On treating the chromosome as the unit of selection.

A simple method is presented for approximating the behavior of a multiple genetic system under the action of selection and linkage. The effects of genetic drift and mutation are left out. It is found that the results are close to those obtained from a computer simulation of the same problem by Franklin and Lewontin (1970). It is possible to describe the system in terms of a correlation length on the chromosome which measures the degree to which the different parts are bound by linkage. There is also a brief discussion about the higher order correlation coefficients as a measure of the interaction effects.