The ultimate size and shape that a child attains as an adult is the result of a continous interaction between genetical and environmental influences during the whole period of growth. Such interaction may be complex. Two genotypes which produce the same adult height under optimal environmental circumstances may produce different heights under circumstances of privation. Thus two children who would be the same height in a well-off community may not only both be smaller under poor economic conditions, but one may be significantly smaller than the other. This type of interaction, called non-additivity of genotype and environment, may be quite detailed and specific in its effects. If a particular environmental stimulus is lacking at a time when it is essential for the child (times known as ‘sensitive periods’), then the child's development may be shunted, as it were, from one line to another. We know, as yet, little of the details of such interactions, but quite enough to make oversimplified models scientifically suspect.

Statements about the relative contributions of heredity and environment to adult size and shape must therefore always specify the circumstances with some exactness. A biologically permissible statement (provided all the evidence is really available), is, for example: ‘Eighty-five per cent of the variance of height in the population of young adults growing up in middle class homes in a London suburb in the 1980s is due to genetical factors’. In a population growing up under famine conditions the percentage of variance due to genetical factors would be expected to be less. A particularly lucid discussion of these points may be found in Thoday (1965).