Relationship of Body Composition to Somatotype M. H. SLAUGHTER AND T. G. LOHMAN Physical Fitness Research Laboratory, University of Illinois, Urbana, Illinois 61 801
KEY WORDS Morphology.
Somatotype
. Body
composition . Body structure
ABSTRACT The purpose of this study is to determine the relationship in college-aged women between somatotype using both Sheldon’s (’69) and Heath and Carter’s (’67) procedures, and body composition, a s measured by whole-body 40K counting and body density. Sheldon’s endomorphy is closely associated with height and weight; Heath and Carter’s first component is significantly related to weight and body fatness. Lean body mass (LBM) as a weight or as a percent is not closely related to Sheldon’s mesomorphy or Heath and Carter’s second component. However, when LBM and height are used as independent variables to estimate somatotype, both variables are significantly related to Heath and Carter’s second component, accounting for 61% of the variance. Thus, Heath and Carter’s second component is significantly associated with LBM for a given body height. Most of the variation in Sheldon’s ectomorphy and Heath and Carter’s third component can be accounted for by weight and height. Sheldon’s somatotype for all three components is not as closely related to body composition as Heath-Carter’s. Body composition, as measured by either 40K counting or body density, is found to be important in accounting for variation in Heath and Carter’s first and second components.
Since Sheldon’s publications (Sheldon et al., ’40; Sheldon et al., ’54) of his somatotyping schema (based primarily on height/weight ratio and subjective evaluations of posed somatotype photographs), attempts have been made to relate the three components (endomorphy, mesomorphy and ectomorphy) to a variety of anthropometric measures and body composition (Cureton, ’47, ’51; Parnell, ’54, ’58; Damon et al., ’64; Heath and Carter, ’66, ’67; Laubach and McConville, ’66; Sheldon et al., ’69; Munroe et al., ’69; Clarke, ’71). In general, endomorphy and ectomorphy have been more closely related than mesomorphy to body composition measures. More recently, Heath and Carter (‘67) modified Sheldon’s method with their somatotype largely determined by anthropometric measures. Conceptually, the Heath-Carter system would appear to be more closely associated with body composition, i.e., body fatness and lean body mass. With the development of more sophisticated techniques for estimating body composition, i.e. densitometry, hydrometry, roentgenogramy and body potassium methods, AM. .I. PHYS. ANTHROP., 44: 237-244.
efforts have been made to relate per cent body fat and lean body mass to somatotype. Dupertuis et al. (‘51) found a significant association between endomorphy and specific gravity. Using the x-ray analysis of tissue development, Reynolds and Asakawa (’50) attempted to correlate body form, evaluated in terms of Sheldon’s system, and the composition of an extremity, taken as an indicator of body composition. When the means were expressed as percentages of the total breadth of calf, the male endomorphs were characterized by highest fat, mesomorphs by highest muscle breadth, and ectomorphs by highest bone breadth. Wilmore (‘70) in an attempt to validate Heath and Carter’s first and second components (using their anthropometric method, ’67) as reflecting body composition, found only a moderate relationship between the first component (endomorphy) and percent body fat and practically no relationship between the second component (mesomorphy) and the absolute weight of lean body mass in college-aged men and women. This latter finding, however, is not surprising since Heath and 237
238
M. H . SLAUGHTER A N D T. G. LOHMAN
Carter’s second component (mesomorphy) is defined as the lean body weight per unit of height, rather than the absolute weight of lean body mass. The purpose of this study is to determine the relationship in college-aged women between somatotype and body composition. Two methods of measuring somatotypes are employed, namely, Sheldon’s modification of his original method (Sheldon et al., ’69) and Heath-Carter’s Anthropometric Method (‘67). Body composition is determined by two methods, namely whole-body 40K counting and body density. It is hypothesized that Heath-Carter’s somatotype (based on their Anthropometric Method) is more closely related to body composition than Sheldon’s somatotype. METHODS
The volunteer subjects were 31 healthy, white college women with a mean age of 19.8 years ranging from 17 to 22 years. They were enrolled in a physical education weight control and conditioning class at the University of Illinois. No nutritional deficiencies were apparent as measured by blood hemoglobin and a dietary survey. Body potassium, body density, and selected skinfold and anthropometric measurements were taken on all subjects twice within seven days. The mean value for the two test sessions was used to compare the body composition measurements to the somatotype ratings. Somatotype procedures The subjects were photographed in twopiece bathing suits in the standard somatotype position (Sheldon et al., ’54). Two methods for determining somatotypes were used: Sheldon’s modification of his somatotyping method (Sheldon et al., ’69), and the Heath-Carter Anthropometric Method (‘67). Sheldon’s somatotyping process (including the use of photographs) consists of deriving three sets of parameters, thought to reflect the individual’s inherited physique. The first parameter, the somatotyp ing ponderal index (SPI) is derived from the subject’s height over the cube root of her weight. The SPI measures the individual’s maximal achieved mass over the surface. The second parameter, the trunk index (TI), is derived from the photographic area of the thoracic trunk over the abdom-
inal trunk, both measured by a planimeter. This index provides a parameter for differentiating quantitatively between endomorphy and mesomorphy. The third parameter is the height of the subject. When the other parameters are known, height provides a measurement for determining the ectomorphic component. The somatotype of the individual is determined from the Basic Tables for Somatotyping (Sheldon et al., ’69) and is expressed in three numerals. The first number refers to the degree of endomorphy in the physique and means relative predominance of soft roundness throughout the various regions of the body. The second number refers to the mesomorphic component in the physique and means relative predominance of muscle, bone, and connective tissue. The third number refers to the ectomorphic component and means relative predominance of linearity and fragility. Heath (‘63) modified Sheldon’s original method (Sheldon et al.,’54) and with Carter (‘67) further objectified the system by incorporating anthropometric measurements and redefining the somatotype and its components. Heath and Catter define the somatotype as a description of the present morphological conformation. This somatotype is expressed by a three-numeral rating of the three primary components of physique which identify individual features of body morphology and body Composition. The first component refers to relative fatness; the second component refers to relative musculo-skeletal development per unit of height; and the third component refers to relative linearity. Essentially, there are three different ways for obtaining the somatotype rating, namely, Heath-Carter Anthropometric Somatotype, the Photoscopic Somatotype and the Anthropometric plus Photoscopic Somatotype. For the purposes of this study, we used the Anthropometric Method.
