"Education, Faculty of"@en . "Curriculum and Pedagogy (EDCP), Department of"@en . "DSpace"@en . "UBCV"@en . "Cameron, Katherine Susan Gacuk"@en . "2010-03-29T16:36:22Z"@en . "1981"@en . "Master of Physical Education - MPE"@en . "University of British Columbia"@en . "Sixty-nine Canadian female gymnasts ages 11.5 to 18.0 years, from varying ability levels (National Elite = Group 1, Pre-National Elite = Group 2, Competitive = Group 3, Recreational = Group 4), were studied to determine relationships between performance and maturity, and between performance and anthropometric characteristics. It was hypothesized\r\nthat there would be significant maturational and anthropometric differences among the ability groups.\r\n\r\n\r\nSkeletal age in reference to chronological age differences, among the ability groups, were assessed using analysis of variance, while differences in the incidence of menarche were assessed using chi-square analysis. Anthropometric differences were assessed using multivariate and univariate analysis of covariance, with chronological age as the covariate.\r\n\r\n\r\nAt a level of significance of p < .01, and using preplanned orthogonal contrasts of Group 1+2+3 vs Group 4, Group 1+2 vs Group 3, and Group 1 vs Group 2, both of the maturational and all five of the anthropometric hypotheses were partially supported, with the following significant differences noted:\r\n\r\n\r\nHighly skilled gymnasts, in comparison to lesser skilled gymnasts (Group 1+2+3 vs Group 4, and Group 1+2 vs Group 3), were maturationally delayed both skeletally and menarcheally. Anthropometrically, they were shorter in trunk length; smaller in triceps, suprailiac, abdominal, front thigh, and medial calf skinfolds; smaller in proportional fat mass; and larger in proportional muscle mass. In addition, highly skilled gymnasts (Group 1 + 2 + 3), in comparison to recreational gymnasts (Group 4), were smaller in bi-epicondylar femur width, thigh girth, and subscapular skinfold. As well, elite gymnasts (Group 1 + 2), in comparison\r\nto lesser skilled competitive gymnasts (Group 3), were smaller in sitting height and larger in proportional skeletal mass.\r\n\r\nNational elite gymnasts (Group 1), in comparison to pre-national elite gymnasts (Group 2), were not maturationally different, skeletally or menarcheally. Anthropometrically, they were shorter in trunk length, longer in thigh length, and smaller in anterior-posterior chest depth.\r\n\r\nThe significant maturational differences noted among the ability groups were considered to be related to gymnastic performance, with higher skilled gymnasts being developmentally less mature than lesser skilled gymnasts.\r\n\r\nThe significant anthropometric differences noted among the ability groups were considered to be related to gymnastic performance. More specifically, these differences were considered to be of biomechanical importance, and reflections of differences in activity level. As well, it was suggested that these anthropometric differences were associated with maturational differences.\r\nThe results of the maturational and anthropometric assessments indicated that there may be a relationship between gymnastic performance and maturity, and between gymnastic performance and anthropometric characteristics. It was proposed that further analysis of the anthropometric\r\nparameters, with respect to proportional assessment, would be necessary before anthropometric characteristics would closely reflect maturational differences."@en . "https://circle.library.ubc.ca/rest/handle/2429/22834?expand=metadata"@en . "ANTHROPOMETRIC AND MATURATIONAL ASSESSMENT OF FEMALE GYMNASTS FROM VARYING PERFORMANCE LEVELS by KATHERINE SUSAN GACUK CAMERON Honours Degree B .P .H .E . , Lakehead University, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHYSICAL EDUCATION in THE FACULTY OF GRADUATE STUDIES School of Physical Education and Recreation We accept this thesis as conforming to the required standards THE UNIVERSITY OF BRITISH COLUMBIA March, 1981 \u00C2\u00A9 Katherine S. Cameron, 1981 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department of S.AC_Jl , f o<3u*oO The U n i v e r s i t y of B r i t i s h Columbia 2075 Wesbrook P l a c e Vancouver, Canada V6T 1W5 Date i ABSTRACT Sixty-nine Canadian female gymnasts ages 11.5 to 18.0 years, from varying ab i l i ty levels (National El i te = Group 1, Pre-National El i te = Group 2, Competitive = Group 3, Recreational = Group 4), were studied to determine relationships between performance and maturity, and between performance and anthropometric character ist ics. It was hypoth-esized that there would be signif icant maturational and anthropometric differences among the ab i l i ty groups. Skeletal age in reference to chronological age differences, among the ab i l i ty groups, were assessed using analysis of variance, while differences in the incidence of menarche were assessed using chi-square analysis. Anthropometric differences were assessed using multivariate and univariate analysis of covariance, with chronological age as the covariate. At a level of significance of p < .01, and using preplanned orthogonal contrasts of Group 1 + 2 + 3 vs Group 4, Group 1 + 2 vs Group 3, and Group 1 vs Group 2, both of the maturational and a l l f ive of the anthropometric hypotheses were part ia l ly supported, with the following signif icant differences noted: Highly sk i l led gymnasts, in comparison to lesser sk i l led gymnasts (Group 1 + 2 + 3 vs Group 4, and Group 1 + 2 vs Group 3), were matur-ationally delayed both skeletal ly and menarcheally. Anthropometrically, they were shorter in trunk length; smaller in t r iceps, suprai l iac , abdominal, front thigh, and medial cal f skinfolds; smaller in proportional fat mass; and larger in proportional muscle mass. In addition, highly ski l led gymnasts (Group 1 + 2 + 3), in comparison to recreational gymnasts (Group 4), were smaller in bi-epicondylar femur width, thigh g i r th , and subscapular skinfold. As wel l , e l i te gymnasts (Group 1 + 2), in compar-ison to lesser ski l led competitive gymnasts (Group 3), were smaller in s i t t ing height and larger in proportional skeletal mass. National e l i te gymnasts (Group 1), in comparison to pre-national e l i te gymnasts (Group 2), were not maturational ly di f ferent , skeletal ly or menarcheally. Anthropometrically, they were shorter in trunk length, longer in thigh length, and smaller in anterior-posterior chest depth. The signif icant maturational differences noted among the ab i l i ty groups were considered to be related to gymnastic performance, with higher ski l led gymnasts being developmentally less mature than lesser ski l led gymnasts. The signif icant anthropometric differences noted among the ab i l i ty groups were considered to be related to gymnastic performance. More spec i f i ca l l y , these differences were considered to be of biomechanical importance, and reflections of differences in act iv i ty leve l . As wel l , i t was suggested that these anthropometric differences were associated with maturational differences. The results of the maturational and anthropometric assessments indicated that there may be a relationship between gymnastic performance and maturity, and between gymnastic performance and anthropometric character ist ics. It was proposed that further analysis of the anthro-pometric parameters, with respect to proportional assessment, would be necessary before anthropometric characterist ics would closely ref lect maturational differences. i i i TABLE OF CONTENTS Page LIST OF TABLES v i i Chapter 1 INTRODUCTION 1 STATEMENT OF THE PROBLEM 2 HYPOTHESES 3 DEFINITION OF TERMS 4 DELIMITATIONS 6 LIMITATIONS 6 2 REVIEW OF THE RELATED LITERATURE 8 INTRODUCTION 8 I , MORPHOLOGICAL DESCRIPTORS 9 Somatotypical Analysis 9 Anthropometric Assessment 9 Body Composition Analysis 10 Summary: Morphological Descriptors 20 II MORPHOLOGY AND PERFORMANCE . . . . 21 Gymnastics and Other Sport Disciplines 21 Descriptive Studies of Gymnasts 24 Biomechanical Advantages of Specif ic Physiques \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 27 Summary: Morphology and Performance 31 III PERFORMANCE AND MORPHOLOGY 32 Fat Mass Content and Act iv i ty Levels 33 iv Page Cr i t ica l Periods for Fat Mass Deposition \u00E2\u0080\u00A2\u00E2\u0080\u00A2\u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 35 Muscle Mass Content and Act ivi ty Levels \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 36 Cr i t ica l Periods for Muscle Mass Deposition \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 37 Skeletal Mass Content and Act ivi ty Levels \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 38 Cr i t ica l Periods for Skeletal Mass Deposition \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 39 Summary: Performance and Morphology 42 IV PERFORMANCE AND MATURITY 43 Maturity Indicators 43 Maturity and Population Studies 47 Maturity and Athletic Populations 48 External Stressors and Maturity 50 Exercise as a Maturity Stressor 51 Growth Rate Intervention and Catch-up Growth \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 52 Summary: Performance and Maturity 53 V MORPHOLOGY AND MATURITY 54 Height, Weight, and Maturity 54 Secular Trends 55 Endomorphy, Mesomorphy, and Maturity 56 The Cr i t ica l Mass Hypotheses 58 Summary: Morphology and Maturity 60 VI MATURITY AND MORPHOLOGY 61 Height, Weight, and Maturity 62 Proportionality, Shape, and Maturity 62 Endomorphy, Mesomorphy, and Maturity 64 Summary: Maturity and Morphology 65 VII MATURITY AND PERFORMANCE 66 Maturity and Muscular Development 66 V Chapter Page Early Maturity and Performance 68 Late Maturity and Performance 68 Maturity and Cr i t ica l Learning Periods . . . . 70 Age and Olympic Participation 71 Summary: Maturity and Performance 72 3 METHODS AND PROCEDURES 73 SUBJECTS 73 PROCEDURES 74 MEASUREMENTS 76 Maturational Measurements 76 I Skeletal age 76 II Menarche 76 Anthropometric Measurements 76 I Height and Length Measurements 78 II Breadth, Width, and Depth Measurements . . . 85 III Girth Measurements 90 IV Skinfold Thickness Measurements 96 V height and Proportional Mass Measurements \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 99 EVALUATION OF MEASUREMENTS . 104 Maturational Evaluation 105 Anthropometric Evaluation \u00E2\u0080\u00A2 105 STATISTICAL ANALYSIS 106 Test Groupings 106 Stat ist ica l Analysis of Maturational Data 106 I Skeletal age data 106 II Menarcheal data 107 Stat ist ica l Analysis of Anthropometric Data \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 107 vi Chapter Page 4 RESULTS AND DISCUSSION 108 RESULTS 110 MATURATIONAL ASSESSMENT 110 I Skeletal Age 110 II Menarche 113 Summary of Results: Maturational Assessment . . . . 115 ANTHROPOMETRIC ASSESSMENT 116 I Height and Length Measures 116 II Breadth, Width, and Depth Measures 121 III Girth Measures 124 IV Skinfold Thickness Measures 129 V Weight and Proportional Mass Measures 132 Summary of Results: Anthropometric Assessment . . . . 136 DISCUSSION 137 MATURATIONAL ASSESSMENT . . . . . . . . 137 ANTHROPOMETRIC ASSESSMENT 143 I Height and Length Measures \u00E2\u0080\u00A2 147 II Breadth, Width, and Depth Measures 157 III Girth Measures 165 IV Skinfold Thickness Measures 171 V Weight and Proportional Mass Measures 174 MATURITY-ANTHROPOMETRIC RELATIONSHIP 183 5 SUMMARY AND CONCLUSIONS 185 SUMMARY 185 CONCLUSIONS 189 SUGGESTIONS FOR FURTHER RESEARCH 190 REFERENCES 193 APPENDIX A 208 v i i LIST OF TABLES Table Page 1. Observed Cell Means and Standard Deviations For Chronological Age, Skeletal Age, and the Chronological Age Minus Skeletal Age Difference . . . . m 2. Univariate Analysis of Variance Results For the Chronological Age Minus Skeletal Age Difference For Each Preplanned Orthogonal Contrast 112 3. Observed Cell Frequencies For the Incidence of Menarche \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 114 4. Chi-Square Analysis Results For the Incidence of Menarche For Each Preplanned Orthogonal Contrast 114 5. Observed Cell Means and Standard Deviations For Height and Length Measures \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 117 6. Multivariate and Univariate Analysis of Covariance Results For Height and Length Measures For Each Preplanned Orthogonal Contrast 119 7. Observed Cell Means and Standard Deviations For Breadth, Width, and Depth Measures 122 8. Multivariate and Univariate Analysis of Covariance Results For Breadth, Width, and Depth Measures For Each Preplanned Orthogonal Contrast 123 9. Observed Cell Means and Standard Deviations For Girth Measures 125 10. Multivariate and Univariate Analysis of Covariance Results For Girth Measures For Each Preplanned Orthogonal Contrast 127 11. Observed Cell Means and Standard Deviations For Skinfold Thickness Measures 130 12. Multivariate and Univariate Analysis of Covariance Results For Skinfold Thickness Measures For Each Preplanned Orthogonal Contrast 131 13. Observed Cell Means and Standard Deviations For Weight and Proportional Mass Measures 133 14. Multivariate and Univariate Analysis of Covariance Results For Weight and Proportional Mass Measures For Each Preplanned Orthogonal Contrast 134 v i i i ACKNOWLEDGEMENTS To the members of my committee, Dr. S. R. Brown, Dr. H. S. Gi l lesp ie , Dr. A. T. Lascari , Dr. R. E. Mosher, and Dr. R. W. Schutz, I would l ike to express my appreciation for their assistance throughout the preparation of this thesis. To the chairman of my thesis committee, Dr. R. E. Mosher, I would l ike to express a special appreciation for his support and guidance throughout my academic career. I would l ike to acknowledge the assistance of M. Gessaroli and T. Wood for computer programming, L. Grayson and M. Harrison for taking the radiographic photos, and J . Thayer for administering the questionnaire. Special thanks are due Dr. W. M. Ross for his continued interest in this project, and due A. Vajda-Janyk for her time and care in taking the anthropometric measurements and in assigning skeletal ratings. A sincere appreciation is extended to the participating gymnasts, their parents and coaches, and part icularly to the Canadian Gymnastics Federation and the National Women's Coach, Boris Bajin. Without their co-operation, consent, and concern, this research would not have been possible. F ina l ly , a warm thank-you is reserved for my husband Brian, for his encouragements, and for allowing me the luxury of such an endeavour. CHAPTER 1 INTRODUCTION The factor of age presents unlimited problems in research involv-ing anthropometric and performance comparisons between former and present Olympic Games and World Class competitors. The ages of the female gymnastic participants in these competitions appear to have decreased substantial ly' with each successive Games. A wide age range of these competitors has made i t d i f f i c u l t to draw any conclusive information from anthropometric compar-isons of former and present e l i t e , female gymnasts, other than the identif -ication of obvious physical differences that normally exist between the adult female f igure, and the younger pre-pubescent or adolescent female f igure. Many of today's e l i te female gymnasts are in the age range when normal pubertal developments are expected to take place. During the pubescent years, dramatic physical changes'in s ize , shape, and body compo-sit ion occur, along with dif ferential changes in the reproductive and other organs. One such development is the onset of menarche, which marks a definite stage of physical maturity in the female progression through adolescence towards ultimate adult, physical status. The recent trend of younger participants at e l i te gymnastic competitions seems to indicate that female gymnasts are reaching their prime or \"peaking\" at younger ages. This appears to be the case since, accompanying this decrease in age, is a progressive increase in sk i l l : complexity and the attainment of very high degrees of performance. 1 2 This age factor may play an important role in the evolution of gymnastic performance. A logical inquiry from these observations i s , \"whether or not certain physical advantages co-exist with the younger, female gymnast, rendering her a more appropriate candidate for gymnastic-type movement than the older, female gymnast\". This inquiry may have some support in the fact that fame for female gymnasts is often short - l ived, with female gymnasts tending to drop out of high caliber competition during, or soon after adolescence. This early retirement may be due to the s t i f f competition imposed by upcoming gymnasts, who \"win\" their way into the e l i te posit ions, or due to the adolescent development of other interests. An overlooked, and possibly unconsidered, factor involved in this early retirement trend may l i e in the fact that the body changes in s ize , shape, proportionality, and composition during puberty. Some of these obvious changes are; increases in height, weight, total body adiposity, and the development of secondary sex characteristics in general. A declining age of part ic ipat ion, concurring with an increasing level of s k i l l complexity, may ref lect the changing nature and evolution of \"women's\" gymnastics. This trend suggests the possib i l i ty that important relationships exist between maturational status, anthropometric character-i s t i c s , and success in gymnastic performance. STATEMENT OF THE PROBLEM It is the purpose of this research to investigate the possib i l i ty that success in gymnastics is related to the concept of maturational age and anthropometric character ist ics. More spec i f i ca l l y , the questions to be investigated are as follows: 3 1. Are there maturational di f ferences among national e l i t e , pre-nat ional e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts? 2. Are there anthropometric di f ferences among national e l i t e , pre-national e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts? HYPOTHESES I t is hypothesized that: 1. The maturational s ta tus , as determined by skeleta l age in reference to chronological age, i s s i g n i f i c a n t l y d i f fe rent among national e l i t e , pre-nat ional e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts. 2. The incidence of menarche i s s i g n i f i c a n t l y d i f fe rent among national e l i t e , pre-national e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts. 3. The measures of height and length are s i g n i f i c a n t l y d i f fe rent among national e l i t e , pre-national e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts. 4 . The measures of width, breadth, and depth are s i g n i f i c a n t l y d i f f -erent among national e l i t e , pre-national e l i t e , lesser s k i l l e d competitive and recreat ional gymnasts. 4 5. The measures of girth are s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. 6. The measures of skinfold thickness are s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts. 7. The measures of weight and proportional body mass are s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser sk i l led compet-i t ive and recreational gymnasts. DEFINITION OF TERMS Anthropometry: as defined by Webster (1970: s .v . ) i s , \"the study of human body measurements, especially on a comparative basis\". Anthropometric Characteristics: refers, in the present study, to the specif ic height, length, width, breadth, depth, g i r th , skinfold thickness, and weight measurements taken, as well as to the length and proportional \u00E2\u0080\u00A2mass measures.derived. Growth: refers ,to the physical act of enlargement of a structure, which may result in volume increases and shape changes. Maturation: refers to the various stages of dif ferentiat ion of c e l l s , t issues, and organs, in preparation for functional and structural changes. It is also the occurrence of physical and physiological changes 5 associated with progression towards ultimate and fu l l adult development and function. The term \"development\" is used synonymously with \"maturation\". Menarche: refers to, \"the in i t ia t ion of menstruation\" (Webster, 1970: s . v . ) . It is an objective and easily identi f iable stage in the maturation sequence of the female. Morphology: is a col lect ive term, referring to the shape, bu i ld , physique, structure, composition, and proportions of the1 body. Proportional Body Masses: refers to, the derived fa t , skeleta l , muscle, and residual weights of the body, expressed in percentages of the total body weight, with total body weight for these derivations calculated as the sum of the four derived weights, and not the conventionally observed scale weight. Pubertal or Adolescent Developments: refers, in the present study, to the init-iation of the growth spurts in height and weight; to speci f ic compositional, s tructural , and shape changes; as well as to physiological and functional developments, such as menarche, that accompany puberty. Skeletal Age: i s , \"a measure of the developmental status of the skeleton as disclosed by an x-ray fi lm\" (Greulich & Pyle, 1970:2), usually of the hand and wrist area. The maturational status of the hand and wrist area \"represents\" the developmental status of the skeleton as a whole (Greulich & Pyle, 1970:11), and \"reflects\" the developmental and functional status of the reproductive system (Greulich & Pyle, 1970:15). The terms 6 \"bone age\" and \"physiological age\" are used synonymously with the term \"skeletal age\". DELIMITATIONS 1. The present study is delimited to maturational and anthropometric assessments of Canadian female gymnasts, between the ages of 11.5 and 18.0 years, with ab i l i ty levels ranging from recreational, through provincial competitive levels A & B, to pre-national and national team cal iber . 2. The concept of maturity is delimited to: a) skeletal age, as evaluated via the Tanner-Whitehouse II Method, twenty bone assessment, bone specif ic approach to assessing the skeletal maturity of the hand and wrist area (Tanner, Whitehouse, Marshall, Healy, & Goldstein, 1975); and b) the occurrence or absence of menarche. 3. The anthropometric assessment is delimited to the 39 measurements evaluated, and more speci f ica l ly to the 11 height and length; 6 breadth, width, and depth; 11 g i r th; 6 skinfold thickness; 1 weight, and 4 proportional body mass measures. LIMITATIONS 1. Sk i l l level c lass i f icat ion into Research Groups 1, 2, and 3, is limited to the Canadian Gymnastics Federation categorization of gymnasts into \"National E l i t e , Pre-National E l i t e , and Provincial Competitive Levels A and B\", respectively. Categorization of gymnasts is based on their performance in a test of \"gymnastic fitness\" comprised of a battery of individual physical fitness and standard gymnastic moves tests (Bajin, 1977). Since the tra ining, competitive, and rest season may not coincide for gymnastic Groups 1, 2, and 3, i t is recognized that the state of physical f itness \"readiness\" may not have been uniform among these groups. This situation may have influenced the difference values, among these groups, in those anthropometric variables readily influenced by act iv i ty level (weight, skinfold thickness, proportional fat mass, g i r th , proportional muscle mass measures). The subjects' involvements in other sporting endeavours and physical pursuits were not control led. 8 CHAPTER 2.\" REVIEW OF THE RELATED LITERATURE Introduction Research involving pre-adolescent and adolescent children involved in high levels of competition must be concerned with many variables in order to isolate those factors that may contribute s igni f icant ly to the success of these young athletes. This thesis is directed toward isolating the contribution of \"maturity\" and \"morphology\", as these two domains relate to success in gymnastics. The related l i terature for this study wil l be presented under the following headings: I MORPHOLOGICAL DESCRIPTORS: conventional methods of describing the body morphologically. II MORPHOLOGY AND PERFORMANCE: the importance of body \"build\" to performance outcomes. III PERFORMANCE AND MORPHOLOGY: the role of exercise in developing and altering body \"build\". IV PERFORMANCE AND MATURITY: physical exercise as i t affects the maturing processes. V MORPHOLOGY AND MATURITY: physical size and \"build\" as these affect the maturing processes. VI MATURITY AND MORPHOLOGY: the time and rate of maturity as they affect the ultimate adult physique. 9 VII MATURITY AND PERFORMANCE: physical changes associated with maturity as they affect performance. I MORPHOLOGICAL DESCRIPTORS Three of the conventional methods of describing the body morphologically are; somatotypical analysis, anthropometric assessment, and body composition analysis. Somatotypical Analysis The concept of somatotyping, an evaluation of the total body's external shape, the degree of endomorphy, mesomorphy, and ectomorphy assessed through visual inspection of a series of photos, was introduced by Sheldon, Stevens, and Tucker in 1940. In order to increase the object-iv i ty of the assessment, the Heath-Carter Method of Somatotyping, incorporating anthropometric elements into the photoscopic assessment, was later developed (Carter, 1975), based on ear l ier works of Cureton (1947, 1951), Parnell (1954, 1958), and Damon et a l . (1962) (all cited in Carter, 1975:2-1). Anthropometric Assessment Anthropometric analysis involves taking surface measurements of lengths, g i r ths , widths, breadths, depths, and skinfold thicknesses. These surface measurements offer a means of assessing the distr ibution of the body's composition; with girth measurements generally indicating muscular development; length, breadth, and width measurements ref lecting-skeletal development; depth, breadth, and width measurements generally encompassing 10 the residual mass; and skinfold thickness measurements representing the subcutaneous fat d istr ibut ion. While the absolute values of anthropometric measurements are used for inter-individual and group comparisons (Alexander, 1976; De Garay, Levine, & Carter, 1974; Eiben, 1972; Hebbelinck & Ross, in Nelson & Morehouse, 1974:546; Lewis, 1969; Nelson, 1974; Novak, Woodward, Best i t , & Mellerowicz, 1977; Ross & Day, 1972; Smit, 1973), conversion into proport-ional values, the ratio of one measure to another or the adjustment of a l l measurements to a single measure, such as height, has also been employed. Proportional analysis has been used effect ively in comparisons between individuals (Ross & Wilson, 1974), between groups (Eiben, Ross, Christensen, & Faulkner, 1976), in assessing biomechanical advantages of speci f ic physiques (Hebbelinck & Ross, in Nelson & Morehouse, 1974:546; Le Veau, Ward, & Nelson, 1974; Tanner, 1964, and Ti t te l & Wutscherk, 1972, cited in Hebbelinck & Ross, in Nelson & Morehouse, 1974:546), and also in the monitoring of physical changes in shape and proportionality due to growth, maturity (Behnke, 1963; Behnke & Wilmore, 1974; Bie l ick i & Waliszko, 1975; Eiben, 1978; Huxley, 1932, cited in Behnke, 1961:953; Medawar, 1945; Ross & Wilson, 1974; Tanner & Whitehouse, 1976; Tosovsky, Prokopec, & Mejsharova, 1976), and altered levels of energy expenditure and nutritional status (Behnke & Wilmore, 1974). Body Composition Analysis A computer search program described by Deutch and Ross (1978, cited in Drinkwater & Ross, 1980:178) provides a basic bibliography of over 800 papers in the area of body composition. Comprehensive reviews of the various techniques developed to assess body composition are presented by Behnke and -11 Wilmore (1974), Brozek (1961), Friis-Hansen (1971), Ismail (1971), and Malina (1969b). Jackson and Pollock (1976) provide the computational framework for determining body composition from skinfold, g i r th , and diameter measurements, through the use of multivariate scal ing. Kowalski (1972) examines the limitations and strengths of such models in anthropometric research. Body composition analyses, especially of children and adolescence, are limited and fraught with problems of val id i ty and r e l i a b i l i t y , with each technique and method having i ts own dist inct and inherent errors. Since direct compositional analysis involves the dissection of cadavers, the majority of the research to date has been conducted through indirect means. While chemical analysis of cadavers provides the framework for developing indirect methods for estimating in vivo body composition (Brozek, 1961; Damon & Goldman, 1964; Malina, 1969b), the majority of indirect methods developed are validated against more widely recognized indirect methods (Behnke & Wilmore, 1974), which themselves \"have not been, validated against direct analysis. As a resul t , these indirect methods also involve major assumptions and errors, with regard to the amount and nature of the body components (Damon & Goldman, 1964). (i) \"Fat .mass. .The body has conventionally been divided into a two component system comprised of a fat mass and a fat-free or lean body mass. The progenitor of most of the current procedures for fractionating the body weight into two compartments was Behnke (1942, cited in Drinkwater & Ross, 1980:178). Reviews of the techniques developed to assess body fat are presented by Keys and Brozek (1953), and Reynolds (1950). The fat mass component has been identif ied through the use of anthropometric formulae (Behnke & Wilmore, 1974; Damon & Goldman, 1964; 12 Yuhasz, 1977), and equations involving skinfold measurements in the prediction of total body fat are essential ly elaborations of the pioneer works of Matiegka (1921, cited in Drinkwater & Ross, 1980:178). Skinfold thickness measurements have been shown to be useful predictors of body fatness in children and adolescents (Durnin & Rahaman, 1967; Hammond, 1955), in adolescents (Durnin & Womersley, 1974; Forbes et a l . , 1975, Michael & Katch, 1968, and Young et a l . , 1968, cited in Slaughter, Lohman, & Boileau, 1978:470), in men and women (Damon & Goldman, 1964; Hammond, 1955; Keys & Brozek, 1953; Matiegka, 1921, cited in Keys & Brozek, 1953:264); and of body density in children (Parizkova, 1961a; Shephard, Jones, I sh i i , Kanek, & Albecht, 1969) and adolescents (Parizkova, 1961a). Tables to calculate body fat from skinfold measurements and body density have also been proposed (Durnin & Rahaman, 1967; Parizkova, 1961a). Skinfold th.rckness and girth measurements were found to be useful predictors of body density in college women (Katch & Michael ,--1968), and Best et a l . (1953, cited in Brozek, 1961:923) found a close association between percent body fat and the ratio of height to abdominal girth in adult men. Previous to Parizkova's investigations (1961a), there appears to have been no systematic investigation of the relationship of body fat to subcutaneous fat during childhood and adolescence, and in a later review of the l i terature (Durnin & Rahaman, 1967), no equations were available for use with children or adolescents to derive body fat from measurements of body density. Although the site at which skinfold thickness had the highest correlation with body density varied with age and sex, the high correlations found suggested that the relationship between density and percent fat in 13 children and adolescents, may be close to that found in adults (Parizkova, 1961a). Ongoing cadaver analysis at the Vri je Universiteit Brussels, in conjunction with Simon Fraser University (Canada), has revealed some interesting observations concerning body compositional analysis: The basic assumptions that there are fixed relationships between external and internal adiposity, and between adiposity and l i p i d , are not supported by cadaver analysis. (Ross, 1980) Previous accounts in the l i terature have also indicated the inadequacy of predicting total body fat from subcutaneous fat measurements. A l len , Peng, Chen, Huang, Chang, and Fang (1956) have demonstrated a curvil inear relat ion-ship between external and internal adiposity, with fat persons having approximately two-thirds of their excess adiposity subcutaneously located, and thin persons having most of i t internally located. While Durnin and Womersley (1974) found males to have a higher proportion of body fat situated subcutaneously, than females, cadaver analysis (Alexander, 1964, cited in Durnin & Womersley, 1974:87) revealed that the \"subcutaneous fat accounted for only 0..2;of the total fat in the men and 0.1 in the women\" (referring to the ratio of subcutaneous fat to total body fa t ) . Young et a l . (1964, cited in Shephard et a l . , 1969:1185) found that the triceps skinfold did not correlate well with obesity in females of a given age group. As wel l , this skinfold has been found to show l i t t l e increase with age (Parizkova, 1963; Shephard et a l . , 1969), and i ts use in predicting body density and relative obesity is cautioned (Shephard et a l . , 1969). Wilmore, Royce, Girandola, Katch, and Katch (1970b); Wilmore, Girandola, and Moody (1970a); and Zwiren, Skinner, and Buskirk (1973) found 14 individual skinfold thickness measurements basical ly unsound in assessing changes in body fat with exercise, as signif icant reductions in skinfold thicknesses were not reflected by changes in body density, body fa t , or lean body weight. While Damon and Goldman (1964) found the accuracy of predicting body fat from anthropometric equations to vary with the degree of endomorphy and mesomorphy, Young et a l . (1963, cited in Malina, 1969b:20) and Sker l j , Brozek, and Hunt (1953) proposed that the relationship between subcutaneous skinfold thickness and total body fat may be a function of age and sex, as well as amount of fat . Both these studies indicated an increase in inner fat at the expense of outer fat during the later phase of maturity in females. Skinfold thickness has been found to correlate negatively with body density (Malina, 1969b), and the relationship has been found to be non-linear in both sexes (Durnin & Womersley, 1974). However, a higher correlation found in boys than g i r ls (Parizkova, 1961a) suggests the possib i l i ty of separate equations' and site selections for the sexes. From cadaver analyses, thickness of skin was found to vary with age, sex, and region of the body, with correlations between caliper and actual fat varying from .61 to .92 for females (Lee & Ng, 1965). For the same actual fat thickness, caliper readings were lower for females than males (Lee & Ng, 1965), with skinfold compressibility found to decrease with age (Brozek & Kinsey, 1960, cited in Durnin & Womersley, 1974:91). There is also some indication that the compressibility of thick skinfolds is greater than thin ones (Clegg & Kent, 1967). Himes, Roche, and Siervogel (1979) found signif icant differences among individuals in the compressibility of skinfolds, and reported that the difference between actual fat thickness 15 and skinfold readings is systematic due to compression, with skinfold readings always underestimating actual fat thickness, but inconsistently so, depending on the site and perhaps sex. These studies serve to indicate that there is a degree of error introduced into results when constant pressure calipers are used in the estimation of total body fa t , in assessments of the distr ibution of fa t , and in the appraisal of variation in skinfold thickness and total body fat between individuals. Until the exact relationships among subcutaneous, internal , total body fa t , and adiposity are ident i f ied , the skinfold caliper technique wil l continue to be accepted as a valid and rel iable method for estimating total body fatness, based on the firmly established observations that: 1. A considerable amount of body fat l ies within the subcutaneous tissue (Malina, 1969b). 2. Results are highly reproducible and therefore re l iab le , although not necessarily va l id . Inter-rater differences in cal iper readings have led to reported maximum errors of 4% (Burkinshaw, Jones, & Krwpowicz, 1973) and 6% (Womersley & Durnin, 1973) in the estimation of total body fa t , with no signif icant difference in variance, between observers, when the sites were marked (Burkinshaw et a l . , 1973). 3. Skinfold cal iper measurements have a high correlation with measurements obtained through other indirect techniques, such as ultrasonic depth (Bullen, Quaade, Olesen, & Lund, 1965) and roentgenogrammetric measurements (Garn & Gorman, 1956). ( i i ) Lean body mass. Because of i ts labi le nature and i ts association with nutr i t ion, exercise, disease, and mortality, the fat mass component of the body has received far more attention in the l i terature 16 than the lean body mass component. The major obstacle in the current estimates of body composition is the lack of a ver i f ied in vivo method for quantifying skeletal weight and muscle mass (Baker, 1961, Garn, 1963, and Moore et a l . , 1963, cited in Malina, 1969b:21,25). While the majority of the l i terature refers to a single def init ion of lean body mass for both sexes, Behnke and Wilmore (1974) term the anthro-pometrically calculated lean body weight in women \"minimal weight\". This measure is associated with the leanest individual for a given stature, and incorporates a certain amount of \"essential fat\" in mammary and other t issue. In the male, minimal weight is tantamount to lean body weight. Malina (1969b) refers to lean body mass as an in vivo concept, , and fat-free body mass as an in vitro concept, with the difference between the two masses being in the amount of essential l i p i d s . Moore et a l . (1963, cited in Malina, 1969b:10) introduced the concept of body ce l l mass, \"the working, energy metabolizing portion of the human body in relation to i ts supporting structure\". Morales, Rathbun, Smith, and Pace (1945) regard the mammalian body as consisting of f ive tissue components; fa t , muscle, skin, v iscera l , and nervous t issue. From a biochemist's point of view, the body is reducible to fa t , osseous and non-osseous protein, mineral, and intra-cel lu lar and extra-cel lular water (Malina, 1969b). In many of the early studies, and in most studies where the lean body mass is considered subservient and secondary to the fat mass, the lean body mass is derived by subtraction of the fat weight from the total body weight (Behnke & Wilmore, 1974). Such a procedure leaves the prediction and estimation of this mass, subject to, affected by, and dependent on, the fat mass value and the accuracy with which this mass was obtained. Ideally the two mass components should be assessed independent of one another, 17 although, changes in the composition of the non-fat component of the body have been reported to occur with changing fatness (Keys & Brozek, 1953). Equations involving anthropometric surface measurements in the estimation of the lean body mass have been proposed and developed (Behnke, 1961; Behnke & Royce, 1966; Behnke, 1963, Hampton et a l . , 1966, and Roessler & Dunavant, 1967, cited in Forbes, 1972:336; Forsyth & Sinning, 1974, cited in Sinning, 1974:140; Matiegka, 1921, cited in Keys & Brozek, 1953:264; Wilmore & Behnke, 1970), and are often validated against more widely accepted, but not necessarily val id indirect methods. The most popular indirect method used in these val idat ions, is the densitometric determinant of the lean body mass (Durnin & Rahaman, 1967; Cowgill , 1957, and Von Dobeln, 1959, cited in Bakker & Struikenkamp, 1977:194). However, the s tab i l i ty of the assumptions upon which densitometry is based; notably^that the lean body mass has a constant density and a constant proportion of water, bone is a constant pro-portion of the lean body mass, and cel l water is a constant proportion of cel l mass (Wilmore et a l . , 1970a) has been questioned (Bakker & Struikenkamp, 1977; Wedgewood, 1963, cited in Wilmore et a l . , 1970ar316). The density of the lean body mass in normal humans has been shown to be dependent on age, sex, race, intensity of muscular ac t iv i ty , and nutritional states (Bakker & Struikenkamp, 1977; Parizkova, 1961a). Much of the change in the contribution of the skeleton to body weight during growth has been attributed to the maturation of the skeleton, the relative amount of bone to carti lage (Malina, 1969b), with a decrease in bone mineralization noted accompanying ageing (Behnke & Wilmore, 1974; Durnin & Womersley, 1974; Trotter, 1960, cited in Malina, 1969b:26). The estimated muscular mass in females shows relative s tab i l i ty from 15 through 60 years, with a range of 23.0 to 24.3 kg (Malina, 1969b). 18 There is a lack of knowledge concerning body composition in chi ldren, part icularly with respect to differences in the composition of the muscle and skeletal masses, and in their relationship to one another (Durnin & Rahaman, 1967). Although the estimative and predictive equations for body composition have inherent weaknesses^, the basic relationship that exists between body dimensions and weight allows for the description of the body's configuration in quantitative terms (Behnke & Wilmore, 1974). High corre-lations between selected body circumferences, stature, and body weight (Behnke, 1961); and between skeletal diameters, stature, arid lean body weight (Behnke, 1961, cited in Behnke, 1963:191) have been demonstrated. Edwards (1950) noted a close relationship between subcutaneous tissue thickness and body weight however, Shephard et al . (1969) reported this relationship to be poor in adult women. The lean body mass has been successfully predicted from skinfold thicknesses, muscular g i r ths, and skeletal width measurements; in pre-pubescent children (Slaughter et a l . , 1978; Wilmore & Behnke, 1970); from wrist breadth and height in children and adolescents (Bugyi, 1972); from skinfold and diameter measurements in wrestlers (Sinning, 1974); and from body diameters in college men (Wilmore & Behnke, 1968). Parizkova (1963) has also noted a positive relationship in children and adolescents, between thorax width, pelvis width, and the proportion~of lean body mass. The Behnke and Wilmore (1974) estimates of lean body weight are based on the assumption that a certain amount of lean body mass, \"muscle mass\", is associated with a given skeletal size (Behnke, 1963, cited in Behnke & Royce, 1966:76). The studies of Maresh'(1961) and Stuart et a l . (1940) (Both cited in Malina, 1969b:32) indicated moderate correlations be-tween bone and muscle during the early ages. The relationship between 19 skeleta l measures and true lean body mass has however, been found by other invest igators to be poor (Bakker & Struikenkamp, 1977). Furthermore, the studies of Baker (1961) and Tanner (1965) (both c i ted in Bakker & Struikenkamp, 1977:198) have shown no important degree of re la t ionsh ip between muscle diameter and bone diameter. Hebbelinck and Ross (1972, c i ted in Ross, Marsha l l , Vajda, & Roth, 1978:4) found the bone widths of young g i r l s to deviate in a pos i t i ve d i rec t ion and limb g i r ths in a negative d i rec t ion from re la t i ve height values, and Ross, McKim, and Wilson ( in Taylor , 1976:257) speculated that th is indicated that chi ldren may have a propor t ional ly greater amount of ske leta l t issue to muscle mass than adu l ts , or less muscle mass per uni t of s i z e . Drinkwater and Ross (1980) have developed the \"phantom model\", proposed by Ross and Wilson (1974), to include an anthropometric f r ac t i on -ation of the body mass into a four component system, comprised of f a t , muscle, s k e l e t a l , and residual masses. The system i s , e s s e n t i a l l y based on a proposit ion of Matiegka's (1921, c i ted in Drinkwater & Ross, 1980:178), in which the body i s d i v i s i b l e into an osseous, muscular, and skin plus fa t component. Fract ionat ion of the lean body components were \" a r b i t r a r i l y \" derived from cadaver analysis c i ted by Behnke (1974, c i ted in Drinkwater & Ross, 1980:183). The ''Drinkwater Tact ic \" of f rac t ionat ing the body mass into a four component system has a number of a t t rac t i ve features: 1. It accounts for to ta l body mass with an absolute error tolerance of 5%, and has the theoret ical advantage of permitt ing a l l four components to be derived independent of one another and of to ta l body mass, with the tota l mass serving as a major v a l i d i t y c r i t e r i o n . 