Melons (Cucumis melo L. Maazoun cultivar) were harvested in Skhira, 80 Km from the governorate of Sfax in Tunisia (geographic position 38°94’N; 9°21’E). The rinds were removed immediately and the seeds were carefully separated by hand from the pulp, cleaned and rinsed of any residual materials. The seeds were then dried in the oven at 40 °C for 24 hours. Dried seeds were finely milled in a grinding type moulinex to a fine powder. The particle-size of the powder has a maximum of 500 μm. The seed powder was stored and preserved in hermetic bags at -20 °C until further analysis and oil extraction.

The crude oil content of the sample was determined with a Soxhlet apparatus using hexane (99% purity, Merck, Darmstadt, Germany) for extraction (method 920.39c, AOAC, 2000).

50 g of melon seed powder were macerated for 4h and then mixed with a solvent of n-hexane and placed in 1 L dark glass bottles at a ratio of seed powder weight to solvent volume of 1:5. The mixture was centrifuged for 15 min at 1000 g. Then, the extract was filtered through Whattman filter paper. This procedure of maceration was done with two replications. The organic phase was eliminated in a rotary evaporator at 40 °C. A stream of gas was used to dry the seed oil, which was stored at -20 °C in amber flasks until analysis.

The acid value (method Cd 3d-63), peroxide value (Cd 8-53), saponification index (Cd 3-25), iodine index (Cd 1-25), refractive index (Cc 7-25) and specific gravity (Cc 10 a-25) of the melon seed oil were determined according to AOCS Official methods (AOCS, 2012). The extinction coefficients from absorbances at 232 and 270 nm were obtained with a UV spectrophotometer using oil in cyclohexane (1%) and a path length of one centimeter. Viscosity was calculated with a Rheologica Rheometer (Rheologica Instruments AB, Lund, Sweden) at 25 °C with a steel cone-plate (C40/4) at a constant shear rate of 100/s.

The pigments (chlorophylls and carotenoids) were performed by measuring the absorbance at 670 and 470 nm, in cyclohexane using UV spectrophotometer (Shimadzu Co., Kyoto, Japan). The chlorophyll and carotenoid contents were calculated as follows (Chtourou et al., 2013):

Where A is the absorbance of the oil at the respective wavelength and d is the spectrophotometer cell thickness (1 cm). The chlorophyll and carotenoid concentrations are expressed as mg per kg of oil, respectively.

The CIELab coordinates of melon seed oil were generated directly by a spectrophotometer Konica Minolta colorimeter previously calibrated with a standard white calibration plate. In this coordinate system, L* is an estimation of the relative luminosity of the seed oil; this parameter varies from 0 to 100 (black-white). A negative a* number designates green color; whereas an a* positive value indicates red-purple. The value of b* gives the yellow–blue character with positive values for yellow shades and negative values for blues.

The Regulation EEC/2568/91(EEC, 1991) was adopted to determine fatty acid methyl esters (FAMEs) which were prepared by the vigorous shaking of the seed oil solution in heptane with 0.2 mL of 2N methanolic potassium hydroxide. Gas chromatography of FAMEs was performed on a Gas chromatograph (HP 6890 N, Agilent Technologies) fitted with a split/splitless injector at 230 °C and a flame ionization detector at 250 °C. The column used was a capillary Agilent CP-Sil88 (length 50 m; internal diameter 0.32 mm; film thickness 0.20 μm). The temperature of the column was held at 180 °C; the carrier gas was helium at a flow rate of 1 mL/min; the split ratio was 1:50, with an injection volume of 1 μL. FAME peak identification was determined by the comparison of retention times of the authentic samples in the same conditions. FAMEs were quantified according to their percentage area, obtained by integration of the peaks. The results were expressed as a percentage of individual fatty acids in the lipid fraction.

The triacylglycerol profile was determined using a high-performance liquid chromatography (HPLC) apparatus, with an HP1100 Series instrument chromatograpic system (Agilent Technologies, Waldbronn, Germany) fitted with a differential refractometer detector (model RID-10Avp; Shimadzu). A 0.25 g of oil sample was transferred to a graduated flask (5 mL) and dissolved with acetone (9:91, v/v). After that, separation was made with an analytical spherisorb column (250×4.6 mm) with a particle size of 5 μm from Supelco (Bellefonte, PA, USA). The mobile phase was acetone/acetonitrile (60:40, v/v) at a 1.50 mL/min flow rate. All solvents were of high performance liquid chromatography grade. The column temperature was 30 °C; the injection volume was 20 μL of sample solution prepared as indicated above. The total run time was one hour. The triacylglycerol peaks were identified based on the retention time of triacylglycerol standards: LLL, LLLn, OLL, OOL, POL, PLL, PLLn, PLP, OOO, POO, POP, PPO, POS, SOO, SLO and SLL (S, L, Ln, O and P are stearic, linoleic, linolenic, oleic and palmitic acids, respectively) purchased from Sigma (St Louis, MO, USA). Peak areas produced by the data integrator were used for quantification.

