Journal of Sports Sciences, 1996, 14, 301-309

Anthropometric, strength, endurance and flexibility

characteristics of elite and recreational climbers
S. GRANT,1* V. HYNES, 1 A. WHITTAKER2 and T. AITCHISON 3
1
Institute of Biomedical and Life Sciences, 2Department of Mechanical Engineering and 3Department of Statistics,
University of Glasgow, Glasgow, UK

Accepted 3 January 1996

There has been remarkable development in the scope and quality of rock climbing in recent years. However,
there are scant data on the anthropometry, strength, endurance and flexibility of rock climbers. The aim of this
study was to compare these characteristics in three groups of subjects -- elite rock climbers, recreational
climbers and non-climbers. The 30 male subjects were aged 28.8 ± 8.1 (x ± S.D.) years. Group 1 (n = 10)
comprised elite rock climbers who had led a climb of a minimum standard of 'El' (E1-E9 are the highest
climbing grades) within the previous 12 months; Group 2 (n = 10) comprised rock climbers who had achieved
a standard no better than leading a climb considered 'severe' (a low climbing grade category); and Group 3
(n = 10) comprised physically active individuals who had not previously done any rock climbing. The test
battery included tests of finger strength [grip strength, pincer (i.e. thumb and forefinger) strength, finger
strength measured on climbing-specific apparatus], body dimensions, body composition, flexibility, arm
strength and endurance, and abdominal endurance. The tests which resulted in significant differences
(P < 0.05) between the three groups included the bent arm hang (elite 53.1 ± 1.32 s; recreational
31.4 ± 9.0 s; non-climbers 32.6 ± 15.0 s) and pull-ups (elite 16.2 ± 7.2 repetitions; recreational 3.0 ± 4.0
reps; non-climbers 3.0 ± 3.9 reps); for both tests, the elite climbers performed significantly better than the
recreational climbers and non-climbers. Regression procedures (i.e. analysis of covariance) were used to
examine the influence of body mass and leg length. Using adjusted means (i.e. for body mass and leg length),
significant differences were obtained for the following: (1) finger strength, grip 1, four fingers (right hand) (elite
447 ± 30 N; recreational 359 ± 29 N; non-climbers 309 + 30 N), (2) grip strength (left hand) (elite 526 ± 21
N; recreational 445 ± 21 N; non-climbers 440 ± 21 N), (3) pincer strength (right hand) (elite 95 ± 5 N;
recreational 69 ± 5 N; non-climbers 70 ± 5 N) and (4) leg span (elite 139 ± 4 cm; recreational 122 ± 4 cm;
non-climbers 124 ± 4 cm). For tests 3 and 4, the elite climbers performed significantly better than the
recreational climbers and non-climbers for any variable. These results demonstrate that elite climbers have
greater shoulder girdle endurance, finger strength and hip flexibility than recreational climbers and non-
Climbers. Those who aspire to lead 'El' standard climbs or above should consider training programmes to
enhance their finger strength, shoulder girdle strength and endurance, and hip flexibility.

Keywords: Elite athletes, endurance, flexibility, rock climbing, strength.

Introduction
ing to professional rock climbing competitions with
There has been remarkable development in the scope huge cash prizes (Bollen, 1988).
and quality of rock climbing recently. In the UK, indi- Despite the increasing numbers of rock climbers and
vidual membership of the British Mountaineering die improvement in climbing standard, there are scant
Council increased from 4350 in 1988 to around 8000 scientific data on this particular sport. Much of the sci-
in 1993, with a further 20,000 affiliated club members. entific literature relating to rock climbing centres
There has been a vast improvement in the standard of around climbing injuries (Bollen, 1988; Bollen and
climbing, ranging from increases in the grade of climb- Gunson, 1990; Cole, 1990). Climbing ability has been
considered to consist of two components, namely
knowledge and application of technique (Birkett,
1988). Recommendations for an appropriate training
* Address all correspondence to Stanley Grant, Institute of Biomed-
ical and Life Sciences, University of Glasgow, 64 Oakfield Avenue, regimen for the development of some physiological
Glasgow G12 8LT, UK. variables and potentially the enhancement of climbing
0264-0414/96 © 1996 E. & F.N. Spon
302 Grant et al.

