Advertisement

Rootstock effects on seed yield and quality in watermelon

  • Mohamed Dhamir Kombo
  • Nebahat SariEmail author
Open Access
Research Report
  • 965 Downloads

Abstract

This research was conducted to investigate rootstock effects on seed yield and quality in watermelon. The work was conducted in the experimental fields and laboratories of the Department of Horticulture of the University of Cukurova in the 2016 and 2017 growing seasons. Watermelon cv. Crimson sweet (CS) scion was grafted onto three different rootstocks (Cucurbita ‘NUN-9075’, Lagenaria ‘Argentario’, and citron watermelon ‘PI296341’). Plants were assessed according to their main stem length, main stem diameter, number of nodes, biomass, pollen production and development, fruit yield and quality, and seed yield and quality. NUN-9075/CS and Argentario/CS graft combinations resulted in higher average stem length, plant biomass, fruit yield and quality, seed yield, seed emergence, and germination percentage. There was no significant difference observed between graft combinations in pollen viability, pollen germination, and normal pollen production. No significant difference was observed between graft combinations in accelerated ageing (AA), and the seed germination decreased after AA of 192 h. In the second year (2017), there was higher main stem length, main stem diameter, total fruit yield, seed yield, and number of seeds per fruit compared to the first year (2016). Based on the results of this study, NUN-9075 rootstock performed better than other rootstocks; hence, it is recommended as the best rootstock.

Keywords

Fruit analysis Grafting Graft combinations Pollen production Watermelon seeds 

1 Introduction

Watermelon [Citrullus lanatus (Thunb.) Matsum. & Nakai] of the family Cucurbitaceae is one of the most important horticultural crops worldwide nutritionally and economically. It is grown in both temperate and tropical regions (Bosgnin 2002). Watermelon species, native to Africa, are believed to be from the Kalahari Desert of Africa and have been cultivated all over the world since ancient times (Meeuse 1962; Paris 2017). Watermelon has three subspecies: C. lanatus subsp. lanatus L., which represents the group of ancient cultigens, such as citron watermelon; C. lanatus subsp. mucosospermus L.; and C. lanatus subsp. vulgaris, which represents the modern cultivated sweet watermelon group (Ren et al. 2014). Watermelon contains important bioactive compounds for health, including phenolic compounds, carotenoids, citrulline, vitamins, and flavonoids (Gunn et al. 2015). The lycopene in red flesh watermelon is believed to be almost 40% higher (4.81 mg per 100 g) than that in tomato (3.03 mg per 100 g) (Naz et al. 2014).

Watermelon grafting faces many challenges, such as low number of rootstocks (Elazar and Zoran 2014), graft incompatibility (Andrews and Marquez 1993), and high costs (Djidonou et al. 2013). The most common rootstocks for watermelon grafting are wild watermelon (C. lanatus var. citroides) (King et al. 2010; Kong et al. 2014), Cucurbita interspecific hybrids (C. moschata × C. maxima), and the bottle gourd accessions (Lagenaria siceraria) that are highly resistant to soil-borne fungi (Liu et al. 2015). The use of suitable rootstocks offers many benefits; however, rootstocks are associated with negative effects of grafting. Therefore, it is important to choose the appropriate rootstocks for successful grafting.

Besides the agronomic practices and environmental conditions, the successful production of the crop also depends on the quality of the seeds used for sowing. Ambika et al. (2014) showed that quality seeds can increase yield by 15–20%. Seed yield and quality can be increased using different methods, such as vine control, pollination techniques, and maintenance to improve fruit set. Numerous studies were conducted to investigate the impact of these techniques on seed yield and quality. Some of the factors shown to increase seed yield and quality were fruit-set order (Nerson 2004), fruit shape and plant density (Nerson 2005), plant spacing and pollen quality (Lima et al. 2003), effect of fertilizers (Olaniyi and Tella 2011), and seed size (Ambika et al. 2014). However, so far, no study has investigated the effect of rootstocks on seed yield and quality; therefore, the aim of this work was to study the best rootstock that increases seed yield and quality.

2 Materials and methods

This study was conducted at Horticulture Experimental fields and Biotechnology Center of the Cukurova University in Adana, while seed sowing and grafting was done at the Antalya Seed Company in Antalya,Turkey in the 2016 and 2017 spring seasons.

2.1 Plant materials

We used four plant species of the family Cucurbitaceae for this research: 2 hybrids, Cucurbita ‘NUN - 9075’ (Cucurbita maxima × Cucurbita moschata) and Lagenaria ‘Argentario’ (Lagenaria siceraria); and 2 watermelon species, citron watermelon (Citrullus amarus Schard) ‘PI 296341’ and ‘Crimson sweet’ (CS) cultivar (Citrullus lanatus). ‘NUN 9075’, ‘Argentario’, and PI 296341 were used as rootstocks, while watermelon cv ‘Crimson sweet’ was used as control (nongrafted) and as a scion grafted onto all other rootstocks. Seeds for rootstock production were obtained from the Department of Horticulture seed stock, and those for scions were bought from Antalya Seedlings Company. Latin square design was used in assigning rootstocks to the plots during experimentation.

2.2 Seed sowing, grafting, and transplanting

Seeds of citron watermelon were sown on January 29, 2016, and those of Crimson sweet (scion) were sown on January 26, 2016. Seeds of NUN-9075 and Argentario were sown later (February 1, 2016) for the 2016 growing season because citron and Crimson sweet watermelon germinate more slowly than NUN-9075 and Argentario. For 2017, sowing was done on February 3, 2017 for citron watermelon, February 13, 2017 for Crimson sweet (control/scion), and February 16, 2017 for NUN-9075 and Argentario. Grafting was done using the splice/one cotyledon technique on February 23, 2016 for the first year and February 22, 2017 for the second year. All grafted seedlings were healed and maintained in healing chambers at 23–24 °C and 60–70% relative humidity for 10 days, and then all grafted seedlings hardened for 20 days.

