ISSN:
1573-5060
Keywords:
Oryza sativa L.
;
rice
;
pure lines
;
hybrids
;
rough rice yield
;
head rice
;
total milled rice
;
selection index
;
path analysis
Source:
Springer Online Journal Archives 1860-2000
Topics:
Agriculture, Forestry, Horticulture, Fishery, Domestic Science, Nutrition
Notes:
Abstract High rough rice yields are important for the profitable production of rice (Oryza sativa L.). The value of rough rice is determined primarily by head rice (whole milled grains) and broken rice, together referred to as total milled rice. The objective of this study was to optimize selection for rough rice yield, head rice, and total milled rice using path-coefficient analysis and selection indices. Thirty-seven long grain pure lines were grown in an experiment at three Arkansas locations during 1993 and 1994. In another test, twenty-eight F1 hybrids obtained from an eight-parent half diallel of long-grain rices were grown at two locations in Arkansas during 1991. For the pure lines, path analysis revealed that rough rice yield was approximately twice as important in determining total milled rice/ha than head rice. Path analysis for the hybrids revealed that rough rice yield was an even greater component in determining total milled rice/ha. Based on yield components for hybrid rice, panicle density had the largest direct effect determining total milled rice/ha. Filled grain/panicle, grain weight, and head rice, in that order, were secondary, but positive, factors determining total milled rice/ha. A selection index for pure lines based only on selection for head rice was 19.2 percent as effective as selection for total milled rice/ha directly. When selection for total milled rice/ha was based solely on rough rice yield, selection was 82.7 percent as effective as selecting for total milled rice/ha directly. Selection to improve total milled rice/ha should concentrate on increasing both rough rice yield and head rice. Rough rice yield should receive greater priority than head rice in selection strategies.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1023/A:1018331809955
