To evaluate the effect of soil tillage and management of winter cover crops (black oat+radish intercrop) on the soil temperature, an experiment was conducted in a Nitossol (Alfisol) in Botucatu, state of Sao Paulo, Brazil, in the 2000 autumn/winter season. A design in randomized blocks was used in a 3*3 factorial scheme (three tillage and three cover crop managements). Soil tillage consisted of: conventional tillage, conservation tillage with chiselling, and no-tillage. The cover crops managements included plant killing with post-emergence herbicide, rolling, or shredding. The soil temperature (thermocouples) was evaluated at a depth of 5 cm, every hour, 7, 14, 30, 45, and 60 days after plant emergence; the soil water content at a depth of 10 cm, at the same dates and, the soil surface coverage (dry mass and cover index) were measured immediately after treatment application. Lower soil temperatures were observed in the no-tillage system than under conventional tillage until the 14th day after plant emergence (DAE). From the 30th DAE, the temperature was no longer influenced by the treatments due to the soil cover and sufficient water availability in the soil. The cover managements with plant rolling, shredding, or herbicide had no influence on the soil temperature. The soil temperature did not affect the development of the cover crops.