Boros István (szerk.): A Magyar Természettudományi Múzeum évkönyve 52. (Budapest 1960)

Thoma, A.: Anthropometric characters and selective survival

This equation assumes that if the phenomenon of selective survival fails to show in the population, the variance obtained by the generalizing method must be equal to that determined by the individual one. In this case R. I. = O. The overplus of the variance calculated by the generalizing method is due to selective survival. If only the effect of selective survival is perceptible, R. I. = 1. Equation (3) holds true in the case of positive selection. By positive selection I understand that, where selective survival acts in the same direction as the changes resulting from aging. The same equation (3) is valid in the case of mixed selection too, only the variances have to be subjected to certain adjustments. Mixed selection is taken here to mean that, where in a certain phase of life the individual curve turns to the opposite of its former trend, while the generalized curve follows all along a uniform one. In such cases selection operates in one phase counter to aging, in the other parallel to it. In the latter phase the effects of aging and selective survival cumulate on the generalized curve. The variance, calculated with unchanged degrees of freedom from the square deviations relative to this portion of the individual curve, represents the manifested aging variances. With the remaining square deviations obtained by the individual method we then similarly adjust the generalized variance, since in the opposite phase the selection first had to cover up the effect of aging in order to be able to bring about a converse variance. (As regards the adjustment by class size, see below.) The parameter of the mixed selection shall be indicated by the i sign. Negative selection counters the effect of aging. In this case the curve of age changes plotted by the generalizing method follows the trend of the individual curve, but more slowly, — or again it follows it for a while, but halts under the influence of counterselection, without further changes. In this case s^ ^> s%. The effect of selective survival is proportional to the lack of variation found by the generalizing observation and defined in relation to the individual method. Thus : The subcase of negative selection presents itself when the generalized curve counters in one of its phases the direction followed by the individual curve. If so, the formula showing the intensity of selection remains unchanged, and we have again to submit only the variances to some adjustments. Indeed, the effect of aging makes itself felt only in the parallel phase of the generalized curve. In the opposite phase selection has not onlv ,,knocked out" the effect of aging, but has caused also a differently directed surplus variation. The plus variation thus resulting must similarly be ascribed to selective survival. Accordingly, we subtract from the sum of the generalized square deviations those referable to the converse phase, add them to the (corrected ; see below) total of the square deviations obtained by the individual method, and eventually we once more calculate the variances with unchanged degrees of freedom. In order to estimate the component of aging, we shall utilize the data of the Swiss survey made by Büchi(l) at Untertoggenburg according to the individual method. The author in question has examined the same population at inter-