Two years after establishment, areas of low vigour (LV) were noticed within a white clover (Trifolium repens)/ryegrass (Lolium perenne) pasture. These areas became progressively larger, and, two years later, the pasture had low vigour, comprising only 3% white clover. Clover tissue nitrogen and phosphorus levels were greater in high vigour (HV) than in LV herbage. There were no differences (P > 0.05) in "total" bacteria, fluorescent pseudomonads, or proportions of deleterious bacteria in LV and HV white clover roots (mean logio CFUs for total count and fluorescent pseudomonads were 9.4 and 8.8, respectively, and the proportions of deleterious bacteria encountered in these groups were 6.9% and 17.5%). There was a 9-fold difference (P < 0.01) in the white clover dry matter between LV areas and HV areas in February, increasing to 25-fold (P< 0.01) in March. The total dry matter yield differences, which included grasses and clover, were 2.7-and 2-fold (P < 0.01) for the two months. More fungi were isolated (P < 0.05) from LV white clover roots than from HV roots (164 and 104 isolates, respectively). This was mostly due to greater numbers (P < 0.01) of Codinaea fertilis in LV roots (111 cf. 41 in HV). In ryegrass roots, there were also greater numbers (P < 0.01 ) of total fungi in LV roots (97 cf. 50 total A00024 isolates in HV), mostly due to higher populations of Fusarium oxysporum in LV roots (29 isolates cf. 5 in HV). There was no difference (P > 0.05) in the Soil Pathogenicity Index between LV and HV soil. The plant parasitic nematode populations in 250 ml of soil in LV and HV areas were 42.7 and 9.7 for Heterodera and 28.1 and 3.9 for Meloidogyne, respectively.