California Agriculture

Transcription

1 California Agriculture Volume 62, Number Page 91 San Joaquin Valley blueberries evaluated for quality attributes Vanessa Bremer Gayle Crisosto Richard Molinar Manuel Jimenez Stephanie Dollahite Carlos H. Crisosto Copyright c 2008 by Regents of the University of California, unless otherwise noted. This article is part of the collected publications of California Agriculture. California Agriculture is archived by the escholarship Repository of the California Digital Library.

2 Abstract Blueberry production in California was estimated in 2007 at around 4,500 acres and is rapidly increasing. Common southern highbush cultivars with low chilling-hour requirements are being grown from Fresno County southward, including Misty, O Neal, Emerald, Jewel, Star and others. We characterized the quality parameters (soluble solids concentration, titratable acidity, ratio of soluble solids concentration to titratable acidity, firmness and antioxidant capacity) of six southern highbush blueberry cultivars grown at the UC Kearney Agricultural Center in Parlier, in the San Joaquin Valley, for three seasons ( ). We also conducted in-store tests to evaluate their acceptance by consumers who eat fresh blueberries. We found that the southern blueberry cultivars currently grown under warm San Joaquin Valley conditions are producing blueberry fruit that is of acceptable quality to consumers and profitable to growers. Keywords: quality attributes, firmness, titratable acidity, antioxidant, In store consumer tests Suggested Citation: Vanessa Bremer, Gayle Crisosto, Richard Molinar, Manuel Jimenez, Stephanie Dollahite, and Carlos H. Crisosto (2008) San Joaquin Valley blueberries evaluated for quality attributes, California Agriculture: Vol. 62: No. 3, Page 91.

3 Research Article t San Joaquin Valley blueberries evaluated for quality attributes by Vanessa Bremer, Gayle Crisosto, Richard Molinar, Manuel Jimenez, Stephanie Dollahite and Carlos H. Crisosto Blueberry production in California was estimated in 2007 at around 4,500 acres and is rapidly increasing. Common southern highbush cultivars with low chilling-hour requirements are being grown from Fresno County southward, including Misty, O Neal, Emerald, Jewel, Star and others. We characterized the quality parameters (soluble solids concentration, titratable acidity, ratio of soluble solids concentration to titratable acidity, firmness and antioxidant capacity) of six southern highbush blueberry cultivars grown at the UC Kearney Agricultural Center in Parlier, in the San Joaquin Valley, for three seasons ( ). We also conducted in-store tests to evaluate their acceptance by consumers who eat fresh blueberries. We found that the southern blueberry cultivars currently grown under warm San Joaquin Valley conditions are producing blueberry fruit that is of acceptable quality to consumers and profitable to growers. Highbush blueberries (Vaccinium corymbosum), native to the northeastern United States, are important commercial fruit (Jimenez et al. 2005) and are the most planted blueberry species in the world (Strik and Yarborough 2005). In the United States, blueberries traditionally have been grown in cooler northern regions; however, the development of new southern cultivars with low chilling-hour requirements (the accumulated number of hours below 45 F [7.2 C] necessary to break dormancy) has made possible the expansion of blueberry production to the southern United States and California (Jimenez et al. 2005). New blueberry cultivars that require fewer hours of chilling have made it possible to grow this specialty crop profitably in hot, dry places such as the San Joaquin Valley. Above, a blueberry field day at the UC Kearney Agricultural Center in Parlier. Blueberry production in California was estimated in 2007 at around 4,500 acres (1,821 hectares) and is rapidly increasing. Common southern cultivars grown include Misty and O Neal, but other improved southern highbush cultivars are now being grown from Fresno southward, such as Emerald, Jewel and Star (Hashim 2004). Southern highbush low-chill cultivars are notable for their productivity, fruit quality and adaptation (Draper 2007), and require only 150 to 600 chillhours, making them promising cultivars for the San Joaquin Valley s mild winters (600 to 1,200 chill-hours annually). Since 1998, we have conducted long-term productivity and performance evaluations of these cultivars at the University of California s Kearney Agricultural Center in Parlier (Jimenez et al. 2005). North American production of highbush blueberry has been increasing since 1975, due to expansion of harvested area and yields through improvements in cultivars and production systems. In 2005, North America represented 69% of the world s acreage of highbush blueberries, with 74,589 acres (30,185 hectares) producing million pounds (139,000 metric tons). Acreage and production increased 11% and 32%, respectively, from 2003 to The U.S. West, South and Midwest experienced the highest increases in acreage. In 2005, 63% of the world s production of highbush blueberries went to the fresh market. North America accounts for a large part of global high- July September

4 Consumers at a Fresno supermarket participated in taste tests of new southern highbush blueberry cultivars. Blueberry samples were presented in random order for consumers to taste and rate on a 9-point hedonic scale (dislike extremely to like extremely). bush blueberry production, representing 67% of the fresh and 94% of the processed markets (Brazelton and Strik 2007). Blueberry consumption is increasing, which is encouraging increased production. As a result, fresh blueberries are becoming a profitable specialty crop, especially in early production areas such as the San Joaquin Valley (Jimenez et al. 2005). In general, a consumer s first purchase is dictated by fruit appearance and firmness (texture). However, subsequent purchases are dependent on the consumer s satisfaction with flavor and quality, which are related to fruit soluble solids (mainly sugars), titratable acidity (organic acids), the ratio of soluble solids to titratable acidity, flesh firmness and antioxidant activity (Kader 1999). Vaccinium species differ in chemical composition, such as sugars and organic acids. The sugars of the larger highbush blueberry cultivars that are grown in California are fructose, glucose and traces of sucrose. Lowbush blueberries (V. angustifolium) which are wild, smaller and grow mostly in Maine lack sucrose. (Kalt and McDonald 1996). The composition of organic acids is a distinguishing characteristic among species. In highbush cultivars, the predominant organic acid is usually citric (~ 83%), while the percentages of succinic, malic and quinic acids are 11%, 2% and 5%, respectively. However, in rabbiteye blueberries (V. ashei) the predominant organic acids are succinic and malic, with percentages of 50% and 34%, respectively, while citric acid accounts for only about 10% (Ehlenfeldt et al. 1994). These different proportions of organic acids affect sensory quality; the combination of citric and malic acids gives a sour taste, while succinic acid gives a bitter taste (Rubico and McDaniel 1992). In addition to flavor, consumers also value the nutritional quality of fresh fruits and their content of energy, vitamins, minerals, dietary fiber and many bioactive compounds that are beneficial for human health (Kader 1999). Fruits, nuts and vegetables are of great importance for human nutrition, supplying vitamins, minerals and dietary fiber. For example, they provide 91% of vitamin C, 48% of vitamin A, 27% of vitamin B6, 17% of thiamine and 15% of niacin consumed in the United States (Kays 1997). The daily consumption of fruits, nuts and vegetables has also been related to reductions in heart disease, some forms of cancer, stroke and other chronic diseases. Blueberries, like other berries, provide an abundant supply of bioactive compounds with antioxidant activity, such as flavanoids (flavonols, anthocyanins and others) and phenolic acids (Schotsmans et al. 2007). For example, a study performed in rats showed that when they were fed diets supplemented with 2% blueberry extracts, age-related losses of behavior (Alzheimer s disease and other) and signal transduction were delayed or even reversed, and radiation-induced losses of spatial learning and memory were reduced (Shukitt-Hale et al. 2007). Some studies have shown that the effects of consuming whole foods are more beneficial than consuming compounds isolated from the food, such as dietary supplements and nutraceuticals. Because fruit consumption is mainly related to visual appearance, flavor and antioxidant properties, we Fresh blueberries are becoming a profitable specialty crop, especially in early production areas such as the San Joaquin Valley. 92 CALIFORNIA AGRICULTURE VOLUME 62, NUMBER 3

5 decided to evaluate fruit quality attributes, antioxidant capacity and consumer acceptance of the early-season blueberry cultivars currently being grown in California. We characterized the quality parameters (soluble solids concentration, titratable acidity, ratio of soluble solids to titratable acidity, firmness and antioxidant capacity) of six southern highbush blueberry cultivars grown in the San Joaquin Valley for three seasons ( ), and evaluated their acceptance by consumers who eat fresh blueberries. Highbush blueberry evaluation Field plots. For the quality evaluations at UC Kearney Agricultural Center, we used three patented southern highbush blueberry cultivars Emerald (US Plant Patent 12165), Jewel (US Plant Patent 11807) and Star (US Plant Patent 10675), and three nonpatented cultivars Reveille, O Neal and Misty. The plants were started from tissue culture and then grown for two seasons by Fall Creek Farm and Nursery in Lowell, Ore. Before planting these cultivars in 2001, the trial plot was fumigated to kill nut grass (Cyperus rotundus and C. esculentus). Because blueberries require acidic conditions, the plot s soil was acidified with sulfuric acid, which was incorporated to a depth of 10 to 12 inches (25.4 to 30.5 centimeters) with flood irrigation, resulting in a ph ranging from 5.0 to 5.5. A complete (NPK) granular fertilizer ( ) was broadcast-applied at a rate of 400 pounds per acre (448 kilograms per hectare). The plants were mulched with 4 to 6 inches (10.2 to 15 centimeters) of pine mulch and irrigated with two drip lines on the surface of the mulch, one on each side of the plant row. Irrigation frequency was two to three times per After harvest, blueberries were tested for, left, soluble solids concentration (shown, Gayle Crisosto with refractometer), center, titratable acidity (shown, Vanessa Bremer with automatic titrator) and, right, firmness (shown, fruit texture analyzer), as well as other qualities. week in the spring and daily during June and July. The emitter spacing was 18 inches (45.7 centimeters), with each delivering 0.53 gallon (2 liters) per hour of water acidified with urea sulfuric acid fertilizer to a ph of 5.0. The plot received an application of nitrogen in the first season, as well as in subsequent growing seasons. The rate was 80 pounds (36.3 kilograms) nitrogen per acre at planting, 60 pounds (27.2 kilograms) the second year, 90 pounds (40.8 kilograms) the third year and 120 pounds (54.4 kilograms) the fourth year. Annual pest control was limited to one application of Pristine fungicide (a combination of the active ingredients pyraclostrobin and boscalid) in February for botrytis management, and two or three herbicide treatments of paraquat (Gramoxone). In year three, the plants received one insecticide treatement of spinosad (Success) for thrips management. Twenty-eight plants per cultivar were planted in a randomized block design using seven plants per block (row) as an experimental unit, replicated in four rows. Rows were spaced 11 feet (3.4 meters) apart, with the plants in the rows spaced 3 feet (0.9 meter) apart, with a space of 4 feet (1.2 meters) between plots. Fruit was harvested at times when it would have been commercially viable if it had been in a commercial field. Fruit from each of the seven plant blocks was harvested and a composite sample of 80 random berries per each replication was used for quality evaluations. Quality measurements. Berries were randomly selected from each replication for quality evaluation at the first harvest time for each respective season ( ). During the 2007 season, in addition to the initial quality evaluations, harvested berries were stored at 32 F (0 C) in plastic clam shells, and measured for firmness 15 days after harvest and for antioxidant capacity 5, 10 and 15 days after harvest. Three replications per cultivar ( seasons) were measured for each quality parameter. The initial firmness of 10 individual berries per replication was measured with a Fruit Texture Analyzer (FTA) (Güss, GS.14, Strand, South Africa) (Slaughter and Rohrbach 1985). Each berry was compressed on the cheek with a 1-inch (2.5 centimeters) flat tip at a speed of 0.2 inch per second (5 millimeters) to a depth of 0.16 inch (4 millimeters) and the maximum value of force was expressed in pounds force (lbf) (1 lbf = 4.5 Newtons). Sixty berries per replication were then wrapped together in two layers of cheesecloth and squeezed with a hand press to obtain a composite juice sample. The juice was used to determine soluble solids concentration (SSC) with a temperature-compensated handheld refractometer (model ATC-1, Atago Co., Tokyo, Japan) and expressed as a percentage. Twenty-one hundredths of an ounce (6 grams) of the same juice sample was used to determine titratable acidity (TA) with an automatic titrator (TIM850 auto-titrator, Radiometer Analytical, Lyon, France) and reported as a percentage of citric acid. Some samples that had a high viscosity were centrifuged with a superspeed centri- July September

6 TABLE 1. Production of six southern highbush blueberry cultivars ( )* fuge (SerVall type SS-1, U.S.A.) at 15,000 rpm for 5 minutes, in order to get liquid juice for soluble solids concentration and titratable acidity measurements (both methods were compared and no differences were observed [data not published]). The ratio of soluble solids concentration to titratable acidity was calculated. Antioxidant analysis. Antioxidant capacity (Trolox Equivalent Antioxidant Capacity [TEAC]) was measured in the 2005 and 2007 seasons. Eighteenhundredths of an ounce (5 grams) of berries (not used for quality measurements) per replication was used to determine the level of antioxidants by the DPPH free-radical method (Brand- Williams et al. 1995). Samples were extracted in methanol to assure a good phenolic representation, homogenized using a polytron (Ultra-Turrax TP 18/101 S1, Junke & Kunkel, Staufen, Germany) and centrifuged (Sorvall Lab assistant Megan Bishop prepares blueberry samples for antioxidant analysis. RC5C, Du Pont Company, Wilmington, Del.) for 25 minutes. The supernatant was analyzed against the standard, Trolox, a water-soluble vitamin E analogue, and reported in micromoles Trolox equivalents per gram of fresh tissue (µmoles TE/g FW). Consumer tests. An in-store consumer test was conducted on Jewel, O Neal and Star blueberry cultivars in 2006, and on the six blueberry cultivars studied in 2007, using methods described previously (Crisosto and Crisosto 2001). The fruit samples were held for 2 days after harvest at 32 F (0 C) prior to tasting. One hundred consumers who eat fresh blueberries, representing a diverse combination of ages, ethnic groups and genders, were surveyed in a major supermarket in Fresno County. Each consumer was presented with a sample of each blueberry cultivar in random order at room temperature, 68 F (20 C). A sample consisted of three fresh whole blueberries presented in a 1-ounce (30 milliliters) soufflé cup labeled with a three-digit code. At the supermarket, the samples were prepared in the produce room out of sight from the testing area. For each sample, the consumer was asked to taste it, and then asked to indicate which statement best described how they felt about the sample on a 9-point hedonic scale (dislike extremely to like extremely). Consumers were instructed to sip bottled water between samples to cleanse their palates. Consumer acceptance was measured as both degree of liking (on a scale of 1 to 9) and percentage acceptance, which was calculated as the number of consumers liking the sample (score > 5.0) divided by the total number of consumers within that sample (Lawless and Heymann 1998). In a similar manner, the percentage of consumers disliking (score < 5.0) and neither liking nor disliking (score = 5.0) the sample was calculated. Statistical analysis. Quality values (firmness, SSC, TA, SSC:TA and TEAC) and data on degree of liking were analyzed with analysis of variance (multifactor ANOVA) and LSD mean separation (P 0.05) with the SAS program. Cultivar lb/acre Emerald 10,747 18,494 19,623 Jewel 8,411 26,966 23,228 Star 3,821 9,968 17,198 Reveille 7,081 7,039 8,313 O Neal 3,830 7,232 9,708 Misty 7,375 8,128 11,157 * Yield was calculated from 21 feet of row including seven plants spaced 3 feet apart. Blueberry cultivar performance Production. Among the studied cultivars, Emerald and Jewel had the highest productivity for 2005 to 2007 (table 1). However, Star had an unexpectedly high productivity in Yield increases for all varieties were due to the maturity of the plants. At planting, the tissue-culture plants were 2 years old; as they matured, they all produced larger yields. The harvest period for Star began the first week of May and ended after the third harvest. Most other cultivars required five or more harvests, 1 week apart. Based on the berry size (table 2), the cultivars studied would be separated into large berry ( Emerald, Jewel and Star ) and medium berry ( Reveille, O Neal and Misty ). The cultivars studied have an erect plant stature, except for Misty, which has a spreading stature that makes hand-harvest difficult. Fruit quality. Quality attributes such as soluble solids concentration, titratable acidity, soluble-solids-totitratable-acidity ratio and firmness were significantly different among cultivars and seasons (table 3). There was wide variability in soluble solids concentration among cultivars. Reveille had the highest average value (14.4%) of the 2005 to 2007 seasons, followed by Misty (12.3%), Emerald (12%) and Star (11.9%). Jewel (11.7%) and O Neal (11.4%) had the lowest soluble solids concentration within this group. Titratable acidity within cultivars was less variable, and only O Neal had a significantly lower average value (0.54%) than the rest of the tested cultivars. Titratable acidity varied from 0.70% to 0.80% within this group with the exception of O Neal. Cultivars segregated into three groups based on their soluble-solids-to-titratable-acidity 94 CALIFORNIA AGRICULTURE VOLUME 62, NUMBER 3

7 TABLE 2. Characteristics of six southern highbush blueberry cultivars ratio. Because of its low titratable acidity, O Neal had the highest ratio, while Jewel had the lowest ratio due to its high titratable acidity. The rest of the cultivars formed an intermediate group in which the soluble-solids-to-titratableacidity ratio ranged from 17 to Jewel and O Neal also had the lowest firmness (1.2 lbf), while Reveille and Misty had the highest (1.6 lbf). Emerald and Star were significantly different than these two groups, forming an intermediate group (1.5 lbf). Quality attributes were also significantly affected by the season. Soluble solids concentration across all cultivars was highest in 2007 and lowest in 2006, while titratable acidity was highest in Soluble-solids-to-titratableacidity ratio and firmness were significantly higher in 2007 than the other years. There was a significant interaction between cultivar and season for all these quality attributes (table 3). The lowest soluble solids concentration was 10.8% in 2006 for O Neal and the highest was 15.8% for Reveille in During this 3-year period, all of the cultivars yielded soluble solids concentrations higher than 10%, which has been proposed as a minimum quality index for blueberries (Kader 1999). Titratable acidity was similar among cultivars in these three seasons except for O Neal in 2007, which reached 0.3%, and Jewel and Emerald in 2006 with about 1.0%. O Neal (40.5) and Reveille (22.9) had the highest soluble-solids-to-titratableacidity ratio, followed by the rest of the cultivars with ratios from 11.4 to During this 3-year period, Jewel and O Neal were the softest cultivars, and Misty and Reveille the firmest. Antioxidant capacity was significantly different among the cultivars but not between seasons (table 3). There was a wide variability of TEAC within cultivars. Misty had the highest average TEAC (19.6 µmol TE/g FW) followed by Reveille (17.3) and Emerald (16.1). Star (12.4), O Neal (12.6) and Jewel (11.0) had the lowest TEAC within this group. Like the rest of the quality attributes, there was a significant interaction between cultivars and seasons for antioxidant Cultivar Plant stature Harvest period* Fruit size Berry grade Hand-harvest ease Emerald Erect Early/midseason Large Moderately easy Jewel Erect Early Large Moderately easy Star Erect Early Large Very easy Reveille Very erect Early Medium Moderate O Neal Erect Early Medium Easy Misty Spreading Early/midseason Medium Difficult * Harvest period specifies initiation of harvest. Early = initial harvest; early/midseason = 7 days later. Number of berries per 6 oz. (0.18 liter) cup. Based on average fruit size: extra large, < 64 berries/cup (6 oz.); large, berries/cup (6 oz.); medium, berries/cup (6 oz.). Source: Jimenez et al TABLE 3. Quality attributes of six southern highbush blueberry cultivars growing in the San Joaquin Valley, (values per cultivar and season) Cultivar SSC * TA SSC:TA Firmness TEAC % % citric acid ratio lbf (µmol TE/g FW) 2005 Emerald 12.1 cdef 0.63 ab 18.4 b 1.60 ab 19.1 ab Jewel 11.9 cdef 0.67 ab 18.1 b 1.07 d 10.3 d Misty 12.2 cdef 0.70 ab 16.6 b 1.57 ab 21.9 a O Neal 11.8 def 0.60 ab 19.0 b 1.30 bcd 13.6 cd Reveille 14.3 ab 0.80 a 18.1 b 1.40 bcd 13.8 cd Star 12.9 bcde 0.77 a 16.4 b 1.57 ab 12.1 d 2006 Emerald 11.6 def 0.90 a 13.2 b 1.43 bcd N/A Jewel 10.9 ef 1.00 a 11.4 b 1.13 cd N/A Misty 11.1 ef 0.57 ab 20.6 b 1.37 bcd N/A O Neal 10.8 f 0.77 a 14.5 b 1.13 cd N/A Reveille 13.3 bdf 0.70 ab 20.0 b 1.57 ab N/A Star 11.1 ef 0.70 ab 17.4 b 1.50 abc N/A 2007 Emerald 12.3 cdef 0.60 ab 20.0 b 1.60 ab 13.2 cd Jewel 12.3 cdef 0.73 ab 17.2 b 1.30 bcd 11.7 d Misty 13.7 bc 0.83 a 17.3 b 1.87 a 17.4 bc O Neal 11.5 def 0.27 b 40.5 a 1.27 bcd 11.7 d Reveille 15.8 a 0.70 ab 22.9 ab 1.90 a 20.7 ab Star 11.6 def 0.67 ab 17.1 b 1.50 abc 12.7 d LSD P value < < * Soluble solids concentration. Titratable acidity. Firmness represents the maximum value of force expressed in pounds force (lbf) (1 lbf = 4.5 Newtons) required to compress the fruit 0.16 inches (4 mm) using a fruit texture analyzer with a 1-inch (2.5-centimeters) flat tip at a speed of 0.2 inch/sec (5 mm/sec). TEAC (Trolox Equivalent Antioxidant Capacity). Same letters within the same column indicate no significant difference between means. TABLE 4. Acceptance of six southern highbush blueberry cultivars by U.S. consumers in consumer test during 2007 season Cultivar Degree of liking* Acceptance Neither like nor dislike Dislike % Emerald 6.2 cd Jewel 6.7 b Misty 6.9 b O Neal 5.9 d Reveille 7.4 a Star 6.6 bc LSD P value < * Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. Same letters within the same column indicate no significant difference between means. July September

8 capacity (data not shown). Storage of the six blueberry cultivars at 32 F (0 C) for 15 days did not affect either antioxidant capacity or firmness, except for O Neal and Misty, whose firmness was reduced slightly but not significantly (data not shown). Consumer acceptance. During the 2006 season, our in-store test results indicated that consumers liked the three tested cultivars slightly to moderately, with an acceptance range of 73.3% to 80%. There were no significant differences in degree of liking between Jewel, O Neal and Star. In these three cultivars the percentage of consumers disliking these fruit reached about 17% (data not shown). During the 2007 season, there were significant differences in degree of liking between the six cultivars tested (table 4). In this test, degree of liking varied from liking slightly to moderately. Reveille had the highest (7.4) and O Neal the lowest (5.9) degree of liking with an acceptance of 92.1% and 67.3%, respectively. Degree of liking of Misty and Jewel was significantly lower than Reveille, but higher than Star and Emerald. Acceptance was near 80% for Jewel, Misty and Star, while only 67% for O Neal and 72% for Emerald. The percentage of consumers that disliked these cultivars varied from 5.9% to 19.8%; Reveille and Misty had the lowest dislike percentage and O Neal the highest. Degree of liking for Jewel and Star were similar (from slight to moderate) during the two seasons. For O Neal, the degree of liking decreased from like slightly-moderately to like slightly. This reduction in consumer acceptance can be explained by the change of titratable acidity from 0.6% to 0.8% in previous years down to 0.3% in 2007 that only occurred in O Neal. This reduction of titratable acidity for O Neil was independent of soluble solids concentration, which remained between 10.8% and 11.8% for the 2005 to 2007 seasons. These results indicated that blueberries with very low titratable acidity (0.3%), despite soluble solids concentrations between 10% and 12%, are not acceptable to consumers. A similar situation has been observed in white and yellow flesh peaches and nectarines with very low acidity (less than 0.4%) (C. and G. Crisosto, personal communication). This reduction in consumer acceptance also points out that the ratio of soluble solids to titratable acidity is not a good indicator for blueberry taste when titratable acidity is low. We are not sure of the reasons for the low titratable acidity in 2007 of O Neal fruit, which appears to be independent of other cultivars. The 2007 season was characterized by high chilling accumulation and a hotter than normal spring, which could have affected O Neal ripening. Choosing a variety The six southern highbush blueberry cultivars studied ( Emerald, Jewel, Star, Reveille, O Neal and Misty ) growing in the San Joaquin Valley had soluble solids concentration levels above the 10% proposed for a minimum quality standard. Titratable acidity ranged from 0.6% to 0.9%, with the exception of 0.3% (2007) for O Neal. Firmness ranged from 1.2 to 1.6 lbf. Reveille was the cultivar with the highest soluble solids concentration, firmness and degree of liking. Antioxidant capacity ranged from 10 References Brand-Williams W, Cuvelier ME, Berset C Use of a free radical method to evaluate antioxidant activity. Lebensmittel Wissenschaft Technologie 28: Brazelton D, Strik BC Perspective on the U.S. and global blueberry industry. J Am Pom Soc 61(3): Crisosto CH, Crisosto GM Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol Technol 22: Draper A Blueberry breeding: Improving the unwild blueberry. J Am Pom Soc 61(3): Ehlenfeldt MK, Meredith FI, Ballington JR Unique organic profile of rabbiteye vs. highbush blueberries. HortScience 29(4): Hashim J Blueberry production gaining in California. Western Farm Press. com/mag/farming_blueberry_production_gaining. Jimenez M, Carpenter F, Molinar RH, et al Blueberry research launches exciting new California specialty crop. Cal Ag 59(2):65 9. Kader AA Fruit maturity, ripening, and quality relationship. Acta Horticulturae (ISHS) 485: to 22 µmoles TE/g FW, with Misty and Reveille the cultivars with higher antioxidant capacity for the 2005 and 2007 seasons. Antioxidant capacity and firmness of the cultivars studied was not affected by storage up to 15 days at 32 F (0 C). Blueberries with very low titratable acidity, despite acceptable soluble solids concentration, had lower consumer acceptance and degree of liking, indicating that the solublesolids-to-titratable-acidity ratio is not a good indicator of consumer acceptance for blueberries. For San Joaquin Valley conditions, these cultivars are all good options for our fast-growing, early fresh blueberry market. V. Bremer is Research Assistant, G. Crisosto is Associate Specialist, S. Dollahite is Research Assistant, and C.H. Crisosto is Postharvest Physiologist, Department of Plant Sciences, UC Davis, located at the UC Kearney Agricultural Center; and R. Molinar and M. Jimenez are Farm Advisors, UC Cooperative Extension, Fresno and Tulare counties, respectively. Thanks to the UC Small Farm Center and Kearney Agricultural Center for their support on the development of this work. Kalt W, McDonald JE Chemical composition of lowbush blueberry cultivars. J Am Soc Hort Sci 121(1): Kays SJ Postharvest Physiology of Perishables Plant Products. Athens, GA: Exon Pr. 532 p. Lawless HT, Heymann H Sensory Evaluation of Food: Principles and Practices. Food Science Texts Series. NY: Chapman Hall. 827 p. Rubico SM, McDaniel MR Sensory evaluation of acids by free-choice profiling. Chemical Senses 17(3): Schotsmans W, Molan A, MacKay B Controlled atmosphere storage of rabbiteye blueberries enhances postharvest quality aspects. Postharvest Biol Technol 44: Shukitt-Hale B, Carey AN, Jenkins D, et al Beneficial effects of fruit extracts on neuronal function and behavior in a rodent model of accelerated aging. Neurobiol Aging 28(8): Slaughter DC, Rohrbach RP Developing a blueberry firmness standard. Trans Am Soc Agric Eng 28(3): Strik B, Yarborough D Blueberry production trends in North America, 1992 to 2003 and predictions for growth. HortTechnol 15: CALIFORNIA AGRICULTURE VOLUME 62, NUMBER 3

9 Using In Store Consumer Tests to Understand Fresh Fruit Consumer Acceptance G.M. Crisosto, K.R. Day and C.H. Crisosto Department of Plant Sciences University of California Davis, CA USA Keywords: peaches, nectarines, plums, sensory evaluation techniques, in store consumer acceptance, organoleptic groups, flavor segregation, chilling injury Abstract Cultivar segregation according to their organoleptic perception was achieved by using trained panel and principal component analysis of 27 peach, 28 nectarine and 12 plum cultivars, respectively, as a part of our three year program to develop minimum quality indexes. Source significantly affected cultivar ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) but, it did not affect sensory perception of sourness and aroma by the trained panel. Based on this information, cultivars were consistently clustered in organoleptic groups with dominant perception of sweetness, sourness, aroma and flavor (peach, plum, or nectarine). This preliminary work demonstrated that high consumer acceptance was attained on ripe cultivars with stronger sensory characteristics than traditional ones without any specific sensory characteristic (balanced group) or cultivars in the sour group. Thus, developing and releasing new cultivars with specific sensory characteristics, low acid and free of chilling injury are desirable to increase consumption. However, proper postharvest temperature management and fast marketing within the market life of each cultivar are requirements to allow consumers to perceive the flavor potential of each cultivar. INTRODUCTION In our previous work we have consistently segregated peach, nectarine and plum cultivars into organoleptic groups. This trained panel sensory work was triggered by the ready availability of cultivars with different flesh colors, flavors, soluble solids concentrations (SSC), and titratable acidities (Neri et al., 1996; Byrne, 2003; Hilaire, 2003). Principal component analysis was used to segregate cultivars into different organoleptic groups. The perception of the four sensory attributes was reduced to three principal components that explained a high percentage of the variation in the sensory characteristics of the cultivars tested (Crisosto and Crisosto, 2006; Crisosto et al., 2007). As cultivars were segregated in different groups with strong perception of sourness, aroma/ flavor (peach, plum, or nectarine), sweetness, or with a lack of them (balanced) we decided to check if consumer acceptance was different among cultivars representing these established organoleptic groups. In these three commodities, consumer acceptance was higher in cultivars that had a dominant organoleptic characteristic than cultivars representing the balanced-traditional group. MATERIALS AND METHODS In store consumer tests were conducted on Ivory Princess (aroma, white flesh peach); Honey Kist (sweet, low acid, yellow flesh nectarine); Spring Bright (balanced, moderate acid, yellow flesh nectarine); Blackamber (sour, high acid plum); Fortune (aroma, moderate acid plum); and October Sun (sweet, low acid pluot) because of their differences in organoleptic characteristics reported in our previous work (Crisosto and Crisosto, 2006; Crisosto et al., 2007). Fruit were collected from trees of each cultivar grown under standard commercial practices at the Kearney Agricultural Center (Parlier, CA). Fruit were harvested at their ideal commercial maturity based on previously published work (Crisosto et al., 2004a, b). Harvested fruit were commercially handled Proc. EURASIA Sym. on Quality Management in Postharvest Systems Eds.: S. Kanlayanarat et al. Acta Hort. 804, ISHS

10 and forced-air cooled to 0-1 C (within 12 h of picking) with an air flow of 0.5 L per minute per kilogram of fruit using a portable cooling tunnel and stored at 0 C up to 7 days prior to ripening. For each sample, the consumer was asked to taste it, then to indicate if he/she likes, neither likes nor dislikes, or dislikes the sample. Then the consumer was asked to indicate his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The consumer s response was recorded using a 9-point hedonic scale (1- dislike extremely to 9-like extremely). Consumer acceptance was measured as both degree of liking (1-9) and percentage acceptance. Percentage acceptance was calculated as the number of consumers liking the sample (score >5.0) divided by the total number of consumers within that sample (Lawless and Heymann, 1998). RESULTS Consumer acceptance calculated as the number of consumers liking the sample varied among the tested cultivars (Table 1). Spring Bright nectarine, a good representative of the balanced group, had a degree of liking of 6.9 and 712% consumer acceptance. Blackamber, a high acid cultivar in the sour group, had the lowest degree of liking (6.3) and consumer acceptance (62%), while Fortune plum and Ivory Princess peach, representatives of the aroma group, reached about 80% consumer acceptance with a degree of liking near 7.2. Finally, Honey Kist and October Sun, in the sweet and flavor groups, reached the highest consumer acceptance (88%) with a degree of liking of 7.5. This preliminary work points out that the presence of a strong sensory characteristic such as aroma (peach, plum or nectarine), flavor (peach, plum or nectarine), or sweet that is perceived by consumers results in high consumer acceptance. However, cultivars which lack strong sensory characteristics (balanced) or with a strong sour characteristic appear to have lower consumer acceptance than cultivars with a characteristic flavor/aroma. It is important to point out that these preliminary in store consumer tests were carried out in a specific and limited consumer population. Thus, we should pursue testing the role of the ethnicity in consumer acceptance using large and diverse populations. This new information may help marketing by matching organoleptic groups (flavor code) with consumer acceptance according to ethnic background. As new cultivars with strong characteristic flavors/aromas are becoming commercially available, consumer acceptance can increase in the near future. Although it is important to point out that flavor potential should be protected during postharvest handling by proper temperature management and fast marketing within the market life of each cultivar (Crisosto et al., 1999). FINAL COMMENTS Higher consumer acceptance was attained on ripe cultivars with a predominant flavor than on traditional ones. Thus, developing and releasing new cultivars with a characteristic flavor component and free of chilling injury are desirable to increase consumption. Intensive research to identify important sensory attributes of cultivars, and in store consumer acceptance should be pursued. The role of orchard factors including ideal maturity should be studied further to assure maximum flavor potential for the consumer. Correct postharvest temperature management and fast marketing within the market life of each cultivar are requirements to allow consumers to perceive the flavor potential of each cultivar. Literature Cited Byrne, D Breeding peach and nectarines for mild-winter climate areas: State of the art and future directions. p In: F. Marra and F. Sottile (eds.), Proc. First Mediterranean Peach Symposium. Agrigento, Italy. Crisosto, C.H. and Crisosto, G.M Segregation of peach and nectarine (Prunus 52

11 persica (L.) Batsch) cultivars according to their organoleptic characteristics. Postharvest Biol. Technol. 39: Crisosto, C.H., Crisosto, G.M., Echeverria, G. and Puy, J Segregation of plum and pluot cultivars according to their organoleptic characteristics. Postharvest Biol. Technol. 44: Crisosto, C.H., Garner, D., Crisosto, G.M. and Bowerman, E. 2004a. Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34: Crisosto, C.H., Mitchell, F.G. and Ju, Z Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. HortScience 34: Crisosto, C.H., Slaughter, D. and Garner, D. 2004b. Developing maximum maturity indices for tree ripe fruit. Adv. HortScience 18: Hilaire, C The peach industry in France: State of art, research and development. p In: F. Marra and F. Sottile (eds.), Proc. First Mediterranean Peach Symposium. Agrigento, Italy. Lawless, H.T. and Heymann, H Sensory evaluation of food: Principles and practices. Food Science Texts Series. Chapman & Hall, New York p.827. Neri, F., Vassalli, P. and Brigati, S Valutazione organolettica di alcune cultivar di pesche e nettarine. Rivista di Frutticoltura 7/8: Tables Table 1. Consumer acceptance of different ripe tree fruit cultivars harvested at their ideal commercial maturity according to their organoleptic group. Cultivar Organoleptic group Degree of liking Acceptance (1-9) 1 (%) Ivory Princess Aroma Honey Kist Sweet Spring Bright Balanced Fortune Aroma October Sun Sweet and plum flavor Blackamber Sour Degree of liking: 1=dislike extremely, 2=dislike very much, 3=dislike moderately, 4=dislike slightly, 5=neither like nor dislike, 6=like slightly, 7=like moderately, 8=like very much, and 9=like extremely. 53

12 54

13 Understanding Tree Fruit Quality Based on Consumer Acceptance C.H. Crisosto and G. Crisosto Department of Plant Sciences University of California Davis, CA USA F. Neri Criof Dept Protect & Improvement Agr Food Prod University of Bologna, Via Gandolfi 19 Bologna, I Italy Keywords: peaches, nectarines, plums, sensory evaluation techniques, in-store consumer acceptance tests, trained panel, preconditioning treatment, soluble solids concentration, ripe titratable acidity, industry surveys, flavor segregation. Abstract In recent years, we have been developing stone fruit quality indexes based on consumer acceptance and fruit market life with the main goal of increasing fruit consumption. To reach this goal we have taken the following steps: First, we conducted soluble solids concentration (SSC) and titratable acidity (TA) surveys, which indicated the potential fruit quality range for these cultivars within the industry. Second, we investigated the potential role of preharvest factors on these quality attributes. Third, we segregated cultivars according to a trained panel s perception of the predominant sensory attributes: sweetness, sourness, and fruit flavor and aroma intensity. Finally, we used the above data to design large in store consumer tests within each sensory classification group. After completion of these steps, the industry will have the solid information it needs to propose a flavor code classification with fruit quality standards specific to each potential sensory classification group. INTRODUCTION After solving the internal breakdown (IB) or chilling injury (CI) problem and lack of ripening issues (Crisosto, 1997;2000, Crisosto et al., 1999;2004a), a detailed research program focusing on understanding flavor and consumer acceptance for different tree fruit cultivars should be pursued (Crisosto, 2002;2003a). As production of new cultivars with different flesh colors, flavors, soluble solids concentrations (SSC), and titratable acidities (TA) is increasing in California and in other areas of the world (Byrne, 2003; Neri et al., 1996, Liverani et al., 2002 and Hilaire, 2003), understanding the relationship between SSC, TA, and texture and consumer acceptance is more critical. Early work associated high consumer acceptance of fruit with high soluble solids concentration (SSC) in sound peaches (Bruhn et al., 1991). However, the role of titratable acidity, SSC: TA and/or peach, nectarine or plum flavor in consumer acceptance has not been well established. SSC and TA in tree fruit are determined by several factors such as cultivar (Ravaglia et al., 1996; Ventura et al., 2000), environmental conditions, canopy position (Crisosto et al., 1997), crop load, ripening, fruit maturity (Testoni, 1995; Crisosto et al., 1997), and rootstocks. PROTECTING FRUIT FLAVOR DURING POSTHARVEST HANDLING A preconditioning treatment to limit internal breakdown (IB) on a short term basis and allow flavor expression is being successfully carried out with consistent results under California and Chilean conditions (Crisosto, 2004a). Currently, there are several companies that offer a considerable volume of high quality fruit to domestic and overseas consumers using this controlled delayed cooling treatment as part of their delivery system. This new system allows the potential of delivering to retail stores tree fruit that are ready to buy with a low occurrence of internal breakdown symptoms and high consumer acceptance. Preconditioned fruit will extend fruit market life even when exposed to temperatures between 36 and 48ºF (temperature killing range) during postharvest handling. In general, a well applied preconditioning treatment extends market Proc. IV th IC on MQUIC Eds. A.C. Purvis et al. Acta Hort. 712, ISHS

14 life by about one to two weeks (Crisosto et al., 2004a). Due to physical and chemical changes occurring in the fruit during the preconditioning treatment, fruits are allowed to express their potential eating quality. A controlled preconditioning/preripening treatment induces fruit softening to the ready to buy stage (approximately 0.36 N for peaches). If the program is not properly monitored, decay development, shriveling, and excessive softening may become commercial problems. Fast cooling and maintaining temperature prior to and during shipment are essential to protect fruit quality from fast deterioration and a high refrigeration capacity may be needed for this program. The product should be packed to specific marketing requirements depending on customer desires, such as being stickered with PLU (tree-ripe codes) and packed in attractive display-ready pre-printed cartons that are either single layer or double layer in depth. An aggressive marketing and promotion program is required. Retailer and consumer education on the handling of preconditioned/preripened fruit is important to increase the demand for this new high quality fruit delivery system. A long term solution to the IB problem is the development of flavorful cultivars free of IB through breeding programs using conventional and molecular genetics techniques to assure high consumer acceptance. PROPOSING QUALITY INDEX (ES) Currently, we do not have enough reliable information from different peach cultivars over different seasons to justify the establishment of SSC as a single quality index. For example, we have such a diversity of potential flavors in our current 200+ peach cultivars that we cannot simplify this complex issue by choosing SSC without the support of solid research over a long period of time. Therefore, to define a high quality eating fruit, we are following these steps: First, conduct an industry quality survey of initial fruit quality attributes. This information will reveal the range of fruit quality attributes within the industry. Second, preliminary studies on the role of preharvest factors in relation to these parameters should be continued. It is very important to realize the cultivar quality potential, thus, our suggested quality indexes will be attainable by these industries. Third, utilize a trained panel to identify the predominant sensory attributes for peach cultivars such as sweetness, sourness, peach flavor and aroma intensity and determine the interrelationship among them. Understanding the role of these quality attributes in consumer acceptance is important to design the in store consumer tests to propose a quality index(es). These in store consumer tests should be designed based on the results of the trained panel and the industry survey. After the completion of this program, the tree fruit industry will have more information to determine if the establishment of this/these index(es) will help to consistently deliver fruit of high eating quality. From our previous sensory work using a single fruit technique and in store consumer tests, we determined that our current cultivars can be classified into three groups according to their consumer acceptance. In the first group, consumer acceptance was related to ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA). Consumer acceptance for these cultivars increased rapidly as RSSC increased reaching ~90%. The cultivars in this group had high acidity (>0.8%) and were predominantly early season with some mid-late season cultivars (Crisosto et al., 2004b; Hilaire and Mathieu, 2004). In the second group of cultivars, consumer acceptance was significantly related to RSSC but not to RTA within the quality attribute ranges used for this test. Consumer acceptance for these cultivars increased rapidly as RSSC increased, reaching ~90%, and then it reached a plateau (Fig. 1) above which it became insensitive to any additional increase in RSSC (saturation point) (Crisosto and Crisosto, 2005). Unfortunately, the saturation point was cultivar dependent, thus, the RSSC to satisfy a minimum of 85% of the consumer population varied between cultivars making it very difficult to propose a minimum RSSC across cultivars. In the third group of cultivars, consumer acceptance was also related to RSSC but in this case consumer acceptance did not reach a plateau with increased RSSC and consumer acceptance reached nearly 100% at the highest level of RSSC tested (Fig. 2). This acceptance level is very high compared to other 184

15 commodities such as table grapes (Crisosto and Crisosto, 2002b) and kiwifruit (Crisosto and Crisosto, 2001) using the same single fruit technique and similar consumer populations. Similar levels of consumer acceptance using the in store consumer test were obtained on Brooks and Bing cherries (Crisosto et al., 2003a & b). THE FLAVOR CODE In our previous sensory work we also found that consumer acceptance was highly related to consumers that chose the neither like nor dislike option rather than disliked the fruit (Crisosto and Crisosto, 2005). For example, the fact that 33% of the consumers picked the neither like nor dislike option, while only 4.2% disliked the fruit suggested that a large number of consumers may be confused. We believe that clear identification and consistent delivery of fruit groups based on their flavor components may help this confused consumer group to make a decision about whether or not they like the fruit and increase consumption. For this reason and the fact that in the last decade a large number cultivars with new notable flavor characteristics and high antioxidant capacity have been released, we tested the hypothesis that cultivars could be consistently segregated based on their predominant sensory characteristics: sweetness, sourness, flavor and aroma intensity (peach, nectarine, plum) using our trained panel. Principal component analysis (PCA) was used to segregate cultivars into different organoleptic groups (Crisosto and Crisosto, 2006). The perception of the four sensory attributes (sweetness, sourness, peach or nectarine flavor intensity, and peach or nectarine aroma) was reduced to three principal components, which accounted for 92% for peaches and 94% for nectarines of the variation in the sensory attributes of the tested cultivars. By plotting the cultivars sensory attributes in the two most important principal components (PC1 and PC2), they were segregated into four groups named balanced, tart (sour), peach aroma/flavor, and sweet (Fig. 3). Our results support our hypothesis that cultivars can be consistently segregated into groups dominated by a specific flavor component such as sweetness, tartness, or strong peach flavor, or peach aroma. Cultivars plotted in between all four sensory vectors were classified in the balanced group. This proposed classification of cultivars into flavor groups and development of a minimum quality index within each group rather than proposing a single generic minimum quality index based on RSSC across cultivars should be pursued. This cultivar flavor classification will help to match ethnic preferences and enhance current promotion and marketing programs. Furthermore, representative cultivars from each organoleptic group could be used to describe biochemical compounds related to the perception of their specific sensory attributes. After identification of these compounds, a candidate gene approach could be used to develop marker(s) to establish an early breeding (seedling) program screening for specific flavors. FINAL COMMENTS The expression of chilling injury symptoms (lack of flavor, mealiness, flesh browning, and uneven ripening) and lack of ripening are the main cause of fruit rejection by consumers. Our in store consumer tests indicated that high consumer acceptance is attained with our mid-season cultivars when fruit are free of internal breakdown and ready to eat prior to consumption. Higher consumer acceptance was attained on cultivars with a predominant flavor than traditional ones. Thus, developing and releasing new cultivars with a characteristic flavor component is desirable to increase consumption. Cultivars were classified into three groups according to consumer acceptance and their quality attributes. Consumer acceptance was mainly related to RSSC and RTA but it was cultivar dependent. A classification of cultivars into flavor groups prior to developing a minimum quality index should be pursued. The use of proper cultural practices and the careful determination of the harvest date should be applied properly to assure that fruit express their quality attributes and the 185

16 majority of fruit would exceed any proposed minimum quality index. Intensive research to identify cultivars important sensory attributes, in store consumer acceptance, industry quality potential, and the role of preharvest factors (orchards and climatic conditions) to meet these potential quality standards should be studied further. Literature Cited Bruhn, C.M., Feldman, N., Garlitz, C., Harwood, J., Ivans, E., Marshall, M., Riley, A., Thurber, D., Williamson, M Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14: Byrne, D Breeding peach and nectarines for mild-winter climate areas: State of the art and future directions. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy 10 September, pp Crisosto, C.H Stone fruit ripening protocol for receivers. 97/101, slide set with cassette tape. Division of Agriculture and Natural Resources. Crisosto, C.H., Johnson, R.S., DeJong, T., Day, K.R Orchard factors affecting postharvest stone fruit quality. HortScience 32(5): Crisosto, C.H., Mitchell, F.G., Ju, Z Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. HortScience 34(6): Crisosto, C.H., Optimum procedures for ripening stone fruit. Management of Fruit Ripening. Postharvest Horticulture Series. No. 9, April. Crisosto, C.H., Crisosto, G.M Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22: Crisosto, C.H. 2002a. How do we increase peach consumption? Acta Hort. 592: Crisosto, C.H., Crisosto, G.M. 2002b. Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24: Crisosto, C.H., Crisosto, G., Bowerman, E. 2003a. Searching for consumer satisfaction: New trends in the California peach industry. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, 10 September, pp Crisosto, C.H., Crisosto, G.M., Metheney, P. 2003b. Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol. Technol. 28: Crisosto, C.H., Garner, D., Andris, H.L., Day, K.R. 2004a. Controlled delayed cooling extends peach market life. HortTechnology 14: Crisosto, C.H., Garner, D., Crisosto, G.M., Bowerman, E. 2004b. Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34: Crisosto, C.H. and G.M. Crisosto Relationship between ripe soluble solids concentration (RSSC) and consumer acceptance of high and low acid melting flesh peach and nectarine (Prunus persica (L.) Batsch cultivars. Postharvest Biology and Technology 38(3): Crisosto, C.H. and G.M. Crisosto Segregation of peach and nectarine (Prunus persica (L.) Batsch cultivars according to their organoleptic characteristics. Postharvest Biology and Technology 39(1): Hilaire, C., The peach industry in France: State of art, research and development. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, 10 September, pp Hilaire, C., Mathieu, V Le Point sur la qualite gustative des peches et nectarines. Infos-Ctifl. 201: Liverani, A., Giovannini, D., Brandi, F Increasing fruit quality of peaches and nectarines: the main goals of ISF-FO (Italy). Acta Hort. 592: Neri, F., Vassalli, P., Brigati, S Valutazione organolettica di alcune cultivar di pesche e nettarine. Rivista di Frutticoltura 7/8: Ravaglia, G., Sansavini, S., Ventura, M., Tabanelli, D Indici di maturazione e 186

17 miglioramneto qualitative delle pesche. Frutticoltura. 3: Testoni, A Momento di raccolta, qualità, condizionamento e confezionamento delle pesche. Proc. Symposium La peschicoltura Veronese alle soglie del Verona, 25 February, p Ventura, M., Sama, A., Minguzzi, A., Lazoni, S., Sansavini, S Ottimizzazione del carico di frutti per migliorare la produzione e la qualità delle nettarine Supercrimson e Venus. In: Sansavini, S. (Ed.), Per una nuova peschicoltura: produzione, organizzazione, mercato. XXIV Convengo Peschicolo, Cesena Febbraio, p Figurese DEGREE OF LIKING (1-9) (A) c c bc ab ab a ab 2 1 CONSUMER ACCEPTANCE (%) (B) RSSC (%) Fig. 1. Relationship between Elegant Lady peach ripe soluble solids concentration (RSSC) and (A) consumer degree of liking measured on 9-point hedonic scale (1- dislike extremely, 5-neither like nor dislike and 9-like extremely) or (B) percentage consumer acceptance. Different letters within a given RSSC level indicate a significant difference between means by LSD

18 DEGREE OF LIKING (1-9) (A) f f ef de cde bcd abc abc a ab 2 1 CONSUMER ACCEPTANCE (%) (B) RSSC (%) Fig. 2. Relationship between Ivory Princess peach ripe soluble solids concentration (RSSC) and (A) consumer degree of liking measured on 9-point hedonic scale (1- dislike extremely, 5-neither like nor dislike and 9-like extremely) or (B) percentage consumer acceptance. Different letters within a given RSSC level indicate a significant difference between means by LSD

19 Fig. 3. Segregation of 23 peach cultivars originating from different breeding programs according to their organoleptic characteristics as perceived by a trained panel and determined by Principal Component Analysis (PCA). PC1 (44%) is plotted on the X-axis and PC2 (33%) on the Y-axis with the vectors representing the loadings of sensory data along with the principal component scores. 189

20 190

21 Postharvest Biology and Technology 22 (2001) Understanding consumer acceptance of early harvested Hayward kiwifruit Carlos H. Crisosto *, Gayle M. Crisosto Department of Pomology, Uni ersity of California at Da is, Kearney Agricultural Center, 9240 South Ri erbend A e., Parlier, CA 93648, USA Received 3 October 2000; accepted 28 January 2001 Abstract During two seasons, in-store consumer acceptance tests were performed to determine the relationship between ripe soluble solids concentration (RSSC) and/or ripe titratable acidity (RTA) on Hayward kiwifruit consumer acceptance. For this, 252 consumers were presented kiwifruit slice samples with RSSC of 11.0, 12.0, 13.0, and 14.0% with a RTA ranging from 0.8 to 1.2%. Kiwifruit with RSSC that ranged from 11.6 to 13.5% were always liked by consumers but with different degrees of liking. A 12.5% RSSC is proposed as a minimum quality index for early-marketed kiwifruit. RTA played a significant role in consumer acceptance only on kiwifruit that had RSSC 11.6% with RTA 1.17% ( sour ). This kiwifruit quality survey indicated that some vineyards had kiwifruit with RSSC higher than 12.5% before they met the recommended minimum maturity standard of 6.5% HSSC. Kiwifruit picked with SSC 6.2% developed flesh breakdown. This work demonstrated that some California vineyards could produce high taste quality (12.5% RSSC) kiwifruit before they reach 6.5% SSC measured when kiwifruit are hanging on the vines Elsevier Science B.V. All rights reserved. Keywords: Actinidia deliciosa; In store consumer tests; Sensory evaluation; Flesh breakdown; Soluble solids concentration; Titratable acidity; Survey; Degree of liking 1. Introduction During the last 5 years, use of the preconditioning protocol by shippers, packers, and buyers, and the ripening protocol by receivers and produce managers has allowed the California kiwifruit industry to safely deliver ready to buy fruit to * Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). consumers during the early and late season (Crisosto et al., 1997). Because of the current kiwifruit market situation, it is economically advantageous to harvest early. However, repeat purchases are dependent upon good eating quality (taste) and it is essential that consumers are satisfied. Consumers consider high quality fruit to be those with nice appearance, high nutritional value, and good taste (Bruhn et al., 1991). According to current University of California recommendations, Hayward kiwifruit harvest can start /01/$ - see front matter 2001 Elsevier Science B.V. All rights reserved. PII: S (01)

22 206 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) when kiwifruit reach a minimum of 6.5% SSC. In Chile and New Zealand, kiwifruit harvest is recommended to start when kiwifruit have at least 6.2% SSC. Earlier studies (Crisosto et al., 1984) demonstrated that late harvested kiwifruit have a long storage potential and are highly accepted by consumers. Consumer acceptance, rather than potential storage, is the most important factor to consider for kiwifruit that will be sold immediately. There is limited information on the relationship between consumer preference or acceptance and chemical composition of ripe kiwifruit (Gorini and Lasorella, 1990; Mitchell et al., 1990; Crisosto, 1992). In other fruit, it has been reported that consumer preferences or acceptance are mainly based on SSC but other quality attributes may also be important (Sozzi et al., 1980; Ben-Arie et al., 1982; Matsumoto et al., 1983; Young and Paterson, 1985; Robertson and Meredith, 1989; MacRae et al., 1990). Among these other quality attributes, TA may play an important role in consumer preferences or acceptance (Lawes and Sawanobori, 1984; Pehrson and Ivans, 1988; Crisosto et al., 1997). The purpose of this work was to determine consumer acceptance for early harvested California Hayward kiwifruit in relation to RSSC, RTA and RSSC/RTA. 2. Materials and methods 2.1. Industry sur ey During the 1998 season, Hayward kiwifruit were collected at four different maturity stages (harvest dates) from eight vineyards in two production areas (San Joaquin Valley and Sacramento Valley). Ten vines (experimental unit) were labeled in each of five replicates per vineyard. In both seasons, 120 kiwifruit from each replicate were sampled at each maturity stage (harvest date). Sixty fruit were used for initial quality and 60 for quality evaluation on ripe kiwifruit. Kiwifruit samples from the four Sacramento Valley vineyards were harvested on 9/25/98, 10/7/98, 10/ 15/98, and 10/29/98, while those from the four San Joaquin Valley vineyards were harvested on 9/22/98, 10/5/98, 10/15/98, and 10/22/98. During the 1999 season, kiwifruit were harvested on 9/30/ 99, 10/7/99, and 10/18/99 from three vineyards located in the San Joaquin Valley production area. Quality attributes measured at harvest included soluble solids concentration (HSSC), firmness (HF), and titratable acidity (HTA). Ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) were measured on kiwifruit ripened at 20 C with 100 ppm ethylene until flesh firmness reached Newtons (N) Storage quality In both the 1998 and 1999 seasons, 100 Hayward kiwifruit from each vineyard/maturity stage/replicate were packaged and stored at 0 C. In 1998, all fruit were evaluated after 4 months storage. In 1999, 15 fruit/vineyard per maturity stage from each of three replicates were removed from storage monthly, and ripened at 20 C until flesh firmness reached N. Kiwifruit were then cut in half longitudinally from sepal end to stem end and evaluated for the presence or absence of flesh breakdown (FB). Flesh breakdown (Harman, 1981), also called internal breakdown (Lallu, 1997), was observed as a ring or zone of granular, water soaked tissue in the outer pericarp beginning at the stylar end of the fruit In store consumer test Groups of 110 and 142 consumers at a major supermarket, located in Fresno County, were tested during the 1998 and 1999 seasons, respectively. Based on an industry-wide survey in 1998 and previous records, each consumer was presented with four ripe kiwifruit samples at four targeted RSSC levels (11.0, 12.0, 13.0, and 14.0%) to begin to define a minimum quality index for early harvested Hayward kiwifruit. Target RTA levels were 1.00 and 0.60%. In 1998, there was a wide range of RTA at the targeted RSSC levels. To ripen previously cold stored kiwifruit samples, cold kiwifruit (0 C) were exposed to 100 ppm ethylene for 12 h following the preconditioning

23 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) protocol (Crisosto et al., 1997; Ritenour et al., 1999). This process was carried out 2 3 days prior to the consumer test and kiwifruit were allowed to ripen at 20 C until they reached N firmness. Firmness was measured on each piece of ripe fruit before the consumer test. Any kiwifruit with firmness outside the N range were not used in these tests. In both seasons, each consumer that said he/she ate kiwifruit was asked to taste four ripe kiwifruit wedges presented in random order in coded ounce soufflé cups at room temperature. The consumer was asked if he/she liked, disliked, or neither liked nor disliked the sample. Then the consumer was asked his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The response was recorded using a nine-point hedonic scale (1, dislike extremely to 9, like extremely). The consumer was instructed to sip bottled water in between samples to cleanse his/her palate. The RSSC and RTA were measured on each sample tasted. Consumer acceptance was measured as both a degree of liking and a percentage. The percentage of consumers liking the kiwifruit sample was calculated as number of consumers liking the kiwifruit sample (score 5.0) divided by the total number of consumers within the sample (Lawless and Heymann, 1998). The percentage of consumers disliking the kiwifruit sample (score 5.0) was calculated as the number of consumers disliking the kiwifruit sample divided by the total number of consumers within the sample. The percentage of consumers that neither liked nor disliked the kiwifruit sample was calculated as the number of consumers that neither liked nor disliked the kiwifruit sample (score=5.0) divided by the total number of consumers within the sample. Degree of liking data was subjected to analysis of variance (ANOVA) prior to Least Significant Differences (LSD) mean separation using the SAS program. In addition to the previous analysis, the Friedman multiple means comparison (Newell and MacFarlane, 1987; Basker, 1988) procedure for ranked data for the two combined seasons was used to compare degree of liking. Percentage consumer acceptance was also calculated in all of the cases. 3. Results and discussion 3.1. Har est quality Growing season In both growing areas, kiwifruit HF decreased as HSSC increased but HTA did not change during the sampling period (Tables 1 3). Average HF of kiwifruit growing in Sacramento Valley vineyards (Table 1) decreased from 94.5 to 70.7 N and HSSC increased from 4.9 to 6.9% between September 25 and October 29, During this sampling period, HTA remained between 1.87 and 2.00%. HF and HSSC values were similar among vineyards. After ripening fruit to a flesh firmness of N average, vineyard kiwifruit RTA decreased to % while RSSC increased to %. In this growing season, three out of the four vineyards in the Sacramento Valley reached minimum maturity (6.5% HSSC) between October 15 and October 29, Vineyard c4 did not reach 6.5% HSSC or have a RSSC 12.5% by October 29. When the three vineyards met the minimum maturity of 6.5% HSSC (measured at harvest time), RSSC was approximately 13%, and kiwifruit HF was below 7.7 N. Kiwifruit from two of the four vineyards had a RSSC 12.5% by October 15 when HSSC was still below 6.5%. Average HF for kiwifruit growing in San Joaquin Valley vineyards (Table 2), decreased from 93.6 to 81.0 N and HSSC increased from 5.4 to 6.9% between September 22 and October 22, During this sampling period, HTA remained near 2.00%. With the exception of vineyard c5 on the last harvest date, there were no important differences in HF, HSSC, HTA, RTA and RSSC among locations. Kiwifruit in vineyard c5 appear to have ripened faster than fruit from the other locations. After ripening to a flesh firmness of N, kiwifruit RTA decreased to % while RSSC increased to %. In this growing season, vineyard c5 in the San Joaquin Valley reached minimum maturity (6.5% HSSC) by October 15. Vineyards c6 and c7 reached minimum maturity by October 22, and vineyard c8 did not meet the minimum maturity standard during the sampling period. When the

24 208 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) three vineyards met the minimum maturity standard of 6.5% HSSC, HF was below 80.6 N. Vineyard c6 was 6 years old and had an average HF of 87.3 N when it reached minimum maturity. Kiwifruit from all vineyards sampled in the San Joaquin Valley had a RSSC 12.5% by October 5 when HSSC were still 6.0%. Kiwifruit from vineyards c5, c6 andc7 had a RSSC 14.2% when they reached 6.5% HSSC. Kiwifruit from vineyard c8 only had 13.5% RSSC on the last sampling date of October Growing season In this short crop season, we only evaluated kiwifruit growing in the San Joaquin Valley. Kiwifruit HF decreased, HSSC increased, and HTA remained the same during the sampling period (Table 3). Average HF among vineyards decreased from 77.0 to 73.8 N, while HSSC increased from 5.8 to 7.1% from September 30 to October 18, During this sampling period, HTA remained between 1.97 and 2.08%. There were no important differences in HF and HSSC values among vineyards. After ripening fruit to a flesh firmness of N, kiwifruit RTA decreased to %, while RSSC ranged from 12.9 to 15.0%. In this growing season, two of the three vineyards reached minimum maturity (6.5% HSSC) between October 7 and October 18, Vineyard c6 had a HSSC of 6.4% on the last sampling date. At the time the two vineyards met the minimum maturity standard, kiwifruit HF Table 1 Hayward kiwifruit quality attributes from four vineyards in the Sacramento Valley at harvest and after ripening, 1998 growing season Vineyard Initial Ripe ( N) Firmness (N) HSSC (%) HTA (%) RSSC (%) RTA (%) Harvest 1 (9/25/98) a a 5.0 b 1.88 a 10.9 a 0.65 a a 4.9 b 1.89 a 11.5 a 0.89 b b 5.0 b 1.91 a 11.0 a 0.79 b b 4.6 a 1.78 a 10.8 a 0.78 b Mean Harvest 2 (10/7/98) a 5.2 a 1.81 a 11.8 a 0.99 a a 5.3 a 1.96 b 12.3 a 0.92 a a 5.3 a 1.90 ab 12.3 a 1.06 a b 5.0 a 1.80 a 11.6 a 1.05 a Mean Harvest 3 (10/15/98) a 5.8 b 1.93 a 12.6 a 0.64 a a 5.7 b 2.03 b 12.5 a 0.57 a a 5.7 b 1.91 a 12.1 a 0.64 a b 5.1 a 1.98 ab 11.7 a 0.66 a Mean Harvest 4 (10/29/98) a 7.2 a 1.97 a 13.1 b 0.80 a a 6.6 a 2.07 a 13.0 b 0.76 a a 7.5 a 1.98 a 13.8 c 0.81 a a 6.3 a 1.97 a 12.3 a 0.65 a Mean a Mean separation within each harvest date by LSD at P Different letters within columns in the same harvest date indicate significant differences.

25 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) Table 2 Hayward kiwifruit quality attributes from four vineyards in the San Joaquin Valley at harvest and after ripening, 1998 growing season Vineyard Initial Ripe ( N) Firmness (N) HSSC (%) HTA (%) RSSC (%) RTA (%) Harvest 1 (9/22/98) a a 5.5 a 1.91 a 12.3 ab 0.76 a a 5.4 a 2.28 a 12.8 b 0.67 a a 5.5 a 1.86 a 11.9 a 0.80 a b 5.1 a 1.97 a 11.9 a 0.75 a Mean Harvest 2 (10/5/98) a 5.8 b 1.85 a 13.2 a 0.99 a b 5.8 b 2.09 c 13.5 a 1.02 a b 6.0 b 1.96 b 13.4 a 0.93 a a 5.4 a 2.04 bc 13.2 a 1.06 a Mean Harvest 3 (10/15/98) a 6.8 c 1.87 a 14.3 b 0.76 bc bc 6.0 ab 2.14 b 14.2 b 0.57 a c 6.4 bc 2.08 b 13.8 b 0.69 ab ab 5.9 a 2.12 b 12.9 a 0.86 c Mean Harvest 4 (10/22/98) a 8.0 c 2.08 ab 14.9 b 0.69 c c 6.5 ab 2.18 bc 14.2 ab 0.54 a b 7.0 b 2.08 ab 14.2 ab 0.57 ab bc 6.2 a 2.26 c 13.5 a 0.68 bc Mean a Mean separation within each harvest date by LSD at P Different letters within columns in the same harvest date indicate significant differences. was 71.6 N. Kiwifruit from the three vineyards had an average RSSC of 12.9% on the first sampling date (September 30) while HSSC was 5.8%. Kiwifruit from vineyards c5 and c7 had a RSSC 15.0% when they reached minimum maturity (6.5% HSSC). This survey agrees with the one we conducted in 1991 (data not shown) Storage quality Growing season Flesh breakdown (FB) symptoms were detected after 4 months at 0 C. Kiwifruit storage quality was not evaluated prior to 4 months cold storage. Incidence of FB was related to HSSC (Fig. 1). There was a high incidence of FB after storage in kiwifruit picked before they reached 6.3% HSSC. During this growing season, kiwifruit reached approximately 6.3% HSSC after October 6 in vineyards in the San Joaquin Valley and after October 13 in vineyards in the Sacramento Valley Growing season During the 1999 growing season, kiwifruit from the San Joaquin Valley were removed from cold storage monthly and evaluated for incidence of FB. FB did not become commercially important until after 3 months at 0 C for these early harvested kiwifruit. Incidence of FB was also well related to HSSC (Fig. 1). During the first 3 months of cold storage, incidence of FB was very low. Kiwifruit from one vineyard had a higher

26 210 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) Table 3 Hayward kiwifruit quality attributes from three vineyards in the San Joaquin Valley at harvest and after ripening, 1999 growing season Vineyard Initial Ripe ( N) Firmness (N) HSSC (%) HTA (%) RSSC (%) RTA (%) Harvest 1 (9/30/99) b a 5.8 a 2.02 a 13.0 a 0.92 b b 5.8 a 1.97 a 12.8 a 1.02 b b 5.7 a 1.92 a 13.0 a 0.60 a Mean Harvest 2 (10/7/99) b 6.3 b 2.05 a 13.8 a 0.81 b a 6.1 a 2.13 b 13.3 a 0.90 bc b 6.4 b 2.05 a 14.9 b 0.66 a Mean Harvest 3 (10/18/99) b 7.8 c 2.10 b 15.9 a 1.08 b b 6.4 a 2.19 b 14.2 a 1.20 c a 7.0 b 1.80 a 15.0 a 0.85 a Mean a Mean separation within each harvest date by LSD at P Different letters within columns in the same harvest date indicate significant differences. incidence of FB than kiwifruit from the others. Differences in FB incidence can be explained by seasonal influences (Lallu, 1997). The 1998 and 1999 growing seasons had differences in weather patterns and crop loads Consumer test Season Consumer acceptance, expressed as degree of liking, was significantly related to RSSC, RTA and RSSC: RTA (Fig. 2). Kiwifruit with RSSC that ranged from 11.6 to 13.5% were always liked by consumers but with different degrees of liking. Consumers liked these kiwifruit from slightly (5.9) to moderately (6.6), and acceptance varied from 70 to 84%. The neither like nor dislike option was selected by only a few consumers, varying from 6.9 to 1.0% depending on RSSC level (Table 4). The percentage of consumers that chose the neither like nor dislike option decreased from 6.9 to 1.0% as RSSC increased. For kiwifruit with equal to or higher than 13.5% RSSC, the percentage of consumers that chose the neither like nor dislike option was approximately 1.0%. There was a significant interaction between RSSC and RTA on the degree of liking. For this reason, data were further analyzed using three levels of RTA (Fig. 3). Degree of liking was the Fig. 1. Relationship between harvest SSC and Hayward kiwifruit flesh breakdown from eight vineyards measured after 4 months cold storage at 0 C, then ripened at 20 C until soft, 1998 and 1999 growing seasons.

27 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) Fig. 2. Relationship between Hayward kiwifruit ripe soluble solids concentration (RSSC), consumer degree of liking (1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, dislike slightly; 5, neither like nor dislike; 6, like slightly; 7, like moderately; 8, like very much; 9, like extremely), and percentage consumer acceptance, 1998 growing season. Fig. 3. Relationship between Hayward kiwifruit ripe soluble solids concentration (RSSC), degree of liking (1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, dislike slightly; 5, neither like nor dislike; 6, like slightly; 7, like moderately; 8, like very much; 9, like extremely), and percentage consumer acceptance at three different levels of titratable acidity, 1998 growing season. Table 4 Relationship between ripe soluble solids concentration (RSSC) and consumer acceptance of ripe Hayward kiwifruit RSSC range (%) Degree of liking a Like Neither like nor dislike Dislike 1998 growing season consumer acceptance (%) b ab a a P-value LSD growing season consumer acceptance (%) c b ab a P-value LSD a Degree of liking with 1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, dislike slightly; 5, neither like nor dislike; 6, like slightly; 7, like moderately; 8, like very much; 9, like extremely.

28 212 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) same for kiwifruit with RSSC 11.6% combined with high ( 1.17%), moderate ( %) and low ( 1.04%) RTA (Fig. 3). However, when kiwifruit had RSSC 11.6%, RTA played a significant role in consumer acceptance (Fig. 3 and Table 5). Kiwifruit with RSSC 11.6 and RTA 1.17% ( sour ) had a significantly lower degree of liking than kiwifruit with 11.6 and RTA 1.17%. Consumer degree of liking of kiwifruit with RSSC 11.6 and RTA 1.17% was not significantly different. Fig. 4. Relationship between Hayward kiwifruit ripe soluble solids concentration (RSSC), consumer degree of liking (1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, dislike slightly; 5, neither like nor dislike; 6, like slightly; 7, like moderately; 8, like very much; 9, like extremely), and percentage consumer acceptance, 1999 growing season. Table 5 Combined 1998 and 1999 seasons, degree of liking according to ripe soluble solids concentration (RSSC) means separation based on rank sums for multiple comparison for ripe Hayward kiwifruit Average 1998/1999 season RSSC level (%) b Degree of liking (score 1 9) a a Degree of liking with 1, dislike extremely; 2, dislike very much; 3, dislike moderately; 4, dislike slightly; 5, neither like nor dislike; 6, like slightly; 7, like moderately; 8, like very much; 9, like extremely. b Same letters within the column indicate no significant differences (P 0.001) between the means according to the critical values of differences among rank sums for multiple comparisons (Friedman multiple means comparison, Newell and MacFarlane, 1987) Season Kiwifruit with RSSC that ranged from 11.5 to 13.5% were always accepted by consumers (Fig. 4), but with different degrees of liking. Consumers liked these kiwifruit from moderately (6.2) to very much (7.3) and acceptance ranged from 72 to 90% (Fig. 4). There was no significant interaction between RTA and RSSC with degree of liking. This agreed with the fact that during this season, kiwifruit RTA never reached values higher than 1.08%. The neither like nor dislike option was selected by only a few consumers (Table 4). The percentage of consumers that chose the neither like nor dislike option decreased from 7.4 to 3.8% as RSSC increased. For kiwifruit which had equal to or higher than 13.2% RSSC, the percentage of consumers that chose the neither like nor dislike option was approximately 3.8. Two years of in store consumer tests indicated that consumers liked very much ripe kiwifruit with RSSC 12.5% (Table 5). However, consumers disliked kiwifruit with 11.6% RSSC and 1.17% RTA. Some vineyards had RSSC 12.5% before they reached the minimum maturity index of 6.5% HSSC. We propose 12.5% RSSC as a quality index for early harvested California Hayward kiwifruit; thus, kiwifruit in vineyards that reach 12.5% RSSC could be harvested and consumed early in the season. Determination of RSSC can be accomplished by subjecting fruit to warm ethylene (100 ppm) treatments. By following this ripening protocol (Crisosto, 1999; Ritenour et al., 1999), kiwifruit should ripen within 7 10 days. Fast prediction of RSSC is being investigated by using the relationship between dry

29 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 22 (2001) weight (DW) and RSSC (Crisosto, 1992; Slaughter and Crisosto, 1998), although the correlation between DW and RSSC has not been acceptable under Californian conditions. Further work on consumer acceptance and kiwifruit dry weight is in progress. This work also showed that kiwifruit picked before they reach 6.2% HSSC do not store well because they develop commercially important flesh breakdown by 3 months at 0 C. These results agree with earlier work done in New Zealand (Harman, 1981) and California (Mitchell et al., 1990). Based on this industry survey and others, it is clear that there is an opportunity for some specific growing areas to market high quality tasting Hayward kiwifruit early in the season. Acknowledgements We thank Dr Ann Noble and Paul Metheney for the help in the sensory evaluation work. This work was funded by the California Kiwifruit Commission. Fruit used in the survey were donated by the California growers. References Basker, D., Critical values of differences among rank sums for multiple comparisons. Food Technol. 42, Ben-Arie, R., Gross, J., Sonego, L., Changes in ripening parameters and pigment of the Chinese gooseberry during ripening and storage. Sci. Hortic. 18, Bruhn, C.M., Feldman, N., Garlitz, C., Harwood, J., Ivans, E., Marshall, M., Riley, A., Williamson, M., Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14, Crisosto, C.H., Predicting high quality kiwifruit. Calif. Grower 16, Crisosto, C.H., Optimum procedures for ripening kiwifruit. Management of fruit ripening. Postharvest Hortic. Ser. 9, Crisosto, C.H., Garner, D., Crisosto, G.M., Kaprielian, R., Kiwifruit preconditioning protocol. Acta Hortic. 444, Crisosto, G.M., Mitchell, F.G., Arpaia, M.L., Mayer, G., The effect of growing location and harvest maturity on the storage performance of Hayward kiwifruit. J. Am. Soc. Hortic. Sci. 109, Gorini, F., Lasorella, M., Sensory and objective evaluation of kiwifruit. Acta Hortic. 282, Harman, J., Kiwifruit maturity. Orchardist N. Z. 54, , 130. Lallu, N., Low temperature breakdown. Acta Hortic. 444, Lawes, G.S., Sawanobori, S., Maturation, ripening and storage of kiwifruit, Orchardist N. Z., August, p Lawless, H.T., Heymann, H., Acceptance and preference testing. In: Lawless, H.T., Heymann, H. (Eds.), Sensory Evaluation of Food, Principles and Practices. Chapman and Hall, New York, pp MacRae, E., Stec, M.G.H., Triggs, C., Effects of postharvest treatment on the sensory qualities of kiwifruit harvested at different maturities. J. Sci. Food Agric. 50, Matsumoto, S., Obara, T., Luh, B.S., Changes in chemical constituents of kiwifruit during postharvest ripening. J. Food Sci. 48, Mitchell, F.G., Mayer, G., Biassi, W., Gulli, D., Estimating kiwifruit maturity through total solids measurements. Report to the California Kiwifruit Commission, pp Newell, G.J., MacFarlane, J.D., Expanded tables for multiple comparison procedures in the analysis of ranked data. J. Food Sci. 52, Pehrson, J.E., Ivans, E.M., Variability in early season navel orange clone maturity and consumer acceptance. Proceedings of the 6th International Citrus Congress, pp Ritenour, M.A., Crisosto, C.H., Garner, D.T., Cheng, G.W., Zoffoli, J.P., Temperature, length of cold storage and maturity influence the ripening rate of ethylene-preconditioned kiwifruit. Postharvest Biol. Technol. 15, Robertson, J.A., Meredith, F.I., Characteristics of fruit from high- and low-quality peach cultivars. HortScience 23, Slaughter, D.C., Crisosto, C.H., Nondestructive internal quality assessment of kiwifruit using near-infrared spectroscopy. Semin. Food Anal. 3, Sozzi, A., Testoni, A., Youssef, J., Deluisa, A., Nardin, C., Conservazione dell Actinidia in atmosfera controllata. J. Ann. Ist. Sperimentale Valorizzazione Tecnol. Prod. Agric. 11, Young, H., Paterson, V.J., The effects of harvest maturity, ripeness and storage on kiwifruit aroma. J. Sci. Food Agric. 36,

30 Postharvest Biology and Technology 24 (2002) Understanding American and Chinese consumer acceptance of Redglobe table grapes Carlos H. Crisosto *, Gayle M. Crisosto Department of Pomology, Kearney Agricultural Center, Uni ersity of California at Da is, 9240 South Ri erbend A enue, Parlier, CA 93648, USA Accepted 8 September 2001 Abstract In-store consumer acceptance tests were performed on a group of 400 American consumers and 250 native Chinese consumers to determine the relationship between ripe soluble solids concentration (SSC) and titratable acidity (TA) on consumer acceptance of Redglobe grapes for different ethnic groups. For this, a group of Caucasian, Hispanic and Chinese-American consumers and a group of native Chinese consumers tasted monadically four individual berry half samples at room temperature with SSC ranging from 10 to 20% and TA from 0.30 to 1.80%. Grapes with SSC 16.1% were accepted by 70% of the American consumers and 47% of the Chinese consumers. However, the percentage of consumers that disliked these grapes was the same ( 21) for both ethnic groups. The difference in the percentages of Chinese and American consumers accepting the grapes was due to the neither like nor dislike category. Chinese consumers chose the neither like nor dislike category in approximately 34% of the cases, while approximately only 7% of the American consumers chose it. Our data confirmed that Redglobe consumer acceptance is highly related to SSC:TA ratio but within a given SSC and TA level. In contrast to American consumers, Chinese consumer acceptance was not related to SSC:TA ratio when TA 0.80%. Thus, our work indicates that high TA affects consumer acceptance in relation to SSC:TA ratio depending on ethnic background Elsevier Science B.V. All rights reserved. Keywords: Ethnic; Consumer acceptance; In-store consumer acceptance tests; Soluble solids concentration; Titratable acidity; Degree of liking 1. Introduction Consumers consider high quality fruit to be those with nice appearance, high nutritional value, and good taste (Bruhn et al., 1991). The * Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). relationship between consumer preference and fruit chemical composition has been studied in different commodities. Several researchers have established that consumer preference is well related with fruit soluble solids concentration (SSC), titratable acidity (TA) and SSC:TA ratio in kiwifruit (Gorini and Lasorella, 1990; Crisosto and Crisosto, 2001); peaches (Parker et al., 1991); cherries (Kappel et al., 1996; Drake and Fellman, /02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S (01)

31 156 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) ); grapes (Rodan, 1988; Nelson et al., 1963, 1972, 1973; Guelf-Reich and Safran 1971); and oranges (Pehrson and Ivans, 1988). Recently, it has been reported in mango that sugars and acid enhance human perception of specific flavor notes, including aromatics (Malundo et al., 2001). Controlled sensory studies on the relationship between SSC and TA fruit composition and consumer preferences were performed on Thompson Seedless and Perlette table grapes in California (Nelson et al., 1972, 1973), and in Israel (Guelf- Reich and Safran, 1971). Based on these studies, it was concluded that consumers were more sensitive to SSC:TA ratio than SSC. Furthermore, there is no information on the relationship between SSC and TA composition of Redglobe table grapes (seeded, dark red skin cultivar) and consumer acceptance. In addition, the potential role of ethnic background on table grape consumer acceptance has not been reported. Since production of California and Chilean Redglobe table grapes is still increasing, and these grapes are mainly consumed in Hong Kong and Taiwan, understanding the role of fruit SSC and/or TA on native Chinese consumer acceptance has become important. The availability of this information will help to maintain or expand both the American and Chinese markets long term. 2. Materials and methods 2.1. Consumer test A group of 400 American consumers consisting of Caucasian, Hispanic, and Chinese-American ethnic backgrounds were interviewed at two major supermarkets located in the Fresno area, California. The Caucasian and Hispanic consumers were interviewed at the Reedley Save Mart and the Chinese-American consumers were interviewed at the R-N Market in north Fresno with a predominantly Asian clientele of Chinese descent. The native Chinese consumer test was conducted at the Doong Kong Seafood Restaurant, Honolulu, Hawaii (caters to native Chinese tour groups) over 2 days. A group of 250 native Chinese consumers were interviewed. Of the Chinese consumers, 80 of them were sailors from two commercial vessels. Chinese interpreters were used during the interviews. The SSC and TA combinations of the Redglobe grape samples tested were selected based on previous seasons California (Dokoozlian et al., 1999) and Chilean (Crisosto unpublished data) SSC and TA surveys. In all of the tests, all samples were prepared away from the testing area prior to the consumer test. Firmness was measured non-destructively on each berry using a Durafel-10 (Copa Informatique, S.A., France) with a 2.3 mm diameter tip. Any berries with firmness outside the Durafel range were not used in these sensory evaluation tests. For each sample, the berry was cut in half longitudinally, i.e. from the stem end to the blossom end. Each half was placed in a labeled 44.3-ml soufflé cup. One half was used for tasting and the other for chemical analysis. The juice was extracted from one half of each berry (approximately, 6 g) and analyzed for SSC and TA expressed as percentage of tartaric acid, which is the predominant acid in this species. This allowed us to study the single consumer response relative to the SSC and TA. Each consumer, after being asked if he/she was willing to taste fresh table grapes in a University of California research project, was asked to indicate his/her age range on a chart and the interviewer recorded his/her gender and ethnic group. Each consumer that said he/she ate grapes was asked to taste monadically four one-half grape samples with the seeds intact. For each test, each consumer was presented the four proposed SSC:TA combinations of Redglobe table grapes in random order in coded 44.3 ml soufflé cups at room temperature. The consumer was asked if he/she liked, disliked, or neither liked nor disliked the sample. Then the consumer was asked his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The consumer s response was recorded using a 9-point hedonic scale (1 dislike extremely to 9 like extremely). In between samples the consumer was instructed to sip bottled water to cleanse his/her palate. Consumer acceptance was measured as both degree of liking and a percentage. The percentage of consumers liking the grape sample (consumer

32 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) acceptance) was calculated as the number of consumers liking the grape sample (score 5.0) divided by the total number of consumers within the sample (Lawless and Heymann, 1998). The percentage of consumers disliking the grape sample (score 5.0) was calculated as the number of consumers disliking the grape sample divided by the total number of consumers within the sample. The percentage of consumers that neither liked nor disliked the grape sample was calculated as the number of consumers that neither liked nor disliked the grape sample (score=5.0) divided by the total number of consumers within the sample. The degree of liking data was subjected to analysis of variance (ANOVA) prior to the least significant difference (LSD) means separation using the SAS program. 3. Results and discussion while grapes within the SSC:TA ratio range of 20.1 to 22.5 were accepted by 76% of the American consumers (Fig. 2B). American consumer acceptance decreased as TA increased in the berries although they were always liked (Fig. 3A and B). Redglobe grapes degree of liking data was divided into three groups according to a statistical means separation (Fig. 3A): grapes with TA 0.70% ( moderately ), grapes with TA between 0.71 and 0.80% ( slightly ), and grapes with TA 0.80% ( slightly, lowest consumer acceptance). Percentage consumer acceptance of grapes decreased from 84 to 59% as the TA increased (Fig. 3B). As SSC and TA affected degree of liking and consumer acceptance, the interaction between different SSC and TA levels on consumer acceptance was studied. American consumers liked grapes with TA 0.70% from slightly to moderately American consumer test For the American consumer test, individual Redglobe berry SSC varied from 10.0 to 18.8% and TA ranged from 0.36 to 1.40%. Degree of liking and percentage acceptance of Redglobe table grapes was related to SSC, TA and SSC:TA ratio. These relationships were not different between the ethnic groups interviewed in the American consumer test. Grapes with 12.1% SSC were accepted by American consumers but with different degrees of liking (Fig. 1A). Consumer acceptance increased from 62 to 88% according to SSC (Fig. 1B). American consumers degree of liking varied from like slightly for grapes within the % SSC range (average 66% consumer acceptance), to like moderately for grapes within 15.1 to 17.0% (average 84% consumer acceptance). Based on the SSC:TA ratio, American consumer acceptance of Redglobe grapes ranged from like slightly to moderately (Fig. 2A). Redglobe with a SSC:TA ratio greater than 25.0 were liked the most ( moderately ) and percentage consumer acceptance ranged from 67 to 92% (Fig. 2B). Percentage consumer acceptance for Redglobe with a SSC:TA ratio 20.1 reached 67%, Fig. 1. Degree of liking and percentage consumer acceptance of Redglobe table grapes at different levels of SSC by American (A) and (B) and Chinese (C) and (D) consumers. Degree of liking measured on 9-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; and 9, like extremely). Different letters within a given SSC range indicate a significant difference between means by LSD 0.05.

33 158 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) % also all had the same degree of liking (Table 1). Consumer acceptance of grapes at this level of TA varied from 58 to 88%. The percentage of consumers disliking these grapes varied from 32 to 7%. The neither like nor dislike option ranged from 5 to 10% of the consumers at this TA level (Table 1). American consumers liked Redglobe grapes with TA 0.80%, from neither like nor dislike to like moderately. Consumer acceptance of these grapes varied from 45 to 88%. Only 45% of American consumers accepted Redglobe grapes with SSC 14.1% and TA 0.80%. The percentage of consumers disliking these grapes varied from approximately 45 to 6%. The neither like nor dislike option ranged from 10 to 4% of the consumers at this TA level (Table 1). Fig. 2. Degree of liking and percentage consumer acceptance of Redglobe table grapes at different SSC:TA ratio ranges by American (A) and (B) and Chinese (C) and (D) consumers. Degree of liking measured on 9-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; and 9, like extremely). Different letters within a given SSC range indicate a significant difference between means by LSD Degree of liking was significantly higher for grapes with SSC 15.1% than for grapes with SSC 15.1%. There was a significant difference in the degree of liking of grapes in the SSC range 14.1 to 15.0% and SSC range 14.1% (Table 1). Consumer acceptance of grapes at this level of TA in relation to SSC ranged from 73 to 89%. The percentage of consumers disliking these grapes varied from 18 to 7%. The neither like nor dislike option ranged from 3 to 9% of the consumers at this TA level (Table 1). American consumers liked Redglobe grapes with TA between 0.70 and 0.80%, from slightly to moderately. Grapes with SSC 15.1% had the highest degree of liking within this TA range. Grapes within the SSC ranges below 15.1% all had the same degree of liking. Grapes within the SSC ranges between 14.1 Fig. 3. Degree of liking and percentage consumer acceptance of Redglobe table grapes at different TA ranges by American (A) and (B) and Chinese (C) and (D) consumers. Degree of liking measured on 9-point hedonic scale (1, dislike extremely; 5, neither like nor dislike and 9, like extremely). Different letters within a given SSC range indicate a significant difference between means by LSD 0.05.

34 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) Table 1 Degree of liking and percentage consumer acceptance of Redglobe table grapes by American consumers at different levels of SSC and TA measured as percentage tartaric acid SSC range Mean SSC (%) (%) TA TA 0.80 TA 0.80 Degree of Acceptance Neither Dislike Degree of Acceptance Neither Dislike Degree of Acceptance Neither Dislike liking a (%) like nor (%) liking (%) like nor (%) liking (%) like nor (%) (1 9) dislike (%) (1 9) dislike (%) (1 9) dislike (%) c b c c b bc b ab ab ab a a a a NA c NA NA NA NA NA NA NA LSD P-value a Degree of liking measured on 9-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; and 9, like extremely). b Same letters within the same column indicate no significant difference between means. c NA, data not available due to an insufficient number of responses.

35 160 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) Chinese consumer test For the Chinese consumer test, individual Redglobe berry SSC varied from 12.0 to 20.0% and TA ranged from 0.30 to 1.80%. Chinese consumer degree of liking and percentage acceptance of Redglobe table grapes was related to SSC, TA and SSC:TA ratio. Grapes with 12.1% SSC were accepted by Chinese consumers but with different degrees of liking (Fig. 1C). Consumer acceptance increased from 45 to 89% depending on SSC (Fig. 1D). Chinese consumers liked slightly grapes within the % SSC range (average 50% consumer acceptance); liked moderately grapes within the 15.1 to 17.0% SSC range (average 78% consumer acceptance); and liked moderately to very much grapes within 17.0% SSC range (89% consumer acceptance). Based on the SSC:TA ratio, Chinese consumers liked Redglobe grapes from neither like nor dislike to moderately (Fig. 2C). Grapes were liked moderately with ratios Grapes were liked slightly within the SSC:TA ratio range of , and neither liked nor disliked when they had SSC:TA ratios Overall, percentage acceptance ranged from 37 91% (Fig. 2D). Only 37% of Chinese consumers accepted Redglobe grapes with a SSC:TA ratio 20.1 and 42% of Chinese consumers when the grapes had a SSC:TA ratio in the range of (Fig. 2D). Chinese consumer acceptance decreased as TA increased in the berries although they were always liked; except when the grapes had a TA 0.80% (Fig. 3C and D). Redglobe grapes degree of liking data was divided into three groups according to a statistical means separation (Fig. 3C): grapes with TA 0.60% ( moderately ); grapes with TA between 0.61 and 0.80%( slightly ); and grapes with TA 0.80% ( neither like nor dislike ). TA significantly affected consumer acceptance of Redglobe grapes; as TA increased, consumer acceptance decreased from 80 to 25% (Fig. 3D). SSC and TA affected degree of liking and consumer acceptance the same as for American consumers, so the interaction between different SSC and TA levels on consumer acceptance was studied. Chinese consumers liked Redglobe grapes with TA 0.60% from slightly to moderately. Degree of liking was significantly higher for grapes with 16.1% SSC ( moderately ) than for grapes with 16.1% ( slightly to moderately ). There were no significant differences between the degrees of liking of grapes at the SSC ranges 16.0% (Table 2). Consumer acceptance of grapes with TA 0.60% in relation to SSC level ranged from 57 to 88% (Table 2). The percentage of consumers disliking these grapes varied from 13 to 1% while the neither like nor dislike option ranged from 32 to 11% of the consumers (Table 2). Chinese consumers liked Redglobe grapes with TA between 0.60 and 0.80% from slightly to moderately. Grapes with SSC 16.1% had the highest degree of liking within this TA range. Grapes within the SSC ranges less than 16.1% all had the same degree of liking. Also, grapes in the SSC ranges between 15.1 and 17.0% all had the same degree of liking. The percentage of consumers disliking these grapes varied from 20 to 5%, while the neither like nor dislike option ranged from 36 to 15% of the Chinese consumers (Table 2). Consumer acceptance of grapes with TA between 0.60 and 0.80% varied from 45 to 77%. At the same level of SSC, Chinese consumer acceptance of Redglobe grapes was higher with TA 0.60% than for grapes within the 0.60 to 0.80% TA range (Table 2). Redglobe grapes with 16.1 SSC were accepted by 70% of the American consumers and 47% of the Chinese consumers. If we use the mean SSC for the SSC range to calculate a SSC:TA ratio of 20 within the three TA levels, American consumer liking and acceptance did not change across TA level. The degree of liking ranged from 6.4 to 6.7 and acceptance from 83 to 79%. In contrast, Chinese consumer acceptance decreased if TA increased with a SSC:TA ratio of 20. The degree of liking varied from 6.2 to 5.0 and acceptance from 57 to 20%. For both American and Chinese consumers, grapes with 16.0% SSC, degree of liking varied according to TA level indicating that consumer acceptance is more sensitive to the SSC:TA ratio than SSC alone. However, for grapes with 16.0% SSC, TA level did not affect acceptance.

36 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) Table 2 Degree of liking and percentage consumer acceptance of Redglobe table grapes by Chinese consumers at different levels of SSC and TA measured as percentage tartaric acid SSC range Mean TA 0.60 (%) SSC (%) 0.60 TA 0.80 TA 0.80 Degree of Acceptance Neither Dislike Degree of Acceptance Neither Dislike Degree of Acceptance Neither Dislike liking a (%) like nor (%) liking (%) like nor (%) liking (%) like nor (%) (1 9) dislike (%) (1 9) dislike (%) (1 9) dislike (%) c b b c b bc ab ab a NA NA c NA NA a NA NA NA NA NA NA NA NA LSD NS d P-value a Degree of liking measured on 9-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; and 9, like extremely). b Same letters within the same column indicated no significant difference between means. c NA, data not available due to an insufficient number of responses. d NS, no significant differences.