One month after grafting, all plants were transported to Adana from Antalya. The next day, seedlings were transplanted to the main experimental field. Grafted and nongrafted (control) plants were planted 3 m × 0.75 m apart on March 30, 2016 for the first year and April 6, 2017 for the second year. Each graft combination replicated randomly 4 times, and 20 plants were planted in each plot. Soon after transplanting, low plastic tunnels were made to protect the young plants from cold temperature and heavy rains; 2 weeks later, the tunnels were removed. Fertilizer was applied 3 times during the growing period as follows: after transplanting, during plant development, and at flowering-young fruit formation. Fertilizer was applied at a rate of 15:15:20 kg per 1000 m2, with pure nutrients as N:P2O5:K2O. Monoammonium phosphate (MAP) and potassium nitrate (KNO3) with ratio of 18:18:18 and 20:20:20, respectively, were used as fertilizers.

Observation and measurements were done on various days: when main stem formation reached 10 cm long, at the time of first flowering of male and female flowers, and at 50% flowering of male and female flowers. Other measurements were done on plant biomass, fruit and seed yield, and quality analysis. To assess the capacity of pollen production among rootstocks, 10 flowers were used in every replication. Five mature flowers before opening when pale-yellow color started to develop (approximately one day before anthesis) were selected, closed in the afternoon using clips, and picked the next morning for pollen germination and viability. In the morning, after picking 5 closed flowers from each replicate, 5 other unopened flowers close to anthesis were picked and sent to the laboratory for anther counting and determination of pollen production. Pollen viability, pollen germination, and normal pollen production assays were conducted according to Norton (1966) and Eti and Stosser (1988). Harvesting was done on July 7, 2016 for the first year (100 days after transplanting) and on June 28, 2017 for the second year (83 days after transplanting). Fruits were left for a longer time in the field before harvest in order to get maximum maturation of the seeds in the fruits. All fruits from all plots were harvested and weighed and then fruits were analyzed for single fruit weight, fruit height, fruit diameter, fruit rind thickness, and total soluble solid content. Seeds were extracted manually, fermented for 3 days, and then well washed and put on fine wire mesh for drying at room temperature. Seeds were left to dry for 1 week and then they were taken to Biotechnology Laboratory of the Cukurova University for seed analysis tests. Before being used for germination, emergence, and aging tests, seeds were sterilized by soaking in vials with 3% sodium hypochloride for 10 min (Wang et al. 2013; Souza et al. 2013).

2.3 Statistical analysis

Measured and observed data were recorded in Microsoft Excel, and averages for each parameter were calculated. Significant difference between means was calculated using Tukey’s multiple range test at a significance level of P ≤ 0.05. Statistical analysis was performed using JMP statistics software (v8.00, SAS Institute Inc., NC 27513-2414, USA).

3 Results and discussion

3.1 Plant growth and flowering

Results pertaining to number of days observed when the main stem reached 10 cm long after transplanting and flowering are presented in Table 1 for the 2016 and 2017 growing seasons. There was significant difference among rootstocks in number of days to 10-cm-long main stem formation in both growing seasons; rootstocks reached that length earlier in the 2017 growing season than the 2016 growing season. In the 2017 season, the NUN-9075/CS graft combination attained 10-cm main stem length earlier compared to other rootstocks; it reached that length after 8.25 days, while ‘PI296341/CS’ did so after 11.25 days. Also, in the 2016 growing season, the NUN-9075/CS graft combination was the earliest rootstock to attain 10-cm main stem length at an average of 14.25 days, while Argentario/CS was the latest rootstock to attain that main stem length (17.50 days). NUN-9075/CS resulted in early flowering for all measured flowering parameters in all growing seasons. The first male flower was observed 33.00 days after transplanting in the 2016 growing season. The 50% male flowers were observed after 36.00 days in the 2017 growing season, while 50% female flowers were observed 41 days after transplanting. Argentario/CS and PI296341/CS rootstocks exhibited late flowering in both seasons. Male flowers appeared earlier compared to female flowers with an average of 11 days between first male and first female flower, while the interval between 50% male and 50% female flower occurrence was 18 days for the 2016 growing season and 5 days for the 2017 growing season in the NUN-9075/CS graft combination. The difference in flowering between the first and second growing seasons may be explained by the difference in planting time and the disturbance on plant growth caused by heavy snow and rain accompained by strong wind in mid May 2016, which tore the plant leaves and chopped off the growing twigs. This delayed regrowth and flowering of the plants in the first year.
Table 1

Days to 10-cm-long main stem formation and days to flowering of first male flower (FMF), first female flower (FFF), 50% male flower (50% MF), and 50% female flower (50% FF) for the 2016 and 2017 growing seasons

Graft combination

DSL 10 (cm)

FMF (days)

FFF (days)

50% MF (days)

50% FF (days)

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

Control

13.25 c

10.25 ab

35.75 ab

32.00

53.75 a

39.75

43.75 b

36.50

59.75 a

44.75

PI 296341/CS

15.75 ab

11.25 a

34.50 ab

34.00

55.00 a

40.75

43.00 bc

37.00

59.50 a

45.00

Argentario/CS

17.50 a*

9.00 b

40.00 a

33.75

53.50 a

39.25

53.25 a

37.75

59.00 a

42.50

NUN-9075/CS

14.25 bc

8.25 b

33.00 b

33.50

44.75 b

37.75

37.25 c

36.00

55.00 b

41.50

Prob > f

0.0003

0.0056

0.0181

0.2451

0.0060

0.0644

0.0002

0.3604

0.0063

0.0819

LSD (0.05)

