Variations in Stand as Sources of Experimental Error in Yield Tests with Corn 1

The aim of investigators concerned with agronomic plat trials is to eliminate as many sources of error as possible. Variations in stand in test plats of corn are commonly obtained in spite of most careful attention to details of planting, thinning, and cultivation. It is essential to know the effect of such variations in order to obviate, so far as possible, such effect at harvest. Reduction of stand in varietal trials may result from seed-borne infection with the seedling blight organisms; and also from other uncontrolled sources of error, such as insect or rodent injury, variations in germination, particularly in a dry spring, and accidental destruction of plants during cultivation. Variations in stand in varietal trials may be obviated to a considerable extent by planting thick and later thinning to a uniform stand. Where studies on the effect of seedling blight diseases are being conducted, such a practice is obviously undesirable. Reduced stand as an important source of error in yield trials was demonstrated by Kiesselbach3 in studies conducted" for two years at the Nebraska Station. He found, in ~9~4, that the relative grain yields of 3-plant, 2-plant, and i-plant hills surrounded by hills having a full stand of 3 plants were ~oo, 82, and 74, respectively. In 1917, the corresponding relative yields were ioo, 83, and 50. In I9~4, 3-plant hills adjacent to one hill with 2 plants, one hill with ~ plant, one blank hill, and two blank hills, respectively, and otherwise surrounded by a full stand, were increased in yield 3, 5, 13, and 43%, respectively. In i917, the corresponding increases were ~, 9, 15, and 25%. Olson4 found variations in stand closely correlated with yield per acre. In the course of a seed study experiment in which ear rows of