37 162 C.H. Crisosto, G.M. Crisosto / Posthar est Biology and Technology 24 (2002) Changes in consumer acceptance due to TA levels were more sensitive for Chinese than American consumers. For example, Chinese consumers liked slightly to moderately grapes with TA 0.60%, while American consumers liked slightly to moderately grapes with 0.70% TA. Chinese consumers neither liked nor disliked grapes with TA 0.80% despite their SSC level. Chinese consumers liked Redglobe grapes moderately (score 6.5) when they had 14.1% SSC and 0.60% TA. In the TA range of 0.60 to 0.80%, a SSC of 15.1% was needed to maintain this level of liking (score 6.5). American and Chinese consumers chose the neither like nor dislike option in different ways. American consumers chose the neither like nor dislike option only a few times (average 7%), whereas Chinese consumers chose it frequently (average 34%). Our data confirmed that Redglobe consumer acceptance is more sensitive to SSC:TA ratio than SSC alone as has been reported for Perlette and Thompson Seedless grown in California and Israel (Nelson et al., 1963, 1972, 1973; Guelf-Reich and Safran, 1971). However, the relationship between consumer acceptance and SSC:TA ratio is highly related to consumer acceptance within a TA or SSC range. For example, Redglobe consumer acceptance for both American and Chinese consumers was not related to SSC:TA ratio when SSC was 16.1%. In spite of the SSC:TA ratio, Chinese consumer acceptance was low for grapes with TA 0.80%. This is the first report that points out that Chinese consumers are more sensitive to TA than the SSC:TA ratio. Our work also demonstrates the importance to evaluate the dislike, like, and neither like nor dislike options to understand consumer acceptance. Acknowledgements We would like to thank Dr Ann Noble for her help in the planning of the sensory evaluation work. References Bruhn, C.M., Feldman, N., Garlitz, C., Harwood, J., Ivans, E., Marshall, M., Riley, A., Williamson, M., Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14, Crisosto, C.H., Crisosto, G.M., Understanding kiwifruit consumer acceptance. Postharvest Biol. Technol. 22, Dokoozlian, N.K., Boling, E., Luvisi, D., Ebisuda, N., Influence of fruit maturity on the fresh and postharvest quality of Redglobe table grapes. Research summary to the California Table Grape Commission, pp. 19. Drake, S.R., Fellman, J.K., Indicators of maturity and storage quality of Ranier sweet cherry. HortScience 22, Gorini, F., Lasorella, M., Sensory and objective evaluation of kiwifruit. Acta Hortic. 282, Guelf-Reich, S., Safran, B., Indices of maturity for table grapes as determined by variety. Volcani Institute of Agric. Research, Bet-Dagan, Israel. XVIII International Horticulture Congress, Tel-Aviv, Israel, March Series No E, pp Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. HortScience 31, Lawless, H.T., Heymann, H., Acceptance and preference testing. In: Lawless, H.T., Heymann, H. (Eds.), Sensory Evaluation of Food, Principles and Practices. Chapman & Hall, New York, pp Malundo, T.M.M., Shewfelt, R.L., Ware, G.O., Baldwin, E.A., Sugars and acids influence flavor properties of mango (Mangifera indica). J. Am. Soc. Hort. Sci. 126, Nelson, K.E., Schutz, H.G., Ahmedullah, M., McPherson, J., Flavor preferences of supermarket customers for Thompson Seedless grapes. Am. J. Enol. Viticult. 24, Nelson, K.E., Allen, J.W., Schutz, H.G., Effect of grape maturity, sample order, and sex of the taster on the flavor response of supermarket customers. Am. J. Enol. Viticult. 23, Nelson, K.W., Baker, G.A., Winkler, A.J., Amerine, M.A., Richardson, H.B., Jones, F.R., Chemical and sensory variability in table grapes. Hilgardia 34, Parker, D., Ziberman, D., Moulton, K., How quality relates to price in California fresh peaches. Calif. Agric. 45, Pehrson, J.E., Ivans, E.M., Variability in early season navel orange clone maturity and consumer acceptance. Proceedings of the International Society for Citriculture, pp Rodan, L., Consumer acceptance of Florida grapes and grape products. Proceedings of the Viticulture Science Symposium Tallahassee, FL. Florida A&M University, Center for Viticultural Sciences, pp

38 Postharvest Biology and Technology 24 (2002) Testing the reliability of skin color as an indicator of quality for early season Brooks (Prunus a ium L.) cherry Carlos H. Crisosto a, *, Gayle M. Crisosto a, Mark A. Ritenour b a Department of Pomology, Kearney Agricultural Center, Uni ersity of California, Da is, 9240 South Ri erbend A e., Parlier, CA 93648, USA b Indian Ri er Research and Education Center, Uni ersity of Florida, 2199 S. Rock Road, Fort Pierce, FL , USA Accepted 8 September 2001 Abstract During the seasons, we investigated the relationship between Brooks cherry skin color at harvest (full light red, 50% bright red, full bright red and full dark red) and consumer acceptance using fruit grown in different geographic locations in the San Joaquin Valley (SJV). Soluble solids concentration (SSC) increased, but titratable acidity (TA) levels did not decrease as cherries matured from the full light red to full dark red skin color. The perception of sweetness, sourness and cherry flavor intensity by a trained taste panel was highly correlated to skin color, SSC and SSC:TA at harvest. There were no differences in the level of correlation between SSC or SSC:TA and the perception of sweetness, sourness or cherry flavor by trained judges. In-store consumer tests indicated that Brooks cherries with SSC 16.1% had the highest consumer acceptance (ca %) and cherries with SSC 16.0%, the lowest (ca. 48%). Gender and ethnicity (Caucasian, Asian American, Hispanic, or African American) did not affect American consumer acceptance of Brooks cherries Elsevier Science B.V. All rights reserved. Keywords: Sensory evaluation; In-store consumer acceptance; Demographics; Soluble solids concentration; Orchard survey 1. Introduction Early May cherry shipments to Japan, Hong Kong and within the USA have become increasingly attractive to California cherry growers and shippers because of the high revenue returns. In the past, cherry production has been restricted to growing areas with mild summers; cherries grown in areas such as the San Joaquin Valley (SJV), * Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). with hot summers traditionally produce excessive double or spur fruits. However, in the last 10 years, new cherry cultivars such as Brooks, Tulare, Garnet, and King that are not susceptible to double or spur fruits have been developed in the SJV. These cultivars are normally harvested the last week in April or the first week in May in California, and marketed to Asian countries and different areas within the USA. Of these cultivars, Brooks is not a full dark red cherry in contrast to Tulare, King, Early Burlat and Bing. Brooks was selected in the University of California cherry-breeding program among the progeny /02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S (01)

39 148 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) of a cross between two commercial sweet cherry cultivars Ranier (white flesh) and Early Burlat (Hansche et al., 1988). It is characterized as a very high quality, early maturing sweet cherry that possesses the outstanding ability to develop large fruit. In general, Brooks ripens 5 7 days after Early Burlat and about 7 12 days before Bing. A better understanding of the relationships between cherry quality attributes and consumer acceptance is needed to assure that Brooks cherries are harvested with the quality that meets consumer demands without loss of potential postharvest life. Controlled trained panel sensory studies on the relationship between cherry quality attributes and sensory quality perception have been performed on other cherry cultivars such as Bing (Dever et al., 1996; Drake and Fellman, 1987; Drake et al., 1989; Guyer et al., 1993; Schotzko, 1993; Kappel et al., 1996). However, no in-store consumer tests have been reported. Skin color is a practical and simple indicator with which to instruct harvesters on what to harvest. Therefore, it is important to establish a relationship between skin color and/or soluble solids concentration (SSC) on consumer acceptance to identify the skin color that cherries should be harvested to assure minimum taste consumer acceptance. The objective of this study was to understand the relationship between consumer acceptance and SSC or skin color for early harvested Brooks cherries. To meet this objective, three steps were followed. First, an industry-wide survey of Brooks cherry quality attributes was conducted over 4 years for cherries picked at different skin colors. Second, a trained taste panel was used to determine if SSC is a key component of Brooks cherry sensory quality. Third, based on data from the first two steps, an in-store consumer acceptance test was conducted over 3 years. 2. Materials and methods 2.1. Industry sur ey During the 1995 through 1998 seasons, Brooks cherries were harvested at four different maturity stages based on skin color: full light red (Hue= 26.15, L=41.35, C=42.30), 50% bright red (Hue=21.96, L=36.24, C=37.69), full bright red (Hue=16.90, L=32.54, C=30.79), and full dark red (Hue=11.85, L=29.11, C=23.77) from five orchards in the SJV. Fruit at the four different skin colors were harvested from five healthy trees in each orchard. A completely randomized design with five replicates of 10 fruit per maturity stage was used to evaluate quality attributes. Quality attributes measured at harvest for each skin color included SSC, firmness (F), titratable acidity (TA), weight, ph and color. SSC, ph, and TA were measured on juice extracted with a food press and filtered through cheesecloth. SSC was measured with a temperature compensated refractometer previously calibrated with distilled water. TA was measured by adding 5 g of sample juice to 50 ml of distilled water and titrating with 0.1 N sodium hydroxide (NaOH) to an end point of ph 8.2. TA is expressed as percent of malic acid, which is the predominant acid in this species. Average color was evaluated according to the Commission Internationale de l Eclairage (CIE), with a Minolta colorimeter (Minolta, CR-200, Japan). Cherry skin color is expressed as hue angle (h ), lightness (L) and chroma (saturation). The hue angle is expressed in degrees and is a measure of color that, for example, from 0 to 90 spans from red to orange to yellow. Lightness separates color into bright and dark. Chroma is a measure of color intensity with low values representing dull colors and high values representing vibrant colors. F was measured using a U.C. firmness tester with a 3 mm tip. Skin from opposite cheeks of each fruit was removed and F calculated as the average of two measurements per fruit expressed as Newtons. Data were subjected to ANOVA and correlation analysis. Means were separated using LSD means separation at the 5 or 1% level using the SAS statistical software (SAS Institute, Cary, NC) Sensory e aluation Trained panel During the 1996 season, a trained panel of 15 judges was selected for their taste acuity (O Ma-

40 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) hony, 1986). Sensory perception of sweetness, sourness and cherry flavor intensity was evaluated on Brooks cherries picked at different skin colors as previously described on fruit collected from two of the same orchards used for the fruit quality evaluations. Physical measurements of weight, SSC, ph, TA, and F were obtained from the corresponding fruit quality evaluations. Each judge tasted three replicates from each skin color per orchard. Samples consisted of two whole cherries at room temperature (20 C) presented in random order in individual booths illuminated by fluorescent lighting. To minimize potential color differences judges were instructed to wear dark green glasses during the tasting. Judges scored cherries by circling a hatch mark placed at increments of 0.5 cm on a 10 cm horizontal line anchored 1 cm from both ends of the line by less and more for each sensory attribute (sweetness, sourness, and cherry flavor intensity). Data were subjected to ANOVA and correlation analysis. Means were separated using LSD means separation at the 5 and 1% levels using the SAS statistical software (SAS Institute, Cary, NC) In-store consumer test Groups of 100, 400, and 150 consumers representing a diverse combination of ages, ethnic groups and genders were tested at two major supermarkets in Fresno County, California during the 1997, 1998 and 1999 seasons, respectively. Based on our industry-wide survey, each consumer was presented four Brooks cherry samples to test the four targeted skin colors: full light red, 50% bright red, full bright red, and full dark red. Each consumer was asked if he/ she ate fresh cherries; the interviewer would then note their gender and ethnic group. Then, the consumer was instructed to wear dark glasses (to mask the skin color) during the tasting. Each consumer was presented, in random order, the four cherry half samples without pits in coded 88.7 ml soufflé cups at room temperature (20 C). Each consumer was instructed to sip bottled water between samples to cleanse his/her palate. The interview process for all three seasons varied. In 1997, the consumer was asked if he/ she liked or disliked the sample. In 1998, the consumer was asked if he/she liked or disliked the sample or was not sure. In 1999, the consumer was asked if he/she liked, disliked or neither liked nor disliked the sample. Then the consumer was asked to indicate his/her degree of liking or disliking: slightly, moderately, very much or extremely. In all 3 years, SSC was measured on the second half of each cherry sample tasted as previously described. In 1997 and 1998, consumer acceptance was measured as a percentage. Percentage of consumers liking the cherry sample was calculated as the number of consumers liking the sample divided by the total number of consumers within the sample. Percentage of consumers disliking the cherry sample was calculated as the number of consumers disliking the cherry sample divided by the total number of consumers within the sample. In 1998, the percentage of consumers indicating not sure about the cherry sample liking or disliking was calculated as the number of consumers not sure divided by the total number of consumers within the sample. In 1999, consumer acceptance was measured as both a degree of liking and a percentage. Percentage of consumers liking the cherry sample was calculated as the number of consumers liking the cherry sample (score 5.0) divided by the total number of consumers within the sample (Lawless and Heymann, 1998). Percentage of consumers disliking the cherry sample (score 5.0) was calculated as the number of consumers disliking the cherry sample divided by the total number of consumers within the sample. Percentage of consumers that neither liked nor disliked the cherry sample (score=5.0) was calculated as the number of consumers that neither liked nor disliked the cherry sample divided by the total number of consumers within the sample. For all 3 years, data was subjected to analysis of variance (ANOVA) prior to the Least Significant Differences (LSD) mean separation using the SAS program.

41 150 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) Results and discussion 3.1. Industry sur ey In general, fruit weight, SSC, and SSC:TA increased as cherry skin color changed from full light red to full dark red (Table 1). The largest increase in fruit weight occurred between the light red and 50% bright red skin color. SSC increases were relatively evenly distributed between the different skin colors, increasing by about 1.0% SSC with each successive maturity stage. TA did not change as Brooks cherries changed color from full light red to full bright red. However, TA decreased form 0.89 to 0.77% when cherries changed color from full bright red to full dark red. Thus, SSC:TA increased greatly when cherry skin color changed from bright red to dark red. This lack of TA decrease during final cherry maturation was previously observed in Brooks, Tulare and King cherries growing in the SJV (Crisosto et al., 1993, 1997). Fruit ph remained relatively unchanged during cherry maturation and among orchards. Fruit quality attributes of weight, SSC, TA, and SSC:TA also varied significantly among orchards (Table 1). The large range in the SSC:TA (18 28) among orchards was mainly due to differences in TA. Year also influenced fruit quality attributes. Of the 4 years, cherries grown in 1995 were among the largest (8.0 g), had the highest SSC:TA (29.1), had the lowest TA (0.59%), and one of the lowest SSC (16.6%). During the 1996 season, SSC and TA were the highest (19.4 and 1.20%, respectively), and SSC:TA and fruit weight were the lowest (16.6 and 7.5 g, respectively). During the 4-year evaluation period, there were interactions between skin color, orchard and year for most of the quality attributes evaluated. This indicates that at a given skin color, fruit quality Table 1 Effect of skin color, orchard and year on Brooks cherry quality attributes measured at harvest ( ) Sources of variability Fruit weight (g) SSC (%) TA (%) SSC:TA ph Skin color (SC) ** ** ** ** * Full light red 7.1d 15.3d 0.81c 20.3d 3.9a 50% Bright red 8.0c 16.8c 0.85b 21.5c 3.9ab Full bright red 8.2b 18.7b 0.89a 22.6b 3.9ab Full dark red 8.5a 20.4a 0.77d 27.6a 3.8b Orchard (Or) ** ** ** ** ** No e 17.6b 0.85b 22.5c 3.8 No b 17.0cd 0.79c 22.2c 3.9 No a 17.9a 1.00a 17.9d 3.8 No c 16.9d 0.63d 28.4a 4.0 No d 17.2c 0.79c 23.2b 3.9 Year (Y) ** ** ** ** NS a 16.6c 0.59d 29.1a c 19.4a 1.20a 16.6d b 17.4b 0.97b 18.1c a 16.5c 0.72c 23.2b 3.8 Interactions SC*Or * ** * ** NS SC*Y ** * ** ** * Or*Y ** ** ** ** ** SC*Or*Y ** ** * ** NS Values within each column and source of variability followed by different letters are significantly different according to the LSD means separation. NS, no significant differences; * and ** significant differences at the 0.05 and 0.01 levels, respectively. No letters within a column indicates no significant differences between treatments.

42 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) Table 2 Quality attributes according to the interaction between orchard and skin color for Brooks cherries during the 1997 season Weight (g) Firmness a (N) SSC (%) TA (%) SSC:TA Orchard skin color No. 1 Full light red 5.6c b 2.02a 16.8d 1.22c 13.7c 50% Bright red 6.4b 1.81ab 18.8c 1.27b 14.8b Full bright red 7.2a 1.63b 21.3b 1.30ab 16.4a Full dark red 7.2a 1.75b 22.2a 1.35a 16.4a No. 2 Full light red 8.1c 2.13c 18.8d 1.00c 18.8c 50% Bright red 9.2a 2.59b 22.2c 1.08ab 20.6b Full bright red 8.6b 2.67b 23.2b 1.05bc 22.1a Full dark red 8.6bc 2.99a 25.3a 1.12a 22.6a No. 3 Full light red 7.6c 2.64a 18.1c c 50% Bright red 8.1b 2.17b 18.8c c Full bright red 8.3ab 1.94c 19.9b b Full dark red 8.5a 1.95c 21.6a a No. 4 Full light red 6.8b d 1.15b 16.3c 50% Bright red 7.5a c 1.19ab 16.9c Full bright red 7.8a b 1.19ab 18.3b Full dark red 7.1b a 1.23a 19.7a No. 5 Full light red 6.5c d 1.05b 16.0c 50% Bright red 7.2b c 1.08b 16.6c Full bright red 7.5ab b 1.14a 17.5b Full dark red 8.0a a 1.14a 20.4a a Flesh F measured with a 3 mm tip. b Values within each column and source of variability (orchard skin color) followed by different letters are significantly different according to the LSD means separation. NS, no significant differences; * and ** significant differences at the 0.05 and 0.01 levels, respectively. No letters within a column indicates no significant differences between treatments. attributes will be different depending on the orchard and year. Thus, a given skin color cannot assure a minimum SSC and/or SSC:TA. For example, in 1997, SSC measured in cherries with 50% bright red skin color from orchard No. 2 was higher than the SSC in cherries with the full bright red skin color from orchards No. 1, 3, 4 and 5 (Table 2). Full bright red cherries from orchard No. 2 had a SSC equal to or higher than the SSC for full dark red cherries from orchards No. 1, 3, and 5. A similar situation occurred for the other quality attributes tested such as SSC:TA, fruit weight, and F Sensory e aluation Trained panel Trained judges were able to detect differences in sweetness, sourness and cherry flavor intensity between different skin colors of Brooks cherries (Table 3). Cherries at each successive skin color, from full light to full dark red, were progressively sweeter and had more cherry flavor intensity. Cherries harvested at the full light red color had the highest perception of sourness and cherries harvested at the full dark red color, the lowest. Trained judges did not detect differences in sourness between the 50% bright red and the full bright red skin colors. All of the physical and chemical quality attributes were correlated to sensory sweetness, sourness and cherry flavor intensity of Brooks cherries. Of these attributes, skin color, SSC, and SSC:TA had the highest positive correlation with sweetness and cherry flavor intensity and the highest negative correlation with sourness (Table 4). There were no differences in perception of fruit quality sensory attributes based on either SSC or SSC:TA; in all of these

43 152 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) Table 3 Relationship between skin color and perception of sensory quality attributes evaluation for Brooks cherry Sweetness Sourness Cherry flavor intensity Full light red 3.8a a 5.4a 3.7a 50% Bright red 4.7b 4.7b 4.6b Full bright red 5.5c 4.3b 5.5c Full dark red 6.6d 3.5c 6.4d P-value LSD a Scores ranged from 0.5 to 9.5 in 0.5 increments on a 10-cm horizontal line anchored 1 cm from both ends by less (1) and more (9). Values within each column followed by the same letters are not significantly different according to the LSD means separation at P cases, the correlation coefficients (r) were highly significant between and As TA measurements involve careful laboratory analysis, we recommend only the use of SSC measurements Consumer test In all 3 years, as the SSC increased the percent consumer acceptance increased. Combining data, consumer acceptance was related to cherry SSC and there was no interaction between year and SSC on consumer acceptance (Table 5). Based on consumer acceptance, fruit SSC was segregated into three groups. Brooks cherries Table 4 Correlation coefficients (r) between physical and chemical attributes and sensory quality attributes for Brooks cherry Sweetness Sourness Cherry flavor intensity Skin color 0.958*** 0.904*** 0.983*** Weight 0.512** 0.123** 0.559** SSC 0.842*** 0.824*** 0.904*** SSC:TA 0.842*** 0.932*** 0.890*** TA 0.344** 0.072** 0.404** ph 0.185** 0.276** 0.188** F 0.584** 0.835*** 0.614** L 0.910*** 0.838*** 0.956*** Hue 0.902*** 0.906*** 0.951*** Chroma 0.976*** 0.857*** 0.995*** **, *** Significant at 1 and 0.1% levels, respectively. Table 5 Relationship between SSC and consumer acceptance for Brooks cherries measured during three consecutive in-store consumer tests Treatments SSC (%) a a b b LSD Year SSC year NS Acceptance (%) a Values within each column and source of variability followed by different letters are significantly different according to the LSD means separation at P No letters within a column indicates no significant differences between treatments. with SSC 20.1% had the highest consumer acceptance (ca. 93%) and cherries with SSC 16.0%, the lowest (ca. 48%). Cherries within the % SSC range were accepted by approximately 83% of consumers. Consumer acceptance was related to cherry skin color as has been reported in other commodities (Bruhn et al., 1991). In this case, there was interaction between skin color and year on consumer acceptance (Table 6). In all 3 years, cherries harvested at the full light red skin color had the lowest consumer acceptance (ca. 33%). Consumer acceptance for cherries harvested at the 50% bright red color reached approximately 76%, except for the 1998 season that only attained 53% consumer acceptance. For cherries harvested at the full bright red and full dark red skin colors, consumer acceptance was approximately 82 and 91%, respectively. In general, full dark red cherries had higher consumer acceptance than full bright red cherries. The interaction between skin color and year on consumer acceptance can be explained by the variability due to year/orchard on the relationship between SSC and skin color. The reason why consumer acceptance of the 1998 cherries harvested at the 50% bright red color was lower than for the same skin color in 1997 and

44 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) Table 6 Interaction between skin color and year on Brooks cherry consumer acceptance measured during a 3-year period Skin color year Acceptance (%) SSC (%) Full light red a a a % Bright red a 74.7a b a 15.5 Full bright red a a a 17.5 Full dark red a a a 21.6 a Values within each column and source of variability (skin color year) followed by different letters are significantly different according to the LSD means separation at P No letters within a column indicates no significant differences between treatments may be explained in part because the cherries had only 14.1% SSC (1998) compared with 18.4 and 15.5% in 1997 and 1999, respectively. Other sensory quality attributes such as cherry flavor may also have been a factor. In the 3 years of these in-store consumer tests, Brooks cherries harvested at the full bright red skin color assured acceptance by approximately 80% of the consumers. In general, fruit weight and SSC increased as skin color darkened but postharvest life decreased (Crisosto et al., 1993, 1997). Based on our 4-year data, only cherries with a minimum SSC 16.0% should be harvested to assure consumer satisfaction. In most of the cases, Brooks cherry, which is not a full dark red cherry like Bing, reaches a minimum of 16% SSC before it develops a full bright or dark red skin color. We recommend growers measure SSC in Brooks cherries at different skin colors. According to their specific orchard and growing season relationship, growers can instruct harvesters as to which skin color they should be picking. However, as the consumer decision in many markets is currently based on full dark red color, promotional and educational programs to demonstrate that high consumer acceptance of Brooks cherries is independent of full dark red color development should be pursued. Using this proposed harvesting protocol and educational programs will allow Brooks orchards with this high SSC potential to be harvested before they develop a full dark red skin color. Acknowledgements We thank Anne Noble for her help and advice for the sensory work. The California Cherry Growers Association funded this work. Fruit used in the survey and in the sensory evaluations were donated by Chris Crossland and Jeff Hildebrand. References Bruhn, C.M., Feldman, N., Garlitz, C., Harwood, J., Ivans, E., Marshall, M., Riley, A., Williamson, D., Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14, Crisosto, C.H., Garner, D., Doyle, J., Day, K.R., Relationship between fruit respiration, bruising susceptibility, and temperature in sweet cherries. Hortscience 28 (2), Crisosto, C.H., Garner, D., Crisosto, G.M., Wiley, P., Southwick, S., Evaluation of the minimum maturity index for new cherry cultivars growing in the San Joaquin Valley. California Cherry Growers Association, Visalia, CA, 21 pp. Dever, M.C., MacDonald, R.A., Cliff, M.A., Lane, W.D., Sensory evaluation of sweet cherry cultivars. Hortscience 31, Drake, S.R., Fellman, J.K., Indicators of maturity and storage quality of Ranier sweet cherry. Hortscience 22, Drake, S.R., Williams, M.W., Fountain, J.B., Stemless sweet cherry (Prunus a ium L.) fruit quality and consumer purchase. J. Food Qual. 11, Guyer, D.E., Sinha, N.K., Chang, T.S., Cash, J.N., Physiochemical and sensory characteristics of selected Michigan sweet cherry (Prunus a ium L.) cultivars. J. Food Qual. 16, Hansche, P.E., Beres, W., Doyle, J., Micke, W.C., Brooks sweet cherry. Hortscience 23, 644.

45 154 C.H. Crisosto et al. / Posthar est Biology and Technology 24 (2002) Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. Hortscience 31, Lawless, H.T., Heymann, H., Acceptance and preference testing. In: Lawless, H.T., Heymann, H. (Eds.), Sensory Evaluation of Food, Principles and Practices. Chapman and Hall, New York, pp O Mahony, M., Sensory Evaluation of Food. Marcel Dekker, Inc, New York. Schotzko, R.T., Fresh sweet cherry eating characteristics: some baseline data. Washington State Univ. Res. Bull. XB1028, pp

46 Postharvest Biology and Technology 44 (2007) Segregation of plum and pluot cultivars according to their organoleptic characteristics Carlos H. Crisosto a,, Gayle M. Crisosto a, Gemma Echeverria b, Jaume Puy c a Department of Plant Sciences, University of California, Davis, CA 95616, USA 1 b Centre UdL-IRTA, Rovira Roure, 191, Lleida 25198, Spain c Departament de Química, Rovira Roure, 191, Lleida 25198, Spain Received 19 June 2006; accepted 9 December 2006 Abstract Cultivar segregation according to the sensory perception of their organoleptic characteristics was attempted by using trained panel data evaluated by principal component analysis of 12 plum and four pluot cultivars as a part of our program to understand plum minimum quality. The perception of the four sensory attributes (sweetness, sourness, plum flavor intensity, plum aroma intensity) was reduced to three principal components, which accounted for 98.6% of the variation in the sensory attributes of the tested cultivars. Using the Ward separation method and PCA analysis (PC1 = 49.8% and PC2 = 25.6%), plum and pluot cultivars were segregated into groups (tart, plum aroma, and sweet/plum flavor) with similar sensory attributes. Fruit source significantly affected cultivar ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA), but it did not significantly affect sensory perception of plum flavor intensity, sourness, sweetness, and plum aroma intensity by the trained panel on fruit harvested above their physiological maturity. Based on this information, we recommend that validation of these organoleptic groups should be conducted using in store consumer tests prior to development of a minimum quality index within each organoleptic group based on ripe soluble solids concentration (RSSC). This organoleptic cultivar classification will help to match consumer preferences and enhance current promotion and marketing programs Elsevier B.V. All rights reserved. Keywords: Sensory attributes; PCA; Ripe fruit quality attributes; Minimum quality index 1. Introduction In the last decade, plum per capita consumption has remained the same or even decreased in the USA (Anon., 2004) and some European countries (Liverani et al., 2002). In the USA, plum consumption is lower than in European countries declining from 0.64 kg per capita per year in 2001 to 0.50 kg per capita per year in In Italy, plum consumption varies from 1.2 to 0.90 kg per capita per year according to geographic location; for example, 1.2 kg for Italian consumers that live in the central area, 1.14 kg for consumers in the northwestern area and 0.90 kg for consumers in the south and islands (Macchi, 2006). On both continents, consumers complaints included off flavor, lack of ripening, astringent, flesh browning and textural characteris- Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). 1 Located at Kearney Agricultural Center, 9240 S. Riverbend Avenue, Parlier, CA 93648, USA. tics associated with low quality and chilling injury symptoms (Ceponis and Cappellini, 1987; Streif, 1989; Taylor and Jacobs, 1993; Taylor and Rabe, 1995; Abdi et al., 1998; Crisosto et al., 1999; Plich, 1999; Argenta et al., 2003). At the same time, costs of production are increasing while prices are not. Postharvest handling practices with an emphasis on temperature management recommendations to avoid plum chilling injury have been proposed as part of the solution in California (Mitchell, 1987; Crisosto et al., 1999) and in other areas of the world (Streif, 1989; Taylor and Rabe, 1994). Ripening protocols at the shipping and receiving end have been developed, promoted and established as an attempt to enhance flavor or even give an added value to plums (Crisosto, 2005). The creation and establishment of a generic minimum quality index based on ripe soluble solids concentration (RSSC) and/or ripe titratable acidity (RTA) as an approach to protect consumers and increase consumption is being pursued by several postharvest physiologists and private companies. However, it has been claimed that for some plums titratable acidity, characteristic flavor, aroma, astringency and texture become as important /$ see front matter 2006 Elsevier B.V. All rights reserved. doi: /j.postharvbio

47 272 C.H. Crisosto et al. / Postharvest Biology and Technology 44 (2007) as RSSC in determining consumer acceptance. The interaction between RSSC and RTA has been well illustrated for a high acid, early dark plum (Crisosto et al., 2004), i.e. within the same RSSC range ( %) combined with three RTA ranges (RTA 0.60%, RTA %, and RTA 1.00%) the mean degree of liking by consumers decreased significantly (p < ) as the RTA range increased, 6.2, 5.3 and 3.3, respectively. This relationship has also been reported for early grapes (Nelson et al., 1963; Crisosto and Crisosto, 2002), cherries (Kappel et al., 1996; Crisosto et al., 2003) and Hayward kiwifruit (Crisosto and Crisosto, 2001; Marsh et al., 2004). Other approaches to fuel consumption have been taken by plant breeders, who are developing and introducing new plum and plum type (pluot) cultivars with different chemical and sensory characteristics than current commercial plum cultivars. Pluots are interspecific hybrids of complex crosses of plum and apricot with predominantly plum parentage typically with smooth skin. Thus, in the last decade a large number of cultivars with notable flavor characteristics, i.e. strong plum flavor, a blend of apricot and plum flavors, very sweet and high antioxidant capacity have been released (Wills et al., 1983; Tomas-Barbera et al., 2001). In this study, we tested the hypothesis that plum and plum type cultivars could be consistently segregated based on their predominant sensory characteristics: sweetness, sourness, flavor and aroma intensity (plum), using a trained panel (Crisosto et al., 1998, 2003, 2006). As plums and pluots are currently reaching new domestic and overseas markets with diverse consumer ethnic groups, this proposed organoleptic classification may help to match fruit flavor characteristics to consumers specific flavor characteristic requirements, i.e. sweet, balanced sweet/tart, strong plum flavor and/or aroma, tart, etc., thus enhancing marketing and promotion activities. Our sensory research program involved the following steps: verify the variability of sensory attributes in current plum and pluot cultivars, segregate cultivars into organoleptic groups, validate organoleptic groups with consumer perception of sensory attributes, describe the chemical attributes of each organoleptic group, propose a minimum quality index within each organoleptic group, and understand the relationship between consumer preferences and these proposed organoleptic groups. In this work we reported information on the first two steps. 2. Materials and methods 2.1. Trained panel Cultivar segregation studies focused on the organoleptic description of 12 plum and four pluot cultivars (Table 1) were carried out by a trained panel of nine (2003) or 10 (2004) trained judges selected for their taste acuity (O Mahony, 1986; Lawless and Heymann, 1998). Four sources (orchards) per cultivar were used for nine plum and one pluot cultivar and only one source per cultivar for three plum and three pluot cultivars. Plum and pluot cultivars with diverse quality attributes, (low and high acidity, high soluble solids concentration (SSC) and strong plum flavor) originating from different breeding programs were selected for this study. Training sessions were conducted to instruct the judges on measuring the perception of sweetness, sourness, plum flavor intensity and plum aroma intensity using references (O Mahony, 1986). At each session, judges evaluated no more than a maximum of eight cultivar-source combinations for aroma and taste attributes. All testing was carried out at room Table 1 Means (X) and standard deviations (S.D.) of ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA as percent malic acid) for plum and pluot cultivars from one to four different sources per cultivar Cultivar code Plant breeding program RSSC a (%) RTA a (%) X S.D. X S.D. Plum cultivar Betty Anne BA Zaiger Blackamber BK USDA/UC Catalina b CA Krause Earliqueen EQ Zaiger Fortune FO USDA Friar b FR Weinberger Hiromi Red HR Zaiger Joanna Red JR Zaiger October Sun OS Chamberlin, Sr. (Met. Life Ins.) Purple Majesty PM Bradford Royal Zee RZ Zaiger Simka b SI Kazarian Pluot cultivar Dapple Dandy b DD Zaiger Flavor Grenade b FG Zaiger Flavorich FLR Zaiger Flavorosa b RS Zaiger a RSSC and RTA measured on ripe fruit (8.8 N) using a penetrometer with an 8 mm tip. b One source per cultivar.

48 C.H. Crisosto et al. / Postharvest Biology and Technology 44 (2007) temperature (20 C) in individual booths illuminated with fluorescent lighting. Samples were presented in random order in ml soufflé cups labeled with three digit random numbers. For each cultivar-source, fruit were harvested at the peak size and California Well-mature for that cultivar, and then held at 0 C for approximately 7 10 days until ripened. Prior to testing, the fruit were ripened at 20 C in a temperature-controlled room for 1 5 days until a subsample measured N flesh firmness. On each fruit for tasting, a piece of skin 2 cm in diameter was removed from one cheek and the flesh firmness measured with a UC firmness tester (Western Industrial Supply, San Francisco, CA) equipped with an 8 mm tip. If the fruit was ripe, i.e N, it was labeled, the firmness recorded and used for taste. A sample for aroma consisted of one whole, ripened (selected by touch for ripeness), unblemished fruit of the cultivar-source to be tested. A sample for taste consisted of two longitudinal slices cut from the stem end to the blossom end of the fruit on the cheek opposite the flesh firmness measurement of the cultivar-source to be tested. Judges scored a sample for each sensory attribute by circling a hatch mark placed at increments of 0.5 on a 10-cm horizontal line anchored 1 cm from both ends of the line by none and more (plum aroma and flavor intensity) or less and more (sweetness and sourness). Labeled references at room temperature (20 C) were provided at each session: sweet less (SSC = 8.1%, TA = 0.72%), sweet more (SSC = 16.0%, TA = 0.71, sour less (SSC = 11.0%, TA = 0.31%), sour more (SSC = 11.0%, TA = 1.19, flavor none (water), flavor more (100% Kern s peach nectar), aroma none (water), and aroma mid (100% Kern s peach nectar). Judges cleansed their nostrils between samples by inhaling and exhaling deeply two to three times. Judges cleansed their palates between samples and references with drinking water. After the aroma and taste evaluation, flesh firmness was measured on the aroma samples (2004) as previously described. Then, on all of the previously labeled fruit samples (aroma and taste), a longitudinal wedge was removed from the same area as the flesh firmness measurement, placed between two layers of cheesecloth and the juice expressed for subsequent soluble solids concentration (SSC) and titratable acidity (TA) measurements. TA was measured with an automatic titrator (Radiometer, Copenhagen, Denmark) and expressed as percent malic acid Data analysis The relationship between cultivar-source and perception of sensory attributes by a trained panel and fruit chemical composition (SSC and TA) was calculated by using the SAS program. Data were subjected to ANOVA and correlation analysis. After that means were separated using LSD means separation at the 5% and 1% levels using the SAS statistical software (SAS Institute, Cary, NC). As source did not affect the perception of sensory attributes, cultivars were segregated into groups according to the average of their sources by organoleptic characteristics by using an agglomerative hierarchical clustering (AHC) protocol (XLSTAT version 5.1, Addinsoft, New York, USA). This analysis was applied to the data in order to identify particular clusters of cultivars with similar sensory attributes. These calculations were made using the Euclidian distance with Ward s method as segregation criterion. The coordinates of the cluster centroids were used to calculate a principal component analysis (PCA) in order to characterize the attributes of each cluster of cultivars (Serrano-Megías and López-Nicolás, 2006; Ward, 1963). 3. Results and discussion In general, early season plum cultivars had lower RSSC than late season plum cultivars while RTA was not related to time of season (Table 1). Mean RSSC measured on the ripe plum cultivars ( N) ranged from 9.0% ( Earliqueen ) to 19.8% ( October Sun ) and mean RTA ranged from 0.18% ( Friar ) to 0.87% ( Earliqueen ). In the four pluot cultivars tested, RTA was similar among them at approximately 0.40% and the RSSCs were greater than or equal to 12.4%. Flavorosa, an early season pluot, had a higher RSSC and lower RTA than the early season plums, Earliqueen and Royal Zee (Table 1). In all these cultivars, the RTAs were lower than the values ( %) measured on other typical early season California plums ( Blackamber ) over many years. Source significantly affected the RSSC and RTA levels within each cultivar, except for Purple Majesty in which source did not affect RSSC and October Sun in which source did not affect RTA (Table 2). In all of the cultivars tested, harvesting above their physiological maturity and within the tested quality attribute parameters, source did not significantly affect sweetness, plum flavor inten- Table 2 Significance (p-values) of correlation between four sources each per cultivar of plum and pluot cultivars and perception of sensory attributes by a trained panel and fruit chemical composition Species Cultivar Sweetness Sourness Flavor Aroma RSSC RTA Plum Betty Anne < < Plum Blackamber < < Plum Earliqueen < < Plum Fortune < < < Plum Hiromi Red < < Plum Joanna Red < Plum October Sun < Plum Purple Majesty < Plum Royal Zee < Pluot Flavorich < <0.0001

49 274 C.H. Crisosto et al. / Postharvest Biology and Technology 44 (2007) Fig. 1. Dendrogram of 12 plum and four pluot cultivars originating from different breeding programs according to their organoleptic characteristics as perceived by a trained panel and segregated according to Ward methodology. (Letter codes represent cultivars listed in Table 1). sity, plum aroma intensity or sourness perception even though sources differed significantly in RSSC and RTA (Table 2) Organoleptic segregation Principal component analysis was used to segregate cultivars into different organoleptic groups after using the agglomerative hierarchical clustering (Fig. 1). The perception of the four sensory attributes (sweetness, sourness, plum flavor intensity, and plum aroma intensity) was reduced to three principal components, which accounted for 98.6% for plums and pluots of the variation in the sensory attributes of the tested cultivars. PC1 accounted for 49.8% of the variability and it was positively loaded for sweetness (0.697) and plum flavor (0.668). In this model, sourness ( 0.212) and plum aroma ( 0.150) had little representation. Similar to PC1, PC2, which accounted for 25.5% of the variation, had high positive loading (0.731) for plum aroma and low for sweetness (0.160), while sourness was very high but negative ( 0.678) and plum flavor ( 0.068) was very low. Contrary to PC1 & PC2, in the PC3 model, all of the loading components were positive; sweetness (0.686) and plum aroma (0.662) had the highest values (Table 3). Fig. 2. Segregation of 12 plum and four pluot cultivars originating from different breeding programs according to their organoleptic characteristics as perceived by a trained panel and determined by principal component analysis (PCA). PC1 (49.8%) is plotted on the X-axis and PC2 (25.5%) on the Y-axis with the vectors representing the loadings of sensory data along with the principal component scores. By plotting the 12 plum and four pluot cultivars sensory attributes in the two most important principal components (PC1 = 49.8% and PC2 = 25.6%) they were segregated into three groups: tart, plum aroma, and sweet/plum flavor (Fig. 2). Cultivars plotted near the vector representing the sensory loading for sourness and distant from the sweetness and plum flavor vectors were classified in the tart group. These cultivars, Earliqueen, Purple Majesty, Blackamber, Simka, Betty Anne, and Flavorich were characterized by high sourness and low sweetness/plum flavor. Cultivars plotted near the vector representing the plum aroma were classified in the plum aroma group. These Table 3 Component loadings for sensory attributes and component scores for 12 plum and four pluot cultivars Attribute Component loadings Component scores PC1, λ = 49.8% PC2, λ = 25.5% PC3, λ = 23.3% Cultivar PC1 PC2 PC3 Sweetness BA Sourness BK Plum flavor CA Plum aroma DD EQ FG FLR FO FR HR JR OS PM RS RZ SI

50 C.H. Crisosto et al. / Postharvest Biology and Technology 44 (2007) Table 4 Coefficients of determination (r 2 ) between ripe fruit chemical attributes and sensory attributes as perceived by a trained panel for 12 plum and four pluot cultivars Sweetness Sourness Flavor intensity Aroma RSSC 0.32 ** NS 0.27 ** 0.48 *** RTA NS NS NS NS RSSC/RTA 0.29 ** NS 0.28 ** NS ** Significant at 1% level. *** Significant at 0.1% level. cultivars, Royal Zee, Joanna Red, Fortune, and Flavorosa were characterized by a strong plum aroma. Cultivars plotted near the vectors representing the sensory loadings for sweetness and plum flavor vectors were classified in the sweet/plum flavor group, and included cultivars Catalina, Dapple Dandy, October Sun, Hiromi Red, Friar and Flavor Grenade. The relationships between fruit chemical composition and perception of sensory attributes were significant and similar for plums and pluots (Table 4). For cultivars picked above their physiological maturity, RSSC was significantly correlated with the perception of sweetness, plum flavor and plum aroma intensity, but not to sourness. The RSSC:RTA was correlated only with sweetness and plum flavor intensity but not with aroma intensity or sourness. These relationships were significant, but their levels were very low. The fact that RTA did not affect any sensory attribute can be explained by the low RTA of the cultivars used in this study. We previously reported that for Blackamber, an early season dark plum with high acidity, RTA ( 1.00%) played a significant role in consumer acceptance only on fruit with RSSC less than 12.0% (Crisosto et al., 2004). However, the RTA influence disappeared on fruit with RSSC 12.0%. For another important plum cultivar, Fortune a full red plum with moderate acidity, RTA did not play a role at all. Thus, our recent in store consumer tests carried out using Blackamber (tart group) and Fortune (plum aroma group) indicated that fruit with RSSC 12.0% resulted in high consumer acceptance and it was controlled by RSSC (Crisosto et al., 2004). Further execution of more in store consumer tests to develop a minimum quality index and a survey of the chemical composition of other cultivars should be tested in our proposed system. In the meantime, a minimum quality index of 12.0% RSSC is being proposed for California plum cultivars based on our in store consumer work (Crisosto et al., 2004). Our information suggests that an electronic sorting device that measures RSSC could be used to segregate fruit populations based on consumer acceptance during the packing line operation. Near infrared (NIR) spectroscopic techniques have been developed for nondestructive assessment of internal quality characteristics of produce such as soluble solids content and total solids content (Slaughter and Abbott, 2004). This NIR technology to sort fruit based on RSSC is commercially available but implies an extra cost. Therefore, prior to installing a NIR detector, it is important to determine how much of your cultivars fruit population will exceed any proposed quality index. For example, according to the RSSC variability measured in our cultivar samples, 100% of the fruit from 10 out of the 14 tested cultivars would exceed this 12.0% RSSC. However, 60% of the Blackamber and Flavorosa fruit had RSSC 12.0%, while only 27% of the Royal Zee and none of the Earliqueen fruit were above our proposed quality index. Several techniques such as harvest date, crop load adjustments, training system, irrigation and others can be used to increase the population of fruit exceeding a proposed RSSC, but each cultivar has a limited RSSC and/or RTA range (Crisosto et al., 1997). Currently, breeding programs are producing early ripening cultivars with low RTAs and high RSSCs to increase the sweetness and plum flavor. Because fruit source did not affect cultivar organoleptic classifications when fruit were harvested above their physiological maturity, this proposed organoleptic classification should be attempted after validation with in store consumer tests. According to our current knowledge on plum quality attributes, we recommend a postharvest handling system that delivers plum and pluot cultivars that exceed 12.0% RSSC, grouped according to their organoleptic characteristics and delivered to consumers close to their ready to eat stage. The establishment of this delivery system will assure satisfaction of a high percentage of consumers, help match consumer preferences, enhance consistent flavor delivery, and, therefore, increase plum and pluot consumption. Acknowledgements The California Department of Food and Agriculture (CDFA) and the California Agricultural Technology Institute, California State University, Fresno provided economic support for this work. References Abdi, N., McGlasson, W.B., Holford, P., Williams, M., Mizrahi, Y., Responses of climacteric and suppressed-climacteric plums to treatment with propylene and 1-methylcyclopropene. Postharvest Biol. Technol. 14, Anon., Fruit and tree nuts: situation and outlook yearbook. United States Department of Agriculture, Economic Research Service, TFS-271. Argenta, L.C., Krammes, J.G., Megguer, C.A., Amarante, C.V.T., Mattheis, J., Ripening of Laetitia plums following harvest and cold storage as affected by inhibition of ethylene action. Pesqui. Agropecu. Bras. 38 (10), Ceponis, M.J., Cappellini, R.A., Disorders in plum, peach, and nectarine shipments to the New York market, Plant Dis. 71 (10), Crisosto, C.H., Plum ripening. Cent. Valley Postharvest Newslett. 14 (4). Crisosto, C.H., Crisosto, G.M., Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22, Crisosto, C.H., Crisosto, G.M., Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24, Crisosto, C.H., Johnson, R.S., Day, K.R., DeJong, T., Orchard factors affecting postharvest stone fruit quality. HortScience 32 (5), Crisosto, C.H., Crisosto, G., Watkins, M., Chemical and organoleptic description of white flesh nectarines and peaches. Acta Hortic. 465, Crisosto, C.H., Mitchell, F.G., Ju, Z., Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. HortScience 34,

51 276 C.H. Crisosto et al. / Postharvest Biology and Technology 44 (2007) Crisosto, C.H., Crisosto, G.M., Metheney, P., Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol. Technol. 28, Crisosto, C.H., Garner, D., Crisosto, G.M., Bowerman, E., Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34, Crisosto, C.H., Crisosto, G.M., Echeverria, G., Puy, J., Segregation of peach and nectarine (Prunus persica (L.) Bastch) cultivars according to their organoleptic characteristics. Postharvest Biol. Technol. 39, Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. HortScience 31, Lawless, H.T., Heymann, H., Sensory evaluation of food: principles and practices. In: Food Science Texts Series. Chapman & Hall, New York, 827 pp. Liverani, A., Giovannini, D., Brandi, F., Increasing fruit quality of peaches and nectarines: the main goals of ISF-FO (Italy). Acta Hortic. 592, Macchi, E., I Consumi delle famiglie. In calo I consume. Acquista al dettaglio di susine in Italia. Frutticoltura 3, 4 5. Marsh, K., Attanayake, S., Walker, S., Gunson, A., Boldingh, H., Macrae, E., Acidity and taste in kiwifruit. Postharvest Biol. Technol. 332, Mitchell, F.G., Influence of cooling and temperature maintenance on the quality of California grown stone fruit. Rev. Int. Froid 10, Nelson, K.E., Baker, G.A., Winkler, A.J., Amerine, M.A., Richardson, H.B., Jones, F.R., Chemical and sensory variability in table grapes. Hilgardia 34 (1), O Mahony, M., Sensory Evaluation of Food. Marcel Dekker Inc., New York. Plich, H., The effect of storage conditions and date of picking on storability and quality of some plum (Prunus domestica L.) cultivars. Acta Hortic. 485, Serrano-Megías, M., López-Nicolás, J.M., Application of agglomerative hierarchical clustering to identify consumer tomato preferences: influence of physiochemical and sensory characteristics on consumer response. J. Sci. Food Agric. 86, Slaughter, D.C., Abbott, J.A., Analysis of fruits and vegetables. In: Near- Infrared Spectroscopy in Agriculture. Agronomy Monograph No. 44. Am. Soc. of Agronomy, Crop Sci. Soc. of Am. Soil Sci. Soc. of Am. Madison, WI, USA. Streif, J., Storage behavior of plums. Acta Hortic. (Wageningen) 258, Taylor, M.A., Jacobs, G., Physiological factors associated with overripeness, internal breakdown and gel breakdown in plums stored at low temperature. J. Hortic. Sci. 68 (5), Taylor, M.A., Rabe, E., Effect of storage regimes on pectolytic enzymes, pectic substances, internal conductivity and gel breakdown in cold stored Songold plums. J. Hortic. Sci. 69 (3), Taylor, M.A., Rabe, E., Effect of harvest maturity on pectic substances, internal conductivity, soluble solids and gel breakdown in cold stored Songold plums. Postharvest Biol. Technol. 5 (4), Tomas-Barbera, F.A., Gil, M.I., Kader, A.A., HPLC DAD ESIMS analysis of phenolic compounds in nectarines, peaches, and plums. J. Agric. Food Chem. 49, Ward, J.H., Hierarchical grouping to optimize an objective function. J. Am. Stat. Assoc. 58, Wills, R.B.H., Scriven, F.M., Greenfield, H., Nutrient composition of stone fruit (Prunus spp.) cultivars: apricot, cherry, nectarine, peach and plum. J. Sci. Food Agric. 34 (12),

52 Postharvest Biology and Technology 39 (2006) Segregation of peach and nectarine (Prunus persica (L.) Batsch) cultivars according to their organoleptic characteristics Carlos H. Crisosto a,, Gayle M. Crisosto a, Gemma Echeverria b, Jaume Puy c a Department of Plant Sciences, University of California, Davis, CA 95616, USA 1 b Centre UdL-IRTA, Rovira Roure, 191, Lleida 25198, Spain c Departament de Química, Rovira Roure, 191, Lleida 25198, Spain Received 4 March 2005; accepted 22 September 2005 Abstract Cultivar segregation according to the sensory perception of their organoleptic characteristics was attempted by using trained panel data evaluated by principal component analysis of four sources per cultivar of 23 peach and 26 nectarine cultivars as a part of our program to develop minimum quality indexes. Fruit source significantly affected cultivar ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA), but it did not significantly affect sensory perception of peach or nectarine flavor intensity, sourness or aroma by the trained panel. For five out of the 49 cultivars tested, source played a role in perception of sweetness. In all of these cases when a source of a specific cultivar was not classified in the proposed organoleptic group it could be explained by the fruit having been harvested outside of the commercial physiological maturity (immature or over-mature) for that cultivar. The perception of the four sensory attributes (sweetness, sourness, peach or nectarine flavor intensity, peach or nectarine aroma intensity) was analyzed by using the three principal components, which accounted for 92 and 94% of the variation in the sensory attributes of the tested cultivars for peach and nectarine, respectively. Season did not significantly affect the classification of one cultivar that was evaluated during these two seasons. By plotting organoleptic characteristics in PC1 and PC2 ( 76%) for peach and nectarine, cultivars were segregated into groups (balanced, tart, sweet, peach or nectarine aroma and/or peach or nectarine flavor intensity) with similar sensory attributes; nectarines were classified into five groups and peaches into four groups. Based on this information, we recommend that cultivars should be classified in organoleptic groups and development of a minimum quality index should be attempted within each organoleptic group rather than proposing a generic minimum quality index based on the ripe soluble solids concentration (RSSC). This organoleptic cultivar classification will help to match ethnic preferences and enhance current promotion and marketing programs Elsevier B.V. All rights reserved. Keywords: Sensory attributes; PCA; Ripe fruit quality attributes; Minimum quality index 1. Introduction In the last decade, peach and nectarine per capita consumption has remained the same or even decreased in the USA (Anon., 2004) and some European countries (Liverani et al., 2002; Hilaire and Mathieu, 2004). Consumer com- Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). 1 Located at Kearney Agricultural Center, 9240 S. Riverbend Avenue, Parlier, CA 93648, USA. plaints for peaches center on lack of flavor and textural characteristics associated with ripening (Bruhn et al., 1991), in addition to chilling injury symptoms such as off flavor, mealy texture and flesh browning (Von Mollendorff et al., 1992). At the same time, costs of production are increasing while prices are not. Postharvest handling practices with an emphasis on temperature management recommendations to avoid chilling injury have been proposed as part of the solution (Mitchell, 1987; Crisosto et al., 1999). Ripening protocols at the shipping and receiving end have been developed, promoted and established as an attempt to enhance /$ see front matter 2005 Elsevier B.V. All rights reserved. doi: /j.postharvbio

53 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) flavor or even give an added value to peaches and nectarines (Crisosto, 1997, 2000; Crisosto et al., 2004a). Other approaches to increase consumption have been taken by plant breeders, who are developing and introducing new peach and nectarine cultivars with different chemical characteristics than cultivars previously planted (Byrne, 2003). These recently released white or yellow flesh cultivars with low, medium or high acidity and/or high soluble solids concentration (SSC) (measured at harvest), more peach or nectarine flavor and/or aroma have the potential of being perceived as cultivars with different sensory attributes (Neri et al., 1996). Unfortunately, sensory classification of peach and nectarine cultivars based on the organoleptic perception of these characteristics by consumers has not been investigated. Another approach to increase consumption that is being pursued by several postharvest physiologists is the establishment of a generic single minimum quality index based on a level of SSC (Ravaglia et al., 1966; Kader, 1994; Testoni, 1995; Ventura et al., 2000; Hilaire, 2003). Following this generic SSC approach, agricultural engineering companies are introducing nondestructive sensors to segregate fruit based on SSC or other fruit quality attributes prior to and during packaging that will help to enforce any proposed minimum quality standard (Chen, 1996; Shmulevich, 2003). However, it is well accepted by postharvest physiologists, but not well documented for peaches and nectarines, that there are some commodities or situations in which titratable acidity, characteristic flavor, aroma, astringency and texture become as important as SSC in determining consumer acceptance. For example, the interaction between RSSC and ripe titratable acidity (RTA) has been reported for an early dark plum (Crisosto et al., 2004b) and RTA may also be involved in consumer acceptance for early season peach and nectarine cultivars with high acidity and/or low RSSC situations. It has been reported that RTA plays an important role in consumer acceptance for grapes (Nelson et al., 1973; Crisosto and Crisosto, 2002), cherries (Kappel et al., 1996; Crisosto et al., 2003b) and kiwifruit (Crisosto and Crisosto, 2001; Marsh et al., 2004). The establishment of a generic single quality index based on SSC (Neri et al., 1996; Crisosto, 2002; Crisosto et al., 2003a; Hilaire and Mathieu, 2004) may create more confusion in the market without contributing to the solution of the consumption problem. For this reason, we believe that it is important to segregate cultivars according to their most dominant organoleptic characteristic (i.e. sweetness, sourness, peach or nectarine flavor intensity, or peach or nectarine aroma) and then develop a reliable minimum quality index within each organoleptic group. As peaches and nectarines are currently reaching new domestic and overseas markets with diverse consumer ethnic groups (Liverani et al., 2002; Crisosto et al., 2003a), this proposed organoleptic classification may help to match fruit characteristics to consumers specific characteristic requirements and enhance marketing and promotion activities. Our sensory research program involved the following steps: verify the variability of sensory attributes in current peach and nectarine cultivars, segregate cultivars into organoleptic groups, describe the chemical attributes of each organoleptic group, propose a minimum quality index within each organoleptic group, and understand the relationship between ethnic preferences and these proposed organoleptic groups. In this work we reported information on the first two steps. 2. Materials and methods 2.1. Trained panel Cultivar segregation studies focused on the organoleptic description of 23 peach and 26 nectarine cultivars from four sources per cultivar (Table 1) were carried out by a panel of nine (2003) or ten (2004) trained judges selected for their taste acuity (O Mahony, 1986; Lawless and Heymann, 1998). The same sources of Sweet Dream peaches were evaluated during these two seasons as an internal control for taste panel and/or environmental conditions. Yellow and white flesh peach and nectarine cultivars with diverse combinations of fruit quality attributes (low and high acidity, high soluble solids concentration (SSC) and high peach or nectarine aroma) originating from different breeding programs were selected for this study. Both years, training sessions were conducted to instruct the judges on measuring the perception of sweetness, sourness, peach or nectarine flavor intensity and peach or nectarine aroma intensity using references (O Mahony, 1986). At each session, judges evaluated no more than a maximum of eight cultivar source combinations for aroma and taste attributes. All testing was carried out at room temperature (20 C) in individual booths illuminated with fluorescent lighting. Samples were presented in random order in ml soufflé cups labeled with three digit random numbers. For each cultivar source, fruit were harvested at the peak size and California Well-Mature for that cultivar, then held at 0 C for approximately 7 10 days until ripened. Prior to testing, the fruit were ripened at 20 C in a temperature-controlled room for 1 5 days until a subsample measured N flesh firmness. On each fruit for tasting, a piece of skin 2 cm in diameter was removed from one cheek and the flesh firmness measured with a UC firmness tester (Western Industrial Supply, San Francisco, CA) equipped with an 8 mm tip. If the fruit was ripe (i.e N) it was labeled, the firmness recorded and used for taste. A sample for aroma consisted of one whole, ripened (selected by touch), unblemished fruit of the cultivar source to be tested. A sample for taste consisted of two longitudinal slices cut from the stem end to the blossom end of the fruit on the cheek opposite the flesh firmness measurement of the cultivar source to be tested. Judges scored a sample for each sensory attribute by circling a hatch mark placed at increments of 0.5 cm on a 10 cm horizontal line anchored 1 cm from both ends of the line by none and more (peach or nectarine aroma and flavor intensity) or less and more

54 12 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) Table 1 Means (X) and standard deviations (S.D.) of ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) for peach and nectarine cultivars from four different sources per cultivar Cultivar code Plant breeding program RSSC a RTA a X S.D. X S.D. Peach cultivar Autumn Flame AF Doyle Brittney Lane BL Zaiger Country Sweet CS Zaiger Diamond Princess DP Bradford Elegant Lady EL Merrill July Flame JF Burchell Kaweah KA Zaiger May Sweet MS Zaiger O Henry OH Merrill Rich May RM Zaiger Saturn SA Bailey Snow Fire SWF Zaiger Snow Kist SK Zaiger Spring Snow SPW Zaiger Sugar Lady SL Zaiger Summer Sweet SS Zaiger Summer Zee SZ Zaiger Sunlit Snow SUL Zaiger Super Rich SR Zaiger Sweet Dream-2003 SD-03 Zaiger Sweet Dream-2004 SD-04 Zaiger Tra-Zee TZ Zaiger White Lady WL Zaiger Zee Lady ZL Zaiger Nectarine cultivar Arctic Jay AJY Zaiger Arctic Snow ASOW Zaiger Arctic Star ARS Zaiger Arctic Sweet ASW Zaiger August Fire AUF Waldner August Glo AUG Zaiger Bright Pearl BGP Bradford Diamond Bright DBG Bradford Diamond Ray DR Bradford Fire Pearl FIP Bradford Fire Sweet FRW Bradford Grand Pearl GP Bradford Grand Sweet GSW Bradford Honey Blaze HB Zaiger Honey Kist HK Zaiger Honey Royale HR Zaiger Red Diamond RED Anderson Royal Glo RG Zaiger Ruby Diamond RUD Bradford Ruby Pearl RP Bradford Ruby Sweet RSW Bradford September Free SFR USDA Spring Bright SPBG Bradford Summer Blush SBL Bradford Summer Bright SBG Bradford Zee Glo ZG Zaiger a RSSC and RTA measured on ripe fruit (8.8 N) using a penetrometer with an 8 mm tip. (sweetness and sourness). Labeled references at room temperature 20 C were provided at each session: sweet less (SSC = 8.1 ± 0.1%, TA = 0.72%), sweet more (SSC = 16.0%, TA = 0.71 ± 0.02%), sour less (SSC = 11.0%, TA = 0.31%), sour more (SSC = 11.0%, TA = 1.19 ± 0.02%), flavor none (water), flavor more (100% Kern s peach nectar), aroma none (water), and aroma mid (100% Kern s peach nectar). Judges cleansed their nostrils between samples by inhaling

55 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) and exhaling deeply two to three times. Judges cleansed their palates between samples and references with drinking water. After the aroma and taste evaluation, flesh firmness was measured on the aroma samples (2004) as previously described. Then, on all of the previously labeled fruit samples (aroma and taste), a longitudinal wedge was removed from the same area as the flesh firmness measurement, placed between two layers of cheesecloth and the juice expressed for subsequent soluble solids concentration (SSC) and titratable acidity (TA) measurements Data analysis The relationship between cultivar source and perception of sensory attributes by a trained panel and fruit chemical composition (SSC and TA) was calculated by using the SAS program. Cultivars were segregated into groups according to the average of their sources according to organoleptic characteristics by using the principal component analysis program (CAMO ASA, 1997). In five of the tested cultivars in which the source played a significant (p-value 0.05) role in the perception of sensory attributes, PCA was also carried out by source within each of these cultivars. 3. Results and discussion Even though all of the sources within a cultivar had significantly different RTAs, and for most of the cultivars RSSC was significantly different between sources, most of the sources for a given cultivar did not deviate from the sensory attributes of that cultivar. For Autumn Flame, Brittney Lane, May Sweet, and Sugar Lady peaches, RSSC was not related to their sources. Two of these cultivars ( Brittney Lane and May Sweet ) are early peaches that ripen during the last 2 weeks of May and their RSSCs have been historically between 9.5 and 10.5% with little variability between orchards (less than 1.0%). May Sweet is a recent introduction so there is no historical data available for it. Thus, this small variability in RSSC can explain the lack of source significance for these cultivars. Within the cultivars tested, source had a significant effect on perception of sweetness for five cultivars, but source was not related to perception of peach or nectarine flavor intensity, aroma or sourness (Table 2). In all of the cultivars tested, source did not significantly affect flavor, aroma or sourness perception even though sources differed significantly in RTA within a given cultivar. It has been our experience over the last 10 years that Table 2 Significance (p-values) of correlation between four sources each per cultivar of peach and nectarine cultivars and perception of sensory attributes by a trained panel and fruit chemical composition Fruit Cultivar Sweetness Sourness Flavor Aroma RSSC RTA Peach Autumn Flame Peach Brittney Lane < Peach Country Sweet Peach Elegant Lady < Peach Kaweah Peach May Sweet Peach Saturn < Peach Spring Snow <.0001 < Peach Sugar Lady Peach Summer Sweet < Peach Sunlit Snow Peach Super Rich < Peach Sweet Dream Nectarine Arctic Jay < Nectarine Arctic Snow < < Nectarine Arctic Star < < Nectarine Arctic Sweet < < Nectarine Diamond Bright Nectarine Diamond Ray Nectarine Fire Pearl Nectarine Grand Pearl < Nectarine Honey Blaze < Nectarine Honey Kist < < Nectarine Royal Glo < < Nectarine Ruby Pearl < Nectarine Ruby Sweet < Nectarine September Free < Nectarine Spring Bright < Nectarine Summer Blush < < Nectarine Summer Bright < Nectarine Zee Glo <

56 14 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) harvest titratable acidity (HTA) or RTA variability within a cultivar is less than for other fruit quality attributes such as RSSC, color, or firmness (Crisosto et al., 1997). This low RTA variability between sources for a given cultivar explains the fact that trained judges did not perceive sensory sourness differences between sources within a given cultivar. However, because there were large RTA differences between cultivars with low variability within the same cultivar, trained judges were able to segregate cultivars according to their perception of sourness. For Spring Snow peach and Arctic Jay, Arctic Star, Arctic Sweet, and Honey Kist nectarines, source was significantly correlated to sweetness perception (pvalue 0.05). Because source played a significant role in the perception of sweetness for this group of cultivars, we plotted PC1 and PC2 for these cultivar source combinations to test if sources for the same cultivar were segregated into the same organoleptic group(s) (Figs. 1 and 2). For Spring Snow peach, three sources were in the balanced group but fruit from source #1 (11.5% RSSC) were in the sweet group. For Arctic Jay nectarine, sources #1, #2 and #3 were classified in the balanced group ( % RSSC), while fruit from source #4 (18% RSSC) were segregated into the sweet group. Fruit from Arctic Star sources #1 (11.1% RSSC) and #4 (12.7% RSSC) were classified in the balanced group, while fruit from sources #2 (10.1% RSSC) and #3 (9.6% RSSC) were not classified in this group. For Arctic Sweet nectarine, three sources (12.1, 10.9, 9.7% RSSC) were classified in the balanced group and the source with the lowest RSSC (9.1%) was not classified in this group. A different situation occurred with Honey Kist in which the sources with RSSC between 10.9 and 12.8% were classified in the bal- Fig. 2. Segregation of 26 nectarine cultivars originating from different breeding programs according to their organoleptic characteristics as perceived by a trained panel and determined by principal component analysis (PCA). PC1 (44%) is plotted on the X-axis and PC2 (31%) on the Y-axis with the vectors representing the loadings of sensory data along with the principal component scores. anced group but the source with the highest RSSC (17.0%) and lowest RTA (0.38%) was classified in the sweet group. In our in store consumer tests, Honey Kist consumer acceptance significantly increased from 72% (10 14% RSSC) to 88% for fruit with RSSC 14%. For these two RSSC ranges, the percentage of consumers that chose the dislike option was the same ( 2%). This suggests that consumers that chose the neither like nor dislike option for fruit with 10 14% RSSC changed to liking the fruit with RSSC 14% and therefore increasing the acceptance Organoleptic segregation Fig. 1. Segregation of 23 peach cultivars originating from different breeding programs according to their organoleptic characteristics as perceived by a trained panel and determined by principal component analysis (PCA). PC1 (44%) is plotted on the X-axis and PC2 (33%) on the Y-axis with the vectors representing the loadings of sensory data along with the principal component scores. Principal component analysis was used to segregate cultivars into different organoleptic groups. The perception of the four sensory attributes (sweetness, sourness, peach or nectarine flavor intensity, and peach or nectarine aroma) was reduced to three principal components, which accounted for 92% for peaches (Table 3) and 94% for nectarines (Table 4) of the variation in the sensory attributes of the tested cultivars. By plotting the 23 peach cultivars sensory attributes in the two most important principal components (PC1 and PC2), they were segregated into four groups named balanced, tart (sour), peach aroma/flavor, and sweet (Fig. 1) in which the cultivars in a given group had sensory attributes of the first two components clustered closely together, which accounted for 77% of this peach model (Table 3). PC1 accounted for 44% of the variability and it was positively loaded for peach flavor intensity, peach aroma and sweetness. In this model, sourness had little representation (component loading of 0.14). In contrast to PC1, PC2, which accounted for 33% of the variation, had high positive loading (0.88) for sourness and negative

57 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) Table 3 Component loadings for sensory attributes and component scores for 23 peach cultivars Attribute Component loadings Cultivar Component scores PC1, λ = 44% PC2, λ = 33% PC3, λ = 15% PC1 PC2 PC3 Sweetness Brittney Lane Sourness Saturn Peach Flavor Country Sweet Peach Aroma Summer Sweet Sweet Dream Kaweah Autumn Flame Super Rich Rich May May Sweet Sunlit Snow Snow Kist Sugar Lady Elegant Lady Snow Fire White Lady Diamond Princess Zee Lady Sweet Dream July Flame Summer Zee O Henry Tra-Zee Spring Snow Spring Snow Spring Snow Spring Snow loading for sweetness and the other attributes (Table 3). Cultivars plotted near the vectors representing the sensory loading data for peach flavor were classified in the peach flavor group. Only Spring Snow -2 was classified in the aroma group. Cultivars plotted in between all four sensory vectors were classified in the balanced group. Cultivars plotted near either the sweetness or sourness vectors were classified in the sweet and tart groups, respectively. O Henry, July Flame, Elegant Lady and Zee Lady were classified in the peach flavor group. Kaweah, Autumn Flame, Country Sweet, Spring Snow -2, Spring Snow -3, Spring Snow -4, Summer Sweet, May Sweet, Snow Kist, and Sunlit Snow peaches were classified in the balanced group. Snow Fire, White Lady, Sweet Dream -2003, Sweet Dream -2004, Saturn, Sugar Lady and Spring Snow -1 were classified in the sweet group. Brittney Lane, Diamond Princess, Rich May, Super Rich, Summer Zee, and Tra-Zee were classified in the tart group. For nectarines, judges perception of the fruit sensory attributes (sweetness, sourness, nectarine flavor intensity and nectarine aroma) were represented by PC1 = 44%, PC2 = 31% and PC3 = 24%. Using PC1 and PC2 plotting analysis, which accounts for 75% of this model (Table 4), the 26 nectarine cultivars were segregated into five groups (balanced, sweet, tart (sour), nectarine flavor, nectarine aroma) in which the cultivars in a given group had sensory attributes clustered closely together (Fig. 2). PC1 accounted for 44% of the variability and it was negatively loaded for nectarine flavor intensity, aroma and sweetness, and positively loaded for sourness. PC2, which accounted for 31% of the variation included cultivars that contrast to PC1 with high positive loading for sourness, nectarine flavor intensity, and aroma and negative loading for sweetness (Table 4). Honey Kist -1, Honey Kist -2, Honey Kist -3, Diamond Bright, Summer Bright, Arctic Star -1, Arctic Jay -1, Arctic Jay -2, Arctic Jay -3, Arctic Sweet -1, Arctic Sweet -2, Arctic Sweet -4, Grand Pearl, Ruby Pearl, and Ruby Sweet nectarines were classified in the balanced group. Arctic Sweet -3, Arctic Star -2, Arctic Star -3 were not segregated into any of these groups but were near the balanced group. Ruby Diamond, Red Diamond, Diamond Ray, Royal Glo, Spring Bright, August Fire, September Free and Zee Glo were classified in the tart group. Arctic Snow, Arctic Star -4, Arctic Jay -4, Fire Sweet, Honey Kist -4, Bright Pearl, Grand Sweet, Arctic Sweet -1, Fire Pearl, Ruby Sweet, Honey Blaze, and Honey Royale were classified in the sweet group. August Glo, Summer Blush, and Zee Glo were classified in the nectarine flavor group. Within this group of cultivars, Fire Pearl, Honey Blaze and Honey Royale were cultivars classified in the nectarine aroma group. Correlation coefficients between fruit chemical composition and perception of sensory attributes were significant and similar for peaches and nectarines (Table 5). For cultivars

58 16 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) Table 4 Component loadings for sensory attributes and component scores for 26 nectarine cultivars Attribute Component loadings Cultivar Component scores PC1, λ = 44% PC2, λ = 31% PC3, λ = 24% PC1 PC2 PC3 Sweetness Diamond Bright Sourness Honey Blaze Nectarine Flavor Ruby Sweet Nectarine Aroma Ruby Pearl Grand Pearl Summer Bright Royal Glo Spring Bright Diamond Ray Zee Glo August Glo Summer Blush Arctic Snow September Free Fire Pearl Grand Sweet Ruby Diamond Bright Pearl Honey Royale Red Diamond Fire Sweet August Fire Arctic Star Arctic Star Arctic Star Arctic Star Arctic Jay Arctic Jay Arctic Jay Arctic Jay Arctic Sweet Arctic Sweet Arctic Sweet Arctic Sweet Honey Kist Honey Kist Honey Kist Honey Kist picked above their physiological maturity, RSSC had a higher correlation with sweetness, peach or nectarine flavor intensity, and aroma perception than RSSC/RTA. The fact that only 40% of the relationship between RSSC and sweetness perception is controlled by RSSC demonstrates that other Table 5 Correlation coefficients (r) between ripe fruit chemical attributes and sensory attributes as perceived by a trained panel for 23 peach and 26 nectarine cultivars Sweetness Sourness Flavor intensity Aroma Peach RSSC 0.68 *** NS 0.52 ** NS Peach RTA NS 0.90 *** 0.50 ** NS Peach RSSC/RTA 0.47 ** 0.86 *** NS NS Nectarine RSSC 0.65 *** NS NS NS Nectarine RTA 0.48 ** 0.86 *** NS NS Nectarine RSSC/RTA 0.45 ** 0.76 *** NS NS ** Significant at 1% level. *** Significant at 0.1% level. quality attribute factors such as RTA, flavor and aroma are also involved in the perception of sweetness. A similar relationship has been reported previously in mangos (Malundo et al., 2001). RTA had a low correlation with sweetness but it had a significant and strong correlation ( 80%) with perception of sourness and was equally important as RSSC in perception of flavor ( 40%). Because season did not affect organoleptic classification and fruit source only affected five out of 49 cultivar organoleptic classifications (all five of these cultivars had RSSC below or higher than their normal maturity/quality range) this proposed organoleptic group classification should be attempted. In order to create reliable organoleptic cultivar groups, the cultivar s potential quality attributes should be defined and RSSC or other quality attribute limits within each group should be established. Several techniques such as crop load adjustments, irrigation and others can be used to modify SSC but each cultivar has a limited SSC and/or TA range

59 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) (Crisosto et al., 1997). Our recent in store consumer tests carried out using Honey Kist, a low acid, yellow flesh nectarine (balanced group), Elegant Lady, a high acid, yellow flesh peach (peach flavor group), and Spring Bright, a high acid, yellow flesh peach (tart group) indicated that these cultivars have high consumer acceptance when fruit are above specific RSSC levels regardless of acidity or the proposed organoleptic group. According to these results, we recommend that cultivars should be classified in organoleptic groups and development of a minimum quality index should be attempted within each organoleptic group rather than proposing a generic minimum quality index based on RSSC. This organoleptic cultivar classification will help to match ethnic preferences and enhance the current promotion and marketing programs. Future work should be pursued to describe the chemical attribute requirements for each organoleptic group to propose a minimum quality index. Furthermore, representative cultivars from each organoleptic group could be used to describe biochemical compounds related to the perception of their sensory attributes. After identification of these compounds, a candidate gene approach can be used to develop marker(s) to establish an early breeding (seedling) program screening for high quality fruit. After that, the relationship between trained panel data and consumer acceptance with an emphasis on ethnic preferences for these proposed organoleptic groups should be investigated. Acknowledgements The California Department of Food and Agriculture (CDFA) and the California Agricultural Technology Institute, California State University, Fresno provided economic support for this work. References Anon., Fruit and Tree Nuts: Situation and Outlook Yearbook, TFS United States Department of Agriculture, Economic Research Service. Bruhn, C.M., Feldman, N., Garlitz, C., Harwood, J., Ivans, E., Marshall, M., Riley, A., Thurber, D., Williamson, M., Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14, Byrne, D., Breeding peaches and nectarines for mild-winter climate areas: state of the art and future directions. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, September 10, pp CAMO ASA, Unscrambler Users Guide, ver. 6.11a. Programme Package for Multivariate Calibration, Trondheim, Norway. Chen, P., Quality evaluation technology for agricultural products. In: Proceedings of the International Conference on Agriculture Machinery Engineering, vol. 1, Seoul, Korea, November 12 15, pp Crisosto, C.H., Stone fruit ripening protocol for receivers. 97/101, slide set with cassette tape. Division of Agriculture and Natural Resources. Crisosto, C.H., Optimum procedures for ripening stone fruit. Management of Fruit Ripening. Postharvest Horticulture Series No. 9, April. Crisosto, C.H., How do we increase peach consumption? Acta Hort. 592, Crisosto, C.H., Crisosto, G.M., Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22, Crisosto, C.H., Crisosto, G.M., Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24, Crisosto, C.H., Johnson, R.S., Day, K.R., DeJong, T., Orchard factors affecting postharvest stone fruit quality. HortScience 32 (5), Crisosto, C.H., Mitchell, F.G., Ju, Z., Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. HortScience 34, Crisosto, C.H., Crisosto, G., Bowerman, E., 2003a. Searching for consumer satisfaction: new trends in the California peach industry. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, September 10, pp Crisosto, C.H., Crisosto, G.M., Metheney, P., 2003b. Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol. Technol. 28, Crisosto, C.H., Garner, D., Andris, H.L., Day, K.R., 2004a. Controlled delayed cooling extends peach market life. HortTechnology 14, Crisosto, C.H., Garner, D., Crisosto, G.M., Bowerman, E., 2004b. Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34, Hilaire, C., The peach industry in France: state of art, research and development. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, September 10, pp Hilaire, C., Mathieu, V., Le Point sur la qualite gustative des peches et nectarines. Infos-Ctifl 201, Kader, A.A., Fruit maturity, ripening, and quality relationships. Perishable Handling Newslett. 80, 2. Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. HortScience 31, Lawless, H.T., Heymann, H., Sensory Evaluation of Food: Principles and Practices. Food Science Texts Series. Chapman & Hall, New York, 827 pp. Liverani, A., Giovannini, D., Brandi, F., Increasing fruit quality of peaches and nectarines: the main goals of ISF-FO (Italy). Acta Hort. 592, Malundo, T.M.M., Shewfelt, R.L., Ware, G.O., Baldwin, E.A., Sugars and acids influence flavor properties of mango (Mangifera indica). J. Am. Soc. Hort. Sci. 126, Marsh, K., Attanayake, S., Walker, S., Gunson, A., Boldingh, H., Macrae, E., Acidity and taste in kiwifruit. Postharvest Biol. Technol. 332, Mitchell, F.G., Influence of cooling and temperature maintenance on the quality of California grown stone fruit. Rev. Int. Froid. 10, Nelson, K.E., Schutz, H.G., Ahmedullah, M., McPherson, J., Flavor preferences of supermarket customers for Thompson Seedless grapes. Am. J. Enol. Viticult. 24, Neri, F., Vassalli, P., Brigati, S., Valutazione organolettica di alcune cultivar di pesche e nettarine. Rivista di Frutticoltura 7/8, O Mahony, M., Sensory Evaluation of Food. Marcel Dekker, Inc., New York. Ravaglia, G., Sansavini, S., Ventura, M., Tabanelli, D., Indici di maturazione e miglioramneto qualitative delle pesche. Frutticoltura 3, Shmulevich, I., Nondestructive texture assessment of fruits and vegetables. Acta Hort. (ISHS) 599,

60 18 C.H. Crisosto et al. / Postharvest Biology and Technology 39 (2006) Testoni, A., Momento di raccolta, qualità, condizionamento e confezionamento delle pesche. In: Proceedings of the Symposium La peschicoltura Veronese alle soglie del 2000, Verona, February 25, pp Ventura, M., Sama, A., Minguzzi, A., Lazoni, S., Sansavini, S., Ottimizzazione del carico di frutti per migliorare la produzione e la qualità delle nettarine Supercrimson e Venus. In: Sansavini, S. (Ed.), Per una nuova peschicoltura: produzione, orrganizzazione, mercato. XXIV Convengo Peschicolo, Cesena, Febbraio 24 25, pp Von Mollendorff, L.J., Jacobs, G., de Villiers, O.T., Cold storage influences internal characteristics of nectarines during ripening. HortScience 27,

61 Postharvest Biology and Technology 38 (2005) Relationship between ripe soluble solids concentration (RSSC) and consumer acceptance of high and low acid melting flesh peach and nectarine (Prunus persica (L.) Batsch) cultivars Carlos H. Crisosto, Gayle M. Crisosto Department of Plant Sciences, University of California, Davis, CA 95616, USA Received 8 February 2005; accepted 16 July 2005 Abstract The minimum RSSC needed to reach high consumer acceptance for peach and nectarine was determined by using in-store consumer tests of low and high RTA melting flesh cultivars as a part of our program to develop minimum quality indexes. For Ivory Princess, a low acid, white flesh peach, Honey Kist, a low acid, yellow flesh nectarine, Elegant Lady, a high acid, yellow flesh peach, and Spring Bright, a high acid, yellow flesh nectarine, degree of liking and consumer acceptance were associated with ripe soluble solids concentration (RSSC) regardless of ripe titratable acidity (RTA). For the two high acid ( % RTA) cultivars tested, consumer acceptance increased rapidly as RSSC increased, reaching 90%. In these cultivars, consumer acceptance reached a plateau and above which, it became insensitive to any additional increase in RSSC. For Elegant Lady and Spring Bright, the plateau was reached at 11 12%, and 10 11% RSSC with 90% consumer acceptance, respectively. For the low acid cultivars ( % RTA), Ivory Princess and Honey Kist, consumer acceptance progressively increased as RSSC increased without reaching a plateau, and attained nearly 100% acceptance with RSSC of 16 and 15%, respectively. For these low acid and high acid cultivars, consumer acceptance was closely related to RSSC but maximum consumer acceptance was attained at different RSCC levels depending on the cultivar. The fact that these cultivars reached high consumer acceptance with different RSSC levels indicates that a single generic RSSC quality index would not be reliable to assure consumer satisfaction across all cultivars Elsevier B.V. All rights reserved. Keywords: In-store consumer test; Quality attributes; Ripe titratable acidity; Quality survey; Peach; Nectarine 1. Introduction Corresponding author at. Kearney Agricultural Center, 9240 S. Riverbend Avenue, Parlier, CA 93648, USA. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). New peach and nectarine cultivars are being planted that have a wide range of titratable acidity (TA) and soluble solids concentration (SSC) potentials when /$ see front matter 2005 Elsevier B.V. All rights reserved. doi: /j.postharvbio

62 240 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) harvested mature (Giovannini et al., 2000; Liverani et al., 2002; Byrne, 2003; Crisosto et al., 2003). Early studies have associated high consumer acceptance with high SSC in peaches (Ravaglia et al., 1966; Parker et al., 1991). Based on these early studies, a minimum quality index of 10% SSC has been proposed for a group of yellow flesh peaches and nectarines in California (Kader, 1994). In France, with a large diversity of characteristics available in the peach genotype (white/yellow flesh, low/high acid, melting/nonmelting flesh), a minimum of 10% SSC for peaches with a low TA, and 11% SSC for peaches with a high TA are being evaluated as part of a quality standard (Hilaire, 2003). In this work, harvest titratable acidity (HTA) measured in mature fruit less than 0.90% was considered low and HTA 0.90% was classified in the high TA group. Our experience with Californian and Chilean yellow flesh peaches is that up to approximately 30% of the HTA in a mature fruit (harvest) will be lost during ripening. For low acidity white and/or yellow flesh peaches, this acidity loss varied (Crisosto, unpublished). In Italy, with an industry that has a high proportion of yellow flesh cultivars, a minimum SSC of 10% for early season, 11% for mid-season and 12% for late season cultivars was previously proposed as a quality standard (Testoni, 1995; Ventura et al., 2000). There is limited information on the relationship between consumer acceptance and the interaction between ripe titratable acidity (RTA) and ripe soluble solids concentration (RSSC) of peaches and nectarines. In this work, we investigated the minimum RSSC required to reach high consumer acceptance for peach and nectarine cultivars with similar texture but low ( %) and high ripe titratable acidity ( %). 2. Materials and methods 2.1. Fruit collection Ivory Princess, a low acid, white flesh peach; Honey Kist, a low acid, yellow flesh nectarine; Elegant Lady, a high acid, yellow flesh peach; and Spring Bright, a high acid, yellow flesh nectarine were selected for this study for their commercial importance, differences in titratable acidity and similar texture after ripening (melting flesh). Fruit was collected from trees of each cultivar grown under standard commercial practices at the Kearney Agricultural Center (Parlier, CA) during the 2002 season. Fruit were harvested at commercial first pick from different canopy positions on trees previously thinned at different crop levels to assure the potential range of soluble solids concentrations and titratable acidities for each cultivar. Samples of 100 fruit each were selected from each tree. Harvested fruit were commercially handled; fast air cooled down to 0 1 C (within 12 h of picking) with an air flow of 0.5 l min 1 per kilogram of fruit using a portable cooling tunnel and stored at 0 C up to 7 days prior to ripening Ripening Fruit were ripened in a temperature-controlled room at 20 C (85% RH) until the flesh firmness reached 17.8 N. For each cultivar, firmness changes were monitored during ripening on samples of 10 fruit per day per cultivar/canopy position/crop load with a UC firmness tester equipped with an 8 mm tip Fruit selection and preparation On the day of the consumer test in the morning, on each ripened fruit of the cultivar to be used for the consumer test, a 2 cm diameter-sized piece of skin was removed from one cheek and the flesh firmness measured with a UC firmness tester equipped with an 8 mm tip. If the fruit was ripe ( 17.8 N flesh firmness) a numerical code was written on the tip of the fruit and the flesh firmness recorded. We chose this firmness range ( N) based on our previous sensory work that demonstrated that plum consumer acceptance reaches its maximum potential when fruit have been ripened to a flesh firmness of N (80 90% consumer acceptance). If the fruit are consumed at a higher firmness (less ripe, N) consumer acceptance is reduced from 90 to 25% (Crisosto, 1999). Only fruit within the N firmness range were used for this in-store consumer test. After the firmness measurement of the coded fruit was recorded a longitudinal wedge was cut from the same area used for the flesh firmness, placed between two layers of cheesecloth and the juice expressed for subsequent SSC and TA measurements. The SSC of the juice was measured with a temperature compensated

63 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) refractometer (model ATC-1, Atago Co., Tokyo, Japan) and the TA was measured with an automatic titrator (Radiometer, Copenhagen, Denmark). The coded fruit was placed in one of four boxes of the correct RSSC class (A D) based on the SSC measurements In-store consumer tests In-store consumer tests were conducted on Ivory Princess peach, Honey Kist nectarine, Elegant Lady peach and Spring Bright nectarine according to our previous work (Crisosto and Crisosto, 2001; Crisosto et al., 2004). Groups of 100, 100, 120, and 100 consumers representing a diverse combination of ages, ethnic groups and genders were surveyed in a major supermarket in Fresno County, California for the cultivars listed above on June 15, July 12, July 19, and June 21, 2005, respectively. The ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) were measured on each previously labeled ripe fruit sample presented to the consumers and correlated with their responses. For each cultivar, each consumer was presented in random order four samples; one from each of four ripe SSC classes. These classes were selected to include the historical range of SSC levels reported in our previous industry-wide fruit quality surveys for that cultivar (Table 1). Then at the supermarket, the samples were prepared in the produce room out of sight from the testing area. A sample consisted of one longitudinal slice cut from the stem end to the blossom end on the cheek opposite the flesh firmness measurement of the fruit. Consumers who responded that they ate fresh peaches/nectarines were surveyed. For each sample, the consumer was asked to taste it, then to indicate if Table 1 Means and standard deviations for ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) of fruit for high acid Elegant Lady peach and Spring Bright nectarine, and low acid Ivory Princess peach and Honey Kist nectarine during the 2002 season Cultivar Plant breeding RSSC a (%) RTA a,b (%) program Elegant Lady Merrill 12.4 (1.3) 0.74 (0.13) Spring Bright Bradford 11.0 (1.8) 0.92 (0.15) Ivory Princess Bradford 12.6 (1.9) 0.22 (0.04) Honey Kist Zaiger 12.0 (3.0) 0.46 (0.08) a Fruit ripened at 20 C and 85% RH until flesh firmness reached 17.8 N. b RTA expressed as percentage malic acid. he/she likes, neither likes nor dislikes, or dislikes the sample. Then the consumer was asked to indicate his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The consumer s response was recorded using a nine-point hedonic scale (one-dislike extremely to nine-like extremely). Consumer acceptance was measured as both degree of liking (1 9) and percentage acceptance. Percentage acceptance was calculated as the number of consumers liking the sample (score > 5.0) divided by the total number of consumers within that sample (Lawless and Heymann, 1998). In a similar manner, the percentage of consumers disliking (score < 5.0) and neither liking nor disliking (score = 5.0) the sample was calculated. The degree of liking data was subjected to analysis of variance (ANOVA) prior to the LSD mean separation using the SAS program. 3. Results and discussion For these two low acid and two high acid melting flesh cultivars, degree of liking was significantly related to RSSC but not to RTA, and there was no significant interaction between RSSC and RTA. For the high acid Elegant Lady and Spring Bright, which account for about 30% of the peach and nectarine volume in California, the degree of liking and percentage consumer acceptance increased constantly as RSSC increased until it reached a plateau (Fig. 1). Degree of liking and percentage consumer acceptance then became insensitive to any additional increase in RSSC (i.e. reached the saturation point). At the saturation point, percentage consumer acceptance was similar between these two cultivars, reaching approximately 90%. The percentage consumer acceptance of Elegant Lady and Spring Bright reached the saturation point at and 10 11% RSSC, respectively. There was a significant difference in degree of liking between Elegant Lady peaches with RSSC < 11.0% and RSSC 11.0% (Table 2). Elegant Lady peaches with RSSC 11.0% were liked moderately (7.1) with a consumer acceptance of 71.8% while peaches with RSSC < 11.0% were liked slightly (5.9) with consumer acceptance of 47.8%. The neither like nor dislike option was selected by several consumers, and it decreased from 37.8 to 25.0% for RSSC < 11.0% and RSSC 11.0%, respectively. For Elegant Lady

64 242 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) Fig. 1. Degree of liking and percentage consumer acceptance of Elegant Lady (EL), a high acid, yellow flesh peach and Spring Bright (SB), a high acid, yellow flesh nectarine at different levels of ripe soluble solids concentration (RSSC) by American consumers. Degree of liking was measured on a nine-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; 9, like extremely). Different letters within a given RSSC level indicate a significant difference between means by LSD peaches with RSSC 11.0%, the percentage of consumers that chose the dislike option was only 3.2% while 14.4% disliked Elegant Lady peaches with RSSC < 11.0%. Elegant Lady RSSC ranged from 9.0 to 15.0% with a mean of 12.4% and RTA ranged from 0.45 to 0.90% with a mean of 0.74% within the population of fruit used in this consumer test. In the previous 5 years, the average RSSC has ranged from 11.4 to 11.7% based on California industry-wide fruit quality surveys (Crisosto, unpublished). Consumer acceptance expressed as degree of liking was significantly different for Spring Bright nectarines with RSSC < 10.0% and RSSC 10.0% (Table 2). Spring Bright nectarines with RSSC 10.0% were liked moderately (6.9) with an acceptance of 70.5% while nectarines with RSSC < 10% were neither liked nor disliked (5.4) with an acceptance of 34.9%. The neither like nor dislike option was selected by several consumers, and it varied from 39.4 to 24.0% for fruit with RSSC < 10.0% and RSSC 10.0%, respectively. For Spring Bright nectarines with RSSC 10.0%, the percentage of consumers that chose the dislike option was only 5.6%, while 25.7% of consumers disliked Spring Bright nectarines with RSSC < 10.0%. Spring Bright RSSC ranged from 8.0 to 16.0% with a mean of 11.0% and RTA ranged from 0.60 to 1.20%, with a mean of 0.92% within the population of fruit used in this consumer test. For the low acid Ivory Princess and Honey Kist, recent releases that are being widely planted in California, the degree of liking and percentage consumer acceptance also increased as RSSC increased (Fig. 2), reaching 100%. Ivory Princess RSSC ranged from 8.0 to 17.0% with a mean of 12.6% and RTA ranged from 0.15 to 0.45% with a mean of 0.22% within the population of fruit used in this consumer test. Since Ivory Princess is a recently released cultivar, no industry wide quality data was available at the time of this Table 2 Acceptance of Elegant Lady peaches and Spring Bright nectarines by American consumers at different levels of ripe soluble solids concentration (RSSC) Elegant Lady quality attributes Degree of liking (1 9) a Acceptance (%) Neither like nor dislike (%) Dislike (%) RSSC < 11.0% 5.9 b b RSSC 11.0% 7.1 a LSD P-value Spring Bright quality attributes Degree of Liking (1 9) a Acceptance (%) Neither like nor dislike (%) Dislike (%) RSSC < 10.0% 5.4 b b RSSC 10.0% 6.9 a LSD P-value a Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. b Same letters within the same column indicate no significant difference between means.

65 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) Fig. 2. Degree of liking and percentage consumer acceptance of Ivory Princess (IP), a low acid, white flesh peach and Honey Kist (HK), a low acid, yellow flesh nectarine at different levels of ripe soluble solids concentration (RSSC) by American consumers. Degree of liking measured on a nine-point hedonic scale (1, dislike extremely; 5, neither like nor dislike; 9, like extremely). Different letters within a given RSSC range indicate a significant difference between means by LSD test. Within this range of quality attributes degree of liking for Ivory Princess peaches varied from neither like nor dislike to like extremely and consumer acceptance varied from 33.3 to 100% (Fig. 2). There was a significant difference in consumer acceptance expressed as degree of liking between peaches with RSSC < 12.0% and RSSC 12.0% (Table 3). Peaches with RSSC < 12.0% were liked slightly (5.8) with a consumer acceptance of 47.1%, while peaches with RSSC 12% were liked very much (7.3) with a consumer acceptance of 80%. The neither like nor dislike option was selected by several consumers, and it decreased from 34.8 to 15.9% for RSSC < 12.0% and RSSC 12.0%, respectively. Peaches with RSSC 12.0% were disliked only by 4.1% of the consumers while 18.1% disliked peaches with RSSC < 12.0%. Honey Kist RSSC ranged from 7.0 to 16.0% with a mean of 12.0% and RTA ranged from 0.30 to 0.60% with a mean of 0.46% within the population of fruit used in this consumer test. Since Honey Kist is also a recently released cultivar, no industry wide quality data was available at the time of this test. Within the RSSC range tested, degree of liking varied from neither like nor dislike (5.0) to like very much (7.5). The lowest (23.8%) and highest (87.7%) consumer acceptance was attained in fruit that had RSSC < 9.0% and fruit that had RSSC 14.0%, respectively (Table 3). Consumer acceptance was nearly the same ( 65%) for fruit within the % RSSC range. The percentage of consumers that chose the neither like nor dislike option decreased from 57.1 to 10.2% as the RSSC increased from <9.0 to 14.0%. For Honey Kist nectarines with 9.0 to 14.0% RSSC, the percentage of consumers that Table 3 Acceptance of Ivory Princess peaches and Honey Kist nectarines by American consumers at different levels of ripe soluble solids concentration (RSSC) Ivory Princess quality attributes Degree of liking (1 9) a Acceptance (%) Neither like nor dislike (%) Dislike (%) RSSC < 12.0% 5.8 b b RSSC 12.0% 7.3 a LSD P-value Honey Kist quality attributes Degree of liking (1 9) a Acceptance (%) Neither Like nor Dislike (%) Dislike (%) RSSC < 9.0% 5.0 d b RSSC % 6.6 c RSSC % 7.0 b,c RSSC > 14.0% 7.5 a LSD P-value a Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. b Same letters within the same column indicate no significant difference between means.

