Genetic variation, heritability and genetic gain in seed purpose watermelon
Various modern and traditional methods are used to create useful variations in plants for the crop improvement23,24,25,26,27,28,29,30. Genetic variability is very important for the crop improvement to incorporate new traits in existing cultivars12. The main purpose of this study was to evaluate genetic diversity among seed purpose watermelon genotypes collected from Thar desert (located North West of India and South West of Pakistan) for their various qualitative and quantitative traits. The success of breeder in selecting suitable quality and yield parameters lies largely on existence and exploitation of genetic variability to the fullest extent. In the present study, significant differences among the genotypes under evaluation were observed via the analysis of variance. The traits showed significant variation among accessions, with the coefficient of variation of various parameters had a range value, indicating that the trial and data recording was carried out with sufficient precision. The coefficients of variations measures magnitude of variability present in the population. In present research, high values of coefficient of variability were recorded, thereby suggesting better improvement scope for these traits by selection of parents with desirable attributes for breeding program. The frequency distribution of genotypes for yield and related attributes in different groups indicated the presence of high level of genotypic diversity. Further, higher range and more differences among the genotypes and traits means indicated the significant diversity present among the genotypes and traits. The current study’s findings support the previous study by Solmaz and Sari (2009)31, Szamosi et al. (2009)32, and Ngwepe et al. (2021)33, who reported considerable variations in fruit and seed related traits in sweet desserts and citron watermelon, respectively. Singh et al. (2018)34 reported similar findings for mean, range and variation for 88 watermelon genotypes of the 17 morphological traits pertaining to seed, plant and fruit characteristics. Elbekkay et al. (2021)35 revealed considerable variations (p 33.
Success of any crop improvement programme lies in exploiting genetic variability and partitioning of total genetic variability into genetic and non-genetic components15. High PCV and GCV for SY, FY and FPP indicated genetic variation for these traits and scope for genetic improvement through phenotypic selection. Earlier, Ngwepe et al. (2021)36 have also reported higher PCV for flower, fruit and seed traits and higher GCV for the flower and fruit traits in citron watermelon. The previous study of Mahla and Choudhary (2013)37 for the fruits and seed related traits in seed purpose watermelon accessions for GCV and PCV value sync with the present findings. High heritability for the SY, FY, HSW, DFI, FD, and FPP showed that these traits were governed by additive genes. Ngwepe et al. (2021)36 in citron watermelon also reported higher heritability for the plant height and most of the fruits and seeds traits. Likewise, for SY, FY, FPP, HSW, DFI, and FD, the highest genetic advance as a percentage of mean was noted, suggesting the existence of additive gene effects. The heritability and genetic advance values observed in the present study corresponds with the report of Mahla and Choudhary (2013)37 in seed purpose watermelon for the fruits and seed related traits. To achieve selection gains, genetic advance that is strongly associated with yield gains is helpful. High heritability and high genetic advance in present study corresponds with high heritability and genetic advance were recorded for fruit length, seed length, hundred- seed weight, fruit diameter and fruit weight by Ngwepe et al. (2021)36 in citron watermelon landrace accessions. This indicates that these traits have a high level of heritable genes for efficient selection38.
Clustering, principal component and trait association
Significant variations among 138 seed purpose watermelon genotypes for traits linked to fruit, seed yield and phenological traits were distributed into nine distinct clusters. Presence of 124 genotypes in two clusters (clusters V and VIII) delineated most of the accessions originated from the common ancestry as previously hypothesised by Soghani et al. (2018)39 in watermelon. The high genetic similarity in most of the genotypes in the present study could be due to artificial selection for high seed volume by breeders over the years as postulated by Kwon et al. (2010)40 to the improvement through the sequential selection of the targeted traits. The ability to identify potential parents from distinct clusters for a hybridization programme that produces desired recombinants in segregating generations or inbred lines for the heterosis breeding is made possible by the genotype clustering. For this reason, cluster analysis based on quantitative morphological data is easy, cost-effective and an effective tool for assessing genetic variation among genotypes15. Furthermore, genotypes developed from introduced exotic materials and local collections from farmers’ fields, along with genotypes from different sources and origins, such as North Gujarat (Banaskantha, Patan, and Mehsana districts) and the Thar Desert of Rajasthan (Bikaner, Nagaur, Churu, Jodhpur, and Jaisalmer districts), were not clustered based on traits and geographical origins and instead displayed a mixed ancestry. Mixed ancestry might support the presence of outcrossing and inbreeding kind of genotypes due to presence of the monoecious and andromonoecious nature of sex expression pattern, respectively in wild watermelon41. Elbekkay et al. (2021)35 in Tunisian watermelon cultivars; Ngwepe et al. (2021)33 in citron watermelon and Mahla and Choudhary (2013)37 in seed purpose watermelon reported similar clustering pattern by hierarchical cluster method for the quantitative traits based on the most of the fruits and seed related traits. Similar to our study, genetic variability for morphological and qualitative traits of 26 Coccinia genotypes reported eight clusters44. On the contrary, newly developed red fleshed guava accessions had only three clusters which could be due to less variation in the germplasm45. Therefore, hybridization programs in seed purpose watermelon should be directed towards the selection of diverse parents based on both the genetic distance as well as choice based on the geographical origin would be more promising. Hence, the genotypes from the distinct cluster should be used for varietal improvement and parent selection in future breeding programs.
