Journal of Applied Botany and Food Quality 89, 191 - 201 (2016), DOI:10.5073/JABFQ.2016.089.024 1Department of Horticulture, Mustafa Kemal University, Antakya, Hatay, Turkey 2 Department of Horticulture, Erciyes University, Melikgazi, Kayseri, Turkey 3Alata Horticultural Research Institute, Erdemli, Mersin, Turkey 4Agriculture County Directorate, Silifke, Mersin, Turkey Effects of rootstocks on storage and shelf life of grafted watermelons Ahmet Erhan Özdemir1, Elif Çandır1*, Halit Yetişir2, Veysel Aras3, Ömer Arslan4, Özay Baltaer1, Durmuş Üstün1, Mustafa Ünlü2 (Received December 17, 2015) * Corresponding author Summary Watermelon fruits from non-grafted or grafted ‘Crimson Tide’ (CT) and ‘Crisby’ (CR) onto Ferro, RS841, Argentario and Macis rootstocks were compared for their postharvest quality during storage at 7 °C for 21 days and additional 7 days at 21 °C. Non-grafted and grafted CT and CR fruits did not exhibit chilling injury (CI) symptoms, but the 1-2 % of fungal decay occurred after shelf life period following storage. Watermelons grafted on Ferro and RS841 rootstocks had higher flesh firmness thicker rind, lower ripening rating, more intense (higher C*) brighter red (lower h° value) color and higher lycopene content after shelf life period following storage, compared to non- grafted fruits. All of the fruit tested by the panelists received high taste scores of >7.9 out of 8.5 at the beginning, but the scores decreased to >6.8 out of 7.7 at the end of shelf life period. Watermelons could successfully be kept for 21 days at 7 °C and additional 7 days at 21 °C. Watermelons grafted on Ferro and RS841 rootstocks had higher postharvest quality, compared to the non-grafted fruits for both cultivars. Introduction Turkey is an important vegetable producing country with total of 27 billion metric tons vegetable production (FAOSTAT, 2014). The Cucurbitaceae family comprises 28 % of total vegetable production of Turkey. Watermelon accounts for 14 % of total vegetable produc- tion. Turkey is the second largest watermelon-producing country in the world after China, with production of 4 million tons per year. Recently, grafted watermelon production has become widespread in Turkey, because soil borne diseases are severe during early cultivation under plastic tunnels or later in the season in field production due to continuous and intensive cropping (KURT et al., 2002). Soil borne diseases cause a decrease in yield and quality. There are different ways to prevent soil-borne diseases such as crop rotation, breeding programs, soil fumigant (YETISIR and SARI, 2004). However, the use of methyl bromide has been banned due to its effect on depletion of ozone layer. The use of seedlings grafted on Cucurbita and Lagenaria rootstocks, which have an acquired resistance to soil borne diseases, was suggested by several researchers as an environmentally safe alternative to methyl bromide (MIGUEL et al., 2004). Owing to environmental restrictions imposed on the use of chlorofluorocarbon-based soil fumigants, use of rootstocks resistant to soil borne pathogens has become an established practice in regions growing watermelon (DAVIS and PERKINS-VEAZIE, 2005). The most common rootstocks for watermelon are bottle gourd, interspecific hybrids between C. maxima and C. moschata, and wild watermelon (Citrullus lanatus var. citroides) (DAVIS et al., 2008). Grafting of watermelon scions on squash, pumpkin, or bottle gourd rootstocks is practiced in all the major watermelon production regions of the world (LEE, 1994; 2003). These rootstocks influenced resistance to soil borne diseases, plant growth, yield, and fruit quality. Graft in- compatibility and decrease in the fruit quality appeared depending on the scion-rootstock combination (LEE and ODA, 2003). Although effects of rootstocks on yield have been reported, the information on fruit quality has been incidental and does not account for all correlative aspects of quality. Abnormal fruit quality has been reported due to grafting for watermelon (YAMASAKI et al., 1994; LEE and ODA, 2003; LIU et al., 2006; ALEXOPOULOS et al., 2007). However, there are other reports of positive effects of grafting on watermelon fruit quality (YETISIR and SARI, 2003; DAVIS and PERKINS-VEAZIE, 2005; KARACA et al., 2012; ÇANDIR et al., 2013). The effect of the rootstocks on plant growth, fruit yield and quality of watermelon cv. Crispy grafted onto TZ-148 and RS841 of commercial hybrid of Cucurbita maxima × Cucurbita moschata and 64-18 of experimental bottle gourd rootstocks were studied. Grafting resulted in higher yield by increasing in both fruit number and weight, however, no detrimental effect on fruit quality such as fruit index, rind thickness, and soluble solid contents on grafted plants was observed (ALAN et al., 2007). Grafting on the local bottle gourd rootstocks improved plant growth parameters, yield (KARACA et al., 2012) and the total soluble solid (TSS), titratable acidity (TA), TSS/TA ratio, sugar, organic acid, and carotenoid (β-carotene and lycopene) contents of Crimson Tide fruits (ÇANDIR et al., 2013). For short-term storage or transit to distant markets (>7 days), most recommendations use 7.2 °C (45 °F) and 85-90 % R.H. as the ac- ceptable handling conditions for watermelons (SUSLOW, 1997). Watermelons are, however, prone to chilling injury at this temperature and extended holding at this temperature will induce chilling injury, rapidly evident after transfer to typical retail display temperatures (SUSLOW, 1997). Short-term storage at near ambient temperatures is the common practice for watermelon (CHISHOLM and PICHA, 1986; PERKINS-VEAZIE and COLLINS, 2006; RADULOVIĆ et al., 2007). Watermelons in most of the world are customarily handled posthar- vest under non-refrigerated conditions. Shelf life for watermelon is 2-3 weeks at 10-15 °C depending on cultivar (RUSHING et al., 2001). There are few reports on the effects of grafting on storage and shelf life of watermelons. Effects of grafting on postharvest quality of watermelon are not completely known and cause some speculation. Previous studies have shown that grafting increased in flesh firmness of watermelon fruits in most scion-rootstock combinations (SALAM et al., 2002; YETISIR et al., 2003; DAVIS and PERKINS-VEAZIE, 2005; ROBERTS et al., 2007; CUSHMAN and HUAN, 2008; BRUTON et al., 2009; SOTERIOU and KYRIACOU, 2015). Firmer fruits are more likely to retain a desirable consistency and are expected to have a better shelf life than are softer fruit (ROBERTS et al., 2007; KING et al., 2010). Therefore, effects of grafting on storage and shelf life performances of watermelon fruits have gained importance. The objective of this study was to determine quality changes of watermelon fruits cv. ‘Crimson Tide’ (CT) and ‘Crisby’ (CR) grafted onto Ferro, RS841, Argentario, and Macis rootstocks during storage at 7 °C for 21 days and additional 7 days at 21 °C and compare graf- 192 A.E. Özdemir, E. Çandır, H. Yetişir, V. Aras, Ö. Arslan, Ö. Baltaer, D. Üstün, M. Ünlü ted and non-grafted ‘Crimson Tide’ and ‘Crisby’ fruits for posthar- vest quality. Materials and methods The experiment was conducted at the Alata Horticultural Research Institute, Erdemli, Mersin, Turkey. ‘Crimson Tide’ (CT) and Crisby (CR) watermelon [Citrullus lanatus (Thunb.) Matsum. and Nakai] cultivars were grafted onto Ferro and RS841 (C. maxima × C. moschata) and Argentario and Macis (Lagenaria siceraria) rootstocks by using slunt cut grafting method (LEE and ODA, 2003). The grafted plants were supplied by the commercial seedling company of Grow Fide (Antalya, Turkey). The non-grafted CT and CR were used as control. Fruits were harvested at full maturity when the 75 % of tendril and stipule on the same node with peduncle were desiccated. After har- vest, fruits were stored at 7±0.5 ºC and 90±5 % relative humidity for 21 days in cold store and hold 21 days at 7 °C and subsequent 7 days at 21±0.5 ºC and 70±5 % relative humidity for shelf life. Changes in fruit weight (g), diameter (mm) and length (mm), rind thickness (mm), weight loss (%), fruit flesh firmness (N), taste (1-9), total sol- uble solids (%), juice pH, titratable acidity (%), chilling injury and fungal decay (1-5), flesh color (L* C* hº) values, hallow heart (1-5), ripening (1-7), citric and malic acid (%), glucose (%), fructose (%), sucrose (%), total sugar (%), β-carotene (μg g-1), lycopene (μg g-1) were determined after shelf life period following storage at a weekly interval. Fruit weight (g); 30 fruits were weighted by a laboratory balance with an accuracy of 0.01 g for grafted and non-grafted fruits of both culti- vars; it was calculated by taking the arithmetic mean. Diameter (mm) and length (mm); 5 fruits in each replicate were measured with a ruler. Rind thickness was measured at two points on each fruit cross-section using an electronic caliper. Weight loss (%); 30 fruits were numbered and they were weighted by a laboratory balance with an accuracy of 0.01 g for grafted and non-grafted fruits of both cultivars, the loss was calculated by subtracting the final weigh from the initial weigh in percent. Fruit flesh firmness (N); Flesh firmness of the middle of each fruit was determined with a penetrometer (Now FHR-5 Nippon Opti- cal Works Co. Ltd. Tokyo, Japan). This involved measuring the force in kilograms required for a 12-mm conical probe to penetrate the cut surface to a depth of 5 mm at 3 locations in the mesocarp tissue; the results were then converted to newton (N). As described by MOTSEN- BOCKER and PICHA (1996), a 2.5-cm cross-sectional slice was taken from the middle of each fruit. The flesh of each slice was carefully re- moved from the outer rind and quartered. Two opposite quarter pieces were combined and homogenized for one minute in a Waring Blender and filtered under suction through Whatman #4 paper. Portions of the homogenate were used to determine total soluble solid (TSS) content, juice pH, titratable acidity (TA). TSS content was measured using a digital refractometer (Atago Model ATC-1E Atago Co. Ltd., To- kyo, Japan) and juice pH with digital pH-meter (Orion 5-Star model Thermo Fisher Scientific Inc., MA, ABD) at 20 °C. TA was mea- sured by potentiometric method. The 5 ml of juice sample was di- luted with distilled water to 100 ml and this sample was titrated with 0.1 N NaOH until the pH value of 8.1 was read on digital pH-meter. The results were calculated as “g malic acid 100 ml-1 juice” in per- cent. The fruits were also scored at each evaluation for chilling injury (CI) and decay (1 = none, 2 = <10 % of surface area, 3 = 11 % to 25 %, 4 = 26 % to 50 %, and 5 = >50 %) (RISSE et al., 1990). Incidence of CI and decay were determined after 7 days at 21 °C following each sto-rage period. For sensory evaluation, two opposite quarter pieces from the middle of each fruit were prepared as described above. Ten trained panelists (non-smoker 7 male and 3 female, ages 20 to 45) evaluated taste (1-9) of fruits on hedonic scale of 1= disliked (lowest value) to 9 = liked (the best), hallow heart (1-5) of fruits on a scale of 1= none to 5 = very severe (50 % “more than hallow heart) and ripen- ing (1-7) of fruits on a scale of 1= raw fruit and 3 = mature to 7= over- ripe extremely. Fruit flesh color was measured using the CIELAB (L*a*b*) color space by a CR-300 Minolta Chroma Meter (Konica Minolta, Osaka, Japan), calibrated using the manufacturer’s standard white plate. Two readings were performed from the flesh of verti- cally cut fruits. L* represents lightness, ranging from 0 (black) to 100 (white). Chroma (C*) represents color saturation, which varies from dull (low value) to vivid color (high value) and was calculated using the formula (a2 + b2)1/2. Hue angle (h°) represents a color wheel with red-purple at an angle of 0°, yellow at 90°, bluish green at 180°, and blue at 270°, and it was calculated by h° = tan−1 (b/a) (MCGUIRE, 1992). Sugars and organic acids were extracted of the method described by ÇANDIR et al. (2013). Briefly, frozen samples were homogenized using an Ultra-Turrax T25 model homogenizer (IKALabortechnik) at low speed with a 10-mm shaft. The resulting slurry was filtered through Whatman No. 4 paper with a Buchner funnel under vacuum. Exactly 1 mL of sample was diluted with deionized distilled water to a total volume of 10 mL. After vortexing for 1 min, 20 μL of sample was injected directly into the HPLC equipment after filtration through a Millex-HV 0.45 μm filter (Millipore). HPLC analysis of sugars and organic acids was performed on LC-10A equipment consisting of LC-10AD pumps, an in-line degasser, a CTO-10A column oven, an SCL-10A system controller, an SPD 10Avp, a photo diode array (PDA) detector, and a refractive index detector (RID), all operated by LC solution software (Shimadzu). Sugars were separated on an EC NUCLEOSIL Carbohydrate 250 mm × 4 mm i.d. column (Macherey- Nagel) at 25 °C. The mobile phase was acetonitrile and water (80:20, v/v) at a flow rate of 2 mL min–1. Organic acids were separated on a TransgenomicTM ICSep ION300 300 mm × 7.8 mm i.d. column (Transgenomic) at 65 °C. The mobile phase was 0.0085 N H2SO4 at a flow rate of 0.4 mL min–1. Sugars and organic acids were detected using the RID and PDA detectors at 210 nm, respectively. The quan- tification was performed according to external standard solution cali- brations. The results were expressed as g 100 g–1 fresh weight. Carotenoids were extracted following a modified version of the me- thod described by PERKINS-VEAZIE and COLLINS (2006). Brefly, fro- zen samples were homogenized using the Ultra-Turrax homogenizer at low speed with a 10-mm shaft. Three grams of puree were weighed into the centrifuge tube and extracted with HPLC-grade solvents of 10 mL of hexane, 5 mL of ethanol, and 5 mL of acetone contain- ing 0.05 % butylated