Introduction:
Pumpkin (Cucurbita moschata Duch. ex Poir.) is locally known as ‘Misti kumra’ or ‘Mistilau’ or ‘Misti kudu’, is an important common vegetable in Bangladesh. Pumpkin originated in equatorial and sub-equatorial America (Whitaker and Davis, 1999). It starts from Southern part of USA and continues up to Peru of South America. It grows throughout the entire tropical and sub-tropical regions of the world and milder areas of the temperate zones of hemispheres. It is widely cultivated in India, China, Malaysia, Taiwan, and Bangladesh. It is distributed widely in Southeast Asia, tropical Africa, tropical South and Central America (Peru and Mexico), the Caribbean and most part of tropics.

Pumpkin belongs to the family Cucurbitaceae. There are 27 species under the genus Cucurbita, five of which are in cultivation. These are C. moschata, C. maxima, C. ficifolia, C. pepo and C. mixta, commonly known as pumpkin. Pumpkin is highly cross-pollinated crop having chromosome number 2n=40. C. moschata is probably the most widely grown species of Cucurbita and this species is cross compatible with C. maxima, C. pepo and C. mixta. It is insect- pollinated and 1000 m isolation distance is necessary to maintain purity of cultivars plants and vine crop. It is an annual crop having a climbing or trailing habit (Katyal and Chadha, 2000).

Pumpkin is relatively high in energy and carbohydrates and a good source of vitamins, especially high carotenoid pigments and minerals (Bose and Som, 1998). The nutrient per 100 g edible portions of fruit is cited in appendix 1. Night-blindness is a serious problem in Bangladesh which happens due to vitamin A deficiency. It may certainly contribute to improve nutritional status of the people, particularly the vulnerable groups in respect of vitamin A requirement. Encouraging the mass people to take more pumpkin can easily be solved the problem. As a matter of fact, there is a program from Health Department to encourage feeding mature pumpkin to the children.

The delicate shoots and leaves are used as delicious vegetables. The fleshy large fruits can be
consumed at mature and immature stages. It is one of the main vegetables in a wedding party
or on other occasional party in northern India (Chauhan, 1995). The sweet pie and pumpkin
haluwa are the delicious items prepared from matured fruit (Shanmungavelu, 1998). The seed are very nutritious (it contains 40-50% oil and 30% protein) and eaten as food in many countries of the world (Tindall, 1998).

The pumpkin has good medicinal value. It is used against many diseases like gonorrhea, urinary problem. The paste of the dried fruit-stalk, which is in immediate contact with the ripe gourd, is used as the remedy for the bites of venomous insects of all kinds, especially for the centipedes (Chauhan, 1995).

It is a very common vegetable in Bangladesh and particularly popular among the rural people. It is grown round the year in our country. It becomes available even in the lean period when other vegetables are scarce in Bangladesh. It has the longest storability among all the cucurbits. The well-matured fruits (ripe fruits) can be stored for 2 to 4 months (Yawalkar, 1991). Due to its good taste and keeping quality, nutritional status, easier cooking quality, reasonable market price and year-round availability, its demand is increasing day by day in the country.

Vegetable production rate in Bangladesh is very low; yearly only 8.542 million M tones (BBS, 2010). Vegetables consumption rate is 104 g per day per adult, against the optimum amount of about 300 g per day per adult (Rashid, 2006). The total area under cultivation of pumpkin is 27,935 ha with a total production of 2, 63, 000 MT having national average of 9.42 ton/ha in a year of this country (BBS, 2012).

The productivity of local genotypes ranged from 6.93 t/ha to 19.07t/ha (Hamid et al., 1991). On the other hand, there are many exotic genotypes, which have short life cycle but high yield potential. Some of these exotic genotypes bear deep green long fruits, which are attractive. Flower buds of these genotypes appear 20 to 25 days earlier than the local genotypes. The exotic genotypes do not need big trellis because of their medium climbing habit. However, the exotic types are more susceptible to different virus diseases than the local genotypes. These variabilities among the indigenous and exotic genotypes are genetic attributes, which can be combined through hybridization to develop short vine type varieties with high yield, with smaller fruit type with high carotene content virus resistance and high number of female flowers. Being a cross-pollinated crop, it seems easy to transfer suitable traits by crossing appropriate genotypes of sweet gourd.