The study by Elbekkay et al. (2021)35 based on 15 morphological traits in commercial and local Tunisian watermelon varieties reported the 67.85% of variance was explained by genotypes. The variance of 69.57% was explained in the first three principal components fully aligned with previous findings. DFI with FD, FPP, FY, HSW, and SY was shown in the PCA biplot, exhibiting a negative correlation. The traits that were positively associated were DM and HSW, FD and SY, SY and FY, FY and FPP, FPP and SY, SY and HSW, FD and FY, FD and FPP, FY and HSW, DM and FD, and FD and HSW. The DFI and DM, FY and DM, FPP and HSW, SY and DM, and FPP and DM vectors imply that there was a little correlation between these attributes. The current correlation was consistent with the morphological quantitative traits study of Assefa et al. (2020)42 in varying watermelon germplasm collections, and the fruits, flower, and seed traits analysis of Ngwepe et al. (2021b)33 in citron watermelon.
An understanding of inter-character correlation is essential to successful selection of useful genotypes from the whole population but intensive selection for any characteristic might result in losses in others. Association analysis revealed most important economic part of seed purpose watermelon i.e. seed yield could be enhanced when fruit yield, fruits per plant, fruit diameter/girth and 100-seed weight increase. Generally short growth period is correlated with reduced yield potential, but in present study seed yield has nonsignificant negative correlation with days to fruit initiation and very weak positive association with maturity duration in seed purpose watermelon. So, productivity could be enhanced in early maturity genotypes in seed purpose watermelon without yield penalty. Our results congruence with Mahla et al. (2014)10 as pointed out that seed yield has nonsignificant positive association with number of fruits per plant, fruits yield per plant and negative association with fruit diameter but significant positive association with 1000-seed weight. The correlation of fruit morphology, citrulline, and arginine levels in diverse collection of watermelon germplasm previously investigated by Assefa et al. (2020)42.
Identification of trait specific genotypes in seed purpose watermelon
Furthermore, the current study highlighted trait-specific superior seed purpose watermelon genotypes that might be deployed to future breeding endeavours. The Thar desert frequently experienced very erratic, scanty, unpredictable and recurrent drought during the rainy season. Cropping pattern of this region mainly monocropping and mixed farming types43. So, breeding the genotypes which are extra early in fruit initiation and maturity could be useful genetic resource to transform mono-cropped into multi-cropped ones, enhance the productivity and profitability of the resource poor farmers of this region. The extra early genotypes such as RMK2313, RMK2345 and RMK2353, RMK2324 in terms of DFI and DM traits could hold great promise for expanding seed purpose watermelon cultivation in arid regions of India, where crop is exclusively sown as rainfed crop10. The multiple trait specific genotypes such as RMK23123, RMK23127, RMK23130 along with check GK-2 identified for the FPP, FY and SY could be potential donor for utilization in breeding program. Genotypes RMK2348 and RMK2365 found highest seed yielder could be used in crossing program to develop mapping population for quantitative trait loci (QTLs) identification or might be useful for the development of lines for the higher productivity in seed purpose watermelon. Notably, genotype RMK2355 exhibited exceptional in both early days to fruit inhiation, and seed yield could be strategically utilized in hybridization program. Trait-specific genotypes were identified for seed purpose watermelon, which will help increase consumer availability and productivity while also taking advantage of the natural genetic variation present in the available germplasm. In order to use this genetic diversity and accelerate the overall advancement of seed purpose watermelon, the genotypes exhibiting specific features will be deliberately included in hybridization initiatives. The goal of this strategy is to promote the growth of high-yielding cultivars with enhanced agronomic traits, potential for yield, and overall resilience to changing environmental factors. It is possible to increase production, food security, and nutritional security through seed purpose watermelon breeding programmes.
Presence of large number of round fruit genotypes have comparatively more fruit diameter associated with large number of fruits per plant and fruits yield and ultimately to the seed yield. Presence of diverse seed coat color viz., black, dark black, brown, light brown, dark brown, blackish brown and whitish with black margin might be used as visual descriptors to select the suitable lines for the breeding and evaluation purpose. Additionally, currently, restaurants and hotels use the seed kernels, or Magaz, as a flavouring and thickening agent while preparing vegetables, beverages, snacks, bakery, ice creams and sweets3. In dry regions, people typically eat its roasted seeds as a snack10. Therefore, seed coat color has preferred market trait that determine the Magaz quality (generally light brown or cream color preferred trait by stakeholders) as well as roasted snack (preferred color is blackish brown and black by local people) in seed purpose watermelon. Immature, small, green and tender fruits of about 100Â g weight (Loiya) are used extensively as vegetables and to sell in the neighbourhood market3. A key fruit quality characteristic that contributes to consumer appeal is flesh color. Presence of absolute large number of white flesh color genotypes in the present study is beneficial to the local consumers/farmers in the Thar region of India as white flesh color with tender fruits consumed as vegetable through out year. Furthermore, a number of researches highlighted the importance of white-fleshed watermelon varieties for biotic stress tolerance in breeding, and ongoing attempts to collect and conserve them are essential, particularly in the area of origin as well regions where diversity of their wild forms are still present36.