UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Physical and chemical properties of apple juice and apple juice particulate McKenzie, Darrell-Lee 1988

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1988_A6_7 M32_4.pdf [ 10.08MB ]
Metadata
JSON: 831-1.0097691.json
JSON-LD: 831-1.0097691-ld.json
RDF/XML (Pretty): 831-1.0097691-rdf.xml
RDF/JSON: 831-1.0097691-rdf.json
Turtle: 831-1.0097691-turtle.txt
N-Triples: 831-1.0097691-rdf-ntriples.txt
Original Record: 831-1.0097691-source.json
Full Text
831-1.0097691-fulltext.txt
Citation
831-1.0097691.ris

Full Text

PHYSICAL AND CHEMICAL PROPERTIES OF A P P L E JUICE AND A P P L E JUICE PARTICULATE.  By  DARRELL-LEE MCKENZIE B.Sc, The University of British Columbia, 1983  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in  THE FACULTY OF GRADUATE STUDIES Department of Food Science We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA February, 1988 (c)Darrell-Lee McKenzie, 1988  In  presenting  degree freely  at  this  the  available  copying  of  department publication  of  in  partial  fulfilment  University  of  British  Columbia,  for  this or  thesis  reference  thesis by  this  for  his thesis  and  scholarly  or for  her  Department  DE-6(3/81)  Columbia  I  I  further  purposes  gain  the  requirements  agree  that  agree  may  be  It  is  representatives.  financial  permission.  The University of British 1956 Main Mall Vancouver, Canada V6T 1Y3  study.  of  shall  not  that  the  Library  permission  granted  by  understood be  for  allowed  the  an  advanced  shall for  make  extensive  head  that without  it  of  copying my  my or  written  ii.  ABSTRACT  In order to prevent enzymatic oxidation of phenols during the centrifugal extraction of juice from apple puree, a sulfite or ascorbic acid treatment followed by blanching has been proposed. However, juice from blanched puree is more turbid and difficult to clarify than juice processed without blanching. In order to better understand the effect of blanching on juice turbidity as well as to provide more information concerning the effect of cultivar, post-harvest storage and enzyme treatment on juice quality, the chemical and physical properties of apple juice and apple juice particulate from fresh and stored Mcintosh, Red Delicious and Spartan apples processed with and without enzyme digestion and with and without blanching were examined. Analysis of juice included measurement of: soluble solids, total sugars (by HPLC), sucrose, glucose, fructose, sorbitol, pH, titratable acidity, total acidity (by HPLC), citric acid, galacturonic acid, malic acid, quinic acid, succinic acid, pectin (as anhydrogalacturonic acid by HPLC) and turbidity (as absorbance at 600 nm). Analysis of particulate included measurement of dry matter weight, pectin, protein and zeta potential as well as thin sectioning, negative staining and shadow casting transmission electron microscopy. Chemical analysis of apple juice showed that the levels of organic acids, sugars and soluble pectin differed between cultivars. However, no varietal differences were observed in the chemical or microscopic analysis of cloud material. Blanching of apple puree, on the other hand, increased apple juice turbidity by increasing the amount of particulate suspended in the juice. Furthermore, blanching stabilized suspended particulate by what appeared to be the formation of a protective colloid which prevented particle aggregation through electrostatic repulsion. Post-harvest apple storage also resulted in changes to juice particulate, which were observed as gel formation during juice storage at 0 C and as a web-like aspect in the microscopic appearance of juice particulate. Treatment of apple puree with Irgazyme 100 decreased juice turbidity, resulting in the concomitant decrease  iii.  in both the level of soluble juice pectin and the amount of suspended cloud material.  Use of enzyme treatment and blanching in the processing of  apple juice was demonstrated by stepwise discriminant analysis to allow production of four unique apple juice products.  TABLE OF CONTENTS ABSTRACT  i  LIST OF TABLES  v  LIST OF FIGURES  v  ACKNOWLEDGEMENTS INTRODUCTION LITERATURE REVIEW I.  Apple Fruit (a) . Anatomical and Cellular Structure of Apple Fruit (b) . Fruit Growth and Development (c) . Biochemistry of Apple Development and Storage  II.  Apple Juice (a) . Apple Juice Processing (b) . Other Juice Products (c) . Changes During Processing (d) . Apple Juice Composition (e) . Changes During Storage (f) . Emerging Techniques in Apple Juice Production  1 1 1 1 2 2 2  METHODS AND MATERIALS I.  Apple Juice Preparation  3  n.  Preliminary Turbidity Studies  3  ELL  Cloud and Colour Stability  3  IV.  Chemical Analyses (a) . Alcohol Insoluble Solids (b) . Organic Acids (c) . Pectin (d) . pH and Titratable Acidity (e) . Protein (f) . Soluble Solids ( g . Sugars Zeta Potential  3 3 3 3 3 3 3 3 3  V.  Transmission Electron Microscopy (a) . Thin Sectioning (b) . Negative Staining (c) . Shadow Casting  3 3 ....3  V.  VI.  Statistical Analyses  3 9.  RESULTS AND DISCUSSION I. n.  IH.  IV.  V.  Preliminary Turbidity Studies Cloud and Colour Stability During Juice Storage (a) . Cloud Stability (b) . Colour Stability Chemical Analyses (a) . Chemical Composition of Apple Juice (b) . Composition or Centrifuged Cloud Material from Apple Juice (c) . Composition of Apple Juice After Removal of Cloud Material i Analysis by Transmission Electron Microscopy (a) . Thin Sectioning (b) . Negative Staining (c) . Shadow Casting (d) . Conclusions-.. |  Stepwise Discriminant Analysis i  40. 4 2. 4 2. 4 9. 5 o. 5 0. 61. 6 4. 6 6. 66. 67. 8 0. 86. 8 7.  CONCLUSIONS  9 2.  REFERENCES  9 4.  APPENDIX A  105.  vi.  LIST OF T A B L E S Table I. II. UJ.  IV.  V.  VI. VII.  A-L  Page Composition of Several Varieties of Canadian Apple and Apple Juice  21.  Composition of Centrifuged Material from Apple Juice  24.  Composition of Apple Juice as Affected by Cultivar, Post-Harvest Storage, Blanching and Treatment with Irgazyme 100  5 1.  Composition and Yield of Cloud Material from Apple Juice as Affected by Cultivar, Post-Harvest Storage, Blanching and Treatment with Irgazyme 100  6 2.  Comparison of Juice Composition Before Removal of Cloud Material with Serum Composition After Removal of Cloud Material .'  6 5.  Characterization of Particles in Thin Sections of Cloud Material  6 8.  The Effect of Cultivar, Post-Harvest Storage, Blanching and Treatment with Irgazyme 100 on Particle Dimensions of Negatively Stained and Shadow Cast Preparations of Apple Juice  7 2.  ANOVA of Apple Juice Soluble Solids Data  108.  vii.  LIST OF FIGURES Figure  Page  1. Diagram of a mature apple fruit  4.  2. Diagram of a mature parenchymatic plant cell  4.  3. Schematic representation of the progress of respiration or an apple fruit  11.  4. The effect of blanching on cloud formation in the juice from apples incubated with 1 % Irgazyme 100 for 45 minutes at 35 C  41.  5. The effect of storage at 0 C on the stability of cloud in juice from fresh apples not treated with enzyme  4 3.  6. The effect of storage at 0 C on the stability of cloud in the juice from stored apples not treated with enzyme  4 4.  7. The effect of storage at 24 C on the stability of cloud in the juice from fresh apples not treated with enzyme  4 6.  8. The effect of storage at 24 C on the stability of cloud in the juice from stored apples not treated with enzyme  4 7.  9. Visible region spectra of juice from blanched Red Delicious apples treated with Irgazyme 100 and stored at 24 C  4 8.  10. Thin section of pellets of cloud material from oxidized juice of fresh Spartan apples. Aggregate (a); granule (g); sphere (s); vesicle (v)  69.  11. Thin section of pellets of cloud material from natural juice or fresh Spartan apples. Aggregate (a); granule (g); sphere (s)  69.  12. Negatively stained particles from oxidized juice of fresh Spartan apples not treated with enzyme. Aggregate (a); sphere (s)  74.  13. Negatively stained particles from natural juice of fresh Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  74.  Figure  Pag  14. Negatively stained particles from oxidized juice of stored Mcintosh apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  74  15. Negatively stained particles from natural juice of stored Red Delicious apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  74  16. Negatively stained particles from oxidized juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); sphere (s) ,  77  17. Negatively stained particles from natural juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s)  77  18. Negatively stained particles from oxidized juice of stored Spartan apples treated with enzyme. Aggregate (a); sphere (s)  71  19. Negatively stained particles from natural juice of stored Mcintosh apples treated with enzyme. Aggregate (a), granule (g); sphere (s)  77  20. Shadow cast particles from oxidized juice of fresh Spartan apples not treated with enzyme. Aggregate (a); sphere (s)  81  21. Shadow cast particles from natural juice of fresh Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  81  22. Shadow cast particles from oxidized juice of stored Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  81  23. Shadow cast particles from natural juice of stored Red Delicious apples not treated with enzyme. Aggregate (a); granule (g); sphere (s)  81  24. Shadow cast particles from oxidized juice of fresh Spartan apples treated with enzyme. Aggregate (a); granule (g); sphere (s)  83  25. Shadow cast particles from natural juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s)  83  ix.  Figure  Page  26. Shadow cast particles from oxidized juice of stored Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s)  83.  27. Shadow cast particlesfromnatural juice of stored Mcintosh apples treated with enzyme. Aggregate (a); granule (g); sphere (s)  8 3.  28.  Canonical plot of 24 apple juice samples categorized according to processing treatment.  29. Canonical plot of the same apple juice samples after elimination of the data from fresh Mcintosh apples. Juices were categorized according to processing treatment A-l. A-2.  HPLC determination of sugars: (1) sucrose, (2) glucose, (3) fructose, (4) sorbitol HPLC deteimination of organic acids: (1) citric, (2) galacturonic, (3) malic, (4) quinic, (5) succinic  89.  9 o. 1 • 06  107.  X.  ACKNOWLEDGEMENTS  The author wishes to express her gratitude to Dr. T. Beveridgefor his a tance, support and encouragement throughout this study. The advice o Professors S. Nakai, W.D. Powrie, BJ. Skura and J. Vanderstoep is also si cerely appreciated. A further word of thanks goes to Okanagan College Kelowna for the loan of their publishing equipment. Finally, the author gra fully acknowledges the generous donation offacilities by Agriculture Cana and wishes to thank the staff of the Agriculture Canada Research Station Summerland and Vanouver, British Columbia for their technical advice an guidance throughout this investigation.  1. INTRODUCTION  The Canadian apple juice manufacturer relies heavily on the the availability of surplus and cull grade fruit for production of juice (Moyer and Aitken, 1980). Since juice quality is a reflection of apple quality, the manufacturer must consider the effect of fruit variety and maturity, as well as seasonal and locational differences in climate and weather on the quality of the final juice product. In Canada however, very little information is available concerning the effect of these factors on juice composition. Juice quality is further influenced by manufacturing practices. Production of apple juice in Canada has centered mainly around the clarified, amber type of juice (Atkinson and Strachan, 1949a; Beveridge et al., 1986a). In Japan, however, the majority of fruit juices are sold in the cloudy, unoxidized, 'natural' state (Yamasita, Personal Communication, 1987). A 'natural' apple juice can be produced through inactivation of polyphenol oxidase by blanching of apple puree at 90 C in combination with an ascorbic acid or sulfite pretreatment (Holgate et al., 1948; Atkinson and Strachan, 1949b; Beveridge et al., 1986a). Blanching successfully inhibits the development of the brown colour and cider-like flavour characteristic of oxidized juice, producing a naturally coloured, opalescent juice with a fresh apple flavour characteristic of the variety  processed  (Bauernfeind, 1958). The opalescence or cloud formed as a result of blanching is very stable (Montogomery and Petropakis, 1980). On the other hand, the cloud formed in oxidized juice is very unstable, readily flocculating to form an undesirable thick layer of sediment at the bottom of the container (Atkinson and Strachan, 1949b). The thermal stabilization of apple juice cloud by blanching offers the possibility of marketing a 'natural', unoxidized juice in either the opalescent form, or upon cloud destabilization, in the clarified form allowing for further expansion of the apple juice market (Carpenter and Walsh, 1932; Atkinson and Strachan, 1949b). An understanding of the factors contributing  2.  to stabilization or destabilization of the cloud formed in apple juice is required to enable the manufacturer to efficiently produce either a cloudy or clarified unoxidized 'natural' apple juice. In addition, more information concerning the effect of cultivar and fruit maturity on juice quality is required to assist manufacturers in choosing appropriate varieties for blending and/or appropriate techniques for processing based on the availability of raw materials. The objectives of the current study were to establish whether or not blanching increased juice turbidity and stabilized the suspension of cloud material in unclarified apple juice. Furthermore, the nature of the differences existing between juice from blanched and unblanched pur6e, and the effect of cultivar, post-harvest storage, and enzyme treatment on the chemical and physical properties of apple juice and apple juice particulate were also examined. Finally, stepwise discriminant analysis was used to objectively categorize  apple juice  into groups based on processing treatment,  demonstrating the potential for the production of four distinct apple juice products.  3. LITERATURE REVIEW  I. APPLE FRUIT Apples, classified as pome fruits belonging to the family Rosaceae, are the most important and widely grown fruit in Canada (Hulme and Rhodes, 1971; Fruit and Vegetable Production, 1985; Handbook of Selected Agricultural Statistics, 1986) . While the apple is believed to have existed from prehistoric times, apple production in Canada began only in the seventeenth century when the first known orchard of apple trees was planted in Quebec by Louis Hebert, a pharmacist who sailed from France to Canada with Champlain (Smock and Neubert, 1950; Proctor, 1979). From these early beginnings, apple production in Canada has steadily increased until in the years 1978 - 1985 production averaged 515,000 tons (Handbook of Selected Agricultural Statistics, 1986). Apple production in British Columbia alone accounted for approximately 36% of the total crop in the years 1983 - 1985, with the major cultivars being Red Delicious, Mcintosh and Spartan (Fruit and Vegetable Production, 1985). Increased production together with improved processing and storage techniques has made the apple available for consumption in either the fresh or processed form all year round (Ryall and Pentzer, 1982).  (a). ANATOMICAL AND CELLULAR STRUCTURE OF APPLE FRUIT The apple fruit consists of a variety of cellular and chemical constituents. As described by Smock and Neubert (1950), the outer skin of the apple is covered on the surface by a waxy coating referred to as the cuticle (Fig. 1). As the fruit matures, the cuticle continues to increase in thickness eventually separating and surrounding the outer layer of cells known as the epidermis. The epidermal cells, which lengthen tangentially as the fruit enlarges, are thin-walled and flattened in appearance with the cell contents being completely disintegrated by the time the fruit is mature. Below the epidermis and derived from the floral tissue is the parenchyma, comprising  4.  Figure  1.  Diagram of the mature apple fruit (Smock  and N e u b e r t ,  1950).  Nucleus Cell wall Cytoplas m Middle  lamella  Mitochondria Chloroplast Vacuole Intercellular s p a c e  F i g u r e 2.  Diagram of a mature p a r e n c h y m a t i c plant cell ( S c h w i m m e r , 1981).  5. the major portion of the edible part of the mature fruit and consisting of variably sized cells separated by a high proportion of intercellular spaces (25%). The core region of the apple is composed of parenchyma derived from the carpel tissue and forms the ovary walls and the core lining of the seed cavaties. The core is separated from the edible portion of the fruit by the vascular bundles. The edible portion of the apple fruit consists mainly of parenchyma cells. Closer examination of the parenchyma cell provides a general understanding of the chemical constituents of the apple and their location within the individual cell. The major physical structures of the parenchyma cell are the cell wall and the vacuole (Fig. 2).  The cell wall appears to consist of  cellulose microfibrils attached to two interacting but semi-independent networks, a glycoprotein-phenolic network and a polysaccharide network. The polysaccharide network involves hemicellulose and the xylan and glucan polymers of branched polygalacturonate (pectin) both hydrogen bonded to cellulose microfibrils (Knee et al., 1975; Darvill et al., 1980; Wilson and Fry, 1983). The cell wall is further associated with unbranched polygalacturonate, the major component of the middle lamella which forms an amorphous matrix acting to cement adjacent cell walls together (Knee, 1978a; Jewell, 1979; McFeeters, 1985). The glycoprotein network consists mainly of extensin, a glycoprotein containing a polypeptide backbone with side chains of galactose and arabinose. The glycoprotein appears to function mainly in a structural capacity where accumulation of extensin is necessary to render walls inextensible and thereby halt cell elongation  (Wilson and  Fry, 1983). There are probably few strong bonds existing between the components of the cell wall and the associated region of the middle lamella as linkages are continually broken and reformed to allow interpolation of new material during cell enlargement. During ripening and senescence bonds are broken irreversibly, weakening the wall structure (Knee, 1973b; Preston, 1979). Knee (1973a) and Darvill et al. (1980) suggest that components of the cell wall and middle lamella are held together by a combination of covalent bonds and regions of extensive hydrogen bonding, ultimately form-  6. ing a semi-elastic structure which serves as the skeletal framework of the tissues (Schwirnmer, 1981). The plasma membrane, found immediately inside the cell wall, serves to separate the protoplast of the cell from its environment by specializing in specific transport of substances in and out of the cell. In addition, the plasma membrane plays a role in both the synthesis and assemblage of the cell wall and the reception and translation of external signals. The plasma membrane consists primarily of lipids (40-50%) and proteins (35-40%).  Metabolic energy is  coupled to ion transport across the membrane by membrane bound enzymes such as K -stimulated adenosine triphosphatase (ATPase). The plasma membrane +  also contains specific binding sites for transport of the plant hormone auxin, and photoreceptors for translation of environmental signals into physiological and developmental responses (Leonard and Hodges, 1980). Contained within the plasma membrane but exterior to all other cell membranes is the cytoplasm, which is composed mainly of water and enzymes. The cytoplasm is the site of amino acid and protein synthesis, gluconeogenesis, glycolysis and the conversion of sucrose to glucose and fructose. Metabolite levels within the cytoplasm are maintained by adjusting the levels of metabolite in vacuoles (Kelly and Latzko, 1980). Vacuoles contain a variety of acid hydrolases such as proteinase which are probably involved in the turnover of cytoplasmic components (Marty et al., 1980). Yamasaki (1984) conducted studies on the vacuoles of immature apple fruit and found that almost all fructose, glucose, malic acid and phenolic components present in the cell were compartmentalized within the vacuoles. Other components such as sorbitol were located in both the free space and the vacuoles, while sucrose was found only in the free space and the cytoplasm. Vacuoles function mainly as storage sites, allowing cell expansion without the synthesis of large quantities of cytoplasm (Marty etal., 1980). Suspended within the cytoplasm of the cell are the organelles: the nucleus, golgi bodies, endoplasmic reticulum, ribosomes and most impor-  7. tantly  to the developing fruit,  the mitochondria  (Jewell,  1979).  Mitochondria are the site of the tricarboxylic acid (TCA) cycle which produces adenosine triphosphate (ATP), the principle energy source used by the cell for synthesis and transport of metabolites, as well as for building and maintenance of the protoplasm, cell membranes and cell walls (Hanson and Day, 1980; Ryall and Pentzer, 1982). Overall, coordination of the individual components of cellular metabolism is a complex process involving a series of intricate biochemical reactions affected by the action of hormones, environmental and cultural factors (Crane, 1969). Environmental factors include light, water and temperature, while cultural factors involve soil fertilization, pruning, thinning, spraying and rootstocks (Smock and Neubert, 1950). Successful management of these components is required for the development of good quality apple fruit and high quality processed products such as apple juice and apple sauce.  (b). FRUIT GROWTH AND DEVELOPMENT Fruit growth and development is reviewed by Smock and Neubert (1950), Ryall and Pentzer (1982) and Lau (1985). Gortner et al. (1967) also provide excellent biochemical definitions of the terminology used in the discussion of fruit development. The life of a fruit, which by definition is a mature or ripened ovary or ovaries, begins with initiation of the fruit bud in the year before harvest and continues in the subsequent season of growth through the stages of cell division, cell enlargement, maturation, ripening and senescence (Hulme and Rhodes, 1971). Fruit physiology begins at the time of flowering with pollination and fertilization of the ovules of the blossom. After petal drop, the apple fruit undergoes a period of rapid cell division lasting 3 - 5 weeks in which most of the cells of the apple are produced. The number of cells produced depends on ambient temperatures, cultivar, tree nutrition and vigor, endogenous cytokinin concentration and number of fruitlets on the tree. The majority of growth is achieved during  8. the vegetative phase due in part to cell enlargement and in part to an increase in the size of intercellular spaces. As a result, the fruit continues to increase in size until harvest. Maturation, which must take place while the fruit is still attached to the tree, is the stage during which the fruit attains a fullness of growth. The terminal period of maturation referred to as ripening, may take place either before or after harvest and invloves chemical changes which cause the fruit to develop the flavor, texture and aroma typical of an apple at optimum eating quality.  Once initiated, the ripening  process is irreversible and is rapidly followed by senescence where fruit growth is replaced by the biochemical process of aging.  (c). BIOCHEMISTRY OF APPLE DEVELOPMENT AND STORAGE As described above, the apple fruit is basically composed of the surrounding tissue and the five mature ovaries of the apple flower, consequently its development is linked to that of the ovules and is therefore subject to hormonal control (Smock and Neubert, 1950; Nitsch, 1970). Crane (1969), using hormones and radioactive tracers of the movement and distribution of organic materials in plants, concluded that the high levels of hormones in developing seeds served as mobilization centers to attract metabolites to the fruit. This allows the fruit to successfully compete with other growing organs of the plant. Hormones are produced in fluctuating levels throughout the development of the apple fruit and include auxins, gibberellins and cytokinins (Nitsch, 1970; Ryall and Pentzer, 1982). Auxins and gibberellins are involved in fruit setting, auxins and cytokinins are responsible for cell division. Auxins as well as ethylene are involved in increased vascular development and cell expansion and abscisic acid is responsible for the cessation of the period of cell elongation. Auxins also stimulate production of ethylene which is essential for initiation of the ripening process. These hormones have been shown to control apple growth through the mediation of hormone-directed transport systems which mobilize metabolites and other materials into the developing fruits (Crane, 1969; Nitsch, 1970).  9. The leaf is the main source of nutrition for the developing fruit, converting water, carbon dioxide and light energy to glucose and oxygen (Whiting, 1970). Complex biochemical processes, including glycolysis and the TCA cycle, then allow the fruit to produce the energy it requires for metabolic processes such as the building and maintenance of protoplasm, cell membranes and cell walls through the oxidation of organic molecules like glucose and malic acid (Lehninger, 1973; Ryall and Pentzer, 1982). Studies by Webb and Burley (1962) and Hansen (1967; 1970) using C02 applied to the leaves of apple trees showed 14  that sorbitol and sucrose are the two main forms in which carbohydrate is transported in the phloem of apple trees. Sorbitol, the major translocate, is imported to the fruit in response; to hormonal action which causes the fruit to act strongly as a "sink", removing up to 90% of the photosynthates from the leaves on the fruit spur (Hansen, 1967). Once in the fruit, sorbitol is rapidly metabolized into other compounds during the main part of the season with only a slight respiratory loss (Hansen, 1967; 1970; 1979). Sorbitol iconversion within the fruit varies both during and between growing seasons and also between apple cultivars (Hansen, 1970; Chan et al., 1972). Beruter and Kalberer (1983) used disks of apple tissue incubated in C-sorbitol to demostrate that the rate of sorbitol uptake is controlled main14  ly by the rate of sorbitol conversion to fructose. Sorbitol is first transported to the free space of the cell wall, diffusing freely into the metabolic space of the cell where conversion to sucrose, fructose and glucose occurs. Any unmetabolized sorbitol is subsequently transported into a vacuole by facilitated diffusion, a process which requires the presence of both a concentration gradient and a specific carrier protein at the surface of the vacuolar membrane. Highest concentrations of sorbitol coincide with the period of cell division during which translocated sorbitol is rapidly converted to sucrose, the preferred source of carbon for meeting the metabolic needs of the young fruit (Chan et al., 1972; Beruter, 1985). Respiration per unit weight of fruit is also highest during the period of cell division, but slowly declines throughout the stage of cell enlargement,  10. remaining relatively steady as maturity is approached until it rises again as the respiration climacteric develops (Fig. 3; Hulme and Rhodes, 1971). After the period of rapid cell division, entry into the cell expansion stage is marked by the beginning of sugar accumulation, mainly in the form of fructose (Hansen, 1979; Beruter, 1985). Also at this time, an increase in glucose is found to precede starch accumulation (Hansen, 1979). Sugars stored in the vacuolar sugar pool during the period of cell expansion are available to passively diffuse through the vacuolar membrane and supply the metabolic cell compartment with carbohydrate when this compartment is depleted of sugars (Beruter and Kalberer, 1983). Synthesis of organic acids followed by the synthesis of permanently insoluble materials mainly associated with the cell wall also occurs during the beginning of the cell expansion stage (Hansen, 1979). Malic acid, the main substrate for respiration, is available either for energy (ATP) synthesis via the TCA cycle in the mitochondria or for synthesis of the reducing compound nicotinamide adenine dinucleotide phosphate (NADPH2), a requisite of many synthetic processes, via decarboxylation by malic enzyme in the cytoplasm (Hulme and Rhodes, 1971; Conn and Stumpf, 1976; Hanson and Day, 1980). The initial accumulation of organic acids during the early stages of the cell expansion period is followed by a decrease in organic acids, measured as a decrease in titratable acidity, resulting from a breakdown of previously synthesized acids in combination with a decline in new acid synthesis (Knee, 1978b; Hansen, 1979). Phenolic compounds, which contribute to the astringency, skin colour and browning of apple tissue are also present in high concentrations during the early stages of cell expansion and may be involved in cell wall synthesis (Hulme, 1958; Hulme and Rhodes, 1971; Wilson and Fry, 1983). Throughout the remaining period of growth, phenolics, total and soluble pectin and nitrogen levels stay fairly constant, while sugars, especially fructose, and anthocyanins increase and starch, ascorbic acid and organic acids decrease in content (Hulme, 1958; Hulme and Rhodes, 1971; Hansen, 1979).  11.  Figure 3.  Schematic representation of the progress of respiration of an apple (Fidler et al., 1973).  12. The final stage of fruit growth is the ripening process which is accompanied by a large increase in ethylene production (Lau, 1986) stimulated by auxins (Crane, 1969; Nitsch, 1970; Leiberman, 1979). Ethylene in turn triggers the characteristic rise in respiration (Fig. 3) referred to as the climacteric (Rhodes, 1970; Hulme and Rhodes, 1971; Ryall and Pentzer, 1982). Rhodes (1970) defined the climacteric as that "period in the ontogeny of certain fruits [including apples], during which a series of biochemical changes is initiated by the autocatalytic production of ethylene, marking the change from growth to senescence and involving an increase in respiration and leading to ripening." Fruit ripening and the extent of the climacteric can be controlled through modification of ethylene production by a variety of chemical and environmental treatments. Ripening is delayed through inhibition of ethylene action as a result of increased presence of carbon dioxide, low levels of oxygen and temperatures below 5 C , as well as by growth retardants such as N-dimethylaminosuccinamic acid (Alar, B9). On the other hand, growth regulators such as naphthalene  acetic acid (NAA) and 2:4-  dichlorophenoxyacetic acid (2:4D) can be used to stimulate ripening by stimulating ethylene production (McGlasson, 1970; Hulmes and Rhodes, 1971; Leiberman, 1979; Ryall and Pentzer, 1982; Lidster et al., 1983). Metabolically, the climacteric is characterized by a rise in respiration due in part to an increase in mitochondrial number and in part to an increase in the activity of the TCA cycle enzymes within the mitochondria (Hulme and Rhodes, 1971). Further changes occurring during the climacteric include changes of colour due to the loss of chlorophyll, alterations in flavour resulting from changes in acidity, astringency and sweetness, changes in texture, changes in cell permeability and increases in ribonucleic acid (RNA) and protein synthesis (Rhodes, 1971). Frenkel et al. (1968) using C-labelled phenylalanine ad14  ministered to intact fruit demonstrated that increased levels of protein parallelled the increased synthesis of malic enzyme, an enzyme which catalyzes the reversible formation of pyruvic acid and N A D P H 2 from malic acid (Conn and Stumpf, 1976). This suggests that new enzyme synthesis early in the climacteric contributes greatly to the rise of metabolic activity. The decarboxylation of  13. malic acid gives rise to acetaldehyde and ethanol and also contributes to the climacteric rise in CO2, although the significance of this in quantitative terms is still not clear (Rhodes, 1970; Hulme and Rhodes, 1971). After the climacteric, the cells of the apple follow the characteristic pattern of senescence. The onset of senescence is denoted by a decrease in ribosome numbers, the start of chloroplast degradation, degeneration of the mitochondria and tonoplast, along with the swelling, vesiculation and ultimate disappearance of the endoplasmic reticulum. Eventually degradation of the plasmalemma leads to cell death and the breakdown of the nucleus (Jewell, 1979). Storage of the apple fruit is aimed at delaying the onset and controlling the pace ofripeningand senscence for as long as necessary to satisfy market requirements.  After harvest the fruit continues to respire, only interchanging  water, oxygen and carbon dioxide with the environment (Fidler et al., 1973). The gradual change in environmental conditions resulting from fruit respiration in a sealed chamber causes the qualitative nature of respiration to progressively change during storage (Hansen, 1966). Respiration can be modified by controlling the action of ethylene, the production of which is inhibited during storage through use of refrigeration temperatures (< 5 C) alone or in combination with low levels of oxygen (1-3%) and carbon dioxide (1.5-5%) (Ryall and Pentzer, 1982; Lau, 1983; Lidster et al., 1983). Both refrigerator storage, which consists of temperatures of 0-4 C and normal atmosphere (20.9% O2, 78.1% N, 0.9% Ar, 0.03% CO2) and controlled atmosphere (CA) storage, which utilizes low temperatures (0-4 C) in combination with carefully modified levels of O2 (1-3%) and CO2 (1.5-5%) (Lau, 1983; Lidster et al., 1983), have enabled producers to provide apples for consumption in every month of the year. CA storage has generally resulted in better retention of quality and less damage resulting from product breakdown and water loss, while allowing for longer storage than refrigeration alone (Fidler et al., 1973; Ryall and Pentzer, 1982). The main problem with eitherform of storage is the possibility of physiological disorders which may result from abnormal functioning of the respiratory mechanism due to adverse internal or external conditions (Hansen, 1966).  14. The chemical changes occurring during storage are mainly degradative in nature. A continued decrease in organic acid content is observed, as well as a decrease in insoluble pectin with a concomitant decrease in textural firmness and increase in soluble pectin (Hulme, 1958; Hansen, 1979; Van Woensel and De Baerdemaeker, 1983; Rouchaud et al., 1985). In addition, studies by Hansen (1979) and Rouchaud et al. (1985) showed the breakdown of starch and conversion of sorbitol in the early stages of storage increased concentrations of glucose, fructose and sucrose. Although far from complete, an understanding of the complex biochemical changes occurring during the development and storage of the apple has enabled producers to provide a high quality product throughout the year. However, further knowledge is required especially with respect to the biochemistry of the climacteric and fruitripeningif the quality of stored fruit and processed apple products is to be improved. Further work aimed at the development of varieties with specific characteristics designed for the processed market is also needed to improve quality of processed products such as apple juice and sauce.  II. APPLE JUICE. Commercial processing of apples began in North America as an outgrowth of orcharding, utilizing the supply of low cost cull and surplus apples to extend the season of fruit consumption as well as to equalize supplies from year to year (Smock and Neubert, 1950; Pollard and Timberlake, 1971; Mover and Aitken, 1980). Today, apples are processed in many ways: canned, frozen, dried, glaced and spiced, as well as being made into sauce, butter, jam, jelly, cider, vinegar and confections (Smock and Neubert, 1950; Woodruf, 1975). Processing continues to utilize an increasing percentage of the total apple crop and juice production is now a firmly established industry in its ownright(Duymovic, 1975). Canadian consumption of apple juice is steadily increasing and apple juice predominates single strength juice markets (Pollard and Timberlake, 1971; Cumming, 1983).  15. (a). APPLE JUICE PROCESSING Apple juice is obtained by expressing the liquid from the edible portion of the apple fruit. The quality of the juice is a reflection of the apple quality used in juice production and is affected by the genetic make-up and maturity of the fruit, cultural and nutritional conditions, climate and weather (Pollard and Timberlake, 1971). Apple maturity is probably the single most important factor affecting juice quality.  Immature apples produce a 'starchy' tasting juice  lacking sweetness and body, while the decrease in organic acids and increase in soluble pectin during storage cause overmature apples to produce low yields of juice high in suspended solids and sweetness, but lacking flavour (Tressler and Pederson, 1938; Pollard and Timberlake, 1971; Mover and Aitken, 1980). Blending has been used to overcome varietal and maturity differences, providing juice with an optimum combination of acid, sugar, aroma and astringency (Smock and Neubert, 1950). In British Columbia, only Mcintosh apples at optimum eating maturity can be used to produce a high quality varietal juice. Other varieties, such as Red Delicious which is grown in largest volume, are not suitable for juicing alone (Atkinson and Strachan, 1949b). Regional preferences for blends also exist, so sugar-acid ratios and astringent-aromatic levels must be balanced accordingly (Smock and Neubert, 1950; Moyer and Aitken, 1980). After good quality apples are selected, they are prepared for processing by passing over a leaf eliminator, acid washed to remove spray residues and microorganisms, then fresh water washed and trimmed to remove defects (Atkinson and Strachan, 1949b; Moyer and Aitken, 1980). Apples are then ground in a hammer mill to mechanically disrupt the cells (Smock and Neubert, 1950). Grinding provides better flow and press effect than cutting, but pulp must have proper consistency since decreased yield will result from insufficient grinding .while overgrinding increases juice solids and makes pressing difficult (Hurler and Wey, 1954; Moyer and Aitken, 1980). Extraction of juice from pulp is achieved through the combined action of gravity and applied pressure which cause juice to be transported from the vacuoles, through membranes and outside the cell wall (Glunk, 1981). Extraction is the most laborious step in production of apple juice, with juice yield  16. dependent on: variety, condition and maturity of fruit, degree of grinding, number of open channels for juice drainage and method of extraction (Moyer and Aitken, 1980; Binnig and Possman, 1984). Extraction systems can be categorized as batch type, automated and continuous, with the rack and cloth press providing best juice yields and quality. Unfortunately, this press is a labour intensive, batch type system and is ineffective for processing large volumes of fruit The screw press is the most commonly used extraction system in North America (Cumming, 1984; 1985). However, despite the advantages of high volume, low labour and continuous operation, press aids are required to achieve adequate yields. Also, extensive oxidation occurs during processing and juice is high in suspended solids (Pollard and Timberlake, 1971; Glunk, 1981). Currently, yields of 70-83% are achieved with the use of enzymes, press aids and second pressing techniques. However, storage of apples reduces yield to 65% due to increases in soluble pectin as the pulp softens (Smock and Neubert, 1950; Glunk, 1981). This ultimately leads to the collapse of cell structure and an increase in viscosity, making the pulp extremely difficult to press-out (Pollard and Timberlake, 1971). Extraction techniques which provide continuous processing with production of high quality juice from a variety of fruits at different stages of maturity are still being sought, although advances are being made as will be discussed later (Section n (f)). After extraction, juice contains a suspension of particles and by-products of processing such as pectin, which are assumed to be stabilized by particle hydration and electrical charges. If the juice is to remain turbid, clarification by endogenous pectolytic enzymes must be prevented either by heat treatment to inactivate enzymes or by low temperature storage (Pollard and Timberlake, 1971). If juice is to be clarified, the colloidal system must be disrupted prior to filtration either by use of enzymes to hydrolyze pectin, by treatment with gelatin or bentonite to effect colloidal precipitation through binding of oppositely charged colloids, or by heat to coagulate suspended particles (Smock and Neubert, 1950; Moyer and Aitken, 1980). After clarification, filtration is used for the complete removal of suspended material to produce crystal clear apple juice. This is accomplished through use of filter presses, leaf filters or drum precoat filters in  17. combination with filter aids such as diatomaceous earth (Smith et al., 1984). Centrifugation at 2700 x g with a decanter centrifuge may also be used to achieve similar results (Beveridge etal., 1987a). After clarification, all apple juice made in Canada is then fortified with at least 35 mg ascorbic acid /100 mL juice followed by preservation with either heat, freezing or chemicals such as sulfite and sorbate (Smock and Neubert, 1950; Pollard and Timberlake, 1971).  (b). OTHER APPLE JUICE PRODUCTS In North America, apple juice is produced mainly as single strength juice of the clarified, amber type (Cumming, 1985; Beveridge et al., 1986a). However, with the increasing use and popularity of apple juice, new products such as crushed, carbonated, concentrated and natural, opalescent apple juices have been developed to capture a part of this expanding market (McClellan et al., 1984). Concentration of apple juice to be used as reconstituted juice or as a base for the manufacture of marmalades, fruit butters, jams and jellies provides considerable savings in shipping, container and storage costs (Atkinson and Strachan, 1949b; Smock and Neubert, 1950). Concentration above 65 °Brix is required for the product to befreefrom microbial spoilage and is usually achieved by vacuum distillation, with aroma removed before or during evaporation and returned unchanged upon dilution. Juice used for concentration is treated with pectin decomposing enzymes to avoid pectin gellation, followed by filtration to avoid development of haze during concentration. Concentrates also must be rapidly cooled and stored at low temperatures under anaerobic conditions in order to minimize nonenzymatic browning and to prevent oxidative changes and mold growth (Smock and Neubert, 1950; Pollard and Timberlake, 1971). Juice from concentrate, as well as crystal clear single strength apple juice, can be used further to produce a carbonated apple-based fruit drink similar in taste to ginger ale (Carpenter and Walsh, 1932). The carbonation procedure is simple, although a specific optimum level of carbonation must be determined  18. for each level of soluble solids in order not to mask the delicate apple flavour with CO2 (McClellan et al., 1984). The juices discussed up to this point, i.e. single strength, concentrated and carbonated apple juice, are all of the clarified, amber type (Atkinson and Strachan, 1949a; Beveridge et al., 1986a). An alternative product in the form of an unoxidized 'natural' juice can be produced through the treatment of puree with ascorbic acid or sulfite followed by thermal inactivation of polyphenol oxidase (Holgate, 1948; Atkinson and Strachan, 1949b; Beveridge etal., 1986a). Taking this process one step further to include 3 -10 % suspended solids results in the formation of crushed apple juice, also referred to as 'liquid apple' (Atkinson and Strachan, 1949b). Crushed juice can be produced using enzymatic maceration followed by blanching to get dissolution of cell structure (Schmitt, 1983). A stable suspension and juice with the body, viscosity and texture of fresh, natural apples results (Asti, 1970). Unfortunately, the suspended solids adhere to glass and lead to a 'gritty' mouth feel (Atkinson and Strachan, 1949b). Of the five types of juice mentioned, production of an unoxidized 'natural' juice requires the least equipment and is easiest to make, eliminating the need for fining agents and enzymes. In addition, it omits all filtering problems while at the same time producing a full-flavoured apple juice without high levels of suspended solids (Atkinson and Strachan, 1949b).  (c). CHANGES DURING PROCESSING The object of juice processing is to disrupt cell structure sufficiently to allow for extraction of the liquid contents from the cell interior. However, disorganization of the ceU structure during milling and extraction also brings together components which are normally segregated. As a result, enzymes normally associated with the cell walls and membranes interact with oxygen from the air and intercellular spaces to oxidize polyphenols, ascorbic acid and other components of the juice. Such interactions ultimately affect juice quality and flavour (Pollard and Timberlake, 1971). The most obvious change to occur during processing is the browning of apple tissues and juice. Browning occurs  19. mainly through the action of polyphenol oxidase which catalyzes the oxidation of ortho-diphenols to ortho-quinones, with subsequent conversion to brown melanin pigments carried out by nonenzymatic reactions (Richardson, 1976). Although Mihalyi et al. (1978) report enzyme activities to be negligible at the pH values of apple juice, unpublished data by Beveridge and Harrison indicate the presence of sufficient polyphenol oxidase activity in pear juice to cause extensive browning. In apples, browning is affected by varietal, seasonal and locational factors, with activity depending mainly on polyphenol oxidase and not substrate concentration (Vamos-Vigyazo et al., 1976; 1977; Nadudvari-Markus and Vamos-Vigyazo, 1984; Cummingetal., 1986). Browning is further affected by application of growth regulators such as gibberellic acid and ethephon, which appear to decrease browning by diluting cell constituents through increased cell expansion (Paulson et al., 1979). Browning mediated by polyphenol oxidase can be prevented through heat inactivation of the enzyme, use of sulfites to form addition products with quinone, removal of oxygen by vacuumization, chelation of copper or addition of enzymes to alter ortho-diphenols (Embs and Markakis, 1965; Pollard and Timberlake, 1971; Richardson, 1976). Degradation of cell structure during processing is further enhanced by use of either heat or enzymes. Enzyme treatment produces greater yields of juice by degrading high molecular weight compounds solubilized during milling, thus decreasing viscosity and facilitating juice flow out of the puree (de Vos and Pilnik, 1973; Pilnik, 1982; Junker, 1987). Juice yield is increased further through the enzyme- mediated release of the water which hydrates molecules of dissolved pectin (Junker, 1987). Pectinase treated juice is cloudy due to the suspension of intact cells and other fragments of apple tissue (Voragen et al., 1979; Pilnik, 1982) . However, clarification is possible by addition of hemicellulase, oligomerase and glucosidase enzymes individually or in combination (Dorreich, 1983) . Heating has an effect on apple tissue similar to that of enzyme treatment. Studies by Reeve and Leinbach (1953), Reeve (1954) and Sterling (1953) showed heating causes cell separation before disruption of cell walls, which suggests that dissolution of pectic substances in the middle lamella between adjoin-  20. ing cells is the major cause of tissue softening and sloughing during cooking. More recently, Loh et al. (1982) demonstrated the texture of plant tissue to depend on the combined mechanical strength of cell walls and the adhesive forces cementing them together. Thermal degradation of apple tissue results from the destruction of hydrogen bonds between pectins and other cell wall polysaccharides. This ultimately causes a breakdown in the organization of protoplasmic structure, rupturing of plasma membranes, loss of turgor pressure and the degradation and separation of cell walls (Jewell, 1979). As with pectinase treated puree, heat treatment produces a juice that is cloudier than the juice from unheated puree but which is readily clarified by subsequent enzyme treatment (Montogomery and Petropakis, 1980; Beveridge and Harrison, 1986). Heating also stimulates the nonenzymatic breakdown of ascorbic acid and other compounds such as sugars to form hydrogen peroxide and hydrox^emylfurfural (HMF) (Pollard and Timberlake, 1971).  (d). APPLE JUICE COMPOSITION Apple juice basically consists of the solublized constituents of the original fruit with chemical composition affected by cultivar, growing region, climate, maturity, cultural practices, post-harvest storage, processing and juice storage (Mattick and Moyer, 1983; Lee and Wrolstad, 1988). As reviewed by Lee and Wrolstad (1988), many compositional studies have been performed on commercial apple juice primarily with the aim of establishing standards for the detection of adulterated juices. However, as most commercial juices are made from a blend of three or more cultivars, little information is available concerning the influence of individual cultivars on the final juice composition (Moyer and Aitken, 1980). At present, cultivar effects can only be inferred from the compositional data reported for the apple fruit (Ryan, 1972). Despite the lack of specific information regarding the constituents of varietal apple juice, apple juice in general has approximately the following composition: 85% water, 10-12% carbohydrate, 1% pectin, 0.5-1% organic acids, 0.5% potassium, phenols, amino acids and a small amount of flavouring compounds  Table I .  Composition  o f S e v e r a l V a r i e t i e s of Canadian  Apple  and Apple  Juice. A p p l e  J u i c e Clarified  T i t r a t a b l e A c i d s (as m a l i c ) ( X ) T o t a l A c i d s (mg/100 ml) Malic Quinic Citric Phosphoric Ascorbic  Potassium Pectin  (mg/100 ml)  (X)  Centrifuged Solids 1  2  3  4  5  6 7  5  0.27 N.D. N.D. N.D. N.D. N.D.  0.30 N.D. N.D. N.D. N.D. N.D.  N.D.  2.2  5.4  2.7  10.89  11.79  11.32  11.13  6 8.30°  6 8.81°  6 9.00  3.04 N.D.  2.61 N.D.  2.89 N.D.  1.95 N.D.  N.D.  0.067  0.038  0.026  0.020  9  N.D.  N.D.  N.D.  N.D.  N.D.  N.D.  0.52  0.42  0.54  N.D.  N.D.  N.D.  N.D.  N.D.  2  2  7  (X)  0.54 N.D. N.D. N.D. N.D. N.D.  Mean  6 8. 16°  O.U  Total Solids ( X )  0.57 N.n. N.D. N.D. N.D. N.D.  N.D.  130  4  0.23 13.1 3.81 N.D. N.D.  0.048 59 59 7.4 N.D. 4.5  0.049  Tannin ( X )  Spartan  0.21 13.2 3.90 N.D. N.D.  0.23 12.7 3.45 N.D. 1.053  1.45 0.70 0.42 0.57 0.26  Delicious  0.19 12.7 3.34 N.D. N.D.  0.015 0.73 0.06 0.085 0.0033  3  Red  Mean  0.238 11.20 3.45 0.30ft 1.0456 0.488 682 628 29.9 0-15 15.1  Mcintosh  Mean  Mean  10.74 5.48 2.50 1.75 1.01  T o t a l Sugars ( X ) Fructose Glucose Sucrose Sorbitol  2  S.D.  41.5  (mg/100 g )  Crushed  Mean  Component Aah ( X ) S o l u b l e S o l i d s ( g / 1 0 0 g) pH Amino A c i d s (meq./100 ml) Density (g/100 ml)  1  3  Ryan (1972). Atkinson and Strachan (1949b). Strachan et a l . (1951). P e c t i n as c a l c i u m p e c t a t e . Percent o f pulp by volume a f t e r c e n t r i f u g i n g Determined as r e d u c i n g sugars. N.D. - Not determined.  15 ml i u i c e  13.7  13.90 N.D.  f o r 5 min at 2200 rpm.  14.85 N.D.  6  13.95 N.D.  22. (Table I). The water content of the juice is important to juice quality as it affects both soluble solids and juice density. Pectin and pectin-like compounds also affect body or viscosity of the juice, as well as juice appearance through the colloidal suspension of cloud particles (Moyer and Aitken, 1980; Rouau and Thibault, 1984). Apple juice flavour is another important aspect of juice quality and is dependent on several components, namely sugars, organic acids, aromatic volatiles and astringent tannins (Smock and Neubert, 1950). Approximately 90% of the soluble solids are sugars, with fructose typically present in amounts three times greater than glucose. Still smaller amounts of sucrose and sorbitol are also present (Ryan, 1972; Li and Schuhmann, 1983; Melton and Laas, 1985; Lee and Wrolstad, 1988). Organic acid levels, although much smaller than sugar levels, are extremely important in balancing apple juice flavour. Juice may be either flat or sharp depending on the acid content (Atkinson and Strachan, 1949b). The major organic acid present is malic acid, accounting for 71-94% of the total acids. Smaller amounts of quinic and citric acid are also usually present, while succinic, galacturonic, lactic, glyoxylic, oxaloacetic, shikimic and still other acids have likewise been detected. Additionally, small amounts of tannins such as chlorogenic acid, phloridzin, catechin and epicatechin have a marked effect both on flavour due to their astringency, as well as on appearance by acting as major substrates in enzymatic browning reactions and as precursors to haze and sediment formation. The delicate flavour of apple juice is further affected by the presence of alcohols like methanol, ethanol and propanol; carboxyl compounds such as acetone, acetaldehyde and hexenal; esters like ethyl butyrate and ethyl caproate. These components, because of their volatility, are extremely sensitive to processing conditions and are easily lost especially through heating (Moyer and Aitken, 1980; Lee and Wrolstad, 1988). Only with careful blending of all the aforementioned components can the producer obtain a juice with maximum aroma and flavour (Smock and Neubert, 1950). The Canadian apple juices presented in Table I are similar in content to those reported by Mattick (1984) and Lee and Wrolstad (1988). Unoxidized crushed juice has a substantially higher tannin content perhaps due to destruction of polyphenol oxidase during processing or increased contact with  23. the skin. The values obtained for Canadian apple juices are also consistent with those reported for Canadian grown apples (Table I). The low sugar and high tannin levels of Mcintosh apples appear readily balanced by blending with low acid, low tannin and high sugar varieties such as Red Delicious and Spartan. Of nutritional significance are the extremely low levels of ascorbic acid of all three cultivars, a condition which has made fortification of all apple juice products mandatory in Canada since 1943 (Smock and Neubert, 1950). A knowledge of the composition of apple juice cloud is necessary for the development of techniques which either stabilize or destabilize the cloudy suspension. Analysis of centrifuged cloud material by Carpenter and Walsh (1932) and Yamasaki et al. (1964) showed cloud material to consist mainly of pectic material (31-40%) and protein (23-39%), with lesser amounts of sugar, tannin, fiber, minerals and ether soluble substances such as gums, resins and waxes (Table II). Examination of the data in Table II reveals cloud material to be quite heterogenous in nature, with a high degree of variability existing between the suspended materials of different cultivars (Carpenter and Walsh, 1932; Johnson et al., 1968). Yamasaki et al. (1964; 1968) further demonstrated that juice viscosity was influenced mainly by pectin, where increasing levels of pectin increased viscosity and made juice more difficult to clarify, while decreases in viscosity generally resulted in coagulation and precipitation of suspended materials. Carpenter and Walsh (1932) and Yamasaki et al. (1964) also reported the surface of suspended materials to be negativley charged based on colloidal precipitaion of cloud with gelatin and on electrophoretic measurements. From these studies, Yamasaki et al. (1964; 1967) proposed a model for the suspension of cloud in apple juice where particles exist as protein- carbohydrate complexes surrounded by negatively charged protective colloids mainly in the form of pectin. Degradation of the protective colloid by pectic enzymes would expose positively charged protein in the interior of the complex, allowing electrostatic interaction of the positively charged interior with the negatively charged pectin-coated surface not yet degraded by enzymes. Such an interaction  TABLE  I[.  COMPOSITION OF CENTRIFUGED MATERIAL FROM APPLE JUICE  Composition, %  Northern Spy  Dry material, grams . 2  22.00  Rome Beauty  37.40  Russet  43.30  Ash Fiber Protein (N X 6.25). Protopectin . Pectic material . Tannin Ether extract Total sugars  2.80 4.24 19.05 1.58 33.02 1.58 8.19 11.60  4.85 2.66 30.70 3.64 24.00 1.46 6.57 3.84  5.50 2.80 24.75 2.72 22.80 0.96 5.06 2.91  Total  82.06  77.72  67.50  3  4  1  * D. C. Carpenter and W. F. Walsh, N. Y. Agr. Expt. Sta. Tech. Bull., 202, 1932. Centrifuged material was obtained in all cases from 5 gallons of raw juice and -dried to constant weight at 102°C. ^Protopectin (pectose) extracted by M/75 HCl and reported as calcium pectate. Pectic material extracted by Af/75 NaOH and reported as calcium pectate. 4^  25. would ultimately result in flocculation and destabilization of the suspended material.  (e). CHANGES OCCURRING DURING STORAGE OF JUICE Storage of juice is accompanied by a variety of changes which ultimately have a deleterious effect on juice quality. Chemical reactions taking place during storage are largely nonenzymatic in nature, as enzymes are previously inactivated and most reactions have already gone to completion (Pollard and Timberlake, 1971). One of the most frequently encountered problems during juice storage is the development of post-bottling hazes which are usually not detected until several weeks of storage and may involve a whole production run or more (Van Buren, 1984). Hazes are due to suspended matter in the size range of 10 -10 nm, particles which are too small to sediment but large enough to 2  4  refract light (Whitaker, 1984). Nonbiological, post-bottling hazes are mainly caused by oxidation of polymerized phenolics, by proteins or by carbohydrate polysaccharides individually or in combination with each other and metals such as copper. Hazes can be removed or avoided by a variety of techniques depending on their composition. Protein-phenolic haze, the most common post-bottling haze, can be avoided through reduction of protein by minimizing use of enzyme and gelatin, as well as by treatment with bentonite. Pectin, starch and arabinose hazes can be destabilized through the action of pectin lyase, amylase and endoarabinase enzymes, respectively. Slow developing tannin hazes may be impossible to avoid, but can be inhibited by mimmizing oxidation during processing and by storage at low temperatures (Heatherbell, 1976a; Van Buren, 1984; Whitaker, 1984). Further changes during juice storage include degradation of flavour, aroma, colour and ascorbic acid mainly through a series of reactions referred to as nonenzymatic browning (Smock and Neubert, 1950; Pollard and Timberlake, 1971). Nonenzymatic browning can occur by three major pathways: carmelization, ascorbic acid oxidation and the Maillard reaction, with only the latter two being of consequence in apple juice. Ascorbic acid oxidation is catalyzed by low pH  26. and elevated temperatures, conditions which decompose ascorbic acid to products ultimately forming brown pigments (Labuza and Saltmarch, 1981). Nonenzymatic browning via the Maillard reaction follows the general pathway described below: sugar + amino acid—^Amadori compound—^->Brown pigment (Hashiba, 1982). As outlined by this pathway, the Maillard reaction has three phases: a condensation reaction between an amino group contributed by a free amino acid or protein with the carbonyl group of a reducing sugar to produce Amadori compounds; an intermediate phase involving the removal of the amino group with subsequent dehydration, cyclization, fragmentation or amine condensation to produce reductones, alpha-dicarbonyls and HMF; a final phase involving the complex polymerization of products in the intermediate phase to form soluble and insoluble melanoidin pigments as well as other compounds affecting juice flavour and aroma (Hodge and Osman, 1976; Saltmarch and Labuza, 1982). Nonenzymatic browning contributes little to the colour of single strength apple juice since dilution of reaction products is substantial and the majority of browning has already occurred through the enzymatic oxidation of polyphenols during processing (Johnson et al., 1969). Nonenzymatic browning is more significant in concentrated apple juice due to the increase in browning reaction rate with increasing soluble solids (Toribio et al., 1984). As a result, colour develops very rapidly at commercial levels of concentration (Toribio and Lozano, 1984). Nonenzymatic browning can be inhibited by: pH levels below the amino acid isoelectric point; low temperatures during processing and storage; dilution of reactants; use of nonreducing sugars; reduction of reducing sugars to alditols; use of bisulfites to form addition products with sugars, thus preventing formation of brown pigments (Hodge and Osman, 1976; Saltmarch and Labuza, 1982).  (f). EMERGING TECHNIQUES IN APPLE JUICE PRODUCTION In the development of new fruit processing techniques, manufacturers attempt to create continuous systems which minimize degradative changes such  27. as oxidation while at the same time improving process efficiency and juice yield. The Winklepress, decanter centrifuge and diffuser extraction systems are examples of new innovations designed to improve upon existing processes. The Winklepress is a horizontal press with two endless polyester belts wound in serpentine fashion around perforated rollers, giving continuous extraction and yields of up to 80% if two presses are operated in series (Binnig and Possman, 1984). The decanter centrifuge also offers a continuous process. A slurry, pumped in at one end of the centrifuge, is subjected to the centrifugal action of the rotating bowl which causes solids to be deposited on the bowl surface while clarified liquids flow out of the bowl through an adjustable discharge port. Solids are removed by the action of a rotating screw which conveys the solids to a second discharge port (Moyer, 1984; Beveridge et al., 1986b). The decanter centrifuge offers the advantage of continuous operation, ability to handle hard to press apples and the ability to process a variety of fruit (Beveridge et al., 1987a). However, capital cost is high and similar to the screw press, processing capacity is low. Good yields (74-80%) are only achieved with the use of enzymes, juice is high in suspended solids and resulting pomace may have a high moisture content requiring second pressing or extensive drying (Moyer, 1984; Beveridge et al., 1987a; 1987b). Still another method offering a continuous process and high juice yields is the diffuser extraction system. Diffuser extraction may be used either alone or as a secondary process with other press systems and involves movement of sliced apples or pressed residues counter-current to a warm or cold water liquid phase. Water transports soluble solids away from the apple tissue with the extraction stream becoming more concentrated as it moves along the diffuser. Diffuser systems provide yields of 92-96% based on soluble solids content, but the equipment is expensive, the effort required to dry press residues is increased and the addition of water prevents legal use of this system in many countries, including Canada (Glunk, 1981; Wucherpfennig, 1981). The filtration and concentration aspects of processing have also undergone recent improvement with the introduction of ultrafiltration and reverse osmosis. Both systems utilize membranes to achieve separation of juice components, dif-  28. fering mainly in the size of molecules being separated. Ultrafiltration systems are available in several different configurations all of which separate the influent stream into a permeate stream and a concentrate stream, with separation based on molecule size (Milnes, 1984; Paulson et al., 1984). As the influent stream flows along the membrane, high molecular weight compounds such as pectin, starch, yeasts and molds are retained by the membrane, while low molecular weight solutes and solvents pass through the membrane to be collected in an outer cartridge (Short, 1983). With the use of mineral membranes, ultrafiltration offers the advantages of increased recovery, enhanced product, decreased material costs, 'cold sterilization' for aseptic packaging and long membrane life (Milnes, 1984; Terre, 1987). Ultrafiltration systems have also been developed that can obtain 85% yields of juice direcdy from enzyme-treated puree, eliminating the need for pressing equipment and consolidating the extraction, clarification and filtration processes; into a single pass operation (Thomas et al., 1986). The crystal clear juice produced by ultrafiltration is ideally suited for concentration by reverse osmosis. The small pore size of the membranes used in reverse osmosis combined with an applied pressure sufficient to overcome the osmotic pressure of the dissolved solids, results in the preferential permiation of water and the formation of concentrates up to 20-30 °Brix (Sheu and Wiley, 1983; Paulson et al., 1984). Reverse osmosis systems operate at low temperatures, providing increased aroma and flavour retention, decreased energy consumption and lower capital costs (Timbers, 1975; Sheu and Wiley, 1983) Advances in enzyme technology also allow juice processors to increase yields and improve juice quality. The importance of enzymes in the clarification of apple juice has been recognized since the early 1930's (Willaman, 1933). Today, enzymes are used much more extensively to facilitate pressing and extraction operations, as well as to aid in clarification and filtration after extraction (Kilara, 1982). Furthermore, enzymes are required to prevent gel formation during concentration of apple juice and to ensure commercially acceptable yields when processing apples of poor pressing quality (Smock and Neubert, 1950; de Vos and Pilnik, 1973). Use of immobilized enzyme systems requiring decreased amounts of enzyme and lower incubation temperatures to allow greater reten-  29. tion of flavour and aroma may soon be available to the juice industry (Vijayalakshmi et al., 1980). Until that time however, the need still exists for the production of enzymes that will degrade pectin and cellulose at low temperatures, optimize juice yields and complete clarification in 2-4 hours rather than the currently required 10-18 hours (Kilara, 1982).  30. METHODS AND MATERIALS  S. APPLE JUICE PREPARATION This study examined oxidized apple juice processed without a blanching step and natural apple juice processed with a blanching step. Oxidized juice was obtained from Mcintosh, Red Delicious and Spartan apples harvested at the Summerland Research Station in 1985. 