Forms of newly retained phosphorus in mineral soils

The distribution of soluble phosphate in various fractions of soil phosphorus was studied by treating 1 g-samples of 180 mineral soils with 50 ml of a KH2PO4- solution containing P 5 mg/l for 24 hours, and carrying out the fractionation by the method of CHANG and JACKSON after the solution was removed and the moist samples had stood for 3 days at room temperature. The amount of retained phosphorus in the different fractions was computed by taking the difference between the treated and check samples. In the 70 samples of clay soils, the mean proportion of the retained phosphorus was 57 per cent of the 250 mg/kg applied, in the 62 samples of the sand and fine sand soils the corresponding part was 45 per cent, and in the 48 samples of loam and silt soils it was 44 per cent. The higher retention in the clay soils was mainly due to a higher retention in the alkali-soluble fraction. The net increase in the fluoride-soluble forms was of the same order in these three soil groups. On the average, more than 95 per cent of the sorbed phosphorus was found in the fluoride-soluble and alkali-soluble fractions. In one third of the samples a low net increase in the acid soluble fraction was detected, but this may be partly due to changes in the solubility of the native phosphorus in the treated samples. Owing to the fairly large variation, the tendency to somewhat higher mean values for the sorption in the subsoils compared with those of the topsoils was not statistically significant. The ratio between the sorbed amounts of fluoride-soluble and alkali soluble forms was higher in the sand and fine sand soils than in the clay soils. Only in 15 samples, most of them Litorina-soils, the net increase in the alkali-soluble forms was higher than in the fluoride-soluble fraction. Probably, because an equilibrium in the phosphorus conditions was not yet reached at the end of the treatment, the attempt failed to find any clear connection between the distribution of the sorbed phosphorus and such soil properties as pH, the contents of acid oxalate soluble aluminium and iron, organic carbon, the phosphate sorption capacity and the degree of phosphate saturation. Only in the subsoil samples, 76 per cent of the variation in the net increase in the fluoride-soluble fraction could be explained by the variation on the content of oxalate-soluble aluminium, and in the topsoil samples the oxalate-soluble iron and pH determined 61 per cent of the variation in the net increase in the alkali-soluble phosphorus. The ratio of oxalate-soluble aluminium to iron was more closely correlated with the ratio between the total amounts of fluoride-soluble and alkali-soluble phosphorus than with the ratio between the corresponding sorbed amounts. In the topsoils, it explained 70 per cent of the variation in the former. The distribution of the retained phosphorus did not depend on the soil pH, its content of organic carbon, or its degree of phosphate saturation but there was some tendency to a higher accumulation of alkali-soluble phosphorus compared with the fluoride-soluble forms with an increase in the phosphate sorption capacity of the soil