Random Selection

Abstract

THIS memoir1 is the first of a series dealing with the problem of selection, namely the measurement of the changes in the characters of a race, when selection has acted upon any one, two, or more of them. The problem mathematically differs considerably according to the nature of the selection. But in all cases the general result is the same, the selection of any organ, whether by size, variability, or correlation with other organs, changes the sizes, variabilities, correlations of all other organs, whether directly correlated with the first organ, or only indirectly correlated with it owing to correlation with other organs which are correlated with the first organ. (A and C may have no correlation with each other, but both be correlated with B, e.g. two parents in the absence of sexual selection and their offspring.) The chief types of selection which have to be treated independently are:—