Montmorillonite clay type soils have been implicated in cases where the excessive persistence of atrazine caused damage to sensitive follow-up crops. A study was conducted to evaluate the influence of temperature and soil water content on the persistence of atrazine in a clay soil (total clay 55%, 80% montmorillonite, 0.7% organic C, CEC = 48 me 100 g−1, pH = 7.3) and a loamy sand soil (total clay 7%, 0.34% organic C, CEC = 0.8 me 100g−1pH = 5.4). The latter soil represented soils in which atrazine carry-over is not expected to occur. A factorial experiment (2×2×3×3) comprising the two soil types; atrazine rates of 1 and 2 mg kg−1; soil water contents of about 0 (air-dry), the amount at field capacity and 2×field capacity; and temperature regimes (day/night) of 30/16°C, 30/8°C and 16/8°C, was conducted in growth cabinets. Samples of soil incubated in the dark were analysed for atrazine on day 0 (herbicide applied), day 30 and day 60 by means of high performance liquid chromatography (HPLC). By day 30, distinct but not significant degradation of atrazine had occurred in the light soil only. Subsequent determinations on day 60 after atrazine application, however, revealed that soil water at the field capacity level significantly increased degradation in both soils, with a difference in magnitude such that the half-life of atrazine in the loamy sand was reached after about 60 days, while at the same stage 70–75% of the applied atrazine remained in the clay soil. The increase in soil water from field capacity to 2×field capacity had little or no effect on persistence. Virtually no degradation of atrazine occurred in air-dry soil. The lowest temperature regime significantly reduced the rate of degradation in the light soil only. It is suggested that the higher pH and higher adsorptive capacity of the clay soil caused atrazine to persist longer in that soil than in the loamy sand soil.