Inheritance of Yield and its Related Traits in Bitter Gourd (Momordica charantia L.)

A study on understanding the inheritance of yield and its related traits in bitter gourd by using generation means analysis involving six generations in two crosses (IC-044417×IC-045339 and IC-045339×IC-470560) revealed the presence of non-allelic interaction effects in expression of all the traits studied. All the traits were inherited by dominance (h) among the main effects and additive×additive (i) among the interaction effects in IC-044417×IC-045339 while dominance, additive×additive (i) and additive×dominance (j) in IC-045339×IC-470560 except fruit girth which was inherited by dominance (h) and dominance×dominance (l) type of interaction effect. Therefore, recurrent selection and diallel selective mating system in addition to heterosis breeding may be used to improve the traits in bitter gourd.


Introduction
Bitter gourd (Momordica charantia L.) is extensively cultivated for its tender fruits owing to its nutritive and medicinal values.In India, a wide range of variability is available in bitter gourd for fruit characters in terms of size, shape, flesh thickness, colour etc. Yield is a complex character influenced by various component characters inherited polygenically and highly subjected to environmental variations.The exploitation of genetic variability of these traits through hybridization and selection is the primary focus of most crop improvement programmes.A good knowledge of genetic systems inheriting the expression of these characters facilitates the choice of the most efficient breeding and selection procedures.In addition to additive and dominance variation, it has been suggested that epistasis may also be involved in the inheritance of many quantitative characters in bitter gourd (Dey et al., 2010).But the information available on non-allelic interactions for quantitative traits in bitter gourd is very limited.Therefore, the present study was undertaken using generation mean analysis to partition the genetic variation into additive, dominance and epistasis which helps in formulating an effective and sound breeding programme.

Results and Discussion
The significance of A, B, C and D scales for different characters of two crosses were presented in Table 1.The A and B scaling tests provide the evidence for the presence of all types of interactions viz., additive× additive [i], additive×dominance [j] and dominance× dominance [l]. the C scaling test provide information about dominance×dominance type of interaction effect whereas D scaling test gave information about presence of additive×additive type of non-allelic interaction effect.Based on the significance of these scaling tests, six parameter model was employed according to Jinks and Jones (1958) to estimate six components (m, d, h, I, j and l) of genetic variation.The results (Table 1) revealed the presence of non-allelic interaction effects in expression of all the traits studied in both the crosses.These results are in comparable with that of Celine and Sirohi (1998) in bitter gourd.
The genetic parameters viz., m, d, h, i, j and l of two crosses were presented in Table 2. Significant positive dominance effect was exhibited by cross IC-044417× IC-045339 while additive and dominance effects were significant and positive (but dominance was in higher For number of laterals, both crosses exhibited significant and positive dominance (h) effect and additive×additive (i) interaction effect indicating the role of both dominance and additive gene effects in inheritance of this trait.
Earliness in terms of days to 1 st female and node number at 1 st female flower appeared in bitter gourd, influenced by dominance (h) among main effects and additive × additive (i) and additive×dominance (j) among interaction effects.These results showed that importance of heterosis breeding and simple selection methods in improvement of earliness in bitter gourd.These findings are in conformity with the results of Devaraju et al (2010) in cauliflower.However, in contrary to this Singh and Ram (2003) reported additive component was more than that of dominance effect for days to first female flower anthesis and days to first fruit set in bitter gourd.
The cross IC-045339×IC-470560 expressed significant and positive effect of dominance (h) and additive× additive and additive×dominance interaction gene effects while IC-044417×IC-045339 showed significant dominance and additive×additive type of interaction gene effect only for number of fruits/vine.These results were similar to findings of Mandal et al (2002) in ash gourd.
For average fruit weight significantly positive dominance (h) effect was notice in both the crosses.Among the interaction effects the cross IC-045339× IC-470560 recorded significant additive×additive (i) and additive×dominance (j) effect while cross IC-044417×IC-045339 recorded significantly positive additive×additive (i) interaction effect only.These results revealed that both number of fruits and fruit weight were inherited by both additive and dominance gene effects in bitter gourd.
The two crosses expressed significant and positive dominance (h) and additive×additive type of non-allelic interaction in inheritance of fruit length.Note: * Significant at 0.05% probability; ** Significant at 0.01% probability For fruit girth, the cross IC-044417×IC-045339 exhibited significant d, h and i in desired direction while cross IC-045339×IC-470560 recorded significant and positive l gene effect only.These results indicated the importance of heterosis breeding for improvement of this trait.Similar results reported by Tewari et al (1998).
The genetic parameters h and i were found significantly positive in cross IC-044417×IC-045339 while h, i and j were significant in desired direction in IC-045339×IC-470560 for yield/vine indicating the involvement of dominance and all epistatic interaction effects in controlling this trait in bitter gourd.Patel et al (2005) also reported additive and additive× additive type of epistasis in inheritance of fruit yield in bitter gourd.
The components of dominance (h) and dominance× dominance (l) having opposite sign implied the involvement of duplicate epistasis in the inheritance of all the traits studied in both the crosses.The presence of duplicate epistasis would be detrimental for rapid progress, making it difficult to fix genotype with increase level of character manifestation because the positive effect of one parameter would be cancelled out by the negative effect of another (Devaraju et al., 2010).Hence, early generation intermating besides accumulating the favourable genes and maintaining heterozygosity in the population are likely to though out desirable recombinants.Ram et al (1997) reported complimentary epistasis for total yield per plant.
From the above results it can be concluded that heterosis breeding may be used where large magnitude of non-fixable gene effects is observed.Considering the importance of dominance as well as epistasis observed in the present study, recurrent selection and diallel selective mating system may be used to exploit both type of gene effects in bitter gourd.

