Genetic and environmental correlations between yield components and popping expansion in popcorn hybrids --Burak, R, Broccoli, AM Six field trials under a three replication complete randomized block design were conducted; 14 simple popcorn hybrids were evaluated for: sowing date (early and late) and three plant densities ( 62.500, 74.000 and 85.000 plants per hectare), with standard experimental units. Triple factorial design was used for estimating sowing date, genotype and plant density effects, according with the following model:

yijqk = m + Bk(iq) + Gi + Dj + Fq + (G*D)ij + (G*F)iq + (D*F)jq + (G*D*A)ijq + eijqk

where:

yijqk = Ith treatment of gth plant density of the qth sowing date of the kth replication; m = overall mean; Gi = randomized effect of genotype i th; PDj = fixed effect of plant density jth; SDq = fixed effect of sowing date qth ; (G*F)ij =.G X SD interaction effect; (G*D)ij =.G x PD interaction effect; (PD*SD)ij =.PD x SD interaction effect; (G*PD*SD)ijq = triple interaction; Rk(jq) = replications nested in PD and SD effect; eijqk = experimental error randomized variable.

Genetic and environmental variances for each variable were estimated from expected mean square (Table 1) and covariances were calculated following the Kempthorne procedure.

MPxy = 1/2 ( MSx+y - MSx -MSy )

where:

MPxy = X e Y variables mean product; MSx y MSy = mean squares of X and Y variables; MSx+y = mean squares of the sum between X and Y variables.

Table 1. Expected means squares for factorial design, considering simple hybrids a randomized sample of the commercially available ones.
 
S. V. D. F. M.S Expected M.S
(R/PD)/SD (r-1) f d MS1 s2E + g s2r
G (g-1) MS2 s2E + r d f s2G
SD (f-1) MS3 s2E + g s2r + (f / f-1) r d s2GF + r d g fF
PD (d-1) MS4 s2E + g s2r + (d/d-1) r f s2GD + r f g fD
G*SD (g-1) (f-1) MS5 s2E + (f / f-1) r d s2GF
G*PD (g-1) (d-1) MS6 s2E + (d/d-1) r f s2GD
SD*PD (f-1) (d-1) MS7 s2E + g s2r + (f / f-1) (d/d-1) r s2GDF + r g fFD
G*SD*PD (g-1)(f-1)(d-1) MS8 s2E + (f / f-1) (d/d-1) r s2GDF
Error (b-1)(g-1) d f MS9 s2E 

Variables analyzed were: Grain yield (kg/ experimental unit) (YIELD); Expansion volume (cc/gr) (EXVOL); Grain roundness index (GRI); Harvest index (HI); % of cob (COB); Kernel density not expanded (gr/cc) (KNED); Kernel density expanded (gr/cc) (KED).

Higher expansion volumes usually obtained from samples with medium to small kernels, rounder than the average was reported in classic papers, and indicates the roundness index RI (relation between thickness, width and length of the seed) is an adequate parameter to measure this phenotypic correlation.

(RI) = KTH / (KW + KL)

Harvest index was calculated as no. of ears/no. of plants by experimental unit. % of cob (COB), by the rate weight of the kernels/weight of the ears

Table 2 shows significant interaction for (G*PD*SD) and (G*SD) for the variables (YIELD) and (KED), therefore it is necessary to make the analysis within sowing date, for both variables. For the other variables, interactions were not significant and therefore the mean analysis was done for the factors genotype, sowing date and plant density. (EXVOL), (RI) and (COB) were affected by the three factors and (HI) by sowing date and genotype, while (KNE) was only affected by sowing date. Based upon this results, for this no traditional region under study the correct management of these three factors is priority for ensure the popcorn quality measure as expansion volume. For yield stability of the genotypes is a very important factor.

