Physical activity has been associated with a variety of cognitive and behavioural changes which, collectively, allow the conclusion that regular exercise contributes to psychological as well as physical health (Folkins & Sime, 1981; Veale, 1987). Thus, while concern has not moved away entirely from whether certain types of exercise have psychological benefits (e.g. Hughes, 1984) considerable attention is now being paid to how these benefits are mediated, and at whom they might best be targeted. Retired and elderly people in particular are being seen increasingly as appropriate candidates for health promotion initiatives which, by increasing levels of physical activity (and, by implication, levels of physical fitness), aim broadly to improve quality of life.

In this context, levels of customary or habitual activity (as distinct from levels of formal exercise participation) are presumed to play an important part, and are now beginning to receive research attention (Shephard & Montelpare, 1988). At present, however, the empirical basis for activitybased health promotion initiatives among elderly people owes much to information derived from younger age groups. Certainly, as regards mental health, relatively little research has directly addressed the assumption that customary physical activity reliably contributes to psychological wellbeing in later life.

The Nottingham Longitudinal Study of Activity and Ageing was set up in 1983 to assess the role of lifestyle and Customary Physical Activity (CPA) in promoting and maintaining mental health and psychological well-being in later life.

What is health?

Health is a term that is widely used (for example, in the title of this symposium) but the underlying assumption that readers share a common understanding of its meaning is rarely justified (Seedhouse, 1986). Before the validity of health measurements can be discussed, the meaning of health must be clarified. A useful starting point is the definition of health included in the charter of the World Health Organization (WHO, 1946): ‘Health is a state of complete physical, mental and social well being and not merely the absence of disease or physical infirmity’.

As a statement of ideals it cannot be faulted, but it is doubtful that it describes an attainable goal and it does not offer a useful basis for measurement. A different definition is ‘Health is a state in which a person is enabled to work to fulfil their realistic chosen and biological potential’ (Seedhouse, 1986). From a practical point, this is much to be preferred, recognizing as it does that different individuals have different potentials. However, for purposes of measurement, it complicates things even further requiring not only the present state of each individual to be assessed but also their theoretical potential.

None the less, three important characteristics of health emerge from these definitions:

I am fortunate to have been able to write this book. This good fortune began with my appointment to the Fels Staff 23 years ago. It has taken me all that time to understand the complexities of the Fels Longitudinal Study. Writing this book about the study has been a pleasure. I trust the reader will find it pleasant also. Quoting Daniel Defoe: ‘If this work is not both pleasant and profitable to the reader… the fault… cannot be any deficiency in the subject.’

Many helped. Some added to the quality of the study, particularly the generous participants and their relatives, outstanding collaborators and consultants, efficient secretaries and dedicated research assistants. Particular thanks are due to Ruth Bean and Lois Croutwater who, during a joint span of more than 56 years of extraordinary effort, have organized the examinations and maintained contact with the participants. Doctors Lester Sontag and Frank Falkner, the past Directors, provided effective leadership from 1929 to 1979. The Fels Longitudinal Study, supported in its early years by the Samuel S. Fels Fund, continues as part of the Division of Human Biology of the Department of Community Health at Wright State University. This division receives enthusiastic support from the Departmental Chairman (Dr Robert Reece), the School of Medicine and the central administration of Wright State University. Continuation of this support is important because the story is not complete. The significance of the Fels Longitudinal Study increases as the data base enlarges, the serial records become longer, and new techniques are introduced. Additionally, the focus of the Study is shifting rapidly to more applied areas.

Many helped write this book, but they are not to blame for its defects.

The future of the Fels Longitudinal Study is uncertain. One can look ahead with great expectations but it is ‘foolish to look further than you can see,’ as stated by Winston Churchill. The study is more exciting than ever: more data, more brilliant young investigators, more health-related and socially significant questions to be addressed. Consequently, the near future is viewed with optimism that is based, in part, on the knowledge that what we have done in the last few decades has contributed to knowledge, met with the approval of review groups and has received support from the Federal Government and others.

It is impossible, however, to predict how long this happy state of affairs will last. The National Institutes of Health, which provide most of the funding for the Fels study, have support cycles that do not extend longer than 5 years and funding is highly competitive. Therefore, there is some apprehension. Our view of the Fels study is biased and scientific progress becomes more difficult when all the low apples have been picked. Anything may occur. As a Swedish proverb states: ‘The afternoon knows what the morning never suspected.’ Despite uncertainties, we will continue as long as possible. An observational longitudinal study becomes more valuable as its duration increases and more complete descriptions of natural changes become possible, leading to more complete understanding of human development.

