Experiments at high pressures and temperatures were carried out (1) to investigate the crystal-chemical behaviour of Fe4O5-Mg2Fe2O5 solid solutions and (2) to explore the phase relations involving (Mg, Fe)(2)Fe2O5 (denoted as O-5-phase) and Mg-Fe silicates. Multi-anvil experiments were performed at 11-20 GPa and 1100-1600 degrees C using different starting compositions including two that were Si-bearing. In Si-free experiments the O-5-phase coexists with Fe2O3, hp-(Mg, Fe)Fe2O4, (Mg, Fe)(3)Fe4O9 or an unquenchable phase of different stoichiometry. Si-bearing experiments yielded phase assemblages consisting of the O-5-phase together with olivine, wadsleyite or ringwoodite, majoritic garnet or Fe3+-bearing phase B. However, (Mg, Fe) 2Fe2O5 does not incorporate Si. Electron microprobe analyses revealed that phase B incorporates significant amounts of Fe2+ and Fe3+ (at least similar to 1.0 cations Fe per formula unit). Fe-L-2,L-3-edge energy-loss near-edge structure spectra confirm the presence of ferric iron [Fe3+/Fe-tot = similar to 0.41(4)] and indicate substitution according to the following charge-balanced exchange: [4]Si4+ + [6]Mg2+ = 2Fe(3+). The ability to accommodate Fe2+ and Fe3+ makes this potential "water-storing" mineral interesting since such substitutions should enlarge its stability field. The thermodynamic properties of Mg2Fe2O5 have been refined, yielding H degrees(1bar, 298) = -1981.5 kJ mol(-1). Solid solution is complete across the Fe4O5-Mg2Fe2O5 binary. Molar volume decreases essentially linearly with increasing Mg content, consistent with ideal mixing behaviour. The partitioning of Mg and Fe2+ with silicates indicates that (Mg, Fe) 2Fe2O5 has a strong preference for Fe2+. Modelling of partitioning with olivine is consistent with the O-5-phase exhibiting ideal mixing behaviour. Mg-Fe2+ partitioning between (Mg, Fe)(2)Fe2O5 and ringwoodite or wadsleyite is influenced by the presence of Fe3+ and OH incorporation in the silicate phases.