Physical.

Each whole berry was weighed and the firmness was measured immediately after harvest. Then, the seeds of each berry were removed, weighed, and counted. Berry weight and total seed weight were measured on a digital scale (PA224 Analytic Balance; Ohaus Corporation, Parsippany, NJ). Individual seed weight was calculated (total seed weight/number of seeds). Firmness of the berries was measured using a Stable Micro Systems TA.XT.plus texture analyzer (Texture Technologies Corporation, Hamilton, MA). The berries were placed on the texture unit vertically, stem scar down. Using the 2-mm diameter probe, at a rate of 2 mm/s with a trigger force of 0.02 N, the flesh firmness and skin firmness were measured. Skin firmness, the force required to puncture the skin of the berry divided by the distance traveled before the berry skin was ruptured, was measured in N·mm⁻¹. Flesh firmness was measured consecutively as the probe entered the flesh and was measured as force in N.

Composition.

The juice from the berries was measured for composition (soluble solids, pH, and titratable acidity). The whole berries were frozen (−10 °C), thawed, and squeezed to extract the juice for analysis. The pH and titratable acidity were measured using the Titrino plus 862 compact titrosampler (Metrohm AG, Herisan, Switzerland) with the electrode standardized to pH 4.00, 7.00, and 10.00 buffers. Titratable acidity was determined using ≈6 g of juice diluted with 50 mL deionized, degassed water with titration using 0.1 N sodium hydroxide to an end point of pH 8.2. Titratable acidity was expressed as percentage of tartaric acid. Soluble solids (expressed as percent) were measured using an Abbe Mark II refractometer (Bausch and Lomb; Scientific Instrument, Keene, NH). The ratio of soluble solids to titratable acidity was calculated.

Organic acids and sugars of the juice were determined using high-performance liquid chromatography (HPLC). The juice was filtered through a 0.45-μm nylon filter (VWR International, Radnor, PA) and analyzed using HPLC. Glucose, fructose, tartaric acid, isocitric acid, and malic acid of the juice were measured using previously established HPLC procedures (Segantini et al., 2018; Walker et al., 2003). The HPLC was equipped with a Bio-Rad HPLC Organic Acid Analysis Aminex HPX-87H ion exclusion column (300 × 7.8 mm), a Bio-Rad HPLC fast acid analysis column (100 × 7.8 mm), and a Bio-Rad HPLC column for fermentation monitoring (150 × 7.8 mm) in series (Bio-Rad, Hercules, CA). A Bio-Rad Micro-Guard Cation-H refill cartridge (30 × 4.5 mm) was used for a guard column. The columns were maintained at 65 °C by a temperature control unit. The mobile phase consisted of a pH 2.28 solution of sulfuric acid and water with a resistivity of 18 m, obtained from a Millipore Milli-Q reagent water system, with a 45 mL/min flow rate. The solvent delivery system was a Waters 515 HPLC pump equipped with a Waters 717 plus autosampler (Waters Corporation, Milford, MA). Injection volumes were 10 μL for all samples and run time for completion was 35 min. A Waters 410 differential refractometer to measure refractive index connected in series with a Waters 996 photodiode array detector monitored the eluting compounds. Tartaric, isocitric, and malic acids were detected by photodiode array at 210 nm, and glucose and fructose were detected by the differential refractometer. Peaks were quantified using external standard calibration based on peak height estimation with baseline integration. Individual sugars and acids, total sugars (glucose + fructose), and total organic acids (tartaric + isocitric + malic) were expressed as g/100 g.

Physiochemical attributes.

The six muscadine genotypes were evaluated for physiochemical attributes (physical and composition), and physiochemical attributes varied significantly for most attributes. AM-74 (14.38 g) had the highest berry weight and seed weight (0.12 g) (Table 2). ‘Summit’ (9.25 g) had the lowest berry weight. Threlfall et al. (2007) found similar berry weight for ‘Summit’, but slightly lower weights for ‘Ison’ and ‘Nesbitt’ grapes also grown in Arkansas. AM-83 (0.09 g) had the lowest seed weight. Berry size may have been affected by sex of the plant as ‘Summit’ and AM-74 were female vines, whereas the other four genotypes were perfect-flowered vines. Although Williams (1957) found that female vines produce larger berries, the female vines in this study produced both large and small berries. There were no significant differences in seed number between these genotypes, but berries had about three seeds (ranging from one to four).

Table 2.Physical attributes for fresh-market muscadine genotypes, Clarksville, AR (2017).