Body composition procedures Body composition and structure were assessed by two methods: whole-body potassium-40 (4OK) and densitometry (underwater weighing). Potassium is distributed throughout the intracellular fluid of all tissues and is mostly concentrated in the skeletal muscle (Conway, ’57). Considerable evidence has been accumulated to indicate that the weight of the lean body
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
mass is proportional to the total amount of body potassium in several animal species. Therefore, the measurement of body potassium can be used as an index of lean body mass and in the case of humans, can be calculated from the constants proposed by Forbes and Hursch ('63) and modified for women from 2.66 gm K/kg LBM to 2.50 gm K/kg LBM (Forbes et al., '68). With the development of highly sensitive whole-body counters over the past 15 years, total body potassium can be estimated from the measurement of the body's naturally-occurring 40K gamma radiation; potassium-40 being found in a constant proportion of stable potassium (0.012%). In our study 40Kwas measured using a 477 liquid scintillation whole-body counter described in detail by Twardock et al. ('66). Body density (Db) was measured by underwater weighing (Buskirk, '61) with procedures and instrumentation modified according to Akers and Buskirk ('69). Body weight in air was taken on a Howe-Richardson scale (Model 5600) and recorded to the nearest 2 5 g. The subject was seated on a platform in a tank of water warmed to approximately 36°C for the underwater weight determination. Underwater weight was recorded from two linear voltage differential transformer transducers, upon which the weighing platform rested, at the end of a forced expiration prior to the func-
tional pulmonary residual volume measurement. Functional residual volume was measured instantaneously with the underwater weight by an open circuit sevenminute nitrogen washout (02dilution) as described by Darling et al. ('40). Relative body fatness (% fat) was computed from body density accordingly (Brozek et al., '63): % fat = (4.570/D - 4.142)lOO. Correlation coefficients were computed to assess week to week reliability of all body composition measures. Zero order and multiple correlations were computed to show relationships of body composition to somatotype. Where appropriate, multiple regression equations were calculated to show the association of lean body mass and somatotype holding constant body weight and height. RESULTS
The age and physical characteristics of the subjects are presented in table 1. The subjects range on body fatness from 14.4% to 38.2% with a mean of 25.2% and a S.D. of 5.6% as estimated by body density. These results are comparable to those found by Wilmore ('70) and Katch and McArdle ('73) in college-aged women. Both studies estimate the fat content for women to be approximately 25% with a S.D. between 4% and 6 % . The mean fat content was 31.6% and a S.D. of 7.1% based on
TABLE 1 Age and physical characteristics of women ( N = 3 1 ) Variable
Age (yrs.) Height (cm) Weight (kg) % fat (4oK) % fat (density) LBM (4oK) lean body mass % lean body mass (4OK) Lean body mass (density) % LBM (density) Triceps skinfold (cm) Subscapular skinfold (cm) Suprailiac skinfold (cm) Calf skinfold (cm) Biceps circumference (cm) Calf circumference (cm) Elbow width (cm) Knee width (cm) Endomorphic component Mesomorphic component Ectomorphic component First component Second component Third component
239
Mean
S.D.
Range
19.8 165.9 59.73 31.6 25.2 40.5 68.4 44.5 74.8 19.1 15.3 17.9 13.0 26.5 34.3 6.1 8.6 5.18 3.45 3.35 5.10 3.16 2.76
1.3 5.8 7.4 7.1 5.6 4.7 6.9 4.5 5.5 5.2 6.3 6.0 4.0 2.1 2.2 2.8 0.4 0.93 0.86 1.25 1.30 1.04 1.26
17.3-22.4 152.8-175.9 38.7-77.0 18.340.6 14.4-38.2 26.648.9 59.4-81.7 29.6-51.8 61.845.6 9.4-28.3 6.8-31.1 6.0-32.8 5.5-21.6 19.6-30.7 30.0-37.6 5.7-6.7 7.7-9.5 2.5-7.0 2.05.5 1.0-6.0
2.0-6.5 0.55.5 0.5-6.0
240
M. H. SLAUGHTER AND T. G. LOHMAN
the 40K method, ranging from 18.3% to ponent (r = 0.88). Much less association 40.6%. These results are comparable to is found between Sheldon’s mesomorphy those found by Novak (‘72) for 18- and 25- and Heath and Carter’s second component year old women (33.0 % ) using whole body (r = 0.54). These correlations are similar 40Kcounting. The mean difference between to those by Morton (‘67) who compared the two methods of estimating body fatness Sheldon’s Trunk Index method of somato(40K and body density) in our study may be typing and Heaths Anthroposcopic Method due to the lower potassium content of the on children. lean body mass (2.33 gmfkg) as suggested Sheldon’s Trunk-Index Method produced by the data for Wormesley et al. (’72) higher somatotype designations for the rather than the 2.66 gmlkg used by Novak three components than did the Heath-Car(’72) and the 2.50 gm/kg proposed by ter Method (table 1). Morton (’67) also Forbes et al. (‘68) and used in our study. obtained similar results in comparing ShelThe anthropometric measures show general don’s Trunk-Index method and Heaths agreement with those reported by Katch Anthroposcopic Method of somatotyping. and McArdle (‘73) and Sloan et al. (‘62) The differences between the means of comfor college-aged women. parable somatotyping components as asAll reliability coefficients for body com- sessed by Sheldon’s method and Heathposition measures between the two test Carter’s Anthropometric Method show that days (N = 31) are above 0.90; 0.92 for per- Sheldon’s mesomorphy and ectomorphy are cent fat from both 40K and body density, significantly different (p < 0.05) from 0.94 for LBM from density and 0.93 for Heath and Carter’s parallel components, LBM from 40K. The correlation between whereas the mean difference between percent fat by the two methods (body den- Sheldon’s endomorphy and Heath and Carsity and 40K) was 0.82 and between LBM ter’s first component does not attain sta(kg) by the two methods, 0.86. The reli- tistical significance. ability coefficients for the anthropometric Zero-order correlation coefficients for measures used for the Heath-Carter so- selected body composition measurements matotyping process are as follows: height and somatotype ratings from both Sheldon 0.99; weight, 0.99; triceps skinfold, 0.94; and Heath-Carter are presented in table 2. subscapular skinfold, 0.95; suprailiac skin- Percent fat, estimated from 40K measurefold, 0.94; calf skinfold, 0.92; biceps cir- ments, correlates 0.66 with Sheldon’s encumference, 0.97; calf circumference, domorphy and 0.69 with Heath and Car0.92; elbow width, 0.73; and knee width, ter’s first component. Percent fat, estimated 0.79. from body density, correlates 0.75 with The zero-order correlations between par- Sheldon’s endomorphy and 0.74 with allel components of Sheldon and Heath- Heath and Carter’s first component. AbCarter’s somatotypes indicate considerable solute lean body mass (LBM), estimated association for Sheldon’s endomorphy and from 40K measurements, correlates 0.34 Heath and Carter’s first component (r = with Sheldon’s mesomorphy and 0.15 with 0.79) and ectomorphy and the third com- Heath and Carter’s second component. TABLE 2
Zero-order coefficients for body composition and somatotype ratings from Sheldon, and Heath and Carter ( N = 31 subjects) Heath-Carter
Sheldon
Variable
Age Height Weight % fat (*OK) % fat (density) LBM (4OK) LBM (density) 7c LBM (4OK) % LBM (density) 1
endo
mew
ecto
0.12
-0.23 0.02 0.48
-0.06
- 0.08 0.80
1
1
0.66 1