'Fresh' apples were stored at 4 C for two weeks after harvest and then processed, while 'stored' apples were stored for nine months at 4 C before processing. Apple maturity was measured by testing firmness and starch content (Lau, 1985). Fresh Mcintosh, Red Delicious and Spartan apples had an average starchratingof five, three and five respectively, and an average firmness of 69.5 N, 80.7 N and 69.3 N respectively as tested by a Magness-Taylor pressure tester with a 7.8 mm probe. Stored Mcintosh, Red Delicious and Spartan apples all had an average starch rating of nine and an average firmness of 40.8 N, 47.1 N and 44.4 N respectively as tested by a Magness-Taylor pressure tester with a 7.8 mm probe. The starch test involved transversely bisecting 10 apples perpendicular to the core and immersing the freshly cut surface of the top half of the fruit in a dilute iodine solution for one minute. The starch test gave a measure of apple maturity based on a nine- point scale where a rating of one represented an immature apple with the whole cut surface reacting to turn blue, while a rating of nine corresponded to an overmature apple in which none of the cut surface turned blue (Lau, 1985). A smooth puree was produced from 10 kg of apple by blending batches of 500 g of destemmed apple cut into two centimeter cubes with 200 ml of a 200500 ppm sulfite solution (sulfite as potassium metabisulfite). The concentration of sulfite was adjusted within the range described so that browning was only just inhibited prior to blanching. The following correction factors were used to compensate for the dilution effect introduced by addition of sulfite: fresh apples, Mcintosh, 1.42; Red Delicious, 1.33; Spartan, 1.39; stored apples, Mcintosh, 1.43; Red Delicious, 1.38; Spartan, 1.42. Unless otherwise stated, juice was expressed from the puree by centrifugation at 7970 x g for 10 min with a Sorvall  31. RC-5 centrifuge equipped with a GSA (145.6 cm) rotor. For the chemical analyses and storage trials, the supernatant was then decanted into a buchner funnel and filtered through Whatman 514 filter paper (particle retention 20-25 um). Enzyme-treated juice was prepared by incubation of the puree with 0.5% Irgazyme 100 (CIBA-GEIGY Corp.), a polygalacturonase with lyase and pectinesterase side activities, for 1 h at 45 C. Juice was then expressed from the enzyme-treated puree as described above. Juice and puree were frozen and stored at -18 C until required. Natural juice was produced as above with batches of apple puree being fed continuously into the blanching apparatus after blending. Blanching was done in 316 stainless steel tubing (222 cm x 0.775 cm) immersed in boiling water, while cooling was achieved with an identicle tube immersed in flowing tap water and connected to the first by a tygon holding tube (Beveridge et al. 1986a). The heating and cooling coils were fitted with recording copper-constantan thermocouples. During processing, temperatures from the thermocouples were monitered at 5 min intervals using a Hewlett- Packard 3421 A data acquisition unit. The puree was heated to over 90 C for at least 25 sec, which was sufficient to destroy apple polyphenoloxidase (Beveridge and Harrison, 1986). After processing, separate batches of the puree were mixed together in a Hobart H 600 mixer, frozen and stored at -18 C.  II. PRELIMINARY TURBIDITY STUDIES For the preliminary turbidity studies, 300 g of puree was measured into a 500 ml beaker and held in an ice-water bath for 30 min. The puree was then combined with 3.0 g Irgazyme 100 (CIBA-GEIGY Corp.) and blended for 1 min at low speed in a chilled Waring blender. The puree was returned to the icewater bath and a duplicate sample prepared. The enzyme-treated puree was then weighed into 50 ml centrifuge tubes, making sure filled tubes were of equal weight. Centrifuge tubes were covered with a rubber stopper and kept in an icewater bath until all the tubes were filled. Time zero tubes were placed directly from the ice-water bath to a Sorvall RC-5 centrifuge equipped with an SS-34  32. (10.7 cm) rotor and centrifuged at 7700 x g for 10 min, while at the same time the remaining tubes were placed into a water bath heated to the desired temperature. After incubation for the desired time, the tubes were removed from the water bath and centrifuged as described above.  Time of incubation was  measured from the time the tubes were placed in the heated water bath to the time at which the tubes were removed from the water bath and placed in the centrifuge. After centrifugation, the supernatant was decanted into test tubes, the tubes were placed in an ice-water bath and the absorbance at 600 nm was immediately measured against a distilled water blank using a Varian DMS 100 spectrophotometer fitted with a 1 cm cell.  III. CLOUD AND COLOUR STABILITY To each 1 L of juice prepared as described above (Section I, p. 30-31), was added 1000 ppm potassium sorbate to prevent spoilage by yeasts and molds (Moyer and Aitken, 1980). Duplicate 200 mL aliquots of juice contained in sealed 220 mL plastic containers (Falcon) were stored at 0 C and 24 C for 84 days. Colour stability was followed by visible scans (400-700 nm) of juice against a distilled water blank using a Varian DMS 100 spectrophotometer equipped a 1 cm cell. Turbidity or cloud was measured in duplicate as absorbance at 600 nm using a Varian DMS 100 spectrophotometer fitted with a Routine Sampler. Colour ratio was determined as described by Wrolstad (1976): (A530 - A600)/(A420 - A600).  IV. CHEMICAL ANALYSES Chemical analyses were performed on four fractions of apple puree: 1. Juice - prepared as described in Section I (p. 30-31). 2. Pulp  - sedimented after centrifugation at 7970 x g for 10 min.  3. Cloud - sedimented after centrifugation at 37,000 x g for 60 min.  33. 4. Serum - clarified liquid (absorbance at 600 nm 0.100 absorbance units) remaining after centrifugation at 37,000 x g for 60 min.  (a) ALCOHOL INSOLUBLE SOLIDS (AIS) The alcohol extraction procedure was based on the method of Robertson (1979). 15 mL juice or serum was measured into a 50 mL graduated, conical bottom centrifuge tube. Hot (75 C) 95% ethyl alcohol was added to a volume of 40 mL and the mixture heated for 10 min in an 85 C water bath, stirring occasionally. The volume of the mixture was made up to 45 mL and the tubes centrifuged at 190 x g for 30 min with an MBE counter-top centrifuge. After centrifugation the supernatant solution was decanted and discarded. The leaching was repeated with hot 63% ethyl alcohol for 10 min at 85 C followed by centrifugation as described above. The supemate was discarded and the pellet was removed to an evergreen vial and stored at -18 C. The pellet was frozen further to -50 C, fbreezedried in a Virtis RePP freeze-drier for 72 h and weighed.  (b) ORGANIC ACIDS Citric, galacturonic, malic, quinic and succinic acids were determined by high pressure liquid chromatography (HPLC) using a Hewlett-Packard 1084 B HPLC equipped with either a Hp 1084 B Variable Wavelength detector (UV 210:430) or a Waters 410 Differential Refractometer at 40 C with a sensitivity of 32 and a scale factor of 20. The integrator was set at a slope sensitivity of 0.1 with a recording attenuation of 4-8. The recorder was operated at a chart speed of 0.5 cm/min. A Polypore H PPH-GU (styrene-divinyl benzene copolymer, hydrogen form) 10 um guard column (Brownlee) was used in combination with an Aminex HPX-87H (sulfonated divinyl benzene-styrene copolymer) column, 300 mm x 7.8 mm (Bio-Rad) either individually or in series with a second Aminex HPX-87H column and maintained at 60 C. Injected  34. samples (20 uL) were eluted with 0.0IN H2SO4 at 65 C flowing at 0.6 mL/min if one column used or at 0.8 mL/min if two columns mounted in series. A standard solution containing 1.0 mg/mL each of the components was prepared. The standard solution was further diluted to 0.8, 0.6, 0.4, 0.2, 0.1, 0.08, 0.06, and 0.04 mg/mL and standard curves were prepared for each component. A 20 uL sample was injected and the response to each individual component was calibrated using the external standard method. Apple juice and serum samples were prepared as reported by Beveridge et al. (1986c) by passing 1 mL sample through a 5 mm x 5 cm bed of anion exchange resin (AG1-X8, acetate form, Bio-Rad), and eluting to a volume of 25 mL with double distilled water. The eluate was collected in a volumetric flask and retained for sugar analysis. Organic acids bound to the resin were removed with either 1.0N H2SO4 collected in a 25 mL volumetric flask or 1.5N H2SO4 collected in a 10 mL volumetric flask. A 25 times dilution of the sugars and a 25 or 10 times dilution of organic acids results. Samples were filtered through a GF/B glass microfibre filter, 2.5 cm (Whatman) in combination with an HVLP filter, 0.45 um (Millipore) into 2 mL crimp-capped sample vials for analysis. Percent recovery of organic acids was as follows: citric, 95 %; galacturonic, 100 %; malic, 98 %; quinic, 103 %; succinic, 101 %. The concentration of the components in the original sample was calculated as follows: mgAcid/100 mL sample = mg determined from standard curve x dilution factor x 100. Total acids were measured as the sum, in g/100 mL, of citric, galacturonic, malic, quinic and succinic acids in each sample as determined by HPLC.  (c) PECTIN Methodology was adapted from Voragen et al. (1983). 5 mg samples of freeze-dried apple juice pulp, cloud and AIS were measured into test tubes to which 1 mL of the following enzyme mixture was added:  35. 2.0% Celluclast 200L, exo-B-l,4-glucanase (NOVO Laboratories, Inc.) 1.0% Diazyme L-150, amyloglucosidase (Miles Laboratories, Inc.) 0.2% Novozyme 188, cellobiase (NOVO Laboratories Inc.) 0.1% Pectinex 3XL, polymethylgalacturonase, polygalacturonase, pectinesterase (Swiss Ferment Ltd.). The above enzyme preparation was made to 100 mL in a volumetric flask just prior to addition to sample tubes. A similar preparation used by Voragen et al. (1979) and Voragen and Pilnik (1981) was reported to be the most effective mixture for apple cell wall digestion. After incubation, the enzyme solution was inactivated by placing the tubes in a boiling water bath for 1-2 min, a process which caused the enzyme proteins to precipitate. To prevent this precipitate from interfering with the separation of components on an anion exchange column, the samples were centrifuged for 5 min at 150 x g with an MBE counter- top centrifuge. The supernate was quantitatively transferred to an anion exchange column described previously (p. 34). The pellet was washed with 10 mL distilled water and re-centrifuged. This procedure was repeated 2 more times, with the supernate transferred after each centrifugation to the anion exchange column. The samples were prepared in duplicate along with an enzyme solution control and a 0.8 mg/mL standard organic acid solution for recovery determinations. Galacturonic acid was deteiTriinedby HPLC as outlined previously. Recovery of galacturonic acid averaged 101%. Pectin was reported as mg galacturonic acid/100 mL juice or serum, or on a percentage basis (w/w) for pulp and cloud.  (d) pH and TITRATABLE ACIDITY A 2 mL sample of juice or serum was placed in a 50 mL titration vessel and pH was measured using a G2040C glass electrode in combination with a K4040 calomel electrode and a Radiometer ETS822 Autotitrator system. Following the pH measurement, 18 mL distilled water was added and the sample  36. was stirred while being titrated with 0.1 N NaOH to an end point of pH 8.1. Titratable acidity was calculated as mg malic acid/100 mL. Measurements were performed in triplicate.  (e) PROTEIN 50 mg cloud, AIS, or 100 mg pulp were combined with 40 mg HgO on cigarette paper, folded together and placed in 75 mL digestion tubes along with 2-3 hengar granules, 1.9g K2SO4 and 5 mL concentrated H2SO4 added using an Eppendorf pipette. A blank consisting of 40 mg HgO wrapped in cigarette paper was also prepared. 2 mL H2O2 was added to pulp and 4 mL H2O2 to cloud and AIS in 1 mL aliquots, with reaction subsiding before addition of more H2O2. Tubes were alternately placed for 5 sec on digestion block at 350 C and off for 60 sec until smooth boiling was achieved. When boiling smoothly, tubes were placed for 60 min on digestor at 350 C. After digestion, tubes were removed from digestor and cooled, made up to 75 mL with distilled water, mixed and poured into sample vials for analysis. Samples were analyzed for nitrogen using a Technicon Autoanalyzer II according to Industrial Method No. 329-74W/A. Protein was determined from nitrogen analysis as N x 6.25 (Yamasaki et al., 1964) and expressed on a percentage basis (w/w).  (f) SOLUBLE SOLIDS Soluble solids of the juice and serum were measured in triplicate as °Brix using a Reichert Abbe Mark II refractometer at 25 C.  (g) SUGARS Sucrose, glucose, fructose and sorbitol were separated from organic acids in juice and serum as detailed for organic acids (p. 34). Sugars were determined as described by Beveridge et al. (1986c) using a Hewlett-Packard 1084 B HPLC equipped with a Waters 410 Differential Refractometer at 40 C. Detector sen-  37. sitivity was 64 with a scale factor of 20. A Polypore CA PPH-Gu (styrene-divinyl benzene, hydrogen form) 10 um guard column (Brownlee) preceded by a Polypore A PPA-Gu (styrene-divinyl benzene, anionic form) 10 um anion exchange column was used in combination with a Polypore CA PPC-224 (styrenedivinyl benzene, calcium form), 4.8 mm X 22 cm column (Brownlee) maintained at 60 C. Injected samples (20 uL) were eluted with double distilled water at 65 C flowing at 0.3 mL/min. Percent recovery of sugars was as follows: sucrose, 100 %; glucose, 108 %; fructose, 100 %; sorbitol, 85 %. The concentration of the components in the original sample was calculated as follows: gSugar/100 mL sample = g determined from standard curve x dilution factor x 100. Total sugars were measured as the sum, in g/100 mL, of sucrose, glucose, fructose and sorbitol in each sample as determined by HPLC.  (h) ZETA POTENTIAL Zeta potential measurements of juice were made using a Lazer Zee Meter model 501 equipped with a 25 mL chamber and operating at 150 mV and 24 C. Triplicate measurements were performed on each sample.  V. TRANSMISSION ELECTRON MICROSCOPY (TEM) All samples were examined with a Philips E M 300 transmission electron microscope operating at 60 kV. With the exception of thin sectioning, each treatment was performed twice; duplicate samples were prepared within treatments, then representative sections were photographed. Unless otherwise stated, measurements of particle dimensions were taken of 10 randomly selected particles within each particle category. The range and mean of these measurements were recorded along with descriptions of particle stain density.  38. (a) THIN SECTIONING Duplicate 2 mL aliquots of juice were dialyzed overnight against 4L distilled water at room temperature. The water was changed after the first hour of dialysis and again before the last hour of dialysis. Dialyzed juice was stored refrigerated (4 C) until used. Pellets of juice cloud were obtained by centrifugation of dialyzed juice at 343,000 x g for 30 min with a Beckman ultracentrifuge. The pellets were fixed in a 2% osmium tetroxide - 0.01 M cacodylate buffer solution at pH 7.0 for 1 h, dehydrated in 50%, 70%, 95%, absolute ethanol and propylene oxide (two times each for 10 min), embedded in Epon 812 and cured at 60 C. Thin sections (60-90 nm; Hunter,1984) were obtained using a Reichert OM U2 ultramicrotome. The sections were stained first with 5% uranyl acetate for 20 min, washed in distilled water and then stained with Reynold's lead citrate in combination with 0.01 N NaOH (1:1 (v/v)) for 10 min (Sjostrand, 1967), washed with distilled water and placed on filter paper in a covered petri plate to dry.  (b) NEGATIVE STAINING Copper grids (400 mesh) were prepared with a collodion support film (0.5% collodion in amyl acetate) and coated with carbon in an Edwards E306A high vacuum coating unit. The grid was placed on a drop of dialyzed juice for 5 min, transferred to a drop of distilled water for 1 min and washed with 10 drops 2% uranyl acetate. Excess stain was removed by touching the edge of the grid with a piece of filter paper. Grids were then air dried in a covered petri plate.  (c) SHADOW CASTING Copper grids (400 mesh) were prepared as described above. Dialyzed juice was sprayed onto grids using a high velocity spray gun and air dried. The grids were shadowed with platinum-palladium (4:l(w/w)) on a tungsten filament at an angle of 23.6 (tan 1/2) in an Edwards E306A high vacuum coating unit.  39.  VI. STATISTICAL ANALYSES The experiments were set up in a 2 x 2 x 2 x 3 completely randomized factorial design. The main effects to be tested were: Variety  - Mcintosh, Red Delicious, Spartan;  Condition  -fresh,stored;  Process  - blanched, unblanched;  Maceration - enzyme-treated, not enzyme-treated. Analysis of variance and paired comparison T-Tests were performed by the S AS Version 5 ANOVA and means procedures, respectively, using a VAX/VMS V02 computer. Discriminant analysis was used to classify juices according to i  method of production based on the qualitative measurement of turbidity, colour ratio and total acidity. The discriminant analysis was performed by the SAS Version 5 Stepdisc, Candisc and Discrim procedures (SAS, 1986).  40.  RESULTS AND DISCUSSION  I. PRELIMINARY TURBIDITY STUDIES In previous studies by Montogomery and Petropakis (1980) and Beveridge and Harrison (1986) inactivation of polyphenol oxidase by blanching of pear tissue caused juice from blanched puree to be cloudier and more difficult to clarify than juice from unblanched puree. Similar results were found in the present study using apple puree not treated with enzyme. Processing stored apples also increased juice turbidity, supporting the findings of Lidster et al. (1984). Without enzyme treatment, juice yields were low (50-55%, w/w), separation between pellet and supernatant was poor and turbidity readings were greater than 1.0 absorbance units indicating increased suspended solids. Treatment with Irgazyme 100 provided more consistent reproduction of particle size and number by increasing pellet firmness after centrifugation and thus allowing better separation of supernatant, increased yields (74-82%, w/w) and turbidity readings less than 0.6 absorbance units. A comparison of juice from enzyme-treated fresh apples with the juice from enzyme-treated stored apples (Fig. 4) indicates that juice from stored apples was clarified to a greater extent by treatment with Irgazyme 100 than juice from fresh apples, perhaps due to the degradation of particulate forming components during storage. The effect of blanching on turbidity of juice from enzyme-treated fresh apples was variable, perhaps as a result of the variable levels of starch at harvest (Mcintosh, 5; Red Delicious, 3; Spartan, 5). Blanching fresh Spartan apple puree resulted in a more turbid juice, while there was little difference between turbidity of juices from blanched and unblanched Mcintosh purees. Juice from unblanched Red Delicious puree was slighdy more turbid than juice from blanched puree. In stored apples where starch levels are negligible, blanching increased the turbidity of juice from all three cultivars. Differences were larger at lower incubation temperatures (15 C, 25 C) and smaller at higher temperatures (35 C, 45 C). The increased turbidity of juice from enzyme-treated  41.  Blanched - Not Blanched .4 —  • Fresh • Stored  E c  o CD  "°  cu o c _Q  .2  v_ O CO  <  .0 Mcintosh  Spartan  Red Delicious  Figure 4. The effect of blanching on cloud formation in the juice f r o m apples incubated with 1% Irgazyme 100 for 45 minutes at 35 C.  42. blanched puree suggests blanching either increases particle numbers or alters the nature of the particles present in the juice. Charley (1961) also noted that heating of freshly milled puree to 80 C or higher inactivates not only polyphenol oxidase, but innate pectic enzymes as well. The presence of these naturally occurring pectic enzymes may be required to promote the action of commercial enzyme preparations and their inactivation may result in the observed turbidity differences between juices from enzyme-treated blanched and unblanched puree. Although differences in turbidity were noted, it must be remembered that turbidimetric measurements are based on the scattering of light by particles and are therefore highly empirical, providing evidence only for the presence of particles and the existence of differences between measurements (Charalambous, 1984). However, these measurements parallel visual appearances and provide a means for quantitating observed differences in juice turbidity. Since the possibility exists that the physical properties of suspended particles may change with no change in turbidity, electron microscopy in combination with other physical and chemical analyses was used to gain a better understanding of the effect of blanching, refrigerated (4 C) post- harvest storage and enzyme treatment with Irgazyme 100 on the structural and chemical nature of cloud particulate in apple juice.  -  II. CLOUD AND COLOUR STABILITY DURING APPLE JUICE STORAGE  (a). CLOUD STABILITY Results of apple juice storage at 0 C and 24 C are shown in Figs. 5-8. Evaporation from cups used to store juice samples was 0.4% at the end of the storage period and thus considered to have a negligible effect on results. Generally all juice preparations showed a slight, unexpected increase in turbidity  Figure  5. The effect of storage at 0 C on the stability of cloud i n the juice from fresh apples not treated with enzyme.  Figure  6. The effect of storage at 0 C on the s t a b i l i t y of cloud i n the juice f r o m stored apples not treated with enzyme.  