Evaluation of Local Pumpkin Genotypes in the Central Region of Saudi Arabia

A.A. Alsadon, H.H. Hegazi, and I.A. Almousa

Department of Plant Production, College of Agriculture, King Saud University,
P.O. Box 2460, Riyadh 11451, Saudi Arabia

Additional index words. Cucurbita moschata, squash, phenotypic correlation, growth characteristics

Abstract. Many local pumpkin (Cucurbita moschata Poir.) landraces are spread over a wide range of environments in the Kingdom of Saudi Arabia and used in many local dishes and food industries. Fruit of the most common genotypes, Hasawi-1, Hasawi-2, Masri, Najdi, and Skaka were collected from different regions of the Kingdom. An evaluation trial was conducted near Riyadh in the central region where vegetative traits, yield, and yield components were studied. Genotypes showed variation for most of these traits. However, nonsignificant differences in yield were found among the studied genotypes. Average fruit weight was the highest in Najdi and the lowest in Skaka, which had the highest seed number and seed weight. On the other hand, Hasawi-1, Masri, and Najdi showed the highest total soluble solids and reducing sugar content. Significant correlation coefficients among vegetative traits, yield, and yield components were discussed. The results of this study should be useful in breeding programs and improving local pumpkin cultivars under the arid conditions of the central region of Saudi Arabia.

We thank Safwat M. Othman for his technical help and Saleh El Hadary for his assistance in data analysis.

Saudi Arabia comprises a major part of the Arabian Peninsula with an estimated area of 2,250,000 km2. The climate varies from one region to another. Saudi Arabia is located between 16 and 32 degrees north latitudes, which means that it is mostly located in the dry tropical desert region (MAW-JECOR, 1988).

In Saudi Arabia, local pumpkin (Cucurbita moschata Poir.) is one of the major vegetable crops. It is consumed in different local dishes and some food industries such as jams, purees, and cakes.

There is no specific statistics on the world production of pumpkin alone. It is generally mentioned with squash. Pumpkin has been grown widely in Saudi Arabia for many decades. The total area of pumpkin and squash harvested in 1994 was 7,420 ha with total production of 69,687 ton (MAW, 1996) and average yield of 9.391 ton·ha­1.

Variations among cucurbit species in vegetative and yield characteristics have been reported. Examples are the studies on cucumber (Prasad and Singh, 1992), melon (Swamy and Dutta, 1991), watermelon (Mondal et al., 1989), muskmelon

(Swamy and Dutta, 1991), and pumpkin (Damarany et al., 1995; Doijode et al., 1982; Doijode and Sulladmath, 1986; Rana et al., 1985).

Researchers and farmers have noticed variations and instability among fruit characteristics of local pumpkin genotypes grown in Saudi Arabia (Al-Sulaiman, 1992). There has been limited work on the improvement of cucurbits in Saudi Arabia. Ibrahim and coworkers were among the pioneers to initiate the cucurbitaceae improvement program. They have released two sour-sweet melon (Ibrahim and Al-Zeir, 1992) and two desert-adapted winter squash (Ibrahim et al., 1996) genotypes.

The objectives of this work were a) to evaluate the performance of local pumpkin genotypes for their growth, yield and fruit composition and b) to determine the phenotypic correlation among vegetative, yield and fruit composition characteristics of local pumpkin genotypes.

Materials and methods

Plant materials. Fruit of local pumpkin genotypes were collected during the period from November 1996 to February 1997 from different re

 

Cucurbitaceae '98


gions of Saudi Arabia where they have been commonly grown for several decades (Fig. 1); Hasawi-1 and Hasawi-2 from the eastern region, Masri and Najdi from the central region, and Skaka from the northern region. The distinction between Hasawi-1 and Hasawi-2 was based on fruit shape and seed size. Local pumpkin genotypes exhibit various fruit size and shape. Fruit characteristics are shown in Fig. 2 and Table 1.

Seeds were collected, washed with water, air-dried, and kept at room temperature until time of planting. Seeds were planted in JV-7 pellets on 28 Mar. 1997 and were placed in the greenhouse (Average day/night temperatures were 24/18 °C).

Field layout. Five hundred seedlings, 14 days old, representing five different genotypes, were

transplanted out in the field on 9 Apr. 1997. The average temperature and relative humidity during the growing season are shown in Table 2.

