Red clover plants inoculated with Glomus mosseae were grown in a sterile pasture soil containing 50 mg Zn kg(-1) in 'Plexiglas' (acrylic) containers with nylon net partitions (30 microm mesh) designed to separate the soil into a central root zone and two outer zones for hyphal growth with no root penetration. Two porous plastic soil moisture samplers were installed in each pot, one in the root compartment and the other in one of the hyphal compartments. The soil in the outer compartments was amended with one of the four application rates of Zn (as ZnSO4) ranging from 0 to 1000 mg kg(-1). Non-mycorrhizal controls were included, and there were five replicates of each treatment in a randomised block in a glasshouse. Uninoculated plants received supplementary P to avoid yield limitation due to low soil P status. Plants grew in the central compartment for nine weeks. Soil moisture samples were collected 4, 24 and 62 days after sowing to monitor changes in the Zn concentration and pH of the soil solution. At harvest, the mean mycorrhizal infection rate of inoculated plants ranged from 29% to 34% of total root length and was little affected by Zn application. Root and shoot yields were not affected by mycorrhizal infection. Plant Zn concentration and uptake were lower in mycorrhizal plants than non-mycorrhizal controls, and this effect was more pronounced with increasing Zn application rate to the soil. Soil solution Zn concentrations were lower and pH values were higher in mycorrhizal treatments than non-mycorrhizal controls and the mycorrhiza effect was more pronounced at higher Zn application rates. The protective effect of mycorrhiza against plant Zn uptake may have been associated with changes in Zn solubility mediated by changes in the soil solution pH, or by immobilisation of Zn in the extraradical mycelium.