We analyze the possibility that polarized gluons could make a substantial contribution to the value of ʃ 0 1 dx g 1 p ( x) estimated by the EMC. We show that with a physically motivated kernel for γ ∗g→ q q, where the q are light quarks, a very substantial part of the possible polarized gluon contribution to g 1 p ( x) is in the region x<0.01 not measured by the EMC. This means that the polarized gluon distributions proposed so far are too small to explain the EMC data via their effects on light quarks alone. However, they do generate a significant charm contribution to g 1 p ( x) for x>0.01 which could help explain the data. We present quantitative predictions for this charm cross section asymmetry in future polarized lepton-nucleon scattering experiments.