For the extraction of the phenolic fraction, an aliquot of 2 mL of n-hexane and 4 mL of a solution of 60:40 methanol/water were homogenized with 4 g of sample. After vigorous mixing, they were centrifuged at 1490 g for 3 min. The extraction was performed twice and the hydroalcoholic phase was collected. Hydroalcoholic fractions were then united, washed with 4 mL of n-hexane to eliminate the residue of oil, concentrated and dried using a rotary evaporator at 35 °C.

The total phenolic content was determined according to the the Folin Ciocalteu reagent method (Singleton and Rossi, 1965). An aliquot of 0.5 mL of the Folin reagent was added to 0.5 mL of extract. The mixture was stirred and incubated for 3 min in the dark. Then, 10 mL of a solution of Na₂CO₃ (75 g/L) were added. The mixture was shaken and incubated for 1h in the dark. Finally, the absorbance was measured at 750 nm. The results were expressed as mg of gallic acid equivalents per 100 g extract (mg GAE/100 g extract). The value presented the average of three measurements.

For the identification and quantification of melon seed oil phenolic compounds, thirty-four commercial standards were used. Standards were acquired from Sigma-Aldrich (St. Louis, MO, USA) and Extrasynthèse (Genay, France).

A stock solution of each phenolic compound standard was prepared at a concentration of 1 mg/mL and stored in methanol. An intermediate solution containing all standard compounds was prepared in methanol for the calibration curve. A stock solution of cinnamic acid, used as an internal standard (1 mg/mL), was also prepared and stored in methanol. Standard stock solutions were filtered through a 0.45 μm membrane and stored at −20 °C.

For the HPLC analysis, an aliquot of 100 μL of cinnamic acid (1 mg/mL) (internal standard) was added to 100 μL of melon seed oil extract. Samples were passed through a 0.45 μm membrane filter before injection.

The qualitative evaluation of phenolic compounds was carried out using a high performance liquid chromatography technique, Hewlett-Packard system (Waldbronn, Germany) comprising a HP-1100 pump, a Rheodyne model 7725 injector (Cotati, CA), a UV detector (280 nm) and a C18 analytical column (250 mm × 8 mm). The mobile phase consisted of acetonitrile (solvent A) and water with 0.2% sulphuric acid (solvent B). The flow rate was kept at 0.5 mL/min. The gradient program was as follows: 15% A/85% B, 0–12 min; 40% A/60% B, 12–14 min; 60% A/40% B, 14–18 min; 80%A/20% B, 18–20 min; 90% A/10% B, 20–24 min; 100% A, 24–28 min. The injection volume was 20 μL. The data were stored and processed by a HPLC Chemstation (Dos Series) (Hewlette Packard). Peaks were identified by comparison of their retention times with the corresponding standards. Analyses were performed in triplicate. The amounts of phenolic compounds were expressed as mg of phenols per 100 g of dry weight (DW).

The oxidative-stability was measured using a Rancimat apparatus (MetrohomSeries 743). The stage of stabilization was deliberated by the oxidative induction time using 3 g of sample. The temperature was set at 100 °C and the air flow rate was 20 L/h. For the duration of the oxidation procedure, the formation of volatile acids in a receiving flask filled with distilled water (60 mL) was measured by conductometry. The values obtained were defined as induction time.

Thermal analysis of the oil was carried out using a differential scanning calorimeter (Thermal analysis DSC-204, Germany). Seed oil samples were weighed in a differential scanning calorimeter aluminium pan. An unfilled DSC-pan was used as an inert reference. The reference and sample pans were then positioned within the calorimeter and cooled to -50 °C at 10 °C/min and held for 10 min. Then the temperature was increased to 100 °C at 10 °C/min. This process was repeated and the DSC thermographs were traced in melting transition.

