Procedures for the determination of between 0 and 5 µg l–1 of total iodine and iodate-iodine present in natural freshwaters are described. Total iodine is determined using the catalytic effect that iodide has on the reaction between ammonium cerium(IV) sulphate and arsenic(III) oxide. An earlier method has been modified rigorously and according to Mark's (1973) recommendations. The analysis of sets of eleven replicate samples showed that the total iodine concentration could be resolved to within ±0.1 µg l–1(±2 × standard error) at the 95% confidence level. The iodate-iodine in the sample is determined by the same procedure after any iodide-iodine has been extracted into chloroform as an ion pair with the tetraphenylarsonium cation. The optimisation of the extraction procedure is described. The analysis of sets of six replicate samples encompassing the 0 to 5.0 µg l–1 range of iodate-iodine concentrations and containing also 1.0 µg l–1 of iodide-iodine returned essentially the same standard errors as for total iodine determination. The effect upon the methods of various substances that could be present in freshwaters is described together with tests conducted upon some organic iodine compounds. The latter experiments suggest that the unsuspected presence of organic iodine compounds in natural waters would be problematic. Analysis for total iodine can be conducted at the rate of approximately 50 samples per hour and that for iodate-iodine at about 20 per hour.