2. The fa t mass value, when compared with those obtained through anthro-20 p o m e t r i c r e g r e s s i o n e q u a t i o n s , i s not a \"maverick\" e s t i m a t e , and l i e s i n the mid-range o f t h e s e p r e d i c t i v e v a l u e s . 3. The approach i s g e n e r a l , whereby, any measurement r e l a t i n g to a p a r t i c u l a r t i s s u e mass, may be used as i t s p r e d i c t o r , by v i r t u e o f i t s d e p a r t u r e from a s p e c i f i e d , s i n g l e , r e f e r e n c e human, which can be r e g a r d e d as a measuring d e v i c e f o r c o m p a r a t i v e p u r p o s e s . ( D r i n k w a t e r & Ross, 1980:186) Summary: M o r p h o l o g i c a l D e s c r i p t o r s The a c c u r a c y and v a l i d i t y o f a n t h r o p o m e t r i c e q u a t i o n s , to p r e d i c t the c o m p o s i t i o n o f the body, a r e a t t h i s time s u s p e c t because o f the absence o f v a l i d a t i o n a g a i n s t d i r e c t means. Fu r t h e r m o r e , the g e n e r a l a p p l i c a b i l i t y o f these e q u a t i o n s , t o p o p u l a t i o n s o t h e r than those from which they were o r i g i n a l l y d e r i v e d , s h o u l d not be assumed (Behnke & Wilmore, 1974; C u r e t o n , B o i l e a u , & Lohman, 1975; Damon & Goldman, 1964; M a l i n a , 1969b; P a r i z k o v a , 1961a; Steinkamp e t a l . , 1965, c i t e d i n M a l i n a , 1969b: 19). P r e d i c t i v e f o r m u l a e tend t o be s p e c i f i c t o t e c h n i q u e , i n s t r u m e n t , s i t e , sex, age,' and sample; and t h e s e f a c t o r s most l i k e l y a c c o u n t f o r d i s c r e p a n c i e s o b s e r v e d i n the l i t e r a t u r e . In view o f r e c e n t f i n d i n g s from d i r e c t c o m p o s i t i o n a l a n a l y s e s , p r e -v i o u s l y r e p o r t e d r e s u l t s and r e l a t i o n s h i p s s h o u l d be a c c e p t e d and c o n s i d e r e d w i t h some r e s e r v a t i o n s s i n c e , the b i o l o g i c a l c o n s t a n t s u n d e r l y i n g many i n -d i r e c t body c o m p o s i t i o n e s t i m a t e s are q u e s t i o n a b l e (Damon & Goldman, 1964). I t s h o u l d a l s o be r e c o g n i z e d t h a t most o f the a v a i l a b l e d a t a on body c o m p o s i t i o n i s d e r i v e d from c r o s s - s e c t i o n a l s t u d i e s which i n h e r e n t l y p o s s e s s a wide range o f sampling v a r i a b i l i t y ( M a l i n a , 1969b). 21 II MORPHOLOGY AND PERFORMANCE Th e r e i s voluminous d a t a t o s u b s t a n t i a t e the c l a i m t h a t morph-o l o g i c a l c h a r a c t e r i s t i c s have a v e r y r e a l and me a s u r a b l e e f f e c t on s p o r t p e r f o r m a n c e . I t has a l s o been e s t a b l i s h e d t h a t the \" n a t u r e \" o f the s p o r t d i c t a t e s , t o a d e g r e e , t h o s e p h y s i c a l c h a r a c t e r i s t i c s n e c e s s a r y f o r s u c c e s s and u l t i m a t e i n c l u s i o n i n e l i t e c o m p e t i t i o n s . Gymnastics and Other S p o r t D i s c i p l i n e s By c o l l e c t i n g a n t h r o p o m e t r i c d a t a on N a t i o n a l , World C l a s s , -and Olympic a t h l e t e s , and comparing a t h l e t e s w i t h one a n o t h e r , o r t o a r e f e r e n c e p o p u l a t i o n , t h e m o r p h o l o g i c a l c h a r a c t e r i s t i c s o f s p e c i f i c a t h l e t i c p o p u l a t -i o n s have been i d e n t i f i e d . I t i s w e l l documented and g e n e r a l l y a c c e p t e d t h a t d i f f e r e n c e s i n s i z e , shape, c o m p o s i t i o n , and \" p h y s i q u e \" i n g e n e r a l , e x i s t between e l i t e p a r t i c i p a n t s o f v a r i o u s s p o r t s , and i t i s not u n r e a s o n a b l e t o p o s t u l a t e t h a t t h e s e d i f f e r e n c e s d e monstrate the r e l a t i o n s h i p between s t r u c t u r e and f u n c t i o n : The s t u d y o f champion a t h l e t e s , t h e r e f o r e may p r o v i d e i n f o r m a t i o n on the s t r u c t u r a l r e q u i r e -ments f o r s u c c e s s i n the s p e c i f i c , t a s k s as w e l l as measures o f the d i f f e r e n c e s between t a s k s . ( C a r t e r , 1970:535) In a n t h r o p o m e t r i c comparisons o f Olympic f e m a l e a t h l e t e s , gymnasts have c o n s i s t e n t l y been d e s c r i b e d as the s h o r t e s t and l i g h t e s t p a r t i c i p a n t s , i n c omparison t o swimmers, c a n o e i s t s , s p r i n g b o a r d and h i g h d i v e r s , f e n c e r s , e q u e s t r i a n , and t r a c k and f i e l d a t h l e t e s o f the 1964 Olympics ( H i r a t a , 1966); s p r i n t e r s , swimmers, d i v e r s , c a n o e i s t s , and w e i g h t t h r o w e r s o f the 1968 Olym-p i c s (De Garay e t a l . , 1974); r u n n e r s and swimmers o f the 1972 Olympics (Novak e t a l . , 1977); and i n comparison t o r o w e r s , swimmers, c a n o e i s t s , 22 f e n c e r s , and t r a c k s p r i n t e r s , but not 1500 meter t r a c k a t h l e t e s , who were th e s m a l l e s t and l i g h t e s t o f t h o s e f e m a l e a t h l e t e s sampled a t the 1976 Olympic Games (Ross, 1980). The 1968 Olympic gymnasts a l s o had the s m a l l e s t b i a c r o m i a l and b i i l i o c r i s t a l b r e a d t h s ; the s h o r t e s t arm, l e g , and t r u n k l e n g t h s ; t h e s m a l l e s t s k i n f o l d v a l u e s , and the l o w e s t endomorphic r a t i n g , o f the a t h l e t e s sampled. They were more mesomorphic than the swimmers, d i v e r s , and s p r i n t -e r s , and more e c t o m o r p h i c than the c a n o e i s t s and w e i g h t thr o w e r s (De Garay e t a l . , 1974). The 1972 Olympic gymnasts a l s o had a s i g n i f i c a n t l y s m a l l e r b i -c r i s t a l b r e a d t h than the r u n n e r s and swimmers, and o f the 33 a n t h r o p o m e t r i c measurements t a k e n they had the s m a l l e s t v a l u e s , e x c e p t f o r lower l e g l e n g t h , f o r e a r m and upper arm g i r t h s , c o r r e c t e d upper arm d i a m e t e r , and f o r e a r m and i l i a c c r e s t s k i n f o l d s . In t h e s e l a t t e r measurements, the r u n n e r s had the s m a l l e s t v a l u e s . In the t r i c e p s and b i c e p s s k i n f o l d s , and p e r c e n t l e a n body mass, the gymnasts had the l a r g e s t v a l u e s . The gymnasts d i f f e r e d s i g n i f -i c a n t l y from the r u n n e r s o n l y i n h a v i n g more f a t o v e r the biceps.. Compared w i t h the swimmers, the gymnasts had s i g n i f i c a n t l y s m a l l e r f o r e a r m and upper arm g i r t h s ; b i - e p i c o n d y l a r femur w i d t h ; f o r e a r m , c a l f , i l i a c c r e s t , and u m b i l i c u s s k i n f o l d s . As w e l l , the gymnasts had s i g n i f i c a n t l y s m a l l e r ab-s o l u t e and p e r c e n t f a t masses., and a s i g n i f i c a n t l y l a r g e r p e r c e n t l e a n body mass and b i c e p s s k i n f o l d , than the swimmers (Novak e t a l . , 1977). The t e n d e n c y towards s h o r t n e s s \" i n s t a t u r e and l i g h t n e s s i n w e i g h t , noted f o r Olympic f e m a l e gymnasts, has a l s o been r e p o r t e d f o r l e s s e r s k i l l e d gymnasts, i n c omparison t o swimmers, d i v e r s , g o l f e r s , t r a c k and f i e l d , b a s k e t b a l l , f i e l d h ockey, S o f t b a l l , and t e n n i s a t h l e t e s ( M o r r i s , 1960, c i t e d i n C a r t e r , 1970:559); m i d d l e d i s t a n c e swimmers and r u n n e r s (Novak e t a l . , 23 1973); swimmers (S p r y n a r o v a & P a r i z k o v a , 1969); f i g u r e s k a t e r s , r owers, swimmers, s k i e r s , c l i m b e r s , b o w l e r s , h a n d b a l l , v o l l e y b a l l , b a s k e t b a l l , t a b l e t e n n i s , and t r a c k and f i e l d a t h l e t e s (Medved, 1966); t r a c k and f i e l d a t h l e t e s ( N e l s o n , 1974), and i n comparison to p r o f e s s i o n a l and amateur g o l f e r s , b a s k e t b a l l , and t r a c k and f i e l d a t h l e t e s ( C a r t e r , 1970). M o r r i s (1960, c i t e d i n C a r t e r , 1970:560) a l s o r e p o r t e d t h a t of. the a t h l e t e s sampled, the d i v e r s , gymnasts, and t r a c k a t h l e t e s were the h i g h e s t , i n mesomorphy. Novak e t a l . (1973) a l s o r e p o r t e d t h e i r gymnasts to have s i g n i f -i c a n t l y s m a l l e r t h i g h , h i p , c a l f , and maximal c h e s t c i r c u m f e r e n c e s ; and b i c r i s t a l , f e m u r a l , c o r r e c t e d t h i g h , and c a l f d i a m e t e r s than the r u n n e r s . As w e l l , the gymnasts had a s i g n i f i c a n t l y s m a l l e r body weight and f a t - f r e e mass, and a s i g n i f i c a n t l y s h o r t e r s t a t u r e than the r u n n e r s . The t r i c e p s s k i n f o l d was s i g n i f i c a n t l y s m a l l e r i n the gymnasts, compared w i t h the r u n n e r s and the swimmers. Of the 31 a n t h r o p o m e t r i c measurements t a k e n , the gymnasts had the s m a l l e s t v a l u e s , e x c e p t f o r the f o r e a r m , t h i g h , and c a l f c i r c u m f e r -ences and d i a m e t e r s , where the swimmers had the s m a l l e s t v a l u e s , and e x c e p t f o r the c o r r e c t e d upper arm d i a m e t e r , where the gymnasts had a l a r g e r v a l u e than the r u n n e r s . The gymnasts o f S p r y n a r o v a and P a r i z k o v a ' s s t u d y (1969) were a l s o s i g n i f i c a n t l y s h o r t e r and l i g h t e r , p o s s e s s e d a s m a l l e r a b s o l u t e l e a n body mass, p e r c e n t f a t mass, and a b s o l u t e f a t mass; and a h i g h e r p e r c e n t l e a n body mass than the swimmers. N e l s o n (1974) a l s o found the gymnasts to have a s i m i l a r c h e s t c i r c u m f e r e n c e t o the t r a c k and f i e l d a t h l e t e s , w i t h t h e i r l e g c i r c u m f e r e n c e -l e s s than the t r a c k and f i e l d a t h l e t e s . C a r t e r (1970) found t h a t a l m o s t a l l groups o f champion a t h l e t e s 24 are rated high on mesomorphy, and of the females, the track and f i e ld jumpers and runners have the lowest mesomorphic rat ing, and the gymnasts the highest. Descriptive Studies of Gymnasts While descriptive studies of female gymnasts from e l i te ab i l i ty levels ex is t , documented studies of female gymnasts from varying ab i l i ty levels are sparse. Most studies concerned with ab i l i ty and morphology have involved e l i te level gymnasts, and have compared the winners of a com-petition with the lesser placing participants: (Montpetit, in Salmela, 1976: 183; Pool, Binkhorst, & Vos, 1969; Youngren, 1969). While these studies are valuable, they do not provide a descriptive demonstration of morpholog-ical changes that may accompany progressive increases in ab i l i ty from recreational through to e l i te levels . Disparity in anthropometric descriptions of female gymnasts, that arise from inter-study comparisons, may be due to the fact that the ab i l i ty leve l , age of the gymnasts, and the year that the study was conducted, varies from one study to another. In studies relating structure and function, a simple categorization such as \"gymnasts\", is misleading and inadequate for both descriptive and comparative purposes. . . . Adequate struct-ural description of samples is necessary for correct interpretation of findings related to function. (Carter, Sleet, & Martin, 1971:162) Pool et a l . (1969) studied the anthropometric measurements of the female competitors of the 1967 European Gymnastic Championship. Concerning anthropometric dimensions and performance, low correlations were found for height and weight, and performance. A positive correlation between thorax 25 width and performance1 was found however, indicating that \"possibly the best gymnasts have more muscle mass and are therefore more strongly bui l t\" . As was expected, since \"in gymnastics . . . body weight is a handicap and body fat a superfluous load\", a negative correlation between the skinfold measure and performance was reported (Pool et a l . , 1969:336). The Eastern gymnasts identif ied as the \"better\" performers in comparison to the Western gymnasts, were younger, shorter, l ighter , and had a smaller mean skinfold value. However, none of these differences was s ta t i s t i ca l l y signif icant. . Pool et a l . (1969) also compared the results of their study with those of six other studies of female gymnasts, from varying ab i l i ty levels , in the same age range of 19 to 25 years. These gymnasts were; (a) from the 1964 Olympic Games, the mongolian and non-mongolian participants; (b) from the 1961 World Championships of Students; (c) from the Russian selection of top gymnasts, 1964; (d) from the Dutch selection of top gymnasts, 1966; (e) Dutch gymnasts of low capacity, 1966; and (f) Yugoslavian gymnasts of modern capacity, 1966. Al l top gymnasts from the various countries had .. similar mean heights and weights, ranging from 156.3 to 160.3 cm, and 51.0 to 55.3 kg, except the mongolian competitors who were shorter and l ighter at 152.0 cm and 46.8 kg. Even though-the Dutch mean performances were lower than the Europeans, they had similar mean heights and weights. However, the mean height and weight values of the Dutch top gymnasts were s i g n i f i -cantly shorter and l ighter than lower caliber Dutch gymnasts, with a mean height of 167.7 cm and mean weight of 59.5 kg. The U.S. National College Gymnastics Champions of 1970-71 were studied by Sinning and Lindberg (1972). With a mean age of 20 years, height :of 158.5 cm, and weight of 51.1 kg, these female gymnasts were similar to the gymnasts of the European Championships described in Pool et a l . ' s study (1969). 26 The female gymnasts from Novak et a l . ' s study (1977) were a l l from one country that did not reach a signif icant place during the 1972 Olympic Games. These gymnasts, with a mean height of 163.5 cm, were substantially ta l le r than the European and top ranking gymnasts from the various countries described in Pool et a l . ' s study (1969), and ta l ler than the U.S. National College champions studied by Sinning and Lindberg (1972). With a mean weight of 52.5 kg, they were however, similar to the gymnasts from these studies. Youngren (1969) studied the relationship of selected anthropometric measurements to gymnastic performance, in the 1968 U.S. National Women's Olympic T r i a l s . These participants ranged in age -from 15 to 30 years, with a mean age of 18.5 years, weight of 52.4 kg, and height of 159.7 cm. No signif icant differences were found between the selected anthropometric measurements and placement, although many anthropometric variations were found among the top performers, and in general: Those who were shorter and of thinner skinfolds had a tendency to rank higher. . . . weight had l i t t l e effect on how a top gymnast placed. . . . Body type [described in terms of the ponderal index] was not signif icant to the performance of top women gymnasts . . . However, body type was very similar in the women gymnasts. (Youngren, 1969:41-42) Montpetit (in Salmela, 1976:183) noted the range in age, 18.5 to 19.9 years; in height, 158.9 to 162.3 cm; and weight, 46.8 to 52.5 kg, of the top f ive female gymnastic teams and the tbp s ix , a l l around winners, at the 1972 Olympics, and concluded that once international prominence is attained, size is no longer decisive in determining gymnastic success. The correlation coeff icient between rank and height, calculated for the top 36 female gymnasts in these Games, resulted in a low \"r\" of 0.01. Drazil (1971) has noted that the percentage of body fat in trained 27 female gymnasts does not exceed 10 to 15%, and Grossfeld has observed that: with rare exceptions, the maximum height for a female gymnast is 5 feet 5 inches Q~65.1 cm]. The other end of the scale is about 4 feet 10 inches [T47.3 c l l \u00E2\u0080\u00A2 (Grossfeld, cited in Nichols, 1979:18) Ross (1980) compiled anthropometric data on 15 of the female gymnastic competitors at the 1976 Olympic Games, and described them in terms of over 51 variables, including body compositional analysis. In general, i t has been noted that the higher the level of competition, the narrower the var iab i l i ty of physical character ist ics. Carter et a l . ' s (1970) comprehensive investigations have led them to the general consensus that; athletes are somatotypically different from the general population; certain athletic groups are somatotypically different from one another; and performers at the same level of competition, in the same sport, tend to be of similar physique. Biomechanical Advantages of Specific Physiques While most descriptive and comparative studies have succeeded in identifying the physical characteristics common to participants of a part-icular sport (Alexander, 1976; Chovanova, 1972; Ross & Day, 1972), there is usually only general, i f any, comment on how these characteristics aid or deter the athlete in performing specif ic elements of the sport. The biomechanical advantages afforded by certain physiques, in specif ic sporting endeavours, are very real (Hebbelinck & Ross, in Nelson & Morehouse, 1974:546; Lewis, 1969); and Khosla (1968) has gone as far as to suggest that certain events in the Olympic Games are unfair for specif ic competitors, because of their \"bui ld\". In a later a r t i c l e , Khosla (1977) discusses the potential of nations to win medals at the Olympic Games, based 28 on the distr ibution of height found in the populations. Since gymnastic performance requires high re la t ive , rather than absolute, strength, smallness is an advantage resulting in a high strength-to-mass ratio (Le Veau et a l . , 1974) since: Strength is proportional to the square of any length (height) . . . we should expect strength to increase as the power 2 and mass as the power 3 of a chosen height value. (Ross & Marshall, 1979:13) The smaller stature observed in gymnasts also allows them to perform free f l igh t spins and rotations more easily than larger athletes since: The small stature results in a smaller moment of inert ia about an axis in the transverse plane through the mass center. . . . Further the moment of inert ia of the total body about other axes such as the horizontal bar would also be less. (Le Veau et a l . , 1974:150) and; When examining linear motion the inertia of the object is direct ly proportional to i ts weight. However, in angular (rotary) motion, not only is the weight important but also i ts d ist r ibu-tion in relation to the axis of rotation . . . . A smaller person has a smaller moment of inertia when rotation occurs about his center of gravity in free f l igh t or when he rotates about his hands. (Nelson, 1974:46) The domination of men's gymnastics by the Japanese, in the 1972 Olympic Games, has led Le Veau et a l . (1974) to speculations concerning advantages that they may have as a result of their physique. In comparison to the American male gymnasts, the Japanese gymnasts were shorter; smaller in hand, foot, shank, and arm lengths; greater in trunk and extremity c i r -cumferences; and greater in chest circumference and body weight relat ive to 29 height. These differences appear to be conducive to movements in which the body is supported by the arms and moved about the hands, with the shorter shank and smaller feet making i t easier to control the lower extremities (Le Veau et a l . , 1974). Furthermore, Nelson (1974) contends that a shorter arm would be an asset for swings performed on the horizontal bar; since: [with a shorter arm] the center of gravity of the body is nearer the axis of rotation in swings on the bars. . . . fand this] tends to reduce the moment of inertia of the body when rotation occurs about the hands. (Nelson, 1974:46) Similar ly , Rozin (1974) noted that a gymnast with relat ively long arms would need to expend more effort in order to maintain his body in the iron cross held position on the r ings. The biomechanical requirements of the long horse vault may have created a disadvantage for the Japanese, and thus contributed to their low placement in this event: Since the height of the horse is standard, i t follows that a shorter person must raise his centre of gravity to a higher point before contacting the horse, which requires a greater vertical velocity at takeoff. (Le Veau et a l . , 1974:150) Furthermore, Nelson (1974) claims that a shorter lower limb is less suited to running, and this may affect the running approach to the vault . The fact that this was the only event in which the Japanese men did not win at least two medals, and in which their highest f in ish was fourth, perhaps supports these contentions. In addition to these factors, the takeoff in vaulting is not completely dependent on the ab i l i t y of the gymnast, as a comparative study of f ive internationally approved gymnastic vaulting boards revealed: There would be an advantage for those gymnasts 30 who use a beat board whose quality of e las t ic i ty was commensurate with their execution potential , that is their body mass, their speed and their force of takeoff. (George, in Salmela, 1976:96) Furthermore, Val l iere introduced the concept of \"performance discrimination\", notably that: The structure as well as the internal qualit ies of the apparatus do not provide equal advantages to a l l gymnasts during performance. Does not the diver adjust his diving board in relation to his execution potential , while the pole vaulter s e l -ects a pole that is compatible with his weight and speed? There s t i l l remains much to do to ensure the fact that the gymnast wil l become the only determining factor responsible for his performance. (Val l iere, in Salmela, 1976:96-97) It appears that the best body type for a gymnast, who competes in a l l events, is a compromise of the ideal morphological determinants best suited for a one-event spec ia l is t , such that \"deficiencies in one component may be rect i f ied by a surplus in another\" (Salmela, Halle, Petiot, & Samson, 1976:169). Youngren (1969) noted that those gymnasts that placed well on the f loor exercise, also placed well on the beam, and those placing well on the uneven bars, also tended to place well on the vault. Rozin (1974) considers long arms a disadvantage on the r ings, but an asset in performing on the side horse, where swinging movements predominate. Hi rata (1966) claims that, in the Olympic Games the method of training has l i t t l e incidence on the resul ts , because i t has reached the top level in each country: Physique and consti tut ion, which can by no means be reconstructed by training, seem to have an im-portant effect . . . . when training has reached the top l eve l , the most adequate physique wil l win. (Hirata, 1966:222) 31 Tanner (1964) and Eiben (1972) support the credence that athletes are both \"born\" and \"made\", and that top athletes have already been selected according to their body build by the time they have arrived at the Olympics: Physique is a factor in the sort of success that may lead to inclusion in an Olympic Team; more negatively, that the lack of proper physique may make i t impossible for an athlete to reach that degree of success. But we do not suppose that winning the Olympic event has much to do with physique, except perhaps in some rare cases where one single man is altogether outstanding. By the time the f inals are reached even the physiques are becoming matched one with another. (Tanner, 1964:14) Summary: Morphology and Performance It has long been recognized that body \"build\" is an important prerequisite for successful participation in high level sporting endeavours. In comparison to other athletic groups, the e l i te female gymnast has been identif ied as short in stature, l ight in weight, and possessing a high percent lean body weight accompanied by a low percent fat weight, as well as a high mesomorphic component. These characteristics have been ass-ociated with biomechanical advantages, such as a reduced moment of iner t ia , when rotation occurs about the transverse axis, around the center of gravity, or around the hands. As wel l , these characteristics result in a high strength-to-mass ra t io , which is also conducive to gymnastic-type movements. In comparison to lesser ski l led gymnasts, the e l i te gymnast is shorter and l ighter , and has smaller skinfold thicknesses. Among e l i te female gymnasts a narrow var iab i l i ty in physique is observed. However, the more successful gymnasts tend to be shorter, l ighter , and have smaller skinfold values. A large thorax width has also 32 been identif ied with success at this leve l . Relationships between body morphology and the properties of standard gymnastic equipment have revealed the occurrence of a \"performance discrimination factor\", in that, the equipment does not offer the same advantage or potential to a l l individuals. The best morphological physique, for a gymnast competing in a l l events, has been identi f ied as a compromise between the best build for each apparatus. I l l PERFORMANCE AND MORPHOLOGY While i t is recognized that successful participation in a part-icular sporting endeavour entails specif ic \"physique requirements\", i t is al so 'r_ea.li=zed: -tha-t\" ^ performance or training in that sport has the effect of modifying specif ic physical parameters. It is well accepted that exercise has a potentially great influence on modifying the body, internally through compositional and physiological changes, which are reflected externally in altered contours and dimensions (Brozek, 1961; Malina, 1969a), The potential or ab i l i ty of an individual to respond to environ-mental pressures, such as exercise, with short term phenotypic modifications, such as biochemical, physiological , and morphological al terat ions, is refer-red to as the concept of p last ic i ty (Malina, 1976). Individuals vary considerably in their plast ic responses. P last ic i ty can be viewed in two ways, that operating during the period of growth and development (developmental plast ici ty) / and that operating during adulthood. . . . and modifications in.the growth and development processes to envi-ronmental stress may become permanent and are 33 thus irreversible when adulthood is attained. . . . changes associated with regular physical act iv i ty ref lect plast ic responses and the l imits of an individual 's p last ic i ty is set by his genotype. (Malina, 1976:157) Fat Mass Content and Act iv i ty Levels The consistently lower fat mass values of athlet ic females, in comparison to re lat ively sedentary female reference populations, i l lust ra te the effect of exercise on the fat mass component of the body. Percent fat mass values of 20 to 30% have been noted for young adult women (Ljungren, 1965, cited in Novak et a l . , 1977:282; Malina, 1969b; Wilmore & Behnke, 1970); 14.8%, 19.2%, 23.4%, 24.9%, 21.7%, 24.0%, 22.2%, 23.2%, 23.6%, 28.1%, and 22.7%, for 8 to 18 year old females, inclusive; and 22.8 to 24.7% for 18.5 to 29.0 year olds (Forbes, 1972). In comparison to these values, the percent fat mass values of athlet ic females, have been reported as 13.0% for e l i te figure skaters (Faulkner, 1977); 11.1% (Ross, 1980), and 12.9% (Novak et a l . , 1977) for Olympic gymnasts, 14.6% for prominent gymnasts (Novak et a l . , 1973), 16.8% for low caliber gymnasts (Sprynarova & Parizkova, 1969); 13.3% for Olympic runners (Novak et a l . , 1977), 16.8% for prominent runners (Novak et a l . , 1973); 18.9% for Olympic swimmers (Novak et a l . , 1977), 17.0% for prom-inent swimmers (Novak et a l . , 1973), and 19.2% for high caliber swimmers (Sprynarova & Parizkova, 1969). Since females are reported to possess more subcutaneous fat than males at almost a l l ages (Bonnet, Rocour-Brumioul, & Heuskin, 1979; Edwards, 1951; Parizkova, 1963; Reynolds, 1950), i t is of interest to note how extreme act iv i ty may reduce this difference. Smit (1973) compared the skin-fold thicknesses of adult male and female gymnasts, and noted remarkably 34 small differences. Signif icant reductions in the fat \"content\" of adolescent g i r l s have been noted accompanying programs of regular physical act iv i ty (Wells et a l . , 1963), and Smit (1973) reported that skinfold values of gymnasts decreased as their number of act iv i ty hours per week increased. Johnson (1969) noted that children involved in daily physical education classes, in comparison to children participating 2 and 3 times a week, have smaller fat mass contents. Parizkova (1959) studied a group of adolescent female gymnasts, and noted an increase in weight with no change in fat content, following a 6 week period of training. However, after a 10 week period of relative inact iv i ty , a further increase in weight, accompanied by a substantial increase of 35.7% in subcutaneous fat (sum of the skinfolds) , was noted. In comparison, a control group of females of the same age experienced a 6.9% increase in subcutaneous fat , during the same 10 week period. Simi lar ly , in a later study, Parizkova and Poupa (1963) studied a group of female gymnasts from the Czechoslovak!'an National Team, with a mean age of 23 years, and two groups of gymnastic sports students, with mean ages of 16 years. After periods of intense training, the weight remained unchanged in the groups, while a definite f a l l in fat content and increase in lean body mass was noted. After periods of relat ive rest , body weight rose s ign i f icant ly , along with increases in fat content. These changes were reported to have occurred in the younger gymnasts during considerably shorter periods. These studies demonstrate the effect of exercise in \"reducing\" and \"checking\" or maintaining the fat \"content\" of the body. 35 With weight reduction, as adipose tissue de-creases in total s ize , the total number of ce l ls remains constant. Hence weight reduction is achieved by a reduction in cel l volume. (Hirsch, 1972:84) Cr i t ica l Periods for Fat Mass Deposition Current evidence suggests that specif ic periods in the growth sequence of the human are associated with deposition of adipose t issue, hyperplasia, and with increases in adipocyte s ize , hypertrophy. Hirsch (1972) claims that adipose cel ls are laid down late in gestation, in the f i r s t year of l i f e , and in early adolescence. Bonnet et a l . (1979), in a cross-sectional study of normal weight infants, chi ldren, adolescents, and adults, noted that cel l size s igni f icant ly enlarged during the f i r s t six months of l i f e , but beyond the f i r s t year there were no signif icant changes until puberty, when there was a substantial increase in adipocyte s ize. Concerning cel l number, i t was observed that no signif icant increase of mature adipocytes occurred in the f i r s t year of l i f e . However, the number progressively rose during childhood, and there was a signif icant and very great increase during puberty, with the adult value being reached at the end of adolescence. Behnke and Wilmore (1974) suggest that random development of new adipocytes - probably occur -when fat eel Is-\"normally present\" attain-'maxfrnal' saturation (about 85%) with t r ig lycer ide. There is some speculation that i f the periods of adipocyte multi-pl icat ion can be pinpointed, control over \"fat potential\" may be possible (Hirsch, 1972). Furthermore, according to Rarick (1947, cited in Barnes, 1979:117), percent body fat wil l be lower in young adults who exercised as chi ldren, and:. ' -36 exercise is more effective in influencing develop-ment i f i t takes place during the growth spurts of adolescence than i t is during preadolescence. (Rarick, 1947, cited in Barnes, 1979:117) Muscle Mass Content and Act iv i ty Levels Physical training results in muscular hypertrophy and an increase in contracti le proteins, whereas physical inact iv i ty results in atrophy and a reduction of contracti le proteins. (Malina, 1969a:21) The effects of exercise, on the muscle mass, can be observed direct ly in the increased muscular girths that resul t after prolonged periods of- intense training. Using a variety of weight training techniques, increased muscular girths have been reported in elementary children (Bready, 1961), in adolescent boys (Kusinitz et a l . , 1958), and in adults (Tanner, 1952, a l l cited in Malina, 1969a:24). Increased muscle mass can only result from muscle work, supported by an appropriate increase in dietary intake. Without muscle work, no food, vitamin, hormone or drug wil l increase muscle mass. (Smith, 1976:151) Bulky, bulging muscles are almost impossible for females to attain through strenuous exercise, such as weight l i f t i n g , because they possess low testosterone levels. Testosterone is a \"powerful hormone involved in the deposition of protein in the formation of muscle tissue\" (Johnson, Updyke, Schaefer, & Stolberg, 1975:124). As wel l , i t should also be noted that deterioration in eff iciency of gymnasts and basketball players, undergoing strenuous training, has been associated with breakdown of the lean body mass (Zhdanova & Parizkova, 1962, cited in Parizkova, 1968b:275). 37 In general, the type and intensity of the exercise, and hereditary influences, determine the appearance and shape of the muscles, as well as muscular development (Church, 1976). Cr i t ica l Periods for Muscle Mass Deposition During training there is an increase in muscle mass by an enlargement of the already existing f ibers. Most investigators find no increase in the number of muscle ce l ls with training, through a division of already existing c e l l s . . . . The total amount of protein in the muscle increases with training and decreases with inact iv i ty . Disuse atrophy is associated with a decline in myofibr i ls , and the proportion of sarcoplasm proteins r i ses . (Astrand & Rodahl, 1970:399) Whether or not the above is true during the active years of growth is not clear. Cheek (1968, cited in Ross, McKim, & Wilson, in Taylor, 1976: 257) presents longitudinal data which indicates an increase in skeletal muscle cel l population for both boys and g i r l s , with a markedly greater increase in boys after I0h years. The superior strength.performance of adult males over adolescent males, and the lack of a comparable difference in females (Jones, 1949, cited in Reynolds, 1950:107), suggests the occurrence of di f ferent ial increases in the male and not the female (Ross, McKim, & Wilson, in Taylor, 1976:257). It has been speculated that exercise may be a determinant in cel l d i f ferent iat ion: If during the growth spurt exercise of the proper kind and amount is not forthcoming, stem cel ls may differentiate as fat ce l ls rather than muscle c e l l s . If large percentages of our population are unexercised at the c r i t i c a l period of muscle cel l increase, we should expect the deleterious effect on strength and stamina would be greater in males than females. (Ross, McKim, & Wilson, in Taylor, 1976:257) 38 Furthermore, Bowden and Goyer (1960, cited in Malina, 1969a:19) suggest that the size differences, observed in the fibers of different muscles during normal growth, are direct ly related to functional act iv i ty . Malina (1969a) has commented on the necessity of distinguishing the growth of skeletal muscle due to exercise, from the increase in muscle due to normal growth. Jokl et a l . (1941, cited in Malina, 1969a:22) reported striking increases in body weight in 16 to 21_year old male recruits subjected to six months of systematic ac t iv i ty . They attributed this increase to muscle tissue gains and concluded that: Physical training may not be postponed until the age of developmental r ig id i ty is reached. . . . t ra ining, i f applied to younger age groups, is capable of producing a much greater influence upon development. Had the training which our recruits received, been applied eight or ten years ear l i e r , the effect would have been more marked; they would have developed a better physique. (Jokl et a l . , 1941, cited in Malina, 1969a:22) Skeletal Mass Content and Act iv i ty Levels The effect of exercise on the skeletal structure, mass, and composition, is d i f f i c u l t to assess. Morphological variables, that are genetically endowed, are probably more obviously expressed in the skeletal mass than in the fat and muscle masses, which appear to be more malleable to the influences of exercise. It is known however, that moderate exercise has a positive effect on bone growth by strengthening the organic matrix of the bone, through stimulating osteoblastic deposition: Bone is continually being deposited by osteoblasts, and i t is continually being absorbed where osteo-clasts are active bone ordinari ly adjusts i ts strength in proportion to the degree of bone stress. Consequently bone thickens when subjected to heavy loads. . . . Even the shape of the bone can be 39 rearranged for proper support of mechanical forces by deposition and absorption of bone in accord with stress patterns the bones of children in whom the rate of deposition and absorption is rapid, show l i t t l e britt leness compared with the bones of old age, at which time the rates of deposition and absorption are slow Bone is deposited in proportion to the compressional load that the bone must carry. . . . the bones of athletes become considerably heavier than those of nonathletes Therefore continual phys-ical stress stimulates osteoblastic deposition of bone. (Guyton, 1976:1058-1060) Inactivity has been shown to have detrimental effects on the strength and composition of bones, and bones subjected to immobilizing casts become thin and decalcif ied through inact iv i ty (Bullough, Goodfellow, & O'Connor, 1973; Guyton, 1976). Although resumption of physical act iv i ty corrects this d is -turbance, many years may be needed to restore the loss (Kotte, 1966, cited in Malina, 1969a:17). While the mechanical stresses of normal weight-bearing, and the tension and compression of muscular forces, are generally recognized as essential for normal bone formation, development, and growth, the l imits of tolerance, beyond which an increase in pressure or tension leads to destruct-ion of bone, by resorption, remains undetermined (Malina, 1969a). Cr i t ica l Periods for Skeletal Mass Deposition The potential size of the skeletal structure is predetermined by genetic factors (Church, 1976; Johnson et a l . , 1975), and the effect that stressors such as exercise have on this potential is not entirely understood. Since bone is continually undergoing change through osteoblastic deposition and osteoclastic absorption (Guyton, 1976), experiences periods of \"stepped-up growth\" (Tanner, 1962), and does not entirely stop growing until approxim-ately the second decade of l i f e (Greulich & Pyle, 1970; Tanner et a l . , 1975), 40 i t is probably more vulnerable to \"stressors\" during the developmental years, before fu l l maturity is reached. Arnold (1930) and Correnti (1941, both cited in Tanner, 1962: 134-135) claim that exercise increases the rate of growth, and Prives (1960, cited in Malina, 1969a:24) found physical exercise to favour growth of bone in length. On the other hand, Rarick (1960, cited in Johnson et a l . , 1975:123) claims that heavy pre-pubertal exercise may result in a somewhat heavier and shorter skeletal structure, and Goodings and Neuhauer (1965, cited in Malina, 1969a:18) noted increased vertical growth of the vertebral bodies, in the absence of normal weight-bearing function, during the growing years of humans. Beyer (1896\", cited in Malina, 1969a :22) studied a group of 16 to 20 year old male cadets subjected to six months of systematic \"gymnastic\" ac t iv i ty , and reported an average increase in stature for this group, of about one inch over controls. Since the greatest gain occurred between 16 and 17 years of age, Beyer inferred that factors capable of influencing growth would exert their effect at a time when the growth impulse was strong. The specialized use of body parts, and lateral size differences, provide a possible means of assessing the effects of physical act iv i ty on bone growth and development. Van Dusen (1939, cited in Malina, 1969a:23) found that the right upper extremity measurements in chi ldren, one through eight years, were generally larger than the le f t . The measurements of the right tended to be larger more frequently with increasing age, suggesting development through specialized use. From radiographs, Vicinus (1962, cited in Malina, 1969a:23) noted a tendency for the right hand of adults to be larger than the l e f t , in both length and width, with differences in breadths more marked. Buskirk et a l . (1956, cited in Malina, 1969a:24) 41 found nationally ranked tennis players to have greater musculature and osseous development, in both length and width of the radius and ulna, in the dominant hand and forearm than in the non-dominant members. Since the players had participated extensively during their teen years, this la tera l i ty difference was attributed to the effects of exercise on bone growth during adolescence. While there is some speculation that participation in sports, during adolescence, may have an unfavourable effect on the adolescent female boney pe lv is , Erdelyi (1962) did not confirm this hypothesis, and the data of Ivata and Kadsuo (cited in Erdely i , 1962:177) also disputes this claim. In a comparative study of female gymnasts, and a corresponding reference population, Smit (1973) found the intercr ista l width to be far smaller in gymnasts. The biacromial width, in comparison, was found to be nearer to the norms in spite of the relative shortness of the gymnasts. Buckler and Brodie (1977) reported similar results for 10 to 19 year old boys involved in gymnastics. Parizkova (1968a), in a study of active and inactive boys followed longitudinally from 11 to 15 years, found the active-boys after a f ive year period, developed a larger biacromial breadth, and a s igni f icant ly narrower pelvis in relation to their height, and to their biacromial breadth, than the less active boys. These results suggest that: Gymnastics, in addition to i ts obvious influence on the musculature has a favourable influence upon skeletal growth in the region of the shoulders. (Smit, 1973:484) The increased biacromial diameter is secondary to the prolonged physical ac t iv i ty , implying that muscular growth and use can influence the way in which bones develop. (Buckler & Brodie, 1977:462) 42 Adams (1938, cited in Malina, 1969a:22) found Negro women, ages 17 to 21 years, subjected to strenuous physical labor during their childhood years, to be t a l l e r , heavier, and larger in muscle g i r th , chest breadth and depth, and hip and knee width, than women of the same age not subjected to such stress. Simi lar ly , Godin (1920, cited in Malina, 1969a:22) found \"gymnasts\" (active youths), ages 14% to 18 years, to be t a l l e r , heavier, and larger in thoracic and forearm measurements than \"non-gymnasts\". The size differences in both these studies were attributed to the programs of heavy physical work. Summary: Performance and Morphology Exercise has been shown to dynamically affect the body for short terms, and possibly to the extent of resulting in some permanent tissue changes as c r i t i ca l periods have passed. The effects of exercise on the fat and muscular components of the body are easily observed and accepted. It is known that exercise has a positive effect on strengthening the organic matrix of the bone. However, whether or not exercise can substantially alter the development, structure, and size of the skeleton s t i l l remains an unanswered question. Type of exercise, as well as genetic contributory factors, dictate to a degree, the extent of morphological changes possible. Genetic factors may well dictate the vulnerabil i ty of the body to specif ic exercise \"stressors\", and may affect the same individual in different ways at different times, depending on the stage of maturity reached. While moderate exercise may favourably affect and stimulate growth, excessive, strenuous exercise may deter and negatively affect normal growth patterns and potentials. The maximum tolerance l im i ts , beyond which the 43 positive effects of exercise and training cease to ex ist , and the negative destructive forces come into operation, have not been determined. IV PERFORMANCE AND MATURITY As the number of pre-adolescent and adolescent children par t i c i -pating in strenuous training programs increases, concern over the effect of exercise on the growth and maturing processes has taken on a new emphasis: In 1966, 5 year olds weren't running marathons or training as Olympic gymnasts in the numbers they are today. Can cumulative microtrauma of some sort change the picture when preadolescents train long and hard? (Barnes, 1979:116) Krustev (1977:25), in an Olympic Committee Bul le t in , raised the question of setting age l imits for participants in the Olympic Games, with concern for \"the potential danger to health which such rigorous training at a very early age might enta i l\" . These concerns are well founded since: Puberty by i t s e l f is for the organism a stress which may cause troubles; i f during this period, the individual undergoes training to i ts extreme l imi ts , the stimulation which puberty exercises spontaneously on the endocrine glands is summed with that caused by motor ac t iv i ty ; by conse-quence the functional load of a fast physical development is added to that of the training; i t is a vicious c i rc le with cumulative action, whose negative effects sometimes appear pre-maturely, but more often later on. (La Cava, 1974:163) Maturity Indicators While the term \"growth\" usually refers to the physical act of enlargement of a structure, such as a bone length, muscle g i r th , or volume increase; the term \"maturity\" is reserved for \"qualitative\" changes in the structure, and involves differentiat ion of the c e l l s , leading to shape and/or 44 functional changes: While growth and development proceed concomitant-ly in the normal c h i l d , they are to some degree potentially independent processes. (Greulich & Pyle, 1970:2) There are a number of ways in which the level of maturity of the body, and the rate of maturity can be assessed and monitored. Among the techniques presently ut i l ized for.females are: (a) morphological age: the changing shape and proportions of the body (Behnke & Wilmore, 1974; Bie l ick i & Waliszko, 1975; Eiben, 1978; Ross & Wilson, 1974; Tanner & Whitehouse, 1976; Tosovsky et a l . , 1976). (b) secondary sex character ist ics: the stages of development of the breasts, and pubic and axi l lary hair (Marshall, 1972; Tanner, 1962, 1978), as well as the number of maturity indicators or developmental milestones experienced, such as peak height spurt and menarche (Eveleth & Tanner, 1976; Tanner, 1962, 1978). (c) dental age: the number of primary and secondary teeth erupted (Anderson, 1975; Demerjian et a l . , 1973, cited in Tanner, 1978: 82; Eveleth & Frei tas, 1965; Eveleth & Tanner, 1976; Filippson & Ha l l , 1976; McGregor, Thomson, & Bi l lewicz, 1968). (d) endocrine secretion patterns: The presence and level of hormones circulat ing in the blood (Reiter & Kulinh, 1972; Tanner, 1978). (e) skeletal age: assessing the stages of development of the bones from radiographs (Greulich & Pyle, 1970; Tanner et a l . , 1975). (i) Skeletal age as a maturity indicator. Skeletal age as a maturity indicator has a number of attractive features: 45 1. It can be assessed with relat ively objective c r i t e r i a , and can be used throughout the entire lifespan until completion of skeletal develop-ment (Greulich & Pyle, 1970; Tanner et a l . , 1975). 