("’"'""".P'.““'-'°' ° “"’"‘ "t ' “‘z.o~ ' ”0-“ .‘1P9v!nmlo.oovla-roo.¢ .. “-. —oo ‘..O.-o--.o~4 h. urvo o... g-o u-ogo-o-q -g94.;.*,."qou “ouqu—‘0100m3 ”-O9'9 .i' BEAN ANB MEBBLENGS AS A SOURCE 0F BERRY FEBER EN LAYER 0%ES ' E‘uQ'SSfC‘ViT'S 8g 0W 3 E E3 233% ail SfAFEg UNE‘JERSITY _ CMDAEE HNNE :WGHEL i ’ 1975 ' A n ‘ I l ‘ l A. . . f. . - .- ‘H '1 V' " ' . y ‘ I II ‘. I - . v ‘ a ‘ ~ A -' . n O-JI'. 1:). 'n.‘ .0 4”,]; '. I'L ‘,n:_og II .‘ .0 v C " I'l'DO' ........ . ‘ ‘ u. .1125'.'\t‘.'.'9 . 1 ‘ RE? ’1‘]. ' : b .4 “ti. ~’ I O - I» u U A. I "I O . I. A 3 r ‘ , , "y J ' u: on . 0 l ‘\ .‘ ' . ". .-'.°. - I 0'»: ~.\ ‘ ' .' '. .‘ ‘” |:,'.~\ ‘ a n. . .... .'.v . .'. \.‘ ‘ - u. . a O ' . ‘7 9 . ' 4}:,vt.;‘.“cg‘ - ¢. WW H 2‘ ‘ 0 :‘ gl‘ ". ~. '0‘ *\9| 4“ n3?“ ' \J 0 ABSTRACT BRAN AND MIDDLINGS AS A SOURCE OF DIETARY FIBER IN LAYER CAKES By Candace Lynne Rajchel Medical studies have recently indicated that a relation- ship exists between a number of diseases and a lack of dietary fiber in the diet. In an attempt to increase the level of dietary fiber intake white layer cakes were prepared with seven substitution levels of bran and middlings for both 60 and 70% extraction flours. Substitution levels included A, 8, and 16% bran, 12% midds bleached or unbleached, and 16% bran plus 12% midds bleached or unbleached. Flavored layer cakes including nut, spice, banana, and chocolate, were then prepared using substitution levels of 16% bran, 12% unbleached midds, and 16% bran plus 12% unbleached midds for a 70% extraction flour. In order to determine substitution levels consistent with high quality the white cake batter and all cakes were evaluated by both objective and subjective measure- ments. The results of this study indicate that bran and midds may be successfully substituted for flour in the layer cake system. The batter viscosity increased as did the cake volumes with increasing substitution levels of fiber for flour Candace Lynne Rajchel in the white cakes. Substitution of fiber in the white cake formula resulted in increased tenderness and the color scores of both the white batter and cake were affected in that yellow— ness and redness increased as lightness decreased. The quality characteristics of specific gravity, pH, symmetry, uniformity, shrinkage, moisture, and compressibility, as well as sensory scores, were not adversely affected by fiber sub- stitutions in the white cake formula. On the whole, the quality of the flavored cakes was not adversely affected by substitutions of fiber in the cake formulas for either the objective or subjective measure- ments. All sensory scores were 7 or better on a 10-point scale for all characteristics evaluated. The fact that so few cake characteristics were affected by a fiber for flour substitution in the cake formula indi- cates that such substitutions may become a feasible method of increasing dietary fiber levels in the diet. BRAN AND MIDDLINGS AS A SOURCE OF DIETARY FIBER IN LAYER CAKES By Candace Lynne Rajchel A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1975 ACKNOWLEDGEMENTS I wish to express my sincere appreciation to Dr. Mary Ellen Zabik for her guidance and help throughout the course of my graduate study and especially during the preparation of this manuscript. I want to thank Mr. John Donelson and the USDA Soft Wheat Quality Research Laboratory in Wooster, Ohio for their help in the analysis of the flours. In addition, I am grate- ful to Dr. Everett Everson, Dr. Kusum Patel, and Miss Jean McFadden for their review of this paper. I am also grateful for the use of the equipment borrowed from the departments of Agricultural Engineering, Biochemistry, and Crop and Soil Science and for the ingredients donated by Mennel Mills, Gerber, and Seymour Foods. Finally, I want to thank my parents for their constant encouragement and finanacial support which made it possible for me to obtain this degree and Mr. Dean Brockmole for his never-failing interest and help. ii TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . REVIEW OF LITERATURE . . . . . Dietary Fiber . . . . . . . Fiber Deficiency Disorders Diverticular Disease Atherosclerosis . . . . Colonic Cancer . . . . . Appendicitis . . . . Crude Fiber vs. Dietary Fiber Dietary Fiber Components . . . . Cellulose . . . . . . . . Lignin . . . . . . . . . Hemicellulose Pectic Substances . . . Dietary Fiber in Foods . . Composition of the Wheat Kernel Bran . . . . . . . . . . . . Middlings . . Formation of the Layer Cake Structure EXPERIMENTAL PROCEDURE . . . . . . . . . . Ingredient Procurement . . . . . Layer Cake Formulations . . . . . . . Method of Preparation . . . . . . . Flour Preparation . . . . Cake Mix Preparation . . . . . White Cake Preparation . . . . . Flavored Cake Preparation Nut Cakes . Spice Cakes Banana Cakes . . . Chocolate Cakes . . . . . . Baking Procedure . . . . . . . . . . Preparation of Samples . . . . . Objective Measurements Batter Viscosity Specific Gravity . . . . . pH . . . . . . . . . . . . . . . iii TABLE OF CONTENTS--continued Color Volume . . . Tenderness . . . . . . . . . . Compressibility Moisture . . Crude Fat . . Alkaline Water Retention Capacity (AWRC) Flour Fractionation Subjective Evaluation . . . . White Layers . . . . . . . . . . Flavored Layers . . . . . . . . Analysis of Data . . . . RESULTS AND DISCUSSION White Layers . . . . . . Objective Measurements . . . . . Batter and Volume Parameters Shrinkage and Symmetry and Uniformity Indices . . . Tenderness and Compressibility Batter and Cake Color . . Moisture and Alkaline Water Retention Capacity . . . . . . . . . pH . . . Dietary Fiber by Difference Sensory Evaluation . . Flavored Layers . . . Objective Measurements Shrinkage and Volume- Related Indices Cake Color . . . . . . . . Moisture . . . . . . . . Sensory Evaluation SUMMARY AND CONCLUSIONS PROPOSALS FOR FURTHER RESEARCH REFERENCES APPENDICES I. CITED Proximate Analysis of Bran, Middlins, Flours Obtained from Mennel Mills iv and TABLE OF CONTENTS—-continued II. III. IV. VI. VII. VIII. Average Particle Size of Flour and Cyclone Milled Bran and Midds Taste Panel Evaluation of White Layer Cake Characteristics . . . . . . . Taste Panel Evaluation of Flavored Layer Cake Characteristics . . . . . . . . . . Scoring of Individual Characteristics by Taste Panel Members for White Layer Cakes Prepared with 60% Extraction Flour . . Scoring of Individual Characteristics by Taste Panel Membensfor White Layer Cakes Prepared with 70% Extraction Flour . . Results of Flour Fractionation Procedure for 60 and 70% Extraction Flours . . . . Crude Fat Analyses of 60 and 70% Extraction Flours . . . . . . . . . . . . . . . . . . Page 87 88 89 9O 91 92 93 TABLE 10. 11. LIST OF TABLES Trends in annual per capita food availability (Friend, 1967) . . . . . . . . . . . . Estimated composition of the cell wall poly- saccharides in some foods (Southgate, 1969) Nutrient contribution of some wheat kernel constituents (Wheat Flour Institute, 196”) Formula for standard high-ratio white layers Formualtions for flavored layer cakes Means and standard deviations for specific grav- ity, viscosity, and volume index values of white layers prepared with 60 and 70% extraction flours . . . . . . . . . . Analyses of variance of specific gravity, viscos- ity, and volume indices of white layer cakes pre- pared with 60 and 70% extraction flours . . . . Means and standard deviations for shrinkage and symmetry and uniformity indices of white layer cakes prepared with 60 and 70% extraction flours Analyses of variance of shrinkage and symmetry and uniformity indices of white layer cakes prepared with 60 and 70% extraction flours Means and standard deviations for tenderness and compressibility values of white layer cakes pre- pared with 60 and 70% extraction flours . . Analyses of variance of tenderness and compressi- bility values of white layer cakes prepared with 60 and 70% extraction flours . . . . . . . . . vi Page 15 17 2A 25 38 39 A2 “3 AA “5 LIST TABLE l2. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. OF TABLES--continued Means and standard deviations for L, a , and b values of white layer cake batter prepared wIth 60 and 70% extraction flours . . . . . . Analyses of variance of L, a , and b values of white layer cake batter prepared Nith 60 and 70% extraction flours . . . . . . . . . . . . Means and standard deviations for L, aL, and bL values of white layer cakes prepared with 60 70% extraction flours . . . . . . . . . Analyses of variance of L, aL, and b values of white layer cakes prepared with 60 and 70% extraction flours . . . . . . . . . . . . . . . Means and standard deviations for percentage moisture and single determination value for AWRC for white layer cakes prepared with 60 and 70% extraction flours . . . . . . . . . . Analyses of variance for percentage moisture of white layer cakes prepared with 60 and 70% extraction flours . . . . . . . . . . . . . Mean values for white cake batter and cake pH prepared with 60 and 70% extraction flours . Dietary fiber determinations by difference in 60% extraction flour . . . . . . . . . . . Dietary fiber determination by difference in 70% extraction flour . . . . . . . . . . . Means and standard deviations of sensory scores for white layers prepared with 60 and 70% extraction flours . . . . . . . . . Analysis of variance of sensory scores for white layer cakes prepared with 60 and 70% extraction flours . . . . . . . . . . . . . . . . . . . Means and standard deviations for shrinkage and volume, symmetry, and uniformity indices of flavored layer cakes vii Page “7 A8 A9 50 52 53 55 56 57 6O 61 6A u.- _‘L n'uq .- OI-yl pl. J D//_ n (L A\. Al/L LIST OF TABLES-—continued TABLE Page 2A. Analyses of variance of shrinkage and volume, symmetry, and uniformity indices for flavored cakes . . . . . . . . . . . . . . . . 65 25. Means and standard deviations for color scores of flavored layer cakes . . . . . . . . . . . . 67 26. Analyses of variance of color scores for flavored layer cakes . . . . . . . . . . . . . 68 27. Means and standard deviations for moisture determinations of flavored layer cakes . . . . 7O 28. Analyses of variance of moisture determinations for flavored layer cakes . . . . . . . . . . . 71 29. Means and standard deviations for sensory scores of flavored layer cakes . . . . . . . . . . . . 72 30. Analyses of variance for sensory scores of flavored layer cakes . . . . . . . . . . . . . 73 viii LIST OF FIGURES FIGURE Page 1. Pathogenesis of diverticular disease . . . . . . 7 2. Cross section of a wheat kernel . . . . . . . . l6 3. Cutting diagram for cakes . . . . . . . . . . . 30 A. The color and appearance of white layer cakes prepared with increasing substitution levels of bran and midds for flour . . . . . . . . . . 