0.16
0.75 1 0.23 0.41 1
0.2 1 0.34 0.42
-0.66 1
-0.16
-0.75
-0.21
Significant at the 0.05 level
1
1
0.54 1 -0.43 1 -0.45 1 -0.54 1 - 0.03 -0.11 0.45 1 0.54 1
1st comp
0.21 0.03 0.65 1
2nd comp
3rd comp
0.08 -0.45 0.47
1
0.03 0.33
1
-0.69 1
0.691
0.361
-0.54
0.74 1 0.07 0.25
0.45 0.15 0.20
1
-0.66 1
- 0.691
- 0.361
-0.23 - 0.34 - 0.54 1
-0.74
-0.45
1
1
1
0.661
24 1
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
Similar results are found using LBM estimated from body density. The correlation between LBM (40K) and Sheldon’s ectomorphy is -0.03, and Heath and Carter’s third component, - 0.23. Similar results are found using LBM estimated from body density. To better study the association of body composition measurements with Sheldon’s and Heath and Carter’s somatotype components, multiple regression analysis was used to determine the relation of various combinations of age, body weight, height, and body composition as predictors of somatotype (tables 3, 4). The coefficients of determination (R2X 100) from age, height, and weight as independent variables are 83% for Sheldon’s endomorphy but only 58% for Heath and Carter’s first component (table 3). When either percent fat or LBM as measured by either 40K or body density in addition to age, height and weight is used to estimate somatotype, the coefficients increased to 85% (not significantly different from 83% without percent fat) for Sheldon’s endomorphy and to 65% for Heath and Carter’s first component
(significantly different from 58% without percent fat [table 31 ). The coefficients of determination from age, height, and weight as independent variables are 32% for Sheldon’s mesomorphy and 68% for Heath and Carter’s second component (table 3). Adding LBM (body density) as an independent variable increases the variation accounted for only slightly to 37% for Sheldon’s mesomorphy and to 72% for Heath and Carter’s second component. The coefficients of determination from age, height, and weight as independent variables are 79% for Sheldon’s ectomorphy and 90% for Heath-Carter’s third component. Adding LBM (body density) as an independent variable increases the variation accounted for slightly for Sheldon’s ectomorphy (81%) but not for Heath and Carter’s third component (table 3). When LBM (body density) and height are used to estimate somatotype, both independent variables are significantly associated with Sheldon’s mesomorphy and Heath and Carter’s second component (table 4). The coefficient of determination
TABLE 3
Coefficients of determination between body composition and somatotypes ( N = 3 1 ) Sheldon endo
meso
ecto
1st comp
2nd comp
3rd comp
83 85 85 85 85 85 85
32 35 37 35 37 35 37
79 79 79 79
58 66 66 66 64 66 66
69 71 72 71 72 71 72
90 90 90 90 90 90 90
Age, height, weight Plus % fat (4OK) Plus % fat (density)
Plus Plus Plus Plus
Heath-Carter
LBM (4OK) LBM (density) % LBM (40K) % LBM (density)
81
79 79
TABLE 4
Regression coefficients for the association of height and lean body mass and height, weight and lean body mass w i t h Sheldon’s mesomorphy and Heath and Carter’s Second Component ( N = 3 1 ) 1 Sheldon’s Mesomorphy
sb
b
Height
LBM Weight Intercept
RZ X 100 SE 1 2
-0.074 0.1402
0.030 -0.038
9.47 32 0.73
LBM estimated from body density. p < 0.05.
Heath and Carter’s second component b
sb
-0.178 2 0.193 2
0.028 0.035
24.1 61
0.67
Sheldon’s Mesomorphy b
sb
-0.066 2 0.031 0.096 0.059 0.030 0.030 8.37 35 73
Heath and Carter’s second component b
sb
-0.158 2 0.025 0.081 0.047 0.0752 0.024 21.27 72 59
242
M. H. SLAUGHTER AND T. G. LOHMAN
was 32% for mesomorphy and 61 % for the second component. When weight in addition to height and LBM is used to estimate somatotype, coefficients of determination increased to 72% for Heath and Carter’s second component but only to 35% for Sheldon’s mesomorphy (table 4). DISCUSSION
Because Heath-Carter’s somatotype describes body morphology as well as body composition, not all of the variation in somatotype is expected to be accounted for by measures of body composition. Furthermore, both potassium-40 and body density methods have inherent error from both the biological and technical standpoints. From biological sources both methods are based on the assumption that the lean body mass has a constant potassium content and a constant density for all individuals. While this assumption has been found to be relatively accurate, the exact limitations of both methods are still unknown. Technical errors also contribute to measurement error for both methods but were reduced to some extent by using the mean of two determinations, one week apart. Percent fat (estimated from either body density or 4OK) is as closely related with Sheldon’s endomorphy based on somatotyping ponderal index and trunk index as with Heath and Carter’s first component, determined directly from skinfolds. Similar relationships between percent fat and Sheldon’s endomorphy, as well as Heath and Carter’s f i s t component (based on the sum of three skinfolds), have been reported by other investigators (Brozek and Keys, ’51; Durnin and Rahaman, ’67). Also high negative correlations have been found between endomorphy and specific gravity (Dupertuis et al., ’51; Brozek and Keys, ’52). The correlations between LBM (by both methods) and both Sheldon’s mesomorphy and Heath and Carter’s second component are quite low and indicate that LBM as an absolute weight is not significantly related to Sheldon’s mesomorphy or Heath and Carter’s second component. The correlations between percent LBM and mesomorphy as well as the second component are also quite low and in all cases negative. Thus neither the absolute nor the relative amount of LBM is associated with meso-
morphy. Both Dupertuis et al. (‘51) and Brozek and Keys (‘52) also found low positive correlations between mesomorphy and specific gravity (an index of the fat content and lean body mass content of the body). As Carter (’72) has stated “the second component refers to relative musculoskeletal development per unit of height” and Sheldon’s mesomorphy is based on the relation of the abdominal trunk to thoracic trunk. Wilmore (‘71) interpreted Carter and Heath’s definition of their second component incorrectly when he attempted to validate it by use of the absolute weight of lean body mass. The correlation of 0.16 between LBM (kg) and Heath and Carter’s second component (Wilmore, ’70) is similar to the correlation found in this study from both methods of estimating LBM (0.15 and 0.20). Lean body mass estimated by either method is not significantly correlated (p > 0.05) with Sheldon’s ectomorphy or with Heath and Carter’s third component. Dupertuis et al. (‘51) and Brozek and Keys (’52) also found that ectomorphy was not closely related to specific gravity. In studying the coefficients of determination which resulted from the use of age, height, weight and percent fat in multiple regression analysis, Sheldon’s endomorphy is seen to be largely associated with weight and height, whereas Heath and Carter’s first component is more closely associated with weight and body fatness. Heath and Carter’s second component can be predicted closer from age, height and weight than Sheldon’s mesomorphy. In a study by Damon et al. (’62) Sheldon’s mesomorphy (rated by his 1954 method) also could not be predicted as closely using a combination of anthropometric measures in 369 white male soldiers with 44% of the variation accounted for in mesomorphy as compared to 61% and 81% of the variance accounted for in endomorphy and ectomorphy, respectively (Damon et al., ’62). When estimating Sheldon’s ectomorphy and the third component of Heath and Carter, body weight and height are the only two significant independent variables. In studying the coefficients of determination when only the independent variables of height and LBM are used in multiple regression analysis, Heath and Carter’s second component is found to be signifi-
