You are on page 1of 5

Journal of Strength and Conditioning Research, 2006, 20(1), 136–140

q 2006 National Strength & Conditioning Association

THE RELATIONSHIP OF BODY SEGMENT LENGTH


AND VERTICAL JUMP DISPLACEMENT IN
RECREATIONAL ATHLETES
D. SCOTT DAVIS, ERIN E. BOSLEY, LINDSAY C. GRONELL, SARAH A. KEENEY,
ANDREA M. ROSSETTI, CORRIE A. MANCINELLI, AND JOHN J. PETRONIS
Division of Physical Therapy, Department of Human Performance and Exercise Science, West Virginia University,
Morgantown, West Virginia 26505.

ABSTRACT. Davis, D.S., E.E. Bosley, L.C. Gronell, S.A. Keeney, Aragon-Vargas and Gross (3) depicts a 4-segment model
A.M. Rossetti, C.A. Mancinelli, and J.J. Petronis. The relation- consisting of the trunk, thigh, leg, and foot connected by
ship of body segment length and vertical jump displacement in 3 joints between each segment. During a VJ, muscle force
recreational athletes. J. Strength Cond. Res. 20(1):136–140. production primarily from the back extensors, gluteus
2006.—The purpose of this study was to determine if segmental
skeletal length contributes to vertical jump (VJ) displacement
maximus, quadriceps, gastrocnemius, and soleus results
in recreational athletes. Skeletal length measurements of the in a powerful ground reaction force that propels the body
trunk, femur, tibia, and foot were obtained by palpation of bony upward against gravity. As in any lever system, the
landmarks and a standard tape measure. A pilot study (n 5 10) length of the lever arm affects joint torque (19). There-
examined the intratester and intertester reliability for each fore, it is theorized that body segment length (trunk, fe-
skeletal measure. The pilot investigation revealed fair to excel- mur, tibia, and foot) may influence VJ displacement.
lent intratester and intertester reliability. Seventy-eight recre- Body segment length has been measured in several
ational athletes (55 men and 23 women) with a mean age of 21.9 investigations by identification of bony landmarks via
6 2.9 years participated in the investigation. Multiple regres- palpation (1, 11, 13, 16, 20, 26). Biomechanical studies
sion analysis with gender as a categorical indicator variable re-
using motion analysis often use reflective markers that
vealed a significant gender difference; therefore, men and wom-
en were analyzed separately. Regression analysis for men iden- are placed on bony landmarks by palpation to measure
tified foot length (p , 0.033, R2 5 0.08) as the only significant body segment length (16, 20). Al-Eisa et al. (1) measured
skeletal length predictor of VJ displacement. None of the skel- tibia length as the distance from the medial joint line of
etal length measures was predictive of VJ displacement in wom- the knee to the medial malleolus. They measured femur
en. Based on the results of this investigation, intrinsic skeletal length from the greater trochanter of the femur to the
length is not a strong predictor of VJ displacement in young lateral joint line of the knee. Duyar and Pelin (11) also
adult recreational athletes. measured tibia length from the medial joint line of the
KEY WORDS. skeletal measurements, biomechanics, functional knee to the medial malleolus. Fry et al. (13) measured
testing tibial height by measuring the distance from the lateral
tibial condyle to the ground. Al-Eisa et al. (1) measured
foot length from the posterior heel to the longest toe,
INTRODUCTION
whereas McWhorter et al. (22) measured foot length us-
he vertical jump (VJ) test is a common func- ing a Brannock device that measured the distance from