Though it is a very common crop, it may be mentioned that until to date there is no released variety of pumpkin with high yield potential and better nutritional quality. Further, a very limited attempt had been made for the genetic improvement of this crop, particularly with quality traits. An understanding of the nature and magnitude of variability among the genetic stocks is of prime importance to the breeder. Good knowledge of genetic resources might also help in identifying desirable cultivars for commercial cultivation. Because of its high cross-pollination, genetically pure strain is available hardly to the growers. Lack of high yielding, disease and pest tolerant variety is the main constraint towards its production. Among the cultivated landraces, a wide range of genetic variability exists in this crop that can be exploited for its improvement. It is the touchstone to a breeder to develop high-yielding varieties through selection, either from the existing genotypes or from the segregates of a cross. Hence, information on gene action, its nature and magnitude in respect of quality characters aspects is required to be properly assessed for its improvement.

Heterosis breeding is a potential tool to achieve improvement in the quality, quantity, and productivity of pumpkins (Tamilselvi et al., 2015). Heterosis and combining ability is a powerful tool in identifying the best combiner that may be used in crosses either to exploit heterosis or to accumulate fixable gens and obtain desirable segregates. It will help to understand the genetic architecture of various characters that enable the breeder to design effective breeding plan for future up gradation of the existing materials. The information may also be useful to breeders for genetic improvement of the existing genotypes on the basis of performance in various hybrid combinations.

There have been not many studies on heterosis and combining ability of pumpkin in Bangladesh particularly with quality traits except Rana et al., (2015). Though pumpkin as a vegetable is becoming an important ingredient in daily diet, relatively little attention has been paid towards the development of hybrids/varieties which are rich in beta-carotene with high Brix content; high reducing sugar with high yielding capacity. Therefore, considering the above facts the present investigation was carried out to achieve the following objectives:

1. to identify potential parents and productive hybrids of pumpkin
2. to estimate the combining ability effects and variances for quality traits in pumpkin
3. to identify of best cross combination for higher yield and other quality characters
4. to estimate the heterosis against mid and better parents of different characters

Materials and Methods:

Experimental site:

The experimental site is located at the centre of Modhupur tract (24.09 °N latitude and 90.26 °E longitude), which is 8.4 m above the sea level. It is about 40 km North of Dhaka, the site was previously under shal forest and developed later for research purpose.

Climate:

The Experimental site is situated in the sub-tropical climate zone, characterized by heavy rainfall during the month of May to September and scanty rainfall during rest of the year. During crossing of parents, the average temperature, relative humidity and rainfall was 27.26°C (max) and 18.18 °C (min), 85.03% and 3.48 mm per month, respectively (Appendix 3). During studying combining ability of the parents and hybrids, the average temperature, relative humidity and rainfall was 28.05°C (max) and 16.28 °C (min), 78.86 % and 26.31 mm per month, respectively.

Soil:

The soil is terrace soil, which is nearly equivalent to Ochrept sub order of USDA soil taxonomy and belongs to the locally termed Salna series of Shallow Red Brown Terrace soil (Brammer, 1971). The soil is silt loam in texture having acidic (pH=5.5) in nature, poor fertility status, and impeded internal drainage.

Materials:

Five (5) advanced inbreds of pumpkin viz IBD 23, IBD 40, IBD 47, IBD 50 and IBD 57 developed by the GPB, BSMRAU was used in the study for combining ability analysis in a 5x5 diallel population. The inbreds were synthesized in the previous year.

Design and layout:

The experiment was laid out in a randomized complete block design (RCBD) with three replications.

Raising of seedlings:

The seeds were sown in 9cm x 15 cm sized polyethylene bags. Two seeds were sown in each bag. The growth medium was prepared by mixing compost and soil in 50:50 proportions. Intensive care was taken for production of healthy seedlings.

Preparation of land and pits:

The experimental land was prepared by deep and cross ploughing and harrowing followed by laddering. The plots were raised 10 cm above the ground level. Pits of 50 x 50 x 50 cm size were dug at a spacing of 2 x 2 m.

Manure and fertilizer applied in each pit:

Around 10 kg Cow dung, 52 g TSP, 60 g Urea and 40 g MP were applied in each pit. Cow dung and pit soils were mixed together. The fertilizers were applied on the top and worked up to 10 cm soil of the pits.

Transplanting:

Twenty-four days old seedlings were transplanted in well-prepared experimental plot on 17^th December 2013. The seedlings were watered immediately after transplanting. Four plants of each genotype were accommodated in each replicated plot maintaining 2x2 m spacing. 