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
cantly related to height and LBM (body density). However, when body weight is used along with height and LBM as independent variables, the regression coefficient for LBM is no longer statistically significant. Thus, height and weight appear to be major contributors to accounting for variation in the second component with a somewhat smaller contribution from LBM (table 4). The coefficient of determination for Sheldon’s mesomorphy when only the independent variables of height and LBM are used, is quite low (32 % ) and increases only 3% when weight is added. Furthermore, the regression coefficient of LBM is no longer significant. Thus, Sheldon’s mesomorphy does not appear to be associated with height, LBM and weight. In summary, Sheldon’s endomorphy is largely associated with weight and height whereas Heath and Carter’s first component is more closely associated with weight and body fatness. Neither the absolute nor the relative amount of LBM is associated with Sheldon’s mesomorphy or Heath and Carter’s second component. Height and lean body mass together are found to be more closely related to Heath and Carter’s second component than to Sheldon’s mesomorphy. ACKNOWLEDGMENTS
This research was conducted in cooperation with the Physical Fitness Research Laboratory, Department of Physical Education, University of Illinois. We wish to express appreciation to Richard Boileau, Kirk Cureton, Tanya Hildebrandt, Jerry Mayhew, and Benjamin Massey for their contribution to this study. LITERATURE CITED Akers, R., and E. R. Buskirk 1969 An underwater weighing system utilizing forcecube transducers. J. Appl. Physiol., 26; 649-652. Allen, T. H., M. T. Peng, K. P. Chen, T. F. Huang, C. Chang and H. S. Fang 1956 Prediction of total adiposity from skinfolds and the curvilinear relationship between external and internal adiposity. Metabolism, 5 ; 346352. Boileau, R. A,, B. H. Massey and J. E. Misner 1973 Body composition changes in adult men during selected weight training and jogging programs. Res. Quart., 44; 158-168. Brozek, J., F. Grande, J. T. Anderson and A. Keys 1963 Densitometric analysis of body composition: revision of some quantitative assumptions. Ann. N. Y.Acad. Sci., 110; 113-140. Brozek, J., and A. Keys 1951 The evaluation of
243
leanness-fatness in man: norms and interrelationships. Brit. J. Nutr., 5; 194-206. 1952 Body build and body composition. Science, 1 1 6; 140-142. Buskirk, E. R. 1961 Underwater weighing and body density: a review of procedures. In: Techniques for Measuring Body Composition. J. Brozek, ed. Nat. Acad. of Sci., Washington, D. C., pp. 90-107. Carter, J. E. L. 1972 The Heath-Carter Somatotype Method. J. E. Carter, publisher, San Diego. Clarke, H. 1971 Physical and Motor Tests in the Medford Boy’s Growth Study. Prentice-Hall, New York. Conway, E. J. 1957 Nature and significance of concentration relations of potassium and sodium ions in the skeleton muscle. Physiol. Rev., 37; 84-132. Cureton, T. K., Jr. 1947 Physical Fitness, A p praisal and Guidance. Henry Kimpton, London. 1951 Physical Fitnessof Champion Athletes. Univ. of Illinois Press, Urbana, Ill. Damon, A,, H. K. Bleibtreu, 0. Elliot and E. Giles 1962 Predicting somatotype from body measurements. Am. J. Phys. Anthrop., 20; 461-474. Darling, R. C., A. Cournand and D. W. Richards 1940 Studies on intrapulmonary mixture of gases. 111. An open-circuit method for measuring residual air. J. Clin. Invest., 19: 606-618. Dupertuis, C. S., G. C. Pitts, E. F. Osserman, W. C. Welham and A. R. Behnke 1951 Relation of specific gravity to body build in a group of healthy men. J. Appl. Physio., 3; 676-680. Durnin, T. V . G. A,, and M. M. Rahaman 1967 The assessment of the amount of fat in the human body from measurements of skinfold thickness. Brit. J. Nutr., 21 ; 681-689. Forbes, G. B., and J. B. Hursh 1963 Age and sex trends in lean body mass calculated from K40 measurements: with a note on the theoretical basis for the procedure. Ann. N. Y. Acad. Sci., 1 1 0 ; 255-263. Forbes, G., F. Schultz, C. Cafarelli and G. Amirhakimi 1968 Effects of body size potassium-40 on measurement in the whole-body counter. Health Physics, 15: 4 3 5 4 4 2 . Heath, B. H. 1963 Need for modification of somatotype methodology. Am. J. Phys. Anthrop., 21; 227-233. Heath, B. H., and J. E. L. Carter 1966 A comparison of somatotype methods. Am. J. Phys. Anthrop., 24: 87-99. 1967 A modified somatotype method. Am. J. Phys. Anthrop., 27: 57-74. Katch, F. I., and W. McArdle 1973 Prediction of body density from simple anthropometric measurements in collegeage men and women. Human Biology, 45: 445-454. Keys, A., and J. Brozek 1953 Body fat in adult man. Physio. Rev., 33; 2 4 5 3 2 5 . Morton, A. R. 1967 Comparison of Sheldon’s Trunk-Index and Anthroposcopic Methods of Somatotyping in Their Relationships to the Maturity, Structure, and Motor Ability of the Same Boys Nine through Sixteen Years of Age. Microcard doctoral dissertation, University of Oregon. Munroe, R., H. H. Clarke and B. H. Heath 1969 A somatotype method for young boys. Am. J. Phys. Anthrop., 30; 195-202. Novak, L. 1972 Aging, total body potassium,
244
M. H. SLAUGHTER AND T. G. LOHMAN
fat-free mass and cell mass in males and females between ages 18 and 85 years. J. Geron., 27:
438443. Parnell, R. W. 1954 Somatotyping by physical anthropometry. Am. J. Phys. Anthrop., 12: 209-
239. 1958 Behavior and Physique. Edward Arnold, Ltd., publishers, London. Reynolds, E. L., and T. Asakawa 1950 A comparison of certain aspects of body structure and body shape in 200 adults. A m . J. Phys. Anthrop., 8: 343365. Sheldon, W. H., C. W. Dupertuis and E. McDermott 1954 Atlas of Men. Harper Bros., New York. Sheldon, W. H., N. D. C. Lewis and A. M. Tenny 1969 Psychotic Patterns and Physical Constitution: Schizophrenia, Current Concepts and Research. D. V. Siva Sankar, ed. PJD Publications, Hicksville, N. Y. Sheldon, W. H., J. J. Stevens and W. B. Tucker 1940 The Varieties of Human Physique. Harper Bros., New York. Sinclair, G. D. 1966 Stability of Physique Types
of Boys Nine through Twelve Years of Age. Microcard master’s thesis, University of Oregon. 1969 Stability of Somatotype Components of Boys Ages Twelve through Seventeen Years and Their Relationships to Selected Physical and Motor Factors. Microcard doctoral dissertation, University of Oregon. Sloan, A. W., J. J. Burt and C. S. Blyth 1962 Estimation of body fat in young women. J. Appl. Physio., 1 7 : 967-970. Twardock, A. R., T. G. Lohman, G. S. Smith and B. C. Breidenstein 1966 The Illinois animal science counter: performance characteristics and animal radioactivity measurement procedures. J. Animal Sci., 25: 1209-1217. Wilmore, J. 1970 Validation o f the first and second components of the Heath-Carter modified somatotype method. Am. J. Phys. Anthrop., 32:
369372. Womersley, J., K. Boddy, P. King and J. V. G. A. Durnin 1972 A comparison of the fat-free mass of young adults estimated by anthropometry, body density, and total body potassium content. Clin. Sci., 43: 468475.