ability have been proposed (Grant, 1985; Fyffe and groups. The intensity of the warm-up progressed from
Peter, 1990). However, these authors did not refer to low to high throughout the 5 min warm-up period.
any scientific studies, but stressed that strength, endur- The order of the tests was such that the effect of
ance and flexibility are of major importance for rock previous tests would not adversely affect performance
climbing. Grant (1985) emphasized that exercises in subsequent tests. There was no formal recovery time
should be as specific as possible to rock climbing, while between the different tests. Testing was carried out in a
Cappon (1983) considered that technique was essential standardized order. Recovery consisted of walking
for climbing success. Cappon believed that appropriate (when necessary) to the location of the next test and
training programmes were of great importance. having the forthcoming test explained. Rest periods of
We predicted that if certain characteristics are essen- 45 s duration were allowed between trials on the finger
tial for the attainment of a high standard of rock climb- strength tests (i.e. 45 s between the left and right
ing, differences would be identified using an hands) and 90 s between tests on the same limb.
appropriate battery of tests among elite climbers, recre- The tests were carried out in the following order:
ational climbers and physically active non-climbers.
The aim of this study, therefore, was to determine
1. Grip strength
which characteristics, if any, can distinguish between
elite rock climbers, recreational climbers and physically A grip dynamometer (Takei Kiki Koyo, Japan), cali-
active individuals who have never climbed. brated by the suspension of weights, was used to meas-
ure grip strength. The size of the grip was adjusted for
each subject so that the second joint of the middle fin-
Method ger of the hand holding the dynamometer was bent at a
90° angle. The subjects stood upright with their feet
A total of 30 males aged 28.8 ±8.1 (x ± S.D.) years slightly apart and the dynamometer was grasped with
volunteered as subjects for this investigation. The the indicator scale facing away from the body. With the
climbing standard of individuals in a number of climb- arm extended downwards, the dynamometer was held
ing clubs was determined. If the climbers accepted the away from the body and clothes. The subjects were
invitation to take part in the study, they were assigned asked to squeeze the dynamometer gradually and with
to the appropriate group. Recreationally active indi- maximal force for at least 2 s without swinging their
viduals (i.e. subjects who took part in regular physical arm around. One practice trial was followed by three
activity) were invited to take part as members of the tests on both hands. Measurements were taken alterna-
control group. Self-reported lead capability was used to tively between the left and right hands. The best score
allocate individuals to the following categories: for each hand was taken as the final result, with meas-
urements being made to the nearest 0.5 kg.
1. Elite rock climbers (n —10)'. experienced rock climb-
ers who had led to a minimum standard of grade
' E l ' within the previous 12 months. Grades E1-E9 2. Leg span
are the highest climbing grades (Fyffe and Peter,
To stimulate the bridging movement performed on the
1990).
rock face, the leg span of the subjects was measured
2. Recreational climbers (n -10): rock climbers who
while they lay flat in a supine position. This procedure
had achieved a standard no better than leading a
reduced the risk of muscle strain injury due to support-
climb considered 'severe' (a relatively easy climbing
ing the body weight while ensuring the plane of move-
grade) within the previous 12 months (Fyffe and
ment was similar to the actual position performed on
Peter, 1990).
the rock face. The subjects placed their feet as far apart
3. Non-climbers (n = 10): individuals who had not pre-
as possible while keeping their knees straight. With the
viously done any rock climbing. The non-climbers
subjects lying flat on their back, leg span was measured
were regularly involved in a range of activities,
from medial calcaneus to medial calcaneus. One meas-
including endurance running, 'keep fit' and endur-
urement was taken to the nearest 0.5 cm using a tape
ance cycling.
measure.
The subjects were asked to rest or undertake only
light training on the day prior to testing. Before testing,
3. Body dimensions
the subjects were asked to warm up. They were per-
mitted to choose their own warm-up or were given a 5 The length of both the arm and leg on the right side of
min warm-up consisting of running, press-ups, bench the body only was measured to the nearest 0.5 cm
stepping and other movements involving large muscle using a tape measure.
Anthropometry, strength, endurance and flexibility of rock climbers 303