Intercultural operations:

Intercultural operations were done as necessary during the growing period for proper growth and development of the plants and to protect the fruits from rotting.

Mulching:

Immediately after planting, the field was covered with straw to ensure optimum moisture for easy emergence of buds.

Weeding:

Routine weeding was done to keep the field free from weeds and to pulverize the soil.

Irrigation and drainage:

Irrigation was applied as and when required.

Harvesting:

The fruits were harvested when the peduncle dried on maturity.

Data collection:

Three plants were selected at random from each plot for recording data. Both quantitative and qualitative characters were recorded.

Quantitative characters:

1. Days to first male and female flower: The number of days to first male and female flower was recorded.
2. Days to first male and female flower opening: The number of days to first male and female flower opening was also recorded.
3. Nodes for first male and female flowers: The nodes at the ground level to the nodes of first blooming of male and female flowers were recorded.
4. Number of male and female flower per plant: From the first blooming of male and female flowers were counted.
5. Number of nodes for first fruit setting: The numbers of nodes from the first fruit setting were counted.
6. Fruit yield per plant (Kg): Total numbers of fruits from three randomly selected plants were weighed and their average value was taken.
7. Fruit length: Fruit length was measured using scale.
8. Fruit breadth: Fruit breadth was measured using scale.
1. Hollowness: Fruit hollowness was measured using scale after cutting the whole fruit into two pieces.
10. Flesh Thickness: Flesh thickness was measured using scale.

Qualitative Characters:

1. Dry matter (%): Dry matter percentage was also calculated from matured fruit. For dry matter content, 200gm of matured pumpkin was cut into small pieces and dried in the sun for 3 to 4 days. After that it was again kept in an oven at 60^oc for 72 hours. Then the weight was taken using electric balance.
2. Brix (%): It was measured with the help of a Brix meter (Model: ATAGONI Brix 0-32%, Made in Japan).
3. Carotene (mg/g): Three fruits of each genotype were used for carotene analysis and their average value was taken. At first 10 g flesh of pumpkin was taken and crushed by mortar and pastel. After then 10 ml mixture (Acetone: Hexane=2:3) was added in the paste. Then the solution of pumpkin paste and acetone hexane was filtered in a vial having air tight lid. Then the nascent spectrophotometer reading was recorded at four different nano meter length viz 663 nm, 645 nm, 505 nm and 453 nm. Finally, β-carotene was calculated by the following formula:

B-carotene (mg) = (Reading of 663 nm) +(Reading of 453 nm) -(Reading of 645 nm) - (Reading of 505 nm).

4. Sugar (gm/100gm) Estimation

One fruits of each genotype from each replication were used for reducing sugar analysis and their average value was taken.

Reducing Sugar:

Ten ml of each of Bertrand A (40g of CuSO₄. 5H₂O dissolved in water and diluted to 1 liter) and Bertrand B (200g of sodium-potassium tartarate and 150g of NaOH dissolved in water and diluted to 1 liter) solutions were added to 5ml of sample solution. The conical flask was placed on a hot plate (sand bath) and boiled for about 3 minutes and kept overnight for cooling. The supernatant was decanted and discarded very carefully by keeping the precipitation. The precipitation was washed repeatedly until blue color was present. Then 10ml of Bertrand C [50g of Fe₂ (SO₄)₃ and 115ml of concentrated H₂SO₄ was added and diluted to 1 litre] solution was added to dissolve the precipitation (Cu₂O). Finally, it was titrated with 0.4% KMnO₄ solution. Reducing sugar was calculated comparing tabulated values. Before calculation of reducing sugar factor of 0.4% KMnO₄ was determined.

Total Sugar:

Ten ml of extract solution was taken in a volumetric flask and 2-3 drops of 4N HCl was added. The flask was then boiled for about 3 minutes on a hot plate for hydrolysis. After cooling in tap water, the extract was neutralized with 0.1N NaOH. The rest of the procedure was same as mentioned in reducing sugar.

Non-Reducing Sugar:

Non- reducing sugar was calculated by deducting reducing sugar from total sugar.

Statistical Analysis:

Method 1 model II of Griffing (1956b) was followed for combining ability analysis. The recorded data were analyzed by using Diallel analysis and simulation program by Mark, D. Burow and James G. Coors, copyright 1993 and version 1.1. Vr-Wr graph of different traits was drawn according to Hayman (1954).

