Much attention has been given to this subject during the last few years, and it has been shown that as often as not two plots of the same variety of wheat grown in the same field under conditions which are made as uniform as possible will differ in yield by 5 per cent. or more. Obviously it is impossible to make comparisons of the cropping power of different varieties of wheat as the result of trials in which single plots of each variety are grown. It is a deplorable fact however that the results of most of the trials which are published are based on single plots only of the varieties compared. Such results can have no claim to reliability. Single plots tests are excellent as local demonstrations, to give the farmers a chance of seeing the general characters of the various wheats in the field, but for the determination of cropping power their results are misleading. For the comparison of two varieties however an accuracy of about 1 per cent., which is good enough for the purpose in view, can be obtained by growing, harvesting and weighing separately, five separate plots of each variety under experiment, provided the plots are distributed in pairs over the experimental field. Still greater accuracy can be attained by growing very large numbers of very small plots of each variety in a bird-proof enclosure. The illustration shows such an enclosure at Cambridge where five varieties were tested, each on 40 plots. Each plot was one square yard, and the whole 200 plots occupied so small an area that uniformity of soil could be secured by hand culture. Several experimenters are now at work on these lines, and it is to be hoped that all who wish to carry out variety tests will either follow suit, or content themselves with using their single plots only for demonstrating the general characters of the varieties in the field. So far we have confined our discussion to the standard varieties, and we must now turn our attention to the work which has been done in recent years on the breeding of new varieties which will yield heavier crops than any of the varieties hitherto in cultivation. [Illustration: Fig. 2. Part of bird-proof enclosure containing many small plots for variety testing]