4. Arm length 7. Height

The length of the upper limb was measured from the Each subject was asked to remove his shoes and stand
inferior border of the acromion process to the head of upright, hands by his side and eyes fixed at a point level
the radius, while the forearm was measured from the on the. opposite wall. His height was then measured to
head of the radius to the tip of the radial styloid the nearest 0.5 cm using a stadiometer (Holtain Ltd).
(Weiner and Lowrie, 1984). However, in practical
terms, it is not only the length of the arm which may be
a limiting factor at the rock face, but how far the arm 8. Bent arm hang
can extend the hand to reach a hold which determines The subject gripped the 'jug holds' on either side of the
success or failure. Therefore, for the purposes of this climbing board (specialized climbing apparatus, Nevis
investigation, arm length was measured from the in- Sport, Glasgow). The subject then lifted himself up
ferior border of the acromion process to the tip of the until his chin was just above the bottom edge of the
middle finger while the arm was straight, abducted to fingerboard (see Fig. 1). This position corresponds
shoulder height and with the forearm supinated. The approximately to a right angle at the elbow joint. The
subjects were instructed to stand up straight with their
subject was then instructed to hold that position as long
back and legs against a wall and arm length was then
as possible without resting his chin on the fingerboard.
measured to the nearest 0.5 cm.
A short familiarization trial was allowed. This was a
maximally timed test with the watch started the
moment the chin was in the correct position; the test
5. Leg length ended when the subject's eyes went below the bottom
The subjects lay in the supine position with their legs end of the fingerboard (Eurofit, 1988).
flat on the mat. With their feet placed approximately
shoulder width apart (Ellis, 1983), the length of the
9. Abdominal curl-ups
right leg was then measured from the anterior superior
iliac spine to the apex of the medial malleolus. Again, in Abdominal endurance was assessed using the method
reality on the rock face it is the placing of the foot on outlined by Faulkner et al. (1989).
the rock which is vital. Hence, the length of the lower
limb was taken from the anterior superior iliac spine to
the tip of the medial malleolus and, with the foot plan- 10. Sit-and-reach test
tar flexed as much as possible, the measurement was Though similar to the Eurofit (1988) test, the proce-
extended to the tip of the big toe. dure for scoring was amended. A score of 30 cm was
considered equivalent to the subject being able to touch
his toes. The subjects were asked to stretch as far as
6. Pincer strength they could and hold that position for 3 s. They were
given three practice trials followed by three tests. The
The grip dynamometer (Takei Kiki Koyo, Japan) was highest score was recorded.
adjusted so that the grip was 4.5 cm wide. The grip to
be used was then demonstrated. Each subject was
instructed to hold the cross bars of the dynamometer 11. Finger strength
between the thumb and forefinger; the other three fin-
gers were not to be used in this test. The subject then Measurements of finger strength were made using an
stood upright with his feet slightly apart and arm exten- innovative apparatus developed by the Department of
ded forward to shoulder height. The experimenter then Mechanical Engineering at the University of Glasgow.
stood in front of the subject holding the dynamometer The machine was developed in an attempt to simulate
level with the subject's shoulder and at arms length. the positions which climbers encounter at the rock
With a straight arm, the subject was asked to grip the face. When the finger muscles in the hand contract, the
dynamometer with his thumb and forefinger and horizontal plate of the apparatus bends. The degree of
squeeze with maximum effort for 2 s. The subject was strength in the fingers is determined by how much a
given one practice trial followed by three tests, alternat- subject can bend the plate and influence the reading on
ing between arms until three results were recorded for the strain gauge attached to the plate (see Fig. 2). In
both arms. Measurements were made to the nearest the early stages of its development, it was noted that the
0.5 kg with the highest score achieved for each hand apparatus may have applications in other areas, includ-
being taken as the final result. ing medicine.
 304 Grant et al.

