Several types of soil analysis require pretreatment of samples with hydrogen peroxide for removal of the organic matter. The present work was carried out in order to obtain more especific information about the effect of the hydrogen peroxide treatment on the organic matter in soil samples. Hydrogen peroxide did not destroy all the organic matter present in the soil samples. Usually from 59 to 94 per cent of it was destroyed, these results varying according to kind of soil, method of treatment, and in some cases with the concentration of the hydrogen peroxide solution used. Samples leached after each treatment with hydrogen peroxide had more carbon removed than unleached samples. The behavior of soil nitrogen was different from that of carbon. The nitrogen level in treated samples that were not leached was increased or maintained at the same level in relation to their initial content. When the samples were leached, from 72 to 100 per cent of their nitrogen content was removed. The C/N ratio followed closely the removal of carbon and nitrogen. It was wide for leached samples and narrow for the non-leached ones. The destruction of the organic matter in samples rich in organic carbon was independent of the peroxide concentration, but varied to a certain extent with the amount added. Low concentrations of peroxide solutions gave better results with samples from poor soils than the same amount added in higher concentration. It is believed that some peroxide was decomposed before acting on the organic matter. Removal of the nitrogen fraction was more dependent on leaching the samples after treatment than on the amount or concentration of the reagent. Repeated treatments of the soil samples with small amounts of hydrogen peroxide gave results different from those given by a single treatment with the same total volume of the reagent. When leaching was employed, the repeated treatments were more effective for removal of carbon and nitrogen from soil samples rich in organic matter than the single treatment. In case of samples with low organic matter content, the method of repeated treatments was more effective than the single treatment only for nitrogen removal. When no leaching was done after treatment, both methods were equally effective.