Variability and Correlation between the Seed Yield and its Component in Alfalfa (Medicago sativa L.) Populations under Dry Land Farming System, Hamadan, Iran

Document Type : Research and Full Length Article


1 Research Institute of Forests and Rangelands, Tehran.

2 Islamic Azad University, Borujerd Branch

3 Agricultural Research Center, Hamadan


. In order to study the variation for seed yield and its components, 200 accessions of alfalfa (Medicago sativa L.) were sown as drilled plots, using alpha designs/unreplicated with two repeated entries with in all of 10 blocks under dry land farming system in Kabodarahang Research Station, Hamadan, Iran, during 2010 to 2011. Data were analyzed for descriptive statistics, correlation, regression, path analysis, factor analysis and cluster analysis. Results showed considerable variation for all of traits. The best 20 entries were identified. Among them, Cody, Kerissary, Italian, Australian, Hamadani, Bami, Jirofti and Faridonshahr with average values of 900 to 1250 kg ha-1 forage yield and 85 to 107 kg ha-1 seed yield were introduced for breeding improved synthetic varieties. Results of correlation analysis showed that DM yield was positively correlated with node number, plant height, stem number, vegetation score, seed yield, and pod weight. Similarly, seed yield had positive correlation with pod weight, DM yield, vegetation score and plant height. Results of path analysis showed that DM yield and pod yield had higher direct and total effects on seed yield. Results of factor analysis were accounted for 77% of total variation for first five factors. Factor 1 which was accounted for 25% of variation was associated with DM yield, seed yield and pod yield. This factor was regarded as productive factor. Factor 2 accounted for 14% of variation was known as stem elongation factor, since it was included for plant height and node number. Based on ward cluster analysis method, 200 entries were divided into 5 groups. Accessions in cluster 4 were averaged well above the overall mean for DM and seed yield. The results of this study indicated that selection of variables in productivity factor (factor1 and cluster 4) could enable breeders to release the desirable increment in forage yield of alfalfa.