Inbreeding Depression in Maize by Selfing and Full‐sibbing¹

Three series of unselected lines from ‘Iowa Stiff Stalk Synthetic’ (BSSSCO) were developed by three methods inbreeding: selfing, ful‐sibbing, and full‐sibbing followed by selfing. Data were collected for 12 traits in nine environments to determine: 1) the rate of inbreeding depression by selfing, full‐sibbing, and full‐sibbing and selfing; 2) the differences among means of lines developed by the three methods of inbreeding at comparable levels of homozygosity; and 3) the phenotypic correlation between the traits with the change in homozygosity by inbreeding. Rates of inbreeding depression were significantly different among the three methods of inbreeding for plant height, cob diameter, yield, kernel weight, stand, and barrenness. In most instances the differences were small. Predicted yields by the linear regression model for F = 1.0 were 23.2, 24.6, and 23.9 q/ha for the selfed, full‐sib, and full‐sib and selfed series of lines, respectively, whereas with F = 0, yield was 69.7 q/ha. Rate of inbreeding depression for yield by selfing was b 1 = ‐‐46.53 q/ha and by full‐sibbing, b L = ‐‐45.11 q/ha. At comparable levels of homozygosity, means of the lines developed by selfing and full‐sibbing were not significantly different except for yield at 99% homozygosity. All correlations of traits with percentageo f homozygosityw ere very high. Linear regression of yield on percentage of homozygosity for 22 levels of homozygosity is BSSSCO accounted for 99.1% of the total variation.