62 ix INTRODUCTION Although fiber has been recognized as an important com- ponent of animal feeds, it has long been considered an insig- nificant part of the human diet. Fiber was considered neces- sary exclusively for the relief of constipation. During the past decade, however, fiber has been studied extensively and recent advances in the knowledge of its chemistry in addition to numerous epidemiological observations have led to the belief that fiber plays an essential role in the intestine and in maintaining man's health. Since the epidemiological evidence appears to point an incriminating finger at the lack of foods containing fiber in the diet, it is essential to examine the effects of fiber on intestinal content and behavior. Cowgill and Anderson (1932) found that the addition of fiber-containing foods to the diet increased fecal bulk and improved bowel habits. Dietary fiber has also been shown to shorten transit time (McCance et al., 1953), that is, the time it takes for a food to pass through the digestive tract. The role that dietary fiber plays in the intestine has been correlated with epidemiological data to form the hypothesis which relates the prevalence of a wide group of diseases to a lack of fiber in the diet. Much of the present interest in the role of dietary fiber 1 B1 In t .uv ‘m‘ 3“ HA, ‘II LV v‘“ C!" (D ,. ‘1'. a in relation to disease arises from the fact that both the source and the amount of fiber intake have changed during the past century. Between 1860 and 1960 fat consumption increased by less than 50% and sugar consumption more than doubled (Burkitt, 1973). In Britain in 1835 flour consump- tion was approximately 3601b/person/year. By 1880 it had dropped to 280lb/person/year and in l970'to lASlb/person/ year (Cummings, 1973). In the United States consumption of whole wheat flour has decreased from l60lb/capita in 1900 to less than lOOlb/capita in 1970 (Scala, 197A). The fiber Content of flour has also been reduced considerably since the beginning of the century due to a change in the milling process whereby flour is now more highly refined. The evidence appears to substantiate the theory that an inc31"¢Eiase in dietary fiber intake would be very beneficial. Fiber- which will need to be added to the daily diet to aid in disease prevention must be added to foods in an appealing manner in order to assure consumer acceptance of the product. Because fiber will need to be added to foods in amounts greater than that of the level of present food additives in order to be beneficial, it may have a significant effect on product quality characteristics. Because flour is a major component of any baked product, these Systems could become feasible fiber carriers. Possible DPOdUCtS for such fiber supplementation would include cakes, cookies, crackers, doughnuts, pastry, muffins, and biscuits. Ho Wevep, due to the physical and chemical properties Of the V‘ V A r_n ..x.‘. ‘r’v' ' “ ., ~...Jn.> r ‘7'” 41“ ' l ‘->‘ . uni, fiber components, product quality may be impaired without a change in the product formulation. The purpose of this research was to determine the differ- ential effects of the incorporation of varying levels of bran and middlings on the physical and sensory quality character— istics of layer cakes. REVIEW OF LITERATURE Since the primary emphasis of this research was to :irlxreastigate the effects of the addition of dietary fiber on :Lélbfexr cake quality characteristics, the need for dietary fZ’Lbe-rl" in the diet and the composition of the dietary fiber C“Gristituents are reviewed in the first portion. The second p(31"tion of this review summarizes the formation of the layer C a ke structure . Dietary Fiber Cereals, fruits, and vegetables are the primary sources CDf‘ f‘Ziber in the diet but a marked reduction in the consumption CDf‘ these foods has occurred during the past century. Associated ““1131? this reduction is increased intake of foods low in fiber Corltent such as animal products and sweetening agents as ShOWn in Table 1. Not only has flour and cereal consumption declined but techlhological advances in the milling process have caused a reduction in flour fiber content. Flour refining changed from Stc"Tie—grinding to roller-milling in the 1880's. As a result (bf ‘this change, the flour extraction rate dropped from 80% to 7(yz (Jones, 1958), while the fiber content was reduced by mmOpmpoa pmmzm woodeocHe Nee woe oee oee mm .mne .meowpmozm a emwsm pee wee eee eom emm .mee .memmaee e mesoem mom mom mmm mmm mom .mee .meeemeeme> bezeo eoe oee mme eee mom e.mee .meoemeoa wee mme mom mme wee .mee .eeapm em em em mm mm .mee .meeo e meme emm oem emm mom nee .mee .meeseoad eeema mm me we em em .mee .mmmm mom mme wee mee Nee .mee .Ewee .AEeesoa .eemz meme mmuemme meueeme mmummme menmome meadow eoom .Amwme .ncmeemv zpeeenmeem>w boom mpwqmo emu emeecm me mesmee .e eeeee 6 nearly half (Pomeranz, 196”). Fiber Deficiency Disorders The close association between the incidence of certain diseases and their geographical distribution has led to the hypothesis that environment and diet play an important role in the development of these physical disorders. Many diseases are thought to be characteristic of Western civilization because they are rarely found in communities which exist on high resi— dLIe diets and are common in the Western world which exists on a low residue diet. Cleave gt a_1_. (1969) drew attention to the relationship between some diseases and implicated the OVer—consumption of refined carbohydrates as a major causal fac tor. This relationship has been supported by a number of I‘eSearchers for diverticular disease (Painter and Burkitt, 1971), atherosclerosis (Trowell, 1972), colonic cancer (Burkitt, 1971), and appendicitis (Walker et a}, 1973). Stool bulk and content, transit time, bacterial flora, and intra—abdominal pressures are all markedly affected by dietary changes and par— t1(It‘llarly so by the removal of the unabsorbable fiber component from food. This over-consumption of refined carbohydrates, esDecially white flour and sugar, is characteristic of Western civilization. Diverticular disease. The role of fiber in the diet is to act as a bulking agent. In the absence of bulk increased motor activity can develop in the sigmoid colon and the I‘esulting high contractile pressures may cause the formation 7 of an abnormal muscle. This muscle abnormality is the primary factor in the pathology of diverticular disease as seen in Figure l (Painter and Burkitt, 1971). 11- B. m 7% Segmented colon—-one Relaxation of contracting segment producing ring on one side—-contents pressure by contracting move to next segment which is an area of lower pressure 0' W. A... 0- ea Contraction rings halt Outflow obstruction at both feces--contents slowed ends—~diverticula develop and finally halted with due to segment pressure-- resultant pressure diverticula can burst thus Change producing infection Figure l. Pathogenesis of diverticular disease Additions of fiber to the diet in the form of bran were found to relieve the symptoms of diverticulosis in 62 Of the 70 patients for whom this dietary treatment was ppe‘Soribed (Painter g a” 1972). The added fiber produced a. 1ess viscous stool due to the high water absorption capa- City of the bran thereby curtailing segmentation and reducing the contractile pressures . Atherosclerosis. A positive correlation has been found 8 ‘to exist between blood cholesterol levels and atherosclerosis arui both have then been related to a lack of fiber in the (iijat. Studies conducted with human subjects indicate that f23r* populations consuming a low fiber diet and which have high blood cholesterol levels, an increase in fiber intake Significantly reduces these levels (deGroot _e_t 11., 1963). TIWDIB hypothesis for this mechanism involves the increased bile Eiczzici excretion noted to occur with high fiber diets. Bile Eiczzlcis are secreted into the intestine during digestion to aid iilfi» 1:he emulsification of fats after which they are reabsorbed into the system. Bile acids form complexes with the fiber C3<>Inrmonents and the subsequent excretion of these complexes I)I‘events reabsorption. To resupply bile acids necessary for fuI‘ther fat digestion, available cholesterol is converted to t’i—lxe acids. The net effect of this reaction is the reduction C’f‘ 1:he size of the bile acid pool thereby lowering blood (31‘<321esterol levels. Leveille and Sauberlich (1966) showed that: pectin reduced plasma cholesterol levels in rats from 128:3mg/100m1 with a 1% cholesterol addition to a basal diet It(3 3116:5mg/100m1 when 1% cholesterol plus 5% pectin was added. SL1~1'1-daravalli et a1. (1971) found that plasma cholesterol in rats fed a basal diet plus 1.5% cholesterol was increased by 831Tlg/lOOml over the basal diet fed alone while the same basal cti‘at supplemented with 1.5% cholesterol and 20% cellulose decreased plasma cholesterol from 1117 to 91mg/100ml. Colonic cancer. The high incidence of colonic cancer in the Western world has been related to carcinogens produced by 9 bacterial flora action on the bile acids. Fiber in the diet increases bile acid excretion; it also reduces the amount of fecal deoxycholate (Pomare and Heaton, 1973), a substance sus- pected of being carcinogenic in the human colon (Hill _e_t al., 1971). Cancer development depends upon the time and concen— tration of a carcinogen in contact with a tissue. The decreased transit time induced by increased dietary fiber intake limits the time for this interaction to occur and the increased water Content in the colon due to the high water absorption capacity Of the fiber can dilute the concentration of the carcinogen. For these reasons the opportunity for colonic cancer formation is thought to be reduced. Appendicitis. Short (1920) postulated that the increas— ing incidence of appendicitis cases seen in Britain was directly related to a fiber depleted diet. This theory has recently been substantiated by extensive epidemiological observations (Burkitt, 1971). Increased fecal viscosity can result in the deVelopment of fecaliths and segmentation of the appendix. It has been theorized that this obstruction causes increased intra- lumlnal pressures with the resultant effect of bacterial infil- tr“cation to the appendix (Burkitt _et al., 19714). At the turn of the century Van Zavalenburg (19014) hypothesized that bac— tel-.1511 invasion in the appendix gives rise to inflammation and SubSequent disease. Based on the available evidence it would now appear that this hypothesis is what actually does occur (Burkitt , 1971) . ‘ J Mr». “J ‘4 U 5n “3 (L4 {1‘ '{Ln‘l n...'