T tional measure of an athlete’s current and po-


tential level of athletic performance (2, 5, 6, 12).
Several studies have documented the influence
of VJ displacement on the playing status, level of play,
and position in college football players (5, 6, 12). As a
the posterior heel to the longest toe.
To date, no investigation has attempted to determine
the relationship between body segment length and VJ
displacement. Therefore, the purpose of this study was to
determine if body segment lengths (trunk, femur, tibia,
result, coaches and athletes are interested in improving foot) are predictors of VJ displacement in male and fe-
VJ displacement in those sports in which vertical jumping male recreational athletes.
is required. If the factors that predict VJ displacement
are clearly identified, then targeted training programs METHODS
can be investigated to determine the most efficacious pro-
Experimental Approach to the Problem
gram. Many studies have attempted to identify the fac-
tors that influence VJ displacement (2, 4–10, 12, 15, 17, This investigation used an exploratory correlational de-
21, 24, 28, 30). Despite the efforts of many investigators, sign and regression analysis to test the null hypothesis
a strong prediction model for VJ displacement has not that there is no predictive relationship between the skel-
been identified. Previous studies have focused on train- etal length variables and VJ displacement in recreational
able variables, such as muscular strength and power, athletes. A pilot study was conducted before performing
flexibility, balance, coordination, body weight and com- the main investigation to establish intratester and inter-
position, and jumping technique (2–10, 12, 15, 17, 18, 21, tester reliability for each skeletal measure. The investi-
24, 28, 30). To date, no studies have examined the influ- gation was approved by the university’s Institutional Re-
ence of nontrainable variables such as the length of in- view Board for Human Subjects and all subjects complet-
dividual body segments in predicting VJ displacement. ed a health history questionnaire and signed a written
The biomechanical model of the body, as described by informed consent before testing.
RELATIONSHIP OF BODY SEGMENT LENGTH 137

pect of the C7 spinous process. The base of the sacrum


was identified by palpation using the sacral sulcus and
the posterior superior iliac spine as easily identified land-
marks. All measurements were taken in centimeters and
were repeated 3 times by 4 examiners in random order.
The examiners were blinded from the subject’s identity
by a curtain throughout the testing session.
Testing Procedures. The main study was conducted at
the University Student Recreation Center in a large aer-
obics room on 2 separate occasions. After completing a
health history and signing an informed consent docu-
ment, each subject rotated through a series of 6 stations.
For each of the lower extremity measurements, only the
right lower extremity was measured. The examiner at the
first station measured the total body height in centime-
ters using a standard scale (Continental Scale Corp.,
Bridgeview, IL) with height attachment. The second
through fifth stations included foot, tibia, femur, and
trunk length measurements. Each of the skeletal length
measurements was performed as described previously for
the pilot investigation. The examiners were chosen to per-
form a measurement at a given station based on their
intratester reliability. Each measurement was conducted
by an investigator with a minimum intratester reliability
coefficient of 0.76.
FIGURE 1. Measurement of foot length with a modified Bran- VJ displacement was measured at the sixth and final
nock device. station. VJ was performed with shoes on using a coun-
termovement but no step. Blue chalk was applied to the
long finger of the subject’s chosen hand. The subject was
Subjects then instructed to stand with the self-selected side next
Seventy-eight recreationally active men (n 5 55) and to a wall covered with white paper. Baseline reach was
women (n 5 23) volunteered to participate in this study recorded by having the subject reach as far as possible
(age: men 21.9 6 2.9 years, women 21.6 6 2.1 years). A and touch the chalked finger to the wall, while maintain-
recreational athlete was defined as any individual who ing both feet in contact with the floor. VJ displacement
participates in nonprofessional or noncollegiate athletics was determined as the difference between the superior
3 or more times per week. No attempt was made to sub- aspect of the best VJ mark and the superior aspect of the
divide the sample based on type of athletic participation. baseline standing reach mark. Measurements were ini-
All subjects were informed of the purpose and risks of the tially recorded in inches but later converted to centime-
study before signing an informed consent document. Sub- ters for data analysis. Each subject performed 3 jumps.
jects completed a health-screening questionnaire to iden- The jump with the greatest displacement was recorded
tify conditions that would limit their ability to participate and used for analysis (31).
in the study.
Statistical Analyses
Procedures Data from the pilot study were analyzed using a repeated
Preliminary Testing. A pilot investigation was conducted measures analysis of variance (ANOVA) and intraclass
on 2 separate occasions (men: n 5 4; women: n 5 6) to correlation coefficient (ICC) (2,1) for intertester reliability
establish intratester and intertester reliability of the and ICC (3,1) for intratester reliability according to the
skeletal length measures (age: 22.1 6 1.69 years). The formula described by Shrout and Fleiss (25). The main
first testing session examined right tibia and femur investigation was initially analyzed using multiple re-
length. Femur length was measured in a sidelying posi- gression analysis (least squares) with gender as a cate-
tion, with a standard clinical tape measure. The length gorical indicator variable. Based on the results of the ini-
of the femur was measured from the inferior aspect of the tial analysis, men and women were analyzed separately
subject’s greater trochanter to the distal lateral femoral using multiple regression analysis to identify predictors
condyle, identified by the proximal lateral joint line of the of VJ displacement. In each case, the response variable
knee. Tibia length was measured in a supine position was VJ displacement and the regressor variables were
from the medial joint line of the knee to the inferior as- age, body height, trunk, femur, tibia, and foot length. Par-
pect of the medial malleolus. tial regression coefficients were used in hypothesis test-
The second testing session measured right foot and ing.
trunk length. Foot length was measured in the seated
position using a Brannock device (Figure 1) that was RESULTS
modified to include a standard clinical tape measure. The The intratester and intertester reliability values for all
length of the foot was measured from the posterior heel skeletal measurements are listed in Table 1. The intra-
to the longest toe. Trunk length was measured with the tester reliability was found to be excellent (ICC . 0.81)
subject in a prone position with a standard clinical tape for all testers for tibia, foot, and trunk length measure-
measure from the base of the sacrum to the inferior as- ments. Intratester reliability for femur length was found
138 DAVIS, BOSLEY, GRONELL ET AL.

