Volume 1

Sex and repression in savage society / by Bronislaw Malinowski.

  • Bronisław Malinowski
Date:
[1960]
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: Sex and repression in savage society / by Bronislaw Malinowski. Source: Wellcome Collection.

    43/484 (page 31)
    these studies have employed data from physical, rather than from mental traits, they supply the cues for investigations of mental heredity. Pearson, 16 turning in 1904 to the newly rediscovered Mendelian theory, and assuming that a trait was due to N cumulative genes showing perfect dominance, calculated that under random mating the correlation between parent and offspring would be .33. As this value was out of harmony with the coefficients that had been found for parent-child measurements on a large number of con tinuous traits, Pearson definitely turned away from the theory which he had found inconsistent with experimental results, and never again made any serious attempt to interpret his own data by means of it. It was subsequently pointed out by Yule, 17 and later by others, that parental correlations higher than .33 and quite consistent with actual values would follow from a generalized Mendelian theory if complete dominance in genes were not assumed. It remained for Fisher 18 to work out a scheme by which the correlations between parents and offspring, between siblings, and between fathers and mothers are used to infer a coefficient of en vironment, “the ratio of the variance [of an unselected population on a continuous trait] with environment absolutely uniform to that when difference of environment also makes its contribution.” Fisher shows that when a trait is due to Mendelian factors, “the effect of dominance is to reduce the fraternal correlation ( i.e., its genetic value of one half) to only half the extent to which the parental correlation is reduced,” and that this effect “is inde pendent of the relative importance of different factors or of their different degrees of dominance.” Making the important assump tion that environment works in a random manner, thus reducing rather than raising correlations between relatives, and reducing fraternal correlations to the same extent as parent-child correla tions, Fisher then utilizes the differences actually found between fraternal and parent-child correlations to distinguish between the effects of dominance and those of environment, finally arriving at the coefficient of environment defined above. ™Phil. Trans., 203 A, 1904, 53-87. 17 1906 Conference on Genetics, Horticultural Society’s Report. 18 Trams. Royal Soc. Edinburgh, 52:1918, 399-433.