View Fullscreen

AM-83 had the highest skin firmness (1.48 N·mm⁻¹) and flesh firmness (2.14 N). AM-9 (0.85 N·mm⁻¹) had the lowest skin firmness and ‘Nesbitt’ (0.89 N) had the lowest flesh firmness (Table 2). Barchenger et al. (2015b) found that the force required to penetrate the muscadine skin ranged from 6 to 13 N at harvest, similar to findings in this study of 5.31–7.50 N (data not shown). Conner (2013) reported flesh firmness ranging from 0.65 to 3.06 N in a study of 26 muscadine grape genotypes grown in Georgia. In that study, they also determined that flesh force or firmness was one of the most useful characteristics for screening in a breeding program to determine Vitis vinifera–like characteristics. Vitis vinifera grapes typically had higher flesh firmness and lower skin firmness than muscadine grapes. Therefore, Conner (2013) suggested selecting for increased flesh firmness and lower skin firmness in muscadine breeding programs. In this study, AM-83 was the most V. vinifera-like with respect to flesh firmness, but AM-9 was the most V. vinifera-like with respect to skin firmness.

‘Summit’ had the highest soluble solids (15.40%) and lowest titratable acidity (0.54%). ‘Nesbitt’ had the lowest soluble solids (12.73%) and ‘Ison’ had the highest titratable acidity (1.01%) (Table 2). AM-83 (3.33) had the highest pH and ‘Ison’ (2.88) had the lowest. Threlfall et al. (2007) found similar soluble solids, pH, and titratable acidity levels for eight muscadine genotypes grown in Arkansas. The ratio of soluble solids to titratable acidity has proven useful for understanding how consumers perceived the balance of sugar to acid in fruit such as peaches and nectarines (Crisosto and Crisosto, 2005). ‘Summit’ (28.49) had the highest soluble solids/titratable acidity ratio and ‘Ison’ (13.12) the lowest (Fig. 1). Walker et al. (2001) indicated preferred soluble solids/titratable acidity ratios of 24–33 for muscadine grapes. Flora (1979) and Threlfall et al. (2007) also found similar preferred ratios for muscadine juice. Using these established parameters, three of the six genotypes (AM-9, AM-74, and ‘Summit’) in this study had an ideal ratio, two had almost ideal ratios (AM-83 and ‘Nesbitt’), and one less than ideal (‘Ison’). ‘Ison’ had a higher titratable acidity than the other genotypes. Although the soluble solids were similar, the greater titratable acidity led to the lower ratio observed.

View Figure

• Download as Powerpoint
• Download Figure

Total organic acids and total sugars (Table 2) and individual organic acids and sugars (data not shown) were not significantly affected by the genotype. Overall, total sugars ranged from 6.17 to 9.75 g/100 g, and the total organic acids ranged from 0.50 to 0.84 g/100 g. Glucose and fructose were present in the fruit in an ≈1:1 ratio, with an average glucose content of 4.14 g/100 g and fructose content of 3.81 g/100 g (data not shown). Tartaric acid was the predominant acid in the muscadines with an average tartaric acid content of 0.37 g/100 g, isocitric acid content of 0.11 g/100 g, and malic acid content of 0.21 g/100 g (data not shown). Similar organic acid and sugar contents have been observed in Arkansas-grown muscadines (Striegler et al., 2005).

Sensory attributes.

Sensory analysis has been shown to explain cultivar characteristics better than instrumental analysis measurements alone. A descriptive sensory lexicon for Arkansas-grown muscadine grapes was created and used to evaluate the six genotypes (Table 1). The panelists evaluated the fruit on seven categories of attributes that included aroma, external appearance, internal appearance, aromatics, basic tastes, feeling factors, and texture. Within each of the seven categories, multiple attributes were evaluated. The results are reported in the order that the panelists evaluated the seven categories.

The panelists evaluated aroma (grape/overall, grape/muscadine, grape/other, fruity, floral, earthy/dirty, mold/mildew, green/unripe, and overripe) of five whole, intact berries (Table 3). All of the aroma attributes were less than 6.5 on the 15-point scale, indicating low-mid aroma intensity. Of the aroma attributes, the panelists detected differences between genotypes in grape/overall, grape/muscadine, and fruity. The panelists did not detect differences in grape/other, floral, earthy/dirty, green/unripe, or mold/mildew, which were low values (0.8), and did not detect any overripe aromas. AM-74 had the highest grape/overall and grape/muscadine aroma and AM-83 had the least. AM-9 had the greatest fruity aroma, and AM-83 had none.

Table 3.Descriptive sensory aroma and external and internal appearance attributes of muscadine genotypes evaluated on a 15-point scale (0 = less of the attribute; 15 = more of the attribute in terms of intensity), Clarksville, AR (2017).

