Field experiments were conducted in southern New South Wales to determine the effect of surface-retained wheat stubble on the emergence, growth and yield of canola. The 5 experiments included treatments to investigate the impact of stubble load, stubble cultivar and level of decomposition as well as the impact of different environments on the crop response. Overall, 5 t/ha of surface-retained wheat stubble reduced the rate of emergence, plant establishment (mean reduction 33%), vegetative biomass (–56%) and yield (–23%), although the impact varied with site and season. Laboratory experiments assessing the phytoxicity of stubble revealed the possible role of allelopathy in the growth response at 1 site; however, there was no correlation between laboratory phytotoxicity of different stubble cultivars and their impact on canola growth at any other site. Wheat stubble comprising thinner stems (lower straw linear density) had a greater impact on emergence at 2 of the sites, indicating a possible role of reduced light penetration in the growth response. Colder temperatures on the surface of the stubble also reduced emergence and growth, and caused seedling death at the coldest sites. The experiments confirm the widely observed phenomenon of poor canola growth in surface-retained wheat stubble, and suggest several possible mechanisms for the effect, although further studies are required to determine their relative importance in different environments.