Results and discussion:

Analysis of variance (ANOVA) for heterosis and combining ability:

The mean sum of squares from analysis of variance due to heterosis and combining ability for fruit length, fruit breadth, hollowness, flesh thickness, dry matter (%), brix (%), reducing sugar, non-reducing sugar, total sugar and fruit yield  have been shown in Table 1.

Analyses of variance for genotypes were highly significant for fruit length, fruit breadth, hollowness, brix (%) and fruit yield. The analysis of variance for general combining ability variances were found highly significant for the traits fruit length, fruit breadth, hollowness, brix (%), dry matter percentage and fruit yield (Table 1) which indicated that additive gene actions played significant role for the expression of these characters.

General combining ability was found non-significant for the character such as flesh thickness, β carotene, reducing sugar, non-reducing sugar and total sugar. Specific combining ability and reciprocal variances were found significant for fruit length; fruit breadth, hollowness, brix (%) and fruit yield which indicated that non additive gene action was involved in the inheritance of these traits.  The characters viz. flesh thickness, β carotene, reducing sugar, non-reducing sugar and total sugar showed non-significant specific combining ability and reciprocal variation.

The ratio of GCA and SCA variance was high and more than one for the characters fruit length, hollowness, dry matter, β carotene and reducing sugar revealed that the preponderance of additive gene action over the non-additive gene action. This indicated the limited scope of heterosis breeding for these characters and population improvement through recurrent selection should be adopted for exploiting the genetic variations (Kushwaha and Ram, 1996; Pandey et al., 2005, Jha et al., 2009). The gca variances was lower than the sca variance for fruit breadth, flesh thickness, brix (%), non-reducing sugar, total sugar and fruit yield, may be improved through hybridization (heterosis) as indicating the predominance of non-additive gene effects which was opined by Jha et al. (2009).  However, both additive and non-addictive gene actions were observed by Suneal Kumar (1984) and Sirohi et al., (1986).

FL= Fruit length, FB= Fruit breadth, HN= Hollowness, FT=Flesh thickness, DRM= Dry matter (%), BRX=Brix%, BCAR= Beta carotene (mg/100g), RS= Reducing sugar (g/100 gm), NRS=Non reducing sugar (g/100g), TS=Total sugar (g/100g), FY=Fruit yield (kg/plant)

Total Sugar:

The positive and significant GCA effects (Table 4b) was found in IBD50 (0.606**), IBD40 (0.106**) and IBD23 (0.073**) for this trait i.e. the rest three parents were not suitable for improvement of total sugar content in pumpkin.Yang et al. (2006) also found similar results in pumkin.

Fruit yield:

Significant GCA effects were ascertained from three parents for fruit yield (Table 4b) but other two parents had negative value. From GCA effects analysis it implied that among five parents, IBD40 (0.653**), IBD47 (0.103**), and IBD50 (0.262**) might be selected as a good combiner for increasing fruit yield per plant. Nisha and Veeraragavathatham (2014) had similar results in their experiments.

Table 4b: GCA effects for β carotene, reducing sugar, non-reducing sugar, total sugar and fruit yield in 5 X 5 full diallel populations in pumpkin

Specific combining ability (SCA) effect and reciprocal effect

Non-additive gene action is signified by the SCA effects in the expression of the characters of a crop. Specific combining ability indicated the performance of some specific cross combination. That is why it is related to a particular cross. High SCA effects may arise not only in crosses involving high general combiners but also in those involving low combiners. Estimates on SCA effects of the crosses in F₁ generation revealed that there were a good number of crosses having significant positive or negative SCA effects on different important traits of pumpkin. The SCA effect of 10 F₁ for eleven different characters studied are presented in the Table 5a, 6a. Reciprocal effects are presented in the Table 5b, 6b.

Table 5a: Specific combining ability (SCA) effects for fruit length, fruit breadth, hollowness, flesh thickness and dry matter in 5 x 5 full diallel population of pumpkin

FL= Fruit length, FB= Fruit breadth, HN= Hollowness, FT=Flesh thickness, DRM= Dry matter

The SCA values provide important information about the performance of the hybrid relative to its parents. However, Arunga et al. (2010) found that the SCA effect alone has limited value for parental choice in breeding programs. They, therefore, suggested that the SCA effects should be used in combination with other parameters, such as hybrid mean and the GCA of the respective parents such that a hybrid combination with both high mean and favorable SCA estimates and involving at least one of the parents with high GCA, would tend to increase the concentration of favorable alleles; which is desired by any breeder (Tamilselvi et al., 2015).

