Gene flow among herbicide-resistant (HR) canola varieties can lead to the development of multiple HR canola plants, creating volunteer canola management challenges for producers. In western Canada, escaped populations of HR canola are ubiquitous outside of cultivated fields, yet the extent of gene flow resulting in herbicide resistance trait stacking in individuals within these populations remains unknown. The objectives of this study were to document the presence of single and multiple herbicide resistance traits and assess the extent of gene flow within escaped canola populations. Seed was collected from 16 escaped canola populations along the verges of fields and roadways in four agricultural regions in southern Manitoba from 2004 to 2006. Glyphosate resistance was found in 14 (88%) of these populations, glufosinate resistance in 13 (81%) populations, and imidazolinone resistance in five (31%) populations. Multiple herbicide resistance was observed at levels consistent with previously published canola outcrossing rates in 10 (62%) of the tested populations. In 2005 and 2006, maternal plants from two escaped populations were tested using trait indicator test strips for glyphosate and glufosinate resistance to confirm outcrossing events. In 2005, two of 13 tested maternal plants with single herbicide resistance traits produced progeny with both glyphosate and glufosinate resistance. In 2006, of 21 tested plants, 10 single HR maternal plants produced multiple HR progeny, and five nonresistant maternal plants produced resistant offspring. This is the first report indicating that intraspecific gene flow results in stacking of herbicide resistance traits in individuals within escaped canola populations, confirming that multiple HR canola volunteers are not confined to agricultural fields. Results of this study suggest that escaped populations of crop plants can contribute to the spread of genetically engineered novel traits, which has important implications for containment, especially for highly controversial pharmaceutical and industrial traits in crop plants.

Observational studies of weed emergence provide essential data for the creation and testing of predictive emergence models, with data ideally being collected from a wide range of sites representing a range of environments under which the seedlings of a given species emerge. The spring emergence of genetically engineered glyphosate-resistant volunteer canola was monitored in 20 farmers' commercial wheat fields over 2 yr, 2003 and 2004, in the southwestern region of Manitoba, Canada. Three different tillage systems, low-disturbance direct-seeded, high-disturbance direct-seeded, and conventionally tilled fields, were represented in this sample. Tillage system did not significantly influence the emergence periodicity of volunteer canola, but did influence total densities observed. Total volunteer canola densities in the surveyed fields were variable and ranged from 6 to 2,015 seedlings m−2. Conventionally tilled fields had the lowest densities of volunteer canola seedlings, and high-disturbance direct-seeded fields (no autumn tillage) had the highest densities. Volunteer canola is a relatively early spring emerging plant species requiring minimal accumulated growing degree days for emergence (calculated using a base temperature of 5 C). Emergence curves were steep, with a short window of volunteer canola emergence either prior to or within a spring-seeded wheat crop. To determine what proportion of autumn-broadcast canola seed (simulated harvest losses) would recruit the following spring and the influence of fall or spring tillage on this recruitment, a small plot experiment also was conducted at three sites. The following spring the percentage of canola seedlings that emerged ranged from 1.3 to 9.4% of the seed broadcast, depending upon the tillage treatment. The effect of tillage treatment on canola densities in the small plot experiment was similar to the field survey results.

Variation in spring emergence periodicity (both before and after crop seeding) of summer annual weeds is a potentially exploitable attribute that may be applied to weed management in canola. Tillage intensity, which is decreasing in the Great Plains of North America, may influence emergence periodicity of summer annual weeds. Emergence periodicity of common lambsquarters, field pennycress, green foxtail, redroot pigweed, wild buckwheat, wild mustard, and wild oat were monitored during the spring of 2000 in 17 producers' canola fields across southern Manitoba, Canada. The fields represented a region of approximately 2 million ha and included a broad range of soil types, agronomic practices, environmental conditions, and seedbank distributions. Fields were grouped into one of two broad tillage classifications (conventional or conservation). For most species, except redroot pigweed and wild mustard, conservation tillage promoted earlier emergence than conventional tillage in terms of both thermal and chronological time. The differences were significant even though there was only a limited range of tillage intensity for the two tillage classes within this region. Onset of canola crop emergence preceded that of all but one weed species in the conservation-tillage fields and five weed species in the conventional-tillage fields. This suggests that canola seeded in conservation- vs. conventional-tillage systems may have a competitive advantage by way of an earlier relative time of crop emergence. The influence of tillage system on weed emergence periodicity is likely due to the influence of tillage on the vertical origin of weed seedling recruitment because measurements of soil temperature and soil moisture did not help to fully explain the differences in emergence periodicity between tillage systems. The results from this study will facilitate weed control timing decisions in canola and provide validation data for weed emergence models.