The field plot is located in the Agricultural Research and Experiment Station (ARES), King Saud University at Dirab near Riyadh, Saudi Arabia. Soil is sandy loam in texture with the pH of 7.6 and EC of 4.9 dS·m­1. The experimental plot consisted of two rows, each one was 12 m long and 0.8 m wide. The distance between seedlings in a row was 1 m. The field plot layout was a randomized complete block design with five replications. Fertilization, furrow irrigation and other cultural practices were carried out as recommended.

Vegetative and growth characteristics. Plant length (cm), number of leaves per plant, number of

FIG 1

Figure 1. Map of Saudi Arabia showing the different regions from where fruit of local pumpkin genotypes have been collected.

Cucurbitaceae '98


vines per plant and leaf area (cm2) were measured at 30, 45, 60 days from transplanting. Fruit were harvested at full maturity, then they were transferred to the laboratory and kept at room temperature until they were evaluated for number of fruit per plant, fruit weight (g), fruit length (cm), fruit diameter (cm), flesh thickness at the middle of the fruit (cm), number of seeds per fruit, seed weight per fruit (g), and weight of 100 seeds (g). Percent (%) total soluble solids (TSS) was measured with a hand refractometer (HRN-32, Kruss, Germany). Analysis of reducing sugars (in mg/100 g fresh weight) was performed on Shimadzu HPLC (Shimadzu LC-10, Shimadzu, Kyoto, Japan) according to AOAC method (1995). Determination of total sugars was carried out according to the procedure of Lane-Eynon method in AOAC (1995).

Statistical analyses. Data were analyzed

Figure 2. Fruit characteristics of five local pumpkin genotypes grown in the central region of Saudi Arabia.

Table 1. Fruit characteristics of five local pumpkin genotypes grown in the central region of Saudi Arabia.

Genotype Skin color Size Shape Shell Flesh color

Hasawi-1 Yellow Medium (1­2 kg) Pyriform­globular Semi-grooved Yellow

Hasawi-2 White Large (>2kg) Globular Semi-grooved White

Masri Light yellow Large (>2kg) Pyriform Grooved Light yellow

Najdi Light yellow Large (>2kg) Pyriform Grooved Light yellow

Skaka Orange Small (<1kg) Globular Semi-grooved Yellow

Table 2. Temperature and relative humidity data during the growing season.z

Temperature (°C) Relative humidity (%)

Max Min Max Min

Period Mean Range Mean Range Mean Range Mean Range

April 1997 34 27­42 15 9­24 33 13­70 9 6­15

May 1997 40 31­45 22 15­29 16 11­68 7 5­10

June 1997 45 41­48 23 20­27 13 11­19 5 4­6

July 1997 44 42­47 23 20­29 12 11­16 5 4­6

zData were collected by the weather station of the ARES. Data cover the period from transplanting to harvest.

Cucurbitaceae '98


Table 3. Statistics of vegetative, fruit, seed and quality characteristics of five local pumpkin genotypes grown in the central region of Saudi Arabia.

