Rapid-cycling Brassica species: inbreeding and selection of B. campestris for anther culture ability

A study was made to determine the feasibility of producing, by inbreeding and selection, lines of rapid-cycling Brassica campestris with high or low potential for anther culture. Since the base population of rapid-cycling B. campestris is self-incompatible, inbreeding was achieved by a combination of bud-pollination and the application of pollen to the cut surfaces of decapitated stigmas. Three inbred generations were raised, and in each generation plants were selected for high or low potential for anther embryogenesis. The proportion of viable pollen present in anthers, as indicated by a fluorochromatic reaction and a germination test, was also determined at each stage. Lines of rapid-cycling B. campestris with clearly defined high or low potential for anther embryogenesis were isolated in these experiments. Within each line, however, continuous variation was always observed. Pollen viability and anther efficiency were not correlated. Although inbreeding depression caused a significant decrease in pollen viability over the three generations, there were no obvious deleterious effects on anther efficiency. In general, over the three generations of inbreeding, no segregation in plant morphological characters was observed, although many developmental abnormalities were seen in the 3rd inbred generation and there was a marked reduction in the number of seeds set. No association between plant vigour and high or low anther efficiency was noted. All plants regenerated from anther embryoids of rapid cycling B. campestris were haploid. By treating anther-derived embryoids, axillary buds and whole plants with colchicine, dihaploid plants were produced, but these failed to set seed after self-pollination. The diploid nature of the plants was confirmed, however, when they produced normal seeds after cross-pollination with plants of the base population.