1.88

2.02

5.55

ns

7.24

ns

6.27

ns

3.45

ns

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

3.2 Plant growth measurements

Measurements on main stem length (MSL), main stem diameter (MSD), and number of leaves (NL) on the main stem were conducted 1 month after transplanting and repeated 1 month later. The results of these plant growth parameter measurements for 2016 and 2017 are shown in Table 2.
Table 2

Main stem length (MSL), main stem diameter (MSD), and number of leaves (NL) for the 2016 and 2017 growing seasons

Graft combination

MSL 1 (cm)

MSD 1 (mm)

NL 1

MSL 2 (cm)

MSD 2 (mm)

NL 2

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

Control

39.25 c

52.19 b

8.04

7.72 b

11.70 b

11.81 ab

127.95 b

211.06 c

11.96 b

13.30 b

25.15

22.19 b

PI 296341/CS

45.90 bc

52.31 b

8.59

9.71 a

12.95 b

10.19 b

174.35 b

242.63 c

14.33 ab

15.98 a

26.20

25.06 b

Argentario/CS

55.15 b

57.63 ab

7.91

9.26 a

13.05 b

11.13 ab

178.45 ab

292.25 b

13.79 ab

15.96 a

22.02

25.44 b

NUN-9075/CS

90.60 a*

76.31 a

9.19

10.16 a

17.45 a

12.94 a

234.50 a

376.06 a

16.06 a

17.70 a

25.90

37.06 a

Prob > f

0.0001

0.0101

0.052

0.0040

0.0001

0.0079

0.0022

0.0001

0.0062

0.001

0.5046

0.0015

LSD (0.05)

10.14

19.11

ns

1.54

2.25

1.86

57.84

49.62

2.60

2.15

ns

8.25

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

There was a significant difference between graft combinations on main stem length, main stem diameter, and number of leaves in all growing seasons. The ‘NUN-9075/CS’ graft combination resulted in highest values in the first measurement and second measurement with highest average MSL of 376.06 cm obtained in the second year. Control (non grafted) plants had the lowest average values in all parameters, followed by the Argentario/CS graft combination on NL1 and NL2 in the first year. Nongrafted plants had a main stem length of 39.25 cm, main stem diameter of 8.04 mm, and 11.70 leaves for the first measurement and 127.95 cm, 11.96 mm, and 25.15 for main stem length, main stem diameter, and number of leaves in the second measurement, respectively. However, there was no significant difference between graft combinations in average number of leaves for the second measurement of the first year, although the PI296341/CS graft combination was observed to have a higher number of leaves (26.20).

Stem length and number of leaves in this study are higher than those found by Yetisir and Sari (2004), who observed the highest stem diameter value of 26.27 cm in Lagenaria hybrid/CT and the lowest value of 6.0 cm in Luffa (landrace)/CT, and most number of leaves in Cucurbita maxima/CT (21.67) and lowest in Luffa cylindrica/CT (10.17). Indeed, results obtained from this study for main stem length, stem diameter, and number of leaves are higher compared to that found by Yetisir et al. (2006) and Gómez et al. (2017), who used Chitosan-PVA hydrogen with copper nanoparticles to improve growth of grafted watermelon. This may be because of the high vigor of varieties of rootstocks and scions used in this study.

3.3 Plant biomass content

Plant biomass content provides additional information on the growth characteristics of a particular plant. In this study, the importance of rootstock also has been assessed in plant vigor through leaves and stem on both a fresh and dry weight basis. There was a significant difference between rootstocks on plant fresh weight (kg plant−1) and plant dry weight (kg plant−1) at 5% level of significance with Prob > f = 0.0001 for both parameters. Results of plant fresh weight and plant dry weight are presented in Fig. 1.
Fig. 1

Effect of rootstocks on plant fresh weight (a) and dry matter content (b). Data were analyzed at 5% level of significance and are the average of four replications. Letters above the bars indicate statistically significant differences using Tukey’s multiple range test at p > 0.05

The NUN-9075/CS graft combination resulted in the highest average value of total plant fresh weight and total plant dry weight (3.39 and 0.36 kg plant−1, respectively). The lowest value for both total fresh weight and total plant dry weight was found in the control, Crimson sweet (0.43 and 0.06 kg plant−1). In the 2017 growing season, the same graft combination (NUN-9075/CS) resulted in a higher average value in total plant fresh and total dry weight (3.84 and 0.55 kg plant−1, respectively) compared to control and other graft combinations. Again, nongrafted plants (control) had the lowest value in total plant fresh weight and total dry weight (0.90 and 0.17 kg plant−1, respectively). Similar to other parameters observed in this study, values of the second year were higher than that of the first year.

The results of this study showing that grafted plants result in higher biomass than nongrafted plants, are in accordance with the results of Mohamed et al. (2014) and many others who concluded that grafted plants produce much more foliage and vigorous plants compared to nongrafted plants. The great biomass in grafted plants is due to the strong root system of the rootstocks, which results in enhanced water and nutrient uptake. The more water and nutrients, especially nitrogen, are transported to the sink, the more leaf and stem growth that results in plant vigor.