I wish to conclude with three quotations that are apposite. Venerable Bede wrote: ‘It is better never to begin a good work than, having begun it, to stop.’

Most of this chapter will be concerned with bones and teeth which share a common embryological origin. The Fels studies of these organs will be grouped as follows: (i) skeletal growth, (ii) growth of specific skeletal regions, (iii) skeletal variations, (iv) skeletal mass, and (v) teeth.

Skeletal growth

Studies of skeletal growth at Fels have attempted to determine the amounts of growth at specific locations and to examine changes in overall dimensions such as length and width. In an early attempt to establish a natural bone marker for use in studies of bone elongation, based on serial radiographs, Pyle (1939) observed the nutrient foramen of the radius. She found the groove leading to the foramen made it difficult to locate a fixed radiographic point, although the foramen could be recognized in about 90% of radiographs of children. Some difficulties were associated with the presence of multiple nutrient foramina in about 20% of radii and the changing location of points at the external end of the obliquely aligned nutrient canal as the cortex thickened. This interesting exploratory study did not establish the nutrient foramen as a suitable fixed point from which length measurements could be made.

Fels data have been analyzed to provide reference data derived for bone lengths and widths measured in serial radiographs. Garn et al. (197293) presented such data for the bones of the hand at annual intervals from 2 through 18 years. These workers defended the use of Fels data on the basis that these data were derived from a well-nourished contemporary population and, therefore, they should be applicable fairly generally to US whites.

The original mission of the Fels Research Institute included the serial study of individuals before birth; this aspect of the mission has not been neglected. Particularly during the early years of the Fels Longitudinal Study, strenuous efforts were made to perform prenatal studies and successes were achieved although the available technology allowed only a narrow range of investigations. Better methods are now available that could assist prenatal studies of growth, maturation and body composition, but some involve radiation (computerized tomography), and others are expensive (ultrasonography, magnetic resonance imaging). Imaging procedures have not been applied serially in normal pregnancies although these studies have great potential.

Some investigations made within the Fels Longitudinal Study that relate to the fetal period are described with physical growth (Chapter 4) and skeletal and dental studies (Chapter 6). The prenatal investigations described in this Chapter have been grouped under the headings: prenatal studies, and familial and genetic studies.

Prenatal studies

Diet and nutrition

During the 1930s and early 1940s, the relationships between prenatal maternal diets and the size of the infant at birth were studied. This was a Herculean task. Daily dietary records were kept by 205 mothers for 4 to 7 months. These mothers were not given dietary advice; this was a ‘natural experiment,’ as is generally true for the observations made in the Fels Longitudinal Study. These, and other dietary data, provided some important information.

Maturation is the process that leads to the achievement of adult maturity. Maturation is a part of development. Both relate to progressive increases in complexity, but maturation is restricted to those developmental changes that lead to the same end point in all individuals. For example, the percentage of adult stature achieved by a child is a measure of maturity: all reach 100% in adulthood. Levels of maturity that are intermediate between the absence of measurable indicators and the adult state indicate the extent to which a child, or a group of children, has proceeded toward the completion of maturation in a particular body system. Maturation occurs in all body systems, organs, and tissues. For example, ‘skeletal maturation’ refers to a set of radiographically visible changes that culminate in the achievement of adult skeletal status by the early 20s.

This broad subject has received considerable sustained attention from Fels scientists. For example, Roche (1974c180) published an introduction to a symposium on adolescent physiology in which variation in maturation was the dominant theme. Much of this variation is associated with differences in rates of maturation among individuals. Attention was also given to the hormonal control of adolescence and the factors that regulate the timing of menarche. Maturation has also been the subject of a review by Chumlea (in press13).

In the Fels Longitudinal Study, as should occur in any long-term serial study, great efforts were made to ensure that the data collected were reliable and that this reliability was retained during the transfer of the data to computers. High levels of data quality can be achieved in a prospective longitudinal study but not in a retrospective study. Additionally, the hypotheses posed and the analyses made in the Fels Longitudinal Study ensured, as far as possible, that the maximum information was derived from the serial nature of the data. Aspects of data management and analysis in the study will be described in the sequence: (i) the need for accurate data, (ii) quality control, (iii) data management, (iv) interpolation, (v) derivation of variables, (vi) transformation of variables, and (vii) statistical analyses.

The need for accurate data

In a cross-sectional study, errors in the measurement of some individuals have little effect on the results of analyses unless these errors are large and common. If it is concluded that outlying values denote abnormal individuals or that large errors occurred during data collection, these data points can be excluded from cross-sectional analyses. This exclusion should be documented, and based on objective rules.