KEY WORDS Morphology.
Somatotype
. Body
composition . Body structure
ABSTRACT The purpose of this study is to determine the relationship in college-aged women between somatotype using both Sheldon’s (’69) and Heath and Carter’s (’67) procedures, and body composition, a s measured by whole-body 40K counting and body density. Sheldon’s endomorphy is closely associated with height and weight; Heath and Carter’s first component is significantly related to weight and body fatness. Lean body mass (LBM) as a weight or as a percent is not closely related to Sheldon’s mesomorphy or Heath and Carter’s second component. However, when LBM and height are used as independent variables to estimate somatotype, both variables are significantly related to Heath and Carter’s second component, accounting for 61% of the variance. Thus, Heath and Carter’s second component is significantly associated with LBM for a given body height. Most of the variation in Sheldon’s ectomorphy and Heath and Carter’s third component can be accounted for by weight and height. Sheldon’s somatotype for all three components is not as closely related to body composition as Heath-Carter’s. Body composition, as measured by either 40K counting or body density, is found to be important in accounting for variation in Heath and Carter’s first and second components.
Since Sheldon’s publications (Sheldon et al., ’40; Sheldon et al., ’54) of his somatotyping schema (based primarily on height/weight ratio and subjective evaluations of posed somatotype photographs), attempts have been made to relate the three components (endomorphy, mesomorphy and ectomorphy) to a variety of anthropometric measures and body composition (Cureton, ’47, ’51; Parnell, ’54, ’58; Damon et al., ’64; Heath and Carter, ’66, ’67; Laubach and McConville, ’66; Sheldon et al., ’69; Munroe et al., ’69; Clarke, ’71). In general, endomorphy and ectomorphy have been more closely related than mesomorphy to body composition measures. More recently, Heath and Carter (‘67) modified Sheldon’s method with their somatotype largely determined by anthropometric measures. Conceptually, the Heath-Carter system would appear to be more closely associated with body composition, i.e., body fatness and lean body mass. With the development of more sophisticated techniques for estimating body composition, i.e. densitometry, hydrometry, roentgenogramy and body potassium methods, AM. .I. PHYS. ANTHROP., 44: 237-244.
efforts have been made to relate per cent body fat and lean body mass to somatotype. Dupertuis et al. (‘51) found a significant association between endomorphy and specific gravity. Using the x-ray analysis of tissue development, Reynolds and Asakawa (’50) attempted to correlate body form, evaluated in terms of Sheldon’s system, and the composition of an extremity, taken as an indicator of body composition. When the means were expressed as percentages of the total breadth of calf, the male endomorphs were characterized by highest fat, mesomorphs by highest muscle breadth, and ectomorphs by highest bone breadth. Wilmore (‘70) in an attempt to validate Heath and Carter’s first and second components (using their anthropometric method, ’67) as reflecting body composition, found only a moderate relationship between the first component (endomorphy) and percent body fat and practically no relationship between the second component (mesomorphy) and the absolute weight of lean body mass in college-aged men and women. This latter finding, however, is not surprising since Heath and 237
238
M. H . SLAUGHTER A N D T. G. LOHMAN
Carter’s second component (mesomorphy) is defined as the lean body weight per unit of height, rather than the absolute weight of lean body mass. The purpose of this study is to determine the relationship in college-aged women between somatotype and body composition. Two methods of measuring somatotypes are employed, namely, Sheldon’s modification of his original method (Sheldon et al., ’69) and Heath-Carter’s Anthropometric Method (‘67). Body composition is determined by two methods, namely whole-body 40K counting and body density. It is hypothesized that Heath-Carter’s somatotype (based on their Anthropometric Method) is more closely related to body composition than Sheldon’s somatotype. METHODS
The volunteer subjects were 31 healthy, white college women with a mean age of 19.8 years ranging from 17 to 22 years. They were enrolled in a physical education weight control and conditioning class at the University of Illinois. No nutritional deficiencies were apparent as measured by blood hemoglobin and a dietary survey. Body potassium, body density, and selected skinfold and anthropometric measurements were taken on all subjects twice within seven days. The mean value for the two test sessions was used to compare the body composition measurements to the somatotype ratings. Somatotype procedures The subjects were photographed in twopiece bathing suits in the standard somatotype position (Sheldon et al., ’54). Two methods for determining somatotypes were used: Sheldon’s modification of his somatotyping method (Sheldon et al., ’69), and the Heath-Carter Anthropometric Method (‘67). Sheldon’s somatotyping process (including the use of photographs) consists of deriving three sets of parameters, thought to reflect the individual’s inherited physique. The first parameter, the somatotyp ing ponderal index (SPI) is derived from the subject’s height over the cube root of her weight. The SPI measures the individual’s maximal achieved mass over the surface. The second parameter, the trunk index (TI), is derived from the photographic area of the thoracic trunk over the abdom-
inal trunk, both measured by a planimeter. This index provides a parameter for differentiating quantitatively between endomorphy and mesomorphy. The third parameter is the height of the subject. When the other parameters are known, height provides a measurement for determining the ectomorphic component. The somatotype of the individual is determined from the Basic Tables for Somatotyping (Sheldon et al., ’69) and is expressed in three numerals. The first number refers to the degree of endomorphy in the physique and means relative predominance of soft roundness throughout the various regions of the body. The second number refers to the mesomorphic component in the physique and means relative predominance of muscle, bone, and connective tissue. The third number refers to the ectomorphic component and means relative predominance of linearity and fragility. Heath (‘63) modified Sheldon’s original method (Sheldon et al.,’54) and with Carter (‘67) further objectified the system by incorporating anthropometric measurements and redefining the somatotype and its components. Heath and Catter define the somatotype as a description of the present morphological conformation. This somatotype is expressed by a three-numeral rating of the three primary components of physique which identify individual features of body morphology and body Composition. The first component refers to relative fatness; the second component refers to relative musculo-skeletal development per unit of height; and the third component refers to relative linearity. Essentially, there are three different ways for obtaining the somatotype rating, namely, Heath-Carter Anthropometric Somatotype, the Photoscopic Somatotype and the Anthropometric plus Photoscopic Somatotype. For the purposes of this study, we used the Anthropometric Method.
