MSc

Twenty watermelon genotypes were evaluated in RBD with three replication at College of Agriculture, Vellayani during 1994-95. The study was conducted to assess the variability in growth, production and quality parameters and to identify the superior types for further improvement programme. Data on nineteen characters were collected and subjected to analysis of variance and co-variance. Genetic parameters like heritability, genetic advance and correlation were estimated. The path analysis was done to assess the cause –effect relationship and discriminant function analysis to formulate selection index.

The analysis of variance revealed significant difference with respect all the characters. High genotypic and phenotypic variances were observed for number of seeds/fruit and percentage of germination. Highest phenotypic and genotypic coefficients of variations were observed for 100 seed weight and weight of fruits/ plant indicating the presence of large amount of genetic variability.

High heritability estimates were recorded for all the characters viz. days to germinate, percentage of germination, days to first male flower production, node at which first male flower was produced, days to first female flower production , node at which first female flower was produced, inter nodal length, length of vine node at which first fruit produced number of branches/plant, number of fruits/plant, weight of fruits /plants, weight of individual fruit, flesh thickness, rind thickness, TSS, number of seeds/fruit, 100 seed weight and days to first harvest. Number of seeds/fruit only recorded high heritability along with high genetic advance indicating additive gene action.

High positive genotypic correlation and significant positive phenotypic correlation were obtained for yield with number of fruits per plant, weight of individual fruit, flesh thickness and number of branches / plant. TSS with number of branches/plant, percentage of germination and number of fruits/plants, earliness with days to first female flower production, node at which first female flower was produced, node at which first fruit was produced, length of vine and hundred seed weight.

Path coefficient analysis revealed that number of fruits/plant had the maximum direct effect an yield followed by weight of individual fruits. Length of vine recorded negative direct effect on yield.

Based on the result of variability, correlation and path analysis it is suggested that yield, TSS and earliness can not be improved simultaneously. Yield and TSS can be improved by selecting the characters viz. number of fruits/plant, weight of individual fruit number of branches /plant and percentage of germination.

A selection index was constructed and by practicing twenty percent selection, genotypes viz. Sugar Baby, Asahi Yamato, HW 1 and Fuken were identified as the superior ones for the southern tracts of Kerala.