Heights and weights of infants (142 boys and 117 girls) who were physically examined at an infant health center in Osaka Prefecture during 1964 and 1967 were followed up monthly for one year from birth. The following results were obtained by the analysis of the data.
1) The average height and weight of the male and female infants investigated were slightly larger at every age of month than the standard set by the Ministry of Welfare in 1960. However, the infants investigated can be considered as a relatively unbiased sample of the population of Japanese healthy infants at least as far as size and growth of body build, as the body build of infants during the period of the investigation were slightly larger than those in 1960.
2) Relatively large positive correlations were found between height (weight) of a infant measured at different age of month. Moreover, when male and female infants were classified into three groups (large, normal, small) according to their height (weight) at the time of birth, and height (weight) of each group was followed up for a year, it was found that in both boys and girls, the absolute values of difference of mean height (weight) among three groups at the time of one year old was almost the same as that observed at the time of birth.
3) However, a relatively large variation of relative body size of a infant at different month ages and large coefficients of variation observed on the monthly increasement of height (weight) in every interval between two successive age of month suggested that there were relatively large individual variation in the growth of infants.
4) Using Rao's method, the standard growth curves of height and weight of infants and their 99% confidence bands were obtained for groups of male and female infants classified by their height (weight) at the time of birth, on the assumption that the deviations of any individual growth process from population growth curve have a multivariate normal distribution. It was found that in every group, the standard growth curve of height (weight) was expressed in the three degree polynomial in age of month and it corresponded well with the observations.
