Genotype by environment interaction on popping expansion and yield in popcorn hybrids cultivated in Argentina --Burak, R, Broccoli, AM The open-pollinated popcorn varieties traditionally cultivated in Argentina were replaced, during the last years, by introducing popcorn hybrids principally from the USA. Fourteen of these hybrids commercially available were grown in 1997 and 1998 at three locations of the "Milk Belt" (34º 38 South and 58º 48 L , 23m altitude) . The trials were set up in a completely random block design with three replications and the statistical methods applied were AMMI (additive main effects and multiplicative interaction) using biplots in interpreting variety by enviroment interactions, for the variables yield and expansion volume (EV).

AMMI Model

Yge = m + ag + be + Â ln g gn rgn + ege

Yge = yield for the g genotype in e enviroment

m = general mean

ag = genotype effect

be = environment effect.

ln = n-th axis singular value (eigenvalue)

tgn and ren = corresponding genotype and environment scores (eigenvectors)

ege = normally distributed error.

Statistical model for the preliminary ANOVA procedure:

Yijk = m + ai + bj + (a x b )ij + rk(j) + eijk

Table 1: analysis of variance results for yield
 
Source DF SS MS
Environments 5 24650 4930**
Rep (Env) 12 23640 1970
Hybrids 13 20540 1580**
Hyb x Env 65 30550 470**
Error 156 40560 260

(*, **) Significant at the 0.05 and 0.01 probability, respectively.

Table 2: analysis of variance results for popping expansion
 
Source DF SS MS
Environments 5 180.45 36.09**
Rep (Env) 12 100.32 8.36
Hybrids 13 307.84 23.68 ** 
Hyb x Env 65 716.30 11.02**
Error 156 765.05 5.08

(*, **) Significant at the 0.05 and 0.01 probability, respectively.

The interaction that can be described by two multiplicative components can be represented in a biplot (see biplots I and II). Cultivars are represented by lines, trials by points, both end points are the representation of the end points of vectors starting at the origin. The length of a cultivar line reflects the amount of interaction for that cultivar, thus according with Figure1, most interaction is due to genotypes 1, 6, 10 and 14. The angle between cultivar lines corresponds to the interaction betwen the interaction residuals. Genotypes 3 and 12, the nearest distributed to the center of the biplot, have more stability. Genotypes 1 and 9, 13 and 8, 2 and 5 are very similar with a high positive correlation. 1 and 10 have a negative correlation. 14 and 6 are very dissimilar with a correlation close to zero. Genotype 14 with a high residual interaction with D environment. The same occurs with 6 in E environment and genotype 1 in A.

In Figure 2, expansion volume, there is less stability for 3, 5 and 6 and the most for genotypes 10 and 14. Similar performance is shown by 7-8, 1-13 and 14. In contrast 5 and 6 are quite alike. There is a particular performance of the genotypes, analysing both figures jointly, that confirms the negative correlation between yield and expansion volume.

Figure 1. Biplot of the AMMI model for grain yield, with 14 genotypes (1,....14) and 6 environments (A,....F). PCA1 and PCA2 are represented by eje1 and eje2.

Figure 2. Biplot of the AMMI model for expansion volume, with 14 genotypes (1,....14) and 6 environments (A,....F). PCA1 and PCA2 are represented by eje1 and eje2.
 
 


Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors.

Return to the MNL 74 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page