative to trunk markers in x, y and z
direction in 12 horses
Mean distance to BCM (m) Regression equation R2
Th16 TSS TM
xBCM 0.10 0.49 0.72 xBCM=xTh16+0.403 (xTh5xTh16) 0.0562 0.37*
(0.05) (0.05) (0.05)
yBCM 0.01 0.17 0.29
(0.03) (0.06) (0.10)
zBCM 0.22 0.11 0.02 zBCM=0.723 zTh5 + 0.122 0.86*
(0.04) (0.03) (0.03)

Data are presented as mean (sd). Th16: back marker at level of the dorsal spinal process of the 16th thoracic vertebra;
TSS: Tuber spinae scapulae; TM: Trochanter major; *: P < 0.05
 CENTRE OF MASS IN THE HORSE 229

Vertical movement (m) Table II

0.00 Displacement of the body centre of mass (BCM)
a
and the trunk (Th16) in x, y and z direction in 12
horses at the walk and the trot
0.01
axis BCM (m) Trunk (m)

0.02 Walk X 0.014 (0.005) 0.040 (0.009)*

Y 0.020 (0.002) 0.058 (0.016)*
Z 0.020 (0.008) 0.035 (0.013)*
0.03
Trot X 0.013 (0.003) 0.017 (0.004)*
caudal cranial Y 0.019 (0.004) 0.029 (0.011)*
0.04 Z 0.053 (0.006) 0.073 (0.008)*
0.025 0.020 0.015 0.010 0.005 0.000 0.005
Forwardbackward movement (m) Data are presented as mean (sd).
*: P < 0.05
Vertical movement (m)
0.00
b marker as external reference length. Roughly, this
position is near to the dorsal process of Th13 or at
0.01 the lowest point of the back in the saddle region.
The vertical position can be estimated using the
0.02 regression equation (Table I) using withers height
as reference length and is approximately 2 cm
beneath the hip joint. The position in the medio-
0.03 lateral direction was slightly out of the median
plane, a distance of 0.01 m to the left of the median
right left
0.04 line.
0.05 0.04 0.03 0.02 0.01 0.00 The movement of the BCM was found to be
Medio - lateral movement (m)
smaller than Th16 movement in all three move-
ment planes both at the walk and the trot. Figure 2
Forward-backward movement shows the 2D movement of the BCM and Th16 for
0.00 xz (sagittal) plane, yz (mediolateral/vertical) plane
c
and xy (horizontal) plane at the trot. All graphs
start (arrow) with the suspension phase before the
0.01
stance phase of the left fore limb. In the sagittal
plane (Fig. 2a), as seen from the right side of the
0.01
horse, both BCM and Th16 show two nearly sym-
metrical oval circles. Both points move down and
slightly forward at the beginning of the stance,
0.02 then back and down until the midstance position.
Slightly after starting the upwards movement, the
right left propulsion caused a forward movement in the sec-
0.02 ond half of the stance phase. Looking from the
0.05 0.04 0.03 0.02 0.01 0.00 front of the horse (Fig. 2b) the BCM movement
Mediolateral movement (m) showed in addition to the vertical movement, a
slight leftright movement directed to the weight-
Fig. 3. Mean relative displacement of the body centre of bearing limb during the stance phase. Th16
mass (BCM) within the trunk of 12 horses in the xz plane marker showed a mediolateral movement in oppo-
(a, seen from the right side), yz plane (b, seen from the site direction. Accordingly, Figure 2c combines for-
front side) and xy plane (c, seen from the ground) dur- wardbackward and mediolateral movement
ing one complete stride. The arrow indicates start at sus-
pension before landing of the left fore limb and pattern in one graph.
direction of the movement during the stride. Figure 3 shows the movement of the BCM within
the trunk during one stride at the trot. Owing to
230 THE VETERINARY JOURNAL, 160, 3

walk Vertical movement walk

Forwardbackward movement
1.6 1.4
(m) (m)
b
head and neck
1.4
a
head and neck
1.3
1.2
trunk
right forelimb
left forelimb
1

1.2
0.8 BCM
both forelimbs

0.6
1.1
BCM
0.4

trunk both forelimbs

0.2
all limbs 1
right forelimb left forelimb
% of stride
0
0 10 20 30 40 50 60 70 80 90 100

