Premium
Simultaneous Selection for High Yielding and Stable Crop Genotypes
Author(s) -
Kang M. S.,
Pham H. N.
Publication year - 1991
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1991.00021962008300010037x
Subject(s) - yield (engineering) , statistics , rank (graph theory) , stability (learning theory) , mathematics , index (typography) , statistic , selection (genetic algorithm) , variance (accounting) , combinatorics , economics , computer science , artificial intelligence , materials science , accounting , machine learning , world wide web , metallurgy
Integration of stability of performance with yield is essential in yield trials. Methods that select for high yield and stability have been developed, but have not been compared for their usefulness. Our objective was to compare these methods and to study their relationship to yield and the stability‐variance statistic (σ 2 i ). We compared Kang's (1988b) rank‐sum method (Index 1; equal weights for yield and σ 2 i ) and four additional rank‐sum indices [two (Index 2), three (Index 3), four (Index 4), and five (Index 5) times more weight for yield than for stability variance] with Hühn's (1979) S 3 i and S 6 i statistics, and Lin and Binns' (1988) P i . All statistics were calculated for each of five sets of data from international maize ( Zea mays L.) yield trials. In Set 1, low σ 2 i 's (indicative of stable performance) were generally associated with high yield ( r 3 = 0.73), but in Set 2 low σ 2 i 's appeared to be associated with low yield ( r 3 = −0.46). Index 1 ranks were positively correlated with σ 2 i ranks in Sets 2 to 5 as were S 3 i ranks. Index 1 and S 3 i offered an opportunity to select for both stability based on σ 2 i and yield. Indices 2, 3, 4, and 5, and P i favored selection primarily for yield. It was assumed that the top 50% genotypes would be selected, in Sets 1 and 2, on the basis of yield rank alone or individual statistics. In Set 1, Index 1 was slightly more conservative than S 3 i , in that Index 1 selected a higher yielding genotype from the two lowest yielding genotypes than did S 3 i . S 6 i was slightly more conservative than Index 1. Index 1 was intermediate between S 3 i and S 6 i . In Set 2, Index 1 and Index 2 were more conservative than σ 3 i , whereas σ 6 i was more conservative than Index 1, but less conservative than Index 2. P i favored selection for yield only. We concluded that Kang's rank‐sum method (Index 1 here) and Hühn's S 3 i and S 6 i statistics would be useful for simultaneously selecting for yield and yield stability.