View Fullscreen

Of the seven external appearance attributes, the panelists detected differences among the genotypes for all attributes except the amount of blemishes/deformities (Table 3). The panelists found that the muscadines had a low amount of blemishes/deformities (3.9). AM-83 had the most color-purple, glossiness, and oblong shape. AM-74 had the most color-bronze and largest size. ‘Summit’ had the least glossiness (most dull exterior) and smallest size along with ‘Nesbitt’. AM-9 had the roundest shape. In terms of stem scar, AM-9 had no visible stems scars, and AM-74 and ‘Nesbitt’ had more visible stem scars than the other genotypes (data not shown).

After the panelists evaluated external appearance, the muscadine berry was sliced in half from the stem scar down, and internal appearance (visual separation, number of seeds, and seed size) was evaluated (Table 3). Seed size (6.9) was not different among the genotypes, with seed size closest to the reference B = 4.9 × 7.1 mm. The descriptive number of seeds ranged from 2.6 to 3.6 and was similar to physicochemical results (average of three seeds). Visual separation was the detachability of pulp from the skin of the berry. AM-83 had the least visual separation and most seeds. ‘Ison’ had the most visual separation, and AM-74 had the fewest seeds. In a consumer study by Degner and Mathis (1980), the primary reason consumers did not purchase muscadines in Florida was the presence of seeds. Current muscadine breeding efforts are aimed at maintaining traditional muscadine traits such as aroma, flavor, and texture, while obtaining the complete absence of seeds.

Basic tastes (sweet, sour, and bitter) were evaluated (Table 4). Panelists detected differences among the genotypes for sweetness and sourness but not bitterness (1.0). Sweetness ranged from 6.3 to 7.9, with a score of five which was equal to a 5% solution of sucrose. Sourness ranged from 2.7 to 3.9, with a reference of two which was equal to a 0.05% solution of citric acid. AM-74 was the most sweet and AM-83 was the least. ‘Ison’ was the sourest and ‘Summit’ was the least.

Table 4.Descriptive sensory basic tastes, aromatics, and feeling factor attributes of muscadine genotypes evaluated on a 15-point scale (0 = less of the attribute; 15 = more of the attribute in terms of intensity), Clarksville, AR (2017).

View Fullscreen

Panelists then evaluated the aromatics (olfactory perception caused by volatile substances released from a product in the mouth; retronasal) of the muscadines (Table 4). All of the aromatic attributes were less than 8.5 on the 15-point scale, indicating mid-low aromatic intensity. Of the attributes, the panelists detected differences between genotypes in overall aromatic impact, grape/overall, and overripe, but all other attributes were not significantly different. AM-83 had the least overall aromatic impact and grape/overall aromatics. ‘Summit’ had the most overall aromatic impact; however, AM-74 had the most grape/overall aromatics. Although the overripe attribute was significant, these values were mostly zero (data not shown). There were no differences in grape/muscadine (7.5) aromatics for these genotypes. The panelists found that the fruit had low levels of grape/other (0.4), fruity (0.9), floral (0.7), earthy/dirty (1.0), green/unripe (1.8), and mold/mildew (0.3) aromatics.

Feeling factors (astringent and metallic) were evaluated (Table 4). Panelists detected differences among the genotypes for metallic but not astringent (6.7). The reference for metallic is tin foil to bite. Metallic feeling factor was low (1.2–1.6). AM-74 was the least metallic, and ‘Ison’ was the most metallic.

The panelists evaluated texture (berry hardness, berry crispness, moisture release, awareness of skins, detachability, fibrousness between teeth, and seed separation) for four berries in the five-berry sample. However, no differences were found among the genotypes for these attributes. All genotypes had a mid-high intensity with respect to awareness of skins (13.0), detachability (11.9), and seed separation (10.3). Panelists found a medium intensity for all genotypes with respect to berry hardness (8.3) and moisture release (9.9). Finally, these genotypes had low to mid intensity for berry crispness (3.6) and fibrousness (4.1). Because the panelists were unable to detect differences in texture of muscadine grapes, possibly reestablishing the texture standards within a narrower range could lead to more discrimination between the texture attributes or having the panelists evaluate a shorter lexicon with texture only.

Correlations between physiochemical and sensory attributes.

Physical attributes with more than one correlation to sensory attributes were presented in Table 5 and in text, whereas attributes with one correlation are only discussed in the text. Of the physical attributes, berry weight was positively correlated with the size of the muscadine (r = 0.95), floral aromatics (r = 0.86), and moisture release (r = 0.83), and negatively correlated with the descriptive number of seeds (r = −0.84). Seed weight was positively correlated with green/unripe aromas (r = 0.87). Seed number was correlated with stem scar tear (r = −0.82), descriptive number of seeds (r = 0.98), floral aromatics (r = −0.94), and mold/mildew aromatics (r = 0.92).

