BREEDING FOR YIELD IN SPRING CEREALS

Tests were conducted from 1956 to 1961 with barley and oats to study the relationship among tillers, grains per head, and grain weight, and their influence on yield under Ontario conditions. It was found that grains per head, rather than tillers per plant, was the most reliable component to use in estimating yield. However, both tillering rate and grains per head were variable and were found to be influenced by variety, seeding rate, seeding date, and soil fertility levels. Grain weight was less influenced by these factors and contributed little to the yield variation. Tillers per plant were negatively correlated with grains per head.It is suggested that in Ontario when selecting for yield, in addition to components of yield, the morphological plant characteristics determining photo-synthetic efficiency per unit area could also be considered.