2. The radiograph is easily attained without involving personal questioning or examination. 3. The data is portable, and can be re-evaluated. 4. As well as making the dist inct ion between growth and maturity, the method is able to differentiate between \"slow growers\" and those who are inherently short; as well as between \"fast growers\" and those who are inherently ta l l (Johnston, 1962; Tanner et a l . , 1975). 5. The maturational changes in the skeleton are intimately related to those of the reproductive system (Tanner et a l . , 1975), and are assoc-iated with other growth and maturity indicators: Menstruation generally occurs between the skeletal ages of 13 and 14 years and i t is said never to have been seen in a g i r l with a skeletal age of less than 12.5 years or more than 14.5 years. (Marshall, 1974:310) The ulnar sesamoid of the f i r s t metacarpophalangeal joint . . . showed a very close relationship with the onset of secondary sex characteristics and with ages at in i t ia t ion and peak of adolescent height velocity. Usually i t became v is ib le radiographic-a l ly .5 years after the onset of pubic hair devel-opment, .75 years after the in i t ia t ion of the height spurt and .7 years before peak height velo-c i ty . It indicates that puberty has already started and that height velocity is in the accel-erating phase and that 88% of adult height is reached. (Onat & Numan-Cebeci, 1976:659) 6. The method is useful in the prediction of adult height from height during childhood or early adolescence (Acheson & Dupertuis, 1957; Tanner et a l . , 1975), and in the prediction of menarche. Menarche usually occurs between 10.0 and 16.5 years chronologically (Tanner, 1962), and between 46 12.5 and 14.5 years, skeletal ly (Marshall, 1974). ( i i ) Skeletal age rating systems. The most commonly used methods of assessing skeletal maturity are those developed by Greulich and Pyle (1970) and Tanner et a l . (1975), in which the le f t hand and wrist area is rated. Systematic differences between the two methods have been reported, with the Tanner-Whitehouse II method, bone speci f ic approach, yielding a substantially-higher value than the Greulich-Pyle method, atlas approach, for the same radiograph (Ross, McKim, & Wilson, in Taylor, 1976:257). The relative s tab i l i ty of the rating systems is based on the s tab i l i ty of the maturing sequences of the skeleton: The bone stages and their individual sequences are the same in a l l populations. (Tanner et a l . , 1975:18) and th is , is evident both before and after b i r th . (Greulich & Pyle, 1970:24) Each bone passes through al l stages, although stages last for varying times. (Tanner et a l . , 1975:4) and, The bone stages and their individual sequences are unaffected by starvation. (Tanner et a l . , 1975:18) ( i i i ) Menarche as a maturity indicator. Menarche, as a maturity indicator, is commonly used to assess the rate of maturity of most female populations (Eveleth & Tanner, 1976). Menarcheal surveys are usually conducted using one of three methods: (a) prospective method - recording the onset of menarche in a longitudinal study; (2) retrospective method -recording recalled age of menarche; (3) status quo method - recording whether or not menarche has occurred at the time of investigation (De Wijn, 1966). 47 The strengths and weaknesses of each method are reviewed in Atwood and Taube (1976) and De Wijn (1966). The time of menarche is also valuable as a maturity indicator because of i ts relationship to other developmental milestones, such as peak height velocity. \"Menarche occurs almost invariably after the apex of the height spurt has passed\" (Deming, 1957, cited in Tanner, 1962:39). Greulich ' and Pyle (1970) noted ;this invariable pattern in a study of eight menarcheal age groups, where the maximum annual increment in height occurred.consist-ently during the-year preceeding that in which the menarche took place. Maturity and Population Studies The most conventional method of assessing the rate of maturity of female populations is through the age at menarche (Eveleth & Tanner, 1976). In every population there appears to be early, la te , and average maturing females (Tanner, 1962), and differences in rate of maturation, as measured by age at menarche, exist between national or racial groups i r res-pective of environmental and geographical differences (Eveleth & Tanner, 1976). Ages of menarche of 12.3 years, for middle class European descendants l iv ing in Santiago; 12.8 years, for the United States population in general; 15.1 years, for Asiat ics l iv ing in Maya; and a much later age of menarche of 18.4 years, for females of New Guinea; have been reported (Eveleth & Tanner, 1976). Mean ages at menarche; in six recent (1965-1973) non-athletic, American samples, ranged from 12.20 to 12.65 years (Malina, Harper, Avent, & Campbell, 1973:12). The standard deviation of most samples is about + 2 years (Eveleth & Tanner, 1976). Because there is a re lat ively high correlation between the onset of menarche in mothers and daughters (Damon, 1974; Damon, Damon, Reed, & 48 Valadian, 1969; Tanner, 1962), and between sisters (Damon et a l . , 1969; Tanner, 1962), and a very much higher one between identical twins (Damon et a l . , 1969; Petersen, 1979; Tanner, 1962), there is a genetic implication connected with age of menarche: Distribution of age of menarche in the population is Gaussian . ' . ' .the time of menarche is to a con-siderable extent under hereditary control and depends on the combined actions of genes at several different loci rather than on any single a l l e l e . . . . . t h i s genetic control evidently operates through-out the whole process of growth and the conclusions regarding age at menarche apply equally to rate of development in general. /(Tanner, 1962:114-115) Genetic factors currently account for only 10 to 15% of variation in age at menarche, \"a proportion that is increasing as growing uniformity of nutrition and health eliminates other variables\" (Petersen, 1979:47). Maturity and Athletic Populations The ages of menarche of athlet ic populations have been studied with the purpose of evaluating whether or not e l i te performers, and those involved in physical training programs, di f fer with respect to the average rate of maturity, as defined by the normal population. Erdelyi (1962) and Rarick (1973, cited in Ross, Brown, Faulkner, & Savage, 1976:191) found the age of menarche of young female athletes to be about the same as non-athletes. Furthermore, Erdelyi (1962) concluded that participation in active competitive sports does not disturb the onset of menarche. Astrand, Engstrom, Eriksson, Karl berg, Nylander, Salt-in, and 49 Thoren (1963) found Swedish swimmers to have an ear l ier menarche in relation to Swedish norms, and Bugyi and Kausz (1970) found eight of the best Hungarian swimmers to be advanced in skeletal age, in relation to their chronological age, by approximately 5.1 months. Malina et a l . (1973) found the mean age of menarche of college track and f ie ld athletes (13.58 years) to be signi f icant ly later than that found in a reference population of non-athletes (12.23 years). The ages of menarche of the athlet ic groups did not d i f fer s igni f icant ly among themselves. However, increasing lateness in menarche for these athletes was reported in the following order; shot putters (13.44 years), sprinters (13.54 years), distance runners (13.58 years), discus and javel in throwers (13.60 years), and jumpers and hurdlers (13.73 years). In a later study, Malina, Spirduso, Tate, and Baylor (1978) found Olympic volleyball candidates to have a s igni f icant ly later age of menarche (14.18 years) than high school (13.02 years), and college (13.05 years) athletes, with the high school and college athletes having a signi f icant ly later menarche than a non-athletic sample (12.29 years). In the college athletes, increasing lateness in menarche was reported in the following order: golfers (12.50 years), volleyball players (12.54 years), swimmers (12.84 years), basketball players (12.89 years), gymnastic and track athletes,(13.21 years, n = 6, combined), and tennis players (13.73 years). The Olympic volleyball candidates attained menarche signi f icant ly later than a l l the sport speci f ic groups, except the gymnastic and track, and tennis athletes. Faulkner (1977) reported an age of menarche of 14.0 years for e l i te female figure skaters, and noted that none of the outstanding singles under 12 years had experienced menarche. Ross et a l . (1976) reported ages of menarche of 14.0 years, for e l i te Canadian junior and senior figure 50 skaters; and 12.9 years, for e l i te Canadian alpine racers. In comparison to this latter study, two high school reference samples had ages of menarche of 12.9 and 12.4 years, and a university sample had an age of menarche of 12.9 years. External Stressors and Maturity External \"stressors\", such as climate (Eveleth & Tanner, 1976), season (Bojlen & Bentzon, 1974; Burre l l , Tanner, & Healy, 1961; Hillman, Slater, & Nelson), alt itude (Frisch, 1973; Petersen, 1979), a r t i f i c i a l l ighting (Jafarey, Khan, & Jafarey; McClintock, 1971, cited in Johnston, 1974:167), parental age (Hillman et a l . , 1970; Newton & De Issekutz-Wolsky, 1969), and sleep (Goldfarb, 1977), to name but a few, have been shown to have an effect on the rate of maturity. In comparison to these external stressors, i t has been observed that nutrit ion has a very strong and signif icant influence on the rate of maturity, and especially in influencing the time of adolescence (Goldfarb, 1977; Tanner, 1962): Skeletal maturation is slower everywhere in the worse-off compared to the better-off socio-economic groups. (Tanner et a l . , 1975:19) Tanner et a l . (1962:121) have found that malnutrition in pre-adolescence delays the appearance of the adolescent growth spurt. Dreizen et a l . (1967, cited in Johnston, 1974:165) and Frisch and Revelle (1969) have found that undernutrition delays menarche, while Charzewska, Ziemlanski, and Laseck (1975), Kralj-Cercek (1956), and Skerlj (1947, cited in Damon et a l . , 1969: 170) have reported lower ages of menarche in populations with high consump-tions of animal proteinous foods. 51 There appear to be many factors influencing the age of menarche, and Kralj-Cercek (1956) presents age of menarche as a mathematical function incorporating environmental, nutr i t ional , s o c i a l , and physique factors. Exercise as a Maturity Stressor The impact of exercise on the body is very dramatic, and al l of i ts effects may not have been ident i f ied. Continuous,\"strenuous'exercise may act on the body and the maturing systems in much the same way as other external stressors, with i ts effect and influence subject to \"cr i t ica l growth periods\" and genetic vulnerabi l i ty . During moderate to intense exercise, growth hormone circulates in the blood stream (1965, U.S. Department of Health Report, cited in Smit, 1973:484). Although this hormone is necessary for normal growth from birth to adulthood, i ts function in adult l i f e is not entirely understood (Tanner, 1978), and furthermore: GH does not act direct ly on the bones to make them grow . . . but on the l iver to stimulate production of another hormone, called somato-medin . . . a smaller molecule, that acts on the growing carti lage cel ls at the ends of bones, and probably on muscle c e l l s , whose growth is also stimulated by GH. (Tanner, 1978:93) Increased concentrations of growth hormone in the blood, as a result of exercise, does not necessarily indicate the potential for an increase in s ize . Differences, between normally large and normally small children and adults, are not caused by differences in GH secretion (Tanner,1978:93): Perhaps i t is the receptors in the carti lage cel ls which control size . . . normal children have plenty of GH and are not turned into normally big ones by being given GH in excess. (Tanner, 1978:93) 52 Two longitudinal studies on the effects of rigorous training on growth were located. Astrand et a l . (1963) studied g i r l swimmers, who had commenced training before thirteen years of age. The growth curves from the 7th to 16th year were normal, and at 18 years a medical examination disclosed no harmful effects of the training. Motajova (1974) studied g i r l gymnasts involved in strenuous training, and a control group of females, from their'11th to 15th year. At the end of the 4th year no signif icant differences, between the groups, in bone age were found. Since the mechanism that triggers menarche may not be the same as that which causes resumption of the menses after exercise, weight loss , or starvation induced amenorrhea (Bil lewicz, Fellowes, & Hytten, 1976), the contribution of exercise to the disruption of the menstrual cycle can not be considered as evidence that exercise disrupts the maturing processes. Growth Rate Intervention and Catch-up Growth The body is continually undergoing change due to the growth, maturing, and ageing processes, which are constantly being influenced by such factors as d iet , exercise, and external stressors in general. The existence of a \"catch-up\" growth phenomenon that occurs when adverse conditions are removed, strongly suggests the existence of a genetically established and predetermined plan or course of growth that is highly resistant to change (Tanner, 1963): When recovery takes place a \"catch-up\" period of growth ensues during which growth may proceed at as much as twice i ts normal rate for the age concerned. (Tanner,1962:133) Furthermore, females show a greater resistance to growth changes caused by environmental stressors than males (Crimson & Turner, 1953, and Greulich, 1951, 53 cited in Tanner, 1962:127). Summary: Performance and Maturity There is substantial evidence that factors, such as nutr i t ion, affect the growth and maturing processes of the female, and interfere with normal development. Whether or not strenuous, prolonged exercise, stresses the body in much the same way, is not known. It has not been established whether the mechanism controll ing the onset of the menses is the same as that controll ing resumption of the menses following exercise and weight loss induced amenorrhea. The effect on growth and maturity, of growth hormone secretions, in.response to exercise, is not known. Skeletal age and age of menarche are two useful maturity indicators often used in population surveys of rate of maturity. Although there is large variat ion, in the population, in the age of menarche, there is a high correlation between the ages of menarche of related females. This suggests that there is a genetic control over this variable. Furthermore, the existence of a \"catch-up\" growth phenomenon suggests the existence of a strong genetic control .over the rate of growth, and possibly maturity, and suggests a control that is highly resistant to external stressors, that may include the stress of exercise. Consistent trends concerning age of menarche and sport participation have been reported, with gymnasts and figure skaters tending to be late maturers, and swimmers tending to be early maturers. The l i terature to date tends to indicate that exercise does not substantially affect the maturing processes. However, this relationship has not been investigated in l ight of the present trend toward younger par t i c i -pation in strenuous training programs. 54 V MORPHOLOGY AND MATURITY While the maturing process per se involves changes in s ize , shape, composition, and proportionality, i t has been speculated that these para-meters are closely related to the in i t ia t ion of the maturing processes. Height, Weight, and Maturity Women who mature early were reported to be less slender at maturity (McNeill & Livson, 1963), with a lower ponderal index (height-to-weight ratio) (Hillman et a l . , 1970) than late maturing women. Children with an early puberty were found to be ta l le r and heavier than late maturers, some time before puberty (McNeill & Livson, 1963). Richey (1937, cited in Tanner, 1962:87) found early maturers to be ta l ler and heavier than late maturers, at ages 6, 7, arid 8 years, and Dupertuis and Michael (1953, cited in Tanner, 1962:102) found that \"as early as two years old those who at puberty will be late maturers weightess ' for ' the i r\" height than do early maturers\". Simi lar ly , Shuttleworth (1939, cited in Tanner, 1962:102) reported this relationship as far back as six years. Acheson and Hewett (1954, cited in Acheson and Dupertuis, 1957: 167) found children who reached a specif ic skeletal maturation stage \"late\", were ta l ler at that stage than children who reached the same stage \"early\". Richey (1937, cited in Tanner, 1962:97) found no difference in height between early, la te , and average maturing females, while Stone and Barker (1937, cited in Tanner, 1962:96) have shown late maturers to be s l ight ly ta l ler at maturity than early maturers, tFrisch'(1969) reported late maturers to be s igni f icant ly ta l le r than early maturers, at the in i t ia t ion of the adolescent growth spurt, and at the maximum rate of growth peak. 55 Frisch and Revelle (1971) reported that the mean height at menarche increased signi f icant ly as the menarcheal age increased. Zacharias, Rand, and Wurtman (1976) found that a g i r l who experiences menarche at an early age, is l ike ly to be shorter than a g i r l who is older when this event occurs. For g i r ls of the same height at menarche, the older ones tended to be l ighter and thinner; and the younger ones tended to be heavier (Johnston, Malina, & Galbraith, 1971; Zacharias et a l . , 1976). At a constant menarcheal weight, the early maturers were found to be shorter than the late maturers (Johnston et a l . , 1971). In general, a late menarche is associated with thinness, and a t a l l e r , l ighter body; and an early menarche is associated with stoutness, and a shorter, heavier body (Johnston et a l . , 1971; Zacharias et a l . , 1976). Linear people, both men and.women, develop la te , . . . [and i t is] l i ke ly that l inear individuals are less advanced in growth at a l l ages, and proceed in a more le isurely way along their (Tanner, 1962:102). Secular Trends An association between height, weight, and rate of maturity, appear to be evident in the secular trends that have been ident i f ied. Analysis of growth data has revealed that the whole process of growth and maturity has been progressively \"stepped-up\", with children getting progressively t a l l e r , and heavier, sooner; and with the growth spurt and time of menarche occurring ear l ier (Bakwin, 1964, Towns et a l . , 1966, Roberts & Dann, 1967, Harper & Co l l ins , 1972, and Zacharias et a l . , 1970, cited in Johnston, 1974:168; Eveleth & Tanner, 1976; Tanner, 1962). Recognizing the existence of racial differences in rate of maturity, i t has been noted that: In Norway the average g i r l begins to menstruate 56 at just over 13 years of age, as opposed to 17 years in the 1840's. In the United States . . . the average age at f i r s t menstruation has de-clined from 14.2 in 1900 to about 12.45 today. (Petersen, 1979:45) On the other hand, the data of Damon et a l . , (1969) suggests that, in a general way, those factors affecting secular trends in adult height are not necessarily the same factors affecting secular trends to age at menarche. The secular trend of an earl ier menarche has also been associated with a number of other variables, such as improved nutr i t ional , environmental, and health care factors (Johnston, 1974; Petersen, 1979; Tanner, 1962), the breaking down of genetic isolates (Broman et a l . , 1942, cited in Tanner, 1962:150), and the occurrence of heterosis (Huise, 1957, and Lasker, 1960, cited in Tanner, 1962:151). Recent investigations have suggested that, in some countries, the downward trend in age of menarche is coming to a halt (Brundtland & Walloe, 1973, and Zacharias et a l . , 1973, cited in Dann & Roberts, 1973:266; Dann & Roberts, 1973). Damon (1974) and Zacharias et a l . (1976) have also reported secular trends of an ear l ier menarche to be no longer observable, and the data of Maresh (1972) indicates the possib i l i ty of a trend in the opposite direct ion. Endomorphy, Mesomorphy, and Maturity \"Relative fatness in children has been associated with an early menarche\" (Ross et a l . , 1976:191), and \"at least in countries where nutr i -tion may be more than adequate, fatness and maturity go together\" (Tanner, 1962:102). Furthermore, early maturers have been found to be s igni f icant ly more endomorphic than late and average maturers (Garn & Haskell , 1960; Frisch & Revelle, 1971; Reynolds, 1950; Rona & Pereira, 1974; Zuk, 1958). 57 Garn and Haskell (1960) found that the fatter the c h i l d , the ta l ler he/she i s , and the more advanced the skeletal age. Fat and developmental progress appear to be l inearly related (Garn & Haskell , I960; Reynolds, 1950; Rona & Pereira, 1974). However, an asymptotic point, beyond which increased fatness is no longer associated with accelerated development, is speculated (Garn & Haskell , 1960). Bruch (1941, cited in Johnston, 1974:161) found c l i n i c a l l y obese children to be as much as 1% years advanced in age of men-arche, and Hammar et a l . (1972, cited in Johnston, 1974:162) found obese g i r ls to reach menarche at an average age of 11.3 years, and non-obese at 12.8 years. Reynolds' comprehensive study (1946, cited in Reynolds, 1950:11) of fat patterning and maturity suggests the existence of developmental differences in subcutaneous fat in early, la te , and average maturing females. Early maturing females were found to have relat ively greater fat measurements in childhood, and prior to puberty, than late and average maturing females. Simi lar ly , Garn and Haskell (1960) found g i r ls who were relat ively fat at 8.5 to 9.5 years, to be advanced in the age of menarche, and in the age of attainment of t ib ia l union. Children between 1% and 12% years, who were one standard deviation above the mean in normalized fat scores, were advanced skeletal ly by approximately .4 years, and were ta l ler than the average for their age, by about 6 months growth. Barker and Stone (1936) and Wallau (1939, both:icited in Tanner, 1962:101) have shown \"pyknic\" (endomorphic) women to experience menarche approximately eight months ahead of \"leptosomic\" (ectomorphic) women. Kraij-Cercek (1956) reported that g i r ls of the \"Baroque\" type (pyknic, broad-bui l t , feminine) reach menarche ear l ier - than-gir ls of the \"renaissance1^ type (medium bui ld) , with the \"gothic\" type (linear or boyish build) having 58 the latest menarche. McNeill and Livson (1963) found that both endomorphy and ectomorphy were related to maturity, while mesomorphy was found to be essential ly unrelated. Johnston (1974) cautions against in terpre t ing \"cause-effect\" relationships between body build and rate of maturity: Increased amounts of depot fat do not \"cause\" menarche to be early . . . Rather these would seem to be the result of other determinants of growth var iat ion, be they hormonal, nutr i -t iona l , hereditary . . . Thus, in a l l probabil-i t y , a late maturing female wi l l tend to have a more linear physique due to a longer period of pre-adolescent growth, characterized by relat ively greater amounts of growth of the shafts of the long bones. (Johnston, 1974:162) The Cr i t ica l Mass Hypotheses Frisch and Revelle (1970) proposed the hypothesis that c r i t i c a l body weights may trigger certain adolescent events in females.. From their analyses of data from three longitudinal growth studies, they interpolated mean heights and weights, at three maturational events; at menarche, at the in i t ia t ion of the weight spurt, and at the time of maximum rate of weight gain, and proposed that these events occurred at \"invariant\" mean weights of 48 kg, 31 kg, and 39 kg, respectively. They proposed that menarche occurs when: attainment of a body weight in the c r i t i ca l range causes a change in metabolic rate, which in turn, reduces the sensi t iv i ty of the hypothalamus to estrogen, thus altering the ovarian-hypothalamus feedback. (Frisch & Revelle, 1970:398) Fr isch, Revelle, and Cook (1973) expanded the hypothesis to include 59 discussion of a c r i t i ca l \"metabolic\" mass component. From the height and weight values previously interpolated (Frisch & Revelle, 1970), they calculated total body water, lean body weight, and fat values, and noted the change in these components, between the maturational events. They proposed that: Total water and lean body weight are more closely correlated with metabolic rate than body weight . . . as is expected since they represent the meta-bolic mass, as a f i r s t approximation. (Frisch et a l . , 1973:479) In 1974, Frisch and McArthur revised the c r i t i ca l mass hypothesis to include c r i t i c a l fat mass values, calculated from total body water, necessary for the onset of the menses, and for the maintenance of menstruation. They proposed that: the minimal weight for body height for the onset of menstrual cycles . . . is equivalent to about 17% fat of body weight. . . . about 22% fat of body weight, indicates a minimal weight for height necessary for the restoration and maintenance of menstrual cycles for women of age 16 years and over. (Frisch -\u00E2\u0080\u00A2& McArthur, 1974:949) In 1976, Frisch presented a nomogram for estimating total body water as a percentage of body weight (an index of fatness) from height and weight values of females. The concept of \"cr i t ica l masses\" is not a new one. In 1923, Moulton (cited in Friss-Hansen, 1971:272; and Parizkova, 1961a:805-806) introduced the concept of \"chemical maturity\", indicating that at a certain age a constant composition of water, protein, and ash content would be achieved. The Frisch-Revelle (1970), Frisch-Revelle-Cook (1973), and the Frisch-McArthur (1974) hypotheses have been tested against other long-itudinal growth data, and have found l i t t l e support and much skeptical 60 cr i t ic ism (Billewicz et a l . , 1976; Cameron, 1976; Crawford & Osier, 1975; Johnston et a l . , 1971; Johnston, Roche, Sche l l , & Wettenhall, 1975). One of the major shortcomings of the hypotheses is that they are based on interpolated and not direct ly observed data. The new data are then treated as i f they are observed values. Furthermore, the process of a l lot t ing a l l g i r ls of a given height and weight the same body water value, and thus the same \"fatness\" value, is questioned (Billewicz et a l . , 1976), as this implies that there is no compositional variation between individuals of similar s i ze . As welT, the invariant mean values specified in the original hypothesis appear to be representative of central tendencies, and applicable to group means, and not to individuals, who demonstrate their own unique patterns of growth (Billewicz et a l . , 1976; Crawford & Osier, 1975; Johnston et a l . , 1975). Summary: Morphology and Maturity It appears that females who are above average for their age, in height, weight, and fat \"content\", are also advanced in maturity, and reach menarche at an ear l ier age than their peers. In general, an early menarche is associated with stoutness, shortness, and a heavy body; while a late menarche is associated with th innessand_a tal ler , '1 ighter .body. The secular trend of an ear l ier menarche has been associated with the secular trend of children getting progressively ta l le r and heavier sooner. The concept of \"cr i t ica l metabolic masses\" triggering certain adolescent events, has not received much support. Although exercise has the effect of modifying the \"physique\" and .composition of the body, i t is not known at this time, whether a modification 61 of the body's physical morphology can, in turn, substantially affect the rate of maturity, which appears to be a predetermined genetic t ra i t . VI MATURITY AND MORPHOLOGY While i t has been demonstrated that early and late maturers possess d is t inct ly different physiques, at the in i t ia t ion of specif ic adolescent events, there is some speculation that the time and rate of maturity may also affect.the ultimate, adult physique, with early maturers possessing different physical character ist ics, as adults, than late maturers. Pract ical ly a l l of our knowledge on morphological changes accomp-anying growth and maturity is substantiated from longitudinal studies. Since most of these studies are concerned primarily with the dramatics surrounding adolescence, they usually delve descriptively into these events, and comment br ie f ly , i f at a l l , on the persistence of certain growth trends into adult-hood. The effect of specif ic growth characteristics and phenomena, upon ultimate, adult physique, has not received much attention in the l i terature. Although early, average, and late maturers pass through the same maturational and growth events, and in relat ively the same order, the characteristics of the adolescent events appear to d i f fer depending on the time of maturity: The ear l ier the spurt occurs, the more intense i t appears to be . . . the ear l ier the menarche, the greater the peak veloci ty; and the ear l ier the menarche, the less time elapses between i t and peak velocity. . . . as a result of the great-er peak the total contribution of the spurt to adult height is also greater in early maturing chiIdren. (Tanner, 1962:94) 62 Height, Weight, and Maturity Richey (1937, cited in Tanner, 1962:96-97) found late maturers to be l ighter in weight than early maturers, both at menarche and at 18 years, with no difference in adult height reported. Stone and Barker (1973) and Shuttleworth (1939, both cited in Tanner, 1962:97) found late maturing females to be s l ight ly ta l ler as adults. Frisch and Revelle (1969) reported that the adult height, at age 18 years, of four maturity groups, with ages of menarche of 11.4, 12.5, 13.5, and 14.4 years, was approximately the same, except for the latest maturing group which was shorter by 8 centimeters. This group showed a mean peak height increase that was lower than the others, and was also found to be l ighter in weight at age 18 years.than the others. Tanner (1962:102) reported no signif icant relationship between age at menarche and adult height, weight, or triceps skinfold thickness. However, Hillman et a l . (1970) and Tanner (unpublished data, as cited in Tanner, 1962:102) reported a strong association between menarcheal age and adult body form, and women with a high ponderal index as adults, had a s igni f icant ly later onset of the menses. Zuk (1958) also found early maturers to be less ectomorphic at age 17 years than late and average maturers. Proportionality, Shape, and Maturity While some differences in physique are the result of variations in the magnitude of the growth spurt and i ts time of i n i t i a t i o n , others, such as the wider shoulders of the male and the wider hips of the female, result at adolescence through dif ferential growth rates, which account for differences in shape and tissue structure. S t i l l other differences, such as the relat4 ively longer male forearm, develop continuously throughout the whole period 63 of growth (Tanner, 1962:40). Bayley (1943b, cited in Tanner, 1962:102) noted that late maturing g i r ls have wider shoulders, as adults, than early maturing g i r l s . Malina (1978, cited in Malina et a l . , 1978:221) found the late maturing g i r l to be character ist ical ly longer-legged and narrower hipped, with a more linear physique and less weight for height, with less relative fatness, than the early maturing female. Tanner (unpublished, cited in Tanner, 1962:102) similar ly confirmed a l inear i ty relationship between height and b i - i l i a c diameter for late maturers. Shuttleworth (1939, cited in Tanner, 1962:97) attributed the greater adult height observed in late maturing females to a greater leg length. These results may indicate that late maturing g i r ls have an adolescent spurt appoximating the boys' spurt, in composition, as well as in time (Tanner, 1962). The difference in height, and the relat ively longer legs of the male, in comparison to the female, is due to a later age of maturity, and more spec i f i ca l l y , to a longer growing period prior to the adolescent spurt (Tanner, 1962): In g i r ls the spurt begins about 2 years ear l ier than in boys . . . and is somewhat smaller in magnitude. . . . the peak height velocity averag-ing about 8 cm. per year [Tn females] . . . . In the immediate preadolescent years, i t is the legs which are growing relat ively fastest of a l l skeletal dimensions . . . and so i f allowed to grow for an extra 2 years before the spurt, the legs become relat ively longer. (Tanner, 1962:1,46) Smit (1973:484) attributed the longer lower limbs in proportion to total height, observed in gymnasts, to their \"reaching a pubertal growth spurt sooner than those from the population\". However, i t is apparent from the above quotation, that this is due, in fact , to a delayed spurt. 64 The degree of androgyny; \"the masculinity of form in the female, or femininity in the male\" (Tanner, 1962:102), has been studied in an attempt to associate i t with certain developmental events, such as the time of in i t ia t ion of the growth spurt, and menarche. Deming (1957, cited in Tanner, 1962:104) did not find g i r ls with growth spurts resembling that of boys, to have a more masculine physique, as adults, than the average. Malina and Zavaleta (1976) investigated the relationship between androgyny scores and age of menarche in female track and f ie ld athletes, and found low and non-significant resul ts . Olympic female swimmers and divers, in a study by Hebbelinck et a l . (1975, cited in Malina & Zavaleta, 1976:444), were found to have androgyny scores in the masculine range. The androgyny scores however, are said to confuse mesomorphy with masculinity, and endomorphy with femininity (Tanner, 1962:104). Endomorphy, Mesomorphy, and Maturity Zuk (1958) studied the changes in physique with maturity, in an attempt to determine the degree of s tab i l i ty of the somatotypes. Mesomorphy and endomorphy in females, were found to increase consistently from 12 to 17 years, and from 17 to 33 years, while ectomorphy declined consistently. In general, the somatotype tended to be fa i r l y stable and consistent from early adolescence through adulthood. The endomorphic component fluctuated randomly from late adolescence to adulthood, but tended to be stable within the adolescent years, and females, as they grew older, tended to take on endomorphy. Boothby et a l . (1952, cited in Tanner, 1962:102) found a clear tendency for those with an early menarche to have more subcutaneous fat at age 18 years than those with a late menarche. However, Zuk (1958) found 65 late maturers to have a higher degree of endomorphy at age 33 years, than early and average maturers. This difference however, was reported as not s ta t i s t i ca l l y s igni f icant . L i t t l e research effort has been directed towards monitoring shape changes from childhood, through adolescence, and into adulthood. The general impression however, is that; \"shape changes far less than size at least from 2 years onwards\" (Tanner, 1962:91). Summary: Maturity and Morphology The time of maturity has been shown to affect the physical characteristics of the adolescent events. Although the l i terature is non-conclusive, the rate of maturity appears to have some effect on the ultimate adult morphology, with late maturers tending, as adults, to have a more linear physique, a higher ponderal index, narrower hips, and longer legs, but not necessarily a ta l le r stature. Differences in adult physique, and especially the longer lower limbs found in late maturers, may result both from differences in the rate of growth, and the time of in i t ia t ion of the growth spurt. Somatotype components appear to be fa i r l y stable from early adolescence throughout adulthood. However, the endomorphic component fluctuates randomly and tends to increase with age. In view of these findings: It would certainly be wrong to leave any impress-ion that the adolescent spurt, whether late or early, causes any radical changes in body bui ld; i t certainly does not. It adds only the f inishing touches to a physique which is recognized years before. (Tanner, 1962:104) 66 VII MATURITY AND PERFORMANCE The process of physical growth and maturity involves dramatic changes in s i ze , shape, composition, and proportionality (Edwards, 1950, 1951; Friis-Hansen, 1971; Garn & Haskell, 1960; Malina, 1974; Parizkova, 1961a; Reynolds, 1950; Skerlj et a l . , 1953; Tanner, 1962). These changes are especially marked at adolescence: Every muscular and skeletal dimension of the body seems to take part in the adolescent growth spurt. Even the head diameters, pract ical ly dormant since a few years after b i r th , accelerate somewhat in most individuals. The cartilages of the wrist grow and ossify more rapidly. The heart grows faster; so also do the abdominal viscera. The re-productive organs in particular enlarge. (Tanner, 1962:10) Since each stage of maturity presents a different and changed physique, there may also exist , concurrent with these physical a l terat ions, changes in physical potential and performance. In view of the signif icant contribution of physical morphology to performance outcomes, i t is not unreasonable to expect performance potential to vary in relation to the stage of maturity reached. Maturity and Muscular Development In males, puberty is accompanied by an increase in muscular development (Tanner, 1965, as cited in Malina, 1969b:22) and strength, and these are attributable to an increase in hormonal secretions (Clarke, 1968, 1973; Jones, 1949, cited in Reynolds, 1950:107). In females, l i t t l e , i f any, strength increase is detected accompanying pubertal development, and although there is an increase in hormone production, i t is not as great as that experienced in males (Church, 1976). 