High resolution MC-ICP-MS is used for the precise measurement of variations in the isotopic composition of Fe in ferromanganese concretions and sediments relative to IRMM-014 standard. The sensitivity for 56Fe in high resolution mode was 3 V per mg l−1 Fe, a figure that is comparable to those from other MC-ICP-MS instruments operated at low resolution. Incorporation of a guard electrode and the efficient ion transmission capabilities of the Neptune MC-ICP-MS instrument are responsible for the high sensitivity. It was observed that the use of HCl resulted in the formation of ClOH+, causing interference with 54Fe in particular. This acid has been preferred in some cases over HNO3 to minimize formation of ArN+, the major interferent for 54Fe. Using the high resolution mode of the Neptune, the nature of spectral interferences is unimportant as all are completely resolved and will not affect the accuracy of the determined Fe isotope ratios. As the instrument also provides flat-topped peaks, high resolution operation does not necessarily result in impaired precision, providing that higher concentrations are used to compensate for the loss in sensitivity compared with the low resolution mode. In the present work, external reproducibilities of 56Fe/54Fe and 57Fe/54Fe isotope ratios were better than 50 ppm (one standard deviation) at a concentration of 5 mg l−1. The level of instrumental mass discrimination observed for raw ratios drifted by as much as 0.09% per mass unit over a measurement session, but could be corrected on-line by simultaneous monitoring of the 62Ni/60Ni isotope ratio. Variations in the Fe concentrations or the acid strength of measurement solutions were found to affect the apparent mass discrimination. Increasing the Fe concentration caused a relative decrease in the raw 56Fe/54Fe and 57Fe/54Fe isotope ratios, thus ruling out the space charge effect as the explanation for this phenomenon. Instead, it is suggested that the larger dry aerosol particles formed at higher Fe concentrations are not completely vaporized until later in the plasma, thus reducing the relative rate of diffusional losses of lighter 54Fe from the central channel. However, application of on-line correction using Ni could adequately account for this effect. From the results for a variety of sedimentary geological materials, analysis of three-isotope data revealed that equilibrium fractionation of Fe occurred during deposition. To be able to distinguish between equilibrium and kinetic fractionation processes, it is imperative to collect accurate and precise data for the 56Fe/54Fe and 57Fe/54Fe isotope ratios. These requirements are readily fulfilled by applying high resolution MC-ICP-MS and on-line correction for instrumental mass discrimination using Ni.