A comparison of several procedures for mass selection in winter rye I. The relative quality of the procedures

Abstract

The HS-family offspring from each of 463 plants occurring in a population of winter rye were grown in a randomized complete block design. According to 4 sets of requirements, 4 groups of HS-families were created. The HS-families belonging to a certain group were considered to be superior with regard to the specified requirements. Independently from this, selection among the 463 parental plants was simulated. The selection was according to 3 methods (truncation selection, grid selection and honeycomb selection) applied with regard to a single plant trait or with regard to two plant traits. Altogether 9 different single-trait and two-trait criteria were studied for each selection method. The quality of some selection procedure or the other, i.e. a combination of a selection method and a selection criterion, was measured on the basis of the frequency of plants giving rise to a superior HS-family among all plants selected by the procedure. It is concluded that grid selection and honeycomb selection are more effective than truncation selection. Further it is tentatively concluded that honeycomb selection is more reliable than grid selection.