The possibility that Spearman's Hypothesis is a statistical artifact is tested with studies of deaf children's intelligence. Six published and unpublished sources yielded 37 instances in which the magnitude of deaf-hearing differences, and the g loading of the test for deaf children, could be determined on intelligence tests. The magnitude of differences between deaf and normal-hearing children are negatively correlated with g loadings (r = −.14, NS), and are significantly different from positive correlations reported for black-white differences (zs > 2.51, ps < .01). Post hoc analyses show the lack of positive relationship between deaf-hearing differences and g loadings is consistent across psychometric and subject samples. The results clearly refute Schonemann's (1985) proposal that the positive relationship between black-white differences and g loadings is a statistical artifact, and the results provide evidence of divergent validity for Spearman's Hypothesis.