Not to excuse the shoddy scholarship of rushton and jensen, but I'd just like to add that a cursory examination of the nisbett article indeed shows some highly dubious claims. In several places he assumes a hypothesis of the form "If the hereditary model is true, then we should see X". But for many of these it seems that X does not necessarily follow from the hereditary hypothesis. the hereditary hypothesis is not a monolithic structure. it is a spectrum of correlation from 0.0 to 1.0. both ends seem equally implausible to me.