45. after the initial day of storage. Turbidity of juice from fresh apples stored at 0 C decreased only slightly throughout the storage period, with very little difference between juice from blanched and unblanched puree (Fig. 5). A greater difference was observed between processing treatments in juice from stored apples (Fig. 6). Turbidity of juice from blanched puree was greater than that of juice from unblanched puree and increased slightly during storage. On the other hand, turbidity of juice from unblanched puree decreased during storage. Upon storage at 0 C, oxidized juice from stored Mcintosh and Red Delicious apples separated into two phases consisting of a clear liquid and an opaque gel. Gel from oxidized juice of Red Delicious apples ultimately precipitated, while that of Mcintosh apples remained stable throughout the 84 day storage period. Although the oxidized juice of stored Spartan apples did not form distincdy separate phases, a marked increase in viscosity was observed. Gel formation in oxidized juice of stored apples is presumably due to increases in pectin solubilized during post-harvest storage (Section IHa, p. 50), although structural changes in pectin molecules may also contribute to gel formation (Section IVb, p. 67). Storage of juice from fresh apples at 24 C caused little change in the turbidity of juice from blanched and unblanched Mcintosh and Red Delicious apple puree (Fig. 7). The turbidity of juice from blanched Spartan apple puree showed a more definite decrease, but a larger decrease was observed in oxidized juice from the unblanched puree of Spartan apple due to the precipitation of suspended material to form a sediment at the bottom of the container. During storage of juice from post-harvest stored apples at 24 C there was increased turbidity in juices from Mcintosh and unblanched Spartan apples, and decreased turbidity in juice from Red Delicious and blanched Spartan apples (Fig. 8). Again, a large decrease in turbidity was observed, this time in the oxidized juice of Red Delicious apples.  The precipitation of cloud material in oxidized  juice from unblanched puree is indicative of the presence of active endogenous pectic enzymes capable of degrading pectic material and destabilizing suspended cloud particles. In these studies, blanching was generally found to improve the  Figure  7. The e f f e c t of s t o r a g e a t 24 C on t h e s t a b i l i t y of c l o u d i n t h e j u i c e f r o m f r e s h apples n o t t r e a t e d w i t h enzyme.  2.50  2.25  —  • • A  Mcintosh Red Delicious Spartan Oxidized Juice Natural Juice  0.50  0.00  —  i  1  10  Figure  i 20  1  i— —i 1  30  1  40  i  1  50  r  STORAGE TIME (DAYS)  60  70  8. The effect of storage at 24 C on the stability of cloud in the juice from stored apples not treated with enzyme.  80  90  48. a  stability of cloud material suspended in apple juice by preventing precipitation or coagulation of suspended components.  (b). COLOUR STABILITY After storage for 84 days at 24 C all oxidized juice from unblanched puree was brown in colour, with the exception of juice from stored Mcintosh apples which had too much sulfite added during processing. The development of brown pigments was observed as an increase in absorbance at 420 nm. Formation of brown pigments results mainly from oxidation of phenolic compounds during juice processing and storage (Cumming et al., 1986; Vamos-Vigyazo and Nadudvari-Markus, 1983). In agreement with Beveridge et al. (1986a), blanching extracted anthocyanin pigments from apple skin, causing juice from the blanched puree of both Red Delicious and Spartan apples to be pink in colour with an absorbance peak at 513 nm (Fig. 9). However, storage of natural juice at 24 C resulted in the gradual development of a peak at 460 nm (Fig. 9), and a shift in juice colouration from pink to brown. These results are in agreement with those of Beveridge et al. (1986a), who suggest that the browning reaction in juice from blanched puree is different from both the enzymatic browning observed in juice from unblanched puree and the nonenzymatic browning which occurs in concentrates. Rather, they propose a condensation reaction between tannins and anthocyanins aided by the acid hydrolysis of anthocyanin glycosides to form yellow-brown condensation pigments. S tabilization of the pink colour observed in natural juice from blanched puree was achieved by storage at 0 C, conditions which also minimized the development of brown pigments in oxidized juice.  49. mrruL  DAY  or  STORAGE  Juict Juict  J •00  Figure  490  f r o m f r e s h a p p l e s f r o m t t o r a d a p p l * *  t_  900  960  •60  r  "i  1  900  990 l i f t L I X C T H  •SO  100  700  ( u i )  9. Visible region spectra of juice from blanched Red Delicious apples treated with Irgazyme 100 and stored at 24 C.  50.  III. CHEMICAL ANALYSES  (a). CHEMICAL COMPOSITION OF APPLE JUICE The chemical composition of varietal apple juice as affected by post-harvest storage, blanching and enzyme treatment is presented in Table m. Examples of chromatograms and analysis of variance are available in Appendix A. Examination of Table IH shows sugars to be the predominant soluble constituents in apple juice. Soluble solids, irrespective of treatment, averaged 12.83 °Brix (Table V, p. 65) which is in general agreement with literature values for both apple juice and fruit (Table I, p. 21). Soluble solids varied significantly with variety, storage, processing and enzyme treatment. Juice from Mcintosh apples had significantly lower (p < 0.01) soluble solids than juice from either Spartan or Red Delicious apples. Similar results were obtained for these cultivars by Strachan et al. (1951, Table I). Juice from stored apples had significantly higher (p < 0.01) soluble solids than juice from fresh apples, presumably due to the degradation of starch to glucose during storage (Pollard and Timberlake, 1971). Juice from unblanched puree had significantly higher (p < 0.05) soluble solids than juice from blanched puree. This however, was unexpected based on the analysis of individual sugars and organic acid constituents. Treatment with Irgazyme 100 significantly increased (p < 0.01) soluble solids due to release of galacturonic acid during enzymatic degradation of pectic material (deVries et al., 1981). Similar to soluble solids, total sugars varied significantly with enzyme treatment, where treatment with Irgazyme 100 significandy increased (p < 0.05) total sugars. Analysis of the individual sugars (Table HI) showed fructose to be present at levels two to three times greater than glucose, with lower levels of sucrose and sorbitol also present. The relative proportions of these constituents were in general agreement with those reported by Lee and Wrolstad (1988), Melton and Laas (1985), Li and Schuhmann (1983) and Moyer and Aitken (1980). One peak remained unidentified and analyses showed it was not: cellobiose, melezitose, raffinose or stachyose. Fructose did not vary significantly between  51. Table III. Composition of apple juice as affected by cultivar, post-harvest storage, blanching and treatment with Irgazyme 100. 1  Component Soluble solids  Significant Treatment Effects i)Variety*  (°Brix)  Spartan Red Delicious Mcintosh 13.59  13.06  11.84  S E M = 0.16 ii)Storage*  iii)Processing*  Stored  Fresh  13.20  12.46  Unblanched Blanched 13.03  iv)Enzyme Treatment* Total Sugars (g/100 mL)  i)Enzyme  Sucrose (g/100 mL)  i)Storage*  Treatment*  ii)Processing*  12.63  S E M = 0.13  Enzyme No enzyme 13.33 12.33  S E M = 0.13  Enzyme No enzyme 11.31 Fresh 1.49 Blanched  10.45 Stored 0.52  i)Variety*  (g/100 mL)  S E M = 0.27  S E M = 0.14  Unblanched  1.25 Glucose  S E M = 0.13  0.76  S E M = 0.14  Red Delicious Spartan 3.37 2.43  Mcintosh 2.29  S E M = 0.11 ii)Storage*  iii)Enzyme * •  Stored  Fresh  2.99  2.41  Enzyme No enzyme 305 2.35  S E M = 0.09  S E M = Q.Q9  52. Table III (continued). Component Glucose (g/100 mL)  Fructose  Significant Treatment Effects iv)Processing by  Enzyme  No enzyme  Enzyme Treatment  Blanched  3.16  Interaction**  Unblanched  2.93 2.64 SEM = 0.12  NSD  2.06  6.73  S E M = 1.33  (g/100 mL) Sorbitol  i)Storage**  (g/100 mL) PH  i)Variety**  Stored 0.42  iii)Enzyme Treatment * *  Stored 3.86  Enzyme  Red  Treatment Interaction**  Delicious Spartan Mcintosh  v)Storage by Treatment Interaction**  S E M = 0.03  Fresh 3.54  S E M = 0.01  S E M = 0.01  No enzyme  Enzyme  4.11  3.63  3.88 3.60 S E M = 0.02  3.55 3.41  Enzyme  No enzyme Stored Fresh  Mcintosh  3.72 3.51 SEM = 0.01  No enzyme Enzyme 3.87 3.53  iv)Variety by  Enzyme  0.27  Red Delicious Spartan 3.87  ii)Storage**  Fresh  4.08  3.64  3.66  3.43  S E M = 0-02  53. Table III (continued). Component Titratable Acidity  Significant Treatment Effects i)Variety*  (g malic/100 mL)  Mcintosh Spartan Red Delicious 0.573  0.463  0.330  S E M = 0.012 ii)Storage**  iii)Processing*  Fresh  Stored  0.535  0.375  Unblanched 0.470  iv)Enzyme Treatment** v)Variety by Storage Interaction**  S E M = 0.010  Blanched 0.441  S E M = 0.010  Enzyme  No enzyme  0.523  0.387  S E M = 0.010  Fresh  Stored  Mcintosh  0.684  0.462  Spartan  0.576  0.350  Red  Total Acidity (g/100 mL)  i)Variety**  Delicious  0.346 0.315 S E M = 0.017  Mcintosh  Spartan Red Delicious 1.084 0.580 S E M = 0.037  1.190  ii)Storage**  Fresh 1.065  iii)Enzyme Treatment**  Stored 0.837  Enzyme No enzyme 0.838 1.064  S E M = 0.030  S E M = 0.030  54.  Table III (continued). Component  Significant Treatment Effects  Total Acidity  iv)Variety by Enzyme  (g/100 mL)  Treatment Interaction**  Enzyme  No Enzyme  Mcintosh  1.442  0.937  Spartan  1.355  0.813  0.764  0.395  Red Delicious  S E M = 0.052 Citric acid  45  2  SEM = 3  (mg/100mL) Galacturonic acid (mg/100 mL)  i)Enzyme  Malic acid  i)Variety**  Treatment**  (mg/100 mL)  Enzyme 510.4  Fresh 719.0  iii)Variety by Storage Interaction**  0.00  S E M = 21.0  Mcintosh Spartan Red Delicious 741.9  ii)Storage**  No enzyme  Mcintosh Spartan  610.0 393.1 S E M = 15.2 Stored 444.3 Fresh 917.3 791.5  S E M = 12.4 Stored 566.5 428.5  Red Delicious  Quinic acid  2  448.3 338.0 S E M = 21.5  111  SEM= 6  122  S E M = 16  (mg/100 mL) Succinic acid (ma/100 mL)  2  55. Table III (contiued). Component Pectin as anhydro-  Significant Treatment Effects i)Variety*  Spartan Mcintosh Red Delicious  galacturonic acid  44.2  35.5  (mg/100 mL)  16.0  S E M = 4.5 ii)Storage*  iii)Enzyme  Stored 53.1  S E M = 3.7  No enzyme Enzyme  Treatment* iv)Variety by Storage Interaction*  Fresh 10.7  63.8  0.0  S E M = 3.7  Stored  Fresh  Spartan  64.5  24.0  Mcintosh  64.9  6.1  Red 29.9 2.1 S E M = 6.3  Delicious  v)Storage by Enzyme Treatment Interaction** Turbidity (Absorbance at 600 nm)  i)Processing*  Enzyme  Blanched  NSD  -9.3  1Where necessary, values were adjusted to correct for dilution due to the Due to the high variability inherent in quantitating  0.0  Unblanched 0.338  S E M = 0.095  0.159  S E M = 0.095 S E M = 0.7  addition of sulfite during processing.  components at low concentrations, statistical evaluation of citric,  quinic and succinic acids was not considered valid and therefore not performed. SEM = standard error of the mean; NSD = no significant difference. * Significant at p < 0.05.; * * Significant at p < 0.01.  0.0 S E M = 5.2  No enzyme Enzyme 0.857  Zeta Potential (rrM  Fresh 21.5  No enzyme  0.677 ii)Enzyme  Stored 106.2  56. treatments and averaged 6.73 g/100 mL juice, falling within the range presented by Mattick and Moyer (1983) and in general agreement with Ryan (1972; Table 1).  Glucose levels on the other hand, varied significantly with variety,  storage and treatment with Irgazyme 100. Glucose levels were within the range reported by Mattick and Moyer (1983) and Ryan (1972, Table I). Juice from Red Delicious apples had significandy higher (p < 0.01) levels of glucose than juice from Spartan apples which in turn had significandy higher (p < 0.01) levels of glucose than juice from Mcintosh apples. Since no significant variety effects were observed in any other sugar determined by this analysis, the varietal differences observed in soluble solids measurements were presumably due to differences in glucose levels. Juice from stored apples also had significantly higher (p < 0.01) glucose levels than juice from fresh apples probably resulting from the degradation of starch and the inversion of sucrose to glucose and fructose (Lee and Wrolstad, 1988). Although a processing by enzyme treatment interaction was significant (p < 0.01), treatment with Irgazyme 100 significantly increased (p < 0.01) glucose content in juice from both blanched and unblanched puree. Sucrose levels were also in general agreement with literature values (Table I). Significant differences in sucrose levels resulted from processing and storage treatments. Juice from stored apples was significandy lower (p < 0.01) in sucrose than juice from fresh apples due to inversion of sucrose to glucose and fructose during post-harvest storage (Lee and Wrolstad, 1988). Juice from blanched puree had significandy higher (p < 0.05) sucrose levels than juice from unblanched puree perhaps due to inactivation of invertase by blanching, although the heat employed during blanching would be expected to cause inversion of sucrose (Beveridge et al., 1986b). Sorbitol levels fell within the range presented by Mattick and Moyer (1983). Sorbitol levels were only affected by apple storage, where juice from stored apples had significantly higher (p < 0.01) sorbitol levels than juice fromfreshapples. This is consistent with the findings of Lee and Wrolstad (1988), in which sorbitol accumulation resultedfrompostharvest storage of apples at 3.5 to 7.5 C. Titratable acidity, pH and organic acid levels (Table HI) were in general agreement with literature (Table I, Lee and Wrolstad, 1988). The pH of juice  57. varied significantly with variety, storage and enzyme treatment. There were significant (p < 0.01) variety by enzyme treatment and storage by enzyme treatment interactions, but as neither of the interactions was crossed, interpretation of main effects remained unchanged. Juice from Red Delicious apples had significandy higher (p < 0.01) pH than juice from Spartan apples which had in turn significantly higher (p < 0.01) pH than juice from Mcintosh apples. Strachan et al. (1951) found similar differences between these cultivars in the analysis of apple fruit (Table I). Juice pH was also influenced by post-harvest storage, where respiratory decreases in organic acid content were reflected in the significandy higher (p < 0.01) pH of juice from stored apples (Lee and Wrolstad, 1988). Juice from puree treated with Irgazyme 100 had significandy lower (p < 0.01) pH than juice from untreated puree as a result of the release of galacturonic acid during pectin degradation. Titratable acidity measurements reflected those of pH. Titratable acidity varied with variety, storage, enzyme treatment and, unlike pH, with processing. A significant (p < 0.01) variety by storage interaction was present, indicating a varietal difference in the metabolism of organic acids during storage with a large decrease in the acidity of juice from Mcintosh and Spartan apples, and only a slight decrease in juice from Red Delicious apples. Juice from Mcintosh apples was significantly more (p < 0.05) acidic than juice from Spartan apples which was in turn significantly more (p < 0.05) acidic than juice from Red Delicious apples, results which are in agreement with those of Strachan et al. (1951, Table I). Juice from fresh apples was significantly higher (p < 0.01) in acidity than juice from stored apples, indicating that a degradation of organic acids occurred during post-harvest storage (Lee and Wrolstad, 1988; Rouchaud et al., 1985; Hansen, 1979; Hulme, 1958). Treatment with Irgazyme 100 caused a significant increase (p < 0.01) in acidity due to the release of galacturonic acid during degradation of pectic material. Juice from unblanched puree was significantly more (p < 0.01) acidic than juice from blanched puree, although a significant difference between processing treatments was not detected in measurement of pH, total acidity or individual organic acids. Total acidity, measured as the sum of citric, galacturonic, malic, quinic and succinic acids determined by HPLC, varied significandy with variety, storage and enzyme treatment and generally followed the  58. same pattern as titratable acidity. A significant (p < 0.01) variety by enzyme interaction was present, but did not change interpretation of main effects. Juice from Mcintosh apples had significantly higher (p < 0.01) total acids than juice from Spartan apples which was in turn significandy higher in total acids than Red Delicious apples. Juice fromfreshapples had significandy higher (p < 0.01) acid levels than juice from stored apples, while treatment with Irgazyme 100 also significandy increased (p < 0.01) total acidity.  |  The organic acids present in apple juice were identified as citric, galacturonic, malic, quinic and succinic acids with trace amounts of fumaric acid also detected. Several unidentified peaks were observed in juice from Mcintosh and Irgazyme 100 treated puree (Appendix A, Fig. A-2). Analysis showed the peaks were not citramlic,oc-ketoglutaric, lactic, maleic, malonic, oxalic, pyruvic or shikimic acids. Malic acid was the predominant acid in juice not treated with Irgazyme 100 with levels up to 81% of the total acids. However, treatment of puree with Irgazyme 100 resulted in the presence of galacturonic acid at levels comparable to those of malic acid, similar to the results of Dorreich (1983). Malic acid content varied significandy with variety and storage.! A significant (p < 0.01) variety by storage interaction was also present, again indicating that metabolic differences exist between varieties during post-harvest storage. However, the interaction was not crossed so interpretation of main effects was not changed. Variations in malic acid content reflected variations in pH, titratable acidity and total acidity measurements as expected, since malic acid is the predominant organic acid in juice from puree not treated with Irgazyme 100. Juice from Mcintosh apples was significandy higher (p < 0.01) in malic acid than juice from Spartan apples which was in turn significantly higher (p < 0.01) than juice from Red Delicious apples. Juicefromfreshapples was also significandy higher (p < 0.01) in malic acid than juice from stored apples. Galacturonic acid varied significandy (p < 0.01) only with enzyme treatment where, as expected, juice from puree treated with Irgazyme 100 had high levels of galacturonic acid while juice from puree not treated with Irgazyme 100 contained no galacturonic acid. The increased analytical variability inherent in quantitating the low concentrations of citric, quinic and succinic acids was evident in the relatively large stand-  59. ard error associated with analysis of these components (Lee and Wrolstad, 1988). This higher variability is most likely a reflection of the ability of the instrument to discriminate between differences in analyte concentration at these lower levels (Skoog, 1985). As a result, statistical evaluation of citric, quinic and succinic acids was not performed. Pooled treatment means for these acids are presented in Table IH. Pectin measurements were also associated with a relatively high standard error perhaps reflecting the error introduced through the large number of sample transfers required by this method of analysis. Pectin levels were lower than those reported by Atkinson and Strachan (1949b, Table I) even considering the 10% overestimation introduced by the calcium pectate method used by these workers (McComb and McCready, 1952). Such a difference may be the result of seasonal pectin variations, although Knee (1973a) has reported that special rehydration treatment of freeze- dried alcohol insoluble solids may be required before satisfactory extraction or enzymatic degradation of pectin can be acheived. Thus, incomplete enzymatic degradation could also contribute to the lower levels of pectin obtained by the method used in the current study. Pectin varied significantly with variety, storage and enzyme treatment, with significant (p < 0.05) variety by storage and significant (p < 0.01) storage by enzyme treatment interactions also apparent. Despite the presence of interactions, juice from stored apples had significantly higher (p < 0.01) amounts of pectin than juice from fresh apples indicating that solubilization of pectin occurred during post-harvest storage as described by Hulme (1958). Juice from apples not treated with Irgazyme 100 was significantly higher (p < 0.01) in pectin than juice from enzyme-treated puree. Varietal effects were influenced by the interaction with storage effects. Juices from stored Spartan and Mcintosh apples were higher in pectin content than juice from Red Delicious apples, while in juice from fresh apples the Spartan cultivar had highest pectin levels followed by Mcintosh then Red Delicious. The existence of such differences ultimately affects the suitability of these cultivars for post-harvest storage and processing. The low levels of soluble pectin present in juice from Red Delicious apples indicates a reduced breakdown during storage and is consistent with the firmer post-storage  60. texture of this fruit as measured by a Magness-Taylor pressure tester (Mcintosh, 40.8 N; Spartan, 44.4 N; Red Delicious, 47.1 N). Turbidity measurements (Table HI) statistically confirmed the findings of the preliminary study described in Section I and also reflected differences in particle density observed microscopically in Section IV. Turbidity varied significandy with processing and enzyme treatment. Juice from blanched puree was significantly more (p < 0.05) turbid than juice from unblanched puree, while treatment with Irgazyme 100 significandy reduced (p < 0.01) turbidity regardless of variety, storage or processing treatment. Zeta potential, determined microelectrophoretically from the movement of suspended juice particles in an applied electric field, indicated that suspended particles in apple juice were negatively charged (Table IE), supporting the electrophoretic results of Yamasaki et al. (1967). The zeta potential of apple juice particulate did not significantly differ between treatments and averaged -9.3 mV, which is in general agreement with zeta potential measurements of suspended particles in orange juice (-7.1 mV atpH 3.4) reported by Mizrahi and Berk (1970). The Laser Zee Model 501 Operating and Service Manual indicates that particles with zeta potential measurements in the range of -6 to - 30 mV form unstable suspensions. However, as seen by storage tests (Section II, Figs. 5-8) suspensions of cloud material in apple juice were very stable, particularly when blanching was employed during processing. This suggests that the state of particle hydration is more important to cloud stability than the charge associated with the suspended particles (Mizrahi and Berk, 1970). In summary, blanching generally did not result in any major compositional changes in the soluble constituents of apple juice. Furthermore, Poll (1983) demonstrated that heating apple juice for up to 20 min at 96 C did not result in reduction of fruit aroma or development of cooked aroma as detected by a trained sensory panel. Other than colour differences, the most significant difference between natural juice and conventional oxidized apple juice was the increase in turbidity as a result of blanching. Greater differences in soluble constituents were observed as a result of cultivar, storage and enzyme treatments. Juice from Mcintosh apples was highest in acidity and lowest in sugar content,  61. while juice from Red Delicious apples was lowest in acidity and highest in sugar content. The degradative processes occurring as a result of post-harvest storage were evident in chemical changes such as the decrease in organic acids and the increase in sugar and soluble pectin. Treatment with Irgazyme 100 caused significant increases (p < 0.01) in soluble solids, glucose and galacturonic acid, with a concomitant decrease in pectin and turbidity.  (b). COMPOSITION OF CENTRIFUGED CLOUD MATERIAL FROM APPLE JUICE The dry matter weight, protein and pectin composition of sediment removed by centrifugation from apple juice is shown in Table IV. Dry matter weight averaged 0.019% (w/v) which was lower than yields obtained by Yamasaki et al. (1964) and Carpenter and Walsh (1932, Table II). Differences probably result from differences in the authors definition of 'cloud material'. Dry matter weight varied significandy with processing and enzyme treatment. Juice from blanched puree had significantly larger (p < 0.05) amounts of cloud material based on dry matter weight than juice from unblanched puree, indicating that the increase in turbidity (Section Ilia, p. 60) as a result of blanching is due to an increase in the amount of cloud material suspended in the juice. Larger amounts of cloud material presumably arise from the increased degradation and solubilization of cloud forming components caused by the high temperatures applied during blanching. Juice from puree not treated with enzyme also had significandy larger (p < 0.01) amounts of dry matter than juice from enzymetreated puree, again supporting turbidity measurements. Studies by Yamasaki et al. (1964) and Carpenter and Walsh (1932) showed composition of centrifuged material from apple juice to consist mainly of pectic material (23-33%) and protein (19-37%). These constituents are derived primarily from fragments of cell wall tissue, although organelles may also contribute to protein content (Klavons and Bennett, 1985). In the current study, the amount of pectic material did not vary significandy with treatments and averaged 13% (w/w) of the dry matter (Table IV). Carpenter and Walsh (1932) found  62.  Table IV. Composition and yield of cloud material from apple juice as affected by cultivar, post-harvest storage, blanching and treatment with Irgazyme 100. Component Dry Matter  1  Significant Treatment Effects i)Processing*  Blanched  Unblanched  0.036 ii)Enzyme Treatment**  0.018  S E M = 0.005  No Enzyme Enzyme 0.041  0.013  S E M = 0.005  Pectic Material Protein (%)'  NSD i)Storage*  Fresh 30.5  i)Processing*  iii)Enzyme Treatment**  13.3  S E M = 3.3  Stored 22.8  S E M = 2.3  Blanched Unblanched 30.9 22.4  S E M = 2.3  No enzyme Enzyme 34.0 19.3  S E M = 2.3  iv)Processing by Blanched Unblanched Enzyme Treatment No enzyme 33.0 35.0 Interaction** Enzyme 28.8 9.8 SEM = 3.3 1  Expressed as percentage (weight/volume) in original apple juice. Values were adjusted to correct for dilution due to sulfite addition during processing.  2  Expressed as percentage (weight/weight) of dry matter.  S E M = standard error of the mean. NSD = no significant difference. * Means significant at p <0.05. " M e a n s significant at p < 0.01.  63. much higher levels of pectic material, but differences could result from differences in the definition of 'cloud material' as well as in methods used for pectin analysis. The calcium pectate method used by the other investigators for determination of pectin has been reported to give 10% higher levels for pectic material than other methods (McComb and McCready, 1952). Also, methods for analysis of cloud material are often not specific and may yield unreliable results largely because of iriteractions betweeen polyphenols, polypeptides and/or polysaccharides (Dadic and Belleau, 1980). In this respect, the method used in the current study is more specific since analysis of pectin is based on the HPLC measurement of galacturonic acid released during enzymatic digestion of alcohol insoluble solids. Protein levels (Table IV) on the other hand, are in general agreement with those of both Yamasaki et al. (1964) and Carpenter and Walsh (1932). Protein varied significandy with storage, processing and enzyme treatment, although a significant (p < 0.01) processing by enzyme treatment interaction was present. As a result of this interaction, protein levels were higher in juice from blanched puree only when Irgazyme 100 was used. When Irgazyme 100 was not employed, there was no difference between protein levels in juice from blanched and unblanched puree. The interaction did not alter the interpretation of enzyme treatment main effects. Juice from puree not treated with Irgazyme 100 had significandy higher (p < 0.01) levels of protein than that treated with Irgazyme 100. This suggests that enzyme treatment results in the precipitation of protein constituents along with the pulp removed in the initial stage of juice preparation. Cloud from the juice of fresh apples also had significantly higher (p < 0.05) protein levels than that from stored apples, suggesting that changes during storage may accelerate the metabolic utilization of protein, prevent release of protein during juice extraction or result in the removal of protein constituents with pulp in the initial stage of juice preparation. It appears that some of the differences in turbidity and yield of cloud based on dry material weights can be accounted for by differences in protein levels. The presence of protein and pectin alone, as determined by the present study, accounts for 23-48% of the composition of cloud material. However, a large per-  64. centage still remains unidentified. Other constituents probably include cell wall components such as cellulose, hemicellulose and lignin, starch, sugars and inorganic constituents such as metals. Generally, the degree of turbidity appears to be associated with differences in the amount of cloud material suspended in apple juice and variations in the amount of this material seem to be related to differences in the amount of protein remaining in the juice after processing, although other components may also be involved. Investigations by Yamasaki et al. (1964; 1967) and by Klavons and Bennett (1985; 1987) strongly suggest that a physical association exists between cloud pectin and cloud protein. Based on enzymatic and electrophoretic studies of cloud material in apple juice, the former proposed a protein-carbohydrate complex surrounded by a negatively charged protective colloid such as pectin. Klavons and Bennet (1987) however, suggest that in lemon juice approximately 50- 67% of the pectin is entrapped within a protein matrix, with the rest of the pectin remaining inherendy insoluble and distinct, aggregating through interchain and intrachain hydrogen bonding and not direcdy associating with other cloud constituents. Cloud in unclarified apple juice is most likely present as a protein-pectin complex, however further information is required to fully understand the effect of treatments such as post-harvest storage, blanching and use of enzymes on the nature of this association.  (c). COMPOSITION OF APPLE JUICE AFTER REMOVAL OF CLOUD MATERIAL The liquid remaining after removal of cloud material by centrifugation was defined earlier as serum (Methods and Materials, V. Chemical Analyses, p. 33). Analysis of serum indicated that removal of cloud material also affected the composition of soluble constituents (Table V) with significant decreases (p < 0.01) in pH,titratableacidity and total acidity mainly as a result of significant losses (p < 0.05) in galacturonic acid and (p < 0.01) in malic acid. Such losses combined with a significandy decreased (p < 0.01) soluble solids due to significantly lower (p < 0.05) glucose levels would be expected to have a deleterious ef-  65. Table V. Comparison of juice composition before removal of cloud marterial with serum composition after removal of cloud material . 1  Component  Juice Mean SEM  Serum Mean SEM  Soluble Solids (Brix)**  12.83  0.22  12.48  0.24  Total Sugars (g/100 mL; NSD) Sucrose (g/100 mL; NSD) Glucose (g/100mL)* Fructose (g/100 mL; NSD) Sorbitol (g/100 mL; NSD)  10.84 1.00 2.70 6.73 0.35  0.22 0.16 0.15 0.12 0.03  10.92 1.07 2.46 6.89 0.37  0.25 0.25 0.15 0.12 0.03  3.70  0.11  3.62  0.11  0.455  0.032  0.450  0.031  Total Acidity (g/100 mL)** Citric (mg/100 mL) Galacturonic (mg/100 mL)* Malic (mg/100 mL)** Quinic (mg/100 mL) Succinic (mg/100 mL)  1.115 45 510 582 111 122  0.068 3 21 43 6 16  1.012 59 446 453 106 154  0.052 2 16 28 8 29  Pectin as Anhydrogalacturonic acid (mg/100 mL)**  53.7  pH** Titratable Acidity (g malic/100 mL; NSD)  2  2  2  10.3  38.6  10.7  Significant differences determined by a paired comparison T-test. Values were adjusted to correct for dilution due to sulfite addition during processing. 'Due to the high variability inherent in quantitating components at low concentrations, statistical evaluation of citric, quinic and succinic acids was not considered valid and was therefore not performed. SEM = standard error of the mean; NSD = no significant difference. *Means significant at p < 0.05. * * Means significant at p < 0.01.  66. feet on flavour similar to the flavour losses that are experienced in commercially clarified apple juice products (Asti, 1970). Pectin levels were significantly reduced (p < 0.01) after removal of cloud material, most likely causing a concommitant reduction in viscosity. Soluble constituents in the serum differed from those in the juice. Proportionately greater losses of galacturonic acid, malic acid, pectin and glucose (as starch) suggests that an interaction, perhaps in the form of hydrogen bonds, may exist between these constituents and components of cloud material. The slight increasein the levels of citric and succinic acids observed after removal of cloud material is presumably an artifact of the increased variability associated with the quantification of these components which are present at low concentrations.  IV. ANALYSIS BY TRANSMISSION ELECTRON MICROSCOPY  (a). THIN SECTIONING Thin sections from pellets of cloud material from oxidized juice cloud of fresh Spartan apples exhibited a high concentration of particles with a variety of structural characteristics and varying affinities for electron dense stains (Fig. 10). The structures were categorized as granules (g), spheres (s), aggregates (a) and vesicles (v) as tabulated in Table VI. Particle classification was complicated by the possible introduction of sectioning artifacts which could affect particle appearance and distribution (Hayat, 1981).  Large vesicles with electron  dense membranes were the most prevelant structures in sections of oxidized juice cloud. Smaller structures such as spheres and granules were found not only individually, but also attached to the surface and within the the interior of the vesicles. Spheres and granules also appeared to combine to form larger electron dense aggregates. The varying affinity of vesicles, aggregates, sheres and granules for the electron dense stains was either an artifact of sectioning or indicated these structures were compositionally different.  Since structures maintained these  67. relative stain densities from section to section and from block to block, the differences were considered to be primarily compositional. The more electron dense aggregates and vesicular membranes may have contained a greater number of exposed heavy metal binding sites than the less densely stained spheres and granules. Of the stains used, osmium reacts with proteins, lipids and membranes whereas lead reacts with hydroxyl groups of carbohydrates and sulfhydryl groups of proteins, and uranium is bound by carboxyl and phosphoryl groups (Hayat, 1972; 1981). Studies have also shown that 2% uranyl acetate followed by lead citrate effectively stains cellulose (Hayat, 1981). Based on chemical analyses of cloud material (Table IV, p. 62), the most probable binding sites in the juice cloud would be hydroxyl and sulfhydryl groups of proteins from cell walls and organelles, with hydroxyl groups of polyanionic carbohydrates such as pectin from the middle lamella accounting for a smaller number of the binding sites. Although the formation of aggregates could be an artifact of the dehydration procedure, the presence of heavy metal binding sites suggests aggregation could also be caused by hydrogen bonding between exposed groups on the surface of spheres and granules. Thin sections of cloud material from natural juicefromthe blanched puree offreshSpartan apples (Fig. 11) had fewer particles with less variety of structural characteristics and a more uniform affinity for the electron dense stains than structures of oxidized juice (Fig. 10 vs 11). Vesicles were not found in cloud from natural juice and a comparison of other structures presented in Table VI indicated that aggregates found in natural juice were significantly smaller (p < 0.05) and less distinct than those in oxidized juice. Blanching appeared to cause the disintegration of aggregates observed in oxidized juice to form relatively smaller structures and a more electron dense background (Fig. 10 vs 11).  (b). NEGATIVE STAINING The dimensions of particlesfromnegatively stained juice samples (Figs. 12 - 19) are presented in Table VII. The structures were again categorized as granules (g), spheres (s) and aggregates (a) based on dimensions and stain den-  68. Table VI. Characterization of particles in thin section of cloud material from the juice of fresh Spartan apples. SAMPLE  PARTICLE  DIMENSIONS (nm) RANGE  STAIN DENSITY 1  AVERAGE  Oxidized  granule  18.4-  31.8  25.1 diam.  slight-moderate  juice cloud  sphere  56.1 -280.5  152.3 diam.  slight-moderate  aggregate 70.4- 678.8 35.2- 274.9 vesicle  95.5- 410.0  2  273.3 diam.  extreme  5.7 -  39.2  20.5 wall  36.9  22.5 diam. slight-moderate  Natural  granule  15.1 -  juice cloud  sphere  50.2 - 335.0  aggregate  12.2-342.2 50.5-157.1  1  366.9 length moderate156.9 width extreme  116.8 diam. moderate-heavy 179.8 length 76.5 width  2  moderate-heavy  Extreme=black ; slight=just discernable over background. By a n a l y s i s of varianceCANOVAD,aggregate particles in cloud from  oxidized juice were significantly longer (p < 0.05) than similar particles in cloud from natural juice.  69.  10. Thin section of pellets of cloud material from oxidized juice of fresh Spartan apples. Aggregate (a); sphere (s); vesicle (v).  11. Thin section of pellets of cloud material from natural juice of fresh Spartan apples. Aggregate (a); sphere (s).  11  »  »  *  •  -  S  5 0 0  n m  71. sity; no vesicular structures were present. Granules, spheres and aggregates on the other hand were observed in each preparation of juice, although the relative proportions of these particles varied dramatically between preparations (Fig. 12 vs 13, 14 vs 15). Statistical analyses showed particle dimensions did not differ between juice preparations with the exception of spheres, which were significandy larger (p < 0.05) in juice from blanched puree than spheres in juice from unblanched puree (Table VII). Examination of Figs. 12 -19, revealed several trends. Generally, there were only slight differences between cultivars with respect to particle dimensions and particle proportions, with more dramatic differences resulting from heat (Fig. 12 vs 13), storage (Fig. 13 vs 15) and treatment with Irgazyme 100 (Fig. 12 vs 16). Natural juice from blanched puree had a larger number of particles than similarly prepared samples of oxidized juice, results that are consistent with increased turbidity and increased yield of cloud material described earlier (Table in, p. 51-55; Table IV, p. 62). Blanching not only altered the amount of cloud material present, but also the nature of the particles. Natural juice from blanched puree not treated with enzyme had a larger number of smaller particles in the form of granules and spheres, whereas similarly prepared samples of oxidized juice consisted of particles mainly in the form of aggregates (Figs. 12 vs 13). The aggregates appeared to be an agglomeration of spheres and granules and could be an artifact of drying. However, aggregates were formed mainly in oxidized juice not treated with enzyme and enzyme-treated natural juice (Fig. 12,17), but not in similarly prepared samples of natural juice not treated with enzyme and enzyme-treated oxidized juice (Fig. 13,16). Since all grids were prepared in the same manner, the formation of aggregates in certain juice preparations and not in others suggests forces other than surface tension effects were involved in particle aggregation. Compositional differences also existed between the particles forming aggregates as indicated by the variation in stain density of the components within aggregates (Fig. 12). Although negative staining was carried out, it appeared that some particles were negatively stained while other particles were stained positively. This in combination with results of chemical analyses (Table IV) and  72.  Table VII. The effect of cultivar, post-harvest storage, blanching and treatment with Irgazyme 100 on particle dimensions of negatively stained and shadow cast preparations of apple juice. Component  Significant Treatment Effects  Negative Staining granule  NSD  1  sphere'  i)Processing*  15.9  SEM= 3.3  Blanched Unblanched 92.1 65.7 S E M = 9.1  aggregate length  NSD  365.2  S E M =37.0  aggregate width  NSD  135.2  S E M =10.3  3  Shadow Casting granule 1  granule height  i) Storage by Enzyme Treatment Interaction*  i) Variety*  ii)Condition*  iii)Processing*  iv)Variety by Processing Interaction*'  Fresh Stored No enzyme 24.6 29.1 Enzyme 21.0 22.9 S E M = 1.7 Red Delicious Mcintosh Spartan 18.0 10.3 9.2 S E M = 0.5 Stored 13.4  Fresh 11.6  S E M = 0.4  Unblanched Blanched 16.2 8.8 S E M = 0.4 Unblanched Blanched Red Delicious Mcintosh Spartan  27.0 10.3 11-3  9.0 10.3 7.0  73. Table VII (continued). Component  Significant Treatment Effects i)Processing*  Blanched Unblanched 112.2 1.5 S E M =7.8  i)Processing**  Blanched Unblanched 55.8 39.1 SEM=3.9  aggregate length  i)Processing*  Unblanched Blanched 497.8 310.6 S E M =52.9  aggregate width  i)Processing by Enzyme Treatment Interaction*  sphere  Shadow cast sphere height  3  aggregate height  i)Processing by Enzyme Treatment Interaction*  Unblanched Blanched No enzyme 197.2 100.5 Enzyme 108.3 120.9 S E M = 21.9 Unblanched Blanched No enzyme 65.0 43.7 Enzyme 38.6 52.3 S E M = 5.8  electron density = slight (just discernable over background), electron density = slight to moderate, with an extremely dense envelope, ^electron density = slight to extreme, ^electron density = moderate to extreme NSD = no significant difference. 1  S E M = standard error of the mean. *means significantly different at p < 0.05. ** means significantly different at p < 0.01.  74.  12. Negatively stained particles from oxidized juice of fresh Spartan apples not treated with enzyme. Aggregate (a); sphere (s). 13. Negatively stained particles from natural juice of fresh Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s). 14. Negatively stained particles from oxidized juice of stored Mcintosh apples not treated with enzyme. Aggregate (a); granule (g); sphere (s). 15. Negatively stained particles from natural juice of stored Red Delicious apples not treated with enzyme. Aggregate (a); granule (g); sphere (s).  76. work by Klavons and Bennett (1985; 1987) on lemon juice, suggests that these aggregates may be a complex of pectin and protein in which the pectin is entrapped within a protein matrix, being released only as a result of conformational changes in the protein. Blanching appears to alter the nature of this proposed protein-pectin relationship through heat induced conf omational changes in protein structure which cause particles in natural juice to be slight or moderately stained structures surrounded by a diffuse layer of densely stained material (Fig. 13,15,17,19). The presence of a negatively charged, heavy metalattracting envelope of pectin would stabilize the cloud! particles in unclarified natural juice by preventing the aggregation of particles through charge repulsion, thus accounting for the increased stability of juice from blanched puree during storage (Section Ha, p. 41-49). This also gives further support to Yamasaki et al. (1964) who, through enzymatic investigation of apple juice clarification, proposed a similar structure for apple' juice particulate in which a protein-carbohydrate complex was surrounded by a negatively charged protective colloid such as pectin. Storage produced a cloudy oxidized juice in which a small number of particles were embedded in a densely stained web-like matrix (Fig. 14). The particles also appeared to become less compact and less distinct than their counterparts in the fresh juice (Fig. 12 vs 14) probably as a result of the gradual disintegration of cell structure during storage (Hulme, 1958; Hulme and Rhodes, 1971). The affinity of the web-like material for the uranyl acetate stain provided further evidence for the presence of uranium binding sites with potential for hydrogen bonding. The formation of a web-like matrix suggests that structural changes occur during storage which alter the nature of protein-pectin complex described above. Such changes combined with the increase in soluble pectin during storage probably promote pectin-pectin interactions and may ultimately lead to gel formation as observed during juice storage (Section Ha). Similar micrographs were obtained by Lewis (1981) from uranyl acetate stained thin sections of pectin gels. The existence of a gel-like network which allowed for entrapment and imbibition of water, might account for the low yields of juice obtained when pressing stored fruit (Powrie and Tung, 1976; Glunk, 1981).  77.  16. Negatively stained particles from oxidized juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); sphere (s). 17. Negatively stained particles from natural juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s). 18. Negatively stained particles from oxidized juice of stored Spartan apples treated with enzyme. Aggregate (a); sphere (s). 19. Negatively stained particles from natural juice of stored Mcintosh apples treated with enzyme. Aggregate (a); granule (g); sphere (s).  