hydroxytoluene (Merck KGaA). Samples were tightly sealed and placed on an orbital shaker (Heidolph Unimax 2010, Heidolph Instruments GmbH Co. KG) for 15 min at 320 rpm, and then 3 mL of deionized distilled water was added and samples were shaken again for 10 min. Afterwards, samples were put in a rack to allow solvent phase separation. The upper hexane layer was also filtered using a Millex-HV 0.45-μm filter (Millipore) and 20 μL of sample was injected directly into Shimadzu HPLC equipment (as pre- viously described). Carotenoids were separated on a YMC carotenoid column, C30 250 mm × 4.6 mm id, 5 μm particle size (YMC Europe GMBH), operating at 30 °C with a flow rate of 1.5 mL min–1. The mobile phase was solvent A (methanol, methyl tertiary butyl ether, and deionized distilled water, 81:15:41) and solvent B (methanol and methyl tertiary butyl ether, 10:90) with elution with a linear gradient of 0-16 min with 100 % A and 16-60 min with 100 % B (LIU et al., 2009). Detection was carried out at 503 nm for lycopene and 452 nm for β-carotene using the PDA detector. Components were identified by comparison of their retention times to those of authentic standards under analysis conditions and were quantified by external standard method and expressed as μg g–1 fresh weight. The study was performed over a 2-year period, in 2009 and 2010. Data are represented as the mean of 2 experimental years. In the wa- termelon production region where the experiment was conducted, Storage and shelf life of grafted watermelons 193 farmers start the season with early season cultivar ‘Crisby’ followed by mid-early/middle season cultivar ‘Crimson Tide’. Crisby has a mealy texture while Crimson Tide is crisp. Our objective in this ex- periment was to determine the commercial rootstock(s) with best sto- rage and shelf life performance for each cultivar, not to determine the best scion/rootstock combination(s) for watermelon producing area. Therefore, the experiment was conducted in completely randomized block design and the data were analyzed by one-way ANOVA using SAS software of SAS Institute, Cary, N.C. (SAS, 1999). The data were obtained from three replicates per scion/rootstock combination. Each replicates contained 5 fruits. Mean separation was performed by Fisher’s Least Significance Test at p<0.05 level. Results and discussion Watermelons grafted on RS841 and Ferro rootstocks had the higher fruit weight for CT and CR cultivars (Tab. 1). An increase in fruit weight due to grafting was reported in previous studies (YETISIR et al., 2003; MIGUEL et al., 2004; ALEXOPOULOS et al., 2007; ALAN et al., 2007; PRIOETTI et al., 2008). Grafting had no significant effect on fruit length and diameter for both cultivars (Tab. 1). Similarly, DAVIS et al. (2008) reported that fruit shape, determined by length, circumference, and their ratio, did not change significantly from wa- termelon fruit harvested from grafted and not grafted plants. Graf- ting affected rind thickness dependent on the scion. Consistent with these findings, SOTERIOU and KYRIACOU (2015) reported that varia- bility in rind thickness was derived mainly from the scion. At har- vest, CT fruits from grafted plants had thicker rind than those from non-grafted plants while grafting did not affect rind thickness for CR watermelon cultivar (Fig. 1). Rind thickness of ‘Crisby’ watermelon fruits grafted on TZ-148, RS-841 and 64-18 of Lagenaria rootstock was similar to non-grafted control fruits (ALAN et al., 2007). Inter- specific hybrid rootstocks (C. maxima × C. moschata) increased rind thickness of four watermelon cultivars (‘Celebration’, ‘Gallery’, ‘Pe- gasus’ and ‘Torpilla’) although their effect was very limited, indi- cating minimal rootstock effect on rind thickness compared with the effect of cultivar (KYRIACOU and SOTERIOU, 2015). ALEXOPOULOS et al. (2007) reported that ‘Crimson Sweet’ fruits from grafted plants on rootstocks (Long gourd, Early Max, Max-2 and F-14 gourd), had a thicker rind than the fruits from non-grafted plants. DAVIS et al. (2008) also found rind thickness increased for both seedless and seeded watermelon fruits when grafted to gourd rootstock ‘451’. In the study with ‘Crimson Tide’ watermelon fruits, all of the grafted plants on bottle gourd rootstocks produced fruit with a thicker rind than the control plants (KARACA et al., 2012). Rind thickness de- creased in non-grafted and at a lesser extent in grafted fruits during storage at 7 °C for 21 days and additional 7 days at 21 °C in both cultivars (Fig. 1). Fruits grafted on RS841, Argentario and Ferro root- stocks had thicker rind compared to non-grafted fruits after shelf life period following storage for both cultivars (Fig. 1). Similarly KYRIA- COU and SOTERIOU (2015) reported that postharvest storage at 25 °C caused thinning of the rind after 7 and 14 days and all hybrid root- stocks resulted in thicker rind than the non-grafted control. Thinning of the rind, known to characterize watermelon maturation (COREY and SCHLIMME, 1988), therefore may indicate an overripe fruit and one subjected to prolonged storage (KYRIACOU and SOTERIOU, 2015). Weight losses of grafted and non-grafted fruits were very low (<1 %) after shelf life period for 7 days at 21° following 21 days of storage at 7 °C for both cultivars. Effect of rootstocks on weight loss was not significant (Fig. 2). Consistent with our results, PERKINS-VEAZIE and COLLINS (2006) reported the <1 % of weight loss in watermelon fruits at all temperatures (5 °C, 13 °C and 21 °C) after 14 days of storage. However, NETO et al. (2000) determined higher weight loss (3.8 %) than our results. Non-grafted CT and CR or CT and CR grafted onto different root- stocks did not exhibit CI symptoms, but the 1-2 % of fungal decay occurred during shelf life period after 21 days of storage (Fig. 3). The decayed areas covered <10 % of rind surface of fruits for both cultivars. The graft combinations did not differ in the incidence of fungal decay for CR cultivar. With CT cultivar, fruits grafted on Ferro Tab. 1: Effects of rootstocks on fruits weight, diameter and length of ‘Crimson Tide’ (CT) and ‘Crisby’ (CR) watermelon fruits at harvest Scion / rootstock Fruit weight Fruit diameter Fruit length (g) (mm) (mm) CR Control 4814.29c 21.07a 21.00a Macis 5653.78ab 21.55a 21.80a Argentario 5251.60bc 21.50a 21.92a RS841 6076.50a 22.01a 22.14a Ferro 6058.54a 22.17a 22.95a CT Control 6316.41c 21.26a 24.66a Macis 6343.75c 22.26a 25.81a Argentario 6668.41b 22.92a 26.37a RS841 6914.47a 21.60a 25.33a Ferro 6814.06ab 21.99a 25.60a X Mean separation was performed by Fisher’s LSD test. Means (n=3) followed by same letters within a column are not significantly different at P<0.05. Fig. 1: Effects of rootstocks on rind thickness of ‘Crisby’ and ‘Crimson Tide’ watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C R in d th ic kn es s (m m ) R in d th ic kn es s (m m ) 194 