SPECIFICATIONS variation of strain gauge reading with load was found to

It was specified that the apparatus should: be linear within a range from zero to loads well beyond
those encountered during the tests. To ensure consis-
i. Be applicable to both hands, allow the use of differ- tency of tests from one day to the next, a calibration
ent numbers of fingers and a variable, though con- weight was attached to the end of the plate before and
trolled, finger length when applying the load. after each test. This is particularly important with the
ii. Measure finger strength only, with the effects of current device, since it uses only one strain gauge and
arm, shoulder and body strength being removed. so cannot be temperature-compensated. This limitation
iii. Simulate the positions which climbers experience could easily be overcome by the application of several
at the rock face. strain gauges, but in the interests of simplicity and low
iv. Be sensitive enough to record differences of less costs, temperature compensation was not considered to
than 1 N, but be strong enough to be robust. be a prime consideration in this work. The results of the
v. Allow accurate and repeatable measurements. calibration tests ensured that Specifications (iv) and (v)
vi. Be inexpensive. had been met.
vii. Be safe. The additional instrumentation required is a DC
power supply, a strain gauge bridge and a form of
DESIGN
recording device. In the current tests a voltmeter was
The main measuring device is a strain gauge attached used, which required the operator to watch one device
to a flexible steel plate. The fingers apply a load to the all of the time during the test and observe the peak
end of the plate and the strain gauge measures the reading. In future, some sort of peak detecting device
amount that the plate bends. The strain gauge reading would make the tests easier to perform and more accu-
can be calibrated to measure applied end load. This rate. Some sort of pen recorder, or a computer equip-
calibration was performed by hanging a series of ped with an analogue-to-digital converter, would allow
weights on a cord attached to the end of the plate. The load-time traces to be produced.

Figure 1 An example of the bent arm hang.

Anthropometry, strength, endurance and flexibility of rock climbers 305

Flexible Steel Plate

Adjustable Finger Stop^ , Fitted with Strain Gauge

Adjustable Wrist Rest

Steel Supporting
Structure

Adjustable Elbow
Support
Wooden Base
\
Support arranged
to keep upper arm
horizontal'-

Figure 2 Diagram of the finger strength apparatus.

The adjustable finger stop ensures that only a con-

trolled portion of the fingers can be used to apply
the load [Specification (i)]. The elbow support allows
the elbow to rest on it and so prevents muscles other
than those in the fingers applying the load.

USE IN THE CURRENT TESTS

The apparatus was placed on a filing cabinet 1.3 m in

height and packing placed beneath the wooden base to
ensure that the subject's upper arm was horizontal. The
elbow support was then adjusted to suit the individual Figure 3 An example of the finger strength grip 1 test.
length of forearm. The subject stood with his hips and
toes touching the filing cabinet with his pelvis square to
the cabinet. The unused hand was rested lightly against limiting calliper and following Durnin and Womersley's
the side of the cabinet. The hand being tested was then
(1974) method. All readings were taken on the right
placed on the finger plate with the fingers in the desired
side of the body with the subject standing upright and
position and the palm of the hand resting against the
adjustable wrist rest. It was ensured that the elbow was relaxed.
positioned directly below the hand. This stance was
used with every test, regardless of finger grip used. 13. Body mass
For the first test of finger strength, the subject was The subject was asked to remove his shoes and his
required to push the tips of all four of his fingers as far body mass was recorded to the nearest 0.1 kg using
along the steel plate as possible; this position was called scales (Avery Beam Balance, Birmingham, UK).
'grip 1' (see Fig. 3). While standing in the correct
stance, the subject was instructed to pull down on the
plate as hard as possible in a maximal contraction. The 14. Finger strength: Grip 2
subject was allowed one familiarization trial followed
by three tests for each hand, switching hands between The subject stood in exactly the same stance as for grip
tests. The best score for each hand was recorded. 1. In this test, however, the grip was used slightly dif-
ferently (see Fig. 4). The bar was moved forward
towards the horizontal plate, which just left room for
12. Body composition only the fingertips and the tip plate. The subject was
Each subject's percent body fat was predicted by taking asked to press his fingernails against the bar with his
four skinfold measurements using a Holtain skinfold fingertips on the edge of the plate. The same procedure
 306 Grant et al.

