Mass selection for low oxidation of root flesh was initiated in the fourth generation of an open-pollinated sweetpotato [Ipomoea batatas (L.) Lam.] population. Two selection schemes were followed which provided different selection pressures by varying effective population sizes. In one (population A), selected plants were randomly intercrossed by insects each cycle. In the other (population D), approximately 10% of the randomly intercrossing population were selected each cycle and their true seed used to plant the next generation. After 2 cycles of selection in A and 3 in D, they were compared to appropriate generations of the base population. Results were in agreement with selection theory and closely paralleled those obtained with other crops. More rapid advance was made with A, which requires 2 seasons per cycle for any trait not measured in the seedling stage. Good advance was made with D, which allows 1 cycle per season. Study of 21 other traits indicated more changes in unselected traits in A than in D, thus favoring the method of D in early generations of mass selection in sweetpotato. The rapid increase of low oxidizing plants in this study suggests that selection for low oxidizing cvs. may reduce associated processing problems.
Received for publication March 27, 1972.
Research Geneticist, Vegetable Breeding Laboratory, Plant Science Research Division.
The author is indebted to F. W. Martin, Agricultural Research Service, U. S. Department of Agriculture, Mayaguez, Puerto Rico, for providing seed increase nurseries; and to E. J. Koch, Biometrical Services, USDA, Beltsville, Maryland, for assistance with the statistical analysis of data.