Published online by Cambridge University Press: 27 March 2009
(1) The method of taking four sub-samples and making in all 160 measurements of these sub-samples gives a satisfactory value for the average diameter of the sample.
(2) The average of the samples from each shoulder gives a good indication of the shoulder for each sheep.
(3) In comparing two sheep A and B we may take as almost certainly significant a difference between their two average shoulder diameters such that the ratio
that is a difference of some 8% of the average shoulder diameter of either of them for measurements taken as here indicated.