• C. Randal Linder

    1. Department of Ecology and Evolutionary Biology, Box G-W, Brown University, Providence, Rhode Island 02912 USA
    Search for more papers by this author
    • Present address: Department of Botany, The University of Texas, Austin, Texas 78713 USA. E-mail: rlinder@mail.utexas.edu


Because wild, weedy-ephemeral species take advantage of periodic disturbance for growth and reproduction, many have considerable seed dormancy and longevity, coupled with germination cued on environmental changes correlated with disturbance. Brassica napus canola is derived from a weedy ephemeral and has sexually compatible wild relatives that are weedy ephemerals, most notably B. rapa. Hence, if transgenes introduced into B. napus canola alter its seed bank dynamics to be more similar to its wild relatives, the chance of transgene escape and persistence is increased. Further, if introgression of the transgene into a wild relative enhances or does not disrupt the seed bank dynamics of wild plants, the transgene could persist in wild populations. I conducted experiments with transgenic, oil-modified B. napus canola (high-stearate and high-laurate types) and wild B. rapa × B. napus canola hybrids (high-laurate type) to determine whether they possessed seed dormancy and germination cuing characteristics favoring persistence of escaped transgenes. Seeds of high-stearate B. napus canola and its untransformed parental type were germinated in growth chambers in a complete factorial design of light (full light, darkness, simulated foliage shade), nutrient concentration (high, low), and temperature (10°, 20°, 35°C). Most combinations of factors did not alter high-stearate canola’s timing of germination or total proportion of seeds germinated relative to its nonpersistent parental control. Where effects were seen, they were disadvantageous, with high-stearate seeds germinating more slowly than controls (35°C:full light:high nutrient) and in lower proportions (35°C:darkness:high nutrient and 35°C:full light:high nutrient). Ungerminated high-stearate seeds in treatment combinations having reduced germination were primarily dormant rather than dead. Although most of their dormancy was enforced by high temperature, under some conditions (35°C:full light:high nutrients and 35°C:darkness:low nutrients), I detected significantly higher levels of induced dormancy for high-stearate seeds relative to controls. Hence, persistent high-stearate canola seed banks could form under some field conditions. For high-laurate B. napus canola and high-laurate wild B. rapa × B. napus canola hybrids and their controls, I germinated seeds in growth chambers, varying light and nutrients using the same treatments as above in a complete factorial design. Temperature was maintained at 10°C. High-laurate canola germinated later than its parental control, with the difference enhanced by low nutrients. In low nutrients, high-laurate canola also germinated at a lower proportion than its control. Both of these altered germination characteristics do not favor population persistence of high-laurate canola. Under some conditions (darkness:high nutrients), high-laurate canola had higher overall dormancy and induced dormancy than its control, suggesting the possibility of increased seed bank persistence. Finally, in spite of expectations that the high-laurate wild × crop hybrid would show strong maternal effects, this hybrid and the hybrid control always germinated at proportions at or near 1.0, whereas the wild parent had very low germination proportions and high levels of dormancy, especially in simulated foliage shade. These results suggest that high-laurate wild–crop hybrids lack germination cuing mechanisms and will germinate primarily at inappropriate times. However, when they do germinate with wild B. rapa, they are likely to compete well with it because the high-laurate hybrids germinated and grew as fast or faster than their wild parental control. This should provide opportunities for backcrossing to wild B. rapa.