The sensorial properties of melon seed oil were compared with those of sunflower and virgin olive oil. Oil samples were presented in a perfectly homogeneous and anonymous way (identical conditions of preparation and presentation) with a simple coding of three numbers. The oil samples were served to untrained panel consisting of sixty panelists of both sexes (30 males and 30 females) from the staff members and students of the National Engineering School of Sfax in Tunisia. Their ages ranged from 23 to 48. The samples prepared for tasting were reserved in tasting glasses at 28 °C with 15 mL of oil.

Detailed directions on how to evaluate the samples were presented verbally before the panelists entered the cubicle area and then also in written form on the score paper. The panelists were given napkins and enough water to rinse their mouths between each sample. Small containers into which samples may be spit were also provided. The prepared oils were evaluated based on a nine-point hedonic scale (1: dislike extremely, 5: neither like nor dislike, 9: like extremely). Each consumer was provided three samples tagged with arbitrary three-digit codes and asked to assess the oils for color, odor, taste, appearance and overall appreciation.

All experiments were performed in triplicate and the results were expressed as the mean values and standard deviation. One-way analysis of variance (ANOVA) was used to determine significant differences among means and Tukey’s test was used to perform multiple comparisons among means using SPSS software (version 20.0; SPSS Inc., Chicago, IL, USA). The significance level was defined as p < 0.05.

The lipid fraction obtained from the extraction of melon seeds (Maazoun variety) presented a yield of 30.65%, on dry weight. Such a yield may be considered economically important when compared to conventional oilseeds such as soy and corn, which contain 18–20% and 3.1–5.7% of oil, respectively (O’Brien, 2004). Belonging to the same botanical family (Cucurbiatceae), the oil content of Maazoun melon seed was higher than agrestis melon (23.33%), bitter melon (19.30%) and Kalahari melon (30.50%) seeds (Mariod et al., 2009; Nyam et al., 2009). Our findings were in agreement with the inodorus Naudin melon seeds with 30.60% of oil (da Silva and Jorge, 2014). According to Nyam et al., (2009), variations in oil yield may be assigned to differences in plant variety, cultivation climate, ripening stage and the extraction method excerted. The notable percentage of oil makes Maazoun melon seeds convenient for application in the oil industry. For this reason, the profile characteristics and thermal properties of this oil should be investigated.

The quality parameters used frequently to evaluate the chemical and physical caracteristics of edible oils are contents in free fatty acids, peroxide value, saponification value, iodine value and refractive index. Table 1 summarizes the physical and chemical properties of melon seed oil (Maazoun variety). Free fatty acid (FFAs) content is an important quality parameter and it gives information about the shelf- life of oil, since FFA can change the physicochemical or organoleptic caracteristics of fat. FFA content in melon seed oil was found to be 0.31% (Table 1). This low value revealed that melon seed oil was edible oil with nutritive value. Moreover, this meant that Maazoun seed oil was nearly free from the hydrolytic rancidity brought about by enzymes.

The peroxide value measures the amount of hydro peroxide in the fat and allows for determining the auto-oxidation of oils. The peroxide value of melon seed oil was equal to 0.5 meq O₂/kg of oil. This value was lower than that reported by Azhari et al., (2014) for melon seed oil (2.5 meq O₂/kg oil) from Sudan. The saponification index indicates the presence or absence of triacylglycerols with high molecular weight in vegetable oils. The saponification index of Maazoun melon seed oil (179.8 mg/g oil) was higher than that described in the literature by Nyam et al., (2009) for Kalahari melon seed oil (173.2 mg KOH/g oil) although lower than that of bitter melon seed oil (190.7 mg KOH/g oil), as observed by the same authors.

Iodine value is used to measure the average number of double bonds or unsaturation in fats. The iodine value of melon seed oil was determined as 139.5 g I₂/100 g oil. Compared to melon seed oils from other geographical origins, this value was higher than the iodine values for melon seed oil from Malaysia (125 g I₂/100g oil) and Sudan (110.99 g I₂/100g oil) (Nyam et al., 2009; Azhari et al., 2014). Such a high value might indicate that melon seed oil from Tunisia possesses higher contents in unsaturated fatty acids than most melon seed oils. Thus, it possessed adequate qualities of drying oil purposes and edible oil. The iodine value was situated within the interval of values for other edible vegetable oils and varied from 95 to 145 g I₂/100 g (Tan and Che Man, 2002).