A.E. Özdemir, E. Çandır, H. Yetişir, V. Aras, Ö. Arslan, Ö. Baltaer, D. Üstün, M. Ünlü (1.43 %) and RS841 (1.54 %) rootstocks had the lower fungal decay than those on Argentario (1.87 %), Macis (1.90 %) and control fruits (2.11 %) after shelf life period for 7 days at 21° following 21 days of storage at 7 °C. In this study, increased in storage time and sub- sequent shelf life period resulted in the surface decay caused by dif- ferent fungi. Botrytis cinerae, Alternaria cucumerina, Cladosporium cucumerinum, Fusaium spp., Penicilliım digitatum and P. italicum, the most encountered fungi identified followed by Colletotrichum or- biculare and Stemphylium spp. Grafted and non-grafted watermelon fruits did not show differential susceptibility to the those pathogen. Out of pathogen identified, B. cinerae, C. cucumerinum Stemphylium spp. were first time described as additional pathogens to other previ- ously known species causing postharvest decay of watermelon. Post- harvest rots caused by Fusarium spp. and Phytophthora capsici are of concern because control measures for these fungi in the field often are inadequate. With good disease control in the field, anthracnose (C. orbiculare) and black rot (Didymella bryoniae) rarely develop on watermelon (RUSHING et al., 2001). Flesh firmness decreased during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars and grafted fruits had firmer comparing to non-grafted fruits (Tab. 2). Our data suggest that effects of rootstocks on flesh firmness varied depending on the rootstock and the scion. Watermelons grafted on Ferro and RS841 rootstocks had the higher fruit flesh firmness for CT and CR culti- vars. Non-grafted fruits had the lowest fruit flesh firmness after shelf life period following storage for CR and CT cultivars. At harvest, an increase in flesh firmness due to grafting has been reported. (SALAM et al., 2002; YETISIR et al., 2003; DAVIS and PERKINS-VEAZIE, 2005; ROBERTS et al., 2007; CUSHMAN and HUAN, 2008; BRUTON et al., 2009; SOTERIOU and KYRIACOU, 2015) while grafting on some root- stocks seems not affect watermelon firmness (KARACA et al., 2012). The findings of KYRIACOU and SOTERIOU (2015) indicated that C. maxima × C. moschata hybrids sustained higher postharvest flesh firmness compared with non-grafted controls, while rootstock effect was superior to that of cultivar and storage. Watermelon fruit flesh firmness did not change or reduced during storage during 4 weeks of storage at 5°, 10°, 15° or 20 °C depending on storage temperature and cultivars (RISSE et al., 1990). Depending on cultivar, seasonal varia- tion and harvest maturity, postharvest decline in flesh firmness may compromise fruit quality within 14 days from harvest (KYRIACOU and SOTERIOU, 2015). Juice pH value slightly decreased during storage at 7 °C for 21 days and additional 7 days at 21 °C. Similarly, lower pH values were re- ported in ‘Charleston Gray’ watermelons fruits after storage at 7 °C for 14-19 days (CHISHOLM and PICHA, 1986) and ‘Fantasy F1’ water- melons after storage at 20 °C for 14 days (RADULOVIĆ et al., 2007) compared to the pH values at harvest. In CR cultivar, non-grafted fruits had higher pH comparing to grafted fruits. In CT cultivar, fruits on RS841 rootstock resulted in lower pH than those on other root- stocks and non-grafted fruits after shelf life period following sto- rage (Tab. 3). Grafting of mini-watermelons on commercial hybrid rootstock PS 1313 had no effect on juice pH (PROIETTI et al., 2008) while grafted ‘Crimson Tide’ watermelons on some local bottle gourd rootstocks had lower juice pH, compared to control (ÇANDIR et al., 2013). TA content slightly increased in parallel with changes in juice pH during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars (Tab. 3). In CR cultivar, fruits on RS841 and Ferro rootstocks had higher TA than those on other rootstocks and non- grafted fruits after 7 days at 21 °C following 21 days storage at 7 °C. In CT cultivar, fruits on RS841 rootstock resulted in higher TA than those on other rootstocks and non-grafted fruits after shelf life period following storage (Tab. 3). Higher TA due to grafting was reported in watermelon fruits (PROIETTI et al., 2008; ÇANDIR et al., 2013). The malic acid content varied from 0.23 % to 0.28 % for CR cultivar and 0.23 % to 0.31 % for CT cultivar and the citric acid content varied Fig. 3: Effects of rootstocks on fungal decay of ‘Crisby’ and ‘Crimson Tide’ watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C Fig. 2: Effects of rootstocks on weight loss of ‘Crisby’ and ‘Crimson Tide’ watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C Storage and shelf life of grafted watermelons 195 Tab. 2: Effects of rootstocks on some quality attributes of ‘Crisby’ (CR) and ‘Crimson Tide’ (CT) watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7°C Scion / rootstock Days in storage at 7 °C +7 days at 21 °C Mean 0+7 7+7 14+7 21+7 (Rootstock) Firmness (N) CR 6.16c 5.47c 5.25c 3.99c 5.22e Control 6.31bc 6.24b 5.62c 5.13b 5.83d Macis 6.86b 6.44b 6.33b 5.54b 6.29c Argentario 7.65a 7.63a 7.40a 7.09a 7.44a RS841 7.47a 6.90b 7.31a 5.83a 6.88b Ferro 6.16c 5.47c 5.25c 3.99c 5.22e CT Control 5.74b 5.60c 4.38e 4.81c 5.13e Macis 6.06b 6.46bc 5.16d 4.79c 5.62d Argentario 7.03a 6.49bc 6.05c 5.72b 6.32c RS841 7.16a 6.88b 6.72b 6.75a 6.88b Ferro 7.35a 8.29a 7.40a 7.24a 7.57a Ripening (1-7) CR 4.0a 3.4a 3.6a 3.7ab 3.7ab Control 3.8a 3.6a 3.8a 3.7ab 3.7ab Macis 3.8a 3.7a 3.9a 4.0a 3.9a Argentario 3.4a 3.5a 3.5a 3.6b 3.5bc RS841 3.5a 3.1a 3.6a 3.5b 3.4c Ferro 4.0a 3.4a 3.6a 3.7ab 3.7ab CT Control 4.6a 5.2a 5.0a 5.2a 5.0a Macis 4.4a 4.2b 4.4b 5.1a 4.5b Argentario 3.3b 4.0b 3.9bc 4.7b 4.0cd RS841 3.7b 4.1b 4.3b 4.6b 4.2c Ferro 3.2b 3.5b 3.7c 4.5b 3.7d TSS(%) CR Control 10.6c 10.1c 10.4b 10.3c 10.4d Macis 10.2bc 10.3c 10.6b 10.1c 10.3d Argentario 10.8b 10.6b 10.6b 10.5bc 10.6c RS841 10.9ab 10.8ab 11.0a 10.8ab 10.9b Ferro 11.4a 11.0a 11.1a 11.0a 11.1a CT Control 10.9a 11.0a 10.8a 11.1a 10.9a Macis 10.6a 10.4a 10.5a 10.7a 10.5b Argentario 10.1a 10.6a 10.7a 10.7a 10.5b RS841 10.9a 10.9a 10.8a 11.2a 11.0a Ferro 10.3a 10.7a 10.8a 11.1a 10.7ab Taste (1-9) CR 8.1a 6.9a 6.2b 5.3a 6.6c Control 8.0a 7.0a 6.3b 5.7a 6.8bc Macis 8.1a 7.3a 6.5ab 5.9a 7.0ab Argentario 8.2a 7.5a 6.9a 6.1a 7.2a RS841 8.4a 7.2a 6.9a 5.9a 7.1a Ferro 8.1a 6.9a 6.2b 5.3a 6.6c CT Control 7.6b 7.3a 6.7c 5.8c 6.8b Macis 7.9b 7.3a 7.1bc 5.4bc 6.9b Argentario 8.3a 7.9a 7.6b 6.3b 7.5a RS841 8.2a 7.8a 7.2bc 6.7bc 7.5a Ferro 8.2a 7.6a 8.3a 6.5a 7.7a X Mean separation was performed by Fisher’s LSD test. Means (n=3) followed by same letters within a column are not significantly different at P<0.05. 