0.2 left hind limb right hind limb 0.9 all limbs
both hind limbs
right hind limb
0.4 both hind limb
left hind limb
% of stride
0.6 0.8
0 10 20 30 40 50 60 70 80 90 100

the larger displacement of the trunk compared to
the BCM, the relative movement was in the oppo-
site direction to the absolute BCM movement in all
three movement planes. The displacement ampli-
tudes in x, y and z direction are shown in Table II.
BCM displacement is generally significantly smaller
than trunk displacement. At the trot, BCM x (for-
wardbackward) displacement was 24% smaller
than Th16 displacement, y (mediolateral) displace-
ment 34% smaller and z (vertical) displacement
27% smaller. At the walk this difference was even
more evident than at the trot: BCM movement was
65, 65 and 43% smaller than Th16 movement in x,
y and z direction respectively.
The share of different body segments and com-
binations of segments of the BCM are illustrated in
Figure 4. Both for the walk (Figs 4 a & b) and for
the trot (Figs 4 c & d) the movement of the four
limbs, the fore-limb and hind-limb combination, all
limbs combined, the head and neck combination,
the trunk and as the sum of all the total BCM are
plotted together for the x-direction and z-direc-
tion. In the figures showing the x-direction move-
ment (Figs 4a & c) the opposed movement of
contralateral limbs is nicely visible. The combined
contralateral limb pair centre of mass is not a con-
stant position, but moves sinusoidally with the most
cranial located position shortly before the begin-
ning of the stance phase of each limb. In the verti-
cal direction (Fig. 4b) the shift of 25% between
fore- and hindlimb pattern at the walk evens out
most vertical movement of the all-limbs-combina-
tion. At the trot (Fig. 4d) the synchronized move-
ment of fore and hind limb caused the same
sinusoidal movement as all other body segments
and the overall BCM movement.
DISCUSSION
The position of the BCM in horses has been of
interest both for equitation schools and for biome-
chanic research for a long time. However, the
 CENTRE OF MASS IN THE HORSE 231

Forwardbackward movement Vertical movement trot

1.8 trot 1.5
(m) (m) d
1.6 head and neck head and neck
c
1.4
1.4

left forelimb right forelimb

1.2 1.3
trunk

1
both forelimbs BCM
1.2
0.8

BCM
0.6
1.1
trunk
0.4
all limbs both forelimb
1 right forelimb left forelimb
0.2

% of stride
0
0 10 20 30 40 50 60 70 80 90 100 0.9
both hind limb both hind limb all limbs
0.2
right hind limb left hind limb
left hind limb % of stride
right hind limb 0.8
0.4
0 10 20 30 40 50 60 70 80 90 100

Fig. 4. Centre of mass movement of various segment combinations and the body centre of mass (BCM) in the
forwardbackward (x) direction at the walk (a) and the trot (c), and the vertical (z) direction at the walk (b) and the trot
(d) during a complete stride in one horse. The graph starts with placement of the right fore limb till the next placement.
The co-ordinate system is calibrated with its origin (x, z) at ground level between fore and hind limbs.