.V* a“ .5, ug‘i ant 10 Crude fiber vs. dietary fiber Crude fiber, the time-honored concept of fiber, is not equivalent to dietary fiber. Furthermore, it is dietary fiber that is believed to contribute to disease prevention (Trowell, 1972). Crude fiber is the material that remains after a food has been treated with sulfuric acid and sodium hydroxide. This remaining material is composed of the struc— tural polysaccharide cellulose and lignin and is what is pre— sent 1y recorded in food composition tables as "fiber." How- ever, dietary fiber encompasses not only the cellulose and lignin but also the other structural polysaccharides hemi- cellulose and pectin. These components can be determined by a number» of methods (Southgate, 1969; Van Soest and Wine, 1967)- Dietary fiber components Qellulose. Cellulose is the major structural poly- saccharide in plant tissues; it is also the most abundant organic material found in nature. Cellulose is acid, alkali, and Water insoluble but is attacked by the enzyme cellulase. Cellulose is a chain structure composed exclusively of EIUCOSe molecules linked. by B—(l-+A) glycosidic linkages. It is believed that cellulose is not a branched molecule but rather is arranged in bundles of parallel chains, or fibers. These fibers are composed of both crystalline and amorphous regions. The molecules of the crystalline area are held together by hydrogen bonds between the hydroxyl groups. ll Portion of cellulose structure A loose arrangement of the molecules is found in the amorphous 1"egions of cellulose; this amorphous region readily adsorbs water and swells which accounts for the increased water absorp- tion capacity of fiber. Lignin. Lignin is the second most abundant organic com- pound found in nature. It is a complex molecule formed by the polymerization of three alcohols--trans—coniferyl, trans— SinaDYl, and trans—coumaryl. This polymerization occurs by free radical formation involving covalent bonding. Softwoods are especially abundant in coniferyl residues, hardwoods pre- dominate in sinapyl residues, and grasses abound in coumaryl residues. Lignin has also been noted to increase in plant tissues as the plant ages. l2 Ho@cmcncwzow CH30 t-coniferyl alcohol cwso HO.CH=CHCH20H CHSO t-sinapyl alcohol HO@CHICHCH20H t-coumaryl alcohol Iiemicellulose. Hemicellulose is found in close asso- ciaCion with cellulose and lignin in the plant tissue though it heis not been clearly defined. Hemicelluloses are synthe- sized! in plants as structural components of the cell wall and are fOund in the middle lamella. 5Phe hemicelluloses have been divided into two categories for'Chescriptive purposes. These categories are hemicelluloses or “fifter—insoluble polysaccharides, and pentosans or water- SQIJHDIG polysaccharides. However, Aspinall (1959) noted that clafiniification according to solubility may lack precision in 13 respect to chemical structure and biological function. None- theless, solubility is a widely used method of classification. By this definition the water—insoluble pentosans are classi— fied as hemicelluloses. The cereal hemicelluloses and pentosans are built of five and six carbon sugar units. The hemicelluloses are composed chiefly of anhydro-D—xylose units and are referred to as the cementing tissue of a plant (Smith and Montgomery, 1959). The monomeric units most common in the cereal pento- sans are D-xylose and L-arabinose. The pentosans are highly H H H H OH H HC) (”1 i1 OH D-xylose L-arabinose branched polymers; they imbibe large quantities of water at room temperature and in turn exhibit high viscosities (Gilles, 1960). Pectic substances. Pectic substances is a general term used to describe complex colloidal carbohydrate derivatives which occur in plants and contain a large proportion of anhydrogalacturonic units. Pectic substances are linear polymers of D-galacturonic acid linked by a—(l-+A) glycosidic 1A bonds. The pectic substances include protopectin, pectinic acid, pectin, and pectic acid. OH COOH o H H H H OH H OH D—galacturonic acid In the chain structure the carboxyl groups are partly esterified with methanol. The hydroxyl groups on C-2 and C-3 may also be acetylated. Many pectic substances have the sugars arabinose, galactose, glucose, xylose, or rhamnose covalently linked as side chains. Bhattacharjee and Timell (1965) found that rhamnose may also occur in the parent chain with the galacturonic acid linked by a-(1-+2) bonds. Dietary fiber in foods The percent dietary fiber in most foods is relatively small. However, from the standpoint of the undigested food residue it is quite significant as shown in Table 2. 15 Table 2. Estimated composition of the cell wall polysaccharides in some foods (Southgate, 1969). Distribution of unavailable carbohydrate + lignin Fiber constituent Wheat Rye Potato Apple Cabbage % % % % % Cellulose 10.0 12.“ 35.8 A1.9 69.1 Lignin 23.3 9.9 32.0 24.8 6.A Hemicellulose 65.0 77.7 32.1 32.3 2A.3 Composition of the wheat kernel Wheat produces dry fruits containing one seed which do not open at maturity to shed the seed. The seed consists basically of the germ, or embryo, and endosperm covered by a seed coat all of which is enclosed in a fruit coat or peri- carp as seen in Figure 2. Milling is the process that breaks down the wheat kernel into separate portions by a series of steps. The kernel is tempered for ease of milling and the bran is then broken off, the wheat germ is flaked, and the endosperm is powdered into different flours possessing varying characteristics. Bran. Ideally, from the miller's standpoint, the bran encom— passes the pericarp, seed coat, a thin layer of nucellar tissue that lies just inside the seed coat, and the aleurone l6 "lll!!.u..‘ GERM OUTER PERICARP ENDOSPERM ) ‘\ CAVITY ' SEEDCOAT A STARCHY ENDOSPERM CREASE Figure 2. Cross section of a wheat kernel, layer, or outer layer of the endosperm. The bran accounts for approximately 1A 1/2% of the wheat kernel (Wheat Flour Institute, 1964). Bran has traditionally been used as an animal feed but recent efforts have been intensified to pro- mote the more extensive use of this mill product as a human food. When the bran and germ are not separated from the endo- sperm the resulting product is whole wheat flour. Whole wheat flour contains more of the essential nutrients required by man than the separated wheat fractions as seen in Table 3. The aleurone cells in the bran account for almost half of the bran weight and most of its protein content (Hutchinson and Martin, 1970). The amino acid pattern of bran indicates a 17 Table 3. Nutrient contribution of some wheat kernel constituents (Wheat Flour Institute, 196“). Nutrient Bran Endosperm Germ % % % Protein 19 73 8 Pantothenic acid 50 A3 7 Riboflavin A2 32 26 Niacin 86 12 2 Pyridoxine 73 6 21 Thiamine 33 . 3 6A lysine content significantly greater than that found in white flour; the biological value of white flour is limited by this low lysine content (Hutchinson gt al., 1962). Crude wheat bran has a lysine content of 656 mg/100 g while all: purpose unenriched white flour has a lysine content of 2A2 mg/100 g (Bowes and Church, 1970). Therefore, if the pro- tein present in bran were digestible the biological value of white flour could be substantially improved by a mixture of the white flour and bran. Hutchinson and Martin (1970) found that addition of bran to white flour in the diets of rats increased protein and essential amino acid intakes. When wholemeal bread was fed to human subjects the proteins of the whole wheat bread were found to be almost completely digested l8 and absorbed (McCance and Walsham, 19A8). In addition to its increased nutrient value bran has been analyzed for percentage fiber constituents and was found to contain 23% pentosans and 21% cellulose (Frasar and Holmes, 1959). Middlings. The middlings are an intermediate mill product. They are that part of the wheat after milling that is too poor for use as a flour but too good for sale as animal feed. The middlings consist of fine particles of wheat bran, wheat germ, wheat shorts, wheat flour, and other materials at the completion of milling, containing not more than 9.5% crude fiber (Wheat Flour Institute, 1965). Middlings differ from shorts only in that the shorts contain no more than 7% crude fiber. Formation of the Layer Cake Structure The formation of the layer cake structure is due primar- ily to the gelatinization of the starch granules present. Howard e£_al, (1968) demonstrated this principle by substi- tuting pure starch for flour in layer cakes. However, other components in addition to starch have been found to be essen- tial for the development of the cake structure. These compon- ents include soluble proteins, surface—active lipids, and poly— valent cations (Howard gt al., 1968; Howard, 1972). These components may be contributed by milk and egg, fat, and egg, milk, and leavening agent, respectively. The functions of 19 other cake ingredients have been described by O'Brien (1972) for emulsifiers, Wilson and Donelson (1963) for water, and Hunter gt gt. (1950) for sugar. A properly mixed batter is an oil-in-water emulsion. The continuous phase is an aqueous sugar solution with suspended solids and gases (MacDonald, 1968). Most of the air is found in the continuous phase. Upon baking the cake batter changes from a fluid emulsion to a solid porous structure due to the gelatinization of the starch granules. In order to form a structure that will not collapse or shrink after removal from the oven the free water must be absorbed. However, in a high- ratio cake formulation (1A0% sugar based on the weight of the flour) the sugar competes with the starch for the available water and gelatinization may be impaired. Matz (1960) suggests that total liquid in the system should exceed the amount of sugar by 25 to 35%. This amount should be sufficient liquid to dissolve the sugar and provide for adequate starch gela- tinization. Miller and Trimbo (1965) found that early starch gelatinization during baking prevents the formation of a sunken center in the layers. A number of researchers have studied the effects of the various fractions of the flour in the cake system follow- ing a flour fractionation procedure. Many methods have been employed for flour fractionation (Sollars, 1958; Donelson and Wilson, 1960) but all methods give the same four fractions as a result of the process. These fractions include starch, gluten, water—solubles, and tailings. The tailings fraction 20 is composed of damaged starch granules, proteins, and insol- uble pentosans. MacMasters and Hilbert (19AA) found the tailings fraction to be composed of 87 to 9A% starch, 1 to 2% protein, 4% pentosan, 0.7% fatty compounds, and 0.3% ash. Yamazaki (1955) dispersed the tailings fraction in water and then fractionated this dispersion through a 325—mesh sieve to obtain "purified tailings." The purified tailings were found to contain 30 to 65% insoluble pentosans, 7 to 23% starch, and 0.5 to 3% total nitrogen. Studies have shown that while the soluble pentosans have an improving effect in dough systems, it is the water-insoluble pentosans that show an improving effect in batter systems. Gilles (1960) found that water-insoluble pentosans added to a marginal quality soft wheat flour produced a yellow cake with quality characteristics more desirable than those of the control cake. Donelson and Wilson (1960) determined that the