67 Espenchade (1940, 1960, cited in Malina, 1974:127) found that performance in a variety of motor tasks requiring power, speed, a g i l i t y , co-ordination, and balance, reached a plateau in females at approximately 14 years of age, with l i t t l e improvement thereafter. Johnson and Buskirk (1974) have observed similar resul ts , with performance in some tasks showing sl ight declines after the age of 14 years. Whether or not these declines are functions of biological ageing is questionable however, since Kriesel (1977) noted-an obvious unwillingness in g i r ls to perform strength related tasks to the best of their a b i l i t y . Furthermore, the cross-sectional studies of Fleishman (1964) and Hunsicker and Reiff (1966, both cited in Malina, 1974:128) indicate a sl ight but continued improvement in running-and jumping performance of g i r l s , through to 17 years. Hebbelinck and Ross (1972, cited in Ross, McKim, & Wilson, in Taylor, 1976:257) found that in female chi ldren, bone widths deviate in a positive direct ion, and comparable girths in a negative direct ion, from relat ive height values, suggesting that children have greater skeletal tissue relat ive to muscle mass than adults. They proposed that this \"reduced\" proportion of muscle mass was a physiological reason why children are at a disadvantage in relat ive strength and stamina. The theoretical expectancy of weakness with increasing size has recently been shown to be non-applicable to growing chi ldren, as the basic assumption of geometrical s imi lar i ty is not tenable, since children change in shape and composition as they gain in height-(Ross & Marshall, 1979:15): The muscular girths increase with age. Thus we have a relat ive increase in muscle mass with growth which must be taken into account when appraising strength and relat ive strength per-formances in growing children. (Ross & Marshall, 1979:15) 68 Early Maturity and Performance Although there appears to be no strength \"advantage\" in pubertal development, for the female, an ear l ier maturation has been associated with improved performance of 12 year old female swimmers (Kanitz & Bar-Or, 1973, cited in Bar-Or, Zwiren, & Ruskin, 1974:214). Asmussen (1966) discusses changes in anthropometric dimensions accompanying puberty, in relation to aerobic capacity and strength, and presents a credible case for attributing some of the improvements in performance, especially of female swimmers, to an ear l ier maturity. During the adolescent growth spurt there is a f a i r l y regular order in which the various dimensions accelerate: hands and feet reach adult size f i r s t , followed by calf and forearm, then hips and chest width, then by shoulders, and last ly by trunk length and chest depth. (Hebbelinck & Ross, 1974:546) Hebbelinck and Ross (in Nelson & Morehouse, 1974:546) have commented on the \"tempting\" speculation that the sequence of events of maturity, especially with the hands and feet being relat ively large with respect to the body and to the shoulders, may provide a biomechanical advantage for young swimmers, with respect to propulsion and resistance in the water. Late Maturity and Performance While certain advantages have been associated with an early or advanced maturity, there are also advantages associated with a delayed or late maturity: In swimming and perhaps other sports where body weight is supported or gives a biomechanical advantage, ear l ier maturation with a concomitant increase in size may be an advantage. However, 69 a post-adolescent increase in body mass due to an increased\"percentage of fat may be a deterrent to sports requiring force to support or move the whole body mass. Thus in some track events and possibly in gymnastics and figure skating, delayed matura-tion may be somewhat of an advantage. (Ross et a l . , 1976:191) In a longitudinal study of female non-athletes, Espenchade (1940, cited in Malina et a l . , 1973:12) found the better performers, on a test battery of track and f ie ld related items, to be late maturers, both menar-cheally and skeletal ly . Faulkner (1977) studied the physique characteristics of nationally ranked senior figure skaters and outstanding skaters under the age of 12 years, and concluded that the growth characteristics of the younger skaters, including their low fat content, small general s i ze , l ight weight, and their proportional length and width measurements, provided biomechanical advant-ages in the'performance of acrobatic moves: The proportionally wider knee widths and longer feet may benefit s tab i l i t y . . . . The relat ively narrow bitrochanteric and mesosternal widths . . . may be advantageous in performing some motor s k i l l s . (Faulkner, 1977:92-93) Al l of the outstanding skaters under 12 years of age were pre-menarcheal, and the differences in proportional measures between these skaters and the senior skaters were attributed to differences in maturity. Concerning shock absorption and joint structure, i t appears that children are well equipped to absorb forces caused by physical act iv i ty : For their s ize , children have proportionately larger knees, ankles and feet. Thus they have proportionately greater area for weight-bearing stress than adults. In addit ion, from early infancy to the adolescent growth spurt children tend to grow ta l le r more quickly than they add weight: this l inear i ty is an advantage in meet-ing the stress of running. On the other hand 70 young tissue may be more vulnerable to stress-induced injury. (Ross, cited in Taunton, 1979:20) Johnston et a l . (1975) reported that a structural factor affect-ing running performance is the width of the pelv is , with a wider pelvis providing a less ef f ic ient angle for weight bearing and for attachment of the tendons of the quadraceps muscle group. Since there are structural and shape changes occurring in the hip joint with growth and development (Bullough et a l . , 1973; Ralis & McKibbin, 1973); and at puberty, \"Girls have a part icularly large spurt in hip width\" (Tanner, 1962:45), i t is possible that these changes also affect the eff iciency of the running str ide. Maturity and Cr i t ica l Learning Periods Ross and Marshall (1979:15) have noted that skating professionals seem resigned to the dictum that, \" i f a g i r l does not double jump by the time she is twelve, she never w i l l \" , and have suggested that there may exist a \"cr i t ica l pre-pubertal learning period\", where skating s k i l l s can be mastered easier than after puberty, when there may be structural d is -advantages. Faulkner (1977) has also commented that although the e l i t e , adult female figure skater possesses physical factors that can be regarded as \"deterrents\" to performance, these disadvantages are offset by her acquisition of required s k i l l s before these factors came into play. The importance of the pre-pubertal growth period to future potential and athlet ic success has been studied by Bannister (1968) and Ekblom (1970, both cited in Bailey, 1973:5). These studies emphasize the importance of pre-pubertal training in the setting of upper l imits of the functional aerobic capacity. The necessity of building world class athletes 71 by beginning training before the ages of 10, 11, and 12 years is emphasized by Councilman (cited in Bai ley, 1973:5). Age and Olympic Participation From observing the ages of the female Olympic participants, i t appears that they are younger in comparison to the male participants. Hi rata noted that the ages of the female participants of the 1964 Olympic Games ranged from 13 to 35 years, with a mean of 22.8 years, and concluded that: The best period of man's physical function is in his age of 20 to 30 years in males and 17 to 25 years in females. And i t may be concluded that the male's development is completed later than the females and maintained longer. (Hirata, 1966:208) In comparison to other Olympic sports, i t appears that the female gymnasts are among the youngest participants. De Garay et a l . (1974) found the female gymnasts of the 1968 Olympic Games, with ages ranging from 13 to 26 years, and a mean age of 17.8 years, to be_among theyoungest participants of these Games. Novak et a l . (1977) has observed that: While track and f ie ld events are dominated by adult women, figure skating, gymnastics and part icularly swimming seems to have more record holders in the teenage group. (Novak et a l . , 1977:275) Krustev (1977) reported a lowering in the average age of both the participants and the medalists of women's gymnastics, in the Olympic Games. The average age of the participants dropped from 23 years 10 months (range 18 to 38 years), in the 1896 to 1956 Games; to 18 years 2 months (range 14 to 31 years), in the 1976 Olympic Games. The average age of the medalists dropped from 25 years 6 months (range 19 to 31 years), to 19 years 2 months 72 (range 15 to 24 years). An association between age and performance appears to be apparent since, comcomitant with these decreases in age, there is also observed an increase in the display and mastery of complex s k i l l s . Summary: Maturity and Performance The physical changes and developments accompanying puberty may affect performance and performance potential . While improvements in strength related s k i l l s are observed accompanying puberty in males, no such improve-ments are seen in the female, with performance tending to plateau at this time. A late maturity in the female may provide an advantage in sports, such as gymnastics and figure skating, where smallness in general is required to perform complex s k i l l s , and where re la t ive , rather than absolute, strength is displayed. In these sports, the longer pre-pubertal period accompanying delayed maturity is associated with an \"extended c r i t i ca l learning period\", a period when specif ic complex s k i l l s can be mastered with ease, and after which mastery may not be possible. In comparison to the males, the female Olympic participants appear to be younger, and in comparison to the other female athletes, the gymnasts are among the youngest participants. There appears to be an association in female gymnasts between age and performance since, concomitant with increasing displays of complex s k i l l s , there is also observed a decrease in the ages of the participating gymnasts and the medalists of the Olympic Games. 73 CHAPTER 3 METHODS AND PROCEDURES SUBJECTS Sixty-nine Canadian female gymnasts between the ages of 11.5 and 18.0 years participated in the study. Descriptions of the test groups, and subject selection for these groups, are as follows: Group 1 Fifteen of the seventeen designated Canadian National Gymnastics Team Members (1977) were available for testing during an o f f i c ia l Canadian Gymnastics Federation Training Camp. These gymnasts, termed \"National E l i t e \" , are c lass i f ied by the Canadian Gymnastics Federation as Level III or Level IV gymnasts, based on their performance results from a battery of physical fitness and standard gymnastic moves tests. These are Canada's best.gymnasts and comprise the Canadian National Gymnastics Team competing internationally in World and Olympic Games competitions. Group 2 Thirteen of approximately th i r ty - f ive e l ig ib le gymnasts were available for testing. These gymnasts, termed \"Pre-National E l i t e \" , are c lass i f ied by the Canadian Gymnastics Federation as Level I or Level II gymnasts, based on their performance results from a battery of physical fitness and standard gymnastic moves tests. 74 These gymnasts have a very real potential of becoming national team members, and are e l ig ib le for selection for some international tournaments (Bajin, 1978:1). Group 3 Twenty gymnasts between the ages of 11.5 and 18.0 years were randomly selected from the Vancouver and Thunder Bay areas to represent this test group. These gymnasts, termed \"Competitive\", are c lass i f ied by the Canadian Gymnastics Federation as Competitive Level A or Level B gymnasts, and are active participants of cer t i f ied Regional and Provincial Competitions. Some of these gymnasts have attempted the Canadian Gymnastics Federation e l i te tests, but none have met the requirements for the lowest leve l , Level I. Group 4 Twenty-one gymnasts between the ages of 11.5 and 18.0 years were randomly selected from the Vancouver and Thunder Bay areas to represent this test group. These gymnasts, termed \"Recreational\", have never trained for , or competed i n , cer t i f ied Provincial or Regional Level A or Level B competitions. These gymnasts have attended gymnastic classes for at least one year, and for at least one hour per week. PROCEDURES Participants selected for the study were sent an introductory letter explaining the nature of the project, i ts testing concerns, and the subject's expected involvement. The parent or guardian of each subject was then contacted by phone, at which time any concerns were discussed, and an appointment for having the radiographic photo taken was arranged. 75 Prior to the actual test session the project was again explained to the participating gymnasts. Al l radiographic photos were taken by one of two quali f ied technicians at one of two medical inst i tut ions, U.B.C. Medical Institution and Health Centre in Vancouver, or St. Joseph's General Hospital in Thunder Bay. The procedures for hand positioning specified in the Tanner-Whitehouse II Radiographic Atlas (Tanner et a l . , 1975:41) were followed. Al l radiographs were taken in random order. Menarcheal data were collected at the time when the radiograph was taken, by one of two females, the author, or a previously instructed assistant. Al l anthropometric measurements were taken by the same Inter-nationally Cert i f ied Criterion Anthropometrist, aided by a recorder previously instructed in recording procedures. The same measurement tape, ca l ipers , and anthropometers were used for a l l subjects. One of two beam scales validated for precision was used in the attainment of body weight. Group 1 subjects had their anthropometric measurements taken as a group, while a l l other subjects were randomly measured. For this phase of the test ing, a l l subjects were att ired in two piece bathing suits or under-garments . In order to accommodate the gymnasts, the anthropometric measurements were taken at a variety of gymnastic training centres, and over several days. The complete series of anthropometric measurements for each individual subject was taken at a single test session. The anthropometric measurements and the radiograph were not taken on the same day, for a l l subjects. However, complete anthropometric and maturational data for a single subject were collected within seven days. 76 Complete data col lection for the study took place from December 28th, 1977 to January 30th, 1978. MEASUREMENTS The present study consisted of maturational and anthropometric measurements. Maturational Measurements I Skeletal age. One radiographic photo of the lef t hand and wrist was taken for each subject to provide a measure of the overall maturation of the skeleton, as expressed in terms of a skeletal age. II Menarche. The status quo method of data col lection was used, in that i t was determined whether menarche had, or had not occurred. Anthropometric Measurements A total of 35 anthropometric variables were direct ly measured, and 8 anthropometric variables were further derived from these direct ly observed measurements. A l i s t of these anthropometric variables is provided on the following pages, after which a detailed explaination of the variables is presented. A detailed description of the anthropometric measurements, and the procedures, techniques, and landmarks employed in taking them, is provided in this text because there is at this time, no internationally recognized single manual or source referencing these spec i f ics , which often vary from one study to another (Ross, 1980). The following series of anthropometric measurements were taken. Heights and Lengths vertex standing height s i t t ing height acromial height * radial height * styl ion height * dactyl ion height * i l iospinale height trochanterion height symphysion height * t ib ia l height foot length Note: * measurements not used in s ta t is t ica l Breadths, Widths and Depths biacromial breadth b i i1 iocr is ta l breadth transverse chest width anterior-posterior chest depth bi-epicondylar humerus width bi-epicondylar femur width Weight body weight Girths relaxed arm girth flexed arm girth forearm girth wrist girth chest girth waist girth thigh girth cal f girth ankle girth head girth neck girth analysis Skinfolds triceps skinfold subscapular skinfold suprail iac skinfold abdominal skinfold front thigh skinfold medial cal f skinfold 78 The following measurements were derived from the direct ly observed measurements previously l i s ted . Lengths: upper arm, forearm, hand, trunk, thigh. Proportional Masses: fa t , skeleta l , muscle, residual. The landmarks and terminology referred to were those estab-lished by, Brown, Carter, Hebbelinck, Ross, Behnke, J r . , and Savage, in a Leon and Thea Koerner Foundation Study Group (Ross, Brown, Hebbelinck & Faulkner,, in Shephard & Lavallee, 1978:44-49). The landmarks were identi f ied by palpation of the skeleton and were marked with a dermatographic penci l . Care was taken to ensure that movement of the underlying skin did not misplace the actual marking. Prior to taking the actual measurement, the anthropometrist again ident-i f ied the point, ensuring that the landmark was identif ied accurately. I Height and Length Measurements Unless otherwise indicated, the following conditions applied to the height and length measurements: Instruments Al l height and length measurements, except foot length, were made with an adaptable Martin design anthropometer, comprised of four, 50 centimeter long segments, and a horizontal pointed crossbar. (Manufactured by G P M and distributed by Siber-Hegner). 79 Accuracy Measurements were read to the nearest 0.1 centimeter. \" Body Position The subject stood erect, feet together, with toes directed forward, and with the weight equally distributed on both feet. The head and eyes were fixed in the Frankfort plane, and the arms were extended loosely down the sides of the body with the fingers extended,' but not held along the thighs. Technique The anthropometer base was stabi l ized oh the standing surface by the anthropometrist and was located in f ront , and to the right side,/\"of the subject. The anthropometer spine was held vertical and stable by the supporting base and the anthropometrist. The anthropometer crossbar was lowered down the vertical spine until the point of the arm was on the designated landmark. \"The anthropo-meter arm indicated the height from the f loor of each designated landmark\" (Whittingham, 1978:72). All measurements were taken from the right side of the body. I (i) Vertex Standing Height Definition of measurement:, \"Maximum height from the soles of the feet to the vertex when the head is in the Frankfort Plane\" (Whittingham, 1978:61). Landmark: \"Vertex - the most superior point on the s k u l l , in the mid-sagittal plane when the head is held in the Frankfort Plane\" (Ross et 80 a l . , in Shephard & Lavallee, 1978:46). Technique: The anthropometer base was stabil ized on the standing surface by the anthropometrist and was located posterior to the subject. The anthropometer spine was held vert ical and stable by the supporting base and the anthropometrist. The horizontal crossbar of the anthropometer was brought down in the mid-sagittal plane to s i t firmly on the subject's vertex. I ( i i ) Sit t ing Height Definition of measurement: \"The maximum height of the vertex while seated with the head in the Frankfort Plane\" (Whittingham, 1978:64). This measurement is also referred to as stem height. Landmark: The vertex (see Vertex Standing Height) Body Position: The subject sat on a table with her knees bent over the edge at approximately 90 degrees, and with her legs suspended. The subject's hands rested on her thighs. With the weight equally d is -tributed on the buttocks, and with the sacral and thoracic spine resting against the anthropometer, the subject was instructed to \"s i t t a l l \" and was manually assisted to do so. The head was positioned in the Frankfort plane. Technique: The technique was the same as that used in taking the Vertex Standing Height measurement, only with the anthropometer base stabi l ized on the s i t t ing surface. I ( i i i ) Acromial Height Definition of measurement'-: The height from the soles of the feet to the acromial point with the arms relaxed and hanging loosely at the sides. 81 Landmark: \"Acromiale - Acromial point - the point at . the superior and external border of the acromial process when the subject i s standing erect with relaxed arms\" (Ross et a l . , in Shephard & Lava l le , 1978:47). This point was located by palpating along the length of the spine of- the scapula, or by having the subject bend the trunk l a t e r a l l y to relax the de l to id muscles (Whittingham, 1978:74). I ( iv) Radial Height Def in i t ion of measurement: The height from the soles of the feet to the rad ia l po in t , with the arms relaxed and hanging loosely at the s ides. Landmark: \"The point at the upper and la te ra l border of the head of the radius\" (Ross et a l . , in Shephard & Lava l lee, 1978:47). This point was located by palpating downward in the la te ra l dimple at the elbow where the rotat ing head of the radius could be f e l t under the stat ionary condyle of the humerus, espec ia l l y when the subject s l i g h t l y pronated and supinated the forearm. (Whittingham, 1978:75) I (v) Sty ! ion Height Def in i t ion of measurement: The height from the soles of the feet to the s t y l i on point with the arms relaxed and hanging loosely at the s ides. Landmark: \"The most d i s ta l point of the s t y l o i d process of the radius\" (Ross et a l . , in Shephard & Lava l lee , 1978:47). By approaching the area from the d i s ta l aspect of the wr is t j o i n t , th is point was located in the \"anatomical snuff box\" or that t r iangular area formed when the thumb was extended l a t e r a l l y . The area was defined by the raised tendons of the muscles of the abductor po l l i cus longus and extensor po l l i cus brevis with the extensor po l l i cus longus. Palpat ion of the point 82 during passive abduction and adduction of the hand and wrist also aided in identifying the landmark (Ross et a l . , in Shephard & Lavallee, 1978:47). I (vi) Dactyl ion Height Definition of measurement: The height from the soles of the feet to the most distal point of the third d ig i t (middle f inger) , with the arms hanging at the sides of the body and the fingers outstretched downwards (Ross et a l . , in Shephard & Lavallee, 1978:47). Landmark: \"Tip of the middle f inger\", also termed Dactyl ion III (Ross et a l . , in Shephard & Lavallee, 1978:47). I (vi i ) I l iospinale Height: Definition of measurement: The height from the soles of the feet to the i l iosp ina le point. Landmark: I l iospinale - the anterior superior i l i a c spine. The most pronounced t i p , and not the most frontal ly curved site of the cr ista i l i a c a , is the designated landmark (Ross et a l . , in Shephard & Lavallee, 1978:48). The anthropometrist palpated anteriorly along the crest of the ilium following the curve downwards until the landmark, a prominent poster-ior ly directed curve, was detected (Whittingham, 1978:76). I ( v i i i ) Trochanterion Height Definition of measurement: The height from the soles of the feet \u00E2\u0080\u00A2to the trochanterion point. Landmark: \"Trochanterion - the most superior point on the greater trochanter of the femur\" (Ross et a l . , in Shephard & Lavallee, 1978:48). 83 T h i s p o i n t was l o c a t e d by h a v i n g the s u b j e c t s l i g h t l y a b d u c t h i s l e g and move i t f o r w a r d and backward, w h i l e the a n t h r o p o m e t r i s t p a l p a t e d the femur p r o g r e s s i v e l y upwards to d e t e r m i n e the landmark,.on the .uppermost p o i n t o f the g r e a t e r t r o c h a n t e r (Ross e t a l . , i n Shephard & L a v a l l e e , 1978:48). I ( i x ) Symph.ysion H e i g h t D e f i n i t i o n o f measurement: The h e i g h t from the s o l e s o f the f e e t t o the symphysion p o i n t . Landmark: \"The s u p e r i o r b o r d e r o f the symphysis p u b i s a t the m i d s a g i t t a l p l a n e \" (Ross e t a l . , i n Shephard & L a v a l l e e , 1978:47). The landmark i s g e n e r a l l y a t the upper l e v e l o f the p u b i c h a i r zone and i s l o c a t e d by p r o c e e d i n g downward from the n a v e l , p a l p a t i n g the abdominal w a l l w i t h the l e f t thumb u n t i l the boney s u r f a c e o f the p u b i s i s l o c a t e d . The f o r e s i d e o f the s u b j e c t ' s symphysis and g e n i t a l i a a r e i n f e r i o r t o the landmark (Ross e t a l . , i n Shephard & L a v a l l e e , 1978:46). I (x) T i b i a l H e i g h t D e f i n i t i o n o f measurement; The h e i g h t from the s o l e s o f the f e e t t o the t i b i a l externum p o i n t . Landmark: T i b i a l e externum p o i n t above the c a p i t u l u m f i b u l a r e on the l a t e r a l b o r d e r o f the head o f the t i b i a (Ross e t a l . , i n Shephard & L a v a l l e e , 1978:48). The p o i n t i s l o c a t e d a t the c e n t r e o f the t r i a d formed by the e p i c o n d y l a r femur, the e p i c o n d y l a r t i b i a , and the head o f the f i b u l a , and i s i n p r a c t i c a l l y the same t r a n s v e r s e p l a n e as the t i b i a l e i n t e r n u m p o i n t on the m e d i a l b o r d e r o f the head o f the t i b i a (Ross e t a l . , i n Shephard & L a v a l l e e , 1978:48). 84 The upper border of the t ib ia was located by palpating the tendon of the quadraceps muscle at the distal end of the patel la , and location of the landmark was fac i l i ta ted by having the subject flex her knee, while palpation continued at.the frontal border of the ligamentum collaterale (Whittingham, 1978:77). I (xi) Foot Length Definition of measurement: \"The distance from the pternion to the akropodion\" (Whittingham, 1978:90). Landmarks: \"Pternion - the most posterior point on the heel of the foot when the subject is standing\" (Ross et a l . , in Shephard & Laval l e , 1978:49). \"Akropodion - the most anterior point on the toe of the foot when the subject is standing. This may be the f i r s t or second phalange\" (Ross et a l . , in Shephard & Lavallee, 1978:49). Instrument: The upper section of the Martin anthropometer, which is comprised of a double scale of measurements in cm. and mm., was used as a large sl iding caliper with the addition of two straight and pointed crossbars. Accuracy: Measurement was read to the nearest .01 centimeters. Technique: \"The caliper was fixed at the pternion and adjusted to the akropodion\" (Whittingham, 1978:91). The crossbars were applied to the points at approximately halfway along their length. Derived Lengths The following lengths were derived from the direct ly observed height measurements, described previously. 85 I (xi i ) Upper Arm Length Definition and Derivation: The distance from the acromial point to the radial point. Upper Arm Length = Acromial Height - Radial Height. I (xi i i) Forearm Length Definition and Derivation: The distance from the radial point to the styl ion point. Forearm Length = Radial Height - Stylion Height. I (xiv) Hand Length Definition and Derivation: The distance from the styl ion point to the dactyl ion point. Hand Length = Stylion Height - Dactyl ion Height. I (xv) Trunk Length Definition and Derivation: The distance from the acromial point to the symphysion point \"projected\" horizontally. Trunk Length = Acromial Height - Symphysion Height. I (xvi) Thigh Length Definition and Derivation: The distance from the symphysion point to the t ib ia l externum point \"projected\" horizontally. Thigh Length = Symphysion Height - Tibial Height. II Breadth, Width, and Depth Measurements Unless otherwise indicated, the following conditions applied to the breadth, width, and depth measurements: Instruments The biacromial and b i i l i o c r i s t a l breadth, and the transverse 86 chest measurements were made with a large, sl iding anthropometer, an adaptation of the Martin anthropometer. The upper section of the Martin anthropometer, which is divided into a double scale of centimeters and mill imeters, was used with the addition of two pointed crossbars. (Manufactured by G P M and distributed by Siber-Hegner). Anterior-posterior chest depth was measured using the wide spreading ca l ipers , an adaptation of the Martin ca l iper , consisting of a crossbar and two curved olive tipped branches. (Manufactured by G P M and distributed by Siber-Hegner). The bi-epicondylar humerus .and the bi-epicondylar femur width measurements were made with the Mitutoyo vernier ca l ipers , as adapted by Carter in 1975 (Whittingham, 1978:57). The caliper consisted of a cross-bar f i t ted with two branches, each .having a 15 millimeter disc at their end to serve as the contact points. Accuracy Al l measurements were read to the nearest 0.1 centimeter, except the bi-epicondylar humerus and femur widths which were read to the nearest 0.01 centimeter. Body Position As stated in Height and Length Measurements. Technique Since the purpose of these measurements was to measure \"skeletal\" dimensions, the instrument was held firm against the subject during readings in order to reduce the influence of the subcutaneous fat . 87 Al l measurements involving the extremities were taken with the right appendages of the body. II (i) Biacromial Breadth Definition of measurement: \"The maximum distance between the most lateral aspect of the acromion processes\" (Whittingham, 1978:89). Landmarks: The most lateral aspects of the right and le f t acromion processes. These aspects were s l ight ly infer ior to the most superior lateral margin defined as the acromiale (Whittingham, 1978:89). Body Posit ion: As stated, and with the shoulder girdle in a relaxed standing position and not abducted or adducted. Technique: As stated, and with the pointed crossbars of the anthropometer directed upwards from the rear as the horizontal bar was held parallel to the f loor and supported on the anthropometrist's forearms (Whittingham, 1978:89). The crossbars were applied to the body at approx-imately halfway along their length. II ( i i ) B i i l i o c r i s t a l Breadth Definition of measurement: \"The maximum distance between the two i l i a c crests or the i l i o c r i s t a l diameter\" (Whittingham, 1978:89). Landmarks: \" I I iocristale - the outermost lateral point of the cr is ta i l i a c a \" . These landmarks are usually encompassed when obtaining b i i l i o c r i s t a l breadth (Ross et a l . , in Shephard & Lavallee, 1978:48). Technique: As stated, and the pointed crossbars were directed upwards, posteriorly from the front, at an angle of approximately 30 degrees (Whittingham, 1978:90). The horizontal bar was held parallel to the f loor and supported on the anthropometrist's forearms. The crossbars were applied. 88 to the body a t a p p r o x i m a t e l y halfway a l o n g t h e i r l e n g t h . II ( i i i ) T r a n s v e r s e Chest Width D e f i n i t i o n o f measurement: \"The t r a n s v e r s e d i s t a n c e from the most l a t e r a l p o i n t s o f the c h e s t a t the l e v e l o f the f o u r t h r i b \" (Whittingham, 1978:90), taken d u r i n g the pause between normal e x p i r a t i o n and i n s p i r a t i o n . T h i s w i d t h i s a l s o r e f e r r e d to as the \" t r a n s v e r s e b r e a d t h o f the t h o r a x \" ' ( C a r t e r 1976:3). Landmarks: The most l a t e r a l p o i n t s o f the r i b s a t the l e v e l o f the f o u r t h c o s t o s t e r n a l a r t i c u l a t i o n ( C a r t e r , 1 9 7 6 : 3 ) . T e c h n i q u e : The s u b j e c t ' s arms were r a i s e d w h i l e the c a l i p e r was p o s i t i o n e d and then lowered f o r the r e a d i n g . The p o i n t e d c r o s s b a r s were d i r e c t e d downwards a t about 30 d e g r e e s , and the measurement was taken d u r i n g the pause between normal e x p i r a t i o n and i n s p i r a t i o n . The c r o s s b a r s were a p p l i e d to the body a t a p p r o x i m a t e l y h a l f w a y a l o n g t h e i r 1ength. Because d i s t o r t i o n o f the r i b cage by a p p l i e d p r e s s u r e i s l i k e l y t o o c c u r when measuring young c h i l d r e n , s p e c i a l a t t e n t i o n was g i v e n t o a p p l y i n g the c o r r e c t p r e s s u r e and c a r e was taken to ensure t h a t the c r o s s -bar arms d i d not s l i p i n t o an i n t e r c o s t a l space (Whittingham, 1978:90). II ( i v ) A n t e r i o r - P o s t e r i o r Chest Depth D e f i n i t i o n o f measurement: \"The depth o f the c h e s t a t the l e v e l o f the f o u r t h i n t e r c o s t a l a r t i c u l a t i o n a t end t i d a l e x p i r a t i o n \" (Whittingham, 1978:92). Landmark: \" M e s o s t e r n a l e - p o i n t l o c a t e d on the corpus s t e r n i a t the i n t e r s e c t o f the m i d s a g i t t a l p l a n e and the h o r i z o n t a l plane a t the 89 midlevel of the fourth chondrosternal art iculat ion\" (Ross et a l . , in i Shephard & Lavallee, 1978:46). Recognizing that the f i r s t art iculat ion was under the c lav ic le , counting was started at the manubriosternal jo in t , which corresponds to the level of the second costal carti lage (Ross et a l . , in Shephard & Lavallee, 1978). Body Position: As stated for Sitt ing Height. Technique: One of the tips of the curved branches was placed on the landmark, and the other on a spinous process judged to be at the same level as the f i r s t point. The measurement was taken during the pause between normal expiration and inspirat ion. As the force generated in this caliper is magnified, l ight pressure was applied to the instrument (Whittingham, 1978:93). II (v) Bi-epicondylar Humerus Width Definition of measurement: \"The width between the medial and lateral epicondyles of the humerus, with the arm horizontal and forming a right angle with the forearm\" (Carter, 1976:3). Landmark: \"The point on either epicondyle of the distal extremity of the humerus most lateral to the medial plane of the bone\" (Carter, 1975:A-2). Body Posit ion: The right arm was raised forward to approximately the level of the shoulders, and the forearm flexed at the elbow forming a right angle with the upper arm (Carter, 1975:A-3). Technique: Starting several centimeters proximal to the elbow joint the condyles were palpated (Whittingham, 1978:91). 90 \"The discs on the branches of the calipers were applied against the epicondyles in such a manner as to bisect the angle of the joint and to l i e in the same plane as the limb\" (Carter, 1975:A-2). II (vi) Bi-epicondylar Femur Width Definit ion of measurement: The width between the medial and lateral epicondyles of the femur, with the thigh horizontal and forming a right angle with the lower leg. Landmark: \"The point on either epicondyle of the distal extremity of the femur most lateral to the medial plane of the bone\" (Carter, 1975:A-2). Body Posit ion: The right foot was raised and placed f l a t on a bench so that the t ib ia was vertical to the f loor , and at right angles to the femur. Technique: Starting several centimeters proximal to the knee joint the condyles were palpated (Whittingham, 1978:92). Also as stated for Bi-epicondylar Humerus Width. I l l Girth Measurements Unless otherwise indicated, the following conditions applied to the girth measurements: Insruments As cloth f iber tape tends to stretch with use, a l l girth measurements were taken with a narrow and f lex ib le steel tape. The tape used was a Wyteface Tiptop, two meter long tape, #860358, manufactured by Keuffel & Esser. 91 Accuracy Al l measurements were read to the nearest 0.1 centimeter. Body Position As stated for Height and Length Measurements. Technique A crossed tape technique was used such that the zero l ine of the tape was in l ine with the measuring as-pect of the tape. The tape was always in the horiz-ontal plane of the body part being measured. (Whittingham, 1978:79) Light pressure suff ic ient to maintain the tape's posit ion, but so as not to produce indentation of the skin, was applied. Al l measurements involving the extremities were taken with the right appendages of the body. I l l (i) Relaxed Arm Girth Definition of measurement: The circumference of the relaxed upper arm, taken at mid-distance between the acromion and the olecranon processes. Landmark: Halfway between the acromion and the olecranon processes of the right arm. With the elbow sl ight ly f lexed, a tape was placed posteriorly on the acromiale, and extended to the point of the elbow. The mid-distance was marked by a horizontal l ine . A vertical l ine joining the acromiale and the point of the elbow was projected, and a vertical mark along this.: projection was drawn to intersect with the horizontal l ine . 92 III ( i i ) Flexed Arm Girth Definition of measurement..: \"The maximum girth of the flexed (and tense) arm when measured at right angles to the long axis of the humerus \" (Whittingham, 1978:81) . Landmark;: On the flexed bicep muscle. Body Position : With the upper arm horizontal, the forearm was supinated and fu l ly flexed at the elbow. The subject clenched her f i s t and contracted the biceps as strongly as possible (Carter, 1975:A-3). Technique : The tape was passed around the muscle and the region of the muscle explored, with the tape always at right angles to i ts long axis (Carter, 1975:A-3). I l l ( i i i ) Forearm Girth Definition of measurement: Maximum girth of the forearm when measured at right angles to the long axis of the radius, with the forearm extended at the elbow and held in the anatomical position (open palm to the front) (Whittingham, 1978:81). Landmark: In the region of the most muscular part of the forearm. Technique: The tape was passed around the forearm one to two centimeters distal to the elbow, and manipulated to obtain the maximum girth -at right angles to the long axis of the radius (Whittingham, 1978:82). I l l (iv) Wrist Girth Definition of measurement: \"Minimal girth of the wrist when measured at right angles to the long axis of the radius at a point proximal to the styloid processes of the radius and ulna \" (Whittingham, 1978:82). 93 Landmark: Proximal to the styloid processes of the radius and ulna. Body Posit ion: The arm and hand were held in l ine and horizontal. Technique: The tape was passed around the wrist , proximal to the styloid processes of the radius and ulna, and manipulated to obtain the minimal girth measured at right angles to the long axis of the radius (Whittingham, 1978:82). I l l (v) Chest Girth Definition of measurement: \"Girth at the level of the fourth costosternal art iculat ion obtained when the subject was in the end-tidal phase of expiration\" (Whittingham, 1978:82). Landmark: Mesosternale - see Anterior-Posterior Chest Depth. Technique: The measurement was made facing the subject. With the subject's arms raised to the horizontal , the tape was placed around the chest at the level of the landmark. The arms were lowered to hang freely at the sides, and after checking to ensure that the tape was in a horizontal posit ion, the measurement was taken (Whittingham, 1978:83). The measurement was taken during the pause between normal expiration and inspirat ion. I l l (vi) Waist Girth Definition of measurement: \"Girth at the level of the noticeable waist narrowing located approximately halfway between the costal border and the i l i a c crest\" (Whittingham, 1978:83). Landmark: At the narrowest point of the waist. Technique: Light pressure was applied to the tape as i t was manipulated to find the smallest girth measurement in this area. 94 I I I (vi i ) Thigh Girth Definition of measurement: The g i r th , \"one centimeter distal to the horizontal gluteal fold at the lower border of the gluteus maximus\" (Whittingham, 1978:84). Landmark: At the upper thigh, just below the gluteus maximus muscle. Body Posit ion: The subject stood with legs s l ight ly parted to permit passage of the tape around the limb (Whittingham, 1978:84). Weight was equally distributed on both limbs. Technique: The tape was placed perpendicular to the axis of the thigh, with i ts upper border about one centimeter distal to the gluteal fold (Carter, 1976:4). I l l ( v i i i ) Calf Girth Definition of measurement: \"Maximum calf girth when measured at right angles to the long axis of the t ib ia\" (Whittingham, 1978:84). Landmark: On the gastrocnemius muscle. Body Posit ion: The subject stood with her feet approximately 10 centimeters apart, and with weight equally distributed on both limbs. Technique: The tape was passed around the leg , near the top of the cal f muscle, then lowered and manipulated to obtain the greatest girth measured at right angles to the long axis of the t ib ia (Carter, 1975:A-3). I l l (ix) Ankle Girth Definition of measurement: \"Minimal girth of the ankle located proximally to the malleoli\" (Whittingham, 1978:85). 95 Landmark: Proximal to the mal leol i . Body Position: As stated for Calf Girth. Technique: \"The tape was passed around the ankle proximal to the malleoli and manipulated to obtain minimal girth at right angles to the long axis of the t ib ia\" (Whittingham, 1978:85). Due to the e l l i p t i c a l shape of the ankle, visual judgement of \"minimal girth\" was inaccurate, and manipulation of the tape was manditory (Whittingham, 1978:85). I l l (x) Head Girth Definition of measurement: \"The maximum girth of the head, immediately superior to the glabellar point\" (Whittingham, 1978:85). Landmark: Glabellar point (brow ridge). Body Position: As stated for Sitt ing Height. Technique: \"The tape was pulled tight to minimize the contr i -bution of soft tissue and hair to the true boney measurements\" (Whittingham, 1978:85). I l l (xi) Neck Girth Definition of measurement: \"The circumference of the neck taken at a level immediately superior to the larynx\" (Whittingham, 1978:86). Landmark: Immediately superior to the larynx. Body Posit ion: As stated for Sitt ing Height. Technique: \"The tape was kept horizontal to the longitudinal axis of the neck with l i t t l e tension applied to the tape\" (Whittingham, 1978:86). 