Analysis of Heterosis and Stability in Brinjal at Multi-locations of Gujarat for Yield and its Attributing Traits

Vegetables, such as brinjal, are a natural source of protective food due to their roughage, vitamins and minerals. Brinjal is self-pollinating and originated in India, allowing breeders to use heterosis breeding to improve yield and economic features. The extensive use of heterosis in vegetable crops has led to the creation of numerous high-yielding hybrid cultivars. Newly developed hybrids should be more economical than standard hybrids and they should be performed similarly across multiple regions to make them popular; or, specific hybrids should perform better than average in specific regions to make them recommended for that region. This information can be obtained through analysis of genotype × environment interactions. Eberhart and Russel (1966) provided a clear-cut idea to test materials and select stable genotypes over environments. These details can serve as a reference for best practices to increase germplasm potential, ultimately benefiting heterosis in brinjal hybridization.

The materials were comprised of 8 lines [‘Arka Komal’, ‘Ph-6’, ‘Ph-9’, ‘PPL’, ‘JDNB-16-1’, ‘ISD-006’, ‘CO₂’ and ‘ABSR-2’], 6 testers [‘GOB-1’, ‘GAOB-2’, ‘GAB-6’, ‘GRB-5’, ‘Arka Harshita’ and ‘P. Anupam’], 48 F1s and standard checks [‘GJBH-4’ (standard check 1) and ‘GABH-3’ (Standard check 2)]. The hybrids were prepared by adopting line × tester mating design during kharif 2021. The evaluation programme was carried out using RBD with three replications during late rabi 2021-22 at three different locations, viz. (i) Horticulture Instructional Farm, Sardarkrushinagar Dantiwada Agricultural University, Sardarkrushinagar, District: Banaskantha (E₁) (ii) Seed Spices Research Station, S.D.A.U., Jagudan, District:  Mehsana (E₂) (iii) Maize Research Station, S.D.A.U., Bhiloda, District: Aravalli (E₃). Five sample plants from each genotype in each replication were marked for recording 14 characters (days to opening of first flower, days to first picking, fruit pedicel length, fruit length, fruit girth, fruit shape index, fruit weight, number of fruits per plant, number of primary branches per plant, plant height, fruit yield per plant, plant spread, total soluble solids and chlorophyll content index) and average data were used for applying statistical procedures. Standard statistical methods were applied viz., pooled analysis of variance (ANOVA) estimation (Comstock and Robinson, 1952); percent heterosis estimation (Fonseca and Patterson,1968) and stability assessment (Eberhart and Russel, 1966).

The ANOVA of individual and pooled environments showed noticeable genetic differences in the material for the majority of the traits (excluding total soluble solids for genotypes at Jagudan). Behaviour of genotypes was different in different locations for the majority of the traits (excluding days to first picking and total soluble solids). Parents vs. hybrids interaction proved the existence of overall heterosis at three locations for the majority of the characters. Also, considerable variations among the environments were observed for the majority of the traits (excluding total soluble solids).
       
Positive heterotic effects were desirable for all the traits (excluding days to first flower and first picking). For a particular trait, some hybrids showed higher heterotic effects while others showed lower, which was due to the variation in the genetic nature of the parents. Out of 48 F₁ hybrids studied, top hybrids based on a pooled analysis of 14 traits for desired standard heterosis along with heterobeltiosis and mean value are displayed in Table 1. Also, 48 hybrids’ heterobeltiosis and economic heterosis for fruit yield per plant over the pooled environments are shown in Fig 1. For days to opening of first flower and days to first picking, ‘ISD-006 × GRB-5’ was found desirable in terms of heterobeltiosis. For fruit pedicel length ‘Arka Komal × GOB-1’, ‘ISD-006 × Arka Harshita’ and ‘PPL × GAOB-2’ were the hybrids of choice. ‘PPL × Arka Harshita’ and ‘Arka Komal × P. Anupam’ were the top hybrids for fruit length and fruit shape index. ‘Ph-9 × GOB-1’, ‘ISD-006 × GRB-5’ and ‘Ph-9 ´ GRB-5’ showed significant desirable standard heterosis for fruit girth. Being an important trait, fruit weight showed ‘ISD-006 × GRB-5’, ‘ISD-006 × GAOB-2’ and ‘ISD-006 × GOB-1’ as the top desired hybrids. There were also sufficient desirable hybrids obtained for a number of fruits per plant, but the same wasn’t noticed for a number of primary branches per plant. ‘JDNB-16-1 × GOB-1’ was the top significant hybrid for plant spread and plant height. Total soluble solids and chlorophyll content index also showed desirable hybrids in terms of standard heterosis and heterobeltiosis.
 