calculation of the BCM position needs a high tech-
nical effort and was possible with reasonable accu-
racy only for the standing horse in longitudinal
direction (Benke, 1934). Knoll (1934) was the first
to estimate the BCM position for particular
instances of moving horses, but the lack of anatom-
ical data and processing capacity impeded the use
of his results for biomechanical analyses. Only the
availability of accurate inertial properties of horses
(Buchner et al., 1997), together with modern
kinematic equipment allow for a thorough assess-
ment of the 3D-BCM movement using a segmental
analysis method. This segmental method using the
weighted sum of all segmental centres of mass
enables not only the calculation of the total body
centre of mass, but also the share of particular
parts of the body such as head and neck, or the
fore- and hind-limb centres of mass. The second
method for the determination of the BCM move-
ment, often used for human locomotion research,
uses force plate data (Simon et al., 1977; Iida &
Yamamuro, 1987). Measuring the ground reaction
forces of a person and integrating the forces twice
gives the BCM-displacement in each moving direc-
tion. This method is relatively simple and non-inva-
sive, but needs the total ground reaction forces of
the subject during the stance phase. With horses
having at most times at least two limbs with simulta-
neous ground contact, this would need a set of
force plates to gather the force data simultaneously
in all supporting limbs.
The segmental method of BCM assessment also
has some limitations. Firstly, the inertial properties
are based on cadaver studies and the data of the
232 THE VETERINARY JOURNAL, 160, 3
living horses could be estimated only using regres-
sion equations. Additionally, the original data base
belong to Dutch Warmbloods with a slightly more
narrow mass range than the Warmbloods used in
this study. This introduces some systematic error in
the absolute position of the BCM, but the displace-
ment pattern and the relative movement of the
BCM are fortunately not influenced. A second limi-
tation lies in the fact that living horses are not rigid
body systems, which is a basic principle of the
segmental method. Muscle shape changes and skin
displacement necessarily cause some additional
error in the results, which have a constant pattern
during all strides. In addition, the use of only two
markers in this study to characterize one segment
does not allow for the assessment of rotational
movements around the longitudinal segmental axis
of each segment. As these movements are only very
small owing to the limb and joint geometry in
equine limbs, these limitations are considered
acceptable for the assessment of the general move-
ment. None of these errors in the results, however,
impedes an assessment of various influences on
BCM movement as, for example, different riding
techniques or changes in the locomotion pattern
owing to lameness.
The position of the BCM in the square standing
horse is as described in 1934 by Benke and was
shown graphically by Seiferle and Frewein (1977).
Benke (1934) found the position in forwardback-
ward direction to be at a distance of 42.86% of the
body length (distance from the cranial shoulder
contour to the caudal contour at the Tuber
ischiadicum) caudal from the shoulder, which is
slightly more caudal than found in this study (Fig.
1). This position near Th13 comes close to the low-
est point of the back line of horses and allows for
an alignment of the riders centre of mass with the
BCM of the horse in the craniocaudal direction
(Benke, 1934). The vertical position of the BCM,
however, proved to be significantly higher than that
estimated by Knoll (1934), who assumed an equal
density distribution in his plastillin model of the
trunk. The dorsal position of trunk segment parts
with high density, such as the spine and back mus-
cles, and the ventral position of the parts with low
density, such as the lung, cause a much more dor-
sally located position of the BCM than has been
estimated up to now. This vertical position can be
calculated using the regression equation given in
Table I and may be located roughly 2 cm beneath
the hip joint. Sprigings and Leach (1986) reported
a lower vertical position of the BCM, basing on
their trunk centre of mass data. However, their ver-
tical trunk reference line is described with a varia-
tion of the dorsal origin between Th10 and 12.
Owing to the curved shape of the back in this
region this reference length definition can cause a
variation in length and origin C which might be
responsible for the high variation within their
three horses and the low position in their
estimation of the trunk centre of mass.
The movement of the BCM both at the walk and
the trot follows in its general pattern the move-
ment of the trunk as shown for Th16 (Figs 2ac).
However, in all three movement axes, the displace-
ment amplitude of the BCM is significantly smaller
than the external visible trunk movement. This
shows a smooth and small BCM movement, which
contributes to the high locomotor efficiency of the
horse. This difference also shows the need to use
the real BCM displacement pattern for the calcula-
tion of the external work for locomotion, as
described by Minetti et al. (1999). An assumption of
similar movement ranges of BCM and external
trunk markers would clearly overestimate the real
energy requirements of horses for external work.
The reasons for these smaller displacements are
different for the walk and the trot. In the walking
horse, the trunk shows two cycles of a rotational
movement around the medio-lateral axis during
one stride. This means while the cranial part of the
trunk reaches a maximal vertical position during
midstance of a fore limb, the caudal part reaches its
vertical maximum 25% of the stride duration later,
during midstance of one hind limb. The BCM,
located near to this mediolateral rotation axis of
the trunk, has consequently less vertical movement
than the peripheral trunk locations. At the trot,
however, fore and hind limb cycles are synchro-
nous and so almost no trunk rotation takes place.
During trotting, a vertical bending of the equine