tailings fraction added to cake batters produced a cake with a significant increase in volume and improved grain and texture. Baldi gt gt. (1965) obtained similar results when they found increasing amounts of tailings up to an optimum level improved cake volume and internal structure. This optimum level was determined to be 9 to 15% of the weight of the flour for a high—ratio (125% sugar based on flour weight) cake formulation and 18% of the weight of the flour for a low-ratio (100% sugar based on flour weight) cake formulation. The optimum amount of tailings was significantly less for the high-ratio than for the low-ratio cakes. The reason for this 21 result was theorized to be due to a greater swelling of the water-insoluble pentosans at the higher sugar-water level, compensating for the smaller amount of this fraction. This resulted in a stronger and more stable matrix. A high batter viscosity seems to be required to hold the shape and structure of the cake during baking (Miller and Trimbo, 1965). There- fore, any change in the cake formulation that allows for an increase in the batter viscosity will result in an improved baked product. The addition of the water-insoluble flour fraction to a flour used in cake baking may be such a formu- lation change. EXPERIMENTAL PROCEDURE This research was initiated to determine whether the fiber constituents of the wheat kernel could be satisfactorily substituted for white flour in a layer cake formulation. Pre- 1iminary experimentation indicated that partial substitution of bran and middlings for flour in a layer cake produced an acceptable product. In order to determine the effects of these substitutions all factors known to affect cake quality were carefully controlled. Ingredient Procurement Common lots of sugar, salt, shortening, non-fat dry milk, and baking powder were obtained from the Michigan State University Food Stores. The baking powder was special ordered to insure freshness. Common lots of soda, cocoa, vanilla, ground cloves, cinnamon, and mace, almond extract, chopped walnuts, brown sugar, and dried buttermilk were obtained from local commercial establishments. Bran, middlings, 50 and 70% flour extractions, and clears were obtained from Mennel Mills, Fostoria, Ohio. A partial proximate analysis, as well as pH, of the bran, middlings, and flours were also obtained from Mennel Mills and is found in the Appendices. Dried egg whites and dried whole eggs were procured from Seymour Foods, Inc.; canned mashed bananas were donated by Gerber Products Company. 22 23 Layer Cake Formulations A standard high-ratio white cake formula and four flavored cake formulas including nut, spice, banana, and chocolate were prepared. Five replications of each variable were evaluated by both objective and subjective measurements. Variables prepared for evaluation in the white layer cake series included a control and substitutions of flour with A, 8, and 16% bran, 12% bleached midds, 12% unbleached midds, 12% bleached midds plus 16% bran, and 12% unbleached midds plus 16% bran. The white cakes were prepared from a 60% extraction flour using the standard AACC formula, method 10-90 (Table A). The series was then repeated using a 70% extraction flour. Variables prepared for evaluation in the flavored cake series included a control and substitutions of flour with 16% bran, 12% unbleached midds, and 16% bran plus 12% unbleached midds. The flavored cake formulations were adapted for use in this study from popular cookbooks (Table 5). The flavored layers were prepared using a 70% extraction flour only. Method of Preparation The bran and middlings were ground to the same particle size as the control flours to eliminate mouthfeel differences using a Udy Cyclone Sample Mill, model MS. Particle sizing was done using a Roto-Tap Testing Sieve Shaker, model H589 (Donelson and Yamazaki, 1972). The resulting particle size 2A Table A. Formula for standard high-ratio white layers. Amount Ingredient % (flour g basis) Flour 150.0 100.0 Sugar 210.0 140.0 Shortening 75.0 50.0 Non-fat Dry milk 18.0 12.0 Dried egg whites 13.5 9.0 Salt ".5 3.0 Water 232.5 155.0 Baking powder1 7.87 - 8.25 5.5 - 5.75 l as outlined in AACC method 10-90. Added according to schedule based on barometric pressure 25 .mzfi .wm .xpom 3oz ..oo xoom Haemizwhooz .HmmH .xoonxooo mthoem m.pmxoopo mpummm .5m ocm em .mmm .xeow smz .mmme .xoonxooo mmxmo ocm moem memopmw out mosom emppmmm .5ee .mo .xso» 3oz ..ose .omsom soosmm .mmme .xoooeooo m.eesooze o.m5e o.mem m.:oe o.o:e m.mme o.ome e.mme o.owe poem: omeeeumeo o.m m.m pomepxm weeecm> 3.0 m.o pomepxm ocose< 0.2m o.em messes: omooono e.o e.o moms ocsoew 5.0 o.e cosmcceo ocsoew 5.0 o.e mo>oeo ocsoeo m.mm o.m5 powSm csoem m.5m 2.55 mcmcmo omammz m.mm e.mm moooo m.e o.m o.e m.e e.m 5.m gonzoo wcexmm m.m e.m :.e m.e m.e :.e moom m.m m.: m.m o.m m.m o.m m.e m.e pewm o.me o.em e.me o.ee gees ass omeusoz . 5.0: o.mm o.eo xeesemppso ocean m.me e.me 5.oe o.:e e.me :.me e.me mwm ooeeo meonz m.mm m.o5 5.mc m.em 5.mo m.me e.mo msesooooem m.w5e o.mmm 5.mme o.5w o.ooe e.moe :.mme Lowsm o.ooe o.mme o.:me o.ooe o.mee o.ooe :.mme psoem nsoem m m ezoem e esoewie m esoem e m acmeoopwcH mopmeoooco oodmm epsz .moxmo Loewe ooeo>wem Lou mcoepmesseom .m memos 26 values are cited in the Appendices. Flour Preparation A 60% extraction flour was prepared for use in the white cakes by mixing 5 parts of 50% extraction flour with 1 part of clears for 30 minutes in a Liquid-Solids Blender, model SAAEXAK-989. The 70% extraction flour from Mennel Mills was used in the second white layer cake series. Due to an insufficient quantity of 70% extraction flour for use in the flavored cakes a 70% extraction flour was prepared for use in this series by combining 5 parts of 50% extraction flour with 2 parts of clears. This mixture was divided in half and each half mixed for 30 minutes in a Hobart mixer, model K-200. To insure adequate mixing half of the contents of each portion of this flour were remixed together for an additional 30 minutes. To insure an even mixing of the different particle sizes of the bran and middlings these were also mixed in the Liquid-Solids Blender for 30 minutes before preparation of the flours for the variables studied. Enough flour for each variable was prepared at the same time. The premixed flour, bran, and middlings were weighed to the nearest 0.1 g and mixed in the Liquid-Solids Blender for 30 minutes to insure an even distribution of the flour components. 27 Cake Mix Preparation The dry ingredients for each cake layer were preweighed to the nearest 0.1 g and stored at 1.700 until used. Cake ingredients for the white layers were stored in glass beakers and the ingredients for each flavored layer were packaged in polyethylene bags with the exception of the brown sugar and chopped nuts which were packaged separately. The mashed bananas were weighed to the nearest 0.1 g in glass sample jars and frozen at -17.7°C until the day before use. The frozen bananas were thawed at 8°C for 2h hours and warmed to room temperature prior to use in the cakes. Baking powder, shor- tening, vanilla, and almond extract were stored at room tem- perature in closed containers and weighed on each day of use. All dry ingredients were sifted into the mixing bowl. White Cake Preparation The white cakes were prepared according to AACC specifi- cations, method 10-90. The preweighed dry ingredients, shor- tening, 60% of the distilled water, and baking powder were mixed in a Kitchen Aid mixer, model K5-A at low speed (1A5 rpm) for 1/2 minute. The bowl was scraped and mixing was continued for an additional A minutes at medium speed (2A9 rpm). Half of the remaining water was added and the batter mixed for 1/2 minute at low speed, scraped, and mixed 2 minutes at medium speed. The remaining water was added and the batter was again 28 mixed for 1/2 minute at low speed, scraped, and then mixed for 2 minutes at medium speed. Flavored Cake Preparation All flavored cakes were mixed using the same mixer as that used for the white layers. Nut Cakes. The preweighed dry ingredients, shortening, and llO—milliliters of distilled water were mixed at low speed for 1/2 minute. The bowl was scraped and the batter was mixed an additional 2 minutes at medium speed. The remaining water and almond extract were added and the batter mixed at low speed for 1/2 minute, the bowl was scraped, and the batter mixed 2 additional minutes at medium speed. Finally, the nuts were added and the batter mixed 10 seconds at low speed. Spice Cakes. The preweighed dry ingredients, shortening, brown sugar, and loo-milliliters of distilled water were mixed at low speed for 1/2 minute. The bowl was scraped and the batter mixed for 2 minutes at medium speed. The remaining water was added and the mixing process was repeated. Banana Cakes. The preweighed dry ingredients, shorten- ing, thawed mashed bananas, and 90-milliliters of distilled water were mixed at low speed for 1/2 minute. The bowl was scraped and the batter was mixed for 2 minutes at medium speed. The remaining water was added and the mixing process was repeated. Chocolate Cakes. The preweighed dry ingredients, shor— tening, and 118-milliliters of distilled water were mixed at 29 low speed for 1/2 minute. The bowl was scraped and the batter was mixed 2 minutes at medium speed. The remaining water and vanilla were added and the mixing process was repeated. Baking Procedure Weighed to the nearest gram, A00, A25, and A50 g of batter were scaled into ungreased 8 in round two piece cake pans for the chocolate, white, and remaining flavored layers respectively. The cake layers were placed on the middle shelf of an Etco forced convection oven, model 186A, and baked for 25 minutes. The oven temperature was maintained at 1781300 by a Honeywell versatronic controller. After removal from the oven the cakes were allowed to cool in the pan for 30 minutes. They were then removed from the pan onto racks and allowed to cool an additional 30 minutes. Objective and subjective evaluations for each layer were obtained on their respective baking day. Preparation of Samples After volume measurements were obtained the layers were cut for evaluation (Figure 3). Specially manufactured cutters were used to cut samples for shear press evaluations. All samples were wrapped in ReynologaFood Service Film to prevent dehydration. The cake remaining after samples were removed was ground in an Osterizer blender on high speed for 30 Color 0111 [Tflices for Sensory White Cake [—Evaluation Series J Tenderness Moisture and pH Edges ground for mnisturz / Pieces for Sensory Flavored Cake Evaluation Series / , i \\ / \¥// Figure 3. Cutting Diagram for Cakes 31 10 seconds for use in pH and moisture determinations of the white layers. In the case of the flavored layers all remain- ing cake was ground and both moisture and color determinations were done on ground samples. Objective Measurements Objective