TABLE 1. Intratester and intertester reliability for skeletal TABLE 3. Predictors of VJ* displacement in men.
measures.
Regressor Partial
Intratester Intertester variables coefficients p value
Measurement ICC* (3,1) ICC (2,1)
Age 20.073 0.8684
Femur SD 5 0.76 0.61 Height 20.463 0.3244
CM 5 0.50 Trunk 20.273 0.5855
SK 5 0.37 Femur 0.598 0.3304
AR 5 0.12 Tibia 20.028 0.9829
Tibia SK 5 0.98 0.61 Foot 3.02 0.0330†
CM 5 0.93
* VJ 5 vertical jump.
SD 5 0.92
† Statistically significant at p ,0.05.
AR 5 0.85
Foot length JP 5 0.99 0.99
CM 5 0.99 TABLE 4. Predictors of VJ* displacement in women.
AR 5 0.99
EB 5 0.99 Regressor Partial
Trunk JP 5 0.93 0.74 variables coefficients p value
CM 5 0.91 Age 20.237 0.7544
AR 5 0.89 Height 0.041 0.9526
EB 5 0.81 Trunk 0.941 0.2383
* ICC 5 intraclass correlation coefficient. Femur 20.575 0.5590
Tibia 20.177 0.9160
Foot 20.165 0.9255
to be poor to fair with only 1 tester having a reliability * VJ 5 vertical jump.
coefficient greater than 0.75. As would be expected, the
intertester reliability was generally less than the intra-
tester reliability. TABLE 5. Pearson correlation coefficients.
Descriptive statistics including means and standard Variables Men Women
deviations are listed in Table 2. Inspection of the descrip-
tive statistics revealed a possible difference between men Age by VJ* 0.02 20.16
Height by VJ 0.09 0.23
and women for all the variables except age, indicating Trunk by VJ 20.03 0.35
that gender should be used as a categorical indicator var- Femur by VJ 0.18 0.11
iable in the initial regression analysis. Multiple regres- Tibia by VJ 0.18 0.15
sion analysis with gender as an indicator variable re- Foot by VJ 0.28 0.13
vealed a significant difference between men and women
* VJ 5 vertical jump.
(p , 0.0001). Because of the heterogeneity of the sample,
men and women were separated for further data analysis.
Multiple regression analysis of the male data (Table can be altered by training. The results of these investi-
3) revealed that foot length was the only significant var- gations revealed that VJ displacement is difficult to pre-
iable (p , 0.0330, R2 5 0.08). This model resulted in a dict; however, 2 variables, power (3, 4, 9) and body com-
positive partial beta coefficient of 3.02 revealing that as position (9, 21, 23, 27–29), appear to hold the greatest
foot length increases, VJ displacement increases. Inspec- promise. In contrast, no investigations have examined the
tion of the residual plot revealed a good fit of the linear importance of intrinsic (nontrainable) variables in the
model. A similar examination of the female data (Table prediction of VJ displacement. Therefore, the present
4) revealed that none of the regressor variables was pre- study attempted to identify the importance of intrinsic
dictive of VJ displacement. Pearson product-moment cor- skeletal length variables in predicting VJ displacement
relation coefficients, for each of the skeletal length vari- in young adult recreational athletes.
ables against VJ, are listed in Table 5. Based on the literature, it was not surprising to find
that gender was a significant indicator variable. Previous
DISCUSSION investigations have identified a difference in VJ displace-
Previous investigations have attempted to determine pre- ment between men and women (14, 31). Additionally, on
dictors of VJ displacement based largely on variables that average, men have greater body height and skeletal seg-

TABLE 2. Descriptive statistics for each variable.


