Ruudun koko perunan lajikekokeissa

Authors

• L. Yllö Kasvinviljelylaitos, Maatalouden tutkimuskeskus, Tikkurila

DOI: https://doi.org/10.23986/afsci.71622

Abstract

In the early years of this century, potato trials at the Department of Plant Husbandry were made using large plots, having a size of 30—40 m2. Later, in order to reduce labour costs, this size was reduced, and in the years 1931—58 the main potato variety trials had plots averaging only 14.8 m2 in area, with 3–8 replicates (usually 5). Generally, the plots were long and narrow and contained two rows each. In the period 1943—58 the trials were arranged according to the row system, which gave good results, with an average error of 2.2 %. In 1963 and 1964 the effect of plot size was investigated in potato variety trials on humus soil. The main plot consisted of four 5.1-metre long rows. Each row was separately weighed at the time of lifting, thus providing data for the study of four different plot sizes. The block method with 4 replicates was used. It was found that the size of plot had a great effect on the order of yields (Table 1). Variance analyses showed that varietal differences were significant even in small plots of 6.6 m2, provided that the divergences were at least 18 % of the standard yield. In general, smaller differences were not significant even when the plot size was increased. In 1963 the standard error distinctly decreased with an increase in the plot size, but in 1964 this effect was considerably less. On the basis of trial results as well as the labour involved, it can be concluded that the optimum plot size depends on many factors, such as growing conditions, experimental arrangement, and accuracy in performing the trials.