back might cause an increased vertical movement
of the middle part of the trunk at Th16 compared
to Th5 or the first coccygeal vertebra (Audigie
et al., 1999). This implies that real BCM movement
might show a larger vertical displacement than cal-
culated in the rigid body model basing on periph-
eral trunk markers (C6 and coccygeal markers),
but less displacement than found for Th16 at the
middle of the back.
The segmental method allows us to analyse all
possible combinations of segmental units and to
assess their share for the overall BCM. The most
 CENTRE OF MASS IN THE HORSE
significant influence is of course the movement of
the trunk which represents about 67% of the total
body mass (Buchner et al., 1997). Thus the general
movement pattern of the BCM and the trunk cen-
tre of mass are similar. The pattern of limb move-
ment is dominated by phase shifts of 50% between
contralateral limb pairs and 25% between fore and
hind limb at the walk. For the x-direction, most for-
wardbackward movements of one limb are evened
out by the contralateral limb. However, there
remains some small sinusoidal movement in the
limb pair combination during both the walk and
the trot, with the caudal extreme in the middle of
the stance phase of one limb and the cranial
extreme before impact of the limb. This influence,
however is almost negligible in the BCM pattern as
the amplitude is small and the fore and hind limb
share of the total body mass is only about 12% each
(Buchner et al., 1997). The vertical movement of
contralateral limbs always shows a synchronized up
and down action (Figs 4b & d) with the minimum
deflection reached after impact, and the maximum
at the midstance of one limb. The fore-limb maxi-
mum coincides with the straight vertical position of
the supporting fore limb and at the walk this also
causes a maximum height of the withers. The same
pattern is also present in the hind limb. At the
walk, fore- and hind-limb movements even out,
when they are summed up to the all-limbs-combi-
nation, as these pairs are 25% out of phase. During
the trot, however, all movements are synchronized
and the all-limbs-combination has about the same
vertical amplitude as the other body segments
(Fig. 4d).
In conclusion, the segmental method can suc-
cessfully be used to determine the 3D movement
pattern of all body segments as well as the total
body centre of mass. The BCM movement has a
generally smaller amplitude compared to trunk
movement and provides an important biomechani-
cal basis for explaining the high locomotor effi-
ciency of horses. The method described forms the
foundation for further research on particular influ-
ences on equine gait such as, for example, changes
due to lameness or the various influences caused by
different riding techniques.
ACKNOWLEDGEMENTS
The study was conducted using equipment sup-
ported by the Austrian Fond zur Frderung der
wissenschaftlichen Forschung, PNr 6904.
233
REFERENCES
AUDIGIE, F., POURCELOT, P., DEGUEURCE, C., DENOIX, J.M.
& GEIGER, D. (1999). Kinematics of the equine back:
flexion-extension movements in sound trotting
horses. Equine Veterinary Journal Supplement 30, 21013.
BENKE, G. (1934). Vizsglatok klnbz fajtajelleg s
hasznosits lovak slypontjnak meghatrozsrl s
vltozsrl (Beitrag zur Bestimmung der
Schwerpunktsvernderung bei Pferden). Thesis,
Budapest.
BUCHNER, H.H.F., SAVELBERG, H.H.C.M., SCHAMHARDT,
H.C., MERKENS, H.W. & BARNEVELD, A. (1994).
Habituation of horses to treadmill locomotion. Equine
Veterinary Journal Supplement 17, 135.
BUCHNER, H.H.F., SAVELBERG, H.H.C.M., SCHAMHARDT,
H.C. & BARNEVELD, A. (1997). Inertial properties in
Dutch Warmblood horses. Journal of Biomechanics, 30,
6538.
CAVAGNA, G.A., SAIBENE, F.P. & MARGARIA, R. (1963).
External work in walking. Journal of Applied Physiology
18, 19.
IIDA, H. & YAMAMURO, T. (1987). Kinetic analysis of the
center of gravity of the human body in normal and
pathological gaits. Journal of Biomechanics 20,
98795.
KNOLL, W. (1934). Kinematographische Bewegungs-
studien. Arbeitsphysiologie 8, 375406.
KRGER, W. (1941). ber das Verhalten des
Schwerpunktes bei der normalen Fortbewegung des
Pferdes. Tierrztliche Rundschau 47, 14751, 1626.
KUBO, K., SAKAI, T., SAKUROKA, H. & ISHII, K. (1992).
Segmental body weight, volume and mass center in
Thoroughbred horses. Japanese Journal Equine Sciences
3, 14955.
LEACH, D.H. & CRAWFORD, W.H. (1983). Guidelines for
the future of equine locomotion research. Equine
Veterinary Journal 15, 10310.
MINETTI, A.E., ARDIGIO, L.P., REINACH, E. & SAIBENE, F.
(1999). The relationship between mechanical work
and energy expenditure of locomotion in horses. The
Journal of Experimental Biology 202, 232938.
NIGG, B.M. (1994). Inerital properties of the human or
animal body. In Biomechanics, ed. B.M. Nigg, & W.
Herzog, pp 33764. Chichester: John Wiley & Sons.
PEHAM, C.H. & SCHEIDL, M. (1998). Statistische Methode
zur kinematischen Bestimmung der Sttzbeinphase.
Biomedizinische Technik 43 (Suppl 2), 779.
ROMASZKAN, VON G. (1957). Reiten Lernen, Schule des
Reitens und der Pferdekunde fr Anfnger und
Fortgeschrittene. pp 24877, Rschlikon: A. Mller
Verlag.
SAUNDERS, J.B., INMAN, V.T. & EBERHARDT, H.D. (1953).
The major determinants in normal and pathological
gait. Journal of Bone and Joint Surgery 35A, 54358.
SEIFERLE, E. & FREWEIN, J. (1977). Statik und Dynamik des
Bewegungsapparates. In: Lehrbuch der Anatomie der
Haustiere I, ed. R. Nickel, A. Schummer & E. Seiferle,
pp 479504. Berlin, Hamburg: Verlag Paul Parey.
SIMON, S.R., KNIRK, J.K., MANSOUR, J.M. & KOSKINA, M.F.
(1977). The dynamics of the centre of mass during
walking and its applicability. Bulletin Hospital Jt Dis 38,
1126.
234 THE VETERINARY JOURNAL, 160, 3

SPRIGINGS, E. & LEACH, D. (1986). Standardized tech- WINTER, D.A. (1979). A new definition of mechanical
nique for determining the centre of gravity of body work done in human movement. Journal of Applied
and limb segments of horses. Equine Veterinary Journal Physiology 46, 7983.
18, 439.
WANLESS, M. (1999). Reiten in Vollendung, Das
Praxisbuch zur Wanless-Methode. pp 22156. Cham:
Mller Rschlikon Verlag. (Accepted for publication 22 July 2000)