196 A.E. Özdemir, E. Çandır, H. Yetişir, V. Aras, Ö. Arslan, Ö. Baltaer, D. Üstün, M. Ünlü Tab. 3: Effects of rootstocks on juice pH, TA and organic acid content of ‘Crisby’ (CR) and ‘Crimson Tide’ (CT) watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C Scion / rootstock Days in storage at 7 °C +7 days at 21 °C Mean 0+7 7+7 14+7 21+7 (Rootstock) Juice pH CR Control 5.80a 5.60a 5.55a 5.70a 5.66a Macis 5.67b 5.48a 5.60a 5.61b 5.59bc Argentario 5.68b 5.57a 5.56a 5.61b 5.60bc RS841 5.59c 5.52a 5.62a 5.48c 5.55c Ferro 5.63bc 5.53a 5.56a 5.44c 5.54c CT Control 5.69a 5.57a 5.62a 5.59a 5.62a Macis 5.60a 5.57a 5.60ab 5.60a 5.59a Argentario 5.62a 5.44b 5.53c 5.48a 5.52b RS841 5.59a 5.45b 5.55bc 5,39b 5.50b Ferro 5.61a 5.47b 5.44d 5.50a 5.50b TA (%) CR Control 0.16b 0.18a 0.17a 0.17b 0.17bc Macis 0.16b 0.17a 0.15a 0.17b 0.16c Argentario 0.16b 0.17a 0.16a 0.16b 0.16c RS841 0.18a 0.19a 0.16a 0.19a 0.18a Ferro 0.17ab 0.18a 0.17a 0.19a 0.18ab CT Control 0.17a 0.18b 0.17b 0.18b 0.18b Macis 0.17a 0.16c 0.15c 0.16c 0.16c Argentario 0.16a 0.18b 0.17b 0.17bc 0.17bc RS841 0.17a 0.20a 0.19a 0.20a 0.17a Ferro 0.17a 0.19a 0.17b 0.17bc 0.18b Citric acid (%) CR Control 0.07a 0.08a 0.04ab 0.06a 0.06a Macis 0.07a 0.06a 0.03b 0.06a 0.06a Argentario 0.08a 0.07a 0.05a 0.07a 0.07a RS841 0.08a 0.09a 0.05a 0.07a 0.07a Ferro 0.09a 0.07a 0.05a 0.07a 0.07a CT Control 0.10a 0.12a 0.10a 0.06bc 0.09a Macis 0.09a 0.07b 0.05b 0.04c 0.06c Argentario 0.09a 0.11ab 0.04b 0.04c 0.07b RS841 0.10a 0.12a 0.06b 0.08a 0.09a Ferro 0.07a 0.14a 0.08a 0.07ab 0.09a Malic acid (%) CR Control 0.22b 0.25bc 0.23bc 0.26bc 0.24b Macis 0.22b 0.24c 0.22c 0.28c 0.24b Argentario 0.21b 0.23c 0.23bc 0.24c 0.23b RS841 0.25a 0.29a 0.27a 0.33a 0.28a Ferro 0.25a 0.27ab 0.26ab 0.35ab 0.28a CT Control 0.24a 0.27c 0.25bc 0.24b 0.25bc Macis 0.23a 0.25c 0.23c 0.21c 0.23c Argentario 0.25a 0.38ab 0.25bc 0.23bc 0.28ab RS841 0.24a 0.43a 0.28ab 0.29a 0.31a Ferro 0.19a 0.36b 0.31a 0.28a 0.29b aX Mean separation was performed by Fisher’s LSD test. Means (n=3) followed by same letters within a column are not significantly different at P<0.05. Storage and shelf life of grafted watermelons 197 from 0.06 % to 0.07 % for CR cultivar and 0.06 % to 0.09 % for CT cultivar after shelf life period following storage (Tab. 3). The citric acid content was not affected by grafting for CR. In CT, fruits grafted on RS841 and Ferro had higher citric acid content after shelf life peri- od for 7 days at 21° following 21 days of storage at 7 °C, compared to other graft combination and control. Malic acid was the predominant organic acid for both cultivars. Compared to the control and fruits from other graft combination, watermelons grafted on RS841 root- stock had higher malic acid content for CR and CT cultivars during storage at 7 °C for 21 days and additional 7 days at 21 °C. We found a slight increase in ripening ratings during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars (Tab. 2), indicating fruits became overripe toward the end of 21 days of storage and subsequent shelf life period. Similar findings were reported by RISSE et al. (1990) for several watermelon cultivars during 4 weeks of storage at 5°, 10°, 15° or 20 °C. Fruits grafted on RS841 and Ferro rootstocks for CR cultivar and fruits grafted on RS841, Argentario and Ferro rootstocks for CT cultivar had the lowest ripening ratings after shelf life period following storage (Tab. 2). Ripening could be retarded by grafting in watermelon fruits at harvest. ROBERTS et al. (2007) suggested that grafting may delay the harvest date by about 7 days. SOTERIOU et al. (2014) found that as grafting retarded the ripening process, optimum harvest maturity in non-grafted plant was reached at 35-40 days post-anthesis (dpa) compared with 40-45 dpa in grafted plants. TSS content remained above 10 % throughout during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars (Tab. 2), rendering fruit acceptable for perceived sweetness as reported by (KYRIACOU and SOTERIOU (2015). In, CR cultivar, fruits grafted on Ferro rootstock had higher TSS content after shelf life period for 7 days at 21° following 21 days of storage at 7 °C, compared to other graft combination and control (Tab. 2). In case of CT cultivar, con- trol and grafted fruits had similar TSS content after shelf life period following storage (Tab. 2). Although some previous studies (MIGUEL et al., 2004; PROIETTI et al., 2008; BRUTON et al., 2009; BALAZS et al., 2011; BEKHRADI et al., 2011; SOTERIOU and KYRIACOU, 2015) showed that grafting had no effect on TSS, grafting on the bottle gourd rootstocks increased TSS contents of watermelon fruits compared to the control (SALAM et al., 2002; KARACA et al., 2012; ÇANDIR et al., 2013). Unlike our findings, in other studies, grafted watermelon fruits had lower TSS content compared to non-grafted controls (DAVIS and PERKINS-VEAZIE, 2005; ROBERTS et al., 2007; KYRIACOU and SOTE- RIOU, 2015). YETISIR et al. (2003) reported that quality (brix, firm- ness, rind thickness, and fruit shape) of watermelon was greatly af- fected by grafting, but the results were dependent on the rootstock used. The differences in reported results may be due in part to differ- ent production environments, type of rootstock, interactions between specific rootstocks and scions, and harvest date (DAVIS et al., 2008). The most abundant sugar was sucrose at harvest and during storage at 7 °C for 21 days and additional 7 days at 21 °C in both cultivars as reported previously (CHISHOLM and PICHA, 1986; KYRIACOU and SOTERIOU, 2015). In general, changes in on total soluble solid, total and individual sugar contents were not significant during storage and shelf life. In contrast to our findings, in previous studies, it was re- ported an accumulation of sucrose accompanied the decline in total soluble carbohydrates and soluble solids content in grafted and non- grafted watermelons during storage for 14 days at 25 °C (KYRIACOU and SOTERIOU, 2015) and a significant decrease soluble solids and total sugar contents of watermelons during storage for 14 days at 20 °C (RADULOVIĆ et al., 2007). In another study showed that soluble solid content, sucrose, glucose, and fructose concentrations of wa- termelons mostly did not change during storage for 14 days at 0 °C plus 5 days at 23 °C, but all generally were reduced at higher storage temperatures (CHISHOLM and PICHA, 1986). Similarly, total soluble solid content of watermelon cultivars decreased with increased sto- rage temperature (RISSE et al., 1990; PERKINS-VEAZIE and COLLINS, 2006). In our study, preservation of sugars at lower storage tempera- ture may be attributed to a presumably lower rate of respiration. In CR cultivar, effect of grafting on total and individual sugar contents was not significant after shelf life period following storage (Tab. 4). In CT cultivar, although sucrose and total sugar contents were not affected by grafting, fructose and glucose content were higher in fruits grafted on RS841, Ferro and Argentario rootstocks than those on Macis and non-grafted fruits after shelf life period for 7 days at 21° following 7 days of storage at 7 °C (Tab. 4). The differences in fructose and glucose contents between grafted and