by Rakesh Kumar and Todd C. Wehne
Department of Horticultural Science
North Carolina State University
Raleigh, NC 27695-7609
One of the major objectives of watermelon breeders is to improve yield. Yield in watermelon is a quantitative trait that is controlled by multiple genes. Yield is measured as fruit weight per unit area (for example, Mg/ha).
Yield in watermelon is a composite trait which depends on other secondary traits such as number of plants per hectare, number of branches per plant, number of nodes per branch, number of flowers per node, percentage of flowers that are pistillate (female), number of days to the first female flower, fruit length, fruit diameter, and weight per fruit. So, it becomes necessary to determine correlation of these characters with yield. We can practice selection for those characters which have higher correlation with yield.
Yield is a quantitative trait, meaning that it cannot have a discrete value, as in the case of red vs. white flowers. Quantitative traits must be measured, rather than classified, and often have many genes controlling them, and a large amount of environmental variation. Loci controlling quantitative traits are called quantitative trait loci (QTL). Gusmini and Wehner in 2005, screened a diverse set of watermelon cultivars for fruit yield. They observed wide variation for yield in watermelon. These cultivars can be used to develop high yielding populations and measure the narrow-sense heritability.
Successful breeding for higher yield depends upon its heritability. Little information is available on the inheritance of yield in watermelon. Most of the research has produced estimates of broad-sense heritability (genetic variance as a fraction of total variance) since those are easier to do. Additive genetic variance (the variance of breeding value) is the portion of total variance that is responsible for genetic advance after recombination.
- Narrow-sense heritability of a trait is h2N = VA/VP
- VA is additive genetic variance (the variance of the breeding value
- VP is phenotypic variance
Recombination occurs when the genes of two or more parents segregate from a cross. High heritability (narrow-sense) indicates that progress can be made using smaller plot sizes tested in fewer replications and environments.
One common method for estimating narrow-sense heritability is parent(x) offspring(y) regression. The slope of the parent-offspring regression line is equal to Cov(x,y)/Var(x).
Kumar and Wehner (unpublished) reported that total fruit weight was highly correlated with marketable fruit weight (0.86) and marketable fruit number (0.74). Correlation of weight per fruit with total fruit weight and marketable fruit weight was low. Total weight had no correlation with percent culls. They also recorded narrow-sense heritability for different yield traits. Marketable fruit weight and weight per fruit had moderate heritability (0.35). This clearly indicates that good genetic gain in yield can be achieved by improving these traits. Percent culls had heritability near zero. Gusmini and Wehner (2007) reported low to intermediate narrow- and broad-sense heritability (mean, 0.59 and 0.41, respectively) for fruit weight. Narrow-sense heritability was greater than broad-sense heritability. They emphasized progeny testing for optimizing genetic gain due to low heritability of the trait.
Further study can be carried out to locate the molecular markers associated with quantitative trait loci (QTL) that controls yield in watermelon. Once association of molecular markers with QTL has been established, we can advance only those lines which have positive alleles for higher yield. Hashizume et al. (2003) detected five QTLs for four agronomical traits in watermelon using RAPD markers. However, DNA polymorphism is rather poor in cultivated watermelon. But, they found polymorphism in 37% of amplified DNA between cultivars and wild accessions originating in Africa.
- Gusmini, G. and T.C. Wehner. 2005. Foundations of yield improvement in watermelon. Crop Sci. 45: 141-146.
- Gusmini, G. and T.C. Wehner. 2008. Heritability and genetic variance estimates for fruit weight in watermelon. HortScience 42: 1332-1336.
- Hashizume, T., I. Shimamoto, and, M. Hirai. 2003. Construction of linkage map and QTL analysis of horticultural traits for watermelon [Citrullus lanatus (Thunb.) Matsum. & Nakai] using RAPD, RFLP, and ISSR markers.