The refraction index increases with the unsaturation level of the oil. The refractive index of melon seed oil (1.47) was found to be higher than the previously reported value (1.44) by Azhari et al., (2014). The obtained value for specific gravity (0.91) was in agreement with the value found for melon seed oil of the tibish variety (Azhari et al., 2014). The oxidative state of every plant oil may be assessed from the specific absorption at 232 nm (K₂₃₂) and 270 nm (K₂₇₀) which designates, respectively, the presence of primary and secondary oxidation products. The extinction coefficients (K₂₃₂ and K₂₇₀) were found to be 1.21 and 0.42, respectively. These results suggested that Maazoun melon seed oil’s stability to oxidation was relatively good.

The viscosity of melon seed oil (90.75 mPa.s) was lower than that reported by Gohari Ardabili et al., (2011) for pumpkin seed oil (93.66 mPa.s), yet, it was higher than those of most vegetable oils such as sunflower (59.7 mPa.s), soybean (63.23 mPa.s) and groundnut (69.04 mPa.s) oils due to the presence of long chain fatty acids such as linoleic and oleic acids (Chouaibi et al., 2012). Similarly, Hellier et al., (2015) reported an obvious trend in the relationship between fatty acids’ chain length and viscosity. The authors confirmed the influence of fatty acid double bonds on viscosity.

The chlorophyll and carotenoid contents of melon seed oil of the Maazoun variety are illustrated in Table 1. Melon seed oil accounted for 2.43 mg/kg of carotenoids which explained the yellow color of seed oil from Maazoun variety. This amount was lower than that of virgin olive oil with 3.70 mg/kg (Issaoui et al., 2010). With regards to chlorophyll, melon seed oil was found to contain 5.70 mg/kg. This value was higher than that of fig achenes oil (1.05 mg/kg) (Soltana et al., 2016). However, it was lower than that of virgin olive oil (9.32–12.67mg/kg) (Aparicio et al., 1999). These pigments are beneficial because they simulate the appearance of annattos and carotenes frequently used in the fat industry (Oomah et al., 2000). From a nutritional point of view, carotenoids are sources of vitamin A and they play an important role in the oxidative stability of vegetable oils.

The CIE-Lab values (L*, a*, b*) of melon seed oil were 48.42, -5.81 and 15.29, respectively. For comparison purposes, Cucumis melo seed oil of the Maazoun variety showed a higher L* value and lower a* and b* values than pumpkin (Cucurbita maxima) seed oil (Rezig et al., 2012). This indicates that melon seed oil was lighter in color. The CIE-Lab coordinates (L*, a*, b*) values for other vegetable oils, such as olive, sunflower, palm, corn and soybean ranged from 63.4 to 69.50, 3.80 to 4.40 and 9.20 to 10.40, respectively (Hsu and Yu, 2002). Hence, melon seed oil was differentiated by a negative value of the a* coordinate which was clearly lower than that of other vegetable oils. Moreover, the melon seed oil’s b* value was higher than those of common vegetable oils. Thereby, Cucumis melo seed oil was more golden than the vegetable oils examined by Hsu and Hu (2002). This means the attendance of yellow pigments like carotenoids.

The phenolic compounds have a significant importance due to their nutritional and technological properties. As shown in table 1, the total phenolic content of melon seed oil was 22.63 mg/100g oil, which was higher than that of melon inodorus of the Naudin variety (13.07 mg/100g oil) (da Silva and Jorge, 2014). However, melon seed oil of the Maazoun variety had a relatively lower phenol content compared to olive oil, which is considered as a rich source of phenolic compounds in the Mediterranean diet. In fact, Ouni et al., (2011) reported that the total phenolic content of olive oil of the Oueslati cultivar ranged from 33.48 to 85.98 mg/100g oil.

Phenolic compounds have a positive effect in the prevention of cancer and coronary heart disease. Furthermore, the high amount of polyunsaturated fatty acids could make melon seed oil prone to oxidation. However, this could be avoided by the natural phenolic compounds that protect the melon seed oil. Phenolics exert a beneficial effect on the oxidative stability of vegetable oils. According to Aparicio et al., (1999), the contribution of polyphenols in the oxidative stability of olive oil was about 51%, fatty acids, 24% and in lower percentages in carotenoids and chlorophylls.