measurements were used to evaluate the quality characteristics of both the batter and the cake for the white layers and the cake for the flavored layers. Batter quality measurements included viscosity, specific gravity, pH, and color. Cake quality characteristics determined were pH, crumb color, volume, tenderness, compressibility, and moisture for the white layers, and crumb color, volume, and moisture for the flavored layers. In addition to these determinations the 16 flours used in the white cake series were analyzed for crude fat content and alkaline water retention capacity and were fractionated into A fractions to determine gluten, prime starch, tailings, and water—solubles. Batter Viscosity. Batter viscosity was determined using a Brookfield Viscometer, model RVF-lOO, equipped with a No. 7 spindle rotating at 10 rpm. The spindle was submerged in a 250—milliliter sample in a SOC-milliliter beaker. The read— ing was taken after the dial was allowed to make one complete revolution and the reading was then multiplied by the appro- priate conversion factor to express viscosity in poise. Specific Gravity. Batter specific gravity was determined 32 by comparing the weights of equal volumes of batter and water at room temperature using the method of Platt and Kratz (1933). EH. The batter and cake pH were determined using the method of Ash and Colmey (1973). gglgt. The Hunter Color Difference Meter, model D25, equipped with a spherical head was used to determine the color of both crumb and batter samples. The instrument was standard- ized before use against a white tile (L=9A.8; aL=0.7; bL=2.7) for the white layers, a yellow tile (L=83.0; aL=-3.5; bL=26.5) for the nut, spice, and banana layers, and a grey tile (L=21.5; aL=-l.7; bL=-0.l) for chocolate layers. One reading was taken for each batch of batter and each cake layer for the L, aL, and bL values. Volume. Cake volume was determined using the AACC method 10-91 to obtain shrinkage and volume, symmetry, and unifor- mity indices. Tenderness. Tenderness values were determined using the standard shear-compression cell of the Allo-Kramer Shear Press, model SP12, equipped with an electronic recorder, model E2EZ. Cake samples 5.73 cm square were weighed to the nearest 0.1 g and placed in the cell. The 3000—pound proving ring and a range of 5 pounds were used for each measurement. The cell assembly was cleaned between each determination. The tender- ness value expressed as pounds force per gram for each layer was determined by a single measurement calculated according to the formula, 33 Reading x Ring x Range Tenderness= Sample Weight x 100 x 100 Compressibility. The 3000-pound proving ring and a range of 5 pounds were used for this determination. The height of each cake sample was determined by averaging readings from a vernier caliper taken at 3 different points. The diameter remained constant at 5.23 cm. The sample was placed at the base of the shear press column, centered, and depressed by a flat plunger 5.23 cm in diameter to a uniform thickness of 0.7 cm. Each compressibility determination was calculated from a single trial and recorded as pounds force per cm com- pressed by substituting cm compressed for sample weight in the formula used to calculate tenderness. Moisture. Cake moisture was determined by drying 2 g samples, weighed to the nearest 0.001 g, for 5 hrs at 90°C under a vacuum of 27—in of Hg in a Hotpack vacuum oven, model 633. The samples were reweighed after cooling in a desiccator. Percentage moisture was calculated according to the formula, Wt. of moisture lost (g) % moisture: Original sample wt. (g) X 100 Crude Fat. Percentage crude fat in the flours was determined using acid hydrolysis followed by petroleum and ethyl ether extraction according to the AACC method 30—10. Alkaline Water Retention Capacity (AWRC). AWRC as an indication of flour hydration properties was done on the flours by researchers at the USDA Soft Wheat Quality Lab, 3A Wooster, Ohio according to the method by Yamazaki gt gt. (1968). Flour Fractionation. The flours were fractionated into gluten, prime starch, tailings, and water-solubles by researchers at the USDA Soft Wheat Quality Lab, Wooster, Ohio according to the method by Donelson (197A). Subjective Evaluation Training sessions were held prior to taste panel evalu— ation to acquaint the panel members with the score cards. Sample score cards appear in the Appendices. White Layers The White layers were evaluated by a panel of 5 trained judges for internal characteristics following the method out— lined in the AACC method 10-90. Flavored Layers The flavored layers were evaluated by a 12 to 15 member taste panel composed of faculty, staff, and graduate students in the Department of Food Science and Human Nutrition. Eight of these members were trained and participated in all evalu- ations. The cakes were examined for color, moisture, tender- ness, texture, flavor, and general acceptability using a 10- point scale with 10 indicating the best value. Additional comments were noted. 35 Analysis of Data The data were analyzed for variance and Duncan's Multiple Range Test (1957) was then used to sort out differences revealed by the analysis of variance. RESULTS AND DISCUSSION This study was designed to determine the effects of sub- stituting various levels of wheat bran and midds for flour on the quality characteristics of a standard high—ratio white layer cake and four flavored cakes. Objective and subjective data were examined to ascertian the effect of these substitu— tions. All numerical data were subjected to analyses of vari- ance and significant differences were pinpointed by use of Duncan's Multiple Range Test (1957). Tables of replicate means and standard deviations, as well as analysis of variance tables, accompany this discussion. White Layers A white layer cake formulation was chosen for fiber substitution studies because changes in the quality character— istics of a white cake are the most stringent test of flour performance. White layers were prepared with 60 and 70% extraction flours using the American Association of Cereal Chemists standard formula and preparation method. 36 37 Objective Measurements Batter and cake quality characteristics were determined by employing a number of different objective tests. Batter and Volume Parameters. Specific gravity, batter viscosity, and volume index means and standard deviations are presented in Table 6 and a summary of the analyses of variance for these parameters is presented in Table 7. Specific gravity is an indication of the amount of batter aeration. The desir- able specific gravity of a layer cake is 0.65 to 0.75 (Ellinger and Shappeck, 1963). A specific gravity below this range is indicative of an overaerated batter while a specific gravity above this range is indicative of a dense batter which may result in a decreased volume. All white cakes, prepared by using the AACC standard formula and method, showed a specific gravity slightly above this range. However, substitutions of bran and/or midds in the cake formula revealed no significant changes in batter specific gravity. A comparison of variable means for viscosity disclosed that cake batters prepared with increasing fiber substitution levels resulted in a significantly greater (p<0.001) viscosity. This increased viscosity could be due to the increased water absorption capacity of the fiber constituents as a result of their large molecular size (Gilles, 1960). Viscosity was also significantly greater (p<0.001) in the batters prepared with the 70% extraction flour. Collins and Sunderline (19A0) demon- strated that a definite relationship exists between batter 38 eoece ee one so possesses sepssoeoesmem go: who poppoe .esmme .smossov eeeeeomooso so mean one an ompoemomemosm mommpm> wEoV xoocH osseo> Anew zuemoome> mpe>meu cameomQWI xooCH mezeo> com .zpemoome> .mpe>mew oemeooom pom .mesoem coepomepxm $05 one om cues commompo memmme been: eo mmzem> H mcoepme>oo oemosmpm osm mcmoz .c oeome 39 .mpHHHQmQOLQ mo Hm>ma ea.o msp om osmoeeesmemesa oe.o o5.mw 00.0 no seapHB 50.0 examm.mem 00.0 e ooeoomooose :m.o *xxmm.emo: 00.0 e esoem ***:q.o ***©m.::MH 00.0 5 mmHQmHLm> me.o am.5om oo.o m5 empoe xmocH msseo> mnemoome> mpe>meu cemeomom no meadow mwhmsvm com: .mp30em coepomppxo e05 ocm om cues omemooea moxMQ Loewe opens mo mmerCH msseo> Ucm .mpemoome> .zpe>wpw oeheomom mo mocmepm> mo mommemc< .5 canoe A0 viscosity and cake structure. They found that thin batters were not viscous enough to hold the air incorporated during mixing or the gas liberated by the baking powder. Gas bubbles were found to be small, numerous, and evenly distributed in more viscous batters. No significant differences were revealed for the volume indices of cakes prepared with 60 or 70% extraction flours but significant differences (p<0.001) did occur among variables. Substitutions of bleached midds for flour in two variables resulted in decreased cake volumes. However, significant dif- ferences were not noted among the other substitution levels. The effects of chlorine bleaching on wheat flour have been widely studied. The reaction of chlorine with flour produces three distinct changes: bleaching of the flour pigments, reduction of the pH, and chemical modification of the flour components (Kulp, 1972). WhiStler and Pyler (1968) studied the effects of bleaching on isolated hemicellulose A. Hemi- cellulose is a constituent of dietary fiber found in both bran and midds. These researchers found that hemicellulose is rapidly depolymerized by chlorine and the rate of this reaction is most rapid with gaseous chlorine. The depolymerization results in a breakdown of high molecular weight compounds and may decrease the water absorption capacity as a result. Insufficient liquid absorption in the cake batter may result in decreased volumes or sunken centers in the cakes. Shrinkage and Symmetry and Uniformity Indices. The mean values for shrinkage as well as the symmetry and uniformity Al indices are presented in Table 8 and the analyses of variance for these data are presented in Table 9. The shrinkage is calculated as the difference between the diameter of the interior of the cake pan and the cake diameter after baking. The term "symmetry" has traditionally been used in the cake industry to indicate cake contour and "uniformity" is a measure- ment of the cake symmetry. Fiber substitutions for flour do not appear to affect these cake quality characteristics since the analyses of variance revealed no significant differences between flours or among variables. Tenderness and Compressibility. Means and standard deviations for tenderness and compressibility scores are presented in Table 10. Tenderness values analyzed for vari- ance revealed very highly significant differences among variables while analyses of compressibility values revealed significant differences (p<0.05) among variables and (p<0.001) between flours as seen in Table 11. Tenderness increased with increasing fiber substitution levels in the cakes. As the fiber substitution levels increase the total amount of starch, as well as the gluten-forming proteins, are diluted. This dilution of the basic components of the layer cake structure may result in a less rigid cake and, therefore, a more tender cake (Howard gt gl., 1968). As would be expected, cakes prepared with the 60% extraction flour required less force for compression than those prepared with the 70% extrac- tion flour since the 70% extraction flour is a stronger flour. However, no trend was established to relate compressibility to A2 .mcoepmoeeooe m co oommm H I I I I I I venomoeoso H.