Components of length growth variation in infants from the same population but different environments

Growth in length of infants in Khartoum is profoundly related to whether or not their mothers are in paid employment. Some of this relationship appears to be due to levels of infectious disease, which are much higher in the infants of working mothers. Exponential decay curves do not fit longitudinal measurements of length growth particularly well in either group of infants, but they do yield distinctive temporal patterns of residuals. Simulation demonstrates that these patterns accord with a "catch-down" form of growth in infants of working mothers and a "catch-up" form in infants of housewives. Attempts to allow for these phenomena by providing for the opportunity of a single pulse of negative sine form in the former and a similar pulse of positive sine form in the latter reduces the level of residuals and tends to reduce temporal pattern in them. Most strikingly, it leads to average growth curves for the two groups of children becoming almost identical. It is hypothesized that these decay curves represent the underlying curves of infant growth in Khartoum free from the effects of environmental variation.