Figure 4 An example of the finger strength grip 2 test. Figure 5 An example of pull-ups, full arm extension.

as for grip 1 was carried out; that is, three tests on each 16. Pull-ups
hand were carried out using four fingers and two fin- Subjects were required to complete the maximum
gers. The hand tested was alternated. number of pull-ups possible without coming off the
climbing board. A pull-up consisted of hanging from
full arm extension (see Fig. 5) and pulling the whole
body up until the chin had reached a point on the fin-
15. Foot raise gerboard and then lowering the body down until the
arms were fully extended again. This whole manoeuvre
A demonstration of the test was given to the subject. was performed with the subject gripping the 'jug hold'
The subject was required to stand facing a wall with his on either side of the fingerboard. A pull-up was not
toes touching a line 23 cm from the wall. Both his counted if the subject did not reach full extension or if
hands were placed on the wall at shoulder height. his chin did not reach the required height.
Standing on his left foot the subject brought his foot
directly up in front of him. The subject was instructed STATISTICAL ANALYSIS
to place the toe of his right foot as high up the wall as
Differences among the three groups of subjects were
possible without allowing it to move laterally. The dis-
investigated for each of the tests in the battery by
tance the subject managed to lift his foot was marked means of one-way analysis of variance or covariance
on the wall. The distance from the floor to the highest correcting for body mass or leg length as applicable.
mark was recorded to the nearest 0.5 cm using a tape Appropriate Bonferroni multiple comparison proce-
measure. Measurements were only taken on the right dures were then carried out if significant differences
leg. among the three groups existed.
Anthropometry, strength, endurance and flexibility of rock climbers 307