Table 3 shows the values of genetic and enviromental correlations between the seven variables under study. These will be useful for selection and development of specific genotypes within this region. Similar negative genetic and environmental correlations (rG = -0.1876**; rE = -0.1641**), with (YIELD) and (EXVOL) agree with many papers reporting this negative association. (YIELD) has positive correlations with (HI), rG = 0.307**; rE = 0.241**, in agreement with classic maize bibliography, and is relevant to indirect selection. Positive genetic correlation also was registered with (KNED), rG = 0.415**. Negative correlations between (YIELD) and (COB) are obvious (rG = - 0.452**; rE = -0.164**) Genetic correlation (which influences kernel length) is higher than environmental, affected by deficient grain filling Strong association exists between (COB) and (RI), rG = 0.928** and rE = 0.169**, being the genetic component more important than the environmental because the filling of grain influences kernel shape. (EXVOL), a measure of popcorn quality, has strong genetic correlation also with kernel shape (RI), rG = 0.613** and with (KNED), rG = 0.691**, without environmental influence. Opposite relationships were verified between (EXVOL) y (KNED), rG = -0.981** ; rE = -0.921**, because both variables (DECA Y DCEX) were inverse. The same criteria is applicable for the correlations between (KNED) and (RI) with (KED) rG = -0.712** and rG = -0.592**, where even the environmental correlation between (KNED) and (KED) is positive rE = 0.187**.

Prolificacy influences kernel shape, based on the negative correlation between (HI) and (RI)), rG = -0.246**. COB proportion influences (KNED) with a negative genetic correlation rG = -0.226**. Kernel shape is modified by (COB) and affects the kernel density measurement procedure, where smaller grains have more density than the bigger ones; there is a high negative genetic correlation between (KNED) and (RI) rG = -0.519**.

These results indicate that kernel shape is an important trait associated with quality. Genetic correlations are higher than environmental, therefore this would be a relevant trait for a popcorn expansion volume breeding program. Breeding strategy for simultaneous quality (popping) and yield improvement could be grounded in selection for the component (HI), strongly associated with yield but without negatively influencing popping expansion.

Table 2. Mean squares from the combining ANOVA for all the variables.

(**p<0.01, *p<0.05).
 
S. V. D. F. YIELD EXVOL R I H I COB KNED KED
(r/PD)/SD 12 0.975 4.43 7.4x10-4 0.096 3.52 8x10-4 1.6x10-5
G 13 **
0.782		 **
16.02		 **
1.7x10-3		 **
0.064		 **
39.87		 1.3x10-3 **
5.8x10-5
SD 1 0.699 **
49.16		 *
6.1x10-3		 **
0.93		 **
67.20		 **
8.3x10-3		 **
2.1x10-4
PD 2 **
59.18		 *
19.12		 **
9.5x10-3		 0.112 *
22.55		 2.2x10-3 5.7x10-5
G*SD 13 **
0.295		 8.18 4.4x10-4 0.013 4.02 6x10-4 *
3.3x10-5
G*PD 26 0.158 5.96 2.4x10-4 0.017 4.87 9x10-4 1.9x10-5
SD*PD 2 1.665 5.35 7.6x10-4 0.22 12.83 1.1x10-3 1.9x10-5
G*SD*PD 26 **
0.277		 6.97 5.5x10-4 0.024 3.45 7x10-4 *
2.5x10-5
Error 156 0.128 4.57 3.2x10-4 0.012 6.00 9x10-4 1.5x10-5
V.C. ( %) 9.98 9.09 6.13 9.79 12.48 3.89 8.99

Table 3. Genetic and environmental correlations (**p<0.01, *p<0.05).
 
Genetic correlations
- YIELD EXVOL R I H I COB KNED KED
YIELD - **
-0.1876		 **
-0.359		 **
0.307		 **
-0.452		 **
0.415		 *
-0.128
EXVOL **
-0,1641		 - **
0.613
0.048		 0.095 **
0.691		 **
-0.981
RI -0.0102 0.045 - **
-0.246		 **
0.928		 **
-0.519		 **
-0.592
HI **
0.241		 -0.104 0.064 - **
-0.411		 -0.118 -0.082
COB **
-0.164		 -0.055 **
0.169		 0.078 - **
-0.226		 -0.111
KNED 0.104 -0.111 0.105 0.007 -0.028 - **
-0.712
KED **
0.152		 **
-0.921		 -0.051 0.102 0.054 **
0.187		 -
Environmental correlations

 
 
 

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

Return to the MNL 75 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage