The cocrystallization coefficient of Mn and Fe (DMn/Fe) in magnetite crystals is determined in hydrothermal-growth experiments with internal sampling at 450 and 500 °C and 100 MPa (1 kbar). It is weakly dependent on temperature in the studied PT-region and is constant over a wide range of Mn/Fe values. This permits using the magnetite composition as an indicator of Mn/Fe in the fluid under equilibrium: (Mn/Fe)aq ≈ 100 (Mn/Fe)mt. Since Mn is often a macrocomponent of the fluid and a microcomponent of magnetite, local analysis of fluid inclusions for Mn might help to determine Fe even in iron minerals. This will permit evaluation of the contents of other ore metals if the DMe/Fe values are known. For fine crystals (< 0.1–0.2 mm) with low contents of Mn (< 0.01–0.02%), it is necessary to take into account the fractionation of Mn into the surficial nonautonomous phase, in which its content can reach several percent. Comparison of these data with earlier data on the distribution of Mn in the system magnetite–pyrite–pyrrhotite–greenockite–hydrothermal solution shows that DMn/Fe remains constant in the presence of sulfur and sulfides. Precipitation of magnetite, in which Mn is a compatible admixture, cannot affect radically Mn/Fe in the solution because of the low DMn/Fe value. This effect is still more unlikely for pyrrhotite and pyrite, in which Mn is an incompatible admixture. The most probable mechanism of Mn fractionation into the solid phase is crystallization of FeOOH at lower temperatures. This is indirectly supported by the strong fractionation of Mn into the nonautonomous oxyhydroxide phase on the surface of magnetite crystals. The necessity of a more rigorous validation of “the new Fe/Mn geothermometer for hydrothermal systems” is substantiated.