The processing volume theory of general intelligence predicts that brain volume imposes general constraints on the overall amounts of cortical substrate available for solving complex problems. Bigger brains contain more processing volume, hence can solve problems generally more efficiently than smaller brains. This theory predicts that the g loading of ability tests should substantially moderate the association between ability test performance and brain volume. This is tested with a bare-bones meta-analysis employing the method of correlated vectors (MCV) on the association between brain volume∗subtest correlations and subtest g loadings. The mean weighted vector correlation across four studies is ρ=0.07 (total N=246), increasing to ρ=0.35 when one potentially outlying study is removed. The magnitude of these Jensen effects suggests that g plays at best only a modest role in moderating the strength of the brain volume∗IQ correlation, contrary to predictions from the processing volume theory. This result is consistent with the observation that brain volume has low evolutionary lability compared with g, suggesting that recent evolutionary changes in population levels of g may have occurred largely independently of overall changes in brain volume.