The fatty acid composition is an essential indicator of the nutritional value of oils. The data about the fatty acid profile of melon seed oil from the Maazoun variety is presented in Table 2. This composition was dominated by C18:2 (68.98%), C18:1 (15.84%), C16:0 (8.76%) and C18:0 (5.64%). Melon seed oil belongs to the linoleic-oleic group of oil. These results are in agreement with previously published studies on other varieties such as Honeydew and inodorus Naudin, in which the main fatty acids were linoleic (59-59.04%), oleic (24.71-26.4%), palmitic (8.7-9.57%) and stearic acid (5.3-5.74%) (da Silva and Jorge, 2014; Górnaś and Rudzińska, 2016). According to Oomah et al., (2000), linoleic acid has favorable nutritional implications and beneficial physiological effects in the prevention of cancer. Hence, melon seed oil has important nutritional value due to the adequate levels of linoleic and oleic acids which constituted more than 84% of total fatty acids. Additionally, melon seed oil contains arachidic, linolenic and gondoic acids, but in much smaller amounts (0.16–0.25%). Trace amounts (<0.1%) of myristic, pentadecanoic, margaric, lignoceric, palmitoleic, heptadecenoic and erucic acids were also detected.

In the current study, the percentage of the different types of fatty acids in melon seed oil recovered from the Maazoun cultivar were as follows: saturated (14.76%), monounsaturated (16.23) and polyunsaturated (69.18%). A similar observation was made by Azhari et al., (2014) in tibish melon seed oil (saturated 19.81%; monounsaturated 18.92%; polyunsaturated 61.26%). The potential reason for the differences in the composition of fatty acids of melon seed oils could be accredited to varietal differences and abiotic factors such as climate conditions, horticultural measures and water availability.

The Maazoun melon seed oil showed an unsaturated/saturated fatty acid ratio of 5.79 (Table 2). A similar ratio of fatty acid content was found in the melon seed oil from the Honeydew variety (UFA/SFA = 5.9) (da Silva and Jorge, 2014). It is well-known that the digestibility and quality of edible vegetable oils are determined by the quantity and composition in unsaturated fatty acids. According to El-Adawy and Taha (2001), the presence of unsaturated fatty acids in sufficient amounts raises great interest. Polyunsaturated fatty acids lower cardiovascular risk and their consumption should be encouraged. Consequently, melon seed oil may be recommended as a dietetic oil and suggested as a dietary supplement for diverse food purposes.

The determination of the triacylglycerol profile has been considered an evaluation of the quality and purity of plant oils. The analysis of triacylglycerol molecular species in Maazoun melon seed oil revealed that trilinolein (LLL), oleo dilinolein (OLL) and palmitoyl dilinolein (PLL) were the most abundant triacylglycerols (Table 3). Therefore, linoleic acid was a major component in the three main types of triacylglycerols (LLL, OLL, and PLL). This can be explained by the fatty acid composition in which linoleic acid presents the highest concentration (68.98%; Table 2). It was remarkable that a close correspondence existed between the fatty acid composition and triacylglycerol profile of the oil. Maazoun melon seed oil also contained POL (Palmitoyl Oleoyl Linoleoylglycerol) in a lower quantity (16.13%). In comparison, these results were similar to the previous study by Yanty et al., (2008) for Honeydew melon seed oil which had the following amounts of major triacylglycerols: LLL (24.90%), OLL (21.50%), PLL (15.90%) and POL (12.40%). Likewise, watermelon seed oil showed a triacylglycerol composition with dominance of LLL (29.5%) and OLL (21.7%) (Conto et al., 2011). However, da Silva and Jorge (2014) reported that OLL (33.10%) was the main triacylglycerol followed by LLL (24.70%) for melon seed oil from the variety inodorus Naudin. The differences in triacylglycerol amounts and types were accredited to the oil source and product processing and were mainly governed by percentages of the individual fatty acids. Furthermore, Maazoun melon seed oil contained relatively small quantities of OOL (4.87%) and OOO (3.60%).

In Table 3, all the triacylglycerols were grouped according to their equivalent carbon number (ECN). Melon seed oil contained four triacylglycerol species (from ECN42 to ECN48) in which triacylglycerols with ECN44 were dominant (44.69%), followed by TAGs ECN42 (28.99%), TAGs ECN46 (22.29%) and TAGs ECN48 (4.02%).

