Mass and line selection can produce equally uniform rice varieties

Abstract

We compared intra-varietal variation of two rice varieties (Ashoka 200F and Ashoka 900F) produced by a very simple bulk breeding method (mass selection with no line selection at any stage) with one line-selected variety (Ashoka 228), all derived from the same cross. Their parents, the upland variety Kalinga III and the irrigated transplanted medium-lowland variety IR64, both originated through line selection and were used as control varieties. Panicle-to-row progenies of all the varieties were evaluated in eastern India under irrigation in the dry season and in the rainy season using two water regimes; entirely rainfed or with supplementary irrigation. Intra-varietal variation for quantitative traits, irrespective of the method of breeding, was low. Only Ashoka 200F had significant variation for grain yield, days to flowering and plant height but only in the dry season. This season was not the target of the breeding programme and selection would be unlikely to produce a worthwhile response for the rainy season. At 43 SSR loci there was more genetic variation between lines within the bulk-selected varieties than within Ashoka 228. Kalinga III was the least variable variety while IR64 had less heterozygosity but greater heterogeneity than the two bulk-selected varieties. Despite the greater simplicity of the method, mass selection in bulk populations produced varieties that met the distinctness, uniformity and stability (DUS) criteria for seed certification in India. Such uniformity was achieved because of the high selection pressures applied to the bulk. We conclude that this very simple bulk-population breeding approach is highly cost-effective and produces sufficient seed for wide testing earlier than any alternative method.