non-grafted fruits disappeared afterwards. In some studies, grafted watermelon fruits had lower (YETISIR et al., 2003; DAVIS and PERKINS-VEAZIE, 2005; ROBERTS et al., 2007) or similar (MIGUEL et al., 2004; PROIETTI et al., 2008; BRUTON et al., 2009; BEKHRADI et al., 2011) sugar content compared to non-grafted controls. Similarly, KYRIACOU and SOTE- RIOU (2015) reported that between the hybrid rootstocks, mean suc- rose concentration was undifferentiated. In some studies, grafting on the local bottle gourd rootstocks increased fructose, glucose, and sucrose contents of ‘Crimson Tide’ watermelon fruits compared to the control and commercial rootstock grafts (ÇANDIR et al., 2013). On the other hand, the fruits of the non-grafted Bonta watermelon plants had higher sucrose content than the fruits from the grafted plants on the interspecific hybrid rootstock RS 841 and the Lagenaria rootstock FR Strong, while the reducing sugar content (glucose and fructose) showed an opposite pattern (BALAZS et al. 2011). Taste scores (1-9) declined to the lowest level after shelf life period for 7 days at 21° following 21 days of storage at 7 °C (Tab. 2). The effect of rootstock on the taste of watermelon fruits was found to be significant. As the storage time extended, taste tented to decrease, all of the fruit tested by the panelists received high taste scores of >7.9 out of 8.5 at the beginning and decreased to scores of >6.8 out of 7.7 at the end of shelf life, with the exception of non-grafted fruits and fruits grafted on Macis rootstock, which had lower taste scores than the other rootstocks after shelf life period following storage for CR and CT cultivars. Lower taste score may be related to becoming of overripe of control fruits and grafted fruits on Macis rootstock. Furthermore, no off-flavors were detected in fruit from grafted plants. The similar results were obtained in another study conducted on the fruit from grafted watermelons (BRUTON et al., 2009). Flesh color lightness (L* value) decreased during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars (Tab. 5). Similarly PERKINS-VEAZIE and COLLINS (2006) indicated darker (lower L* values) fruits after storage at 21 °C than in freshly harves- ted watermelons. Grafting did not affect flesh color lightness during storage, but RS841and Ferro rootstocks resulted in darker fruits after shelf life period for 7 days at 21° following 21 days of storage at 7 °C for both cultivars. KYRIACOU and SOTERIOU (2015) reported that flesh color lightness (L*) of watermelon fruits was affected by root- stock and storage and compared to the non-grafted control, all hybrid rootstocks invariably maintained darker flesh color (lower lightness value) during storage. The overall intensity of flesh color (C* value), hue angle (h° value) and lycopene content were affected by storage time and rootstocks (Tab. 5). Watermelon flesh color changes from bright red (lower h°) to orange red (higher h°) as ripening level progresses (KARACA et al., 2012). Grafted and non-grafted fruits showed a progressive increase in h° value after shelf life period following storage, indicating a shift from red to orange-yellow. This changes in h° value, characteristic of over-ripening and senescence has been reported after prolonged post- harvest storage of watermelons (KYRIACOU and SOTERIOU, 2015). In both cultivars, C* value continuously decreased during shelf life period at 21 °C following storage at 7 °C. Lycopene content peaked after shelf life period for 7 days at 21° following 7 days of storage at 7 °C for CT cultivars, but it tented to decrease for CR cultivars. Fruits 198 A.E. Özdemir, E. Çandır, H. Yetişir, V. Aras, Ö. Arslan, Ö. Baltaer, D. Üstün, M. Ünlü Tab. 4: Effects of rootstocks on sugar contents of ‘Crisby’ (CR) and ‘Crimson Tide’ (CT) watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C Scion / rootstock Days in storage at 7 °C +7 days at 21 °C Mean 0+7 7+7 14+7 21+7 (Rootstock) Fructose (%) CR Control 3.21a 3.10a 3.24a 3.46a 3.25a Macis 3.22a 3.35a 3.08a 3.89a 3.39a Argentario 2.94ab 3.43a 3.34a 3.43a 3.29a RS841 3.15a 3.31a 3.21a 4.23a 3.48a Ferro 2.73b 3.57a 3.48a 3.90a 3.42a CT Control 2.81ab 2.76b 2.67a 2.80a 2.76b Macis 2.63b 2.83b 2.78a 2.70a 2.74b Argentario 3.14a 3.42a 3.24a 3.00a 3.29a RS841 3.01a 3.79a 3.22a 3.01a 3.16a Ferro 3.11a 3.74a 3.10a 3.34a 3.32a Glucose (%) CR Control 1.66a 1.64a 1.75a 1.75a 1.70a Macis 1.66a 1.51a 1.63a 1.91a 1.68a Argentario 1.55a 1.88a 1.81a 1.81a 1.76a RS841 1.54a 1.63a 1.67a 2.22a 1.76a Ferro 1.32a 1.87a 1.85a 2.06a 1.77a CT Control 1.79a 1.56c 1.62a 1.55a 1.63b Macis 1.75a 1.63bc 1.68a 1.55a 1.65b Argentario 1.73a 1.81bc 1.64a 1.61a 1.70b RS841 1.86a 1.86b 1.88a 1.78a 1.85a Ferro 1.81a 2.16a 1.66a 1.80a 1.86a Sucrose (%) CR Control 4.47a 3.72a 3.89a 4.69a 4.19a Macis 4.85a 4.31a 4.16a 4.65a 4.49a Argentario 5.27a 4.33a 3.94a 5.26a 4.70a RS841 4.72a 4.41a 4.54a 3.57a 4.31a Ferro 4.94a 4.06a 4.66a 4.12a 4.45a CT Control 4.90a 5.39a 5.65a 4.81a 5.19a Macis 4.47a 4.89a 4.77a 4.56a 4.67a Argentario 4.35a 6.21a 4.76a 4.82a 5.04a RS841 4.86a 5.23a 4.39a 3.94a 4.61a Ferro 4.27a 5.44a 4.51a 4.66a 4.72a Total sugar (%) CR Control 9.34a 8.45a 8.87a 9.89a 9.14a Macis 9.73a 9.17a 8.87a 10.45a 9.55a Argentario 9.76a 9.64a 9.08a 10.49a 9.74a RS841 9.40a 9.35a 9.42a 10.01a 9.54a Ferro 8.97a 9.49a 9.99a 10.07a 9.63a CT Control 9.50a 9.71b 9.93a 9.15a 9.57a Macis 8.84a 9.34b 9.23a 8.81a 9.05a Argentario 9.22a 11.85a 9.63a 9.42a 10.03a RS841 9.73a 10.45ab 9.49a 8.72a 9.60a Ferro 9.18a 11.34a 9.27a 9.79a 9.90a X Mean separation was performed by Fisher’s LSD test. Means (n=3) followed by same letters within a column are not significantly different at P<0.05. Storage and shelf life of grafted watermelons 199 Tab. 5: Effects of rootstocks on color and total lycopene content of ‘Crisby’ (CR) and ‘Crimson Tide’ (CT) watermelon fruits after shelf life period for 7 days at 21 °C following storage at 7 °C Scion / rootstock Days in storage at 7 °C +7 days at 21 °C Mean 0+7 7+7 14+7 21+7 (Rootstock) L* CR Control 44.02a 45.50a 44.24a 44.04a 44.45ab Macis 46.61a 46.77a 46.91a 42.29ab 45.65a Argentario 44.84a 44.91a 43.48a 44.02a 44.31b RS841 44.99a 45.40a 43.29a 41.10b 43.70b Ferro 43.08a 44.74a 43.60a 41.48b 43.23b CT Control 44.87a 44.28a 43.70a 45.55a 44.60a Macis 46.42a 43.96a 41.51a 44.36ab 44.06a Argentario 44.87a 44.80a 44.72a 40.22c 43.65ab RS841 43.74a 43.02a 42.29a 41.52bc 42.64b Ferro 44.76a 43.25a 41.73a 40.83c 42.64b C* CR Control 32.75b 30.91b 26.95b 27.81b 29.61c Macis 33.32b 30.86b 30.72a 31.40a 31.57b Argentario 32.58b 34.26a 31.58a 31.29a 32.43ab RS841 35.76a 33.99a 31.74a 32.40a 33.47a Ferro 35.86a 34.39a 31.37a 31.94a 33.39a CT Control 35.27c 32.28a 15.49c 28.37c 27.85a Macis 37.01bc 33.98a 16.15bc 30.16b 29.33a Argentario 39.74a 36.33a 16.41bc 31.38ab 30.97a RS841 39.10ab 34.09a 16.94b 32.67a 30.70a Ferro 37.96ab 35.27a 19.38a 31.68ab 31.07a h° CR Control 44.58a 47.14a 47.54b 48.21a 46.87ab Macis 44.27a 46.69ab 49.85a 47.62a 47.11a Argentario 44.08a 46.68ab 46.25bc 47.36b 46.09b RS841 42.72b 44.12c 45.91c 45.04b 44.45c Ferro 42.41b 45.44b 45.31c 44.34b 44.37c CT Control 45.83a 46.33a 46.40a 45.62a 46.05a Macis 44.38ab 44.09b 43.69b 47.56a 44.93ab Argentario 42.72bc 43.72bc 42.95b 46.92a 44.08bc RS841 42.41bc 42.14b 43.15b 42.27b 42.50d Ferro 42.34c 42.58c 42.30b 45.71a 43.23cd Lycopene (μg g-1) CR Control 32.06c 30.12b 25.03c 20.36c 26.89c Macis 34.65bc 30.53b 22.34d 24.04d 27.89c Argentario 40.47ab 31.80b 25.94a 26.06a 31.07b RS841 44.65a 38.04a 29.32b 33.18b 36.30a Ferro 43.19a 36.36a 32.62a 29.75a 35.48a CT Control 35.42a 39.95c 27.60d 26.81c 32.44b Macis 36.91a 40.42bc 34.36c 27.61c 34.82b Argentario 42.99a 43.83abc 40.23ab 33.55b 40.15a RS841 39.05a 44.55ab 35.78bc 40.09a 39.87a Ferro 42.41a 47.35a 42.18a 36.86ab 42.20a X Mean separation was performed by Fisher’s LSD test. Means (n=3) followed by same letters within a column are not significantly different at P<0.05. 