The quantitative analysis showed that melon seed oil from the Maazoun variety contains fourteen phenolic compounds (Table 4). Among these compounds, several flavonoids and phenolic acids were identified. The HPLC characterization of melon seed oil revealed that the amentoflavone, luteolin-7-O-glycoside and gallic acid were the major phenolic compounds accounting for 3.88 mg/100g, 1.26 mg/100g and 0.93 mg/100g, respectively. Defined as plant flavonoids, the presence of amentoflavone has been associated with an antidiabetic function; whereas the luteolin-7-O-glycoside possesses antioxidant and antibacterial properties. Gallic acid seems to have antifungal and antiviral properties. It acts as an antioxidant and helps to protect cells against oxidative damage (Soong and Barlow, 2006). Melon seed oil was also found to contain naringenin, vanillic acid, p-coumaric acid, flavones and syringic acid, although in smaller amounts ranging from 0.44 mg/100 g to 0.67 mg/100g (Table 4). However, compared to the same variety, vanillic, p-coumaric and syringic acids were not detected in the melon seed oil of the Maazoun variety (Mallek-Ayadi et al.,2018). Interestingly, vanillic acid as well as p-coumaric acid are components with important anticancer activity (Rocha et al., 2012); whereas syringic acid possesses antibacterial ability against several microorganisms (Shi et al., 2016). In comparision with other melon varieties, some of the phenolic compounds identified in this study had not been previously reported for Kalahari melon seed oil (Nyam et al., 2009), in particular rosmarinic acid, which exihits important biological activities including antioxidant properties.

The health benefits associated with the intake of polyphenols of plant origin have been scientifically established. Thereby, melon seeds are potentially a source of natural antioxidants which can be used for the prevention of oxidative stress in the human body. It is worth noting that during storage, phenolic compounds protect the polyunsaturated fatty acids of vegetable oils against oxidation by exerting their antioxidant role. Additionally, edible oils rich in natural antioxidants may play a role in reducing the risk of chronic diseases. Thus, the examined oil could be utilized in diverse food preparations to provide nutrition and health benefits.

Oxidative stability is the determination of oxidation induction time of oil and hence its resistance to oxidative phenomena. The oxidative stability index is a significant parameter to evaluate the quality of edible oils. The stability of melon seed oil, evaluated by the Rancimat method, was about 7.20 hours at 100 °C. At the same temperature, the Rancimat induction time of pumpkin seed oil was 18.61 h (Rezig et al., 2012). The melon seed oil was thus less resistant to oxidation than pumpkin seed oil. With regard to the literature, it is observed that tibish seed oil presented lower stability (4.28 h) against oxidation (Azhari et al., 2014). It seems that an arrangement of some composites is probably responsible for the good oxidative stability of unsaturated fatty acids. Decker (1998) reported that the oxidative stability could be improved by controlling the composition and the activity of reaction substrates and minimizing thus the use of food additives.

Differential scanning calorimetry (DSC) is an efficient method to evaluate the physical and thermal caracteristics of oils. The DSC heating parameters assist in discovering the character of the phase evolution which take place when the oils are melted. As shown in figure 1, melon seed oil (Maazoun variety) presented a simple thermogram with two different peaks: a primary temperature melting peak possessing a peak temperature of -46.02 °C, a melting enthalpy 88.41 J/g and an onset temperature -50.76 °C followed by a melting peak with a temperature of -19.5 °C, a melting enthalpy of 138.15 J/g and an onset temperature of -25.81 °C. The melting of the polymorphic forms of triacylglycerols which possess weaker stability could explain the record of the initial endothermic transition observed in the heating thermogram of melon seed oil. The second endothermic peak was clearly discernible in the melon seed oil and may be due to the moresteady polymorphic form or to the presence of higher-melting triacylglycerols. The same trend was found by Cheikh-Rouhou et al., (2007) for the Tunisian variety of Nigella sativa seed oil. The thermal properties of fats were found to correspond with their constitution especially the chain length of fatty acids, level of unsaturation and the distribution of fatty acids in triacylglycerols (Bahloul et al., 2014).

The means of different sensory attributes (color, odor, taste, appearance and overall appreciation) of sunflower oil, melon seed oil and virgin olive oil are presented in table 5. The results of the hedonic analysis show that virgin olive oil is more appreciated by tasters than the other two vegetable oils (melon seed oil and sunflower oil) since it obtained the highest scores for all organoleptic attributes evaluated. In addition, the results from the sensory evaluation indicated that melon seed oil is also appreciated by tasters. In fact, there is no significant difference (p-value > 0.05) in color, taste and appearance between virgin olive oil and melon seed oil. Therefore, this unconventional oil extracted from melon seed of the Maazoun cultivar can be used like other vegetable food oils in culinary preparations and food formulations.