96 IV Skinfold Thickness Measurements Unless otherwise indicated, the following conditions applied to the skinfold measurements: Instruments Al l skinfold measurements were taken with the Harpenden skinfold caliper (Manufactured by Holtain and distributed by Quinton Instruments). This caliper has a spring which exerts a pract ical ly constant tension of 10 gm/mm2 jaw surface over the measuring range 2 to 40 mm. (Tanner, 1962:241) Accuracy Measurements were read to the nearest 0.1 millimeter. Body Position As stated in Height and Length Measurements. Technique \"The objective is to measure the thickness of a complete double layer of skin and subcutaneous tissue without including any underlying muscle tissue \" (Carter, 1975:A-l). In a l l instances the skinfold was picked up at the landmark. A fold of skin plus underlying fat was grasped between the thumb and three f ingers, with the back of the hand facing the measurer. Keeping the jaws of the calipers at right angles to the body surface, the contact faces of the caliper were placed one centimeter below the lowest f inger, and at the depth of the mid-finger t ip . The trigger of the caliper was released and while the pressure on the fingers was maintained the .measurement was taken. The measurement was taken two seconds after the caliper jaws were applied. The arms of the caliper were always at right angles to the skinfold. 97 (Whittingham, 1978:67) The measurement was repeated i f the calipers began to s l ip during the measurement. Al l measurements were made on the right side of the body, except for the umbilicus measurement (abdominal), which was taken on the le f t side. IV (i) Triceps Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken on the triceps muscle, mid-distance between the acromiale and the point of the elbow. Landmark: As stated for Relaxed Arm Girth. Technique: As stated, and the skinfold ran parallel to the long axis of the arm. IV ( i i ) Subscapular Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken below the scapula. Landmark: \"Just beneath the infer ior angle of the scapula in a direction which was obliquely downwards and outwards\" (Carter, 1976:5). Technique: As stated, and \"The forefinger was placed on the landmark and the thumb picked up the natural fold such that the l ine of the fold ran at approximately 40 degrees to the horizontal\" (Whittingham, 1978:69). IV ( i i i ) Suprailiac Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken just above the i l iosp ina le . 98 Landmark: The skinfold was picked up f ive to seven centimeters superior to the i l iosp ina le , and on an imaginary l ine joining the acromiale and i l iospinale (Whittingham, 1978:70). i l iospinale - - see I l iospinale Height acromiale - - see Acromial Height Technique: As stated, and \"the l ine of the fold sloped downward towards the midline at an angle of approximately 45 degrees\" (Whittingham, 1978:70). IV (iv) Abdominal Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken to the le f t of the umbilicus. Landmark: \"The skinfold was selected five to seven centimeters to the le f t of and level with the umbilicus\" (Whittingham, 1978:70). Technique: As stated, and \"the l ine of the fold was vert ical\" (Whittingham, 1978:70). IV (v) Front Thigh Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken on the anterior of the thigh, mid-distance between the i l i a c fold and the patella (Whittingham, 1978:71). Body Position: As stated for Bi-epicondylar Femur Width, and with the thigh muscle relaxed. Technique: As stated, and \"the l ine of the skinfold was parallel! to the long axis of the thigh\" (Whittingham, 1978:71). - With some subjects, the'.skinfold was part icular ly- .di f f icul t to grasp and the procedure was painful . In these cases, the subject was 99 instructed to support the underside of the thigh with the hands. When the weight of the underthigh was heavy, i t was necessary for the anthro-pometrist to further support the knees, with her knee under, the thigh (Whittingham, 1978:71). IV (vi) Medial Calf Skinfold Definition of measurement: A measurement of a double layer of skin and subcutaneous tissue taken on the medial aspect of the ca l f . Landmark: The measurement was made on the most medial aspect of the ca l f , at the level where the girth was seen to be maximal (Whittingham, 1978:71). Body Position: As stated for Bi-epicondylar Femur Wiidth, and with the cal f muscle relaxed. Technique: \"The l ine of the skinfold was vert ical\" (Whittingham, 1978:72). V Weight and Proportional Mass Measurements V (i) Weight Definition of measurement: The weight of the body taken in minimal clothing. Instruments: Body weight was determined for a l l subjects with either a Detecto-Medic beam scale, or a Horns ful1-capacity beam scale (Both distributed by Industrial Scales Ltd. ) . Accuracy: Body weight was recorded to the nearest 0.1 kilogram. Both scales were calibrated for accuracy, and where applicable, corrections 100 were made to the recorded weight. Corrected weight values were used in a l l s ta t is t ica l t es ts ; Body Posit ion: The subject, wearing minimal clothing stood in the center of the scale's platform. Technique: The locking device was turned to the free posit ion. With the subject standing on the platform, the weight indicators were adjusted until a balance was reached. The weight of each subject was then recorded. V ( i i ) Proportional Fat Mass Definition of measurement: The percent of estimated total body weight, that is fat weight (estimated total body weight is calculated as the sum of the anthropometrically derived fat , skeleta l , muscle, and residual masses). Derivation of measurement: The following direct ly observed variables were used in calculating the proportional fat mass: t r iceps, subscapular, supra i l iac , abdominal, front thigh, and medial cal f skinfolds, and vertex standing height. V ( i i i ) Proportional Skeletal Mass Definition of measurement: The percent of estimated total body weight that is skeletal weight (see Proportional Fat Mass). Derivation of measurement: The following direct ly observed variables were used in calculating the proportional skeletal mass: bi-epicondylar humerus and femur widths, wrist and ankle gir ths, and vertex standing height. 101 V (iv) Proportional Muscle Mass Definition of measurement: The percent of total body weight that is muscle weight (see Proportional Fat Mass). Derivation of measurement: The following directly observed variables were used in calculating the proportional muscle mass: relaxed arm,'.chest, thigh, c a l f , and forearm gir ths; t r iceps, subscapular, front thigh, and medial cal f skinfolds; and vertex standing height. V (v) Proportional Residual Mass Definition of measurement: The percent of total body weight that is residual weight (see Proportional Fat Mass). Derivation of measurement: The following direct ly observed variables were used in calculating the proportional residual mass: biacromial and b i i l i o c r i s t a l breadths, transverse chest width, anterior-posterior chest depth, and vertex standing height. Derived Proportional Masses The fa t , skeleta l , muscle, and residual masses were derived from the direct ly observed anthropometric variables, previously described. The Drinkwater Tactic (Drinkwater & Ross, 1979), which provides an anthropometric fractionation of the body mass into fa t , skeleta l , muscle, and residual mass components, was used to determine these masses, and their subsequent proportional mass contribution to total body weight. Total body weight in these equations is the sum of the derived component weights; the skeleta l , residual , muscle, and fat weights, and is referred to as the \"predicted body weight\". This predicted body weight may vary from the observed and direct ly measured \"true body weight\", with the 102 difference between the two weights, indicating the ab i l i ty of the anthro-SEE pometric components to estimate \"true\" or observed body weight (APPENDIX A) The proportional mass contribution of each mass component was calculated as: the mass in kg. X 100 predicted mass (which is the sum of the 4 mass components). formulae 1 Masses were calculated for each subject, in each group, using the M = -z X S + P 170.181 where: M = the estimated mass (Fat, Skeletal , Muscle, Residual), h = the subject's height, d = 3 for a l l masses. P = a specif ic Phantom value which changes for each mass. S = a specif ic Phantom standard deviation value for the mass being estimated Mass fat skeletal muscle residual P_ 12.13 10.49 25.55 16.41 S 3.25 1.57 2.99 1.90 D. T. Drinkwater and W. D. Ross, \"Anthropometric fractionation of body mass\", in Kinanthropometry II, ed. M. Ostyn, G. Beunen and J . Simons, (Baltimore: University Park Press, 1980), p. 183. 103 z = the mean phantom z value, calculated for each subject, from the selected subset of variables. The subset of variables used in calculating the speci f ic masses, and the Phantom specif ication values used in calculating the z score values, for each individual subject, are: 2 Phantom Specification Values Fat Mass Variables ^ -triceps skinfold 15.4 4.47 subscapular skinfold 17.2 5.07'-suprail iac skinfold 15.4 4.47 abdominal skinfold 25.4 7.78 front thigh skinfold 27.0 8.33 medial cal f skinfold 16.0 4.67 Skeletal Mass Variables bi-epicondylar humerus width 6.48 .35 bi-epicondylar femur width 9.52 .48 wrist girth 16.35 .72 ankle girth 21.71 1.33 Residual Mass Variables biacromial breadth 38.04 1.92 transverse chest width 27.92 1.74 b i i l i o c r i s t a l breadth 28.84 1.75 anterior-posterior chest depth 17.50 1.38 2 I b i d . , p. 181. 104 Phantom Specification Values Muscle Mass Variables p_ \u00C2\u00A3 relaxed arm girth - ( TT X triceps skinfold*) 22.05 1.91 chest girth - ( TT X subscapular skinfold*) 82.36 4.86 thigh girth - ( TT X front thigh skinfold*) 47.33 3.59 calf girth - ( TT X medial calf skinfold*) 30.22 1.97 forearm girth 25.13 1.41 Note: *skinfolds expressed in centimeters The Phantom z values for these variables were calculated from the formula 3: z where: z = the Phantom z value (for the subject) for the variable sought h = the subject's height d = 1 for a l l of these calculations p = the Phantom specif ication value for the variable s = the Phantom standard deviation value for the variable v = the subject's value for the variable being converted to a z score EVALUATION OF MEASUREMENTS Chronological Age Chronological age, for each subject, was calculated in decimal years, from the birthdate and the date when the radiograph was taken. 3 I b i d . , p. 183. 105 Maturational Evaluation All skeletal radiographs were assessed by the cr i ter ion anthro-pometrist. Each radiograph was identi f ied by a randomly assigned number, permanently embedded in the fi lm at the time of exposure. The identi t ies of the radiographs, and the chronological ages of the subjects remained anonymous to the rater. The rater was given a l l radiographs at one time, and the radiographs were assessed in no specif ic order. The Tanner-Whitehouse II Method, the 20 bone specif ic approach for the assessment of the skeletal bones of the wrist and hand, was used exclusively in assigning skeletal ages to the radiographs. The rater was trained and practised in this method. A difference value for Chronological Age minus Skeletal Age was computed for each subject. A positive value indicated a younger skeletal than chronological age, or a \"delayed skeletal age\" in reference to the chronological age. A negative value indicated an older skeletal than chronological age, or an \"advanced skeletal age\" in reference to the chronological age,. The absolute difference value indicated the amount of this deviation in decimal years. A difference value of zero indicated that the chronological and skeletal ages were ident ica l . The s ta t is t ics on menarche were compiled by the author. Anthropometric Evaluation Al l anthropometric derivations were calculated by the author. Proportional mass measurements were calculated in a computer program developed by T. Wood (1980). 106 Mean and standard deviation values: for the maturational and anthropometric variables, for each test group, were obtained using the computer program SIMPLE DATA DESCRIPTION (Halm, 1974). STATISTICAL ANALYSIS Test Groupings The research groups, designated as Group 1, Group 2, Group 3, and Group 4, were combined into the following preplanned orthogonal contrasts, for a l l s ta t is t ica l analyses: Group 1 + Group 2 + Group 3 vs Group 4 Group 1 + Group 2 vs Group 3 Group 1 vs Group 2 Stat ist ical Analysis of Maturational Data I Skeletal age data. A difference value for Chronological Age minus Skeletal Age was calculated for each subject. Since the skeletal age analysis of the Tanner-Whitehouse II system (Tanner et a l . , 1975) considers a skeletal age of 16 years as a fu l ly mature skeleton, subjects assessed at fu l l maturity (that is a skeletal age rating of 16 years), and over the chronological age of 16 years, were given a skeletal age rating equal to their chronological age. This adjustment was necessary in order to eliminate the introduction of art ifactual Chronological Age minus Skeletal Age differences, caused exclusively by the rating system's definit ion of \" fu l l \" maturity. It is recognized however, that this adjustment is a conservative procedure since, hypothetically, a chronologically aged 18 year o ld , with^a fu l l maturity rating of 16 years, and an adjusted skeletal age 107 rating of 18 years, may have \"just\" attained fu l l maturity, and thus also \"just\" attained the skeletal rating of 16 years. Chronological Age minus Skeletal Age values were subjected to an analysis of variance test with application of the preplanned comparisons previously stated, using the computer program MULTIVAR (Finn, 1978). II Menarcheal data. The incidence of menarche values were subjected to a chi-square analysis test with application of the preplanned comparisons previously stated. Stat is t ica l Analysis of Anthropometric Data The 39 anthropometric variables analyzed s t a t i s t i c a l l y , for group differences, were assembled into appropriate subset c lassi f icat ions as follows: Heights and Lengths (M variables) Breadths, Widths, and Depths (6 variables) Girths (11 variables) Skinfolds (6 variables) Weight (1 variable) and Proportional Masses (4 variables). Each c lass i f ica t ion group of variables was subjected to a multivariate and univariate analysis of covariance test , with application of the preplanned comparisons stated previously, using the computer program MULTIVAR (Finn, 1978). The mean chronological ages for the various a b i l i t y groups were not s imi lar , and since i t has been shown that the absolute values for specif ic anthropometric variables are functions of age (Burgess, 1937, cited in Greenberg & Bryan, 1951:163), chronological age was used as a covariate. 108 CHAPTER 4 RESULTS AND DISCUSSION Sixty-nine gymnasts of varying ab i l i ty levels were studied to determine i f s ignif icant differences existed among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts, with respect to maturational and anthropometric characterist ics. Fifteen of the designated National Team Members formed testing Group 1, termed \"National E l i te\" . Complete anthropometric, radiographic, and menarcheal data were obtained on al l f i f teen subjects. Thirteen e l i te gymnasts represented the \"Pre-National El i te\" ab i l i ty group and formed testing Group 2. Complete anthropometric and menarcheal data were obtained on a l l thirteen subjects. Radiographic data were obtained on eight of the subjects. Forty-one non-elite gymnasts of varying gymnastic ab i l i t y were randomly selected to represent the lesser s k i l l e d , \"Competitive\", Group 3; and the \"Recreational\", Group 4, ab i l i ty groups. Complete anthropometric, radiographic, and menarcheal data were obtained on al l subjects in these two groups, with the exception of one radiograph for one subject in Group 4. In this chapter the results and discussions are presented in the following manner: Under the heading RESULTS, reference is made only to those results signif icant at a level of p < .01. Based on these resul ts , reference to a l l hypotheses takes place. 109 Reference to the signif icant maturational differences, that emerge from the orthogonal contrasts, takes place under the sub-classif icat ions of Skeletal Age, and Menarche. Reference to the signif icant anthropometric differences, that emerge from the orthogonal contrasts, takes place under the col lect ive categorial term to which that variable belongs. Discussion of a univariate is dependent on a signif icant multivariate. The results are presented under the following headings: MATURATIONAL ASSESSMENT I Skeletal Age, II Menarche, Summary of Results: Maturational Assessment ANTHROPOMETRIC ASSESSMENT I Height and Length Measures, II Breadth, Width, and Depth Measures, III Girth Measures, .IV Skinfold Thickness Measures, ~-V Weight and Proportional Mass Measures, Summary of Results: Anthropometric Assessment Under the Heading DISCUSSION, reference is made to both signif icant and non-significant resul ts . The discussion proceeds under the following headings: MATURATIONAL ASSESSMENT ANTHROPOMETRIC ASSESSMENT I Height and Length Measures, II Breadth, Width, and Depth Measures, III Girth Measures, IV Skinfold Thickness Measures, V Weight and Proportional Mass Measures MATURITY-ANTHROPOMETRIC RELATIONSHIP 110 RESULTS MATURATIONAL ASSESSMENT I Skeletal Age Hypothesis 1 The maturational status, as determined by skeletal age in reference to chronological age, is s i g n i f i -cantly different among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts. Table 1 presents the observed ce l l means and standard deviations for the chronological age, skeletal age, and the chronological age minus skeletal age difference. Table 2 presents the univariate analysis of variance results for the chronological age minus skeletal age difference, for each preplanned orthogonal contrast. Differences, in the mean chronological age minus skeletal age difference, in the following comparisons were found to be signif icant at p < .001 and p < .0003 respectively: Group 1 + 2 + 3 vs Group 4 Group 1 + 2 vs Group 3 The difference in the mean chronological age minus skeletal age difference between Group 1 and Group 2 was not signif icant (p < .03). The results of this study part ia l ly support the above hypothesis that there are signif icant differences in maturational status among the four groups. Group 1, 2, and 3, taken together, in comparison to Group 4, was found to be s igni f icant ly younger in mean skeletal age in reference to TABLE 1 OBSERVED CELL MEANS AND STANDARD DEVIATIONS FOR CHRONOLOGICAL AGE, SKELETAL AGE, AND THE CHRONOLOGICAL AGE MINUS SKELETAL AGE DIFFERENCE Group 1+2 1+2+3 Chronological Age (decimal years) (n=15) (n=13) (n=20) (n=21) \u00E2\u0080\u00A2' (n=28) (n=48) X 15.1 14.3 15.0 15.3 14.7 14.8 s . d . 1.4 1.3 1.4 1.4 1.4 1.4 Chronological Age For Those Subjects With Skeletal Assessments (decimal years) (n=15) (n=8) (n=20) (n=20) (n=23) (n=43) X 15.1 14.1 15.0 15.2 14.7 14.8 's.\"d: 1.4 1.1 1.4 1.4 1.4 1.4 Skeletal Age (decimal years) _ X 12..9 13.0. 14.4 14.8 12.9 13.6 s . d . .8 .9 1.5 1.6 .8 1.4 Chronological Age Minus Skeletal Age (decimal years) ~ X 2,2 . 1.2 .6 ,3 1.8 1.3 s . d . 1.0 .6 1.2 1.1 1.0 1.2 TABLE 2 UNIVARIATE ANALYSIS OF VARIANCE RESULTS FOR THE CHRONOLOGICAL AGE MINUS SKELETAL AGE DIFFERENCE FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 1 vs 2 Chronological Age Minus Skeletal Age Difference (n=63) F l i .48 \u00E2\u0080\u00A2 P < .001 (n=43) 15.20 .0003 (n=23) 5.07 .03 113 chronological age (1.0 decimal years). Similarly Group 1 and 2, taken together, in comparison to Group 3, was found to be s igni f icant ly younger in mean skeletal age in reference to chronological age (1.2 decimal years). Group 1 was found to be younger than Group 2 in mean skeletal age in reference to chronological age (1.0 decimal years). However, due possibly to the small sample s ize , this difference was not signif icant and thus does not support the stated hypothesis. Thus, the national e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, have a s igni f icant ly younger mean skeletal age in reference to chronological age than do recreational gymnasts. Simi lar ly , national e l i te and pre-national e l i te gymnasts, taken together, have a s igni f icant ly younger mean skeletal age in reference to chronological age, than do lesser sk i l led competitive gymnasts. However, national e l i te gymnasts do not d i f fe r s igni f icant ly in mean skeletal age in reference to chronological age from pre-national e l i te gymnasts. II Menarche Hypothesis 2 The incidence of menarche is s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. Table 3 presents the observed cel l frequencies for the incidence of menarche. Table 4 presents the chi-square analysis results for the incidence of menarche for each preplanned orthogonal contrast. Differences in the mean incidence of menarche in the following comparisons were found to be signif icant at p < .01: TABLE 3 OBSERVED CELL FREQUENCIES FOR THE INCIDENCE OF MENARCHE Group 1 2 3 4 1+2 1+2+3 (n=15) (n=13) (n=20) (n=21) (n=28) (n=48) Frequency of Menarche 2 6 13 19 8 21 Percent of Subjects to Have Experienced Menarche 13.3 46.2 65.0 90.5 28.6 43.8 TABLE 4 CHI-SQUARE ANALYSIS RESULTS FOR THE INCIDENCE OF MENARCHE FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 1 vs 2 Incidence of Menarche x 2 21.65 7.45 3.8 p < .01 0^_1 p > .05 115 Group 1 + 2 + 3 vs Group 4 Group 1 + 2 vs Group 3 The difference in the mean incidence of menarche between Group 1 and Group 2 was not signif icant (p > .05). The results of this study part ia l ly support the above hypothesis that the incidence of menarche is s igni f icant ly different among the four groups. Group 1, 2, and 3, taken together, was found to have a s igni f icant ly smaller mean incidence of menarche than Group 4 (46.7%). Similarly Group 1 and 2, taken together, was found to have a s igni f icant ly smaller mean incidence of menarche than Group 3 (36.4%). Group 1 was found to have a smaller mean incidence of menarche than Group 2 (32.9%). However, due possibly to the small.sample s ize , this d i f f -erence was not signif icant and thus does \"not support'the. stated\u00E2\u0080\u00A2hypothesis. Thus, the national e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, have a s igni f icant ly smaller mean incidence of menarche than do recreational gymnasts. Simi lar ly , national e l i te and pre-national e l i te gymnasts, taken together, have a s igni f icant ly smaller mean incidence of menarche than do lesser sk i l led competitive gymnasts. However, national e l i te gymnasts do not have a s igni f icant ly different mean incidence of menarche than do pre-national e l i te gymnasts. Summary of Results: Maturational Assessment Stat ist ica l analysis of the two maturity indicators, skeletal age and menarche, disclosed that Hypotheses 1 and 2 were part ia l ly supported. More spec i f i ca l l y , the s ta t is t ica l analysis indicated that at a level of significance of p < .01: 116 1. National e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are maturationally delayed, both skeletally and menarcheally. 2. National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser ski l led competitive gymnasts, are maturationally delayed, both skeletal ly and menarcheally. 3. National e l i te gymnasts, in comparison to pre-national e l i te gymnasts, are not maturationally di f ferent , skeletal ly or menarcheally. ANTHROPOMETRIC ASSESSMENT I Height and Length Measures Hypothesis 3 The measures of height and length are signi f icant ly different among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. Table 5 presents the observed cel l means and standard deviations for the eleven height and length measures. Table 6 presents the multivariate and univariate analysis of covariance results for the height and length measures for each preplanned orthogonal contrast, using chronological age as the covariate. Group 1 + 2 + 3 vs Group 4. In the orthogonal contrast Group 1 + 2 + 3 vs Group 4, the highly signif icant multivariate F (p < .001) was due primarily to the trunk length difference (univariate p < .0001) . with some contribution from the si t t ing height difference (univariate TABLE 5 OBSERVED CELL MEANS AND STANDARD DEVIATIONS Variables (cm.) FOR HEIGHT AND LENGTH MEASURES (n=15) 2 (n=13) Group 3 (n=20) 4 (n=21) 1+2 (n=28) 1+2+3 (n=48) Vertex Standing Height X s .d . 154.0 6.8 152.6 8.7 156.4 4.5 157.0 6.2 153.4 7.6 154.6 6.6 Tibial Height X s .d . 40.4 2.7 40.1 2.4 40.6 1.5 40.4 2.1 40.3 2.5 40.4 2.2 I l iospinale Height X s .d . 86.3 4.9 84.9 5.3 86.6 3.2 86.4 4.4 85.6 5.0 86.0 4.3 Trochanterion Height X s .d . 79.0 4.5 78.5 4.4 79.8 2.7 79.5 3.9 78.7 4.4 79.2 3.8 Foot Length X s .d . 23.1 1.1 22.8 1.2 22.9 .9 23.2 .9 22.9 1.2 22.9 1.1 Sitt ing Height X s.d. 80.2 3.3 79.5 4.3 82.5 3.0 83.3 3.3 79.9 3.8 81.0 3.7 TABLE 5 CONTINUED 1 (n=15) Variables (cm.) Upper Arm Length X 30.0 s . d . 1.7 Forearm Length X 21.6 s . d . 1.0 Hand Length X 17.8 s . d . 1.3 Trunk Length X 44.9 s . d . 1.9 Thigh Length X 38.9 s . d . 2.6 Group 2 3 4 1+2 1+2+3 (n=13) (n=20) (n=21) (n=28) (n=48) 29.1 1.8 29.9 1.2 29.6 1.9 29.6 1.8 29.7 1.6 21.2 1.5 21.8 1.2 21.5 1.2 21.4 1.2 21.6 1.2 18.0 1.0 17.7 .9 18.3 1.1 17.9 1.1 17.8 1.1 47.0 3.5 49.2 2.3 50.4 2.7 45.8 2.9 47.3 3.1 35.6 2.6 37.1 2.3 36.8 2.2 37.4 3.1 37.2 2.7 TABLE 6 MULTIVARIATE AND UNIVARIATE ANALYSIS OF COVARIANCE RESULTS FOR HEIGHT AND LENGTH MEASURES FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 Multivariate Univariates Vertex Standing Height Tibial Height I l iospinale Height Trochanterion Height Foot Length Sitt ing Height Upper Arm Length Forearm Length Hand Length Trunk Length Thigh Length P < P < P < P < P < P < P < P < P < P < P < .001 .38 .75 .89 .87 .47 .02 .37 .60 .19 .0001 .28 .006 .14 .77 .57 .44 .91 .01 .57 .31 .37 .0001 .57 1 vs 2 .0003 .91 .98 .75 .84 .72 .82 .34 .75 .34 .001 .002 to 120 p < .02). Al l other height and length differences for this contrast were not signif icant (p's > .19). Group 1 + 2 vs Group 3. In the orthogonal contrast Group 1 + 2 vs Group 3, the highly signif icant multivariate F (p < .006) was due to the trunk length difference (univariate p < .0001) and the s i t t ing height difference (univariate p < .01). Al l other height and length differences for this contrast were not signif icant (p's > .14). Group 1 vs Group 2. In the orthogonal contrast Group 1 vs Group 2, the highly signif icant multivariate F (p < .0003) was due to the trunk length difference (univariate p < .001) and the thigh length difference (univariate p < .002). Al l other height and length differences for this contrast were not signif icant (p's > .34). The results of this study part ia l ly support the above hypothesis that measures of height and length are s igni f icant ly different among the four groups, with primary support due to differences noted in the trunk .\u00E2\u0080\u00A2 length measure, and secondary support due to differences noted in the s i t t ing height and the thigh length measures. National e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are s igni f icant ly shorter in mean trunk length (3.1 cm). National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser sk i l led competitive gymnasts, are s igni f icant ly shorter in mean trunk length (3.4 cm) and s i t t ing height (2.6 cm). National e l i te gymnasts in comparison to pre-national e l i t e gymnasts, are s igni f icant ly shorter in mean trunk length (2.1 cm) and s igni f icant ly longer in mean thigh length (3.3 cm). 121 II Breadth, Width, and Depth Measures Hypothesis 4 The measures of breadth, width, and depth are signi f icant ly different among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts. Table 7 presents the observed cel l means and standard deviations for the six breadth, width, and depth measures. Table 8 presents the multivariate and univariate analysis of covariance results for the breadth, width, and depth measures for each pre-planned orthogonal contrast, using chronological age as the covariate. Group 1 + 2 + 3 vs Group 4. In the orthogonal contrast Group 1 + 2 + 3 vs Group 4, the signif icant multivariate F (p < .01) was due to the bi-epicondylar femur width difference (univariate p < .002). Al l other breadth, width, and depth differences for this contrast were not signif icant (p's > .09). Group 1 + 2 vs Group 3. In the orthogonal contrast Group 1 + 2 vs Group 3, the multivariate F was not signif icant (p < .14), indicating that differences in the breadth, width, and depth measures, for this contrast, were not s igni f icant . It is of interest to note however, that irrespective of the multivariate analysis, univariate analysis (univariate p < .007) shows the b i i l i o c r i s t a l breadth difference to be signif icant for this contrast. Group 1 vs Group 2. In the orthogonal contrast Group 1 vs Group 2, the highly signif icant multivariate F (p < .005) was due to the anterior-posterior chest depth difference (univariate p < .0008). Al l TABLE 7 OBSERVED CELL MEANS AND STANDARD DEVIATIONS FOR BREADTH, WIDTH, AND DEPTH MEASURES Variables (cm.) (n=15) Group 2 3 4 1+2 1+2+3 (n=13) (n=20) (n=21) (n=28) (n=48) Biacromial Breadth X s .d . 34.1 1.6 33.4 2.1 34.8 1.5 34.6 1.8 33.8 1.9 34.2 1.8 B i i l i o c r i s t a l Breadth X s .d . 24.0 1.3 23.5 2.1 25.2 1.4 25.3 1.9 23.8 1.7 24.4 1.7 Transverse Chest Width X s .d . 23.0 1.2 22.6 1.5 23.7 1.3 23.4 1.4 22.8 1.4 23.2 1.4 Bi-epicondylar Humerus Width X s .d . 6.1 .4 5.9 .4 6.1 .3 6.1 .3 6.0 .4 6.1 .4 Bi-epicondylar Femur Width X s .d . 8.4 .3 8.3 .5 8.4 .3 8.7 .4 8.3 .4 8.4 .4 Anterior-Posterior Chest Depth X s .d . 14.2 1.0 15.7 1.5 15.5 1.4 15.4 1.2 14.9 1.4 15.2 1.4 ro ro TABLE 8 MULTIVARIATE AND UNIVARIATE ANALYSIS OF COVARIANCE RESULTS FOR BREADTH, WIDTH, AND DEPTH MEASURES FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 Multivariate p < Univariates Biacromial Breadth p < B i i l i o c r i s t a l Breadth p < Transverse Chest Width p < Bi-epicondylar Humerus Width p < Bi-epicondylar Femur Width p < Anterior-Posterior Chest Depth p < .01 .80 .09 .92 .90 ,002 .71 .14 .07 .007 .02 .39 .64 .16 1 vs 2 .005 .66 .98 .85 .27 .66 .0008 124 other breadth, width, and depth differences for this contrast were not signif icant (p's > .27). The results of this study part ia l ly support the above hypothesis that measures of breadth, width, and depth are s igni f icant ly different among the four groups, with primary support due to differences noted in the b i -epicondylar femur width and the anterior-posterior chest depth measures. National e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are s igni f icant ly smaller in mean bi-epicondylar femur width (.3 cm). National e l i te gymnasts in comparison to pre-national e l i te gymnasts, are s igni f icant ly smaller in mean anterior-posterior chest depth (1.5 cm). Although a number of differences, with respect to the breadth, width, and depth measures, were noted between the national e l i te and pre-national e l i te gymnasts, taken together, and the lesser sk i l led competitive gymnasts, none of these differences was signif icant and thus the results of this contrast do not support the stated hypothesis. I l l Girth Measures Hypothesis 5 The measures of girth are s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. Table 9 presents the observed cel l means and standard deviations for the eleven girth measures. Table 10 presents the multivariate and univariate analysis of covariance results for the girth measures for each preplanned orthogonal contrast, using chronological age as the covariate. TABLE 9 Variables (cm.) OBSERVED CELL MEANS AND STANDARD DEVIATIONS FOR GIRTH MEASURES (n=15) 2 (n=13) Group 3 (n=20) 4 (n=21) 1+2 1+2+3 (n=28) (n=48) Relaxed Arm Girth X s .d . 23.4 1.5 23.0 2.3 24.6 ' 2.3 24.5 2.3 23.2 1.9 23.8 2.2 Flexed Arm Girth X s .d . 25.0 1.4 24.7 2.4 26.0 2.1 25.6 2.1 24.8 1.9 25.3 2.1 Forearm Girth X s . d . 22.1 1.1 21.5 1.7 22.7 1.4 22.3 1.2 21.8 1.4 22.2 1.5 Wrist Girth X s .d . 14.5 .6 14.2 .9 14.5 .6 14.4 .6 14.3 .8 14.4 .7 Chest Girth X s .d . 78.3 4.7 78.3 6.1 81.5 5.6 80.5 6.0 78.3 5.3 79.7 5.6 Waist Girth X s . d . 61.1 3.2 62.6 6.5 63.9 4.2 64.9 5.4 61.8 4.9 62.7 4.7 ro TABLE 9 CONTINUED 1 (n=15) Variables (cm.) Thigh Girth X 47.1 s . d . 3.0 Calf Girth X 30.2 s . d . 2.9 Ankle Girth X 19.2 s . d . 1.1 Head Girth X\" 53.5 s . d . .8 Neck Girth X 29.8 s . d . 1.0 Group 2 3 4 1+2 1+2+3 (n=13) (n=20) (n=21) (n=28) (n=48) 47.1 4.3 50.8 5.1 51.9 4.4 47.1 3.6 48.7 4.6 31.3 3.1 31.7 2.0 32.6 1.8 30.7 3.0 31.1 2.7 19.4 1.5 19.7 1.1 20.0 .9 19.3 1.3 19.5 1.2 52.7 1.4 53.2 1.3 53.9 1.4 53.1 1.2 53.1 1.2 29.6 2.2 30.6 1.6 30.3 1.6 29.7 1.6 30.1 1.7 TABLE 10 MULTIVARIATE AND UNIVARIATE ANALYSIS OF COVARIANCE RESULTS FOR GIRTH MEASURES FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 1 vs Multivariate P < .0004 .29 .04 Univariates Relaxed Arm Girth P < .42 .04 .92 Flexed Arm Girth P < .99 .06 .77 Forearm Girth P < .95 .04 .44 Wrist Girth P < .64 .52 .51 Chest Girth P < .87 .05 .34 Waist Girth P < .18 .18 .16 Thigh Girth P < .01 .004 .43 Calf Girth P < .05 .20 .07 Ankle Girth P < .10 .28 .49 Head Girth P < .03 .93 .13 Neck Girth P < .91 .08 .94 128 Group 1 + 2 + 3 vs Group 4. In the orthogonal contrast Group 1 + 2 + 3 vs Group 4, the highly signif icant multivariate F (p < .0004) was due primarily to the thigh girth difference (univariate p < .01) with some contribution from the head girth difference (univariate p < .03). Al l other girth differences for this contrast were not signif icant (p's > .05). Group 1 + 2 vs Group 3. In the orthogonal contrast Group 1 + 2 vs Group 3, the multivariate F was not signif icant (p < .29), indicating . that differences in the girth measures, for this contrast, were not signi f icant . It is of interest to note however, that irrespective of the multivariate analysis, univariate analysis (univariate p < .004) shows the thigh girth difference to be signif icant for this contrast. Group 1 vs Group 2. In the orthogonal contrast Group 1 ' vs Group 2, the multivariate F was not signif icant (p < .04), indicating that differences in the girth measures, for this contrast, were not s igni f icant . The results of this study part ia l ly support the above hypothesis that measures of girth are signi f icant ly different among the four groups, with primary support due to the difference noted in the thigh girth measure. National e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are signi f icant ly smaller in thigh girth (3.2 cm). Although a number of differences, with respect to the girth meas-ures, were noted between national e l i te and pre-national e l i te gymnasts, taken together, and lesser ski l led competitive gymnasts; and between national e l i te a/.nd pre-national ..elite gymnasts, none of these\" differences was signif icant and thus the results of these contrasts do not support.the stated hypothesis. 129 IV Skinfold Thickness Measures Hypothesis 6 The measures of skinfold thickness are signi f icant ly different among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts. Table 11 presents the observed cel l means and standard deviations for the six skinfold thickness measures. Table 12 presents the multivariate and univariate analysis of covariance results for the skinfold thickness measures for each preplanned orthogonal contrast, using chronological age as the covariate. Group 1 + 2 + 3 vs Group 4. In the orthogonal contrast Group 1 + 2 + 3 vs Group 4, the highly signif icant multivariate F (p < .0001) was due to a l l of the skinfold thickness differences; the triceps (univariate p < .0001), subscapular (univariate p < .001), suprail iac (univariate p < .009), abdominal (univariate p < .0002), front thigh (univariate p < .0001), and medial cal f (univariate p < .0001). Group 1 + 2 vs Group 3. In the orthogonal contrast Group 1 + 2 vs Group 3, the signif icant multivariate F (p < .01) was due primarily to the triceps (univariate p < .001), suprail iac (univariate p < .009), abdominal (univariate p < .003), front thigh (univariate p < .004), and medial cal f (univariate p < .002) skinfold thickness differences, with some contribution from the subscapular skinfold thickness difference (univariate p < .02). Group 1 vs Group 2. In the orthogonal contrast Group 1 vs Group 2, the multivariate F was not signif icant (p < .98), indicating that differences in the skinfold measures, for this contrast, were not s igni f icant . TABLE 11 Variables (mm.) OBSERVED CELL MEANS AND STANDARD DEVIATIONS FOR SKINFOLD THICKNESS MEASURES 1 (n=15) 2 (n=13) Group 3 (n=20) 4 (n=21) 1+2 (n=28) 1+2+3 (n=48) Triceps Skinfold X s .d . 7.0 1.5 6.7 1.1 9.3 1.9 11.5 3.5 6.9 1.4 7.9 2.0 Subscapular Skinfold X s .d . 5.5 1.1 5.8 1.2 8.0 2.4 9.9 5.5 6.6 2.1 Suprailiac Skinfold X s .d . 4.0 1.2 4.0 1.0 6.6 3.9 7.5 4.2 5.1 2.9 Abdominal Skinfold X s .d . 5.0 1.3 4.9 1.2 9.0 4.9 11.7 6.1 5.0 1.2 6.7 3.9 Front Thigh Skinfold X s .d . 11.2 3.2 11.4 3.3 15.5 4.6 19.3 6.0 11.3 3.2 13.1 4.3 Medial Calf Skinfold X s . d . 6.4 1.9 6.2 1.6 8.9 1.9 10.9 4.2 6.3 1.7 7.4 2.2 GO o TABLE 12 MULTIVARIATE AND UNIVARIATE ANALYSIS OF COVARIANCE RESULTS FOR SKINFOLD THICKNESS MEASURES FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 Multivariate Univariates Triceps Skinfold Subscapular Skinfold Suprailiac Skinfold Abdominal Skinfold Front Thigh Skinfold Medial Calf Skinfold P < P < P < P < P < P < P < .0001 .0001 .001 .009 .0002 .0001 .0001 .01 ,001 .02 ,009 ,003 .004 .002 1 vs 2 .98 .94 .65 .71 .76 .65 .89 00 132 The results of this study part ia l ly support the above hypothesis that measures of skinfold thickness are signif icant ly different among the four groups, with primary support due to differences noted in the t r iceps, suprai l iac, abdominal, front thigh, and medial cal f skinfold thickness measures, and secondary support due to differences noted in the subscapular skinfold thickness measure. National e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are s igni f icant ly smaller in mean triceps (3.6 cm), subscapular (3.3 cm), suprail iac (2.4 cm), abdominal (5.0 cm), front thigh (6.2 cm), and medial cal f (3.5 cm) skinfold thickness measures. National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser sk i l led competitive gymnasts, are s igni f icant ly smaller in mean triceps (2.4 cm), suprail iac (2.6 cm), abdominal (4.0 cm), front thigh (4.2 cm), and medial cal f (2.6 cm) skinfold thickness measures. Although a number of differences, with respect to the skinfold thickness measures, were noted between national e l i te and pre-national e l i te gymnasts, none of these differences was signif icant and thus the results of this contrast do not support the stated hypothesis. V Weight and Proportional Mass Measures Hypothesis 7 The measures of weight and proportional mass are s igni f icant ly different among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts. Table 13 presents the observed cel l means and standard deviations for the single weight and the four proportional mass measures. TABLE 13 OBSERVED CELL MEANS AND STANDARD DEVIATIONS FOR WEIGHT AND PROPORTIONAL MASS MEASURES Group 1 2 . 3 4 1+2 1+2+3 (n=15) (n=13) (n=20) (n=21) (n=28) (n=48) Variables Body Weight (kg.) X 43.7 44.1 48.7 50.0 43.9 45.9 s . d . 5.3 8.0 6.7 7.8 6.6 7.0 Proportional Fat Mass * ' X 9.5 9.4 11.8 13.9 9.4 ' 10.4 s . d . 1.1 .9 1.6 3.2 1.0 1.8 Proportional Skeletal Mass * X 18.3 17.7 17.0 17.0 18.0 17.6 s . d . 1.5 1.2 1.5 1.5 1.4 1.5 Proportional Muscle Mass * J 46.9 46.9 45.7 44.2 46.9 46.4 s .d . 1.6 1.5 1.6 1.8 1.5 1.6 Proportional Residual Mass * X 25.6 26.1 25.6 25.0 25.8 25.7 s .d . 1.2 1.4 1.3 1.3 1.3 1.3 * Note: Masses expressed in percent TABLE 14 MULTIVARIATE AND UNIVARIATE ANALYSIS OF COVARIANCE RESULTS FOR WEIGHT AND PROPORTIONAL MASS MEASURES FOR EACH PREPLANNED ORTHOGONAL CONTRAST Orthogonal Contrasts 1+2+3 vs 4 1+2 vs 3 1 vs 2 Multivariate p < .0001 .006 .37 Univariates Body Weight p < .07 .02 .32 Proportional Fat Mass p < .0001 .0003 .87 Proportional Skeletal Mass p < .35 .01 .04 Proportional Muscle Mass p < .0001 .009 .82 Proportional Residual Mass p < .04 .48 .35 CO -F=. 135 Table 14 presents the multivariate and univariate analysis of covariance results for the weight and proportional mass measures for each preplanned orthogonal contrast, using chronological age as the covariate. Group 1 + 2 + 3 vs Group 4. In the orthogonal contrast Group 1 + 2 + 3 vs Group 4, the highly signif icant multivariate F (p < .0001) was due primarily to the proportional fat (univariate p < .0001) and muscle mass (univariate p < .0001) differences, with some contribution from the proportional residual mass difference (univariate p < .04). Body weight and proportional skeletal mass differences for this contrast were not signif icant (p's > .07). Group 1 + 2 vs Group 3. In the. orthogonal contrast Group 1 + 2 vs Group 3, the highly signif icant multivariate F (p < .006) was due primarily to the proportional fat (univariate p < .0003), skeletal (univariate p < .01), and muscle (univariate p < .009) mass differences, with some contribution from the body weight difference (univariate p < .02). The proportional residual mass difference for this contrast was not signif icant (univariate p < .48). Group 1 vs Group 2. In the orthogonal contrast Group 1 vs Group 2, the multivariate F was not signif icant (p < .37) indicating that differences in the body weight and the proportional mass measures, for this contrast, were not s igni f icant . The results of this study part ia l ly support the above hypothesis that measures of weight and proportional mass are signi f icant ly different among the four groups, with primary support due to differences noted in the proportional skeletal mass measure. National e l i t e , pre-national e l i t e , and lesser ski l led competitive 136 gymnasts, taken together, in comparison to recreational gymnasts, are signi f icant ly smaller in mean proportional fat mass (3.5%) and signi f icant ly larger in mean proportional muscle mass (2.2%). National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser sk i l led competitive gymnasts, are s igni f icant ly larger in mean proportional muscle (1.2%) and skeletal (1.0%) masses; and s igni f i - i cantly smaller in mean proportional fat mass (2.4%). Although a number of differences, with respect to the weight and proportional mass measures, were noted between national e l i te and pre-national e l i te gymnasts, none of these differences was signif icant and thus the results of this contrast do not support the stated hypothesis. Summary of Results: Anthropometric Assessment Stat ist ica l analysis of the anthropometric measures of height and length; breadth, width, and depth; g i r th ; skinfold; and weight and proportional mass disclosed that Hypotheses 3, 4, 5, 6, and 7 were part ia l ly supported. More spec i f i ca l l y , the s ta t is t ica l analysis indicated that at a level of significance of p < .01: 1. National e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are shorter in mean trunk length, smaller in mean bi-epicondylar -femur width, thigh g i r th , and tr iceps, subscapular, supra i l iac , abdominal, front thigh, and medial cal f skinfolds. They also have a smaller mean proportional fat mass and a larger mean proportional muscle mass .. : than the recreational gymnasts. 137 2. National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser sk i l led competitive gymnasts, are shorter in mean s i t t ing height and trunk length, and smaller in mean t r iceps, suprai l iac , abdominal, front thigh, and medial calf skinfolds. They also have a smaller mean proportional fat mass, and larger mean proportional muscle and skeletal masses, than the lesser sk i l led competitive gymnasts. 