200 A.E. Özdemir, E. Çandır, H. Yetişir, V. Aras, Ö. Arslan, Ö. Baltaer, D. Üstün, M. Ünlü grafted on RS841, Argentario and Ferro rootstocks had more intense (higher C*) brighter red (lower h° value) color and higher lycopene content after shelf life period following storage for both cultivars, compared to non-grafted fruits (Tab. 5). In agreement with our results KYRIACOU and SOTERIOU (2015) reported that h° value increased at 14 days of storage indicating yellowing of the flesh and non-grafted control watermelon fruits presented a greater postharvest transition to yellow than the hybrid rootstocks during storage at 25 °C. Posthar- vest color changes and lycopene biosynthesis in watermelons can be affected by storage temperature and cultivar. PERKINS-VEAZIE and COLLINS (2006) reported that watermelons stored at 21 °C had in- creased C* value and lycopene content compared to fresh fruit where- as no or little change was observed in C* value and lycopene content of fruit held at 5 °C or 13 °C depending on cultivars. The C* value of grafted and non-grafted watermelon fruits peaked after 7 days at 25 °C was not affected by rootstock. However, the intensity of red color in particular, expressed by component a*, was higher in fruits on rootstocks “TZ148” and “N101” than control fruits (KYRIACOU and SOTERIOU, 2015). Consistent with our results, previous studies have typically shown higher lycopene content in watermelon fruit from grafted plants at harvest (DAVIS and PERKINS-VEAZIE, 2005; DAVIS et al., 2008; PROIETTI et al., 2008; ÇANDIR et al., 2013) and dur- ing storage (KYRIACOU and SOTERIOU, 2015). In the current study, changes in lycopene content supports changes in grafted and non- grafted CR and CT watermelon fruits flesh color after shelf life period following storage. The increase in lycopene content, the dominant pigment in watermelon, most likely contributed to the increased C* value as reported by (PERKINS-VEAZIE and COLLINS, 2006). Degrada- tion in lycopene during senescence of non-grafted watermelon fruits of both cultivar and grafted CR fruits after prolonged storage and consequent shelf life period led to decrease in C* and h° value. Al- though grafted and non-grafted watermelons held for 14 days at 25 °C developed a yellowing of flesh (KYRIACOU and SOTERIOU, 2015), we did not observed any yellowing of flesh in grafted or non-grafted fruit held for 7 days at 21° following 21 days of storage at 7 °C. Flesh color changes was observed in non-grafted fruit, suggesting that fruit ripe- ning occurs faster in non-grafted than in grafted fruit during shelf life period after storage. Ripening ratings also confirmed these changes in non-grafted fruits which became overripe toward the end of storage and shelf life. β-carotene content did not significantly changed and was not af- fected by grafting during storage at 7 °C for 21 days and additional 7 days at 21 °C (data not presented). PERKINS-VEAZIE and COLLINS (2006) reported that watermelons stored for 14 days at 21 °C gained 50-139 % in β-carotene compared to fresh fruit, whereas fruit held at 5 and 13 °C changed little in β-carotene content. In our study, lower storage temperature may suppress increase in β-carotene content. In agreement with our results, a similar β-carotene content was reported between fruits grafted on some local bottle gourd rootstocks and non- grafted fruits (ÇANDIR et al., 2013). Effects of grafting on hallow heart was not significant during storage at 7 °C for 21 days and additional 7 days at 21 °C for both cultivars (data not presented). CUSHMAN and HUAN (2008) reported a higher rate of hollow heart incidence in non-grafted watermelon plants than in those that had been grafted. This indicates that hollow heart is af- fected not only by rootstocks but also by other environmental and cultural conditions. Many conflicting results have been reported on the changes in fruit quality resulting from grafting (SALAM et al., 2002; LEE and ODA, 2003; DAVIS and PERKINS-VEAZIE, 2005). BRUTON et al. (2009) ob- served a field and year effect due to soil and climatic conditions on watermelon quality parameters besides rootstock effect. The diffe- rences observed in previous studies may be explained by different production conditions, the type of rootstock/scion combinations used, or the harvest date. Since flowering and harvest time are influenced by grafting, the duration of fruit harvest is prolonged, and the number of fruits per plant is increased in grafted plants (YETISIR and SARI, 2003), it is often difficult to harvest fully ripe fruit from grafted plants in large-scale watermelon production where all fruits are harvested in a single harvest. Watermelons could successfully be kept for 21 days at 7 °C and ad- ditional 7 days at 21 °C. Watermelons grafted on Ferro and RS841 rootstocks had higher postharvest quality compared to the control for both cultivars. Acknowledgments This research was supported by the Scientific and Technological Re- search Council of Turkey (TUBITAK 108 O 391). The authors wish to thank the TUBITAK for their financial support. References ALEXOPOULOS, A.A, KONDYLIS, A., PASSAM, H., 2007: Fruit yield and quality of watermelon in relation to grafting. J. Food Agric. Environ. 5, 178-179. ALAN, Ö., ÖZDEMIR, N., GÜNEN, Y., 2007: Effect of grafting on watermelon plant growth, yield and quality. J. Agron. 6, 362-365. BALAZS, G., KAPPEL, N., FEKETE, D., 2011: The rootstock effect in the Hungarian watermelon production. In: Pospišil, M. (ed.), Proceedings of 46th Croatian and 6th International Symposium on Agriculture, 492-495, Croatia. BEKHRADI, F., KASHI, A., DELSHAD, M., 2011: Effect of three cucurbits rootstocks on vegetative and yield of Charleston Gray watermelon. Int. J. Plant Product. 