66 244 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) disliked the fruit varied from 4.2 to 2.2%. When fruit of this cultivar had RSSC <9.0, 19.0% of consumers disliked the fruit. It is interesting to point out that the in store consumer test results for these cultivars indicated that the percentage consumer acceptance was more sensitive to the consumers selecting the neither like nor dislike option rather than disliking the fruit. For example, 62.1% of the consumers accepted Honey Kist fruit with % RSSC while 4.2% of the consumers disliked these fruit, but 33.7% of the consumers neither liked nor disliked them. As the RSSC ranges increased from 9.1 to 10.0% to RSSC > 14.0%, acceptance increased from 62.1 to 87.7% mainly due to a change by the consumers from neither liking nor disliking the fruit to liking them as the change in disliking the fruit was only from 4.2 to 2.2%. Further work to understand the reason(s) why a large percentage of the consumers selected the neither like nor dislike option for this cultivar and others is needed. We believe that because many new cultivars with diverse fruit quality attributes and flavors are becoming available in the commercial market, a cultivar description identifying specific characteristics during marketing would aid this group of consumers in selecting fruit with the fruit quality attributes they desire. For the high acid cultivars, after they reached their plateau an increase in RSSC did not result in higher consumer acceptance. For example, Elegant Lady peaches with 14% RSSC did not have a higher consumer acceptance than Elegant Lady peaches with 12% RSSC. In the case of the low acid cultivars, after they reached approximately 80% acceptance, the increase in consumer acceptance was small with increased RSSC. For these four cultivars, high consumer acceptance, similar to that previously reported on cherry (Crisosto et al., 2002) and table grape (Nelson et al., 1973; Crisosto and Crisosto, 2002) was reached when ripe fruit free of internal breakdown symptoms were tasted. The relationship between RTA and RSSC has an important role in consumer acceptance of some peach, nectarine, and plum cultivars within a specific range of RTA and RSSC. Based on our previous work, cultivars with RTA > 0.90% and RSSC < 12.0%, consumer acceptance was controlled by the interaction between RTA and RSSC rather than RSSC alone. For example, a significant interaction was reported between RSSC and RTA for Blackamber plum (Crisosto et al., 2004). In this early season plum cultivar, RTA ranged from 0.50 to 1.10% and low consumer acceptance was related to RTA within a given RSSC range. These quality attributes are typical of early season peach, nectarine and plum cultivars and may also be encountered in the case of some mid and late season cultivars harvested with low maturity. The influence of RTA on consumer acceptance has been reported on early season table grapes (Nelson et al., 1972; Crisosto and Crisosto, 2002), kiwifruit (Crisosto and Crisosto, 2001; Marsh et al., 2004) and cherries (Kappel et al., 1996; Crisosto et al., 2002). As the tested cultivars had different consumer acceptance at a given RSSC, a proposed minimum quality index should be specific for each cultivar. Under the conditions of this test (ripe fruit, free of internal breakdown symptoms), the two low acid and the two high acid, melting flesh cultivars attained high acceptance by American consumers. This suggests that the characteristic of high acidity is not a negative quality attribute when the fruit is ripe and handled properly. However, higher consumer acceptance ( 100%) was achieved with the low RTA cultivars than the high RTA cultivars ( 90%). This can be explained because low RTA cultivars also had fruit in the high RSSC range. In our previous sensory work using the same technique with Hayward kiwifruit (Crisosto and Crisosto, 2001), Redglobe table grapes (Crisosto and Crisosto, 2002), and Blackamber plum (Crisosto et al., 2004) we reached a consumer acceptance of nearly 85%, and only in cherries (Crisosto et al., 2002) we attained higher consumer acceptance. We have observed that some of the recently released, low acid cultivars may have the ability to produce fruit with RSSC higher than the traditional yellow flesh cultivars. For example, new releases such as Ivory Princess and May Sweet that are replacing Spring Crest peach had consistently higher RSSC and lower RTA than Spring Crest peach in the last 3 years of our survey from commercial packinghouses in the San Joaquin Valley, CA. This tendency may be verified when more cultivars of this fruit type are produced commercially. It is important to point out that other sensory quality attributes such as aroma, peach or nectarine flavor and texture may also be contributing to their high consumer acceptance. Some of these quality attributes can interact with sweetness and sourness perception, and in some cultivars compensate

67 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) for the low RSSC or high RTA. Unfortunately, in spite of the potential importance of these flavor components and the commercial availability of aromatic, crunchy, nonmelting flesh, flavorful new cultivars, there is a lack of information on this topic. This 10% higher consumer acceptance for these two low RTA cultivars than the two high RTA cultivars implied that new consumers are being satisfied when eating these two low RTA cultivars. Our ongoing industry quality survey revealed that most of the tested cultivars have a high proportion of fruit above the desired RSSC. Several researchers have reported manipulations of orchard factors to increase the number of fruit that exceed these potential quality standards (Prashar et al., 1976; Marini et al., 1991; Crisosto et al., 1997; Forlani et al., 2002; Girona, 2002). For example, Elegant Lady peach RSSC can be increased by utilizing a well-illuminated training system, proper crop load, summer pruning and controlled late harvest (Crisosto et al., 1997). It is important to point out that in these consumer tests, fruit free of chilling injury that were stored at 0 C for short periods of time (<7 10 days) simulating a fast delivery domestically were used. Because development of off flavor has been widely reported as a consequence of chilling injury during cold storage (Smith, 1934; Ben Arie and Lavee, 1971; Von Mollendorff et al., 1992), the four cultivars were tasted prior to the onset of the end of their market life. In our peach storage test, California O Henry peaches developed visual flesh mealiness symptoms after 2 and 4 weeks stored at 5 and 0 C, respectively. However, trained judges were able to detect off flavor or mealy texture 1 or 2 weeks before the visual flesh mealiness symptom was apparent (Crisosto and Labavitch, 2002). Thus, in fruit that exceeds any proposed quality index, the development of off flavor or chilling injury during postharvest handling will prevent consumers from perceiving high eating quality. This in-store consumer test indicated that high consumer acceptance is attained with our mid-season cultivars when peaches are free of internal breakdown and ready to eat (ripe) prior to consumption. These results indicate that the relationship between RSSC and consumer acceptance is cultivar specific, and there is not a single reliable SSC that assures a given percentage of satisfied consumers. As the availability of cultivars with a diversity of sensory characteristics and flavor potentials is increasing, segregation of cultivars into organoleptic groups based on the sensory perception of sweetness, sourness, peach or nectarine flavor, aroma intensity and texture may be beneficial prior to the establishment of a generic minimum quality index based on RSSC using in-store consumer tests. Detailed biochemistry and sensory studies on different texture and aroma components should be carried out to improve the current understanding of peach and nectarine flavor. We expect that cultivars segregated into these organoleptic groups would have similar minimum quality indexes which may help the peach and nectarine industry to consistently describe, identify and deliver fruit of high eating quality. Meanwhile, plant breeders should select for high RSSC potential, strong peach flavor, relatively low acidity, and low chilling injury susceptibility, to assure new releases will have high consumer acceptance. References Ben Arie, R., Lavee, S., Pectic changes occurring in Elberta peaches suffering from woolly breakdown. Phytochemistry 10, Byrne, D., Breeding peaches and nectarines for mild-winter climate areas: state of the art and future directions. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, 10 September, pp Crisosto, C.H., Optimum procedures for ripening stone fruit. In: Management of Fruit, Ripening. University of California, Davis, Postharvest Horticulture Series 9, pp Crisosto, C.H., Crisosto, G.M., Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22, Crisosto, C.H., Crisosto, G.M., Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24, Crisosto, C.H., Labavitch, J.M., Developing a quantitative method to evaluate peach (Prunus persica) flesh mealiness. Postharvest Biol. Technol. 25, Crisosto, C.H., Johnson, R.S., Day, K.R., DeJong, T., Orchard factors affecting postharvest stone fruit quality. HortScience 32 (5), Crisosto, C.H., Crisosto, G.M., Ritenour, M.A., Testing the reliability of skin color as an indicator of quality for early season Brooks (Prunus avium L.) cherry. Postharvest Biol. Technol. 24, Crisosto, C.H., Crisosto, G., Bowerman, E., Searching for consumer satisfaction: new trends in the California peach industry. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First

68 246 C.H. Crisosto, G.M. Crisosto / Postharvest Biology and Technology 38 (2005) Mediterranean Peach Symposium. Agrigento, Italy, 10 September, pp Crisosto, C.H., Garner, D., Crisosto, G.M., Bowerman, E., Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34, Forlani, M., Basile, B., Cirillo, C., Iannini, C., Effects of harvest date and fruit position along the tree canopy on peach fruit quality. Acta Hort. 592, Giovannini, D., Liverani, A., Brandi, F., Indagine sulla qualità commerciale delle pesche prodotte in Romagna: metodi di valutazione e possibilità di miglioramento. In: Sansavini, S. (Ed.), Per una nuova peschi coltura: produzione, orrganizzazione, mercato. XXIV Convegno Peschicolo. Cesena, February, pp Girona, J., Regulated deficit irrigation in peach. A global analysis. Acta Hort. 592, Hilaire, C., The peach industry in France: state of art, research and development. In: Marra, F., Sottile, F. (Eds.), Proceedings of the First Mediterranean Peach Symposium. Agrigento, Italy, 10 September, pp Kader, A.A., Fruit maturity, ripening, and quality relationships. Perishables Handling Newsletter 80, 2. Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. HortScience 31, Lawless, H.T., Heymann, H., Sensory evaluation of food: principles and practices. In: Food Science Texts Series. Chapman & Hall, New York. Liverani, A., Giovannini, D., Brandi, F., Increasing fruit quality of peaches and nectarines: the main goals of ISF-FO (Italy). Acta Hort. 592, Marini, R.P., Sowers, D., Marini, M.C., Peach fruit quality is affected by shade during final swell of fruit growth. J. Am. Soc. Hort. Sci. 116 (3), Marsh, K., Attanayake, S., Walker, S., Gunson, A., Boldingh, H., MacRae, E., Acidity and taste in kiwifruit. Postharvest Biol. Technol. 32, Nelson, K.E., Allen, J.W., Schutz, H.G., Effect of grape maturity, sample order, and sex of the taster on the flavor response of supermarket customers. Am. J. Enol. Viticult. 23, Nelson, K.E., Schutz, H.G., Ahmedullah, M., McPherson, J., Flavor preferences of supermarket customers for Thompson Seedless grapes. Am. J. Enol. Viticult. 24, Parker, D.D., Zilberman, D., Moulton, K., How quality relates to price in California fresh peaches. California Agric. 45 (2), Prashar, C.R.K., Pearl, R., Hagan, R.M., Review on water and crop quality. Sci. Hortic. 5, Ravaglia, G., Sansavini, S., Ventura, M., Tabanelli, D., Indici di maturazione e miglioramneto qualitative delle pesche. Frutticoltura 3, Smith, W.H., Cold storage of Elberta peaches. Ice and Cold Storage 37, Testoni, A., Momento di raccolta, qualità, condizionamento e confezionamento delle pesche. In: Proceedings of the Symposium on La peschicoltura Veronese alle soglie del 2000, Verona, 25 February, pp Ventura, M., Sama, A., Minguzzi, A., Lazoni, S., Sansavini, S., Ottimizzazione del carico di frutti per migliorare la produzione e la qualità delle nettarine Supercrimson e Venus. In: Sansavini, S. (Ed.), Per una nuova peschicoltura: produzione, orrganizzazione, mercato. XXIV Convengo Peschicolo. Cesena, February, pp Von Mollendorff, L.J., Jacobs, G., de Villiers, O.T., Cold storage influences internal characteristics of nectarines during ripening. HortScience 27,

69 1 DETERMINING THE PRIMARY DRIVERS OF LIKING TO PREDICT CONSUMER ACCEPTANCE OF FRESH NECTARINES AND PEACHES DRIVERS OF LIKING OF FRESH NECTARINES AND PEACHES Claudia Delgado a*, Gayle M. Crisosto a, Hildegarde Heymann b, and Carlos H Crisosto a a Department of Plant Sciences, University of California, Davis, CA 95616, USA b Department of Viticulture and Enology, University of California, Davis, CA 95616, USA Corresponding author: * Claudia Delgado, University of California, Davis, Department of Plant Sciences One Shields Avenue, Davis, CA cdelgado@ucdavis.edu Tel.: Fax: New address: Chromocell Corporation 685 U.S. Highway One, North Brunswick, NJ Claudia.Delgado@Chromocell.com Tel.: ext Fax:

70 2 1 ABSTRACT A generic descriptive analysis using 11 judges provided 16 sensory attributes that described the aroma, flavor, and texture characteristics of seven nectarine and peach cultivars selected for their predominant sensory attributes. Simultaneously, the in-store acceptability of these cultivars 5 was evaluated by 120 consumers from northern California. The relationships among instrumental measurements (flesh firmness, ripe soluble solids concentration (RSSC), and ripe titratable acidity (RTA)), sensory panel descriptors, and consumer hedonic responses were studied. In these cultivars, RSSC was the only instrumental measurement significantly related to overall liking. Cultivars with medium acidity and/or flavor-aroma were liked very much, and consumer willingness to pay more was correlated with overall liking without regard to cultivar. External preference mapping revealed three clusters that were associated with ethnicity and consumer preferences within each cluster. Sweetness was the main driver of liking for two consumer clusters; however, for the third cluster, the perception of fruit aromas described as grassy/green fruit and pit aromas were the main drivers of liking. There was a high correlation between instrumental measurements and their sensory perception; however, the sensory attribute measurements explained cultivar characteristics better than instrumental measurements alone. Sweetness correlated positively with overall liking and consumer acceptance. 18 PRACTICAL APPLICATIONS: The main objective of this study was to identify drivers of liking for fresh peaches and nectarines in order to understand consumer preferences for these fruits. This information can be used by postharvest researchers to evaluate the potential of new postharvest technologies and consumer

71 acceptance and for plant breeders to develop new cultivars with desirable sensory attributes driven by the consumer KEYWORDS: Instrumental quality; Descriptive analysis; Fruit sensory attributes; Consumer acceptance; External preference mapping, L-PLS regression.

72 INTRODUCTION Despite increasing knowledge of the health benefits of eating fruits and vegetables, consumption of some tree fruit commodities in the United States, such as peaches, nectarines, and plums, has been static. Since 1980, consumption has averaged 5.5 pounds per capita per year (USDA Economic Research Data, 2009). A recent consumer quality survey of 1,552 consumers (Sterling-Rice Group, 2006) corroborates previous survey results (Bruhn 1995) concluding that lack of flavor and chilling injury symptoms are still the main barriers restricting California peach, nectarine, and plum purchasing in the U.S. market. Some researchers have tried to predict tree fruit consumer acceptance and/or preference using fruit physicochemical quality measurements at harvest, such as soluble solids concentration (SSC) for overall sweetness, penetration force for firmness-texture, and titratable acidy (TA) for sourness (Crisosto et al. 2003). A similar approach was used in Italy (Esti et al. 1997) and Slovenia (Colaric et al. 2005) to predict consumer quality. However, most of these studies did not attempt to relate these measurements to consumer responses. Other groups have attempted to evaluate the efficacy of such physicochemical measurements to explain consumer responses to apples (Hoehn et al. 2003) pears (Predieri and Gatti 2009), pineapples (Schulbach et al. 2007), mangos (Malundo et al. 2001), blueberries (Saftner et al. 2008), and oranges (Obenland et al. 2009). In some instances, fruit physicochemical measurements were related to consumers hedonic ratings or acceptance percentages (Crisosto, Crisosto and Bowerman 2003; Crisosto and Crisosto 2001; Crisosto and Crisosto 2005; Crisosto et al. 2004; Guerra et al. 2009; Gunness et al. 2009). Despite general agreement that measured soluble sugars and/or organic acid concentrations are key components in predicting consumer acceptability of fresh fruits, other fruit quality characteristics also affect liking. In peaches, fruit firmness, color, and aroma were important

73 characteristics consumers used to evaluate fruit quality when selecting fruit to purchase (Bruhn 1995; Bruhn et al. 1991). These physicochemical measurements and sensory techniques can be used by the industry to evaluate the potential effect of new postharvest technologies on consumer acceptance; by shippers to evaluate current postharvest practices; by retail managers to validate their handling practices; and by plant breeders to develop new cultivars with desirable sensory 56 attributes. Although significant correlations have been reported between physicochemical parameters measured instrumentally and sensory properties/hedonic scores (Colaric et al. 2005; Crisosto and Crisosto 2005; Rossiter et al. 2000), these correlations usually do not predict consumer behavior well. Despite the expense of consumer tests, they are more effective in predicting consumer behavior (Bett 2002; Harker et al. 2008; Saftner et al. 2008). In an ideal situation, researchers should simultaneously use physical instrumental fruit quality measurements and sensory methodology to evaluate consumer responses; however, because of budget constraints, fast rotation, intensive preparation, and/or limited quantities of fruit available, it is not always possible to conduct sensory evaluations. Descriptive analysis will provide with the characterization of the most important attributes for the fruit cultivars, while consumer tests will indicate how much those fruit cultivars are liked. External preference mapping is a well known technique used in the sensory field to relate sensory and consumer data with the purpose of identifying the drivers of liking (Lawless and Heymann, 2010, Yenken et al. 2011). L-PLS analysis is a relatively new technique based on partial least square regression (PLS) used to characterize three different datasets: the sensory attributes [X], consumers liking [Y] and consumers information [Z] to identify demographic differences among consumers (Lengard and Kermit, 2006).

74 Several researchers have explored the relationship between instrumental measurements, sensory properties, and consumer perception in other fruit commodities for example: apples (Daillant-Spinnler et al. 1996; Harker et al. 2003; Harker et al. 2008; Harker et al. 2002; Kühn and Thybo 2001; Oraguzie et al. 2009), tomatoes (Causse et al. 2010; Lee et al. 1999; Sinesio et al. 2010), and strawberries (Ares et al. 2009). We believe that the recent release and marketing of tree fruit cultivars with different flavors and the establishment of ripening protocols (Crisosto 1999) justify the expense of developing drivers of liking for tree fruit. Thus, the main goal of this study was to identify drivers of liking for fresh nectarines and peaches that predict consumer acceptance and/or preferences for these fruits during postharvest handling MATERIALS AND METHODS Cultivar selection and fruit preparation Seven peach and nectarine cultivars were selected for this study for their commercial importance, differences in titratable acidity, flavor, and aroma, and similar melting flesh texture after ripening (Table 1). August Pearl is a low acid, white flesh nectarine; Fire Sweet, a medium acid, flavorful, yellow flesh nectarine; August Bright, a high acid, yellow flesh nectarine; Autumn Snow, a low acid, white flesh peach; Ryan Sun, a medium acid, yellow flesh peach with balanced sensory attributes; O Henry, a medium acid, flavorful, yellow flesh peach; and Summer Lady, a medium acid, sweet, yellow flesh peach. The cultivars were selected based on their previously determined sensory attributes (Crisosto and Crisosto 2005; Crisosto et al. 2006; Crisosto et al. 1998). For each cultivar, fruit were harvested at peak size and California Well-mature for that cultivar from commercial orchards in Fresno Co., CA, then held at 0 C (85% RH) for up to 10 d, except for Summer Lady, which was held at 5 C (85%

75 RH) prior to ripening to induce onset of chilling injury (Crisosto and Labavitch 2002). The same fruit from each cultivar was used for both descriptive analysis and the consumer study Instrumental fruit quality measurements Fruit were ripened in a temperature-controlled room at 20 C (85% RH) until a subsample reached a flesh firmness of 17.8 N as described (Crisosto 1999). On the day of the descriptive analysis session or in-store consumer study, a 2-cm diameter piece of skin was removed from one cheek of each ripened fruit of the cultivar to be tested and the flesh firmness (penetration force) was measured with a UC firmness tester (Western Industrial Supply, San Francisco, CA) equipped with an 8 mm tip. If the fruit was ripe ( 17.8 N flesh firmness), a numerical code was written on the tip of the fruit and the flesh firmness recorded. A sample consisted of one longitudinal slice cut from the stem end to the blossom end on the cheek opposite that on which flesh firmness was measured (Crisosto and Crisosto 2005). In addition, a longitudinal wedge was removed from the same area as the flesh firmness measurement, placed between two layers of cheesecloth, and the juice expressed for subsequent soluble solids concentration (SSC) and titratable acidity (TA) measurements. The SSC of the juice was measured with a digital temperature-compensated refractometer (model PR-32α, Atago Co., 113 Tokyo, Japan). TA was measured with an automatic titrator (TitraLab 850, Radiometer 114 Analytical, Copenhagen, Denmark) and expressed as percent malic acid In-Store consumer study One hundred twenty consumers who reported eating fresh nectarines/peaches participated in the study. The experiment was conducted at a major supermarket in Davis, California. Each

76 consumer evaluated seven samples; the experimental design was a Williams Latin square design provided by the FIZZ software. One fruit sample per cultivar was evaluated by each consumer using a written questionnaire. For each nectarine or peach sample, consumers expressed their overall liking using the 9-point hedonic scale (Peryam and Pilgrim 1957). Consumer acceptance was calculated as the percentage of respondents who liked the sample, with scores > Consumer dislike of a sample was calculated as the percentage with scores <5. At the supermarket, the samples were prepared in the produce room out of sight of the testing area as described (Crisosto and Crisosto 2005) Generic descriptive analysis A generic descriptive analysis (Lawless and Heymann 2010) was used to identify sensory descriptors for fresh nectarines and peaches. The panel consisted of 11 judges (six women and five men) with an average age of 32 years. Each judge completed eight training sessions. The first two sessions covered the development of the language; four sessions were intended to achieve concept alignment, provide references, eliminate similar terms or ambiguities, and perfect use of the scale; and the last two sessions evaluated the judges agreement and understanding of the attributes. FIZZ software (Biosystèmes) was used to build an automated session. Sixteen attributes were defined by the panel using standards (Table 2) and evaluated using a continuous, unstructured 10-cm line scale anchored at the ends by low and high intensity, except for firmness, which was anchored by soft and hard, and crunchy, which was anchored by not and very. Samples were evaluated in triplicate with one single fruit used for each judge. The order of presentation of the samples was randomized using a Latin square design provided by the FIZZ software.

77 Statistical analysis The majority of the statistical analyses were executed using SAS version 9.1 (SAS Institute, Cary, NC). To understand the relationships between physicochemical measurements, sensory attributes, and consumer hedonic ratings, univariate analysis (correlation, analysis of variance, and Fisher s LSD multiple mean comparisons) and multivariate analysis (canonical variate analysis (CVA), MANOVA, and preference mapping) were performed. Market clusters were determined using external preference mapping and cluster analysis. Cluster analysis was performed with XL-Stat Version The Unscrambler version 9.8 was used to perform block partial least square regression (L-PLS) analysis RESULTS AND DISCUSSION Instrumental fruit quality measurements RSSC varied from 10 to 13.4% and RTA ranged between 0.21 and 0.77% in the ripe fruit (Table 1). In general, peaches had lower RSSC than nectarines and RTA varied among cultivars. Among the nectarines, white-fleshed August Pearl and yellow-fleshed Fire Sweet had low RTAs (~0.30%), while yellow-fleshed August Bright had a high RTA (0.77%). Autumn Snow, a white-fleshed peach, had the lowest RTA (0.21%) and the yellow-fleshed peaches O Henry, Summer Lady, and Ryan Sun had medium RTAs (0.50%). Our previous 10 years of surveys indicated that RSSC is more variable than RTA for a given cultivar over years or locations. Orchard management and environmental conditions have a strong effect on RSSC but less on RTA. We observed larger changes in fruit TA than in SSC during ripening on and off the tree. RTA measurements reported here are similar to those measured in previous surveys. RTA

78 values reported here for August Bright nectarine (0.77%) and Summer Lady, O Henry, and Ryan Sun (~0.50%) peaches were somewhat lower than previously reported for mature fruit (~0.60 to 0.80%). These differences in RTA are explained by loss of fruit acidity during ripening In-Store consumer study For this consumer population, there were no significant differences in the distribution of female and male ages (p>0.05, chi-square test): 58% were female and 42% male. The average age was 33 years with a standard deviation of 17.5 years. Of the total population, 41% identified themselves as White-Caucasian, 36% as Asian-Asian American, 18% as Hispanic or Latino, 2% American Indian or Alaska Native, 1% Black-African American, 12% Mixed or other, and 9% preferred not to report their ethnicity. The consumption rate of nectarines and peaches for this consumer population was approximately equally distributed among once a month (23% nectarines, 24% peaches); two to three times a month (28% nectarines, 23% peaches); once a week (21% nectarines, 21% peaches), and two to four times a week (12% nectarines, 18% peaches). These consumption rates were low for when nectarines and peaches are in season; this supports reports that indicate a static consumption for nectarines and peaches since 1980 (USDA Economic Research Data, 2009). The 120 consumers differed in their preferences for the seven cultivars (ANOVA 184 p<0.05). Pearson s correlations (p 0.05) between overall liking for each cultivar and instrumental quality measurements of ripe soluble solids concentration (RSSC), ripe titratable acidity (0.20 to 0.80% RTA), and flesh firmness (6.5 to 20.2 N) were not significant (p>0.05), except for RSSC (R 2 = 90.2). Overall liking increased significantly from like slightly to like

79 moderately (positive slope) as RSSC increased from 10.0 to 14.0%. Even though the relationship between RTA and overall liking was not significant, it had a negative slope, suggesting that cultivars with high acidity were less preferred. The significant negative effect of high RTA on overall liking has been reported previously on fruit with RTA higher than 0.80 to 1.00% and RSSC lower than 12.0% (Crisosto and Crisosto 2001; Crisosto and Crisosto 2005). The RTA for the tested cultivars ranged from 0.21 to 0.77%, which may explain the lack of 194 significant correlation between hedonic scores and RTA measurements. The relationship between overall liking and flesh firmness had a flat slope and was not significant in any cultivar. This lack of relationship differs from other fruit commodities such as apples, where texture change is one of the most important drivers of liking (Harker et al, 2008). However, all fruit tested here was ripened to a firmness penetration force of 6.3 to 17.8 N, considered ready to eat with maximum sensory potential based on our previous work (Crisosto and Crisosto 2005). Since acceptance of these cultivars measured as a degree of liking was not affected by changes in RTA, and RSSC was the only significant instrumental measurement affecting overall liking of nectarines and peaches, a detailed statistical analysis between RSSC and degree of liking was pursued. In general, degree of liking increased as RSSC increased and then reached a plateau (Table 3). Among the nectarine cultivars, RSSC did not affect degree of liking within each cultivar and consumer acceptance ranged from 65 to 91% (Table 3). Nectarines with predominant sensory characteristics of low acidity and/or flavor/aroma had high consumer acceptance percentages (72 to 91%) and were liked moderately to very much (6.6 to 7.7). Nectarines with high acidity were liked less (5.7 to 6.7) and less accepted (65 to 82%). In the nectarine cultivar with high acidity (~0.80%), acceptance increased and rejection decreased for fruit with RSSC 12.0%. In the peach with low acidity, RSSC from 10.4 to 14.5% did not

80 significantly affect degree of liking or acceptance. Rejection was around 17% and acceptance ranged from 67 to 84%. In peaches with predominant sensory characteristics of flavor or medium acidity, fruit with RSSC < 9.0% had a low degree of liking (~4.5) and acceptance (25 to 36%). For fruit with RSSC 9.0%, degree of liking increased for peaches with high flavor and for peaches with medium acidity reached a plateau above 9.0% RSSC. In this small population of consumers that tasted nectarines and peaches with predominant flavor and high RSSC, degree of liking and acceptance was very high (91%). This data also suggests that nectarines or peaches with very low acidity may have a low potential consumer acceptance; however, this is affected by ethnicity (Crisosto and Crisosto 2002). It is important to point out that perception of flavor in peaches decreased and off flavor increased during cold storage as a consequence of chilling injury (Crisosto and Labavitch 2002; Infante et al. 2009). In most cultivars, this flavor loss is faster when fruit is stored at 5 C than at 0 C. In this study, all cultivars were handled rapidly to avoid any onset of chilling injury except for Summer Lady, in which onset of loss of flavor or off flavor development may have occurred. A further detailed analysis of nectarine and peach cultivars by hedonic scale categories for purchase intent, price expectation, second consumption, and RSSC was conducted (Tables 4, 5). In general, consumers were willing to pay more for fruit with a higher hedonic score; this trend was independent of the nectarine (Table 4) or peach (Table 5) cultivar. The same trends occurred for purchase intent and willingness to consume the fruit for a second time. Other researchers have found a correlation between overall liking and the price consumers would be willing to pay for specialty food such as extra virgin olive oil (Delgado and Guinard 2011; Stefani et al. 2006) and consumers agreed to pay more when they liked cheeses (Napolitano et al. 2010). These in-store consumer test results agreed with previous studies (Crisosto et al.

81 ), in which peaches and/or nectarines with predominant sensory attributes such as flavor and/or aroma had a slightly higher consumer acceptance (~10%) than the standard ones. These results confirmed our previous sensory study and demonstrate that tree fruit degree of liking is associated with buying habits and even willingness to pay more. These results justify changes in orchard management to produce more fruit with high RSSC and selection of cultivars with predominant sensory attributes by plant breeders during cultivar development and growers for future plantings (Crisosto et al. 1997) Generic descriptive analysis Sixteen attributes were defined by the judges to describe the sensory characteristics of the seven nectarine and peach cultivars (Table 2). These attributes were evaluated through a threeway ANOVA (judges, cultivars, replications, and all two-way interactions). The ANOVA F- ratios confirmed that the panel performance was satisfactory (data not shown). The replication effects were not significant (p>0.05) for the majority of attributes evaluated, except for firmness, 248 crunchy, juicy, and melting. Given the complexity of fresh nectarines and peaches, this difference may be due more to variation in the fruit than to variation among judges. This explanation was confirmed because the replication per cultivar interaction was significant, indicating that there was some variation in the fruit that is reflected in the replication effect. Chilling injury symptoms such as mealy-woolly texture develop in specific areas in the fruit. Despite Summer Lady peaches having the highest mealy texture score, Summer Lady also had the highest overall, grassy/green fruit, and pit aromas; bitter taste; and melting and fibrous textures. This cultivar also had moderate sweetness and sourness, and the least floral aroma, firmness, and crunchy attributes among the cultivars. August Pearl nectarine had the highest

82 floral aroma (Table 6). There were no significant differences in the sweetness of Autumn Snow peach and August Pearl and Fire Sweet nectarines, which had the highest sweetness 259 scores. The sourest cultivar was August Bright nectarine, followed by Ryan Sun and O Henry peaches; the descriptive panel was able to detect some minimal bitterness in these cultivars, highest for Ryan Sun, Summer Lady, and O Henry peaches. August Bright and 262 Fire Sweet nectarines had the firmest flesh. Even though care was taken to follow recommended postharvest ripening practices for stone fruit, these cultivars still may behave differently during ripening and exhibit slight differences in texture. Differences in texture perception have also been observed in blueberries at different ripeness (Saftner et al. 2008). The three nectarine cultivars exhibited no significant differences in crunchy texture, which is somewhat expected because the nectarine cultivars were also the firmest cultivars. Other authors have found an association between firmness and crunchiness in processed tomatoes (Lee et al. 1999) and kiwifruit (Stec et al. 1989). The highest means for mealy, melting, and fibrous textures were found in Summer Lady peach, which makes sense because this cultivar was held at a different temperature than the rest (5 C) to induce mealy texture. We believe that the low floral and high grassy/green fruit and pit aromas, combined with high mealiness and low firmness and crunchy textures detected in Summer Lady were the onset of chilling injury symptoms detected by the judges. A canonical variate analysis (CVA) was conducted to understand sensory similarities and differences among the seven cultivars and characterize their significant sensory attributes (Fig. 1). The variance explained corresponds to 64.5% on the x axis and 22.3% on the y axis. There was a significant simultaneous effect among all attributes and cultivars (MANOVA, Wilks Lambda, F = 9.06; df 72, p < 0.05). Sour, sweet, floral aroma, overall aroma, and mealy were

83 the main attributes associated with the first dimension (CV1), while firmness, grassy/green fruit aroma and bitterness were the attributes related to the second dimension (CV2). Pit aroma, fibrous, and melting have short vectors, an indication of low discrimination for these attributes. Autumn Snow peach was defined as sweet and fibrous, with some grassy/green fruit aroma, and not sour or mealy. August Pearl and Fire Sweet nectarines were sweet, firm, crunchy, and not sour, with a floral aroma. Ryan Sun and O Henry peaches and August Bright nectarine were defined mainly by their sour taste; the main difference among them was in firmness. August Bright was firmer, crunchier, and also had some floral aroma lacking in Ryan Sun and O Henry. Summer Lady did not cluster with any other cultivar and was characterized as melting and mealy, with the highest overall, grassy/green fruit, and pit aromas. The relationship among sweetness, sourness, and firmness, measured both instrumentally and by a descriptive analysis panel, was studied using principal component analysis (PCA). A correlation matrix was used to explain the relationship between the instrumental quality measurements and the descriptive panel sensory attributes measurements (Fig. 2). The total variance explained was 46.4% on the x axis and 28.4% on the y axis. The PCA analysis demonstrated that the sensory attributes sourness, firmness, and sweetness were highly correlated with the instrumental measurements (RTA, firmness, and RSSC). The length of the vector indicates the discrimination among samples provided by a particular attribute; for example, firmness as rated by the descriptive panel and firmness as measured instrumentally have exactly the same length vector, indicating that both methods provide similar discrimination among samples. However, sweetness and sourness as rated by the descriptive panel had slightly longer vectors than those generated by RSSC and RTA; this may be an indication that sensory methods provide better discrimination and characterization among samples for these attributes.

84 External preference mapping was selected to study the relationship between the sensory properties and the hedonic responses; cluster analysis (Wards Method, Euclidean distance) revealed three segments among the consumer population. Consumers in cluster 1 (n=52) preferred O Henry peaches and August Bright nectarines. Cluster 2 (n=54) and cluster 3 (n=14) were similar: consumers in both liked August Pearl and Fire Sweet nectarines and Autumn Snow peaches. However, while consumers in cluster 2 accepted cultivars that were mainly characterized by sourness, consumers in cluster 3 did not like these sour cultivars at all. Verification of the differences among clusters was accomplished by one-way ANOVA applied to overall liking on each cluster per cultivar, with the exception of Fire Sweet nectarine that had no difference in liking among the three clusters. Overall liking was significantly different for each cluster for the rest of the cultivars, so even though the number of consumers in cluster 3 is small, it is important to keep this cluster. August Bright nectarines were liked equally by clusters 1 and 2, but were disliked by cluster 3. August Pearl nectarines received the lowest average hedonic score from cluster 1, but fared better with clusters 2 and 3. O Henry peaches were preferred by consumers in cluster 1 only, while Ryan Sun was liked only by consumers in cluster 2. Consumers in cluster 3 provided the highest hedonic score for Autumn Snow peaches, while cluster 1 provided the highest hedonic score for Summer Lady peaches (Fig. 3). The external preference mapping provided the main drivers of liking and which cultivars were preferred most by consumers. Sweetness was the main driver of liking for clusters 2 and 3, while cluster 1 preferences were mainly driven by the aroma composition (overall, grassy/green fruit and pit aromas). The fact that sweetness was a key driver of liking agrees with other fruit commodities such as apples (Daillant-Spinnler et al. 1996; Thybo et al. 2004), strawberries

85 (Lado et al. 2010), pineapples (Schulbach et al. 2007), tomatoes (Causse et al. 2010), and fruitbased products such as apple juice (Rødbotten et al. 2009) or pear fruit leathers (Huang and Hsieh 2005). The perception of grassy/green fruit and pit aromas that were detected in Summer Lady stored at 5 o C to induce chilling injury could be the first signs of chill injury development; it has been suggested that specific volatiles can be used to detect onset of mealiness (Crisosto and Labavitch 2002). Block-Partial least square regression, (L-PLS) was used to study the relationship between the sensory descriptors given by the panel and the hedonic responses of 333 consumers. L-PLS demonstrated differences in demographics for the clusters (Fig. 4). There were some differences in preferences according to gender; male consumers preferred more firm, crunchy cultivars and females tended to like the nectarine August Bright more. Ethnicity had a strong influence on preferences within each cluster; for example, preferences in cluster 1 were mainly associated with White-Caucasian ethnicity that assigned more importance to overall aroma. Consumers in clusters 2 and 3 were mainly Asian-Asian Americans who preferred sweet nectarines and peaches CONCLUSIONS There were strong correlations between the instrumental measurements of penetration firmness, ripe soluble solids concentration (RSSC), and ripe titratable acidity (RTA), and their 344 respective sensory panel descriptors of firmness-texture, sweet, and sour. The sensory descriptors explained cultivar differences better than instrumental measurements alone. RSSC (sweetness predictor) was the only instrumental measurement that was highly associated with overall liking by consumers. The expected price that consumers were willing to

86 pay and purchase intent increased with the overall degree of liking and was not affected by cultivar. Sweetness perception was the main driver of liking for two consumer clusters; however, for the third cluster the sensory attributes of grassy/green fruit and pit aromas were the main drivers of liking.

87 ACKNOWLEDGEMENTS The authors would like to acknowledge financial support from USDA NIFA grant # , Increasing consumption of specialty crops by enhancing their quality and safety and the California Tree Fruit Agreement for making this work possible. Thanks to Dr. Helene Hopfer, Melanie Lai, Catherine Hsu, Lauren Ginn, Paulo Wu Liu, and John Zuanyuan Chan for their assistance on this work

88 REFERENCES ARES, G., BARRIOS, S., LAREO, C. and LEMA, P Development of a sensory quality index for strawberries based on correlation between sensory data and consumer perception. Postharvest Biology and Technology 52, BETT, K Evaluating sensory quality of fresh-cut fruits and vegetables. In Fresh-Cut Fruits and Vegetables, (O. Lamikanra, ed.) pp , CRC Press, Boca Raton, Florida BRUHN, C.M Consumer and retailer satisfaction with the quality and size of california peaches and nectarines. Journal of Food Quality 18, BRUHN, C.M., FELDMAN, N., GARLITZ, C., HARWOOD, J., IVANS, E., MARSHALL, M., RILEY, A., THURBER, D. and WILLIAMSON, E Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. Journal of Food Quality 14, CAUSSE, M., FRIGUET, C., COIRET, C., LÉPICIER, M., NAVEZ, B., LEE, M., HOLTHUYSEN, N., SINESIO, F., MONETA, E. and GRANDILLO, S Consumer Preferences for Fresh Tomato at the European Scale: A Common Segmentation on Taste and Firmness. Journal of Food Science 75, S531-S541. COLARIC, M., VEBERIC, R., STAMPAR, F. and HUDINA, M Evaluation of peach and nectarine fruit quality and correlations between sensory and chemical attributes. Journal of the Science of Food and Agriculture 85, CRISOSTO, C.H Optimum procedures for ripening stone fruit. In: Management of Fruit, Ripening. Postharvest Horticulture Series No. 9. University of California, Davis, CA, pp CRISOSTO, C.H., CRISOSTO, G. and BOWERMAN, E Understanding consumer acceptance of peach, nectarine and plum cultivars. Acta Hort. (ISHS) CRISOSTO, C.H. and CRISOSTO, G.M Understanding consumer acceptance of early harvested `Hayward' kiwifruit. Postharvest Biology and Technology 22, CRISOSTO, C.H. and CRISOSTO, G.M Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biology and Technology 24, CRISOSTO, C.H. and CRISOSTO, G.M Relationship between ripe soluble solids concentration (RSSC) and consumer acceptance of high and low acid melting flesh peach and nectarine (Prunus persica (L.) Batsch) cultivars. Postharvest Biology and Technology 38, CRISOSTO, C.H., CRISOSTO, G.M., ECHEVERRIA, G. and PUY, J Segregation of peach and nectarine (Prunus persica (L.) Batsch) cultivars according to their organoleptic characteristics. Postharvest Biology and Technology 39,

89 CRISOSTO, C.H., GARNER, D., CRISOSTO, G.M. and BOWERMAN, E Increasing `Blackamber' plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biology and Technology 34, CRISOSTO, C.H., JOHNSON, R.S., DAY, K.R. and DEJONG, T Orchard factors affecting postharvest stone fruit quality. HortScience, CRISOSTO, C.H. and LABAVITCH, J.M Developing a quantitative method to evaluate peach (Prunus persica) flesh mealiness.. Postharvest Biology and Technology, CRISOSTO, G.M., CRISOSTO, C.H. and WATKINS, M Chemical and organoleptic description of white flesh nectarines and peaches. Acta Hort. (ISHS) 465, DAILLANT-SPINNLER, B., MACFIE, H.J.H., BEYTS, P.K. and HEDDERLEY, D Relationships between perceived sensory properties and major preference directions of 12 varieties of apples from the Southern Hemisphere. Food Quality and Preference 7, DELGADO, C. and GUINARD, J.-X How do consumer hedonic ratings for extra virgin olive oil relate to quality ratings by experts and descriptive analysis ratings? Food Quality and Preference 22, ESTI, M., MESSIA, M.C., SINESIO, F., NICOTRA, A., CONTE, L., LA NOTTE, E. and PALLESCHI, G Quality evaluation of peaches and nectarines by electrochemical and multivariate analyses: relationships between analytical measurements and sensory attributes. Food Chemistry 60, GUERRA, M., SANZ, M.A. and CASQUERO, P.A Influence of Harvest Dates on Quality, Storage Capacity and Sensory Attributes of European Plum cv. Green Gage. Food Science and Technology International 15, GUNNESS, P., KRAVCHUK, O., NOTTINGHAM, S.M., D'ARCY, B.R. and GIDLEY, M.J Sensory analysis of individual strawberry fruit and comparison with instrumental analysis. Postharvest Biology and Technology 52, HARKER, F.R., GUNSON, F.A. and JAEGER, S.R The case for fruit quality: an interpretive review of consumer attitudes, and preferences for apples. Postharvest Biology and Technology 28, HARKER, F.R., KUPFERMAN, E.M., MARIN, A.B., GUNSON, F.A. and TRIGGS, C.M Eating quality standards for apples based on consumer preferences. Postharvest Biology and Technology 50, HARKER, F.R., MAINDONALD, J., MURRAY, S.H., GUNSON, F.A., HALLETT, I.C. and WALKER, S.B Sensory interpretation of instrumental measurements 1: texture of apple fruit. Postharvest Biology and Technology 24,

90 HOEHN, E., GASSER, F., GUGGENBÜHL, B. and KÜNSCH, U Efficacy of instrumental measurements for determination of minimum requirements of firmness, soluble solids, and acidity of several apple varieties in comparison to consumer expectations. Postharvest Biology and Technology 27, HUANG, X. and HSIEH, F.-H Physical Properties, Sensory Attributes, and Consumer Preference of Pear Fruit Leather. Journal of Food Science 70, E177-E186. INFANTE, R., MENESES, C. and CRISOSTO, C.H Preconditioning treatment maintains taste characteristic perception of ripe September Sun peach following cold storage. International Journal of Food Science & Technology 44, KÜHN, B.F. and THYBO, A.K The influence of sensory and physiochemical quality on Danish children's preferences for apples. Food Quality and Preference 12, LADO, J., VICENTE, E., MANZZIONI, A. and ARES, G Application of a check-all-thatapply question for the evaluation of strawberry cultivars from a breeding program. Journal of the Science of Food and Agriculture 90, LAWLESS, H.T. and HEYMANN, H Sensory Evaluation of Food, Second Ed., Springer, NY. LEE, S.Y., LUNA-GUZMÁN, I., CHANG, S., BARRETT, D.M. and GUINARD, J.X Relating descriptive analysis and instrumental texture data of processed diced tomatoes. Food Quality and Preference 10, LENGARD, V. and KERMIT, M Way and 3-block PLS regressions in consumer preference analysis. Food Quality and Preference 17, MALUNDO, T.M.M., SHEWFELT, R.L., WARE, G.O. and BALDWIN, A.E.A An alternative method for relating consumer and descriptive data used to identify critical flavor properties of mango (Mangifera indica l.). Journal of Sensory Studies 16, NAPOLITANO, F., BRAGHIERI, A., PIASENTIER, E., FAVOTTO, S., NASPETTI, S. and ZANOLI, R Cheese liking and consumer willingness to pay as affected by information about organic production. Journal of Dairy Research 77, OBENLAND, D., COLLIN, S., MACKEY, B., SIEVERT, J., FJELD, K. and ARPAIA, M.L Determinants of flavor acceptability during the maturation of navel oranges. Postharvest Biology and Technology 52, ORAGUZIE, N., ALSPACH, P., VOLZ, R., WHITWORTH, C., RANATUNGA, C., WESKETT, R. and HARKER, R Postharvest assessment of fruit quality parameters in apple using both instruments and an expert panel. Postharvest Biology and Technology 52,

91 PERYAM, D.R. and PILGRIM, F.J Hedonic scale method of measuring food preferences. Food Technology 11, PREDIERI, S. and GATTI, E Effects of cold storage and shelf-life on sensory quality and consumer acceptance of Abate Fetel pears. Postharvest Biology and Technology 51, RØDBOTTEN, M., MARTINSEN, B.K., BORGE, G.I., MORTVEDT, H.S., KNUTSEN, S.H., LEA, P. and NÆS, T A cross-cultural study of preference for apple juice with different sugar and acid contents. Food Quality and Preference 20, ROSSITER, K.L., YOUNG, H., WALKER, S.B., MILLER, M. and DAWSON, D.M The effects of sugars and acids on consumer acceptability of kiwifruit. Journal of Sensory Studies 15, SAFTNER, R., POLASHOCK, J., EHLENFELDT, M. and VINYARD, B Instrumental and sensory quality characteristics of blueberry fruit from twelve cultivars. Postharvest Biology and Technology 49, SCHULBACH, K.F., PORTIER, K.M. and SIMS, C.A Evaluation of overall acceptability of fresh pineapple using the regression tree approach. Journal of Food Quality 30, SINESIO, F., CAMMARERI, M., MONETA, E., NAVEZ, B., PEPARAIO, M., CAUSSE, M. and GRANDILLO, S Sensory Quality of Fresh French and Dutch Market Tomatoes: A Preference Mapping Study with Italian Consumers. Journal of Food Science 75, S55-S67. STEC, M.G.H., HODGSON, J.A., MACRAE, E.A. and TRIGGS, C.M Role of fruit firmness in the sensory evaluation of kiwifruit (Actinidia deliciosa cv Hayward). Journal of the Science of Food and Agriculture 47, STEFANI, G., ROMANO, D. and CAVICCHI, A Consumer expectations, liking and willingness to pay for specialty foods: Do sensory characteristics tell the whole story? Food Quality and Preference 17, THYBO, A.K., KÜHN, B.F. and MARTENS, H Explaining Danish children's preferences for apples using instrumental, sensory and demographic/behavioural data. Food Quality and Preference 15, YENKET, R., CHAMBERS IV, E. and ADHIKARI, K A comparison of seven preference mapping techniques using four software programs. Journal of Sensory Studies 26,

92 TABLE 1. CODES, MEANS (X) AND STANDARD DEVIATIONS (S.D.) OF FLESH FIRMNESS, RIPE SOLUBLE SOLIDS CONCENTRATION (RSSC) AND RIPE TITRATABLE ACIDITY (RTA) FOR NECTARINE AND PEACH CULTIVARS EVALUATED IN THE DESCRIPTIVE ANALYSIS AND CONSUMER STUDY. Fruit Cultivar Type* Cultivar Flesh Firmness RSSC RTA Code Color (N) (%) (%) X S.D. X S.D. X S.D. Fire N-F FS Yellow Nectarine Sweet August Bright August Pearl N-HA AB Yellow N-LA AP White O'Henry P-F OH Yellow Ryan P-MA RS Yellow Peach Sun Autumn Snow Summer Lady P-LA AS White P-SM SL Yellow * N-F = nectarine flavorful, N-HA = nectarine high acidity, N-LA = nectarine low acidity, P-F = peach flavorful, P-MA = peach medium acidity, P-LA = peach low acidity, P-SM = peach slightly mealy.

93 TABLE 2. SENSORY ATTRIBUTE DEFINITIONS AND STANDARDS USED FOR TRAINING THE DESCRIPTIVE ANALYSIS PANEL TO EVALUATE NECTARINE AND PEACH CULTIVARS. Attributes Description Standard Overall aroma Intensity of aroma (whole aroma) Verbal description Floral Smell of flowers Nectarine cultivars Almond Smell of almonds 1% Artificial flavor (McCormick) over 50 g of peach/nectarine paste * Grassy/green fruit Smell of grass associated with unripe fruit 0.01 grams over 50 g of peach/nectarine paste * Pit Woody aroma associated with fruit with traces of pit 10 pits were removed from the fruit and used as a standard Overall flavor Intensity of flavor (whole flavor) Verbal description Sweet Sweet taste, example sucrose solution Sucrose solutions 0-50 g/l in spring water Fruit with different soluble solids concentrations ranging from 8-16% Sour Sour taste, example citric acid solution Citric acid solutions g/l in spring water Fruit with different titratable acidities ranging from 0.2% to 0.8% Bitter Bitter taste, example like coffee Caffeine solutions 0.7 g/l 2.5 g/l Firmness Flesh only: measured at the first bite ranging from soft to hard Fruit with different firmness, ranging from 4.4 to 44.4 N Crunchy Flesh only: making a crunching sound Verbal description when chewed or pressed Juicy Flesh only: amount of liquid Verbal description Mealy Gritty, sandy texture, dry not juicy Verbal description Fruit with different degrees of mealiness Melting Smooth Fibrous How easy to fracture into mouth (high melting examples ice cream, chocolate) Texture of the fruit related to having a continuous even surface After first two bites, amount of fibers in the sample * A paste (50/50) peach/nectarine was prepared as base. Verbal description and use of ice cream and chocolate to explain concept Verbal description Verbal description 559

94 TABLE 3. DEGREE OF LIKING AND CONSUMER ACCEPTANCE ACCORDING TO RIPE SOLUBLE SOLIDS CONCENTRATION (RSSC) CLASSES FOR DIFFERENT NECTARINE AND PEACH CULTIVARS AND THEIR CORRESPONDING INSTRUMENTAL MEASUREMENTS. Acceptance Cultivar RSSC class n Liking Neutral Rejection RSSC RTA Type * (%) (1-9) ** (%) (%) (%) (%) (%) a *** N-LA a a > a RSSC class n Liking Acceptance Neutral Rejection RSSC RTA a a N-F a a > a RSSC class n Liking Acceptance Neutral Rejection RSSC RTA a N-HA a a > a RSSC class n Liking Acceptance Neutral Rejection RSSC RTA < a a P-LA a a < a RSSC class n Liking Acceptance Neutral Rejection RSSC RTA < a ab P-F bc abc < c RSSC class n Liking Acceptance Neutral Rejection RSSC RTA < a bc P-MA ab bc < c RSSC class n Liking Acceptance Neutral Rejection RSSC RTA P-SM a

95 a a < a * Cultivar type: N-F = nectarine flavorful, N-HA = nectarine high acidity, N-LA = nectarine low acidity, P-F = peach flavorful, P-MA = peach medium acidity, P-LA = peach low acidity, P-SM = peach slightly mealy. ** Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. *** Same letters within the same column indicate no significant difference between means p < 0.05.

96 TABLE 4. PURCHASE INTENT, PRICE EXPECTATION, SECOND CONSUMPTION, AND RIPE SOLUBLE SOLIDS CONCENTRATION (RSSC) BY DEGREE OF LIKING (1-9) FOR NECTARINE CULTIVARS BY NORTHERN CALIFORNIA CONSUMERS. Cultivar type * N-F N-HA N-LA Degree of liking ** (score) Purchase Intent *** Price **** ($) Second Consumption ***** RSSC (%) a ****** 0.00 a 1.0 a a 0.40 a 2.0 a a 0.70 a 2.3 a b 0.60 a 3.0 b b 1.30 b 3.1 b c 1.30 b 3.9 c d 1.60 b 4.5 d d 1.60 b 4.9 d 14.3 P-value < < < a 0.00 a 1.0 a a 0.30 a 1.3 ab b 0.40 a 1.9 bc c 0.40 a 2.2 cd de 0.70 ab 3.0 de d 1.30 b 3.5 e d 1.40 b 3.9 e e 1.50 b 4.3 f f 2.50 c 4.8 g 12.5 P-value < < < ab 0.80 ab 1.9 a a 0.40 a 2.2 ab b 0.70 ab 3.0 bc b 1.10 b 3.5 c c 1.10 b 3.9 d d 1.70 c 4.3 e e 1.90 c 4.8 f 13.1 P-value < < < * Cultivar type: N-F = nectarine flavorful, N-HA = nectarine high acidity, N-LA = nectarine low acidity, P-F = peach flavorful, P-MA = peach medium acidity, P-LA = peach low acidity, P-SM = peach slightly mealy. ** Degree of liking score: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. *** Purchase intent: 1 = definitely would not buy, 2 = probably would not buy, 3 = neither would not buy, nor would buy, 4 = probably would buy, 5 = definitely would buy.

97 **** Price willing to pay per pound of the sample tasted at retail. ***** Second consumption: 1 = certainly will not consume this nectarine again, 2 = probably will not consume this nectarine again, 3 = not sure or undecided, 4 = probably will consume this nectarine again, 5 = certainly will consume this nectarine again. ****** Same letters within the same column indicate no significant difference between means. 600

98 TABLE 5. PURCHASE INTENT, PRICE EXPECTATION, SECOND CONSUMPTION, AND RIPE SOLUBLE SOLIDS CONCENTRATION (RSSC) BY DEGREE OF LIKING (1-9) FOR PEACH CULTIVARS BY NORTHERN CALIFORNIA CONSUMERS. Cultivar Type * P-F P-MA P-LA P-SM Degree of liking ** (score) Purchase Intent *** Price **** ($) Second Consumption ***** RSSC (%) a ****** 0.00 a 1.0 a 9.9 ab a 0.10 a 1.4 a 9.9 ab b 0.40 a 1.7 ab 9.6 ab b 0.50 a 2.1 b 9.5 ab c 0.70 ab 3.0 c 9.3 a c 1.00 bc 3.0 c 9.8 ab d 1.40 cd 3.9 d 10.1 bc e 1.60 d 4.2 de 10.6 cd f 1.5cd 4.8 e 11.8 d P-value < < < a 0.10 a 1.0 a 10.5 ab a 0.40 ab 1.6 a 9.4 a b 0.30 a 1.5 a 10.3 ab b 0.80 abc 2.4 b 10.3 ab c 0.80 abc 2.6 b 11.3 b d 0.90 bc 3.4 c 10.3 ab d 1.20 c 3.7 c 10.8 b e 1.70 d 4.4 d 11.2 b e 2.00 d 5.0 d 10.6 ab P-value < < < a cd ab 0.30 ac 1.5 ab a 0.20 ab 1.3 a a 0.60 bc 2.2 bc bc 0.80 bc 2.6 c c 0.90 c 3.2 d d 1.50 d 3.9 d e 1.30 d 4.7 e e 2.0 e 4.9 e 12.9 P-value < < < a 0.0 a 1.0 a a 0.1 ab 1.0 a b 0.1 ab 1.8 b c 0.7 bcd 2.7 c c 0.8 cde 3.0 cd d 1.0 de 3.5 d e 1.4 e 3.9 e f 1.7 f 4.5 f g 2.0 f 4.7 f 11.8 P-value < < <

99 * Cultivar type: N-F = nectarine flavorful, N-HA = nectarine high acidity, N-LA = nectarine low acidity, P-F = peach flavorful, P-MA = peach medium acidity, P-LA = peach low acidity, P-SM = peach slightly mealy. ** Degree of liking score: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. *** Purchase intent: 1 = definitely would not buy, 2 = probably would not buy, 3 = neither would not buy, nor would buy, 4 = probably would buy, 5 = definitely would buy. **** Price willing to pay per pound of the sample tasted at retail. ***** Second consumption: 1 = certainly will not consume this nectarine again, 2 = probably will not consume this nectarine again, 3 = not sure or undecided, 4 = probably will consume this nectarine again, 5 = certainly will consume this nectarine again. ****** Same letters within the same column indicate no significant difference between means.

100 TABLE 6. OVERALL MEANS FOR THE SIGNIFICANT SENSORY ATTRIBUTES OF AROMA, TASTE AND TEXTURE FOR SEVEN NECTARINE AND PEACH CULTIVARS. Aroma Attributes Overall Floral Grassy/green fruit Pit Cultivar Cultivar Cultivar Cultivar type * Mean type Mean Type Mean type Mean P-SM 5.4 a ** N-LA 4.2 a P-SM 2.5 a P-SM 2.9 a P-F 4.6 a b N-HA 3.1 b P-F 2.1 a b P-F 1.9 b N-LA 4.5 b P-MA 2.9 b N-F 1.7 c b N-HA 1.7 b N-HA 4.4 c b P-LA 2.7 b N-HA 1.4 c b d N-F 1.5 b c d P-MA 4.4 c b N-F 2.5 b P-MA 1.3 c d P-MA 1.3 c d N-F 3.7 c d P-F 2.4 b N-LA 1.0 c d P-LA 0.9 c d P-LA 3.3 d P-SM 2.4 B P-LA 0.9 d N-LA 0.8 LSD 0.78 LSD 0.83 LSD 0.68 LSD 0.61 Taste Attributes Sweet Sour Bitter Cultivar Cultivar Cultivar type Mean type Mean Type Mean P-LA 5.3 a N-HA 6.0 a P-MA 1.3 a N-LA 5.1 a P-MA 4.6 b P-SM 1.2 a b N-F 4.9 a P-F 4.6 b P-F 1.0 a b c P-SM 3.6 b P-SM 3.7 c P-LA 0.8 b c P-MA 3.0 b c P-LA 1.3 d N-F 0.7 b c N-HA 2.7 c N-F 1.3 d N-LA 0.6 c P-F 2.4 c N-LA 0.8 d N-HA 0.5 c LSD 0.69 LSD 0.68 LSD 0.51 Texture Attributes Firmness Crunchy Mealy Melting Fibrous Cultivar Cultivar Cultivar Cultivar Cultivar type Mean Type Mean Type Mean type Mean type Mean N-HA 4.6 a N-HA 3.6 a P-SM 2.2 a P-SM 6.8 a P-SM 4.8 a N-F 4.1 a b N-LA 3.3 a N-LA 1.3 b P-LA 4.7 b N-LA 4.6 a b N-LA 3.8 c b N-F 3.0 a N-HA 1.1 b c P-MA 4.4 b P-F 4.4 a b P-F 3.6 c b d P-F 2.3 b N-F 1.1 b c N-LA 3.8 b c P-LA 4.4 a b P-LA 3.2 c d P-LA 1.9 b P-LA 0.8 b c N-F 3.8 b c N-HA 4.3 a b c P-MA 3.0 d P-MA 1.9 b P-MA 0.8 b c P-F 3.8 b c P-MA 3.8 b c P-SM 1.2 e P-SM 1.0 c P-F 0.5 N-HA 3.4 c N-F 3.4 c LSD 0.63 LSD 0.68 LSD 0.68 LSD 0.97 LSD 0.95

101 * Cultivar type: N-F = nectarine flavorful, N-HA = nectarine high acidity, N-LA = nectarine low acidity, P-F = peach flavorful, P-MA = peach medium acidity, P-LA = peach low acidity, P-SM = peach slightly mealy. ** Same letters within the same column indicate no significant difference between means p<0.05.

102 Figure captions: FIG. 1. CANONICAL VARIATE ANALYSIS (CVA) FOR SEVEN FRESH NECTARINE AND PEACH CULTIVARS. Each sphere represents a cultivar * (refer to Table 1 for cultivar descriptions). There is no significant difference between cultivars (p<0.05) when two spheres overlap. Attributes are represented by vectors. * Cultivar type (cultivar): N-F = nectarine flavorful ( Fire Sweet ), N-HA = nectarine high acidity ( August Bright ), N-LA = nectarine low acidity ( August Pearl ), P-F = peach flavorful ( O Henry ), P-MA = peach medium acidity ( Ryan Sun ), P-LA = peach low acidity ( Autumn Snow ), P-SM = peach slightly mealy ( Summer Lady ). FIG. 2. PRINCIPAL COMPONENT ANALYSIS BY CORRELATION OF INSTRUMENTAL VS. DESCRIPTIVE VARIABLES. Cultivars are represented by ellipses and attributes by vectors. Please refer to Table 1 for a full description of the samples. Cultivar type (cultivar): N-F = nectarine flavorful ( Fire Sweet ), N- HA = nectarine high acidity ( August Bright ), N-LA = nectarine low acidity ( August Pearl ), P- F = peach flavorful ( O Henry ), P-MA = peach medium acidity ( Ryan Sun ), P-LA = peach low acidity ( Autumn Snow ), P-SM = peach slightly mealy ( Summer Lady ). FIG. 3. EXTERNAL PREFERENCE MAPPING INCLUDING CONSUMER SEGMENTS Cultivars are represented by solid circles. Please refer to Table 1 for a full description of the samples. Cultivar type (cultivar): N-F = nectarine flavorful ( Fire Sweet ), N-HA = nectarine high acidity ( August Bright ), N-LA = nectarine low acidity ( August Pearl ), P-F = peach flavorful ( O Henry ), P-MA = peach medium acidity ( Ryan Sun ), P-LA = peach low acidity ( Autumn Snow ), P-SM = peach slightly mealy ( Summer Lady ). Solid square indicates consumers belong to cluster 1(n=52). Consumers in cluster 2 (n=54) are indicated by a triangle, and cluster 3 (n=14) by a cross. FIG. 4. BLOCK-PARTIAL LEAST SQUARE REGRESSION (L-PLS). Descriptive analysis data: sensory attributes per cultivar [X] vs. overall liking by 120 consumers indicated by solid squares [Y] and consumers demographics [Z]. Variables in the outer circle are more important than variables in the inner circle. Cultivar type (cultivar): N-F = nectarine flavorful ( Fire Sweet ), N-HA = nectarine high acidity ( August Bright ), N-LA = nectarine low acidity ( August Pearl ), P-F = peach flavorful ( O Henry ), P-MA = peach medium acidity ( Ryan Sun ), P-LA = peach low acidity ( Autumn Snow ), P-SM = peach slightly mealy ( Summer Lady ).

103 Postharvest Biology and Technology 34 (2004) Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance Carlos H. Crisosto a,, David Garner a, Gayle M. Crisosto a, Earl Bowerman b a Department of Pomology, University of California, Davis, CA 95616, USA b Plant Science Department, California State University, Fresno, CA 93740, USA Received 23 December 2003; accepted 6 June 2004 Abstract Blackamber plum (Prunus salicina Lindell) consumer acceptance and market life were highly dependent on harvest date. For fruit within the most common industry ripe soluble solids concentration (RSSC) range ( %), ripe titratable acidity (RTA) played a significant role in consumer acceptance. Plums within this RSSC range combined with low RTA ( 0.60%) were disliked by 18% of consumers, while plums with RTA 1.00% were disliked by 60% of consumers. Plums with RSSC 12.0% had 75% consumer acceptance, regardless of RTA. Fruit harvested between 35.6 and 17.8 N had high consumer acceptance because of lower RTA and higher RSSC than earlier harvested fruit. Ripening plums before consumption decreased TA by approximately 30% from the TA measured at harvest. In some cases, this decrease in TA during ripening may increase the acceptability of plums that would otherwise be unacceptable. Development of chilling injury (CI) symptoms limited market life of fruit harvested early ( N) and late ( N). Late harvested fruit were more likely to develop flesh translucency (overripe or bladderiness) when stored at 5 C, whereas early harvested fruit had low consumer acceptance and were more prone to develop flesh bleeding/browning during storage at 0 or 5 C. Based on this work, Blackamber plums are well adapted to late harvest but proper postharvest temperature management, including ripening, and marketing within its market life potential are necessary to avoid the onset of storage disorders and maintain flavor Elsevier B.V. All rights reserved. Keywords: In store consumer test; Fruit quality; Market life; Flesh browning; Gel breakdown; Flesh translucency; Chilling injury 1. Introduction Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). Plums (Prunus salicina Lindell) have the potential to contribute greatly to human nutrition because of their richness in fiber and antioxidants (Kim et al., 2003; Stacewicz-Sapuntzakis et al., 2001). Despite news of /$ see front matter 2004 Elsevier B.V. All rights reserved. doi: /j.postharvbio

104 238 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) their benefits to human health, consumption of plums in the USA has remained very low, averaging only 0.64 kg per capita per year over the last 22 years (US Department of Agriculture, 2003). This low rate of consumption has been attributed to a lack of fruit ripening before consumption, poor flavor, and internal flesh problems attributed to chilling injury (CI) (Bruhn et al., 1991). Plums are susceptible to chilling injury (Abdi et al., 1997; Crisosto et al., 1999; Dodd, 1984; Robertson et al., 1991; Taylor et al., 1993, 1994, 1995) similar to many peaches and nectarines (Mitchell and Kader, 1989; Smith, 1934; Von Mollendorff, 1987). We believe that plum consumption may increase by offering fruit that are tasty, ripe and free of chilling injury. Fruit maturity controls quality attributes, such as soluble solids concentration (SSC), titratable acidity (TA), firmness, and market life potential (Crisosto et al., 1995). In general, ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) determine consumer acceptance. Fruit softening (loss of firmness) is related to bruising susceptibility and is an important factor when considering the latest date fruit can be successfully harvested. Unfortunately, there is no detailed information available on the role of maturity in both CI and consumer acceptance for plums (Abdi et al., 1997; Dodd, 1984; Taylor et al., 1993). The objective of this work was to determine the ideal harvesting period for Blackamber plums based on consumer acceptance and market life potential in order to increase plum consumption. Blackamber, a black, high acid plum was used in this work because it is highly susceptible to CI (Crisosto et al., 1999) and has a reputation for low consumer acceptance within the trade. 2. Materials and methods During the 2002 season, studies to understand the relationship between plum maturity, consumer acceptance and market life were carried out using mature Blackamber Japanese plum (P. salicina Lindell) trees growing at the Kearney Agricultural Center (Parlier, CA). The trees were planted on m spacing and trained to a two leader perpendicular Kearney V conformation. Five replications of six trees each were selected for this study, and these trees received standard commercial horticultural care throughout the growing season. Fruit were harvested for quality, storage and sensory analysis on June 20, June 26, July 2 and July 8. On each harvest 25 fruit were selected from each tree and these combined within each replication to yield 150 fruit per replication per harvest date Quality attribute measurements Ten fruit from each of the five replications on each harvest date (June 20, June 26, July 2 and July 8) were used to determine fruit mass (g), flesh color, soluble solids concentration, titratable acidity, and flesh firmness. After weighing the fruit with an electronic balance, the skin was removed from both cheeks, the tip, suture and shoulder using a vegetable peeler. Flesh color was measured on both cheeks with a Minolta colorimeter (Minolta, CR-200, Japan) and is expressed as hue angle (h ), lightness (L*) and chroma (saturation) according to the Commission Internationale de l Eclairage (CIE). Next, at each position where the skin had been removed flesh firmness was measured using a UC firmness tester with a 7.9 mm tip. Then, a longitudinal wedge (from stem end to calyx end) was removed from each fruit, pressed through cheesecloth, and the SSC of the juice was measured with a temperature compensated refractometer (model ATC-1, Atago Co., Tokyo, Japan). Juice from each replication was pooled to form a composite sample, and TA was measured with an automatic titrator (Radiometer, Copenhagen, Denmark). All fruit quality data was subjected to analysis of variance (ANOVA) prior to the LSD mean separation using the SAS program Market life evaluations To measure market life potential, another 120 fruit from each replication on each sampling date were subdivided into two storage treatments with 60 fruit stored at 0 C (85% RH) and 60 fruit at 5 C (85% RH). Plums were stored under these conditions for up to 6 weeks according to our previously published protocol (Crisosto et al., 1999). After 2, 4 and 6 weeks storage, 20 fruit per replication were removed from each storage temperature and warmed to 20 C. Flesh firmness of 10 fruit per treatment-replication was evaluated immediately as previously described. The remaining 10 fruit per treatment-replication were ripened at 20 C and 85% RH until flesh firmness reached N. Extra fruit

105 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) Fig. 1. Blackamber plum chilling injury symptoms observed during cold storage: flesh browning, flesh bleeding, gel breakdown, flesh translucency (overripe or bladderiness). samples from each treatment were used to determine the end of the ripening period for each treatment. When ripe, chilling injury symptoms, such as gel breakdown, flesh browning, flesh bleeding, and flesh translucency (overripe or bladderiness) were visually evaluated immediately after cutting fruit in half (Fig. 1). Fruit were determined to have gel breakdown when the tissues between the pit and the middle of the mesocarp were soft, translucent, and lacking in juice when squeezed. Flesh browning was evaluated as a brown coloration originating near the pit and extending out into the flesh. Flesh bleeding was evaluated as an accumulation of red pigment either around the stone or immediately beneath the epidermis. Plum flesh translucency occurred as water soaked areas observed mainly in the outer mesocarp immediately under the epidermis as a consequence of the fruit being overripe. In these translucent areas, the vascularization within the otherwise clear tissues could easily be observed. In some fruit, translucent areas were also observed within the inner mesocarp, but in most of these cases these tissues turned brown or formed gel. Plums that had a dry appearance and little or no juice expressed upon hand squeezing were considered mealy. These fruit were also informally tasted for a feeling of mealiness (like sand in the mouth) and/or off flavors to corroborate visual assessment. Fruit that remained hard after ripening and from which little or no juice could be expressed upon hand squeezing were considered leathery. Market life potential, based on losses as a consequence of CI symptoms, was subjectively defined as the number of weeks each cultivar could be stored at 0 C (maximum) or 5 C (minimum) before CI became limiting (Crisosto et al., 1999). The end of market life was determined to occur when more than 25% of the fruit had symptoms of chilling injury as described by Mitchell and Kader (1989).

106 240 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) In-store consumer test An in-store consumer test was conducted in which 100 consumers were surveyed in a major supermarket in Fresno Co., CA. Each consumer was presented four ripe Blackamber plum samples one sample from each harvest date as previously described (Crisosto and Crisosto, 2001). At the supermarket, the SSC of each fruit was measured as previously described and the fruit were separated into the following four classes: %, %, %, >13.9%. In this way each consumer was presented fruit samples representing the entire range of SSC. A portion of juice from each fruit was also frozen for analysis of TA in the laboratory. The ripe soluble solids concentration and ripe titratable acidity were measured for each fruit presented to the consumers and correlated with their response. This technique reduces the error that results from fruit to fruit variability that is masked when using average RSSC and RTA values. For each fruit sample, the consumer was asked to taste it, then to indicate if he/she likes, neither likes nor dislikes, or dislikes the sample. Then, the consumer was asked to indicate his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The consumer s response was recorded using a 9-point hedonic scale (1- dislike extremely to 9-like extremely). Consumer acceptance was measured as both degree of liking (1 9) and percentage acceptance. Percentage acceptance was calculated as the number of consumers liking the fruit sample (score > 5.0) divided by the total number of consumers within that sample (Lawless and Heymann, 1998). In a similar way the percentage of consumers disliking and/or indifferent to that sample was calculated. The degree of liking data was subjected to analysis of variance prior to the LSD mean separation using the SAS program. 3. Results and discussion 3.1. Quality attributes At harvest time and during storage, flesh firmness did not differ between positions of measurement on the fruit. During the harvesting period, flesh firmness decreased from 44.9 to 12.9 N during this period of ripening on the tree. Fruit fresh mass increased significantly until the third harvest date, but not significantly thereafter (Table 1). Fruit flesh color measured as hue angle and L* decreased significantly during the harvesting period. This indicates that the flesh color became less pale green and lighter yellow during this period of ripening on the tree. In our previous work, we found that flesh color is a better indicator of maturity than other fruit quality parameters including skin color (Crisosto, 1998). In fact, in Blackamber plum the skin color becomes completely dark purple relatively early during maturation before the beginning of the commercial harvest period. SSC and TA measurements changed significantly during on the tree ripening. SSC increased from 10.3% on June 20 to 11.9% on July 8, while TA decreased significantly from 1.15% to 0.42% during this Table 1 Blackamber plum fruit size, flesh firmness and flesh color measured at harvest; and soluble solids concentration (SSC), titratable acidity (TA as % malic acid) and sugar-to-acid ratio (SSC:TA) measured at harvest and on ripe fruit collected on different harvest dates Harvest date At harvest After ripening a Fruit mass (g) Cheek firmness (N) Flesh color SSC (%) TA (%) SSC:TA SSC (%) TA (%) SSC:TA L* Chroma Hue angle Harvest 1 (6/20/02) Harvest 2 (6/26/02) Harvest 3 (7/2/02) Harvest 4 (7/8/02) P-value < < < < < LSD b NS a Fruit ripened at 20 C and 85% RH until flesh firmness reached N. b Mean separation by LSD test at P > 0.05.

107 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) same period of time. This resulted in a significant increase in the sugar-to-acid ratio (SSC:TA) from 9.2 on June 20 to 28.2 on July 8. SSC did not change significantly during the ripening process; however there was an approximate 30% reduction in TA during ripening (Table 1). As TA decreased during ripening and SSC did not change, the ripe SSC:TA ratio increased from 13.2 for fruit harvested June 20 to 37.5 for fruit harvested July 8. Acidity, skin astringency and firmness decrease during ripening while aroma develops contributing to an increase in consumer acceptance (Crisosto, 1999; Meredith et al., 1992). In a previous maturity study conducted in the same Blackamber research plot in 1989 (Mitchell, 1990), cheek firmness decreased from N on the first harvest date (June 11) to18.25 N on the last (July 2). SSC increased from 8.6% to 11.3% during this same period of time. As in our study, fruit size did not increase after the third harvest date (June 25). Our previous work (Crisosto et al., 2001) demonstrated that Blackamber plums, which are more susceptible to impact bruising than Fortune, Royal Diamond and Angeleno plums, with flesh firmness 13.3 N exposed to impact forces of 245 X G (simulating impacts occurring during standard packingline operations) were susceptible to bruising injury. Plums with flesh firmness >13.34 N were highly resistant to impact injury Market life potential Different expressions of CI, such as flesh browning, flesh mealiness, gel breakdown, flesh translucency (overripe or bladderiness), and flesh bleeding were observed during cold storage (Fig. 1). The onset of these symptoms was associated with fruit maturity and storage temperature. The onset of CI symptoms was delayed when fruit were stored at 0 C rather than at 5 C. Maturity affected market life of Blackamber plums when stored at either 0 or 5 C. Fruit harvested on the first harvest date (June 20) had a shorter market life than fruit harvested later (Table 2). Maximum market life was 3 weeks shorter for fruit harvested on the earliest date than for fruit harvested on any of the later dates. The predominant CI symptoms that developed at 0 C were flesh mealiness and flesh bleeding/browning. Minimum market life was only 2 weeks for fruit harvested on June 20, June 26 and July 8; Table 2 Market life of Blackamber plums harvested on four different dates then stored at 0 or 5 C Harvest date Maximum market life a (weeks at 0 C) Harvest 1 (6/20/02) 2 2,3 <2 3,4 Harvest 2 (6/26/02) ,4 Harvest 3 (7/2/02) ,3,4 Harvest 4 (7/8/02) ,3,4 Minimum market life (weeks at 5 C) a End of market life based on chilling injury (CI) determined when 25% of the fruit became mealy 1 or leathery 2, or had flesh bleeding/browning 3 or gel breakdown/translucency 4. Superscript indicates limiting condition. and 3 weeks for fruit harvested on July 2. For all four harvests, flesh bleeding/browning and gel breakdown/translucency limited minimum market life. In addition, for the two later harvests mealiness was limiting at 5 C. Thus, early harvested fruit had a short market life (up to 2 weeks) under both storage temperatures. Late harvested fruit had limited market life (up to 3 weeks) only when stored at 5 C. These results agreed with our previously reported findings that the onset of CI symptoms occurred after 5 and 2 weeks in California grown Blackamber plums stored at 0 and 5 C, respectively (Crisosto et al., 1999). Previous research in South Africa (Hartmann et al., 1988) pointed out that Songold plum tissue breakdown occurred immediately under the fruit s epidermis. In other work carried out in Australia on Radiant, Gulfruby and Shiro plums, it was concluded that early harvested fruit had poor quality and late harvested fruit were more susceptible to CI (Abdi et al., 1997) In store consumer test Degree of liking was significantly related to RSSC, RTA and the RSSC:RTA. RSSC ranged from 8.0 to 18.0% with a mean of 12.0%, and RTA ranged from 1.21 to 0.15% with a mean of 0.62% within the population of fruit used in this consumer test. Based on industry-wide fruit quality surveys over the previous 5 years, the average RSSC has ranged from 10.8 to 11.7% (data not shown). There was a significant interaction between RSSC, RTA and degree of liking. For this reason, data were further analyzed using three levels of RTA (Table 3). Within this range of quality attributes Blackamber plums were liked from dis-

108 242 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) Table 3 Acceptance of Blackamber plums by American consumers at different levels of ripe soluble solids concentration (RSSC) and ripe titratable acidity (RTA) measured as percentage malic acid Quality attributes Degree of liking (1 9) a Acceptance (%) Neither like nor dislike (%) Dislike (%) RTA 0.60% RSSC < 10.0% RSSC % RSSC % RSSC 14.0% RTA % RSSC < 10.0% RSSC % RSSC % RSSC 14.0% RTA 1.00% RSSC < 10.0% RSSC % LSD b P-value a Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. b Mean separation by LSD test at P > liked moderately (mean score = 3.3) to liked very much (mean score = 7.8) and consumer acceptance increased from 8.3 to 88.6%. RTA did not affect degree of liking when plums had RSSC 12.0%. Plums with RSSC 12.0% were always liked slightly (mean score = 6.3) to liked very much (mean score = 7.8) by consumers. However, when plums had RSSC of %, RTA played a significant role in consumer acceptance. Plums with RSSC of %, combined with low ( 0.60%) or moderate ( %) RTA, were liked slightly (mean score = 6.2) to neither liked nor disliked (mean score = 5.3) with an acceptance of 56.4% and 34.5%, respectively. Blackamber plums with RSSC of % combined with a RTA 0.60% had a degree of liking similar to plums with RSSC of %. Blackamber plums with RSSC of % and a RTA 1.00% had the lowest degree of liking disliked moderately (mean score = 3.3) and had a consumer acceptance similar to plums with a RSSC <10.0% and RTA 1.00%. Plums with RSSC <10.0% had the lowest level of consumer acceptance regardless of RTA. The percentage of consumers that chose the dislike option ranged from 1.3 to 66.7% and this was also related to RSSC and RTA. The percentage of consumers that chose the dislike option is high in comparison to other commodities and it may be explained by a potential bias against dark colored plums. For plums which had RSSC 12.0%, disregarding RTA, the percentage of consumers that chose the neither like nor dislike option was approximately 18%. Less than 7% of the consumers disliked these fruit. Studies have associated high consumer acceptance of fruit with high soluble solids concentration in peaches (Testolini, 1995; Hilaire, 2003), and sweet cherries (Crisosto et al., 2003). However, consumer acceptance was more closely related to SSC:TA than SSC alone within a given range of RSSC for early Hayward kiwifruit (Crisosto and Crisosto, 2001), and early Redglobe table grapes (Crisosto and Crisosto, 2002). A similar situation was observed in this plum cultivar. When RSSC was 12.0%, consumer acceptance was not dependent on RSSC:RTA but when RSSC was 12.0%, SSC influenced consumer responses Conclusions Blackamber is a plum cultivar that is well suited to late harvesting because of its low susceptibility to bruising during postharvest handling operations, such as harvesting, packing and transportation (Crisosto et al., 1999). This recommendation should be extended to

109 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) other Japanese plums, such as Fortune, Friar, Royal Diamond, and Angeleno, that are even less likely to bruise during postharvest operations. Our work suggests the use of firmness as an indicator of how late to safely harvest ( Tree Ripe ), thereby maximizing orchard quality for other plum cultivars too. However, the decision when to harvest should also take into account other factors, such as fruit drop, environmental conditions, hand labor availability, market prices, distance to market, potential transportation damage, and temperature management at the receiving location. To maximize flavor and storage potential, Blackamber plums should be harvested when they reach a minimum SSC within the range of 10 12% and a titratable acidity < 0.70%, but with a firmness 14.0 N. Any fruit in the population with firmness 14.0 N will likely be bruised during standard postharvest operations. This work also pointed out the need for plum ripening at the production site prior to shipment as it has been recommended for peaches (Crisosto et al., 2004) or at the warehouse before retail display (Crisosto, 1999). Our ongoing sensory work on plum, peach and nectarine cultivars suggested that the relationship between consumer acceptance and quality attributes will be cultivar dependent, thus, the establishment of a minimum quality index based on SSC or SSC:TA needs to be evaluated for each cultivar. Also, plums should be marketed and consumed within their potential market life. This information provides guidance for growers, packers, shippers, handlers and retailers in designing their postharvest strategy to increase plum consumption. Acknowledgement The authors would like to acknowledge financial support from the California State University Agricultural Research Initiative and the California Tree Fruit Agreement for making this work possible. References Abdi, N., Holford, P., McGlasson, W.B., Mizrahi, Y., Ripening behaviour and responses to propylene in four cultivars of Japanese type plums. Postharvest Biol. Technol. 12, Bruhn, C.M., Feldman, N., Garlitz, C., Hardwood, J., Ivans, E., Marshall, M., Riley, A., Thurber, D., Williamson, E., Consumer perceptions of quality: apricots, cantaloupes, peaches, pears, strawberries, and tomatoes. J. Food Qual. 14, Crisosto, C.H., Developing maturity indices for full red plum cultivars. California Tree Fruit Agreement Research Report, pp Crisosto, C.H., Optimum procedures for ripening stone fruit. In: Management of Fruit Ripening. University of California, Davis, Postharvest Horticulture Series, vol. 9, pp Crisosto, C.H., Crisosto, G.M., Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22, Crisosto, C.H., Crisosto, G.M., Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24, Crisosto, C.H., Crisosto, G.M., Metheney, P., Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol. Technol. 28, Crisosto, C.H., Garner, D., Andris, H.L., Day, K.R., Controlled delayed cooling extends peach market life. HortTechnology 14 (1), Crisosto, C.H., Mitchell, F.G., Johnson, R.S., Factors in fresh market stone fruit quality. Postharvest News and Information 6, Crisosto, C.H., Mitchell, F.G., Ju, Z.G., Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. Hortscience 34 (6), Crisosto, C.H., Slaughter, D.C., Garner, D., Boyd, J., Stone fruit critical bruising thresholds. J. Am. Pomological Soc. 55, Dodd, M.C., Internal breakdown in plums. Deciduous Fruit Grower, August, Hartmann, P.E., De Kock, V.A., Taylor, M.A., Picking maturities and cold storage requirements of Songold plums. Deciduous Fruit Grower 38, Hilaire, C., The peach industry in France: State of art, research and development. In: Marra, F., Sottile, F. (Eds.), Proceeding of the First Mediterranean Peach Symposium, Agrigento (Italy), 10 September. Kim, D.O., Chun, O.K., Kim, Y.J., Moon, H.Y., Lee, C.Y., Quantification of polyphenolics and their antioxidant capacity of fresh plums. J. Agric. Food Chem. 51, Lawless, H.T., Heymann, H., Sensory Evaluation of Food: Principles and Practices. Food Science Texts Series. Chapman & Hall, New York. Meredith, F.I., Senter, S.D., Forbus Jr., W.R., Robertson, J.A., Postharvest quality and sensory attributes of Byrongold and Rubysweet plums. J. Food Qual. 15, Mitchell, F.G., Selecting and handling high quality stone fruit. California Tree Fruit Agreement, Research Report, 1989, 15 pp. Mitchell, F.G., Kader, A.A., Factors affecting deterioration rate. In: LaRue, J.H., Johnson, R.S. (Eds.), Peaches, Plums and Nectarines-Growing and Handling for Freshmarket. Publication 3331, University of California DANR, Oakland, pp Robertson, J.A., Meredith, F.I., Lyon, B.G., Effect of cold storage on the quality characteristics of Au-Rubrum plums. J. Food Qual. 14, Smith, W.H., Cold storage of Elberta peaches. Ice and Cold Storage 37,

110 244 C.H. Crisosto et al. / Postharvest Biology and Technology 34 (2004) Stacewicz-Sapuntzakis, M., Bowen, P.E., Hussain, E.A., Damayanti- Wood, B.I., Farnsworth, N.R., Chemical composition and potential health effects of prunes: a functional food? Crit. Rev. Food Sci. Nutr. 41 (4), Taylor, M.A., Jacobs, G., Rabe, E., Dodd, M.C., Physiological factors associated with overripeness, internal breakdown and gel breakdown in plums stored at low temperature. J. Hort. Sci. 68 (5), Taylor, M.A., Rabe, E., Dodd, M.C., Jacobs, G., Effect of storage regimes on pectolytic enzymes, pectic substances, internal conductivity and gel breakdown in cold stored Songold plums. J. Hort. Sci. 69 (3), Taylor, M.A., Rabe, E., Jacobs, G., Dodd, M.C., Effect of harvest maturity on pectic substances, internal conductivity, soluble solids and gel breakdown in cold stored Songold plums. Postharvest Biol. Technol. 5, Testolini, A., Momento di racolta, qualita, condizionamento e confezionamento delle pesche. In: Proceedings Symposium La peschicoltura Veronesa alle soglie del Verona, February, 25, pp US Department of Agriculture, Fruit and tree nuts situation and outlook yearbook. Market and Trade Economics Division, Economic Research Service, US Department of Agriculture, Washington, DC. Von Mollendorff, L.J., Woolliness in peaches and nectarines: a review. I. Maturity and external factors. Hort. Sci./Tuinbouwetenskap 5, 1 3.