Means (range) Standard deviation (6)
Variable Men Women Men Women
Age (years) 21.9 (18–36) 21.6 (18–27) 2.9 2.1
Height (cm) 178.5 (165–195) 166.1 (150-177) 7.4 7.2
Trunk (cm) 56.4 (48–64) 51.6 (44.5–57.5) 3.7 3.5
Femur (cm) 41.4 (36.4–49.9) 39.3 (34.2–45.1) 2.8 2.6
Tibia (cm) 39.5 (35.4–45.9) 36.7 (32.2–40.2) 2.5 2.2
Foot (cm) 27.2 (23.4–31.5) 24.6 (22.7–26.4) 1.6 1.1
VJ* (cm) 56.8 (40.0–85.1) 36.8 (26.7–50.8) 9.0 6.3
* VJ 5 vertical jump.
RELATIONSHIP OF BODY SEGMENT LENGTH 139

ment length than women. It is surprising, however, that heavily on the examination of isolated body systems. A
the length of the foot was found to be a predictor in men VJ maneuver is a complex task that requires the coordi-
but not in women. There are several possible reasons that nation of the nervous and musculoskeletal systems and
may explain this difference. The first is that the sample is largely regulated by affective factors such as motiva-
size for men (n 5 55) was larger than for women (n 5 tion and desire. Within the human system, bones, mus-
23), which may account for the low variability among the cles, ligaments, tendons, nerves, and emotions must in-
women. Inspection of Table 2 reveals that the standard teract synchronously for optimal athletic performance.
deviation for foot length in men was larger than for wom- The results of this investigation identified that intrinsic
en. Foot length in women ranged from 22.7 to 26.4 cm for nontrainable variables (age; body height; trunk, femur,
a difference of 3.7 cm, whereas foot length in men ranged and tibia length) are not predictive of VJ displacement in
from 23.4 to 31.5 cm, for a difference of 8.1 cm. It is also recreational athletes, whereas foot length was found to be
possible that women use a different jumping strategy a weak predictor of VJ displacement in male recreational
than men, which requires less ankle joint contribution. athletes. Based on the literature and the results of this
This possible difference in technique may place more em- investigation, future studies should examine the inter-
phasis on back, hip, and knee extension and less empha- action of promising variables in a comprehensive model
sis on ankle plantarflexion. that better captures the complexities of the human sys-
It may be hypothesized that for a given ground reac- tem.
tion force, an individual with a longer foot would generate
greater ankle torque due to a longer lever arm and more PRACTICAL APPLICATIONS
propulsive force would be delivered in the vertical direc-
The importance of VJ in athletes is well documented in
tion. Further research should attempt to determine the
the literature. VJ displacement has been shown to be re-
reason for this difference between men and women. It
lated to performance in several different sports. The
should also be noted that the strength of prediction, as
search for ways to improve VJ displacement has lead in-
determined by the R2 statistic, reveals that foot length is
vestigators to examine many different variables. Howev-
not a strong predictor of VJ displacement. An R2 of 0.08
er, to date, skeletal length measures have not been in-
indicates that 8% of the variability in VJ displacement is
vestigated. The results of this investigation support the
explained by foot length. The relatively low prediction
hypothesis that skeletal length measurements generally
strength of foot length and the fact that none of the other
have no predictive value. The only exception is that foot
skeletal measurements was significant supports the null
length appears to be an intrinsic factor that is predictive
hypothesis that intrinsic skeletal length measures have
of VJ displacement in recreational male athletes. From a
very little influence on VJ displacement in recreational
biomechanical perspective, the length of the foot may
athletes.
serve to provide additional mechanical leverage by which
In this investigation, a recreational athlete was de-
the ankle plantarflexors exert a propulsive force. The re-
fined as any individual who participates in nonprofes-
sults of this investigation revealed that VJ displacement
sional or noncollegiate athletics 3 or more times per week.
is minimally influenced by intrinsic foot length. There-
VJ displacement of this sample is consistent with data
fore, based on the literature, strength and conditioning
presented in other investigation of recreational athletes.
specialists should continue to focus efforts on improving
In this investigation, the mean VJ displacement was 56.8
cm for men and 36.8 cm for women. Davis et al. (9) re- anaerobic muscle power (3, 4, 9), while optimizing body
ported an average VJ displacement of 59.8 cm for male composition to maximize VJ displacement (9, 21, 23, 27–
recreational athletes. Ashley and Weiss (4) reported a 29).
mean VJ displacement of 27.0 cm in college women. This
is compared with a mean VJ displacement of 73.9 cm re- REFERENCES
ported by Davis et al. (8) in Division I college football 1. AL-EISH, E., D. EGAN, AND R. WASSERSUG. Fluctuating asym-
players. No attempt was made to identify the specific metry and low back pain. Evol. Hum. Behav. 25:31–37. 2004.
sport in which the subjects participated. Future investi- 2. ANDERSON, M., J. GIECK, D. PERRIN, A. WELTMAN, R. RUTT,
gations should identify the type of sport in which the sub- AND C. DENEGAR. The relationships among isometric, isotonic,