3.4 Pollen development

Pollen development was studied to assess the capability of fertilizing female flowers so as to have a better prediction of the fruit and seed yield and quality of production. The study further estimated the number of anthers per flower, number of pollen per anther, and number of pollen per flower (Table 3).
Table 3

Anther analysis for pollen production

Graft combination

Total number of pollen in a box with 5 flowers

Number of pollen per anther

Number of pollen per flower

2016

2017

2016

2017

2016

2017

Control

1,662,728

913,767

107,084

60,918

332,546

182,753

PI296341/CS

1,525,003

923,844

101,667

61,590

305,001

184,769

Argentario/CS

1,891,236

1,031,386

124,487

68,759

378,247

206,277

NUN-9075/CS

1,780,409

1,121,775

113,109

74,785

356,082

224,355

Prob > f

0.3

0.42

0.36

0.42

0.3

0.42

LSD (0.05)

ns

ns

ns

ns

ns

ns

ns no significant difference

There was no significant difference among rootstocks on number of anthers per flower, number of pollen per anther, and number of pollen per flower for both experiment years. All graft combinations and the control had the same average number of anthers per flower, and all flowers were found to have an average of 3.00 anthers. Also, there was no significant difference among rootstocks at 5% level on number of pollen per anther and number of pollen per flower. Pollen grain production per flower (pg/f) ranged from 305,001 (pg/f) in PI296341/CS to 378,247 (pg/f) in Argentario/CS for the 2016 growing season, and 182,753 (pg/f) in control (Crimson sweet) to 224,355 (pg/f) in NUN-9075/CS for the 2017 growing season.

There was no significant difference observed between graft combinations and control for both 2016 and 2017 in pollen viability, pollen germination, and normal pollen production. The percentage of pollen viability, percentage of pollen germination and percentage of normal pollen production ranged between 98.3% (PI296341/CS) and 99.7% (control), 90.8% (control) and 96.2% (NU-9075/CS), and 96.9% (control) and 99.2% (Argentario/CS) for the 2016 growing season. For the 2017 growing season, pollen viability percentage was between 98.4% (control) and 100% in all graft combinations. Pollen germination percentage was observed to be between 88.2% (control) and 97.2% (NUN-9075/CS), and normal pollen production ranged between 90.4% (Argentario/CS) and 99.4% (NUN-9075/CS) (Fig. 2).
Fig. 2

Effect of rootstocks on a pollen viability, b pollen germination, and c normal pollen development in watermelon. Data were analyzed at 5% level of significance and are the average of four replications. No letters on bars means that there is no significant difference between graft combinations

Generally, pollen production of the second year was lower compared to that of the first year. This may be explained by few and very small flowers on the plants caused by the heavy rainfall for two consecutive days before the day of the pollen experiment. In the first year, flowers for pollen experiments were collected on June 6, 2016; in 2017, flowers were collected on the same date to avoid a difference between the growing seasons and graft combinations in pollen development experiments.

There was no significant difference between graft combinations for all parameters in the pollen development experiment. These results are in line with those reported by Stanghellini and Schultheis (2005). However, the results obtained in this study are much higher, especially in pollen grain production per flower, reaching a maximum number of 18,125 pollen/flower. Pollen viability results of this study are equivalent to those of Sain and Joshi (2003), who reported a pollen viability range between 18.5% and 84.3% in Citrullus lanatus × C. colocynthis hybrids. McGregor and Waters (2013) also obtained similar results to this study, with pollen viability ranging between 92.9% and 96.9% in wild relatives of watermelon.

3.5 Fruit yield and quality

Results of fruit yield and fruit quality are presented in Table 4 for the 2016 and 2017 growing seasons. Total fruit yield (kg m−2), single fruit weight (kg), fruit height (cm), and fruit diameter (cm) were significantly influenced by rootstocks, while no significant difference was observed in fruit rind thickness (mm) and total soluble solids (TSS %). Grafted plants had higher values in all assessed parameters of fruit yield and quality compared to control. During the 2016 growing season, the NUN-9075/CS graft combination resulted in higher fruit yield of 4.82 kg m−2 followed by Argentario/CS, which had 3.12 kg m−2, while the lowest total fruit yield was obtained in control (1.09 kg m−2). Single fruit weight ranged from 1.58 kg (control) to 8.35 kg (NUN-9075/CS), the highest fruit height average value was found in NUN-9075/CS (26.43 cm), and the lowest value was obtained in control (14.53 cm). The average fruit diameter value was 24.08 cm obtained in NUN-9075/CS, and the lowest fruit diameter value was obtained in control (Crimson sweet). The second year (2017) had a higher yield compared to the first year (2016). There was a significant difference at 5% level among graft combinations and control in total fruit yield, single fruit weight, fruit height, fruit diameter, and TSS, and no significant difference between graft combinations observed in fruit rind thickness (Prob > f = 0.76). The highest yield was 6.39 kg m−2 obtained in NUN-9075/CS, and the lowest yield was 1.49 kg m−2 obtained in nongrafted plants (Crimson sweet). Results obtained in this study showing that grafted watermelons resulted in higher yield and quality are similar to results obtained by Turhan et al. (2012), Ozmen et al. (2015), and Marsic et al. (2016). Colla et al. (2006) reported that the cause of the lowest yield in nongrafted watermelon plants (Citrullus lanatus L. Tex) is the reduction in fruit mean mass and number of fruits per plant. Several studies have reported a negative effect of grafting in TSS, especially the decrease in TSS content (Lopez-Galarza et al. 2004; Alexopoulos et al. 2007; Bie et al. 2010; Turhan et al. 2012), while others reported no difference in TSS between grafted and nongrafted plants (Burton et al. 2009). In contrast, this study revealed a positive effect of grafting on TSS content in both years, in which the average TSS content of 13% was recorded in many fruits.
Table 4

Total fruit yield (TFY), single fruit weight (SFW), fruit height (FH), fruit diameter (FD), fruit rind thickness (FRT), and total soluble solids (TSS) of the fruits from graft combinations and control for the 2016 and 2017 growing seasons

