In Reply.— The comments made by Drs Wember and Manz are of interest and help us to clarify some points made in our paper that may have been obscure. The percentiles in figs 1 and 2 were calculated from the raw data and not from the means and standard deviations on the assumption of normality of the distributions. The data in table 1 for the means suggest that a slow increase in head circumference may be occurring in boys near the age of 18 years.