Fruit length:

Out of 10 F₁s, all hybrids provided non-significant positive or negative SCA effects (Table 5a) on fruit length indicating their unsuitability to increase or decrease fruit length.

Out of 10 F₁s, all cross combination showed reciprocal effect. IBD57 X IBD23 provided the highest negative and significant reciprocal effects (Table 5b) on fruit length followed by IBD40 X IBD23 and IBD47 X IBD23,  having value -3.066 **, -6.150** and -6.550**, respectively indicated their suitability of using maternal effect to increase fruit length (fruit size). On the other hand, the cross IBD50 X IBD40 (5.358) gave the highest positive reciprocal effects followed by IBD57 X IBD47, IBD47 X IBD40 and IBD57 X IBD40  having value (3.500, 3.291and 1.833), respectively which can be used to reduce fruit size to some extent. Monhanty (2000) and Rana et al., (2015) agreed with our findings.

Fruit breadth:

The cross combination IBD50 X IBD57 exhibited the highest significant and positive SCA effects (1.34**) followed by IBD23 X IBD40 (0.552**). Thus, these cross could be considered as good specific combiner for increasing fruit breadth in pumpkin. The negative SCA value (Table 5a) for this parameter was obtained from the crosses viz., IBD23 X IBD 57 and IBD47 X IBD57 could provide decreased fruit breadth in pumpkin.

Among all ten crosses, all crosses had reciprocal effects of which six crosses had significant and positive reciprocal effects (Table 5b) indicated their ineptness (unsuitability) to increase fruit breadth in pumpkin.

Table 5b: Reciprocal effects for fruit length, fruit breadth, hollowness, flesh thickness and dry matter in a 5 x 5 full diallel populations of pumpkin

FL= Fruit length, FB= Fruit breadth, HN= Hollowness, FT=Flesh thickness, DRM= Dry matter

Hollowness:

Among the cross combination, four crosses viz. IBD23 X IBD40 (0.150**), IBD23 X IBD47 (0.093**), IBD23 X IBD57 (0.303**) and IBD50 X IBD57 (0.336**) showed significant positive SCA values (Table 5a). The rest six the cross combinations which had negative values were non-significant for hollowness in pumpkin.

The data about reciprocal effects for hollowness are given in Table 5b. All the crosses showed reciprocal effects. Among the cross, IBD57 X IBD47(0.291**) had the highest positive and significant reciprocal effects on hollowness of pumpkin followed by IBD57 X  IBD 50, IBD50 X IBD40, IBD50 X IBD23 and IBD47 X IBD40 having value  0.233**,0.233**, 0.150**and 0.075**, respectively.  Hence these cross combinations could provide chance to reduce the hollowness in pumpkin which is expected. On the other hand, the hybrid IBD40 X IBD23 (-0.162*) gave the negative significant reciprocal effects. Similar results were also found by Pandey et al., (2010) and Rana et al., (2015).

Flesh thickness:

Positive and significant value is expected in case of flesh thickness (Table 5a) in order to increase flesh thickness and the yield as well. The cross combination IBD23 X IBD40 revealed the highest (0.476**) positive and significant SCA value which could be used as a good specific combiner for flesh thickness in pumpkin.

The cross combination IBD50 X IBD47 (0.476**) had the highest positive and significant reciprocal effects on of flesh thickness of pumpkin followed by IBD57 X  IBD50, IBD47 X IBD23, IDB57 X IBD40, and IBD47 X IBD40 having value  0.128**,0.032*,0.017** and 0.080**, respectively which could be used as the good specific combination to increase the flesh thickness in pumpkin.

Dry matter (%):

The estimates on SCA effects for dry matter are given in (Table 5a).The highest positive and significant effect was found in cross IBD50 X IBD57(1.726**) followed by IBD23 X IBD47 (1.226**),  IBD47 X IDB57 (0.826**) and IBD23 X IBD40(0.560**), respectively which is expected as it indicated increase in dry matter percentage. All other cross combination had non-significant and negative SCA effects for this trait.