Graft combination

TFY (kg m−2)

SFW (kg)

FH (cm)

FD (cm)

FRT (mm)

TSS (%)

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

2016

2017

Control

1.09 b*

1.49 b

1.58 c

4.77 b

14.53 b

21.49 b

13.43 b

20.17 c

7.80

14.31

8.33

11.49 ab

PI 296341/CS

2.51 b

4.01 ab

4.73 bc

4.98 ab

21.36 a

21.43 b

19.53 a

20.80 bc

9.44

13.75

10.05

12.50 a

Argentario/CS

3.12 ab

4.92 a

6.52 ab

7.36 a

22.79 a

25.03 a

22.21 a

23.20 a

10.04

14.96

9.24

12.98 a

NUN-9075/CS

4.82 a

6.39 a

8.35 a

7.12 ab

26.43 a

24.70 a

24.08 a

22.78 ab

11.30

14.88

9.85

8.49 b

Prob > f

0.0028

0.0019

0.001

0.0156

0.0007

0.0201

0.0008

0.0428

0.0911

0.7554

0.101

0.0107

LSD (0.05)

2.11

2.68

3.42

2470.02

5.62

3.71

5.29

3.22

ns

ns

ns

3.39

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

3.6 Seed yield and quality

There was a significant difference between rootstocks in seed yield (g m−2), while no significant difference between rootstocks was observed in number of seeds per fruit and weight of 1000 seeds at 5% significant level for 2016 and 2017 (Table 5). Grafting resulted in higher seed yield compared to nongrafted plants. The highest average seed yield was 16.68 g m−2 obtained in the NUN-9075/CS graft combination, and the lowest seed yield average value was 4.08 g m−2 obtained in Crimson sweet (control) for the first growing season (2016). In the 2017 growing season, the highest average value of seed yield was 30.12 g m−2 observed in NUN-9075/CS, and the lowest seed yield was 7.46 g m−2 obtained in Crimson sweet (control).
Table 5

Seed yield, number of seeds per fruit, and weight of 1000 seeds for the 2016 growing season. Data are the average of four replications

Graft combination

Seed yield (g m−2)

Number of seeds per fruit

Weight of 1000 seed (g)

2016

2017

2016

2017

2016

2017

Control

4.08 c*

7.46 b

241.25 b

483.00 b

34.93

32.42 a

PI296341/CS

8.75 bc

14.48 b

477.83 a

607.00 b

36.14

28.15 b

Argentario/CS

11.27 ab

27.26 a

529.44 a

607.50 b

40.74

32.52 a

NUN-9075/CS

6.68 a

30.12 a

558.11 a

805.00 a

36.46

31.10 a

Prob > f

0.0004

0.0008

0.0021

0.005

0.2886

0.004

LSD (0.05)

5.49

12.20

189.19

197.32

ns

2.95

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

NUN-9075/CS had a larger number of seeds per fruit (558.11 seeds per fruit in the first season and 805.00 seeds per fruit in the second season), and a smaller number of seeds was obtained in Crimson sweet (241.25 and 483.00 seeds per fruit in 2016 and 2017, respectively). As observed in fruit yield and other parameters recorded in this study, the 2017 growing season also had a higher seed yield and number of seeds. The increased seed yield in grafted plants may be explained by the increased plant biomass and number and size of fruits, which are all caused by the effect of a stronger and deeper root system that can absorb more water and nutrients. Nerson and Paris (2002) explained that the increased seed yield is determined by many components, but the main one is the number of fruit per unit area. Seed yield also is related to the size of the seed, fruit-set ratio, plant density, and fruit shape and quality.

3.7 Determination of seed quality

There was no significant difference between graft combinations on seed moisture content (SMC, %) for both two growing seasons. In the 2016 growing season, the seed moisture content ranged between 7.0% and 7.8% (control and NUN-9075/CS) and between 6.7 and 7.1% for the 2017 growing season obtained in PI 296341/CS and control, respectively. Results for seed quality analysis are presented in Tables 6 and 7.
Table 6

Seed analysis results for seed moisture content (SMC), seed emergence percentage (SE), days of seed emergence (SED), seed germination percentage (SGP), days of seed germination (SGD), and accelerated aging (AA) at different times for the 2016 growing season

Graft combination

SMC (%)

SE (%)

SED (days)

SGP (%)

SGD (days)

AA 48 h (%)

AA 96 h (%)

AA 144 h (%)

AA 192 h (%)

Control

7.0

30.5 b*

10.8 a

39.3 b

7.3 a

50.0

50.0

72.0 a

40.0

PI296341/CS

7.8

61.0 a

8.9 a

68.5 a

3.2 a

63.0

53.0

57.3 b

48.0

Argentario/CS

7.6

86.5 a

7.8 b

97.0 a

2.6 b

96.0

87.0

80.0 a

75.0

NUN-9075/CS

7.8

91.0 a

7.1 b

97.5 a

2.6 b

97.0

98.0

97.0 a

84.0

Prob > f

0.100

0.012

0.003

0.02

0.02

0.17

0.17

0.03

0.19

LSD (0.05)

ns

52.14

2.185

53.03

4.4

ns

ns

33.65

ns

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

Table 7

Seed analysis results for seed moisture content (SMC), seed emergence percentage (SE), days of seed emergence (SED), seed germination percentage (SGP), days of seed germination (SGD), and accelerated aging (AA) at different times for the 2017 growing season

Graft combination

SMC (%)

SE (%)

SED (days)

SGP (%)

SGD (days)

AA 48 h (%)

AA 96 h (%)

AA 144 h (%)