Body composition procedures Body composition and structure were assessed by two methods: whole-body potassium-40 (4OK) and densitometry (underwater weighing). Potassium is distributed throughout the intracellular fluid of all tissues and is mostly concentrated in the skeletal muscle (Conway, ’57). Considerable evidence has been accumulated to indicate that the weight of the lean body
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
mass is proportional to the total amount of body potassium in several animal species. Therefore, the measurement of body potassium can be used as an index of lean body mass and in the case of humans, can be calculated from the constants proposed by Forbes and Hursch ('63) and modified for women from 2.66 gm K/kg LBM to 2.50 gm K/kg LBM (Forbes et al., '68). With the development of highly sensitive whole-body counters over the past 15 years, total body potassium can be estimated from the measurement of the body's naturally-occurring 40K gamma radiation; potassium-40 being found in a constant proportion of stable potassium (0.012%). In our study 40Kwas measured using a 477 liquid scintillation whole-body counter described in detail by Twardock et al. ('66). Body density (Db) was measured by underwater weighing (Buskirk, '61) with procedures and instrumentation modified according to Akers and Buskirk ('69). Body weight in air was taken on a Howe-Richardson scale (Model 5600) and recorded to the nearest 2 5 g. The subject was seated on a platform in a tank of water warmed to approximately 36°C for the underwater weight determination. Underwater weight was recorded from two linear voltage differential transformer transducers, upon which the weighing platform rested, at the end of a forced expiration prior to the func-
tional pulmonary residual volume measurement. Functional residual volume was measured instantaneously with the underwater weight by an open circuit sevenminute nitrogen washout (02dilution) as described by Darling et al. ('40). Relative body fatness (% fat) was computed from body density accordingly (Brozek et al., '63): % fat = (4.570/D - 4.142)lOO. Correlation coefficients were computed to assess week to week reliability of all body composition measures. Zero order and multiple correlations were computed to show relationships of body composition to somatotype. Where appropriate, multiple regression equations were calculated to show the association of lean body mass and somatotype holding constant body weight and height. RESULTS
The age and physical characteristics of the subjects are presented in table 1. The subjects range on body fatness from 14.4% to 38.2% with a mean of 25.2% and a S.D. of 5.6% as estimated by body density. These results are comparable to those found by Wilmore ('70) and Katch and McArdle ('73) in college-aged women. Both studies estimate the fat content for women to be approximately 25% with a S.D. between 4% and 6 % . The mean fat content was 31.6% and a S.D. of 7.1% based on
TABLE 1 Age and physical characteristics of women ( N = 3 1 ) Variable
Age (yrs.) Height (cm) Weight (kg) % fat (4oK) % fat (density) LBM (4oK) lean body mass % lean body mass (4OK) Lean body mass (density) % LBM (density) Triceps skinfold (cm) Subscapular skinfold (cm) Suprailiac skinfold (cm) Calf skinfold (cm) Biceps circumference (cm) Calf circumference (cm) Elbow width (cm) Knee width (cm) Endomorphic component Mesomorphic component Ectomorphic component First component Second component Third component
239
Mean
S.D.
Range
19.8 165.9 59.73 31.6 25.2 40.5 68.4 44.5 74.8 19.1 15.3 17.9 13.0 26.5 34.3 6.1 8.6 5.18 3.45 3.35 5.10 3.16 2.76
1.3 5.8 7.4 7.1 5.6 4.7 6.9 4.5 5.5 5.2 6.3 6.0 4.0 2.1 2.2 2.8 0.4 0.93 0.86 1.25 1.30 1.04 1.26
17.3-22.4 152.8-175.9 38.7-77.0 18.340.6 14.4-38.2 26.648.9 59.4-81.7 29.6-51.8 61.845.6 9.4-28.3 6.8-31.1 6.0-32.8 5.5-21.6 19.6-30.7 30.0-37.6 5.7-6.7 7.7-9.5 2.5-7.0 2.05.5 1.0-6.0
2.0-6.5 0.55.5 0.5-6.0
240
M. H. SLAUGHTER AND T. G. LOHMAN
the 40K method, ranging from 18.3% to ponent (r = 0.88). Much less association 40.6%. These results are comparable to is found between Sheldon’s mesomorphy those found by Novak (‘72) for 18- and 25- and Heath and Carter’s second component year old women (33.0 % ) using whole body (r = 0.54). These correlations are similar 40Kcounting. The mean difference between to those by Morton (‘67) who compared the two methods of estimating body fatness Sheldon’s Trunk Index method of somato(40K and body density) in our study may be typing and Heaths Anthroposcopic Method due to the lower potassium content of the on children. lean body mass (2.33 gmfkg) as suggested Sheldon’s Trunk-Index Method produced by the data for Wormesley et al. (’72) higher somatotype designations for the rather than the 2.66 gmlkg used by Novak three components than did the Heath-Car(’72) and the 2.50 gm/kg proposed by ter Method (table 1). Morton (’67) also Forbes et al. (‘68) and used in our study. obtained similar results in comparing ShelThe anthropometric measures show general don’s Trunk-Index method and Heaths agreement with those reported by Katch Anthroposcopic Method of somatotyping. and McArdle (‘73) and Sloan et al. (‘62) The differences between the means of comfor college-aged women. parable somatotyping components as asAll reliability coefficients for body com- sessed by Sheldon’s method and Heathposition measures between the two test Carter’s Anthropometric Method show that days (N = 31) are above 0.90; 0.92 for per- Sheldon’s mesomorphy and ectomorphy are cent fat from both 40K and body density, significantly different (p < 0.05) from 0.94 for LBM from density and 0.93 for Heath and Carter’s parallel components, LBM from 40K. The correlation between whereas the mean difference between percent fat by the two methods (body den- Sheldon’s endomorphy and Heath and Carsity and 40K) was 0.82 and between LBM ter’s first component does not attain sta(kg) by the two methods, 0.86. The reli- tistical significance. ability coefficients for the anthropometric Zero-order correlation coefficients for measures used for the Heath-Carter so- selected body composition measurements matotyping process are as follows: height and somatotype ratings from both Sheldon 0.99; weight, 0.99; triceps skinfold, 0.94; and Heath-Carter are presented in table 2. subscapular skinfold, 0.95; suprailiac skin- Percent fat, estimated from 40K measurefold, 0.94; calf skinfold, 0.92; biceps cir- ments, correlates 0.66 with Sheldon’s encumference, 0.97; calf circumference, domorphy and 0.69 with Heath and Car0.92; elbow width, 0.73; and knee width, ter’s first component. Percent fat, estimated 0.79. from body density, correlates 0.75 with The zero-order correlations between par- Sheldon’s endomorphy and 0.74 with allel components of Sheldon and Heath- Heath and Carter’s first component. AbCarter’s somatotypes indicate considerable solute lean body mass (LBM), estimated association for Sheldon’s endomorphy and from 40K measurements, correlates 0.34 Heath and Carter’s first component (r = with Sheldon’s mesomorphy and 0.15 with 0.79) and ectomorphy and the third com- Heath and Carter’s second component. TABLE 2
Zero-order coefficients for body composition and somatotype ratings from Sheldon, and Heath and Carter ( N = 31 subjects) Heath-Carter
Sheldon
Variable
Age Height Weight % fat (*OK) % fat (density) LBM (4OK) LBM (density) 7c LBM (4OK) % LBM (density) 1
endo
mew
ecto
0.12
-0.23 0.02 0.48
-0.06
- 0.08 0.80
1
1
0.66 1