Characteristics Hasawi-1 Hasawi-2 Masri Najdi Skaka

Plant length (cm) Mean 132.08 126.50 125.82 141.01 59.92

Range 241.75 203.81 188.81 197.20 122.75

cv 62.21 58.06 52.39 52.28 70.23

No. of leaves/plant Mean 74.64 73.04 66.48 79.36 57.45

Range 181.63 180.37 141.38 155.81 114.81

cv 84.37 80.71 80.59 71.34 65.09

No. of vines/plant Mean 6.26 4.82 5.86 6.87 4.62

Range 15.88 9.78 12.16 12.75 8.90

cv 85.03 64.25 75.15 64.54 67.79

Leaf area (cm2) Mean 16,454.86 20,580.36 14,656.58 18,137.65 22,367.02

Range 39,243.64 53,744.58 36,382.90 42,646.50 50,393.10

cv 90.38 86.16 88.82 77.42 75.81

Yield/plant (g) Mean 746.75 542.02 842.50 659.39 221.71

Range 1,960.00 361.43 2,562.50 1,394.44 316.54

cv 108.47 25.22 127.31 87.01 54.51

No. of fruit/plant Mean 0.36 0.25 0.46 0.18 0.23

Range 0.90 0.11 1.20 0.33 0.26

cv 99.36 17.67 99.10 84.46 42.48

Yield (ton/ha) Mean 7.78 5.65 8.78 6.87 2.31

Range 20.42 3.76 26.69 14.53 3.30

cv 108.47 25.22 127.31 87.01 54.51

Ave. fruit weight (g) Mean 1,970.25 2,230.50 2,099.17 2,859.17 952.37

Range 675.00 2,300.00 5,125.00 4,250.00 334.17

cv 14.34 37.56 104.83 60.29 13.85

Fruit length (cm) Mean 21.31 12.78 34.06 37.19 15.45

Range 8.75 6.00 6.67 6.50 4.33

cv 16.35 19.14 9.85 7.94 10.37

Fruit diameter (cm) Mean 13.19 18.96 10.07 11.44 14.73

Range 4.45 5.83 4.63 8.00 6.00

cv 15.20 11.43 23.82 33.22 16.4

Flesh thickness (cm) Mean 2.18 2.59 2.06 2.19 1.86

Range 0.40 1.22 0.75 1.50 1.15

cv 8.04 17.50 19.16 28.57 25.05

Seed number/fruit Mean 144.8 216.55 63.4 91.83 308.75

Range 139.2 140.3 57.87 148 127.5

cv 35.32 31.21 52.01 67.39 19.08

Seed weight/fruit (g) Mean 10.67 10.17 5.59 8.60 12.24

Range 12.97 5.00 5.90 16.35 5.12

cv 49.55 23.16 55.35 85.03 18.45

Weight of 100 seed Mean 7.04 5.21 8.67 10.14 4.16

Range 4.67 2.60 2.7 11.84 2.76

cv 27.84 20.68 16.46 50.44 29.11

Total soluble solids Mean 14.89 10.43 16.22 15.45 12.05

Range 3.65 4.50 0.5 3.25 1.00

cv 9.55 17.43 1.75 9.54 3.57

Reducing sugars Mean 1.5 0.08 1.75 2.42 0.21

Range 1.78 0.25 0.51 1.70 0.47

cv 48.29 133.20 14.69 28.88 84.58

Nonreducing sugars Mean 2.46 2.23 2.43 2.03 0.80

Range 1.85 2.01 0.78 1.30 2.15

cv 32.66 35.19 17.27 26.33 99.86

Total sugars Mean 3.97 2.30 4.18 4.45 1.02

Range 2.01 2.00 0.85 1.58 2.45

cv 20.97 34.10 10.17 18.42 88.66

Cucurbitaceae '98


using SAS (Ray and Sall, 1982) software at the Computer Center of the College of Agriculture, King Saud University. Correlation coefficients and mean separation were done as described by Steel and Torrie (1980).

Results and discussion

The variability magnitudes within the original populations, of the five local pumpkin genotypes were measured for the most important horticultural characteristics (Table 3). In general, data indicated that most of the vegetative and yield characteristics reflected high variability with each of the five studied populations except that for yield per plant trait which had the lowest coefficient of variation (25.22%) within the populations of Hasawi-2. Fruit characteristics such as average fruit weight, fruit length, and diameter showed moderate variability magnitudes except for the average fruit weight trait within Masri and Najdi populations. On the contrary, the flesh thickness trait showed the lowest variability magnitudes and their cv values ranged from (8.04% to 28.57%) for the five populations.

Seed number, seed weight per fruit and weight of 100 seed showed relatively intermediate variability. The variability magnitudes within the original populations for TSS were the lowest among all traits (1.75% to 17.43%). Total sugars and nonreducing sugars showed greater variation within Skaka population. Differences among cucurbit species were also reported. Kasrawi (1995) evaluated 41 half-sib families of summer squash (C. pepo L.) for 17 morphological and horticultural traits. A large diversity was observed for many traits within and among families including fruit shape and fruit number per plant. Mondal et al. (1989) showed wide range of variability in fruit length, fruit diameter, number of fruit per plants, flesh thickness, and fruit yield.

Figure. 3. Development of vegetative growth characteristics of five local pumpkin genotypes grown in the central region of Saudi Arabia. Means within each period followed by the same letter are not significantly different at P < 0.05 (lsd).