AA 192 h (%)

Control

7.1

89.0

7.1

86.0

10.2 a

95.0

86.0

95.0

86.0

PI296341/CS

6.7

90.0

7.4

92.5

7.9 a

75.0

76.0

72.0

48.0

Argentario/CS

6.9

82.5

7.6

70.5

8.5 a

89.0

92.0

86.0

93.0

NUN-9075/CS

7.0

75.5

7.8

71.5

7.1 b

92.0

90.0

83.0

73.0

Prob > f

0.36

0.08

0.59

0.37

0.04

0.33

0.46

0.54

0.09

LSD (0.05)

ns

ns

ns

ns

2.83

ns

ns

ns

ns

ns no significant difference

*Different letters indicate statistical significance using Tukey’s multiple range test at p > 0.05

In the 2016 growing season, there was a significant difference between graft combinations in seed emergence percentage, seed emergence rate (days), seed germination percentage, and seed germination rate (days). NUN-9075/CS had the highest average value in seed emergence percentage (91.0%) and seed germination percentage (97.5%) as well as fewer average days in seed emergence rate and germination rate (7.1 and 2.6 days, respectively) (Table 6).

However, for seed germination rate, the NUN-9075/CS graft combination and Argentario/CS statistically attained the same rate (2.6 and 2.6 days, respectively). Nongrafted plants (control) had the lowest average seed emergence percentage, germination percentage, seed emergence, and germination rate with averages of 30.5%, 39.3%, 10.8 days, and 7.3 days, respectively. Generally, grafting resulted in a more than 200% increase in seed emergence and germination, as well as 3 days earlier seed emergence and 4 days earlier seed germination compared to nongrafted plants. A significant difference between rootstocks in accelerated aging was found after 144 h (6 days) of seed storage at 45 °C and 100% humidity. The average germination percentage ranged between 57.3% and 97.0%, with the highest germination percentage obtained in NUN-9075/CS and the lowest percentage obtained in PI96341/CS. Although no significant difference between rootstocks was observed after 48 h, 96 h and 192 h of seed storage at 45 °C and 100% humidity, yet the grafted plants attained higher germination percentage values compared to non-grafted. The percentages ranged between 50.0 and 97.0% after 48 h, 50.0–98.0% after 96 h and 40.0–84.0% after 192 h. In all days, the highest percentage value was observed in NUN-9075/CS and lowest percentage value was obtained in Crimson sweet (control). The ability of seeds to germinated after storage at 45 °C and 100% humidity was almost the same until at 192 h of seed storage, thereafter, a minimum decrease of germination percentage was observed. In every genotype at the 8th day there was an approximately 10% decrease in germination.

In the second year of cultivation, a significant difference between rootstocks at 5% level was found in seed germination rate, with NUN-9075/CS germinating after an average of 7.1 days and Crimson sweet exhibiting late germination (10.2 days). No significant difference was observed between rootstocks and control in seed moisture content, seed emergence percentage, seed emergence rate, germination percentage, and accelerated aging at all storage time (Table 7).

Seed germination and emergence tests were conducted as a means to determine the seed quality, which can give insight into the seed performance during storage and or in the field (Gupta 1993). This study showed a higher percentage of seed germination and seedling vigor, which has been tested through seed emergence and accelerated aging. Seed emergence results of the grafted plants ranged between 61.0% and 91.0% and 75.5% and 90.0% for the 2016 and 2017 growing seasons, respectively, and seed germination of the grafted plants ranged between 68.5% and 97.5% and 70.5% and 92.50% for the 2016 and 2017 growing seasons, respectively. The average accelerated aging of the grafted plants ranged between 50.0% at the lowest to 98.0% at the highest, with the overall average range between the rootstocks between 80.0% and 90.0%. Decrease in germination after storage for the accelerated aging may be due to the fungal infection of the seeds caused by frequent opening of the incubator and the containers. However, the decrease in germination percentage was not that much lower compared to that reported by Demir and Mavi (2008), who indicated the decrease of germination to be up to 48%. These results are similar to those reported by Jaskani et al. (2005), Barbosa et al. (2016), and Soares et al. (2016). Singh et al. (2001) reported lower results compared to the results obtained in this research. Results of this study on seed germination and emergence are higher than those reported by Mavi and Atak (2016). Baalbaki et al. (2009) concluded that seeds that have been exposed to high temperature between 40 and 45 °C and 100% relative humidity and are able to produce high germination after tolerating this aging treatment have a greater possibility of producing a high percentage of normal seedlings. Marcos-Filho (2015) suggested the high performance of seeds in germination and vigor as an important initiative towards successful crop production and further indicated that germination and vigor are important factors influencing seed physiological potential, which governs the capacity of seeds under both favorable and unfavorable environmental conditions. Therefore, seeds of grafted plants tested in this study have the potential for better storage life, germinability, and a capacity to express their vital functions under any environmental conditions.

4 Conclusion

According to the results of this research, we conclude that seed yield and quality greatly relate to plant size and vigor because the best performing rootstocks (NUN-9075 and Argentario) showed the highest yield and quality and highest values of stem length, stem diameter, number of leaves, and plant biomass. The NUN-9075 rootstock increases the vigor of the plant and always results in robust plant growth and many fruits. However, at the seedling stage, there is a delay in growth, which may be caused by the union effect after grafting, but once well established, plants grow very fast. PI 296341 rootstock has potential in some traits such as high total soluble content and higher average number of seeds per fruit, which is one of the factors that contributes to seed yield and strong rootstock. Number of seeds depends on pollen quantity and viability, and in this research, plants were left to self-pollinate, but the efficiency of pollination may be questionable. Based on the results of this study, NUN-9075 rootstock performed better than other rootstocks and is therefore recommended as the best one. Considering the increased production of grafted watermelons in greenhouses, this research gives another opportunity to investigate the effects of rootstocks on seed yield and quality by using honey bees and/or by hand-pollination in both the open field and greenhouses.

