Work done with herbaceous plants has shown the long-term effectiveness of Fe-3(PO4)(2). 8H(2)O (a synthetic iron(II)-phosphate analogous to the mineral vivianite) and Fe(NH4)PO4. H2O in preventing iron chlorosis. In this work, we investigated the effect of suspensions of vivianite and mixtures of vivianite and Fe(NH4)PO4. H2O injected in the soil to prevent iron chlorosis in pear trees (cv. 'Blanquilla', rootstock EM-A) cultivated in an iron chorosis-inducing soil. In the first three growing seasons (no harvest) the leaf chlorophyll content of the treated trees was significantly higher than that of the untreated (control) trees and not significantly different from that of iron chelate-treated trees. In the fourth year, leaf chlorophyll content, blooming and fruit set in the treated trees were significantly higher than in the control but lower than in the chelate-treated trees. In the first three years, leaf colour (C.I.E. L*a*b* color space) of trees treated with Fe(II)-phosphate and chelate was characterized by lower a*, b*, a*/b* and chroma, and higher hue as compared to the control. In the fourth year, the colour parameters of the trees treated with Fe(II)-phosphate were intermediate between those of control and chelate-treated trees. The average yield (fourth year) for the treatments with Fe(II)-phosphate ranged from 28.5 to 35.3 kg tree-1 and did not differ significantly from that of the trees treated with chelate (32.0 kg tree-1). The untreated trees showed severe chlorosis and low yield (4.6 kg tree-1). There were significant differences in soluble solids but not in firmness and titrable acidity of the fruit. Thus, Fe(II)-phosphates show long-term effectiveness to prevent iron chlorosis under field conditions.