+H.0 m.+0.0 m.+0.0 m.+H.0 m.+:.a 0+m.H move: ewe + cmem ewe I I I I I I Cwfiommflm H.+H.0I H.+0.0 m.+H.0 m.+m.0I m.+H.H 2.+N.H move: ema + swam ewe .l . .l . .l . .l . .l . .l . omsommabco e +0 0 e +0 0 m +m 0 m +m 0 e +m e m +e e nooez ewe .l . .I . .l . .l . .I . .I . omnowmem e +0 0 e +0 0 m +e o m +e 0 m +m e m +e e move: ewe e.Ho.o m.uo.o N.He.o m.um.o e.um.e m.Ho.e seem ewe oHo.o e.He.o e.Ho.o e.Hm.o e.Hm.e m.He.e seem em e.Ho.o e.Ho.o m.Hm.o m.He.o N.He.e m.Hm.o seam e: e.Ho.o e.He.o m.He.o e.Hm.o m.Hm.e e.Hm.e eososoo Ie05 R00 R05 .M00 Ie05 e00 menmeem> xoocH mmespomeco xoocH hepossmm uwwaHenm .mesoem coepomepxm R05 0:0 00 ape: omemooea mmxmo Loewe opens 00 wooeoce mpeseomec: 0:0 mepmssmm 0cm mmmxcHezm pom Hmcoepme>mo cementum cam mcmoz .0 meome A3 H0.0 0H.0 mH.H no Cecpez m0.0 0e.0 00.0 5 coepompmpce No.0 me.0 m:.0 e ezoem e0.0 em.0 me.0 5 mmeomeem> e0.0 HH.0 00.0 05 empoe xmocH mpeEpOMHCD xmoce zemeESm mwmxsepnm 00 mopsom mmemsvm com: .mezoem coepomeuxo e05 0cm e00 hues omemomeo moxmo Loewe opens mo mmoeoce mpeEeomec: 0cm zepmeszm 0cm ommxseehm mo mocmehm> mo mommemc< .0 pence AA .A5mmH .cmossov mpeHHomnoeo mo Hm>mH em on» pm pcohmeeeo mecmoHestHm no: ohm eouuoe meme hue: oopoHeomhmozm mommem>¢3 .eewee .seossmv aoeeeoesoso so eesee eH map pm pcoemeeHo mecmoHechHm pod one poppmH oEmm csz oopoeeomeoQSm mommem>< e (I I I I eesoeeeoss NahaNHm.+mN.: 30m.+mm.m 00H.+m:.H omo.+wm.H mUUHS RNH + Sahm NwH I I I I oesoeeem e.e.eem.+ee.e ewe.+ee.m eeo.+em.e ome.+em.e eeees eee + seem ewe .I . .I . .I . .I . eesoeeeoss x.zms +em m e.x.3ew +ee m ome +wm e ome +ew e move: eme .I . .I . .I . .I . oesoeeem haXa3Nm +©O m xn3m: +w© m DOH +H© H QmH +2® H mUUHE fiNH x.3ew.nwm.w x.2ww.uw5.w owe.uww.e ome.uew.e sesm ewe e.eom.nee.e e.s.smw.umo.w ome.uww.e owe.Hm5.e sesm ew New.uem.e on.zww.Hee.w owe.Hwe.e owo.Hew.e seem es e.s.3ww.uwm.w see.uw5.w emm.Hwe.e eee.wa.e eosesoo eoe eow eoe eow prHHhemmopQEoo mmocemocoe oHanem> .mtone COHpomepxm e05 02m ow cue: oohmuoem mmxmo emmmH mpan eo mcoeume>mo cemocepm one meme: .0H mHomB mo3Hm> mpeeebemmohosoo 0cm mmmcemocmp pom H HS .zpeeepmnopa mo em>me ee.o ms» pm pcmoeeeswem *** .epeeenmnopa mo em>me em map pm pcmueeecmem* mm.o No.0 no Cecpez e:.o mo.o w soepompmpCH ***wm.w mo.o e Lsoem *mh.o ***me.o n meanLm> 23.0 no.0 me eMpOB mpeeenemmmpasoo mmmcpmvaB aw mopsom mmmmswm 2mm: .mpsoem coepompuxm eow Ucm om cue: vmpmqmpa mmxmo pmmme muesz mo mmzem> mpeeepemmmpqsoo new mmmcpmncmp mo mocwepm> mo mmmmemc< .ee menme U6 percentage fiber substitutions. Compressibility is the amount of force necessary to compress a given size slice to a uniform height. High standard deviations were found for this measure— ment and may reflect differences in the textures of the layers. As a result, the compressibility measurement may not be indi- cative of cake quality. Batter and Cake Color. Means and standard deviations for the L, a and bL values are presented in Table 12 and L’ the analyses of variance for these data are presented in Table 13. Similar tables containing data for the cake color scores are presented in Tables 1“ and 15. No significant differ— ences were noted between flours for any of the batter measure- ment color scores but very highly significant differences were revealed among variables for the L, aL, and bL readings. As the percentage fiber substitution increased the L or lightness value decreased while the aL or redness and the bL or yellow- ness values increased. Similarly, no significant differences were revealed between the L and aL color measurements of cakes prepared with either 60 or 70% extraction flour but a signi- ficant difference was revealed between the bL values (yellow- ness) of cakes with the two types of flour. Very highly significant differences were revealed among variables for the L and aL values while no significant differences were revealed among variables for the bL measurement. As the percentage fiber substitution increased the lightness of the cake decreased while the redness increased. Cakes prepared with 60% extraction flour exhibited lower bL values than the 70% “7 .eemme .ceoczov epeeepenoee no eo>oe ee en» pe pnonommeo meaneoeeenwem won one nmppme eEem en» an ompnenomnmnnm memenm>< e I I I I I I emeoeoenc: em.e+o.ee ee.+e.oe m.em.+w.eI m.ee.+m.eI m:.e+w.oe em.+o.ee meeez ewe + nenm ewe I I I I I I UmfloflmHm em.+e.oe em.+e.oe em.+m.oI me.+m.OI em.+e.ee mo.e+e.oe meeez ewe + Sewcem RmH .I . .I . . I . I .I . I . I . .I . eenoeeepco Unoz +5 0 Ugo: +H 0 Uaonflm H+® H Unoaflm +m N UqOH N+m ON Unom +0 m5 mUDHZ RNH . I . .I . . I . I .I . I .I . .I . emnoemem o.pm w+m oe 0.9m +m oe De e+m w De +e w 0.9m +e me 0.9e +m me mace: ewe e.omeeoee e.oe.ue.m m.e.oo.eee.eI o.e.om.um.eI m.ee.ew.me e.em.ue.me seem eee pm.em.m pe.uo.m o.pm.Hm.mI o.nm.um.mI o.pe.He.ee o.ne.ue.ee seem em pe.em.m pe.ee.w ee.em.wI pe.em.wI ne.em.me pe.em.me seem e: ee.ue.e em.ue.e ee.ne.mI em.ue.mI em.uee.m em.uw.wm eoepcoo eon eom eoe .Ieom ecu eomII meneene> emmenzoeeemw en, Ammonooan ee emmecpemeeo e exec nmzee evens mo monee> one .mmsoem nOMpoenuxm eon one om nae: omnenmnn heaven 9 . e .4 now ewnoeuee>mo oneonepm one mneoz .me menee .zpeeeneQOLn mo em>me ee.o en» we pneoemenmem*** h8 25.0 02.0 om.m :0 Cenpez m:.o em.o no.0 e noepoenmpne mm.m mo.o om.H e Lnoem ***em.ze ***mm.n ***em.mme e mmepeepe> mm.e me.e am.be mm eMpOE en ee e no monnom memesom new: .mnnoem nOepoempxm eon one om npes omnenmna whoppen exeo nomee open: mo mmnee> a one . e «e no moneene> mo memmeen< .me menee H9 .Aemme .neonnav monee> ee one A one now zpeeenenonn mo eo>oe ee on» we pnonoemeo meaneoemenmem won one nonpoe oEem nee: oopnenomnoanm moweno> en ammonoon noe monee> ee+ .mmonoonw noe mosee> eeI .mmonpnmee pom monee> em .mnoepeoeenon m mo oweno> .IHD new A e .mnnoem noepoenpxo eon one ow npez ooneqona moxeo noeee opens mo moonee> 9 one .ee .4 now emnoepee>oo oneonepm one mneoz .ne oenee 50 .eeeeesesone eo eo>oe ee.o esp se use .eeeeesesoss no ee>ee ee esp ee use seneswemeee oemenwem ** ww.w em.w mm.m ew seseez mm.m m5.e m5.m 5 noepoenopne *ewm.ew em.m V mm.w e seoee 55.: ***mm.om ***wm.ee: 5 oeneehe> eo.m ee.e wm.mm me eesoe en ee 4 mo oonnom monenom neoz .mnnoe noepoenpxo e05 one ow nee: oonenonn moxeo noeee opens mo monee> 9 one .ee “A mo ooneene> mo momeeen< .me ereB 51 extraction flour. The bran and midds used in these substi- tution levels were from a soft red wheat. Therefore, the bran and midds were dark and brownish-red in color which accounts for the decreased lightness and increased redness measurements in both the batter and the cake. The brown color imparted to the batter and cakes also accounts for the increased yellowness scores. Moisture and Alkaline Water Retention Capacity. Means and standard deviations of percentage moisture and the value for the AWRC as determined by the USDA Soft Wheat Quality Laboratory are presented in Table 16. The analysis of vari- ance of the percentage moisture data is given in Table 17. No significant differences were revealed by the analysis of variance among the variables and no trend for moisture reten— tion appeared to be established. A highly significant differ- ence did occur, however, between flours with the cakes pre- pared with the 60% extraction flour being more moist. AWRC values indicate that the flour retained more water as the levels of fiber substitution increased. It has been established that the fiber components of wheat have a high water absorp- tion capacity (Gilles, 1960). This increased water absorption capacity correlates with the increasing AWRC values with increasing fiber substitution levels. Although the cakes prepared with increased fiber levels did not Show an increase in moisture retention this could be accounted for by the fact that sugar and starch act in competition for the water thereby preventing the fiber components from being hydrated to their 52 .memen onnpmeoE eee no epeom .mnOepeoeenon m wo omeno> .mnnoew noepoenpxo e05 one ow npez oonenonn moxeo nomee opens now om3< now onee> nOepeneEnopoo oewnem one onSpmeoE oweunoonon now emnoepee>oo oneonepm one oneoz .we oenee 53 Table 1?. Analyses of variance for percentage moisture of white layer cakes prepared with 60 and 70% extraction flours. Mean Square Source df % Moisture Total 79 l.U2 Variables 7 1.18 Flour 1 1u.79** Interaction 7 1.37 Within 6“ 1.2“ ** Significant at the 1% level of probability. 5A fullest potential. pH. pH means are presented in Table 18. Although these values were not converted to the logarithm form to permit an analysis of variance the fiber substitutions for flour in the cake formula did not appear to affect the pH of either the batter or cake nor did the pH appear to be affected by the type of flour extraction used in the cakes. Dietary Fiber bnyifference. The percentage dietary fiber in the two flour extractions used to prepare the eight variables was calculated by difference as seen in Tables i9 and 20. Moisture, starch, protein, crude fat and ash were calculated as a percentage and subtracted from a total 100% to give an indication of the amount of dietary fiber contri- buted. All data was calculated on a 1U% moisture basis. Raw data used in these calculations for the flour fractionation and crude fat values are found in the Appendices. As the fiber substitution levels for the flour increased the dietary fiber contribution of the flour also increased. However, difficulties were encountered in two areas in the fractionation procedure used for the flours. First, the substitution level of 16% bran plus 12% midds did not frac- tionate well and the starch recovered at this level was of doubtful purity. Secondly, at this level of substitution the gluten dispersed upon hand kneading and a part of the tailings fraction adhered. Apparently this substitution level altered the density gradient thus making the procedure unsatisfactory for use. Although no data for starch is 55 Table 18. Mean values1 for white cake batter and cake pH prepared with 60 and 70% extraction flours. Batter Cake Variable 60% 70%—8 60% 70% Control 7.1 7.1 7.1 7.0 A% Bran 7.1 7.1 7.0 6.9 8% Bran 7.1 7.1 6.9 6.9 16% Bran 6.9 7.1 6.8 6.9 12% Midds Bleached 7 l 7 l 6'9 6 9 12% Midds Unbleached 7 O 7 1 6'9 6 9 16% Bran + 12% Midds 7.1 7.1 6.7 6.7 Bleached 16% Bran + 12% Midds 7.1 7.0 6.8 6.8 Unbleached 1 Average of 5 replications. 