Table 5.Multivariate pairwise analysis of physiochemical and descriptive sensory attributes of muscadine genotypes, Clarksville, AR (2017).

View Fullscreen

Of the texture attributes, flesh firmness was negatively correlated with the shape of the muscadine (r = −0.82) and visual separation (r = −0.82), and positively correlated with grape/muscadine aromatics (r = 0.83). The firmer the berry, the more oval and less detachability of pulp from the skin of berry. Skin firmness was positively correlated with the amount of blemishes/deformities (r = 0.97). Therefore, higher skin firmness indicated a less visually desirable berry in this study. Blemishes/deformities evaluated by the panel were attributed to naturally occurring dark spots or healed sections of the fruit. These blemishes/deformities were superficially present, and in certain instances, they had a different texture than the skin of the berry. This texture difference may have led to the correlation observed with skin firmness.

Of the composition attributes, soluble solids were negatively correlated with grape/other aromatics (r = −0.87). The pH was negatively correlated with green/unripe aromas (r = −0.86) and positively correlated with berry crispness (r = 0.83). Therefore, increased pH indicated less green/unripe aroma and crisper berries. Titratable acidity was negatively correlated with grape/other aromas (r = −0.83) and positively correlated with sourness (r = 0.90), green/unripe aromatics (r = 0.83), and metallic feeling factor (r = 0.87). Interestingly, titratable acidity was the predominant factor of sour taste in the grapes as indicated by the lack of correlation of pH with sourness. More than soluble solids, pH, or titratable acidity, the soluble solids/titratable acidity ratio had the most correlations to descriptive attributes. The soluble solids/titratable acidity ratio was correlated with five sensory attributes, including a positive correlation with grape/other aromas (r = 0.86) and negative correlations with sour (r = −0.97), grape/other aromatics (r = −0.89), green/unripe aromatics (r = −0.86), and metallic feeling factor (r = −0.85). Composition of the berries, especially the relationships observed between composition and aromas/aromatics, may have been attributed to the overall ripeness of the muscadine berries. A study by Walker et al. (2001) found that sensory panelists were best able to identify ripeness of muscadine grapes by the soluble solids to titratable acidity ratio. As the soluble solids and pH increased, and the titratable acidity decreased, the panelists perceived the fruit as more ripe. Therefore, the positive correlation with aromatic attributes and soluble solids, and the negative correlation with aroma attributes and titratable acidity may indicate that the riper the berries are, the greater the aroma/aromatic impact will be.

Correlations were found with respect to the total sugars, glucose, and fructose with descriptive sensory attributes. Total sugars were correlated with 12 sensory attributes, including positive correlations to grape/overall aroma (r = 0.86), grape/muscadine aroma (r = 0.82), floral aroma (r = 0.87), color-bronze (r = 0.92), sweet (r = 0.84), overall aromatic impact (r = 0.84), grape/overall aromatics (r = 0.81), fruity aromatics (r = 0.94), and negative correlations to color-purple (r = −0.97), bitter (r = −0.86), mold/mildew aromatics (r = −0.88), and astringent feeling factor (r = −0.87). Data for glucose and fructose correlation are not shown in Table 5 but listed in the text. Glucose and fructose were both positively correlated (r = 0.85–0.94) with grape/overall aroma, floral aroma, color-bronze, overall aromatic impact, and fruity aromatics, and negatively correlated (r = −0.85–0.97) with color-purple, bitter, mold/mildew aromatics, and astringent feeling factor. Fructose was positively correlated (r = 0.82–0.87) with grape/muscadine aroma, and sweet, grape/overall aromatics, and negatively correlated with glossiness (r = −0.83). From these results, total and individual sugars were clearly important in increasing the presence of desirable aromas and aromatics, such as floral and fruity, as seen by the positive correlations with overall aromatic impact.

Total organic acids were not correlated with any of the descriptive sensory attributes. Of the individual organic acids evaluated, isocitric acid was negatively correlated with the shape of the muscadine (r = −0.82), tartaric acid was negatively correlated with overripe aromatics (r = −0.82), and malic acid was positively correlated (r = 0.83–0.85) with sour and green/unripe aromatics. Although not the predominant acid, malic acid was the only acid to show correlation with sour taste, indicating that malic acid may be of greater importance than isocitric or tartaric acids, with the perception of sourness in the muscadine grapes. In addition, the presence of tartaric acid had a negative effect on the aromatic attribute of overripe.