3. National e l i te gymnasts, in comparison to pre-national e l i te gymnasts, are shorter in mean trunk length, longer in mean thigh length, and smaller in anterior-posterior chest depth. DISCUSSION MATURATIONAL ASSESSMENT Maturational differences among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts were expected. Those females involved in gymnastics and gymnastic-related sports, such as figure skating, in comparison to normal reference (Faulkner, 1977; Ross et a l . , in Taylor, 1976:257) and athlet ic samples (Malina et a l . , 1973), have consis-tently been assessed as late maturers based on their ages of menarche. Signif icant maturational differences among female gymnasts of varying ab i l i ty levels were found in the present study. These differences indicated that there may be a relationship between gymnastic ab i l i ty and maturity, with highly sk i l led gymnasts being developmentally less mature than lesser sk i l led gymnasts. The fact that signif icant maturational differences were not found between national e l i te and pre-national e l i te gymnasts, possibly serves to further support a maturity-abil i ty relationship 138 since these two groups are considered to be re lat ively close in a b i l i t y . Some of the gymnasts who passed Level II and a l l the gymnasts of Level III, [Level III gymnasts referred to are categorized as pre-national e l i te gymnasts] i f they also show very good results at selection meets, could be e l ig ib le for selection for some international tournaments and dual meets' for junior and senior levels [junior and senior levels referred to, are categorized as national el i te gymnasts]. (Bajin, 1978:1) However, i t should also be noted that the Chronological Age minus Skeletal Age difference, when compared between Group 1 and Group 2 was approaching signif icance, with the former group being developmentally less mature than the latter group. Furthermore, the maturity indicators suggest a consistent trend concerning maturational status and success in gymnastics, such that \u00E2\u0080\u00A2 accompanying progressive increases in gymnastic ab i l i t y are progressive decreases in maturational status (Chronological Age minus Skeletal Age differences increase such that; Group 4 < Group 3 < Group 2 < Group 1; and also Group 1 + 2 + 3 < Group 1 + 2 . The frequency (%) of the incidence of menarche decreases such that; Group 4 > Group 3 > Group 2 > Group 1; and also Group 1 + 2 + 3 > Group 1 + 2 ) . From these observations, i t can be hypothesized that; \"those female gymnasts who are late maturers, are more apt, than average or early maturing female gymnasts, to progress to higher performance levels\" . It is not unreasonable to expect that maturity is related both direct ly and indirect ly to success in gymnastics since: 1. Some of the major physique characteristics of female gymnasts that have reached a high degree of success, a re lat ively short stature and l ight weight; small skinfold thickness, g i r th , b icr ista l breadth, and trunk length measurements; re lat ively long legs, and low percent body fat 139 (De Garay et a l . , 1974; Draz i l , 1971; Grossfeld, in Nichols, 1979:18; Hirata, 1966; Nelson, 1974; Novak et a l . , 1973, 1977; Pool et a l . , 1969; Sinning & Lindberg, 1972; Smit, 1973; Sprynarova & Parizkova, 1979; Youngren, 1969) are also present in the young pre-pubescent and late maturing female (Tanner, 1962). r The late maturing g i r l is character ist ical ly longer-legged and narrower hipped, has a more l inear physique, has less weight for height, and has less relat ive fatness than her early maturing peers. (Malina et a l . , 1978:221) The ultimate adult physique of the late maturing female appears also to resemble that physique associated with success at e l i te gymnastic levels . Although the l i terature is non-conclusive, \"late\" maturing females, as adults, are reported to be shorter than average (Frisch & Revelle, 1969) or of average height (Tanner, 1962:96), to possess a higher ponderal index (Hillman et a l . , 1970), to be more ectomorphic (Zuk, 1958), have wider shoulders (Bayley, 1943b, cited in Tanner, 1962:102), and a smaller b i - i l i a c diameter in relation to height (Tanner, unpublished, cited in lanner, 1962:102) than average and late maturing females. A delayed maturity with i ts concomitantly longer pre-pubescent period has been associated with an extended c r i t i c a l learning period in female figure skaters (Faulkner, 1977; Ross & Marshall, 1979). Whether or not c r i t i c a l learning periods per se, do exist in gymnastics and other gymnastic-related sports, such as figure skating, is not known. It appears reasonable however, to suggest that a physique free of pubertal character ist ics, such as increased weight, fa t , and overall dimensions, (Tanner, 1962:10) would be an advantage in learning and mastering specif ic s k i l l s . 140 4. The l i terature tends to indicate that puberty in females offers no \"strength advantage\", as i t does in males (Church, 1976) and further-more,' is associated with a plateau or decrease in performance (Espenchade, 1940, 1960, cited in Malina, 1974:127; Johnson & Buskirk, 1974). This occurrence however, may well be a social or l i f e - s t y l e related phenomenon rather than a result of biological maturity and ageing, since an obvious unwillingness in females to \"perform\" has been noted (Kriesel , 1977), and the cross-sectional studies of Fleishman (1964) and Hunsicker and Reiff (1966, both cited in Malina, 1974:128) indicate a sl ight but continued improvement in performance through to 17 years. Furthermore, Pool et al . (1969:336) reported that strength>rdid not correlate with performance in female gymnasts, and concluded that; \"this measure lias l ike ly no meaning for selection or training of top gymnasts\". This result is not surprizing in view of the fact that gymnastic-activity for females involves \"high.relat ive, rather than absolute strength\" (Ross & Marshall, 1979:13). Thus, any strength in -crease that puberty may bring, without an associated increase in re la -tive strength, would probably affect gymnastic performance very l i t t l e . 5. Regular, strenuous exercise has a potentially great influence on modifying the body, compositionally, physiological ly, and dimensionally (Astrand & Rodahl, 1970; Behnke & Wilmore, 1974; Brozek, 1961; Malina, 1969a), and i t has been proposed that exercise carried to excess during or previous to the pubertal years, may affect the maturing processes by \"loading\" an organism already \"stressed\" by pubertal changes (La Cava, 1974). Such a proposal is well founded since other external variables, 141 such as sleep (Goldfarb, 1977) and nutrition (Charzewska et a l . , 1975; Kraij-Cercek, 1956; Tanner et a l . , 1975:19), are known to \"stress\" and affect speci f ic adolescent events. As wel l , \"c r i t i ca l metabolic fat masses\" have been associated with-triggering the onset of menarche (Frisch & McArthur, 1974). In the event that such \"cr i t ica l masses\" do exist , and at the present time there is a lack of evidence to suggest that they do (Billewicz et a l . , 1976; Crawford & Osier, 1975; Johnston et a l . , 1971, 1975), i t can be conjectured that exercise can direct ly affect the maturing processes by altering the composition of the body. Gymnasts at e l i te and competitive ab i l i ty leve ls , involved in regular strenuous training regimes, display the effects of exercise in modifying the body (a low percent fat mass in the presence of a high percent lean body mass), to a greater degree than gymnasts with less involvement. Furthermore, without disregarding genetical associations with the rate of maturity, i t can be expected that those gymnasts involved in more frequent and strenuous training programs, would ref lect maturational changes associated with exercise to a greater extent, than gymnasts with less strenuous and regular involvement. If regular., strenuous exercise has the effect of altering the \"normal\" rate of maturity, of an indiv idual , then dramatic changes in growth and/or maturity should be \"expected\" upon cessation of ac t iv i ty , in accordance with the \"catch-up\" phenomenon proposed by Tanner (1963). While there is no supportive documentation, a number of gymnasts have reported experiencing menarche following an extended period of rest , following cessation of strenuous training, and accompanying increases in weight. 142 The advantage of a delayed maturity and an extended pre-pubescent period may be a very real and important factor in success for female gymnasts, since there is some speculation that female gymnasts from the Soviet Union and Eastern Europe may be using a \"brake\" drug to remain petite and l i the , by delaying puberty (Quinn, 1979:50). Dr. Klein, the chief medical examiner at the 1978 World Gymnastic Championships, has reported seeing photos of a leading Soviet gymnast, showing a steady regression of breast development during a four year period. It is suspected that the drug acts on the pituitary gland, and gives the smaller, l ighter females, a higher strength-to-weight ratio allowing them to outperform their competitors (Quinn, 1979: 50). The lowering in age of both the participants and the medalists of women's gymnastics in the Olympicsfrom the 1956 to the 1976 Games (Krustev, 1977), the domination of Olympic gymnastics by females in the teenage group (Novak et a l . , 1977), and the increasing complexity of s k i l l performance, has led to the speculation that the \"younger\" gymnast may possess d ist inct \"advantages\" over the \"older\" performer, by virtue of her unique bui ld. Since a \"performance discriminating factor\", in that the equipment does not offer equal advantages to a l l gymnasts, is recognized as operating in gymnastics (Val l iere , in Salmela, 1976:96), i t follows that those gymnasts possessing physical attributes that \"co-operate\" with the equipment's properties and potentials, are more successful in performing specif ic movements than those gymnasts with physical attributes that do not conform with the equipment's capacit ies. These attributes may well be uncontrollable and unalterable physical expressions as skeletal proportions, s i ze , or \"build\" in general. There is the possib i l i ty then, that as yet unidentified variables-are present in the young female gymnastic competitor, rendering 143 her more \"biomechanically\" equipped to perform gymnastic-type movements than the \"older\" performer, and biasing performance in her favour. Whether or not the pre-pubescent female is better equipped \"biomechanically\" and functionally, to perform complex gymnastic s k i l l s , and whether or not puberty is the \"despoiler of athlet ic maids\" as Cranston (cited in Clark, 1980:6) contends, is at this time only speculative since there is an absence of direct evidence. ANTHROPOMETRIC ASSESSMENT Anthropometric differences among national e l i t e , preTnational e l i t e , lesser sk i l led competitive and recreational gymnasts were expected since: 1. Disparity and inconsistencies in anthropometric descriptions of gymnasts have been attributed, in part, to ab i l i ty differences (Carter et a l . , 1971). 2. Specif ic anthropometric characteristics have been shown to present biomechanical advantages (Hebbelinck & Ross, in Nelson & Morehouse, 1974:546; Khosla, 1968, 1977; Lewis, 1969), and to affect performance outcomes of gymnastic s k i l l s (George, in Salmela, 1976:96; Le Veau et a l . , 1974; Nelson, 1974; Ross & Marshall, 1979; Rozin, 1974; Salmela, 1976; Va l l ie re , in Salmela, 1976:96). These \"advantageous\" characteristics were expected to be present more consistently in the higher sk i l led than in the lesser ski l led gymnasts. However, i t was also considered that while speci f ic physical characteristics would be an advantage to performance in one event, they may serve as deterrents 144 to success in another (Le Veau et a l . , 1974; Rozin, 1974; Youngren, 1969). Therefore, i t was recognized that the best body-type for gymnasts competing in a l l events, would be a compromise of the ideal morphological determinants best suited for a one-event spec ia l is t , as suggested by Salmela et a l . (1976:169). In this respect, i t was expected that \"dramatic differences\" among the ab i l i ty groups would not be displayed, since the better gymnasts, competing in a l l events, would have physique characteristics of a compromising nature. 3. Exercise has the effect of modifying specif ic physical parameters, such as skinfold thickness values, and muscle girth dimensions (Brozek, 1961; Malina, 1969a), and i t was expected that highly sk i l led gymnasts involved in frequent, vigorous training would display exercise induced changes to a greater extent than lesser sk i l led gymnasts. 4. Maturational differences were expected among the ab i l i ty groups, and i t was expected that these differences would be represented anthropometric-a l l y , in those physical parameters, such as b i i l i o c r i s t a l breadth, height, weight, and percent body fa t , undergoing changes associated with maturity (Parizkova, 1959, 1961b; Reynolds, 1950; Tanner, 1962:45). Because of the relative closeness in ab i l i ty of the national e l i te and the pre-national e l i te gymnasts (Bajin, 1978), and because i t has repeatedly been shown that the higher the ab i l i ty level the narrower the var iabi l i ty of physique (Carter, 1970; Montpetit, in Salmela, 1976:183), few signif icant differences between these groups, in those anthropometric variables related to success, were expected. Differences between these groups were expected however, since anthropometric differences between 145 higher and lesser placing gymnasts at e l i te competitions have been reported (Montpetit, in Salmela, 1976:183; Pool et a l . , 1969; Youngren, 1969). It was also considered that signif icant differences would possibly indicate those anthropometric variables distinguishing the consistently \"better\" gymnasts, the national e l i t e s , from the \"very good\" gymnasts, the pre-national e l i t es . Trends concerning anthropometric parameters and gymnastic ab i l i ty were not clearly identi f iable in the l i terature , due primarily to an absence of studies concerned direct ly with varying ab i l i ty levels and anthro-pometric parameters. Of the three studies located, involving these two. variables . (Montpetit, in Salmela, 1976:183; Pool et a l . , 1969; Youngren, 1969), a l l were comparisons between the \"winners\" of a competition and the lesser placing participants. Montpetit (in Salmela, 1976:183) compared the top five placing women's gymnastic teams at the 1972 Olympic Games. Although there were no consistent trends among the five teams, in age, height, or weight, the f i r s t and second placing teams were shorter in height and l ighter in weight than the third placing team. Similar ly , Pool et a l . (1969) compared the higher and lower placing gymnasts of the 1967 European Gymnastic Championships, and found the better performers to be younger, shorter, l ighter , and smaller in the skinfold thickness meaurement taken. Youngren (1969) compared the placing and non-placing gymnasts at the 1968 U.S. .Olympic Gymnastic T r i a l s , and also noted that those gymnasts who were younger, shorter, and thinner in skinfold thickness measures, tended to rank higher. In the absence of studies concerned direct ly with varying levels of gymnastic ab i l i ty and anthropometric parameters, trends concerning these 146 two variables were investigated and interpolated through inter-study comparisons of Olympic (De Garay et a l . , 1974; Hirata, 1966; Montpetit, in Salmela, 1976:187; Novak et a l . , 1977; Pool et a l . , 1969; Ross, 1980), \"highly ski l led\" (Carter, 1970; Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972; Smit, 1973; Sprynarova & Parizkova, 1969; Youngren, 1969), and recreational female gymnasts (Medved, 1966; Pool et a l . , 1969). This procedure was followed with some reservation however, since measurement techniques, and the.cr i ter ion for defining ab i l i ty levels , were not consistent from one study to the next. These inconsistencies may well account for anothropometric differences noted among the ab i l i ty groups, as well as for the concealment of any \"true\" and existing differences. Even within the seemingly well defined ab i l i ty group, \"Olympic gymnasts\", there exists varying ab i l i ty levels. The \"Olympic participants\" of one country may well be, merely, of equal ab i l i ty to the \"highly ski l led\" gymnasts of another. The Olympic gymnasts studies by Novak et a l . (1977:276) were a l l from one country that \"did not reach a signif icant place during the competition\". The Olympic gymnasts studied by Montpetit (in Salmela, 1976:183), comprised the \"top f ive teams\" of the 1972 Olympic Games. Furthermore, the ab i l i ty c lassi f icat ions \"highly ski l led\" and \"recreational\" contain much var iabi l i ty in terms of -gymnastic-skill level and a b i l i t y . Age is another factor that may account for anthropometric differences noted among varying ab i l i ty leve ls , and among gymnasts of the same ab i l i t y . Large age differences among the various studies compared may also mask \"true\" anthropometric differences and s imi la r i t ies . There is much var iabi l i ty in mean ages among the gymnastic groups referred to 147 in the inter-study comparisons. The \"Olympic gymnasts\" vary in mean ages from 17.0 to 23.1 years (De Garay et a l . , 17.8 y r s ; Hirata, 22.7 y r s ; Montpetit, 20.2 y r s ; Novak et a l . , 19.0 y r s ; Pool et a l . , 23.1 and 19.2 y rs ; Ross, 17.0 y r s ) , the \"highly sk i l led gymnasts\" from 15.0 to 20.5 years . (Carter, age not given; Nelson, 20.5 y r s ; Novak et a l . , 14.3 y r s ; Pool et a l . , 20.5 y rs ; Sinning & Lindberg, 20.0 y rs ; Smit, 15.0 yrs ; Sprynarova & Parizkova, 17.2 yrs ; Youngren, 18.5 y r s ) , and the \"recreational gymnasts\" from 20.1 to 22.7 years (Medved, 20.1 y r s ; Pool et a l . , 22.7 y rs ) . The gymnasts in the varying ab i l i ty groups from the present study range in mean ages from 14.3 to 15.3 decimal years. I Height and Length Measures Vertex standing height. The l i terature did not indicate a trend towards shortness in stature with increasing gymnastic a b i l i t y , contrary to what was expected. Olympic gymnasts (De Garay et a l . , 1974; Hirata, 1966; Montpetit, in Salmela, 1976:187; Novak et a l . , 1977; Pool et a l . , 1969; Ross, 1980) did not di f fer notably in mean height from highly sk i l led gymnasts (Carter, 1970; Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972; Smit, 1973; Sprynarova & Parizkova, 1969; Youngren, 1969), or from recreational gymnasts (Medved, 1966; Pool et a l . , 1966). However, in comparison to appropriate female reference populations, the gymnasts described by Carter (1970), Medved (1966), Nelson (1974), Sinning and Lindberg (1972), and Smit (1973) were reported to be shorter in stature, with the Sinning and Lindberg gymnasts s igni f icant ly shorter. On the other hand, the highly ski l led gymnasts of Nelson's study (1974) were reported to be of similar height to an appropriate reference population of females. 148 Significant differences in height between national e l i te and pre-national e l i te gymnasts, taken together, and lesser sk i l led competitive gymnasts; and between national e l i te and pre-national e l i te gymnasts, were not expected based on the observation that e l i te and highly sk i l led gymnasts in general are short, and i t appears that a relat ively short stature is a prerequisite for successful participation at these levels . A signif icant difference between national e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, and recreat-ional gymnasts were expected since, a relat ively short stature does not appear to be a prerequisite for participation at recreational leve ls , where s k i l l level is relat ively low and emphasis is generally placed on par t ic i -pation, rather than.on performance. Although Group 1 + 2 + 3 has a shorter mean height than Group 4 (2.6 cm), Group 1 + 2 a shorter mean height than Group 3 (3.0 cm), and Group 1 a ta l le r mean height than Group 2 (1.4 cm), these differences were not s igni f icant . However, they do indicate a trend towards shortness in stature for higher sk i l led in comparison to lesser sk i l led gymnasts. The non-significant difference in height noted between Group 1 and Group 2 was expected due to the relat ive closeness in ab i l i t y of these two groups (Bajin, 1978). This result is similar to results noted in the studies of Pool et a l . (1969) and Youngren (1969), in which higher placing gymnasts were not s igni f icant ly different in height from lower placing gymnasts of similar abi l i ty . . . However, the tendency for higher piacing .gymnasts to be shorter than lower placing gymnasts noted in these studies, was not evident in the present comparison of national and pre-national e l i te gymnasts. The non-significant difference noted between Group 1 + 2 + 3 and Group 4 is not in agreement with the expected resul t , and with the comparison of 149 gymnasts in Pool et a l . ' s study (1969), in which high caliber Dutch gymnasts were found to be s igni f icant ly shorter in stature than lower ca l iber , non-competing Dutch gymnasts. The non-significant difference in the present study may have occurred as a result of club \"attitudes\". In clubs where the recreational level is viewed as a \"base\" for the competitive programs, participants are often pre-selected on shortness in stature, among other variables, for inclusion in the recreational program. Perhaps too, those g i r ls who are relat ively ta l l pursue act iv i t ies other than gymnastics, while relat ively short g i r ls are attracted to recreational gymnastic classes. Because signif icant height differences among the ab i l i ty groups were not found, this does not simultaneously indicate that height is not an important variable for participation and success in gymnastics. A relat ively small stature, coupled with a l ight weight, has been shown to present a biomechanical advantage in performing gymnastic-type movements (Le Veau et a l . , 1974; Nelson, 1974; Ross & Marshall, 1979). In comparison to Olympic gymnasts described by De Garay et a l . (1974), Hirata (1966), Montpetit (in Salmela, 1976:187), Novak et a l . (1977), and Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have a smaller mean height. The non-mongolian Olympic competitors described by Pool et a l . (1969) are also ta l ler in mean height than both the national e l i te and the pre-national e l i te gymnasts, while the mongolian competitors are the only Olympic gymnasts described in the l i terature to be similar in height to both the national e l i te and the pre-national e l i te gymnasts. In comparison to highly sk i l led gymnasts described by Carter (1970), Nelson (1974), Sinning and Lindberg (1972), Smit (1973), Sprynarova and Parizkova (1969), and Youngren (1969), the competitive gymnasts in the 150 present study have a shorter mean height. However, in comparison to the highly sk i l led gymnasts described by Novak et a l . (1973), the competitive gymnasts have a similar mean height. In comparison to the recreational gymnasts described by Medved (1966) and Pool et a l . (1969), the recreational gymnasts in the present study have a shorter mean height. Trunk length. This measurement appears to have been neglected in the l i terature , with only a single study (De Garay et a l . , 1974) including this measure in an anthropometric series of measurements. Speculations concerning trunk length and ab i l i ty were not made although i t was noted in the l i terature that; \"A very long torso . . . wil l get in your way on the uneven bars\" (Grossfeld, cited in Nichols, 1979:18). The trunk length measure was signi f icant ly smaller in Group 1 + 2 +3 in comparison to Group 4, in Group 1 + 2 in comparison to Group 3, and in Group 1 in comparison to Group 2. This consistency indicates a trend towards shortness in trunk length for higher sk i l led gymnasts in comparison to lesser sk i l led gymnasts, and a possible relationship between ab i l i ty and trunk length, with a shorter trunk length presenting an advantage in gymnastics. Perhaps a short trunk length in relation to height may result in a reduced moment of inert ia about the trunk and head in twisting move- . ments. Furthermore, proportional analysis of the trunk length measure, in relation to leg length or height would have maturational implications: About a year separates the peaks of total leg length and trunk length . . . The spurt in height is due more to an increase in length of trunk than length of leg , however, and the ratio of trunk length/leg length always increases during adolescence. (Tanner, 1962:12-13) 151 These results indicate that the trunk length measure is worthy of further analysis, especially with respect to proportional assessments. In comparison to Olympic gymnasts described by De Garay et a l . (1974), both the national e l i te and the pre-national e l i te gymnasts in the present study have a shorter mean trunk length. It should be noted however, that the suprasternal point, used in obtaining this measure, was not the same landmark point as that used in the present study. Sit t ing height. This measure appears to have been neglected in the l i terature with only two studies (Nelson, 1974; Ross, 1980) including this measure in an anthropometric series of measurements. The Olympic gymnasts described by Ross (1980), in comparison to the highly sk i l led gymnasts described by Nelson (1974), have a smaller mean s i t t ing height. Based on this comparison i t was speculated that in the present study higher sk i l led gymnasts would possess a smaller mean s i t t ing height than lesser sk i l led gymnasts. The s i t t ing height measure was signi f icant ly smaller in Group 1 + 2 in comparison to Group 3, and the s i t t ing height difference between Group 1 + 2 + 3 and Group 4 was approaching signif icance, with this former group having a smaller mean than the lat ter . These resul ts , in conjunction with those obtained for trunk length and vertex height, strongly suggest a trend towards shortness in the torso for higher sk i l led in comparison to lesser sk i l led gymnasts. These results also suggest that a short torso in relation to height may be associated with gymnastic a b i l i t y , either direct ly by provid-ing some biomechanical advantage, or ind i rect ly , possibly by presenting less weight in the upper body; assuming that a long torso would weight more than a short. Such a weight distr ibution may have some implications for angular 152 (rotary) motion where \"not only is the weight important but also i ts distr ibution in relation to the axis of rotation\" (Nelson, 1974:46). Further analysis is necessary however, before such relationships can be . established. In comparison to Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have a shorter mean s i t t ing height. In comparison to highly sk i l led gymnasts described by Nelson (1974), the competitive gymnasts in the present study have a shorter mean s i t t ing height. Thigh length. A trend concerning thigh length and gymnastic ab i l i ty was not evident in an inter-study comparison of Olympic (Novak et a l . , 1977; Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974). The gymnasts described by Ross (1980) were given two values for thigh length, each derived using different equations, with a difference of 13.6 cm in thigh length between the two measurements. This difference serves to i l lust ra te that caution should be used in inter-study comparisons of anthropometric measurements, especially with respect to those measurements l ike ly to have varying landmark points, or derivations. The thigh length measure was signi f icant ly longer in Group 1 in comparison to Group 2. Since this was the only signif icant difference to emerge among the contrasts, i t is possible that this is a spurious result . However, since Group 1 + 2 and Group 1 + 2 + 3 , both have \"sl ight ly\" longer mean thigh lengths than Group 3 and Group 4 respectively, there appears to be a \"suggestion\" that highly ski l led gymnasts, in comparison to lesser sk i l led gymnasts tend to have longer thigh lengths. 153 Considering the thigh length resul ts , those of the trunk length, and the tendency for higher sk i l led gymnasts to be shorter in vertex standing height than lesser sk i l led gymnasts, i t appears that there may be proportional length differences among the ab i l i ty groups, especially with respect to the segmental lengths contributing to stature height. Furthermore, a thigh length, proportionately long in relation to total leg length or height, may indicate a developmental stage in the adolescent growth spurt since: Within the leg there is a definite gradient of timing. The foot has i ts rather small acceler-ation about 6 months before the calf and thigh. . . . . The cal f length accelerates a l i t t l e before the thigh. (Tanner, 1962:12) Proportional analysis is necessary however, before these proportionality hypotheses can be tested among the present ab i l i ty groups. These results may indicate that there is a relationship between gymnastic ab i l i ty and thigh length, with a longer thigh length providing a biomechanical advantage to performance. Possibly this variable is one of the few differentiat ing variables in which \"better\" gymnasts, the national e l i t e s , can be distinguished from \"very good\" gymnasts, the pre-national e l i t e s . In comparison to Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have a longer mean thigh length. However, i t should be noted that the thigh length measure in the present study was derived from an equation different from that used by Ross. In comparison to Olympic gymnasts described by Novak et a l . (1977), both/, the national e l i te and the pre-national e l i te gymnasts have a shorter mean thigh length. However, the derivation of 154 the thigh length measure in this study.\"was\"not specified._' \"\" In comparison to the highly sk i l led gymnasts described by Nelson (1974), the competitive gymnasts in the present study have a smaller mean thigh length. However, the derivation of thigh length in this study was not specif ied. Other height and length measures ( t i b i a l , i l iosp ina le , and trochanterion height; foot, upper arm, forearm, and hand length). Trends concerning t ib ia l height (Nelson, 1974; Novak et a l . , 1977; Ross, 1980), forearm length (Nelson, 1974; Novak et a l . , 1977; Ross, 1980), and foot length (Nelson, 1974; Ross, 1980), and gymnastic a b i l i t y , were not evident in an inter-study comparison of Olympic and highly ski l led gymnasts. The upper arm length was longer in Olympic gymnasts (Novak et a l . , 1977; Ross, 1980) than in highly sk i l led gymnasts (Nelson, 1974). The i l iospinale height was shorter in Olympic gymnasts (Ross, 1980) than in highly sk i l led gymnasts (Smit, 1973). These comparisons suggest a tendency for higher sk i l led gymnasts to have a longer upper arm length and a shorter i l iospinale height than lesser sk i l led gymnasts. The hand length and trochanterion height measurements appear to have been neglected in the l i terature , with only a single study including a hand length measure (Ross, 1980), and only a single study including a trochanterion height measurement (Nelson, 1974). There were also a few studies referring to a \"total leg length\" measurement (De Garay et a l . , 1974; Pool et a l . , 1969; Ross, 1980). In two of these studies, in which the leg length measure was derived using the same equation, Olympic gymnasts (Ross, 1980), in comparison to highly sk i l led gymnasts (Pool et a l . , 1969), were found to have a longer mean total leg length. This comparison suggests a 155 tendency for higher sk i l led gymnasts to have a longer leg length than lesser sk i l led gymnasts. Speculations concerning the t i b i a l , i l iosp ina le , and trochanterion height, the foot, forearm, and hand length measurements, and gymnastic ab i l i ty in the present study were not made. However, i t was noted in the l i terature that highly ski l led gymnasts, in comparison to an appropriate reference population of females, had \"longer lower limbs proportionate to their total height\" (Smit, 1973:484). Signif icant differences among the national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts were not found for any of these height and length variables. Furthermore, differences in these variables, among the various groups do not suggest any trends or tendencies. While s ta t is t ica l analysis indicated non-significant differences in these variables for the various contrasts, further analysis is necessary in order to determine whether or not proportional differences exist . Such differences may,indicate developmental stages in the adolescent growth spurt. A long t ib ia l length, in relation to thigh or total leg length; or a long forearm length, in relation to upper arm or total arm length, may indicate specif ic stages in the growth sequences of these extremities, since during the adolescent growth spurt: The cal f length accelerates a l i t t l e before the thigh. a n d . (Tanner, 1962:12) The forearm has i ts peak velocity about 6 months ahead of the upper arm. It seems that the per-ipheral parts of the limbs are throughout growth more advanced than the proximal. (Maresh, 1955, cited in Tanner, 1962:12) Relatively long legs (as measured by trochanterion or i l iospinale height) 156 may indicate a delayed maturity in which a longer pre-adolescent period would provide \"extra\" growing time before the growth spurt: In the immediate preadolescent years, i t is the legs which are growing relat ively fastest of a l l skeletal dimensions . . . and so i f allowed to grow for an extra 2 years before the spurt, the legs become relat ively long. (Tanner, 1962:46) Proportionately longer feet and hands would also have maturational impli-cations (Hebbelinck & Ross, in Nelson & Morehouse, 1974:546): Foot length is probably the f i r s t of a l l skeletal dimensions below the head to cease growing. (Tanner, 1962:12) as well as biomechanical implications (Faulkner, 1977), such that longer feet and hands would benefit balance, and aid in movements where the body is propelled from them. In comparison to Olympic gymnasts described by Novak et a l . (1977), both the national e l i te and the pre-national e l i te gymnasts in the present study, have a shorter mean upper arm and forearm length, and a similar mean t ib ia l height. However, the landmarks used in deriving these measurements were not speci f ied. In comparison to Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts have a shorter mean t ib ia l and i l iospinale height, foot, upper arm, and forearm length, and a similar mean hand length. The same landmarks were used in obtaining these measurements for this latter study as were used in the present study. In comparison to highly sk i l led gymnasts described by Nelson (1974), the competitive gymnasts in the present study have a shorter mean t ib ia l and trochanterion height, a shorter mean forearm and foot length, and a similar mean upper arm length. However, the landmarks used in obtaining these measurements were not specif ied. 157 II Breadth, Width, and Depth Measures Biacromial breadth. A trend concerning biacromial breadth and gymnastic ab i l i ty was not evident in an .inter-study: comparison of Olympic (De Garay et a l . , 1974; Novak et a l . , 1977; Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972; Smit, 1973). However, in comparison to an appropriate female reference population, the highly sk i l led gymnasts of Sinning and Lindberg's study (1972) were reported to have a s igni f icant ly smaller biacromial diameter. A signif icant difference in biacromial breadth was expected between national e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, and recreational gymnasts, based on the obser-vation that e l i te and highly sk i l led gymnasts appear to have relat ively wider biacromial breadths than recreational gymnasts. It was also expected that exercise would show a positive effect upon skeletal growth in the region of the shoulders, as was suspected to have occurred in previous studies of gymnasts (Buckler & Brodie, 1977; Parizkova, 1968a; Smit, 1973). Further-more, late maturing females, as adults, have been reported to have wider shoulders than early maturing females (Bayley, 1943b, cited in Tanner, 1962: 102). Since i t was speculated that higher sk i l led gymnasts would be develop-mentally less mature than lesser sk i l led gymnasts, i t was expected that this difference would be displayed in the biacromial breadth measure. Although the biacromial breadth was smaller in Group 1 + 2 in comparison to Group 3, \"sl ightly\" smaller in Group 1 + 2 + 3 in comparison to Group 4, and larger in Group 1 in comparison to Group 2, these differences were not signif icant and in contradiction to the expected result . These differences however, \"suggest\" a tendency for higher sk i l led gymnasts to shave a smaller biacromial breadth than lesser sk i l led gymnasts. 158 It is suspected that the \"observed\" wide biacromial breadth measure \"appears\" this way in relation to height, or to a \"seemingly\" narrow b i i l i o c r i s t a l breadth, and is not necessarily \"wide\" when considered alone, in absolute terms. Furthermore, Smit (1973) noted that highly sk i l led gymnasts, in comparison to an appropriate reference sample, had a smaller biacromial breadth in absolute terms however, in relation to their height, they had a relat ively large breadth. Proportional analysis is necessary however, before such a relat ionship, among the present ab i l i ty groups, can be investigated. In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature (De Garay et a l . , 1974; Novak et a l . , 1977; Ross, 1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have smaller mean biacromial breadths. In comparison to gymnasts of similar ab i l i ty (highly sk i l led gymnasts) described in the l i terature , the competitive gymnasts in the present study have a similar mean biacromial breadth to the gymnasts described by Novak et a l . (1973) and Sinning and Lindberg (1972); a smaller mean breadth than the gymnasts of Nelson's study (1974); and a larger mean breadth than the gymnasts of Smit's study (1973). B i i l i o c r i s t a l breadth. A trend concerning b i i l i o c r i s t a l breadth and gymnastic ab i l i ty was not evident in an inter-study comparison of Olympic (De Garay et a l . , 1974; Novak et a l . , 1977; Ross, 1980) and highly ski l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972; Smit, 1973). However, in comparison to an appropriate female reference population, the highly sk i l led gymnasts of Sinning and Lindberg's study (1972) were reported to have a s igni f icant ly smaller b i i l i a c diameter. 159 A signif icant difference in b i i l i o c r i s t a l breadth was expected between national e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, and recreational gymnasts, based on the observation that e l i te and highly sk i l led gymnasts appear to have a relat ively smaller b i i1 iocr is ta l breadth than recreational gymnasts. Since i t was also speculated that higher sk i l led gymnasts would be developmentally less mature than lesser sk i l led gymnasts, i t was expected that this difference would be displayed in the b i i l i o c r i s t a l breadth measure. Late maturing females, as adults, have been reported to have smaller b i i l i a c diameters in relation to their height, than early maturing females (Tanner, unpublished, cited in Tanner, 1962:102). Furthermore, at maturity there are changes in the pelvic bones, with \"a part icularly large spurt in hip width\" (Tanner, 1962:45). Although the b i i l i o c r i s t a l breadth was smaller in Group 1 + 2 + 3 in comparison to Group 4, in Group 1 +2 in comparison to Group 3, and larger in Group 1 in comparison to Group 2.-,-.these differences were not signif icant and thus not \"in agreement with the the expected result . -These differences however, indicate a tendency for higher sk i l led gymnasts to have a smaller b i i1 iocr is ta l breadth than lesser sk i l led gymnasts. Furthermore, i t should be noted that irrespective of the non-significant multivariate analysis, univariate analysis shows the b i i1 iocr is ta l breadth difference to be signif icant for Group 1 + 2 in comparison to Group 3. It is suspected that the \"observed\" narrow b i i l i o c r i s t a l breadth measure \"appears\" this way in relation to height or to a \"seemingly\" wide biacromial breadth, and is not necessarily \"narrow\" when considered alone, in absolute terms. Furthermore, Smit (1973) noted that highly sk i l led gymnasts, in comparison to an appropriate reference sample, had a smaller intercr ista l width in absolute terms, and also in relation to their height. 160 Proportional analysis is necessary however, before such a relationship among the present ab i l i ty groups can be investigated. Such an analysis would also have developmental implications, as a broadening of the hips relative to the shoulders and waist is characteristic of female adolescence (Malina, 1974:119). In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature (De Garay et a l . , 1974; Novak et a l . , 1977; Ross, 1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have smaller mean b i i l i o c r i s t a l breadths. In comparison to gymnasts of similar ab i l i ty (highly sk i l led gymnasts) described in the l i terature , the competitive gymnasts in the present study have a similar mean b i i l i o c r i s t a l breadth to the gymnasts described by Novak et a l . (1973) and Sinning and Lindberg (1972); a smaller mean breadth than the gymnasts of Nelson's study (1974); and a larger mean breadth than the gymnasts of Smit's study (1973). Bi-epicondylar femur width. A trend concerning bi-epicondylar femur width and gymnastic ab i l i ty was not evident in an inter-study comparison of Olympic (Novak et a l . , 1977; Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972). Speculations concerning bi-epicondylar femur width and gymnastic ab i l i ty in the present study were not made. However, the observation of Adam's (1938, cited in Malina, 1969b:22), that women subjected to strenuous physical labor during childhood'had larger knee widths than women not subjected to such stress, was noted. The bi-epicondylar femur width was signi f icant ly smaller in Group 1 + 2 + 3 in comparison to Group 4, Since this was the only signif icant 161 difference.;to emerge among the contrasts, i t is possible that this is a spurious resul t . However, observing the similar mean values for Group.1, Group 2, Group 3, and Group 1 + 2, i t can be conjectured that this measure is related to performance, such that highly sk i l led gymnasts, in general, have a similar bi-epicondylar femur width, and in comparison to recreational gymnasts have a s igni f icant ly smaller width. This difference may be a ref lection of the generally small skeletal structure associated with highly sk i l led gymnasts, and known to present d ist inct advantages in performing gymnastic-type movements. This result may also have implications with respect to total body weight, with a small bi-epicondylar femur width indicating a small skeletal structure and thus a small skeletal weight. This measure should also be viewed in proportion to height or total leg length, since these comparisons have developmental implications: for their s i ze , children have proportionately larger knees, ankles and feet. (Ross, cited in Taunton, 1979:20) as well as biomechanical implications: The proportionally wider knee widths . . . may benefit s tab i l i ty [and provide] a proportion-ately greater area for weight-bearing stress. (Faulkner, 1977:22) In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature (Novak et a l . , 1977; Ross, 1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have similar mean bi-epicondylar femur widths. In comparison to gymnasts of similar ab i l i ty (highly sk i l led gymnasts) described in the l i terature (Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972), the competitive gymnasts in the present study have a similar mean bi-epicondylar femur width. 