56 .mpennn enwunnoo we we: noepenoepoenw nnoew en oono>ooon nonepme eeseeeesss IIII e.w m.m w.ee IIII w.ee eseez ewe + nenm ewe eeseeeem IIII e.w e.w w.ee eIIII w.ee eesez ewe + seem ewe . . . . . . oonoeoenns e we e e w w w oe w mm o we eeeez ewe . . . . . . oonoeoem w ow e e w w w oe w em 0 we eweez ewe e.wm w.e e.w w.ee e.wm w.ee seem ewe w.ee o.e m.w e.we e.ww w.ee seem ew w.ee e.w m.e e.w w.ww w.ee seem ee o.e e.w m.e e.w m.me w.ee eoessoo nonew mnepoea nm< new oonno neoponm nonepm onnpmeoz .nnoew noepoenuxo eow ne oononowweo an wnoeueneEnouoo nonew enepoea .me oenee 57 .mpensa enwpnnoo wo we: noepenoepoenw nsoew en Umhmweoomce SOcempmH seseeeesss IIII e.w e.w e.ee IIII w.ee ences ewe + seem ewe emceeeem IIII e.w 0.: w.ee IIII w.ee eeeez ewe + nenm ewe . . . . . . eeseeeessm w ee e e e w w 0e w em 0 ee eweez ewe . . . . . . eesoeeem e ee e e w w w 0e e em 0 ee essez ewe w.wm w.e e.w w.ee w.wm w.ee seem ewe w.we 0.e e.w m.0e w.ww w.ee seem ew e.w e.w m.w e.we m.mw w.ee seem ee e.w e.w m.e e.w w.we w.ee eoeesoo e e e e e e eeseeee> nonem mnepoeo nee new oosno neoponm nonepm onnumeoz nnoew noepoenpxo e05 ne oononowweo an mnoepeneenopoo nonew mnepoea .om oenee 58 available to calculate dietary fiber contributed at the high— est substitution levels it would appear logical to assume that the percentage dietary fiber contributed would be great- est at this level. Dietary fiber by difference is an estima- tion of the level of dietary fiber present. This method may, however, overestimate the available carbohydrate or under- estimate the structural polysaccharides of the plant cell wall. Some researchers have developed methods to report carbohydrate by determination (Southgate, 1969; Van Soest and Wine, 1967). It may also be possible to determine the fiber polysaccharides by gas-liquid chromatography (Sloneker, 1971). The method of estimating the dietary fiber contributed by difference was chosen for this study over an analytical method of fiber determination because the primary purpose of this research was to observe the effects of fiber on cake quality rather than to study the amount of dietary fiber in terms of the individual constituents of fiber provided by such substitutions. Sensory Evaluation Sensory scores were determined by use of the AACC layer cake score card, method 10-90. This method called for a total sum of all individual characteristics evaluated and this sum total was then evaluated as an indication of cake quality. Quality characteristics evaluated to comprise this total score 59 included cell uniformity, cell size, cell wall thickness, grain, moistness, tenderness, and softness, crumb color, and flavor. A breakdown of the scores for the individual charac- teristics evaluated appear in the Appendices. Means and standard deviations of the sensory score totals appear in Table 21 and an analyses of variance for these data appear in Table 22. Very highly significant differences were revealed among the variables evaluated but no significant differences occurred between cakes prepared with different extraction flours. The taste panelists rated the control cakes as having the most desirable characteristics and the cakes with a flour substitution level of 16% bran plus 12% unbleached midds as having the least desirable characteristics. Many of the other variables were, however, rated as being equally as good and no clear cut trend was established for these intermediate substitution levels in the cakes. The characteristics evaluated that contributed to low scores were crumb color and flavor. The bran and midds sub— stitution changed the cake flavor but all taste panelists agreed that the flavor change was not objectionable. This data tends to indicate that consumer acceptance of cakes high in fiber content may be good. The visual characteristics of the white cakes prepared can be seen in Figure A. 60 Table 21. Means and standard deviations1 of sensory scores for white layers prepared with 60 and 70% extraction flours. Sensory Scores2 Variable 60% 70% a a Control 85.1:2.9 83.7:N.5 ”% Bran 80.11123“b 82.011231,b 8% Bran 80.81-_1.6b’C 79.0:3.6b:0 16% Bra“ 78-813-8b’c 78.2:2.8b’c 12 Midd gleachzd 82-3i2-5a’b’c 78-9i2.83’b’c 12% Midds a,b a,b Unbleached 83.1:5.A 79.5:4.0 16% Bran + 12% Midds 75.8:3.9b’C 79.1:?.ub,0 Bleached 16% Bran + c c 12% Midds 77.9il.6 75.1:5.7 Unbleached 1Average of 5 replications. 2Total score 100 based on cell uniformity (10 pts), cell size (10 pts), cell wall thickness (10 pts), grain (16 pts), moistness (10 pts), tenderness (1H pts), softness (10 pts), crumb color (10 pts), and flavor (10 pts). aAverages superscripted by same letter are not significantly different at the 1% level of probability (Duncan, 1957). 61 Table 22. Analysis of variance of sensory scores for white layer cakes prepared with 60 and 70% extraction|flours. Source df Mean Squares Total 79 16.?“ Variable 7 6A.03*** Flour 1 26.H5 Interaction 7 13.69 Within 6” 11.75 ***Significant at the 0.1% level of probability. 62 60% EXIRACTION 70% EXTRACTION j I I ”(II ‘4 BRHI 5 BRAN 6. D ‘ .. ~ILI BxAPJ ,'.f‘ ‘t . . - . ~ \$§§’ ear; . . I _ .. 4“ . ' ' 9'1.\ ‘ _O... I.‘ 3.“! . g. .. . WT I‘~ - *r‘Y—rv‘f'v O O - ' l I I O ' O O D u I I. I O U C . I2 MIDDS BLEACHED II ”WV“ IIICIO-‘Ofiflfi I2 HIDDS UNBLEACHED 6 BRAN+ ,2 I". I 00 S BLEACHED I I I6 BPAN+ I: MIDDS UNBLEACHED Figure A. The color and appearance of white layer cakes prepared with increasing substitution levels of bran and midds for flour. vii 63 Flavored Layers Because objective and subjective data collected after preparation of the white layers indicated that quality characteristics were not greatly impaired by the substitu- tion of bran and midds for flour in the cake formula four flavored cakes including nut, spice, banana, and chocolate were chosen for further study. All flavored layers were prepared using a 70% extraction flour. Objective Measurements Due to a minimum amount of flour, batter characteristics were not evaluated. The primary evaluation of the flavored layers was sensory but objective data related to volume, color, and moisture were collected. Shrinkage and Volume-Related Indices. Means and stand- ard deviations for shrinkage and volume, symmetry, and unifor— mity indices for all four flavored layers are presented in Table 23 while analyses of variance of these data are presented in Table 2A. The only significant differences revealed by the analyses of variance were for the volume index for the banana (p<0.05) and chocolate (p<0.01) layers. The volume index was highest for the control cakes of the banana series while the control cake and the variable containing the bran substitution exhibited volume indices slightly higher than the other .eewwe .seossmv eeeeeoeooea no eeeee ew one we pnonowweo mepneoewenwem won one nonpoe oEem on» nee: oouaenomnonnm momenoezeeoe .A5mme .neonnov mpeeenenonn wo eo>oe ee one we pnonowweo eepneoewenmem won one noppoe oEem one npez oopaenomnoQSm momeno> omexnennm oeneene> .moxeo noeee oono>eew wo mooeone epeEnowenS one .mnpoEEmm .oESeo> one owexnennm now emnoepee>oo oneonepm one mneoz .mw oeneB 65 .epeeenenonn wo eo>oe ee onp pe pneoewenwemsx .eeeeeseooee no eeeee em esp we eseoeeesweme 02.0 00.0 wm.0 5e.0 we nenuez mm.0 00.0 *tm0.m w0.0 m moeneene> m2.0 00.0 05.0 me.0 me eepoa mB w0.0 00.0 05.0 w0.0 0e eepoe 2e.0 e0. 02.0 :e.0 0e eepoe moemm m0.0 00.0 me.0 w0.0 we nenpez w0.0 00.0 m0.0 50.0 m moeneene> 20.0 00.0 ee.0 w0.0 me eepoe Bbz xoone mpeenowenb xoone mnpoEEmm xoone oESeo> omexnennm wo oonnom monendm neoz .woxeo nomee oono>eew now mooeone mueEnowen: one .mnnoEEmm .oeneo> one omexnennm wo ooneene> wo memmeen< .2w oenee 66 variables in the chocolate series. The banana and chocolate cakes contain a greater percentage of sugar based on the weight of the flour than the nut and spice cakes. A volume decrease resulting with increased levels of fiber substitu- tion may be due to an imbalance of moisture during the baking process since the sugar competes with the other cake ingredi- ents for the water. Cake Color. L, aL, and bL color score means and standard deviations are presented in Table 25 for all four flavored cakes. Analyses of variance of these data are presented in Table 26. Very highly significant differences were revealed among variables in the nut, spice, and banana cakes for the lightness value. As the percentage fiber substitution increased the lightness of the cakes decreased. The bran and midds were brownish-red in color and imparted this darkness to the cakes. However, no significant differences occurred in the lightness of the chocolate cakes at increased fiber substitution levels. Very highly significant differences were revealed in all flavored cakes for the aL or redness color score. As the percentage fiber substitution increased the redness similar- ily increased. This again may be due to the fiber constitu— ents imparting their red color to the cake crumb. Significant differences (p<0.05) were revealed for the bL or yellowness score in the spice and chocolate layers only. As the percentage fiber substitution increased the yellowness of the cake decreased. 67 esp ee esoeeeeeo eeeseoenesmee esp we ssoeeeeeo eeeseoeeeswee won one won one .eewwe .seossmv eseeeoeooes no eeeee ew nonpoe oEem nee: oopnenomnonnm mowenofieoe .A5mme .neonnov upeeenenonn wo eo>oe ee nopuoe osem nee: oopqenomnonnm moweno> .moxeo nozee oono>eew wo monoom nOeoo now mnoepee>oo oneoneum one oneoz .mw oenea .eeeeeseooee eo eoeoe ee.o esp we eseoeeesmemeee .mpeeenenonn wo eo>oe ee on» we pneoewenwemes .eeeeesenoeo no eoeoe em one ee oseoeeesweme 68 5w.0 00.0 50.0 we nenpez ewe. eeeew. ew.0 m eoemeeee> m2.0 02.0 w0.0 0e eepoe meeeooomo wm.0 0e.0 5w.e we nenpez 02.0 eeemw.2 *aaww.0we m moeneene> mm.0 w0.0 we.0w 0e eepoe ww.0 w0.0 ww.0 0e eepoe woemm 0e.0 50.0 w0.0 we nenpez 0e.0 eeewe.me aexow.e0e m moeneene> mw.0 me.w w5.0m 0e eepoe 902 en ee 4 wo oonnom monemmw neoz .ooxeo noeee oono>eew now wonoom noeoo wo ooneene> wo momeeene .ww oenee 69 Moisture. Means and standard deviations for moisture determinations of all four flavored cakes are presented in Table 27 and analyses of variance of these data are given in Table 28. Significant differences (p<0.05) were revealed by the analysis of variance for the nut and banana cakes only. In both cases the variable prepared with midds retained more moisture than the variable prepared with bran. The composi- tion of the midds varies with each milling process (Wheat Flour Institute, 1965). Because the composition of the midds is uncertain it is difficult to relate their presence to the cakes exhibiting the greatest water retention. More research is needed in this area to ascertain the interaction between midds and other cake ingredients. Sensory Evaluation All flavored layers were evaluated organoleptically for the characteristics of color, moisture, tenderness, texture, flavor, and general acceptability. Each character- istic was evaluated on a lO-point scale, with 10 being opti- mum. Means and standard deviations of these sensory scores are given in Table 29. Analyses of variance of these data are given in Table 30. Very highly significant differences for color were revealed in the nut and banana layers whereby color acceptability was decreased by fiber substitution in the nut layers but was improved by fiber substitution in the banana layers. Fiber substitutions changed the crumb color Table 27. Means and standard deviations 1 for moisture determinations of flavored layer cakes. Variable % Moisture NUT X y Control 26.”:O.9y’ 16% Bran 25.2:l.lx 12% Midds 27'1i°°“x 16% Bran + 12% Midds 26.0:0.7 ’y SPICE X Control 23°9il°2x 16% Bran 24.2:l.3x 12% Midds 2L:.2:1.3x 16% Bran + 12% Midds 23.7:0.7 BANANA X y Control 2A.7:1.5y’ 16% Bran 23°“i1°5x 12% Midds 26.311.u 16% Bran + 12% Midds 2u.2:o.7y CHOCOLATE X Control 22'9i2‘3x 16% Bran 22°9i1'7x 12% Midds 23.2115x 16% Bran + 12% Midds 23.5il.5 1Average of 5 replications. XAverages superscripted with same letter are not significantly different at the 5% level of probability (Duncan, 1957). 