The relevant assumptions of normality, equal vari- Table 2 Mean (± S.D.) test variables not adjusted for body
ance and linearity were assessed by means of probabil- mass or leg length for the three groups of subjects
ity plots. Hartley's test and residual plots respectively. Recreational
For the vast majority of variables, these assumptions Elite climbers climbers Non-climbers
were found to be satisfactory. The only assumption
found to be untenable was that of equal variances in the Flexibility
case of the number of curl-ups, when an approximate Sit-and-reach (cm) 37.9 ± 7.8 31.3 ± 6.8 34.5 + 5.2
one-way analysis of variance based on an assumption of Foot raise (cm) 99.9 ± 11.7 96.4 ± 11.0 91.8 ± 7.9
unequal variances was carried out (Brown and For-
Endurance
sythe, 1974).
Bent arm hang (s) 53.1 ± 13.2" 31.4 ±9.0 32.6+15.0
Ratio standardization of scores with regard to body Pull-ups (reps) 16.2 ± 7.2" 3.0 ± 9.0 3.9 ± 3.9
mass was considered to be inappropriate, since the Curl-ups (reps) 66.6 ± 17.2 52.7 ± 21.0 62.0 ± 3.7
ratio standardized scores were found still to depend sig-
nificantly on body mass (Vanderburgh et al., 1995). Significance level P < 0.05.
"The elite climbers performed significantly better than the
The only viable option was correction by a linear recreational climbers and non-climbers.
regression on body mass (Armstrong and Welsman,
1994). Accordingly, regression procedures (i.e. analysis
of covariance) were used to examine the influence, if Table 3 shows that the finger strength (grip 1) meas-
any, of body mass on the results. urements did not appear to be related to body mass; leg
length was not significantly related to leg span in these
subjects. There was a significant body mass effect on
Results grip strength and pincer strength and the suggestion of
an influence of body mass on finger strength (grip 2).
Table 1 gives the subjects' characteristics. There were There were no significant differences between recrea-
no significant differences among the three groups for tional and non-climbers for any variable in the test
age, body mass, height, height:body mass ratio, percent battery.
body fat, arm length and leg length. The test results for
which it was not necessary to correct for body mass or
leg length are given in Table 2. The elite climbers had a Discussion
significantly higher score than the other two groups for
leg span, bent arm hang and pull-ups. The adjusted In recent years, rock climbing has become increasingly
means (± S.D.) for all variables corrected for body mass popular and the standard of climbing continues to rise.
or leg length as appropriate are given in Table 3. The It has been hypothesized that certain characteristics
elite climbers had a significantly higher average score required of rock climbers may be more pronounced in
than the two other groups for grip strength (L), pincer elite performers. From the results of this study, some of
strength (R and L) and leg span, where L and R denote the tests performed produced clear-cut distinctions
left and right respectively. The elite climbers had sig- between the groups investigated.
nificantly higher average scores than the non-climbers The innovative apparatus designed for this investiga-
for finger strength grip 1 (4R, 4L, 2R and 2L). There tion has proved very useful in distinguishing between
were no differences among the groups for ringer elite climbers and non-climbers in measures of finger
strength grip 2. strength. All four results taken for finger strength using
grip 1 showed that the elite climbers had significantly
greater finger strength than non-climbers. However,
Table 1 Mean (± S.D.) physical characteristics of the
subjects with grip 2 there was no significant difference across
any of the groups. With grip 2, several subjects com-
Elite Recreational Non- plained that it was very painful to exert full force, and
Characteristic climbers climbers climbers this factor may be responsible for the lack of any differ-
Age (years) 27.8 ± 7.2 32.0 ± 9.2 26.5 ± 7.4 ence among the groups. The possibility exists that grip
Body mass (kg) 74.5 ± 9.6 72.9 ± 10.3 70.9 ± 5.9 2 was not representative of a climbing grip. However,
Height (cm) 178.9 ±8.5 179.417.9 179.4 ± 4.4 preliminary work with this apparatus involving several
Percent body fat elite climbers suggested that the grips used on the
(%) 14.0 ± 3.7 15.3 ± 3.0 12.7 ± 2.4 apparatus were appropriate for rock climbers. The elite
Arm length (cm) 73.8 ±3.9 76.1 ±4.2 75.4 ± 3.0 climbers who took part in this study commented
Leg length (cm) 114.6 ±4.6 119.2 ±7.0 117.5 ±4.8 favourably on the specificity of the apparatus in relation
Significance level P < 0.05. to high levels of rock climbing.
308 Grant et al.

Table 3 Adjusted means (± S.D.)for all variables corrected for body mass or leg length
Adjusted means P-values

Recreational Effect of Effect of body

Characteristics Elite climbers climbers Non-climbers climbing type mass
Grip strength (R) (N) e 532 ± 2 3 472 ± 23 478 ± 23 0.14 0.001
Grip strength (L) (N)c 526 ± 21* 445 ± 21* 440 1 21* 0.01 0.001
Pincer strength (R) (N) c 95 ± 5 * 69 ± 5* 7015* 0.01 0.001
Pincer strength (L) (N)c 93±6» 75 ± 6 * 7416* 0.04 0.001
Finger strength 1 (4R) (N)c 446 ± 30" 359 ± 29 309 1 30° 0.02 0.19
Finger strength 1 (4L) (N) c 441 ± 34<" 346 ± 33 309 1 34" 0.03 0.13
Finger strength 1 (2R) (N)' 329 ± 24" 249 ± 23 224 1 24* 0.01 0.85
Finger strength 1 (2L) (N)c 313 ± 26" 238 1 25 222 1 26" 0.04 0.89
Finger strength 2 (4R) (N) c 221 ± 16 202 ± 16 184 + 16 0.29 0.11
Finger strength 2 (4L) (N)c 228 ± 17 200 ± 17 182116 0.17 0.05
Finger strength 2 (2R) (N) c 154 ± 10 133110 128 1 10 0.20 0.08
Finger strength 2 (2L) (N)c 151 ± 11 127 111 125110 0.18 0.07
Leg span (cm)'' 139 ± 4 * 122 ± 4* 124 1 4* 0.02 0.10
The elite climbers performed significantly better than the non-climbers.
'The elite climbers performed significantly better than the recreational climbers and non-climbers.
'Adjusted for body mass of each subject by means of an analysis of covariance.
''Adjusted for leg length of each subject by means of an analysis of covariance.

