Handedness, with Special Reference to Twins.

HILE the occurrence of right-and left-handedness does not appear W to conform to any simple Mendelian formula, the familial incidence of left-handedness rather definitely indicates a genetic basis. Children are more likely to be left-handed if one parent is left-handed, than if both are right-handed (CHAMBERLAIN 1928), and in families where both parents are left-handed about 50 percent of the children are also left-handed. Students of twins generally agree (NEWMAN 1928, LAUTERBACH 1925, WILSON and JONES 1934) in finding a higher percentage of left-handedness in both monozygotic and dizygotic twins than in single born populations. This increase is due principally to the frequent occurrence of intra-pair differences in handedness, rather than to any excess of pairs in which both members are left-handed. Both types of twins differ from the single born in that the conditions in utero are necessarily changed. But if the position in utero is responsible for the differences in the expression of handedness in certain pairs of twins, why does it not affect all twin pairs similarly? NEWMAN (1928, 1937) is of the opinion that in monozygotic twins, embryonic division occurs near or during gastrulation. As the axis of the embryo and bilateral asymmetry are established during gastrulation, he assumes that monozygotic twins which do not separate until after gastrulation will exhibit intra-pair differences in handedness, whereas those separating before gastrulation will be of the same handedness. He also attributes intra-pair variations in other bilateral asymmetrical traits, such as dermatoglyphics, direction of head hair whorl and dentition to the same cause. But if this is the correct explanation, there should be a definite stage in embryonic development for the establishment of each trait showing bilateral asymmetry, beyond which separation of the embryo should result in intra-pair variation, but if separation should occur at some previous time the members of the pair would be alike. Thus we should expect definite combinations of intra-pair trait reversals in monozygotic twins. For example, we might expect to find some twins showing no intra-pair differences in either handedness or direction of hair whorl, some to show reversals in handedness and the same direction of head hair whorl (assuming handedness normally to be established first), some to show differences in both traits, but none to show reversals in hair whorl and the same handedness. Actually, however, we find all possible combinations in re-