jects participate. Therefore, the results of this investiga- and isokinetic concentric and eccentric quadriceps and ham-
tion should only be applied to young adult male and fe- string force and three components of athletic performance. J.
Orthop. Sports Phys. Ther. 14:114–120. 1991.
male recreational athletes between the ages of 18 and 40
3. ARAGON-VARGAS, L.F., AND M.M. GROSS. Kinesiological factors
who fit this definition of recreational athlete. Future re- in vertical jump performance: Differences among individuals.
search should attempt to replicate these procedures in J. Appl. Biomech. 13:24–44. 1997.
collegiate, Olympic, and professional athletes who partic- 4. ASHLEY, C.D., AND L.W. WEISS. Vertical jump performance and
ipate in various sporting events. selected physiological characteristics of women. J. Strength
Despite the relatively large number of subjects, the Cond. Res. 8:5–11. 1994.
number of women was low compared with men. The in- 5. BARKER, M., T.J. WYATT, R.L. JOHNSON, M.H. STONE, H.S.
tratester and intertester reliability of the measurements O’BRYANT, C. POE, AND M. KENT. Performance factors, psycho-
were established before the investigation; however, the logical assessment, physical characteristics, and football play-
validity of the measurements remains in question. The ing ability. J. Strength Cond. Res. 7:224–233. 1993.
6. BLACK, W., AND E. ROUNDY. Comparisons of size, strength,
use of a tape measure and palpation of bony landmarks
speed, and power in NCAA Division 1-A football players. J.
by experienced testers has been used by previous inves- Strength Cond. Res. 8:80–85. 1994.
tigators and has face validity; however, criterion-related 7. BLACKBURN, J.R., AND M.C. MORRISSEY. The relationship be-
validity of this technique compared with radiographic tween open and closed kinetic chain strength of the lower limb
measurements was not assessed. and jumping performance. J. Orthop. Sports Phys. Ther. 27:
To date, the study of VJ displacement has relied 430–435. 1998.
140 DAVIS, BOSLEY, GRONELL ET AL.

8. DAVIS, D.S., B. BARNETTE, J. KIGER, J. MIRASOLA, AND S. 20. MCLEAN, S.G., X. HUANG, A. SUE, AND A. VAN DEN BOGERT.
YOUNG. Physical characteristics that predict functional perfor- Sagittal plane biomechanics cannot injure ACL during sidestep
mance in Division I college football players. J. Strength Cond. cutting. Clin. Biomech. 19:828–838. 2004.
Res. 18:115–120. 2004. 21. MCLEOD, W.D., S.C. HUNTER, AND B. ETCHISON. Performance
9. DAVIS, D.S., D. BRISCOE, C. MARKOWSKI, S. SAVILLE, AND C. measurement and percent body fat in the high school athlete.
TAYLOR. Physical characteristics that predict vertical jump per- Am. J. Sports Med. 11:390–397. 1983.
formance in recreational male athletes. Phys. Ther. Sport 4: 22. MCWHORTER, J.W., H. WALLMANN, M. LANDERS, B. ALTENBER-
167–174. 2003. GER, L. LAPORTA-KRUM, AND P. ALTENBERGER. The effect of

