Abstract
Selection of canola (Brassica napus L.) for fatty acid (FA) composition is often based on single seeds or small seed lots, but information is lacking on variation among repeated samples. Variation in FA composition was measured among repeated samples of B. napus seeds by gas chromatography on oil extracted by cold solvents, from a selfed doubled-haploid line (Monty-028DH), and an open-pollinated B. napus cv. Surpass 501TT. The variation in FA composition among repeated samples of 1, 3, 5, and 10 whole seeds was compared with the equivalent weight of ground seeds or Soxhlet-extracted oil. The standard deviation of oleic acid for whole seeds of Monty-028DH decreased from 2.12% (single-seed samples) to 0.12% (groups of 10 seeds), but was constant across an equivalent range in weight of ground seed (0.15%) and in volume of Soxhlet-extracted oil (0.05%). A single-seed sample is not sufficient to separate two B. napus genotypes differing in mean oleic acid content of 60% and 70% at the 95% confidence interval, based on the sampling variance measured in these experiments. Similar trends were found for all the major FAs in both varieties.