Cucurbitaceae '98


Table 4. Yield and yield components of five local pumpkin genotypes grown in the central region of Saudi Arabia.z

No. Wt/ No. Seed 100

fruit/ fruit Yield seed/ wt/ seed

Genotype plant (g) (g/plant) (ton/ha) fruit fruit (g) wt (g)

Hasawi-1 0.364 az 1970.3 ab 746.8 a 7.779 a 144.80 bc 10.67 a 7.04 abc

Hasawi-2 0.254 a 2230.5 ab 542.0 a 5.646 a 216.55 ab 10.17 a 5.21 bc

Masri 0.458 a 2099.2 ab 842.5 a 8.776 a 63.40 c 5.59 a 8.67 ab

Najdi 0.180 a 2859.2 a 659.4 a 6.869 a 91.83 c 8.60 a 10.14 a

Skaka 0.229 a 952.4 b 222.7 a 2.309 a 308.75 a 12.24 a 4.16 c

lsd0.05 ns 1697.8 ns ns 97.42 ns 3.85

zMeans within a column followed by the same letter are not significantly different at P < 0.05.

There is great possibility for future studies on the improvement program in pumpkin. The studied traits of vegetative growth, yield, seed weight per fruit, and fruit composition within Skaka population could be improved through selection. Jaradat (1991) and Singh et al. (1996) reported that population improvement through selection could be performed, to varying degrees, according to the amount of variation present in each population , selection intensity, and heritability of these traits.

Vegetative growth characteristics. The developments of vegetative growth characteristics are shown in Fig. 3. Plant length, number of vines per plant and number of leaves per plant exhibited significant differences among genotypes at 60 days from transplanting. There were significant differences in leaf area. Skaka, the earliest maturing genotype, had the highest leaf area at 30 and 45 days. However, these differences diminished at 60 days. Rahman et al. (1990) evaluated sweet gourd (C. moschata) for their vegetative growth traits. They reported significant variation for the different studied traits. Similar findings were reported in pumpkin by Doijode et al. (1982). They showed variation in stem length and number of lateral branches.

Yield and yield components. There were significant differences among pumpkin genotypes

for average fruit weight, seed number per fruit and weight of 100 seed (Table 4). Although Masri and Hasawi-1 outyielded other genotypes, the differences among them were not significant. The yield of Skaka was the lowest. Hasawi-2 and Najdi had an intermediate yield response. Average fruit number per plant was not significantly different among genotypes. On the contrary, Rahman et al. (1990) reported significant differences in fruit number, length and weight. Average fruit weight was the highest in Najdi followed by Hasawi-2, Masri, and Hasawi-1 genotypes, but the differences did not reach the significant level. Masri had the largest fruit number per plant while Najdi had the lowest. Average fruit weight differed significantly among the studied genotypes. Average fruit weight was the highest in Najdi and lowest in Skaka fruit, which resulted in nonsignificant yield differences. Masri and Najdi had the lowest seed number per fruit and seed weight per fruit with the highest weight of 100 seeds.

Fruit composition. Fruit quality measured as TSS and total, reducing, and nonreducing sugars was significantly different among genotypes (Table 5). Najdi and Masri fruit had the highest TSS, total sugars and reducing sugars. Skaka yielded the lowest content of total, reducing and nonreducing

Table 5. Fruit composition of five local pumpkin genotypes grown in central region of Saudi Arabia.

Sugars (mg/100 g fresh wt)

Genotype Total soluble solids(%) Total Reducing Nonreducing

Hasawi-1 14.89 bz 3.38 bc 1.20 b 2.17 a

Hasawi-2 10.43 d 2.38 c 0.061 c 2.32 a

Masri 16.22 a 4.24 ab 1.79 b 2.46 a

Najdi 15.45 ab 4.72 a 2.50 a 2.22 a

Skaka 12.05 c 1.05 d 0.23 c 0.82 b

lsd0.05 1.31 1.06 0.66 0.76

zMeans within a column followed by the same letter or not significantly different at P < 0.05.

Cucurbitaceae '98


Table 6. Overall correlation coefficient between vegetative growth, yield, seed and fruit composition characteristics of five local pumpkin genotypes grown in the central region of Saudi Arabia.z

Characteristic 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

1. Plant length 0.825** 0.748** 0.331 0.398* 0.155 0.398* 0.417* 0.414* ­0.131 0.527** ­0.514* ­0.138 0.379 0.335 0.500** 0.582** 0.682**

2. No. of leaves/plant 0.825** 0.720** 0.481** 0.355* 0.481** 0.440* 0.265 0.021 0.545** ­0.253 0.111 0.334 0.269 0.345 0.395* 0.467**

3. No. of vines/plant 0.462* 0.444* 0.231 0.444* 0.470** 0.513** ­0.284 0.275 ­0.488** 0.020 0.614** 0.604** 0.617** 0.231 0.564**