111 Describing the Appearance and Flavor Profiles of Fresh Fig (Ficus carica L.) Cultivars Ellena S. King, Helene Hopfer, Megan T. Haug, Jennifer D. Orsi, Hildegarde Heymann, Gayle M. Crisosto, and Carlos H. Crisosto Abstract: Twelve fig cultivars, including cultivars destined for the fresh and dried markets, were harvested from 6 locations and evaluated by a trained panel using descriptive sensory analysis. Instrumental measurements were taken at harvest and also during sensory analysis. Each fresh fig cultivar had a characteristic appearance and flavor sensory profile regardless of the source. The primary flavor attributes used to describe the fig cultivars were fruity, melon, stone fruit, berry, citrus, honey, green, and cucumber. Maturity levels significantly affected the chemical composition and sensory profiles of the fig cultivars. Less mature figs had a higher compression force, a thicker outer skin, and higher ratings for green and latex flavors, firmness, graininess, bitterness, tingling, and seed adhesiveness. Meanwhile, more mature figs had higher soluble solids concentration, and were perceptibly higher in fruit flavors, juiciness, stickiness, sliminess, and sweetness. The specific sensory terminology used for fig appearance and flavor profiles will assist with communication between marketers and consumers, which can increase fresh fig consumption. Keywords: canonical variate analysis, fig cultivar, fruit maturity, sensory analysis Practical Application: The development of a unique set of descriptors for each fresh fig cultivar allows better communication between fig growers, retailers, and consumers. Consumers who are able to correctly anticipate how a particular fig variety tastes are more likely to be satisfied and purchase more figs. This work also demonstrates the need to develop and select cultivars with high flavor attributes for future fresh fig production. Introduction The common fig tree (Ficus carica L.) is grown on approximately 454,000 hectares worldwide, with an annual fruit production of over 1.3 million tons (Food and Agriculture Organization 2007). The majority of fig growing areas have moderate climates, exhibiting low relative humidity (lower than 25%), intense luminosity, high summer temperatures around 32 to 37 C, and moderate winters with temperatures above 1 C (Obenauf and others 1978). The United States of America ranks 7th in the world s fig production after Turkey, representing approximately 3.3% of total production, and California accounts for approximately 98% of production in the U.S., due to its favorable climate. The main cultivars planted in California are Calimyrna, Mission, Brown Turkey, and Kadota (Stover and others 2007). Most fig production in California is used for the dried market; only a small percentage is harvested for fresh fig consumption (Stover and others 2007). Until recently, fresh figs represented less than 5% of total fig production (Stover and others 2007). However, after 2002, fresh fig production experienced nearly a 4-fold increase, reaching 16% of California s fig production in 2006 (U.S. Dept. of Agriculture 2007). Most of the fig cultivars currently grown in California have been selected for their favorable characteristics in the dried fig MS Submitted 7/18/2012, Accepted 9/11/2012. Authors King, Hopfer, and Heymann are with Dept. of Viticulture and Enology, Univ. of California, Davis, One Shields Ave., Davis, CA 95616, U.S.A. Authors Haug, Orsi, G. M. Crisosto, and C. H. Crisosto are with Dept. of Plant Sciences, Univ. of California, Mail Stop 2, One Shields Ave., Davis, CA 95616, U.S.A. Direct inquiries to author C. H. Crisosto ( chcrisosto@ucdavis.edu). market. Therefore, it is necessary for breeding programs to select new fig cultivars with favorable traits for fresh fruit consumption, such as intense fruity flavors, an adequate balance of sweetness and sourness, and good postharvest performance (Stover and others 2007). It is necessary to evaluate cultivars grown in California for fresh fig instrumental quality parameters, since no studies have yet been performed on this subject. Previous studies evaluating fig cultivars have mostly been performed on those native to Turkey (Caliskan and Polat 2008; Polat and Caliskan 2008), which are not presently grown; however, these cultivars are different from those currently grown or tested in California (Crisosto and others 2010). One approach to promote and increase fresh fig consumption is to identify the sensory attributes of fresh fig cultivars and use this information to group cultivars based on similar sensory profiles. This can be achieved using descriptive sensory analysis with a trained panel. Cultivar classification will help match consumer preferences, and enhance current promotion and marketing programs. Also, this sensory technique will create a terminology to describe their diverse sensory attributes. The use of sensory methods to characterize foods and beverages is well documented in the literature. Descriptive sensory analysis techniques have been applied to products in order to characterize sensory properties of different cultivars, growing regions, production methods, and so on. So far, descriptive sensory analysis of figs has been related to dried figs and fig jam (Levaj and others 2010; Haug s personal communication). For fresh fruit, descriptive sensory analyses have been used to characterize cultivars of peaches, nectarines, plums, pluots, apples, and kiwifruits (Crisosto and others 2006; Swahn and others 2010; Jaeger and others 2011). S: Sensory & Food Quality C 2012 The Regents of the University of California, Davis Campus Department of Plant Sciences doi: /j x Further reproduction without permission is prohibited Vol. 77, Nr. 12, 2012 Journal of Food Science S419

112 Sensory attributes of fig cultivars... An in-store consumer acceptance test was used to investigate consumer preferences for 4 commercial California fresh fig cultivars at both the commercial mature and the tree ripe stages (Crisosto and others 2010). The results indicated that consumers are able to perceive differences between fresh fig cultivars and preferred late harvested fresh figs (tree ripe maturity compared with commercial ripe maturity) (Crisosto and others 2010). This indicates that there is a market potential to target consumer groups that prefer different fig cultivars and sensory attributes. In order for the United States and other countries to develop and/or select cultivars for a viable fresh fig industry, research initiatives need to be established to develop flavorful fig cultivars with favorable traits for fresh fruit production. The aim of this study was to create a terminology and characterize the sensory properties of California-grown fresh fig cultivars using a descriptive sensory analysis of 12 fresh fig cultivars, grown in California, from 6 different sources. Materials and Methods Plant materials Thirty-nine fresh fig samples consisting of 11 cultivars and one selection, hereafter called cultivars, were assessed (Figure 1). The fig cultivars were collected from 6 sources in California: commercial growers orchards in Fresno, Madera, and Merced counties (Sources A, B, C, and D), the fig plot at the Kearney Agricultural Research and Extension Center (KAC) in Parlier, CA (Source E) and the USDA, ARS National Clonal Germplasm Repository (NCGR) for Fruit & Nut Crops in Davis, CA (Source F) during the 2011 season (Table 1). All fig samples were packed in labeled, single-layer boxes with trays. The fig samples of each cultivarsource-replicate consisted of 100 fruits. The figs were stored at 0 C for an average of 2 d (±SD 1.3) prior to analysis. Fruit selection and preparation Prior to evaluation, the most representative fruits of each cultivar-source-replicate were selected based on appearance, maturity, and freedom from defects: 12 whole fruits for initial fruit instrumental quality measures and photos; and 48 fruits for sensory analysis, divided into 3 presentation replicates of 16 fruits each. It should be noted that although there was some variation in maturity stages within each source replicate, the same proportions of maturities were distributed evenly among the samples for fruit instrumental quality analysis and sensory analysis, based on nondestructive visual and firmness cues. At the time of analysis, figs were simultaneously removed from cold storage and warmed to room temperature, approximately 20 C, under a fan for approximately 1 h. S: Sensory & Food Quality Figure 1 A visual representation of the 12 California fresh fig cultivars used in the descriptive sensory analysis. S420 Journal of Food Science Vol. 77, Nr. 12, 2012

113 Sensory attributes of fig cultivars... Table 1 Details of the California-grown fresh fig cultivars, sources, number of source-replicates and cultivar-source codes (numbers indicating order of assessment) evaluated in this study. # of Cultivarsource Source Fig cultivar Source a replicates Code b v Black Mission A 1 BM8A B 2 BM7B BM22B C 2 BM10C BM21C D 2 BM32D BM36D Brown Turkey A 4 BT1A BT13A BT16A BT17A B 1 BT6B D 3 BT20D BT23D BT25D Calimyrna B 1 CAL3B C 3 CAL11C CAL12C CAL19C Ischia Black E 1 IB29E Kadota A 1 KAD4A B 3 KAD5B KAD24B KAD27B D 1 KAD33D Panachee A 1 PAN26A E 1 PAN34E F 1 PAN39F Sequoia E 1 SEQ30E Sierra A 5 SI2A SI9A SI14A SI15A SI18A C 1 SI31C Tina E 1 TIN35E UCR F 1 UCR38F Violette de Bordeaux F 1 VDB37F Zidi E 1 ZID28E a Sources A, B, C, and D commercial growers orchards located in Fresno, Madera, and Merced counties; Source E Kearney Agricultural Research and Extension Center, Parlier, CA; Source F USDA, ARS National Germplasm Repository, Davis, CA. b Number corresponds to the order of assessment (1 = 1st fig cultivar-source assessed; 39 = last fig cultivar-source assessed). Initial fruit instrumental quality assessment Initial fruit instrumental quality was measured on 10 figs per source replicate prior to sensory analysis. This included fruit weight, ostiole diameter, skin color, compression force, compression distance, visual maturity stage, flesh color, skin thickness, soluble solids concentration (SSC), and titratable acidity (TA). Individual fresh weight was measured with a digital scale (model PM 4000, Mettler Instrument Corp., Hightstown, N.J., U.S.A.). Ostiole diameter and skin thickness were measured with calipers (SPI 2000, dial caliper). Fruit compression force was measured using a Texture Analyzer (TAXT) (model TA.XT plus, Texture Technologies Corp., Scarsdale, N.Y., U.S.A.) fitted with an aluminum 5 cm diameter 20 mm high cylinder probe (TA-25). Each whole fig was oriented on the stage on its side with the ostiole facing the operator, and then the fig was compressed on the cheek with the probe at a speed of 1 mm per second until the fig split and the maximum value of force (compression force) was recorded. Visual maturity was assessed on each cut fig half as commercial mature (CM), tree ripe (TR), or overripe (OR). A fig was considered commercial mature when the fruit was physiologically mature and the flesh gave a little to the touch; tree ripe maturity was riper and softer than commercial maturity with translucent flesh; while a fig was considered overripe when the flesh was very soft, completely translucent and the skin was beginning to thin and deteriorate. Skin color was measured on both cheeks of the fruit perpendicular to the widest diameter and flesh color was measured on the flesh of both halves of the fig cut longitudinally along the split that resulted from the compression, using a Minolta colorimeter (model CR-300, Osaka, Japan). For SSC and TA, 3 composite juice samples were obtained by pressing the flesh of 3 or 4 figs through cheesecloth with a hand press. The juice was used for the determination of SSC with a temperature compensated digital refractometer (model PR 32α, Atago Co.,Tokyo, Japan). Four grams of each composite juice sample were used for determination of TA with an automatic titrator (model TIM 850, Radiometer Analytical, Lyon, France), reported as a percentage of citric acid. Sensory analysis A descriptive sensory analysis was conducted on 12 fresh fig cultivars in August and September Twelve panelists (8 females) were recruited from the Univ. of California, Davis campus, aged 20 to 32 y old (mean age of 24 y), most with previous experience in descriptive sensory analyses. Panelists were selected based on having no allergic reactions to figs and being available at the necessary times. All panelists participated in three 1-h training sessions. Sensory descriptors were generated through panel consensus, using fresh figs and pictures of figs (for appearance), and also from documented and anecdotal evidence, as not all fig cultivars were available for tasting in the training sessions. Some of the published sources used included (McEachern 1996; IPGRI and CIHEAM 2003; Polat and Caliskan 2008; Gozlekci 2010; Oliveira and others 2010; Podgornik and others 2010). Panelists practiced the formal testing procedure and use of the rating scales, prior to formally rating the figs. Panelists rated 5 appearance attributes on the same whole fig in a grey-background light box (GretagMacbeth, Chicago) under sodium light. In isolated, ventilated tasting booths, also under sodium lighting, panelists were each given one whole fruit, which they assessed for 3 texture attributes. Then, they cut the fig in half lengthwise from the stem end to the ostiole using a small paring knife, to assess 5 internal appearance attributes, 25 aroma attributes (including other ), 6 in-mouth texture attributes, 4 taste attributes, 25 flavor attributes (including other ), and 3 aftertaste attributes. One half of each of the fruits assessed was returned to organizers for half-fruit instrumental quality assessment. Panelists were required to taste the skin and expectorate all samples. Panelists were presented with 9 figs per session, 3 fig cultivarsources in triplicate, in a randomized and balanced order across the panelists. All figs were served at room temperature, on paper plates, with 3 digit random codes that differed for each panelist. Panelists cleansed their palates with water in between samples. For appearance, 2 whole fruits of each presentation replicate were placed adjacent to one another in the light box one fruit positioned vertically, and the other oriented on its side with the ostiole facing panelists. The presentation replicates were each given 3 digit S: Sensory & Food Quality Vol. 77, Nr. 12, 2012 Journal of Food Science S421

114 Sensory attributes of fig cultivars... S: Sensory & Food Quality Table 2 Fresh fig appearance and flavor sensory attributes, definitions and reference standards used in the descriptive sensory analysis. Attribute Definition Reference standard Appearance of exterior Size Colors and percentage Color variation Texture of skin Position of ostiole Texture of exterior Firmness-exterior Skin smoothness Skin hairiness Appearance of interior Colors and percentage of pith Colors and percentage of pulp Colors and percentage of seeds Size of internal fruit cavity Percentage of seeds and pulp Aroma Overall aroma intensity Small, medium, large White, yellow, gold, light green, dark green, orange, pink, red, doark red, maroon, brown, blue, violet, purple, black Striped, spotted Smooth, powdery/white coating, matte, glossy, hairy, waxy, furry Closed, open Low anchor (not firm) High anchor (very firm) Low anchor (not smooth) High anchor (very smooth) Low anchor (not hairy) High anchor (very hairy) White, off-white, yellow, pink, red, green White, yellow, pink, red, dark red, maroon, brown, green, purple White, yellow, pink, red, brown None, small, medium, large Low anchor (low) High anchor (high) 1 Intensity of fig aroma Low anchor (low) dried Mission fig (Trader Joe s) 2 High anchor (high) Fruity Includes generic fruit and grape 1 piece canned yellow papaya + 1 piece canned orange, no juice (Dole tropical fruits) Apple 2 1cm 2 pieces of fresh apple, including peel Banana 2 2mm 2 pieces of fresh banana, no peel Melon 2 cm x 1 cm piece of fresh honeydew melon, without peel Stone fruit Includes peach and plum 2 cm slice of fresh peach, including peel + 2 cm slice of fresh plum, including peel Berry Includes raspberry, strawberry, and dark berries 1 frozen blueberry + 1 frozen blackberry (Best Yet naturally sweet berry medley) + 1 frozen raspberry (Cascadian Farm Organic raspberries) + 1 frozen strawberry (Best Yet naturally sweet strawberries) 1 Floral Includes generic floral, rose perfume, violet tsp. of rose water (Sadaf) Citrus 2 cm slice of fresh lime peel + 2 cm slice of fresh lemon peel + 2 cm slice of fresh orange peel Honey Includes sweet aromas 1 tsp. Clover honey (SueBee) Toffee Includes butterscotch, caramel and brown sugar 1 tsp. Butterscotch caramel topping (Mrs. Richardson s) 1 Coconut tsp. sweetened coconut (Angel Flake, Baker s) 2 1 Chocolate square of unsweetened baking chocolate (Baker s) 4 1 Nutty 2 2 walnut baking pieces (Trader Joe s) 1 Spicy tsp. ground allspice (McCormick) 2 Green Includes generic green, vegetal, herbal, leafy and fruit peel 3 frozen cut green beans (Birds Eye) + 4 frozen sweet peas (Birds Eye) Cucumber 4 mm slice halved of fresh cucumber Grassy 2 tsp. fresh cut grass 1 Minty tsp. mint jelly (Reese) 2 Earthy Includes dirt and soil 1 tsp. all-purpose potting soil (Black Gold) Woody 1 tsp. pine shavings (Nature s Care, Alfalfa Hay mini-bales) Latex - 1 Sulfur Includes garlic, onion and meaty 2 2 solution 1 Vinegar tsp. white vinegar (Best Yet) 2 Flavor Overall flavor intensity Low anchor (low) Intensity of fig flavor In-mouth perception of the same attributes as aroma, without standards High anchor (high) Low anchor (low) High anchor (high) Continued. S422 Journal of Food Science Vol. 77, Nr. 12, 2012

115 Sensory attributes of fig cultivars... Table 2 Continued. Attribute Definition Reference standard Taste Skin thickness Sweetness Sourness Bitterness Texture in-mouth Pith firmness Sliminess Seed crunchiness Juiciness Stickiness Graininess Aftertaste Astringency Seed adhesiveness Tingling Low anchor (thin) High anchor (thick) Low anchor (not sweet) High anchor (very sweet) Low anchor (not sour) High anchor (very sour) Low anchor (not bitter) High anchor (very bitter) Low anchor (not firm) High anchor (very firm) Low anchor (not slimy) High anchor (very slimy) Low anchor (not crunchy) High anchor (very crunchy) Low anchor (not juicy) High anchor (very juicy) Low anchor (not sticky) High anchor (very sticky) Low anchor (not grainy) High anchor (very grainy) Low anchor (not astringent) High anchor (very astringent) Amount of product packed in teeth surfaces after chewing Low anchor (not adhesive) High anchor (very adhesive) Low anchor (not tingly) High anchor (very tingly) random codes that were the same for each panelist, in which the computer session forced panelists to rate the appearance of each replicate in a randomized and balanced order. Panelists were given food at the end of each session, as well as a moderately priced gift card upon completion of the sensory analysis. All attributes, reference standards, and descriptions are listed in Table 2. The intensity of some attributes, such as texture, taste, and aftertaste were rated using an unstructured 15 cm line scale anchored by wordings of absence or low to extreme or high. Two aroma and flavor attributes ( overall aroma/flavor intensity and intensity of fig aroma/flavor ) were also rated using scales. All other aroma, flavor, and appearance attributes were assessed using the check all that apply (CATA) method, used by Campo and others (2010). This method was used, as it was thought that figs have subtle aromas and flavors that would be better assessed on a present or absent basis. The vocabulary used by panelists to describe certain parts of the fig were as follows: the skin was the outer most layer of the fruit; the pith was the layer inside the fruit directly beneath the skin; the pulp was the fleshy layer, located in the middle of the fruit, surrounding the seeds; the seeds were the hard, circular parts in the middle of the pulp; the internal fruit cavity was the size of the hole where the seeds are found, and the percentage of seeds and pulp was the total area of seeds and pulp (colored area) in relation to the whole fruit. Half-fruit instrumental quality assessment Each panelist returned a half of each fig sample to organizers after sensory analysis and placed them in labeled, lidded 5.5 oz. soufflé cups(solo R Cup Co., Lake Forest, Ill., U.S.A.). Halffruit instrumental quality assessment included visual maturity stage, skin thickness, compression distance, SSC, and TA, following the methods previously described, except for compression distance. The compression distance of the individual half fig was measured by placing it flesh side down on the stage and measuring the distance to compress the fig half with a compression force of 10 N at a speed of 1 mm per second. Data analysis FIZZ software (Version 2.1, Biosystemes, France) was used for the collection of all sensory data. All data were analyzed per fig cultivar, source, and source replicate (also known as order of assessment) (that is, the 39 samples separately) (see Table 1 for more details), as not all fig cultivars were assessed an equal number of times, and it was hypothesized that there would be differences in maturity stages between figs assessed at the start and at the end of the study. The scaled sensory data and physicochemical data were analyzed separately from the CATA data. For the scaled sensory data, missing values were imputed using mean presentation replicate values, and the data were analyzed using a three-way analysis of variance (ANOVA) with two-way interactions for judge, presentation replicate, and fig cultivar-source, using a pseudo-mixed test, with mean square (judge fig cultivar-source) as the error. Variance was assessed using the Tukey HSD order of comparison. A canonical variate analysis (CVA) was performed on the raw data, when the Wilks Lambda P value was less than The 95% confidence interval bubbles were calculated and fig cultivars were grouped. The CATA data were statistically analyzed using the methods of Campo and others (2010). In summary, the number of checks or citations was summed and frequency tables were created. Missing data were not imputed. The frequency data were then used for correspondence analysis using symmetrical analysis. Two correspondence analyses were performed, one for all appearance attributes, and the other for all categorical aroma and flavor attributes. S: Sensory & Food Quality Vol. 77, Nr. 12, 2012 Journal of Food Science S423

116 Sensory attributes of fig cultivars... S: Sensory & Food Quality Table 3 Initial fruit instrumental quality measurements for each cultivar, source, and source replicate, averaged across 10 replicates, and an average of each cultivar (in boldface, see Table 1 for details of study design), standard deviation italicized in parentheses. TA Skin Ostiole Compression Cultivar- (% citric thickness size force source Weight (g) SSC (%) acid) (mm) (mm) (N) BM7B 27.0 (4.84) 15.3 (1.76) 0.12 (0.01) 2.8 (0.35) 3.8 (0.90) 8.8 (0.42) BM8A 40.5 (4.89) 14.6 (0.78) 0.23 (0.04) 3.3 (0.57) 5.1 (1.30) 22.5 (0.52) BM10C 35.8 (5.72) 14.9 (0.31) 0.11 (0.01) 3.1 (0.33) 5.7 (0.79) 7.8 (0.17) BM21C 32.9 (4.92) 18.1 (1.21) 0.14 (0.01) 3.0 (0.41) 4.8 (0.82) 7.8 (0.35) BM22B 47.1 (9.78) 20.9 (2.14) 0.18 (0.01) 2.9 (0.46) 6.9 (1.26) 7.8 (0.19) BM32D 30.3 (4.94) 18.5 (4.27) 0.15 (0.02) 3.2 (0.47) 5.8 (1.00) 8.8 (0.17) BM36D 29.7 (8.89) 20.9 (1.12) 0.23 (0.00) 3.1 (0.55) 5.4 (1.01) 11.8 (1.02) Black Mission 34.8 (9.07) 17.6 (3.04) 0.17 (0.05) 3.1 (0.47) 5.4 (1.33) 10.8 (0.68) BT1A 58.7 (9.65) 14.3 (0.81) 0.19 (0.05) 4.8 (0.92) 9.6 (1.13) 23.5 (1.49) BT6B 52.1 (8.63) 18.6 (1.64) 0.12 (0.02) 5.5 (1.05) 7.5 (1.99) 9.8 (0.35) BT13A 46.8 (11.52) 18.4 (0.75) 0.09 (0.01) 5.0 (1.21) 5.5 (0.91) 13.7 (0.72) BT16A 52.1 (7.99) 15.9 (1.18) 0.13 (0.02) 4.7 (0.82) 6.6 (1.15) 13.7 (0.56) BT17A 42.3 (11.48) 19.8 (2.10) 0.14 (0.02) 4.3 (0.97) 5.2 (0.75) 12.7 (0.79) BT20D 63.6 (8.05) 19.4 (0.62) 0.14 (0.03) 4.9 (0.79) 7.7 (1.48) 12.7 (0.80) BT23D 57.9 (9.10) 20.4 (0.69) 0.12 (0.02) 5.1 (0.66) 8.2 (1.91) 14.7 (0.64) BT25D 70.6 (7.52) 22.5 (0.89) 0.16 (0.01) 4.2 (0.66) 10.3 (1.19) 13.7 (0.22) Brown Turkey 55.5 (12.38) 18.6 (2.63) 0.14 (0.03) 4.8 (0.95) 7.6 (2.14) 14.3 (0.84) CAL3B 62.3 (15.72) 16.9 (1.39) 0.23 (0.03) 3.1 (0.81) 7.8 (1.59) 17.6 (0.56) CAL11C 77.1 (8.60) 16.3 (0.06) 0.17 (0.03) 3.9 (0.41) 9.5 (1.12) 13.7 (0.46) CAL12C 74.8 (9.45) 16.0 (1.77) 0.23 (0.06) 4.2 (0.81) 9.6 (1.90) 14.7 (0.48) CAL19C 75.1 (5.09) 18.2 (1.01) 0.21 (0.04) 3.7 (0.67) 10.6 (1.47) 11.8 (0.28) Calimyrna 72.3 (11.67) 16.9 (1.31) 0.21 (0.04) 3.7 (0.78) 9.4 (1.81) 14.5 (0.49) IB29E 31.1 (9.16) 22.0 (1.36) 0.30 (0.05) 2.6 (0.60) 4.2 (0.55) 10.8 (0.37) KAD4A 58.6 (10.07) 17.9 (1.23) 0.27 (0.04) 6.2 (0.87) 9.3 (0.92) 41.2 (1.52) KAD5B 45.4 (10.13) 19.1 (2.34) 0.13 (0.01) 5.3 (1.28) 8.5 (1.21) 22.5 (1.11) KAD24B 31.5 (4.31) 18.1 (0.64) 0.14 (0.03) 3.9 (0.80) 8.5 (0.99) 12.7 (0.48) KAD27B 29.3 (4.51) 18.9 (0.17) 0.12 (0.02) 3.7 (0.62) 9.3 (1.21) 13.7 (0.63) KAD33D 30.9 (5.02) 16.4 (1.37) 0.14 (0.01) 3.9 (0.83) 9.2 (0.79) 15.7 (0.89) Kadota 39.1 (13.43) 18.0 (1.50) 0.16 (0.06) 4.6 (1.32) 9.0 (1.07) 21.2 (1.44) PAN26A 55.8 (10.44) 18.1 (0.89) 0.42 (0.08) 3.2 (0.86) 9.8 (1.95) 20.6 (0.52) PAN34E 40.9 (9.64) 21.9 (2.70) 0.55 (0.05) 2.8 (0.35) 6.1 (1.49) 14.7 (0.45) PAN39F 31.4 (4.08) 17.9 (1.27) 0.50 (0.01) 2.6 (0.47) 6.4 (1.76) 19.6 (0.97) Panachee 36.2 (8.70) 20.1 (2.76) 0.52 (0.04) 2.7 (0.42) 6.3 (1.59) 18.3 (0.77) SEQ30E 48.6 (9.36) 19.7 (0.78) 0.34 (0.04) 3.8 (0.59) 3.6 (0.28) 24.5 (0.95) SI2A 45.0 (13.16) 16.5 (0.99) 0.14 (0.02) 3.5 (0.90) 5.9 (1.17) 20.6 (1.24) SI9A 41.5 (9.64) 15.5 (0.75) 0.22 (0.01) 3.3 (0.53) 3.9 (0.49) 19.6 (0.84) SI14A 46.7 (6.50) 14.9 (5.20) 0.17 (0.01) 3.5 (0.80) 3.8 (0.53) 32.3 (2.63) SI15A 52.5 (11.82) 17.5 (3.20) 0.19 (0.01) 4.0 (1.00) 2.8 (0.53) 17.6 (0.80) SI18A 57.7 (8.84) 18.7 (1.71) 0.21 (0.05) 3.5 (0.64) 3.4 (0.40) 17.6 (0.51) SI31C 38.1 (8.05) 15.9 (3.67) 0.31 (0.04) 2.4 (0.52) 4.5 (0.40) 16.7 (0.72) Sierra 46.9 (11.58) 16.4 (2.82) 0.21 (0.06) 3.4 (0.86) 4.0 (1.17) 20.7 (1.39) TIN35E 29.7 (3.78) 18.1 (1.01) 0.14 (0.02) 2.3 (0.49) 6.8 (0.78) 13.7 (0.36) UCR38F 45.6 (9.48) 16.2 (0.72) 0.32 (0.03) 3.3 (0.79) 5.7 (0.46) 9.8 (0.21) VDB37F 29.5 (8.87) 17.5 (1.79) 0.52 (0.00) 3.2 (0.58) 7.3 (1.61) 9.8 (0.41) ZID28E 53.6 (9.45) 21.4 (1.80) 0.27 (0.03) 3.0 (0.46) 8.0 (1.75) 10.8 (0.38) For the initial fruit instrumental quality data, a two-way ANOVA was performed with one-way interactions for each measurement replicate and fig cultivar-source. The means of scaled sensory and chemical data, and sums of categorical data were correlated using Pearson s pairwise correlations. A multiple factorial analysis was performed using Pearson s correlation matrix on the mean scaled sensory data, all mean physicochemical data, and the standardized coefficients of the 1st 2 of the correspondence analyses for the appearance, aroma and flavor CATA factors data. SAS (Version 9.2, SPSS Inc. IBM, Ill., U.S.A.), JMP (Version 8.0, SAS Inst., Cary, N.C., U.S.A.), and XLSTAT (Version Addinsoft, N.Y., U.S.A.) software were used for all data analysis. Results and Discussion Initial fruit instrumental quality measures for each fig cultivar from each source and order of assessment, as well as an average of each fig cultivar are summarized in Table 3. The SSC, TA, and other initial fruit instrumental quality measurements correlated well (r < 0.77, P < 0.05) to their corresponding half fruit instrumental quality measures. The instrumental quality measurements also correlated well with the associated sensory attributes, such as ostiole size and position of ostiole (r = 0.70, P < 0.05), TA and sourness (r = 0.71, P < 0.05), weight and whole fruit size (r = 0.82, P < 0.05), and compression force and firmnessexterior (r = 0.69, P < 0.05). All scaled sensory attributes were significantly different for the fig cultivar-sources (P < 0.05). There were also significant interactions between presentation replicate and fig cultivar-source (P < 0.05), using the pseudo-mix test for intensity of fig aroma, seed adhesiveness, stickiness, graininess, bitterness, and overall flavor intensity. When an ANOVA was performed testing for the effects of judge, presentation replicate and fig cultivar only (without source or order of assessment), the same sensory attributes were significantly different for fig cultivar and S424 Journal of Food Science Vol. 77, Nr. 12, 2012

117 Sensory attributes of fig cultivars... presentation replicate (data not shown). It should be noted that there were very low scores (less than 10% of scale) for bitterness and tingling. A number of sensory attributes were closely related, as shown by their proximity to one another in Figure 2. Stickiness and sliminess were strongly positively correlated (r = 0.88, P < 0.05), as were astringency and graininess (r = 0.76, P < 0.05), and seed adhesiveness, to a lesser extent (r = 0.57, P < 0.05). Crunchiness was moderately associated with firmnessexterior (r = 0.40, P < 0.05), whereas pith firmness and juiciness were moderately negatively correlated (r = 0.53, P < 0.05), despite their close location in Figure 2. Intensity of fig flavor was driving ratings of overall flavor intensity (r = 0.88, P < 0.05). The main sensory attributes of the 12 fresh fig cultivars tested are presented in Table 4. For the CATA data, these were the attributes rated by more than 40% of panelists, and for the scaled sensory data, these attributes were those significantly different between cultivars based on the Tukey HSD order of comparison. Each of the fresh fig cultivars studied had a different appearance and sensory flavor profile (Table 4), however, there were commonalities among sensory attributes used to describe the fresh fig cultivars. The main categorical aroma attributes rated by the panelists were: fruity, melon, green, cucumber, and grassy. The main categorical flavor attributes rated were: fruity, melon, stone fruit, berry, citrus, honey, green, and cucumber. There were moderate to strong correlations between the same aroma and flavor attributes, however, more flavor attributes were used to describe the fresh fig cultivars than aroma attributes. This suggests that the in-mouth flavors of fresh figs are more intense and/or complex than the aromas. Panelists were required to assess both the skin and pith of the fresh fig during the descriptive sensory analysis, which might have contributed to the intensity of the flavors. The results of a CVA of the scaled sensory data for fig cultivarsource and order of assessment are shown in the biplot in Figure 2. The center of each circle represents the mean of the fig cultivar-source, and the circle represents the 95% confidence interval. Overlapping circles signify that the fig cultivar-sources are not significantly different from each other. The CVA biplot in Figure 2 explains 43% of the variance, with an equal amount explained by the 1st 2 axes. CV3 explains an additional 15% (data not shown). The 12 fresh fig cultivars studied are grouped by color in Figure 2. Most of the fig cultivars are spatially separated on the biplot and do not overlap other cultivars, indicating that the fresh fig cultivars studied had characteristic appearance and flavor sensory profiles, described in Table 4. The fig cultivars with multiple sources in this study are generally located close together on the biplot, indicating that source is not affecting the sensory properties of the figs to a large degree. Other tree fruits have also been shown to differ more between cultivars than growing locations (Crisosto and others 2006). There are, however, a number of outlier cultivar-sources in Figure 2. All Brown Turkey (BT) figs are located in the bottom S: Sensory & Food Quality Figure 2 A canonical variate analysis (CVA) biplot of the scaled sensory data for fig cultivar-source and source replicate. The center of the circle represents the mean of each of the fig cultivar-sources, and the circle represents a 95% confidence interval. Black Mission (orange); Brown Turkey (green); Calimyrna (red); Ischia Black (brown); Kadota (purple); Panachee (light blue); Sequoia (dark blue); Sierra (grey); Tina (light purple); Violette de Bordeaux (yellow); UCD (blue); Zidi (light blue). See Table 1 for cultivar-source codes. Vol. 77, Nr. 12, 2012 Journal of Food Science S425

118 Sensory attributes of fig cultivars... S: Sensory & Food Quality Table 4 Summary of the main appearance and flavor sensory descriptors for the 12 California fresh fig cultivars studied, averaged across judges (n = 12), presentation replicates (n = 3), source and source replicate (see Table 1 for study design). Fig cultivar Appearance descriptors Sensory descriptors Black Mission Small to medium size Thin skin, soft pith Exterior: Aroma: Violet, purple, black High intensity Spotted Fig, green, cucumber, grassy Smooth, powdery Flavor: Closed ostiole Fig, fruity, melon, stone fruit, honey, cucumber Interior: Low sourness Pith color: white, off-white Low graininess Pith thickness: thin Pulp color: white, pink Seed color: yellow Small internal cavity Brown Turkey Large size Aroma: Exterior: High intensity Light green, purple Fig, melon, green, cucumber, grassy Striped Flavor: Hairy Fig, fruity, stone fruit, honey, cucumber Open ostiole Low sourness Interior: Grainy, astringent Pith color: white, off-white Low seed crunchiness Pith thickness: thick Tingling Pulp color: white, pink, red Seed color: yellow Large internal cavity Calimyrna Large size Firm exterior, firm pith Exterior: Aroma: Light green, yellow Cucumber, grassy Striped Flavor: Hairy High intensity Open ostiole Fig, fruity, melon, stone fruit, berry, honey, Interior: cucumber Pith color: white, off-white, yellow Sweet, low bitterness Pith thickness: moderate Grainy, juicy Pulp color: white, yellow, pink, red Crunchy, high seed adhesiveness Seed color: brown Tingling Minimal internal cavity Ischia Black Small size Soft exterior Exterior: Aroma: Violet, purple Cucumber Spotted Flavor: Smooth, powdery Fig, fruity, stone fruit, berry, honey Open ostiole Bitter Interior: Grainy, astringent Pith color: off-white Pith thickness: thin Pulp color: white, pink, dark red, maroon Seed color: yellow Small internal cavity Kadota Small size Thick skin, firm pith Exterior: Aroma: Light green, dark green Low intensity Spotted Green, cucumber, grassy Smooth, glossy, waxy Flavor: Open ostiole Low intensity Interior: Honey, green, cucumber Pith: white, off-white Low sweetness and sourness Pith thickness: thick Low juiciness, sliminess and stickiness Pulp color: white Low seed crunchiness Seed color: yellow Minimal internal cavity Panachee Medium size Thick skin, firm exterior, firm pith Exterior: Aroma: Yellow, light green Cucumber Striped Flavor: Smooth High intensity Open ostiole Fruity, stone fruit, berry, citrus, honey Continued. S426 Journal of Food Science Vol. 77, Nr. 12, 2012

119 Sensory attributes of fig cultivars... Table 4 Continued. Fig cultivar Appearance descriptors Sensory descriptors Interior: Sour, low bitterness Pith color: white, yellow Crunchy seeds Pith thickness: thin Pulp color: white, dark red Seed color: yellow Minimal internal cavity Sequoia Medium size Thick skin, firm exterior, firm pith Exterior: Aroma: Light green, dark green Cucumber, grassy Spotted Flavor: Smooth Fruity, cucumber Closed ostiole Low juiciness, sliminess and stickiness Interior: Pith color: white Pith thickness: thick Pulp color: white, pink Seed color: yellow Minimal internal cavity Sierra Medium size Firm exterior Exterior: Aroma: Light green, dark green Low intensity Striped, spotted Green, cucumber, grassy Hairy Flavor: Closed ostiole Low intensity Interior: Honey, cucumber Pith color: white Bitter, low sweetness Pith thickness: moderate Grainy, astringent Pulp color: white, yellow, pink Moderate seed adhesiveness Seed color: yellow, brown Tingling Minimal internal cavity Tina Medium size Soft exterior, soft pith Exterior: Aroma: Light green, dark green High intensity Striped Fig, cucumber, grassy Smooth, waxy Flavor: Open ostiole Honey, cucumber Interior: Low sourness and bitterness Pith color: white, off-white Slimy, sticky Pith thickness: thin Low tingling Pulp color: white, yellow, pink Seed color: yellow Minimal internal cavity UCR Large size Thin skin, soft exterior, soft pith Exterior: Aroma: Yellow, light green, dark Fig, green, cucumber, grassy green Flavor: Striped, spotted Berry, honey Smooth, powdery Bitter Closed ostiole Grainy, juicy Interior: Low astringency Pith color: white Low sliminess and stickiness Pith thickness: moderate Pulp color: white, pink, red Seed color: yellow Medium to large internal cavity Violette de Bordeaux Small to medium size Thin skin, soft pith Exterior: Aroma: Violet, purple Cucumber Spotted Flavor: Smooth, powdery Fruity, stone fruit, berry, citrus honey Open ostiole Sweet, sour Interior: Juicy, slimy, sticky Pith color: off-white, pink Astringent Pith thickness: thin Low tingling Pulp color: white, dark red Seed color: yellow Minimal internal cavity Continued. S: Sensory & Food Quality Vol. 77, Nr. 12, 2012 Journal of Food Science S427

120 Sensory attributes of fig cultivars... Table 4 Continued. Fig cultivar Appearance descriptors Sensory descriptors Zidi Medium size Aroma: Exterior: Fruity, cucumber Violet, purple Flavor: Spotted, striped Fruity, melon, stone fruit, berry, honey Hairy, furry Sweet, low bitterness Open ostiole Juicy, sticky, slimy Interior: Crunchy seeds Pith color: off-white, yellow Low tingling Pith thickness: thin Pulp color: white, red, dark red Seed color: yellow, brown Minimal internal cavity S: Sensory & Food Quality right quadrant of Figure 2, except for BT1A, which is located in the far right of the quadrant. BT1A was rated higher in firmness, graininess, sourness, astringency, seed adhesiveness, and tingling than the other Brown Turkey figs. Black Mission figs are located mostly in the bottom half of the biplot, however, one sample, BM8A, is located in the top half of the quadrant, being rated higher in seed adhesiveness, astringency, and bitterness than the other Black Mission figs, and lower in sweetness. The Sierra figs are mainly clustered in the center of the biplot, being moderately rated in most of the sensory attributes. One Sierra sample, SI2A, located to the right of the cluster, was rated high in tingling and astringency, while fig SI31C, located to the left of the cluster, was rated high for skin smoothness and low for graininess. The Calimyrna figs are located in the center of the top quadrant, whereas fig CAL3B, located to the right of the cluster, had higher ratings for seed adhesiveness, astringency, and tingling than the other Calimyrna samples. The Kadota figs are mostly clustered slightly to the left of the central axis. In contrast, 2 Kadota samples, KAD4A and KAD5B, are positioned to the right of the cluster, and are rated high in those attributes found low in the other Kadota samples, in particular, astringency, tingling, graininess, intensity of fig aroma, and seed adhesiveness. All of the outliers, except SI31C, were located to the right of their cultivar groups, in the right-hand quadrant of the biplot, associated with increased ratings of astringency, seed adhesiveness, graininess, and tingling (Figure 2). These outliers can be explained by the differences in initial fruit instrumental quality measurements (Table 3). All the outliers, except SI31C, were assessed in the 1st 3 d of analysis (order of assessment 1 to 9, see Table 1 for details), indicating that they were harvested early in the growing season. Of those outliers harvested early in the season (BT1A, SI2A, CAL3B, KAD4A, and BM8A), the initial instrumental quality parameters indicated that they were harvested earlier, commonly with a lower maturity rating and SSC, and a higher compression force compared to the other sources harvested as the season progressed (Table 3). While the outlier harvested later in the season, SI31C, was the last of the Sierra figs to be assessed, and had, as expected, the lowest firmness rating of the Sierra figs (Table 3), exhibiting characteristics of an overripe fig. It should be noted that Sierra is a new fig cultivar with young trees just coming into production and growers have yet to understand the optimal harvest maturity for this cultivar. All fresh figs were harvested between maturity stages commercial mature and tree ripe, with a number of figs at the overripe stage toward the end of the study. There was a moderately positive correlation between the maturity stage and SSC (r = 0.62, P < 0.05) and compression force (r = 0.68, P < 0.05), indicating, as expected, that the SSC and compression force increased as figs matured (Crisosto and others 2010). In contrast, there was a moderate negative correlation between maturity and skin thickness (r = 0.53, P < 0.05), where skin thickness decreased as maturity increased. There were a number of sensory attributes associated with maturity levels. As maturity increased, the following sensory aroma attributes decreased: green (r = 0.67, P < 0.05), cucumber (r = 0.61, P < 0.05), grassy (r = 0.54, P < 0.05), earthy (r = 0.61, P < 0.05), woody, (r = 0.72, P < 0.05), and latex (r = 0.40, P < 0.05) (only 20% of panelists rated the latex attribute). Whereas stone fruit (r = 0.45, P < 0.05), berry (r = 0.52, P < 0.05), and honey (r = 0.49, P < 0.05) aromas increased with maturity, maturity was also related to some of the in-mouth texture attributes, including juiciness (r = 0.62, P < 0.05), stickiness (r = 0.68, P < 0.05), and sliminess (r = 0.78, P < 0.05). In contrast, graininess, tingling, and seed adhesiveness were found to decrease with increasing maturity. Tingling was moderately associated with skin hairiness (r = 0.47, P < 0.05), possibly due to the allergic reactions associated with plants in the Ficus family. As expected, maturity was positively related to sweetness (r = 0.63, P < 0.05), and negatively related to bitterness (r = 0.62, P < 0.05) and astringency (r = 0.74, P < 0.05). These data demonstrate that the level of maturity affects the sensory properties of fresh figs, irrespective of fig cultivar. Fresh figs harvested earlier in the season (less mature) are more likely to exhibit unripe sensory profiles, such as green characters, earthy, woody, latex, bitter, astringent, grainy, tingling, and adhesive seeds. Ripe fresh figs are characterized more by fruity characters, honey, juicy, sticky, slimy, and sweet. To our knowledge, no previous studies have indicated that green characters and latex flavor in fresh figs are associated with lower levels of ripeness. It may be that the skin, stem, and/or pith of the fresh figs contributed to the green characters and latex flavor. Skin thickness was somewhat positively associated with green flavor (r = 0.38, P < 0.05) and latex flavor (r = 0.42, P < 0.05). As skin thickness was found to decrease as figs ripen, this suggests that the green characters and latex flavor in ripe figs may decrease not as a result of changing sensory profiles, but of changing amounts of components in the fresh figs containing these characters. Therefore, a ripe fresh fig with a thicker skin, such as Kadota or Sequoia, will retain some of the green characters (Table 4). A detailed study is being prepared to further answer the questions regarding harvest maturity and the resulting sensory attributes. S428 Journal of Food Science Vol. 77, Nr. 12, 2012

121 Sensory attributes of fig cultivars... Conclusions Fig cultivars vary in appearance and flavor sensory profiles, which, in general, were not affected by source. Specific sensory descriptor terminology for fig appearance and flavor profiles are being proposed based on the results of the trained sensory panel. We believe this terminology will assist with communication between marketers and consumers, which can increase fresh fig consumption. Maturity levels had a large effect on the fresh fig sensory profiles. Fresh figs harvested at lower maturity are perceived as having green and latex flavors, and high astringency, seed adhesiveness, graininess, and tingling, while mature-ripe fresh figs are associated with fruity flavors and higher sweetness, juiciness, and sliminess. The results of this study highlight the importance of selecting cultivars and genotypes with strong flavors, not dependent on maturity stage, and which will remain firm as they mature. Based on this sensory descriptors study and our previous instore consumer test (Crisosto and others 2010), it would be of interest for the fresh fig industry to follow up these sensory studies with a large-scale consumer test, to identify the flavor characteristics driving consumer liking of fresh fig cultivars for promising, new fresh fig cultivars. Acknowledgments We thank the California Fresh Fig Assn. and Mr. Maxwell Norton, for their continued support. We acknowledge Mr. Kevin Scott for helping with data analysis, Mr. Andrew Lampinen for texture analysis, and Dr. John Preece and Mr. Howard Garrison at the USDA Natl. Clonal Germplasm Repository, in Davis, CA for providing samples and their expertise. References Caliskan O, Polat AA Fruit characteristics of fig cultivars and genotypes grown in Turkey. Sci Hortic 115: Campo E, Ballester J, Langlois J, Dacremont C, Valentin D Comparison of conventional descriptive analysis and a citation frequency-based descriptive method for odor profiling: an application to Burgundy Pinot noir wines. Food Qual Pref 21: Crisosto CH, Bremer V, Ferguson L, Crisosto GM Evaluating quality attributes of four fresh fig (Ficus carica L.) cultivars harvested at two maturity stages. HortScience 45: Crisosto CH, Crisosto GM, Echeverria G, Puy J Segregation of peach and nectarine (Prunus persica (L.) Batsch) cultivars according to their organoleptic characteristics. Postharvest Biol Technol 39:10 8. Food and Agriculture Organization. Data archives, Food and agriculture organization of the United Nation (FAO); Available from: 408/DesktopDefault.aspx. Accessed 2011 November 24. Gozlekci S Selection studies on fig (Ficus carica L.) in Antalya Province of Turkey. Afr J Biotechnol 9: IPGRI, CIHEAM. Descriptors of figs. Rome, Italy and Paris, France: International Plant Genetic Resources Institute (IPGRI) and International Centre for Advanced Mediterranean Agronomic Studies (CIHEAM) Available from: org/fileadmin/bioversity/publications/pdfs/907_descriptors_for_fig Ficus_carica_.pdf? cache= Accessed 2011 August 1. Jaeger SR, Axten LG, Paisley AG, Wohlers MW, Marsh KB, Sullivan MB, Harker FR Developing models systems for testing the sensory properties and consumer acceptance of new fruit cultivars: the example of kiwifruit. Food Qual Pref 22: Levaj B, Bunić N, Dragović-Uzelac V, Kovačević DB Gel strength and sensory attributes of fig (Ficus carica) jams and preserves as influenced by ripeness. J Food Sci 75: S McEachern GR. Figs. Texas, USA: AgriLife Extension, Texas A&M System; Available from: Accessed 2011 August 1. Obenauf G, Gerdts M, Leavitt G, Crane J Commercial dry fig production in California. Davis, Calif.: Univ. of California. Oliveira AP, Silva LR, Guedes de Pinho P, Gil-Izquierdo A, Valentão P, Silva BM, Pereira JA, Andrade PB Volatile profiling of Ficus carica varieties by HS-SPME and GC-IT-MS. Food Chem 123: Podgornik M, Vuk I, Vrhovnik I, Mavsar DB A survey and morphological evaluation of fig (Ficus carica L.) genetic resources from Slovenia. Sci Hortic 125: Polat AA, Caliskan O Fruit characteristics of table fig (Ficus carica) cultivars in subtropical climate conditions of the Mediterranean region. NZ J Crop Hortic Sci 36: Stover E, Aradhya M, Crisosto CH, Ferguson L Overview of the California fig industry and new interest in varieties for fresh fruit. In: Proceedings of the California Plant and Soil Conference; 6 7 February; Sacramento, CA. California Chapter, American Society of Agronomy. Swahn J, Ostrom A, Larsson U, Gustafsson IB Sensory and semantic language model for red apples. J Sens Stud 25: U.S. Department of Agriculture: Noncitrus fruits and nuts 2006 summary. National Agricultural Statistics Service, U.S. Department of Agriculture Available from: Accessed 2011 November 29. S: Sensory & Food Quality Vol. 77, Nr. 12, 2012 Journal of Food Science S429

122 Peer Reviewed Title: Early harvest delays berry skin browning of Princess table grapes Journal Issue: California Agriculture, 59(2) Author: Vial, Pablo M. Crisosto, Carlos H. Crisosto, Gayle M. Publication Date: Publication Info: California Agriculture, Division of Agriculture and Natural Resources Communication Services, UC Office of the President Permalink: Keywords: Princess, table grapes, vine, skin browning, flesh browning, internal browning, harvest maturity, cold storage, vineyard management Citation: Vial, Pablo M., Crisosto, Carlos H., & Crisosto, Gayle M.(2005). Early harvest delays berry skin browning of Princess table grapes. California Agriculture, 59(2). Retrieved from: Abstract: Table grapes commonly suffer from tissue browning during harvest, packing, storage and shelf life, resulting in lower prices and reduced access to markets. We evaluated the development of browning symptoms in Princess table grapes. The berries had high skin browning but very low flesh browning incidences. The most skin browning was found in highly mature grapes and appeared after 3 weeks of cold storage. Skin browning was directly related to fruit maturity, but vineyard location had a greater impact on the incidence of skin browning than maturity. In all locations, the skin browning susceptibility of Princess table grapes rapidly increased when the berries reached a titratable acidity of less than or equal to 0.60% and/or a soluble solids concentration greater than or equal to 18.0%. Based on this work, we recommend harvesting Princess at a soluble solids concentration between 16.0% and 18.0%. escholarship provides open access, scholarly publishing services to the University of California and delivers a dynamic research platform to scholars worldwide.

123 RESEARCH ARTICLE Early harvest delays berry skin browning of Princess table grapes Pablo M. Vial Carlos H. Crisosto Gayle M. Crisosto Table grapes commonly suffer from tissue browning during harvest, packing, storage and shelf life, resulting in lower prices and reduced access to markets. We evaluated the development of browning symptoms in Princess table grapes. The berries had high skin browning but very low flesh browning incidences. The most skin browning was found in highly mature grapes and appeared after 3 weeks of cold storage. Skin browning was directly related to fruit maturity, but vineyard location had a greater impact on the incidence of skin browning than maturity. In all locations, the skin browning susceptibility of Princess table grapes rapidly increased when the berries reached a titratable acidity of less than or equal to 0.60% and/or a soluble solids concentration greater than or equal to 18.0%. Based on this work, we recommend harvesting Princess at a soluble solids concentration between 16.0% and 18.0%. Table grapes commonly suffer from variations of tissue browning including stem browning (Crisosto et al. 2001), internal flesh browning (Nelson 1969) and berry (skin and flesh) browning (Crisosto, Badr, et al. 2002; Vial 2003) during harvest, packing, storage and shelf life. Table grapes that suffer from browning disorders normally have a shorter postharvest shelf life and may not be utilized for long-distance markets, which are often the most profitable. In general, table grapes with browning problems garner lower prices than grapes without them. We studied postharvest berry browning in Princess (originally known as Melissa ), a white, seedless, table grape Top and bottom right, fresh-market table grapes are vulnerable to browning of skins and flesh during harvesting, packing, storage and shelf life. Research has shown that this browning can be reduced in Princess table grapes, left, by harvesting when soluble solids range between 16.0% and 18.0%. cultivar developed and released by the USDA s Agricultural Research Service in 1999 (Ramming 1999). Princess resulted from the cross of Crimson Seedless with B in B is a white, seedless selection and is a complex hybrid whose parents include Italia, which is well known to exhibit berry browning symptoms. According to the California Agricultural Statistics Service, in 2003 there were 1,329 acres (325 acres nonbearing) planted in the state. In 2000, we conducted a preliminary postharvest evaluation of Princess and observed various berry browning symptoms similar to those seen in Italia. Berry browning occurred either as an irregular shape and scattering on the surface of the berries restricted to the skin, which we called skin browning, or as partial or total browning restricted to the flesh, which we called flesh browning. These berry browning symptoms were different from those described in a 1992 UC DANR bulletin (Luvisi et al. 1992) as internal browning for Thompson Seedless. The Princess berry browning symptoms are frequently expressed on white table grape cultivars, including Italia, Regal Seedless and others. In the few cases where internal browning has been reported on Thompson Seedless, the browning started parallel to the vascular system in the center of the berry and never developed on the skin as it does in Princess. We investigated if the development of postharvest berry browning problems in Princess was related to cluster maturity parameters such as soluble solids concentration (SSC), titratable acidity (TA), the SSC/TA ratio and juice ph. DRAFT APRIL JUNE Maturity and berry browning To study the relationship between grape maturity and berry browning incidence, Princess table grapes were harvested during the 2001 and 2002 seasons at three soluble solids concentration (SSC) ranges: low (13% to 16%), moderate (16% to 18%) and high (greater than 18%).

124 During the 2001 season, Princess table grapes were harvested at three different maturities at California State University, Fresno, from research plots in a vineyard comprised of 6-year-old, cane-pruned vines. The vines were supported by a wide V trellis system, and vine and row spacings were 7 feet (2.3 meters) and 12 feet (3.7 meters), respectively. A drip irrigation system was installed at planting time in The entire block received a gibberellic acid (1.0 gram per acre) bloom-thinning treatment at 80% bloom. Standard cultural practices of irrigation, pest management and canopy management were applied to the entire block. Twelve healthy vines were selected and labeled for this experiment, with each vine serving as a replicate. Four clusters were harvested from each vine on July 19, July 27 and Aug. 2, 2001, and each cluster was labeled to identify its date of harvest and vine number. During the 2002 season, Princess table grapes were harvested at the same three levels of maturity from three commercial vineyards located in major table grape production regions of the San Joaquin Valley: Parlier (Fresno County), Delano (Tulare County) and Arvin (Kern County). These vineyards were carefully chosen based on their similar age, vigor (moderate to high) and management practices. At all three sites, the spacing was 12 feet (3.7 meters) between rows and 8 feet Princess table grapes were harvested at different maturity levels from three San Joaquin Valley vineyards. Researchers determined that vineyard location and management practices had a greater influence on tissue browning than maturity in this cultivar. Trellis systems, nitrogen fertilization, canopy management and rootstocks could play important roles in controlling excess sunlight and hence browning problems. (2.4 meters) between vines, with drip irrigation; and clusters were tipped and the trunks girdled using a 3/8- inch knife immediately after berry set. Twelve healthy grapevines were randomly selected and labeled at each site for the study (Dokoozlian et al. 2001). In 2002, the Princess grapes were harvested in Parlier on July 23, Aug. 1 and Aug. 16; Delano on July 19, July 30, Aug. 12 and Aug. 27; and Arvin on July 17, July 26 and Aug. 8. In Parlier, 5-year-old Princess vines were grown on their own rootstock in fine sandy loam soil. Six canes, along with four to six two-bud spurs, were retained on each vine at pruning. Gibberellic acid was applied at 1 gram per acre near full bloom to reduce berry set. Vine rows were oriented east-west. In Delano, 8-year-old Princess vines were grown on their own rootstock in clay loam soil. Vines were bilateral-cordon trained and were pruned using a combination of spurs and canes (approximately eight to ten two-bud spurs and six to eight 15- bud canes per vine) on a V trellis system. The vineyard was not treated with gibberellic acid, which is used to reduce berry set and/or increase size. The vines were adjusted to similar crop loads (approximately 20 clusters per vine). Vine rows were oriented east-west. In Arvin, 8-year-old Princess vines were grown on Freedom rootstock in sandy loam soil. Vine rows were oriented north-south. Vines were pruned using six to eight 12-bud canes and trellised to an open gable system. Gibberellic acid was applied twice. First, 1.5 grams per acre was applied at 80% bloom to reduce fruit set, then 20 grams per acre was applied at fruit set (0.2 to 0.3 inches [6 to 8 mm berry size]) to increase berry size. Clusters were tipped and the vines were adjusted to similar crop loads (approximately 35 clusters per vine). During both seasons, four clusters were harvested from each replication (48 clusters per harvest date per location total) on each harvest date. Clusters were harvested in the morning (7 a.m. to 10 a.m.) and labeled with a code that included the harvest date, location, vine number and cluster position. Harvested clusters were placed into plastic boxes with cardboard pads in the bottom to reduce DRAFT 104 CALIFORNIA AGRICULTURE, VOLUME 59, NUMBER 2 Table grapes that suffer from browning disorders normally have a shorter postharvest shelf life and may not be utilized for long-distance markets, which are often the most profitable.

125 Left, clusters of Princess table grapes were harvested between 7 a.m. to 10 a.m., labeled with a code indicating date, location, vine number and cluster position and placed in plastic boxes with cardboard pads. From the vineyards, grapes were taken in an air-conditioned vehicle to the F. Gordon Mitchell Postharvest Center at KREC for visual inspection and analysis, right. abrasion damage and were immediately transported to the F. Gordon Mitchell Postharvest Center at the UC Kearney Research and Extension Center in an air-conditioned vehicle. Quality evaluation at harvest During both seasons, each cluster was visually evaluated for skin browning upon arrival at the postharvest facility. When more than 15% of the berries were discolored (skin browning), clusters were considered visually unacceptable. After the visual quality evaluation, five berries per cluster were carefully removed to determine firmness, soluble solids concentration, titratable acidity and juice ph (Crisosto, Garner, et al. 2002), which are physical and chemical parameters used to assess berry quality. These five berries from each replication were pooled (for a total of 60 berries for each set of 12 experimental vines) and pressed through cheesecloth to extract the juice. Soluble solids concentration was measured with a temperature-compensating refractometer (model ATC-1, Atago Co., Tokyo, Japan). Juice titratable acidity and undiluted ph were measured with an automatic titrator (Radiometer, Copenhagen, Denmark) at a final ph of 8.2 and reported as percentage tartaric acid, which is the predominant organic acid in grapes. At the same time that these measurements were taken, labeled clusters from each harvest date were carefully packed using a plastic pad that slowly released sulfur dioxide (SO 2 ; 7 grams of sodium metabisulfite)(tedmark, South Africa) combined with a perforated, polyethylene box liner (1/4-inch hole, 3-inch center) to reduce water loss and assure Botrytis cinerea control without causing bleaching (Crisosto et al. 1994). Clusters were placed inside a plastic cluster bag and packed in 15.7-by-19.7-inch (40-by-50-centimeter) fiberboard boxes as is done commercially. The slow release, one-phase sulfur-dioxide-generating pad was used in the top of each box, above the cluster plastic bags but inside the box liner. Finally, the boxes were labeled and stored at 32 F (0 C) and 80% relative humidity. DRAFT APRIL JUNE Storage quality evaluation In the 2001 season, clusters from the three harvest dates were removed from cold storage at 12 weeks after harvest for visual browning evaluations. The number of berries in the sample with skin browning covering more than 25% of the berry s surface was recorded. If more than 15% of the berries in the sample had skin browning, the cluster was deemed unacceptable (cull). In the 2002 season, clusters from each harvest date were removed from cold storage 1, 3, 5 and 7 weeks after harvest for visual browning evaluations. At 1, 3 and 5 weeks after harvest, incidences of berry skin browning and flesh browning were visually evaluated. Flesh browning incidence was the percentage of clusters with one or more berries showing symptoms. At about 7 weeks after harvest (7 weeks at 32 F [0 C] plus 2 days at 68 F [20 C]), all berries were removed and weighed, and skin browning incidence was expressed as a percentage of cluster weight. Flesh browning and internal browning were also expressed as a percentage of cluster weight after each berry was cut in half and examined internally. Since harvest dates were different at each location, the interaction between harvest date and location was not studied. In both seasons, harvest date was used independently as a main treatment within each location. Twelve vines were used as replicates and four clusters from each replicate were harvested as experimental units. Data analysis was done by ANOVA, whereas mean comparison was carried out by LSD (P < 0.005) using SAS. Harvest date and maturity In Parlier, there were no significant differences in soluble solids concentration among samples collected on

126 TABLE 1. Soluble solids concentration (SSC), titratable acidity (TA), SSC/TA ratio and juice ph of Princess table grapes Vineyard location and harvest date SSC/TA (2002) SSC TA* ratio ph Parlier % / / / P value < < < < LSD Delano 7/ N.A 7/ / / P value < < < < LSD Arvin 7/ N.A. 7/ / P value < < < < LSD * Titratable acidity (TA) expressed as tartaric acid. LSD 0.05 = least significant difference at the 5% level. TABLE 2. Skin browning and flesh browning after 7 weeks of storage at 32 o F (0 o C) for Princess table grape berries Location and harvest date Skin Flesh (2002) browning* browning % cull clusters Parlier 7/ / / P value < LSD Delano 7/ / / / P value < < LSD Arvin 7/ / / P value < LSD the first two harvest dates (17.2% and 17.7%), while samples collected on the third harvest date had a significantly higher soluble solids concentration (19.7%) than those collected on the first two dates (table 1). Titratable acidity decreased significantly between the first and third harvest dates (from 0.70% to 0.50%). The SSC/TA ratio also increased significantly during this period (from 24.7 to 42.1), but did not change significantly between the first two sampling dates. Similarly, juice ph increased between the first and third harvest dates (from 3.2 to 3.6). In Delano, the soluble solids concentration increased significantly between the first and third harvest dates (from 13.8% to 18.9%)(table 1), as did the SSC/TA ratio (from 14.4 to 48.6). During this period, titratable acidity decreased significantly (from 1.00% to 0.40%). Juice ph increased significantly between the second and third harvest dates (from 3.3 to 3.7). In Arvin, the soluble solids concentration increased significantly between the first and third harvest dates (from 15.6% to 18.4%)(table 1). During this period, titratable acidity decreased significantly (from 0.80% to 0.50%) and the SSC/TA ratio increased significantly DRAFT 106 CALIFORNIA AGRICULTURE, VOLUME 59, NUMBER 2 * Clusters with more than 15% discolored (skin browning) berries were considered visually unacceptable (culls). Percentage of clusters having one or more berries with flesh browning. Berries were cut longitudinally to observe flesh browning. LSD 0.05 = least significant difference at the 5% level. Fig. 1. Relationship between Princess table grape harvest date and (A) soluble solids concentration (SSC) measured at harvest and (B) skin browning measured after 12 weeks of storage at 32ºF (0ºC), 2001 season. (from 19.2 to 37.5). Juice ph increased significantly between the second and third harvest dates (from 3.4 to 3.7). Harvesting late resulted in an increase in soluble solids concentration and a reduction in the titratable acidity, thus the SSC/TA ratio increased in the mature grapes. Harvest maturity and storage time In both seasons, skin browning was related to harvest date and in turn, grape maturity, but was very low at harvest for all three maturities. Flesh browning was so low that it was not commercially important (< 2.0%), and internal browning was not observed at all. Skin browning was related to grape maturity, because after 12 weeks of storage, grapes from the early harvest had significantly less skin browning than grapes harvested later (fig. 1). Significant changes in skin browning Fig. 2. Relationship between harvest date and the development of berry skin browning in Princess table grapes grown in (A) Parlier, (B) Delano and (C) Arvin, during 5 weeks of storage at 32ºF (0ºC), 2002 season.

127 incidence occurred when the soluble solids concentration increased from about 18% to 19% (fig. 1A). In Parlier, skin browning incidence was low after a week of cold storage, varying from 0% to 10.4% depending on the 2002 harvest date (fig. 2). After 3 weeks, skin browning incidence increased from 6.3% to 31.8%, and after 5 weeks reached 8.3% to 45.8%. Similarly, in Delano the incidence of skin browning was low after a week of cold storage, but increased dramatically after 3 weeks for the second harvest (35%) and reached up to 64.6% after 5 weeks for the fourth harvest. In Arvin, skin browning incidence was substantially lower than in Parlier and Delano but followed a similar pattern of development, increasing with time in cold storage. After 7 weeks in cold storage, skin browning expressed as a percentage of cull clusters increased significantly in grapes from all three locations (table 2), increasing from the first to the last harvests from 0% to 58.2% in Parlier, 0.3% to 55.5% in Delano, and 0.1% to 14.4% in Arvin. Soluble solids concentration, SSC/TA ratio and ph were significantly positively correlated with skin browning, while titratable acidity was significantly negatively correlated (data not shown). Furthermore, the flesh browning incidence in grapes at all three locations was low after 3 weeks of cold storage but increased dramatically after 5 weeks (fig. 3). There were no significant differences between harvest dates in a specific location and the duration of cold storage, but there was a clear trend of increasing flesh browning during storage on a specific harvest date. It is important to note that flesh browning during the first 5 weeks of cold storage was measured visually and is therefore subjective. However, even in these evaluations, flesh browning was not a major problem and only reached a maximum of 12.5% among all locations after 5 weeks in cold storage. After 7 weeks in cold storage, flesh browning incidence expressed as a percentage of cull clusters did not in- Fig. 3. Relationship between harvest date and the development of berry flesh browning in Princess table grapes grown in (A) Parlier, (B) Delano and (C) Arvin, during 5 weeks of storage at 32ºF (0ºC), 2002 season. crease significantly during the 2002 harvest period (table 2). Flesh browning incidence was so low (about half a berry per cluster) that it did not have commercial implications and was not related to maturity. In general, Princess table grapes collected from the three locations in California during the 2002 season developed skin browning but not flesh browning during storage. In contrast, our group reported a significantly higher flesh browning incidence in 2001 than 2002 (Crisosto, Badr, et al. 2002). The differences can be explained by the vineyard conditions in the 2002 study, which were carefully chosen for vines with wellbalanced vigor and with healthy and shaded canopies. During the 2001 season, the Princess table grapes were grown in sun-exposed and lowvigor vineyards and subsequently showed more browning problems. Significant development of skin browning occurred during cold storage of Princess table grapes from all three vineyards. Skin browning was observed after 3 weeks in cold storage and it was strongly correlated with maturity. This data agrees with our previous work (Crisosto, Badr, et al. 2002), in which the incidence of skin browning Fig 4. Relationship between Princess table grape berry skin browning and (A) soluble solids concentration (SSC) and (B) titratable acidity (TA), 2002 season. after 3 weeks at 32 F (0 C) was directly associated with increases in fruit harvest maturity. Princess table grapes that were harvested at low soluble solids concentrations exhibited lower skin browning incidence than those harvested at values above about 18%. Skin browning increased sharply when soluble solids concentration was greater than 16.0% to 18.0% (fig. 4A). High values of titratable acidity concur with a low incidence of skin browning, and maturity was strongly correlated with titratable acidity. Grapes harvested at titratable acidity values between 0.60% and 1.00% showed significantly lower skin browning incidence than those harvested at titratable acidity values below 0.60% (fig. 4B). The degradation of organic acids, such as tartaric, citric and malic, occurred along with fruit maturation. SSC/TA ratios below 25 were associated with considerably lower skin browning than those above 25 (table 1). However, vineyard location and management were more important than maturity in tissue browning in Princess table grapes. There are many possible reasons for this. First, the phenolic concentration, type of phenolic compounds and polyphenoloxidase (PPO) activity vary for the same cultivar DRAFT APRIL JUNE

128 Davis Krauter P.M. Vial is Research Associate, C.H. Crisosto is Postharvest Physiologist, and G.M. Crisosto is Staff Research Associate, Department of Plant Sciences, UC Davis, located at the UC Kearney Research and Extension Center, Parlier. We thank Sayed A. Badr, Department of Viticulture and Enology, California State University, Fresno, for his help and advice on this work. Consumers have come to expect unblemished table grapes in the produce section. Postharvest research is helping growers to supply fresh, attractive produce for the marketplace. among locations, and from year to year (Sapis et al. 1983). The browning capacity for a specific cultivar and among locations can also be affected by management practices such as irrigation, fertilization, rootstock, trellis system, pruning systems and canopy management (Sapis et al. 1983; Wissemann and Lee 1981). For example, trellis system, nitrogen fertilization, canopy management and rootstocks could play important roles in controlling excess sunlight and hence browning problems. DRAFT 108 CALIFORNIA AGRICULTURE, VOLUME 59, NUMBER 2 Growing location is important Skin browning incidence at harvest was low but its level depends on the level of physical abuse during harvesting operation. The skin browning incidence reached its highest expression after 3 weeks in cold storage and was strongly related to maturity and vineyard location. The effect of location on skin browning incidence was more important than maturity. Flesh browning and internal browning incidences were not commercially important (< 2.0%). The skin browning susceptibility of Princess table grapes increased when berries were harvested at titratable acidity levels less than 0.60% and/or soluble solids concentrations greater than or equal to 16.0%; grapes harvested at an advanced stage of maturity (> 18.0% SSC) were more susceptible to tissue browning. To maximize storage potential and taste, Princess grapes should be harvested at a soluble solids concentration between 16.0% to 18.0%. Skin browning was highly influenced by vineyard location or management; Princess table grapes grown under the specific management conditions of the plot located in Arvin had two to three times less skin browning incidence than those grown in Delano and Parlier. References Crisosto CH, Badr SA, Garner D, et al Screening for orchard factors that influence Princess table grapes berry browning during cold storage. Research Report to California Table Grape Commission. Fresno, CA. Crisosto CH, Garner D, Crisosto G High carbon dioxide atmospheres affect stored Thompson Seedless table grapes. HortScience 37: Crisosto CH, Smilanick JL, Dokoozlian NK Table grapes suffer water loss, stem browning during cooling delays. Cal Ag 55(1): Crisosto CH, Smilanick JL, Dokoozlian N, Luvisi DA Maintaining table grapes post-harvest quality for long distance markets. In: International Symposium on Table Grape Production, Anaheim, CA. Am Soc Enol Viticul. p Dokoozlian N, Peacock B, Ebisuda N, Macias A Influence of vine training and pruning system on Princess table grapes, California Table Grape Commission Viticulture Research Report, Vol XXX. 5 p. Luvisi DA, Shorey H, Smilanick J, et al Sulfur Dioxide Fumigation of Table Grapes. UC DANR Bull p. Nelson KE Controlled atmosphere storage of table grapes. Proc Natl CA Res Conf, Michigan State Univ. Hort Rpt 9: Ramming D New table grape varieties. Dinuba Table Grape Seminar, Feb. 24, California Table Grape Commission, Fresno, CA. Sapis JC, Macheix JJ, Cordonnier RE The browning capacity of grapes. II. Browning potential and polyphenol oxidase activities in different mature grape varieties. Am J Enol Vitic 34: Vial P Relationship between fruit maturity and berry browning potential in several table grape cultivars. M.S. thesis, Department of Pomology, UC Davis. 68 p. Wissemann KW, Lee CY Characterization of polyphenoloxidase from Ravat 51 and Niagara grapes. J Food Sci 46:506 8.

129 Postharvest Biology and Technology 25 (2002) Developing a quantitative method to evaluate peach (Prunus persica) flesh mealiness Carlos H. Crisosto a, *, John M. Labavitch b a Department of Pomology, Kearney Agricultural Center, Uni ersity of California at Da is, 9240 South Ri erbend A enue, Parlier, CA 93648, USA b Department of Pomology, Uni ersity of California at Da is, One Shields A enue, Da is, CA 95616, USA Received 4 May 2001; accepted 17 October 2001 Abstract A quantitative method to measure peach flesh mealiness based on free juice was developed. Free juice was measured by subjecting fruit tissue to a pressing force of 667 N for 1 min before centrifugation. After this extraction, the juice was collected and centrifuged at 6000g for 10 min. The supernatant was weighed and used to represent the free juice from the total initial tissue. The percentage of free juice was more sensitive and it had a higher correlation to mealiness measured visually and by a trained panel than the extractable juice method. Our work also demonstrated that trained judges were able to perceive mealiness earlier than visual mealiness symptoms were expressed. This free juice measurement provides an objective and accurate method to evaluate mealiness potential of stone fruit breeding lines and fruit biochemical and/or physiological changes during postharvest storage Elsevier Science B.V. All rights reserved. Keywords: Chilling injury; Internal breakdown; Free juice; Extractable juice; Apparent juice; Expressible juice; Sensory evaluation; Stone fruit 1. Introduction Some apricot, nectarine, peach, and plum cultivars develop chilling injury (CI) symptoms during low temperature storage and/or shipping. CI is a physiological disorder whose symptoms are expressed during ripening under non-stressed conditions (Ben Arie and Lavee, 1971). Chilled damaged fruit develop flesh mealiness (FM), a dry * Corresponding author. Tel.: ; fax: address: carlos@uckac.edu (C.H. Crisosto). or mealy flesh texture with a grainy sand-like texture, flesh browning (FB), off flavors, and fail to ripen properly during prolonged cold storage and/or after ripening at room temperature. FM is also named woolliness, especially by researchers from South Africa (Von Mollendorff et al., 1992a,b). These CI symptoms are also known as internal breakdown (IB) (Luza et al., 1992; Smith, 1934). Because these symptoms reduce consumer acceptance, their onset limits a fruit s potential postharvest market life (Crisosto et al., 1999). Peach susceptibility to CI varies according to genetic background (Anderson, 1979; Crisosto et /02/$ - see front matter 2002 Elsevier Science B.V. All rights reserved. PII: S (01)

130 152 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) al., 1999), maturity (Von Mollendorff, 1987), and orchard factors (Crisosto et al., 1997). In our previous work (Crisosto et al., 1999), most of the CI-susceptible cultivars grown in California developed FM prior to FB symptoms; only rarely did FM and FB develop at the same time. Several cultivars developed FM without FB symptoms. Thus, development of postharvest handling treatments or breeding programs to reduce CI problems should be focused on FM rather than FB. A rapid evaluation of FM symptoms is usually a subjective visual determination. The measurement of the most obvious characteristic of the FM problem, decreased fruit juice, can be used as a more objective assessment of FM. Approaches for this analysis have been previously described (Lill and van der Mespel, 1988; Von Mollendorff et al., 1992a,b). Nevertheless, there is still the need for a rapid, sensitive and quantitative method to measure the onset of FM. The roles of polygalacturonase (PG) and pectinesterase (PE) activities on pectin degradation during FM development have been well documented (Ben Arie and Lavee, 1971; Ben Arie and Sonego, 1980; Von Mollendorff and De Villiers, 1988b; Dawson et al., 1992; Lurie et al., 1994; Zhou et al., 1999). Pectins from sound peaches and nectarines are normally poor gel formers due to the presence of acetyl groups, high methoxyl levels, and relatively low molecular masses. The fact that cold storage below 8 C reduced fruit production and/or activity of PG but did not significantly affect PE activity (Ben Arie and Lavee, 1971; Ben Arie and Sonego, 1980) led to the assumption that the altered balance of PG and PE activities at low temperatures or during ripening lead to incomplete or unbalanced cell wall pectin degradation. Pectins with longer polymer backbones and reduced methyl ester content have a high gel formation capacity. Available water and solutes (primary sugars) from surrounding tissues are trapped as apoplast conditions trigger pectin gel formation, inducing a dry mealy fruit texture (Ben Arie and Lavee, 1971; Buescher and Furmanki, 1978; Zhou et al., 1999). The water in fruit can be classified as free or bound water. Free water is the water that gives ripe fruit a juicy character; and bound water is that which is associated with polymer or solute hydration spheres, and pectin gels in mealy fruit. Thus, the total water content in both mealy and juicy fruit can be similar (Von Mollendorff and De Villiers, 1988a), but fruit texture can differ substantially if the proportions of free and bound water are altered during the gel formation. Based on this concept, quantitative methods for mealiness assessment, using the supernatant after centrifuging of homogenized fruit tissue, were developed by Lill and van der Mespel (1988) and Von Mollendorff et al. (1992a,b). In the method of Von Mollendorff et al. (1992a,b), fruit tissue was homogenized for 60 s in a Waring blender before centrifugation for 10 min at 1000g (supernatant liquid was defined as extractable juice ). In the apparent juice method (Lill and van der Mespel, 1988), fruit tissue was gently disrupted by placing it in a syringe and extruding it by forcing it through a luer hub. After disruption, collected tissue was centrifuged at g for 10 min. In most cases it was difficult to force tissue through the luer hub and also to separate the gel phase from the liquid phase. We and others have observed, in some cases, that gel former pectins can be created during juice disruption and/or centrifugation, producing additional undesired gel formation. In both of the methods, the volume of the supernatant is recorded as a percentage of the initial tissue mass. This extractable juice quantitative method has been used to report FM by several researchers; however; some contradictions are found in the literature. In some work with peaches and nectarines, the extractable juice percentage did not correlate well with visual mealiness symptoms. For example, in Flavortop nectarine, during 4 weeks storage at 0.5 C, the extractable juice remained constant at approximately 50 and 30% for mature and ripe fruits, respectively. During this same storage period, visual mealiness of ripe fruit increased from none detectable to 80% (Von Mollendorff et al., 1992a). In Israel, preharvest application of gibberellic acid significantly delayed visual mealiness development in nectarines without affecting their apparent or expressible juice contents (Zilkah et al., 1997). In these cases, woolliness was evaluated visually and by hand squeezing the fruit to collect juice. These

131 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) reports suggest that the extractable juice method may not be a reliable indicator of FM. The research described in this paper demonstrates that free juice assessed after a pressing extraction prior to centrifugation is highly related to mealiness development and can be used as a quantitative indicator of FM. 2. Materials and methods Rich Lady, Elegant Lady, Zee Lady and O Henry peach and Summer Grand nectarine cultivars commercially grown in California and Chile were tested for CI symptoms that developed after fruits were stored at 0 or 5 C and then ripened. Fruit samples (100 fruits) of each cultivar were collected at the California well-mature stage according to the California Tree Fruit Agreement ground-color chips from each of the three trees (replications) growing at the Kearney Agricultural Fig. 1. Scheme of the pressing apparatus. Center (KAC), or from the commercial orchards near the KAC. Fruits were forced-air cooled to 0 2 C within 12 h of harvest and then stored at either 0 or 5 C (with 90% relative humidity) for up to 3 weeks. Chilean tree fruits were collected upon arrival at Long Beach, CA after shipment at 0 C lasting approximately 14 days. Three replications of ten fruit samples each were removed weekly from storage (0 or 5 C), ripened at 20 C until firmness reached between 10 and 18 N (measured with a UC firmness tester, 7.9-mm tip), then evaluated for CI symptoms. The ripening period varied from 3 to 7 days depending on the fruit s initial firmness and its rate of softening. Flesh mealiness and flesh browning symptoms were visually determined immediately after cutting the fruit in half according to Crisosto et al. (1999). Fruits with little or no juice after hand squeezing were considered mealy Free juice measurements To evaluate the effects of CI development on the amount of free juice released by this method, O Henry and Elegant Lady peaches from California and Chile with different degrees of mealiness (visually assessed) were used. Peaches were ripened at 20 C until soft (firmness N) measured with a UC firmness tester, 7.9-mm tip. Forty grams of fruit tissue from each fruit/replication was used. To obtain the tissue, a wedge was cut longitudinally from the fruit with an approximate width of 25 mm at the back. This wedge was then cut into three sections and wrapped with four layers of cheesecloth (20 20 cm) before being compressed with a self-made press. The press (Fig. 1) consisted of a wooden test frame supporting a shaft. On the top of this central shaft was a fixed weight (load) and on the bottom was a pressing head. The pressing head (frustum of a right circular cone) measured 25 mm in diameter on the bottom (fruit contact area), 70 mm in diameter at the top, and had a height of 45 mm. The fruit tissue sample was placed in a sample holder inside a perforated steel basket. The perforations in the basket allowed the expressed juice to pass through a funnel into a collection beaker at the base of the sample holder. The sample

132 154 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) holder was raised hydraulically onto the pressing head, raising it until the load was released from the resting position. The use of this self-made press assured the application of a constant load on the tissue over time which is not possible using our Instron model. After pressing the tissue for 1 min, the sample holder was lowered. Then, the tissue was repositioned in the holder and pressed again. Fruit tissues were exposed to different pressing forces (from 222 to 890 N), for different durations (from 1 to 10 min), and different repetitions of pressing (from 1 to 6). After this, the extracted juice was centrifuged at 6000g for 10 min and the supernatant weight was used to determine the percentage free juice based on the initial fruit tissue sample weight. The percentage extracted juice was measured according to Von Mollendorff and De Villiers (1988a). In this method, fruit tissue (1.5g) was homogenized with a blender for 1 min and allowed to stand for 15 min prior to centrifugation at 6000g for 10 min. The supernatant weight (extracted juice) was used to determine the percentage extractable juice based on the initial fruit tissue sample weight Free juice method sensiti ity To compare the relationships between free juice, extractable juice, and mealiness incidence, Flavorcrest and Elegant Lady peaches with different degrees of mealy fruits were used. Sixty mealy and 60 juicy peaches were chosen from the different storage temperatures. These mealy and juicy fruits were cut into 15 g portions. Then, by combining varying weights of the mealy and juicy fruit portions, different fruit mealiness percentages were created. Free juice and extractable juice measurements were determined in the range of different mealiness percentages artificially created as previously described Sensory e aluation To evaluate the sensory perception of mealiness (the feeling of graininess, like sand) a trained panel consisting of six screened judges was selected. Judges were screened and trained for their acuity in perceiving mealiness using a triangle test. In the triangle test, three samples of sliced mealy or sound peaches were randomly presented simultaneously, with two the same and one different. The respondent was asked to choose the odd sample. Six combinations were presented in random order to each potential judge. Using a onetailed binomial test with a 0.05% significance level, only judges with five or six correct responses were used for the mealiness detection (O Mahony, 1986). Each O Henry peach sample consisted of a slice cut from the bottom half of the peach with the skin removed. Each judge was instructed to cleanse his or her mouth with distilled water, chew the peach sample and mark a yes or a no on the scorecard for juiciness, mealiness and off flavor, then cleanse again before proceeding to the next sample. The parameters were defined as: juiciness=the feeling of juice freely flowing in the mouth; mealiness=the feeling of graininess, like sand; and off flavor =not having the usual or standard peach flavor Statistical analysis Data were subjected to ANOVA and LSD means separation at the 5% level on the free juice pressing force data. Correlation analysis was used for the analysis of data from the free juice sensitivity test and sensory evaluation data. The sensory evaluation data were also analyzed using population distribution comparison. The SAS statistical software (SAS Institute, Cary, NC) was used for these analyses. 3. Results and discussion 3.1. Free juice measurements The percentage of free juice varied according to the pressing force that the fruit tissues were subjected to. In the three cultivars tested, free juice release increased as the pressing force increased, but it reached a plateau above 445 N (Fig. 2). There were no differences in free juice released from fruit tissues subjected to pressing forces between 445 and 890 N within each of the three

133 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) Fig. 2. Free juice released as a function of pressing force from peach and nectarine fruit. Different letters within each cultivar/origin represent significant differences at P 0.05 by LSD means separation test. cultivars tested. The percentage of free juice was highest for the California O Henry peaches, intermediate for the California Elegant Lady peaches and Chilean Summer Grand nectarines, and the lowest for the Chilean Elegant Lady peaches. Both of the Chilean fruit samples were tested after shipment, approximately 21 days after harvest. The different percentages of free juice between cultivars indicated the current amount of chilling damage for each cultivar at that sampling date. Our previous laboratory evaluations of free juice for fresh ripe harvested peaches prior to storage yielded 68, 52, 59, and 57% for O Henry, Rich Lady, Elegant Lady, and Zee Lady, respectively. The differences in the percentages of free juice released between these cultivars can be explained by the different cultivars susceptibility to chilling injury, temperature conditions prior to collection, maturity at harvest, orchard factors during the growing season, storage period, and postharvest handling. The expression of CI symptoms such as lack of juice and FB are dependent on the genetic makeup and cold storage period (Crisosto et al., 1999). Probably CI-susceptible cultivars have differences in pectin structures and/ or PG/PE balance that react in different ways to storage temperatures. The California O Henry peach had the highest percentage free juice because that fruit had a shorter exposure to low temperatures than the other three fruit samples. The percentage of free juice was the lowest in the Chilean Elegant Lady because this cultivar was in cold storage for a longer period of time. Summer Grand nectarine has a longer market life than Elegant Lady peach when stored at 0 or 5 C (Crisosto et al., 1999). This reported difference in potential market life between these two cultivars is consistent with the observation that the Chilean Elegant Lady peach had a lower percentage free juice than the Chilean Summer Grand nectarine even though they arrived on the same ship approximately 21 days after harvest. Even though fruits with different levels of chilling injury were used for this test, it is clear that the maximum amount of free juice will be released from fruit subjected to a pressing force between 445 and 700 N Free juice method sensiti ity There was a highly significant (P 0.001) correlation between the range of artificially created

134 156 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) percentages of FM fruit juice content and the measurements of free juice or extractable juice for Flavorcrest and Elegant Lady. It is important to point out that our sensitivity test is based on the assumption that mixing different proportions of mealy and juicy fruits will give the same results as fruit with different intensities of mealiness. Free juice and FM relationship (R 2 =0.91) had a higher correlation than the relationship between the extractable juice and FM (R 2 =0.39%). Thus, changes in extractable juice only account for 39% of the variability in FM rather than 91%, as in the case of free juice. During our sensitivity test and/ or cold storage time course, free juice values (low gel formation) were always higher than extractable juice values in the same fruit measurements. This suggests that more gel former pectins can be created during juice disruption during blending and/or centrifugation steps in the extractable juice method than in the free juice Fig. 4. Distribution of observed frequencies for O Henry peaches with different free juice percentages in the (A) mealy and (B) juicy categories according to our sensory trained panel. method. Thus, this additional undesired gel formation may be artificially increasing the amount of gel former pectins in the extractable juice method. A higher slope was consistently (P 0.001) determined for the free juice and FM relationship than for the extractable juice and FM relationship (Fig. 3). This indicates that small changes in FM are more easily detected using the free juice method than the extractable juice method Sensory e aluation Fig. 3. Correlations between percentage mealiness artificially created in Elegant Lady and Flavorcrest peaches and (A) free juice and (B) extractable juice percentages. Judges segregated O Henry peach samples into mealy and juicy categories. The free juice distribution in the mealy fruit population ranged from 2 to 32%, with a mean of 16% and a standard deviation of 9.8% (Fig. 4A). The percentage of free juice in the juicy fruit population ranged from 20 to 62% with a mean of 40% and a standard deviation of 8.4% (Fig. 4B). The two free juice populations followed a normal distribution, but

135 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) they overlapped in the 20 32% range. This region of overlap included less than 10% of the juicy fruit population and less than 5% of the mealy fruit population. The extractable juice content in the mealy O Henry fruit population ranged from 32 to 60%, with a mean of 45% and a standard deviation of 8.1% (Fig. 5A). The extractable juice content in the juicy O Henry fruit population ranged from 39 to 65% with a mean of 54% and a standard deviation of 6.7% (Fig. 5B). The two extractable juice populations followed a normal distribution, but they overlapped in the 39 59% range. This range included less than 99% of the juicy fruit population and less than 75% of the mealy fruit population. In our peach storage test, different parameters indicated mealiness development during the 3 weeks cold storage. California O Henry peaches developed visual FM symptoms after 2 weeks storage at 5 C (Fig. 6A). However, trained judges were able to detect off-flavors or mealy Fig. 6. O Henry peach flesh mealiness development during storage at 5 C measured using different methods. Vertical bars represent SE. Fig. 5. Distribution of observed frequencies for O Henry peaches with different extractable juice percentages in the (A) mealy and (B) juicy categories according to our sensory trained panel. texture 1 week before the visual mealiness symptom was apparent. Free juice content declined from 58% at harvest time to 28% after 2 weeks storage. According to our sensory evaluation (Fig. 6B), O Henry peaches with 32% free juice were almost always mealy. Thus, the percentage of free juice indicated that mealiness was likely to develop during ripening if fruits were subjected to 1 2 weeks of storage at 5 C. On the other hand, percentage extractable juice did not change during the first 2 weeks of storage; it remained at approximately 46%, the same as at harvest. By the third week of storage, extractable juice decreased to approximately 35%. In all of the tests, our trained judges were able to perceive mealiness earlier than the visual symptoms were expressed. In some cases our trained judges were able to detect off flavor even before the judges perceived mealy texture. This confirms that sensory evaluation should be added to a visual evaluation of mealiness to confirm chilling injury.

136 158 C.H. Crisosto, J.M. Laba itch / Posthar est Biology and Technology 25 (2002) Our empirical comparison study demonstrates that a measurement of free juice content provides a quantitative, accurate and sensitive method to evaluate the onset of peach mealiness. The use of this quantitative method will likely make more meaningful comparisons of work done in different locations and years with fruit of a given cultivar. This method can be used for fundamental investigations of the onset of fruit mealiness, evaluation of breeding materials, and prediction of market life for different commercial cultivars. References Anderson, R.E., The influence of storage temperatures and warming during storage on peach and nectarine fruit quality. J. Am. Soc. Hortic. Sci. 104, Ben Arie, R., Lavee, S., Pectic changes occurring in Elberta peaches suffering from woolly breakdown. Phytochemistry 10, Ben Arie, R., Sonego, L., Pectolytic enzyme activity involved in woolly breakdown of stored peaches. Phytochemistry 19, Buescher, R.W., Furmanki, R.J., Role of pectinesterase and polygalacturonase in the formation of wooliness in peaches. J. Food Sci. 43, Crisosto, C.H., Johnson, R.S., DeJong, T., Day, K.R., Orchard factors affecting postharvest stone fruit quality. HortScience 32, Crisosto, C.H., Mitchell, F.G., Ju, Z., Susceptibility to chilling injury of peach, nectarine, and plum cultivars grown in California. HortScience 34 (6), Dawson, D., Melton, L., Watkins, C., Cell wall changes in nectarine (Prunus persica). Plant Physiol. 100, Lill, R.E., van der Mespel, G.J., A method for measuring the juice content of mealy nectarines. Sci. Hortic. 36, Lurie, S., Levin, A., Greve, C.L., Labavitch, J.M., Pectic polymers from normally and abnormally ripening nectarines. Phytochemistry 36, Luza, J.G., van Gorsel, R., Polito, V.S., Kader, A.A., Chilling injury in peaches: a cytochemical and ultrastructural cell wall study. J. Am. Soc. Hortic. Sci. 117, O Mahony, M., Sensory evaluation of food. Marcel Dekker, New York. Smith, W.H., Cold storage of Elberta peaches. Ice Cold Storage 37, Von Mollendorff, L.J., Woolliness in peach and nectarine: a review 1. Maturity and external factors. Hortic. Sci. 5, 1 3. Von Mollendorff, L.J., De Villiers, O.T., 1988a. Physiological changes associated with the development of woolliness in Peregrine peaches during low temperature storage. J. Hortic. Sci. 63, Von Mollendorff, L.J., De Villiers, O.T., 1988b. Role of pectolytic enzymes in the development of woolliness in peaches. J. Hortic. Sci. 63, Von Mollendorff, L.J., Jacobs, G., De Villiers, O.T., 1992a. Cold storage influences internal characteristics of nectarines during ripening. HortScience 27, Von Mollendorff, L.J., Jacobs, G., De Villiers, O.T., 1992b. The effects of storage temperature and fruit size on firmness, extractable juice, woolliness and browning in two nectarine cultivars. J. Hortic. Sci. 67, Zhou, H.W., Sonego, L., Ben-Arie, R., Lurie, S., Análisis of cell wall components in juice of Flavortop nectarine during normal ripening and wooliness development. J. Am. Hortic. Sci. 124 (4), Zilkah, S., Lurie, S., Lapsker, Z., Zuthi, Y., David, I., Yesselson, Y., Antman, S., Ben Arie, R., The ripening and storage quality of nectarine fruits in response to preharvest application of gibberellic acid. J. Hortic. Sci. 72,

137 HORTSCIENCE 45(4): Evaluating Quality Attributes of Four Fresh Fig (Ficus carica L.) Cultivars Harvested at Two Maturity Stages Carlos H. Crisosto 1, Vanessa Bremer, Louise Ferguson, and Gayle M. Crisosto University of California, Davis, Department of Plant Sciences, Mail Stop 2, One Shields Avenue, Davis, CA Additional index words. in-store consumer test, DPPH analysis, demographics, cultivar genotype selection, firmness Abstract. The effect of two fruit maturity stages on the quality attributes of four fresh fig cultivars was examined, including consumer acceptance and antioxidant capacity. Fig quality attributes such as weight, soluble solids concentration (SSC), titratable acidity (TA), SSC:TA, firmness, antioxidant capacity, and consumer acceptance varied by cultivar. Fig cultivars harvested at the advanced maturity stage ( tree ripe ) had lower TA and firmness but higher weight, SSC, and SSC:TA than figs harvested at commercial maturity. Fig maturity did not affect antioxidant capacity, but tree ripe figs had higher consumer acceptance than commercial maturity figs. SSC was more highly correlated with consumer acceptance than TA or SSC:TA, but other factors may also be important in controlling this relationship. Cultivars with high SSC and firmness, at a maturity stage high enough to tolerate harvesting and postharvest handling, should be selected to develop the fresh fig industry. Because fig firmness is a concern, changes to packaging should be evaluated to protect the flavor of advanced maturity figs during postharvest handling. California fresh fig production has increased recently. As a result, there is now a market for cultivars with favorable fresh fig quality parameters and high consumer acceptance. Recent research demonstrated fig genotype and maturity stage influence fruit quality. Figs (Ficus carica L.) are a nutritious fruit rich in fiber, potassium, calcium, and iron (Chessa, 1997) with higher levels than other common fruits such as bananas, grapes, oranges, strawberries, and apples (Chessa, 1997; Michailides, 2003). Figs are free of sodium, fat-free, and, like other fruits, cholesterolfree. Additionally, figs are an important source of vitamins, amino acids, and antioxidants (Solomon et al., 2006). Compounds with antioxidant properties such as vitamin C, tocopherols, carotenoids, and phenolics can alter the metabolic activation and detoxification/ disposition of carcinogens, affect processes that modify the development of tumor cells (Kader, 2001), and avoid neurochemical and behavioral changes related with aging (Shukitt-Hale et al., 2007). In addition, fruits and vegetables rich in phenolics have been shown to decrease cardio- and cerebrovascular diseases and cancer death rates (Hertog et al., 1997). Fig varieties with dark skin contain higher levels of polyphenols, anthocyanins, and flavonoids accompanied by higher antioxidant activity compared with fig varieties with lighter skin (Solomon et al., 2006). Received for publication 9 Dec Accepted for publication 25 Feb To whom reprint requests should be addressed; chcrisosto@ucdavis.edu. Fig trees (F. carica L.) are among the earliest cultivated fruit trees in the world (Solomon et al., 2006). Although its origin is not entirely known, F. carica is thought to have originated in western Asia and from there slowly spread through the Mediterranean region (Stover et al., 2007b). Figs were brought to America in 1520 by the Spaniards, and in 1769, they were introduced to California from Mexico. Figs are harvested worldwide on 419,000 ha with an annual production of over 1 million tons. The United States ranks sixth in the world s production, representing 4.6% of the total production (Food and Agriculture Organization, 2007). There are 5100 ha of figs in California, mainly in the San Joaquin Valley with yields triple the world s average yield. The main California cultivars are Calimyrna, Adriatic, Mission, Brown Turkey, and Kadota (Stover et al., 2007a). Until recently, fresh figs represented less than 5% of total fig production (Stover et al., 2007a); most of the California figs are destined for the dried market (Obenauf et al., 1978; Soby, 1997; Stover et al., 2007a; Tous and Ferguson, 1996). However, during the period from 2002 to 2006, fresh fig production increased fourfold, constituting 16% of California s 2006 fig production (U.S. Department of Agriculture, 2007). This increase in fresh fig production is a consequence of increasing consumer demand for fresh quality produce of less familiar fruits (Stover et al., 2007a). In a preliminary survey of 1200 men and women conducted by Synovate, only 67% of the people surveyed were familiar with figs and only 55% with fresh figs. Of those surveyed, only 39% had ever eaten fresh figs, whereas 77% had eaten figs in cookies or bars (Synovate, 2004). The large number of consumers unaware of figs, combined with positive consumer perception, indicates there is potential for development of a fresh fig market. However, most current California fig cultivars were selected for drying, and the growers have little fresh fruit handling experience. If a profitable fresh fig industry is to be developed in California, cultivar selection, fruit maturity, and postharvest technology during marketing should be evaluated to produce the quality fresh fig that will increase consumer consumption. Therefore, this work investigated the impact of two fruit maturity stages on fresh fig quality attributes of four cultivars currently grown in California, including consumer acceptance. Materials and Methods Plant materials. Figs for this experiment were harvested from 8-year-old trees grown in a commercial fig orchard in Madera County, CA, with a tree spacing of 3.7 m 7.3 m. Four fresh fig cultivars (Mission, Brown Turkey, Calimyrna, and Kadota) were harvested at two maturity stages, commercial and tree ripe, in The California fig industry is largely based on these four cultivars and Adriatic (Stover et al., 2007b). Figs were considered commercially mature when the fruit flesh gave slightly when touched, whereas tree ripe fruit was riper and softer than commercial maturity but not overripe. In general, tree ripe green and dark skin-colored fig cultivars were visually greener and darker than those harvested at commercial maturity. A few of the tree ripe fruit well exposed in the canopy had leakage from the ostiole. Fruit selection and preparation. Seventyfive fig fruits per cultivar and maturity stage were used from each of the four replicated fig trees randomly selected in the field. Each fig cultivar and maturity stage was evaluated for initial quality attributes (after harvest) and consumer acceptance. Consumer acceptance was evaluated using an in-store consumer test as described subsequently (Crisosto and Crisosto, 2001). Quality evaluations. Initial fruit quality attribute measurements included fresh weight, firmness, skin color, soluble solids concentration (SSC), titratable acidity (TA), ethylene production (evolved ethylene), respiration (CO 2 production), and antioxidant capacity. Individual fresh weight, firmness, and skin color (L, C, H) were measured on 15 fruit per cultivar and maturity. Individual fresh weight was measured with a digital scale (Model PM 4000; Mettler Instrument Corp., Hightstown, NJ) and expressed in grams. Fruit firmness was measured using a fruit texture analyzer (GS.14; Güss, Strand, South Africa) adapted with a flat tip. Each fig was compressed on the cheek with a 2.5-cm flat tip at a speed of 5mms 1 to a depth of 4 mm and the maximum value of force was expressed in Newtons (N). Skin color was measured using a Minolta colorimeter (Model CR-200; Osaka, Japan). The data were expressed as luminosity [L, ranging from darkness (negative L) to HORTSCIENCE VOL. 45(4) APRIL

138 lightness (positive L)], chroma (C, indicating intensity or saturation of the color), and hue [H, angle that indicates the pure spectrum color (i.e., which wavelength is most dominant)]. For SSC and TA, the flesh of five figs was pressed through cheesecloth with a hand press to obtain a composite juice sample (Crisosto et al., 2004). Four composite juice samples were obtained for each cultivar and maturity. The juice was used for determination of SSC with a temperature-compensated handheld refractometer (Model ATC-1; Atago Co., Tokyo, Japan) and the values were expressed as percentages. Three grams of each composite juice sample were used for determination of TA with an automatic titrator (Model 950; Orion, Boston, MA) and reported as a percentage of citric acid. For determination of ethylene production and respiration at 20 C, four fruit per cultivar and maturity were weighed individually with a digital scale and placed individually in 705-mL sealed plastic containers. Air samples from the containers were withdrawn after 1 h. The concentration of ethylene was determined with a gas chromatograph equipped with a flame ionization detector (Carle AGC-211; EG&G Chandler Engineering, Tulsa, OK), 8% NaCl on Alumina F-1 column, helium (275.7 kpa, 30 mlmin 1 ) as a carrier, and air and hydrogen (124.1 and kpa, respectively) as combustion gases. Ethylene production rates were calculated and expressed as ml C 2 H 4 kg 1 h 1. The respiration rate was determined with an infrared CO 2 analyzer (Model PIR-2000R; Horiba Instruments, Irvine, CA) and calculated andexpressedasmlco 2 kg 1 h 1. Antioxidant activity. Three replicates of 4 g of fig sample per cultivar and maturity stage were used to measure the levels of antioxidant capacity by the DPPH free radical method (Brand-Williams et al., 1995). Samples were extracted in methanol [highperformance liquid chromatography (HPLC) grade, EMD, MX0475P-1] and homogenized using a polytron (Ultra-Turrax TP 18/101 S1; Junke & Kunkel, Staufen, Germany) for 40 s. The homogenate was allowed to stand for 24 h at 4 C in dark conditions and then centrifuged (Sorvall RC5C; Du Pont Company, Wilmington, DE) for 15 min at 14,000 rpm at 4 C. The supernatant was collected and stored in dark conditions at 80 C until analysis. Antioxidant capacity was determined using a spectrophotometer (Thermomax Microplate Reader; Molecular Devices, Sunnyvale, CA). Into each well of a 96-well microtiter plate, 247 ml DPPH (2,2-diphenyl-1-picrylhydrazyl, free radical) solution (121.7 mm in HPLCgrade methanol) was pipetted along with 13 ml of pure methanol (control), Trolox (watersoluble vitamin E analog) standard solution (88 to 800 mm; Cayman Chemical, Ann Arbor, MI), or sample extract. After 24 h at 25 C, absorbance was measured at 515 nm using methanol as a blank. The antioxidant capacity was expressed in micromoles Trolox equivalents per gram of fresh weight tissue (mmol TE/ gfw). In-store consumer test. For consumer acceptance evaluation, figs were stored after harvestat0 C for 1 d until the day of the consumer test. An in-store consumer test was conducted on the two maturity stages of the four fresh fig cultivars according to previous work (Crisosto and Crisosto, 2001). One hundred consumers, including a diverse combination of ages and genders, were surveyed at Whole Foods Market, Fresno, CA, in Aug Each consumer was presented, monadically, with eight fresh fig samples (one sample from each cultivar and maturity stage) in random order at room temperature (20 C). A sample consisted of half of a fig cut longitudinally from the stem end to the ostiole end, and it was presented in a 163-mL soufflé cup labeled with a three-digit code. The samples were prepared at the supermarket in the produce room out of sight from the testing area. Before tasting the samples, the consumers were asked about any possible allergic reactions to figs and how frequently they eat fresh figs; in addition, demographic data (gender, race, age range) were recorded. For each sample, the consumers were asked to taste it and then asked to indicate, all things considered, which statement best describes how they feel about the sample on a 9-point hedonic scale (1 = dislike extremely to 9 = like extremely). The consumers were instructed to sip bottled water in between samples to cleanse their palate. Consumer acceptance was measured as both degree of liking (1 to 9) and percentage of acceptance. Percentage of acceptance was calculated as the number of consumers liking the sample (score greater than 5.0) divided by the total number of consumers within that sample (Lawless and Heymann, 1998). In a similar manner, the percentage of consumers disliking (score less than 5.0) and neither liking nor disliking (score = 5.0) the sample were calculated. Statistical analysis. The data from the fig initial quality (except antioxidant capacity) are presented using descriptive statistics as means and SDs by cultivar and maturity stage. Linear regressions were performed between quality and consumer acceptance parameters, and adjusted R 2 is reported for all significant regressions (P # 0.05). The experimental design of antioxidant capacity was a 4 2 factorial design. The experimental design of the consumer test was repeated measures. The data from the consumer test were analyzed considering cultivar and maturity separately (two variables) as well as the combination cultivar maturity (only one variable). The data from antioxidant capacity and the consumer test were analyzed using a general linear model with cultivar and maturity as fixed factors with the program SPSS (SPSS_16.0, 2008) and significance was tested at P # Mean separation was determined by Tukey s honestly significant difference (P # 0.05). Results and Discussion Quality evaluations. Fig weight, firmness, skin color, SSC, TA, and SSC:TA were all affected by cultivar, maturity stage, and by the interaction between these two factors. Therefore, only the interaction data are discussed (Table 1). Ethylene production was affected only by cultivar or maturity stage (data not shown). Furthermore, the respiration rate (47 to 67 ml CO 2 kg 1 h 1 ) was unaffected by cultivar, maturity stage, or their interaction. Ethylene production was the highest in Mission (5.8 mlkg 1 h 1 ) and Brown Turkey (5.9 mlkg 1 h 1 ) followed by Kadota (4.8 mlkg 1 h 1 ) and Calimyrna (4.12 mlkg 1 h 1 ) with the lowest. Ethylene production was lower at the tree ripe stage (4.7 mlkg 1 h 1 ) than at the commercial stage (5.7 mlkg 1 h 1 ). This decrease in ethylene production from commercial maturity to tree ripe maturity is consistent with observations that the climacteric peak in figs routinely occurs before commercial maturity, perhaps reflecting that the individual fruitlets in the composite fig synconium are physiologically mature before the softening of the fig (C.H. Crisosto, unpublished data). Skin color Table 1. Interaction between cultivar and maturity stage on quality attributes of four fresh fig cultivars harvested from a commercial orchard in Madera County, CA, SSC (%) TA (% citric acid) SSC:TA Cultivar Maturity stage Wt (g) FTA (N) L* Chroma Hue Color z Mission Commercial Tree ripe Brown Turkey Commercial Tree ripe Calimyrna Commercial Tree ripe Kadota Commercial Tree ripe P value < < < < < < < LSD z L*, Chroma, and Hue are color parameters: lightness/darkness, chroma, and hue angle, respectively. FTA = fruit texture analyzer; SSC = soluble solids concentration; TA = titratable acidity; LSD = least significant difference. 708 HORTSCIENCE VOL. 45(4) APRIL 2010

