The changes in phosphate availability that occurred when fertilizer phosphate was incorporated in moist uncropped soil were examined by storing fertilized soil in the open for periods up to 3 years before a final test crop was grown. Superphosphate was the standard source as it was fully available initially and therefore provided a measure of the rate of immobilization of phosphate on the four soils used. The extent to which this rate was modified by slower rates of dissolution of non-water-soluble sources was examined, using dicalcium phosphate (powder and granular), basic slag and ground North African rock phosphate (GNAP).

Powdered dicalcium phosphate and basic slag in general behaved similarly to superphosphate, indicating very rapid dissolution. The only exception was a slightly lower rate of dissolution of slag on the calcareous soil. Other than in the very acid soil, the rate of dissolution of GNAP was slow so that the available phosphate was not similar to the superphosphate treatment until after at least the second year. On the calcareous soil GNAP was practically inert. Granular dicalcium phosphate was the only consistent slow-release source. It gave available phosphate levels which were initially lower, but subsequently higher, than those from superphosphate. In the absence of a crop, cumulative assessments of phosphate availability were not possible.

The pertinence of these results to the problem of maintaining a high soil phosphate status is discussed.