Two experiments were conducted at the Mallee Research Station, Walpeup, between 1985 and 1989 to determine whether increases in wheat yield that occur after long fallows result from improvements in the supply and use of additional soil water conserved during the fallow. Although long fallows increased the amount of water stored in the soil at sowing (average 22 mm) and the yield of wheat (0.26 to 1.37 t/ha) in the first experiment, the results suggest no causal relationship between these increases. Improvements in wheat yield were attributed to increases in soil nitrogen availability and to control of cereal root diseases rather than to any increase in soil water conservation. This was confirmed in the second experiment, which was managed to ensure that nitrogen supply and cereal root diseases were not limiting crop production. Increases in soil water content at sowing resulting from long fallows did not result in higher wheat yields. This study suggests that long fallows cannot be justified on the basis of this increased soil water storage, as much of the additional soil water accumulated during the fallow period is stored in the lower part of the rootzone. Movement of this water below the rootzone during the growing season appears to be the main reason for the additional water stored at sowing, with long fallows failing to increase wheat growth and yield.