Notes

Acknowledgements

This study was supported by Cukurova University Scientific Research Project Unit (Grant No. FDK-2016-6465) (Bilimsel Araştırma Projesi—BAP). We also thank Antalya seedling Company for providing seeds and seedlings production. Indebted thanks to Ph.D. and M.Sc. students who helped during field and laboratory work.

References

  1. Alexopoulos AA, Kondylis A, Passam H (2007) Fruit yield and quality of watermelon in relation to grafting. J Food Agric Environ 5:178–189.  https://doi.org/10.3923/ja.2007.362.365 CrossRefGoogle Scholar
  2. Ambika S, Manonmani V, Somasundaram G (2014) Review on effect of seed size on seedling vigour and seed yield. Res J Seed Sci 7:31–38.  https://doi.org/10.3923/rjss.2014.31.38 CrossRefGoogle Scholar
  3. Andrews PK, Marquez CS (1993) Graft incompatibility. Hort Rev Am Soc Hort Sci 15:183–232Google Scholar
  4. Baalbaki R, Elias S, Marcos-Filho J, McDonald MB (2009) Seed vigor testing handbook. AOSA Ithaca NY USA (contribution to the handbook on seed testing, 32)Google Scholar
  5. Barbosa WFS, Steiner F, de Oliveira LCM, Henrique P, das Chagas M (2016) Comparison of seed priming techniques with regards to germination and growth of watermelon seedlings in laboratory condition. Afr J Biotechnol 15:2596–2602CrossRefGoogle Scholar
  6. Bie Z, Han X, Zhu J, Tang M, Huang Y (2010) Effect of nine squash rootstocks on the plant growth and fruit quality of melon. Acta Hortic 856:77–82Google Scholar
  7. Bosgnin DA (2002) Origin and evolution of cultivated cucurbits. Cienc Rural 32:715–723CrossRefGoogle Scholar
  8. Burton BD, Fish WW, Roberts W, Popham TW (2009) The influence of rootstock selection on fruit quality attributes of watermelon. Open Food Sci J 3:5–34Google Scholar
  9. Colla G, Rouphael Y, Cardarelli M, Rea M (2006) Effect of salinity on yield, fruit quality, leaf gas exchange, and mineral composition of grafted watermelon plants. HortScience 41:622–627CrossRefGoogle Scholar
  10. Demir I, Mavi K (2008) Effect of salt and osmotic stresses on the germination of pepper seeds of different maturation stages. Braz Arch Biol Technol 51:897CrossRefGoogle Scholar
  11. Djidonou D, Gao Z, Zhao X (2013) Economic analysis of grafted tomato production in sandy soils in northern Florida. Horttechnology 23:613–621CrossRefGoogle Scholar
  12. Elazar F, Zoran I (2014) Grafted vegetables—the influence of rootstock and scion on post harvest quality. Folia Hortic 26:79–90CrossRefGoogle Scholar
  13. Eti S, Stosser R (1988) Fertility of the mandarin variety clementine citrus–reticulata blanco I. Pollen quality and pollen tube growth. Hortic Sci 53:160–166Google Scholar
  14. Gómez HG, Godina FR, Ortiz HO, Mendoza AB, Torres VR, De la Fuente MC (2017) Use of chitosan-PVA hydrogels with copper nanoparticles to improve the growth of grafted watermelon. Molecules 22:1031.  https://doi.org/10.3390/molecules22071031 CrossRefGoogle Scholar
  15. Gunn CA, Weber JL, McGill AT, Kruger MC (2015) Increased intake of selected vegetables, herbs and fruit may reduce bone turnover in post-menopausal women. J Nutr 7:2499–2517Google Scholar
  16. Gupta PC (1993) Seed vigour testing. In: Agrawal PK (ed) Handbook of seed testing. DAC, Ministry of Agriculture, Government of India, New Delhi, pp 242–249Google Scholar
  17. Jaskani MJ, Kwon SW, Kim DH (2005) Comparative study on vegetative, reproductive and qualitative traits of seven diploid and tetraploid watermelon lines. Euphytica 145:259–268CrossRefGoogle Scholar
  18. King SR, Davis AR, Zhang XP, Crosby K (2010) Genetics, breeding and selection of rootstocks for Solanaceae and Cucurbitaceae. Sci Hortic 127:106–111CrossRefGoogle Scholar
  19. Kong QS, Chen JL, Liu Y, Ma YH, Liu P, Wu SY, Bie ZL (2014) Genetic diversity of Cucurbita rootstock germplasm as assessed using simple sequence repeat markers. Sci Hortic 175:150–155CrossRefGoogle Scholar
  20. Lima MS, Cardoso AII, Verdial MF (2003) Plant spacing and pollen quantity on yield and quality of squash seed. Hortic Bras 21:443–447CrossRefGoogle Scholar
  21. Liu B, Ren J, Zhang Y, An J, Chen M, Chen H, Xu C, Ren H (2015) A new grafted rootstock against root-knot nematode for cucumber, melon, and watermelon. Agron Sustain Dev 35:251–259CrossRefGoogle Scholar
  22. Lopez-Galarza S, San Bautista A, Perez DM (2004) Effects of grafting and cytokinin-induced fruit setting on color and sugar-content traits in glasshouse-grown triploid watermelon. J Hortic Sci Biotechnol 79:971–976CrossRefGoogle Scholar
  23. Marcos-Filho J (2015) Seed vigor testing: an overview of the past, present and future perspective. Sci Agric 72:363–374CrossRefGoogle Scholar
