His position is twin correlations are correct and the within family gwas estimes are also correct. Both are correct. But measures different things. He seems to agree gwas measures genetic effects. But question is what rMZ measures then in a reared apart twin study ?