139 varied within each cultivar from green to yellow ( Calimyrna, Kadota ) and brown to dark color ( Brown Turkey, Mission ) as expected. Brown and black skin color cultivars turned darker and green yellow skin color cultivars became more yellow at the higher maturity stage (tree ripe). Skin color expressed as luminosity and chroma were and 4.03 for the dark Mission tree ripe and and for the green Calimyrna tree ripe, respectively. The higher values of luminosity and chroma of the green cultivars denote lighter and more intense colors. Hue values were 103 for the green tree ripe cultivars, which is in the range of the green yellow colors, fortreeripe Mission,intherangeof blue colors, and for Brown Turkey, in the range of red yellow colors. Mission harvested at both maturity stages had the lowest weight (36.5 g) followed by Brown Turkey harvested at commercial maturity. Fruit weight was higher for the other cultivar maturity combinations and there was no significant difference among them (Table 1). The cultivar maturity stage combinations were segregated into five groups according to their firmness. Calimyrna, a Smyrna type, had the highest firmness (2.29 N) when harvested at the commercial stage, and Brown Turkey, a common fig type, the lowest (0.65 N) when harvested at the tree ripe stage. Among the rest of the cultivar maturity combinations, firmness ranged from 0.85 to 1.55 N. Commercial Mission, tree ripe Calimyrna, and commercial Kadota were firmer than tree ripe Brown Turkey and Mission. SSC was higher in tree ripe Mission, Brown Turkey, Calimyrna, and Kadota than the rest of the combinations. Commercial Mission, Brown Turkey, and Calimyrna combinations had the lowest SSC. Brown Turkey (both maturities) and tree ripe Kadota had the lowest TA (0.25%) followed by Mission (both maturities) and tree ripe Calimyrna, whereas commercial Kadota and Calimyrna had the highest acidity (0.62%). Therefore, tree ripe Kadota had the highest SSC:TA followed by Brown Turkey (both maturities) and tree ripe Mission. The lowest SSC:TA was measured on Calimyrna (both maturities) and commercial Kadota and Mission. A similar range of SSC in fresh fig has been reported in these cultivars in Italy (Chessa, 1997), in Turkey (Aksoy, 1998; Aksoy et al., 2003; Ilgin and Küden, 1997; Küden et al., 2008), and for fig cultivars from different areas of the world that are growing in California (Bremer, 2008). Antioxidant activity. Antioxidant capacity differed significantly between cultivars but not between maturity stages and there was no significant interaction between cultivars and maturity stages (Table 2). Mission had the highest antioxidant capacity (3.14 mmol TE/g FW) followed by Calimyrna and Brown Turkey (1.80 mmol TE/g FW), whereas Kadota had the lowest antioxidant capacity (1.44 mmol TE/g FW). The higher antioxidant capacity of Mission, which was almost double the others, is likely attributed to its dark skin color. Our fig antioxidant capacity values were similar to the ones reported for cultivars and a selection of strawberries (Battino and Mezzetti, 2006), higher than the ones recently reported for peaches and plums (Wang et al., 2008), and equal to or lower than some reported for blueberry cultivars (Bremer et al., 2008). Similar results were observed in six commercial fig cultivars with different skin colors ( Brown Turkey, Brunswick, Bursa, Chechick, Kadota, and Mission ) growing commercially under Israeli conditions (Solomon et al., 2006). Influence of genotype on antioxidant capacity has been reported in strawberries, apples, peaches, blueberries, and apricots (Bremer et al., 2008; Scalzo et al., 2005; Vizzotto et al., 2007). In this study, it was observed that most of the compounds with antioxidant activity such as polyphenols, anthocyanins, and flavonoids were located in the fig skin. Thus, fig cultivars with dark skin contained higher levels of antioxidant activity compared with fig cultivars with lighter skin (Solomon et al., 2006). Polyphenols analysis carried out by HPLC confirmed that phenolics were mainly concentrated in the peel with black cultivars having the highest content (Piga et al., 2008). In vitro and in vivo studies have demonstrated that fig antioxidants were able to affect processes that alter the development of tumor cells and to avoid neurochemical and behavioral changes related with aging (Shukitt-Hale et al., 2007; Vinson et al., 2005). The lack of significant differences in antioxidant capacity between maturity stages has been previously reported on peaches and plums. In dark color cultivars Brown Turkey and Mission, skin color turned dark during ripening demonstrated by hue angle measurements. This change in hue angle indicated that anthocyanins may have increased in figs harvested at the tree ripe stage; however, this suggested that the anthocyanin increase did not affect antioxidant capacity. In other species, it has been reported that antioxidant capacity was well correlated with the phenolics content but not with the anthocyanin content (Ferreyra et al., 2007). In-store consumer test. The in-store consumer test population (n = 100) consisted of 70.1% female and 29.4% male participants. Seventy percent of the consumer population was in the age range between 18 and 59 years old and 30% were 60 years old or older. The consumer population was formed by 78.3% Caucasians, 11.7% Hispanics, 3.6% Asians, 1.8% African Americans, and 4.6% others. Acceptance of the fresh fig cultivars tested was not significantly different between demographic subgroups in this population (data not shown). The degree of liking of the four fresh fig cultivars evaluated in the in-store consumer test was affected by cultivar and maturity stage at harvest, but there was no significant interaction between cultivar and maturity stage (Table 3). Consumers liked Kadota and Mission moderately with an acceptance of 82% and 72%, respectively. Calimyrna and Brown Turkey were liked slightly with Table 2. Antioxidant capacity of four fresh fig cultivars (n = 3) harvested at commercial and tree ripe maturity stages from a commercial orchard in Madera County, CA, TEAC (mmol TE/g FW) Cultivar z Mission 3.14 a Brown Turkey 1.73 bc Calimyrna 1.88 b Kadota 1.44 c Significance Maturity stage z Commercial 1.98 a Tree ripe 2.12 a Significance Cultivar maturity stage Significance z Within the comparison of cultivars and within the comparison of maturity stages values labeled with different letters are significantly different (P # 0.05, Tukey s honestly significant difference). TEAC = Trolox equivalent antioxidant capacity; FW = fresh weight. an acceptance of 64%. For these four tested cultivars, the percentage of consumers disliking them ranged from 13.5% ( Kadota ) to 27.5% ( Calimyrna ). There was a small percentage of consumers (5% to 11%) that chose the neither like nor dislike option. Tree ripe figs had a higher acceptance than figs harvested at commercial maturity. The average degree of liking for all tree ripe figs was moderately, whereas the average degree of liking for all commercial mature figs was neither like nor dislike. The average consumer acceptance for all tree ripe fig cultivars tested reached 86%, whereas all commercial mature fig cultivars tested had only 66% acceptance. The percentage of consumers disliking the figs was almost three times higher (33%) for commercial mature figs in contrast to only 9.8% for tree ripe figs. Fresh fig consumer acceptance was as high as that reported for sweet cherries (Crisosto et al., 2003), table grapes (Crisosto and Crisosto, 2002), tree fruit (Crisosto and Crisosto, 2005), blueberries (Bremer et al., 2008), and kiwifruit (Crisosto and Crisosto, 2001) using the same in-store sensory technique. There was a significant positive correlation between SSC and degree of liking (adjusted R 2 =0.77;P = 0.000) and between SSC:TA ratio and degree of liking (adjusted R 2 = 0.46; P = 0.022). This coefficient of determination indicates that only 77% of the changes in degree of liking are related to changes in SSC. This suggests that other factors related to cultivar flavor perception by consumers are also important in consumer acceptance. Because these quality attributes do not appear to fully explain the cultivar maturity stage differences in consumer acceptance, further detailed sensory and biochemistry work should be pursued to understand this relationship. For example, the level of latex decreases as fresh figs ripen; thus, this may be an important and unique component of sensory attributes that contribute to the differences in acceptance between these two maturity stages. Significant but negative correlations were determined between TA (adjusted R 2 =0.27; HORTSCIENCE VOL. 45(4) APRIL

140 Table 3. Consumer acceptance of four fresh fig cultivars harvested at commercial and tree ripe maturity stages from a commercial orchard in Madera County, CA, P = ) or firmness (adjusted R 2 =0.37; P = ) and degree of liking. These levels of relationships have also been reported in other commodities such as peach, nectarine, and plum (Crisosto et al., 2006). The low impact of TA (27%) on consumer acceptance can be explained by the low TA values and variability between these cultivars (0.22% to 0.65%). In our previous work, TA had a significant impact on consumer acceptance of plums and kiwifruit only when TA values were higher than 0.90% on ripe fruit (Crisosto and Crisosto, 2001; Crisosto et al., 2004). Conclusions Degree of liking (1 9) z This work demonstrated that fresh fig cultivars currently grown in California have a high level of antioxidants and are highly accepted by consumers. Cultivar and maturity stage had a significant effect on consumer acceptance, but did not affect antioxidant capacity. Degree of liking by consumers and percentage of consumer acceptance were better related to SSC alone than to TA or SSC:TA. Because a large number of consumers are still unfamiliar with fresh figs, educational promotion should be pursued as a result of the large potential for the fresh fig market. This work also demonstrated the need for cultivars better suited for fresh consumption that have better taste at the less mature stage and/or remain firm enough at the tree ripe stage to tolerate postharvest handling during harvesting and marketing. Different methods of packing and marketing this highly perishable fruit should also be investigated. Literature Cited Aksoy, U Why figs? An old taste and new perspective. Proc. First Intl. Symp. Fig. ISHS Acta Hort. 480: Aksoy, U., H.Z. Can, A. Misirli, S. Kara, G. Seferoglu, and N. Sahin Fig (Ficus carica L.) selection study for fresh market in western Turkey. Acta Hort. 605: Battino, M. and B. Mezzetti Update on fruit antioxidant capacity: A key tool for Mediterranean diet. Public Health Nutr. 9: Brand-Williams, W., M.E. Cuvelier, and C. Berset Use of a free radical method to evaluate antioxidant activity. Lebensmittel Wissenschaft + Technologie 28: Acceptance (%) Neither like nor dislike (%) Dislike (%) Cultivar y Mission 6.3 b Brown Turkey 5.7 a Calimyrna 5.9 a Kadota 6.8 b Significance Maturity stage y Commercial 5.3 a Tree ripe 7.0 b Significance Cultivars maturity stage interaction z Degree of liking: 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. y Within the comparison of cultivars and within the comparison of maturity stages, values labeled with different letters are significantly different (P # 0.05, Tukey s honestly significant difference). Bremer, V Use of preharvest ethephon applications to reduce breba crop load and postharvest performance evaluations of fresh fig (Ficus carica) varieties/selections. MS thesis, Calif. State Univ., Fresno, CA. Bremer, V., G. Crisosto, R. Molinar, M. Jimenez, S. Dollahite, and C.H. Crisosto San Joaquin Valley blueberries evaluated for quality attributes. Calif. Agr. 62: Chessa, I Fig, p In: Mitra, S. (ed.). Postharvest physiology and storage of tropical and subtropical fruits. CAB International, Wallingford, UK. Crisosto, C.H. and G.M. Crisosto Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22: Crisosto, C.H. and G.M. Crisosto Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24: Crisosto, C.H. and G.M. Crisosto Relationship between ripe soluble solids concentration (RSSC) and consumer acceptance of high and low acid melting flesh peach and nectarine [Prunus persica (L.) Batsch] cultivars. Postharvest Biol. Technol. 38: Crisosto, C.H., G.M. Crisosto, G. Echeverria, and J. Puy Segregation of peach and nectarine [Prunus persica (L.) Bastch] cultivars according to their organoleptic characteristics. Postharvest Biol. Technol. 39: Crisosto, C.H., G.M. Crisosto, and P. Metheney Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color. Postharvest Biol. Technol. 28: Crisosto, C.H., D. Garner, G.M. Crisosto, and E. Bowerman Increasing Blackamber plum (Prunus salicina Lindell) consumer acceptance. Postharvest Biol. Technol. 34: Ferreyra, R.M., S.Z. Viña, A. Mugridge, and A.R. Chaves Growth and ripening season effects on antioxidant capacity of strawberry cultivar Selva. Sci. Hort. 112: Food and Agriculture Organization Data archives. FAOSTAT. 24 Nov < fao.org/site/408/desktopdefault.aspx>. Hertog, M.G., P.M. Sweetnam, A.M. Fehily, P.C. Elwood, and D. Kromhout Antioxidant flavonols and ischemic heart disease in a Welsh population of men: The Caerphilly Study. Amer. J. Clin. Nutr. 65: Ilgin, M. and A.B. Küden Table fig selection study in the Kahramanmaras Province of Turkey. Acta Hort. 441: Kader, A Importance of fruits, nuts, and vegetables in human nutrition and health. Perishables Handling Qrtly. 106:4 6. Küden, A.B., S. Bayazit, and S. Cxömlekcioglu Morphological and pomological characteristics of fig genotypes selected from Mediterranean and South East Anatolia regions. Acta Hort. 798: Lawless, H.T. and H. Heymann Acceptance and preference testing, p In: Lawless, H.T. and H. Heymann (eds.). Sensory evaluation of food, principles and practices. Chapman and Hall, New York, NY. Michailides, T.J Diseases of fig, p In: Ploetz, R.C. (ed.). Diseases of tropical fruit crops. CABI, Wallingford, UK. Obenauf, G., M. Gerdts, G. Leavitt, and J. Crane Commercial dried fig production in California. Division of Agricultural Sciences, University of California. Piga, A., A. Del Caro, G. Milella, I. Pinna, V. Vacca, and S. Schirru HPLC analysis of polyphenols in peel and pulp of fresh figs. Acta Hort. 798: Scalzo, J., A. Politi, N. Pellegrini, B. Mezzetti, and M. Battino Plant genotype affects total antioxidant capacity and phenolic contents in fruit. Nutrition 21: Shukitt-Hale, B., A.N. Carey, D. Jenkins, B.M. Rabin, and J.A. Joseph Beneficial effects of fruit extracts on neuronal function and behavior in a rodent model of accelerated aging. Neurobiol. Aging 28: Soby, S Fig fact sheet. Fruit & Nut Research and Information Center. 8 Nov < fruitsandnuts.ucdavis.edu/index.html>. Solomon, A., S. Golubowicz, Z. Yablowicz, S. Grossman, M. Bergma, H.E. Gottlieb, A. Altman, Z. Kerem, and M.A. Flaishman Antioxidant activities and anthocyanin content of fresh fruits of common fig (Ficus carica L.). J. Agr. Food Chem. 54: SPSS_ SPSS Base 16.0 for Windows. SPSS Inc., Chicago, IL. Stover, E., M. Aradhya, C. Crisosto, and L. Ferguson. 2007a. Overview of the California fig industry and new interest in varieties for fresh fruit. Proc. California Plant and Soil Conference: Opportunities for California Agriculture, Sacramento, Calif. p Stover, E., M. Aradhya, L. Ferguson, and C. Crisosto. 2007b. The fig: Overview of an ancient fruit. HortScience 42: Synovate Fig exploratory attitude and usage study. Presentation of results. Research reinvented. Prepared for Kraft Foods. Job number P979 (Release # ). Tous, J. and L. Ferguson Mediterranean fruits. Purdue University. NewCROP. 20 Oct < default.html>. U.S. Department of Agriculture Noncitrus fruits and nuts 2006 summary. Agricultural statistics board. NASS, USDA. 29 Nov < viewdocumentinfo.do?documentid=1113>. Vinson, J.A., L. Zubik, P. Bose, N. Samman, and J. Proch Dried fruits: Excellent in vitro and in vivo antioxidants. J. Amer. Coll. Nutr. 24: Vizzotto, M., J. Cisneros-Zevallos, D.H. Byrne, D.W. Ramming, and W.R. Okie Large variation found in the phytochemical and antioxidant activity of peach and plum germplasm. J. Amer. Soc. Hort. Sci. 132: Wang, S.Y., C.-T. Chen, W. Sciarappa, C.Y. Wang, and M.J. Camp Genetic improvement of fruits and vegetables. J. Agr. Food Chem. 56: HORTSCIENCE VOL. 45(4) APRIL 2010

141 Research Article New quality index based on dry matter and acidity proposed for Hayward kiwifruit by Gayle M. Crisosto, Janine Hasey, Jorge A. Zegbe and Carlos H. Crisosto Researchers from various countries have proposed using dry matter at harvest as a worldwide quality index for Hayward kiwifruit, because it includes both soluble (sugars and acids) and insoluble (structural carbohydrates and starch) solids and doesn t change during postharvest handling. Our consumer tests in 1999 and 2008 indicated that dry matter and ripe titratable acidity are related to in-store consumer acceptance of kiwifruit. In most California seasons, when ripe titratable acidity was less than 1.2%, only a dry matter greater than or equal to 15.1% was required for consumer acceptability. Our 6-year quality attribute survey of California kiwifruit at harvest and from cold storage demonstrated that dry matter and ripe soluble solids concentration were highly variable among vineyards and seasons, but ripe titratable acidity values varied more among seasons than between vineyards. Our results provide strong evidence that dry matter would be a reliable quality index candidate for California kiwifruit, especially if ripe titratable acidity were factored in. Measuring the soluble solids concentration of kiwifruit juice at harvest is the official method of assessing its maturity in most kiwifruit-producing countries, including New Zealand, Italy, France, Greece, Chile, Japan and the United States (Beever and Hopkirk 1990; Crisosto and Mitchell 2002). A refractometer is used to make the simple and fast measurement of the concentration of soluble solids such as sugars, organic acids, phenolic compounds and pectins. Minimum harvest maturity standards for Hayward kiwifruit (Actinidia deliciosa [A. In research analyzing the role of various fruit quality attributes in consumer acceptance, kiwifruit slices were dried in a dehydrator in order to measure dry matter content. Chev.] C. F. Liang and A. R. Ferguson) are enforced in several countries including the United States, ranging from 5.5% to 6.5% soluble solids concentration at harvest (HSSC) (Beever and Hopkirk 1990; Crisosto and Mitchell 2002). This range assures a minimum consumer acceptance (greater than or equal to 12.5% ripe soluble solids concentration) and adequate storage potential to avoid flesh breakdown, which occurs when soluble solids are less than 6.2% at harvest (Crisosto and Crisosto 2001). Kiwifruit is usually harvested when mature but unripe, then kept in cold storage for up to 6 months. Ripening starts at various points in the distribution chain on the fruit s way to the consumer (Ritenour et al. 1999). Ripening time depends on how long the fruit has been stored at 32 F, and whether it has been preconditioned with ethylene. Kiwifruit harvested with less than 6.2% soluble solids concentration develop flesh breakdown by 3 months in storage at 32 F. Soluble solids increase slowly during cold storage as starch is converted to sugars, but the complete conversion occurs when the kiwifruit is ripened. At harvest, a mature, unripe kiwifruit has a high content of starch and soluble sugars, but soluble solids concentration readings do not take starch into consideration because it is insoluble. Therefore, soluble solids concentration measurements do not accurately predict final soluble sugars after ripening. As ripening and softening progress during the kiwifruit s postharvest life, starch is hydrolyzed to soluble sugars and consequently the soluble solids concentration increases. The sweetness, eating quality, consumer acceptance and repeat purchases of kiwifruit are strongly associated with the concentration of soluble sugars (Burdon et al. 2004; Crisosto and Crisosto 2001; Harker et al. 2009; Jordan et al. 2000; Rossiter et al. 2000). Following an in-store consumer test, UC researchers proposed a soluble solids concentration after ripening (RSSC) of 12.5% as the minimum quality index for early-marketed California kiwifruit (Crisosto and Crisosto 2001). Online: landingpage.cfm?article=ca.v066n02p70&fulltext=yes DOI: /ca.v066n02p70 Carlos H. Crisosto 70 CALIFORNIA AGRICULTURE VOLUME 66, NUMBER 2

142 Hayward kiwifruit dominate California production, and they are marketed worldwide. California growers export little, but fruit is shipped to the Southern Hemisphere (New Zealand and Chile) and Italian markets when local crops are out of season, and California receives imports from these countries as well. This globalization has created economic advantages for early- and late-harvest kiwifruit sales, when fruit availability is low and prices are high. However, the incentive to harvest early or to hold fruit in long-term storage can result in low-quality kiwifruit in the market, reducing repeat purchases and overall demand. To more reliably assure flavor quality, researchers from various countries including New Zealand have proposed, in addition to the harvest maturity index, the use of dry matter concentrations as a voluntary quality index, which would be measured at harvest and/or shipment to market. Dry matter readings include starch and are highly correlated with ripe soluble solids concentration (Beever and Hopkirk 1990; Harker et al. 2009; Jordan et al. 2000), and they do not change during cold storage (Crisosto et al. 2009). New Zealand and Chile have started using a minimum dry matter standard for retail marketing and wholesale trade. However, as yet, there is no agreement on the minimum dry matter level that should be established as a quality index. To develop a quality index for Hayward kiwifruit, we studied the relationships among dry matter, ripe soluble solids concentration, ripe titratable acidity (RTA) and consumer acceptance. We also investigated dry matter variability in kiwifruit at harvest and from cold storage from several California growing regions, and we surveyed imported kiwifruit during the U.S. low-availability season (March to May and August to November). storage at 32 F [0 C] ), cold kiwifruit were exposed to 100 parts per million ethylene for 12 hours, following the preconditioning protocol (Ritenour et al. 1999). This process was carried out 2 to 3 days prior to the consumer test, and the kiwifruit were allowed to ripen at 68 F (20 C) until they reached 2 to 3 pounds-force (1 pound-force equals 4.45 Newtons) flesh firmness (Crisosto and Crisosto 2001). Flesh firmness and dry matter were measured on the day of the test before the consumers tasted the samples, and juice was extracted from the remaining fruit after tasting on the same day of the test, using previously described methods (Crisosto and Crisosto 2001; Crisosto, Hasey et al. 2008). A kiwifruit sample consisted of a 0.25-inch-thick slice, halved, cut perpendicular to the long axis of the fruit, adjacent to the location where the slice for dry matter assessment was taken. The consumer responses were recorded using a 9-point hedonic scale (1 = dislike extremely, 9 = like extremely). Consumer acceptance was measured as a degree of liking and expressed as a percentage (Lawless and Heymann 2010). Fruit quality attributes survey During the 2006 and 2007 growing seasons, samples of 10 to 15 kiwifruit, replicated three times, were collected at six different maturity stages (harvest dates) from the same vineyards: three in the San Joaquin Valley (southern region) and three in the Sacramento Valley (northern region). All kiwifruit samples from these vineyards and commercial cold-storage facilities and retail stores in California for the low-availability season survey were immediately transported to the F. Gordon Mitchell Glossary Dry matter: The ratio of the weight of a test sample after drying to the fresh weight of the test sample, expressed as a percentage. Kiwifruit dry matter at harvest is composed of both starch and soluble sugars as well as organic acids, minerals, pectins and other components. Soluble solids concentration: A measurement consisting primarily of soluble sugars, as well as soluble pectins, and organic, amino and ascorbic acids. Measured using refractometers or hydrometers. Starch: A white, tasteless, solid carbohydrate (C 6 H 10 O 5 ) n occurring in the form of minute granules in seeds, tubers and other parts of plants. Measured by potassium iodide test or other chemical tests. Titratable acidity: A measure of the total amount of acid present expressed as a percentage of the predominant acid, as determined by titration or chemical tests in a laboratory. Carlos H. Crisosto In-store consumer survey During the 1999 and 2008 kiwifruit growing seasons, we surveyed groups of 142 and 124 consumers, respectively, at a major supermarket in Fresno County. Each consumer was presented with three ripe Hayward kiwifruit samples from California vineyards at targeted dry matter levels (14.0% to 15.9%, 16.0% to 17.9% and 18.0%). To ripen the previously coldstored kiwifruit samples (1 to 5 weeks At the F. Gordon Mitchell Postharvest Center in Parlier, UC Davis master s student Jiaxuan Liu juiced kiwifruit collected from vineyards across the Central Valley, as well as imported fruit. April June

143 Postharvest Center at the Kearney Agricultural Research and Extension Center in Parlier, California, for fruit quality assessments. Fruit quality attributes measured at harvest included flesh firmness, soluble solids concentration, titratable acidity and dry matter (Crisosto, Hasey et al. 2008). Soluble solids concentration and titratable acidity of ripened fruit were measured as described by Crisosto and Crisosto (2001). In addition to the vineyard survey, kiwifruit samples of three replicates of 10 to 15 kiwifruit were collected directly from commercial cold-storage facilities in California during 1998, 1999, 2006, 2007, 2008 and 2009 for dry matter determinations. Imported kiwifruit samples from Chile and New Zealand (three replications of 10 kiwifruit) were collected for dry matter determinations directly from California retail stores in 2009 and Quality and consumer acceptance Dry matter. The in-store consumer tests showed that dry matter content significantly influenced the degree of liking in both seasons (tables 1 and 2). In the 1999 growing season, consumers rated kiwifruit from like slightly (6.1) to like moderately (7.1). The percentage of consumers who said they like (score > 5.0) the kiwifruit varied from 71% to 87% depending on dry matter, while the percentage of consumers who chose dislike (score < 5.0) ranged between 6% and 26%. Only a few consumers (3% to 6%) chose neither like nor dislike (table 1). Degree of liking was significantly higher (6.6 to 7.1) and acceptance was approximately 85% for kiwifruit with dry matter greater than or equal to 16.1% than for kiwifruit with dry matter less than 16.1%. The in-store consumer test results for the 2008 growing season were similar to the 1999 results (table 1). Consumers rated kiwifruit from dislike slightly (4.6) to like slightly-moderately (6.5). The percentage of consumers who said they like the fruit varied from 35% to 76%, increasing as dry matter increased, while the percentage who said they dislike the fruit decreased from 50% to 20% as dry matter increased. Neither like nor dislike was chosen by 4% to 15% of consumers (table 1). Ripe titratable acidity. Dry matter and ripe titratable acidity, which is associated with sourness, significantly influenced the degree of liking in the 2008 consumer test, but not in the 1999 test. In 2008, the F-ratio for ripe titratable acidity was significantly higher than for dry matter. (A significant F-ratio means that there is at least one significant difference among These results demonstrate that consumer acceptance of kiwifruit is affected by dry matter and acidity levels. means being compared.) As a result, we divided the data set into two classes for further analysis using dry matter and ripe titratable acidity as combined factors on degree of liking. With the same dry matter levels, one class had a ripe titratable acidity greater than or equal to 1.2% and the other less than 1.2% (table 2). Consumers rated kiwifruit with ripe titratable acidity greater than or equal to 1.2% from dislike slightly (4.4) to like TABLE 1. Relationship between dry matter, as a percentage of fresh weight, and consumer acceptance of Hayward kiwifruit, 1999 and 2008 TABLE 2. Consumer acceptance of Hayward kiwifruit at different levels of dry matter, as a percentage of fresh weight, and ripe titratable acidity (RTA), as a percentage of citric acid, 2008 Dry matter n* Degree of liking Consumer acceptance Neither like nor Like dislike Dislike % Dry matter n* Degree of liking Consumer acceptance Neither like nor Like dislike Dislike % b b ab a > a LSD P > F < b b a a > a LSD P > F < * Number of samples. 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. Mean separations within a column were by Fisher s LSD test (P 0.05). Mean values followed by the same letters were not significantly different. RTA 1.2% b b a a > a LSD P > F 0.01 RTA < 1.2% b a a a > a LSD P > F 0.05 * Number of samples. 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, 9 = like extremely. Mean separations within a column were by Fisher s LSD test (P 0.05). Mean values followed by the same letters were not significantly different. 72 CALIFORNIA AGRICULTURE VOLUME 66, NUMBER 2

144 slightly (5.7), and acceptance ranged from 34% to 65%. A high percentage of consumers (28% to 57%) said they dislike these kiwifruit. In this high titratable acidity class, degree of liking was significantly higher for kiwifruit with dry matter greater than or equal to 16.1% (acceptance ranged from 54% to 65%) than with dry matter less than 16.1%. Consumers rated kiwifruit with ripe titratable acidity less than 1.2% from dislike slightly (4.5) to like moderately (6.8), and acceptance ranged from 33% to 83%. A high percentage of consumers (13% to 33%) said they dislike these kiwifruit. The degree of liking was significantly higher for kiwifruit with dry matter greater than or equal to 15.1% (acceptance ranged from 74% to 83%). In this class, the percentage of consumers who said they dislike the fruit decreased from 33% to 13% as dry matter increased. In contrast, consumers who chose neither like nor dislike varied among dry matter classes, ranging between 0% and 33% (table 2). These results demonstrate that consumer acceptance of kiwifruit is affected by dry matter and acidity levels. Quality at harvest and after ripening 2006 samples. In the 2006 growing season, the average dry matter of kiwifruit sampled from the three San Joaquin Valley vineyards (A, B, C) increased from 14.6% to 16.2% between Sept. 14 and Oct. 23. During this sampling period, harvest soluble solids concentrations ranged between 5.1% and 6.6%, and harvest titratable acidity was between 1.8% and 2.0%. After ripening to a flesh firmness of 2 to 3 pounds-force (9 to 13.5 Newtons), the kiwifruit exhibited average ripe soluble solids concentrations between 10.4% and 13.1% and titratable acidity between 0.4% and 1.1%. (Data in this section is not shown; it is available from authors upon request.) Minimum quality indexes (dry matter 15.1% and soluble solids concentration 6.2%) are important to consider at harvest because they can predict consumer acceptance. For instance, vineyard A reached the minimum dry matter by Sept. 14 and the minimum harvest soluble solids concentration by Oct. 2, with titratable acidity at harvest of 2.0%. However, vineyard A exceeded a ripe soluble solids concentration of 12.5% by Sept. 25, and its ripe titratable acidity was lower than Some 4,200 acres of kiwifruit are harvested annually in California. In this study, dry matter content at harvest, coupled with ripe titratable acidity, was a reliable consumer acceptance indicator. 1.2%. Vineyard B met the minimum quality indexes at harvest and after ripening in the last harvest (Oct. 23). Vineyard C did not reach the minimum quality indexes during this harvest season. Average dry matter at harvest of kiwifruit grown in the Sacramento Valley vineyards (D, E, F) increased from 16.1% to 17.9% between Sept. 15 and Oct. 23. Soluble solids concentrations at harvest increased from 5.4% to 7.0% during this time, and titratable acidity ranged between 1.8% and 2.0%. After ripening kiwifruit to a flesh firmness of 2 to 3 pounds-force (9 to 13.5 Newtons), average soluble solids concentration and titratable acidity varied between 11.8% and 13.3%, and 0.5% and 1.2%, respectively. Vineyard D fruit had the highest dry matter on the first harvest date (Sept. 15); dry matter was the same in the three vineyards for the rest of the harvest dates. The minimum quality index of dry matter greater than or equal to 15.1% was observed on the first harvest date (Sept. 15) for the three Sacramento Valley vineyards. However, none of the fruit from these vineyards reached the minimum quality index for soluble solids concentration until the last two harvest dates. After ripening, fruit from vineyard D had soluble solids concentrations greater than or equal to 12.5% on all harvest dates, except the third harvest date. Fruit from the other two vineyards had ripe soluble solids concentrations greater than or equal to 12.5% by the fourth harvest date, while titratable acidity remained less than or equal to 1.2% samples. In the 2007 growing season, average dry matter at harvest (16.1% to 16.8%) and soluble solids concentrations at harvest (5.6% to 7.7%) and when ripe (13.3% to 14.5%) of kiwifruit grown in the San Joaquin Valley vineyards increased consistently between Sept. 27 and Oct. 22. (We did not measure titratable acidity at harvest or after ripening on fruit from these vineyards.) Vineyard A fruit had dry matter consistently higher on all harvest dates than fruit from vineyards B and C. Vineyard C fruit had consistently the lowest dry matter and harvest and ripe soluble solids concentrations on all harvest dates, but harvest soluble solids concentrations were similar in the three vineyards on the fourth harvest date (Oct. 22). Fruit from vineyard A met the minimum quality index for dry matter and soluble solids concentration by the first (Sept. 27) harvest date, and fruit from vineyard B met it by the second (Oct. 4) harvest date. Vineyard C fruit did not meet the minimum quality index for dry matter on any harvest date, and it met the minimum quality index for soluble solids concentration on the last harvest date (Oct. 22). After ripening, vineyards A and B fruit met the minimum quality index for soluble solids concentration on the first harvest date (Sept. 27), while vineyard C fruit met that value by the third harvest (Oct. 11). Similar results were observed in the three vineyards in the Sacramento Valley during 2007, although the values were higher compared with those in the San Joaquin Valley. Also, titratable acidity David Gomez, istockphoto.com April June

145 decreased from harvest to ripening, with a smaller decrease in 2006 than in Seasonal variation. After two consecutive growing seasons, the data suggest that variation in dry matter and other fruit quality attributes depends on harvest time (Crisosto et al. 2007, 2009; Crisosto, Garibay et al. 2008), growing area (Crisosto et al. 2007, 2009; Crisosto, Garibay et al. 2008), vineyard and year. For these six vineyards, dry matter variability was higher among vineyards than between seasons, and ripe titratable acidity variability was low among vineyards but high between seasons. Our 6-year California kiwifruit quality attribute survey demonstrated that dry matter and ripe titratable acidity were highly variable among seasons, with ripe titratable acidity values varying more among seasons than dry matter (fig. 1). Proposed kiwifruit quality index Based on our studies, we propose a minimum quality index of dry matter greater than or equal to 16.1% when ripe titratable acidity is greater than or equal to 1.2%, and 15.1% when ripe titratable acidity is less than 1.2%, to maximize consumer satisfaction. Ripe titratable acidity (%) Dry matter (%) Years Fig. 1. Seasonal variation in dry matter and ripe titratable acidity of Hayward kiwifruit. Vertical bars on mean values indicate standard deviations. During the six growing seasons of harvest and cold-storage sample collection from California vineyards (n = 3,156), dry matter ranged from 11% to 22% with an average of 17.1% and a standard deviation of 1.6% (fig. 2A). Of these samples, 76.2% and 90.7% exceeded our proposed minimum quality indexes of dry matter greater than or equal to 16.1% and 15.1%, respectively. In the imported kiwifruit collected in 2009 and 2010 (n = 48), dry matter ranged between 14% and 19%, with an average of 16.0% and a standard deviation of 1.4% (fig. 2B). Of these imported samples, 37.5% and 77.1% exceeded our proposed minimum quality indexes of dry matter greater than or equal to 16.1% and 15.1%, respectively. Although the influence of ripe titratable acidity on consumer acceptance has been previously documented (Crisosto and Crisosto 2001; Rossiter et al. 2000), the direct impact of dry matter and ripe titratable acidity together on kiwifruit acceptance (table 2), using single-fruit measurements, has not been previously reported. In our study, kiwifruit with high ripe titratable acidity ( 1.2%) required high dry matter ( 16.1%) for consumers to like the fruit; for kiwifruit with ripe titratable acidity less than 1.2%, a minimum dry matter of 15.1% was adequate for consumer acceptance. This effect of ripe titratable acidity may be explained by the fact that high dry matter is associated with sweetness (Burdon et al. 2004; Jordan et al. 2000), which may balance a sourness perception, as reported in kiwifruit and mango (Harker et al. 2009; Malundo et al. 2001). When fruit is perceived to be less sour (ripe titratable acidity < 1.2%), consumers are less demanding for high soluble sugars, or dry matter content, and they like kiwifruit with dry matter greater than or equal to 15.1%. Crisosto and Crisosto (2001) reported that total organic acid content, in terms of titratable acidity at harvest, was relatively stable across harvest seasons. Similar to that study, we noticed a decrease in ripe titratable acidity between harvest and ripening in both years of our fruit quality attributes survey (data not shown). In 2006, titratable acidity decreased 54% from harvest to ripening, and in 2007 it decreased 70%. The high ripe titratable acidity observed during the in-store consumer test in the 2008 growing season may explain Frequency (%) (A) California (B) Imported Dry matter (%) Fig. 2. Dry matter distribution of Hayward kiwifruit for (A) California harvest and coldstorage samples in 1998, 1999, 2006, 2007, 2008 and 2009 growing seasons (n = 3,156) and (B) imported samples collected from commercial cold-storage facilities and retail stores during the 2009 and 2010 U.S. low-availability season (March to May and August to November) (n = 48). the high dry matter level (and indirectly the ripe soluble solids concentration) needed for consumers to like the kiwifruit when ripe titratable acidity was greater than or equal to 1.2% (table 2). Therefore, in terms of marketing, dry matter would be a more accurate tool than soluble solids concentration alone for controlling minimum quality at harvest, and for meeting export and import requirements. As only minor changes in dry matter occur during postharvest handling, kiwifruit with low consumer acceptance either harvested early in the season or from potentially low-quality vineyards (dry matter < 15.1%) could be detected. In our study, differences in dry matter among kiwifruit sources remained constant throughout the cold-storage period. This lack of dry matter change is an advantage over the soluble solids concentration maturity index currently in place. In some commodities, soluble sugars and organic acids (important components of dry matter) are reduced during cold storage because of respiratory activity. The kiwifruit s low metabolic respiration favors the maintenance of dry matter and soluble sugars, as well as a low water loss 74 CALIFORNIA AGRICULTURE VOLUME 66, NUMBER 2

146 California Kiwifruit Commission potential when fruit is kept at 32 F (0 C) and with relative humidity of 90% to 95%. Kiwifruit s respiration rate is less than or equal to 2 milliliters carbon dioxide per kilogram per hour at 32 F (0 C), whereas the rates for other commodities, such as peach and apple, are about 1.5 and 3.0 times higher, respectively, at the same temperature (Ritenour et al. 1999). Dry matter content should not be used as the sole index at harvest, since storage potential tests at different dry matter levels have not been investigated extensively. To assure long cold-storage potential, we recommend that the new dry matter quality index be used with the current harvest maturity index between 6.2% and 6.5% soluble solids concentration (a minimum of 6.2% is required to avoid flesh breakdown during long-term cold storage of 3 to 6 months). Our proposed dry matter quality index would segregate out kiwifruit of low consumer quality either harvested early in the season or from low-quality vineyards. While the dry matter values may not be reached during early harvests (September, in California), some exceptions could be allowed depending on growing area, vineyard management and weather conditions in particular years, as suggested by other authors (Burdon et al. 2004; Crisosto and Crisosto 2001). Our data indicated that there was strong variation from one season to the next in attributes such as dry matter, harvest and ripe soluble solids concentrations, and harvest and ripe titratable acidity. Among these attributes, ripe titratable acidity had the lowest variability within a season. These differences in kiwifruit quality attributes from year to year, coupled with consumer acceptance, highlight the need to segregate kiwifruit based on consumer acceptance. Because the proposed minimum dry matter quality index is based on single-fruit measurements and consumer acceptance, it would allow single-fruit segregation during packaging without regard to population variability, perhaps using postharvest technologies such as near-infrared sensors (Slaughter and Crisosto 1998). Even though there were highly significant correlations between dry matter and soluble solids concentration in our studies, we do not recommend the use of dry matter as a maturity index. A maturity index is used to define when a given commodity has reached the stage of development that after harvesting and postharvest handling (including ripening, where required) its quality will be at least the minimum acceptable to the consumer. This maturity index may be an enforced standard; a quality index is a guide that is used voluntarily to assure a certain level of quality. Previously published information demonstrated that immature or overmature kiwifruit, with maturity based on harvest soluble solids concentration, can develop senescent breakdown the breakdown of tissues or the development of granular, mealy or soaked tissues during long cold-storage periods (Crisosto and Crisosto 2001). Our studies demonstrate that dry matter is a reliable consumer quality index, but not a reliable maturity index, so harvest decisions should still include harvest soluble solids concentration and firmness to protect kiwifruit in long-term storage or marketing conditions. Studies are needed to understand the potential limitations of dry matter in predicting long-term cold-storage performance. G.M. Crisosto is Specialist, Department of Plant Sciences, UC Davis; J. Hasey is Farm Advisor, UC Cooperative Extension, Sutter-Yuba Counties; J.A. Zegbe is Visiting Scientist, Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias, Campo Experimental Zacatecas, Mexico; and C.H. Crisosto is UC Cooperative Extension Postharvest Physiologist, Department of Plant Sciences, UC Davis. This research was funded by the California Kiwifruit Commission and USDA NIFA grant # Part of the data analysis was undertaken during J.A. Zegbe s sabbatical leave, supported by the Consejo Nacional de Ciencia y Tecnología (CONACYT, México) No. Ref , Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP, México) and Universidad Autónoma de Zacatecas (México). References Beever DJ, Hopkirk G Fruit development and fruit physiology. In: Warrington IJ, Wetson GC (eds.). Kiwifruit: Science and Management. Auckland, NZ: Ray Richards Pub. p Burdon J, McLeod D, Lallu N, et al Consumer evaluation of Hayward kiwifruit of different at-harvest dry matter contents. Postharvest Biol Tech 34: Crisosto CH, Crisosto GM Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol Tech 22: Crisosto CH, Garibay S, Hasey J, et al Evaluating Kiwifruit Dry Weight Research Report to California Kiwifruit Commission. Crisosto CH, Hasey J, Cantin CM, et al New kiwifruit dry weight protocol. UC Cooperative Extension Central Valley Postharvest Newsletter 17:11 5. Crisosto CH, Hasey J, Crisosto G, Garner D Evaluating Kiwifruit Dry Weight Research Report to California Kiwifruit Commission. Crisosto CH, Hasey J, Slaughter D, Crisosto G Evaluating Dry Weight Sensors to Segregate Kiwifruit According to Consumer Acceptance Research Report to California Kiwifruit Commission. Crisosto CH, Mitchell FG Kiwifruit. In: Kader AA (ed.). Postharvest Technology of Horticultural Crops. UC ANR Pub Oakland, CA. p Harker FR, Carr BT, Lenjo M, et al Consumer liking for kiwifruit flavour: A meta-analysis of five studies on fruit quality. Food Qual Prefer 20: Jordan RB, Walton EF, Klages KU, Seelye RJ Postharvest fruit density as an indicator of dry matter and ripened soluble solids of kiwifruit. Postharvest Biol Tech 20: Lawless HT, Heymann H Acceptance testing. In: Lawless HT, Heymann H (eds.). Sensory Evaluation of Food: Principles and Practices. New York, NY: Springer. p Malundo TMM, Shewfelt RL, Ware GO, et al Sugars and acids influence flavor properties of mango (Mangifera indica). J Am Soc Hort Sci 126: Ritenour MA, Crisosto CH, Garner DT, et al Temperature, length of cold storage and maturity influence the ripening rate of ethylene-preconditioned kiwifruit. Postharvest Biol Tech 15: Rossiter KL, Young H, Walker SB, et al The effects of sugars and acids on consumer acceptability of kiwifruit. J Sens Stud 15: Slaughter DC, Crisosto CH Nondestructive internal quality assessment of kiwifruit using near-infrared spectroscopy. Seminars in food analysis 3: April June

147 Postharvest Biology and Technology 28 (2003) 159/167 Consumer acceptance of Brooks and Bing cherries is mainly dependent on fruit SSC and visual skin color Carlos H. Crisosto *, Gayle M. Crisosto, Paul Metheney Department of Pomology, University of California at Davis Kearney Agricultural Center, 9240 South Riverbend Ave., Parlier, CA 93648, USA Received 2 April 2002; accepted 28 July 2002 Abstract During two seasons, in-store consumer acceptance tests were performed to determine the relationship between soluble solids concentration (SSC), titratable acidity (TA) and visual skin color on Brooks and Bing cherry consumer acceptance. For this, approximately 600 consumers were presented cherry samples at targeted skin colors with SSC in the range of ca. 13.0/20.0% and TA in the range of ca. 0.50/1.00%. For each cherry sample, one half of the whole cherry was tasted and the other half was used to determine SSC and TA. TA/0.60% reduced consumer acceptance on Brooks cherries with B/16.0% SSC compared to cherries with 5/0.60% TA, while in Bing the same situation only occurred on cherries with 5/13.0% SSC. High consumer acceptance was determined on Brooks and Bing cherries when SSC were /16.0% without regard to TA. For both cultivars, the highest percentage of American consumers would buy cherries based on dark skin color without regard to ethnic group (Caucasian, Asian American, Hispanic, or Black) or gender. However, consumer age was related to making the buy or not to buy decision based on cherry skin color. Consumers under 18 years old were less biased to buy cherries based on visual skin color. Thus, this work demonstrated that for Brooks and Bing cherries, a full bright red or dark mahogany skin color should be reached, respectively, in addition to a minimum SSC of 16.0% to satisfy the majority of American consumers. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Prunus avium L.; In-store consumer tests; Soluble solids concentration; Titratable acidity; Degree of liking; Consumer preferences; Demographics; Age 1. Introduction High soluble solids concentration (SSC) and aroma have been associated with high fruit consumer preference and/or acceptance for different * Corresponding author. Tel.: / ; fax: / address: carlos@uckac.edu (C.H. Crisosto). commodities (Pehrson and Ivans, 1988; Robertson and Meridith, 1989; Rodan, 1988; Nelson et al., 1963; Gorini and Lasorella, 1990); other quality attributes may also be important in consumer acceptance (Kader, 1999). For example, in peaches and nectarines, consumers will prefer full red color fruit to less full red color fruit (Bruhn, 1995). Furthermore, it has been pointed out in mango (Malundo et al., 2001) and other fruits (Crisosto, /02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S ( 0 2 )

148 160 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/167 unpublished) that aroma may play an important compensation role when SSC is lacking. Minimum quality standards based on SSC to assure consumer satisfaction have been proposed for Bing (Cliff et al., 1996; Dever et al., 1996; Drake et al., 1989; Kappel et al., 1996; Guyer et al., 1993; Schotzko, 1993) and Ranier (Drake and Fellman, 1987) cherries. We believe that the relationship between SSC and TA and visual appearance plays an important role in consumer acceptance as it is reported for other commodities such as citrus (Pehrson and Ivans, 1988); table grapes (Guelf-Reich and Safran, 1971; Nelson et al., 1973; Crisosto and Crisosto, 2002); kiwifruit (Gorini and Lasorella, 1990; Crisosto and Crisosto, 2001) and mango (Malundo et al., 2001). Therefore, it is important to understand the potential involvement of TA in cherry consumer acceptance prior to proposing a minimum quality standard. This relationship between SSC and TA, and consumer acceptance may also be cultivar specific and even related to ethnic group. Bing has been the dominant commercial cherry cultivar, but recently new cultivars that mature earlier or later than Bing are now becoming commercially important (Crisosto et al., 2002; Kappel et al., 2002). For example, Brooks ripens 5/7 days after Early Burlat and about 7/12 days before Bing (Crisosto et al., 1993). Brooks was selected in the University of California cherrybreeding program among the progeny of a cross between Ranier (white flesh) and Early Burlat (Hansche et al., 1988). The purpose of our research was to determine consumer acceptance of Brooks and Bing cherries in relation to SSC and TA, and the importance of visual cherry skin color on consumers decision to buy. Understanding the relationships between the consumer, cherry quality attributes and industry-wide quality surveys will help to develop a minimum eating quality index for these two cultivars. 2. Materials and methods In-store consumer tests were carried out on Brooks and Bing cherries for two seasons. A total of 581 and 596 consumers at a major supermarket, located in Fresno County, were interviewed for Brooks and Bing cherries, respectively. Based on our previous years industry-wide surveys (Crisosto, 1997; Crisosto et al., 2002; Mitcham et al., 1998, 1999, 2000) to define a minimum quality index, each consumer was presented four cherry samples at the four targeted skin colors for each cultivar. Prior to tasting, nondestructive firmness (Durofel, COPA-TECH- NOLOGY S.A., France) and color (Minolta colorimeter, Minolta, CR-200, Japan) were measured on one cheek of each whole cherry sample before cutting it in half to taste to avoid using too soft cherries. As hue angle was the most important color descriptor, cherry skin color was expressed only as hue angle (h8). The hue angle is expressed in degrees and is a measure of color that, for example, from 0 to 908 spans from red to orange to yellow. Brooks cherries were segregated according to skin color based on our previous work (Crisosto et al., 1993, 2002) as full light red (Hue /26.15), 50% bright red (Hue /21.96), full bright red (Hue /16.90), and full dark red (Hue /11.85), while Bing cherries were segregated using the chart from the Centre Technique Interprofessionel des Fruits et legumes (CTIFL, France). Bing cherry skin color was defined as salmon (Hue /27), red (Hue /20.4), mahogany (Hue /15.0), and dark mahogany (Hue /13.3). One cherry half from each sample in a covered, labeled 29.6 ml soufflé cup was immediately placed in an ice chest and kept cold for immediate transport to the F. Gordon Mitchell Postharvest Laboratory for subsequent chemical analysis of SSC and TA (Crisosto and Crisosto, 2002). For both cultivars, each consumer that said he/she ate fresh cherries was asked to taste four cherry half samples presented in random order in coded 29.6 ml soufflé cups at room temperature. In both seasons, each consumer interviewed was asked to indicate their age range on a chart; the interviewer noted their gender and ethnic group. The consumer was instructed to wear dark glasses during the tasting to mask the skin color. For each cherry sample, the consumer was asked if he/she liked, disliked, or neither liked nor disliked the sample. Then, the consumer was asked to indicate

149 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/ his/her degree of liking/disliking: slightly, moderately, very much, or extremely. The consumer s response was recorded using a 9-point hedonic scale (1*/dislike extremely to 9*/like extremely). The consumer was instructed to sip bottled water in between samples to cleanse his/her palate. Consumer acceptance was measured as both degree of liking (1/9) and percentage acceptance (O Mahony, 1986). The percentage of consumers liking the cherry sample was calculated as the number of consumers liking the cherry sample (score/5.0) divided by the total number of consumers within the sample (Lawless and Heymann, 1998). The percentage of consumers disliking the cherry sample (scoreb/5.0) was calculated as the number of consumers disliking the cherry sample divided by the total number of consumers within the sample. The percentage of consumers that neither liked nor disliked the cherry sample was calculated as the number of consumers that neither liked nor disliked the cherry sample (score/5.0) divided by the total number of consumers within the sample. After tasting the four samples, the consumer was instructed to remove the dark glasses. Then, he/she was presented one of three sets of Brooks or Bing cherries. Each set consisted of four cups in random order of three whole cherries with stems per cup of each of the four Brooks or Bing cherry skin colors previously tasted. The consumer was asked to look at the four cups of cherries and indicate just by looking at them which one he/she would buy. The percentage of consumers that would buy a specific skin color based solely on the visual skin color was calculated as the number of consumers choosing each skin color divided by the total number of consumers. The degree of liking data was subjected to analysis of variance (ANOVA) prior to the Least Significant Differences (LSD) mean separation (Lawless and Heymann, 1998) using the SAS program. 3. Results and discussion On cherries used for these in-store consumer tests, single Brooks cherry SSC measurements Fig. 1. Single cherry soluble solids concentration (SSC) for Brooks across maturity based on skin color using 581 cherries. Central vertical line is the mean of the sample, outer vertical lines represent9/1 S.D. about the mean. (581 cherries) varied from 9.0 to 27.0% (Fig. 1) and TA from 0.24 to 1.02% (Fig. 2) across skin color stages, while single Bing cherry SSC measurements (596 cherries) varied from 11.4 to 27.0% (Fig. 3) and TA from 0.53 to 1.19% (Fig. 4) across skin color stages. For Brooks, average SSC varied from 12.8 to 21.6% and average TA changed from 0.47 to 0.67% as cherries turned from full light red to full dark red. For Bing, average SSC varied from 16.5 to 20.6% and average TA changed from 0.78 to 0.90% as cherries turned from salmon to dark mahogany. In both cultivars, at each given skin color stage, there was a large overlap of SSC and TA among the different skin color stages. For example, the distribution of SSC in Brooks was similar between cherries at the 50% bright red and full bright red stages, while the distribution of SSC in Bing cherries was similar between cherries at the mahogany and dark mahogany stages. An even larger overlap of TA occurred between different skin color stages for Brooks and Bing cherries,

150 162 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/167 Fig. 2. Single cherry titratable acidity (TA) for Brooks across maturity based on skin color using 581 cherries. Central vertical line is the mean of the sample, outer vertical lines represent9/1 S.D. about the mean. although the average TA for Brooks was always lower (ca. 0.30%) than the average TA for Bing at each given skin color stage. This variability in quality attributes (SSC and TA) within a given skin color stage and its relationship to sensory perception variability was reported earlier on table grapes (Nelson et al., 1963). In both cherry cultivars, there was a large increase in SSC and a small increase in TA as skin color turned from light to dark resulting in an increase in SSC:TA. A similar situation occurred in our 3 year industry-wide quality survey on Brooks (Crisosto et al., 2002), which indicated that average TA changes during maturation/ripening (based on skin color changes) are small but depend on the orchard s specific conditions and year influence. For example, average TA for Brooks across orchards and years changed from 0.81 to 0.77% during maturation/ripening while skin color changed from full light red to full dark red. During this same maturity period, SSC increased from 15.3 to 20.4% and SSC:TA increased from 20.3 to This small change in TA Fig. 3. Single cherry soluble solids concentration (SSC) for Bing across maturity based on skin color using 596 cherries. Central vertical line is the mean of the sample, outer vertical lines represent9/1 S.D. about the mean. during maturation/ripening was also observed in our 1997 quality attributes survey for Bing carried out during May 8/22 in 36 orchards located in the Stockton area (Crisosto, 1997). In this survey, SSC ranged from 13.8 to 18.0%, TA varied from 1.00 to 1.20%, and SSC:TA ranged from 12 to 16 for Bing cherries with skin color from red to dark mahogany. Mitcham s group has described similar TA changes for a Bing orchard in their 3 years of work in California. (Mitcham et al., 1998, 1999, 2000). In this orchard, Bing cherry average TA was 0.85% when measured at the salmon or the dark mahogany skin color. TA decreased from 0.85 to 0.81% when the skin color changed from salmon to red, but TA increased from 0.81 to 0.85% when skin color changed from mahogany to dark mahogany. During this same maturity period, SSC increased from 13.0 to 20.6% and SSC:TA from 15.3 to While skin color turned from light to dark, there was a large increase in cherry SSC and a small increase in TA. In all of these cases, the increase in SSC:TA

151 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/ Table 1 Consumer acceptance of Brooks cherries by American consumers at different levels of titratable acidity within the SSC range tested TA range z Degree of liking y Acceptance (%) (1 /9) (%) B/ b / b / a / a / a 95 LSD / P-value / z TA, single cherry measurements expressed as percentage malic acid. y Degree of liking: 1/dislike extremely, 2/dislike very much, 3/dislike moderately, 4/dislike slightly, 5/neither like nor dislike, 6/like slightly, 7/like moderately, 8/like very much, 9/like extremely. Same letters within the same column indicate no significant difference between means. Fig. 4. Single cherry titratable acidity (TA) for Bing across maturity based on skin color using 596 cherries. Central vertical line is the mean of the sample, outer vertical lines represent9/1 S.D. about the mean. that occurred at the end of the maturation/ripening period was mainly related to an increase in SSC rather than a decrease in TA. It appears that TA level varied according to cultivar, environmental and orchard management conditions, but TA changes and the final TA level were not highly influenced by maturation/ripening. During our previous work, we observed a decrease in TA during storage for Brooks, Tulare, King, and Garnet cherries (unpublished data). Similar observations have been published for other sweet cherry cultivars such as Bing, Lapins, Santina, Skeena, Sumnue Cristalina, Sweetheart, etc. (Kappel et al., 2002). Degree of liking in our single cherry samples was significantly related to SSC and TA. In general, consumer acceptance responses were divided into two groups based on the relationship between degree of liking and TA within the SSC range tested. There was a significant separation in liking between Brooks cherries with 5/0.60% TA and /0.60% TA (Table 1). Brooks cherries with TA 5/0.60% were accepted by 49 /64% of consumers while degree of liking score varied from 5.9 to 6.3. On cherries with TA /0.60% acceptance ranged from 76 to 95%, while degree of liking score varied from 6.6 to 7.4. For Bing cherries, consumer acceptance was also divided into two groups based on the relationship between degree of liking and TA across the SSC range tested. There was a significant separation in degree of liking between cherries with TA 5/0.80% and TA /0.80%. Bing cherries with TA 5/0.80% were accepted by 49 to 55% of consumers while degree of liking score varied from 5.5 to 5.7. On cherries with TA /0.80% acceptance ranged from 66 to 80% while degree of liking score varied from 6.0 to 6.6 (Table 2). In general, Brooks cherries with TA/0.60% and Bing cherries with TA/ 0.80% had the highest SSC levels and thus, the highest SSC:TA. The increase in SSC:TA during the maturation/ripening period resulted from the higher increase in SSC than the increase in TA. This explains why consumer acceptance increased as TA increased in these samples. In all of the cases, the increase in TA was accompanied with an increase in SSC:TA. Because there was significant interaction between SSC and TA on the degree of liking, consumer acceptance based on SSC for two levels

152 164 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/167 Table 2 Consumer acceptance of Bing cherries by American consumers at different levels of titratable acidity within the SSC range tested TA range z Degree of liking y Acceptance (%) (1/9) (%) 0.50/ b / b / a / a 80 LSD P-value z TA, single cherry measurements expressed as percentage malic acid. y Degree of liking: 1/dislike extremely, 2/dislike very much, 3/dislike moderately, 4/dislike slightly, 5/neither like nor dislike, 6/like slightly, 7/like moderately, 8/like very much, 9/like extremely. Same letters within the same column indicate no significant difference between means. of TA was analyzed for both cultivars. In both cultivars, consumer acceptance increased as SSC increased. For Brooks cherries with 5/13.0% SSC, degree of liking was not significantly different for cherries with /0.60% TA (score 3.9) and for cherries with 5/0.60% TA (score 3.7). The number of consumers that liked the cherries within this category ranged from 11.1 to 26.6%. Thus for Brooks cherries with SSC 5/13.0% TA did not play a role in consumer acceptance. For cherries within the 13.1/16.0% SSC range, consumers liked slightly (5.9 score) Brooks cherries with 5/ 0.60% TA, while they neither liked nor disliked (4.7 score) cherries with /0.60% TA. For cherries within the 13.1 /16.0% SSC range with 5/0.60% TA consumer acceptance was 66.7% while it was only 42.9% for cherries within the same SSC range with /0.60% TA. For Brooks cherries with / 16.0% SSC, TA did not influence consumer responses. TA played an important role in consumer acceptance of cherries with 5/16.0% SSC and TA/0.60%. Within this range of SSC and TA, high SSC compensated for high TA, or low TA compensated low SSC, thus SSC:TA was more sensitive to consumer acceptance than SSC within this range of SSC and TA. Consumers liked moderately (ca. 7.3 score), cherries with / 16.0% SSC disregarding TA reaching the highest consumer acceptance (86.5 /96.7%). Cherry acceptance did not significantly increase on cherries with /20% SSC (Table 3). In a previous work carried out on Ranier, one of Brooks parents (Drake and Fellman, 1987), a minimum quality index of 16.0% SSC was proposed. The influence of SSC on Bing cherry consumer acceptance with 5/0.80% TA and /0.80% TA was determined (Table 4). For Bing cherries with 5/13.0 SSC, degree of liking was significantly lower on cherries with /0.80% TA (score 2.4) than for cherries with 5/0.80% TA (score 4.1). In both cases, consumer acceptance was 31.8% or lower. Cherries within the 13.1 /16.0% SSC range were neither liked nor disliked (5.3 score) disregarding TA and acceptance varied from 47.7 to 58.3%. The same situation occurred on cherries within the 16.1 /20.0% SSC range; cherries were liked slightly (6.1 score) and acceptance varied from 70.0 to 72.3%. Consumer acceptance increased significantly for cherries with SSC / 20.0%. These cherries were liked moderately disregarding TA (ca. 7.3 score) and acceptance reached approximately 90%. In this cultivar, TA only influenced consumer acceptance for cherries with 5/13.0% SSC. However, consumer acceptance was the highest for cherries with /20% SSC. In both cultivars, the number of consumers that chose the neither like nor dislike option varied from 22.2 to 0%. Based on the results of our work, Bing cherries with /16.0% SSC without regard to TA were always liked by consumers but with different degrees of liking. A minimum SSC of 17.0 /19.0% was considered optimum by trained sensory panels for several sweet cherry cultivars grown in British Columbia (Kappel et al., 1996). For Bing, a minimum of 17.0% SSC has been suggested after an informal consumer test (Schotzko, 1993). The influence of cherry skin color on the decision to buy or not to buy Brooks and Bing cherries was also tested for demographics, i.e. gender, ethnic group, and age range. Consumers decided to buy cherries according to skin color; the darker the skin color the higher the percentage of consumers that would buy them. Gender and ethnic group (Caucasian, Asian,

153 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/ Table 3 Consumer acceptance of Brooks cherry by American consumers at different levels of soluble solids concentration (SSC) and titratable acidity (TA) measured as percentage malic acid Quality attributes Degree of liking (1/9) z Acceptance (%) Neither like nor dislike (%) Dislike (%) TA5/0.60% SSC5/13.0% 3.9c y SSC 13.1/16.0% 5.9b SSC 16.1/20.0% 7.2a SSC/20.0% 7.0a TA/0.60% SSC5/13.0% 3.7c SSC 13.1/16.0% 4.7c SSC 16.1/20.0% 7.2a SSC/20.0% 7.7a LSD / / / P-value z Degree of liking: 1/dislike extremely, 2/dislike very much, 3/dislike moderately, 4/dislike slightly, 5/neither like nor dislike, 6/like slightly, 7/like moderately, 8/like very much, 9/like extremely. y Same letters within the same column indicate no significant difference between means. Hispanic and Black) did not affect the decision to buy according to skin color (data not shown). Approximately 64% of the consumers decided to buy Brooks cherries with full dark red color, approximately 22% of consumers decided to buy full bright red color, and approximately less than 14% of consumers decided to buy cherries with less than full bright red color (data not shown). In general, approximately 80% of the consumers decided to buy Bing cherries with dark mahogany color, approximately 10% decided to buy mahogany color, and approximately less than 10% of consumers decided to buy Bing cherries lighter than mahogany color. Consumer age range influenced the decision to buy according to skin color in both cultivars Table 4 Consumer acceptance of Bing cherry by American consumers at different levels of soluble solids concentration (SSC) and titratable acidity (TA) measured as percentage malic acid Quality attributes Degree of liking (1/9) z Acceptance Neither like nor dislike Dislike TA5/0.80% SSC5/13.0% 4.1d y SSC 13.1/16.0% 5.3cd SSC 16.1/20.0% 6.1bc SSC/20.0% 7.5a TA/0.80% SSC5/13.0% 2.4e SSC 13.1/16.0% 5.4c SSC 16.1/20.0% 6.1bc SSC/20.0% 7.1ab LSD / / / P-value z Degree of liking: 1/dislike extremely, 2/dislike very much, 3/dislike moderately, 4/dislike slightly, 5/neither like nor dislike, 6/like slightly, 7/like moderately, 8/like very much, 9/like extremely. y Same letters within the same column indicate no significant difference between means.

154 166 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/167 Table 5 Percentage of American consumers that will buy Brooks and Bing cherries at each skin color according to age ranges Cultivar Age range Skin color Under 18 18/29 30/39 40/49 50/59 60 or older Brooks z Full light red % Bright red Full bright red Full dark red Bing y Salmon Red Mahogany Dark mahogany z 581 consumers interviewed. y 596 consumers interviewed. (Table 5). For Brooks cherries, the percentage of consumers that decided to buy full dark red cherries ranged from 41.7 to 78.4% as consumer age changed from under 18 to 60 years old or older. Percentage of consumers choosing to buy darker skin color cherries increased as they became older but it reached a plateau (ca. 74%) for consumers 40 years old or older. The percentage of consumers choosing to buy darker color Bing cherries dramatically increased from under 18 years old to 18 years old or older. The percentage of consumers that decided to buy dark mahogany Bing cherries remained close to 85% within the 18 years old or older range, while only 61.2% of consumers under 18 years old chose to buy them. This work points out that SSC, SSC:TA and visual skin color influence consumer acceptance of these two cherry cultivars. Although TA plays a role in consumer acceptance, within a given SSC range the importance of TA measurement is less relevant than SSC because TA changes are small in comparison to SSC changes during the cherry maturation/ripening period within a given orchard. SSC, which is easily measured in contrast to TA, is the major contributor to the SSC:TA, and therefore, consumer acceptance. Because high consumer acceptance and a high percentage of consumers making the decision to buy was based on full bright red ( Brooks ) or dark mahogany ( Bing ) skin color, we propose the use of SSC combined with full color development dictated by the cultivar as a minimum quality index for Brooks and Bing cherries in California. The full skin color requirement will also help to assure a large number of cherries with SSC/16.0% SSC at the picking time for both cultivars. Our proposed minimum quality index is higher than the current US Grade and Standards and the more demanding California Agricultural Code indexes (Kader, 2002). The California Agricultural Code requires a cherry surface with at least a solid light red and/or 14.0 /16.0% SSC, depending on the cultivar. Further work to understand the interaction of cultural practices and storage period, and on consumer acceptance during the maturation/ ripening changes should be pursued. Acknowledgements We would like to thank Dr. Anne Noble for her help in planning the sensory evaluation work. Thanks to the California Cherry Commission and California Cherry Growers Association for funding this work.

155 C.H. Crisosto et al. / Postharvest Biology and Technology 28 (2003) 159/ References Bruhn, C.M., Consumer and retail satisfaction with the quality and size of California peach and nectarines. J. Food Qual. 18, 241/256. Cliff, M.A., Dever, M.C., Hall, J.W., Girard, B., Development and evaluation of multiple regression models for prediction of sweet cherry liking. Food Res. Int. 28, 583/589. Crisosto, C.H., Survey of Bing cherry harvest quality. California Department of Food and Agriculture, Stockton, CA, 3 pp. Crisosto, C.H., Crisosto, G.M., Understanding consumer acceptance of early harvested Hayward kiwifruit. Postharvest Biol. Technol. 22, 205/213. Crisosto, C.H., Crisosto, G.M., Understanding American and Chinese consumer acceptance of Redglobe table grapes. Postharvest Biol. Technol. 24, 155/162. Crisosto, C.H., Garner, D., Doyle, J., Day, K.R., Relationship between fruit respiration, bruising susceptibility, and temperatures in sweet cherries. HortScience 28, 132/135. Crisosto, C.H., Crisosto, G.M., Ritenour, M.A., Testing the reliability of skin color as an indicator of quality for early season Brooks (Prunus avium L.) cherry. Postharvest Biol. Technol. 24, 147/154. Dever, M.C., MacDonald, R.A., Cliff, M.A., Lane, W.D., Sensory evaluation of sweet cherry cultivars. HortScience 31, 150/153. Drake, S.R., Fellman, J.K., Indicators of maturity and storage quality of Ranier sweet cherry. HortScience 22, 283/285. Drake, S.R., Williams, M.W., Fountain, J.B., Stemless sweet cherry (Prunus avium L.) fruit quality and consumer purchase. J. Food Qual. 11, 411/416. Hansche, P.E., Beres, W., Doyle, J., Micke, W.C., Brooks sweet cherry. HortScience 23, 644. Gorini, F., Lasorella, M., Sensory and objective evaluation of kiwifruit. Acta Hortic. 282, 309/314. Guelf-Reich, S., Safran, B., Indices of maturity for table grapes as determined by variety. Volcani Institute of Agricultural Research, Bet-Dagan, Israel. XVIII International Horticulture Congress, Tel-Aviv, Israel, March 17/ 25, Series No. 1672E, 13/18. Guyer, D.E., Sinha, N.K., Chang, T.S., Cash, J.N., Physiochemical and sensory characteristics of selected Michigan sweet cherry (Prunus avium L.) cultivars. J. Food Qual. 16, 355/370. Kader, A.A., Fruit maturity, ripening, and quality relationships. In: Symposium on Effect of Pre and Post Harvest Factors on Storage of Fruit. Acta Hortic. 485, 203/ 208. Kader, A.A., Standardization and inspection of fresh fruit and vegetables. In: Kader, A.A. (Ed.), Postharvest and Technology of Horticultural Crops, chapter 23. University of California, Agricultural and National Resources Publication Kappel, F., Fisher-Fleming, B., Hogue, E., Fruit characteristics and sensory attributes of an ideal sweet cherry. HortScience 31, 443/446. Kappel, F., Toivonen, P., Mckenzie, D.L., Stan, S., Storage characteristics of new sweet cherry cultivars. HortScience 37, 139/143. Lawless, H.T., Heymann, H., Acceptance and preference testing. In: Lawless, H.T., Heymann, H. (Eds.), Sensory Evaluation of Food, Principles and Practices. Chapman and Hall, New York, pp. 430/479. Malundo, T.M.M., Shewfelt, R.L., Ware, G.O., Baldwin, E.A., Sugars and acids influence flavor properties of mango (Mangifera indica). J. Am. Soc. Hort. Sci. 126, 115/121. Mitcham, E., Tayfun, A., Biasi, B., Crisosto, C., Gillespie, D., Determination of the relationship between Bing cherry harvest maturity and postharvest quality. Report to the California Cherry Advisory Board, 29 pp. Mitcham, E., Tayfun, A., Southwick, S., Biasi, B., Determination of the relationship between Bing cherry harvest maturity and postharvest quality. Report to the California Cherry Advisory Board. Mitcham, E., Tayfun, A., Southwick, S., Biasi, B., Determination of the relationship between Bing cherry harvest maturity and postharvest quality. Report to the California Cherry Advisory Board. Nelson, K.W., Baker, G.A., Winkler, A.J., Amerine, M.A., Richardson, H.B., Jones, F.R., Chemical and sensory variability in table grapes. Hilgardia 34, 1/26. Nelson, K.E., Schutz, H.G., Ahmedullah, M., McPherson, J., Flavor preferences of supermarket customers for Thompson Seedless grapes. Am. J. Enol. Viticult. 24, 31/40. O Mahony, M., Sensory Evaluation of Food. Marcel Dekker, New York. Pehrson, J.E., Ivans, E.M., Variability in early season navel orange clone maturity and consumer acceptance. Proc. Int. Soc. Citriculture, pp. 1631/1635. Robertson, J.A., Meridith, F.I., Characteristics of fruit from high- and low-quality peach cultivars. HortScience 23, 1032/1034. Rodan, L., Consumer acceptance of Florida grapes and grape products. Proceedings of the Viticulture Sciences Symposium Tallahassee, FL. Florida A&M University, Center for Viticultural Sciences, pp. 119/121. Schotzko, R.T., Fresh sweet cherry eating characteristics: some baseline data. Washington State Univ. Res. Bull. XB1028, 3 /14.

156

157

158

159

160

161

162

163

164

165 Provided for non-commercial research and educational use only. Not for reproduction, distribution or commercial use. This chapter was originally published in the book Postharvest Handling: A Systems Approach published by Elsevier, and the attached copy is provided by Elsevier for the author s benefit and for the benefit of the author s institution, for non-commercial research and educational use including without limitation use in instruction at your institution, sending it to specific colleagues who you know, and providing a copy to your institution s administrator. All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier s permissions site at: From: Postharvest Handling: A Systems Approach, Second Edition Edited by Wojciech J. Florkowski, Robert L. Shewfelt, Bernhard Brueckner and Stanley E. Prussia ISBN: Copyright 2009, Elsevier Inc. Academic Press.

166 Author s personal copy Nutritional Quality of Fruits and Vegetables 5 Ariel R. Vicente Facultad de Ciencias Agrarias y Forestales. UNLP. Calle 60 y 119 s/n. CP 1900 La Plata Argentina y Centro de Investigación y Desarrollo en Criotecnología de Alimentos (CIDCA) CONICET-UNLP. Calle 47 esq CP La Plata Argentina George A. Manganaris Department of Plant Sciences, University of California, Davis, One Shields Avenue, Davis, CA USA Gabriel O. Sozzi Cátedra de Fruticultura, Facultad de Agronomía, Universidad de Buenos Aires. Avda. San Martín C 1417 DSE Buenos Aires, Argentina CONICET, Argentina Carlos H. Crisosto Department of Plant Sciences, University of California, Davis, One Shields Avenue, Davis, CA USA I. Introduction II. Traditional components A. Water B. Organic acids C. Proteins D. Lipids and fatty acids E. Metabolizable carbohydrates F. Dietary fiber G. Vitamins III. Antioxidants in fruits and vegetables A. Oxidative damage and antioxidants B. Ascorbic acid C. Carotenoids D. Tocopherols and tocotrienols E. Phenolic compounds F. Factors affecting the levels of antioxidants in fruits and vegetables IV. Fruits and vegetables as direct sources of minerals A. General considerations of selected minerals B. Factors influencing mineral content of fruits and vegetables C. Effect of minerals on fruit and vegetable quality and consumer acceptance Bibliography Postharvest Handling: A Systems Approach ISBN: Copyright 2009, Elsevier Inc. All rights reserved

167 Author s personal copy 58 Nutritional Quality of Fruits and Vegetables I. Introduction Horticultural crops are some of the main components of a healthy diet. The constituents obtained by the human body from fruits and vegetables include water, carbohydrates, fats, proteins, fiber, minerals, organic acids, pigments, vitamins and antioxidants, among others. Fruits and vegetables, especially, are a good source of fiber, selected minerals, vitamins and antioxidants. Most fruits and vegetables are available almost year-round in a wide variety and they not only taste good, but they also have favorable attributes of texture, color, flavor and ease of use. They can be fresh, cooked, hot or cold, canned, pickled, frozen or dried. Fruits and vegetables are consumed at all times, and due to their convenient size; they are an excellent between-meal snack. They are relatively low in calories and fat (avocado and olives being the exceptions), they have no cholesterol, they are rich in carbohydrates and fiber, they contain vitamin C and carotene, and some are a good source of vitamin B 6. Fruits and vegetables are relatively low in sodium and high in potassium. Ascorbic acid in fruits and vegetables enhances the bioavailability of iron in the diet. Because of all these characteristics, fruits and vegetables have a unique role in a healthy diet. A growing body of research has shown that fruit and vegetable consumption is associated with reduced risk of major diseases, and possibly delayed onset of age-related disorders, promoting good health. However, in many cases fruit and vegetable consumption is still below the dietary guideline goal of consuming 5 10 servings each day. The nutritional value of fruits and vegetables depends on their composition, which shows a wide range of variation depending on the species, cultivar and maturity stage. The composition of fruits and vegetables includes a great number of metabolites however, it could be predicted that no single commodity might be rich in all these constituents. This chapter describes the general characteristics of the components of fruits and vegetables, related to their benefits as food sources. II. Traditional components A. Water The most abundant single component of fruits and vegetables is water, which may account for up to 90% of the total mass. The maximum water content varies between individual fruits and vegetables, because of structural differences. Cultivation conditions that influence structural differentiation may also have a marked affect. B. Organic acids There are two types of acids, namely aliphatic (straight chain) and aromatic acids. The most abundant acids in fruits and vegetables are citric and malic (both aliphatic) acids. However, large amounts of tartaric acid occur in grapes. Malic acid is the major component in oranges and apples. The acid content of fruits and vegetables generally decreases during maturation. For example, the citric acid content of clingstone peaches

168 Author s personal copy II. Traditional components 59 decreases faster than the malic acid content, while the malic acid content of apples and pears decreases faster than the citric acid content. Aromatic organic acids occur in several fruits and vegetables, but in very low concentrations. Benzoic acid occurs in cranberries, quinic acid in bananas and chlorogenic acid in potatoes. Organic acids play an important role in the sugar to acid ratio, which affects the flavor of fruits and vegetables. The distribution of acids within a fruit is not uniform. C. Proteins Proteins represent less than 1% of the fresh mass of fruit and vegetable tissues. Leguminous seeds are rich in protein, containing 15% to 30%. The proteins of fruits and vegetables are built from amino acids, but other related simple nitrogenous compounds also occur. Fruits, vegetables and legumes account for 1.2%, 5.5% and 6.1%, respectively, of the protein in the US food supply ( Hiza and Bente, 2007 ). Fruits are low in proteins, but tree nuts are a good source of high-quality proteins. The protein content of fresh fruits or vegetables is calculated by multiplying the total nitrogen content by a factor of This calculation uses the fact that protein is comprised of about 16% nitrogen, and the assumption that all nitrogen present is protein. The conversion ignores the fact that appreciable amounts of simple nitrogenous substances can be present in an uncombined form. In potatoes, 50% to 60% of the nitrogen occurs in the form of simple soluble constituents, while in apples the estimates range from 10% to 70% ( Salunkhe et al., 1991 ). Senescent tissues, such as those of overripe fruits, usually contain especially high proportions of non-protein nitrogen. Asparagine is abundant in potatoes and apples as non-protein nitrogen fractions. Pears and oranges are rich in proline, and black and red currants in alanine. D. Lipids and fatty acids Plant lipids represent a very broad group of compounds with functions that vary among products. Lipids are an energy source for plants during germination, forming components of cellular membranes and cuticular waxes, and they are mainly present as triglycerides (esters of glycerol and three fatty acids) or phospholipids (in which one fatty acid has been replaced by a phosphate group). Generally, most postharvest products are relatively low in total lipids, except for avocados, olives and many seeds. The fat content of fruits and vegetables is usually below 1% and varies with the product. Examples of fat content on a dry mass basis are: avocado: 35 70%; olive: 30 70%; grape: 0.2%; banana: 0.1%; and apple: 0.06%. Many of the physical and chemical properties of lipids are due to the fatty acids present in their structure. Fatty acids are aliphatic monocarboxylic acids that may be saturated or unsaturated to varying degrees. Saturated fatty acids do not contain any double bonds along the chain. Monounsaturated fatty acids have a single double

169 Author s personal copy 60 Nutritional Quality of Fruits and Vegetables bond in the hydrocarbon chain, and polyunsaturated fatty acids have more than one double bond. Fatty acids in plants usually range from 4- to 26-carbons in size, but oleic acid (18:1) and linoleic acid (18:2) are the most prevalent in nature. Olive oil and other fats high in monounsaturated fatty acids are becoming well-known for helping to lower LDL-cholesterol (the so-called bad cholesterol), while protecting HDL-cholesterol ( good cholesterol) when consumed in moderation in place of saturated fats. The difference among oils is not in their caloric content, but in their composition. Fats derived from animal sources (e.g. butter, cream, hard cheeses) have a high proportion of saturated fats, while oils from plant sources, such as olive and canola, have the lowest ( Table 5.1 ). Fatty acids are necessary for human bodily functions, where they are used primarily to produce hormone-like substances that regulate a wide range of functions including blood pressure, blood clotting, blood lipid levels, the immune response and the inflammatory response. The human body can produce most fatty acids, except for linoleic acid and α -linolenic acid, which are widely distributed in plant oils. These essential fatty acids are polyunsaturated fatty acid members of the omega-6 and omega-3 fatty acid series. Each double bond, depending on its geometry, can be in either a cis or a trans conformation. In cis bonds, the two carbons next to the unsaturated site bond atoms are oriented to the same side. Therefore, in restricted environments, such as when fatty acids are part of a phospholipid in a lipid bilayer or triglycerides in lipid droplets, cis bonds limit the ability of fatty acids to be closely packed and therefore, could affect the melting temperature of the membrane or the fat. A trans configuration, by contrast, means that the two carbons next to the double bond are oriented to opposite sides. As a result, they do not cause the chain to bend much, and their shape is Table 5.1 Fatty acid, vitamin E and cholesterol composition of some common dietary fats Saturated (%) Monounsaturated (%) Polyunsaturated (%) Cholesterol (mg 100 g 1 ) Animal fats Lard Butter Vegetable fats Coconut oil Palm oil Cottonseed oil Wheat germ oil Soya oil Olive oil Corn oil Sunflower oil Safflower oil Canola oil Source: Kays, S.J

170 Author s personal copy II. Traditional components 61 similar to straight saturated fatty acids. In plant sources, unsaturated fatty acids naturally occur in the cis form. Trans fatty acids might be present in some fats of animal origin, or might be the result of oil processing (e.g. hydrogenation of vegetable oils). The differences in geometry between the various types of unsaturated fatty acids, as well as between saturated and unsaturated fatty acids, play an important role in biological processes and in the construction of biological structures (such as cell membranes). Medical research suggests that amounts of trans fats correlate with circulatory diseases, such as atherosclerosis and coronary heart disease, more than the same amount of non- trans fats, for reasons that are not yet completely understood. E. Metabolizable carbohydrates After water, carbohydrates are the most abundant constituents in fruits and vegetables, representing 50% to 80% of the total dry weight. Carbohydrate functions include, among others, the storage of energy reserves and the make-up of much of the structural framework of cells. Simple carbohydrates, which are also the immediate products of photosynthesis, are important components of sensorial quality attributes. Carbohydrates, like proteins, yield 4 kcal g 1, while fats yield 9 kcal g 1. In many products, monosaccharides comprise a major portion of the total sugars. Glucose and fructose are the predominant forms of simple sugars found, especially, in fruits. Sucrose, the primary transport form of carbohydrate in most plants, is a disaccharide yielding glucose and fructose upon hydrolysis. Glucose, fructose and sucrose are water-soluble and together they comprise most of the sugars associated with the sweet taste of fruits and vegetables. The relative proportions of glucose and fructose vary from fruit to fruit and, to a lower extent, in the same fruit according to maturity. In many fruits (e.g. apple, pear, strawberry, grape) glucose and fructose are present in greater amounts than sucrose, but in certain vegetables, such as parsnip, beetroot, carrot, onion, sweet corn, pea and sweet potato, and in some ripe fruits such as banana, pineapple, peach and melon, the sucrose content is higher. Traces of other mono- and disaccharide sugars such as xylose, arabinose, mannose, galactose and maltose may also be present in small amounts ( Salunkhe et al., 1991 ). Some fruits of the Rosaceae family could also have significant levels of the sugar alcohol sorbitol. Total carbohydrate content also includes starches, which are organized into small grains, either within the chloroplasts or in some cases in specialized plastids (amyloplasts). Some non-starchy root vegetables, such as parsnip, beetroot and carrot, are relatively rich in simple sugars, containing between 8% and 18% of total carbohydrates. However, most vegetables contain smaller amounts of metabolizable carbohydrates. F. Dietary fiber Definition and composition Several definitions of fiber, either physiological or based on the measurement techniques used for its determination, have been put forward ( Slavin, 2005 ). An expert panel adopted the term dietary fiber consisting of non-digestible carbohydrates and lignin that are intrinsic and intact in plants ( Institute of Medicine, 2001 ).

171 Author s personal copy 62 Nutritional Quality of Fruits and Vegetables Dietary fiber includes very diverse macromolecules exhibiting a large variety of physico-chemical properties. The main components included as fiber are cellulose, hemicelluloses, pectins, lignin, resistant starch and non-digestible oligosaccharides. Cellulose is a cell wall polymer of β -1,4-linked glucose ( Brett and Waldron, 1996 ). Within the cell wall, the glucan chains are associated with hydrogen bonds to form assemblages highly resistant to degradation, known as microfibrils ( Carpita and McCann, 2000 ). In fruits and vegetables, the cell wall constitutes 1% to 2% of the fresh weight, and cellulose could be as much as 33% of that amount. In general, with the exception of avocado in which the whole cell wall seems to be degraded ( O Donoghue et al., 1994 ), little change in cellulose content occurs during ripening (Brummell, 2006 ). Hemicelluloses Several cell wall polymers soluble in alkalis are classified as hemicelluloses or cross-linking glycans ( Brummell and Harpster, 2001 ). Within the primary cell wall, hemicellulose levels are usually around 30% ( Carpita and McCann, 2000 ). The most common hemicellulose polymer in dicotyledonous species is known as xyloglucan, composed as cellulose of a backbone of β-1,4-linked glucose, but with lateral chains of the pentose xylose ( α-1,6 linked). These xylosyl residues can be modified further, with galactose, arabinose and/or fucose ( Brummell, 2006 ). Xylans are hemicellulosic compounds more abundant in monocotyledonous species, having a backbone of β -1,4-linked xylose which could be decorated with side chains of arabinose and/or glucuronic acid. Other hemicellulosic compounds usually less abundant include glucomannans, galactomannans and galactoglucomannans ( Carpita and McCann, 2000 ). Pectins Fruit tissues are particularly rich in pectins, which can account for up to 40% of the total cell wall polysaccharides. Pectins are also a diverse group of polymers rich in galacturonic acid ( Ridley et al., 2001 ). The most abundant pectic polysaccharide in the cell wall is homogalacturonan, a homopolymer of α-1, 4-linked galacturonic acid residues, with variable degrees of methyl esterification at C6 ( Willats et al., 2001 ). The degree of polymerization and the proportion of methyl esters affect the solubility of pectins. Pectins are deposited in the cell walls, with a high degree of esterification, and methyl ester usually decreases during ripening. Another modification commonly observed in several fruits during ripening is a reduction in pectin polymer size ( Brummell, 2006 ; Vicente et al., 2007b ). The extent of pectin depolymerization is variable, ranging from fruits such as avocado showing a dramatic downshift in polyuronide size ( Huber and O Donoghue, 1993 ) to products in which these changes are negligible, such as pepper or some berries ( Brummell, 2006 ; Vicente et al., 2007a ). Rhamnogalacturonan I (RG I) and rhamnogalacturonan II (RG II) are pectic polysaccharides which are also present in the plant cell wall. RG I has a backbone of alternating α -1,2-rhamnosyl and α-1,4-galacturonosyl residues ( Willats et al., 2001 ), with side chains rich in arabinose and galactose ( Carpita and McCann, 2000 ). Losses in the side chains are a common feature in fruit ripening, which can also affect pectin solubility and hydration potential ( Gross and Sams, 1984 ; Redgwell et al., 1997 ). RG II is the most complex polysaccharide present in the cell wall; it has the ability to form dimers via borate diester bonds ( O Neill et al., 2004 ; Kobayashi et al., 1996 ). Pectins, which are used in the commercial manufacture

172 Author s personal copy II. Traditional components 63 of jams and jellies, are extracted from certain fruits and vegetables such as citrus, apples and beets. Lignin is one of the most abundant biopolymers in nature ( Boerjan et al., 2003 ). It is an aromatic heteropolymer formed by the association of three hydroxycinnamyl alcohol derivatives (p-coumaryl, coniferyl and sinapyl alcohols) ( Reddy et al., 2005 ). Lignin is a highly resistant polymer present in secondary cell walls, and is associated with fibers and xylem vessels. In the case of fruits and vegetables, lignin content is relatively low. Resistant starch Starches are polysaccharides, composed of a number of glucose molecules linked together with α -D-(1-4) and/or α -D-(1-6) linkages ( Sajilata et al., 2006 ). Resistant starch consists of starch and its degradation products that are not digested in the small intestine ( Asp, 1994 ). Legumes are rich in resistant starch, and as much as 35% of their starch could escape digestion (Marlett and Longacre, 1996). Green bananas and potato are also relatively rich in resistant starch. Very little information is available about the resistant starch content of foods and the amount of resistant starch in a typical diet. Non-digestible oligosaccharides (NDOs) Oligosaccharides are low molecular weight carbohydrates intermediate in nature between simple sugars and polysaccharides ( Mussatto and Mancilha, 2007 ). While several oligosaccharides might be hydrolyzed in the digestive tract, others might resist the digestive process. Some of them include raffinose (trisaccharide composed of galactose, fructose, and glucose), stachyose (two galactose, one glucose and one fructose unit, linked sequentially) and verbascose (three galactose, one glucose and one fructose unit, linked sequentially). Legumes are rich in NDOs ( Mussatto and Mancilha, 2007 ). Benefits of fiber intake One of the most well known benefits of dietary fiber is the modulation of function of the intestinal tract ( Institute of Medicine, 2001 ). Meals rich in fiber promote satiety earlier, and are usually relatively low in calories compared to meals rich in other food types ( Marlett et al., 2002 ). Several works have also associated diets rich in dietary fiber with positive effects in disease prevention (see Institute of Medicine, 2001 ). Some works have established an inverse association between fiber intake and coronary disease ( Rimm et al., 1996 ; Wolk et al., 1999 ). Total fruit and vegetable consumption was inversely associated with colorectal cancer risk ( Terry et al., 2001 ). Current national dietary guidelines recommend an increased dietary fiber intake and suggest that fiber, independent of fat intake, is an important dietary component for the prevention of some diseases. Recommendations for adult dietary fiber intake generally fall in the range of 20 to 35 grams per day. The average fiber intake of adults in the US is less than half of this recommended level ( Marlett and Slavin, 1997 ). Sources of fiber Whole grains (especially the pericarp) and also fruits and vegetables are considered very good sources of fiber ( Anderson et al., 2007 ). In 2004, the primary contributors of fiber to the food supply were fruits and vegetables (37.1%), followed by grain products (36.0%) and legumes (13.3%) ( Hiza and Bente, 2007 ). Fiber content of

173 Author s personal copy 64 Nutritional Quality of Fruits and Vegetables Table 5.2 Fiber content in selected fruits, vegetables and nuts Product Dietary fiber (%) Almond 12.2 Apple 2.4 Asparagus 2.1 Avocado 6.8 Banana 2.6 Broccoli 2.6 Carrot 2.8 Kiwifruit 3.4 Lettuce 2.1 Onion 1.7 Orange 2.4 Pea 2.6 Peach 1.5 Peanut 8.5 Pear 3.1 Pepper 2.1 Pineapple 1.4 Plum 1.4 Potato 2.2 Prunes 7.1 Raisin 3.7 Spinach 2.2 Strawberry 2.0 Tomato 1.2 Walnut 6.7 Source: US Department of Agriculture, fruits and vegetables is usually in the range of 1% to 3% ( Table 5.2 ). Nuts, legumes and dried fruits have higher levels of fiber than fruits and vegetables. The nature of fiber varies among food sources. For instance, pectin is low in grains, but constitutes approximately 20% to 35% of the fiber in fruits, vegetables, legumes and nuts. Hemicelluloses account for about half of the total fiber in grains, and approximately 25% to 35% of the total fiber in other foods. Cellulose is one third or less of the total fiber in most foods ( Marlett, 1992 ). Besides total fiber content, some relevant properties include particle size and bulk volume, surface area characteristics, hydration and rheological properties, and adsorption or entrapment of minerals and organic molecules ( Guillon and Champ, 2000 ). The main modifications during storage of most fruits and vegetables occur because of changes in the solubility and molecular size of the cell wall constituents due to the action of several proteins ( Brummell, 2006 ; Fisher and Bennett, 1991 ). In some products, modification in fiber fractions could negatively affect quality. For instance, asparagus shows rapid hardening of the basal portions of the spears during storage related to modifications of fiber, such as

174 Author s personal copy II. Traditional components 65 increased deposition of lignin ( Saltveit, 1988 ). In general, preparation of fruits and vegetables by typical home methods or commercial processing does not seem to cause great loss of fiber ( Zyren et al., 1983 ). G. Vitamins Vitamins are organic molecules required in trace amounts for normal development, which cannot be synthesized in sufficient quantity by the organism and must be obtained from the diet. The term vitamin derives from the words vital amine because the first vitamin discovered (thiamine) contained an amino group. The 14 vitamins known today are vitamin A (retinol), B complex [B1 (thiamine), B2 (riboflavin), B3 (niacin), B5 (pantothenic acid), B6 (pyridoxine), B9 (folate/folic acid), biotin, choline and B12 (cyanocobalamine)] and vitamins C, D, E and K. They do not have common functions or structure and are usually grouped into fat-soluble (A, D, E and K) and water-soluble (B group and C) molecules. The vitamins present in fruits and vegetables make an important contribution to human nutrition, as they have specific functions in normal body performance. The vitamin content of fruits and vegetables shows a wide variation among species ( Salunkhe et al., 1991 ). Differences within cultivars occur, as well as between different batches of the same cultivar grown under different environmental and orchard conditions ( Rodriguez-Amaya, 2001 ; Lee and Kader, 2000 ). Vitamin A Carotenoids are liposoluble pigments responsible for the yellow, orange and red color of several fruits and vegetables. Carotenoids are terpenoids formed by eight isoprene units (2-methyl-1,3-butadiene) and derived from isopentenyl diphosphate. Those having an unsubstituted β -ring with an 11-carbon polyene chain have provitamin A activity ( Meléndez-Martínez et al., 2007 ), such as α-carotene, β-carotene and cryptoxanthin ( Kopsell and Kopsell, 2006 ). The structural requirement for vitamin A is satisfied by around 60 carotenoids ( Rodriguez-Amaya, 2001 ). Vitamin A plays an important role in vision, cell division and differentiation, bone development and reproduction. The average daily requirement for vitamin A for an adult is estimated at 5000 international units (1 IU 0.3 μg retinol or 0.6 μg β-carotene). Among this group there are, basically, two different classes: carotenes containing C and H (e.g. α-carotene, β -carotene, lycopene, etc.), and oxygenated derivatives known as xantophylls, such as lutein, violaxanthin and zeaxanthin. Carotenoids in plants have functions related to radiation interception, mainly in the blue green region of the spectrum, which may be transferred to the photosynthetic centers ( Kopsell and Kopsell, 2006 ). Moreover, these pigments protect the photosynthetic structures from excessive energy ( Grusak and Della Penna, 1999 ). They are usually present in low concentrations and their levels are highly variable among species. Fruits and vegetables account for only 30% of the vitamin A in the American diet ( Hiza and Bente, 2007 ). Vegetables that can supply useful amounts of carotene include carrots, pumpkins and squashes. Compared to vegetables, fruits are generally not as good a source of carotenoids, although there are a few notable exceptions such as apricot, mango, citrus, papaya

175 Author s personal copy 66 Nutritional Quality of Fruits and Vegetables Table 5.3 Carotene content (mean values) of selected fruits Product Carotene ( μ g 100 g 1 ) Mango 1800 Cantaloupe 1000 Pawpaw 810 Guava 435 Apricot 405 Plum 295 Watermelon 230 Source: Rodríguez-Amaya, and watermelon ( Table 5.3 ). Tomatoes and peppers also contain high levels of carotenoids. Their distribution is not usually uniform and in general, their accumulation is higher in the peel than in the pulp (Rodríguez-Amaya, 2001). To date, over 600 different carotenoids have been identified, but only a few of them are commonly found in produce. β -carotene, the most widely studied carotenoid, accumulates in carrots; lycopene is common in tomato and watermelon. Other pigments within this group include α-carotene, β -carotene, lutein, cryptoxanthin and zeaxanthin. In tomatoes, peaches and carrots the synthesis of carotene can continue after harvest. There is no difference between the carotene content of cooked vegetables and that of raw vegetables. Absorption of carotene can only be effective if the diet includes a minimum of 15% fat. The manner in which the food is prepared also determines the amount of carotene that will be absorbed. Homogenized carrots allow for the best absorption, followed by shredded carrots and whole carrots. Vitamin B complex Thiamine is required in the human body for the metabolism of carbohydrates. A daily intake of 1 2 mg is generally considered as necessary for a normal adult. Legumes are especially rich in thiamine. Compared with ascorbic acid, thiamine is relatively stable at cooking temperatures, especially in a slightly acidic solution. However, losses of 25% to 40% may occur during cooking. The average human requirement for riboflavin is estimated to be 1 2 mg per day. Green vegetables such as beans, beets, peppers and spinach are particularly rich in riboflavin. Starchy vegetables and fruits are relatively poor sources of riboflavin. Niacin, also known as nicotinic acid, is a precursor to NADH, NAD, NAD and NADP, which play essential roles in living organisms. A daily intake of 10 mg to 15 mg niacin is recommended. There is evidence that niacin can be synthesized in the body from tryptophan. Almonds are a rich source, but no fruits or vegetables can be singled out as being rich in niacin except perhaps, cape gooseberry and avocado. Niacin is relatively stable. Vitamin B 6 (pyridoxal phosphate) is a cofactor in many transamination, decarboxylation and deamination reactions (e.g. in plants, formation of ACC by ACC synthase requires pyridoxal phosphate as a cofactor) ( Ramalingam et al., 1985 ). Common

176 Author s personal copy II. Traditional components 67 symptoms of vitamin B 6 deficiency include dermatitis around the eyes, elbows and mouth, along with soreness of the mouth and a red tongue. It can also lead to dizziness, vomiting, weight loss and severe nervous disturbances ( Salunkhe et al., 1991 ). Vitamin B 6 is present in appreciable amounts in beans, cabbage, cauliflower, spinach, sweet potatoes, grapes, prunes, avocados and bananas. It is fairly heat stable. Pantothenic acid can be obtained from fresh, canned or frozen fruits and vegetables containing this vitamin if they are included in the diet. Pantothenic acid occurs widely in peas, beans, nuts, broccoli, mushrooms, potatoes and sweet potatoes. Symptoms of pantothenic acid deficiency in the diet include fatigue, headaches, sleep disturbance, tingling of hands and feet and lack of antibody production. Biotin is stable during cooking, processing and storage of fresh, canned and frozen fruits and vegetables. Deficiency leads to depression, sleeplessness and muscle pains. It is synthesized in the intestinal tract ( Salunkhe et al., 1991 ). Folic acid is essential for reproduction and normal growth. The vitamin is present in fruits, spinach, cabbage and other green vegetables. Lack of folic acid in the diet can cause a red tongue, diarrhea and anemia. Choline is heat-stable and occurs in dried legumes and vegetables. Choline deficiency in humans has never been reported. Vitamin B12 does not occur in fruits and vegetables. Because vitamins of the B group are water-soluble, leaching losses occur during cooking. Vitamin C Ascorbic acid (AsA) and its first oxidation product dehydroascorbic acid (which can be reduced in the human body) might be considered as vitamin C. AsA is a watersoluble carbohydrate-derived compound showing antioxidant and acidic properties due to the presence of a 2,3-enediol moiety ( Figure 5.1 ). Humans and a few other species are not able to synthesize AsA ( Chatterjee, 1973 ), because the gene coding for the last enzyme in the pathway (L-gulono-1,4-lactone oxidase) is nonfunctional ( Valpuesta and Botella, 2004 ). Plants synthesize AsA via a pathway that uses L-galactose as a precursor ( Smirnoff and Wheeler, 2000 ; Smirnoff, 2000 ). Another pathway using galacturonic acid, which might be recycled from cell wall pectin degradation, has been suggested in plants ( Agius et al., 2003 ). AsA has crucial biological functions in humans, such as its participation in collagen biosynthesis ( Murad et al., 1981 ). Even though nutritional deficiencies are rare in modern western cultures, it is generally recognized that dietary AsA also has important health benefits for the consumer, and an increased intake of vitamin C has been associated with a reduced incidence of some diseases and disorders ( Carr and Frei, 1999 ; Hancock HO HO O O HO OH Figure 5.1 Structure of ascorbic acid, a main antioxidant present in fruits and vegetables.