0.16
0.75 1 0.23 0.41 1
0.2 1 0.34 0.42
-0.66 1
-0.16
-0.75
-0.21
Significant at the 0.05 level
1
1
0.54 1 -0.43 1 -0.45 1 -0.54 1 - 0.03 -0.11 0.45 1 0.54 1
1st comp
0.21 0.03 0.65 1
2nd comp
3rd comp
0.08 -0.45 0.47
1
0.03 0.33
1
-0.69 1
0.691
0.361
-0.54
0.74 1 0.07 0.25
0.45 0.15 0.20
1
-0.66 1
- 0.691
- 0.361
-0.23 - 0.34 - 0.54 1
-0.74
-0.45
1
1
1
0.661
24 1
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
Similar results are found using LBM estimated from body density. The correlation between LBM (40K) and Sheldon’s ectomorphy is -0.03, and Heath and Carter’s third component, - 0.23. Similar results are found using LBM estimated from body density. To better study the association of body composition measurements with Sheldon’s and Heath and Carter’s somatotype components, multiple regression analysis was used to determine the relation of various combinations of age, body weight, height, and body composition as predictors of somatotype (tables 3, 4). The coefficients of determination (R2X 100) from age, height, and weight as independent variables are 83% for Sheldon’s endomorphy but only 58% for Heath and Carter’s first component (table 3). When either percent fat or LBM as measured by either 40K or body density in addition to age, height and weight is used to estimate somatotype, the coefficients increased to 85% (not significantly different from 83% without percent fat) for Sheldon’s endomorphy and to 65% for Heath and Carter’s first component
(significantly different from 58% without percent fat [table 31 ). The coefficients of determination from age, height, and weight as independent variables are 32% for Sheldon’s mesomorphy and 68% for Heath and Carter’s second component (table 3). Adding LBM (body density) as an independent variable increases the variation accounted for only slightly to 37% for Sheldon’s mesomorphy and to 72% for Heath and Carter’s second component. The coefficients of determination from age, height, and weight as independent variables are 79% for Sheldon’s ectomorphy and 90% for Heath-Carter’s third component. Adding LBM (body density) as an independent variable increases the variation accounted for slightly for Sheldon’s ectomorphy (81%) but not for Heath and Carter’s third component (table 3). When LBM (body density) and height are used to estimate somatotype, both independent variables are significantly associated with Sheldon’s mesomorphy and Heath and Carter’s second component (table 4). The coefficient of determination
TABLE 3
Coefficients of determination between body composition and somatotypes ( N = 3 1 ) Sheldon endo
meso
ecto
1st comp
2nd comp
3rd comp
83 85 85 85 85 85 85
32 35 37 35 37 35 37
79 79 79 79
58 66 66 66 64 66 66
69 71 72 71 72 71 72
90 90 90 90 90 90 90
Age, height, weight Plus % fat (4OK) Plus % fat (density)
Plus Plus Plus Plus
Heath-Carter
LBM (4OK) LBM (density) % LBM (40K) % LBM (density)
81
79 79
TABLE 4
Regression coefficients for the association of height and lean body mass and height, weight and lean body mass w i t h Sheldon’s mesomorphy and Heath and Carter’s Second Component ( N = 3 1 ) 1 Sheldon’s Mesomorphy
sb
b
Height
LBM Weight Intercept
RZ X 100 SE 1 2
-0.074 0.1402
0.030 -0.038
9.47 32 0.73
LBM estimated from body density. p < 0.05.
Heath and Carter’s second component b
sb
-0.178 2 0.193 2
0.028 0.035
24.1 61
0.67
Sheldon’s Mesomorphy b
sb
-0.066 2 0.031 0.096 0.059 0.030 0.030 8.37 35 73
Heath and Carter’s second component b
sb
-0.158 2 0.025 0.081 0.047 0.0752 0.024 21.27 72 59
242
M. H. SLAUGHTER AND T. G. LOHMAN
was 32% for mesomorphy and 61 % for the second component. When weight in addition to height and LBM is used to estimate somatotype, coefficients of determination increased to 72% for Heath and Carter’s second component but only to 35% for Sheldon’s mesomorphy (table 4). DISCUSSION
Because Heath-Carter’s somatotype describes body morphology as well as body composition, not all of the variation in somatotype is expected to be accounted for by measures of body composition. Furthermore, both potassium-40 and body density methods have inherent error from both the biological and technical standpoints. From biological sources both methods are based on the assumption that the lean body mass has a constant potassium content and a constant density for all individuals. While this assumption has been found to be relatively accurate, the exact limitations of both methods are still unknown. Technical errors also contribute to measurement error for both methods but were reduced to some extent by using the mean of two determinations, one week apart. Percent fat (estimated from either body density or 4OK) is as closely related with Sheldon’s endomorphy based on somatotyping ponderal index and trunk index as with Heath and Carter’s first component, determined directly from skinfolds. Similar relationships between percent fat and Sheldon’s endomorphy, as well as Heath and Carter’s f i s t component (based on the sum of three skinfolds), have been reported by other investigators (Brozek and Keys, ’51; Durnin and Rahaman, ’67). Also high negative correlations have been found between endomorphy and specific gravity (Dupertuis et al., ’51; Brozek and Keys, ’52). The correlations between LBM (by both methods) and both Sheldon’s mesomorphy and Heath and Carter’s second component are quite low and indicate that LBM as an absolute weight is not significantly related to Sheldon’s mesomorphy or Heath and Carter’s second component. The correlations between percent LBM and mesomorphy as well as the second component are also quite low and in all cases negative. Thus neither the absolute nor the relative amount of LBM is associated with meso-
morphy. Both Dupertuis et al. (‘51) and Brozek and Keys (‘52) also found low positive correlations between mesomorphy and specific gravity (an index of the fat content and lean body mass content of the body). As Carter (’72) has stated “the second component refers to relative musculoskeletal development per unit of height” and Sheldon’s mesomorphy is based on the relation of the abdominal trunk to thoracic trunk. Wilmore (‘71) interpreted Carter and Heath’s definition of their second component incorrectly when he attempted to validate it by use of the absolute weight of lean body mass. The correlation of 0.16 between LBM (kg) and Heath and Carter’s second component (Wilmore, ’70) is similar to the correlation found in this study from both methods of estimating LBM (0.15 and 0.20). Lean body mass estimated by either method is not significantly correlated (p > 0.05) with Sheldon’s ectomorphy or with Heath and Carter’s third component. Dupertuis et al. (‘51) and Brozek and Keys (’52) also found that ectomorphy was not closely related to specific gravity. In studying the coefficients of determination which resulted from the use of age, height, weight and percent fat in multiple regression analysis, Sheldon’s endomorphy is seen to be largely associated with weight and height, whereas Heath and Carter’s first component is more closely associated with weight and body fatness. Heath and Carter’s second component can be predicted closer from age, height and weight than Sheldon’s mesomorphy. In a study by Damon et al. (’62) Sheldon’s mesomorphy (rated by his 1954 method) also could not be predicted as closely using a combination of anthropometric measures in 369 white male soldiers with 44% of the variation accounted for in mesomorphy as compared to 61% and 81% of the variance accounted for in endomorphy and ectomorphy, respectively (Damon et al., ’62). When estimating Sheldon’s ectomorphy and the third component of Heath and Carter, body weight and height are the only two significant independent variables. In studying the coefficients of determination when only the independent variables of height and LBM are used in multiple regression analysis, Heath and Carter’s second component is found to be signifi-