10. DOWLING, J.J., AND L. VAMOS. Identification of kinetic and tem- walking, running, and shoe size on volumetrics. Phys. Ther.
poral factors related to vertical jump performance. J. Appl. Bio- Sport 4:87–92. 2003.
mech. 9:95–110. 1993. 23. MILLER, T.A., E.D. MILLER, K.A. KINLEY, J.J. CONGLETON, AND
M.J. CLARK. The effects of training history, player position, and
11. DUYAR, I., AND C. PELIN. Body height estimation based on tibia
body composition on exercise performance in collegiate football
length in different stature groups. Am. J. Phys. Antrop. 122:
players. J. Strength Cond. Res. 16:44–49. 2002.
23–27. 2003.
24. SELBIE, W., AND G. CALDWELL. A simulation study of vertical
12. FRY, A.C., AND W.J. KRAEMER. Physical performance charac-
jumping from different starting postures. J. Biomech. 29:1137–
teristics of American collegiate football players. J. Appl. Sports 1146. 1996.
Sci. Res. 5:126–138. 1991. 25. SHROUT, P.E., AND J.L. FLEISS. Intraclass correlation: Uses in
13. FRY, A.C., W.J. KRAEMER, K.W. BIBI, AND T. EYFORD. Stature assessing rater reliability. Psychol. Bull. 86:420–428. 1979.
variables as discriminators of foot contact during the squat ex- 26. SHULTZ, S.J., C. CARCIA, AND D. PERRIN. Knee joint laxity af-
ercise in untrained females. J. Appl. Sports Sci. Res. 5:77–81. fects muscle activation patterns in the healthy knee. J. Electro.
1991. Kines. 14:475–483. 2004.
14. FUSTER, V., A. JEREZ, AND A. ORTEGA. Anthropometry and 27. THOMAS, T.R., C.J. ZEBAS, M.S. BAHRKE, J. ARAUJO, AND G.L.
strength relationship: Male-female differences. Anthrop. Anz. ETHERIDGE. Physiological and psychological correlates of suc-
56:49–56. 1998. cess in track and field athletes. Br. J. Sports Med. 17:102–109.
15. HARMAN, E.A., M.T. ROSENSTEIN, P.N. FRYKMAN, AND R.M. RO- 1983.
SENSTEIN. The effects of arms and countermovement on verti- 28. UGARKOVIC, D., D. MATAVULJ, M. KUKOLJ, AND J. SLOBODAN.
cal jumping. Med. Sci. Sports Exerc. 22:825–833. 1990. Standard anthropometric, body composition, and strength var-
16. HURD, W.J., T. CHMIELEWSKI, M. AXE, I. DAVIS, AND L. SNY- iables as predictors of jumping performance in elite junior ath-
DER-MACKLER. Difference in normal and perturbated walking
letes. J. Strength Cond. Res. 16:227–230. 2002.
kinematics between male and female athletes. Clin. Biomech. 29. WILLIFORD, H.N., J. KIRKPATRICK, M. SCHARFF-OLSON, D.L.
BLESSING, AND N.Z. WANG. Physical and performance charac-
19:465–472. 2004.
teristics of successful high school football players. Am. J. Sports
17. KUBO, K., Y. KAWAKAMI, AND T. FUKUNAGA. Influence of elastic
Med. 22:859–863. 1994.
properties of tendon structures on jump performance in hu- 30. YOUNG, W., G. WILSON, AND C. BYRNE. Relationship between
mans. J. Appl. Physiol. 87:2090–2096. 1999. strength qualities and performance in standing and run-up
18. LEE, E., B. ETNYRE, H. POINDEXTER, D. SOKOL, AND T. TOON. vertical jumps. J. Sports Med. Phys. Fitness 39:285–292. 1999.
Flexibility characteristics of elite female and male volleyball 31. ZACHAZEWSKI, J.E., D.J. MAGEE, AND W.S. QUILLEN. Athletic
players. J. Sports Med. 29:49–51. 1989. Injuries and Rehabilitation. Philadelphia: W.B. Saunders,
19. LEVANGIE, P.K., AND C.C. NORKIN. Basic concepts in biome- 1996. pp. 250–251.
chanics. In: Joint Structure and Function: A Comprehensive
Analysis (3rd ed.). Philadelphia: F.A. Davis Company, 2001. pp. Address correspondence to D. Scott Davis, dsdavis@hsc.
28–47. wvu.edu.

You might also like