The estimates on reciprocal effects for dry matter are given in Table 5b.Among the cross IBD50 X IBD47 (1.166**) had the highest positive and significant reciprocal effects on dry matter (%) of pumpkin followed by IBD40 X IBD23 and IBD50 X IBD40 with values 0.500** and  0.166**, respectively. The cross combinations IDB57 X IBD40 and IBD57 X IBD 47 revealed similar positive and significant reciprocal effects (0.333**) which could be specific combination to increase the dry matter (%) in pumpkin. Among all ten crosses, all crosses had reciprocal effects except IBD57 X IBD23. Rana et al., (2015) and Tamilselvi et al., (2015) observed similar trend in their experiment in pumpkin.

Table 6a: Specific combining ability (SCA) effects for brix (%), beta carotene, reducing sugar, non-reducing sugar, total sugar and fruit yield in a 5 x 5 full diallel populations of pumpkin

BRX=Brix, RS= Reducing sugar, NRS=Non reducing sugar, TS=Total sugar, FY=Fruit yield, BCAR=Beta carotene

Brix (%):

Among the 10 crosses, four revealed significant SCA effects and six showed non-significant SCA effects (Table 6a). The combination IBD40 X IBD57 exhibited the highest (0.121**) positive and significant SCA effects followed by IBD47 X IBD50, IBD47 X IBD50 and IBD23 X IBD50 with SCA values 0.034**, 0.012** and 0.002**, respectively. Hence these combinations could be regarded as a good specific combiner to increase brix (%) in pumpkin.

Out of 10, only two crosses revealed significant reciprocal effects and the rest eight showed non-significant reciprocal effects (Table 6b). The cross combinations IBD57 X IBD40 and IBD57 X IBD23 exhibited positive and significant reciprocal effects with 0.076** and 0.069*, respectively. Therefore, these combinations could also be used as specific combiner to increase brix (%) in pumpkin.

β carotene:

The SCA effects for this character was significant and positive for the combinations IBD23 X IBD57, IBD40 X IBD47, IBD40 X IBD57, IBD47 X IBD50 and IBD50 X IBD57 (Table 6a). To increase β carotene content in pumpkin, it is necessary to have positive significant value.

The estimates on reciprocal effects for beta carotene are given in (Table 6b).Among the cross, IBD50 X IBD47 (1.085**) had the highest positive and significant reciprocal effects on β carotene content in  of pumpkin followed by IBD40 X IBD23, IBD50 X IBD23 and IBD57 X IBD23, having values 0.338**, 0.228* and 0.210**, respectively.

Reducing Sugar:

Reducing sugar indicated the sweetness of a variety. The highest positive and significant SCA effects (Table 6a) on reducing sugar obtained from the cross combination IBD47 X IBD57 (0.313**) followed by IBD23 X IBD57 (0.122**), IBD23 X IBD47 (0.068**), IBD50 X IBD57 (0.053**) and IBD23 X IBD50 (0.047**) which could be used for further breeding program to increase reducing sugar content in pumpkin. The rest combinations had negative and 

 non-significant SCA effect. There existed reciprocal effects in the crosses. The highest negative and significant reciprocal effects (Table 6b) on reducing sugar obtained from the cross combination IBD50 X IBD23 (0.355*) followed by IBD50 X IBD47, IBD47 X IBD40, IBD57 X IBD23, IBD57 X IBD40 and IBD40 X IBD23 with values 0.291**, 0.184**, 0.128**, 0.115**and 0.014**, respectively which could be used for further breeding program to increase reducing sugar content in pumpkin.

Non reducing sugar:

The counts on SCA effects for non-reducing sugar are given in (Table 6a). Among the 10 crosses six revealed positive and significant SCA effects and four showed negative and non-significant SCA effects. The highest positive and significant SCA effects on non-reducing sugar obtained from the cross combination IBD47 X IBD50 (0.700**) followed by IBD23 X IBD57 (0.234**), IBD47 X IBD57 (0.182**), IBD40 X IBD57 (0.139**), IBD50 X IBD57 (0.091**) and IBD40 X IBD47 (0.012**).

The data on reciprocal effects for non-reducing sugar are given in (Table 6b). The parents showed reciprocal effects in specific crosses. The highest and  positive significant SCA effect was found in the cross combination IBD50 X IBD47 (1.377**) followed by IBD50 X IBD23, IBD40 X IBD23 and  IBD57 X IBD23 with values 0.583**, 0.353** and 0.338**, respectively which demonstrated that these combination could be selected to increase non reducing sugar content in pumpkin. Rest of all combinations had negative and non-significant SCA effect.

