Recent work on one of the major banded iron-formations of the world--that of the Hamersley Basin of Australia--has defined the total amount of iron and the probable time interval of accumulation of the large iron ore deposits in the basin. The inferred rate of influx of iron into the basin is so large that a derivation of the iron from weathering and transport into the basin by rivers is unlikely. For the same reason derivation of the iron from volcanic sources is also unlikely. Derivation of the iron from the ocean floor is quantitatively possible if deep ocean water containing a few mg/1 of dissolved iron welled up, was oxidized in near surface environments, and deposited its iron in the basin. Iron concentrations of a few mg/1 in deep ocean water are quite reasonable if these water masses were saturated with respect to siderite and calcite.