Components of growth variation in human stature

Traditional methods of estimating the heritability of a trait such as height utilize a single observation of the variance. This, however, provides little insight into the way in which the variance is generated. In this paper, we propose that repeat measurements of height over a period of childhood growth can be used to estimate different components of the variation in height increment. By height increment we mean the successively longer increments computed relative to a common initial height. Models are developed which indicate that the genetic component of the variance accumulates as a quadratic function of the mean, while the environmental component accumulates until it reaches a constant. This 'homeostatic constant' represents an equilibrium between the input of environmentally induced variation and its removal through homeostasis. Large constants are indicative of poor environments and poor homeostatic responses. The constant therefore provides a measure of the adaptive status of the children at the level of the population. The models are here applied to growth data collected on Hong Kong infants. The resulting estimate of the genetic component of the variance is 0.0044 cm2 per cm2 of mean height increment. This is somewhat higher than expectation but easily accounted for through the effects of environmental stratification on the variance. The homeostatic constant is estimated to be approximately 1.2 cm2. This is the first such estimate of its kind.