In the early days, the standards for body height and weight of man were usually derived from the anthropometric data of retrospective studies, and the growth standards for children were also constructed from data of the same nature. The population thus measured consisted of the so-called "normals", i.e. excluding the anatomical abnormals and those with apparent chronic diseases. It is evident that the body size, particularly the height, of most of the boys and girls measured might not have attained their optimal stage.It is well recognized now that, within the limits of heredity, the body size of man is strongly influenced by the environment in which he lives, and the nutritive value of the diet plays an important role on growth. So, in nowadays, in the construction of growth standards, emphasis is placed on measuring children living under favorable conditions and in optimal nutritional state.Chin and co-workers published (1957) their data of measurement of nearly 3,000 carefully selected children, from birth to 7 years with known backgrounds as stated above. This seems, therefore, to be the most suitable material for constructing a growth standard for preschool children.The method of construction is as follows: Chin's growth curves of height and weight and the corresponding S.D. curves were smoothed out in order to obtain the "ideal" growth curves of the children measured. From such smooth curves, the corrected average value of heights and weights of each age group were read off. The distances between each two age groups were all 1 month from birth to 7 years.For the convenience of evaluating the growth level of an individual child, the following scheme is proposed: it consists of 6 levels, namely: Level A (high): (M + 2S) and aboveB (middle high); (M + 2S) - (M + 1S)C (upper middle); (M + 1S) -(M)D (lower middle); (M) -(M-1S)E (middlelow); (M - 1S)-(M-2S)F (low); (M-2S) and belowWhere M is the mean value of height or weight, and S is the corresponding S.D. value.By using these figures, a "Table of the growth levels of children from birth to age 7" is constructed. Such a table may be used for the evaluation of the growth levels of both individual child and the "collective growth level" of a group of institutional children, i.e. in nurseries or kindergartens. To complete the later procedure, the percentages of the number of children who fall in each level (irrespective of sex and age) are calculated on the basis of the total number of children measured in an institution. By summing up the percentage value of levels A, B, and C (the sum is used as an index) one may easily judge the "collective growth level" of the group of children in such an institution.