The rate of oxidation of ferrous iron in a seasonally anoxic lake was measured on 39 occasions with respect to both depth and time. Sample disturbance was minimal as only oxygen had to be introduced to initiate the reaction. The data were consistent with the simple rate law for homogeneous chemical kinetics previously established for synthetic solutions. The rate constant for the oxidation reaction in lake water was indistinguishable from that measured in synthetic samples. It did not appear to be influenced by changes in the microbial populations or by changes in any particulate or soluble components in the water, including iron and manganese. Analysis of the errors inherent in the kinetic measurements showed that the estimation of pH was the major source of inaccuracy and that values of the rate constant determined by different workers could easily differ by a factor of six. The present data, together with a comprehensive survey of the literature, are used to suggest a 'universal' rate constant of ca. 2 × 10 13 M -2 atm -1 min -1 (range 1.5-3 × 10 13) in the rate law -d[Fe II]/dt = k[Fe II]pO 2 (OH-) 2 for natural freshwaters in the pH range 6.5-7.4. Discrepancies in the effects of ionic strength and interfering substances reported in the literature are highlighted. Generally substances have only been found to interfere at concentrations which far exceed those in most natural waters.