Efficiency of border rows in replicated sugar cane variety trials
- Cite this article as:
- Skinner, J.C. & Hogarth, D.M. Euphytica (1978) 27: 629. doi:10.1007/BF00043195
Each row was weighed separately in 34 sugar-cane variety trials harvested from 1973 to 1976. The efficiency of weighing only the middle row(s) compared with the whole plot was estimated by calculating coefficients of variation and theoretical gains from selection.
It was not efficient to weigh only the middle row in trials with 3-row plots. Gain from selection for true yield (that is, yield in the absence of competition from other varieties) was reduced by 7% and the coefficient of variation increased by 4%, this representing an increase of about 33% in the variability of the trials.
It was efficient to weigh only the middle two rows of 4-row plots, especially for trials on commercial farms. This gave a 2% increase in gain from selection and the coefficient of variation was increased by only 0.9%. It reduces the weighing time and amount of cane weighed, minimizing the delay to commercial harvesting while trials are being weighed.
On Experiment Stations where land is a limiting factor, the most efficient method is to continue the present practice of planting 3-row plots and weighing the whole plots.