  24. Marsic KN, Znidarcic D, Jakse M (2016) Different scion/rootstock combinations influence the yield and quality of grafted watermelon fruits. Acta Hortic 1142:19–24.  https://doi.org/10.17660/actahortic.2016.1142.3 CrossRefGoogle Scholar
  25. Mavi K, Atak M (2016) Effect of organic priming on seedling emergence of watermelon under low temperature stress. In: Proceedings of the 7th international scientific agriculture symposium, “Agrosym 2016”. Jahorina, pp 1727–1732Google Scholar
  26. McGregor CE, Waters V (2013) Pollen viability of F1 hybrids between watermelon cultivars and disease-resistant, infraspecific crop wild relatives. HortScience 48:1428–1432CrossRefGoogle Scholar
  27. Meeuse AD (1962) The Cucurbitaceae of Southern Africa. Bothalia 8:1–111CrossRefGoogle Scholar
  28. Mohamed FH, Abd El-Hamed KE, Elwan MWM, Hussien MNE (2014) Evaluation of different grafting methods and rootstocks in watermelon grown in Egypt. Sci Hortic 168:145–150CrossRefGoogle Scholar
  29. Naz A, Butt MS, Sultan MT, Qayyum MMN, Niaz RS (2014) Watermelon lycopene and allied health claims. EXCLI J 13:650–660PubMedPubMedCentralGoogle Scholar
  30. Nerson H (2004) Fruit-set order affects seed yield and germinability in melon (Cucumis melo L.). J Hortic Sci Biotechnol 79:985–990CrossRefGoogle Scholar
  31. Nerson H (2005) Effects of fruit shape and plant density on seed yield and quality of squash. Sci Hortic 105:293–304CrossRefGoogle Scholar
  32. Nerson H, Paris HS (2002) Relationship between fruit shape and seed yield in Cucurbita pepo. Rep Cucurbit Genet Cooperative 24:82–86Google Scholar
  33. Norton JD (1966) Testing of plum pollen viability with tetrazolium salts. Proc Am Soc Hortic Sci 89:132–134Google Scholar
  34. Olaniyi JO, Tella BA (2011) Effects of nitrogen and potassium fertilizers on the growth, seed yield and nutritional values of egusi melon (Citrullus lanatus (Thumb) Mans.) in Ogbomoso South west Nigeria. Int Res J Plant Sci 2:328–331Google Scholar
  35. Ozmen S, Kanber R, Sari N, Unlu M (2015) The effects of deficit irrigation on nitrogen consumption, yield, and quality in drip irrigated grafted and ungrafted watermelon. J Integr Agric 14:966–976CrossRefGoogle Scholar
  36. Paris H (2017) Origin of the dessert watermelon, Citrullus lanatus. Acta Hortic 1151:87–94.  https://doi.org/10.17660/ActaHortic.2017.1151.15 CrossRefGoogle Scholar
  37. Ren Y, McGregor C, Zhang Y, Gong G, Zhang H, Guo S, Sun H, Cai W, Zhang J, Xu Y (2014) An integrated genetic map based on four mapping populations and quantitative trait loci associated with economically important traits in watermelon (Citrullus lanatus). BMC Plant Biol 14:33CrossRefPubMedPubMedCentralGoogle Scholar
  38. Sain RS, Joshi P (2003) Pollen fertility of interspecific F1 hybrids in genus Citrullus (Cucurbitaceae). Curr Sci 85:431–434Google Scholar
  39. Singh S, Singh P, Sanders DC, Wehner TC (2001) Germination of watermelon seeds at low temperature. Rep Cucurbit Genet Cooperative 24:59–64Google Scholar
  40. Soares VN, Rodrigues HCS, Gadotti GI, Meneghello GE, Villela FA (2016) Influence of fungi associated with watermelon seeds on physiological and health quality. Aust J Crop Sci 10:852–856CrossRefGoogle Scholar
  41. Souza RHV, Villela FA, Aumonde TZ (2013) Methodologies based on seedling performance for vigor assessment of pumpkin seed. J Seed Sci 35:374–380CrossRefGoogle Scholar
  42. Stanghellini MS, Schultheis JR (2005) Genotypic variability in staminate flower and pollen grain production of diploid watermelons. HortScience 40:752–755CrossRefGoogle Scholar
  43. Turhan A, Ozmen N, Kuscu H, Serbeci MS, Seniz V (2012) Influence of rootstocks on yield and fruit characteristics and quality of watermelon. Hortic Environ Biotechnol 53:336–341CrossRefGoogle Scholar
  44. Wang X, Shang L, Luan F (2013) A highly efficient regeneration system for watermelon (Citrullus Lanatus Thunb.). Pak J Bot 45:145–150Google Scholar
  45. Yetisir H, Sari N (2004) Effect of hypocotyls morphology on survival rate and growth of watermelon seedlings grafted on rootstocks with different emergence performance at various temperatures. Turk J Agric For 28:231–237Google Scholar
  46. Yetisir H, Caliskan ME, Soylu S, Sakar M (2006) Some physiological and growth responses of watermelon [Citrullus lanatus (Thumb.) Matsum. and Nakai] grafted onto Lagenaria siceraria to flooding. Environ Exp Bot 58:1–8CrossRefGoogle Scholar

Copyright information

© The Author(s) 2019

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors and Affiliations

  1. 1.Zanzibar Agricultural Research InstituteZanzibarTanzania
  2. 2.Faculty of Agriculture, Department of HorticultureCukurova UniversityAdanaTurkey

Personalised recommendations