Published online by Cambridge University Press: 27 March 2009
Two uniformity trials were carried out, one using S. 22 and the other Irish Italian ryegrass. In both trials the plants were grown at 6 in spacing and were harvested in basic units of 1 yd. If no allowance was made for guard rows the smallest plots were the most efficient in that they required a smaller total area of ground and fewer plants to detect a specified difference. For comparative purposes, the size of trial required to detect a difference of 7% of the mean was used. Allowing for 1 guard row round each plot, 2 yd plots were as efficient as those of 1 yd and had the advantage of requiring fewer replications. The effects of plot and block shape were considerable. In general long narrow plots in short wide blocks were more efficient. The choice of plot and block shape is most important. For a given plot size a poor shape may be less than half as efficient as a good one.