Soil is a complex habitat where inorganic components, organic molecules including proteic biomolecules and metals may interact to each other giving rise to different aggregates depending on the sequence order in which the several components have reacted. In south of Chile, soils are ash-derived, acidic soils and belong to the Andisol order. One of their most important features is the high phosphate retention in their allophane fractions, and mainly in aluminum- and iron-humus complexes. Furthermore, they have usually high levels of available manganese and deficiency of molybdenum. Therefore, the application of molybdenum to these soils as fertilizer might be essential, since molybdenum is involved in the nitrogen cycling as a cofactor of nitrogenase and nitrate reductase enzymes. The main goal of this work was to investigate the effect of molybdenum and manganese, at quantitative levels similar to those found in Andisols of southern Chile, on the interaction between acid phosphatase and the clay fraction extracted from a Chilean Andisol. The activity and the kinetic properties of the resulting aggregates were measured as well. In these soils, phosphatases may play a key role because they may strongly affect phosphorus availability by favoring the formation of inorganic phosphorus by the hydrolysis of organic P compounds. The clay contained < 1% of organic carbon, and had a point of zero charge(PZC) of 8.5, thus indicating that aluminum and iron oxides were the main surface components. Two different experimental conditions were adopted to simulate naturally occurring soil situations. Phosphatase–clay-molybdenum/manganese complexes were formed either by mixing different levels of molybdenum and manganese ions with the clay and after adding the phosphatase or by earlier immobilizing the enzyme onto the clay and then by adding the different amounts of the metals. Phosphatase-clay complexes served as control. The activities of the insoluble complexes and the supernatants fractions, obtained by centrifugation of the mixtures above described, were measured under standard conditions with p-nitrophenylphosphate at saturing concentrations. The activity of the free phosphatase in the presence of either molybdenum or manganese was measured, as well. The enzyme was completely retained onto the clay and its activity was 30% higher as compared to that of the free enzyme, thus indicated a partial activation of the enzyme after its adsorption to the clay material. When manganese cations or molybdenum anions were added to the clay before the enzyme addition, a general, though not very, high decrease of activity was measured (on average ~16% and 13-19% inhibition with Mn and Mo, respectively). By contrast a detectable reduction of activity was observed when the molybdenum anion was put in contact with the enzyme-clay complex. Indeed, the phosphatase activity decreased from 14 up to 40% of that measured in the absence of Mo, when its concentration increased from 0.23 up to 2.0 mg Mo mL^-1. A detectable increase of the Km constant also occurred, thus indicating a strong influence of Mo on the affinity of the substrate for the clay-immobilized phosphatase. The different results obtained under the two experimental indications might suggest that different adsorptive sites onto the clay surface were exposed to the interaction with the enzyme depending whether the clay had previously interacted with the Mn or Mo. Acknowledgements: This work was supported by the Andes Foundation project no. C13755-28, Fondecyt project no. 1020934, DIDUFRO no. 120316, MECESUP FRO 0309, and partly by Ministero dell'Università e della Ricerca, Italy, PRIN 2002-2003.