Even within an apparently homogeneous admixed group, individual ancestry may remain correlated with environmental risk factors

This is the crux of the identification problem. You still need to control for the "environment" which you can never perfectly do.

If the phenomen of assortative mating is true, then I don't see how the study's methodology would be able to discern the effects of BGH. Low IQ slaveowners may have been "randier" than their smarter brethren. Today, one hotbed of interracial mating are colleges, which can be seen as segregated by test scores and IQ. Most IR couples are BM/WF and who knows how gender can affect genetic contribution? Directly measuring the IQs of interracial families would be too easy of course, but the silliness of beating around the PC bush only works to generate needless statistical ponderings like what is found in this topic.

Something similar is the correlation between skin color and sprinting ability. From watching 100m dash men, I'd say there was maybe a low to moderate (r=0.2 to 0.4) correlation between darkness and sprinting ability. The only top sprinter in recent decades to be not quite dark was Frankie Fredericks, who, oddly enough, was from Namibia.

Also, there's more variation than you might think in skin color in Africa, or even in just West Africa: Senegalese tend to be very dark, and Nigerians less so.

"Most IR couples are BM/WF and who knows how gender can affect genetic contribution?"

Notwithstanding the credulity of your ratio assertion, here is one possible example of gender significance, although as Hawks mentions, it hasn't been cited by anyone since.

New report reveals African-Americans may lack key nutrients for optimal health