1 6  30  1 7  46  a *  © 5 00  nm  1 8  nm  5 00  1  1 9 ,  4fi  |  I  o  ©  4|  e 9  5 00  nm  500  nm  79. Treatment with either heat during blanching or treatment with Irgazyme 100 appeared to greatly reduce or eliminate the formation of a web-like matrix (Fig. 14 vs 15,14 vs 18). The possibility therefore exists for improving juice yields by employing either a blanching step (Beveridge and Harrison, 1986) or enzyme treatment in the processing of stored apples. In natural juice with no enzyme treatment, storage resulted in an increase in the number of spheres and granules (Fig. 13 vs 15), possibly accounting for the increased turbidity described in Section I (p. 40-42). The increase in the number of small particles was evidence for the destructive processes occurring during storage. Results of turbidity analyses (Fig. 4; p. 41) showed juice from stored apples clarified more readily upon treatment with Irgazyme 100 than juice from fresh apples. Microscopical analyses suggest this difference may be due to degradation during storage of components involved in particulate formation. The treatment of apple puree with Irgazyme 100 resulted in a decrease in the number of particles present in the juice (Fig. 12vs 16,14 vs 18). The majority of particles present in enzyme-treated oxidized juice were small spheres or granules most likely formed as a result of the disintegration of larger aggregates (Fig. 16, 18). In contrast, treatment of natural juice with enzyme resulted mainly in the formation of aggregates (Fig. 17, 19). Aggregate formation in natural juice after treatment with enzyme may have resulted from the enzymatic degradation of the intensely stained material surrounding the particles observed in the juice from untreated puree (Fig. 13 vs 17). Degradation of the enveloping material would reduce interparticle repulsion. The resulting interaction of particles would lead to the observed formation of larger numbers of aggregates than in the untreated juice. Enzyme treatment of natural juice seemed to follow two stages of degradation in the process of clarification, thus accounting for the longer incubation times or larger amounts of enzyme required to clarify juice from blanched puree (Beveridge et al., 1986). In oxidized juice, the enzymes appeared to direcdy reduce particle size and number (Fig. 12 vs 16). However, in natural juice the enzymes first reduced the thickness of the enveloping material, followed by the formation of numerous aggregates (Fig. 17,19), before a decrease in particle number and degradation of aggregates became apparent.  80. This supports Yamasaki et al. (1967) who suggest that enzymatic clarification of apple juice consists of two stages involving initial degradation of the protective colloid (pectin), followed by electrostatic neutralization resulting from the attraction between exposed positively charged protein from the interior of the complex with negatively charged pectin not yet enzymatically degraded on the particle surface. Enzyme treatment of stored apples appeared to be more effective for reducing cloud than enzyme treatment of fresh apples. Juice from stored apples treated with Irgazyme 100 generally had fewer particles than similar preparations of juice from fresh apples (Fig. 19 vs 17).  (c). SHADOW CASTING The dimensions of particles from shadow cast juice samples are presented in Table VII. Since shadow casting provides a three- dimensional view of the particles, the additional measurement of height was included in the discussion of particle dimensions. Particle height (H) was calculated from shadow length (L) and shadow angle as described in equation 1: H = L x tan 23.6 x magnification factor  (1).  The structures were again categorized as granules (g), spheres (s) and aggregates (a) based on dimensions and electron density, no vesicular structures were observed. Since vesicles were found only in thin sections of oxidized juice, it is likely that these structures were formed either as artifacts of centrifugation and alcohol dehydration or were cross-sections of large spheres. As in negative staining, each type of particle was observed in each preparation of juice, but the relative proportions of the particles varied substantially between preparations (Fig. 20 vs 21,21 vs 23). Granule diameter was affected by a significant (p < 0.05) storage by enzyme treatment interaction, while granule height varied significandy with variety, storage and processing treatment, with a significant (p < 0.01) variety by processing interaction (Table VII). Sphere diameter and height varied significantly (p < 0.01) with processing treatment. Aggregate length also differed significantly (p <0.05) between  81.  20. Shadow cast particles from oxidized juice of fresh Spartan apples not treated with enzyme. Aggregate (a); sphere (s). 21. Shadow cast particles from natural juice of fresh Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s). 22. Shadow cast particles from oxidized juice of stored Spartan apples not treated with enzyme. Aggregate (a); granule (g); sphere (s). 23. Shadow cast particles from natural juice of stored RedDelicious apples not treated with enzyme. Aggregate (a); granule (g); sphere (s).  83.  24. Shadow cast particles from oxidized juice of fresh Spartan apples treated with enzyme. Aggregate (a); granule (g); sphere (s). 25. Shadow cast particles from natural juice of fresh Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s). 26. Shadow cast particles from oxidized juice of stored Red Delicious apples treated with enzyme. Aggregate (a); granule (g); sphere (s). 27. Shadow cast particles from natural juice of stored Mcintosh apples treated with enzyme. Aggregate (a); granule (g); sphere (s).  85. processing treatments, while aggregate width and height were affected by a significant (p < 0.05) processing by enzyme treatment interaction. Examination of Figs. 20 - 27, revealed several trends similar to those observed in negatively stained juice samples. Again, differences between cultivars were negligible (Table VII), more striking differences resulted from blanching (Fig. 20 vs 21) and treatment with Irgazyme 100 (Fig. 20 vs 24). As observed in negative staining, natural juice from blanched puree not treated with enzyme had a larger number of smaller particles in the form of granules and spheres, whereas similar samples of oxidized juice had fewer particles mainly in the form of aggregates (Fig. 20 vs 21). Spheres present in natural juice from blanched puree were significantly larger (p < 0.01) than spheres in oxidized juice, while granules and aggregates were significandy larger in oxidized juice (p < 0.05) than similar structures in natural juice. Once again, the aggregates appeared to be agglomerations of spheres and granules. The formation of aggregated particles could be an artifact of drying, although surface tension effects were expected to be minimized through use of a spray gun for sample application (Hay at, 1972). Aggregates were formed mainly in oxidized juice not treated with enzyme (Fig. 20) and only to a lesser extent in other juice preparations. Since all grids were prepared in the same manner, the formation of aggregates in certain juice preparations and not in others once more suggests forces other than surface tension effects were involved in particle aggregation. Storage of apples increased the amount of particulate present in natural juice (Fig. 21 vs 23, 25 vs 27) and increased the amount of smaller particulate in oxidized juice (Fig. 20 vs 22). Granules present in juice from stored apples were significandy larger (p < 0.01) than those in juice from fresh apples. The web-like material observed in negative staining of oxidized juice from stored apples (Fig. 14) was not as apparent with shadow casting, indicating this material either had very little physical structure for throwing a shadow or the web-like structure was disrupted when the sample was sprayed onto the grid with a high velocity spray gun. Treatment of the apple puree with Irgazyme 100 resulted in a decrease in the number of particles present in the juice (Fig. 20 vs 24,22 vs 26,21 vs 25,  86. 23 vs 27). As in negative staining, shadow casting revealed treatment of natural juice with enzyme caused the formation of a greater number of aggregates than in puree not treated with enzyme (Fig. 21 vs 25,23 vs 27), while enzyme treatment of oxidized juice gready reduced particle size and number (Fig. 20 vs 24, 22 vs 26).  (d). CONCLUSIONS Particle dimensions as measured in thin sectioned, negatively stained and shadow cast preparations of apple juice showed no significant difference with respect to aggregate dimensions (p < 0.05). However, the dimensions of spheres in thin sections were significandy larger (p < 0.05) than similar particles in negatively stained and shadow cast preparations (Table VI vs Table VTJ), while in negatively stained preparations granules were significantly smaller (p < 0.05) than those in shadow cast or thin sectioned preparations. The fixation procedure employed for thin sectioning is more severe with a greater potential for introduction of artifacts and distortion of particles, possibly accounting for the differences in particle dimensions observed by this and the other two techniques. Dimensions of particles in shadow cast samples could also be distorted by the deposition of a layer of metal on the surface of the particles which was probably thick enough to cause the statistical differences observed between shadow cast and negatively stained preparations. Although statistical differences were noted, overall the appearance (Fig. 10 vs 12 vs 20) and dimensions of the particles (Table VI vs Table VTJ) observed by these three different techniques were very similar. Of the techniques examined, thin sectioning was very time-consuming and therefore impracticle to use with a large number of samples. Shadow casting provided a unique three-dimensional view of the particles, but again was relatively time-consuming. Negative staining on the other hand was simple and rapid, supplying substantial structural information as well as compositional information based on particle-stain interactions.  87.  V. STEPWISE DISCRIMINANT ANALYSIS The introduction of blanching and enzyme treatment offers the possibility of producing four different types of apple juice based on four different processing treatments: 1. cloudy oxidized juice - produced from unblanched puree not treated with enzyme. 2. cloudy 'natural' juice  - produced from blanched puree not treated with enzyme.  3. clarified oxidized juice - produced from enzyme-treated, unblanched puree. 4. clarified 'natural' juice - produced from enzyme-treated, blanched puree. In order to verify the existence of four such distinct categories, pattern recognition or multivariate analysis was used to examine the simultaneous relationship between selected quality parameters. Pattern recognition techniques have been applied very succesfully to the classification of other foods such as cheese, wines and soya sauce (Pham and Nakai, 1984; Kwan and Kowalski, 1980; Aishima, 1978). One of the most widely used of the multivariate techniques is linear discriminant analysis. The objective of discriminant analysis is to find a discriminant function that is a linear combination of predictor variables which maximally correlates with a linear combination of criterion measures (Dillon and Goldstein, 1984). Discriminant functions are generated to maximize the ratio of between-groups to within-groups variability and each new discriminant axis is uncorrelated to any previously obtained axis. Once determined, the discriminant function can be used to assign unknown samples to one of the groupings defined by the analysis. The unkown is assigned to that group which gives the largest discriminant score as determined by equation 2: Z = aixi + a2X2 + ...+a xn n  (2)  88. where ai...a are coefficients of the discriminant function, xi...x are quantitan  n  tive variables measured for the unknown and Z, the discriminant function, is calculated to maximize the ratio of between-groups variation to within-groups variation. The classification of the groups is represented graphically in two dimensional space by a plot of the canonical variables. Thefirstcanonical variable (x-axis) is defined by the linear combination of the maximal multiple correlation coefficients, the canonical coefficients. The second canonical variable (y-axis) is obtained by finding the linear combination uncorrelated with the first canonical variable that has the highest possible correlation with the groups (Amantea, 1984; SAS, 1986). Each sample can be characterized by a pair of canonical variables. Thus, while the discriminant function allows an unknown sample to be categorized into a given group, the canonical variables allow visualization of the position of the new sample relative to the rest of the observations in all the groups (Pham and Nakai, 1984). The data were assigned a priori into four groups determined according to processing: 1. unblanched and not enzyme-treated (oxidized). 2. blanched and not enzyme-treated (natural). 3. unblanched and enzyme-treated (clarified oxidized). 4. blanched and enzyme-treated (clarified natural). The qualitative variables subjected to the stepwise discriminant analysis were: turbidity, colour ratio, titratable acidity, total acidity, total sugars, pectin content and zeta potential. With the significance level to enter and to stay at 0.15, turbidity, colour ratio and total acidity were selected as the variables providing the greatest discriminatory power. The data from these three variables was then used to obtain the discriminant functions for each of the four groups and the canonical plot shown infigure29. Four distinct groups were found with only a few points (circled) overlapping between the groups. Upon closer examination of the data, the overlapping points were found all to belong to juice from fresh Mcintosh apples which had been over-sulfited during processing. Removal of this data resulted in the canonical plot infigure30. As seen infigure30 and as determined by the discriminant functions, removal of these data increased the  89.  • • A •  Oxidized Clarified Oxidized Natural Clarified Natural  CO  w H J  0  1  < < >  <  0  o  Z  g-i o  -2 —  -3  -4  T  -3  -  2  - 1 0 1 CANONICAL VARIABLE 1  Figure 28. Canonical plot of 24 apple juice samples grouped according to processing treatment. Circled data removed due to overuse of sulfite during processing.  90.  Enzyme Treated  Not Enzyme Treated  T3  CD Xi O  a  3 G  •  a;  xi o  fl  H • A •  r—I  OQ  T  -5  -4  -3  Oxidized Clarified Oxidized Natural Clarified Natural T  T  • 2 - 1 0 1 CANONICAL VARIABLE 1  Figure 2 9. Canonical plot of the same apple juice samples after elimination of the data from fresh Mcintosh apples. Juices were categorized according to processing treatment.  91.  generalized distance between the groups and decreased the number of misclassified data. The plot in figure 30 shows the presence of four distinct groups which can be described as either blanched or unblanched and enzymed or unenzymed. Clearly each combination of processing conditions produces a unique type of apple juice based on measurement of turbidity, colour ratio and total acidity. Similar groupings were obtained when total acidity data was replaced with galacturonic acid data, suggesting that galacturonic acid was the major organic acid contributing to the discrimination obtained with total acidity measurments. With removal of the fresh Mcintosh data, the discriminant functions for each group of juice were: oxidized:  0.0026T + 0.5299C + 3.6138TA - 13.9881  natural:  0.0091T - 0.8319C + 4.0905TA -15.8157  clarified oxidized: -0.0032T - 1.0772C + 8.5774TA - 39.6666 clarified natural: -0.0007T - 1.5785C + 8.8851TA - 40.8426 where T is turbidity, C is the colour ratio and TA is the total acidity. Using the above discriminant functions and the hold out method to test the classification, the percent of correct classification was improved from 79.2% to 91.7% when fresh Mcintosh data was removed. The sulfite level used in the initial preparation of fresh Mcintosh puree was relatively more extreme than the treatments used in processing the other batches of puree. The juice resulting from too high a sulfite treatment showed litde difference between unblanched and blanched groups since the presence of excess sulfite prevented oxidation in the unblanched samples. Closer attention to sulfite levels during processing and additional discrimination using parameters such as phenol content could improve the separation of the groups as well as improving the percentage of correct classification.  92.  CONCLUSIONS Blanching had litde effect on juice composition, however turbidity was significandy greater (p < 0.05) in juice from blanched puree. Based on analysis of centrifuged cloud material, increased turbidity was due to increased amounts of proteinaceous cloud material suspended in the juice. Microscopic analysis also suggests that degradation of large aggregate particles similar to those observed in oxidized juice results in the smaller, more numerous spheres and granules observed in natural juice from blanched puree, further contributing to the increased turbidity of this juice. Despite the increase in cloud material, storage of juice at 0 C and 24 C for 84 days demonstrated that the suspension in juice from blanched puree was more stable than the suspension in juice from unblanched puree. Microscopic and chemical analyses indicate that particulate in juice from unblanched puree is most likely a pectin-protein complex with pectin entrapped in a matrix of protein. Blanching appears to stabilize the particulate through heat induced conformational changes, producing particles with a central core of protein surrounded by a protective, negatively-charged coating of pectin which prevents aggregation through electrostatic repulsion. Although juice composition was basically unaltered by blanching, it was influenced by cultivar, post-harvest storage and treatment with Irgazyme 100. Of the varieties examined, juice from Mcintosh apples had highest organic acid content and lowest levels of sugar, while juice from Red Delicious apples was highest in sugar and lowest in organic acid content. Juice from Red Delicious apples also had lowest levels of pectin. No varietal differences were observed in either chemical or microscopic analysis of cloud material. Degradative processes during apple storage produced juice with decreased levels of organic acids and increased levels of soluble solids, glucose, sorbitol and pectin. Storage also caused a decrease in the protein content of apple juice cloud particulate. Microscopic analyses indicate that storage alters the nature of the protein-pectin complex of cloud material, giving a web-like aspect to the particulate in negatively stained preparations. Such structural changes combined with increased levels of soluble pectin probably contributed to the gel formation which occurred  93. during storage at 0 C in juice from stored apples. The web-like aspect of the particulate was disrupted either by blanching or treatment with Irgazyme 100. Treatment with Irgazyme 100 also increased soluble solids due to significant increases (p < 0.01) in glucose and galcturonic acid with concomitant decreases in soluble juice pectin, turbidity, dry matter weight and protein content in cloud material. Microscopically, treatment of unblanched puree with Irgazyme 100 resulted in direct reduction of particle size and density, while in juice from blanched puree enzyme treatment appeared to require the initial degradation of the protective colloid before reduction in particulate was achieved. Further investigations are required in order to more fully elucidate the structure and composition of cloud material in apple juice. In particular, analysis for cellulose, starch and metal ions should be considered. Also, more work is re quired to relate compositional information to microscopic observations, perhaps by examining unprocessed appletissuewith techniques sismilar to those used in the examination ofjuice particulate. The effect of changes in pH and salt concentration may also provide more information concerning the structure and composition of particulate examined by transmission electron microscopy. Furthermore, phenolic and volatile composition of apple juice as affected by cultivar, post-harvest storage, blanching and enzyme treatment, as well as sensory evaluations should also be considered before detailed manufacturing practices can be outlined. However,fromresults of the current study, production of a cloudy 'natural' juice appears possiblefromeither fresh or stored apples. Stored apples produce a juice with more particulate, while juice from fresh apples is more resistant to enzyme clarification. Clarified natural juice could also be made by enzyme treatment of juice followed by filtration and centrifugation. Optimum quality with respect to stability of cloud and colour would be maintained by storage at refrigerated temperatures. The introduction of blanching and enzyme treatment offers the possibility of improved product floavour and appearance. Furthermore, it would allow the processor to produce several different products from a single raw material.  94.  REFERENCES 1. Aishima T. (1979). Objective Evaluation of Soy Sauce By Statistical Analysis of GC Profiles. Agricultural and Biological Chemistry 43: 1905. 2. Amantea GF. (1984). Assessment of Bacteriophage-Sensitive Culture Bacteria for Cheddar Cheese Making and Subsequent Discriminant Analysis for Objective Flavour Evaluation. PhD. Thesis. Dept. Food Science, University of British Columbia. Vancouver, British Columbia. 3. Asti AR. (1971). U.S. Patent No. 3,518,093 in "Fruit Processing". (M. Gutterson, ed.). Noyes Data Corporation, p. 87. 4. Atkinson FE, Strachan CC. (1949a). Natural Apple Juice. Food Industries of South Africa June: 37-39. 5. Atkinson FE, Strachan CC. (1949b). Production of Juices. Agriculture Canada Technical Bulletin No. 68. p. 16-32. 6. Bauernfeind JC. (1958). The Role of Ascorbic Acid in the Browning Phenomenon of Fruit Juice. F. Hoffmann-La Roche and Company, Ltd. Basil, Switzerland, p. 1-72. 7. Beruter J, Kalberer PP. (1983). The Uptake of Sorbitol by Apple Fruit Tissue. Z. Pflanzenphysiol. Bd. 110S: 113-125. 8. Beruter J. (1985). Sugar Accumulation and Changes in the Activities of Related Enzymes During Development of the Apple Fruit. J. Plant Physiol. 121:331-341. 