RELATIONSHIP OF BODY COMPOSITION TO SOMATOTYPE
cantly related to height and LBM (body density). However, when body weight is used along with height and LBM as independent variables, the regression coefficient for LBM is no longer statistically significant. Thus, height and weight appear to be major contributors to accounting for variation in the second component with a somewhat smaller contribution from LBM (table 4). The coefficient of determination for Sheldon’s mesomorphy when only the independent variables of height and LBM are used, is quite low (32 % ) and increases only 3% when weight is added. Furthermore, the regression coefficient of LBM is no longer significant. Thus, Sheldon’s mesomorphy does not appear to be associated with height, LBM and weight. In summary, Sheldon’s endomorphy is largely associated with weight and height whereas Heath and Carter’s first component is more closely associated with weight and body fatness. Neither the absolute nor the relative amount of LBM is associated with Sheldon’s mesomorphy or Heath and Carter’s second component. Height and lean body mass together are found to be more closely related to Heath and Carter’s second component than to Sheldon’s mesomorphy. ACKNOWLEDGMENTS
This research was conducted in cooperation with the Physical Fitness Research Laboratory, Department of Physical Education, University of Illinois. We wish to express appreciation to Richard Boileau, Kirk Cureton, Tanya Hildebrandt, Jerry Mayhew, and Benjamin Massey for their contribution to this study. LITERATURE CITED Akers, R., and E. R. Buskirk 1969 An underwater weighing system utilizing forcecube transducers. J. Appl. Physiol., 26; 649-652. Allen, T. H., M. T. Peng, K. P. Chen, T. F. Huang, C. Chang and H. S. Fang 1956 Prediction of total adiposity from skinfolds and the curvilinear relationship between external and internal adiposity. Metabolism, 5 ; 346352. Boileau, R. A,, B. H. Massey and J. E. Misner 1973 Body composition changes in adult men during selected weight training and jogging programs. Res. Quart., 44; 158-168. Brozek, J., F. Grande, J. T. Anderson and A. Keys 1963 Densitometric analysis of body composition: revision of some quantitative assumptions. Ann. N. Y.Acad. Sci., 110; 113-140. Brozek, J., and A. Keys 1951 The evaluation of
243
leanness-fatness in man: norms and interrelationships. Brit. J. Nutr., 5; 194-206. 1952 Body build and body composition. Science, 1 1 6; 140-142. Buskirk, E. R. 1961 Underwater weighing and body density: a review of procedures. In: Techniques for Measuring Body Composition. J. Brozek, ed. Nat. Acad. of Sci., Washington, D. C., pp. 90-107. Carter, J. E. L. 1972 The Heath-Carter Somatotype Method. J. E. Carter, publisher, San Diego. Clarke, H. 1971 Physical and Motor Tests in the Medford Boy’s Growth Study. Prentice-Hall, New York. Conway, E. J. 1957 Nature and significance of concentration relations of potassium and sodium ions in the skeleton muscle. Physiol. Rev., 37; 84-132. Cureton, T. K., Jr. 1947 Physical Fitness, A p praisal and Guidance. Henry Kimpton, London. 1951 Physical Fitnessof Champion Athletes. Univ. of Illinois Press, Urbana, Ill. Damon, A,, H. K. Bleibtreu, 0. Elliot and E. Giles 1962 Predicting somatotype from body measurements. Am. J. Phys. Anthrop., 20; 461-474. Darling, R. C., A. Cournand and D. W. Richards 1940 Studies on intrapulmonary mixture of gases. 111. An open-circuit method for measuring residual air. J. Clin. Invest., 19: 606-618. Dupertuis, C. S., G. C. Pitts, E. F. Osserman, W. C. Welham and A. R. Behnke 1951 Relation of specific gravity to body build in a group of healthy men. J. Appl. Physio., 3; 676-680. Durnin, T. V . G. A,, and M. M. Rahaman 1967 The assessment of the amount of fat in the human body from measurements of skinfold thickness. Brit. J. Nutr., 21 ; 681-689. Forbes, G. B., and J. B. Hursh 1963 Age and sex trends in lean body mass calculated from K40 measurements: with a note on the theoretical basis for the procedure. Ann. N. Y. Acad. Sci., 1 1 0 ; 255-263. Forbes, G., F. Schultz, C. Cafarelli and G. Amirhakimi 1968 Effects of body size potassium-40 on measurement in the whole-body counter. Health Physics, 15: 4 3 5 4 4 2 . Heath, B. H. 1963 Need for modification of somatotype methodology. Am. J. Phys. Anthrop., 21; 227-233. Heath, B. H., and J. E. L. Carter 1966 A comparison of somatotype methods. Am. J. Phys. Anthrop., 24: 87-99. 1967 A modified somatotype method. Am. J. Phys. Anthrop., 27: 57-74. Katch, F. I., and W. McArdle 1973 Prediction of body density from simple anthropometric measurements in collegeage men and women. Human Biology, 45: 445-454. Keys, A., and J. Brozek 1953 Body fat in adult man. Physio. Rev., 33; 2 4 5 3 2 5 . Morton, A. R. 1967 Comparison of Sheldon’s Trunk-Index and Anthroposcopic Methods of Somatotyping in Their Relationships to the Maturity, Structure, and Motor Ability of the Same Boys Nine through Sixteen Years of Age. Microcard doctoral dissertation, University of Oregon. Munroe, R., H. H. Clarke and B. H. Heath 1969 A somatotype method for young boys. Am. J. Phys. Anthrop., 30; 195-202. Novak, L. 1972 Aging, total body potassium,
244
M. H. SLAUGHTER AND T. G. LOHMAN
fat-free mass and cell mass in males and females between ages 18 and 85 years. J. Geron., 27:
438443. Parnell, R. W. 1954 Somatotyping by physical anthropometry. Am. J. Phys. Anthrop., 12: 209-
239. 1958 Behavior and Physique. Edward Arnold, Ltd., publishers, London. Reynolds, E. L., and T. Asakawa 1950 A comparison of certain aspects of body structure and body shape in 200 adults. A m . J. Phys. Anthrop., 8: 343365. Sheldon, W. H., C. W. Dupertuis and E. McDermott 1954 Atlas of Men. Harper Bros., New York. Sheldon, W. H., N. D. C. Lewis and A. M. Tenny 1969 Psychotic Patterns and Physical Constitution: Schizophrenia, Current Concepts and Research. D. V. Siva Sankar, ed. PJD Publications, Hicksville, N. Y. Sheldon, W. H., J. J. Stevens and W. B. Tucker 1940 The Varieties of Human Physique. Harper Bros., New York. Sinclair, G. D. 1966 Stability of Physique Types
of Boys Nine through Twelve Years of Age. Microcard master’s thesis, University of Oregon. 1969 Stability of Somatotype Components of Boys Ages Twelve through Seventeen Years and Their Relationships to Selected Physical and Motor Factors. Microcard doctoral dissertation, University of Oregon. Sloan, A. W., J. J. Burt and C. S. Blyth 1962 Estimation of body fat in young women. J. Appl. Physio., 1 7 : 967-970. Twardock, A. R., T. G. Lohman, G. S. Smith and B. C. Breidenstein 1966 The Illinois animal science counter: performance characteristics and animal radioactivity measurement procedures. J. Animal Sci., 25: 1209-1217. Wilmore, J. 1970 Validation o f the first and second components of the Heath-Carter modified somatotype method. Am. J. Phys. Anthrop., 32:
369372. Womersley, J., K. Boddy, P. King and J. V. G. A. Durnin 1972 A comparison of the fat-free mass of young adults estimated by anthropometry, body density, and total body potassium content. Clin. Sci., 43: 468475.