 

       
The yield of fruits per plant is the most important plant character. The hybrid ‘ISD-006 × GRB-5’ was the hybrid of choice for this trait. It showed high standard heterosis over ‘GJBH-4’, significant heterobeltiosis in the desired direction and high mean fruit yield. In case of individual environments for fruit yield per plant; ‘ISD-006 × GOB-1’, ‘CO₂ × GOB-1’, ‘Arka Komal × GRB-5’, ‘PPL × P. Anupam’ and ‘ISD-006 × GRB-5’ at Sardarkrushinagar; ‘ABSR-2 × GAB-6’, ‘Ph-9 × Arka Harshita’, ‘ABSR-2 × Arka Harshita’, ‘ISD-006 × GAB-6’ and ‘ISD-006 × GRB-5’ at Jagudan and ‘ISD-006 × GRB-5’ at Bhiloda showed desirable standard heterosis.
       
These hybrids showed desirable heterosis for at least one yield component. Almost the same trend was noticed by Saikia et al., (2019) for fruit pedicel length; by Makani et al., (2013) regarding longer fruit and fruit girth; Chaudhari et al., (2020) regarding number of fruits per plant and fruit yield per plant; Rani et al., (2018) regarding plant height and Singh and Chaudhary (2018) regarding TSS and chlorophyll content index.  
       
Stability ANOVA (Table 2) showed that the genotypes were highly significant and majority of the traits were highly affected by the interaction of genotypes × environments (G × E). But genotypes showed consistent behaviour for days to first picking and total soluble solids over the locations, which prevented their further stability analysis. Singh and Chaudhary (2018) and Bhushan and Samnotra (2017) found the same trends in brinjal. Both unpredictable and predictable components affected the stability of the majority traits, which were noticeable by the presence of significant interaction of pooled deviation (non-linear component) with pooled error. Prasad et al., (2002) and Sivakumar et al., (2017) derived the same type of conclusion.
 

       
Regression coefficient (b_i), mean performance and squared deviation from linear regression (S²d_i) are the stability parameters which were calculated for 12 characters for hybrids and parents to analyse the stability over the locations. Table 3 is showing the analysed results with the top three parents and hybrids for each character in each condition of stability.
 

       
The stable genotypes should be used for various breeding activities (Rai et al., 2000). The criteria of Mehra and Ramanujam (1978) were used for the stability assessment. The ‘b_i’ was considered as a level of response of a genotype and ‘S²d_i’ was treated as a level of stability. The non-significant regression coefficient was treated as a unity. Non-significant S²d_i was the sign of “minimum deviation” i.e., zero. Hence, the genotypes which possess the above values with desirable higher mean were considered stable. A desirable higher mean with significantly higher bi than unity and non-significant S²d_i was considered as less than average stable (adaptable to favourable environments but sensitive to environmental changes). Also, a condition of non-significant S²d_i with desirable higher mean and significantly lower b_i than unity was finalized as more than average stable (adaptable to poor environments). Significant S²d_i was treated as an unpredictable behaviour of genotypes.
       
Fig 2 showing overall distribution of parents and hybrids according to stability parameters. For fruit yield per plant, ‘ISD-006 × GAB-6’ (fruit yield: 3320.55 g), ‘JDNB-16-1 × GOB-1’ (fruit yield: 3152.01 g) and ‘ABSR-2 × GAB-6’ (fruit yield: 3090.76 g) were the top three stable crosses; ‘Ph-9 × Arka Harshita’ and ‘Ph-6 × GAB-6’ were considered as a desirable for unfavourable environments, while ‘ISD-006 × GRB-5’ (highest fruit yield: 3733.38 g), ‘JDNB-16-1 × P. Anupam’ and ‘ABSR-2 × GOB-1’ were appropriate for favourable environments. Among these, the hybrid ‘ISD-006 × GRB-5’ and its parents have shown in Fig 3 due to their superior performance in the experiment.
 

 

       
Stable hybrids could be directly used for yield improvement. The stable component traits always result in a stable fruit yield per plant. The challenge of finding a hybrid with stability for all the traits, showing the scope for the inclusion of more environments for future analysis. Siva et al., (2020) reached a similar type of result.

After the precise analysis of heterosis and stability, we conclude that the hybrid ‘ISD-006 × GRB-5’ and its parents (Fig 3) may be directly used in future breeding program. This hybrid showed the highest desirable mean value and noticeable economic heterosis in individuals and pooled over the environments, for fruit yield per plant and the majority of its attributing traits (fruit girth, fruit weight, total soluble solids, chlorophyll content index). It was also desirable in terms of early flowering and early picking. It was average stable for days to opening of first flower, fruit girth, plant height and plant spread; suitable for favourable environment for fruit weight and fruit yield per plant. Also, the desirable hybrids found in this study can create transgressive segregants in brinjal and should be analysed further for getting more knowledge on their genetic structures and gene actions.

We don’t have any conflict of interest regarding this experiment.