177 Author s personal copy 68 Nutritional Quality of Fruits and Vegetables and Viola, 2005 ). Furthermore, in meat-poor diets, dietary AsA can contribute to the improved uptake of iron ( Frossard et al., 2000 ). The recommended dietary allowance of vitamin C for men is 75 mg daily, while the recommended dietary allowance for young women is higher, at 90 mg daily ( Levine et al., 2001 ). Fruits, vegetables and juices are the main dietary sources of vitamin C. Fruits and vegetables account for 90% of the vitamin C in the US food supply ( Hiza and Bente, 2007 ). Its concentration depends on the product considered ( Noctor and Foyer, 1998 ), ranging from 1 to 150 mg 100 g 1 fresh weight (FW) ( Lee and Kader, 2000 ). Vitamin C is present in fresh fruits and vegetables, as well as in fruit juices. Fruits, particularly tropical species, and leafy vegetables are rich in ascorbic acid. Rosehip, jujube and guava have very high levels of ascorbic acid. Other good sources of AsA include persimmon, strawberry, kiwifruit, peppers, and citrus fruit, and spinach, broccoli and cabbage among vegetables ( Table 5.4 ). Wide variations in vitamin C content also exist within cultivars. For instance, AsA content in Actinidia deliciosa fruit varies from 29 to 80 mg 100 g 1 FW, depending on the cultivar ( Nishiyama et al., 2004 ). Even more dramatic variations were found in berry fruits, with levels of AsA ranging from 14 to 103 mg 100 g 1 FW among cultivars of raspberry, blackberry, red currant, gooseberry and cornelian cherry ( Pantelidis et al., 2007 ). For any given product, the levels of AsA are highly variable, depending on genetic and environmental factors (reviewed in Lee and Kader, 2000 ). A main environmental factor determining the level of ascorbic acid is radiation interception. In general, the greater the amount of sunlight received during growth, the higher the ascorbic acid content. The retention of AsA is also markedly affected by storage and processing. Potatoes lose up to 75% to 80% of the original levels over nine months of storage. In most cases, other fruit and vegetable AsA levels decline during storage, because the losses are accelerated by storage at high temperatures. Bruising and mechanical damage greatly increase the rate of ascorbic acid loss. Ascorbic acid is highly susceptible to oxidation, either directly or through the enzyme ascorbate oxidase catalyzing the oxidation of AsA to dehydroascorbic acid, with the concomitant reduction of molecular oxygen to water ( Sanmartin et al., 2007 ). Ascorbic acid can even be oxidized during eating, while food is being chewed. However, it is important to consider that the first breakdown product of AsA, Table 5.4 Vitamin C content (mean values) of selected fruits Product Vitamin C (mg 100 g 1 fresh weight) Guava, raw 184 Kiwi, raw 118 Litchi, raw 72 Pawpaw, raw 62 Strawberry, raw 57 Citrus fruits Cantaloupe 42 Source: Salunkhe et al., 1991.

178 Author s personal copy III. Antioxidants in fruits and vegetables 69 dehydroascorbic acid, still has vitamin C activity and all activity is lost if oxidation proceeds beyond this stage ( Salunkhe et al., 1991 ). When vegetables are cooked before eating, high losses of vitamin C can occur. For instance, starchy vegetables may lose between 40% and 80% of their vitamin C during cooking, because of leaching and oxidation. Loss of vitamin C can be reduced by steaming or by placing the vegetables directly into boiling water. Freezing reduces vitamin C slightly, but at the end of long-term frozen storage (12 months), a significant decrease (33% to 55%) in vitamin C can occur ( de Ancos et al., 2000 ). Vitamin E Vitamin E includes tocopherols and tocotrienols. They can be in eight different forms (four tocopherols and four tocotrienols). All the isomers have aromatic rings with a hydroxyl group that can donate hydrogen atoms to reduce reactive oxygen species (ROS). The different isomers are named alpha ( α ), beta ( β ), gamma ( γ ) and delta ( δ ), and this is related to the number and position of methyl groups in the ring. Each of the forms has its own vitamin E activity, α -tocopherol being the most active (see Figure 5.2 ). Vitamin E deficiency results in stunted growth. In general, vitamin E levels are more abundant in oily seeds, olives, nuts, peanuts, avocados and almonds. Even though the levels of tocopherol in broccoli and leafy vegetables are lower than in fat-rich products, they are good sources compared to other fruits and vegetables. Vitamin E is highly susceptible to oxidation during storage and processing. Vitamins D and K Vitamin D is a group of fat-soluble compounds. The main forms of vitamin D are ergocalciferol and cholecalciferol. It occurs only in trace amounts in fruits and vegetables. Vitamin K is essential for blood coagulation, but dietary deficiency is uncommon. The recommended daily intake is 120 μ g. It occurs abundantly in lettuce, spinach, cauliflower and cabbage. As well as direct intake, it can also be produced by bacteria in the intestines. III. Antioxidants in fruits and vegetables A. Oxidative damage and antioxidants Imbalance in the production of reactive oxygen species (ROS) leading to negative cellular alterations is known as oxidative damage, which is caused by several molecules HO O Figure 5.2 Structure of tocopherol.

179 Author s personal copy 70 Nutritional Quality of Fruits and Vegetables ( Mittler, 2002 ). Reactive oxygen species are partially reduced forms of oxygen such as singlet oxygen, hydrogen peroxide (H 2 O 2 ), superoxide (O 2 ) or hydroxyl radical (OH ) ( Asada, 1999 ). Some, but not all of the components able to cause oxidative damage are free radicals (i.e. molecules with unpaired electrons, which determine their high reactivity). Currently, there is overwhelming evidence showing that the ROS can alter proteins, lipids and nucleic acids, causing deleterious modifications to normal metabolism, which can lead to several disorders and diseases ( Waris and Ahsan, 2006 ), and eventually to cell death ( Jeremy et al., 2004 ). From a biological perspective, an antioxidant is considered as any compound able to oppose cellular oxidation. Diets rich in fruits and vegetables have been shown to reduce the incidence of cardiovascular disease and some chronic and degenerative diseases associated with oxidative damage ( Ames et al., 1993 ; Dragsted, 2003 ). The incorporation of fruits and vegetables in the diet may also help to eliminate certain toxins. The protective effects have been associated with the presence of antioxidant compounds ( Cao et al., 1996 ; Wang et al., 1996 ). Antioxidants are present in all plant organs and include ascorbic acid, carotenoids, vitamin E and phenolic compounds, among others (Larson, 1988 ) (Figure 5.3 ). Here we briefly describe some characteristics of these components. B. Ascorbic acid As mentioned before (see Section II.G) ascorbic acid is one of the most important compounds for human nutrition present in fruits and vegetables. The role of AsA in disease prevention has been associated with its capacity to neutralize ROS. Main antioxidants in fruits and vegetables Ascorbic acid Carotenoids Vitamin E Phenolics Others Carotenes Xanthophylls Tocopherols Tocotrienols Phenolic acids Benzoic acids Cinnamic acids Sulfur antiox. Flavonoids Flavonols Flavones Isoflavones Flavanols Flavanones Anthocyanidins Proanthocyanidins Others Coumarins Stilbenes Lignans and lignin Figure 5.3 Main dietary antioxidants present in fruits, vegetables and legumes.

180 Author s personal copy III. Antioxidants in fruits and vegetables 71 C. Carotenoids Fruits and vegetables are the main sources of carotenoids in the diet ( Rao and Rao, 2007 ). The presence of conjugated double bonds in carotenoids has a main role in determining their antioxidant properties ( Sandmann, 2001 ). In the last few years, carotenoids have received great attention due to their antioxidant properties and potential to prevent some diseases. The general properties of these compounds were described in Section II.G. D. Tocopherols and tocotrienols These include the fat-soluble compounds grouped as vitamin E, characterized by a high antioxidant capacity. Their distribution in fruits and vegetables was previously described (see Section II.G). E. Phenolic compounds This group encompasses a great diversity of compounds derived from the aromatic amino acids phenylalanine and tyrosine. Their main functions are acting as deterrents of potential predators or antimicrobials, protecting against UV-radiation and contributing to the pigmentation of fruits and flowers. Phenolic compounds can contribute to the astringency and bitter taste of some products. They are generally present in low concentrations, but in certain cases, such as in blueberry, they can reach levels of more than 0.1%. In general, they also accumulate in the peel more than in the pulp of fruits. The general characteristic of the compounds within this group is to have aromatic rings with variable degrees of hydroxylation ( Mattila et al., 2006 ). Phenolic compounds are easily oxidized to quinones. The beneficial properties of berry fruits on human health have been associated in part with the presence of relatively high levels of phenolic compounds ( Seeram et al., 2006 ). There is in vitro evidence showing that these compounds could influence several cellular processes. Information regarding the metabolism and effect in vivo is much more limited ( Duthie et al., 2003 ). A large number of phenolic compounds have been identified in plants ( Tsao and Deng, 2004 ). They have been subdivided into different subclasses, such as phenolic acids, flavonoids and other compounds (e.g. lignans, stilbenes, tannins, coumarins and lignin). Phenolic acids Phenolic acids include derivatives of benzoic and cinnamic acid ( Benbrook, 2005 ) ( Figure 5.4 ). The most common benzoic acid derivatives are p-hydroxybenzoic, vanillic, syringic and gallic acid, while common cinnamic acid derivatives include p- coumaric, caffeic, ferulic and sinapic acid. The derivatives differ in the degree of hydroxylation and methoxylation of the aromatic ring. Caffeic acid is the most abundant phenolic acid in several fruits such as berries ( Mattila et al., 2006 ), while coumaric acid is usually present in lower proportions ( Rice-Evans et al., 1997 ). Ferulic acid represents 90% of total phenolic acids in cereals ( Manach et al., 2004 ; Scalbert and Williamson, 2000 ). The contribution of each of the phenolic compounds to the

181 Author s personal copy 72 Nutritional Quality of Fruits and Vegetables O O OH OH Figure 5.4 Structure of benzoic acid (left) and cinnamic acid (right), precursors of the two main classes of phenolic acids present in fruits and vegetables. O O OH O O Figure 5.5 General structure of flavones (left) and flavonols (right). antioxidant capacity depends on their structure. For instance, the number of hydroxyls present in the molecule can increase the antioxidant capacity. Flavonoids Flavonoids represent a large group of phenolic compounds with two aromatic rings in their structure that are associated together by a 3C-oxygenated heterocycle. Phenolic compounds are usually present as glycosides, which reduce their activity against free radicals and increase their solubility. At the cellular level, they are compartmentalized in the vacuoles ( Rice-Evans et al., 1997 ). There are different classes of flavonoids ( Le Marchand, 2002 ) such as: a) fl avones and flavanols; b) fl avanones, flavanols; c) isoflavones; d) proanthocyanidins; and e) anthocyanidins. Flavones and flavonols Flavonols have a central ring of 3-hydroxypyran-4-one ( Rice-Evans et al., 1997 ). Flavones lack the OH in position 3 ( Figure 5.5 ). Rutin, luteolin and apigenin are common among flavones, while the most abundant flavonols are quercetin and kampferol ( Manach et al., 2004 ). Onions are rich in these compounds. Blueberries also have high levels, especially in the peel, because synthesis is stimulated by exposure to light. Celery is a good source of flavones. Flavones are also present in citrus, but they are associated mainly with the fruit peel. Flavanones and flavanols Flavanones do not have the double bond in position 2,3 of the central ring, while flavanols lack the carbonyl group at position 4 ( Figure 5.6 ). The genus Citrus is characterized by the accumulation of flavanone glycosides.

182 Author s personal copy III. Antioxidants in fruits and vegetables 73 O O OH O Figure 5.6 General structure of flavanones (left) and flavanols (right). Orange juice is a source of the flavanone glycoside hesperidin ( Tripoli et al., 2007 ). The flavanols catechin and epicatechin are common in grapes ( Rice-Evans et al., 1997 ). Isoflavones Isoflavones are phytoestrogens present in legumes. Soybean products are a good source of these compounds ( Manach et al., 2004 ). The three most commonly found isoflavones are genistein, glycitein and daidzein. Proanthocyanidins Proanthocyanidins are oligomeric flavonoids (usually dimers or oligomers of the flavanols catechin and epicatechin). They are common in the peel and seeds of grapes ( Gu et al., 2004 ). Other sources of these compounds include apple, almond and blueberry. Anthocyanidins Anthocyanidins are pigments giving several fruits their characteristic red or purple colors, although in some conditions they can be uncolored. Besides being pigments, anthocyanidins have great relevance due to their contribution to the antioxidant capacity of fruits and vegetables. The basic structure of anthocyanidins is derived from the flavilium cation (2-phenyl-benzopyril). There are six anthocyanidins more common in fruits and vegetables: pelargonidin, cyanidin, delphynidin, peonidyn, petunydin and malvidin. The differences between them are the OH, H and OCH 3 groups associated with the phenolic rings. The distribution of hydroxyls in the molecule influences the antioxidant capacity of the different anthocyanidins. These compounds are usually present as glycosides associated with different sugars, since anthocyanidin glycosylation reduces antioxidant capacity relative to the free aglycons. Others Lignans are diphenolic structures formed by the association of two derivatives of cinnamic acid ( Liu, 2007 ). They are present mainly in linseeds, cereals and legumes, but their levels are low in fruits and vegetables. Stilbenes are also phenolic compounds described in fruits. The most studied compound in this group is resveratrol ( Figure 5.7 ). This compound has been known for quite a while, and is commonly produced in response to pathogens and other stress conditions in grapes ( Langcake and Pryce, 1976 ). It has also been identified in other fruits, such as blueberry. It has been suggested that it may have anticarcinogenic properties. Finally, lignin is a phenolic polymer present in secondary cell walls of plant tissues. It is highly hydrophobic and is formed by three main monomeric precursors: coumarylic, sinapylic and coniferyl alcohols. It is associated with conduction tissues

183 Author s personal copy 74 Nutritional Quality of Fruits and Vegetables HO OH OH Figure 5.7 Resveratrol has been studied in detail in grapes. It has been suggested that this compound has anticarcinogenic properties. Factors affecting the levels of antioxidants in fruits Genetic Environmental Preharvest Harvest Postharvest Species Variety Radiation Stress during development (water, fertility, pathogens, etc.) Maturity Handling Storage Postharvest treatments (UVC, ozone) Processing Figure 5.8 Main factors affecting the level of antioxidants in fruits. (xylem vessels, sclereids, tracheids), and in general it is not abundant in fruits and vegetables. Its contribution, from the antioxidant point of view, is associated only with the products of its potential degradation that are very limited. F. Factors affecting the levels of antioxidants in fruits and vegetables Several factors influence the accumulation and degradation of antioxidant compounds in fruits. In general terms, these variables could be divided into genetic and environmental. Different factors are included within each of these groups ( Figure 5.8 ). Genetic factors Species The species is the first factor determining the prevalence of different antioxidants. Although there are some exceptions, each group is characterized by the accumulation of certain types of antioxidants ( Table 5.5 ). Berries are particularly rich in phenolic compounds ( Zheng and Wang, 2003 ) and vitamin C ( Kevers et al., 2007 ). The main antioxidants in this group seem to be phenolics because, in general, a good correlation between total antioxidant capacity and phenolic compounds has been

184 Author s personal copy III. Antioxidants in fruits and vegetables 75 Table 5.5 Fruits and vegetables rich in the different groups of antioxidants Ascorbic acid Vitamin E Carotenoids Phenolics Strawberry Almond Pineapple Blueberry Pepper Corn Plum Plum Kiwifruit Broccoli Peach Raspberry Orange Spinach Pepper Strawberry Pepper Peanut Mango Apple Broccoli Avocado Melon Blackberry Guava Tomato Rosehip Carrot Persimmon observed. In the case of ripe blueberry, ascorbic acid only contributes 0.4% to 9.0% to the total antioxidant capacity ( Kalt et al., 1999 ). Cultivar For a given species, the levels of antioxidants are also markedly affected by the cultivar considered. For instance, in strawberry, Nelson and co-workers (1972) found variations from 19 to 71 mg of ascorbic acid per 100 g FW in six varieties. Similar differences among varieties have been found for phenolic compounds ( Wang and Lin, 2000 ). The identification of lines or mutants enriched in antioxidants might be useful in breeding programs aimed at improving the nutritional value of fruits and vegetables. The identification of the nature of the genes mutated in lines with altered accumulation of antioxidants might also be of great value. For instance, in the case of tomato the characterization of the high pigment ( hp ) mutants, which accumulate high levels of carotenoids, showed that the mutated gene is associated with plant light responses, and the over-expression of this gene resulted in increased accumulation of carotenoids ( Liu et al., 2004 ). Also, in tomato the level of β -carotene and lycopene were raised by increasing the expression of phytoene synthase and lycopene cyclase, respectively ( Fraser et al., 2002 ; D Ambrosio et al., 2004 ). Similarly in carrot, the over-expression of a β -carotene ketolase isolated from Haematococcus pluvialis led to the accumulation of the ketocarotenoid astaxanthin ( Jayaraj et al., 2008 ). The generation of transgenic plants has also been seen to increase the levels of other antioxidants such as phenolic compounds. Transformation of tomato with a Petunia gene for chalcone isomerase increased the concentration of flavonols in the peel almost 80 times, without altering other phenotypic characteristics ( Muir et al., 2001 ). In the case of ascorbic acid, the elucidation of its biosynthetic pathway opened the way to manipulate ascorbate biosynthesis in plants ( Smirnoff, 2000 ). However, while most of the genes proposed to be involved in these pathways have been cloned and expressed in various plant species, transformation strategies to increase AsA concentrations have had only limited success. Thus, there is a need for alternative approaches to identify the genetic determinants underlying whole plant AsA homeostasis. Environmental factors Radiation In many cases, modifications in the level of phenolic compounds, ascorbic acid and carotenoids have been associated with changes in the radiation interception

185 Author s personal copy 76 Nutritional Quality of Fruits and Vegetables in the field. Sun-exposed sides of fruits have higher levels of phenolics and vitamin C than shaded regions ( Lee and Kader, 2000 ). In the case of leafy vegetables, the levels of flavonols are 10 times higher in the surface leaves than in the internal leaves. In tomato, the level of total phenolics increased twofold in plants exposed to higher irradiance. Similarly, these plants presented higher levels of carotenoids and ascorbic acid ( Gautier et al., 2008 ). This illustrates that maximization of radiation interception is important to obtain products with higher accumulation of antioxidants. However, the optimal irradiance levels required to maximize accumulation of the different groups of antioxidants in fruits and vegetables are not well established. Cultural practices There are several works analyzing the effect of cultural practices on the level of different groups of antioxidants. For instance, strawberry fruit grown with plastic mulch had higher antioxidant capacity than fruits from plants grown in beds without plastic mulch ( Wang et al., 2002 ). High nitrogen fertilization has been associated with reduced levels of ascorbic acid ( Lee and Kader, 2000 ), and compost as a soil supplement significantly enhanced levels of ascorbic acid ( Wang and Lin, 2003 ). Vitamin C accumulation also has been inversely correlated with rainfall ( Toivonen et al., 1994 ). Some authors have found evidence suggesting that organic products might accumulate higher levels of antioxidants and vitamins than those produced conventionally ( Woese et al., 1997 ; Weibel et al., 2000 ; Asami et al., 2003 ; Chassy et al., 2006 ). However, there are also studies that show either results that are opposite, or results that show no difference ( Barrett et al., 2007 ). Winter and Davis (2006) concluded that it is not possible to ensure that, from a nutritional point of view, organically grown products are superior to those obtained by conventional agricultural techniques. Maturity at harvest The developmental stage might affect the antioxidant capacity of fruits ( Prior et al., 1998 ). The nature of these changes depends on the product considered. For instance, in tomato and pepper total antioxidant capacity increases because of the accumulation of carotenoids and vitamin C. In the case of blueberry fruit the concentration of phenolic acids decreases during ripening, while anthocyanins are accumulated ( Castrejón et al., 2008 ), resulting in a net reduction of total antioxidant capacity during development. Similar patterns have been observed in strawberry and blackberry ( Wang and Lin, 2000 ). In the case of carotenoids, in some products (e.g. pepper, tomato, mango) the concentration increases during development ( de Azevedo and Rodriguez-Amaya, 2005 ). In contrast, products in which color is mainly associated with the accumulation of anthocyanins or products that maintain their green color at harvest usually show a reduction in the level of carotenoids as development progresses ( Rodriguez-Amaya, 2001 ). Wounding Mechanical damage may cause alterations in the levels of antioxidants. In the case of AsA, cellular breakage causes an increase in the levels of the internal pressure of oxygen favoring oxidation. Carotenoid degradation is also accelerated by oxygen, but the stability of these compounds is higher than that of AsA. In the case of phenolic compounds, wounding could alter both their synthesis and degradation ( Tomás-Barberán et al., 1997 ; Loaiza Velarde et al., 1997). In lettuce, wounding led to the accumulation of soluble phenolic compounds (e.g. chlorogenic acid) ( Choi et al., 2005 ). From a molecular perspective, wounding has been shown to induce de novo

186 Author s personal copy III. Antioxidants in fruits and vegetables 77 synthesis of phenylalanine ammonia lyase, a key enzyme in phenylpropanoid metabolism ( Choi et al., 2005 ). Besides its role on phenolic biosynthesis, wounding also affects degradation. First, also in response to wounding, an increase in enzymes associated with phenolics oxidation such as polyphenol oxidases (PPOs) and peroxidases (PODs) has been reported. In addition, cell disruption allows direct contact between pre-existing phenolic degrading enzymes. Finally, the production of hydrogen peroxide upon damage provides a secondary substrate of PODs and the reduction of barriers for oxygen diffusion might favor PPO activity. This might promote the oxidation of phenolics, which can then polymerize, leading to the formation of brown colored pigments that may ultimately reduce quality. Consequently, careful handling and minimization of physical damage is recommended. Storage The effect of storage on antioxidants in many cases is related to the role of ethylene in the ripening process. Consequently, the final effect on antioxidant (AOX) levels will depend on the typical modifications observed during development of the species considered. In some cases, ethylene can induce specific antioxidants. For instance, in carrot, ethylene stimulated the accumulation of an isocoumarin (6-methoxymellein). In berries it has been observed that atmospheres with high levels of oxygen (60% and 100%) result in increased antioxidant capacity by favoring anthocyanins and other phenolics accumulation ( Zheng et al., 2003 ). However, the oxidation of AsA might also be favored in these conditions. Besides the effect of any specific group of antioxidants in most fruits, it has been observed that the changes in total antioxidant capacity are not dramatic during postharvest storage. Excluding some products, such as broccoli and banana, fruits and vegetables, in general, lose their visual quality before marked losses in total antioxidants occur ( Kevers et al., 2007 ). In some cases, an increase in total AOX capacity is observed, basically, associated with the accumulation of phenolics. In strawberry, storage at 5 C and 10 C increased the antiradical capacity ( Ayala-Zavala et al., 2004 ). Further studies to evaluate the extent of this increased accumulation of antioxidants in some fruits might be done to determine the potential for increasing the functionality of fruits and vegetables through manipulation of the postharvest environment ( Kalt et al., 1999 ). Other treatments Some studies suggest that manipulation of the metabolism of products by the application of postharvest treatments could be useful to increase the antioxidant capacity, with consequent nutritional benefit ( Kalt et al., 1999 ). Phenolic compounds synthesis might be triggered in response to stress conditions, such as infection by microorganisms or wounding, ultraviolet (UV) irradiation or the exposure of the products to ozone-enriched atmospheres. In grapes, postharvest UV-C and ozone treatments increased the accumulation of resveratrol ( Cantos et al., 2001 ; Versari et al., 2001 ; Gonzalez-Barrio et al., 2006). The elicitation of the accumulation of antioxidant compounds has also been observed in other fruits. In blueberry cv. Bluecrop, besides reducing decay, UV-C radiation exposure (2 or 4 kj/m 2 ) resulted in increased accumulation of anthocyanins and higher levels of antioxidants ( Perkins- Veazie et al., 2008 ). In the case of strawberry, UV-C treatments also increased the level of phenolic compounds and the antiradical capacity ( Ayala-Zavala et al., 2004 ). These results, at a laboratory scale, show an interesting eliciting effect of some

187 Author s personal copy 78 Nutritional Quality of Fruits and Vegetables postharvest treatments on antioxidant accumulation. Further studies would be useful to determine the potential of these strategies on a commercial scale. Processing The effect of processing on the level and bioavailability of antioxidants depends on the treatment intensity, as well as on the component considered ( Bernhardt and Schlich, 2006 ). In some cases, processing could lead to higher availability of antioxidants, due to an increase in the ease of extractability. For instance, with carrot and spinach carotenoids vapor cooking increases assimilation, probably due to a disruption of carotenoid-protein complexes. Similarly, the bioavailability of lycopene increases in heat-treated tomato. However, cooking could cause the isomerization of β -carotene, leading to the formation of cis isomers with lower provitamin A activity ( Deming et al., 2002a,b ). For instance, in the case of fresh carrots, 100% of the β -carotene is present in the trans form, while canning results in a significant formation of cis isomers. Carotenoids are in general susceptible to oxidation. Heat, light and oxygen could accelerate their degradation ( von Elbe and Schwartz, 1996 ). Minimizing the influence of these factors could reduce carotenoids loss. Ascorbic acid is one of the antioxidants more susceptible to degradation. Blanching or even freezing and thawing could cause losses up to 25%. More drastic treatments could lead to losses of up to 90% of AsA. Some of the factors affecting the loss of AsA include the degree of heating, the exposed surface (which affects lixiviation in the cooking media), oxygen levels and product ph ( Eitenmiller and Landen, 1999 ). The stability of AsA could be increased at low ph, reduced oxygen pressure, darkness and presence of chelating agents. Consumption in the fresh state is the best way to minimize AsA losses. Finally, processing can also cause losses of phenolic antioxidants. For instance, peeling or cutting reduces quercetin levels by only 1%, but cooking in water may reduce the content of this component by 75%. IV. Fruits and vegetables as direct sources of minerals Dietary minerals raise concern for health specialists and consumers, due to the number of processes they are involved in and the continuous research highlighting the benefits of their adequate and balanced intake. Although there is no universally accepted definition or classification, the dietary focus on minerals derives from an interest in supporting the biosynthetic apparatus with required elemental components other than carbon, hydrogen and oxygen. Total mineral content is determined by the ash value. Nevertheless, classification of many elements as essential minerals for human nutrition is not definitive, and there is still debate as to the natural biological role of vanadium, chromium, boron, aluminum and silicon in human health. Minerals are normally classified as macroor micronutrients, based on the relative concentration of each nutrient when those concentrations are adequate for normal tissue function. Macronutrients include potassium (K), calcium (Ca), magnesium (Mg), nitrogen (N), and phosphorus (P), and their concentrations in plant tissues range from 1000 to μ g per gram of dry weight. In contrast, the concentrations of micronutrients usually found in plant

188 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 79 tissues are 100- to fold lower than those of macronutrients. Mineral micronutrients considered essential in human nutrition include manganese (Mn), copper (Cu), iron (Fe), zinc (Zn), cobalt (Co), sodium (Na), chlorine (Cl), iodine (I), fluorine (F), sulfur (S), and selenium (Se). Macronutrients can also be classified into those that maintain their identity as ions within plant tissues (e.g. K, Ca2 and Mg 2 ), and those that are assimilated into organic compounds (e.g. N and P). In general, vegetables are a richer source of minerals than fruits, but both vegetables and fruits are considered nutrient-dense foods in that they provide substantial amounts of micronutrients, such as minerals and vitamins, but relatively few calories (Table 5.6 ). Minerals have both direct and indirect effects on human health. The direct effects of minerals focus on the consequences of their consumption on human nutrition, while the indirect effects refer to their incidence in fruit and vegetable quality and subsequent consumer acceptance. From a direct nutrition standpoint, potassium has the biggest presence in both fruits and vegetables, but nitrogen and calcium show major impacts on horticultural crop quality. Table 5.6 Fruit and vegetable sources of potassium, ranked by milligrams of potassium per standard amount, also showing calories in the standard amount *. The dietary reference intake (DRI) for potassium for adults and adolescents is 4700 mg/day. Fruits and vegetables, standard amount Potassium (mg) Calories Sweetpotato, baked, 1 potato (146 g) Tomato paste, ¼ cup Beet greens, cooked, ½ cup Potato, baked, flesh, 1 potato (156 g) White beans, canned, ½ cup Tomato puree, ½ cup Prune juice, ¾ cup Carrot juice, ¾ cup Lima beans, cooked, ½ cup Winter squash, cooked, ½ cup Banana, 1 medium Spinach, cooked, ½ cup Tomato juice, ¾ cup Tomato sauce, ½ cup Peaches, dried, uncooked, ¼ cup Prunes, stewed, ½ cup Apricots, dried, uncooked, ¼ cup Cantaloupe, ¼ medium Honeydew melon, 1/ 8 medium Plantains, cooked, ½ cup slices Kidney beans, cooked, ½ cup Orange juice, ¾ cup Split peas, cooked, ½ cup * US Department of Health and Human Services and US Department of Agriculture, 2005.

189 Author s personal copy 80 Nutritional Quality of Fruits and Vegetables Until recently, nutrition research focused on single-mineral impact on human health, generally with incongruent results. The recognition that minerals are not consumed individually, but as combined constituents of a varied diet, has shifted the efforts in this area to unraveling the role of the overall diet, or dietary patterns, in blood pressure and cardiovascular diseases, bone diseases and a range of chronic disorders. Epidemiological surveys suggest that the total diet has a greater influence on health than do specific components. From these dietary pattern studies, it has become increasingly clear that it is not merely the excess or deficiency of a single mineral, but also deficiencies of multiple nutrients in combination that have the greatest dietary effects on health. Adequate intake of minerals such as potassium specifically derived from foods such as horticultural crops, where they coexist with other essential nutrients contributes to overall health. As described in previous sections, fruits and vegetables provide a milieu of phytochemicals, non-nutritive substances that possess health protective benefits. In contrast, fruits and vegetables may not usually be recognized as primary sources of mineral intakes from a nutritional point of view ( Fairweather-Tait and Hurrell, 1996 ). Nevertheless, the Dietary Approaches to Stop Hypertension (DASH) emphasize fruit, vegetable and low-fat dairy product consumption as a source of minerals. In the DASH dietary pattern, vegetables contribute an average of 14.3%, 15.5%, 16.2% and 10.4% to the intakes of calcium, magnesium, potassium and zinc, respectively ( Lin et al., 2003 ). Correspondingly, fruits and juices contribute an average of 5.8%, 17.3%, 33.0% and 6.6% ( Lin et al., 2003 ). There has been a natural trend towards lower mineral content in fruits and vegetables over the past decades ( Mayer, 1997 ; Ekholm et al., 2007 ) which have not been fully compensated for by the increase in fruit and vegetable consumption. Vegetable contribution of potassium, phosphorus, magnesium, calcium, copper, iron and zinc to the US food supply significantly decreased during the last century, while fruit contribution of potassium, phosphorus, magnesium and copper increased ( Table 5.7 ). Table 5.7 Minerals (%) contributed from fruits and vegetables to the US food supply in selected years Fruit Vegetables Mineral Year/s Year/s Potassium Calcium Phosphorus Magnesium Copper Iron Zinc Sodium Selenium Source: Hiza and Bente, 2007.

190 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 81 Nowadays different postharvest strategies for improving the mineral intake from fruits and vegetables are being implemented. These comprise increasing consumption of fruits and vegetables and increasing levels of essential nutrients through fortification methods. Alternative approaches include improving nutrient bioavailability and retention. A. General considerations of selected minerals Potassium (K) A potassium-rich diet contributes to lower blood pressure, blunting the effects of salt ( Salunkhe et al., 1991 ). Inadequate levels of potassium intake have long been associated with higher blood pressure ( McCarron and Reusser, 2001 ). Potassium also regulates heartbeat, assists in muscle contraction and is needed to send nerve impulses and to release energy from fat, carbohydrates and protein. Different nutrients and phytochemicals in fruits and vegetables, including potassium, may be independently or jointly responsible for an apparent reduction in cardiovascular disease risk ( Ignarro et al., 2007 ). Potassium is a systemic electrolyte and is essential in coregulating ATP with sodium. Potassium favorably affects acid base metabolism, which may reduce the risk of developing kidney stones ( Zerwekh et al., 2007 ), and possibly decrease bone loss with age. Although calcium intake is an important determinant in peak bone mass, and in retarding bone loss in postmenopausal women, findings of higher bone mass and lower bone resorption in women consuming high intakes of potassium, magnesium, zinc and vitamin C emphasizes the importance of considering the impact of variation in other nutrients when focusing on a particular mineral ( Cohen and Roe, 2000 ). In fact, up to 11 different groups of compounds (vitamins, minerals, antioxidants and others) in fruits and vegetables could influence bone health ( MacDonald, 2007 ). Potassium is the most abundant individual mineral element in fruits and vegetables. It normally varies between 60 and 600 mg per 100 g 1 of fresh tissue. It plays a role in a myriad of cellular and whole plant functions: it serves as an osmoticum for cellular growth and stomatal function, balancing the charges of anions, activating almost 60 plant enzymes and participating in numerous metabolic processes, including protein synthesis, oxidative metabolism and photosynthesis. In fruits and vegetables, potassium occurs mainly in combination with various organic acids. Examples of potassium-rich fruits and vegetables include bananas and plantains, leafy green vegetables, many dried fruits, oranges and orange juice, cantaloupes and honeydew melons, tomatoes and root vegetables ( Table 5.7 ). Calcium (Ca) Calcium is essential for bone and tooth formation. Because of this, calcium requirements are higher during adolescence. Calcium is also very important during later adulthood, and of great consequence from a public health perspective, because inadequate intake of calcium may increase the risk of osteoporosis, a condition in which decreased bone mass weakens bone ( Nordin, 1997 ; Cohen and Roe, 2000 ). With nearly half of all American women over 50 years of age demonstrating low

191 Author s personal copy 82 Nutritional Quality of Fruits and Vegetables mineral bone density or osteoporosis, and an estimated 1.3 million osteoporosis-related fractures occurring each year in the US, with a billion dollar estimated annual cost ( DeBar et al., 2004 ), osteoporosis prevention is a major public health target. Calcium fluxes are important mediators of hormonal effects on target organs through the phosphoinositol system, and are closely linked with the cyclic AMP systems. There is also evidence linking hypertension with calcium deficiency ( Appel et al., 1997 ; McCarron and Reusser, 2001 ). In plants, calcium is primarily associated with the pectic materials. It is believed to have a major influence on the rheological properties of the cell wall and, consequently, on the texture and storage life of fruits and vegetables. Ca 2 can interact with the anionic pectic polysaccharides, coordinating with the oxygen functions of two adjacent pectin chains to form the so-called eggbox structure, and cross-linking the chains ( Rose et al., 2003 ). Intracellular Ca 2 also occupies a pivotal role in cell signal transduction ( Sanders et al., 1999 ). The plant signals thought to be transduced through cytosolic Ca 2 include wounding, temperature stress, fungal elicitors, oxidative stress, anaerobiosis, abscisic acid, osmotic stress, red or blue light and mineral nutrition. Intracellular Ca 2 transient increases are often associated with initiation of responses. Thus, Ca 2 is a prominent second messenger, and it must be maintained in the cytoplasm at concentrations many orders of magnitude lower than the Ca 2 in the cell wall. Horticultural crops are considered a secondary source of calcium in comparison to dairy products but, taken as a whole, fruits and vegetables account for almost 10% of the calcium in the US food supply (Table 7, Cook and Friday, 2003 ). Dark green leafy cabbage family vegetables and turnip greens are good calcium sources and most green leafy vegetables are potential calcium sources because of their absorbable calcium content ( Jodral-Segado et al., 2003 ; Titchenal and Dobbs, 2007 ). Projects designed to test the efficacy of a health plan-based lifestyle intervention for increasing bone mineral density propose not only to increase the consumption of high calcium foods, but also of fruits and vegetables ( DeBar et al., 2004 ). Magnesium (Mg) Magnesium is important in protein synthesis, release of energy from muscle storage and body temperature regulation. It is critical for proper heart function and plays a role in bone formation, as previously described. Magnesium activates over 100 enzymes. In plants, magnesium is a constituent of the chlorophyll molecule: the porphyrinlike ring structure of chlorophylls contains a central magnesium atom coordinated to the four pyrrole rings. On the other hand, magnesium is involved in the energetic metabolism as a constituent of the Mg-ATP or Mg-ADP complex. Also, the Calvin cycle the pathway that produces a three-carbon compound as the first stable product in the multistep conversion of CO 2 into carbohydrates is partially regulated via stromal Mg 2 concentration. This nutrient also serves important biochemical functions in protein synthesis ( Mengel and Kirkby, 1982 ). In 2004, vegetable contribution to the total magnesium in the US food supply was an average of 14% ( Table 5.7 ). Using current population standards, magnesium

192 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 83 intake was found to be below adequate levels for both adults and children ( Sigman- Grant et al., 2003 ). Mixed users, who are more likely to consume higher intakes of grains, fruit and milk products, were found to have higher magnesium densities than high-fat users, who consume significantly more servings of meat and higher levels of discretionary fat ( Sigman-Grant et al., 2003 ). Generally, magnesium levels are significantly higher in vegetables than in fruits, but nuts are good sources of this nutrient. Dry fruits and legumes are the food groups that rank higher in magnesium content ( Jodral-Segado et al., 2003 ). Phosphorus (P) Inorganic phosphate is essential for skeletal mineralization and for multiple cellular functions, including glycolysis, gluconeogenesis, DNA synthesis, RNA synthesis, cellular protein phosphorylation, phospholipid synthesis and intracellular regulatory roles ( DiMeglio et al., 2000 ). Phosphorus is a primary bone-forming mineral. In western countries, isolated dietary phosphate deficiency is exceedingly rare, because most westerners eat high-phosphate diets, except for occasional metabolic disorders such as hyperphosphatemia ( DiMeglio et al., 2000 ). Phosphorus can exist in plants as both inorganic phosphate anions and organophosphate compounds ( Raghothama, 1999 ). Unlike sulfate and nitrate, phosphate is not reduced in plants during assimilation, but remains in its oxidized state forming phosphate esters in a wide variety of organic compounds. Inorganic phosphorus constitutes a main structural component of nucleic acids and phospholipids, and plays a critical role in energy conversion in the form of high-energy phosphoester and diphosphate bonds. It is important both as a substrate and as a regulatory factor in oxidative metabolism and photosynthesis, it participates in signal transduction, and regulates the activities of an assortment of proteins by way of covalent phosphorylation/ dephosphorylation reactions. In 2004, the primary contributor of phosphorus to the food supply was the dairy group (31.3%), followed by the meat, poultry and fish group (24.9%) and grain products (19.4%) ( Hiza and Bente, 2007 ). Fruit and vegetable contribution to the total phosphorus in the US food supply was an average of 9.5% ( Table 5.7 ). Among tree fruits, nuts are natural sources of phosphorus. Nitrogen (N) The largest requirement for nitrogen in eukaryotic organisms is the biosynthesis of amino acids, building blocks of proteins and precursors of many other compounds. Proteins represent a large percentage of the human body and carry out many different cell functions. Therefore, protein synthesis is central to cell growth, differentiation, and reproduction. Nitrogen is also an essential component of nucleic acids, cofactors and other metabolites. Several plant hormones (indole-3-acetic acid, zeatine, spermidine, etc.) contain nitrogen, or are derived from nitrogenous precursors. Alkaloids and other secondary compounds contain nitrogen, and various phenolics derive from phenylalanine and are therefore linked with amino acid metabolism. Moreover, nitrogen is a major constituent of chlorophyll. The characteristic preharvest yellow color of

193 Author s personal copy 84 Nutritional Quality of Fruits and Vegetables nitrogen-starved vegetables a physiological disorder called chlorosis reflects their inability to synthesize adequate amounts of green chlorophyll under nitrogen-limited conditions. Sulfur (S) Sulfur is an essential nutrient required for growth, primarily used to synthesize cysteine and methionine. The sulfur-containing amino acids play pivotal roles in the structural and catalytic functions of proteins. Cysteines are important because oxidizing the thiol groups of two cysteine residues can form disulfide bonds, important covalent linkages involved in establishing tertiary and, in some cases, quaternary protein structures. The dithiol disulfide interchange can be a regulatory mechanism, as well as a mediator of redox reactions. Sulfur is also a component of numerous essential and secondary metabolites derived from these amino acids. Sulfur nutrition is important in the species within the order Brassicales (e.g. white cabbage, broccoli, cauliflower, capers) for the synthesis of anticarcinogenic glucosinolate compounds (reviewed in Sozzi, 2001 ). In caper ( Capparis spinosa L.) flavor, 160 components were identified, including elemental sulfur (S 8 ) and more than 40 sulfur-containing compounds, among them thiocyanates and isothiocyanates. Although of key importance in human and plant life, sulfur is a relatively minor component in comparison with nitrogen. Generally, it is not a growth-limiting nutrient, since sulfate, the oxidized anion, is relatively abundant in the environment. Manganese (Mn) Manganese is a key component of enzyme systems, including oxygen-handling enzymes. It supports brain function and reproduction and is required for blood sugar regulation. In addition, it is part of bone structure. Manganese is a cofactor in function of antioxidant enzymes, such as the mitochondrial superoxide dismutase. In plants, manganese atoms appear to undergo successive oxidations to yield a strongly oxidizing complex that is capable of water oxidations during photosynthesis. Also like magnesium, manganese is required in enzyme reactions involving carbon assimilation. Chloroplasts are most sensitive to manganese deficiency. Among horticultural crops, spinach is a good source of manganese. Copper (Cu) Copper, a redox active metal, plays an important role in the oxidative defense system. In fact, oxidative stress is a characteristic of copper deficiency ( Uriu- Adams and Keen, 2005 ). Copper is necessary for the formation of hemoglobin and is required for the function of over 30 proteins, including superoxide dismutase, ceruloplasmin, lysyl oxidase, cytochrome c oxidase, tyrosinase and dopamineβ -hydroxylase (Arredondo and Nuñez, 2005). During the past decade, there has been increasing interest in the concept that marginal deficits of this essential nutrient can contribute to the development and progression of a number of disease states, including cardiovascular disease and diabetes. Deficits of this nutrient during pregnancy can result in gross structural malformations in the fetus, and persistent neurological and immunological abnormalities in the offspring ( Uriu-Adams and Keen, 2005 ).

194 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 85 In plants, copper is required for chlorophyll synthesis and in several coppercontaining enzymes involved in the reduction of molecular oxygen. The availability of copper to plants, as with other trace minerals, markedly decreases as ph rises above seven. At high ph copper is strongly adsorbed to clays, iron and aluminum oxides, and organic matter. Of the micronutrients required by plants, copper often has the lowest total concentration in soil. Between 1909 and 1919 in the US, the vegetable group was the leading source of copper (30%). In 2004, the grain group (21%) and the legumes, nuts and soy group (20%) replaced the vegetable group (17%) as the leading sources of copper (Table 7, Hiza and Bente, 2007 ). Iron (Fe) The metabolic fates of copper and iron are intimately related. The essentiality of iron, as well as that of copper, resides in its capacity to participate in one-electron exchange reactions. Systemic copper deficiency generates cellular iron deficiency that, in humans, results in diminished work capacity, reduced intellectual capacity, diminished growth, alterations in bone mineralization, and diminished immune response. Iron is required in numerous essential proteins, such as the heme-containing proteins, electron transport chain and microsomal electron transport proteins, and iron-sulfur proteins and enzymes such as ribonucleotide reductase, prolyl hydroxylase phenylalanine hydroxylase, tyrosine hydroxylase and aconitase (Arredondo and Nuñez, 2005). Iron is a constituent of the haem complex, a naturally occurring plant chelate involved in electron transfer in a number of important plant enzymes ( Mengel and Kirkby, 1982 ). The plant plastid stroma may contain deposits of phytoferritin, a storage form of iron similar to the ferritin of animal cells. Phytoferritin occurs almost exclusively in plastids and most abundantly in the plastids of storage organs ( Briat and Lobreaux, 1997 ). In green vegetable leaves, there is a good correlation between iron supply and chlorophyll content. Inadequate iron nutrition results in abnormal chlorophyll development, so that deficiency begins as an interveinal chlorosis on younger leaves resulting in prominent green veins. The resultant reduction in photosynthetic capability also reduces the weight and area of affected leaves. Descriptions of causes of iron deficiency have been extensively reviewed for horticultural crops ( Korcak, 1987 ). Adult users of lower-fat foods consume more nutrient-dense diets, with higher intakes of iron ( Kennedy et al., 2001 ; Sigman-Grant et al., 2003 ). The predominant source of iron in the American food supply is grain products, followed by the meat, poultry and fish group. Between 1909 and 1919, the vegetable group furnished an average of 18% of the iron in the food supply, but in 2004 that share dropped to an average of 10% ( Table 5.7 ). This is partially due to a decrease in the use of white potatoes after Although potatoes are not a good source of iron, their contribution to the food supply increases when eaten in large quantities ( Hiza and Bente, 2007 ), particularly if the skin is consumed (specifically, baked potato skin is 20-fold richer in iron than the flesh). Almonds, pistachio nuts, walnuts, pecans, etc., are very good sources of iron. Different vegetables (e.g. parsley, broccoli, kale, turnip greens and collards) and legumes (e.g. green peas and beans) are also considered good sources of iron.

195 Author s personal copy 86 Nutritional Quality of Fruits and Vegetables Zinc (Zn) Zinc is a pervasive microelement that plays a catalytic or a structural role in more than 200 enzymes (e.g. carboxypeptidase, liver alcohol dehydrogenase and carbonic anhydrase) involved in digestion, metabolism, reproduction, and wound healing. Zn 2 is a cation with various coordination possibilities and several potential geometries. Thus, it is easily adaptable for different ligands. The main role of structural Zn 2 in proteins is to stabilize tertiary structures. In addition, zinc has a critical role in immune response, and is an important antioxidant. Zinc activates a number of plant cell enzymes ( Romheld and Marschner, 1991 ), but only a few of them (i.e. alcohol dehydrogenase, superoxide dismutase, carbonic anhydrase, RNA polymerase) contain the micronutrient. Zinc can affect carbohydrate metabolism because different Zn-dependent enzymes participate in biochemical reactions involving sugars. Zinc also plays a role in the maintenance of cell membrane integrity, in the protection from O 2 damage, and the synthesis of RNA and tryptophan, a precursor of indole-3-acetic acid. A comprehensive review of soil, plant and management factors associated with zinc nutrition in horticultural crops has been developed by Swietlik (1999). Fruits and vegetables account for only 1.2% and 6.4%, respectively, of the zinc in the American food supply ( Hiza and Bente, 2007 ). As is the case for magnesium, zinc intakes may be below the adequate levels for both adults and children ( Sigman- Grant et al., 2003 ). Fruits are poor in zinc, but pecans and walnuts are good sources of this essential mineral. Parsley is also a good source of zinc. Sodium (Na) Sodium is a systemic ion. It is important in electrolyte balance and essential in coregulating ATP with potassium. In addition, it has an important role in the regulation of blood pressure. Sodium contributed from vegetables increased during the last decades ( Table 5.7 ), due to the increased consumption of processed vegetables (largely tomatoes and white potatoes). With the exception of canned vegetables, sodium estimates in the food supply do not account for sodium added in processing. Thus, the relative contribution of vegetables to sodium reported in the food supply is likely overstated ( Hiza and Bente, 2007 ). Table salt (NaCl) is by far the main dietary source for sodium. Olives and spinach are horticultural sources of sodium. In general, fruits are poor in sodium, and are recommended for low-sodium dietary patterns. B. Factors influencing mineral content of fruits and vegetables Influence of the species and the cultivar Mineral composition varies widely in raw fruits ( Table 5.8 ) and vegetables because of genetics. Leafy vegetables tend to have higher concentrations of nutrients that are less mobile in the plant (e.g. calcium) and depend on direct water flow rather than recycling from leaves. Tissues with higher transpiration rates generally have higher tissue calcium concentrations ( Witney et al., 1990b ). Concentrations of minerals may also vary widely with the cultivar. For example, both Dwarf Brazilian bananas

196 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 87 Table 5.8 Mineral composition of a range of fruit species. Results are in mg 100 g 1 fresh weight. Fruit Mineral Apples, raw, with skin K Ca Mg P Mn Cu Fe Zn Na Se Apricots, raw Avocado, raw (California) Avocado, raw (Florida) Bananas, raw Blackberries, raw Blueberries, raw Cherries, sweet, raw Figs, raw Grapefruit, raw, pink and red (California and Arizona) Grapefruit, raw, pink and red (Florida) Grapes, red or green (euro type, e.g. Thompson seedless ), raw Kiwifruit, fresh, raw Lemons, raw, without peel Mangos, raw Melons, Cantaloupe, raw Oranges, raw, California, Valencia Papayas, raw Peaches, raw Pears, raw (Continued)

197 Author s personal copy 88 Nutritional Quality of Fruits and Vegetables Table 5.8 Continued Fruit Mineral K Ca Mg P Mn Cu Fe Zn Na Se Pineapples, raw, all varieties Plums, raw Pomegranates, raw Raspberries, raw Strawberries, raw Watermelon, raw US Department of Agriculture, (Santa Catarina Prata, Musa sp. AAB) and Williams (Cavendish subgroup, Musa sp. AAA) are considered as a good source of potassium. Nevertheless, Dwarf Brazilian bananas have higher P, Ca, Mg, Mn and Zn contents than Williams bananas ( Wall, 2006 ). In contrast, no strawberry variety was statistically superior as a source of minerals ( Hakala et al., 2003 ). Because of the distribution of vascular tissue, sink characteristics and metabolic rates, higher mineral concentrations are usually found in the skin and seeds, with lower concentrations in the flesh of fruits. Tissues with higher metabolic rates (epicarp, core) may have higher requirements for nitrogen and phosphorus. Rapidly expanding or large-celled tissues are unlikely to have high calcium concentrations. In mature fruit, the calcium concentration is highest in the peel ( Saure, 2005 ). Influence of preharvest factors and practices Orchard location has proved to have important effects on fruit and vegetable mineral content ( Table 5.8 ). For example, potassium content in bananas markedly differs between different locations in Hawaii, from 288 mg 100 g 1 in Kapaa to 485 mg 100 g 1 in Waimanalo ( Wall, 2006 ). Papaya cv. Rainbow is not very rich in potassium, but its content also varies between locations, from 113 mg 100 g 1 on the island of Hawaii to 203 mg 100 g 1 on the island of Maui ( Wall, 2006 ). Mineral composition fluctuates widely in raw fruits and vegetables, because of preharvest factors (soil fertility including ph and availability of nutrients moisture content of the soil, growth temperature) and cultural practices (amount and timing of fertilization and irrigation, application of plant growth regulators, pruning and thinning of tree fruit species, etc.). Most of these practices have been established primarily for productivity goals, and not as a medium to better human health, horticultural crop postharvest life or flavor quality ( Crisosto and Mitchell, 2002 ). Usually, fertilizers are applied directly to the soil to raise nutrient levels, if they are inadequate for

198 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 89 the successful growth of the crop, and to maintain soil fertility, which will decline if nutrient removal from the soil (via crop uptake, leaching, volatilization or denitrification) exceeds nutrients added via weathering of minerals and mineralization of organic matter. Nitrogen is the most frequently deficient and most commonly applied fertilizer in orchards, while addition to the soil of phosphorus and potassium is warranted when soil-test results, plant response or tissue analysis indicate a requirement. N-P-K addition with irrigation water (fertigation) has several advantages, including the ability to transport soluble nutrients directly to the root zone whenever the plant is watered. Thus, fertilizer amounts and timing can be precise and adjusted to coincide more closely with actual plant demand. Calcium additions can be large when lime is applied to increase soil ph. Most micronutrients are rarely applied via soil and can be directly supplied via spray application of dilute concentrations of minerals to the canopy. In the case of fruits, the quantity of nutrients capable of being absorbed through the waxy cuticle is often small relative to nutrient demand, but can ameliorate deficiency symptoms and improve fruit quality ( Swietlik and Faust, 1984 ). An excessive supply of nutrients relative to photosynthesis can develop when the rate of nutrient assimilation is high relative to net photosynthesis. In this case, an accumulation of nutrients in fruits and vegetables can reach levels that are toxic either to the plant or to consumers. For example, excessive nitrogen application can lead to potentially harmful accumulations of nitrate nitrogen, especially in leafy greens and potatoes ( Eppendorfer, 1978 ; Blom-Zandstra, 1989 ). These nutrient imbalances also affect horticultural crop quality, as discussed above. Many other factors influencing nutrient accumulation are related to nutrient transport and source-sink relations. For example, alterations in water economy affect calcium input. Since calcium is transported mainly in the transpiration stream ( Grange and Hand, 1987 ), bagging fruit may result in lower calcium concentrations and higher calcium-related disorders ( Witney et al., 1991 ; Hofman et al., 1997 ), due to increased relative humidity. Nevertheless, evidence is not conclusive ( Saure, 2005 ). Canopy position and crop load also influence calcium input. Tree vigor is usually associated with lower calcium and magnesium content in fruits ( Witney et al., 1990a,b ). Fruit from upper parts of the canopy tend to show lower calcium contents ( Ferguson and Triggs, 1990 ), and heavy cropping trees have fruit with higher calcium and lower potassium concentrations ( Ferguson and Watkins, 1992 ). Nevertheless, calcium transport to fruit may be based on a hormonal control; gibberellins have been shown to inhibit calcium translocation ( Saure, 2005 ). Tree size, spacing, row orientation, canopy shape and training system influence light distribution within fruit trees, which in turn may affect mineral composition. In grapes, improvement of light penetration into the canopy enhanced anthocyanin and soluble phenol levels, but reduced potassium content ( Prange and DeEll, 1997 ). In kiwifruit, light promoted calcium accumulation ( Montanaro et al., 2006 ). The finding was not fully explained by fruit transpiration, a regulatory mechanism governed by phytohormones, which could play a role in determining calcium concentrations. Besides, the effect of sunlight does not seem to be universal: avocado fruit from the sunny side of trees did not contain significantly more calcium than fruit from the shaded side ( Witney et al., 1990a ).

199 Author s personal copy 90 Nutritional Quality of Fruits and Vegetables The mineral content of some horticultural species seems to be affected under intensive culture systems (e.g. glasshouse) or organic conditions. Tomato fruit showed higher calcium and lower potassium, magnesium and sodium concentrations when grown on organic (compost/soil mix) versus hydroponic substrates ( Premuzic et al., 1998 ). Smith (1993) reported higher mineral contents in organically cultivated apples, pears, potatoes and corn in comparison to conventionally cultivated ones. In contrast, Petersen and Pedersen (1991) did not find differences in mineral content between organically and conventionally cultivated vegetables. Hakala and co-workers (2003) reported that organic cultivation did not affect strawberry mineral contents consistently. Postharvest practices influencing mineral content of fruits and vegetables Postharvest treatments with minerals, primarily calcium, are used to improve the storage life and quality of different fruits and vegetables. In the last decade, the industry has been encouraged to fortify food and beverages with calcium. Increasing the calcium content of horticultural crops may give consumers new ways to enhance their calcium intake without resorting to supplements. In addition, the use of phosphorous-free sources of calcium can help to obtain a good balance of calcium and phosphorus in the diet ( Martín-Diana et al., 2007 ). Two major methods of postharvest application of calcium in horticultural crops are used: (1) dipping-washing and (2) impregnation processes ( Martín-Diana et al., 2007 ). Dipping treatments are used for fresh, sensitive products, such as leafy vegetables. The delicate texture of berries prevents the use of vacuum infiltration, and dips in a solution of CaCl 2 are used ( García et al., 1996 ), followed by the removal of excess washing solution. On the other hand, impregnation modifies the composition of food material through partial water removal and impregnation of solutes, with no impairment of the material integrity. The process-driven forces can be osmotic gradient between the sample and solution, application of vacuum followed by atmospheric condition restoration, or both. Calcium chloride has been widely used as firming agent and preservative for both whole and fresh-cut fruits and vegetables, as discussed above. C. Effect of minerals on fruit and vegetable quality and consumer acceptance Consumers buy certain items as good sources of specific minerals: potatoes and sweet potatoes for potassium, bananas for magnesium and potassium, spinach for iron, potassium, magnesium and as a non-dairy source of calcium. Mineral content of products is usually determined by ashing and atomic absorption ( Pomeranz and Meloan, 1987 ). Without advanced analytical equipment, the consumer cannot detect differences in individual products at the point of purchase ( Institute of Food Technologists, 1990 ). These attributes are considered credence attributes (see also Chapter 3), because they cannot be detected readily either by visual inspection or by consumption. Therefore, there is little or no incentive to measure mineral content in a quality control program, unless specific nutritional claims can be made.

200 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 91 Nevertheless, the consumer uses other criteria to judge quality. Quality attributes (see Chapter 3) include purchase attributes (i.e. size, color, firmness to the touch, aroma and absence of defects) and consumption attributes (i.e. flavor, mouth feel). Many of these quality characteristics are also affected by the mineral content and constitute part of a wider range of factors affecting fruit and vegetable acceptability. Acceptability, which is defined as the level of continued purchase or consumption by a specific population ( Land, 1988 ), determines the consumption levels of many hidden essential nutrients: vitamins, antioxidants, fiber. Thus, the effect of minerals on horticultural crop quality attributes and consumer acceptance should be considered. Effect of minerals on color In apples and pears, both leaf and fruit nitrogen positively correlates with fruit green background color ( Raese, 1977 ; Marsh et al., 1996 ), regardless of the rootstock used ( Fallahi et al., 1985 ). Manganese has also been associated with green ground color in apples ( Deckers et al., 1997 ). Excessive nitrogen application inhibits background color change from green to yellow and induces deficient reddish blush development and poor edible quality of peaches ( Sistrunk, 1985 ; Crisosto et al., 1995 ; Crisosto et al., 1997 ). High nitrogen application also decreases fruit color in grapes ( Kliewer, 1977 ). In Citrus, nitrogen is associated with an undesirable retardation of endogenous chlorophyll catabolism ( Koo et al., 1974 ) and postharvest treatments with ethylene may be required to accelerate the loss of the green color (de-greening). In apples, amelioration of potassium deficiencies can increase red fruit color, but such an effect is often not apparent when tree potassium status is adequate ( Neilsen and Neilsen, 2003 ). In tomatoes, potassium deficiency is associated with lower levels of lycopene and higher levels of β -carotene ( Trudel and Ozbun, 1971 ). Effect of minerals on flavor Nitrogen status negatively correlates with soluble solids, both in apples ( Fallahi et al., 1985 ; Dris et al., 1999 ) and in pears ( Raese, 1977 ). In contrast, soluble solid content increases with increasing fertilizer nitrogen levels in tomatoes ( Barringer et al., 1999 ). Apple calcium and phosphorus were both negatively correlated with fruit soluble solids at harvest, and after six months of 0 C storage, while fruit K/Ca ratio was positively correlated with titratable acidity ( Fallahi et al., 1985 ). In mango, total soluble solids increased when zinc sulfate fertilizer was applied to the soil ( Bahadur et al., 1998 ). In Fino 49 lemons, salinity reduces juice percentage and impairs juice quality by decreasing the total soluble solids and titratable acidity ( García-Sánchez et al., 2003 ). Reduction of titratable acidity could be due to the greater accumulation of Cl, compared to Na, which could be compensated for by the degradation of organic acids for charge balance. Minerals are also known to affect the production of several classes of volatile compounds in pome fruit (reviewed in Mattheis and Fellman, 1999 ). In fresh onions, increased sulfur availability enhances pungency and total sulfur flavor, but decreases the amounts of precursors for volatiles imparting green and cabbage notes (Randle, 1997 ).

201 Author s personal copy 92 Nutritional Quality of Fruits and Vegetables Effect of minerals on firmness Excess nitrogen fertilization can result in a decrease in firmness ( Reeve, 1970 ; Prange and DeEll, 1997 ). Low phosphorus may also result in a loss of firmness in low-calcium content fruit ( Sharples, 1980 ). The relationship between calcium and fruit firmness has been extensively studied and reviewed ( Ferguson, 1984 ; Poovaiah et al., 1988 ; Harker et al., 1997 ; Sams, 1999 ). Higher firmness values and/or slower softening rates after harvest/storage have been associated with higher calcium concentrations, or with calcium applications in different fruit species, such as apples and pears ( Fallahi et al., 1985 ; Raese and Drake, 1993, 2000a,b, 2002 ; Gerasopoulos and Richardson, 1999 ; Benavides et al., 2001 ); kiwifruit ( Hopkirk et al., 1990 ; Gerasopoulos and Drogoudi, 2005 ); and strawberries ( Chéour et al., 1990 ). Calcium foliar sprays on peaches and nectarines lead to a slight increase of calcium content ( Manganaris et al., 2005a, 2006 ). Under Californian conditions, no consistent effect on fruit quality of mid- or late-season peach and nectarine varieties was found (reviewed in Crisosto et al., 1997 ). Postharvest calcium treatments have been reported to retain fruit firmness in different horticultural products, among them, apples ( Wang et al., 1993 ; Conway et al., 1994 ), peaches ( Manganaris et al., 2005b, 2007 ), strawberries ( Morris et al., 1985 ; García et al., 1996 ), lemons ( Valero et al., 1998 ; Martínez-Romero et al., 1999 ), sliced pears and strawberries ( Rosen and Kader, 1989 ). Calcium effects on fruit firmness are attributable to calcium s ability to cross-link with the pectic polysaccharide network by ionic association. Calcium binding may reduce the accessibility of cell wall degrading enzymes to their substrates. Effect of minerals on rots, physiological disorders and nutritional value In calcium-treated fruit, the association between firmness retention and reduced rot incidence suggests that calcium may affect both processes simultaneously through its cellular role in strengthening plant cell walls ( García et al., 1996 ; Fallahi et al., 1997 ; Conway et al., 1999 ). On the other hand, high nitrogen fertilization increases susceptibility to decay caused by Monilinia fructicola (brown rot) in nectarines ( Daane et al., 1995 ). Wounded and brown rot inoculated Fantasia and Flavortop nectarines from trees having more than 2.6% leaf nitrogen are more susceptible to Monilinia fructicola than fruit from trees with 2.6% or less leaf nitrogen ( Michailides et al., 1993 ). Consumers consider that fruits have less predictable eating quality than manufactured snacks. In fact, the effect of nutrients on the final quality of horticultural products may not become evident until harvest, distribution or consumption. The expression latent damage was coined by Peleg (1985) and later defined by Shewfelt (1986) as damage incurred at one step but not apparent until a later step to describe the result of non-visible quality loss. Physiological disorders may be a type of latent damage. Some physiological disorders relate to the imbalance between nutrients. Calcium is the nutrient most commonly associated with postharvest disorders. A calcium-deficient status is considered an important preharvest factor related to numerous physiological disorders of fruits and vegetables, such as bitter pit in pome fruit, blossom-end rot in tomato, blackheart in celery, cracking and cavity spot

202 Author s personal copy Bibliography 93 in carrot and tip burn in lettuce and cabbage (reviewed in Ferguson et al., 1999 ), although some authors have questioned the role of calcium in these disorders ( Saure, 1998, 2001 ). Other calcium-related disorders are associated with long-term cold storage, such as chilling injury in muskmelon ( Combrink et al., 1995 ) and avocado ( Chaplin and Scott, 1980 ). Postharvest calcium applications limited the incidence of chilling injury in peach fruit, expressed as flesh browning, after four weeks cold storage at 5 C ( Manganaris et al., 2007 ). Nevertheless, preharvest calcium applications showed no effect on the onset of chilling injury in peaches and nectarines (reviewed in Lurie and Crisosto, 2005 ). Magnesium and potassium have been considered as part of an index to predict bitter pit ( Bramlage et al., 1985 ; Autio et al., 1986 ). Fallahi and Righetti (1984) proposed the relation between nitrogen and calcium as an important component of a diagnosis and recommendation system (DRIS) for apple. High rates of nitrogen application exacerbate the incidence of many physiological disorders, such as apricot pit burn ( Bussi and Amiot, 1998, 2003). In addition, minerals can influence the concentrations of other nutrients in horticultural crops. Nitrogen fertilizers at high rates tend to decrease the concentration of vitamin C in fruits (citrus juices) and vegetables (potatoes, cauliflower, white cabbage, crisphead lettuce, etc.) while increased potassium fertilization increases ascorbic acid content (reviewed in Lee and Kader, 2000 ). Bibliography Agius, F., Gonzalez-Lamothe, R., Caballero, J.L., Muñoz-Blanco, J., Botella, M.A., Valpuesta, V. (2003 ). Engineering increased vitamin C levels in plants by overexpression of a D-galacturonic acid reductase. Nature Biotechnol., 21, Ames, B.M., Shigena, M.K., Hagen, T.M. (1993 ). Oxidants, antioxidants and the degenerative diseases of aging. Proc. Natl. Acad. Sci. U.S.A., 90, Anderson, J.S., Perryman, S., Young, L., Prior, S. (2007). Dietary fiber. Colorado State University Nutrition Resources. N FOODNUT/09333.html Accessed April Appel, L.J., Moore, T.J., Oberzanek, E., Vollmer, W.M., Svetkey, L.P., Sacks, F.M. (1997 ). A clinical trial of the effect of dietary patterns on blood pressure. N. E. J. Med., 336, Arredondo, M., Núñez, M.T. ( 2005 ). Iron and copper metabolism. Mol. Aspects Med., 26, Asada, K. (1999 ). The water water cycle in chloroplasts: scavenging of active oxygen and dissipation of excess photons. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50, Asami, D.K., Hong, Y.H., Barrett, D.M., Mitchell, A.E. (2003 ). A comparison of the total phenolic and ascorbic acid contents of freeze-dried and air-dried marionberry, strawberry and corn grown using conventional, organic and sustainable agricultural practices. J. Agric. Food Chem., 51,

203 Author s personal copy 94 Nutritional Quality of Fruits and Vegetables Asp, N.G. (1994 ). Nutritional classification of food carbohydrates. Am. J. Clin. Nutr., 59, S679 S681. Autio, W.R., Bramlage, W.J., Weis, S.A. (1986 ). Predicting disorders of Cox s Orange Pippin and Bramley s Seedling apples by regression equations. J. Am. Soc. Hort. Sci., 111, Ayala-Zavala, J.F., Wang, S.Y., Wang, C.Y., Gonzalez-Aguilar, G.A. ( 2004 ). Effect of storage temperatures on antioxidant capacity and aroma compounds in strawberry fruit. Lebensm. Wiss. Technol., 37, Bahadur, L., Malhi, C.S., Singh, Z. ( 1998 ). Effect of foliar and soil applications of zinc sulphate on zinc uptake, tree size, yield, and fruit quality of mango. J. Plant Nutr., 21, Barrett, D.M., Weakley, C., Diaz, J.V., Watnik, M. (2007 ). Qualitative and nutritional differences in processing tomatoes grown under commercial organic and conventional production systems. J. Food Sci., 72, C441 C450. Barringer, S.A., Bennett, M.A., Bash, W.D. (1999 ). Effect of fruit maturity and nitrogen fertilizer levels on tomato peeling efficiency. J. Veg. Crop Prod., 5, Benavides, A., Recasens, I., Casero, T., Puy, J. (2001 ). Chemometric analyses of Golden Smoothee apples treated with two preharvest calcium spray strategies in the growing season. J. Sci. Food Agric., 81, Benbrook, C.M. (2005). Elevating antioxidant levels in food through organic farming and food processing. An Organic Center state of science review, org/reportfiles/antioxidant_ssr.pdf Accessed April Bernhardt, S., Schlich, E. (2006 ). Impact of different cooking methods on food quality: Retention of lipophilic vitamins in fresh and frozen vegetables. J. Food Eng., 77, Blom-Zandstra, M. (1989 ). Nitrate accumulation in vegetables and its relationship with quality. Annals Appl. Biol., 115, Boerjan, W., Ralph, J., Baucher, M. ( 2003 ). Lignin biosynthesis. Annu. Rev. Plant Biol., 54, Bramlage, W.J., Weis, S.A., Drake, M. (1985 ). Predicting poststorage disorders of McIntosh apples from preharvest mineral analyses. J. Am. Soc. Hort. Sci., 110, Brett, C.T., Waldron, K.W. (1996 ). Physiology and biochemistry of plant cell walls. In: 1. Topics in plant functional biology, M. Black, B. Charlwood (eds). Chapman and Hall, London, UK, p Briat, J.-F., Lobreaux, S. ( 1997 ). Iron transport and storage. Trends Plant Sci., 2, Brummell, D.A. (2006 ). Cell wall disassembly in ripening fruit. Funct. Plant Biol., 33, Brummell, D.A., Harpster, M.H. (2001 ). Cell wall metabolism in fruit softening and quality and its manipulation in transgenic plants. Plant Mol. Biol., 47, Bussi, C., Amiot, M.J. (2003 ). Effects of fertilizer rates and dates of application on apricot (cv. Bergeron) cropping and pitburn. Sci. Hort., 98, Bussi, C., Amiot, M.J. (1998 ). Effects of nitrogen and potassium fertilization on the growth, yield and pitburn of apricot (cv. Bergeron). J. Hort. Sci. Biotechnol., 73,

204 Author s personal copy Bibliography 95 Cantos, E., Espín, J.C., Tomás-Barberán, F.A. (2001 ). Postharvest induction modeling method using UV irradiation pulses for obtaining resveratrol-enriched table grapes: a new functional fruit?. J. Agric. Food. Chem., 49, Cao, G., Sofic, E., Prior, R.L. ( 1996 ). Antioxidant capacity of tea and common vegetables. J. Agric. Food Chem., 44, Carpita, N., McCann, M. (2000 ). The plant cell wall. In: Biochemistry and Molecular Biology of Plants, B. Buchanan, W. Gruissem, R. Jones (eds). American Society of Plant Physiologists, pp Carr, A.C., Frei, B. (1999 ). Toward a new recommended dietary allowance for vitamin C based on antioxidant and health effects in humans. Am. J. Clin. Nutr., 69, Castrejón, A.D., Eichholz, I., Rohn, S., Kroh, L.W., Huyskens-Keil, S. (2008 ). Phenolic profile and antioxidant activity of highbush blueberry ( Vaccinium corymbosum L.) during fruit maturation and ripening. Food Chem., 109, Chaplin, G.R., Scott, K.C. ( 1980 ). Association of calcium in chilling injury susceptibility of stored avocados. HortScience, 15, Chassy, A., Bui, L., Renaud, E.N.C., Van Horn, M., Mitchell, A. ( 2006 ). Three-year comparison of the content of antioxidant microconstituents and several quality characteristics in organic and conventionally managed tomatoes and bell peppers. J. Agric. Food Chem., 54, Chatterjee, I.B. (1973 ). Evolution and the biosynthesis of ascorbic acid. Science, 182, Chéour, F., Willemot, C., Arul, J., Desjardins, Y., Makhlouf, I., Charest, P.M., Gosselin, A. (1990 ). Foliar application of calcium chloride delays postharvest ripening of strawberry. J. Am. Soc. Hort. Sci., 115, Choi, Y.J., Tomás-Barberán, F.A., Saltveit, M.E. ( 2005 ). Wound-induced phenolic accumulation and browning in lettuce ( Lactuca sativa L.) leaf tissue is reduced by exposure to n-alcohols. Postharvest Biol. Technol., 37, Cohen, A.J., Roe, F.J.C. (2000 ). Review of risk factors for osteoporosis with particular reference to a possible aetiological role of dietary salt. Food Chem. Toxicol., 38, Combrink, N.J.J., Jacobs, G., Maree, P.C.J. (1995 ). The effect of calcium and boron on the quality of muskmelons. J. Southern African Soc. Hort. Sci., 5, Conway, W.S., Janisiewicz, W.J., Klein, J.D., Sams, C.E. ( 1999 ). Strategy for combining heat treatment, calcium infiltration, and biocontrol to reduce postharvest decay of Gala apples. HortScience, 34, Conway, W.S., Sams, C.E., Wang, C.Y., Abbott, J.A. (1994 ). Additive effects of postharvest calcium and heat treatment on reducing decay and maintaining quality in apples. J. Am. Soc. Hort. Sci., 119, Cook, A.J., Friday, J.E. (2003 ). Food mixture of ingredient sources for dietary calcium: Shifts in food group contributions using four grouping protocols. J. Am. Diet. Assoc., 103, Crisosto, C.H., Mitchell, J.P. (2002 ). Preharvest factors affecting fruit and vegetable quality. In: Postharvest technology of horticultural crops, A.A. Kader (ed.). Publication University of California, Agriculture and Natural Resources, CA, USA, pp

205 Author s personal copy 96 Nutritional Quality of Fruits and Vegetables Crisosto, C.H., Johnson, R.S., DeJong, T., Day, K.R. (1997 ). Orchard factors affecting postharvest stone fruit quality. HortScience, 32, Crisosto, C.H., Mitchell, F.G., Johnson, R.S. (1995 ). Factors in fresh market stone fruit quality. Postharvest News and Information, 6, 17N 21N. D Ambrosio, C., Giorio, G., Marino, I., Merendino, A., Petrozza, A., Salfi, L., Stigliani, A.L., Francesco, L. (2004 ). Virtually complete conversion of lycopene into β- carotene in fruits of tomato plants transformed with the tomato lycopene β-cyclase (tlcy-b ) cdna. Plant Sci., 166, Daane, K.M., Johnson, R.S., Michailides, T.J., Crisosto, C.H., Dlott, J.W., Ramirez, H.T., Yokota, G.T., Morgan, D.P. (1995 ). Excess nitrogen raises nectarine susceptibility to disease and insects. California Agriculture, 49 ( 4 ), de Ancos, B., Gonzalez, E.M., Cano, M.P. (2000 ). Ellagic acid, vitamin C, and total phenolic contents and radical scavenging capacity affected by freezing and frozen storage in raspberry fruit. J. Agric. Food Chem., 48, de Azevedo, C.H., Rodriguez-Amaya, D.B. (2005 ). Carotenoid composition of kale as influenced by maturity, season and minimal processing. J. Sci. Food Agric., 85, Debar, L.L., Rittenbaugh, C., Vuckovik, N., Stevens, V.J., Aickin, M., Elliot, D., Moe, E., Orwoll, E., Ernst, D., Irvin, L.M. (2004 ). YOUTH: decisions and challenges in designing an osteoporosis prevention intervention for teen girls. Prev. Med., 39, Deckers, T., Daemen, E., Lemmens, K., Missotten, C., Val, J. (1997 ). Influence of foliar applications on Mn during summer on the fruit quality of Jonagold. Acta Hort., 448, Deming, D.M., Baker, D.H., Erdman, J.W. (2002a ). The relative vitamin A value of 9-cis β-carotene is less and that of 13-cis β-carotene may be greater than the accepted 50% that of all-trans β-carotene in gerbils. J. Nutr., 132, Deming, D.M., Teixeira, S.R., Erdman, J.W. (2002b ). All-trans-β-carotene appears to be more bioavailable than 9-cis or 13-cis-β-carotene in gerbils given single oral doses of each isomer. J. Nutr., 132, Dimeglio, L.A., White, K.E., Econs, M.J. ( 2000 ). Disorders of phosphate metabolism. Endocrinology and Metabolism Clinics of North America, 29, Dragsted, L.O. ( 2003 ). Antioxidant actions of polyphenols in humans. Int. J. Vitam. Nutr. Res., 73, Dris, R., Bennett, M.A., Bash, E. (1999 ). Relationship between leaf and fruit minerals and fruit quality attributes of apples grown under northern conditions. J. Plant Nutr., 22, Duthie, G.G., Gardner, P.T., Kyle, J.A.M. (2003 ). Plant polyphenols: are they the new magic bullet?. Proc. Nutr. Soc., 62, Eitenmiller, R.R., Landen, W.O. ( 1999 ). Vitamin analysis for the health and food sciences. CRC Press, Boca Raton, London, New York, Washington, pp Ekholm, P., Reinivuo, H., Mattila, P., Pakkala, H., Koponen, J., Happonen, A., Hellström, J., Ovaskainen, M.-L. (2007 ). Changes in the mineral and trace element contents of cereals, fruits and vegetables in Finland. J. Food Compos. Anal., 20, Eppendorfer, W.H. (1978 ). Effects of N-fertilisation on amino acid composition and nutritive value of spinach, kale, cauliflower, and potatoes. J. Sci. Food Agric., 29,

206 Author s personal copy Bibliography 97 Fairweather-Tait, S., Hurrell, R.F. ( 1996 ). Bioavailability of minerals and trace elements. Nutr. Res. Rev., 9, Fallahi, E., Righetti, T.L. (1984 ). Use of diagnosis and recommendation integrated system (DRIS) in apple. HortScience, 19, 54. Fallahi, E., Conway, W.S., Hickey, K.D., Sams, C.E. (1997 ). The role of calcium and nitrogen in postharvest quality and disease resistance of apples. HortScience, 32, Fallahi, E., Richardson, D.G., Westwood, M.N., Chaplin, M.H. (1985 ). Relationships among mineral nutrition, ethylene and post-harvest physiology in apples on six rootstocks. Sci. Hort., 25, Ferguson, I.B. ( 1984 ). Calcium in plant senescence and fruit ripening. Plant Cell Environ., 7, Ferguson, I.B., Triggs, C.M. (1990 ). Sampling factors affecting the use of mineral analysis of apple fruit for prediction of bitter pit. N. Z. J. Crop Hort. Sci., 18, Ferguson, I.B., Watkins, C.B. (1992 ). Crop load affects mineral concentrations and incidence of bitter pit in Cox s Orange Pippin apple fruit. J. Am. Soc. Hort. Sci., 117, Ferguson, I., Volz, R., Woolf, A. (1999 ). Preharvest factors affecting physiological disorders of fruit. Postharvest Biol. Technol., 15, Fisher, R.L., Bennett, A.B. (1991 ). Role of cell wall hydrolases in fruit ripening. Annu. Rev. Plant Physiol. Plant Mol. Biol., 42, Fraser, P.D., Romer, S., Shipton, C.A., Mills, P.B., Kiano, J.W., Misawa, N., Drake, R.G., Schuch, W., Bramley, P.M. (2002 ). Evaluation of transgenic tomato plants expressing an additional phytoene synthase in a fruit-specific manner. Proc. Natl. Acad. Sci. U.S.A., 99, Frossard, E., Bucher, M., Machler, F., Mozafar, A., Hurrell, R. (2000 ). Potential for increasing the content and bioavailability of Fe, Zn and Ca in plants for human nutrition. J. Sci. Food Agric., 80, García, J.M., Herrera, S., Morilla, A. (1996 ). Effects of postharvest dips in calcium chloride on strawberry. J. Agric. Food Chem., 44, García-Sánchez, F., Carvajal, M., Porras, I., Botía, P., Martínez, V. (2003 ). Effects of salinity and rate of irrigation on yield, fruit quality and mineral composition of Fino 49 lemon. Eur. J. Agron., 19, Gautier, H., Diakou-Verdin, V., Bernard, C., Reich, M., Buret, M., Bourgaud, R., Poessel, J.L., Caris-Veyrat, C., Genard, M. ( 2008 ). How does tomato quality (sugar, acid, and nutritional quality) vary with ripening stage, temperature, and irradiance?. J. Agric. Food. Chem., 56, Gerasopoulos, D., Drogoudi, P.D. (2005 ). Summer-pruning and preharvest calcium chloride sprays affect storability and low temperature breakdown incidence in kiwifruit. Postharvest Biol. Technol., 36, Gerasopoulos, D., Richardson, D.G. (1999 ). Storage temperature and fruit calcium alter the sequence of ripening events of d Anjou pears. HortScience, 34, González-Barrio, R., Beltrán, D., Cantos, E., Gil, M.I., Espín, J.C., Tomás-Barberán, F.A. (2006 ). Comparison of ozone and UV-C treatments on the postharvest stilbenoid monomer, dimer, and trimer induction in var. Superior white table grapes. J. Agric. Food Chem., 54,

207 Author s personal copy 98 Nutritional Quality of Fruits and Vegetables Grange, R.I., Hand, D.W. (1987 ). A review of the effects of atmospheric humidity on the growth of horticultural crops. J. Hort. Sci., 62, Gross, K.C., Sams, C.E. (1984 ). Changes in cell wall neutral sugar composition during fruit ripening: a species survey. Phytochemistry, 23, Grusak, M.A., Della Penna, D. (1999 ). Improving the nutrient composition of plants to enhance human nutrition and health. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50, Gu, L., Kelm, M., Hammerstone, J., Beecher, G., Holde, J., Haytowitz, D., Gebhardt, S., Prior, R. (2004 ). Concentrations of proanthocyanidins in common foods and estimations of normal consumption. J. Nutr., 134, Guillon, F., Champ, M. (2000 ). Structural and physical properties of dietary fibres, and consequences of processing on human physiology. Food Res. Int., 33, Hakala, M., Lapveteläinen, A., Huopalahti, R., Kallio, H., Tahvonen, R. (2003 ). Effects of varieties and cultivation conditions on the composition of strawberries. J. Food Compos. Anal., 16, Hancock, R.D., Viola, R. (2005 ). Improving the nutritional value of crops through enhancement of L-ascorbic acid (vitamin C) content: Rationale and biotechnological opportunities. J. Agric. Food Chem., 53, Harker, F.R., Redgwell,R.J., Hallett, I.C., Murray, S.H. (1997 ). Texture of fresh fruit. Hort. Rev., 20, Hiza, H.A.B., Bente, L. (2007). Nutrient content of the U.S. food supply, : A summary report. Home Economics Research Report Number 57. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion, Washington, DC, USA. Hofman, P.J., Smith, L.G., Joyce, D.C., Johnson, G.I., Meiburg, G.F. (1997 ). Bagging of mango (Mangifera indica cv. Keitt ) fruit influences fruit quality and mineral composition. Postharvest Biol. Technol., 12, Hopkirk, G., Harker, F.R., Harman, J.E. (1990 ). Calcium and firmness of kiwifruit. N. Z. J. Crop Hort. Sci., 18, Huber, D.J., O Donoghue, E.M. (1993 ). Polyuronides in avocado ( Persea americana ) and tomato ( Lycopersicon esculentum ) fruits exhibit markedly different patterns of molecular weight downshifts during ripening. Plant Physiol., 102, Ignarro, L.J., Balestrieri, M.L., Napoli, C. (2007 ). Nutrition, physical activity, and cardiovascular disease: An update. Cardiovasc. Res., 73, Institute of Food Technologists (1990 ). Quality of fruits and vegetables. A scientific status summary by the Institute of Food Technologists Expert Panel on Food Safety and Nutrition. Food Technol., 44, Institute of Medicine ( 2001 ). Dietary reference intakes proposed definition of dietary fiber. A report of the panel on the definition of dietary fiber and the Standing Committee on the scientific evaluation of dietary reference intakes. Food and Nutrition Board, National Academy Press, Washington, DC, USA. Jayaraj, J., Devlin, R., Punja, Z. (2008 ). Metabolic engineering of novel ketocarotenoid production in carrot plants. Transgenic Res. 17, Jeremy, J.Y., Shukla, N., Muzaffar, S., Handley, A., Angelini, G.D. (2004 ). Reactive oxygen species, vascular disease and cardiovascular surgery. Curr. Vasc. Pharmacol., 2,

208 Author s personal copy Bibliography 99 Jodral-Segado, A.M., Navarro-Alarcón, M., López-G de la Serrana, H., López Martínez, M.C. (2003 ). Magnesium and calcium contents in foods from SE Spain: Influencing factors and estimation of daily dietary intakes. Sci. Total Environ., 312, Kalt, W., Forney, C.F., Martin, A., Prior, R.L. (1999 ). Antioxidant capacity, vitamin C, phenolics, and anthocyanins after fresh storage of small fruits. J. Agric. Food Chem., 47, Kays, S.J. (1997 ). Postharvest Physiology of Perishables Plant Products. Exon Press, Athens, GA, USA. Kennedy, E., Bowman, S., Powell, R. (2001 ). Assessment of the effect of fat-modified foods on diet quality in adults, years, using data from the Continuing Survey of Food Intake by Individuals. J. Am. Diet. Assoc., 101, Kevers, C., Falkowski, M., Tabart, J., Defraigne, J.O., Dommes, J., Pinc , J. (2007 ). Evolution of antioxidant capacity during storage of selected fruits and vegetables. J. Agric. Food Chem., 55, Kliewer, W.M. (1977 ). Influence of temperature, solar radiation, and nitrogen on coloration and composition of Emperor grapes. Am. J. Enol. Vit., 28, Kobayashi, M., Matoh, T., Azuma, J. (1996 ). Two chains of rhamnogalacturonan II are cross-linked by borate-diol ester bonds in higher plant cell walls. Plant Physiol., 110, Koo, R.J.C., Young, T.W., Reese, R.L., Kesterson, J.W. (1974 ). Effect of nitrogen, potassium and irrigation on yield and quality of lemon. J. Am. Soc. Hort. Sci., 99, Kopsell, D.A., Kopsell, D.E. (2006 ). Accumulation and bioavailability of dietary carotenoids in vegetable crops. Trends Plant Sci., 11, Korcak, R.F. (1987 ). Iron deficiency chlorosis. Hort. Rev., 9, Land, D.G. (1988 ). Negative influences on acceptability and their control. In: Food Acceptability, D.M.H. Thomson (ed.). Elsevier, New York, NY, USA, pp Langcake, P., Pryce, R.J. (1976 ). The production of resveratrol by Vitis vinifera and other members of the Vitaceae as a response to infection and injury. Physiol. Plant Pathol., 9, Larson, R.A. (1988 ). The antioxidants of higher plants. Phytochemistry, 27, Le Marchand, L. (2002 ). Cancer preventive effects of flavonoids a review. Biomed. Pharmacother., 56, Lee, S.K., Kader, A.A. (2000 ). Preharvest and postharvest factors influencing vitamin C content of horticultural crops. Postharvest Biol. Technol., 20, Levine, M., Wang, Y.H., Padayatty, S.J., Morrow, J. (2001 ). A new recommended dietary allowance of vitamin C for healthy young women. Proc. Natl. Acad. Sci. U.S.A., 98, Lin, P.-H., Aickin, M., Champagne, C., Craddick, S., Sacks, F.M., McCarron, P., Most- Windhauser,M.M., Rukenbrod,F., Haworth, L. (2003 ). Food group sources of nutrients in the dietary patterns of the DASH-Sodium trial. J. Am. Diet. Assoc., 103, Liu, R.H. (2007 ). Whole grain phytochemicals and health. J. Cereal Sci., 46,

