A new type of analysis of variance for grazing experiments with bovines, taking in account groups of animals

Abstract

A new type of analysis of variance is suggested for grazing experiments with bovines. In an experiment with 18 plots, six blocks of 18 steers (called Groups) were used, one steer for each plot. The consideration of Groups of animals and Blocks of pasture plots, similarly to rows and columns in Latin square experiments, increases the number of degrees of freedom for error from 8 to 93, in the example discussed, with a corresponding increase in the precision of the experiment. Another consequence of the method is the possibility of reduction of the cost of grazing experiments, through the decrease of the number of plots.