Large errors may be detected by comparing observed data with the distribution of values for the same variable in other groups of children of the same sex and age. For example, a recorded stature of 90 cm for a 6-year-old boy can be recognized as an outlier by comparison with the 5th percentile level (108.5 cm).

The above quotation from the writings of Sir William Osier is fully appropriate. It has been an honor for me to work in the Fels Longitudinal Study. Indeed, all the members of the Fels staff are fully conscious of their debt to those who laid the foundations for the studies that are summarized in this volume. Although much has been achieved, all realize that those who follow will greatly extend our present limited horizons.

This book describes the progress that has been made during the first 60 years of the Fels Longitudinal Study of Growth, Maturation and Body Composition. The remarkable nature of the study justifies this volume. Very few, if any, investigations of human beings are so longlived. Despite its longevity, the Fels Longitudinal Study continually becomes more vigorous and active. Few studies have been responsible for equally important research related to serial changes in physical growth and maturation and body composition in normal individuals.

The Fels Research Institute was founded in 1929 with a single complex research project that came to be called the Fels Longitudinal Study. As the name implies, this was a serial study and it was multidisciplinary.

Interest in body composition at Fels is reflected in the data collection protocol from the first years of the study. This early interest was related to the amounts of adipose tissue, muscle and bone at local sites. The possible practical application of this work was soon realized. Garn (1962c62) wrote: ‘Body composition attracts interest from clinical and preclinical disciplines,’ but later (1963a63) he noted that, despite the established relationship of total body fatness to the probability of death (longevity) and of the amount of bone mineral to osteoporosis, little had been done to associate variations in regional body composition with the risk of disease or the probability of death. Garn (1963a63) also claimed that an excess of muscle mass predisposed to coronary atherosclerosis but tended to protect against osteoporosis.

The Fels Longitudinal Study was transformed in 1976 by placing an increased emphasis on serial changes in total and regional body composition in relation to risk factors for cardiovascular disease. The initial plan for the body composition aspect of the study was complex and further complicated by the irregularity of federal funding. As an overview, data are now collected annually from participants aged from 8 to 18 years who live within 80 km (50 miles) of Yellow Springs and then at 3-year intervals until 40 years, after which the examinations are at 2-year intervals. The intervals between examinations are longer for those who live further from Yellow Springs.

This chapter could be disproportionately long because so much has been achieved at Fels in the area of physical growth. To reduce its length, the growth of bones and teeth and studies of body tissues are described in Chapters 6 and 7 respectively. The present chapter will describe research concerning (i) the development and standardization of anthropometric methods, (ii) age changes in anthropometric variables, (iii) methods of growth assessment, (iv) secular changes in anthropometric variables, (v) determinants of growth, (vi) the final phase of growth, (vii) the prediction of adult stature, (viii) associations between growth and behavioral variables, and (ix) future directions. Fels research related to these topics that extends beyond the period of adolescent growth will be included.

Development and standardization of anthropometric methods

The Fels Research Institute has long been recognized as a center of excellence in anthropometry. Few of the methods used at Fels are novel but they are described in considerable detail in the research protocol and they are applied with unusually high reliability. In discussing the assessment of nutritional status, Garn (1962c62) complained that anthropometry was often regarded as a set of crude procedures and little or no attention was given to the need for standardization and training. To meet the need for better standardization of anthropometric techniques between studies, a North American Consensus Conference was held in 1985 under the leadership of Tim Lohman. Fels staff members contributed significantly to the success of this conference, at which agreement was reached on the procedures for measuring 40 body dimensions (Lohman, Roche & Martorell, 1988).

In this chapter and the following ones D'Arcy Thompson is struggling against the notion that all form can simply be explained by heredity, and that therefore changes in form inevitably map out phylogenetic relations. Instead he repeatedly suggests that physical forces (such as those which produce the variations of shapes of snow-flakes) are of prime importance and relationships of shape may not justify any family tree or a sequence in time, but simply show mathematical kinship. Today we are inclined to combine the two and say that the genes, the units of heredity, do control shapes, but that the activities of genes are constrained by the physicochemical properties of the chemical substances and the configuration of these substances present in the organism. This does not touch upon the question of whether or not all the shapes produced are adaptively significant. D'Arcy Thompson's strong arguments that they are not is a reaction against those who see a selective advantage to all structures. But this issue cannot be resolved without an ecological study of each example, a Gargantuan task that is unlikely ever to be achieved. All we can say at the moment is that there is no a priori reason why some structures, which have been initiated by mutation and formed within the confines of physicochemical laws, should not utterly lack any adaptive significance, and yet remain fixed in the population.