9. Beveridge T, Harrison J. (1986). Pear Juice Production from Heated Pear Mashes. Can. Inst. Food Sci. Technol. J. 19(1): 12-16. 10. Beveridge T, Franz K, Harrison J. (1986a). Clarified Natural Apple Juice: Production and Storage Stability of Juice and Concentrate. J. Food Sci. 51(2): 411-414 & 433. 11. Beveridge T, Harrison JE, Bains D. (1986b). Pilot Scale Production and Composition of Juice From Heated Pear Mashes. Lebensm.-Wiss. u.Technol. 19:432-436. 12. Beveridge T, Harrison JE, McKenzie D-L. (1987a). Juice Extraction with the Decanter Centrifuge - A Reveiw. Accepted for Publication in Can. Inst. Food Sci. Technol. J. 13. Beveridge T, Harrison JE, Kitson J. (1987b). Pear Juice from Bardett Peels and Cores. Accepted for Publication in J. Food Sci. 14. Binnie R, Possman Ph. (1984). First Experience with Bellmer Winklepress in Fruit Juice Plants. Flssiges Obst. 51(1): 2-8,13.  15. Carpenter DC, Walsh WF. (1932). The Commercial Processing of Apple Juice. New York State Experimental Station Technical Bulletin No. 202. p. 3-21. 16. Chan WW, ChongC, Taper CD. (1972). Sorbitol and Other Carohydrate Variations During Growth and Cold Storage of Mcintosh Apple Fruits. Can. J. Plant Sci. 52:743-750. 17. Charalambous G. (1984). Analysis of Foods and Beverages: Modern Techniques. Academic Press. Orlando, p. 586. 18. Charley VLS. (1961). Production and Application of Enzyme Preparations in Food Manufacture. SCI Monograph No. 11. Soc. Chem. Ina. London, p. 107 in 'Pectic Substances in Fresh and Preserved Vegetables.'(1965). (J.J. Doesburg, ed.). I.B.V.T. Wageningen. p. 121. 19. ConnEE,StumpfPK. (1976). Oudines of Biochemistry. John Wiley and Sons, Inc. New York. p. 341. 20. Crane JC. (1969). The Role of Hormones in Fruit Set and Development. HortSci. 4(2): 108-111. 21. Cumming, DB. (1984). A General Overlook on Fruit Juice Winning Systems. Confructa Studien March/April: 132-146. 22. Cumming, DB. (1985). Latest Developments in Mechanical Presses. Flssiges Obst. 52(12): 651-653. 23. Cumming DB, Beveridge T, Gayton R. (1986). Manipulation and Control of Brown Pigmentation in Juices Prepared from Dessert Apples. Can. Inst Food Sci. Technol. J. 19(5): 223-226. 24. Dadic M. Belleau G. (1980). Beer Hazes I. Isolation and Preliminary Analysis of Phenolic and Carbohydrate Components. Am. Soc. Brew. Chem., Inc. 38(4): 154-158. 25. Darvill M, MacNeil M, Albersheim P, Delmer DP. (1980). The Primary Cell Walls of Flowering Plants in "The Biochemistry of Plants - A Comprehensive Treatise. Vol.1. The Plant Cell." (N.E. Tolbert, ed.). Academic Press. New York. p. 137-151. 26. Dillon WR, Goldstein M. (1984). Multivariate Analysis: Methods and Applications. John Wiley and Sons. New York. p. 560-364. 27. Dorreich K. (1983). Total Liquefaction of Apples. Flssiges Obst. 50(7): 308-316. 28. Duymovic AA. (1975). Processed Apple Utilization Trends - Outlook and Competition on the Domestic Markets in "Third National Apple Conference". (D.L. Downing and R.L. LaBelle, eds.). Geneva Proceedings, p. 14-15.  96. 29. Embs RJ, Markakis P. (1965). The Mechanism of Sulfite Inhibition Browning Caused by Polyphenol Oxidase. J. Food Sci. 30: 753-758. 30. Fidler JG, Wilkinson BG, Edney KL, Sharpies RO. (1973). The Biology of Apple and Pear Storage. Commonwealth Agricultural Bureaux. Farnnam Royal. England, p. 3-35. 31. Frenkel C, Klein I, Dilley DR. (1968). Protein Synthesis in Relation to Ripening of Pome Fruits. Plant Physiol. 43:1146-1153. 32. Fruit and Vegetable Production. (1985). Minister of Supply and Services Canada. Catalogue 22-003, Vol 54, No. 5, p. 10-13. 33. Glunk U. (1981). Increasing the Juice Yield of Pressing with Special Remarks on the Warm Extraction of Apples. Flssiges Obst 8: 248-255. 34. Gortner WA, Dull GG, Kraus BH. (1967). Fruit Development, Maturation, Ripening and Senescence: A Biochemical Basis for Horticultural Terminology. HortSci. 2(4): 141-144. 35. Handbook of Selected Agricultural Statistics. (1986). Minister of Supply and Services Canada, p. 51. 36. Hansen E. (1966). Postharvest Physiology of Fruits. Ann. Rev. Plant Physiol. 17:459-480. 37. Hansen P. (1967). C - Studies on Apple Trees. I. The Effects of the Fruit on the Translocation and Distribution of Photosynthates. Physiologia Plantarum 20: 382-391. 1 4  38. Hansen P. (1970). C-Studies on Apple Trees. V. Translocation of Labelled Compounds from Leaves to Fruit and their Conversion within the Fruit. Physiologia Plantarum 23: 564-573. 14  39. Hansen P. (1979). C-Studies on Apple Trees. IX. Seasonal Changes in the Formation of Fruit Constituents ana their Subsequent Conversions. Physiol. Plant. 47:190-194. 14  40. Hanson JB, Day DA. (1980). Plant Mitochondria in " The Biochemistry of Plants - A Comprehensive Treatise. Vol. 1. The Plant Cell." (N.E. Tolbert, ed.) Academic Press. New York. p. 315-344. 41. Hashiba H. (1982). The Browning Reaction of Amadori Compounds Derived from Various Sugars. Agric. Biol. Chem. 46(2): 547-548. 42. Hayat MA. (1972). Principles and Techniques of Electron Microscopy: Biological Applications. Vol. 2. Van Norstrand Reinhold Co. New York, p. 30-151. 43. Hayat MA. (1981). Principles and Techniques of Electron Microscopy: Biological Applications. Aspen Publishers, Lie. Rockville, Ma. p. 301-316.  97. 44. Heatherbell DA. (1976a). Haze and Sediment Formation in Clarified Apple Juice and Apple Wine. Alimenta 15:151-154. 45. Heatherbell DA. (1976b). Haze and Sediment Formation in Clarified Apple Juice and Apple Wine: The Role of Polyvalent Cations, Polyphenolics and Proteins. Food Tech. in New Zealand 11(6): 17,25. 46. Hodge JE, Osman EM. (1976). Carbohydrates in "Principles of Food Science. Part I. Food Chemistry". (O.R. Fennema, ed.). Marcel Dekker Inc. New York. p. 41-137. 47. Holgate KC, Moyer JC, Pederson CS. (1948). The Use of Ascorbic Acid in Preventing Oxidative Changes in Apple Juice. The Fruit Products J. and American Food Manufacturer 28:100-112. 48. Hulme AC. (1958). Some Aspects of the Biochemistry of Apple and Pear Fruits. Advances in Food Research 8: 297-413. 49. Hulme AC, Rhodes JC. (1971). Pome Fruits in "The Biochemistry of Fruits and Their Products" (A.C. Hulme, ed.). Academic Press. London, p. 333-373. 50. Hunter EE. (1984). Practical Electron Microscopy: A Beginners Ilustrated Guide. Praeger Publishers. New York. p. 65. 51. Hurler A, Wey R. (1984). Techniques of Grinding and Making the Mash. Confructa Studien March/April: 125-130. 52. Jewell GG. (1979). Fruits and Vegetables in "Food Microscopy". (J.G. Vaughan, ed.). Academic Pres, Inc. London, p. 1-34. 53. Johnson G, Donnelly BJ, Johnson DK. (1968). The Chemical Nature and Precursors of Clarified Apple Juice Sediment. J. Food Sci. 33: 254-257. 54. Johnson G, Donnelly BJ, Johnson DK. (1969). Proanthocyanidins as Related to Apple Juice Processing and Storage. Food Tech. 23: 1312-1316. 55 . Junker R. (1987). Does it Pay to Invest in an Apple Mash Enzyme. Flssiges Obst. 54(8): 435,441-442,443-444. 56. Kelly Gj, Latzko E. (1980). The Cytosol in "The Biochemistry of Plants - A Comprehensive Treatise. Vol. 1. The Plant Cell." (N.E. Tolbert, ed.). Academic Press. New York. p. 184-204. 57. Kilara A. (1982). Enzymes and their Uses in the Processed Apple Inustry: A Review. Process Bioc. July/August: 35-41. 58. Klavons JA, Bennett RD. (1985). The Nature of the Protein Constituent of Commercial Lemon Juice Cloud. Food Chem. 33: 708-712.  98. 59. Klavons JA, Bennett RD. (1987). Nature of the Pectin Content of Commercial Lemon Juice Cloud. J Agri. Food Chem. 35(1): 159-161. 60. Knee M. (1973a). Polysaccharides and Glycoproteins of Apple Fruit Cell Walls. Phytochemistry 12:637-653. 61. Knee M. (1973b). Polysaccharide Changes in Cell Walls of Ripening Apples. Phytochem. 12:1543-1549. 62. KneeM. (1975). Soluble and Wall-Bound Glycoproteins of Apple Fruit Tissue. Phytochem. 14: 2181-2188. 63. KneeM. (1978a). Properties of Polygalacturonate and Cell Cohesion in Apple Fruit Cortical Tissue. Phytochem. 17:1257-1260. 64. KneeM. (1978b). Metabolism of Polymethylgalacturonate in Apple Fruit Cortical Tissue During Ripening. Phytochem. 17:1261-1264. 65. Knee M, Fielding AH, Archer SA, Laborda F. (1975). Enzyme Analysis of Cell Wall Structure in Apple Fruit Cortical Tissue. Phytochem. 14: 2213-2222. 66. Kwan W, Kowalski BR. (1980). Correlation of Objective Chemical Measurements and Subjective Sensory Evaluations. Wines of Vitus vinifera variety 'Pinot Noir' from France and the United States. Anal. Chfm. Acta 122:215. 67. Labuza TP, Saltmarch M. (1981). The Nonenzymatic Browning Reaction as Affected by Water in Foods in "Water Activity: Influences on Food Quality". (L.B. Rockland and G.F. Stewart, eds.). Academic Press. London, p.611-612. 68. Lau OL. (1983). Effects of Storage Procedures and Low 02 and CO2 Atmospheres on Storage Quality of 'Spartan' Apples. J. Amer. Soc. Hort. Sci. 108(6): 953-957. 69. Lau OL. (1985). Harvest Indices for British Columbian Apples. British Columbian Orchardist July: 7-13. 70. Lau OL, Liu Y, Yang SF. (1986). Effects of Fruit Detachment on Ethylene Biosynthesis and Loss of Flesh Firmness, Skin Colour, and Starch in Ripening 'Golden Delicious' Apples. J. Amer. Soc. Hort. Sci. 111(5): 731-734. 71. Lee HS, Wrolstad RE. (in Press). Detection of Adulteration of Apple Juices in "Adulteration of Fruit Juice Beveridges". (S. Nagy, ed.). Marcell Dekker, Inc. New York. 72. Lehninger AL. (1973). Bioenergetics - The Molecular Basis of Biological Energy Transformations. Second Edition. W.A. Benjamin, Inc. Menlo Park, Cal. p. 3.  99. 73. LeibermanM. (1979). Biosynthesis and Action of Ethylene. Ann. Rev. Plant Physiol. 30: 533-91. 74. Leonard RT, Hodees TK. (1980). The Plasma Membrane in "The Biochemistry of Plants - A Comprehensive Treatise. Vol. 1. The Plant Cell." (N.E. Tolbert, ed.). Academic Press. New York. p. 163-179. 75. Lewis DF. (1981). The Use of Microscopy to Explain the Behaviour of Foodstuffs - A Review of the Leatherhead Food Research Association. Scanning Electron Microscopy HI: 391 - 404. 76. Li BW, Schuhmann PJ. (1983). Sugar Analysis of Fruit Juices: Content and Method. J. Food Sci. 48(2): 633-635. 77. Lidster PD, Lightfoot HJ, MacRae KB. (1983). Fruit Quality and Respiration of Mcintosh' Apples in Response to Ethylene, Very Low 02 and CO2 Storage Atmospheres. Scientia Horticulturae 20:71-83. 78. Lidster PD, Sanford KA, McRae KB, Stark R. (1984). Apple Juice Quality and Recovery from Mcintosh Apples Stored in Controlled Atmospheres and Air. Can. Inst. Food Sci. Technol. J. 17(2): 86-91. 79. Loh J, Breene WM, Davis EA. (1982). Between-Species Differences in Fracturability Loss: Microscopic and Chemical Comparison of Potato and Chinese Waterchestnut. J. Texture Studies 13: 325-347. 80. MacMillan JD, Sheiman MT. (1974). Pectic Enzymes. Advances in Chemistry Series 136:101-130. 81. Marty F, Branton DB, Leigh RA. (1980). Plant Vacuoles in "The Biochemistry of Plants - A Comprehensive Treatise. Vol. 1. The Plant Cell." (N.E. Tolbert, ed.). Academic Press. New York. p. 627-654. 82. Mattick LR. (1984). Chemical Composition of Apple Juice in "Third National Processed Apple Conference". (D.L Downing and R.L. LaBelle, eds.). Geneva Proceedings, p. 2-4. 83. Mattick LR, Moyer JC. (1983). Composition of Apple Juice. J. AOAC. 66(5): 1251-1255. 84. McClellan MR, Barnard J, Queale DT. (1984). A Study of Certain Sensory Characteristics of Carbonated Apple Juice in' Apple Juice Worksnop". (D.L. Downing, ed.). Special Report No. 54. New York State Experimental Station. Geneva, New York. p. 8-11. 85. McComb EA, McCready RM. (1952). Colorimetric Determination of Pectic Substances. Anal. Chem. 24:1630-1632. 86. McFeeters RF. (1985). Changes in Pectin and Cellulose during Processing in "Chemical Changes in Food During Processing." AVI Publishing Co. Westport, Conn. p. 347-372.  100.  87. McGlasson WB. (1970). The Ethylene Factor in "Biochemistry of Fruits and their Products." (A.C. Hulme, ed.). Vol.1. Academic Press. London, p. 478-512. 88. Melton LD, Laas A M . (1985). Determination of Sugars in New Zealand Apple Juice. New Zealand J. Food Tech. 1:191-193. 89. Milnes BA. (1984). Application of Ultrafiltration to Apple Juice Processing in "Apple Juice Workshop". (D.L. Downing, ed.). Special Report No. 54. New York State Experimental Station. Geneva, New York. p. 33-38. 90. Mizrahi S, Berk Z. (1970). Physico-Chemical Characteristics of Orange Juice Cloud. J. Sci. Food Agric. 21: 250-253. 91. Montogomery MW, Petropakis HJ. (1980). Inactivation of Pear Polyphenol Oxidase with Heat in Presence of Ascorbic Acid. J. Food Sci. 45:1090-1091. 92. Moyer JC, Aitken HC. (1980). Apple Juice in "Fruit and Vegetable Juice Technology". (P.E. Nelson and D.K. Tressler, eds.). AVI Publishing Co., he. Westport, Conn. p. 212-267. 93. Nadudvari-Markus V, Vamos-Vigyazo L. (1984). Enzymatic Browning Substrates in Apple Cultivars. Acta Alimentaria 13(1): 97-106. 94. Nitsch JP. (1970). Hormone Factors in Growth and Development in "Biochemistry of Fruits and their Products." Vol. 1. (A.C. Hulme, ed.). Academic Press. London, p. 427-459. 95. Paulson AT, Vanderstoep J, Eaton GW. (1979). Effects of Gibberellic Acid and Ethephon on Enzymatic Browning of 'Red Haven' Peaches. HortSci. 14(6): 711-713. 96. Paulson DA, Wilson RL, Spotz DD. (1984). Crossflow Membrane Technology and its Applications. Food Tech. 38(12): 77-87,111. 97. Pham AM, Nakai S. (1984). Application of Stepwise Discriminant Analysis to High Pressure Liquid Chromatography Profiles of Water Extracts forjudging Ripening of Cheddar Cheese. J. Dairy Sci. 67: 1390-1396. 98. Pilnik W. (1982). Enzymes in the Beverage Industry in "Use of Enzymes in Food Technology". (P. Dupuy, ed.). Proc. Int. Symposium. Versatile, France, p. 425-450. 99. Poll L. (1983). The Influence of Apple Varitey on Sensory Quality of Heated Apple Juices. Lebensm. -Wiss. u.-Tecnnol. 16:119-121. 100. Pollard A, Timberlake CF. (1971). Fruit Juices in "Biochemistry of Fruits and their Products". Vol. I. (A.C. Hulme, ed.). Academic Press. London, p. 573-621.  101.  101. Powrie WD, Tung MA. (1976). Food Dispersions in "Principles of Food Science Part I: Food Chemistry." (O.R. Fennema, ed.). Marcel Dekker, Inc. New York. p. 554. 102. Preston RD. (1979). Polysaccharide Conformation and Cell Wall Function. Ann. Rev. Plant Physiol. 30: 55-78. 103. Proctor JTA. (1979). Apple Cultivars Grown in Canada. Fruit Varieties J. 33(1): 12-13. 104. Reeve RM. (1953). Histological Investigations of Texture in Apples. Part I. Composition and Influence of Heat on Structure. Food Research 18:592-603. 105. Reeve RM, Leinbach LR. (1953). Histological Investigations of Texture in Apples. Part n. Structure and Intercellular Spaces. Food Research 18: 604-617. 106. Rhodes JC. (1970). The Climacteric and Ripening of Fruits in "Biochemistry of Fruits and their Products." Vol. 1. (A.C. Hulme, ed.). Academic Press. London, p. 521-533. 107. Richardson T. (1976). Enzvmes in "Principles of Food Science. Part I. Food Chemistry". Marcell Dekker, Inc. New York. p. 325-329. 108. Rouau X, Thibault J.-F. (1984). Apple Juice Pectic Substances. Carbohydrate Polymers 4:111 -125. 109. Rouchaud J, Moons C, Meyer JA. (1985). Cultivar Differences in the Influence of Harvest Date and Cold Storage on the Free Sugars and Acids Contents, and on the Eating Quality of Apples. J. Hort. Sci. 60(3): 291-296. 110. Ryall AL, Pentzer WT. (1982). Handling, Transportation and Storage or Fruits and Vegetables. Vol. 2. Fruits and Tree Nuts. AVI Publishing Co., Inc. Westport, Conn. p. 1-395. 111. Ryan JJ. (1972). Chemical Composition of Canadian Apple Juice. J. AOAC. 55:1104-1108. 112. Saltmarch M, Labuza TP. (1982). Nonenzymatic Browning Via the Maillard Reaction in Foods. Diabetes 31(3): 29-35. 113. SAS (1986). User's Guide: Statistics. Version 5. SAS Institute Inc. Cary, North Carolina. 114. Schmitt R. (1983). New Ways to Enzymic Liquefaction of Fruits and Vegetables. Flssiges Obst. 50(1): 23-27. 115. Schwimmer S. (1981). Source Book of Food Enzymology. AVI Pub. Co., Inc. Westport, Conn. p. 512-513.  102.  116. Sheu MJ, Wiley RC. (1983). Preconcentration of Apple Juice by Reverse Osmosis. J. Food Sci. 48:422-429. 117. Short JL. (1983). Juice Clarification by Ultrafdtration. Pro Bio Tech. October, p. vi. 118. Siostrand FS. (1967). Electron Microscopy of Cells and Tissues. Vol. 1. Instrumentation and Techniques. Academic Press. New York, p. 300. 119. Skoog DA. (1985). Principles of Instrumental Analysis. Third Edition. Saunders College Publishing. Philadelphia, p. 22-23,379-380. 120. Smith GRS, Gore WL and Associates, Inc. (1984). Filter Aids and Equipment in "Apple Juice Workshop". (D.L. Downing, ed.). Special Report No. 54. New York State Experimental Station. Geneva, New York, p.27-32. 121. Smock RM, Neubert AM. (1950). Apples and Apple Products. Interscience Publishers, Inc. New York. 122. Sterling C. (1955). Effects of Moisture and High Temperature on Cell Walls in Plant Tissues. Food Research 20:474-479. 123. Strachan CC, Moyls AW, Atkinson FE, Britton JE. (1951). Chemical Composition and Nutritive Value of British Columbia Tree Fruits. Dept. Agriculture Publication 862. Ottawa, p. 18-19. 124. Terre E. (1987). Extended Experience with Ultrafiltration in Fruit Juice Plant. Flssiges Obst. 54(8): 421-424,437-438. 125. Thomas RL, WestfaU PH, Louvier ZA, Ellis ND. (1986). Production of Apple Juice by Single Pass Metallic Membrane Ultrafiltration. J. Food Sci. 51(3): 559-563. 126. Timbers GE. (1975) Reverse Osmosis Concentration of Second Press Apple Juice in "Third National Processed Apple Conference". (DL. Downing and R.L. LaBelle, eds.). Geneva Proceedings, p. 41-44. 127. Toribio JL, Lozano JE. (1984). Nonenzymatic Browning in Apple Juice Concentrate During Storage. J. Food Sci. 49: 889-892. 128. Toribio JL, Nunes RV, Lozano JE. (1984). Influence of Water Activity on the Nonenzymatic Browning of Apple Juice Concentrate During Storage. J. Food Sci. 49:1630-1631. 129. Tressler DK, Pederson CS. (1938). Pasteurization of Apple Juice. Circular No. 181. New York State Experimental Station. Geneva, New York.  103. 130. Vamos-Vigyazo L, Gajazgo I, Nadudvari-Markus V, Mihalyi K. (1976). Studies into the Enzymic Browning and the Polyphenol- Polyphenol Oxidase Complex or Apple Cultivars. Confructa 21(1/2): 24-35. 131. Vamos-Vigyazo L, Mihalyi K, Gajazgo I, Nadudvari-Markus V. (1977). Role of Enzyme- Subsrtate Ratio m the Enzymatic Browning of Fruit Tissue Homogenates. Acta Alimentaria 6(4): 379-388. 132. Vamos-Vigayz6 L, Nadudvari-Markus V. (1983). Inactivation of Polyphenol Oxidase and Depletion of ortho-Diphenol Content During the Enzymatic Reaction of Fruit Tissues. Acta Alimentaria 12(1): 1-9. 133. Van Buren JP. (1984). Haze in Bottled Apple Juice in "Apple Juice Workshop". (D.L. Downing, ed.). Special Report No. 54. New York State Experimental Station. Geneva, New York. p. 18-24. 134. VanWoenselG, De Baerdemaeker J. (1983). Mechanical Properties of Apples During Storage. Leben. -Wissen. u.-Technol. 16: 367-372. 135. Vijayalakshmi MA, Picque D, Jaumouille R, Segard E. (1980). Immobilized Pectinesterase in "Food Process Engineering. Vol. 2. Enzyme Engineering in Food Processing". (P. Linko and J. Larinkari, eds.). Applied Science Publishers Ltd. London, p. 152-158. 136. Voragen FGJ, Krist R, Heutink R, Pilnik W. (1979). Apple Cell Wall Digestion by Polysaccharide Degrading Enzymes in "Food Process Engineering. Vol. 2. Enzyme Engineering in Food Processing". (P. Linko and J, Larinkari, eds.). Applied Science Publishers. London, p. 264-275. 137. Vos L de, Pilnik W. (1973). Pectolytic Enzymes in Apple Juice Extraction. Process Bioc. 8(8): 18-19. 138. Vries JA de, Voragen AGJ, Rombouts FM, Pilnik W. (1981). Extraction and Purification or Pectins from Alcohol Insoluble Solids from Ripe and Unripe Apples. Carbohydrate Polymers 1:117-127. 139. Webb KL, Burley JWA. (1962). Sorbitol Translocation in Apple. Science 137:766. 140. Whitaker JR. (1984). Pectic Substances, Pectic Enzymes and Haze Formation in Fruit Juices. Enzyme and Microbiological Tech. 6: 341-349. 141. Whiting GC. (1970). Sugars in "Biochemistry of Fruits and their Products." Vol. 1. Academic Press. London, p. 12-16. 142. Willaman JJ. (1933). Economical Clarification Through Use of Enzymes. Food Industries August: 294-295,301.  104.  143. Wilson LG, Fry JC. (1983). Plant Cell Wall Glycoproteins - A Literature Review. Scientific and Technical Surveys No. 141. p. 1-17. 144. Woodroof JG. (1975). Seasonal Suitability of Fruits for Processing in "Commercial Fruit Processing". (J.G. Woodroof and B.S. Luh, eds.). AVI Publishing Co., Inc. Westport, Conn. p. 100-148. 145. Wrolstad RE. (1976). Colour and Pigment Analysis in Fruit Products. Station Bulletin No. 624. Agriculture Experimental Station, Oregon State University, Corvallis. p. 1-17. 146. Wucherpfennig K. (1981). Fruit Extraction by Using the Diffuser Process. Flssiges Obst. 48(8): 232,241-247. 147. Yamasaki M, Yasui T, Arima K. (1964). Pectic Enzymes in the Clarification of Apple Juice. Part I: Study on the Clarification Reaction in a Simplified Model. Agri.Biol. Chem. 28(11): 779-787. 148. Yamasaki M, Kato A, Chu S-Y, Arima K. (1967). Pectic Enzymes in the Clarification of Apple Juice. Part II: The Mechanism of Clarification. Agri. Biol. Chem. 31(5): 552-560. 149. Yamasaki S. (1984). Isolation of Vacuoles from Immature Apple Fruit Flesh and Compartmentation of Sugars, Organic Acids, Phenolic Compounds and Amino Acids. Plant and Cell Physiology 25(1): 151-166 in Hort. Abstracts 54(8) :496.  105.  APPENDIX A  106.  3  Figure  A-1. HPLC stored (2)  determination Spartan  g l u c o s e , C33  of  apples:  s u g a r s in j u i c e (1)  fructose,  sucrose, (4)  sorbitol.  from  107.  a  Figure  A-2.  b  HPLC juice  determination from  stored  the  of  organic  enzyme-treated  Red Delicious a p p l e s :  (2)  galacturonic,  (5)  succinic.  Ca) s u l f u r i c acid from  acids  (3)  Other  puree CD  malic, ( 4 ) peaks  acid from  of  citric,  quinic,  were:  eluant,  ion e x c h a n g e  in  (b)  resin.  acetic  Table A-l. ANOVA of apple juice soluble solids data. Source  Degrees of Freedom  Sums of Squares  Mean Squares  F  Variety (V)  2  1Z91  6A5  30.7**  Condition (C)  1  3.23  3.23  1 5.4**  Process (P)  1  0.93  0.93  4.4*  Maceration (M)  1  6.10  6.10  29.0**  VxC  2  1.19  0.59  2.8  VxP  2  0.02  0.01  0.05  VxM  2  0.79  0.39  1.9  CxP  1  0.24  0.24  1.1  CxT  1  0.26  0.26  1.2  PxT  1  0.07  0.07  0.3  Error  9  1.85  0.21  Total  23  27.59  *Significant at p < 0.05. "Significant at p < 0.01.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097691/manifest

Comment

Related Items