4. Leaf area 0.260 0.196 0.260 0.234 ­0.155 0.319 0.518** 0.394* 0.318 ­0.133 ­0.141 ­0.080 ­0.038 ­0.078

5. Yield (g/plant) 0.428* 0.999** 0.732** 0.220 0.085 0.357 ­0.160 0.077 0.179 0.110 0.146 0.173 0.200

6. Ave. No. fruit/plant 0.428* 0.006 0.070 ­0.034 0.109 ­0.038 0.099 0.042 0.029 ­0.029 0.042 0.006

7. Yield (ton/ha) 0.732** 0.220 0.085 0.357 ­0.160 0.077 0.178 0.110 0.146 0.173 0.200

8. Average fruit weight 0.319 0.069 0.372 ­0.248 ­0.055 0.263 0.191 0.212 0.258 0.295

9. Fruit length ­0.740** ­0.112 ­0.739** ­0.468* 0.592** 0.781** 0.824** 0.192 0.686**

10. Fruit diameter 0.503* 0.508* 0.504* ­0.225 ­0.642** ­0.574** ­0.248 ­0.543**

11. Flesh thickness 0.156 0.227 0.007 ­0.126 ­0.026 0.302 0.154

12. Seed number/fruit 0.584** ­0.649** ­0.833** ­0.696** ­0.292 ­0.681**

13. Seed weight/fruit 0.131 ­0.373 ­0.258 ­0.168 ­0.286

14. Weight of 100 seeds 0.702** 0.652** 0.044 0.509*

15. Total soluble solids 0.800** 0.107 0.621

16. Reducing sugars 0.234 0.833**

17. Nonreducing sugars 0.732**

18. Total sugars

*,**Significant at P = 0.05 or 0.01, respectively.

sugars, while Hasawi-2 had the lowest TSS. Skaka was characterized by both short vegetative growth period and low TSS content. This is in agreement with the positive association between the two traits reported by Welles and Buitelaar (1988).

Correlations. Correlation coefficients were positively significant between vegetative and fruit characteristics (Table 6). Yield significantly correlated with plant length and number of vines per plant. Significant (r = 0.43) and highly significant (r = 0.73) correlations were found between yield and each number of fruit per plant and average fruit weight, respectively. Results of this study indicated that, for increasing fruit yield, selection might be directed toward plants having a higher number of fruit with larger fruit size. These results agree with Rastogi and Deep (1990) and Dehua et al. (1995) in cucumber, Swamy and Dutta (1991) in muskmelon; Mondal et al. (1989) in watermelon, and Rana et al. (1985) in pumpkin. Seed weight per fruit was positively correlated with fruit length and fruit diameter. Same results were found by Mao et al. (1996). It is concluded that fruit length and diameter could be used as selection criteria in breeding pumpkin for seed production.

Fruit quality measured as TSS, total sugars (reducing and non reducing sugars) was significantly and positively correlated with vegetative growth characteristics such as number of vines per plant and

seed characteristics such as average seed weight per fruit and weight of 100 seeds. These findings are in agreement with results obtained by Swamy and Dutta (1991). On the other hand, highly significant negative correlation coefficients were found among fruit quality traits; seed number per fruit and each of TSS (r = ­0.833), reducing sugars (r = ­0.696) and total sugars (­0.681). Fruit diameter had also correlated negatively with fruit quality traits. Seed withdraw sugars from fruit reserve to utilize it for their development. Total soluble solids are related with sugar contents and are important quality components in cucurbit fruit (Robinson and Decker-Walters, 1997). Generally, C. maxima and C. moschata plants produce fruit with highest solids and deepest flesh color which make them preferred for commercial canning (Robinson and Decker-Walters, 1997).

The high values of TSS and total sugars in Hasawi-1, Masri and Najdi genotypes indicated their high quality attributes. Maintaining quality of fruit while at the same time increasing yield has been one of the primary goals of plant breeders. Fruit quality is an important criterion in the production of pumpkin.

Literature cited

Al-Sulaiman, A.I. 1992. All about growing cucurbits in Saudi Arabia. National Center for Agriculture and Water Research, Ministry of Agriculture and Water,

Cucurbitaceae '98


Riyadh, Saudi Arabia (in Arabic).

Association of Official Analytical Chemists. 1995. Official methods of analysis. 16th ed. Assn. Official Anal. Chemists. Washington, D.C.