209 Author s personal copy 100 Nutritional Quality of Fruits and Vegetables Liu, Y., Roof, S., Ye, Z., Barry, C., van Tuinen, A., Vrebalov, J., Bowler,C., Giovannoni, J. (2004 ). Manipulation of light signal transduction as a means of modifying fruit nutritional quality in tomato. Proc. Natl. Acad. Sci. U.S.A., 101, Loaiza-Velarde, J.G., Tomás-Barberán, F.A., Saltveit, M.E. (1997 ). Effect of intensity and duration of heat-shock treatments on wound-induced phenolic metabolism in Iceberg lettuce. J. Am. Soc. Hortic. Sci., 122, Lurie, S., Crisosto, C.H. ( 2005 ). Chilling injury in peach and nectarine. Postharvest Biol. Technol., 37, MacDonald, H.M. (2007 ). Influence of organic salts of potassium on bone health: Possible mechanisms of action for the role of fruit and vegetables. Int. Congr. Ser., 1297, Manach, C., Scalbert, A., Morand, C., Remesy, C., Jimenez, L. (2004 ). Polyphenols: Food sources and bioavailability. Am. J. Clin. Nutr., 79, Manganaris, G.A., Vasilakakis, M., Mignani, I., Diamantidis, G., Tzavella-Klonari, K. ( 2005 a ). The effect of preharvest calcium sprays on quality attributes, physicochemical aspects of cell wall components and susceptibility to brown rot of peach fruits (Prunus persica L. cv. Andross). Sci. Hort., 107, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. ( 2005 b ). Effect of post-harvest calcium treatments on the physicochemical properties of cell wall pectin in nectarine fruit during ripening after harvest or cold storage. J. Hort. Sci. Biotechnol., 80, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. (2006 ). Effect of inseason calcium applications on cell wall physicochemical properties of nectarine fruit ( Prunus persica var. nectarina Ait. Maxim) after harvest or cold storage. J. Sci. Food Agric., 86, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. (2007 ). The effect of postharvest calcium application on tissue calcium concentration, quality attributes, incidence of flesh browning and cell wall physicochemical aspects of peach fruits. Food Chem., 100, Marlett, J.A. (1992 ). Content and composition of dietary fiber in 117 frequently consumed foods. J. Am. Diet. Assoc., 92, Marlett, J.A., Longacre, M.J. (1997 ). Comparisons of in vitro and in vivo measures of resistant starch in selected grain products. Cereal Chem., 73, Marlett, J.A., McBurney, M.I., Slavin, J.L. (2002 ). Position of the American Dietetic Association health implications of dietary fiber. J. Am. Diet. Assoc., 102, Marlett, J.A., Slavin, J.L. (1997 ). Position of the American Dietetic Association health implications of dietary fiber. J. Am. Diet. Assoc., 97, Marsh, K.B., Volz, R.K., Cashmore, W., Reay, P. (1996 ). Fruit colour, leaf nitrogen level, and tree vigour in Fuji apples. N. Z. J. Crop Hort. Sci., 24, Martín-Diana, A.B., Rico, D., Frías, J.M., Barat, J.M., Henehan, G.T.M., Barry-Ryan, C. ( 2007 ). Calcium for extending the shelf life of fresh whole and minimally processed fruits and vegetables: A review. Trends Food Sci. Technol., 18, Martínez-Romero, D., Valero, D., Serrano, M., Riquelme, F. (1999 ). Effects of postharvest putrescine and calcium treatments on reducing mechanical damage and polyamine and ABA levels during lemon storage. J. Sci. Food Agric., 79,

210 Author s personal copy Bibliography 101 Mattheis, J.P., Fellman, J.K. (1999 ). Preharvest factors influencing flavor of fresh fruit and vegetables. Postharvest Biol. Technol., 15, Mattila, P., Hellstrom, J., Torronen, R. (2006 ). Phenolic acids in berries, fruits, and beverages. J. Agric. Food Chem., 54, Mayer, A.M. ( 1997 ). Historical changes in the mineral content of fruits and vegetables. British Food J., 99, McCarron, D.A., Reusser, M.E. (2001 ). Are low intakes of calcium and potassium important causes of cardiovascular disease?. Am. J. Hyp. 14, S206 S212. Meléndez-Martínez, A.J., Vicario, I.M., Heredia, F.J. (2007 ). Provitamin A carotenoids and ascorbic acid contents of the different types of orange juices marketed in Spain. Food Chem., 101, Mengel, K., Kirkby, E.A. ( 1982 ). Principles of Plant Nutrition. International Potash Institute, Worblaufen-Bern, Switzerland. Michailides, T.J., Ramirez, H.T., Morgan, D.P., Crisosto, C.H., Johnson, R.S. (1993). Effects of nitrogen fertilization on susceptibility of stone fruits (peach and nectarine) to brown rot. In: 1992 Research Reports for California Peaches and Nectarines. California Tree Fruit Agreement, Sacramento, CA, USA. Mittler, R. ( 2002 ). Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci., 7, Montanaro, G., Dichio, B., Xiloyannis, C., Celano, G. ( 2006 ). Light influences transpiration and calcium accumulation in fruit of kiwifruit plants ( Actinidia deliciosa var. deliciosa ). Plant Sci., 170, Morris, J.R., Sistrunk,W.A., Sims, C.A., Main, G.L. (1985 ). Effects of cultivar, postharvest storage, preprocessing dip treatments and style of pack on the processing quality of strawberries. J. Am. Soc. Hort. Sci., 110, Muir, S.R., Collins, G.J., Robinson, S., Hughes, S., Bovy, A., Ric De Vos, C.H., van Tunen, A.J., Verhoeyen, M.E. (2001 ). Overexpression of petunia chalcone isomerase in tomato results in fruit containing increased levels of flavonols. Nat. Biotechnol., 19, Murad, S., Grove, D., Lindberg, K.A., Reynolds, G., Sivarajah, A., Pinnell, S.R. (1981 ). Regulation of collagen synthesis by ascorbic acid. Proc. Natl. Acad. Sci. U.S.A., 78, Mussatto, S.I., Mancilha, I.M. (2007 ). Non-digestible oligosaccharides: A review. Carbohydr. Polym., 68, Neilsen, G.H., Neilsen, D. (2003 ). Nutritional requirements of apple. In: Apples: Botany, production and uses, D.C. Ferree, I.J. Warrington (eds). CABI Publishing, Wallingford, Oxon, UK, pp Nelson, J.W., Barritt, B.H., Wolford, E.R. (1972 ). Influence of location and cultivar on color and chemical composition of strawberry fruit. Wash. Agric. Exp. Stn. Tech. Bull., 74, 1 7. Nishiyama, I., Yamashita, Y., Yamanaka, M., Shimohashi, A., Fukuda, T., Oota, T. (2004 ). Varietal difference in vitamin C content in the fruit of kiwifruit and other Actinidia species. J. Agric. Food Chem., 52, Noctor, G., Foyer, C. (1998 ). Ascorbate and glutathione: Keeping active oxygen under control. Annu. Rev. Plant Physiol. Plant Mol. Biol., 49,

211 Author s personal copy 102 Nutritional Quality of Fruits and Vegetables Nordin, B.E.C. (1997 ). Calcium and osteoporosis. Nutrition, 13, O Donoghue, E.M., Huber, D.J., Timpa, J.D., Erdos, G.W., Brecht, J.K. (1994 ). Influence of avocado ( Persea americana ) Cx-cellulase on the structural features of avocado cellulose. Planta, 194, O Neill, M.A., Ishii, T., Albersheim, P., Darvill, A.G. (2004 ). Rhamnogalacturonan II: Structure and function of a borate cross-linked cell wall pectic polysaccharide. Annu. Rev. Plant Biol., 55, Pantelidis, G.E., Vasilakakis, M., Manganaris,G.A., Diamantidis, G. (2007 ). Antioxidant capacity, phenol, anthocyanin and ascorbic acid contents in raspberries, blackberries, red currants, gooseberries and cornelian cherries. Food Chem., 102, Peleg, K. ( 1985 ). Produce handling, packaging, and distribution. AVI/Van Nostrand Reinhold, New York, NY, USA. Perkins-Veazie, P., Collins, J.K., Howard, L. (2008 ). Blueberry fruit response to postharvest application of ultraviolet radiation. Postharvest Biol. Technol., 47, Petersen, J., Pedersen, G. (1991). Sporelementer I økologisk dyrkede grønsaker. Levnedsmiddelstyrelsen. Rapport nr. F Levnedsmiddelstyrelsen, Copenhagen, Denmark (in Danish). Pomeranz, Y., Meloan, C.E. (1987 ). Food Analysis Theory and Practice, 2nd edn. AVI/ Van Nostrand Reinhold, New York, NY, USA. Poovaiah, B.W., Glenn, G.M., Reddy, A.S.N. (1988 ). Calcium and fruit softening: physiology and biochemistry. Hort. Rev., 10, Prange, R., DeEll, J.R. (1997 ). Preharvest factors affecting quality of berry crops. HortScience, 32, Premuzic, Z., Bargiela, M., García, A., Rendina, A., Iorio, A. (1998 ). Calcium, iron, potassium, phosphorus, and vitamin C content of organic and hydroponic tomatoes. HortScience, 33, Prior, R.L., Cao, G., Martin, A., Sofic, E., McEwen, J., O Brien, C., Lischner, N., Ehlenfeldt, M., Kalt, W., Krewer, G., Mainland, C.M. (1998 ). Antioxidant capacity as influenced by total phenolic and anthocyanin content, maturity, and variety of Vaccinium species. J. Agric. Food Chem., 46, Raese, J.T., Drake, S.R. (1993 ). Effects of preharvest calcium sprays on apple and pear quality. J. Plant Nutr., 16, Raese, J.T., Drake, S.R. (2000a ). Effect of calcium sprays, time of harvest, cold storage, and ripeness on fruit quality of Anjou pears. J. Plant Nutr., 23, Raese, J.T., Drake, S.R. ( 2000 b ). Effect of calcium spray materials, rate, time of spray application, and rootstocks on fruit quality of Red and Golden Delicious apples. J. Plant Nutr., 23, Raese, J.T., Drake, S.R. (2002 ). Calcium spray materials and fruit calcium concentrations influence apple quality. J. Am. Pom. Soc., 56, Raese, J.T. ( 1977 ). Response of young d Anjou pear trees to triazine and triazole herbicides and nitrogen. J. Am. Soc. Hort. Sci., 102, Raghothama, K.G. (1999 ). Phosphate acquisition. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50,

212 Author s personal copy Bibliography 103 Ramalingam, K., Lee, K.M., Woodard, R.M., Bleecker, A.B., Kende, H. (1985 ). Stereochemical course of the reaction catalyzed by the pyridoxal phosphate-dependent enzyme 1-aminocyclopropane-1-carboxylate synthase. Proc. Natl. Acad. Sci. USA, 82, Randle, W. (1997). Onion flavor chemistry and factors influencing flavor intensity. In: Spices: Flavor Chemistry and Antioxidant Properties, S.J. Risch, C.T. Ho (eds). ACS Symposium Series 660. ACS Press, Washington, DC, USA, pp Rao, A.V., Rao, L.G. (2007 ). Carotenoids and human health. Pharmacol. Res., 55, Reddy, M.S., Chen, F., Shadle, G., Jackson, L., Aljoe, H., Dixon, R.A. ( 2005 ). Targeted down-regulation of cytochrome P450 enzymes for forage quality improvement in alfalfa ( Medicago sativa L.). Proc. Nat. Acad. Sci. USA, 102, Redgwell, R.J., MacRae, E., Hallett, I., Fisher, M., Perry, J., Harker, R. (1997 ). In vivo and in vitro swelling of cell walls during fruit ripening. Planta, 203, Reeve, R.M. (1970 ). Relationships of histological structure to texture of fresh and processed fruits and vegetables. J. Text. Stud., 1, Rice-Evans, C.A., Miller, N., Paganga, G. (1997 ). Antioxidant properties of phenolic compounds. Trends Plant Sci., 2, Ridley, B.L., O Neill, M.A., Mohnen, D. (2001 ). Pectins, structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry, 57, Rimm, E.B., Ascherio, A., Giovannucci, E., Spiegelman, D., Stampfer, M.J., Willett, W.C. (1996 ). Vegetable, fruit, and cereal fiber intake and risk of coronary heart disease among men. J. Am. Med. Assoc., 275, Rodriguez-Amaya, D.B. (2001 ). A Guide to Carotenoid Analysis in Foods. ILSI Press, International Life Sciences Institute, One Thomas Circle, N.W. Washington, DC, USA. Romheld, V., Marschner, H. ( 1991 ). Function of micronutrients in plants. In: Micronutrients in agriculture, J.J. Mortvedt, F.R. Cox, L.M. Shuman, R.M. Welch (eds), 2nd edn. Soil Science Society of America, Madison, WI, USA, pp Rose, J.K.C., Catalá, C., Gonzalez-Carranza, Z.H., Roberts, J.A. ( 2003 ). Cell wall disassembly. In: The Plant Cell Wall, J.K.C. Rose (ed.). Blackwell Publishing Ltd., Oxford, UK, pp Rosen, J.C., Kader, A.A. (1989 ). Postharvest physiology and quality maintenance of sliced pear and strawberry fruit. J. Food Sci., 54, Sajilata, M.G., Singhal, R.S., Kulkarni, P.R. ( 2006 ). Resistant starch A review. Compr. Rev. Food Sci. Food Safety, 5, Saltveit, M.E. (1988 ). Postharvest glyphosate application reduces toughening fiber content and lignification of stored asparagus spears. J. Am. Soc. Hort. Sci., 113, Salunkhe, D.K., Bolin, H.R., Reddy, N.R. (eds) (1991 ). Storage, processing, and nutritional quality of fruits and vegetables. Volume I. Fresh Fruits and Vegetables. CRC Press, Boston, MA, USA. Sams, C.E. (1999 ). Preharvest factors affecting postharvest texture. Postharvest Biol. Technol., 15, Sanders, D., Brownlee, C., Harper, J.F. (1999 ). Communicating with calcium. Plant Cell, 11,

213 Author s personal copy 104 Nutritional Quality of Fruits and Vegetables Sandmann, G. (2001 ). Genetic manipulation of carotenoid biosynthesis, strategies, problems and achievements. Trends Plant Sci., 6, Sanmartin, M., Pateraki, I., Chatzopoulou, F., Kanellis, A.K. (2007 ). Differential expression of the ascorbate oxidase multigene family during fruit development and in response to stress. Planta, 225, Saure, M.C. ( 1998 ). Causes of the tipburn disorder in leaves of vegetables. Sci. Hort., 76, Saure, M.C. (2001 ). Blossom-end rot of tomato (Lycopersicon esculentum Mill.) a calcium or stress-related disorder?. Sci. Hort., 90, Saure, M.C. (2005 ). Calcium translocation to fleshy fruit: Its mechanism and endogenous control. Sci. Hort., 105, Scalbert, A., Williamson, G. (2000 ). Dietary intake and bioavailability of polyphenols. J. Nutr., 130, Seeram, N.P., Adams, L.S., Zhang, Y., Lee, R., Sand, D., Scheuller, H.S., Heberg, D. (2006 ). Raspberry, blueberry, cranberry, red raspberry, and strawberry extracts inhibit growth and stimulate apoptosis of human cancer cells in vitro. J. Agric. Food Chem., 54, Sharples, R.O. (1980 ). The influence of orchard nutrition on the storage quality of apples and pears grown in the United Kingdom. In: Mineral Nutrition of Fruit Trees, D. Atkinson, J.E. Jackson, R.O. Sharples, W.M. Waller (eds). Butterworths, London, UK, pp Shewfelt, R.L. (1986 ). Postharvest treatment for extending the shelf life of fruits and vegetables. Food Technol., 40, Sigman-Grant, M., Warland, R., Hsieh, G. (2003 ). Selected lower-fat foods positively impact nutrient quality in diets of free-living Americans. J. Am. Diet. Assoc., 103, Sistrunk, W.A. ( 1985 ). Peach quality assessment, fresh and processed. In: Evaluation of Quality of Fruits and Vegetables, H.E. Patee (ed.) AVI, Westport, pp Slavin, J.L. (2005 ). Dietary fiber and body weight. Nutrition, 21, Smirnoff, N. (2000 ). Ascorbic acid, metabolism and functions of a multifaceted molecule. Curr. Opin. Plant Biol., 3, Smirnoff, N., Wheeler, G.L. (2000 ). Ascorbic acid in plants, Biosynthesis and function. Crit. Rev. Biochem. Mol. Biol., 35, Smith, B. (1993 ). Organic food vs. supermarket food, Element levels. J. Appl. Nutr., 45, Sozzi, G.O. ( 2001 ). Caper bush, botany and horticulture. Hort. Rev., 27, Swietlik, D., Faust, M. ( 1984 ). Foliar nutrition of fruit crops. Hort. Rev., 6, Swietlik, D. ( 1999 ). Zinc nutrition in horticultural crops. Hort. Rev., 23, Terry, P., Giovannucci, E., Michels, K.B., Bergkvist, L., Hansen, H., Holmberg, L., Wolk, A. (2001 ). Fruit, vegetables, dietary fiber, and risk of colorectal cancer. J. Nat. Canc. Instit., 93, Titchenal, C.A., Dobbs, J. (2007 ). A system to assess the quality of food sources of calcium. J. Food Compos. Anal., 20, Toivonen, P.M.A., Zebarth, B.J., Bowen, P.A. (1994 ). Effect of nitrogen fertilization on head size, vitamin C content and storage life of broccoli ( Brassica oleracea var. italica). Can. J. Plant Sci., 74,

214 Author s personal copy Bibliography 105 Tomás-Barberán, F.A., Loaiza-Velarde, J., Bonfanti, A., Saltveit, M.E. ( 1997 ). Early wound- and ethylene-induced changes in phenylpropanoid metabolism in harvested lettuce. J. Am. Soc. Hort. Sci., 122, Tripoli, E., La Guardia, M., Giammanco, S., Di Majo, D., Giammanco, M. ( 2007 ). Citrus flavonoids, molecular structure, biological activity and nutritional properties, a review. Food Chem., 104, Trudel, M.J., Ozbun, J.L. ( 1971 ). Influence of potassium on carotenoid content of tomato fruit. J. Am. Soc. Hort. Sci., 96, Tsao, R., Deng, Z. (2004 ). Separation procedures for naturally occurring antioxidant phytochemicals. J. Chromatogr. B., 812, Uriu-Adams, J.Y., Keen, C.L. ( 2005 ). Copper, oxidative stress, and human health. Mol. Aspects Med., 26, U.S. Department of Agriculture (2008). Composition of foods, raw, processed, prepared. USDA national nutrient database for standard reference, release 20. USDA-ARS, Beltsville Human Nutrition Research Center, Nutrient Data Laboratory, Beltsville, Maryland. Accessed April U.S. Department of Health and Human Services and U.S. Department of Agriculture ( 2005 ). Dietary Guidelines for Americans, 2005, 6th edn. U.S. Government Printing Office, Washington, DC, USA. Valero, D., Martínez-Romero, D., Serrano, M., Riquelme, F. (1998 ). Influence of postharvest treatment with putrescine and calcium on endogenous polyamines, firmness, and abscisic acid in lemon ( Citrus limon L. Burn cv. Verna). J. Agric. Food Chem., 46, Valpuesta, V., Botella, M.A. (2004 ). Biosynthesis of L-ascorbic acid in plants, new pathways for an old antioxidant. Trends Plant Sci., 9, Versari, A., Parpinello, G.P., Tornielli, G.B., Ferrarini, R., Giulivo, C. (2001 ). Stilbene compounds and stilbene synthase expression during ripening, wilting, and UV treatment in grape cv. Corvina. J. Agric. Food. Chem., 49, Vicente, A.R., Ortugno, C., Rosli, H., Powell, A.L.T., Greve, C., Labavitch, J.M.L. (2007a ). Temporal sequence of cell wall disassembly events in developing fruits. 2. Analysis of blueberry ( Vaccinium Species). J. Agric. Food Chem., 55, Vicente, A.R., Saladié, M., Rose, J.C.K., Labavitch, J.M. ( 2007 b ). The linkage between cell wall metabolism and fruit softening, looking to the future. J. Sci. Food Agric., 87, von Elbe, J.H., Schwartz, S.J. (1996 ). Colorants. In: Food Chemistry, O.R. Fennema (ed.) 3rd edn. Marcel Dekker, New York, NY, USA. Wall, M.M. (2006 ). Ascorbic acid, vitamin A, and mineral composition of banana (Musa sp.) and papaya ( Carica papaya ) cultivars grown in Hawaii. J. Food Compos. Anal., 19, Wang, C.Y., Conway, W.S., Abbott, J.A., Kramer, G.F., Sams, C.E. ( 1993 ). Postharvest infiltration of polyamines and calcium influences ethylene production and texture changes in Golden Delicious apples. J. Am. Soc. Hort. Sci., 118, Wang, H., Cao, G., Prior, R.L. ( 1996 ). Total antioxidant capacity of fruits. J. Agric. Food Chem., 44, Wang, S.Y., Lin, H.S. (2000 ). Antioxidant activity in fruits and leaves of blackberry, raspberry, and strawberry varies with cultivar and developmental stage. J. Agric. Food Chem., 48,

215 Author s personal copy 106 Nutritional Quality of Fruits and Vegetables Wang, S.Y., Lin, H.S. ( 2003 ). Compost as a soil supplement increases the level of antioxidant compounds and oxygen radical absorbance capacity in strawberries. J. Agric. Food Chem., 51, Wang, S.Y., Zheng, W., Galletta, G.J. (2002 ). Cultural system affects fruit quality and antioxidant capacity in strawberries. J. Agric. Food Chem., 50, Waris, G., Ahsan, H. (2006 ). Reactive oxygen species, role in the development of cancer and various chronic conditions. J. Carcinog., 6, Weibel, F., Bickel, R., Leuthold, S., Alfoldi, T. (2000 ). Are organically grown apples tastier and healthier? A comparative field study using conventional and alternative methods to measure fruit quality. Acta Hort., 57, Willats, W.G.T., McCartney, L., Mackie, W., Knox, J.P. ( 2001 ). Pectin, cell biology and prospects for functional analysis. Plant Mol. Biol., 47, Winter, C.K., Davis, C.F. (2006 ). Organic foods. J. Food Sci., 71, R117 R124. Witney, G.W., Hofman, P.M., Wolstenholme, B.N. (1990a ). Effect of cultivar, tree vigour and fruit position on calcium accumulation in avocado fruits. Sci. Hort., 44, Witney, G.W., Hofman, P.J., Wolstenholme, B.N. (1990b ). Mineral distribution in avocado trees with reference to calcium cycling and fruit quality. Sci. Hort., 44, Witney, G.W., Kushad, M.M., Barden, J.A. ( 1991 ). Induction of bitter pit in apple. Sci. Hort., 47, Woese, K., Lange, D., Boess, C., Bogel, K.W. (1997 ). A comparison of organically and conventionally grown foods results of a review of the relevant literature. J. Sci. Food Agric., 74, Wolk, A., Manson, J.A.E., Stampfer, M.J., Colditz, G.A., Hu, F.B., Speizer, F.E., Hennekens, C.H., Willett, W.C. (1999 ). Long-term intake of dietary fiber and decreased risk of coronary heart disease among women. J. Am. Med. Assoc., 281, Zerwekh, J.E., Odvina, C.V., Wuermser, L.-A., Pak, C.Y.C. (2007 ). Reduction of renal stone risk by potassium-magnesium citrate during 5 weeks of bed rest. J. Urol., 177, Zheng, W., Wang, S.W. (2003 ). Oxygen radical absorbing capacity of phenolics in blueberries, cranberries, chokeberries, and lingonberries. J. Agric. Food Chem., 51, Zheng, Y., Wang, C.Y., Wang, S.Y., Zheng, W. (2003 ). Effect of high-oxygen atmospheres on blueberry phenolics, anthocyanins, and antioxidant capacity. J. Agric. Food Chem., 51, Zyren, J., Elkins, R., Dudek, J.A., Hagen, R.E. (1983 ). Fiber contents of selected raw and processed vegetables, fruits and fruit juices as served. J. Food Sci., 48,

216 Provided for non-commercial research and educational use only. Not for reproduction, distribution or commercial use. This chapter was originally published in the book Postharvest Handling: A Systems Approach published by Elsevier, and the attached copy is provided by Elsevier for the author s benefit and for the benefit of the author s institution, for non-commercial research and educational use including without limitation use in instruction at your institution, sending it to specific colleagues who you know, and providing a copy to your institution s administrator. All other uses, reproduction and distribution, including without limitation commercial reprints, selling or licensing copies or access, or posting on open internet sites, your personal or institution s website or repository, are prohibited. For exceptions, permission may be sought for such use through Elsevier s permissions site at: From: Postharvest Handling: A Systems Approach, Second Edition Edited by Wojciech J. Florkowski, Robert L. Shewfelt, Bernhard Brueckner and Stanley E. Prussia ISBN: Copyright 2009, Elsevier Inc. Academic Press.

217 Author s personal copy Nutritional Quality of Fruits and Vegetables 5 Ariel R. Vicente Facultad de Ciencias Agrarias y Forestales. UNLP. Calle 60 y 119 s/n. CP 1900 La Plata Argentina y Centro de Investigación y Desarrollo en Criotecnología de Alimentos (CIDCA) CONICET-UNLP. Calle 47 esq CP La Plata Argentina George A. Manganaris Department of Plant Sciences, University of California, Davis, One Shields Avenue, Davis, CA USA Gabriel O. Sozzi Cátedra de Fruticultura, Facultad de Agronomía, Universidad de Buenos Aires. Avda. San Martín C 1417 DSE Buenos Aires, Argentina CONICET, Argentina Carlos H. Crisosto Department of Plant Sciences, University of California, Davis, One Shields Avenue, Davis, CA USA I. Introduction II. Traditional components A. Water B. Organic acids C. Proteins D. Lipids and fatty acids E. Metabolizable carbohydrates F. Dietary fiber G. Vitamins III. Antioxidants in fruits and vegetables A. Oxidative damage and antioxidants B. Ascorbic acid C. Carotenoids D. Tocopherols and tocotrienols E. Phenolic compounds F. Factors affecting the levels of antioxidants in fruits and vegetables IV. Fruits and vegetables as direct sources of minerals A. General considerations of selected minerals B. Factors influencing mineral content of fruits and vegetables C. Effect of minerals on fruit and vegetable quality and consumer acceptance Bibliography Postharvest Handling: A Systems Approach ISBN: Copyright 2009, Elsevier Inc. All rights reserved

218 Author s personal copy 58 Nutritional Quality of Fruits and Vegetables I. Introduction Horticultural crops are some of the main components of a healthy diet. The constituents obtained by the human body from fruits and vegetables include water, carbohydrates, fats, proteins, fiber, minerals, organic acids, pigments, vitamins and antioxidants, among others. Fruits and vegetables, especially, are a good source of fiber, selected minerals, vitamins and antioxidants. Most fruits and vegetables are available almost year-round in a wide variety and they not only taste good, but they also have favorable attributes of texture, color, flavor and ease of use. They can be fresh, cooked, hot or cold, canned, pickled, frozen or dried. Fruits and vegetables are consumed at all times, and due to their convenient size; they are an excellent between-meal snack. They are relatively low in calories and fat (avocado and olives being the exceptions), they have no cholesterol, they are rich in carbohydrates and fiber, they contain vitamin C and carotene, and some are a good source of vitamin B 6. Fruits and vegetables are relatively low in sodium and high in potassium. Ascorbic acid in fruits and vegetables enhances the bioavailability of iron in the diet. Because of all these characteristics, fruits and vegetables have a unique role in a healthy diet. A growing body of research has shown that fruit and vegetable consumption is associated with reduced risk of major diseases, and possibly delayed onset of age-related disorders, promoting good health. However, in many cases fruit and vegetable consumption is still below the dietary guideline goal of consuming 5 10 servings each day. The nutritional value of fruits and vegetables depends on their composition, which shows a wide range of variation depending on the species, cultivar and maturity stage. The composition of fruits and vegetables includes a great number of metabolites however, it could be predicted that no single commodity might be rich in all these constituents. This chapter describes the general characteristics of the components of fruits and vegetables, related to their benefits as food sources. II. Traditional components A. Water The most abundant single component of fruits and vegetables is water, which may account for up to 90% of the total mass. The maximum water content varies between individual fruits and vegetables, because of structural differences. Cultivation conditions that influence structural differentiation may also have a marked affect. B. Organic acids There are two types of acids, namely aliphatic (straight chain) and aromatic acids. The most abundant acids in fruits and vegetables are citric and malic (both aliphatic) acids. However, large amounts of tartaric acid occur in grapes. Malic acid is the major component in oranges and apples. The acid content of fruits and vegetables generally decreases during maturation. For example, the citric acid content of clingstone peaches

219 Author s personal copy II. Traditional components 59 decreases faster than the malic acid content, while the malic acid content of apples and pears decreases faster than the citric acid content. Aromatic organic acids occur in several fruits and vegetables, but in very low concentrations. Benzoic acid occurs in cranberries, quinic acid in bananas and chlorogenic acid in potatoes. Organic acids play an important role in the sugar to acid ratio, which affects the flavor of fruits and vegetables. The distribution of acids within a fruit is not uniform. C. Proteins Proteins represent less than 1% of the fresh mass of fruit and vegetable tissues. Leguminous seeds are rich in protein, containing 15% to 30%. The proteins of fruits and vegetables are built from amino acids, but other related simple nitrogenous compounds also occur. Fruits, vegetables and legumes account for 1.2%, 5.5% and 6.1%, respectively, of the protein in the US food supply ( Hiza and Bente, 2007 ). Fruits are low in proteins, but tree nuts are a good source of high-quality proteins. The protein content of fresh fruits or vegetables is calculated by multiplying the total nitrogen content by a factor of This calculation uses the fact that protein is comprised of about 16% nitrogen, and the assumption that all nitrogen present is protein. The conversion ignores the fact that appreciable amounts of simple nitrogenous substances can be present in an uncombined form. In potatoes, 50% to 60% of the nitrogen occurs in the form of simple soluble constituents, while in apples the estimates range from 10% to 70% ( Salunkhe et al., 1991 ). Senescent tissues, such as those of overripe fruits, usually contain especially high proportions of non-protein nitrogen. Asparagine is abundant in potatoes and apples as non-protein nitrogen fractions. Pears and oranges are rich in proline, and black and red currants in alanine. D. Lipids and fatty acids Plant lipids represent a very broad group of compounds with functions that vary among products. Lipids are an energy source for plants during germination, forming components of cellular membranes and cuticular waxes, and they are mainly present as triglycerides (esters of glycerol and three fatty acids) or phospholipids (in which one fatty acid has been replaced by a phosphate group). Generally, most postharvest products are relatively low in total lipids, except for avocados, olives and many seeds. The fat content of fruits and vegetables is usually below 1% and varies with the product. Examples of fat content on a dry mass basis are: avocado: 35 70%; olive: 30 70%; grape: 0.2%; banana: 0.1%; and apple: 0.06%. Many of the physical and chemical properties of lipids are due to the fatty acids present in their structure. Fatty acids are aliphatic monocarboxylic acids that may be saturated or unsaturated to varying degrees. Saturated fatty acids do not contain any double bonds along the chain. Monounsaturated fatty acids have a single double

220 Author s personal copy 60 Nutritional Quality of Fruits and Vegetables bond in the hydrocarbon chain, and polyunsaturated fatty acids have more than one double bond. Fatty acids in plants usually range from 4- to 26-carbons in size, but oleic acid (18:1) and linoleic acid (18:2) are the most prevalent in nature. Olive oil and other fats high in monounsaturated fatty acids are becoming well-known for helping to lower LDL-cholesterol (the so-called bad cholesterol), while protecting HDL-cholesterol ( good cholesterol) when consumed in moderation in place of saturated fats. The difference among oils is not in their caloric content, but in their composition. Fats derived from animal sources (e.g. butter, cream, hard cheeses) have a high proportion of saturated fats, while oils from plant sources, such as olive and canola, have the lowest ( Table 5.1 ). Fatty acids are necessary for human bodily functions, where they are used primarily to produce hormone-like substances that regulate a wide range of functions including blood pressure, blood clotting, blood lipid levels, the immune response and the inflammatory response. The human body can produce most fatty acids, except for linoleic acid and α -linolenic acid, which are widely distributed in plant oils. These essential fatty acids are polyunsaturated fatty acid members of the omega-6 and omega-3 fatty acid series. Each double bond, depending on its geometry, can be in either a cis or a trans conformation. In cis bonds, the two carbons next to the unsaturated site bond atoms are oriented to the same side. Therefore, in restricted environments, such as when fatty acids are part of a phospholipid in a lipid bilayer or triglycerides in lipid droplets, cis bonds limit the ability of fatty acids to be closely packed and therefore, could affect the melting temperature of the membrane or the fat. A trans configuration, by contrast, means that the two carbons next to the double bond are oriented to opposite sides. As a result, they do not cause the chain to bend much, and their shape is Table 5.1 Fatty acid, vitamin E and cholesterol composition of some common dietary fats Saturated (%) Monounsaturated (%) Polyunsaturated (%) Cholesterol (mg 100 g 1 ) Animal fats Lard Butter Vegetable fats Coconut oil Palm oil Cottonseed oil Wheat germ oil Soya oil Olive oil Corn oil Sunflower oil Safflower oil Canola oil Source: Kays, S.J

221 Author s personal copy II. Traditional components 61 similar to straight saturated fatty acids. In plant sources, unsaturated fatty acids naturally occur in the cis form. Trans fatty acids might be present in some fats of animal origin, or might be the result of oil processing (e.g. hydrogenation of vegetable oils). The differences in geometry between the various types of unsaturated fatty acids, as well as between saturated and unsaturated fatty acids, play an important role in biological processes and in the construction of biological structures (such as cell membranes). Medical research suggests that amounts of trans fats correlate with circulatory diseases, such as atherosclerosis and coronary heart disease, more than the same amount of non- trans fats, for reasons that are not yet completely understood. E. Metabolizable carbohydrates After water, carbohydrates are the most abundant constituents in fruits and vegetables, representing 50% to 80% of the total dry weight. Carbohydrate functions include, among others, the storage of energy reserves and the make-up of much of the structural framework of cells. Simple carbohydrates, which are also the immediate products of photosynthesis, are important components of sensorial quality attributes. Carbohydrates, like proteins, yield 4 kcal g 1, while fats yield 9 kcal g 1. In many products, monosaccharides comprise a major portion of the total sugars. Glucose and fructose are the predominant forms of simple sugars found, especially, in fruits. Sucrose, the primary transport form of carbohydrate in most plants, is a disaccharide yielding glucose and fructose upon hydrolysis. Glucose, fructose and sucrose are water-soluble and together they comprise most of the sugars associated with the sweet taste of fruits and vegetables. The relative proportions of glucose and fructose vary from fruit to fruit and, to a lower extent, in the same fruit according to maturity. In many fruits (e.g. apple, pear, strawberry, grape) glucose and fructose are present in greater amounts than sucrose, but in certain vegetables, such as parsnip, beetroot, carrot, onion, sweet corn, pea and sweet potato, and in some ripe fruits such as banana, pineapple, peach and melon, the sucrose content is higher. Traces of other mono- and disaccharide sugars such as xylose, arabinose, mannose, galactose and maltose may also be present in small amounts ( Salunkhe et al., 1991 ). Some fruits of the Rosaceae family could also have significant levels of the sugar alcohol sorbitol. Total carbohydrate content also includes starches, which are organized into small grains, either within the chloroplasts or in some cases in specialized plastids (amyloplasts). Some non-starchy root vegetables, such as parsnip, beetroot and carrot, are relatively rich in simple sugars, containing between 8% and 18% of total carbohydrates. However, most vegetables contain smaller amounts of metabolizable carbohydrates. F. Dietary fiber Definition and composition Several definitions of fiber, either physiological or based on the measurement techniques used for its determination, have been put forward ( Slavin, 2005 ). An expert panel adopted the term dietary fiber consisting of non-digestible carbohydrates and lignin that are intrinsic and intact in plants ( Institute of Medicine, 2001 ).

222 Author s personal copy 62 Nutritional Quality of Fruits and Vegetables Dietary fiber includes very diverse macromolecules exhibiting a large variety of physico-chemical properties. The main components included as fiber are cellulose, hemicelluloses, pectins, lignin, resistant starch and non-digestible oligosaccharides. Cellulose is a cell wall polymer of β -1,4-linked glucose ( Brett and Waldron, 1996 ). Within the cell wall, the glucan chains are associated with hydrogen bonds to form assemblages highly resistant to degradation, known as microfibrils ( Carpita and McCann, 2000 ). In fruits and vegetables, the cell wall constitutes 1% to 2% of the fresh weight, and cellulose could be as much as 33% of that amount. In general, with the exception of avocado in which the whole cell wall seems to be degraded ( O Donoghue et al., 1994 ), little change in cellulose content occurs during ripening (Brummell, 2006 ). Hemicelluloses Several cell wall polymers soluble in alkalis are classified as hemicelluloses or cross-linking glycans ( Brummell and Harpster, 2001 ). Within the primary cell wall, hemicellulose levels are usually around 30% ( Carpita and McCann, 2000 ). The most common hemicellulose polymer in dicotyledonous species is known as xyloglucan, composed as cellulose of a backbone of β-1,4-linked glucose, but with lateral chains of the pentose xylose ( α-1,6 linked). These xylosyl residues can be modified further, with galactose, arabinose and/or fucose ( Brummell, 2006 ). Xylans are hemicellulosic compounds more abundant in monocotyledonous species, having a backbone of β -1,4-linked xylose which could be decorated with side chains of arabinose and/or glucuronic acid. Other hemicellulosic compounds usually less abundant include glucomannans, galactomannans and galactoglucomannans ( Carpita and McCann, 2000 ). Pectins Fruit tissues are particularly rich in pectins, which can account for up to 40% of the total cell wall polysaccharides. Pectins are also a diverse group of polymers rich in galacturonic acid ( Ridley et al., 2001 ). The most abundant pectic polysaccharide in the cell wall is homogalacturonan, a homopolymer of α-1, 4-linked galacturonic acid residues, with variable degrees of methyl esterification at C6 ( Willats et al., 2001 ). The degree of polymerization and the proportion of methyl esters affect the solubility of pectins. Pectins are deposited in the cell walls, with a high degree of esterification, and methyl ester usually decreases during ripening. Another modification commonly observed in several fruits during ripening is a reduction in pectin polymer size ( Brummell, 2006 ; Vicente et al., 2007b ). The extent of pectin depolymerization is variable, ranging from fruits such as avocado showing a dramatic downshift in polyuronide size ( Huber and O Donoghue, 1993 ) to products in which these changes are negligible, such as pepper or some berries ( Brummell, 2006 ; Vicente et al., 2007a ). Rhamnogalacturonan I (RG I) and rhamnogalacturonan II (RG II) are pectic polysaccharides which are also present in the plant cell wall. RG I has a backbone of alternating α -1,2-rhamnosyl and α-1,4-galacturonosyl residues ( Willats et al., 2001 ), with side chains rich in arabinose and galactose ( Carpita and McCann, 2000 ). Losses in the side chains are a common feature in fruit ripening, which can also affect pectin solubility and hydration potential ( Gross and Sams, 1984 ; Redgwell et al., 1997 ). RG II is the most complex polysaccharide present in the cell wall; it has the ability to form dimers via borate diester bonds ( O Neill et al., 2004 ; Kobayashi et al., 1996 ). Pectins, which are used in the commercial manufacture

223 Author s personal copy II. Traditional components 63 of jams and jellies, are extracted from certain fruits and vegetables such as citrus, apples and beets. Lignin is one of the most abundant biopolymers in nature ( Boerjan et al., 2003 ). It is an aromatic heteropolymer formed by the association of three hydroxycinnamyl alcohol derivatives (p-coumaryl, coniferyl and sinapyl alcohols) ( Reddy et al., 2005 ). Lignin is a highly resistant polymer present in secondary cell walls, and is associated with fibers and xylem vessels. In the case of fruits and vegetables, lignin content is relatively low. Resistant starch Starches are polysaccharides, composed of a number of glucose molecules linked together with α -D-(1-4) and/or α -D-(1-6) linkages ( Sajilata et al., 2006 ). Resistant starch consists of starch and its degradation products that are not digested in the small intestine ( Asp, 1994 ). Legumes are rich in resistant starch, and as much as 35% of their starch could escape digestion (Marlett and Longacre, 1996). Green bananas and potato are also relatively rich in resistant starch. Very little information is available about the resistant starch content of foods and the amount of resistant starch in a typical diet. Non-digestible oligosaccharides (NDOs) Oligosaccharides are low molecular weight carbohydrates intermediate in nature between simple sugars and polysaccharides ( Mussatto and Mancilha, 2007 ). While several oligosaccharides might be hydrolyzed in the digestive tract, others might resist the digestive process. Some of them include raffinose (trisaccharide composed of galactose, fructose, and glucose), stachyose (two galactose, one glucose and one fructose unit, linked sequentially) and verbascose (three galactose, one glucose and one fructose unit, linked sequentially). Legumes are rich in NDOs ( Mussatto and Mancilha, 2007 ). Benefits of fiber intake One of the most well known benefits of dietary fiber is the modulation of function of the intestinal tract ( Institute of Medicine, 2001 ). Meals rich in fiber promote satiety earlier, and are usually relatively low in calories compared to meals rich in other food types ( Marlett et al., 2002 ). Several works have also associated diets rich in dietary fiber with positive effects in disease prevention (see Institute of Medicine, 2001 ). Some works have established an inverse association between fiber intake and coronary disease ( Rimm et al., 1996 ; Wolk et al., 1999 ). Total fruit and vegetable consumption was inversely associated with colorectal cancer risk ( Terry et al., 2001 ). Current national dietary guidelines recommend an increased dietary fiber intake and suggest that fiber, independent of fat intake, is an important dietary component for the prevention of some diseases. Recommendations for adult dietary fiber intake generally fall in the range of 20 to 35 grams per day. The average fiber intake of adults in the US is less than half of this recommended level ( Marlett and Slavin, 1997 ). Sources of fiber Whole grains (especially the pericarp) and also fruits and vegetables are considered very good sources of fiber ( Anderson et al., 2007 ). In 2004, the primary contributors of fiber to the food supply were fruits and vegetables (37.1%), followed by grain products (36.0%) and legumes (13.3%) ( Hiza and Bente, 2007 ). Fiber content of

224 Author s personal copy 64 Nutritional Quality of Fruits and Vegetables Table 5.2 Fiber content in selected fruits, vegetables and nuts Product Dietary fiber (%) Almond 12.2 Apple 2.4 Asparagus 2.1 Avocado 6.8 Banana 2.6 Broccoli 2.6 Carrot 2.8 Kiwifruit 3.4 Lettuce 2.1 Onion 1.7 Orange 2.4 Pea 2.6 Peach 1.5 Peanut 8.5 Pear 3.1 Pepper 2.1 Pineapple 1.4 Plum 1.4 Potato 2.2 Prunes 7.1 Raisin 3.7 Spinach 2.2 Strawberry 2.0 Tomato 1.2 Walnut 6.7 Source: US Department of Agriculture, fruits and vegetables is usually in the range of 1% to 3% ( Table 5.2 ). Nuts, legumes and dried fruits have higher levels of fiber than fruits and vegetables. The nature of fiber varies among food sources. For instance, pectin is low in grains, but constitutes approximately 20% to 35% of the fiber in fruits, vegetables, legumes and nuts. Hemicelluloses account for about half of the total fiber in grains, and approximately 25% to 35% of the total fiber in other foods. Cellulose is one third or less of the total fiber in most foods ( Marlett, 1992 ). Besides total fiber content, some relevant properties include particle size and bulk volume, surface area characteristics, hydration and rheological properties, and adsorption or entrapment of minerals and organic molecules ( Guillon and Champ, 2000 ). The main modifications during storage of most fruits and vegetables occur because of changes in the solubility and molecular size of the cell wall constituents due to the action of several proteins ( Brummell, 2006 ; Fisher and Bennett, 1991 ). In some products, modification in fiber fractions could negatively affect quality. For instance, asparagus shows rapid hardening of the basal portions of the spears during storage related to modifications of fiber, such as

225 Author s personal copy II. Traditional components 65 increased deposition of lignin ( Saltveit, 1988 ). In general, preparation of fruits and vegetables by typical home methods or commercial processing does not seem to cause great loss of fiber ( Zyren et al., 1983 ). G. Vitamins Vitamins are organic molecules required in trace amounts for normal development, which cannot be synthesized in sufficient quantity by the organism and must be obtained from the diet. The term vitamin derives from the words vital amine because the first vitamin discovered (thiamine) contained an amino group. The 14 vitamins known today are vitamin A (retinol), B complex [B1 (thiamine), B2 (riboflavin), B3 (niacin), B5 (pantothenic acid), B6 (pyridoxine), B9 (folate/folic acid), biotin, choline and B12 (cyanocobalamine)] and vitamins C, D, E and K. They do not have common functions or structure and are usually grouped into fat-soluble (A, D, E and K) and water-soluble (B group and C) molecules. The vitamins present in fruits and vegetables make an important contribution to human nutrition, as they have specific functions in normal body performance. The vitamin content of fruits and vegetables shows a wide variation among species ( Salunkhe et al., 1991 ). Differences within cultivars occur, as well as between different batches of the same cultivar grown under different environmental and orchard conditions ( Rodriguez-Amaya, 2001 ; Lee and Kader, 2000 ). Vitamin A Carotenoids are liposoluble pigments responsible for the yellow, orange and red color of several fruits and vegetables. Carotenoids are terpenoids formed by eight isoprene units (2-methyl-1,3-butadiene) and derived from isopentenyl diphosphate. Those having an unsubstituted β -ring with an 11-carbon polyene chain have provitamin A activity ( Meléndez-Martínez et al., 2007 ), such as α-carotene, β-carotene and cryptoxanthin ( Kopsell and Kopsell, 2006 ). The structural requirement for vitamin A is satisfied by around 60 carotenoids ( Rodriguez-Amaya, 2001 ). Vitamin A plays an important role in vision, cell division and differentiation, bone development and reproduction. The average daily requirement for vitamin A for an adult is estimated at 5000 international units (1 IU 0.3 μg retinol or 0.6 μg β-carotene). Among this group there are, basically, two different classes: carotenes containing C and H (e.g. α-carotene, β -carotene, lycopene, etc.), and oxygenated derivatives known as xantophylls, such as lutein, violaxanthin and zeaxanthin. Carotenoids in plants have functions related to radiation interception, mainly in the blue green region of the spectrum, which may be transferred to the photosynthetic centers ( Kopsell and Kopsell, 2006 ). Moreover, these pigments protect the photosynthetic structures from excessive energy ( Grusak and Della Penna, 1999 ). They are usually present in low concentrations and their levels are highly variable among species. Fruits and vegetables account for only 30% of the vitamin A in the American diet ( Hiza and Bente, 2007 ). Vegetables that can supply useful amounts of carotene include carrots, pumpkins and squashes. Compared to vegetables, fruits are generally not as good a source of carotenoids, although there are a few notable exceptions such as apricot, mango, citrus, papaya

226 Author s personal copy 66 Nutritional Quality of Fruits and Vegetables Table 5.3 Carotene content (mean values) of selected fruits Product Carotene ( μ g 100 g 1 ) Mango 1800 Cantaloupe 1000 Pawpaw 810 Guava 435 Apricot 405 Plum 295 Watermelon 230 Source: Rodríguez-Amaya, and watermelon ( Table 5.3 ). Tomatoes and peppers also contain high levels of carotenoids. Their distribution is not usually uniform and in general, their accumulation is higher in the peel than in the pulp (Rodríguez-Amaya, 2001). To date, over 600 different carotenoids have been identified, but only a few of them are commonly found in produce. β -carotene, the most widely studied carotenoid, accumulates in carrots; lycopene is common in tomato and watermelon. Other pigments within this group include α-carotene, β -carotene, lutein, cryptoxanthin and zeaxanthin. In tomatoes, peaches and carrots the synthesis of carotene can continue after harvest. There is no difference between the carotene content of cooked vegetables and that of raw vegetables. Absorption of carotene can only be effective if the diet includes a minimum of 15% fat. The manner in which the food is prepared also determines the amount of carotene that will be absorbed. Homogenized carrots allow for the best absorption, followed by shredded carrots and whole carrots. Vitamin B complex Thiamine is required in the human body for the metabolism of carbohydrates. A daily intake of 1 2 mg is generally considered as necessary for a normal adult. Legumes are especially rich in thiamine. Compared with ascorbic acid, thiamine is relatively stable at cooking temperatures, especially in a slightly acidic solution. However, losses of 25% to 40% may occur during cooking. The average human requirement for riboflavin is estimated to be 1 2 mg per day. Green vegetables such as beans, beets, peppers and spinach are particularly rich in riboflavin. Starchy vegetables and fruits are relatively poor sources of riboflavin. Niacin, also known as nicotinic acid, is a precursor to NADH, NAD, NAD and NADP, which play essential roles in living organisms. A daily intake of 10 mg to 15 mg niacin is recommended. There is evidence that niacin can be synthesized in the body from tryptophan. Almonds are a rich source, but no fruits or vegetables can be singled out as being rich in niacin except perhaps, cape gooseberry and avocado. Niacin is relatively stable. Vitamin B 6 (pyridoxal phosphate) is a cofactor in many transamination, decarboxylation and deamination reactions (e.g. in plants, formation of ACC by ACC synthase requires pyridoxal phosphate as a cofactor) ( Ramalingam et al., 1985 ). Common

227 Author s personal copy II. Traditional components 67 symptoms of vitamin B 6 deficiency include dermatitis around the eyes, elbows and mouth, along with soreness of the mouth and a red tongue. It can also lead to dizziness, vomiting, weight loss and severe nervous disturbances ( Salunkhe et al., 1991 ). Vitamin B 6 is present in appreciable amounts in beans, cabbage, cauliflower, spinach, sweet potatoes, grapes, prunes, avocados and bananas. It is fairly heat stable. Pantothenic acid can be obtained from fresh, canned or frozen fruits and vegetables containing this vitamin if they are included in the diet. Pantothenic acid occurs widely in peas, beans, nuts, broccoli, mushrooms, potatoes and sweet potatoes. Symptoms of pantothenic acid deficiency in the diet include fatigue, headaches, sleep disturbance, tingling of hands and feet and lack of antibody production. Biotin is stable during cooking, processing and storage of fresh, canned and frozen fruits and vegetables. Deficiency leads to depression, sleeplessness and muscle pains. It is synthesized in the intestinal tract ( Salunkhe et al., 1991 ). Folic acid is essential for reproduction and normal growth. The vitamin is present in fruits, spinach, cabbage and other green vegetables. Lack of folic acid in the diet can cause a red tongue, diarrhea and anemia. Choline is heat-stable and occurs in dried legumes and vegetables. Choline deficiency in humans has never been reported. Vitamin B12 does not occur in fruits and vegetables. Because vitamins of the B group are water-soluble, leaching losses occur during cooking. Vitamin C Ascorbic acid (AsA) and its first oxidation product dehydroascorbic acid (which can be reduced in the human body) might be considered as vitamin C. AsA is a watersoluble carbohydrate-derived compound showing antioxidant and acidic properties due to the presence of a 2,3-enediol moiety ( Figure 5.1 ). Humans and a few other species are not able to synthesize AsA ( Chatterjee, 1973 ), because the gene coding for the last enzyme in the pathway (L-gulono-1,4-lactone oxidase) is nonfunctional ( Valpuesta and Botella, 2004 ). Plants synthesize AsA via a pathway that uses L-galactose as a precursor ( Smirnoff and Wheeler, 2000 ; Smirnoff, 2000 ). Another pathway using galacturonic acid, which might be recycled from cell wall pectin degradation, has been suggested in plants ( Agius et al., 2003 ). AsA has crucial biological functions in humans, such as its participation in collagen biosynthesis ( Murad et al., 1981 ). Even though nutritional deficiencies are rare in modern western cultures, it is generally recognized that dietary AsA also has important health benefits for the consumer, and an increased intake of vitamin C has been associated with a reduced incidence of some diseases and disorders ( Carr and Frei, 1999 ; Hancock HO HO O O HO OH Figure 5.1 Structure of ascorbic acid, a main antioxidant present in fruits and vegetables.

228 Author s personal copy 68 Nutritional Quality of Fruits and Vegetables and Viola, 2005 ). Furthermore, in meat-poor diets, dietary AsA can contribute to the improved uptake of iron ( Frossard et al., 2000 ). The recommended dietary allowance of vitamin C for men is 75 mg daily, while the recommended dietary allowance for young women is higher, at 90 mg daily ( Levine et al., 2001 ). Fruits, vegetables and juices are the main dietary sources of vitamin C. Fruits and vegetables account for 90% of the vitamin C in the US food supply ( Hiza and Bente, 2007 ). Its concentration depends on the product considered ( Noctor and Foyer, 1998 ), ranging from 1 to 150 mg 100 g 1 fresh weight (FW) ( Lee and Kader, 2000 ). Vitamin C is present in fresh fruits and vegetables, as well as in fruit juices. Fruits, particularly tropical species, and leafy vegetables are rich in ascorbic acid. Rosehip, jujube and guava have very high levels of ascorbic acid. Other good sources of AsA include persimmon, strawberry, kiwifruit, peppers, and citrus fruit, and spinach, broccoli and cabbage among vegetables ( Table 5.4 ). Wide variations in vitamin C content also exist within cultivars. For instance, AsA content in Actinidia deliciosa fruit varies from 29 to 80 mg 100 g 1 FW, depending on the cultivar ( Nishiyama et al., 2004 ). Even more dramatic variations were found in berry fruits, with levels of AsA ranging from 14 to 103 mg 100 g 1 FW among cultivars of raspberry, blackberry, red currant, gooseberry and cornelian cherry ( Pantelidis et al., 2007 ). For any given product, the levels of AsA are highly variable, depending on genetic and environmental factors (reviewed in Lee and Kader, 2000 ). A main environmental factor determining the level of ascorbic acid is radiation interception. In general, the greater the amount of sunlight received during growth, the higher the ascorbic acid content. The retention of AsA is also markedly affected by storage and processing. Potatoes lose up to 75% to 80% of the original levels over nine months of storage. In most cases, other fruit and vegetable AsA levels decline during storage, because the losses are accelerated by storage at high temperatures. Bruising and mechanical damage greatly increase the rate of ascorbic acid loss. Ascorbic acid is highly susceptible to oxidation, either directly or through the enzyme ascorbate oxidase catalyzing the oxidation of AsA to dehydroascorbic acid, with the concomitant reduction of molecular oxygen to water ( Sanmartin et al., 2007 ). Ascorbic acid can even be oxidized during eating, while food is being chewed. However, it is important to consider that the first breakdown product of AsA, Table 5.4 Vitamin C content (mean values) of selected fruits Product Vitamin C (mg 100 g 1 fresh weight) Guava, raw 184 Kiwi, raw 118 Litchi, raw 72 Pawpaw, raw 62 Strawberry, raw 57 Citrus fruits Cantaloupe 42 Source: Salunkhe et al., 1991.

229 Author s personal copy III. Antioxidants in fruits and vegetables 69 dehydroascorbic acid, still has vitamin C activity and all activity is lost if oxidation proceeds beyond this stage ( Salunkhe et al., 1991 ). When vegetables are cooked before eating, high losses of vitamin C can occur. For instance, starchy vegetables may lose between 40% and 80% of their vitamin C during cooking, because of leaching and oxidation. Loss of vitamin C can be reduced by steaming or by placing the vegetables directly into boiling water. Freezing reduces vitamin C slightly, but at the end of long-term frozen storage (12 months), a significant decrease (33% to 55%) in vitamin C can occur ( de Ancos et al., 2000 ). Vitamin E Vitamin E includes tocopherols and tocotrienols. They can be in eight different forms (four tocopherols and four tocotrienols). All the isomers have aromatic rings with a hydroxyl group that can donate hydrogen atoms to reduce reactive oxygen species (ROS). The different isomers are named alpha ( α ), beta ( β ), gamma ( γ ) and delta ( δ ), and this is related to the number and position of methyl groups in the ring. Each of the forms has its own vitamin E activity, α -tocopherol being the most active (see Figure 5.2 ). Vitamin E deficiency results in stunted growth. In general, vitamin E levels are more abundant in oily seeds, olives, nuts, peanuts, avocados and almonds. Even though the levels of tocopherol in broccoli and leafy vegetables are lower than in fat-rich products, they are good sources compared to other fruits and vegetables. Vitamin E is highly susceptible to oxidation during storage and processing. Vitamins D and K Vitamin D is a group of fat-soluble compounds. The main forms of vitamin D are ergocalciferol and cholecalciferol. It occurs only in trace amounts in fruits and vegetables. Vitamin K is essential for blood coagulation, but dietary deficiency is uncommon. The recommended daily intake is 120 μ g. It occurs abundantly in lettuce, spinach, cauliflower and cabbage. As well as direct intake, it can also be produced by bacteria in the intestines. III. Antioxidants in fruits and vegetables A. Oxidative damage and antioxidants Imbalance in the production of reactive oxygen species (ROS) leading to negative cellular alterations is known as oxidative damage, which is caused by several molecules HO O Figure 5.2 Structure of tocopherol.

230 Author s personal copy 70 Nutritional Quality of Fruits and Vegetables ( Mittler, 2002 ). Reactive oxygen species are partially reduced forms of oxygen such as singlet oxygen, hydrogen peroxide (H 2 O 2 ), superoxide (O 2 ) or hydroxyl radical (OH ) ( Asada, 1999 ). Some, but not all of the components able to cause oxidative damage are free radicals (i.e. molecules with unpaired electrons, which determine their high reactivity). Currently, there is overwhelming evidence showing that the ROS can alter proteins, lipids and nucleic acids, causing deleterious modifications to normal metabolism, which can lead to several disorders and diseases ( Waris and Ahsan, 2006 ), and eventually to cell death ( Jeremy et al., 2004 ). From a biological perspective, an antioxidant is considered as any compound able to oppose cellular oxidation. Diets rich in fruits and vegetables have been shown to reduce the incidence of cardiovascular disease and some chronic and degenerative diseases associated with oxidative damage ( Ames et al., 1993 ; Dragsted, 2003 ). The incorporation of fruits and vegetables in the diet may also help to eliminate certain toxins. The protective effects have been associated with the presence of antioxidant compounds ( Cao et al., 1996 ; Wang et al., 1996 ). Antioxidants are present in all plant organs and include ascorbic acid, carotenoids, vitamin E and phenolic compounds, among others (Larson, 1988 ) (Figure 5.3 ). Here we briefly describe some characteristics of these components. B. Ascorbic acid As mentioned before (see Section II.G) ascorbic acid is one of the most important compounds for human nutrition present in fruits and vegetables. The role of AsA in disease prevention has been associated with its capacity to neutralize ROS. Main antioxidants in fruits and vegetables Ascorbic acid Carotenoids Vitamin E Phenolics Others Carotenes Xanthophylls Tocopherols Tocotrienols Phenolic acids Benzoic acids Cinnamic acids Sulfur antiox. Flavonoids Flavonols Flavones Isoflavones Flavanols Flavanones Anthocyanidins Proanthocyanidins Others Coumarins Stilbenes Lignans and lignin Figure 5.3 Main dietary antioxidants present in fruits, vegetables and legumes.

231 Author s personal copy III. Antioxidants in fruits and vegetables 71 C. Carotenoids Fruits and vegetables are the main sources of carotenoids in the diet ( Rao and Rao, 2007 ). The presence of conjugated double bonds in carotenoids has a main role in determining their antioxidant properties ( Sandmann, 2001 ). In the last few years, carotenoids have received great attention due to their antioxidant properties and potential to prevent some diseases. The general properties of these compounds were described in Section II.G. D. Tocopherols and tocotrienols These include the fat-soluble compounds grouped as vitamin E, characterized by a high antioxidant capacity. Their distribution in fruits and vegetables was previously described (see Section II.G). E. Phenolic compounds This group encompasses a great diversity of compounds derived from the aromatic amino acids phenylalanine and tyrosine. Their main functions are acting as deterrents of potential predators or antimicrobials, protecting against UV-radiation and contributing to the pigmentation of fruits and flowers. Phenolic compounds can contribute to the astringency and bitter taste of some products. They are generally present in low concentrations, but in certain cases, such as in blueberry, they can reach levels of more than 0.1%. In general, they also accumulate in the peel more than in the pulp of fruits. The general characteristic of the compounds within this group is to have aromatic rings with variable degrees of hydroxylation ( Mattila et al., 2006 ). Phenolic compounds are easily oxidized to quinones. The beneficial properties of berry fruits on human health have been associated in part with the presence of relatively high levels of phenolic compounds ( Seeram et al., 2006 ). There is in vitro evidence showing that these compounds could influence several cellular processes. Information regarding the metabolism and effect in vivo is much more limited ( Duthie et al., 2003 ). A large number of phenolic compounds have been identified in plants ( Tsao and Deng, 2004 ). They have been subdivided into different subclasses, such as phenolic acids, flavonoids and other compounds (e.g. lignans, stilbenes, tannins, coumarins and lignin). Phenolic acids Phenolic acids include derivatives of benzoic and cinnamic acid ( Benbrook, 2005 ) ( Figure 5.4 ). The most common benzoic acid derivatives are p-hydroxybenzoic, vanillic, syringic and gallic acid, while common cinnamic acid derivatives include p- coumaric, caffeic, ferulic and sinapic acid. The derivatives differ in the degree of hydroxylation and methoxylation of the aromatic ring. Caffeic acid is the most abundant phenolic acid in several fruits such as berries ( Mattila et al., 2006 ), while coumaric acid is usually present in lower proportions ( Rice-Evans et al., 1997 ). Ferulic acid represents 90% of total phenolic acids in cereals ( Manach et al., 2004 ; Scalbert and Williamson, 2000 ). The contribution of each of the phenolic compounds to the

232 Author s personal copy 72 Nutritional Quality of Fruits and Vegetables O O OH OH Figure 5.4 Structure of benzoic acid (left) and cinnamic acid (right), precursors of the two main classes of phenolic acids present in fruits and vegetables. O O OH O O Figure 5.5 General structure of flavones (left) and flavonols (right). antioxidant capacity depends on their structure. For instance, the number of hydroxyls present in the molecule can increase the antioxidant capacity. Flavonoids Flavonoids represent a large group of phenolic compounds with two aromatic rings in their structure that are associated together by a 3C-oxygenated heterocycle. Phenolic compounds are usually present as glycosides, which reduce their activity against free radicals and increase their solubility. At the cellular level, they are compartmentalized in the vacuoles ( Rice-Evans et al., 1997 ). There are different classes of flavonoids ( Le Marchand, 2002 ) such as: a) fl avones and flavanols; b) fl avanones, flavanols; c) isoflavones; d) proanthocyanidins; and e) anthocyanidins. Flavones and flavonols Flavonols have a central ring of 3-hydroxypyran-4-one ( Rice-Evans et al., 1997 ). Flavones lack the OH in position 3 ( Figure 5.5 ). Rutin, luteolin and apigenin are common among flavones, while the most abundant flavonols are quercetin and kampferol ( Manach et al., 2004 ). Onions are rich in these compounds. Blueberries also have high levels, especially in the peel, because synthesis is stimulated by exposure to light. Celery is a good source of flavones. Flavones are also present in citrus, but they are associated mainly with the fruit peel. Flavanones and flavanols Flavanones do not have the double bond in position 2,3 of the central ring, while flavanols lack the carbonyl group at position 4 ( Figure 5.6 ). The genus Citrus is characterized by the accumulation of flavanone glycosides.

233 Author s personal copy III. Antioxidants in fruits and vegetables 73 O O OH O Figure 5.6 General structure of flavanones (left) and flavanols (right). Orange juice is a source of the flavanone glycoside hesperidin ( Tripoli et al., 2007 ). The flavanols catechin and epicatechin are common in grapes ( Rice-Evans et al., 1997 ). Isoflavones Isoflavones are phytoestrogens present in legumes. Soybean products are a good source of these compounds ( Manach et al., 2004 ). The three most commonly found isoflavones are genistein, glycitein and daidzein. Proanthocyanidins Proanthocyanidins are oligomeric flavonoids (usually dimers or oligomers of the flavanols catechin and epicatechin). They are common in the peel and seeds of grapes ( Gu et al., 2004 ). Other sources of these compounds include apple, almond and blueberry. Anthocyanidins Anthocyanidins are pigments giving several fruits their characteristic red or purple colors, although in some conditions they can be uncolored. Besides being pigments, anthocyanidins have great relevance due to their contribution to the antioxidant capacity of fruits and vegetables. The basic structure of anthocyanidins is derived from the flavilium cation (2-phenyl-benzopyril). There are six anthocyanidins more common in fruits and vegetables: pelargonidin, cyanidin, delphynidin, peonidyn, petunydin and malvidin. The differences between them are the OH, H and OCH 3 groups associated with the phenolic rings. The distribution of hydroxyls in the molecule influences the antioxidant capacity of the different anthocyanidins. These compounds are usually present as glycosides associated with different sugars, since anthocyanidin glycosylation reduces antioxidant capacity relative to the free aglycons. Others Lignans are diphenolic structures formed by the association of two derivatives of cinnamic acid ( Liu, 2007 ). They are present mainly in linseeds, cereals and legumes, but their levels are low in fruits and vegetables. Stilbenes are also phenolic compounds described in fruits. The most studied compound in this group is resveratrol ( Figure 5.7 ). This compound has been known for quite a while, and is commonly produced in response to pathogens and other stress conditions in grapes ( Langcake and Pryce, 1976 ). It has also been identified in other fruits, such as blueberry. It has been suggested that it may have anticarcinogenic properties. Finally, lignin is a phenolic polymer present in secondary cell walls of plant tissues. It is highly hydrophobic and is formed by three main monomeric precursors: coumarylic, sinapylic and coniferyl alcohols. It is associated with conduction tissues

234 Author s personal copy 74 Nutritional Quality of Fruits and Vegetables HO OH OH Figure 5.7 Resveratrol has been studied in detail in grapes. It has been suggested that this compound has anticarcinogenic properties. Factors affecting the levels of antioxidants in fruits Genetic Environmental Preharvest Harvest Postharvest Species Variety Radiation Stress during development (water, fertility, pathogens, etc.) Maturity Handling Storage Postharvest treatments (UVC, ozone) Processing Figure 5.8 Main factors affecting the level of antioxidants in fruits. (xylem vessels, sclereids, tracheids), and in general it is not abundant in fruits and vegetables. Its contribution, from the antioxidant point of view, is associated only with the products of its potential degradation that are very limited. F. Factors affecting the levels of antioxidants in fruits and vegetables Several factors influence the accumulation and degradation of antioxidant compounds in fruits. In general terms, these variables could be divided into genetic and environmental. Different factors are included within each of these groups ( Figure 5.8 ). Genetic factors Species The species is the first factor determining the prevalence of different antioxidants. Although there are some exceptions, each group is characterized by the accumulation of certain types of antioxidants ( Table 5.5 ). Berries are particularly rich in phenolic compounds ( Zheng and Wang, 2003 ) and vitamin C ( Kevers et al., 2007 ). The main antioxidants in this group seem to be phenolics because, in general, a good correlation between total antioxidant capacity and phenolic compounds has been

235 Author s personal copy III. Antioxidants in fruits and vegetables 75 Table 5.5 Fruits and vegetables rich in the different groups of antioxidants Ascorbic acid Vitamin E Carotenoids Phenolics Strawberry Almond Pineapple Blueberry Pepper Corn Plum Plum Kiwifruit Broccoli Peach Raspberry Orange Spinach Pepper Strawberry Pepper Peanut Mango Apple Broccoli Avocado Melon Blackberry Guava Tomato Rosehip Carrot Persimmon observed. In the case of ripe blueberry, ascorbic acid only contributes 0.4% to 9.0% to the total antioxidant capacity ( Kalt et al., 1999 ). Cultivar For a given species, the levels of antioxidants are also markedly affected by the cultivar considered. For instance, in strawberry, Nelson and co-workers (1972) found variations from 19 to 71 mg of ascorbic acid per 100 g FW in six varieties. Similar differences among varieties have been found for phenolic compounds ( Wang and Lin, 2000 ). The identification of lines or mutants enriched in antioxidants might be useful in breeding programs aimed at improving the nutritional value of fruits and vegetables. The identification of the nature of the genes mutated in lines with altered accumulation of antioxidants might also be of great value. For instance, in the case of tomato the characterization of the high pigment ( hp ) mutants, which accumulate high levels of carotenoids, showed that the mutated gene is associated with plant light responses, and the over-expression of this gene resulted in increased accumulation of carotenoids ( Liu et al., 2004 ). Also, in tomato the level of β -carotene and lycopene were raised by increasing the expression of phytoene synthase and lycopene cyclase, respectively ( Fraser et al., 2002 ; D Ambrosio et al., 2004 ). Similarly in carrot, the over-expression of a β -carotene ketolase isolated from Haematococcus pluvialis led to the accumulation of the ketocarotenoid astaxanthin ( Jayaraj et al., 2008 ). The generation of transgenic plants has also been seen to increase the levels of other antioxidants such as phenolic compounds. Transformation of tomato with a Petunia gene for chalcone isomerase increased the concentration of flavonols in the peel almost 80 times, without altering other phenotypic characteristics ( Muir et al., 2001 ). In the case of ascorbic acid, the elucidation of its biosynthetic pathway opened the way to manipulate ascorbate biosynthesis in plants ( Smirnoff, 2000 ). However, while most of the genes proposed to be involved in these pathways have been cloned and expressed in various plant species, transformation strategies to increase AsA concentrations have had only limited success. Thus, there is a need for alternative approaches to identify the genetic determinants underlying whole plant AsA homeostasis. Environmental factors Radiation In many cases, modifications in the level of phenolic compounds, ascorbic acid and carotenoids have been associated with changes in the radiation interception

236 Author s personal copy 76 Nutritional Quality of Fruits and Vegetables in the field. Sun-exposed sides of fruits have higher levels of phenolics and vitamin C than shaded regions ( Lee and Kader, 2000 ). In the case of leafy vegetables, the levels of flavonols are 10 times higher in the surface leaves than in the internal leaves. In tomato, the level of total phenolics increased twofold in plants exposed to higher irradiance. Similarly, these plants presented higher levels of carotenoids and ascorbic acid ( Gautier et al., 2008 ). This illustrates that maximization of radiation interception is important to obtain products with higher accumulation of antioxidants. However, the optimal irradiance levels required to maximize accumulation of the different groups of antioxidants in fruits and vegetables are not well established. Cultural practices There are several works analyzing the effect of cultural practices on the level of different groups of antioxidants. For instance, strawberry fruit grown with plastic mulch had higher antioxidant capacity than fruits from plants grown in beds without plastic mulch ( Wang et al., 2002 ). High nitrogen fertilization has been associated with reduced levels of ascorbic acid ( Lee and Kader, 2000 ), and compost as a soil supplement significantly enhanced levels of ascorbic acid ( Wang and Lin, 2003 ). Vitamin C accumulation also has been inversely correlated with rainfall ( Toivonen et al., 1994 ). Some authors have found evidence suggesting that organic products might accumulate higher levels of antioxidants and vitamins than those produced conventionally ( Woese et al., 1997 ; Weibel et al., 2000 ; Asami et al., 2003 ; Chassy et al., 2006 ). However, there are also studies that show either results that are opposite, or results that show no difference ( Barrett et al., 2007 ). Winter and Davis (2006) concluded that it is not possible to ensure that, from a nutritional point of view, organically grown products are superior to those obtained by conventional agricultural techniques. Maturity at harvest The developmental stage might affect the antioxidant capacity of fruits ( Prior et al., 1998 ). The nature of these changes depends on the product considered. For instance, in tomato and pepper total antioxidant capacity increases because of the accumulation of carotenoids and vitamin C. In the case of blueberry fruit the concentration of phenolic acids decreases during ripening, while anthocyanins are accumulated ( Castrejón et al., 2008 ), resulting in a net reduction of total antioxidant capacity during development. Similar patterns have been observed in strawberry and blackberry ( Wang and Lin, 2000 ). In the case of carotenoids, in some products (e.g. pepper, tomato, mango) the concentration increases during development ( de Azevedo and Rodriguez-Amaya, 2005 ). In contrast, products in which color is mainly associated with the accumulation of anthocyanins or products that maintain their green color at harvest usually show a reduction in the level of carotenoids as development progresses ( Rodriguez-Amaya, 2001 ). Wounding Mechanical damage may cause alterations in the levels of antioxidants. In the case of AsA, cellular breakage causes an increase in the levels of the internal pressure of oxygen favoring oxidation. Carotenoid degradation is also accelerated by oxygen, but the stability of these compounds is higher than that of AsA. In the case of phenolic compounds, wounding could alter both their synthesis and degradation ( Tomás-Barberán et al., 1997 ; Loaiza Velarde et al., 1997). In lettuce, wounding led to the accumulation of soluble phenolic compounds (e.g. chlorogenic acid) ( Choi et al., 2005 ). From a molecular perspective, wounding has been shown to induce de novo

237 Author s personal copy III. Antioxidants in fruits and vegetables 77 synthesis of phenylalanine ammonia lyase, a key enzyme in phenylpropanoid metabolism ( Choi et al., 2005 ). Besides its role on phenolic biosynthesis, wounding also affects degradation. First, also in response to wounding, an increase in enzymes associated with phenolics oxidation such as polyphenol oxidases (PPOs) and peroxidases (PODs) has been reported. In addition, cell disruption allows direct contact between pre-existing phenolic degrading enzymes. Finally, the production of hydrogen peroxide upon damage provides a secondary substrate of PODs and the reduction of barriers for oxygen diffusion might favor PPO activity. This might promote the oxidation of phenolics, which can then polymerize, leading to the formation of brown colored pigments that may ultimately reduce quality. Consequently, careful handling and minimization of physical damage is recommended. Storage The effect of storage on antioxidants in many cases is related to the role of ethylene in the ripening process. Consequently, the final effect on antioxidant (AOX) levels will depend on the typical modifications observed during development of the species considered. In some cases, ethylene can induce specific antioxidants. For instance, in carrot, ethylene stimulated the accumulation of an isocoumarin (6-methoxymellein). In berries it has been observed that atmospheres with high levels of oxygen (60% and 100%) result in increased antioxidant capacity by favoring anthocyanins and other phenolics accumulation ( Zheng et al., 2003 ). However, the oxidation of AsA might also be favored in these conditions. Besides the effect of any specific group of antioxidants in most fruits, it has been observed that the changes in total antioxidant capacity are not dramatic during postharvest storage. Excluding some products, such as broccoli and banana, fruits and vegetables, in general, lose their visual quality before marked losses in total antioxidants occur ( Kevers et al., 2007 ). In some cases, an increase in total AOX capacity is observed, basically, associated with the accumulation of phenolics. In strawberry, storage at 5 C and 10 C increased the antiradical capacity ( Ayala-Zavala et al., 2004 ). Further studies to evaluate the extent of this increased accumulation of antioxidants in some fruits might be done to determine the potential for increasing the functionality of fruits and vegetables through manipulation of the postharvest environment ( Kalt et al., 1999 ). Other treatments Some studies suggest that manipulation of the metabolism of products by the application of postharvest treatments could be useful to increase the antioxidant capacity, with consequent nutritional benefit ( Kalt et al., 1999 ). Phenolic compounds synthesis might be triggered in response to stress conditions, such as infection by microorganisms or wounding, ultraviolet (UV) irradiation or the exposure of the products to ozone-enriched atmospheres. In grapes, postharvest UV-C and ozone treatments increased the accumulation of resveratrol ( Cantos et al., 2001 ; Versari et al., 2001 ; Gonzalez-Barrio et al., 2006). The elicitation of the accumulation of antioxidant compounds has also been observed in other fruits. In blueberry cv. Bluecrop, besides reducing decay, UV-C radiation exposure (2 or 4 kj/m 2 ) resulted in increased accumulation of anthocyanins and higher levels of antioxidants ( Perkins- Veazie et al., 2008 ). In the case of strawberry, UV-C treatments also increased the level of phenolic compounds and the antiradical capacity ( Ayala-Zavala et al., 2004 ). These results, at a laboratory scale, show an interesting eliciting effect of some

238 Author s personal copy 78 Nutritional Quality of Fruits and Vegetables postharvest treatments on antioxidant accumulation. Further studies would be useful to determine the potential of these strategies on a commercial scale. Processing The effect of processing on the level and bioavailability of antioxidants depends on the treatment intensity, as well as on the component considered ( Bernhardt and Schlich, 2006 ). In some cases, processing could lead to higher availability of antioxidants, due to an increase in the ease of extractability. For instance, with carrot and spinach carotenoids vapor cooking increases assimilation, probably due to a disruption of carotenoid-protein complexes. Similarly, the bioavailability of lycopene increases in heat-treated tomato. However, cooking could cause the isomerization of β -carotene, leading to the formation of cis isomers with lower provitamin A activity ( Deming et al., 2002a,b ). For instance, in the case of fresh carrots, 100% of the β -carotene is present in the trans form, while canning results in a significant formation of cis isomers. Carotenoids are in general susceptible to oxidation. Heat, light and oxygen could accelerate their degradation ( von Elbe and Schwartz, 1996 ). Minimizing the influence of these factors could reduce carotenoids loss. Ascorbic acid is one of the antioxidants more susceptible to degradation. Blanching or even freezing and thawing could cause losses up to 25%. More drastic treatments could lead to losses of up to 90% of AsA. Some of the factors affecting the loss of AsA include the degree of heating, the exposed surface (which affects lixiviation in the cooking media), oxygen levels and product ph ( Eitenmiller and Landen, 1999 ). The stability of AsA could be increased at low ph, reduced oxygen pressure, darkness and presence of chelating agents. Consumption in the fresh state is the best way to minimize AsA losses. Finally, processing can also cause losses of phenolic antioxidants. For instance, peeling or cutting reduces quercetin levels by only 1%, but cooking in water may reduce the content of this component by 75%. IV. Fruits and vegetables as direct sources of minerals Dietary minerals raise concern for health specialists and consumers, due to the number of processes they are involved in and the continuous research highlighting the benefits of their adequate and balanced intake. Although there is no universally accepted definition or classification, the dietary focus on minerals derives from an interest in supporting the biosynthetic apparatus with required elemental components other than carbon, hydrogen and oxygen. Total mineral content is determined by the ash value. Nevertheless, classification of many elements as essential minerals for human nutrition is not definitive, and there is still debate as to the natural biological role of vanadium, chromium, boron, aluminum and silicon in human health. Minerals are normally classified as macroor micronutrients, based on the relative concentration of each nutrient when those concentrations are adequate for normal tissue function. Macronutrients include potassium (K), calcium (Ca), magnesium (Mg), nitrogen (N), and phosphorus (P), and their concentrations in plant tissues range from 1000 to μ g per gram of dry weight. In contrast, the concentrations of micronutrients usually found in plant

239 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 79 tissues are 100- to fold lower than those of macronutrients. Mineral micronutrients considered essential in human nutrition include manganese (Mn), copper (Cu), iron (Fe), zinc (Zn), cobalt (Co), sodium (Na), chlorine (Cl), iodine (I), fluorine (F), sulfur (S), and selenium (Se). Macronutrients can also be classified into those that maintain their identity as ions within plant tissues (e.g. K, Ca2 and Mg 2 ), and those that are assimilated into organic compounds (e.g. N and P). In general, vegetables are a richer source of minerals than fruits, but both vegetables and fruits are considered nutrient-dense foods in that they provide substantial amounts of micronutrients, such as minerals and vitamins, but relatively few calories (Table 5.6 ). Minerals have both direct and indirect effects on human health. The direct effects of minerals focus on the consequences of their consumption on human nutrition, while the indirect effects refer to their incidence in fruit and vegetable quality and subsequent consumer acceptance. From a direct nutrition standpoint, potassium has the biggest presence in both fruits and vegetables, but nitrogen and calcium show major impacts on horticultural crop quality. Table 5.6 Fruit and vegetable sources of potassium, ranked by milligrams of potassium per standard amount, also showing calories in the standard amount *. The dietary reference intake (DRI) for potassium for adults and adolescents is 4700 mg/day. Fruits and vegetables, standard amount Potassium (mg) Calories Sweetpotato, baked, 1 potato (146 g) Tomato paste, ¼ cup Beet greens, cooked, ½ cup Potato, baked, flesh, 1 potato (156 g) White beans, canned, ½ cup Tomato puree, ½ cup Prune juice, ¾ cup Carrot juice, ¾ cup Lima beans, cooked, ½ cup Winter squash, cooked, ½ cup Banana, 1 medium Spinach, cooked, ½ cup Tomato juice, ¾ cup Tomato sauce, ½ cup Peaches, dried, uncooked, ¼ cup Prunes, stewed, ½ cup Apricots, dried, uncooked, ¼ cup Cantaloupe, ¼ medium Honeydew melon, 1/ 8 medium Plantains, cooked, ½ cup slices Kidney beans, cooked, ½ cup Orange juice, ¾ cup Split peas, cooked, ½ cup * US Department of Health and Human Services and US Department of Agriculture, 2005.

240 Author s personal copy 80 Nutritional Quality of Fruits and Vegetables Until recently, nutrition research focused on single-mineral impact on human health, generally with incongruent results. The recognition that minerals are not consumed individually, but as combined constituents of a varied diet, has shifted the efforts in this area to unraveling the role of the overall diet, or dietary patterns, in blood pressure and cardiovascular diseases, bone diseases and a range of chronic disorders. Epidemiological surveys suggest that the total diet has a greater influence on health than do specific components. From these dietary pattern studies, it has become increasingly clear that it is not merely the excess or deficiency of a single mineral, but also deficiencies of multiple nutrients in combination that have the greatest dietary effects on health. Adequate intake of minerals such as potassium specifically derived from foods such as horticultural crops, where they coexist with other essential nutrients contributes to overall health. As described in previous sections, fruits and vegetables provide a milieu of phytochemicals, non-nutritive substances that possess health protective benefits. In contrast, fruits and vegetables may not usually be recognized as primary sources of mineral intakes from a nutritional point of view ( Fairweather-Tait and Hurrell, 1996 ). Nevertheless, the Dietary Approaches to Stop Hypertension (DASH) emphasize fruit, vegetable and low-fat dairy product consumption as a source of minerals. In the DASH dietary pattern, vegetables contribute an average of 14.3%, 15.5%, 16.2% and 10.4% to the intakes of calcium, magnesium, potassium and zinc, respectively ( Lin et al., 2003 ). Correspondingly, fruits and juices contribute an average of 5.8%, 17.3%, 33.0% and 6.6% ( Lin et al., 2003 ). There has been a natural trend towards lower mineral content in fruits and vegetables over the past decades ( Mayer, 1997 ; Ekholm et al., 2007 ) which have not been fully compensated for by the increase in fruit and vegetable consumption. Vegetable contribution of potassium, phosphorus, magnesium, calcium, copper, iron and zinc to the US food supply significantly decreased during the last century, while fruit contribution of potassium, phosphorus, magnesium and copper increased ( Table 5.7 ). Table 5.7 Minerals (%) contributed from fruits and vegetables to the US food supply in selected years Fruit Vegetables Mineral Year/s Year/s Potassium Calcium Phosphorus Magnesium Copper Iron Zinc Sodium Selenium Source: Hiza and Bente, 2007.

241 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 81 Nowadays different postharvest strategies for improving the mineral intake from fruits and vegetables are being implemented. These comprise increasing consumption of fruits and vegetables and increasing levels of essential nutrients through fortification methods. Alternative approaches include improving nutrient bioavailability and retention. A. General considerations of selected minerals Potassium (K) A potassium-rich diet contributes to lower blood pressure, blunting the effects of salt ( Salunkhe et al., 1991 ). Inadequate levels of potassium intake have long been associated with higher blood pressure ( McCarron and Reusser, 2001 ). Potassium also regulates heartbeat, assists in muscle contraction and is needed to send nerve impulses and to release energy from fat, carbohydrates and protein. Different nutrients and phytochemicals in fruits and vegetables, including potassium, may be independently or jointly responsible for an apparent reduction in cardiovascular disease risk ( Ignarro et al., 2007 ). Potassium is a systemic electrolyte and is essential in coregulating ATP with sodium. Potassium favorably affects acid base metabolism, which may reduce the risk of developing kidney stones ( Zerwekh et al., 2007 ), and possibly decrease bone loss with age. Although calcium intake is an important determinant in peak bone mass, and in retarding bone loss in postmenopausal women, findings of higher bone mass and lower bone resorption in women consuming high intakes of potassium, magnesium, zinc and vitamin C emphasizes the importance of considering the impact of variation in other nutrients when focusing on a particular mineral ( Cohen and Roe, 2000 ). In fact, up to 11 different groups of compounds (vitamins, minerals, antioxidants and others) in fruits and vegetables could influence bone health ( MacDonald, 2007 ). Potassium is the most abundant individual mineral element in fruits and vegetables. It normally varies between 60 and 600 mg per 100 g 1 of fresh tissue. It plays a role in a myriad of cellular and whole plant functions: it serves as an osmoticum for cellular growth and stomatal function, balancing the charges of anions, activating almost 60 plant enzymes and participating in numerous metabolic processes, including protein synthesis, oxidative metabolism and photosynthesis. In fruits and vegetables, potassium occurs mainly in combination with various organic acids. Examples of potassium-rich fruits and vegetables include bananas and plantains, leafy green vegetables, many dried fruits, oranges and orange juice, cantaloupes and honeydew melons, tomatoes and root vegetables ( Table 5.7 ). Calcium (Ca) Calcium is essential for bone and tooth formation. Because of this, calcium requirements are higher during adolescence. Calcium is also very important during later adulthood, and of great consequence from a public health perspective, because inadequate intake of calcium may increase the risk of osteoporosis, a condition in which decreased bone mass weakens bone ( Nordin, 1997 ; Cohen and Roe, 2000 ). With nearly half of all American women over 50 years of age demonstrating low

242 Author s personal copy 82 Nutritional Quality of Fruits and Vegetables mineral bone density or osteoporosis, and an estimated 1.3 million osteoporosis-related fractures occurring each year in the US, with a billion dollar estimated annual cost ( DeBar et al., 2004 ), osteoporosis prevention is a major public health target. Calcium fluxes are important mediators of hormonal effects on target organs through the phosphoinositol system, and are closely linked with the cyclic AMP systems. There is also evidence linking hypertension with calcium deficiency ( Appel et al., 1997 ; McCarron and Reusser, 2001 ). In plants, calcium is primarily associated with the pectic materials. It is believed to have a major influence on the rheological properties of the cell wall and, consequently, on the texture and storage life of fruits and vegetables. Ca 2 can interact with the anionic pectic polysaccharides, coordinating with the oxygen functions of two adjacent pectin chains to form the so-called eggbox structure, and cross-linking the chains ( Rose et al., 2003 ). Intracellular Ca 2 also occupies a pivotal role in cell signal transduction ( Sanders et al., 1999 ). The plant signals thought to be transduced through cytosolic Ca 2 include wounding, temperature stress, fungal elicitors, oxidative stress, anaerobiosis, abscisic acid, osmotic stress, red or blue light and mineral nutrition. Intracellular Ca 2 transient increases are often associated with initiation of responses. Thus, Ca 2 is a prominent second messenger, and it must be maintained in the cytoplasm at concentrations many orders of magnitude lower than the Ca 2 in the cell wall. Horticultural crops are considered a secondary source of calcium in comparison to dairy products but, taken as a whole, fruits and vegetables account for almost 10% of the calcium in the US food supply (Table 7, Cook and Friday, 2003 ). Dark green leafy cabbage family vegetables and turnip greens are good calcium sources and most green leafy vegetables are potential calcium sources because of their absorbable calcium content ( Jodral-Segado et al., 2003 ; Titchenal and Dobbs, 2007 ). Projects designed to test the efficacy of a health plan-based lifestyle intervention for increasing bone mineral density propose not only to increase the consumption of high calcium foods, but also of fruits and vegetables ( DeBar et al., 2004 ). Magnesium (Mg) Magnesium is important in protein synthesis, release of energy from muscle storage and body temperature regulation. It is critical for proper heart function and plays a role in bone formation, as previously described. Magnesium activates over 100 enzymes. In plants, magnesium is a constituent of the chlorophyll molecule: the porphyrinlike ring structure of chlorophylls contains a central magnesium atom coordinated to the four pyrrole rings. On the other hand, magnesium is involved in the energetic metabolism as a constituent of the Mg-ATP or Mg-ADP complex. Also, the Calvin cycle the pathway that produces a three-carbon compound as the first stable product in the multistep conversion of CO 2 into carbohydrates is partially regulated via stromal Mg 2 concentration. This nutrient also serves important biochemical functions in protein synthesis ( Mengel and Kirkby, 1982 ). In 2004, vegetable contribution to the total magnesium in the US food supply was an average of 14% ( Table 5.7 ). Using current population standards, magnesium

243 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 83 intake was found to be below adequate levels for both adults and children ( Sigman- Grant et al., 2003 ). Mixed users, who are more likely to consume higher intakes of grains, fruit and milk products, were found to have higher magnesium densities than high-fat users, who consume significantly more servings of meat and higher levels of discretionary fat ( Sigman-Grant et al., 2003 ). Generally, magnesium levels are significantly higher in vegetables than in fruits, but nuts are good sources of this nutrient. Dry fruits and legumes are the food groups that rank higher in magnesium content ( Jodral-Segado et al., 2003 ). Phosphorus (P) Inorganic phosphate is essential for skeletal mineralization and for multiple cellular functions, including glycolysis, gluconeogenesis, DNA synthesis, RNA synthesis, cellular protein phosphorylation, phospholipid synthesis and intracellular regulatory roles ( DiMeglio et al., 2000 ). Phosphorus is a primary bone-forming mineral. In western countries, isolated dietary phosphate deficiency is exceedingly rare, because most westerners eat high-phosphate diets, except for occasional metabolic disorders such as hyperphosphatemia ( DiMeglio et al., 2000 ). Phosphorus can exist in plants as both inorganic phosphate anions and organophosphate compounds ( Raghothama, 1999 ). Unlike sulfate and nitrate, phosphate is not reduced in plants during assimilation, but remains in its oxidized state forming phosphate esters in a wide variety of organic compounds. Inorganic phosphorus constitutes a main structural component of nucleic acids and phospholipids, and plays a critical role in energy conversion in the form of high-energy phosphoester and diphosphate bonds. It is important both as a substrate and as a regulatory factor in oxidative metabolism and photosynthesis, it participates in signal transduction, and regulates the activities of an assortment of proteins by way of covalent phosphorylation/ dephosphorylation reactions. In 2004, the primary contributor of phosphorus to the food supply was the dairy group (31.3%), followed by the meat, poultry and fish group (24.9%) and grain products (19.4%) ( Hiza and Bente, 2007 ). Fruit and vegetable contribution to the total phosphorus in the US food supply was an average of 9.5% ( Table 5.7 ). Among tree fruits, nuts are natural sources of phosphorus. Nitrogen (N) The largest requirement for nitrogen in eukaryotic organisms is the biosynthesis of amino acids, building blocks of proteins and precursors of many other compounds. Proteins represent a large percentage of the human body and carry out many different cell functions. Therefore, protein synthesis is central to cell growth, differentiation, and reproduction. Nitrogen is also an essential component of nucleic acids, cofactors and other metabolites. Several plant hormones (indole-3-acetic acid, zeatine, spermidine, etc.) contain nitrogen, or are derived from nitrogenous precursors. Alkaloids and other secondary compounds contain nitrogen, and various phenolics derive from phenylalanine and are therefore linked with amino acid metabolism. Moreover, nitrogen is a major constituent of chlorophyll. The characteristic preharvest yellow color of

244 Author s personal copy 84 Nutritional Quality of Fruits and Vegetables nitrogen-starved vegetables a physiological disorder called chlorosis reflects their inability to synthesize adequate amounts of green chlorophyll under nitrogen-limited conditions. Sulfur (S) Sulfur is an essential nutrient required for growth, primarily used to synthesize cysteine and methionine. The sulfur-containing amino acids play pivotal roles in the structural and catalytic functions of proteins. Cysteines are important because oxidizing the thiol groups of two cysteine residues can form disulfide bonds, important covalent linkages involved in establishing tertiary and, in some cases, quaternary protein structures. The dithiol disulfide interchange can be a regulatory mechanism, as well as a mediator of redox reactions. Sulfur is also a component of numerous essential and secondary metabolites derived from these amino acids. Sulfur nutrition is important in the species within the order Brassicales (e.g. white cabbage, broccoli, cauliflower, capers) for the synthesis of anticarcinogenic glucosinolate compounds (reviewed in Sozzi, 2001 ). In caper ( Capparis spinosa L.) flavor, 160 components were identified, including elemental sulfur (S 8 ) and more than 40 sulfur-containing compounds, among them thiocyanates and isothiocyanates. Although of key importance in human and plant life, sulfur is a relatively minor component in comparison with nitrogen. Generally, it is not a growth-limiting nutrient, since sulfate, the oxidized anion, is relatively abundant in the environment. Manganese (Mn) Manganese is a key component of enzyme systems, including oxygen-handling enzymes. It supports brain function and reproduction and is required for blood sugar regulation. In addition, it is part of bone structure. Manganese is a cofactor in function of antioxidant enzymes, such as the mitochondrial superoxide dismutase. In plants, manganese atoms appear to undergo successive oxidations to yield a strongly oxidizing complex that is capable of water oxidations during photosynthesis. Also like magnesium, manganese is required in enzyme reactions involving carbon assimilation. Chloroplasts are most sensitive to manganese deficiency. Among horticultural crops, spinach is a good source of manganese. Copper (Cu) Copper, a redox active metal, plays an important role in the oxidative defense system. In fact, oxidative stress is a characteristic of copper deficiency ( Uriu- Adams and Keen, 2005 ). Copper is necessary for the formation of hemoglobin and is required for the function of over 30 proteins, including superoxide dismutase, ceruloplasmin, lysyl oxidase, cytochrome c oxidase, tyrosinase and dopamineβ -hydroxylase (Arredondo and Nuñez, 2005). During the past decade, there has been increasing interest in the concept that marginal deficits of this essential nutrient can contribute to the development and progression of a number of disease states, including cardiovascular disease and diabetes. Deficits of this nutrient during pregnancy can result in gross structural malformations in the fetus, and persistent neurological and immunological abnormalities in the offspring ( Uriu-Adams and Keen, 2005 ).

