Risk evaluation of herbicide-tolerant (HT) crops includes the assessment of the presence and impact of volunteers. In this study, estimates of frequency and persistence of volunteer populations of rape were obtained from field surveys on farms in eastern (Quebec) and western (Manitoba, Saskatchewan and Alberta) provinces of Canada, and from field experiments conducted in both agroecological regions. Field surveys were conducted from June to August 2000 in eastern Canada, during 1995 in Saskatchewan, and during 1997 in Manitoba and Alberta. As expected, the density of rape volunteers decreased with time since the last rape crop. An average of 0.2 and 0.5 volunteer rape plants/m 2 were found in fields 5 and 4 years after rape production, in eastern and western Canada, respectively. Experimental results suggest the presence of a persistent rape seed bank, although it is likely that volunteers contribute to the replenishment of the seed bank over the years. Overwintering rape plants were also observed in eastern Canada in no-till cropping systems. Weed management in Canadian cropping systems generally provides adequate control of volunteer rape but still allows long-term persistence of low density populations.