5, 105-110. BRUTON, B.D., FISH, W.W., ROBERTS, W., POPHAM, T.W., 2009: The Influ- ence of rootstock selection on fruit quality attributes of watermelon. Open Food Sci. J. 3, 15-34. ÇANDIR, E., YETISIR, H., KARACA, F., ÜSTÜN, D., 2013: Phytochemical cha- racteristics of grafted watermelon on different bottle gourds (Lagenaria siceraria) collected from the Mediterranean region of Turkey. Turk. J. Agric. For. 37, 443-456. CHISHOLM, D.N., PICHA, D.H., 1986: Effect of storage temperature on sugar and organic acid contents of watermelons. HortScience 21, 1031-1033. COREY, K.A., SCHLIMME, D.V., 1988: Relationship of rind gloss and ground- spot color to flesh quality of watermelon fruits during maturation. Sci. Hortic. 34, 211-218. CUSHMAN, K.E., HUAN, J., 2008: Performance of four triploid watermelon cultivars grafted onto five rootstock genotypes: yield and fruit quality under commercial growing conditions. Acta Hortic. 782, 335-341. DAVIS, A.R., PERKINS-VEAZIE, P., 2005: Rootstock effects on plant vigor and watermelon fruit quality. Cucurbit Genet. Coop. Rep. 28, 39-42. DAVIS, A.R., PERKINS-VEAZIE, P., HASSELL, R., LEVI, A., KING, S.R., ZHANG, X., 2008: Grafting effects on vegetable quality. HortScience 43, 1670-1672. FAOSTAT, 2014: FAO Statistics Division. http://faostat.fao.org., accessed 15.10.2015. KARACA, F., YETIŞIR, H., SOLMAZ, İ., ÇANDIR, E., KURT, Ş., SARI, N., GÜLER, Z., 2012: Rootstock potential of Turkish Lagenaria siceraria germplasm for watermelon: plant growth yield and quality. Turk. J. Agric. For. 36, 1-11. KING, S.R., DAVIS, A.R., ZHANG, X., CROSBY, K., 2010: Genetics, breeding and selection of rootstocks for Solanaceae and Cucurbitaceae. Sci. Hortic. 127, 106-111. KURT, S., BARAN, B., SARI, N., YETISIR, H., 2002: Physiological races of Fusarium oxysporum f. sp. melon is in the Southeastern Anatolia region of Turkey and varietal reactions to races of the pathogen. Phytoparasitica 30, 395-402. KYRIACOU, M.C., SOTERIOU, G., 2015: Quality and postharvest performance of watermelon fruit in response to grafting on interspecific cucurbit Storage and shelf life of grafted watermelons 201 rootstocks. J. Food Quality 38, 21-29. LEE, J.M., 1994: Cultivation of grafted vegetables. I. Current status, grafting methods and benefits. HortScience 29, 235-239. LEE, J.M., 2003: Vegetable grafting: advances in vegetable grafting. Chron. Horticult. 43, 13-19. LEE, J.M., ODA, M., 2003: Grafting of herbaceous vegetable and ornamental crops. Hortic. Rev. 28, 61-124. LIU, H.Y., ZHU, Z.J., DIAO, M., GUO, Z.P., 2006: Characteristic of the sugar metablolism in leaves and fruits of grafted watermelon during fruit de- velopment. Plant Physiol. Commun. 42, 835-840. LIU, L.H., ZABARAS, D., BENNETT, L.E., AGUAS, P., WOONTON, B.W., 2009: Effects of UV-C red light and sun light on the carotenoid content and physical qualities of tomatoes during post-harvest storage. Food Chem. 115, 495-500. MCGUIRE, R.G., 1992: Reporting of objective colour measurement. Hort- Science 27, 1254-1255. MIGUEL, A., MAROTO, J.V., SAN BAUTISTA, A., BAIXAULI, C., CEBOLLA, V., PASCUAL, B., LOPEZ, S., GUARDIOLA, J.L., 2004: The grafting of triploid watermelon is an advantageous alternative to soil fumigation by methyl bromide for control of fusarium wilt. Sci. Hortic. 103, 9-17. MOTSENBOCKER, C.E., PICHA, D.H., 1996: Quality parameters of triploid watermelons. J. Veg. Crop Produc. 2, 3-14. NETO, A., HAFLE, O.M., GURGEL, F.L., MENEZES, J.B., SILVA, G.G., 2000: Quality and postharvest shelf life of crimson sweet watermelon marketed in Mossoro. Rev. Bras. Eng. Agric. Ambient. 4, 235-239. PERKINS-VEAZIE, P., COLLINS, J.K., 2006: Carotenoid changes of intact water- melons after storage. J. Agric. Food Chem. 54, 5868-5874. PROIETTI, S., ROUPHAEL, Y., COLLA, G., CARDARELLI, M., DE AGAZIO, M., ZACCHINI, M., REA, E., MOSCATELLO, S., BATTISTELLI, A., 2008: Fruit quality of mini-watermelon as affected by grafting and irrigation regimes. J. Sci. Food Agric. 88, 1107-1114. RADULOVIĆ, M., BAN, D., SLADONJA, B., LUSETIĆ-BURSIĆ, V., 2007: Changes of quality parameters in watermelon during storage. Acta Hortic. 731, 451-456. RISSE, L.A., BRECHT, J.K., SARGENT, S.A., LOCASCIO, S.J., CRALL, J.M., ELMSTROM, G.W., MAYNARD, D.N., 1990: Storage characteristics of small watermelon cultivars, J. Amer. Soc. Hort. Sci. 115, 440-443. ROBERTS, W., BRUTON, B., FISH, W., TAYLOR, M., 2007: Using grafted transplants in watermelon production. In: Kelley, W.T. (ed.), Proceedings of the 2007 Southeast Regional Vegetable Conference, 33-36, Savannah, Georgia. RUSHING, J.W., FONSECA, J.M., KEINATH, A.P., 2001: Harvesting and post- harvest handling. In: Maynard, D.N. (ed.), Watermelons − Characteristics, Production, and Marketing, 156-164, Alexandria, VA, American Society for Horticultural Science Press. SALAM, M.A., MASUM, A.S.M.H., CHOWDHURY, S.S., DHAR, M., SAD- DEQUE, A., ISLAM, M.R., 2002: Growth and yield of watermelon as influenced by grafting. Online J. Biol. Sci. 2, 298-299. SAS, 1999: SAS/STAT User’s Guide, Version 8. SAS Institute Inc., Cary, NC, USA. SOTERIOU, G.A., KYRIACOU, M.C., 2015: Rootstock mediated effects on watermelon field performance and fruit quality characteristics. Int. J. Veg. Sci. 21, 344-362. SOTERIOU, G.A., KYRIACOU, M.C., SIOMOS, A.S., GERASOPOULOS, D., 2014: Evolution of watermelon fruit physicochemical and phytochemical composition during ripening as affected by grafting. Food Chem. 165, 282-289. SUSLOW, T.V., 1997: Watermelon, recommendations for maintaining posthar- vest quality, http://rics.ucdavis.edu/postharvest2/Produce/ProduceFacts/ Fruit/watermelon. shtml, accessed 14.10.2015. YAMASAKI, A., YAMASHITA, M., FURUYA, S., 1994: Mineral concentrations and cytokinin activity in the xylem exudate of grafted watermelons as affected by rootstocks and crop load. J. Japan Soc. Hort. Sci. 62, 817- 826. YETISIR, H., SARI, N., 2003: Effect of different rootstock on plant growth, yield and quality of watermelon. Aust. J. Exp. Agric. 43, 1269-1274. YETIŞIR, H., SARI, N., YÜCEL, S., 2003: Rootstock resistance to Fusarium wilt and effect on watermelon fruit yield and quality. Phytoparasitica 31, 163-169. YETISIR, H., SARI, N., 2004: Effect of hypocotyl morphology on survival rate and growth of watermelon seedlings grafted on rootstocks with different emergence performance at various temperatures. Turk. J. Agric. For. 28, 231-237. Address of the authors: Assoc. Prof. Dr. Ahmet Erhan Özdemir, Prof. Dr. Elif Çandır (corresponding author), Özay Baltaer, Durmuş Üstün, Department of Horticulture, Mustafa Kemal University, 31034 Antakya, Hatay, Turkey. E-mail: ecandir@mku.edu.tr Prof. Dr. Halit Yetişir, Department of Horticulture, Erciyes University, 38039 Melikgazi, Kayseri, Turkey. Veysel Aras, Mustafa Ünlü, Alata Horticultural Research Institute, 33740 Erdemli, Mersin, Turkey Ömer Arslan, Agriculture County Directorate, Silifke, Mersin, Turkey © The Author(s) 2016. This is an Open Access article distributed under the terms of the Creative Commons Attribution Share-Alike License (http://creative- commons.org/licenses/by-sa/4.0/).