| dc.description.abstract | The aims of this experiment are to study the geneUc variation, heritability of the wlleat germplesm collection and cotrelation among chafllctelS using pathcoefficient analysis. This experiment was conducted at Rice Research Station in Kuningan, West Java on June until October 2002 using RandOmized Completely Block Design (RCBD) with three replications. each of the twenty-seven genotypes of wheat was planted on four rows with 3 m lenght. The interrow and interplant spacing were 25 em and 10 em, respectivity. Deta obtained from the twenty-seven genotypes were subjected to variance analysis and the heritability of the traits _mated by component of variance method. Correlation coefflcients were computed according to the formula of Goulden (1959). Direct and indirect path coefflcients were calculated as described by Dewey and Lu (1959). High geneUc varianon displayed by l00-grain weight, numberofUllelS per plant and number of spikes per plant The esUmation of heritability ranged from 0 to 94.58%. The days to nower, plant height, spike length, number of greins per spike, and l00-grain weight were uncotrelated with grain yield per plant. In the other hand, the days to harvest, number of tillelS per plant, number of Spikes per plant, dry spikes weight, and harvest spikes weight were positively and significanty correlated with grain yield per plant. The number of graints per plant, number of spikes per plant, number of grains per spike, and l00-grein weight each had a posiwe direct effect on grain yield per plant. The indireci effect on grain yield per plant via number of tillers and number of Spikes (productive tillers ) per plant showed by days to harvest, dry Spikes weight, and harvest Spikes weight. The number of productive tillers per plant was a good criteria to be used towards improvement of grain yield. | id |
