<p>Two selection procedures are examined and compared in the breeding for quantitative traits in self -fertilizing crops. They represent two more or less extreme breeding schemes: a) Early Selection (ES), with early generation cross selection, followed by line selection. The cross selection is based on F <sub><font size="-2">3</font></sub> estimates of the relevant genetic parameters predicting the distribution of F∞inbred lines; b) Single Seed Descent (SSD), where a quick advancement towards the F <sub><font size="-2">5</font></sub> is combined with line selection only in the F <sub><font size="-2">6</font></sub> . Both field trials and computer simulation studies show that the early cross selection is not an efficient way of breeding. Cross prediction will often be erroneous due to severe bias on estimates of the genetic parameters. This bias is caused by non-additive genetic effects, environmental errors and, especially, intergenotypic competition. The procedure of single seed descent can produce superior inbred lines in a more consistent, cheaper and faster way. It appears though that, with maximum input, ES may produce better lines than SSD. Which procedure is more preferable therefore depends on the effort the plant breeder is willing to spend on a relatively small genetic gain.
|Qualification||Doctor of Philosophy|
|Award date||5 Jan 1993|
|Place of Publication||S.l.|
|Publication status||Published - 1993|
- self compatibility
- selection methods