Three silty clay loams from the classical field experiments at Rothamsted Experimental Station (UK) with low phosphorus (P) status were treated with phosphate fertilizer and incubated for 15 days at field capacity with added acid [hydrochloric acid (HCl)] or base [sodium hydroxide (NaOH)] to give pH ranges measured in 1 :5 suspensions of soil in 10 mM calcium chloride (CaCl2) of 6.3-8.7 (Hoosfield A, pH 8.2), 3.8-8.2 (Hoosfield B, pH 6.3), and 2.8-7.1 (Geescroft pH 5.3). Extractable P was measured by the 10 mM CaCl2 extraction procedure. For the Hoosfield B and Geescroft soils without added phosphate fertilizer, extractable P was decreased at both 'low' and 'high' pH values, the maximum being at about 5.4 in both soils. In the calcareous Hoosfield soil, extractable P decreased with increasing pH over the range studied. These changes in extractable P were magnified in soils treated with phosphate fertilizer but the maxima and trends were unaltered. The results indicate that native soil P appears to be changed by pH in the same way as added P to the soil. In the calcareous Hoosfield A soil, added acid did not reduce soil pH to less than 6.4 (because of its high buffer capacity) and so the pH level for maximum P solubility was not found.