Total Sugar:

The highest and  positive significant SCA effects (Table 6a) was found in cross IBD40 X IBD47 (1.760**)  followed by IBD47 X IBD57, IBD23 X IBD50, IBD47 X IBD50 and IBD23 X IBD57 with SCA values of 0.926**, 0.293** ,0.260** and 1.760**, respectively for this trait. The cross combinations IBD23 X IBD40 and IBD23 X IBD57 revealed similar positive and significant effects (0.126**) which could be specific combination to increase the total sugar content in pumpkin.  To increase total sugar content, it is expected to have positive and significant effects. There existed reciprocal effects for total sugar. The highest  positive and significant reciprocal effects (Table 6b) was found  in cross combinations IBD57 X IBD23 and IBD47 X IBD40  with similar value of 1.16** for this trait followed by  IBD50 X IBD40 (1.00**), IBD57 X IBD40(0.833**) and IBD57 X IBD47(0.166**), respectively.  Among all ten crosses, all crosses had reciprocal effects except IBD40 X IBD23.

Table 6b: Reciprocal effect for brix (%), beta carotene, reducing sugar, non-reducing sugar, total sugar and fruit yield in 5 x 5 full diallel populations of pumpkin

BRX=Brix, RS=reducing sugar, NRS=Non reducing sugar, TS=Total sugar, FY=Fruit yield, BCAR= Beta carotene

Fruit yield:

For fruit yield significant and positive SCA effects was ascertained from the cross IBD40 X IBD47 with a value of 6.245*(Table 6a). All of ten cross combinations were significant but four crosses were negatively significant which reduced the fruit yield. Hence the positive and significant combinations could be selected for increasing fruit yield per plant in pumpkin.

The significant and negative reciprocal value (Table 6b) for this parameter was obtained from the crosses viz., IBD57 X IBD50, IBD40 X IBD23, IBD47 X IBD23, IBD50 X IBD47 and IBD50 X IBD23 which could provide decreased fruit yield per plant in pumpkin. The cross combination IBD47 X IBD40 demonstrated the highest significant and positive reciprocal effects (4.295*). Hence the positive and significant cross combination IBD50 X IBD40, IBD57 X IBD47, IBD57 X IBD40 and IBD57 X IBD23, could be used for further breeding program to increase fruit yield per plant in pumpkin.

Vr-Wr graph and Wr+Vr/parental mean graph:

Analysis of variance in Hayman’s analysis (following Morley Jones):

Analysis of variance showed that the significant values of ‘a’ for the character fruit length, hollowness, brix (%), beta carotene, reducing sugar and non-reducing sugar suggested that additive components were involved in the regulation of these characters. The dominance component (b) was highly significant indicated that this component was important in genetic control of most of the character studied except fruit length, dry matter, reducing sugar and fruit yield. (Table 7). Item ‘b₁’ was highly significant for two characters-flesh thickness and beta carotene, which detected unidirectional dominance and significant difference between parental and hybrid grand mean for these two characters. An asymmetrical distribution of dominant genes was suggested by the significant ‘b₂’ value for the characters beta carotene, non-reducing sugar and total sugar. The ‘b₃’values were also significant for most of the character studied except fruit length, dry matter, reducing sugar and fruit yield which indicated the dominance deviations which are not attributable to ‘b₁’ and ‘b₂’and showed important contribution to the non-additive gene action. Significant ‘c’ value for the trait fruit length, brix (%), beta carotene, reducing sugar and total sugar indicated the presence of maternal effect. Significant‘d’ value indicated presence of reciprocal difference in the character viz. fruit length, hollowness, brix (%), reducing sugar and non-reducing sugar.

FL= Fruit length, FB= Fruit breadth, HN= Hollowness, FT=Flesh thickness, DRM= Dry matter (%), BRX=Brix%, BCAR= Beta carotene (mg/100g), RS= Reducing sugar (g/100g), NRS=Non reducing sugar (g/100 gm), TS=Total sugar (g/100g), FY=Fruit yield (kg/plant)

The analysis of variance due to diallel progenies indicated significant differences among themselves, which warrants further analysis. Hayman’s graphic approach to diallel analysis is based on monogenic additive model. In this approach Hayman’s graphical analysis (Vr-Wr graph) was done and the findings are presented as Vr-Wr graphs, the two dimensional depiction made based on the parental variance (Vr) and parent offspring co-variance (Wr) which are presented in the Fig.(1-11, a) for eleven characters studied.