Comparison of statistical methods for evaluating clones in a potato breeding program
The evaluation of clones in early stages of a potato (Solanum tuberosum L.) breeding program is characterized by limited availability of seed tubers, and by large numbers of clones. The purpose of this study was to compare statistical methods for such evaluation, taking these limitations into account. One hundred clones were evaluated at two locations, Maria da Fé and Lavras, in Minas Gerais State, by the use of a square lattice and an augmented block design. The layout of the latter has been used also to evaluate the methods of moving average and check plots. The individual analysis of Maria da Fé showed similar coefficients of variation among designs, as well as similar rankings of treatment means, but higher differences were observed in Lavras. The techniques of moving average and check plots had low efficiency on removing environmental effects. The joint analysis of variance showed that precision was similar for both designs and the use of moving average did not alter substantially the coefficients of variation and heritability. The augmented block design appeared to be an efficient alternative to the square lattice for selection of clones in early stages of a potato breeding program, as it demanded less resources, with acceptable loss of precision in the estimation of genetic parameters.
| Main Authors: | , , |
|---|---|
| Format: | Digital revista |
| Language: | por |
| Published: |
Pesquisa Agropecuaria Brasileira
1997
|
| Online Access: | https://seer.sct.embrapa.br/index.php/pab/article/view/4738 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|