Pincer strength in both hands and grip strength (left der girdle strength was considered to be one of the sig-
hand) was a distinguishing feature between the elite nificant factors in predicting climbing ability. It has
climbers and the other two groups. This rinding may be been suggested that good climbing technique requires
a result of the frequent use of a pincer grip while elite that the legs are the main means of progress when
climbers are climbing. Watts et al. (1993) considered climbing but, as the severity of the climb increases, the
that male elite climbers may not need to develop demands placed on the arms increases (Birkett, 1988).
extremely high levels of grip strength. The absolute It would therefore seem plausible that these qualities
values for the elite climbers in the present study were (i.e. arm and shoulder endurance) are essential charac-
slightly higher than those ofWatts et al. (1993). How- teristics of a high standard of climbing performance.
ever, grip strength scores (left hand) were significantly Several flexibility measurements were taken in this
higher for the elite climbers compared with the recrea- investigation, both traditional and sports-specific.
tional and non-climbers and may suggest that grip Using a sit-and-reach test, no distinction among the
strength is important for elite rock climbers. groups was found, indicating that hamstring and low
These results suggest that the muscles of the forearm back flexibility are not required to reach a high stand-
and hand are stronger in elite climbers than recrea- ard of climbing, or that all three groups have high ham-
tional and non-climbers. The elite climbers performed string and low back flexibility. It should be noted that a
significantly better as regards grip strength as well as sit-and-reach test is recognized as not being the best
specific finger strength values, whereas Donnelly et al. measure of hamstring flexibility. While there were no
(1991) found that elite climbers had significantly significant differences among the groups, there was a
higher climbing-specific strength (i.e. the index and tendency for the elite climbers to perform best. Never-
middle fingers using a climbing hold mounted to a theless, the score for all three groups in this study
force transducer) compared with non-climbers only would be categorized as 'average' according to Pollock
when strength was expressed per unit body mass. There et al. (1984), who presented norm tables for a US pop-
were no differences in Donnelly and co-workers' study ulation. One of the climbing-specific flexibility tests
between elite climbers and non-climbers for grip used was the leg span, which gives an indication of a
strength. subject's ability to abduct the hips. Elite climbers had a
The pull-up and bent arm hang tests clearly distin- significantly higher leg span score than the recreational
guished the elite climbers from the other two groups and non-climbers. The leg span test was designed to
and suggest that these factors are important for climb- simulate the bridging movement performed when
ing ability at El standard and above. Watts et al. (1993) climbing. It could be postulated that this type of flex-
noted in their discussion with elite climbers that shoul- ibility is important for the elite climber. Certainly some
Anthropometry} strength, endurance and flexibility of rock climbers 309

