Note: This paper is dedicated to Aaron and Elizabeth Waters on the occasion of Dr. Waters' retirement.
Recent analytical data on the compositions of natural hematite-ilmenite solid solutions suggest that the miscibility gap of Carmichael (1961) cannot be correct. Attempts to redetermine the miscibility gap under hydrothermal conditions have been only partly successful, for with the oxygen buffers currently available, it is not possible to retain both hematite and ilmenite phases in a given experiment. Despite the fact that the hydrothermal experiments do not give the maximum solubilities of Fe2O3 in ilmenite and FeTiO3 in hematite, they nevertheless show that the published miscibility gap is too wide (by approximately 15 mole percent at 600°C) and that the consolute temperature is at least 150°C too high.