Cover Image
close this bookCauses and Mechanisms of Linear Growth Retardation (International Dietary Energy Consultative Group - IDECG, 1993, 216 pages)
close this folderBetween-population variation in pre-adolescent growth
View the document(introductory text...)
View the document1. Classifying human populations
View the document2. Population differences in growth patterns
View the document3. The validity of the concept of an international growth reference
View the documentReferences
View the documentDiscussion


The Y chromosome may be a determinant of stature, and some of the differences that have been observed between populations are more pronounced in females than in males. It was therefore suggested that in Latin America, for example, the Y chromosome may come from the European conquerors, while the X chromosome comes from the Amerindian population (Uauy). In North American blacks, 25% of the population is carrying European genetic markers. The difficulty with this idea is that growth is polygenic, determined by many genes on different chromosomes as well as the Y chromosome.

Reference was then made to the hypothesis of Harrison and his colleagues at Oxford that the variability of height in children may be a measure of environmental stress. Ulijaszek's answer to this was that it may be correct for a sample of children from a large population that is out-breeding, but not if the sample is drawn from a small, inbred population, such as a tribal group. Geneticists use variability as a measure of inbreeding.

The problem that always arises in discussion of the genetic origin of ethnic differences is that different groups have different diets. Thus the dietary patterns are quite different in Northern and Southern Europe, or at least have been until recently. (The question of possible relationships of individual nutrients to linear growth is discussed below, in relation to the papers of Allen, Neumann & van Dusseldorp.)

Another possible approach is through correlations between socio-economic status and growth. Such correlations break down in countries where the socio-economic status is rather uniform, as in Scandinavia. These countries have reached an end-stage when the secular trend has come to a stop, and this might reasonably be regarded as a population that has fully expressed its genetic potential for growth. We cannot be certain about the Asian populations, where the secular trend has not yet come to an end.

Some relevant information may be obtained from studies of migrants. Children who were brought to Norway or Sweden from North Korea or India grew exactly as the Scandinavians, provided that they came before the age of 6 months (Karlberg). On the other hand, Pakistanis in the UK do not show the same convergence to the norm, but this could be attributed to their retaining their original dietary habits. In the UK, in populations of different racial groups, substantial differences in length emerge in the first few months of life (Skuse). This finding recalls the data from Hong Kong presented by Davies at the previous workshop, showing that linear growth diverged significantly from the NCHS standard by about 6 months. The question of whether or not this is a genetic effect remains still open.

The paper had touched on the appropriateness of the NCHS reference and the inexactness of the centiles. There is also the well-known problem that it is based on two different data sets, overlapping from 2 to 3 years, with a constant difference over that period between the measurements of length and height. A new reference is expected to be available from the USA in 4-5 years, based on a new survey currently being carried out. (One might add that a new reference, if it is to be international, should also take account of the very comprehensive data bases that have been published in various European countries in recent years. Ed.)