Testosterone levels, handedness and sex ratio at birth

Geschwind has hypothesized that levels of androgens in utero affect postnatal laterality preferences, high levels being associated with left-handedness of offspring. I have hypothesized that high parental androgen levels at the time of conception are associated with subsequent male births. However, left-handed parents seem to produce an excess of female, not male, offspring. This finding could be reconciled with the two hypotheses if individuals who were exposed to high levels of androgens in utero had low circulating levels in adulthood. Some evidence for this loophole is reviewed here. Lyster & Lloyd (1987, J. theor. Biol. 126, 125) have reported that (consistent with the hypotheses) index left-handed men have an excess of brothers; but, contrary to the hypotheses, index left-handed women have an excess of sisters. Accordingly these authors proposed that left-handedness in females is associated not with high levels of androgen in utero, but with a "sensitivity" to it. If such sensitivity were associated with low androgen levels, it might explain an excess of sisters among the sibs of left-handed women. It is shown here that evidence in agreement with this hypothesis (though not discriminating between it and Geschwind's) suggests that the sex ratios of offspring of LH male x RH female and of RH male x LH female matings, are both low.