245 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 85 In plants, copper is required for chlorophyll synthesis and in several coppercontaining enzymes involved in the reduction of molecular oxygen. The availability of copper to plants, as with other trace minerals, markedly decreases as ph rises above seven. At high ph copper is strongly adsorbed to clays, iron and aluminum oxides, and organic matter. Of the micronutrients required by plants, copper often has the lowest total concentration in soil. Between 1909 and 1919 in the US, the vegetable group was the leading source of copper (30%). In 2004, the grain group (21%) and the legumes, nuts and soy group (20%) replaced the vegetable group (17%) as the leading sources of copper (Table 7, Hiza and Bente, 2007 ). Iron (Fe) The metabolic fates of copper and iron are intimately related. The essentiality of iron, as well as that of copper, resides in its capacity to participate in one-electron exchange reactions. Systemic copper deficiency generates cellular iron deficiency that, in humans, results in diminished work capacity, reduced intellectual capacity, diminished growth, alterations in bone mineralization, and diminished immune response. Iron is required in numerous essential proteins, such as the heme-containing proteins, electron transport chain and microsomal electron transport proteins, and iron-sulfur proteins and enzymes such as ribonucleotide reductase, prolyl hydroxylase phenylalanine hydroxylase, tyrosine hydroxylase and aconitase (Arredondo and Nuñez, 2005). Iron is a constituent of the haem complex, a naturally occurring plant chelate involved in electron transfer in a number of important plant enzymes ( Mengel and Kirkby, 1982 ). The plant plastid stroma may contain deposits of phytoferritin, a storage form of iron similar to the ferritin of animal cells. Phytoferritin occurs almost exclusively in plastids and most abundantly in the plastids of storage organs ( Briat and Lobreaux, 1997 ). In green vegetable leaves, there is a good correlation between iron supply and chlorophyll content. Inadequate iron nutrition results in abnormal chlorophyll development, so that deficiency begins as an interveinal chlorosis on younger leaves resulting in prominent green veins. The resultant reduction in photosynthetic capability also reduces the weight and area of affected leaves. Descriptions of causes of iron deficiency have been extensively reviewed for horticultural crops ( Korcak, 1987 ). Adult users of lower-fat foods consume more nutrient-dense diets, with higher intakes of iron ( Kennedy et al., 2001 ; Sigman-Grant et al., 2003 ). The predominant source of iron in the American food supply is grain products, followed by the meat, poultry and fish group. Between 1909 and 1919, the vegetable group furnished an average of 18% of the iron in the food supply, but in 2004 that share dropped to an average of 10% ( Table 5.7 ). This is partially due to a decrease in the use of white potatoes after Although potatoes are not a good source of iron, their contribution to the food supply increases when eaten in large quantities ( Hiza and Bente, 2007 ), particularly if the skin is consumed (specifically, baked potato skin is 20-fold richer in iron than the flesh). Almonds, pistachio nuts, walnuts, pecans, etc., are very good sources of iron. Different vegetables (e.g. parsley, broccoli, kale, turnip greens and collards) and legumes (e.g. green peas and beans) are also considered good sources of iron.

246 Author s personal copy 86 Nutritional Quality of Fruits and Vegetables Zinc (Zn) Zinc is a pervasive microelement that plays a catalytic or a structural role in more than 200 enzymes (e.g. carboxypeptidase, liver alcohol dehydrogenase and carbonic anhydrase) involved in digestion, metabolism, reproduction, and wound healing. Zn 2 is a cation with various coordination possibilities and several potential geometries. Thus, it is easily adaptable for different ligands. The main role of structural Zn 2 in proteins is to stabilize tertiary structures. In addition, zinc has a critical role in immune response, and is an important antioxidant. Zinc activates a number of plant cell enzymes ( Romheld and Marschner, 1991 ), but only a few of them (i.e. alcohol dehydrogenase, superoxide dismutase, carbonic anhydrase, RNA polymerase) contain the micronutrient. Zinc can affect carbohydrate metabolism because different Zn-dependent enzymes participate in biochemical reactions involving sugars. Zinc also plays a role in the maintenance of cell membrane integrity, in the protection from O 2 damage, and the synthesis of RNA and tryptophan, a precursor of indole-3-acetic acid. A comprehensive review of soil, plant and management factors associated with zinc nutrition in horticultural crops has been developed by Swietlik (1999). Fruits and vegetables account for only 1.2% and 6.4%, respectively, of the zinc in the American food supply ( Hiza and Bente, 2007 ). As is the case for magnesium, zinc intakes may be below the adequate levels for both adults and children ( Sigman- Grant et al., 2003 ). Fruits are poor in zinc, but pecans and walnuts are good sources of this essential mineral. Parsley is also a good source of zinc. Sodium (Na) Sodium is a systemic ion. It is important in electrolyte balance and essential in coregulating ATP with potassium. In addition, it has an important role in the regulation of blood pressure. Sodium contributed from vegetables increased during the last decades ( Table 5.7 ), due to the increased consumption of processed vegetables (largely tomatoes and white potatoes). With the exception of canned vegetables, sodium estimates in the food supply do not account for sodium added in processing. Thus, the relative contribution of vegetables to sodium reported in the food supply is likely overstated ( Hiza and Bente, 2007 ). Table salt (NaCl) is by far the main dietary source for sodium. Olives and spinach are horticultural sources of sodium. In general, fruits are poor in sodium, and are recommended for low-sodium dietary patterns. B. Factors influencing mineral content of fruits and vegetables Influence of the species and the cultivar Mineral composition varies widely in raw fruits ( Table 5.8 ) and vegetables because of genetics. Leafy vegetables tend to have higher concentrations of nutrients that are less mobile in the plant (e.g. calcium) and depend on direct water flow rather than recycling from leaves. Tissues with higher transpiration rates generally have higher tissue calcium concentrations ( Witney et al., 1990b ). Concentrations of minerals may also vary widely with the cultivar. For example, both Dwarf Brazilian bananas

247 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 87 Table 5.8 Mineral composition of a range of fruit species. Results are in mg 100 g 1 fresh weight. Fruit Mineral Apples, raw, with skin K Ca Mg P Mn Cu Fe Zn Na Se Apricots, raw Avocado, raw (California) Avocado, raw (Florida) Bananas, raw Blackberries, raw Blueberries, raw Cherries, sweet, raw Figs, raw Grapefruit, raw, pink and red (California and Arizona) Grapefruit, raw, pink and red (Florida) Grapes, red or green (euro type, e.g. Thompson seedless ), raw Kiwifruit, fresh, raw Lemons, raw, without peel Mangos, raw Melons, Cantaloupe, raw Oranges, raw, California, Valencia Papayas, raw Peaches, raw Pears, raw (Continued)

248 Author s personal copy 88 Nutritional Quality of Fruits and Vegetables Table 5.8 Continued Fruit Mineral K Ca Mg P Mn Cu Fe Zn Na Se Pineapples, raw, all varieties Plums, raw Pomegranates, raw Raspberries, raw Strawberries, raw Watermelon, raw US Department of Agriculture, (Santa Catarina Prata, Musa sp. AAB) and Williams (Cavendish subgroup, Musa sp. AAA) are considered as a good source of potassium. Nevertheless, Dwarf Brazilian bananas have higher P, Ca, Mg, Mn and Zn contents than Williams bananas ( Wall, 2006 ). In contrast, no strawberry variety was statistically superior as a source of minerals ( Hakala et al., 2003 ). Because of the distribution of vascular tissue, sink characteristics and metabolic rates, higher mineral concentrations are usually found in the skin and seeds, with lower concentrations in the flesh of fruits. Tissues with higher metabolic rates (epicarp, core) may have higher requirements for nitrogen and phosphorus. Rapidly expanding or large-celled tissues are unlikely to have high calcium concentrations. In mature fruit, the calcium concentration is highest in the peel ( Saure, 2005 ). Influence of preharvest factors and practices Orchard location has proved to have important effects on fruit and vegetable mineral content ( Table 5.8 ). For example, potassium content in bananas markedly differs between different locations in Hawaii, from 288 mg 100 g 1 in Kapaa to 485 mg 100 g 1 in Waimanalo ( Wall, 2006 ). Papaya cv. Rainbow is not very rich in potassium, but its content also varies between locations, from 113 mg 100 g 1 on the island of Hawaii to 203 mg 100 g 1 on the island of Maui ( Wall, 2006 ). Mineral composition fluctuates widely in raw fruits and vegetables, because of preharvest factors (soil fertility including ph and availability of nutrients moisture content of the soil, growth temperature) and cultural practices (amount and timing of fertilization and irrigation, application of plant growth regulators, pruning and thinning of tree fruit species, etc.). Most of these practices have been established primarily for productivity goals, and not as a medium to better human health, horticultural crop postharvest life or flavor quality ( Crisosto and Mitchell, 2002 ). Usually, fertilizers are applied directly to the soil to raise nutrient levels, if they are inadequate for

249 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 89 the successful growth of the crop, and to maintain soil fertility, which will decline if nutrient removal from the soil (via crop uptake, leaching, volatilization or denitrification) exceeds nutrients added via weathering of minerals and mineralization of organic matter. Nitrogen is the most frequently deficient and most commonly applied fertilizer in orchards, while addition to the soil of phosphorus and potassium is warranted when soil-test results, plant response or tissue analysis indicate a requirement. N-P-K addition with irrigation water (fertigation) has several advantages, including the ability to transport soluble nutrients directly to the root zone whenever the plant is watered. Thus, fertilizer amounts and timing can be precise and adjusted to coincide more closely with actual plant demand. Calcium additions can be large when lime is applied to increase soil ph. Most micronutrients are rarely applied via soil and can be directly supplied via spray application of dilute concentrations of minerals to the canopy. In the case of fruits, the quantity of nutrients capable of being absorbed through the waxy cuticle is often small relative to nutrient demand, but can ameliorate deficiency symptoms and improve fruit quality ( Swietlik and Faust, 1984 ). An excessive supply of nutrients relative to photosynthesis can develop when the rate of nutrient assimilation is high relative to net photosynthesis. In this case, an accumulation of nutrients in fruits and vegetables can reach levels that are toxic either to the plant or to consumers. For example, excessive nitrogen application can lead to potentially harmful accumulations of nitrate nitrogen, especially in leafy greens and potatoes ( Eppendorfer, 1978 ; Blom-Zandstra, 1989 ). These nutrient imbalances also affect horticultural crop quality, as discussed above. Many other factors influencing nutrient accumulation are related to nutrient transport and source-sink relations. For example, alterations in water economy affect calcium input. Since calcium is transported mainly in the transpiration stream ( Grange and Hand, 1987 ), bagging fruit may result in lower calcium concentrations and higher calcium-related disorders ( Witney et al., 1991 ; Hofman et al., 1997 ), due to increased relative humidity. Nevertheless, evidence is not conclusive ( Saure, 2005 ). Canopy position and crop load also influence calcium input. Tree vigor is usually associated with lower calcium and magnesium content in fruits ( Witney et al., 1990a,b ). Fruit from upper parts of the canopy tend to show lower calcium contents ( Ferguson and Triggs, 1990 ), and heavy cropping trees have fruit with higher calcium and lower potassium concentrations ( Ferguson and Watkins, 1992 ). Nevertheless, calcium transport to fruit may be based on a hormonal control; gibberellins have been shown to inhibit calcium translocation ( Saure, 2005 ). Tree size, spacing, row orientation, canopy shape and training system influence light distribution within fruit trees, which in turn may affect mineral composition. In grapes, improvement of light penetration into the canopy enhanced anthocyanin and soluble phenol levels, but reduced potassium content ( Prange and DeEll, 1997 ). In kiwifruit, light promoted calcium accumulation ( Montanaro et al., 2006 ). The finding was not fully explained by fruit transpiration, a regulatory mechanism governed by phytohormones, which could play a role in determining calcium concentrations. Besides, the effect of sunlight does not seem to be universal: avocado fruit from the sunny side of trees did not contain significantly more calcium than fruit from the shaded side ( Witney et al., 1990a ).

250 Author s personal copy 90 Nutritional Quality of Fruits and Vegetables The mineral content of some horticultural species seems to be affected under intensive culture systems (e.g. glasshouse) or organic conditions. Tomato fruit showed higher calcium and lower potassium, magnesium and sodium concentrations when grown on organic (compost/soil mix) versus hydroponic substrates ( Premuzic et al., 1998 ). Smith (1993) reported higher mineral contents in organically cultivated apples, pears, potatoes and corn in comparison to conventionally cultivated ones. In contrast, Petersen and Pedersen (1991) did not find differences in mineral content between organically and conventionally cultivated vegetables. Hakala and co-workers (2003) reported that organic cultivation did not affect strawberry mineral contents consistently. Postharvest practices influencing mineral content of fruits and vegetables Postharvest treatments with minerals, primarily calcium, are used to improve the storage life and quality of different fruits and vegetables. In the last decade, the industry has been encouraged to fortify food and beverages with calcium. Increasing the calcium content of horticultural crops may give consumers new ways to enhance their calcium intake without resorting to supplements. In addition, the use of phosphorous-free sources of calcium can help to obtain a good balance of calcium and phosphorus in the diet ( Martín-Diana et al., 2007 ). Two major methods of postharvest application of calcium in horticultural crops are used: (1) dipping-washing and (2) impregnation processes ( Martín-Diana et al., 2007 ). Dipping treatments are used for fresh, sensitive products, such as leafy vegetables. The delicate texture of berries prevents the use of vacuum infiltration, and dips in a solution of CaCl 2 are used ( García et al., 1996 ), followed by the removal of excess washing solution. On the other hand, impregnation modifies the composition of food material through partial water removal and impregnation of solutes, with no impairment of the material integrity. The process-driven forces can be osmotic gradient between the sample and solution, application of vacuum followed by atmospheric condition restoration, or both. Calcium chloride has been widely used as firming agent and preservative for both whole and fresh-cut fruits and vegetables, as discussed above. C. Effect of minerals on fruit and vegetable quality and consumer acceptance Consumers buy certain items as good sources of specific minerals: potatoes and sweet potatoes for potassium, bananas for magnesium and potassium, spinach for iron, potassium, magnesium and as a non-dairy source of calcium. Mineral content of products is usually determined by ashing and atomic absorption ( Pomeranz and Meloan, 1987 ). Without advanced analytical equipment, the consumer cannot detect differences in individual products at the point of purchase ( Institute of Food Technologists, 1990 ). These attributes are considered credence attributes (see also Chapter 3), because they cannot be detected readily either by visual inspection or by consumption. Therefore, there is little or no incentive to measure mineral content in a quality control program, unless specific nutritional claims can be made.

251 Author s personal copy IV. Fruits and vegetables as direct sources of minerals 91 Nevertheless, the consumer uses other criteria to judge quality. Quality attributes (see Chapter 3) include purchase attributes (i.e. size, color, firmness to the touch, aroma and absence of defects) and consumption attributes (i.e. flavor, mouth feel). Many of these quality characteristics are also affected by the mineral content and constitute part of a wider range of factors affecting fruit and vegetable acceptability. Acceptability, which is defined as the level of continued purchase or consumption by a specific population ( Land, 1988 ), determines the consumption levels of many hidden essential nutrients: vitamins, antioxidants, fiber. Thus, the effect of minerals on horticultural crop quality attributes and consumer acceptance should be considered. Effect of minerals on color In apples and pears, both leaf and fruit nitrogen positively correlates with fruit green background color ( Raese, 1977 ; Marsh et al., 1996 ), regardless of the rootstock used ( Fallahi et al., 1985 ). Manganese has also been associated with green ground color in apples ( Deckers et al., 1997 ). Excessive nitrogen application inhibits background color change from green to yellow and induces deficient reddish blush development and poor edible quality of peaches ( Sistrunk, 1985 ; Crisosto et al., 1995 ; Crisosto et al., 1997 ). High nitrogen application also decreases fruit color in grapes ( Kliewer, 1977 ). In Citrus, nitrogen is associated with an undesirable retardation of endogenous chlorophyll catabolism ( Koo et al., 1974 ) and postharvest treatments with ethylene may be required to accelerate the loss of the green color (de-greening). In apples, amelioration of potassium deficiencies can increase red fruit color, but such an effect is often not apparent when tree potassium status is adequate ( Neilsen and Neilsen, 2003 ). In tomatoes, potassium deficiency is associated with lower levels of lycopene and higher levels of β -carotene ( Trudel and Ozbun, 1971 ). Effect of minerals on flavor Nitrogen status negatively correlates with soluble solids, both in apples ( Fallahi et al., 1985 ; Dris et al., 1999 ) and in pears ( Raese, 1977 ). In contrast, soluble solid content increases with increasing fertilizer nitrogen levels in tomatoes ( Barringer et al., 1999 ). Apple calcium and phosphorus were both negatively correlated with fruit soluble solids at harvest, and after six months of 0 C storage, while fruit K/Ca ratio was positively correlated with titratable acidity ( Fallahi et al., 1985 ). In mango, total soluble solids increased when zinc sulfate fertilizer was applied to the soil ( Bahadur et al., 1998 ). In Fino 49 lemons, salinity reduces juice percentage and impairs juice quality by decreasing the total soluble solids and titratable acidity ( García-Sánchez et al., 2003 ). Reduction of titratable acidity could be due to the greater accumulation of Cl, compared to Na, which could be compensated for by the degradation of organic acids for charge balance. Minerals are also known to affect the production of several classes of volatile compounds in pome fruit (reviewed in Mattheis and Fellman, 1999 ). In fresh onions, increased sulfur availability enhances pungency and total sulfur flavor, but decreases the amounts of precursors for volatiles imparting green and cabbage notes (Randle, 1997 ).

252 Author s personal copy 92 Nutritional Quality of Fruits and Vegetables Effect of minerals on firmness Excess nitrogen fertilization can result in a decrease in firmness ( Reeve, 1970 ; Prange and DeEll, 1997 ). Low phosphorus may also result in a loss of firmness in low-calcium content fruit ( Sharples, 1980 ). The relationship between calcium and fruit firmness has been extensively studied and reviewed ( Ferguson, 1984 ; Poovaiah et al., 1988 ; Harker et al., 1997 ; Sams, 1999 ). Higher firmness values and/or slower softening rates after harvest/storage have been associated with higher calcium concentrations, or with calcium applications in different fruit species, such as apples and pears ( Fallahi et al., 1985 ; Raese and Drake, 1993, 2000a,b, 2002 ; Gerasopoulos and Richardson, 1999 ; Benavides et al., 2001 ); kiwifruit ( Hopkirk et al., 1990 ; Gerasopoulos and Drogoudi, 2005 ); and strawberries ( Chéour et al., 1990 ). Calcium foliar sprays on peaches and nectarines lead to a slight increase of calcium content ( Manganaris et al., 2005a, 2006 ). Under Californian conditions, no consistent effect on fruit quality of mid- or late-season peach and nectarine varieties was found (reviewed in Crisosto et al., 1997 ). Postharvest calcium treatments have been reported to retain fruit firmness in different horticultural products, among them, apples ( Wang et al., 1993 ; Conway et al., 1994 ), peaches ( Manganaris et al., 2005b, 2007 ), strawberries ( Morris et al., 1985 ; García et al., 1996 ), lemons ( Valero et al., 1998 ; Martínez-Romero et al., 1999 ), sliced pears and strawberries ( Rosen and Kader, 1989 ). Calcium effects on fruit firmness are attributable to calcium s ability to cross-link with the pectic polysaccharide network by ionic association. Calcium binding may reduce the accessibility of cell wall degrading enzymes to their substrates. Effect of minerals on rots, physiological disorders and nutritional value In calcium-treated fruit, the association between firmness retention and reduced rot incidence suggests that calcium may affect both processes simultaneously through its cellular role in strengthening plant cell walls ( García et al., 1996 ; Fallahi et al., 1997 ; Conway et al., 1999 ). On the other hand, high nitrogen fertilization increases susceptibility to decay caused by Monilinia fructicola (brown rot) in nectarines ( Daane et al., 1995 ). Wounded and brown rot inoculated Fantasia and Flavortop nectarines from trees having more than 2.6% leaf nitrogen are more susceptible to Monilinia fructicola than fruit from trees with 2.6% or less leaf nitrogen ( Michailides et al., 1993 ). Consumers consider that fruits have less predictable eating quality than manufactured snacks. In fact, the effect of nutrients on the final quality of horticultural products may not become evident until harvest, distribution or consumption. The expression latent damage was coined by Peleg (1985) and later defined by Shewfelt (1986) as damage incurred at one step but not apparent until a later step to describe the result of non-visible quality loss. Physiological disorders may be a type of latent damage. Some physiological disorders relate to the imbalance between nutrients. Calcium is the nutrient most commonly associated with postharvest disorders. A calcium-deficient status is considered an important preharvest factor related to numerous physiological disorders of fruits and vegetables, such as bitter pit in pome fruit, blossom-end rot in tomato, blackheart in celery, cracking and cavity spot

253 Author s personal copy Bibliography 93 in carrot and tip burn in lettuce and cabbage (reviewed in Ferguson et al., 1999 ), although some authors have questioned the role of calcium in these disorders ( Saure, 1998, 2001 ). Other calcium-related disorders are associated with long-term cold storage, such as chilling injury in muskmelon ( Combrink et al., 1995 ) and avocado ( Chaplin and Scott, 1980 ). Postharvest calcium applications limited the incidence of chilling injury in peach fruit, expressed as flesh browning, after four weeks cold storage at 5 C ( Manganaris et al., 2007 ). Nevertheless, preharvest calcium applications showed no effect on the onset of chilling injury in peaches and nectarines (reviewed in Lurie and Crisosto, 2005 ). Magnesium and potassium have been considered as part of an index to predict bitter pit ( Bramlage et al., 1985 ; Autio et al., 1986 ). Fallahi and Righetti (1984) proposed the relation between nitrogen and calcium as an important component of a diagnosis and recommendation system (DRIS) for apple. High rates of nitrogen application exacerbate the incidence of many physiological disorders, such as apricot pit burn ( Bussi and Amiot, 1998, 2003). In addition, minerals can influence the concentrations of other nutrients in horticultural crops. Nitrogen fertilizers at high rates tend to decrease the concentration of vitamin C in fruits (citrus juices) and vegetables (potatoes, cauliflower, white cabbage, crisphead lettuce, etc.) while increased potassium fertilization increases ascorbic acid content (reviewed in Lee and Kader, 2000 ). Bibliography Agius, F., Gonzalez-Lamothe, R., Caballero, J.L., Muñoz-Blanco, J., Botella, M.A., Valpuesta, V. (2003 ). Engineering increased vitamin C levels in plants by overexpression of a D-galacturonic acid reductase. Nature Biotechnol., 21, Ames, B.M., Shigena, M.K., Hagen, T.M. (1993 ). Oxidants, antioxidants and the degenerative diseases of aging. Proc. Natl. Acad. Sci. U.S.A., 90, Anderson, J.S., Perryman, S., Young, L., Prior, S. (2007). Dietary fiber. Colorado State University Nutrition Resources. N FOODNUT/09333.html Accessed April Appel, L.J., Moore, T.J., Oberzanek, E., Vollmer, W.M., Svetkey, L.P., Sacks, F.M. (1997 ). A clinical trial of the effect of dietary patterns on blood pressure. N. E. J. Med., 336, Arredondo, M., Núñez, M.T. ( 2005 ). Iron and copper metabolism. Mol. Aspects Med., 26, Asada, K. (1999 ). The water water cycle in chloroplasts: scavenging of active oxygen and dissipation of excess photons. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50, Asami, D.K., Hong, Y.H., Barrett, D.M., Mitchell, A.E. (2003 ). A comparison of the total phenolic and ascorbic acid contents of freeze-dried and air-dried marionberry, strawberry and corn grown using conventional, organic and sustainable agricultural practices. J. Agric. Food Chem., 51,

254 Author s personal copy 94 Nutritional Quality of Fruits and Vegetables Asp, N.G. (1994 ). Nutritional classification of food carbohydrates. Am. J. Clin. Nutr., 59, S679 S681. Autio, W.R., Bramlage, W.J., Weis, S.A. (1986 ). Predicting disorders of Cox s Orange Pippin and Bramley s Seedling apples by regression equations. J. Am. Soc. Hort. Sci., 111, Ayala-Zavala, J.F., Wang, S.Y., Wang, C.Y., Gonzalez-Aguilar, G.A. ( 2004 ). Effect of storage temperatures on antioxidant capacity and aroma compounds in strawberry fruit. Lebensm. Wiss. Technol., 37, Bahadur, L., Malhi, C.S., Singh, Z. ( 1998 ). Effect of foliar and soil applications of zinc sulphate on zinc uptake, tree size, yield, and fruit quality of mango. J. Plant Nutr., 21, Barrett, D.M., Weakley, C., Diaz, J.V., Watnik, M. (2007 ). Qualitative and nutritional differences in processing tomatoes grown under commercial organic and conventional production systems. J. Food Sci., 72, C441 C450. Barringer, S.A., Bennett, M.A., Bash, W.D. (1999 ). Effect of fruit maturity and nitrogen fertilizer levels on tomato peeling efficiency. J. Veg. Crop Prod., 5, Benavides, A., Recasens, I., Casero, T., Puy, J. (2001 ). Chemometric analyses of Golden Smoothee apples treated with two preharvest calcium spray strategies in the growing season. J. Sci. Food Agric., 81, Benbrook, C.M. (2005). Elevating antioxidant levels in food through organic farming and food processing. An Organic Center state of science review, org/reportfiles/antioxidant_ssr.pdf Accessed April Bernhardt, S., Schlich, E. (2006 ). Impact of different cooking methods on food quality: Retention of lipophilic vitamins in fresh and frozen vegetables. J. Food Eng., 77, Blom-Zandstra, M. (1989 ). Nitrate accumulation in vegetables and its relationship with quality. Annals Appl. Biol., 115, Boerjan, W., Ralph, J., Baucher, M. ( 2003 ). Lignin biosynthesis. Annu. Rev. Plant Biol., 54, Bramlage, W.J., Weis, S.A., Drake, M. (1985 ). Predicting poststorage disorders of McIntosh apples from preharvest mineral analyses. J. Am. Soc. Hort. Sci., 110, Brett, C.T., Waldron, K.W. (1996 ). Physiology and biochemistry of plant cell walls. In: 1. Topics in plant functional biology, M. Black, B. Charlwood (eds). Chapman and Hall, London, UK, p Briat, J.-F., Lobreaux, S. ( 1997 ). Iron transport and storage. Trends Plant Sci., 2, Brummell, D.A. (2006 ). Cell wall disassembly in ripening fruit. Funct. Plant Biol., 33, Brummell, D.A., Harpster, M.H. (2001 ). Cell wall metabolism in fruit softening and quality and its manipulation in transgenic plants. Plant Mol. Biol., 47, Bussi, C., Amiot, M.J. (2003 ). Effects of fertilizer rates and dates of application on apricot (cv. Bergeron) cropping and pitburn. Sci. Hort., 98, Bussi, C., Amiot, M.J. (1998 ). Effects of nitrogen and potassium fertilization on the growth, yield and pitburn of apricot (cv. Bergeron). J. Hort. Sci. Biotechnol., 73,

255 Author s personal copy Bibliography 95 Cantos, E., Espín, J.C., Tomás-Barberán, F.A. (2001 ). Postharvest induction modeling method using UV irradiation pulses for obtaining resveratrol-enriched table grapes: a new functional fruit?. J. Agric. Food. Chem., 49, Cao, G., Sofic, E., Prior, R.L. ( 1996 ). Antioxidant capacity of tea and common vegetables. J. Agric. Food Chem., 44, Carpita, N., McCann, M. (2000 ). The plant cell wall. In: Biochemistry and Molecular Biology of Plants, B. Buchanan, W. Gruissem, R. Jones (eds). American Society of Plant Physiologists, pp Carr, A.C., Frei, B. (1999 ). Toward a new recommended dietary allowance for vitamin C based on antioxidant and health effects in humans. Am. J. Clin. Nutr., 69, Castrejón, A.D., Eichholz, I., Rohn, S., Kroh, L.W., Huyskens-Keil, S. (2008 ). Phenolic profile and antioxidant activity of highbush blueberry ( Vaccinium corymbosum L.) during fruit maturation and ripening. Food Chem., 109, Chaplin, G.R., Scott, K.C. ( 1980 ). Association of calcium in chilling injury susceptibility of stored avocados. HortScience, 15, Chassy, A., Bui, L., Renaud, E.N.C., Van Horn, M., Mitchell, A. ( 2006 ). Three-year comparison of the content of antioxidant microconstituents and several quality characteristics in organic and conventionally managed tomatoes and bell peppers. J. Agric. Food Chem., 54, Chatterjee, I.B. (1973 ). Evolution and the biosynthesis of ascorbic acid. Science, 182, Chéour, F., Willemot, C., Arul, J., Desjardins, Y., Makhlouf, I., Charest, P.M., Gosselin, A. (1990 ). Foliar application of calcium chloride delays postharvest ripening of strawberry. J. Am. Soc. Hort. Sci., 115, Choi, Y.J., Tomás-Barberán, F.A., Saltveit, M.E. ( 2005 ). Wound-induced phenolic accumulation and browning in lettuce ( Lactuca sativa L.) leaf tissue is reduced by exposure to n-alcohols. Postharvest Biol. Technol., 37, Cohen, A.J., Roe, F.J.C. (2000 ). Review of risk factors for osteoporosis with particular reference to a possible aetiological role of dietary salt. Food Chem. Toxicol., 38, Combrink, N.J.J., Jacobs, G., Maree, P.C.J. (1995 ). The effect of calcium and boron on the quality of muskmelons. J. Southern African Soc. Hort. Sci., 5, Conway, W.S., Janisiewicz, W.J., Klein, J.D., Sams, C.E. ( 1999 ). Strategy for combining heat treatment, calcium infiltration, and biocontrol to reduce postharvest decay of Gala apples. HortScience, 34, Conway, W.S., Sams, C.E., Wang, C.Y., Abbott, J.A. (1994 ). Additive effects of postharvest calcium and heat treatment on reducing decay and maintaining quality in apples. J. Am. Soc. Hort. Sci., 119, Cook, A.J., Friday, J.E. (2003 ). Food mixture of ingredient sources for dietary calcium: Shifts in food group contributions using four grouping protocols. J. Am. Diet. Assoc., 103, Crisosto, C.H., Mitchell, J.P. (2002 ). Preharvest factors affecting fruit and vegetable quality. In: Postharvest technology of horticultural crops, A.A. Kader (ed.). Publication University of California, Agriculture and Natural Resources, CA, USA, pp

256 Author s personal copy 96 Nutritional Quality of Fruits and Vegetables Crisosto, C.H., Johnson, R.S., DeJong, T., Day, K.R. (1997 ). Orchard factors affecting postharvest stone fruit quality. HortScience, 32, Crisosto, C.H., Mitchell, F.G., Johnson, R.S. (1995 ). Factors in fresh market stone fruit quality. Postharvest News and Information, 6, 17N 21N. D Ambrosio, C., Giorio, G., Marino, I., Merendino, A., Petrozza, A., Salfi, L., Stigliani, A.L., Francesco, L. (2004 ). Virtually complete conversion of lycopene into β- carotene in fruits of tomato plants transformed with the tomato lycopene β-cyclase (tlcy-b ) cdna. Plant Sci., 166, Daane, K.M., Johnson, R.S., Michailides, T.J., Crisosto, C.H., Dlott, J.W., Ramirez, H.T., Yokota, G.T., Morgan, D.P. (1995 ). Excess nitrogen raises nectarine susceptibility to disease and insects. California Agriculture, 49 ( 4 ), de Ancos, B., Gonzalez, E.M., Cano, M.P. (2000 ). Ellagic acid, vitamin C, and total phenolic contents and radical scavenging capacity affected by freezing and frozen storage in raspberry fruit. J. Agric. Food Chem., 48, de Azevedo, C.H., Rodriguez-Amaya, D.B. (2005 ). Carotenoid composition of kale as influenced by maturity, season and minimal processing. J. Sci. Food Agric., 85, Debar, L.L., Rittenbaugh, C., Vuckovik, N., Stevens, V.J., Aickin, M., Elliot, D., Moe, E., Orwoll, E., Ernst, D., Irvin, L.M. (2004 ). YOUTH: decisions and challenges in designing an osteoporosis prevention intervention for teen girls. Prev. Med., 39, Deckers, T., Daemen, E., Lemmens, K., Missotten, C., Val, J. (1997 ). Influence of foliar applications on Mn during summer on the fruit quality of Jonagold. Acta Hort., 448, Deming, D.M., Baker, D.H., Erdman, J.W. (2002a ). The relative vitamin A value of 9-cis β-carotene is less and that of 13-cis β-carotene may be greater than the accepted 50% that of all-trans β-carotene in gerbils. J. Nutr., 132, Deming, D.M., Teixeira, S.R., Erdman, J.W. (2002b ). All-trans-β-carotene appears to be more bioavailable than 9-cis or 13-cis-β-carotene in gerbils given single oral doses of each isomer. J. Nutr., 132, Dimeglio, L.A., White, K.E., Econs, M.J. ( 2000 ). Disorders of phosphate metabolism. Endocrinology and Metabolism Clinics of North America, 29, Dragsted, L.O. ( 2003 ). Antioxidant actions of polyphenols in humans. Int. J. Vitam. Nutr. Res., 73, Dris, R., Bennett, M.A., Bash, E. (1999 ). Relationship between leaf and fruit minerals and fruit quality attributes of apples grown under northern conditions. J. Plant Nutr., 22, Duthie, G.G., Gardner, P.T., Kyle, J.A.M. (2003 ). Plant polyphenols: are they the new magic bullet?. Proc. Nutr. Soc., 62, Eitenmiller, R.R., Landen, W.O. ( 1999 ). Vitamin analysis for the health and food sciences. CRC Press, Boca Raton, London, New York, Washington, pp Ekholm, P., Reinivuo, H., Mattila, P., Pakkala, H., Koponen, J., Happonen, A., Hellström, J., Ovaskainen, M.-L. (2007 ). Changes in the mineral and trace element contents of cereals, fruits and vegetables in Finland. J. Food Compos. Anal., 20, Eppendorfer, W.H. (1978 ). Effects of N-fertilisation on amino acid composition and nutritive value of spinach, kale, cauliflower, and potatoes. J. Sci. Food Agric., 29,

257 Author s personal copy Bibliography 97 Fairweather-Tait, S., Hurrell, R.F. ( 1996 ). Bioavailability of minerals and trace elements. Nutr. Res. Rev., 9, Fallahi, E., Righetti, T.L. (1984 ). Use of diagnosis and recommendation integrated system (DRIS) in apple. HortScience, 19, 54. Fallahi, E., Conway, W.S., Hickey, K.D., Sams, C.E. (1997 ). The role of calcium and nitrogen in postharvest quality and disease resistance of apples. HortScience, 32, Fallahi, E., Richardson, D.G., Westwood, M.N., Chaplin, M.H. (1985 ). Relationships among mineral nutrition, ethylene and post-harvest physiology in apples on six rootstocks. Sci. Hort., 25, Ferguson, I.B. ( 1984 ). Calcium in plant senescence and fruit ripening. Plant Cell Environ., 7, Ferguson, I.B., Triggs, C.M. (1990 ). Sampling factors affecting the use of mineral analysis of apple fruit for prediction of bitter pit. N. Z. J. Crop Hort. Sci., 18, Ferguson, I.B., Watkins, C.B. (1992 ). Crop load affects mineral concentrations and incidence of bitter pit in Cox s Orange Pippin apple fruit. J. Am. Soc. Hort. Sci., 117, Ferguson, I., Volz, R., Woolf, A. (1999 ). Preharvest factors affecting physiological disorders of fruit. Postharvest Biol. Technol., 15, Fisher, R.L., Bennett, A.B. (1991 ). Role of cell wall hydrolases in fruit ripening. Annu. Rev. Plant Physiol. Plant Mol. Biol., 42, Fraser, P.D., Romer, S., Shipton, C.A., Mills, P.B., Kiano, J.W., Misawa, N., Drake, R.G., Schuch, W., Bramley, P.M. (2002 ). Evaluation of transgenic tomato plants expressing an additional phytoene synthase in a fruit-specific manner. Proc. Natl. Acad. Sci. U.S.A., 99, Frossard, E., Bucher, M., Machler, F., Mozafar, A., Hurrell, R. (2000 ). Potential for increasing the content and bioavailability of Fe, Zn and Ca in plants for human nutrition. J. Sci. Food Agric., 80, García, J.M., Herrera, S., Morilla, A. (1996 ). Effects of postharvest dips in calcium chloride on strawberry. J. Agric. Food Chem., 44, García-Sánchez, F., Carvajal, M., Porras, I., Botía, P., Martínez, V. (2003 ). Effects of salinity and rate of irrigation on yield, fruit quality and mineral composition of Fino 49 lemon. Eur. J. Agron., 19, Gautier, H., Diakou-Verdin, V., Bernard, C., Reich, M., Buret, M., Bourgaud, R., Poessel, J.L., Caris-Veyrat, C., Genard, M. ( 2008 ). How does tomato quality (sugar, acid, and nutritional quality) vary with ripening stage, temperature, and irradiance?. J. Agric. Food. Chem., 56, Gerasopoulos, D., Drogoudi, P.D. (2005 ). Summer-pruning and preharvest calcium chloride sprays affect storability and low temperature breakdown incidence in kiwifruit. Postharvest Biol. Technol., 36, Gerasopoulos, D., Richardson, D.G. (1999 ). Storage temperature and fruit calcium alter the sequence of ripening events of d Anjou pears. HortScience, 34, González-Barrio, R., Beltrán, D., Cantos, E., Gil, M.I., Espín, J.C., Tomás-Barberán, F.A. (2006 ). Comparison of ozone and UV-C treatments on the postharvest stilbenoid monomer, dimer, and trimer induction in var. Superior white table grapes. J. Agric. Food Chem., 54,

258 Author s personal copy 98 Nutritional Quality of Fruits and Vegetables Grange, R.I., Hand, D.W. (1987 ). A review of the effects of atmospheric humidity on the growth of horticultural crops. J. Hort. Sci., 62, Gross, K.C., Sams, C.E. (1984 ). Changes in cell wall neutral sugar composition during fruit ripening: a species survey. Phytochemistry, 23, Grusak, M.A., Della Penna, D. (1999 ). Improving the nutrient composition of plants to enhance human nutrition and health. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50, Gu, L., Kelm, M., Hammerstone, J., Beecher, G., Holde, J., Haytowitz, D., Gebhardt, S., Prior, R. (2004 ). Concentrations of proanthocyanidins in common foods and estimations of normal consumption. J. Nutr., 134, Guillon, F., Champ, M. (2000 ). Structural and physical properties of dietary fibres, and consequences of processing on human physiology. Food Res. Int., 33, Hakala, M., Lapveteläinen, A., Huopalahti, R., Kallio, H., Tahvonen, R. (2003 ). Effects of varieties and cultivation conditions on the composition of strawberries. J. Food Compos. Anal., 16, Hancock, R.D., Viola, R. (2005 ). Improving the nutritional value of crops through enhancement of L-ascorbic acid (vitamin C) content: Rationale and biotechnological opportunities. J. Agric. Food Chem., 53, Harker, F.R., Redgwell,R.J., Hallett, I.C., Murray, S.H. (1997 ). Texture of fresh fruit. Hort. Rev., 20, Hiza, H.A.B., Bente, L. (2007). Nutrient content of the U.S. food supply, : A summary report. Home Economics Research Report Number 57. U.S. Department of Agriculture, Center for Nutrition Policy and Promotion, Washington, DC, USA. Hofman, P.J., Smith, L.G., Joyce, D.C., Johnson, G.I., Meiburg, G.F. (1997 ). Bagging of mango (Mangifera indica cv. Keitt ) fruit influences fruit quality and mineral composition. Postharvest Biol. Technol., 12, Hopkirk, G., Harker, F.R., Harman, J.E. (1990 ). Calcium and firmness of kiwifruit. N. Z. J. Crop Hort. Sci., 18, Huber, D.J., O Donoghue, E.M. (1993 ). Polyuronides in avocado ( Persea americana ) and tomato ( Lycopersicon esculentum ) fruits exhibit markedly different patterns of molecular weight downshifts during ripening. Plant Physiol., 102, Ignarro, L.J., Balestrieri, M.L., Napoli, C. (2007 ). Nutrition, physical activity, and cardiovascular disease: An update. Cardiovasc. Res., 73, Institute of Food Technologists (1990 ). Quality of fruits and vegetables. A scientific status summary by the Institute of Food Technologists Expert Panel on Food Safety and Nutrition. Food Technol., 44, Institute of Medicine ( 2001 ). Dietary reference intakes proposed definition of dietary fiber. A report of the panel on the definition of dietary fiber and the Standing Committee on the scientific evaluation of dietary reference intakes. Food and Nutrition Board, National Academy Press, Washington, DC, USA. Jayaraj, J., Devlin, R., Punja, Z. (2008 ). Metabolic engineering of novel ketocarotenoid production in carrot plants. Transgenic Res. 17, Jeremy, J.Y., Shukla, N., Muzaffar, S., Handley, A., Angelini, G.D. (2004 ). Reactive oxygen species, vascular disease and cardiovascular surgery. Curr. Vasc. Pharmacol., 2,

259 Author s personal copy Bibliography 99 Jodral-Segado, A.M., Navarro-Alarcón, M., López-G de la Serrana, H., López Martínez, M.C. (2003 ). Magnesium and calcium contents in foods from SE Spain: Influencing factors and estimation of daily dietary intakes. Sci. Total Environ., 312, Kalt, W., Forney, C.F., Martin, A., Prior, R.L. (1999 ). Antioxidant capacity, vitamin C, phenolics, and anthocyanins after fresh storage of small fruits. J. Agric. Food Chem., 47, Kays, S.J. (1997 ). Postharvest Physiology of Perishables Plant Products. Exon Press, Athens, GA, USA. Kennedy, E., Bowman, S., Powell, R. (2001 ). Assessment of the effect of fat-modified foods on diet quality in adults, years, using data from the Continuing Survey of Food Intake by Individuals. J. Am. Diet. Assoc., 101, Kevers, C., Falkowski, M., Tabart, J., Defraigne, J.O., Dommes, J., Pinc , J. (2007 ). Evolution of antioxidant capacity during storage of selected fruits and vegetables. J. Agric. Food Chem., 55, Kliewer, W.M. (1977 ). Influence of temperature, solar radiation, and nitrogen on coloration and composition of Emperor grapes. Am. J. Enol. Vit., 28, Kobayashi, M., Matoh, T., Azuma, J. (1996 ). Two chains of rhamnogalacturonan II are cross-linked by borate-diol ester bonds in higher plant cell walls. Plant Physiol., 110, Koo, R.J.C., Young, T.W., Reese, R.L., Kesterson, J.W. (1974 ). Effect of nitrogen, potassium and irrigation on yield and quality of lemon. J. Am. Soc. Hort. Sci., 99, Kopsell, D.A., Kopsell, D.E. (2006 ). Accumulation and bioavailability of dietary carotenoids in vegetable crops. Trends Plant Sci., 11, Korcak, R.F. (1987 ). Iron deficiency chlorosis. Hort. Rev., 9, Land, D.G. (1988 ). Negative influences on acceptability and their control. In: Food Acceptability, D.M.H. Thomson (ed.). Elsevier, New York, NY, USA, pp Langcake, P., Pryce, R.J. (1976 ). The production of resveratrol by Vitis vinifera and other members of the Vitaceae as a response to infection and injury. Physiol. Plant Pathol., 9, Larson, R.A. (1988 ). The antioxidants of higher plants. Phytochemistry, 27, Le Marchand, L. (2002 ). Cancer preventive effects of flavonoids a review. Biomed. Pharmacother., 56, Lee, S.K., Kader, A.A. (2000 ). Preharvest and postharvest factors influencing vitamin C content of horticultural crops. Postharvest Biol. Technol., 20, Levine, M., Wang, Y.H., Padayatty, S.J., Morrow, J. (2001 ). A new recommended dietary allowance of vitamin C for healthy young women. Proc. Natl. Acad. Sci. U.S.A., 98, Lin, P.-H., Aickin, M., Champagne, C., Craddick, S., Sacks, F.M., McCarron, P., Most- Windhauser,M.M., Rukenbrod,F., Haworth, L. (2003 ). Food group sources of nutrients in the dietary patterns of the DASH-Sodium trial. J. Am. Diet. Assoc., 103, Liu, R.H. (2007 ). Whole grain phytochemicals and health. J. Cereal Sci., 46,

260 Author s personal copy 100 Nutritional Quality of Fruits and Vegetables Liu, Y., Roof, S., Ye, Z., Barry, C., van Tuinen, A., Vrebalov, J., Bowler,C., Giovannoni, J. (2004 ). Manipulation of light signal transduction as a means of modifying fruit nutritional quality in tomato. Proc. Natl. Acad. Sci. U.S.A., 101, Loaiza-Velarde, J.G., Tomás-Barberán, F.A., Saltveit, M.E. (1997 ). Effect of intensity and duration of heat-shock treatments on wound-induced phenolic metabolism in Iceberg lettuce. J. Am. Soc. Hortic. Sci., 122, Lurie, S., Crisosto, C.H. ( 2005 ). Chilling injury in peach and nectarine. Postharvest Biol. Technol., 37, MacDonald, H.M. (2007 ). Influence of organic salts of potassium on bone health: Possible mechanisms of action for the role of fruit and vegetables. Int. Congr. Ser., 1297, Manach, C., Scalbert, A., Morand, C., Remesy, C., Jimenez, L. (2004 ). Polyphenols: Food sources and bioavailability. Am. J. Clin. Nutr., 79, Manganaris, G.A., Vasilakakis, M., Mignani, I., Diamantidis, G., Tzavella-Klonari, K. ( 2005 a ). The effect of preharvest calcium sprays on quality attributes, physicochemical aspects of cell wall components and susceptibility to brown rot of peach fruits (Prunus persica L. cv. Andross). Sci. Hort., 107, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. ( 2005 b ). Effect of post-harvest calcium treatments on the physicochemical properties of cell wall pectin in nectarine fruit during ripening after harvest or cold storage. J. Hort. Sci. Biotechnol., 80, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. (2006 ). Effect of inseason calcium applications on cell wall physicochemical properties of nectarine fruit ( Prunus persica var. nectarina Ait. Maxim) after harvest or cold storage. J. Sci. Food Agric., 86, Manganaris, G.A., Vasilakakis, M., Diamantidis, G., Mignani, I. (2007 ). The effect of postharvest calcium application on tissue calcium concentration, quality attributes, incidence of flesh browning and cell wall physicochemical aspects of peach fruits. Food Chem., 100, Marlett, J.A. (1992 ). Content and composition of dietary fiber in 117 frequently consumed foods. J. Am. Diet. Assoc., 92, Marlett, J.A., Longacre, M.J. (1997 ). Comparisons of in vitro and in vivo measures of resistant starch in selected grain products. Cereal Chem., 73, Marlett, J.A., McBurney, M.I., Slavin, J.L. (2002 ). Position of the American Dietetic Association health implications of dietary fiber. J. Am. Diet. Assoc., 102, Marlett, J.A., Slavin, J.L. (1997 ). Position of the American Dietetic Association health implications of dietary fiber. J. Am. Diet. Assoc., 97, Marsh, K.B., Volz, R.K., Cashmore, W., Reay, P. (1996 ). Fruit colour, leaf nitrogen level, and tree vigour in Fuji apples. N. Z. J. Crop Hort. Sci., 24, Martín-Diana, A.B., Rico, D., Frías, J.M., Barat, J.M., Henehan, G.T.M., Barry-Ryan, C. ( 2007 ). Calcium for extending the shelf life of fresh whole and minimally processed fruits and vegetables: A review. Trends Food Sci. Technol., 18, Martínez-Romero, D., Valero, D., Serrano, M., Riquelme, F. (1999 ). Effects of postharvest putrescine and calcium treatments on reducing mechanical damage and polyamine and ABA levels during lemon storage. J. Sci. Food Agric., 79,

261 Author s personal copy Bibliography 101 Mattheis, J.P., Fellman, J.K. (1999 ). Preharvest factors influencing flavor of fresh fruit and vegetables. Postharvest Biol. Technol., 15, Mattila, P., Hellstrom, J., Torronen, R. (2006 ). Phenolic acids in berries, fruits, and beverages. J. Agric. Food Chem., 54, Mayer, A.M. ( 1997 ). Historical changes in the mineral content of fruits and vegetables. British Food J., 99, McCarron, D.A., Reusser, M.E. (2001 ). Are low intakes of calcium and potassium important causes of cardiovascular disease?. Am. J. Hyp. 14, S206 S212. Meléndez-Martínez, A.J., Vicario, I.M., Heredia, F.J. (2007 ). Provitamin A carotenoids and ascorbic acid contents of the different types of orange juices marketed in Spain. Food Chem., 101, Mengel, K., Kirkby, E.A. ( 1982 ). Principles of Plant Nutrition. International Potash Institute, Worblaufen-Bern, Switzerland. Michailides, T.J., Ramirez, H.T., Morgan, D.P., Crisosto, C.H., Johnson, R.S. (1993). Effects of nitrogen fertilization on susceptibility of stone fruits (peach and nectarine) to brown rot. In: 1992 Research Reports for California Peaches and Nectarines. California Tree Fruit Agreement, Sacramento, CA, USA. Mittler, R. ( 2002 ). Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci., 7, Montanaro, G., Dichio, B., Xiloyannis, C., Celano, G. ( 2006 ). Light influences transpiration and calcium accumulation in fruit of kiwifruit plants ( Actinidia deliciosa var. deliciosa ). Plant Sci., 170, Morris, J.R., Sistrunk,W.A., Sims, C.A., Main, G.L. (1985 ). Effects of cultivar, postharvest storage, preprocessing dip treatments and style of pack on the processing quality of strawberries. J. Am. Soc. Hort. Sci., 110, Muir, S.R., Collins, G.J., Robinson, S., Hughes, S., Bovy, A., Ric De Vos, C.H., van Tunen, A.J., Verhoeyen, M.E. (2001 ). Overexpression of petunia chalcone isomerase in tomato results in fruit containing increased levels of flavonols. Nat. Biotechnol., 19, Murad, S., Grove, D., Lindberg, K.A., Reynolds, G., Sivarajah, A., Pinnell, S.R. (1981 ). Regulation of collagen synthesis by ascorbic acid. Proc. Natl. Acad. Sci. U.S.A., 78, Mussatto, S.I., Mancilha, I.M. (2007 ). Non-digestible oligosaccharides: A review. Carbohydr. Polym., 68, Neilsen, G.H., Neilsen, D. (2003 ). Nutritional requirements of apple. In: Apples: Botany, production and uses, D.C. Ferree, I.J. Warrington (eds). CABI Publishing, Wallingford, Oxon, UK, pp Nelson, J.W., Barritt, B.H., Wolford, E.R. (1972 ). Influence of location and cultivar on color and chemical composition of strawberry fruit. Wash. Agric. Exp. Stn. Tech. Bull., 74, 1 7. Nishiyama, I., Yamashita, Y., Yamanaka, M., Shimohashi, A., Fukuda, T., Oota, T. (2004 ). Varietal difference in vitamin C content in the fruit of kiwifruit and other Actinidia species. J. Agric. Food Chem., 52, Noctor, G., Foyer, C. (1998 ). Ascorbate and glutathione: Keeping active oxygen under control. Annu. Rev. Plant Physiol. Plant Mol. Biol., 49,

262 Author s personal copy 102 Nutritional Quality of Fruits and Vegetables Nordin, B.E.C. (1997 ). Calcium and osteoporosis. Nutrition, 13, O Donoghue, E.M., Huber, D.J., Timpa, J.D., Erdos, G.W., Brecht, J.K. (1994 ). Influence of avocado ( Persea americana ) Cx-cellulase on the structural features of avocado cellulose. Planta, 194, O Neill, M.A., Ishii, T., Albersheim, P., Darvill, A.G. (2004 ). Rhamnogalacturonan II: Structure and function of a borate cross-linked cell wall pectic polysaccharide. Annu. Rev. Plant Biol., 55, Pantelidis, G.E., Vasilakakis, M., Manganaris,G.A., Diamantidis, G. (2007 ). Antioxidant capacity, phenol, anthocyanin and ascorbic acid contents in raspberries, blackberries, red currants, gooseberries and cornelian cherries. Food Chem., 102, Peleg, K. ( 1985 ). Produce handling, packaging, and distribution. AVI/Van Nostrand Reinhold, New York, NY, USA. Perkins-Veazie, P., Collins, J.K., Howard, L. (2008 ). Blueberry fruit response to postharvest application of ultraviolet radiation. Postharvest Biol. Technol., 47, Petersen, J., Pedersen, G. (1991). Sporelementer I økologisk dyrkede grønsaker. Levnedsmiddelstyrelsen. Rapport nr. F Levnedsmiddelstyrelsen, Copenhagen, Denmark (in Danish). Pomeranz, Y., Meloan, C.E. (1987 ). Food Analysis Theory and Practice, 2nd edn. AVI/ Van Nostrand Reinhold, New York, NY, USA. Poovaiah, B.W., Glenn, G.M., Reddy, A.S.N. (1988 ). Calcium and fruit softening: physiology and biochemistry. Hort. Rev., 10, Prange, R., DeEll, J.R. (1997 ). Preharvest factors affecting quality of berry crops. HortScience, 32, Premuzic, Z., Bargiela, M., García, A., Rendina, A., Iorio, A. (1998 ). Calcium, iron, potassium, phosphorus, and vitamin C content of organic and hydroponic tomatoes. HortScience, 33, Prior, R.L., Cao, G., Martin, A., Sofic, E., McEwen, J., O Brien, C., Lischner, N., Ehlenfeldt, M., Kalt, W., Krewer, G., Mainland, C.M. (1998 ). Antioxidant capacity as influenced by total phenolic and anthocyanin content, maturity, and variety of Vaccinium species. J. Agric. Food Chem., 46, Raese, J.T., Drake, S.R. (1993 ). Effects of preharvest calcium sprays on apple and pear quality. J. Plant Nutr., 16, Raese, J.T., Drake, S.R. (2000a ). Effect of calcium sprays, time of harvest, cold storage, and ripeness on fruit quality of Anjou pears. J. Plant Nutr., 23, Raese, J.T., Drake, S.R. ( 2000 b ). Effect of calcium spray materials, rate, time of spray application, and rootstocks on fruit quality of Red and Golden Delicious apples. J. Plant Nutr., 23, Raese, J.T., Drake, S.R. (2002 ). Calcium spray materials and fruit calcium concentrations influence apple quality. J. Am. Pom. Soc., 56, Raese, J.T. ( 1977 ). Response of young d Anjou pear trees to triazine and triazole herbicides and nitrogen. J. Am. Soc. Hort. Sci., 102, Raghothama, K.G. (1999 ). Phosphate acquisition. Annu. Rev. Plant Physiol. Plant Mol. Biol., 50,

263 Author s personal copy Bibliography 103 Ramalingam, K., Lee, K.M., Woodard, R.M., Bleecker, A.B., Kende, H. (1985 ). Stereochemical course of the reaction catalyzed by the pyridoxal phosphate-dependent enzyme 1-aminocyclopropane-1-carboxylate synthase. Proc. Natl. Acad. Sci. USA, 82, Randle, W. (1997). Onion flavor chemistry and factors influencing flavor intensity. In: Spices: Flavor Chemistry and Antioxidant Properties, S.J. Risch, C.T. Ho (eds). ACS Symposium Series 660. ACS Press, Washington, DC, USA, pp Rao, A.V., Rao, L.G. (2007 ). Carotenoids and human health. Pharmacol. Res., 55, Reddy, M.S., Chen, F., Shadle, G., Jackson, L., Aljoe, H., Dixon, R.A. ( 2005 ). Targeted down-regulation of cytochrome P450 enzymes for forage quality improvement in alfalfa ( Medicago sativa L.). Proc. Nat. Acad. Sci. USA, 102, Redgwell, R.J., MacRae, E., Hallett, I., Fisher, M., Perry, J., Harker, R. (1997 ). In vivo and in vitro swelling of cell walls during fruit ripening. Planta, 203, Reeve, R.M. (1970 ). Relationships of histological structure to texture of fresh and processed fruits and vegetables. J. Text. Stud., 1, Rice-Evans, C.A., Miller, N., Paganga, G. (1997 ). Antioxidant properties of phenolic compounds. Trends Plant Sci., 2, Ridley, B.L., O Neill, M.A., Mohnen, D. (2001 ). Pectins, structure, biosynthesis, and oligogalacturonide-related signaling. Phytochemistry, 57, Rimm, E.B., Ascherio, A., Giovannucci, E., Spiegelman, D., Stampfer, M.J., Willett, W.C. (1996 ). Vegetable, fruit, and cereal fiber intake and risk of coronary heart disease among men. J. Am. Med. Assoc., 275, Rodriguez-Amaya, D.B. (2001 ). A Guide to Carotenoid Analysis in Foods. ILSI Press, International Life Sciences Institute, One Thomas Circle, N.W. Washington, DC, USA. Romheld, V., Marschner, H. ( 1991 ). Function of micronutrients in plants. In: Micronutrients in agriculture, J.J. Mortvedt, F.R. Cox, L.M. Shuman, R.M. Welch (eds), 2nd edn. Soil Science Society of America, Madison, WI, USA, pp Rose, J.K.C., Catalá, C., Gonzalez-Carranza, Z.H., Roberts, J.A. ( 2003 ). Cell wall disassembly. In: The Plant Cell Wall, J.K.C. Rose (ed.). Blackwell Publishing Ltd., Oxford, UK, pp Rosen, J.C., Kader, A.A. (1989 ). Postharvest physiology and quality maintenance of sliced pear and strawberry fruit. J. Food Sci., 54, Sajilata, M.G., Singhal, R.S., Kulkarni, P.R. ( 2006 ). Resistant starch A review. Compr. Rev. Food Sci. Food Safety, 5, Saltveit, M.E. (1988 ). Postharvest glyphosate application reduces toughening fiber content and lignification of stored asparagus spears. J. Am. Soc. Hort. Sci., 113, Salunkhe, D.K., Bolin, H.R., Reddy, N.R. (eds) (1991 ). Storage, processing, and nutritional quality of fruits and vegetables. Volume I. Fresh Fruits and Vegetables. CRC Press, Boston, MA, USA. Sams, C.E. (1999 ). Preharvest factors affecting postharvest texture. Postharvest Biol. Technol., 15, Sanders, D., Brownlee, C., Harper, J.F. (1999 ). Communicating with calcium. Plant Cell, 11,

264 Author s personal copy 104 Nutritional Quality of Fruits and Vegetables Sandmann, G. (2001 ). Genetic manipulation of carotenoid biosynthesis, strategies, problems and achievements. Trends Plant Sci., 6, Sanmartin, M., Pateraki, I., Chatzopoulou, F., Kanellis, A.K. (2007 ). Differential expression of the ascorbate oxidase multigene family during fruit development and in response to stress. Planta, 225, Saure, M.C. ( 1998 ). Causes of the tipburn disorder in leaves of vegetables. Sci. Hort., 76, Saure, M.C. (2001 ). Blossom-end rot of tomato (Lycopersicon esculentum Mill.) a calcium or stress-related disorder?. Sci. Hort., 90, Saure, M.C. (2005 ). Calcium translocation to fleshy fruit: Its mechanism and endogenous control. Sci. Hort., 105, Scalbert, A., Williamson, G. (2000 ). Dietary intake and bioavailability of polyphenols. J. Nutr., 130, Seeram, N.P., Adams, L.S., Zhang, Y., Lee, R., Sand, D., Scheuller, H.S., Heberg, D. (2006 ). Raspberry, blueberry, cranberry, red raspberry, and strawberry extracts inhibit growth and stimulate apoptosis of human cancer cells in vitro. J. Agric. Food Chem., 54, Sharples, R.O. (1980 ). The influence of orchard nutrition on the storage quality of apples and pears grown in the United Kingdom. In: Mineral Nutrition of Fruit Trees, D. Atkinson, J.E. Jackson, R.O. Sharples, W.M. Waller (eds). Butterworths, London, UK, pp Shewfelt, R.L. (1986 ). Postharvest treatment for extending the shelf life of fruits and vegetables. Food Technol., 40, Sigman-Grant, M., Warland, R., Hsieh, G. (2003 ). Selected lower-fat foods positively impact nutrient quality in diets of free-living Americans. J. Am. Diet. Assoc., 103, Sistrunk, W.A. ( 1985 ). Peach quality assessment, fresh and processed. In: Evaluation of Quality of Fruits and Vegetables, H.E. Patee (ed.) AVI, Westport, pp Slavin, J.L. (2005 ). Dietary fiber and body weight. Nutrition, 21, Smirnoff, N. (2000 ). Ascorbic acid, metabolism and functions of a multifaceted molecule. Curr. Opin. Plant Biol., 3, Smirnoff, N., Wheeler, G.L. (2000 ). Ascorbic acid in plants, Biosynthesis and function. Crit. Rev. Biochem. Mol. Biol., 35, Smith, B. (1993 ). Organic food vs. supermarket food, Element levels. J. Appl. Nutr., 45, Sozzi, G.O. ( 2001 ). Caper bush, botany and horticulture. Hort. Rev., 27, Swietlik, D., Faust, M. ( 1984 ). Foliar nutrition of fruit crops. Hort. Rev., 6, Swietlik, D. ( 1999 ). Zinc nutrition in horticultural crops. Hort. Rev., 23, Terry, P., Giovannucci, E., Michels, K.B., Bergkvist, L., Hansen, H., Holmberg, L., Wolk, A. (2001 ). Fruit, vegetables, dietary fiber, and risk of colorectal cancer. J. Nat. Canc. Instit., 93, Titchenal, C.A., Dobbs, J. (2007 ). A system to assess the quality of food sources of calcium. J. Food Compos. Anal., 20, Toivonen, P.M.A., Zebarth, B.J., Bowen, P.A. (1994 ). Effect of nitrogen fertilization on head size, vitamin C content and storage life of broccoli ( Brassica oleracea var. italica). Can. J. Plant Sci., 74,

265 Author s personal copy Bibliography 105 Tomás-Barberán, F.A., Loaiza-Velarde, J., Bonfanti, A., Saltveit, M.E. ( 1997 ). Early wound- and ethylene-induced changes in phenylpropanoid metabolism in harvested lettuce. J. Am. Soc. Hort. Sci., 122, Tripoli, E., La Guardia, M., Giammanco, S., Di Majo, D., Giammanco, M. ( 2007 ). Citrus flavonoids, molecular structure, biological activity and nutritional properties, a review. Food Chem., 104, Trudel, M.J., Ozbun, J.L. ( 1971 ). Influence of potassium on carotenoid content of tomato fruit. J. Am. Soc. Hort. Sci., 96, Tsao, R., Deng, Z. (2004 ). Separation procedures for naturally occurring antioxidant phytochemicals. J. Chromatogr. B., 812, Uriu-Adams, J.Y., Keen, C.L. ( 2005 ). Copper, oxidative stress, and human health. Mol. Aspects Med., 26, U.S. Department of Agriculture (2008). Composition of foods, raw, processed, prepared. USDA national nutrient database for standard reference, release 20. USDA-ARS, Beltsville Human Nutrition Research Center, Nutrient Data Laboratory, Beltsville, Maryland. Accessed April U.S. Department of Health and Human Services and U.S. Department of Agriculture ( 2005 ). Dietary Guidelines for Americans, 2005, 6th edn. U.S. Government Printing Office, Washington, DC, USA. Valero, D., Martínez-Romero, D., Serrano, M., Riquelme, F. (1998 ). Influence of postharvest treatment with putrescine and calcium on endogenous polyamines, firmness, and abscisic acid in lemon ( Citrus limon L. Burn cv. Verna). J. Agric. Food Chem., 46, Valpuesta, V., Botella, M.A. (2004 ). Biosynthesis of L-ascorbic acid in plants, new pathways for an old antioxidant. Trends Plant Sci., 9, Versari, A., Parpinello, G.P., Tornielli, G.B., Ferrarini, R., Giulivo, C. (2001 ). Stilbene compounds and stilbene synthase expression during ripening, wilting, and UV treatment in grape cv. Corvina. J. Agric. Food. Chem., 49, Vicente, A.R., Ortugno, C., Rosli, H., Powell, A.L.T., Greve, C., Labavitch, J.M.L. (2007a ). Temporal sequence of cell wall disassembly events in developing fruits. 2. Analysis of blueberry ( Vaccinium Species). J. Agric. Food Chem., 55, Vicente, A.R., Saladié, M., Rose, J.C.K., Labavitch, J.M. ( 2007 b ). The linkage between cell wall metabolism and fruit softening, looking to the future. J. Sci. Food Agric., 87, von Elbe, J.H., Schwartz, S.J. (1996 ). Colorants. In: Food Chemistry, O.R. Fennema (ed.) 3rd edn. Marcel Dekker, New York, NY, USA. Wall, M.M. (2006 ). Ascorbic acid, vitamin A, and mineral composition of banana (Musa sp.) and papaya ( Carica papaya ) cultivars grown in Hawaii. J. Food Compos. Anal., 19, Wang, C.Y., Conway, W.S., Abbott, J.A., Kramer, G.F., Sams, C.E. ( 1993 ). Postharvest infiltration of polyamines and calcium influences ethylene production and texture changes in Golden Delicious apples. J. Am. Soc. Hort. Sci., 118, Wang, H., Cao, G., Prior, R.L. ( 1996 ). Total antioxidant capacity of fruits. J. Agric. Food Chem., 44, Wang, S.Y., Lin, H.S. (2000 ). Antioxidant activity in fruits and leaves of blackberry, raspberry, and strawberry varies with cultivar and developmental stage. J. Agric. Food Chem., 48,

266 Author s personal copy 106 Nutritional Quality of Fruits and Vegetables Wang, S.Y., Lin, H.S. ( 2003 ). Compost as a soil supplement increases the level of antioxidant compounds and oxygen radical absorbance capacity in strawberries. J. Agric. Food Chem., 51, Wang, S.Y., Zheng, W., Galletta, G.J. (2002 ). Cultural system affects fruit quality and antioxidant capacity in strawberries. J. Agric. Food Chem., 50, Waris, G., Ahsan, H. (2006 ). Reactive oxygen species, role in the development of cancer and various chronic conditions. J. Carcinog., 6, Weibel, F., Bickel, R., Leuthold, S., Alfoldi, T. (2000 ). Are organically grown apples tastier and healthier? A comparative field study using conventional and alternative methods to measure fruit quality. Acta Hort., 57, Willats, W.G.T., McCartney, L., Mackie, W., Knox, J.P. ( 2001 ). Pectin, cell biology and prospects for functional analysis. Plant Mol. Biol., 47, Winter, C.K., Davis, C.F. (2006 ). Organic foods. J. Food Sci., 71, R117 R124. Witney, G.W., Hofman, P.M., Wolstenholme, B.N. (1990a ). Effect of cultivar, tree vigour and fruit position on calcium accumulation in avocado fruits. Sci. Hort., 44, Witney, G.W., Hofman, P.J., Wolstenholme, B.N. (1990b ). Mineral distribution in avocado trees with reference to calcium cycling and fruit quality. Sci. Hort., 44, Witney, G.W., Kushad, M.M., Barden, J.A. ( 1991 ). Induction of bitter pit in apple. Sci. Hort., 47, Woese, K., Lange, D., Boess, C., Bogel, K.W. (1997 ). A comparison of organically and conventionally grown foods results of a review of the relevant literature. J. Sci. Food Agric., 74, Wolk, A., Manson, J.A.E., Stampfer, M.J., Colditz, G.A., Hu, F.B., Speizer, F.E., Hennekens, C.H., Willett, W.C. (1999 ). Long-term intake of dietary fiber and decreased risk of coronary heart disease among women. J. Am. Med. Assoc., 281, Zerwekh, J.E., Odvina, C.V., Wuermser, L.-A., Pak, C.Y.C. (2007 ). Reduction of renal stone risk by potassium-magnesium citrate during 5 weeks of bed rest. J. Urol., 177, Zheng, W., Wang, S.W. (2003 ). Oxygen radical absorbing capacity of phenolics in blueberries, cranberries, chokeberries, and lingonberries. J. Agric. Food Chem., 51, Zheng, Y., Wang, C.Y., Wang, S.Y., Zheng, W. (2003 ). Effect of high-oxygen atmospheres on blueberry phenolics, anthocyanins, and antioxidant capacity. J. Agric. Food Chem., 51, Zyren, J., Elkins, R., Dudek, J.A., Hagen, R.E. (1983 ). Fiber contents of selected raw and processed vegetables, fruits and fruit juices as served. J. Food Sci., 48,

267 20 Preharvest Factors Affecting Peach Quality C.H. Crisosto 1 and G. Costa 2 1 University of California, Davis, Department of Plant Sciences; located at Kearney Agricultural Center, Parlier, California, USA 2 Department of Fruit Trees and Woody Plant Sciences, University of Bologna, Bologna, Italy 20.1 Introduction Quality Definition Maturity and Quality Genotype Mineral Nutrition 540 Nitrogen 540 Calcium 541 Potassium 542 Iron Irrigation Canopy Manipulation Introduction This chapter describes and discusses exclusively the impact of preharvest factors on peach fruit flavour and postharvest life (storage and shelf-life). It does not include the role of environmental factors and the use of plant growth regulators on peach quality; these topics are covered well in other chapters in this book. The present chapter begins by emphasizing quality definitions from the consumer point of view and follows with a series of short sections that update knowledge on preharvest orchard factors. The relationship between maturity and quality is covered, and the role of genotype (cultivar and rootstock) on fruit flavour and postharvest life potential is described. Then the effect of mineral nutrition on peach quality is discussed with detailed attention on N and Ca as the most studied nutrients. Updated information on the effect of foliar nutrient application on fruit quality including foliar Ca sprays is also reported. Detailed practical information on the effect of different irrigation regimes on fruit quality is described next, followed by a section on canopy management. The canopy management section describes practical information on leaf removal, girdling techniques, and the use of reflective materials to improve fruit size and enhance red skin colour. Examples that illustrate the influence of crop load 536 CAB International The Peach: Botany, Production and Uses (eds D.R. Layne and D. Bassi)

268 Preharvest Factors Affecting Quality 537 and canopy position on fruit quality and postharvest storage potential are presented Quality Definition Fruit quality is a concept encompassing sensory properties (appearance, texture, taste and aroma), nutritive value, mechanical properties, safety and defects. Altogether, these attributes give the fruit a degree of excellence and an economic value (Abbott, 1999). Everyone in the peach production and marketing chain from the grower to the consumer looks for fruit with no or few defects. However, in each step of this chain, the term quality takes on different meanings and the economic relevance of the various quality traits is largely variable. For example, the grower is interested in high yield, in fruit with large size and high disease resistance, and in the opportunity to reduce the number of pickings. The definition of quality for packers, shippers, distributors and wholesalers is mainly based on flesh firmness, which is considered a good indication to predict fruit potential storage and market life. Peaches and nectarines ripen and deteriorate quickly at ambient temperature and cold storage is required to slow down these processes, especially for some cultivars and/or long-distance market situations. However, firmness is an erroneous and incomplete way to estimate peach postharvest potential for domestic distribution. In fact, in some production areas such as California and Chile, the development of internal breakdown symptoms such as lack of flavour, flesh mealiness and flesh browning limits the storage life and the postharvest quality of tasty cultivars. For retailers, red colour, size and firmness have historically represented the main components of fruit quality, as they need fruit that are attractive to the consumer, resistant to handling and have a long shelflife. From the consumer s point of view, in general peach fruit quality has declined, mainly because of premature harvesting, chilling injury and lack of ripening prior to consumption, resulting in consumer dissatisfaction. In addition, quality is badly defined and the only parameters being considered are fruit size and skin colour. Other characters such as flesh firmness, sugar content, acidity and aroma, which are perceived by the consumer as fruit quality, are completely disregarded by the grower and other individuals along the chain. In fact, the grower, identifying fruit quality almost exclusively with the fruit size, does not consider that these are only the first characters perceived by the consumer and they orient him just in his very first choice. As soon as he realizes that the fruit, even with good size and attractive colour, is tasteless, with low sugar content, poor aroma and rapidly perishable, he redirects his interest towards other types of fruit. As a consequence, it is imperative for the grower and other individuals in the delivery chain to direct their attention to fruit quality from the consumer s perspective in order to regain the confidence of the consumer. In addition, there is now an increasing appreciation that quality of fruit also includes nutritional properties (e.g. vitamins, minerals, dietary fibre) and health benefits (e.g. antioxidants); and these are becoming important factors in consumer preferences. Experimental, epidemiological and clinical studies provide evidence that diet has an important role in the prevention of the chronic degenerative diseases such as tumours, cardiovascular diseases and atherosclerosis. It is supposed, in fact, that the consumption of fresh fruit and vegetables exerts a protective role against the development of such pathologies (Doll, 1990; Ames et al., 1993; Dragsted et al., 1993; Anderson et al., 2000). Changes in quality definition that are focusing more on consumer demand can increase peach consumption if marketing promotion and education programmes are well executed. Because the consumer quality of peaches cannot be improved after harvest, it is important to understand the role of preharvest factors in consumer acceptance and market life (Kader, 1988; Crisosto et al., 1997) Maturity and Quality Peaches and nectarines are climacteric fruit characterized by a sharp rise in ethylene biosynthesis at the onset of ripening, which is associated with changes in sensitivity to the hormone itself and changes in colour, texture, aroma and other biochemical features (Fig. 20.1/Plate 226).

269 538 C.H. Crisosto and G. Costa Ethylene plays a key role in peach fruit ripening by coordinating the expression of ripening-related genes responsible for flesh softening, colour development and sugar accumulation, as well as other processes such as abscission (Ruperti et al., 2002; Trainotti et al., 2003, 2006). The definition of the proper harvest time is essential, as fruit maturity at harvest greatly influences peach fruit market life potential and quality. Recently, the most important peach-producing countries in Europe have lost considerable market share mainly due to excessive early harvesting. A delayed harvest could lead to a better fruit organoleptic quality but also to faster softening and a shorter shelf-life. In fact, different from other species, in peach fruit there is a close link between on-tree physiological maturity and evolution of key traits responsible for peach quality during the postharvest phase. On the other hand, melting flesh peaches and nectarines undergo a rapid softening after harvest, which leads to dramatic losses in the marketing chain, as soft fruit are easily bruised during handling and more susceptible to decay. Therefore, they are often picked at an early stage of ripening, and they never reach their full flavour and aroma potential. Modulation of pre- and postharvest peach fruit ripening by the means of chemicals that interfere with ethylene biosynthesis and/or perception, such as aminoethoxyvinylglycine and 1-methylcyclopropene, has already been reported (Mathooko et al., 2001; Bregoli et al., 2002, 2005; Ziosi et al., 2006). A better understanding of the physiological basis of the peach fruit ripening process should make it possible to develop further strategies to regulate ripening. Such strategies need objective parameters able to accurately describe fruit ripeness stages and internal quality changes occurring in pre- and postharvest conditions. Until recently, few studies have been carried out on this topic, and mainly by using traditional fruit quality traits (flesh firmness, soluble solids concentration (SSC) and titratable acidity (TA)) which are assessed with simple devices such as penetrometers, refractometers and titrators. Early studies carried out in Europe and the USA have associated peach fruit consumer acceptance with high SSC (Mitchell et al., 1990; Parker et al., 1991; Ravaglia et al., 1996; Anon., 1999). In California, Fig Peaches picked at different maturity levels.

270 Preharvest Factors Affecting Quality 539 a minimum of 10% SSC for yellow-fleshed peaches and nectarines was proposed as a quality standard (Kader, 1995). In France, a minimum of 10% SSC for low-acidity (TA <0.9%) and 11% SSC for high-acidity (TA 0.9%) peaches was proposed as part of their quality standard (Hilaire, 2003). In Italy, a minimum of 10% SSC for early-season, 11% for mid-season and 12% for late-season cultivars was suggested for yellow-fleshed peaches (Testoni, 1995; Ventura et al., 2000). In preliminary studies carried out in California by using trained panels and in-store consumer acceptance tests on Ivory Princess (white flesh/low TA), Elegant Lady, O Henry and Spring Bright (yellow flesh/high TA) peaches and Honey Kist (yellow flesh/low TA) nectarine, it was shown that acceptance correlated well with ripe soluble solids concentration or the ratio of ripe soluble solids concentration to ripe titratable acidity; it was also shown that the relationship was strictly dependent on cultivar and/or maturity and that consumer acceptance was not a linear relationship (Crisosto and Crisosto, 2005). The analyses of traditional fruit quality traits are cheap and fast, but they do not consider other fundamental aspects of quality, such as antioxidant capacity, aroma volatile emission, soluble sugars and organic acids content. A more accurate definition of fruit quality would require sophisticated analyses (high-performance liquid chromatography, gas chromatography or mass spectrometry) that are not usually run because they should be carried out only in well-equipped laboratories with trained personnel. In any case, simple or more complex destructive analyses can be performed only on samples of a limited number of fruit, often not fully representative of the entire lot (Costa et al., 2002, 2003b). In recent years, extensive research has been focused on the development of non-destructive techniques for assessing internal fruit quality attributes. These techniques offer a number of advantages, including: the possibility to extend the assessments on a large number of, or even on all, the fruit in a lot; to repeat the analysis on the same samples, monitoring their physiological evolution; and to achieve real-time information on several fruit quality parameters at the same time (Abbott, 1999). Among the non-destructive techniques, near infrared spectroscopy can be used efficiently for determining traditional peach fruit quality traits and concentrations of the main organic acids and simple sugars. In addition, this technique allowed definition of a new maturity index strictly related to the fruit ethylene emission and ripening stage. This index, called absorbance difference (AD), can be effectively used for determining harvest date and for grouping harvested fruit into homogeneous classes which show different ripening rates during shelf-life (Costa et al., 2006). As a final consideration, as new plantings are based on new cultivars with different organoleptic characteristics (low- and highacid, high SSC, highly aromatic, non-melting, etc.) and since new markets and consumer groups with different ethnic backgrounds are being reached (Liverani et al., 2002; Crisosto, 2003), it is important to understand which characters are determining consumer acceptance and segregate cultivars into different organoleptic categories prior to proposing any quality index (Crisosto, 2002, 2003). As a long-term solution, it is expected that breeding programmes will include quality characteristics in their screening process. The creation of peach categories with their own quality indices according to an organoleptic description may help marketing and promotion Genotype Genotype (cultivar and/or rootstock) has an important role in flavour quality, nutrient composition and postharvest life potential. SSC and acidity are determined by several factors such as cultivar (Crisosto et al., 1995, 1997; Frecon et al., 2002; Liverani et al., 2002; Byrne, 2003) and rootstock (Reighard, 2002). Reduction of physiological disorders and even decay and insect losses can be achieved by choosing the correct genotype for given environmental conditions. Extensive harvest quality and postharvest storage potential evaluations have been carried out since 1970 by several researchers in all the main important peach cultivation areas, such as the USA,

271 540 C.H. Crisosto and G. Costa Italy, Spain, France, Chile and South Africa. Brown rot and grey mould resistance have not been successfully included in recently released cultivars. These are the main diseases, although other ones have been investigated (Frecon et al., 2002; Reighard, 2002), but current breeding programmes are constantly creating new cultivars with improved production and visual appearance attributes. Unfortunately, an ideal cultivar(s) with all of the current consumer quality attributes for domestic and long-distance shipping has (have) not been developed yet Mineral Nutrition Nutritional status is an important factor of quality and postharvest life potential. Deficiencies, excesses or imbalances of various nutrients may result in disorders that can limit storage life. Fertilization rates vary widely among growers, locations and cultivars, and generally depend upon soil type, cropping history and field testing results. Nitrogen This is the nutrient that has been studied the most. N has the single greatest effect on peach quality. Detailed and extensive research performed since the early 1990s at the Kearney Agricultural Center (Parlier, California, USA) has evaluated the role of N in peach and nectarine production and quality under California conditions (Daane et al., 1995). Based on this work, in California, N should be kept between 2.6 and 3.0% leaf N for best fruit quality without reduction in production or size (Table 20.1). Similarly, optimal fruit quality in nectarines in the Eastern Po Valley area (Italy) was obtained in trials having 3.0% leaf N concentration (Tagliavini et al., 1997; Scudellari et al., 1999). Response of peach and nectarine trees to N fertilization is dramatic; high N levels stimulate vigorous vegetative growth, causing shading out and death of lower fruiting wood. Although high-n trees may look healthy and lush, excess N does not increase fruit size, production or SSC. Furthermore, excessive N delays peach maturity because it induces poor visual red colour development (Fig. 20.2/Plate 227) and inhibits ground colour change from green to yellow. As growers delay harvest waiting for fruit colour changes from green to yellow and red colour development, high-n fruit are picked soft especially when measured on the softest position on the fruit such as tips, which generally ripen faster than the rest of the fruit in warm production areas. These fruits then have fast softening rates during postharvest handling and are more susceptible to bruising and decay development. N deficiency leads to small fruit with poor flavour and unproductive trees. Fruit water loss from fruit with the highest N rate (3.6% leaf N) was greater than that from the lowest rate (2.6% leaf N). The relationship between fruit N concentration and fruit susceptibility to decay produced by brown rot (Monilinia fructicola (Wint.) Honey) has been studied extensively on stored fruit (Daane et al., 1995). Wounded and brown rotinoculated fruit from Fantasia and Flavortop Table Relationship between leaf nitrogen and per cent of fruit surface that is red, yield and fruit size (mean for 3 years) on Fantasia nectarine. (Adapted from Daane et al., 1995.) N-fertilization treatment (kg N/ha) Leaf N (%) Fruit visual redness (%) Yield (kg/tree) Fruit weight (g) a 92 a 132 a 131 a b 80 b 207 b 166 b c 72 c 193 b 168 b d 69 c 222 b 169 b d 70 c 197 b 167 b a,b,c Values within columns with unlike superscript letters were signifi cantly different by the Least Signifi cant Difference test (P < 0.05).

272 Preharvest Factors Affecting Quality 541 Fig Influence of increased nitrogen fertilization (kg/ha) on red skin coloration of Fantasia nectarine. trees having more than 2.6% leaf N were more susceptible to brown rot than fruit from trees with 2.6% leaf N or less. Fruit anatomical observations and cuticle density measurements indicated differences in cuticle thickness among Fantasia fruit from the low, middle and high N treatments. These changes in cuticle and epidermis anatomy can partially explain the differences in fruit susceptibility to this disease and water loss. Calcium The nutrient Ca is involved in numerous biochemical and morphological processes in plants and has been implicated in many disorders of considerable economic importance to production and postharvest quality. While Ca accumulation in apple, kiwifruit and grape occurs predominantly in the first stages of fruit development, in peaches, owing to their ability to maintain significant transpiration rates, Ca continues to accumulate until harvest (Tagliavini et al., 2000). Foliar Ca sprays have not been successful and are not used commercially to maintain peach fruit quality. Over the last decade, trials carried out in California using several commercial Ca foliar sprays on peach and nectarine (applied every 14 days, starting 2 weeks after full bloom and continuing until 1 week before harvest) showed no effect on fruit quality of mid- or late-season cultivars (Crisosto et al., 2000). These foliar spray formulations and new formulations did not affect fruit SSC, firmness, decay incidence, fruit flesh Ca concentration or postharvest disorders. Fruit flesh Ca concentration measured at harvest varied among cultivars from 200 to 300 µg/g dry weight basis. A lack of decay control was also reported on Jerseyland peaches, grown in Pennsylvania, treated with ten weekly preharvest Ca sprays of CaCl 2 at 0,

273 542 C.H. Crisosto and G. Costa 34, 67 or 101 kg/ha (Conwall, 1987). Even fruit treated at a rate of 101 kg/ha, which had 70% more flesh Ca (490 versus 287 µg/g dry weight basis) than untreated fruit, showed no reduction in decay severity. Our recent research suggests that any Ca spray formulations and timing on peaches and nectarines should be treated with caution because their heavy metal content (Fe, Al, Cu, etc.) may contribute to peach and nectarine skin discoloration (Crisosto et al., 1999). A moderate and cultivar-dependent effect of Ca sprays on the reduction of skin russeting development has been reported for nectarines in Italy (Scudellari et al., 1995). Potassium K is the major nutrient present in peaches (about kg/t fresh weight basis), where it accumulates progressively as fruit approach maturity (Tagliavini et al., 2000). Optimal K nutrition usually leads to high photosynthetic rates and reallocation of sugars and organic acids that will enhance fruit quality. Iron Fe, as a micronutrient, is taken up by fruit trees in relatively small amounts; however, its deficiency not only affects fruit yields but also peach fruit quality (Álvarez-Fernández et al., 2003). In a study carried out in Spain, only 47% of fruits from Fe-deficient trees had optimal fruit size compared with 95% from green trees (Álvarez-Fernández et al., 2003). Peach fruit colour could also be affected by Fe deficiency: in a red-skin peach cultivar ( Babygold ) Fe deficiency caused decreases in the mean a colour coordinate and increases in the mean L and b colour coordinates (Álvarez- Fernández et al., 2003) Irrigation Despite the important role of water in fruit growth and development, few specific studies have been done on the influence of the amount and the timing of water applications on peach quality at harvest and postharvest performance (Prashar et al., 1976). An early report indicated that when trees were allowed to grow without irrigation during the growing season on a shallow soil under California conditions, yield and fruit size were reduced, SSC increased and fruit developed an abnormal texture (Uriu et al., 1964). Reducing the amount of applied water after harvest of earlyseason peaches (postharvest stress) has shown no negative effects on yield in California; however, timing of the water deficit interval is important. An increase in fruit defects such as deep suture and double-fruit formation has been reported for early-season Regina peaches as a consequence of imposing a postharvest water stress (50% evapotranspiration; ET) in mid- and late summer during the previous season (Fig. 20.3/Plate 228). These defects reduced the final packout for the next season s crop (Johnson et al., 1992). The regulated irrigation deficit (RID) technique has been evaluated for peach performance in different production areas (Chalmers et al., 1981; Ben Mechlia et al., 2002; Girona, 2002; Goldhamer et al., 2002). In general, this technique imposes a moderate stress (30 50% ET) to reduce vegetative growth and save water use (4 30%) at a given physiological stage without affecting yield. Researchers agree that the water stress-tolerant phases in peach, which has a double-sigmoid fruit development pattern, have been identified as stage II, the lag phase of fruit growth and the postharvest period (Goldhamer et al., 2002). In some situations, besides saving water, the RID technique also increased fruit size and SSC. Researchers claim that consistency of the benefits of the RID technology will depend on the understanding of local climatic conditions, soil depth and composition, identification of the fruit growth stages and fruit crop load (Berman and DeJong, 1996; Girona, 2002). In California, during three seasons, the influence of three different irrigation regimes applied 4 weeks before harvest on O Henry peach quality and postharvest performance was evaluated: (i) normal irrigation (100% evapotranspiration); (ii) overirrigation (150% ET); and (iii) RID irrigation (50% ET) (Crisosto et al., 1994; Johnson and

274 Preharvest Factors Affecting Quality 543 Fig Water stress late in the summer causes fruit defects such as deep sutures and double-fruit formation. Handley, 2000). Yield, flesh firmness, per cent red surface, acidity and ph were not altered at harvest by any of these irrigation regimes in any season. Average fruit size, measured as fruit weight, was lower but SSC was higher for fruit from 50% ET than from the other treatments. Ripe yellow-fleshed peaches and nectarines with 10% SSC or higher with low to moderate TA (<0.7%) are highly acceptable to consumers. Although fruit from the 50% ET treatment were smaller, they had higher SSC and consumers would probably prefer their eating quality over fruit from the other two treatments. An economic study showed that peaches with a higher SSC may have a higher retail value (Parker et al., 1991). The irrigation regimes (100%, 50% and 150% ET applied 4 weeks before harvest) did not affect O Henry peach postharvest storage potential based on internal breakdown development during 2, 4 and 6 weeks in cold storage at 0 C or 5 C. Fruit from 50% ET had a lower water loss rate than fruit from 150% ET or 100% ET. Fruit from 150% ET lost nearly 35% more water than fruit from 50% ET or 100% ET after 24 h. Light microscopy studies indicated that fruits from 50% ET and 100% ET had a continuous and much thicker cuticle and a higher density of trichomes than fruits from the 150% ET. These differences in exodermis structure may explain the higher percentage of water loss from fruit from 150% ET compared with the others (Crisosto et al., 1994). Recently, RID and partial root zone drying (PRD) were evaluated on white-fleshed peach growing under California conditions (Goldhamer et al., 2002). PRD involves inducing partial stomatal closure by exposing some part of the root zone to continual soil drying. After 2 years of evaluations, yield and fruit quality were affected equally by the PRD and the RID treatments. Except for a few studies which have comprehensively tested a broad range of water management practices and conditions and their impacts on postharvest quality, it is often difficult to generalize about the effects of water management from the

275 544 C.H. Crisosto and G. Costa site-specific irrigation regimes that have been reported (a) O Henry r 2 = Canopy Manipulation In most cultivars, fruitlet thinning increases fruit size while also reducing total yield, thus a balance between yield and fruit size must be achieved. Some cultivars must not be thinned too much because their fruit will crack easily. In some cases, fruit size, SSC and TA are modified without affecting fruit cracking. In other cultivars the fruit do not ripen properly when trees are carrying too high a fruit load. In general, the number of fruit that can ripen on a tree will depend on the cultivar and orchard conditions. Thus detailed information about cultivar response to crop load adjustment and potential benefits should be developed for each specific situation. Historically, maximum profit does not occur at maximum marketable yield since larger fruit bring a higher market price. Furthermore, new market trends for highly tasty fruit may force a review of this topic. The crop load and fruit quality relationship has been studied by researchers in various countries (Forlani et al., 2002; Giacalone et al., 2002; Luchsinger et al., 2002; Costa et al., 2003a). Leaving too many fruit on a tree reduces fruit size and SSC in the early-season May Glo nectarine and late-season O Henry peach (Fig. 20.4). Crop load on O Henry peach trees affected the incidence of internal breakdown. In general, the overall incidence of mealiness and flesh browning in fruit from the high crop load was low, intermediate in fruit from the commercial crop load, and the highest in fruit from the low crop load (Crisosto et al., 1997). Fruit quality measured at harvest and during storage for several peach and nectarine cultivars varied according to fruit canopy position in different production areas (Marini et al., 1991; Crisosto et al., 1997; Iannini et al., 2002). Large differences in SSC, acidity and fruit size were detected between fruit obtained from the outside and inside canopy positions of open-vase trained trees (Marini et al., 1991; Crisosto et al., 1997). During the last decade, we have observed that fruit grown under a high light environment (outside canopy) has SSC (%) Fruit weight (g) (b) 100 May Glo r 2 = 0.67 O Henry May Glo Crop load (1000s fruit/ha) r 2 = 0.82 r 2 = Fig Relationship between (a) crop load and soluble solids concentration (SSC) and (b) crop load and fruit weight for O Henry peach and May Glo nectarine. (Adapted from Crisosto et al., 1997.) a longer shelf-life (storage and market) than fruit grown under a low light environment (inside canopy). During our work, we found that fruit that developed in the more shaded inner canopy positions have a greater incidence of internal breakdown than fruit from the high light, outer canopy positions (Fig. 20.5/ Plate 229). Thus, fruit from the outer canopy have a longer potential market life, especially for cultivars susceptible to internal breakdown. The use of more efficient training systems which allow more sunlight penetration into the centre and lower canopy areas is recommended to reduce the number of shaded fruit, thus extending postharvest life (Crisosto et al., 1997). Summer pruning and leaf removal around the fruit increase fruit light exposure and, when performed properly, can increase fruit colour without affecting fruit size and

276 Preharvest Factors Affecting Quality 545 Fig Canopy position affects fruit size, red colour development and storage potential. SSC (Fig. 20.6/Plate 230). Excessive leaf pulling or leaf removal executed too close to harvest can reduce both fruit size and SSC in peaches and nectarines (Crisosto et al., 1997; Day, 1997). Girdling (removal of bark) 4 6 weeks before harvest is performed to increase peach and nectarine fruit size (Fig. 20.7/Plate 231) and to advance and synchronize maturity (Day, 1997). Girdling increases fruit SSC in some cases but also increases fruit acidity and phenolics, so the flavour resulting from the additional SSC may be masked. Girdling can also cause the pits of peaches and nectarines to split, especially if it is done too early during pit hardening. Fruit with split pits soften more quickly than intact fruits and are more susceptible to decay. Reports on the benefits of using different reflective materials to improve peach red colour and fruit size and speed up maturation varied according to cultivar, orchard situation and location (Layne et al., 2001; Fiori et al., 2002). Under California s long and hot growing season, canopy manipulations including water sprout removal and leaf removal around fruit become necessary to achieve the benefit of red colour development in vigorous orchards. Also, even when reflected light was reaching fruit in the canopy, but temperatures remained high during that maturation period, improvement in red colour development was not observed. In spite of the limited literature available on the role of preharvest factors in consumer quality, there is strong evidence that fruit flavour quality, market life and physiological disorders are related to preharvest factors. We therefore encourage more detailed work on

277 546 C.H. Crisosto and G. Costa Fig Leaf removal around the fruit improves red colour but may decrease fruit size. Fig Peach girdling (removal of a strip of scaffold bark) at the main scaffolds advances maturity and increases fruit size.