of the elite climbers reported that they regularly under- Birkett, B. (1988). Technique in Modern Rock and Ice Climbing.
took flexibility sessions, including hip flexibility exer- London: A.N.C. Black.
cises. Examination of the relationship between leg Bollen, S.R. (1988). Soft tissue injury in extreme rock climb-
length and leg span score revealed that leg length was ers. British Journal of Sports Medicine, 22, 145-147.
not significantly related to leg span scores in these Bollen, S.R. and Gunson, C.K. (1990). Hand injuries in
competition climbers. British Journal of Sports Medicine,
subjects.
24, 16-18.
No differences were found among the groups for the Brown, M.B. and Forsythe, A.B. (1974). The small sample
curl-up test. Thus abdominal endurance would appear behaviour of some statistics which test the equality of sev-
to be either not a characteristic of elite climbers or a eral means. Technometrics, 16, 129-132.
general requirement for all three groups of subjects. Cappon, M. (1983). Rock and Ice Climbing: The History, Prac-
Several anthropometric measures were taken because tice and Techniques. London: Orbis.
body characteristics can be important in determining Cole, A.T. (1990). Fingertip injuries in rock climbers. British
sporting success (Reilly et al., 1990). There were no Journal of Sports Medicine, 24, 14.
significant differences for limb length, height and body Donnelly, J., Byrnes, W.C., Kearney, J.T. and Fleck, S.
mass. In activities where body mass is repeatedly lifted (1991). A comparison of muscular strength and endur-
against gravity, extra 'mass' in the form of fat or large ance measures between rock climbers and non rock climb-
muscle mass is disadvantageous. Thus, it might have ers. In Abstracts of the International Congress and Exposition
been anticipated that elite climbers would have had a on Sports Medicine and Human Performance, p. 45,
lower percent body fat than the other two groups. Watts Vancouver, April.
et al. (1993) reported extremely low estimated percent Durnin, J.V.G.A. and Womersley, J. (1974). Body fat assessed
body fat (5%) for competitive world-class climbers. from total body density and its estimation from skinfold
thickness: Measurements on 481 men and women aged
The present study was conducted in January/February
16-72 years. British Journal of Nutrition, 32, 77-97.
when a high percent body fat might have been antici- Ellis, H. (1983). Measurement of lower limb. In Clinical
pated, as the climbers were less active. However, all Anatomy, 7th edn, pp. 248-249. Oxford: Blackwell.
elite climbers reported that they were training regularly. Eurofit (1988). Handbook for the Eurofit Test of Physical Fitness.
The fact that the subjects in Watts and co-workers' Rome: Committee of Experts on Sports Research.
study were competitive climbers and testing was car- Faulkner, R.A., Sprigings, E.J., McQuarrie, A. and Bell, R.D.
ried out during the World Cup sport climbing competi- (1989). A partial curl-up protocol for adults based on an
tion, may have prompted the climbers to 'peak' for this analysis of two procedures. Canadian Journal of Sports Sci-
event. ence, 14, 135-141.
We conclude that elite climbers have greater shoulder Fyffe, A. and Peter, I. (1990). The Handbook of Climbing.
girdle endurance, finger strength and hip flexibility London: Pelham Books.
than recreational climbers and non-climbers. These Grant, S. (1985). Training for rock climbing. Climber and
results suggest that climbers who aspire to lead El Rambler, July, pp. 30-32.
standard (or above) should consider training pro- Pollock, M.L., Wilmore, J.H. and Fox, S.M. (1984). Exercise
in Health and Disease. Philadelphia, PA: W.B. Saunders.
grammes to enhance performance in these variables.
Reilly, T., Secher, N., Snell, P. and Williams, C. (1990).
What cannot be concluded from this study is whether Anthropometry. In Physiology of Sports (edited by N.
the qualities seen in the elite climbers are a prerequisite Secher, P. Snell and C. Williams), pp. 470-483. London:
for a high standard of climbing. E. and F.N. Spon.
Vanderburgh, P.M., Mahar, M.T. and Has Chou, C. (1995).
Allometric scaling of grip strength by body mass in col-
lege-age men and women. Research Quarterly for Exercise
References and Sport, 66, 80-84.
Watts, P.B., Martin, D.T. and Durtschi, S. (1993). Anthropo-
Armstrong, N. and Welsman, J.R. (1994). Assessment and metric profiles of elite male and female competitive sport
interpretation of aerobic fitness in children and adoles- rock climbers. Journal of Sports Sciences, 11,
cents. In Exercise and Sport Sciences Reviews (edited by J.O. 113-117.
Holloszy), Vol. 22, pp. 435-476. Baltimore, MD: Williams Weiner, J.S. and Lowrie, J.A. (1984). Practical Human
and Wilkins. Biology. London: Academic Press.