Elemental ratios Zr/Hf and Nb/Ta are expected to be constant and of chondritic value (∼36.30 and ∼17.57, respectively) in mantle and mantle-derived rocks. Studies in recent years have shown, however, that these two ratios do vary in some of these rocks. For example, MORB-like seamount lavas from flanks of the East Pacific Rise (EPR) show a correlated Zr/Hf (∼25–50) and Nb/Ta (∼9–18) variation. These two ratios are also correlated with ratios of more incompatible over less incompatible elements (e.g., La/Sm, Rb/Cs, Th/U, Nb/U, Sm/Yb) and with radiogenic isotope ratios (e.g., 87Sr/86Sr, 143Nd/144Nd). Furthermore, abyssal peridotites, which are melting residues for MORB, also show a huge correlated variation between Zr/Hf (∼2.5–335) and Nb/Ta (∼1–170). All these observations plus a correlated variation between Zr/Hf (∼22–48) and Nb/Ta (∼10–23) in lunar rocks are consistent with the Zr-Hf and Nb-Ta fractionation being of magmatic origin. This contrasts with the common view that geochemical processes cannot readily fractionate them. As charges and ionic radii are the principal factors in the general theory of elemental fractionation, this theory cannot explain the fractionation of these two element pairs with the same charges (i.e., 5+ for Nb and Ta, and 4+ for Zr and Hf) and essentially the same ionic size (i.e., RNb/RTa=1.000, RZr/RHf=1.006 to ∼1.026 for coordination numbers of 6, 7, 8 and 12). We explore the possibilities of other factors and processes (e.g., mass-dependent fractionation during magmatism) that may cause the observed Nb-Ta and Zr-Hf fractionation. We emphasize that understanding the correlated Nb-Ta and Zr-Hf fractionation “known” to take place during magmatism is fundamental for improved understanding of elemental fractionations through other earth processes in various tectonic environments, including the origin and evolution of continental crust, which has a characteristic subchondritic Nb/Ta value of ∼11-12.