Material and Methods
Two F 1 hybrids involving three diverse parents viz., IC-044417, IC-045339 and IC-470560 were produced during summer 2009.These F 1 were selfed to get respective F 2 generation and simultaneously crossed with their first and second parents (P 1 and P 2 ) to obtain BC 1 and BC 2 generations respectively during summer 2010.All six generations viz., parent 1(P 1 ), parent 2(P 2 ), F 1 , F 2 , BC 1 and BC 2 were evaluated in a randomized block design with three replications at Model Orchard, College of Horticulture, Rajendranagar, Hyderabad during summer 2011.Seeds were sown by adopting a spacing of 2.0 m between rows and 0.5 m between plants and good crop was raised by following the recommended package of practices.Data were recorded on nine quantitative characters viz., vine length (m), number of laterals/vine, days to 1 st female flower appeared, node number at 1 st female flower appeared, number of fruits/vine, average fruit weight (g), fruit length (cm), fruit girth (cm) and yield/vine (kg) from five randomly selected plants in all generations except F 2 and Back crosses (BC 1 & BC 2 ) where 20 and 15 pants respectively selected.The means of each of six generations i.e., P 1 , P 2 , F 1 , F 2 , BC 1 and BC 2 for each cross , averaged over the replications were taken as generation means which were used for calculation of various genetic parameters.The estimates of six genetic parameters viz., m (mean), d (additive), h (dominance), i (additive×additive), j (additive×dominance) and l (dominance×dominance) were calculated using the formula given by Hayman (1958) and Jinks and Jones (1958).Scaling test was performed according to Mather (1949) to know the presence of non-allelic interaction effects governing the traits in their expression.

Table 1
Test of significance of A, B, C, and D scales for different characters in bitter gourd * Significant at 0.05% probability; ** Significant at 0.01% probability magnitude) in cross IC-045339×IC-470560 for vine length.These results indicating the importance of dominance gene effect in expression of this character indicating the possibility of heterosis breeding for the improvement of this parameter in bitter gourd.The interaction effects revealed significantly positive i component in IC-044417×IC-045339 and both i and l in IC-045339×IC-470560.

Table 2
Estimates of components of generation means for different characters in bitter gourd