Damarany, A.M., M.M. Abdulla, and M.H. Abdul-Nasr. 1995. Yield and yield component of some Cucurbita spp. Cultivars and hybrids under Assiut conditions. II. Pumpkin (Cucurbita spp.). Assiut J. Agr. Sci. 26(1):59­71.

Dehua, M.A., L.U. Shuzhen, and U.N. Shen-weny 1995. Correlation and path analysis on some agricultural characters in cucumber. Acta Agr. Boreali-Sinica 10(2):34­37.

Doijode, S.D., U.V. Sulladmath, and R.S. Kulkarki. 1982. Graphic analysis and genetic diversity for vegetative and reproductive traits in pumpkin (Cucurbita moschata Poir.). Mysore J. Agr. Sci. 16:4, 439­442.

Doijode, S.D. and U.V. Sulladmath. 1986. Genetic variability and correlation studies in pumpkin. (Cucurbita moschata Poir.). Mysore J. Agr. Sci. 20(1):59­61.

Ibrahim, A.M and K.A. Al-Zeir. 1992. 'Najd I' and Najd II.' two sour-sweet melon cultivars. HortScience 27(3):276­277

Ibrahim, A.M., A.I. Al-sulaiman, and K.A. Al-Zeir. 1996. 'Hamdan' and 'Qasim' desert-adapted winter squashes. HortScience 31(5):889­890.

Jaradat, A.A. 1991. Phenotypic divergence for morphological and yield-related traits among landrace genotypes of durum wheat from Jordan. Euphytica 52:155­164.

Kasrawi, M.A. 1995. Diversity in landraces of summer squash from Jordan. Genet. Resources Crop Evol. 42(3):223­230.

Mao, C.S., C.Y. Min, X.H. Ru, and Z.Q. Yan. 1996. Analysis of seeds per fruit and effective factors in Indian pumpkin. Acta Agr. Boreali-Sinica 11(1):114­117.

Ministry of Agriculture and Water. 1996. Agriculture statistical yearbook. vol. 9. Riyadh, Saudi Arabia.

Ministry of Agriculture and Water and the Saudi Arabian­United States Joint Commission on Economic

Cooperation. 1988. Climate atlas of Saudi Arabia. Riyadh, Saudi Arabia.

Mondal, S.N., A. Rashid, K.A.K.M.A. Hossain, and M.A. Hossain 1989. Genetic variability, correlation and path-coefficient analysis in watermelon. Bangaladesh J. Plant Breeding Genet. 2(1,2):31­35.

Prasad, V.S. and D.P. Singh 1992. Estimated of heritability, genetic advance and association between yield and its components in cucumber (Cucumis sativus L.). Indian J. Hort. 49:62­69.

Rahman, M.M., S.K. Dey, and M. Wazuddin. 1990. Yield, yield component and plant characters of several bitter gourd, ribbed gourd, bottle gourd and sweet gourd genotypes. Bangaladesh Agricultural Univ., Mymensingh (Bangladesh). Proc. workshop on Bangladesh Agricultural Univ. Res. Progress. Bau. p. 117­127.

Rana, T.K., R.N. Vashistha, and M.L. Pandita 1985. Correlation and coefficient studies in pumpkin (Cucurbita pepo L.). Haryana Hort. Sci. 14:108­113.

Rastogi, K.B. and A. Deep. 1990. A note on inter-relationship between yield and important plant characters of cucumber (Cucunis sativus L.). Veg. Sci. 17:102­104.

Ray, A.A. and J.P. Sall. 1982. SAS user's guide: Statistics. SAS Inst., Cary, N.C.

Robinson, R.W. and D. Decker-Walters. 1997. Cucurbits. CAB Intl., New York, New York.

Singh, S.P., N.K. Singh and I.B. Maurya. 1996. Genetic variability and correlation studies in bottlegourd (Lagenaria siceraria (Molina) Stand1). PKV Res. J. 20(1):88­89.

Steel, R.G. and J.H. Torrie. 1980. Principles and procedures of statistics. McGraw-Hill, New York.

Swamy, K.R. and O.P. Dutta. 1991. Coheritable variation in muskmelon (Cucumis melo L.). Indian J. Agr. Res. 25(3):149­153.

Welles, G.W.H. and K. Buitelaar. 1988. Factors affecting soluble solids content of muskmelon (Cucumis melo L.). Netherlands J. Agr. Sci. 36:239­246.

Cucurbitaceae '98