Evaluation of yield stability of flue cured tobaccos using some univariate and multivariate stability statistics

Introduction:

The genotype × environment interaction is one of the most important factors in limiting breeding programs. The compatibility of a given genotype determines its ability and genetic capacity to produce high production and yield stability in different environments. Therefore, it is necessary to study the genotype × environment interaction in order to introduce stable genotypes in different environments. The current research was carried out to determine the compatibility and stability of yield for 10 tobacco genotypes and estimate the most stable genotype for tobacco growing areas in Iran (i.e. Guilan, Mazandaran and Golestan provinces). 

Materials and Methods:

In this study, 10 male sterile flue cured tobacco genotypes, which they completely tested in terms of quality and taste, including 7 internal modified hybrids named RVH5, RVH6, RVH8, RVH27, RVH30, RVH48, RVH70 along with 3 imported genotype DVH2101, PVH19 and NC100 in 6 tobacco growing regions of Guilan, Mazandaran and Golestan in two years of experiment (2018-2019). In total, 12 environments were studied in a randomized complete block design with 3 replications. Combined analysis was performed by considering the environments as a random factor and genotypes as a fixed factor. Then the stability analysis of genotypes was done by 10 stability statistics and the first 3 components of AMMI (Additive Main effects and Multiplication Interaction) analysis including IPCA1, IPCA2 and IPCA3. Analysis of variance was done using SAS.9 software, AMMI analysis and related graphs were performed using IRRISTAT software. 

Results and Discussion:

The results of combined analysis of variance showed that the effect of location and year × location interaction, the effect of genotype and genotype × year × location interaction were significant at the 1% probability level for yield. Genotype × environment interaction analysis using the first three main components of the AMMI analysis justified 69.47% of the total variance. The lowest coefficient of variation was related to DVH2101 and RVH27 genotypes. RVH30, RVH27 and RVH8 genotypes were selected as the most stable genotypes. According to Rick equivalence and Shoklaʹs stability variance parameters. The least squares deviation from the regression line were related to RVH27, RVH30 and RVH8 genotypes. In terms of regression coefficient, RVH27, RVH30 and DVH2101 genotypes had a line slope close to 1. Based on the coefficient of determination and Jenkins and Perkins statistics, RVH27, RVH30 and RVH8 genotypes were in the same rank in terms of stability. RVH27 and RVH30 were the most stable genotypes with superiority index. RVH27, RVH30 and DVH2101 genotypes were selected as the most stable hybrids based on the AMMI first model due to their proximity to the plot center, low interaction and high yield. Based on the second model of AMMI analysis, RVH27 and DVH2101 genotypes were selected as stable genotypes. In the third AMMI model, the RVH27 genotype was selected as the stable genotype due to its closeness to the biplot center and high yield. 

Conclusion:

Considering most of the stability statistics and AMMI analysis, the most stable genotype was RVH27 hybrid. RVH30 and DVH2101 hybrids were placed in the next categories of stability. In addition, genotypes RVH6 and NC100 showed high private compatibility to Tirtash tobacco growing area and genotype RVH27 to Rasht region.