Magnetite (Fe3O4) is a commonly found in the environment and can form via several pathways, including biotic and abiotic reduction of Fe3+ oxides and the oxidation of Fe2+ and Fe0. Despite extensive research, the redox behavior of magnetite is poorly understood. In previous work, the extent and kinetics of contaminant reduction by magnetite varied by several orders of magnitude between studies, two fundamentally different models are used to explain magnetite oxidation (i.e., core-shell diffusion and redox-driven), and reported reduction potentials vary by almost 1 V. In other fields of science (e.g., physics), magnetite stoichiometry (x = Fe2+/Fe3+) is a commonly measured property, however, in environmental studies, the stoichiometry is rarely measured. The stoichiometry of magnetite can range from 0.5 (stoichiometric) to 0 (completely oxidized), with intermediate values (0 < x < 0.5) referred to as nonstoichiometric or partially oxidized magnetite. To determine the relationship between magnetite stoichiometry and contaminant fate, the reduction rates of three substituted nitrobenzenes (ArNO2) were measured. The kinetic rates varied over five orders of magnitude as the particle stoichiometry increased from x = 0.31 to 0.50. Apparent 15N kinetic isotope effects (15N-AKIE) values for ArNO2 were greater than unity for all magnetite stoichiometries investigated, and indicated that mass transfer processes are not controlling the reaction rate. To determine if the reaction kinetics were redox-driven, magnetite open circuit potentials (EOCP) were measured. EOCP values were linearly related to the stoichiometry, with more stoichiometric magnetite having a lower potential, in good agreement with redox-driven models. The reaction of aqueous Fe2+ and magnetite was investigated. Similar to previous findings for other Fe3+ oxides, the formation of a stable sorbed Fe2+ species was not observed; instead, the sorbed Fe2+ underwent interfacial electron transfer to form a partially oxidized magnetite phase, which was accompanied by reduction of the underlying magnetite. The lack of a stable sorbed Fe2+ species on magnetite indicated that the traditional surface complexation model was incorrect; instead, the uptake of Fe2+ by magnetite appeared to be limited by the whole particle (i.e., the sorbed and underlying phases combined) reaching a stoichiometry of 0.5.