162 Anterior-posterior chest depth. A tendency for Olympic gymnasts (Ross, 1980) to have a smaller anterior-posterior chest depth than highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972) was evident in an inter-study comparison of these two ab i l i ty levels. This comparison was made with some reservation however, since the value obtained for this measurement is subject to the landmark points used, as well as to the technique followed. With a difference of 8.3 cm in anterior-posterior chest depth between two of the studies describing highly sk i l led gymnasts, i t is speculated that one of these measurements may have been taken at fu l l expiration, while the other at fu l l inspirat ion. Speculations concerning anterior-posterior chest depth and gymnastic ab i l i ty in the present study were not made. Anterior-posterior chest depth was signi f icant ly smaller in Group 1 in comparison to Group 2. Since this was the only signif icant difference to emerge among the contrasts, i t is possible that this is a spurious result . Conceivably however, this may well be one of the few differentiat ing variables distinguishing the \"better\" gymnasts, the national e l i t e s , from the \"very good\" gymnasts, the pre-national e l i t e s . Furthermore, since Group 1 + 2 has a smaller mean anterior-posterior chest depth than Group 3, and Group 1 + 2 + 3 has a \"sl ightly\" smaller mean than Group 4, there appears to be a tendency for higher sk i l led gymnasts to have a smaller mean chest depth than lesser sk i l led gymnasts. This difference may be a ref lect ion of the generally small skeletal structure associated with highly sk i l led gymnasts, and known to present d ist inct advantages in performing gymnastic-type movements. This result may also have implications with respect to the body mass components, with a small anterior-posterior chest depth indicating a 163 relat ively small skeletal structure and thus, a small skeletal mass. Since the anterior posterior chest depth measure encompasses the organs of the chest, a small measure may also indicate a small residual mass. Although none of the individual subject's values for this variable are in discord with those of the other subjects, i t should not be overlooked that this signif icant difference may be the result of measurement error. Since this measurement is taken at that \"instance\" before normal inspirat ion, i t is possible that this \"point\" may have been misjudged for some individuals. In comparison to Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have a smaller mean anterior-posterior chest depth. In comparison to gymnasts of similar ab i l i ty (highly sk i l led gymnasts) described in the l i terature (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972), the competitive gymnasts in the present study have a smaller mean anterior-posterior chest depth. Other breadth, width, and depth measures (transverse chest and bi-epicondylar humerus widths). A trend concerning transverse chest width and gymnastic ab i l i ty was not evident in an inter-study comparison of Olympic (Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972). A tendency for Olympic gymnasts (Novak et a l . , 1977; Ross, 1980) to have a larger bi-epicondylar humerus width than highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972), was evident in an inter-study comparison of these two ab i l i ty levels . Speculations concerning the transverse chest and bi-epicondylar 164 humerus widths, and gymnastic a b i l i t y , in the present study, were not made. However, the observation of Adam's (1938, cited in Malina, 1969b:22), that women subjected to strenuous physical labor during childhood had larger chest breadths than women not subjected to such stress, was noted. Signif icant differences among the national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts were not found for the transverse chest or the bi-epicondylar humerus widths. Furthermore, differences in these variables, among the various groups, do not suggest any trends. The non-significant transverse chest width differences-found . among the ab i l i ty groups are in discord with the findings of Pool et a l . (1969), in that thorax width was found to correlate s igni f icant ly with gymnastic performance. While s ta t is t ica l analyses indicated non-significant differences in the bi-epicondylar humerus width for the various contrasts, further analysis is necessary in order to determine whether or not proportional differences, especial ly with respect to height or total arm length, are. in existence among the ab i l i ty groups. Furthermore, proportional assessments of this variable may have developmental as well as biomechanical implications. A proportionally wide bi-epicondylar humerus width while indicating a develop-mentally immature physique (Ross, 1980) would also benefit s tab i l i ty of balances performed on the hands by presenting a larger surface area for weight-bearing stress. A proportionally wide bi-epicondylar humerus width, especially in a post-pubescent gymnast, may also indicate the positive effect of exercise upon skeletal growth in this region since: bone thickens'when subjected to heavy loads . . . [and] is deposited in proportion to the compressional load that the bone must carry. (Guyton, 1976:1058) 165 A proportionally wide transverse chest width may also indicate the effects of exercise in developing this area. In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature , both the national e l i te and the pre-national e l i te gymnasts in the present study have similar mean bi-epicondylar humerus widths (Novak et a l . , 1977; Ross, 1980), and smaller mean transverse chest widths (Ross, 1980). In comparison to gymnasts of similar ab i l i ty (highly sk i l led gymnasts) described in the l i terature (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972), the competitive gymnasts in the present study have a similar mean bi-epicondylar humerus width. The competitive gymnasts also have a similar mean transverse chest width to the gymnasts from Pool et a l . ' s study (1969); a smaller mean width than the gymnasts described by Nelson (1974) and Novak et a l . (1973); and a larger mean width than the gymnasts of Sinning and Lindberg's study (1972). I l l Girth Measures Since i t is well established that exercise has the effect of increasing muscular girths (Bready, 1961, Kusinitz et a l . , 1958, Tanner, 1952, cited in Malina, 1969a:24), i t was expected that highly sk i l led gymnasts involved in frequent and strenuous tra ining, would display larger exercise induced muscle girths than lesser sk i l led gymnasts involved in relat ively fewer hours of concentrated act iv i ty . Thus, signif icant differences among national e l i t e , pre-national e l i t e , lesser ski l led competitive and recreational gymnasts were expected. Thigh g i r th . A trend concerning thigh girth and gymnastic abi l i ty 166 was not evident in an inter-study comparison of Olympic (Novak et a l . , 1977; Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972). The thigh girth measure was signi f icant ly smaller in Group 1 + 2 + 3 in comparison to Group 4. In a multivariate sense, the thigh girth measure was not s igni f icant ly smaller in Group 1 + 2 in comparison to Group 3. However, irrespective of the insignif icant multivariate analysis, univariate analysis shows the thigh girth difference to be signif icant for this contrast. These results indicate a trend towards smallness in thigh girth for higher sk i l led gymnasts in comparison to lesser sk i l led gymnasts. Although these results are in the opposite direction to the expected resul ts , they do not indicate that the effects of exercise are displayed to a greater degree in the lesser sk i l led compared to the higher sk i l led gymnasts, or that this former group has a larger \"muscle\" girth than the latter group. Lesser sk i l led gymnasts consistently have a larger mean thigh skinfold than higher sk i l led gymnasts, and since girth measures were not \"corrected\" for skinfold thickness, i t is suspected that this former group has a larger \"absolute\" thigh girth comprised of a smaller \"muscle\" g i r th , relative to a larger layer of subcutaneous fat plus skin, than the latter group. These results do however, ref lect the general smallness in physique associated with highly sk i l led gymnast, in comparison to lesser sk i l led gymnasts. Furthermore, this \"smallness\" has been known to present d ist inct advantages in performing gymnastic-type movements. \u00E2\u0080\u00A2 Proportional assessments of the thigh g i r th , especially with respect to height or total leg length, would more graphically identify shape differences .among the ab i l i ty groups.- .Such, an assessment would also have 167 developmental implications, since major changes in the thigh girth are evident with growth: The principle changes in shape . . . from 2.5 years to 12 years, are relative decrease in size of the waist, and enlargement of hips and thighs. Thus development to maturity can be followed progress-ively as the deviations of the converted dimens-ions migrate to the midline [which represents the fu l ly mature figure of a 20 - 24 year old female]. (Behnke & Wilmore, 1974:88-89) In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature (Novak et a l . , 1977; Ross, 1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have smaller mean thigh gir ths. In comparison to gymnasts of similar ab i l i ty (highly ski l led gymnasts) described in the l i terature , the competitive gymnasts in the present study have a similar mean thigh girth to the gymnasts described by Nelson (1974); a smaller mean girth than the gymnasts of Sinning and Lindberg's study (1972); and a larger mean girth than the gymnasts of Novak et a l . ' s study (1973). Other girth measures (relaxed arm, flexed arm, forearm, wrist , chest, waist, ca l f , ankle, head, and neck gir ths) . Trends concerning the relaxed arm, calf (Nelson, 1974; Novak et a l . , 1973, 1977; Pool et a l . , 1969; Ross, 1980; Sinning & Lindberg, 1972), flexed arm (Novak et a l . , 1973, 1977; Ross, 1980; Sinning & Lindberg, 1972), forearm (Nelson, 1974; Novak et a l . , 1973, 1977; Ross, 1980; Sinning & Lindberg, 1972), and waist girths (Nelson, 1974; Ross, 1980; Sinning & Lindberg, 1972), and gymnastic ab i l i ty were not evident in an inter-study comparison of Olympic and highly sk i l led gymnasts. 168 A tendency for Olympic gymnasts (Ross, 1980) to have a larger wrist girth than highly sk i l led gymnasts (Nelson, 1974; Sinning & Lindberg, 1972) was evident in an inter-study comparison of these two abi l i ty leve ls . Nelson (1974) noted a tendency for highly sk i l led gymnasts to have larger chest circumferences than females from an appropriate reference population! However, there appears to be no trend concerning chest girth and gymnastic ab i l i ty in an inter-study comparison of Olympic (Ross, 1980) and highly sk i l led gymnasts (Nelson, 1974; Novak et a l . , 1973; Sinning & Lindberg, 1972). This latter comparison was made with some reservation, since the value obtained for this measurement is subject to the height of the tape, as well as to the technique followed. Ross (1980) took the chest girth measurement at the height of the fourth costosternal ar t iculat ion. Sinning and Lindberg (1972) recorded two values, one taken from a measurement made at the level of the a x i l l a , and the other from a measurement made at a level just below the breasts, while Novak et a l . (1973) recorded a maximum and a minimum chest girth measurement. Specifics on the chest girth measure-ment taken in Nelson's study (1974) were not given. Furthermore, only the study by Ross (1980) indicated the breathing phase in which the chest girth measurement was taken. The ankle, head, and neck girth measurements appear to have been neglected in the l i terature with only a single study including a head girth measurement (Nelson, 1974), and only a single study including a neck girth measurement (Sinning & Lindberg, 1972). Two studies describing highly sk i l led gymnasts included an ankle girth measurement (Nelson, 1974; Sinning & Lindberg, 1972). Signif icant differences among the national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts were expected, 169 part icularly in the relaxed arm, flexed arm, forearm, and calf gir ths. Since these are relat ively muscular s i t es , i t was expected that the effects of exercise would be displayed at these s i t es , to a greater extent in those highly sk i l led gymnasts involved in frequent and strenuous training. Signif icant differences in wrist , chest, waist, ankle, head, and neck gi r ths, among'the ab i l i ty groups, were expected to a lesser degree. Signif icant differences among the national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts were not found for the relaxed arm, flexed arm, forearm, wrist , waist, c a l f , ankle, chest, head, and neck gir ths. Differences among the ab i l i ty groups, in the flexed arm, forearm, wrist , ankle, head, and neck girths do not suggest any consistent trends. However, the relaxed arm, ca l f , chest, and waist girths are smaller in Group 1 + 2 + 3 in comparison to Group 4, and in Group 1 + 2 in comparison to Group 3, suggesting a tendency for higher sk i l led gymnasts to have smaller girth measurements, in these variables, than lesser sk i l led gymnasts. These resul ts , especially those noted for the relaxed arm and calf g i r ths, are in the opposite direction to the expected trend, and no tenden-cies concerning the flexed arm and forearm gir ths, and gymnastic ab i l i ty were found, in contradiction to what was expected. It is possible that, similar to the thigh girth resul t , the \"muscular\" girth differences of the relaxed arm, flexed arm, forearm, and calf girths were obscured by the subcutaneous fat plus skin layer. Furthermore, lesser sk i l led gymnasts in comparison to higher sk i l led gymnasts consistently have larger mean t r iceps, subscapular, supra i l iac , abdominal, and medial cal f skinfolds. Since girth measurements were not \"corrected\" for skinfold thickness, i t is suspected that this former group has larger \"absolute\" g i r ths, comprised of smaller 170 \"muscle\" girths relat ive to larger subcutaneous fat layers, than the latter group. Proportional assessments of the girth measurements, especially \u00E2\u0080\u00A2 with respect to height, would more graphically identify shape differences among the ab i l i t y groups. Such an assessment would also have developmental implications, since during growth and maturity the head, waist, and cal f girths become progressively smaller in proportion to height (Behnke & Wilmore, 1974:84-85). It should be noted that the head girth difference between Group 1 + 2 + 3 and Group 4, with the former group having a smaller mean than the la t ter , was approaching signif icance. This difference may be attributable to the thickness of the hair, since this measure encompasses the hair as well as the sku l l . This difference may also indicate a developmental difference between the two groups since during the adolescent growth spurt: the head diameters, pract ical ly dormant since a few years after bir th , accelerate somewhat in most individuals. (Tanner, 1962:10) Furthermore, in general, the circumference of the head in relation to height becomes progressively smaller with growth and maturity (Medawar, 1945; Ross, 1980). Proportionally large wrist, ankle, and chest g i r ths, in addition to having developmental implications, may also ref lect the effects of exercise in developing these areas. In comparison to gymnasts of similar ab i l i t y (Olympic gymnasts) described in the l i terature , both the national e l i te and the pre-national e l i te gymnasts in the present study have smaller mean relaxed arm (Novak et a l . , 1977; Ross, 1980), flexed arm (Novak et a l . , 1977; Ross, 1980), 171 forearm (Novak et a l . , 1977; Ross, 1980), wrist (Ross, 1980), chest (Ross, 1980), waist (Ross, 1980), and calf girths (Novak et a l . , 1977; Ross, 1980). In comparison to highly sk i l led gymnasts described by Nelson (1974), the competitive gymnasts in the present study have a smaller mean relaxed arm, forearm, chest, ca l f , and ankle g i r th; a similar mean wrist and head gi r th; and a larger mean waist g i r th . In comparison to the gymnasts described by Novak et a l . (1973), the competitive gymnasts have a smaller mean relaxed arm gi r th; a larger mean flexed arm, forearm, and chest g i r th; and a similar mean calf g i r th . In comparison to Sinning and Lindberg's gymnasts (1972), the competitive gymnasts have a larger mean relaxed arm and waist g i r th; a similar mean flexed arm, forearm, wrist , neck, and ankle g i r th; and a smaller mean calf g i r th . In comparison to the highly sk i l led gymnasts of Pool et a l . ' s study (1969), the competitive gymnasts have a smaller mean relaxed arm and cal f g i r th . IV Skinfold Thickness Measures It is well established that exercise has the effect of reducing the thickness of the subcutaneous fat layer (Johnson, 1969; Parizkova & Poupa, 1963; Smit, 1973; Well, Jok l , & Bohranen, 1963); and Smit (1973: 480) noted that; \"the skinfolds of gymnasts decreased as their hours of act iv i ty per week increased\". Furthermore, exercise has also been noted to \"check\" or maintain the level of fat in growing adolescence, so that as growth proceeds, the skinfold thickness measures remain relat ively constant (Parizkova, 1959). For these reasons, i t was expected that highly sk i l led gymnasts, involved in frequent and rigorous training, would display smaller skinfold thickness measures than lesser sk i l led gymnasts, involved in comparatively fewer hours of concentrated act iv i ty . 172 Triceps, subscapular, suprai l iac , abdominal, front thigh, and medial cal f skinfold thickness measures. A tendency for Olympic gymnasts to have smaller suprail iac (Novak et a l . , 1973, 1977; Ross, 1980) and abdominal skinfolds (Novak et a l . , 1973; 1977; Ross, 1980; Smit, 1973; Youngren, 1969) than highly sk i l led gymnasts was evident in an inter-study comparison of these two abi l i ty levels. A tendency for Olympic gymnasts (Novak et a l . , 1977; Ross, 1980) to have a larger triceps skinfold than highly sk i l led gymnasts (Novak et a l . , 1973; Pool et a l . , 1969; Smit, 1973; Youngren, 1969) was evident in an inter-study comparison of these two ab i l i ty levels. Trends concerning the subscapular (Novak et a l . , 1973, 1977; Pool et a l . , 1969; Ross, 1980; Smit, 1973), front thigh (Novak et a l . , 1973, 1977; Ross, 1980; Youngren, 1969), and medial cal f skinfolds (Novak et a l . , 1973, 1977; Ross, 1980), and gymnastic ab i l i ty were not evident in an inter-study comparison of Olympic and highly sk i l led gymnasts. Although i t was expected that the Olympic gymnasts would consistently have smaller skinfold thicknesses than highly sk i l led gymnasts, these results are not surprising in view of the findings of Wilmore et a l . (1970a, 1970b) and Zwiren et a l . (1973), in that skinfold measurements were found to be basical ly unsound in assessing changes in body fat with exercise. Furthermore, Young et a l . (1964, cited in Shephard et a l . , 1969:1185) noted that the triceps skinfold did not correlate well with obesity. It should also be noted that trends concerning skinfold thickness and a b i l i t y , in the inter-study comparisons, may have been masked by variations in the landmark si te at which the measurement was taken. Ross (1980) takes the suprail iac skinfold measurement approximately f ive to seven centimeters superior to the i l iosp ina le , while 173 Novak et a l . (1973, 1977) takes this measurement at the i l i a c crest. Techniques and landmark points for the skinfold measurements were not \" specified in the majority of the studies reviewed. Al l of the skinfold thickness measures, the t r iceps, subscapular, suprai l iac , abdominal, front thigh, and medial ca l f , were found to be signi f icant ly smaller in Group 1 + 2 + 3 in comparison to Group 4, as was expected. Similarly a l l of the skinfold thickness measures, with the exception of the subscapular skinfold, were signi f icant ly smaller in Group 1 + 2 in comparison to Group 3. However, even the subscapular skinfold was smaller in Group 1 + 2 in comparison to Group 3, with the skinfold difference for this contrast approaching signif icance. These differences indicate a definite trend towards smaller skinfold thickness measurements for highly sk i l led in comparison to lesser sk i l led gymnasts. None of the skinfold thickness measures, the t r iceps, subscapular, suprai l iac, abdominal, front thigh, or medial c a l f , was s igni f icant ly different in Group 1 in comparison to Group 2. This result is not surprising in view of the fact that these groups are very close in a b i l i t y , and are assumed to have training programs of similar intensity and frequency. Furthermore, this result is in agreement with those results reported by Pool et a l . (1969) and Youngren (1969), in which higher placing gymnasts did not have signi f icant ly different skinfolds than lower placing gymnasts. However, the \"tendency\" for higher placing gymnasts to have smaller skinfold thickness measures than lower placing gymnasts, noted in these two studies, was not evident in the present comparison of national e l i te and pre-national e l i te gymnasts. In comparison to the Olympic gymnasts described by Novak et a l . (1977) and Ross (1980) both the national e l i te and the pre-national e l i te 174 gymnasts in the present study have a smaller mean tr iceps, subscapular, suprai l iac , abdominal, front thigh, and medial cal f skinfold. In comparison to the highly sk i l led gymnasts described by Novak et a l . (1973), the competitive gymnasts in the present study have a larger mean t r iceps, subscapular, medial ca l f , and front thigh skinfold; a similar mean abdominal skinfold; and a smaller mean suprail iac skinfold. In compar-ison to the gymnasts of Smit's study (1973), the competitive gymnasts have a similar mean triceps skinfold, and a smaller mean subscapular and abdominal skinfold. In comparison to the gymnasts described by Youngren (1969), the competitive gymnasts have a similar mean triceps and front thigh skinfold, and a smaller mean abdominal skinfold. In comparison to the gymnasts measured by Pool et a l . (1969), the competitive gymnasts have a larger mean triceps and subscapular skinfold. V Weight and Proportional Mass Measures Practical and workable anthropometric equations for fractionating the body mass into muscular, skeleta l , residual , and fat components have just recently been presented (Behnke & Wilmore, 1974; Drinkwater & Ross, in Ostyn et a l . , 1980:177). As a resul t , there were few studies located referring to \"muscular, skeleta l , and residual masses\" distinguishable from a lean body mass. Since exercise tends to reduce the fat \"content\" of the body as i t encourages the \"deposition\" of muscle tissue (Parizkova, 1959; Parizkova & Poupa, 1963), higher sk i l led gymnasts were expected to possess a smaller \"proportional\" fat mass, in relation to a larger \"proportional\" muscle mass, than lesser sk i l led gymnasts. It was expected that higher sk i l led gymnasts would be shorter in 175 stature, and have generally smaller skeletal and girth dimensions, as well as smaller skinfold thicknesses,than lesser ski l led gymnasts. Therefore, i t was speculated that these differences would be reflected in smaller \"absolute\" fa t , skeleta l , muscular, and residual masses for higher sk i l led in comparison to lesser ski l led gymnasts. However, in .proportional terms, due to the expected smaller \"proportional\" (percent) fat mass of the higher s k i l l e d , in comparison to the lesser sk i l led gymnast, there would thus be a larger proportion '.of the body mass due to the other three mass components. Therefore, differences in the \"proportional\" skeleta l , muscle, and residual masses among the ab i l i ty groups were expected to emerge, with higher ski l led gymnasts having larger \"proportional\" muscle, skeleta l , and residual masses than lesser sk i l led gymnasts. Furthermore, these differences were expected to be the direct result of a smaller proportional fat mass in the higher s k i l l e d , in comparison to the lesser sk i l led gymnasts. It should also be emphasized that the \"proportional masses\" incorporated a \"calculated\" mass, which is the sum of the four fractional masses, in their derivation, and not the direct ly measured scale weight. Weight. A trend concerning weight and gymnastic ab i l i ty was not evident in an inter-study comparison of Olympic (De Garay et a l . , 1974; Hirata, 1966; Montpetit, in Salmela, 1976:187; Novak et a l . , 1977; Pool et a l . , 1969; Ross, 1980), highly sk i l led (Carter, 1970; Nelson, 1974; Novak et a l . , 1973; Pool et a l . , 1969; Sinning & Lindberg, 1972; Smit, 1973; Sprynarova & Parizkova, 1969; Youngren, 1969), and recreational gymnasts (Pool et a l . , 1969). However, in comparison to appropriate reference populations, the gymnasts described by Carter (1970), Sinning and Lindberg (1972), and Smit (1973) were reported to be l ighter in weight, and the 176 Sinning and Lindberg gymnasts were s igni f icant ly l ighter . On the other hand, the highly sk i l led gymnasts of Nelson's study (1974) were found to be similar in weight to an appropriate reference population of females, to whom they were compared. Signif icant differences in weight among national e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts were not expected, based on the observation that e l i te and highly sk i l led gymnasts, in general, \"appear\" l ight in weight. It also appears that a relat ively l ight weight is a prerequisite for successful participation at these levels. Highly ski l led gymnasts were expected to possess a low \"proportional\" fat mass in relation to a high \"proportional\" lean body mass (muscle, skeleta l , and residual masses), while lesser sk i l led gymnasts were expected to possess a higher \"proportional\" fat mass in relation to their proportional lean body mass. These differences were expected as a product of the varying degrees of exercise involvement of the groups. However, since adipose tissue is less dense than muscle tissue (Behnke & Wilmore, 1974), and since \"the bones of athletes become considerably heavier than those of non-athletes\" (Guyton, 1976:1060), i t was conjectured that the compositional differences among.-the ab i l i ty groups would cancel weight differences, and \"absolute\" weight would thus show no difference. It was expected that national e l i t e , pre-national e l i t e , and lesser sk i l led competitive gymnasts, taken together, would have a s i g n i f i -cantly l ighter mean body weight than recreational gymnasts. This speculat-ion was based on the assumption that the higher sk i l led gymnasts would be smaller in height.and overall skeletal dimensions (indicating less absolute skeletal weight), and possess smaller skinfold thickness values (indicating -less absolute fat weight) and smaller girth measures (indicating less 177 absolute muscle weight), than the recreational gymnasts. Furthermore, i t was speculated that the \"absolute\" muscle mass (but not the proportional muscle mass) would be less in the higher ab i l i ty group, due to their general overall smallness in skeletal dimensions, which according to Behnke (1963, cited in Behnke & Royce, 1966:76) would indicate a smaller muscle mass since, a specif ic amount of lean body mass is associated with a given skeletal s ize . Although Group 1 + 2 + 3 has a smaller mean weight than Group 4 (4.1 kg), Group 1 + 2 a smaller mean weight than Group 3 (4,8 kg), and Group 1 a smaller mean weight than Group 2 (0.4 kg), these differences were not s igni f icant . It should be noted however, that the difference in weight between Group 1 + 2 and Group 3 was approaching signif icance. These differences indicate a trend towards lightness in weight for higher sk i l led in comparison to lesser sk i l led gymnasts. These resul ts , and especially the non-significant weight difference noted between1 Group 1 and Group 2, are in agreement with the results of Pool et a l . ' s study (1969), in that higher placing gymnasts tended to weigh less than lower placing gymnasts, but not s igni f icant ly so. Although these results are not as expected, they are not \" surprising in view of the fact that the anthropometric indicants of weight differences, height, and most of the other skeletal dimensions, as well as a l l of the girth measures, with the exception of the thigh g i r th , were not s igni f icant ly different among the ab i l i ty groups. Differences in these variables, among the ab i l i ty groups, were not of suff ic ient magnitude to affect the weight measures. However, since there was a tendency for higher sk i l led gymnasts to be both shorter and l ighter than lesser sk i l led gymnasts, perhaps a ponderal index (height-to-weight) comparison, among 178 the ab i l i ty groups, would be a more meaningful comparison. The non-significant weight difference between Group 1 + 2 + 3 , and Group 4, may be a ref lect ion of club \"attitudes\". In clubs where the recreational level is viewed as a \"base\" for the competitive program, participants are often pre-selected on lightness in weight, among other variables, for inclusion in the recreational program. In comparison to gymnasts of similar ab i l i ty (Olympic gymnasts) described in the l i terature (De Garay et a l . , 1974; Hirata, 1966; Montpetit, in Salmela, 1976:187; Novak et a l . , 1977; Pool et a l . , 1969; Ross, 1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have smaller mean weights. In comparison to gymnasts of similar ab i l i ty (highly ski l led gymnasts) described by Carter (1970), Nelson (1974), Pool et a l . (1969), Sinning and Lindberg (1972), Smit (1973), Sprynarova and Parizkova (1969), and Youngren (1969), the competitive gymnasts in the present study have a smaller mean weight. However, in comparison to the highly sk i l led gymnasts described by Novak et a l . (1973), the competitive gymnasts have a larger mean weight. In comparison to the recreational gymnasts referred to in Pool et a l . ' s study (1969), the recreational gymnasts in the present study have a smaller mean weight. Proportional fat mass. A tendency for Olympic gymnasts to have a smaller percent fat mass (Novak et a l . , 1977; Ross, 1980) than highly sk i l led gymnasts (Novak et a l . , 1973; Sinning & Lindberg, 1972; Sprynarova & Parizkova, 1969), was evident in an inter-study comparison of these two ab i l i ty levels. 179 The proportional fat mass was signi f icant ly smaller in Group 1 + 2 + 3 in comparison to Group 4, and in Group 1 + 2 in comparison to Group 3, as expected. The non-significant proportional fat mass difference between Group 1 and Group 2 was also expected, due to the relative closeness of the two groups, in training intensity and regime in general. These results suggest a trend towards smallness in proportional fat mass in higher sk i l led gymnasts in comparison to lesser sk i l led gymnasts. While these results may ref lect the effects of exercise in reducing the fat \"content\" of the body, they may also have developmental implications, especially since signif icant differences in maturity were also reported for these contrasts. Around the time of puberty there are marked changes in the composition of the female body, with a noted increase in the fat \"content\" (Edwards, 1951; Parizkova, 1959; Reynolds, 1950). Those gymnasts that are further advanced in maturity, would l ike ly exhibit this increase to a greater extent than those gymnasts who are less mature. Furthermore, in the present study, lesser sk i l led gymnasts in comparison to higher sk i l led gymnasts were s igni f icant ly more mature, and also had signi f icant ly larger proportional fat masses. Since the proportional fat mass value (also referred to as the percent fat mass value) .varies considerably with the derivative equation used (Damon & Goldman, 1964; Malina, 1969b; Steinkamp et a l . , 1965, as cited in Malina, 1969b: 19), i t is with much reservation that the percent fat mass values in the present study are compared with those reported in the l i terature . Keeping this point in view, i t is noted that in comparison to the Olympic gymnasts described by Novak et a l . (1977) and Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have a smaller mean proportional fat mass. This latter study employed the same equations as 180 the present study, in the calculation of proportional fat mass. In comparison to the highly sk i l led gymnasts described by Novak et a l . (1973), Sinning and Lindberg (1972), and Sprynarova and Parizkova (1969), the competitive gymnasts in the present study have a smaller proportional fat mass. Proportional muscle mass. The proportional muscle mass was signi f icant ly larger in Group 1 + 2 + 3 in comparison to Group 4, and in Group 1 + 2 in comparison to Group 3, as expected. The non-significant proportional muscle mass difference between Group 1 and Group 2 was also expected, due to the relative closeness of the two groups in ab i l i ty and in training regime. These results suggest a trend towards larger proport- .. ional muscle masses in higher sk i l led in comparison to lesser sk i l led gymnasts, and generally ref lect the effect of exercise in developing the muscular component of the body. The proportional muscle mass value was derived from height, \"calculated\" body weight, skinfold thickness, and girth measures. Of these measures, only the skinfold thicknesses and the thigh girth were, found to.be s igni f icant ly smaller in higher sk i l led in comparison to recreational gymnasts. However, calculation of the proportional muscle mass measure incorporated muscle girths \"corrected\" for the subcutaneous fat layer. Furthermore, while larger girths indicated larger \"absolute\" muscle masses for lesser sk i l led in comparison to higher sk i l led gymnasts/\"corrected\" g i r ths, led to smaller \"proportional\" muscle masses for the lesser sk i l led in comparison to the higher ski l led gymnasts. In comparison to the Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present 181 study have similar mean proportional muscle masses. This study employed the same equation as the present study, in the calculation of proportional muscle mass. Proportional skeletal mass. The proportional skeletal mass was signi f icant ly larger in Group 1 + 2 in comparison to Group 3. Although Group 1 + 2 + 3 has a larger mean proportional skeletal mass than Group 4, and Group 1 a larger mean mass than Group 2, these differences were not s igni f icant . However, these results do indicate a tendency for higher sk i l led gymnasts to have larger proportional skeletal masses than lesser sk i l led gymnasts. The proportional skeletal mass value was derived from height, \"calculated\" body weight, wrist and ankle g i r ths, and bi-epicondylar humerus and femur widths. Of these measures, only the bi-epicondylar femur width was s igni f icant ly smaller in higher sk i l led in comparison to recreational gymnasts. While these results may indicate a smaller \"absolute\" skeletal mass, they do not simultaneously indicate a smaller \"proportional\" skeletal mass. The fact that there was a s igni f icant ly smaller proportional fat mass in higher sk i l led in comparison to lesser sk i l led gymnasts, indicates that \"automatically\" the other mass components would have larger proportional weightings, with respect to the \"calculated\" weight. Thus, the proportional skeletal mass measure has emerged as having a s igni f icant ly larger contribution to the \"calculated\" weight in higher sk i l led in comparison to lesser sk i l led gymnasts. In comparison to the Olympic gymnasts described by Ross (1980), both the national e l i te and the pre-national e l i te gymnasts in the present study have larger mean proportional skeletal masses. This study employed 182 the same equations as the present study, in the calculation of proportional skeletal mass. Proportional residual mass. Although Group 1 + 2 + 3 has a larger mean proportional residual mass than Group 4, Group 1 + 2 a larger mean mass than Group 3, and Group 1 a smaller mean than Group 2, these . differences were not s igni f icant . However, the residual mass difference between Group 1 + 2 + 3 and Group 4 was approaching signif icance. Furthermore, these results suggest a tendency for higher sk i l led gymnasts to have larger proportional residual masses than lesser sk i l led gymnasts. The proportional residual mass was derived from height, the biacromial and b i i l i o c r i s t a l breadths, the transverse chest width, and the anterior-posterior ,chest depth. Of these measures, only the anterior-posterior chest depth was signi f icant ly smaller in Group 1 in comparison to Group 2. Considering these resul ts , i t is not surprising that the proportional residual mass measures did not di f fer s igni f icant ly for any of the contrasts. In comparison to the Olympic gymnasts described by Ross (1980), the national e l i te gymnasts have a similar mean proportional residual mass, and the pre-national e l i te gymnasts have a larger mean proportional residual mass. This study employed the same equations as the present study, in the calculation of proportional residual mass. 183 MATURITY-ANTHROPOMETRTC RELATIONSHIP Because s ign i f i can t di f ferences in maturity were found among the a b i l i t y groups, i t was expected that these maturational di f ferences would also be re f lec ted in the anthropometric cha rac te r i s t i c s , since maturity and morphology are so c lose ly re lated (Fr isch & McArthur, 1974; Fr isch & Reve l le , 1970, 1971; Garn & Haske l l , 1960; Johnston et a l . , 1971; Maresh, 1972; McNeill & L ivson, 1963; Reynolds, 1950). While absolute values of the anthropometric var iables indicate quant i tat ive growth, proportional values more graphica l ly r e f l ec t developmental growth. In the present study, a l l anthropometric va r iab les , with the exception of the proportional mass measures, were assessed in terms of the i r absolute values only . While anthropometric di f ferences among the gymnastic a b i l i t y groups were found, i t was not possible to in fer maturational developments from these d i f ferences. I t is possible however, to impl icate maturational developments from proportional di f ferences. ' Furthermore, proportional assessments are concerned with shape and not amount, and therefore quantity i s not confused with representing a mature physique, as may eas i l y occur when maturity is implied from absolute values alone. The maturity-anthropometric re la t ionsh ip i s i l l u s t r a t e d by Ross and Wilson (1974) in a shape comparison between a pre-pubescent and a post-pubescent female. Using anthropometric var iables proport ional ly adjusted to a standard height, the pre-pubescent female in contrast to the post-pubescent female, has a shorter trunk length and a longer tota l leg length, comprised of a longer lower leg and thigh length. These resu l ts are in terest ing in view of the f indings noted in the present study. Highly s k i l l e d gymnasts in comparison to lesser s k i l l e d gymnasts were found to be s i g n i f i c a n t l y 184 delayed in maturity (Group 1 + 2 + 3 vs Group 4, Group 1 + 2 vs Group 3), and anthropometrics!ly, were s igni f icant ly shorter in trunk length (Group 1 + 2 + 3 vs Group 4, Group 1 + 2 vs Group 3, Group 1 vs Group 2), and thigh length (Group 1 vs Group 2), while signif icant differences in height were not found. From the discussion of the anthropometric variables, i t is evident that further analysis with respect to proportional assessments, is necessary to exp l ic i t ly i l lust ra te shape differences among gymnasts from varying ab i l i ty levels. Furthermore, such analysis is necessary before anthropometric variables can be closely associated with maturational status. 185 CHAPTER 5 SUMMARY AND CONCLUSIONS SUMMARY Fame for female gymnasts is often short - l ived, with many highly sk i l led gymnasts dropping out of e l i te caliber competitions during, or soon after adolescence. Many of today's e l i te female gymnasts are in the age range when normal pubertal developments, such as increases in weight, height, adiposity, and dimensions in general, are expected to occur. The recent trend towards younger participation at e l i te gymnastic competitions, coupled with an increasing display of movement perfection and task complexity, seems to indicate that female gymnasts are reaching their prime, or \"peaking\" at younger ages. This trend also suggests the poss ib i l -i ty that important relationships exist among maturational status, anthro-pometric character ist ics, and success in gymnastic performance. It was the purpose of this study to investigate the possib i l i ty that success in gymnastics is related to the concept of maturity and anthro-pometric character ist ics. Furthermore, i t was hypothesized that there would be signif icant maturational (skeletal age in reference to chronological age; incidence of menarche) and anthropometric differences (height and length measures; width, breadth, and depth measures; girth measures; skinfold thickness measures; weight and proportional mass measures) among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. 