Several methods of extracting phosphate from soils for use in constructing phosphate desorption isotherms were evaluated by applying them to each of four contrasting soils. Preparation of the samples for extraction included phosphate addition, both in constant and in varying amounts. For the former, removal of phosphate was accomplished either by successive extractions with 0·02M citric acid at pH 4·0 (washing isotherms) or by single extractions with chloride-resin at specific soil/resin ratios. For the latter, desorption was achieved by extracting with (i) citric acid, 0·02 and 0·0025M, respectively (ii) 0·02M KCl at pH 4·0 and (iii) chloride-resin. The isotherm constructed using different oil: resin ratios differentiated the oils according to their desorption characteristics. Although similar results are obtained by extracting at various rate of added phosphate, this method is more time consuming and requires more preparatory treatment of the soil sample prior to extraction. Hence, the traditional method of using a constant rate of phosphate is recommended for assessing labile pool phosphate in different soil types. The effect of the different extractants on phosphate desorption is discussed within the framework of the theory of specific adsorption of anions on soil and hydrous oxide surfaces.