Accurate and precise determination of the redox state of iron (Fe) in spinels presents a significant challenge due to their refractory nature. The resultant extreme conditions needed to obtain complete dissolution generally oxidize some of the Fe(II) initially present and thus prevent the use of colorimetric methods for Fe(II) measurements. To overcome this challenge we developed a hybrid oxidimetric/colorimetric approach, using Ag(I) as the oxidimetric reagent for determination of Fe(II) and 1,10-phenanthroline as the colorimetric reagent for determination of total Fe. This approach, which allows determination of Fe(II) and total Fe on the same sample, was tested on a series of four geochemical reference materials and then applied to the analysis of Fe(Ni) spinel crystals isolated from simulated high-level-waste (HLW) glass and of several reagent magnetites. Results for the reference materials were in excellent agreement with recommended values, with the exception of USGS BIR-1, for which higher Fe(II) values and lower total Fe values were obtained. The Fe(Ni) spinels showed Fe(II) values at the detection limit (ca. 0.03 wt% Fe) and total Fe values higher than obtained by ICP-AES analysis after decomposition by lithium metaborate/tetraborate fusion. For the magnetite samples, total Fe values were in agreement with reference results, but a wide range in Fe(II) values was obtained indicating various degrees of conversion to maghemite. Formal comparisons of accuracy and precision were made with 13 existing methods. Accuracy for Fe(II) and total Fe was at or near the top of the group. Precision varied with the parameter used to measure it but was generally in the middle to upper part of the group for Fe(II) while that for total Fe ranged from the bottom of the group to near the top.