186 Sixty-nine Canadian female gymnasts, ages 11.5 to 18.0 years, with ab i l i ty levels ranging from recreational through to national e l i t e , were participants in the study. Performance levels and research groups were identif ied as; Group 1, National E l i te Gymnasts; Group 2, Pre-National E l i te Gymnasts; Group 3, Competitive Gymnasts; and Group 4, Recreational Gymnasts. Al l available subjects were participants in Groups 1 and 2, while subjects were randomly selected to represent Groups 3 and 4. Maturational assessment for each subject consisted of a skeletal age rat ing, based on a radiographic photo of the le f t hand and wrist , and an indication that menarche had, or had not occurred. Anthropometric assessment consisted of 39 selected variables. Spec i f ica l ly , 11 height and length; 6 breadth, width, and depth; 11 g i r th ; 6 skinfold thickness; 1 weight and 4 proportional mass measures were evaluated. In a l l s ta t is t ica l analyses the following preplanned orthogonal contrasts were used; Group 1 + 2 + 3 vs Group 4, Group 1 + 2 vs Group 3, and Group 1 vs Group 2. The chronological age minus skeletal age d i f fer -ence, when compared between groups, was assessed using analysis of variance. The incidence of menarche difference between groups was assessed using chi-square analysis. The anthropometric differences were assessed using multivariate and univariate analysis of covariance, using chronological age as the covariate. The level of significance for a l l s ta t is t ica l tests was set at p < .01. The results of the s ta t is t ica l tests indicated that the two maturational hypotheses, and the f ive anthropometric hypotheses were part ia l ly supported. More spec i f i ca l l y , at a level of significance of p < .01, the following results were noted. 187 Highly sk i l led gymnasts in comparison to lesser ski l led gymnasts (Group 1 + 2 + 3 vs Group 4, Group 1 + 2 vs Group 3) were maturationally delayed both skeletal ly and menarcheally. Anthropometrically, they were shorter in trunk length; smaller in t r iceps, suprai l iac , abdominal, front thigh, and medial cal f skinfolds; smaller in proportional fat mass and larger in proportional muscle mass. In addition, highly sk i l led gymnasts (Group 1 + 2 + 3 ) , in comparison to recreational gymnasts (Group 4), were smaller in bi-epicondylar femur width, thigh g i r th , and subscapular skinfold. As wel l , e l i te gymnasts (Group 1 + 2), in comparison to lesser sk i l led competitive gymnasts (Group 3), were smaller in s i t t ing height and larger in proportional skeletal mass. National e l i te gymnasts (Group 1), in comparison to pre-national e l i t e gymnasts (Group 2), were not maturationally di f ferent , skeletal ly or menarcheally. Anthropometrically, they were shorter in trunk length, longer in thigh length, and smaller in anterior-posterior chest depth. The results of the maturational assessment indicated that there may be a relationship between gymnastic ab i l i t y and maturity, with highly sk i l led gymnasts being developmentally less mature than lesser sk i l led gymnasts. The results of the anthropometric assessment indicated that there may be a relationship between gymnastic ab i l i ty and anthropometric parameters. It was suggested that the shorter s i t t ing height, trunk and thigh lengths, observed in higher ski l led in comparison to lesser sk i l led gymnasts, provided specif ic biomechanical advantages in performing gymnastic-type movements. The smaller anterior-posterior chest depth, thigh g i r th , and bi-epicondylar femur width, observed in higher ski l led in comparison to lesser sk i l led gymnasts, were considered to be reflections of the generally small physique 188 associated with biomechanical advantages for gymnastic-type movements. As well i t was considered that, due to the nature of the measurement, the anterior-posterior chest depth difference may have been a product of measurement error. The skinfold thickness, proportional fat and muscle mass d i f f -erences, among the ab i l i ty groups, were presumed related to differences in training regime, with higher ski l led gymnasts involved in consider-ably more hours of intense training displaying smaller skinfold thickness values, a lower proportional fat mass, and a higher proportional muscle mass, than lesser ski l led gymnasts. The larger proportional skeletal mass, observed in higher ski l led in comparison to lesser ski l led gymnasts, was attributed to the low fat mass value of this former group, which consequently led to a higher weighting of the skeletal mass, as well as of the other lean body mass components, when assessed in proportional terms of \"calculated\" total body weight. It was conjectured that the differences noted, among the ab i l i ty groups, in thigh and trunk length, s i t t ing height, bi-epicondylar femur width, thigh g i r th , and proportional fat mass, were reflections of matur-ational differences. However, i t was postulated that proportional analysis of these variables, in relation to height, would more closely indicate developmental differences. Furthermore, the signif icant thigh and trunk -length, and s i t t ing height differences, and the non-significant vertex standing height differences, suggested proportional differences among the ab i l i t y groups, in those segmental lengths comprising the vertex standing height measure. Since height and weight were not s igni f icant ly different among the ab i l i t y groups, and since there was a tendency for higher ski l led gymnasts to be shorter and l ighter than lesser ski l led gymnasts, i t was 189 speculated that a ponderal index comparison would be a meaningful variable to investigate. It was suggested that girth measurements \"corrected\" for the subcutaneous layer of fat plus skin, would be more appropriate in assessing muscular development at specif ic s i tes . F ina l ly , i t was proposed that further analysis of the anthropometric variables, with respect to proportional assessments, would be meaningful in identifying shape differences among the ab i l i ty groups, and would more graphically i l lust ra te maturity-anthropometric relationships. CONCLUSIONS The two maturational and the f ive anthropometric hypotheses were part ia l ly supported at a level of significance of p < .01. Thus, there are maturational and anthropometric differences among national e l i t e , pre-national e l i t e , lesser sk i l led competitive and recreational gymnasts. More spec i f i ca l l y , at ;p < .01, the following results were noted: 1. National e l i t e , pre-national e l i t e , and lesser ski l led competitive gymnasts, taken together, in comparison to recreational gymnasts, are maturationally delayed, both skeletal ly and menarcheally. Anthropo-metrical ly, they are shorter in trunk length, smaller in bi-epicondylar femur width, thigh g i r th , t r iceps, subscapular, supra i l iac , abdominal, front thigh, and medial cal f skinfolds; have a smaller proportional fat mass, and a larger proportional muscle mass. 2. National e l i te and pre-national e l i te gymnasts, taken together, in comparison to lesser ski l led competitive gymnasts, are maturationally delayed, both skeletal ly and menarcheally. Anthropometrically, they are shorter in s i t t ing height and trunk length; smaller in t r iceps, 190 suprai l iac , abdominal, front thigh, and medial cal f skinfolds; have a smaller proportional fat mass; and larger proportional muscle and skeletal masses. 3. National e l i te gymnasts, in comparison to pre-national e l i te gymnasts, are not maturationally dif ferent, skeletal ly or menarcheally. Anthropometrically, they are shorter in trunk length, longer in thigh length, and smaller in anterior-posterior chest depth. It is both necessary and vital to the advancement of the sport of gymnastics that coaches understand the relationships among anthropometric character ist ics, maturity, and performance, since the best performers may not necessarily possess the potential for future promise, and to select teams on the grounds of present performance and physique alone is misleading, and may deter those with true potential from part icipating: We wonder how much Canadian competitive talent is lost by fa i lure to recognize that youngsters with ideal physique characteristics for these sports have a tendency to be late maturers. (Ross et a l . , in Taylor, 1976:277) SUGGESTIONS FOR FURTHER RESEARCH There are many ways in which anthropometric and maturational data can be observed and compared. The present study focused on comparing gymnasts from varying ab i l i ty levels , on maturational and anthropometric parameters. It is proposed that conversion of the absolute anthropometric values into proportional values, would result in the nascency of new information, that would be interpretable in maturational and possibly 191 biomechanical terms. Furthermore, such assessments may aid in the ident i f icat ion of optimum body proportions for success in gymnastics. It would also be of interest to compare the values of the anthro-pometric variables-from each ab i l i ty group, with the s ta t is t ics from normative data, in order to identify the degree of deviation of the gymnastic populations from the norm. Such comparisons would also offer further insights into the \"uniqueness\" of the female gymnast's physique. For example, recognizing that the present sample of gymnasts is comprised of females 11.5 to 18.0 years, with a mean age of 14.9 years, the anthropome-t r i c s ta t is t ics of this population are compared with those of 14 year olds from a reference population described by Ross, Drinkwater, Whittingham, and Faulkner (in Berg & Erikson, 1980:3). Compared with these 14 year olds, highly sk i l led gymnasts (Group 1 + 2 + 3 ) are at the 10th percentile for height and the 20th percentile for weight; while recreational gymnasts (Group 4) are at the 20th percentile for height and the 45th percentile for weight. These results indicate that while the gymnasts are comparatively shorter in stature than the average 14 year o ld , they are heavier for their height in comparison to those 14 year olds of the same height. However, the higher sk i l led gymnasts (Group 1 + 2 + 3 ) are less heavy for their height than the recreational gymnasts (Group 4). Furthermore, this \"heaviness\" is probably due to muscular development and not excess adiposity since, muscular tissue is more dense than adipose tissue (Behnke & Wilmore, 1974), and exercise tends to encourage the \"deposition\" of muscle tissue (Smit, 1976) while i t discourages the \"accumulation\" of adipose tissue (Parizkova & Poupa, 1963). The skinfold thickness values for these Groups further support this contention. Most anthropometric and maturational studies describing the female 192 gymnastic population have included only college aged gymnasts, and have been concerned with gymnasts from only one ab i l i t y leve l . As the number of young female gymnasts undergoing serious gymnastic t ra ining, and participating'and succeeding in e l i te competitions increases, the need for maturational and anthropometric descriptions of this \"new\" gymnastic population becomes evident. REFERENCES 193 194 Acheson, R. M., & D u p e r t u i s , C. W. The r e l a t i o n s h i p between p h y s i q u e and r a t e o f s k e l e t a l m a t u r a t i o n i n boys. Human B i o l o g y , 1957, 29_, 167-193. A l e x a n d e r , M. J . The r e l a t i o n s h i p o f somatotype and s e l e c t e d a n t h r o p o m e t r i c measures to b a s k e t b a l l performance i n h i g h l y s k i l l e d f e m a l e s . Research Q u a r t e r l y , 1976, 4 7 ( 4 ) , 575-585. A l l e n , T. H., Peng, M. T., Chen, K. P., Huang, T. F., Chang, C , & Fang, H. S. P r e d i c t i o n o f t o t a l a d i p o s i t y from s k i n f o l d s and the c u r v i l i n e a r r e l a t i o n s h i p between e x t e r n a l and i n t e r n a l a d i p o s i t y . M e t a b o l i s m , 1956, 5, 346-352. American P s y c h o l o g i c a l A s s o c i a t i o n : P u b l i c a t i o n Manual (2nd e d . ) . B a l t i m o r e : Pridemark P r e s s I n c . , 1979. Anderson, D. L. I n t e r r e l a t i o n s h i p s o f d e n t a l m a t u r i t y , s k e l e t a l m a t u r i t y , h e i g h t , and weig h t from age 4 t o 14 y e a r s . Growth, 1975, 3 9 ( 4 ) , 453-462. Asmussen, E. Growth and a t h l e t i c performance. In K. Kato ( E d . ) , Post-War t r e n d s o f growth and development o f males and females i n each c o u n t r y . P r o c e e d i n g s o f the I n t e r n a t i o n a l Congress o f S p o r t S c i e n c e s , Denmark, 1966, pp. 103-110. A s t r a n d , P. 0., Engstrom, L., E r i k s s o n , B. 0., K a r l b e r g , P., N y l a n d e r , I . , S a l t i n , B., & Th o r e n , C. G i r l Swimmers: With s p e c i a l r e f e r e n c e to r e s p i r a t o r y and c i r c u l a t o r y a d a p t a t i o n and g y n a e c o l o g i c a l and p s y c h i a t r i c a s p e c t s . ACTA P a e d i a t r i c s Supplementum, 1963, 147, 1-75 A s t r a n d , P. 0., & Rodahl, K. Textbook o f work p h y s i o l o g y . New York: McGraw H i l l , 1970. Atwood, C. L., & Taube, A. E s t i m a t i n g mean time to re a c h a m i l e s t o n e , u s i n g r e t r o s p e c t i v e d a t a . B i o m e t r i c s , 1976, 32, 159-172. B a i l e y , D. A. E x e r c i s e , f i t n e s s and p h y s i c a l e d u c a t i o n f o r the growing c h i l d : A c o n c e r n . Paper p r e s e n t e d a t the meeting o f the N a t i o n a l C o n f e r e n c e on the C h i l d i n S p o r t and P h y s i c a l A c t i v i t y , K i n g s t o n , May 1973, pp. 1-23. B a j i n , B. P h y s i c a l a b i l i t y and s t a n d a r d moves manual. U n p u b l i s h e d b u l l e t i n , Canadian Gymnastics F e d e r a t i o n , S p e c i a l b u l l e t i n f o r c o a c h e s , No. 3, September 1977. ( A v a i l a b l e from Women's T e c h n i c a l Committee JJ333 R i v e r Rd., V a n i e r , Ottawa, K1L 8 B 9 ] ) . B a j i n , B. E l i t e program IV. U n p u b l i s h e d b u l l e t i n , Canadian Gymnastics F e d e r a t i o n , 1978. ( A v a i l a b l e from Women's T e c h n i c a l Committee [333 R i v e r Rd., V a n i e r , Ottawa, K1L 8 B 9 ] ) . Baker, I . A., Hughes, J . , & J o n e s , J1. Temporal v a r i a t i o n i n the h e i g h t o f c h i l d r e n d u r i n g the day. The L a n c e t , June 17, 1978, p. 1320. Bakker, H. K., & Struikenkamp, R. S. B i o l o g i c a l v a r i a b i l i t y and l e a n body mass e s t i m a t e s . Human B i o l o g y , 1977, 4 9 ( 2 ) , 187-203. 195 Barnes, L. Preadolescent training: How young is too young? The Physician and Sportsmedicine, 1979, 7.(10), 114-119 Bar-Or, 0 . , Zwiren, L. D., & Ruskin, H. Anthropometric and developmental measurements of 11 and 12 year old boys as predictors of performance 2 years later . ACTA Paediatrica Belgica Supplement, 1974, 28, 214-220 Behnke, A. R. Anthropometric fractionation of body weight. Journal of Applied Physiology, 1961, 16(6), 949-954. Behnke, A. R. Anthropometric estimate of body s ize , shape, and fat content. Postgraduate Medicine, 1963, 34, 190-198. Behnke, A. R., & Royce, J . Body s i ze , shape and composition of several types of athletes: Original papers. Journal of Sports Medicine and Physical Fitness - Quarterly Review, 1966, 6.(2), 75-88. Behnke, A. R., & Wilmore, J . H. Evaluation and regulation of body build and composition, Eaglewood C l i f f s : Prentice Ha l l , 1974. B i e l i c k i , T . , & Waliszko, A. Wroclaw growth study: Part 1, females. Studies in Physical Anthropology, * 1975, Z, 53-83. Bi l lewicz, W. Z . , Fellowes, H. M. & Hytten, C. A. Comments on the c r i t i ca l metabolic mass and the age of menarche. Annals of Human Biology, 1976, 3(1), 51-59. Bojlen, K., & Bentzon, M. W. Seasonal variation in the occurrence of menarche. Danish Medical Bul le t in , 1974, 21(5), pp. 161-168. Bonnet, F. P. , Rocour-Brumioul, D., & Heuskin, A. Regional variations of adipose cel l size and local ce l lu lar i ty in human subcutaneous fat during normal growth. ACTA Paediatrica Belgica, 1979, 32, 17-27. Brozek, J . Body composition: The relative amounts of fa t , t issue, and water vary with age, sex, exercise, and nutritional state. Science, 1961, 134, 920-930 Buckler, J . M. H. , & Brodie, D. A. Growth and maturity characteristics of schoolboy gymnasts. Annals of Human Biology, 1977, 4_(5), 455-463 Bugyi, B. Lean body weight estimation in 6 to 16 year old chi ldren, based on wrist breadth and body weight. Journal of Sports Medicine and Physical Fitness, 1972, 12(1), 171-173. Bugyi, B . , & Kausz, I. Radiographic determination of the skeletal age of young swimmers. Journal of Sports Medicine and Physical Fitness, 1970, 10, 269-270. Bullen, B. A . , Quaade, F . , Olesen, E . , & Lund, S. A. Ultrasonic reflections used for measuring subcutaneous fat in humans. Human Biology, 1965, 37, 375-384. 196 Bui lough, P. , Goodfellow, J . , & O'Connor, J . The relationship between degenerative changes and load-bearing in the human hip. Journal of Bone and Joint Surgery, 1973, 55(B), 746-758. Burkinshaw, L . , Jones, P. R. M., & Krupowicz, D. W. Observer error in skin-fold thickness measurements. Human Biology, 1973, 45(2), 273-279. Burre l l , R. J . W., Tanner, J . M., & Healy, M. J . R. Age at menarche in South African Bantu g i r ls l iv ing in the Transkei reserve. Human Biology, 1961, 33, 250-261. Cameron, N. Weight and skinfold variation at menarche and the c r i t i ca l body weight hypothesis. Annals of Human Biology, 1976, 3_(3), 279-282. Carter, J . E. L. The somatotypes of athletes: A review. Human Biology, 1970, 42(4), 535-569. Carter, J . E. L. The Heath-Carter somatotype method: Manual. San Diego: San Diego State University, 1975. Carter, J . E. L. Montreal Olympic Games anthropological project: Anthropometric instruments and measurements. Unpublished manuscript, San Diego State University, May 12, 1976 (revised November 1976). (Available from Kinesiology Department, [\"Simon Fraser University, Burnaby, B. C . , Canada, V5A 1S6]'). Carter, J . E. L . , Sleet, D. A . , & Martin, G. N. Somatotypes of male gymnasts. Journal of Sport Medicine and Physical Fitness, 1971, 11(3), 162-171. Charzewska, J . , Ziemlanski, S . , & Lasecka, E. Menarcheal age, nutr i t ion, and socio-economic environment. Studies in Physical Anthropology, 1975, 2, 47-51. Chovanova, E. Anthropometric characterization of the best Czechoslovak Ice-Hockey Players. ACTA Facultatis Rerum Naturalium Universitatis Comenianae Anthropologic, 1972, 20, 87-100. Church, K. R. Children's exercise: Facts and f i c t i o n . Elementary School Journal, 1976, 77(2), 110-115. Clark, S. \"Tracey\": Diamond on ice . Canadian Weekend Magazine, March 8, 1980, pp. 2-4;6. Clarke, H. H. Characteristics of the young athlete: A longitudinal look. In Council of Kinesiology, Kinesiology Review. Washington: American Association of Health, Physical Education and Recreation, 1968, pp. 33-42. Clarke,,H. H. Characteristics of athletes. Physical Fitness Research Digest, 1973, 3(2), 1-20. Clegg, E. J . , & Kent, C. Skinfold compressibility in young adults. Human Biology, 1967, 39, 418-429. 197 Crawford, J . D., & Osier, D. C. Body composition at menarche: The Frisch-Revelle hypothesis rev is i ted. Pediatr ics, 1975, 56(3), 449-458. Cureton, K. J . , Boileau, R. A . , & Lohman, T. G. A comparison of densito-metric, potassium-40, and skinfold estimates of body composition in prepubescent boys. Human Biology, 1975, 47(3), 321-336. Damon, A. Larger body size and ear l ier menarche: The end may be in sight. Social Biology, 1974, 21(1), 8-11. Damon, A . , Damon, R., Reed, R., & Valadian, I. Age at menarche of mothers and daughters with a note on accuracy of r e c a l l . Human Biology, 1969, 41, 160-175. Damon, A . , & Goldman, R. F. Predicting fat from body measurements: Densitometric validation of ten anthropometric equations. Human Biology, 1964, 36, 32-44. Dann, T. C , & Roberts, D. F. End of the trend? A 12 year study of age at menarche. Br i t ish Medical Journal, 1973, 3_, 265-267. De Garay, A. L . , Levine, L . , & Carter, J . E. L. Genetic and anthropological studies of Olympic athletes. New York: Academic Press, 1974. De Wijn, J . F. Estimation of age at menarche in a population. In H. E. Stenfert, & N. V. Kroese, Somatic growth of the ch i ld . Proceedings of a Boerhaave Course for postgraduate medical teaching, Leiden, Holland, 1966, pp. 16-24. Draz i l , V. Gymnastics: Physical ac t iv i ty , sports, games, and exercises. In L. Larson, & D. Herrmann (Eds.) , Encyclopedia of sports sciences and medicine. New York: MacMillan, 1971, pp. 353-354. Drinkwater, D. T . , & Ross, W. D. Anthropometric fractionation of body mass. In M. Ostyn, G. Beunen, & J . Simons (Ed.) , Kinanthropometry 2. Baltimore: University Park Press, 1980, pp. 177-188. Durnin, J . V. G . , & Rahaman, M. M. The assessment of the amount of fat in the human body from measurements of skinfold thickness. Bri t ish Journal of Nutri t ion, 1967, 21, 681-689. Durnin, J . V. G . , & Womersley, J . Body fat assessed from total body density and i ts estimation from skinfold thickness measurements on 481 men and women aged from 16 to 72 years. Bri t ish Journal of Nutri t ion, 1974, 32, 77-97. Edwards, D. A. W. Observations on the distr ibution of subcutaneous fat . Cl inical Science, 1950, 9_, 259. Edwards, D. A. W. Differences in the distr ibution of subcutaneous fat with sex and maturity. Cl in ical Science, 1951, 10_, 305-315. 198 Eiben, 0. G. The physique of woman athletes. Budapest: Hungarian Scient i f ic Council for Physical Education, 1972. Eiben, 0. G. Changes in body measurements and proportion of children: Based on Kormend growth study. In L. Gedda, & P. Pari si (Ed.) , Auxology: Human growth in health and disorder. Proceedings of the Serno Symposia, London: Academic Press, 1978, 13, pp. 187-198. Eiben, 0. G . , Ross, W. D., Christensen, W., & Faulkner, R. A. Proportionality characteristics of female athletes. Anthropologiai Kozlenenyek, 1976, 20, 55-67. Erdely i , G. J . Problems and inquir ies: Gynecological survey of female athletes. Journal of Sports Medicine and Physical Fitness, 1962, 2, 174-179. Eveleth, P. B. , & De Souza Frei tas, J . A. Tooth eruption and menarche of Brazil ian born children of Japanese ancestry. Human Biology, 1965, 37, 176-184. Eveleth, P. B. , & Tanner, J . M. Worldwide variation in human growth. Cambridge: Cambridge University Press, 1976. Faulkner, R. Physique characteristics of Canadian figure skaters. Unpublished master's thesis, Simon Fraser University, 1977. F i l ipsson, R., & Ha l l , K. Correlation between dental maturity, height development and sexual maturation in normal g i r l s . Annals of Human Biology, 1976, 3_(3), 205-210. Finn, J . D. Multivariance (Code: GENL-02). Faculty of Education, Research Service Centre, University of Br i t ish Columbia, November 1978.' Forbes, G. B. Growth of the lean body mass in man. Growth, 1972, 36, 325-338. Friis-Hansen, B. Body composition during growth. Pediatr ics, 1971, 47(1), 264-274. Fr isch, R. Influences on age of menarche. The Lancet, May 5, 1973, p. 1007. Fr isch, R. E. Fatness of g i r ls from menarche to age 18 years with a nomogram. Human Biology, 1976, 48_(2), 353-359. Fr isch, R. E . , & McArthur, J . W. Menstrual cycles: Fatness as a. determinant of minimum weight for height necessary for their maintenance or onset. Science, 1974, 185(4155), 949-951. Fr isch, R. E . , & Revelle, R. The height and weight of adolescent boys and g i r ls at the time of peak velocity of growth in height and weight: Longitudinal data. Human Biology, 1969, 41_, 536-559. 199 Fr isch, R. E . , & Revelle, R. Height and weight at menarche and a hypothesis of c r i t i ca l body weights and adolescent events. Science, 1970, 169, 397-399. Fr isch, R. E . , & Revelle, R. Height and weight at menarche and a hypothesis of menarche. Archives of Disease in Childhood, 1971, 46, 695-701. Fr isch, R. E . , Revelle, R., & Cook, S. Components of weight at menarche and the in i t ia t ion of the adolescent growth spurt in g i r l s : Estimated total water, lean body weight, and fat . Human Biology, 1973, 45_(3), 469-483. Garn, S. M., & Gorman, E. L. Comparison of pinch-caliper and teleroentgeno-grammetric measurements of subcutaneous fat . Human Biology, 1965, 28, 407-413. Garn, S. M., & Haskell , J . A. Fat thickness and developmental status in childhood and adolescence. Journal of Diseases of Children, 1960, 99, 746-751. George, G. S. Selected execution techniques common to aerial gymnastic s k i l l s . In J . Salmela, The advanced study of gymnastics. Springfield: Charles, C. Thomas, 1976, pp. 138-147. Goldfarb, A. F. Puberty and menarche. Cl in ical Obstetrics and Gynecology, 1977. 20(3), 625-631. Greenberg, B. G . , & Bryan, A. H. Methodology in the study of physical measurements of school children. Human Biology, 1951, 23j l ) , 160-179. Greulich, W. W., & Pyle, S. I. Radiographic atlas of skeletal development of the hand and wrist (2nd ed.) . Stanford: Stanford University Press, 1970. Guyton, A. C. Textbook of medical physiology (5th ed. ) . Philadelphia: W. B. Saunders, 1976. Halm, J . Simple data description (adapted from UCLA BMD documentation). University of Br i t ish Columbia Computing Center, (September 1971), revised April 1974. Hammond, W. H. Measurement and interpretation of subcutaneous fat with norms for children and young adult males. Bri t ish Journal of Preventive Social Medicine, 1955, 9, 201-211. Hebbelinck, M., & Ross, W. D. Kinanthropometry and biomechanics. In R. C. Nelson, & C. A. Morehouse (Eds.) , Biomechanics 4, Baltimore: University Park Press, 1974, pp. 537-552. Hillman, R. W., Slater, P. , & Nelson, M. J . Season of b i r th , parental age, menarcheal age and body form: Some inter-relationships in young women. Human Biology, 1970, 42, 570-580. 200 Himes, J . H. , Roche, A. F . , & Siervogel, R. M. Compressibility of skinfolds and the measurement of subcutaneous fatness. American Journal of Cl in ical Nutr i t ion, 1979, 32, 1734-1740. Hirata, K. Physique and age of Tokyo champions. Journal of Sports Medicine and Physical Fitness, Quarterly Review, 1966, 6, 207-222. Hirsch, J . Can we modify the number of adipose cel ls? Postgraduate Medicine, 1972, 51, 83-86. Ismail, A. H. Body composition, In L. Larson, & D. Herrmann (Eds.) , Encyclopedia of sports sciences and medicine. New York: MacMillan, 1971, pp. 1209-1211. Jackson, A. S . , & Pollock, M. L. Factor analysis and multivariate scaling of anthropometric variables for the assessment of body composition. Medicine and Science in Sports, 1976, 8_(3), 196-203. Jafarey, N. A . , Khan, M. Y . , & Jafarey, S. N. Role of a r t i f i c i a l l ighting in decreasing the age of menarche. The Lancet, August 29, 1970, p. 471. Johnson, L. V. Effects of 5-day a week, vs 2, and 3-day a week physical education class on f i tness , s k i l l , adipose t issue, and growth. Research Quarterly, 1969, 40(1), 93-98. Johnson, W. R., & Buskirk, E. R. Developmental aspects: Maturing and aging in relation to exercise and sport. In W. R. Johnson (Ed.) , Science and medicine of exercise and sport (2nd ed. ) . New York: Harper and Row, 1974, pp. 330-333. Johnson, P. B. , Updyke, W. F . , Schaefer, M., & Stolberg, D. C. Sport, exercise and you. New York: Holt Rhinehart, and Winston, 1975. Johnston, F. E. The concept of skeletal age. Cl in ical Pediatr ics, 1962, 1_(3), 133-144. Johnston, F. E. Control of age at menarche. Human Biology, 1974, 46(1), 159-171. Johnston, F. E . , Malina, R. M., & Galbraith, M. A. Height, weight, and age of menarche and the c r i t i ca l weight hypothesis. Science, 1971, 174, 1148-1149. Johnston, F. E . , Roche, A. F . , Schel l , L. M., & Wettenhall, H. N. Cr i t ica l weight at menarche: Original a r t i c les . American Journal of Diseases of Children, 1975, 129, 19-23. Katch, F. I., & Michael, E. D., J r . Prediction of body density from skin-fold and girth measurements of college females. Journal of Applied Physiology, 1968, 25, 92-94. 201 Keys, A . , & Brozek, J . Body fat in adult man. Physiological Reviews, 1953, 33(3), 245-325. Khosla, T. Unfairness of certain events in the Olympic Games. Bri t ish Medical Journal, 1968, 4, 111-113. Khosla, T. Will the Olympics survive? Bri t ish Journal of Sports Medicine 1977, 11(1), 20-25. Kowalski, C. J . A commentary on the use of multivariate s ta t is t ica l methods in anthropological research. American Journal of Physical Anthropology, 1972, 36, 119-132. Kraij-Cercek, L. The influence of food, body bui ld , and social origin on the age at menarche. Human Biology, 1956, 28(4), 393-406. Kr iesel , G. Interrelationships between some somatic characteristics and hand strength of 14 to 18 year old boys and g i r l s . Symposia Biologica Hungarica (Budapest), 1977, 20, 291-298. Krustev, K. Age limits for the Olympics. Comite Olympique Bulgare, Bullet in D'Information, July 1977, 25, pp. 25-29. La Cava, G. Sport and puberty. The Journal of Sports Medicine and Physical Fitness, Quarterly Review, 1974, 14(3), 163-164. Lee, M. M. C . , & Ng, C. K. Postmortem studies of skinfold caliper measure-ments and actual thickness of skin and subcutaneous t issue. Human Biology, 1965, 37, 91-103. Le Veau, B. , Ward, T . , & Nelson, R. C. Body dimensions of Japanese and American gymnasts. Medicine and Science in Sport, 1974, 6_(2), 146-150. Lewis, A. S. Physique as a determinant of success in sport with particular reference to Olympic oarsmen. New Zealand Journal of Health, Physical Education and Recreation, 1969, Z{3), 5-21. Malina, R. M. Exercise as an influence upon growth: Review and crit ique of current concepts. Cl in ical Pediatr ics, 1969a, 8_, 16-26. Malina, R. M. Quantification of fa t , muscle, and bone in man. Cl inical Orthopedics and Related Research, 1969b, 6_5, 9-38. Malina, R. M. Adolescent changes in s ize , bui ld , composition, and performance. Human Biology, 1974, 46(1), 117-131. Malina, R. M. Physical anthropology, physical ac t iv i ty , and sport. Canadian Journal of Applied Sport Sciences, 1976, 1_, 155-161. Malina, R. M., Harper, A. B . , Avent, H. H., & Campbell, D. E. Age at menarche in athletes and non-athletes. Medicine and Science in Sports, 1973, 5_(1), 11-13. 202 Malina, R. M., Spirduso, W. W., Tate, C . , & Baylor, A. M. Age at menarche and selected menstrual characteristics in athletes at different competitive levels and in different sports. Medicine and Science in Sports, 1978, 10(3), 218-222. Malina, R. M., & Zavaleta, A. N. Androgyny of physique in female track and f ie ld athletes. Annals of Human Biology, 1976, 3(5), 441-446 Maresh, M. M v A for ty- f ive year investigation for secular changes in physical' maturation. American Journal of Physical Anthropology, 1972, 36, 103-110. Marshall, W. A. Growth in children: Puberty in normal g i r l s . Update, 1972, 4, 1345-1352. Marshall, W. A. Late onset of puberty. Br i t ish Medical Journal, 1974, 4_, 310. McGregor, I. A . , Thomson, A. M., & Bi l lewicz, W. Z. The development of primary teeth in children from a group of Gambian vil lages and c r i t i ca l examination of i ts use for estimating age. Bri t ish Journal of Nutr i t ion, 1968, 22, 307-314. McNeill , D., & Livson, N. Maturation rate and body build in women. Chi Id Development, 1963, 34, 25-32. Medawar, P. B. The shape of the human being as a function of time. Proceedings of the Royal Society of London, 1945, 132, pp. 133-141. Medved, R. Body weight and predisposition for certain sports. Journal of Sports Medicine and Physical Fitness, 1966, 6{2), 89-91. Montpetit, R. R. Physiology of gymnastics. In J . Salmela, The advanced study of gymnastics. Springfield: Charles, C. Thomas, 1976. pp. 183-217. Morales, M. F . , Rathbun, E. N., Smit, R. E . , & Pace, N. Studies on body composition: Theoretical considerations regarding the major body tissue components with suggestions for application to man. Journal of Biological Chemistry, 1945, 158, 677-691. Motajova, J . Effect of special stress over a four-year period on certain morphological parameters and bone age in growing children. Folia Morphologica Praha, 1974, 22(4), 358-361. Nelson, R. C. Comparison of the body dimensions of Japanese, Hungarian, and American gymnasts. Gymnast Magazine, October 1974, 16J10), pp. 44-46, Newton, M. B . , & De Issekutz-Wolsky, M. The effect of parental age on the rate of female maturation. Gerontologia, 1969, _1_5( 16), 328-331. Nichols, H. Peti te, e l i t e : Good gymnasts come in small packages. Women's Sports, March 1979, pp. 17-19. 203 Novak, L. P. , Bierbaum, M., & Mellerowicz, H. Maximal oxygen consumption, pulmonary function, body composition, and anthropometry of adolescent female athletes. Internationale Zei tschr i f t fur Anglewandte Physiologie (einschlieBlich Arbeitsphysiologie), 1973, 31, 103-119. Novak, L. P . , Woodward, W. A . , Best i t , C , & Mellerowicz, H. Working capacity, body composition, and anthropometry of Olympic female athletes. Journal of Sports Medicine and Physical Fitness, 1977, 17_, 275-283. Onat, T . , & Numan-Cebeci, E. Sesamoid bones of the hand: Relationships to growth, skeleta l , and sexual development in g i r l s . Human Biology, 1976, 48(4), 659-676. Parizkova, J . The development of subcutaneous fat in adolescents and the effect of physical training and sport. Physiologia Bohemoslovenica, 1959, 8, 112-119. Parizkova, J . Total body fat and skinfold thickness in children. Metabolism, 1961a, 10, 794-807. Parizkova, J . Age trends in fat ; in normal and obese children. Journal of Applied Physiology, 1961b, 16, 173-174 Parizkova, J . Impact of age, d ie t , and exercise on man's body composition. Annals of the New York Academy of Sciences, 1963, 110, 671-673. Parizkova, J . Longitudinal study of the development of body composition and body build in boys of various physical act iv i ty . Human Biology, 1968a, 40, 212-225. Parizkova, J . Body composition and physical f i tness. Current Anthropology, 1968b, 9(4), 273-287. Parizkova, J . , & Poupa, 0. Some metabolic consequences of adaptation to muscular work. Br i t ish Journal of Nutri t ion, 1963, 17^ , 341-345. Petersen, A. C. Can puberty come any earl ier? Psychology Today Magazine, 1979, 11(9), pp. 45-47. Pool, J . , Binkhorst, R. A . , & Vos, J . A. Some anthropometric and physiological data in relation to performance of top female gymnasts. Internationale Zei tschr i f t fuer angewandte Physiologie Einschl iessl ich Arbeitsphysiologie, 1969, 27(4), 329-338. Popham, W. J . , & Sirotnik, K. A. Educational s ta t i s t i cs : Use and Inter- pretations (2nd ed. ) . New York: Harper and Row, 1973, Quinn, H. With a l i t t l e help from their fr iends! MacLean's Magazine, January 8, 1979, p. 50 . Ra l is , Z . , & McKibbin, B. Changes in shape of the human hip joint during i ts development and their relation to i ts s tab i l i t y . Journal of Bone and Joint Surgery, 1973, 55(B), 780-785. 204 Reiter, E. 0 . , & Kulinh, E. Sexual maturation in the female. Pediatric Cl inics of North America, 1972, 19(3), 581-587. Reynolds, E: L. The distr ibution of subcutaneous fat in childhood and adolescence. Monographs of the Society for Research in Child Development, 1950, 15(2, Serial No.52), pp. 1-189. Rona, R., & Pereira, G. Factors that influence age of menarche in g i r ls in Santiago, Chile. Human Biology, 1974, 46(1), 33-42. Ross, W. D. Personal communication, unpublished data, 1980. Ross, W, D., Brown, S. R., Hebbelinck, M., & Faulkner, R. A. Kinanthro-pometric terminology and landmarks: A report based on agreements reached by a Leon and Thea Koerner Foundation study group held at the University of Br i t ish Columbia, July 1973: Presented to the International Committee for the Standardization of Physical Fitness Tests (Trois Rivieres, July 1976). In R. J . Shephard, & H. Lavallee, Physical f itness assessment. Springfield: Charles, C. Thomas, 1978, pp. 44-49. Ross, W. D., Brown, S. R., Faulkner, R. A . , & Savage, M. V. Age of menarche of e l i te Canadian skaters and skiers. Canadian Journal of Applied Sport Sciences, 1976, 1(3), 191-193. Ross, W. D., & Day, J . A. P. Physique and performance of young skiers. Journal of Sports Medicine and Physical Fitness, 1972, VI, 30-37. Ross, W. D., Drinkwater, D. T . , Whittingham, N., & Faulkner, R. A. Anthropometric prototypes: Ages 6 to 18 years. In K. Berg, & B. 0. Erikson (Ed.) , Pediatric Work Physiology 9. Baltimore: University Park Press, 1980, pp. 3-12. Ross, W. D. & Marshall, G. R. Physique characteristics and the child athlete. Recreational Research Review, February 1979, pp. 13-17. Ross, W. D., Marshall, G. R., Vajda, A . , & Roth, K, D, Some comments on kinanthropometry and theoretical dimensional relationships. Dimensionalitat Und Proportionalitat Von Kraftleistungen: Leiftungsstort, 1978, 8, 1-8. Ross, W. D., McKim, D. R., & Wilson, B. D. Kinanthropometry and young skiers. In A. Taylor (Ed.) , Application of science and medicine in sport. Springfield: Charles, C. Thomas, 1976, pp. 257-277. Ross, W. D., & Wilson, N. C. Growth and development: A stratagem for proportional growth assessment. ACTA Paediatrica Belgica Supplement, 1974, 28, 169-182. Rozin, E. U. The influence of anthropometric parameters on successful learning in gymnastics. Yessis Review, 1974, 9_(1), 16-21. Salmela, J . H. The advanced study of gymnastics. Springfield: Charles, C. Thomas, 1976. 205 Salmela, J . , Halle, M., Petiot, B . , & Samson, R. Determinants of Olympic gymnastic performance. Proceedings of the International Congress of Physical Act iv i ty Sciences, Quebec, July 1976, 1_, pp. 169-177. Shephard, R. J . , Jones, J . , I sh i i , K., Kanek, M., & Albecht, A. J . Factors affecting body density and thickness of subcutaneous fat . American Journal of Cl in ical Nutri t ion, 1969, 22, 1175-1189. Sinning, W. E. Body composition assessment of college wrestlers. Medicine and Science in Sports, 1974, 6(2), 139-145. Sinning, W. E . , & Lindberg, G. D. Physical characteristics of college age women gymnasts. Research Quarterly, 1972, 43(2), 226-234. Sker l j , B . , Brozek, J . , & Hunt, E . , J r . Subcutaneous fat and age changes in body build and body form in women. American Journal of Physical Anthropology, 1953, 11, 577-600. Slaughter, M. H. , Lohman, T. G . , & Boileau, R. A. Relationship of anthro-pometric dimensions to lean body mass in children. Annals of Human Biology, 1978, 5.(5), 469-482. Smit, P. J . Anthropometric observations on South African gymnasts. South Africa Medical Journal, 1973, 47, 480-485. Smith, N. J . Gaining and losing weight in athlet ics. Journal of the American Medical Association, 1976, 236(2), 149-151. Sprynarova, S . , & Parizkova, J . Comparison of the functional circulatory and respiratory capacity in g i r l gymnasts and swimmers. Journal of Sports Medicine, 1969, 9_, 165-172. Tanner, J . M. Growth at adolescence (2nd ed. ) , Oxford: Blackwell Sc ient i f ic Publications, 1962. Tanner, J . M. The regulation of human growth. Child Development, 1963, 34, 817-847. Tanner, J . M. The physique of the Olympic athlete. London: George Al len , and Unwin L t d . , 1964. Tanner, J . M. Foetus into man: Physical growth from conception to maturity. Cambridge: Harvard University Press, 1978. Tanner, J . M, & Whitehouse, R. H. Cl in ical longitudinal standards for height, weight, height veloci ty , weight veloci ty , and stages of puberty. Archives of Disease in Childhood, 1976, 51(3), 170-179. Tanner, J . M., Whitehouse, R. H., Marshall, W. A . , Healy, M. J . R., & Goldstein, H. Assessment of skeletal maturity and prediction of adult height (TW2 Method). London: Academic Press, 1975. 206 Taunton, C. Children and running: The experts speak. B. C. Runner Magazine, Winter 1979/80, 5_, pp. 15-23. Tosovsky, J . , Prokopec, M., & Mejsnarova, B. Growth curves and their application to the growth of anthropometric characteristics of an individual during ontogenesis. ACTA Facultatis Medicae Universitatis Brunensis, 1976, 57, 243-252. Va l l ie re , A. Biomechanical misconception in a r t i s t i c gymnastics. In J . Salmela, The advanced study of gymnastics. Springfield: Charles, C. Thomas, 1976, pp. 93-107. Webster's seventh new collegiate dictionary: Based on Webster's third new international dictionary. Springfield: G. & C. Merriam Co. , 1970. Well, J . B . , Jok l , E . , & Bohranen, J . The effect of intensive physical training upon body composition of adolescent g i r l s . American Corrective Therapy Journal, 1963, 1_7(3), 68-72. Whittingham, N. Anthropometric prototypes for boys and g i r ls aged 6 to 18, exemplified by structure analysis of sub-12 year old figure skaters. Unpublished master's thesis, Simon Fraser University, 1978. Wilmore, J . H, , & Behnke, A. R. Predictabi l i ty of lean body weight through anthropometric assessment in college men. Journal of Applied Physiology, 1968, 25(4), 349-355. Wilmore, J . H. , & Behnke, A. R. An anthropometric estimation of body density and lean body weight in young women. American Journal of Cl in ical Nutri t ion, 1970, 23, 267-274. Wilmore, J . H., Girandola, R. N., & Moody, D. L. Val idity of skinfold and girth assessment for predicting alterations in body composition. Journal of Applied Physiology. 1970a, 29(3), 313-317. Wilmore, J . H., Royce, J . , Girandola, R. N., Katch, F. I., & Katch, V. L. Body composition changes with a 10 week program of jogging. Medicine and Science in Sports, 1970b, 2(3), 113-117. Womersley, J . , & Durnin, J . V. G. An experimental study on var iab i l i ty of measurements of skinfold thickness on young adults. Human Biology, 1973, 45(2), 281-292. Wood, T. Computer program: Calculation of proportional mass values, January 1980. Youngren, D. H. A relationship of selected anthropometric measurements to performance of the U. S. women gymnastic athletes in the 1968 Olympic t r i a l s . Unpublished master's thesis, California State College at Long Beach, 1969. 207 Yuhasz, M. S. Fat measurements and c lass i f i ca t ion , and a method for the individual prediction of optimal body-weight. In M. S..Yuhasz, Physical f itness laboratory manual. London: University of Western Ontario, 1977, p. 158. Zacharias, L . , Rand, W. M., & Wurtman, R. J . A prospective study of sexual development and growth in American g i r l s : The s ta t is t ics of menarche. Obstetrical and Gynecological Survey, 1976, 31(4), 325-337. Zuk, G. H. The p last ic i ty of the physique from early adolescence through adulthood. Journal of Genetic Psychology, 1958, 92, 205-214. Zwiren, L . , Skinner, J . S . , & Buskirk, E. R. Use of body density and various skinfold equations for estimating small reductions in body fatness: Original papers. Journal of Sports Medicine and Physical Fitness, 1973, 13(1), 213-218. APPENDIX A 208 APPENDIX A THE ABILITY OF THE DRINKWATER TACTIC* (ANTHROPOMETRIC FRACTIONATION OF BODY MASS) TO ESTIMATE OBSERVED BODY MASS IN THE PRESENT SAMPLE OF GYMNASTS OBM = Obtained Body Mass = Scale Weight PBM = Predicted Body Mass = FAT MASS + MUSCLE MASS + SKELETAL MASS + RESIDUAL MASS DIFF = (PBM - OBM) = Difference % ERR = DIFF Y 1 n n = % Error OBM A ; i U U Group 1 2 3 4 1+2 1+2+3 OBM X 43.7 44.1 48.7 50.0 43.9 45.9 PBM X 41.9 41.5 46.1 47.1 41.7 43.5 DIFF X -1.8 -2.6 -2.6 -2.9 -2.2 -2.4 % ERR X 4.3 6.3 5.6 6.2 5.3 5.5 * Drinkwater & Ross, 1980 ro o 210 APPENDIX A CONTINUED The Drinkwater Tactic systematically underestimated obtained body mass (scale weight) in the present sample of female gymnasts. The original formulae (applied to the present sample) were based on deviations from a unisex adult model, and i t is hypothesized that the specif ic anthropometric variables, selected to represent the individual masses, do not ref lect these masses in children as closely as they do in adults. "@en . "Thesis/Dissertation"@en . "10.14288/1.0077235"@en . "eng"@en . "Physical Education"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Anthropometric and maturational assessment of female gymnasts from varying performance levels"@en . "Text"@en . "http://hdl.handle.net/2429/22834"@en .