Potato haploid performance in 2x × 4x crosses

Abstract

Nineteen haploids (2n = 2x = 24) extracted from four tetraploids ofSolanum tuberosum (2n = 4x = 48) with high 2n egg frequencies and known mode of 2n egg formation were evaluated via 4x progeny from 2x × 4x crosses (haploid × cultivar) for total tuber yield, tuber appearance and tuber set. Ninety-eight, 4x families (15-20 genotypes each) were planted in the randomized complete block design with two replications at two locations. Analyses of variance were conducted for all haploids over three common testers, and each of the four groups of haploids individually over a variable number of testers. The performance of haploids, based on general combining ability effects, was consistent for all traits, when tested with either three or more testers. Diploid clones can be evaluated in 2x × 4x crosses using three unrelated adapted 4x testers. Nonorthogonal comparisons indicated no significant differences for all traits among four groups of haploids, and between modes of 2n egg formation. Significant variation within each group of haploids indicates that selection at the haploid level should be carried out among haploids from either the same or different cultivars. Superior haploids can be utilized in 2x × 4x and 2x × 2x breeding schemes to generate 4x clones.