71 Table 28. Analyses of variance of moisture deter- minations for flavored layer cakes. Source df Mean Squares NUT Total 19 1.12 Variable 3 3.23* Within 16 0.72 SPICE Total 19 1.20 Variable 3 0.33 Within 16 1.36 BANANA Total 19 2.60 Variable 3 7.27* Within 16 1.73 CHOCOLATE Total 19 2.82 Variable 3 0.37 Within 16 3.28 *Significant at the 5% level of probability. 72 .Aemme .emocsov speeenmooso co eo>me mm on» pm pompoumen zepcwoemecwem uo: ohm poppoe meow sue: ooerLompmasm mmwmpo>oe Re map 00 pcmmommeo zepcmoemecwem 00: ohm poupoe memo cues oopaepompoosm mowmpo>¢m .ESEequ wceoo 0e nae; oeMom pceoonoem .mcoepmoeeomp m mo mwmnm> w:.+0.0 eogpcoo I I I I I I meeqooomo ms.wm.~ xw.Hw.m sw.uw.w mu.ne.m m~.Hm.m oA.Hm.A moves Ame + swam Awe am.Hm.A sm.H:.m xa.nm.A m~.HH.m m:.Ho.m n:.He.m mooflz “we mm.Hm.A xm.HA.m xm.Hw.A mm.Hm.m am.H:.m nz.mm.m swam awe m:.+o.m xm.+m.m x:.+~.s mw.+m.0 mm.+m.0 ee.+s.w Hopscoo I I I I I I «zezem mm.H0.5 mm.HH.m mz.wm.s m:.H:.m mm.Hm.0 m.Hs.m meofiz Rme + comm «we «w.ww.e mm.Hm.m m:.Hm.s mm.n:.m am.Hm.m m.wm.m memes ewe «w.ww.» «e.ww.m mm.Hm.A mm.Hm.m am.He.m m.Hm.m seem “we m:.+w.s mm.+w.m mm.+m.s mm.+o.0 mm.+0.~ m.+m.m Hopscoo MOHmm m:.Hw.~ o:.H:.0 «e.wm.e om.H:.0 mm.Hw.0 m:.Hm.0 move: awe + zoom Awe am.H:.~ om.H:.m «e.wm.n o.m:.ns.w mm.ne.0 om.He.m meoez ewe MH.H©.N. DM.H©.w mm.H©.N. anN.HO.® GN.HH.m DmoHO.® Emcem ROH mm.+m.~ mm.+m.0 mm.+m.> m:.+m.> mm.+m.0 ww.+:.w eoppcoo 902 epeeeQMpomooa Haremflmo hog/Mam meeSmeB mmchemUCmB mLSUWHOE .HOHOO mHDchem> .mmxmo pomme oopo>Mem mo mmopoom maomcom mom emcoepme>mo upmocmum 0cm memos .mm meowe 73 .mpeeenmnopd mo eo>me we.0 on» no peacemecwemxxx .mueeebmoopo mo em>oe we on» no pseuemecwemxx .mpeeenmoopa mo eo>oe am map pm pcmoemecmemw ee. me. mo. mo. em. ow. we segues **es. *ww. mm. no. em. eo. m mesmepm> em. Am. ee. 00. em. mo. me empoa me em. em. em. mm. oe. :5. me empoe «zezem o we. om. em. em. me. we segue: 0 mm. me. :e. em. mm. m mesmepm> 0 Ne. we. om. em. me. me empoe moemm mo. mo. om. we. me. mo. we sense: no. **mw. we. **mm. mm. ***mm. m mmeomepm> we. me. me. mm. we. so. me empoe 502 epeeeompomoo< empmcom Lo>Mem whopxoe mmoCLoocmB mLSpmeoE Loeoo mo mesmepm> mopmSUm coo: .moxMo Loewe oopo>mem no mopoom mpomcmm pom mocmepm> mo mommemc< .om ment 7A of the nut cakes from creamy white to tan while crumb color of the banana cakes changed from grey to tan. No significant differences were revealed in any of the cakes for the moisture evaluation and tenderness was affected by fiber substitution in the nut layers only. Tenderness was significantly decreased (p<0.01) in the variable containing the substitution of both bran and midds as compared to the control. Significant texture differences (p<0.05) were revealed in the banana layers only. The control layer and the layer substituted with the 16% bran were scored as having signifi- cantly better texture than the layer containing the substitu— tion of 16% bran plus 12% midds. Flavor was unaffected by fiber substitutions in the spice layers but was significantly different in the nut (p<0.01), banana (p<0.05), and chocolate (p<0.05) layers. Flavor was slightly impaired at any fiber substitution level in the nut cakes; however, flavor was scored highest in the cakes substi- tuted with 12% midds for the banana layers. Chocolate cake flavor was slightly impaired at only the 16% bran plus 12% midds level of flour substitution. However, in all cases, no score was below 8 on a lO-point scale. Significant differences (p<0.01) in general acceptability were revealed for the chocolate layers only. The cake contain- ing a 16% bran substitution was the most acceptable chocolate layer while the cake containing a flour substitution of 16% bran plus 12% midds was least acceptable. Nevertheless, all 75 of these cakes had average general acceptability scores of 7 to 9. An overall review of sensory scores tends to indicate that flavored layer cakes prepared with fiber substitutions for flour result in good quality products. SUMMARY AND CONCLUSIONS The purpose of this study was to investigate the feasi- bility of incorporating dietary fiber into a baked product system while maintaining the original product quality. The baked product chosen for this research was layer cakes. White layer cakes were prepared from 60 and 70% extrac- tion flours with increasing substitution levels of bran and middlings using a standard white layer cake formulation. Substitution levels included A, 8, and 16% bran, 12% midd- lings either bleached or unbleached, and the combination of 16% bran with either 12% bleached or unbleached middlings to approximate whole wheat flour. Flavored cakes were also prepared from a 70% extraction flour employing levels of 16% bran, 12% unbleached middlings, and the combination of the 16% bran plus 12% unbleached middlings. The flavors prepared were nut, spice, banana, and chocolate. The bran and midd— lings were ground with a cyclone mill to the same particle size as the control flour to eliminate mouthfeel differences. Physical characteristics of the white cake batters examined were specific gravity, viscosity, color, and pH. Cake quality characteristics examined were volume indices, moisture, tenderness, compressibility, color and pH. Physical characteristics of the flavored cakes examined were volume 76 77 indices, moisture, and color. Organoleptic studies were also conducted on all cake variables. Objective measurements of the quality characteristics of the white cakes indicated that specific gravity of the batter and uniformity, symmetry, and shrinkage of the cakes were not significantly changed by either the flour extraction or the level of dietary fiber used. High fiber replacement levels in the 70% extraction flour appeared to form a more viscous batter than any of the batters prepared with the 60% extraction flour. Cakes prepared with the 60% extraction flour showed greater moisture retention than those prepared with 70% extraction flour. The level of dietary fiber used, however, did not affect moisture retention of the cake. The volume index was not adversely affected except by substitu- tions containing the bleached midds. Substitution of fiber in the cake formulations resulted in increased tenderness values but no trend for compressibility was established. Both the batter and cakes prepared with increased levels of dietary fiber had higher yellowness and redness but decreased light- ness values. Sensory data scores showed that the incorpora- tion of bran and midds into the layer cakes did not adversely affect the quality characteristics of the product. Out of a possible 100 points all cakes were scored 75 or above. Objective measurements of the quality characteristics of the flavored cakes indicated that volume was slightly impaired in the banana and chocolate cake layers as the fiber substitution levels increased. This was attributed to the 78 higher sugar levels in these cakes and the fact that sugar acts in competition for the water. As the percentage fiber substitution increased the lightness value decreased in the nut, spice, and banana layers. The redness value increased in all four flavored layers as fiber substitution levels 'increased while the yellowness value decreased in only the spice and chocolate layers. Moisture retention was affected in the nut and banana cakes where the substitution of 12% midds retained the greatest amount of water. This may have been due to the chemical interaction of the midds with the other cake ingredients but further research is necessary to pinpoint the cause. Sensory data indicated that bran and midds incorporation into the layer cakes.did not affect cake quality. All characteristics evaluated were rated 7 or better on a 10-point scale. From an evaluation of the data obtained in this inves— tigation, it can be concluded that dietary fiber in the form of bran and midds can be increased in layer cakes without adversely affecting the cake quality characteristics. PROPOSALS FOR FURTHER RESEARCH Although the results of this study indicate that sub- stitution of wheat bran and middlings for flour in a layer cake system is a feasible method of increasing dietary fiber intake, further investigation is warranted. The following research areas are proposed: 1. The effect of fiber substitutions in other cake formulations needs to be studied. The effects of ingredient changes, especially changes in the water level and/or the addition of emulsifiers, need further investigation. 2. Fiber substitutions in other baked products, such as cookies, biscuits, or pastry, need to be studied. In order to increase the intake of dietary fiber in the diet a variety of fiber carriers need to be available for consumption. 3. Studies concerning the effects of the substitution of levels of fiber greater than 16% bran plus 12% midds need to be conducted on cakes as well as other baked products. A. The effect of dietary fiber on the absorption of vitamins and minerals needs to be investigated. Controlled feeding studies would indicate whether or not fiber consti- tuents decrease nutrient absorption and utilization in the body. 79 8O 5. Studies are needed to determine the effect of fiber substitutions on product stability and shelf life. The increased lipid levels at high fiber substitution levels may cause rancidity to occur at an increased rate. 6. Finally, investigations into the physical and chemi— cal characteristics of middlings need to be made. Only then can the interactions of midds with the other baked product ingredients be clearly understood. REFERENCES C ITED REFERENCES CITED American Association of Cereal Chemists, 1962. AACC approved methods. 7th ed. The Association. St. Paul, Minn. Ash, D. J. and J. C. Colmey, 1973. The role of pH in cake baking. Baker's Dig. A7:2, 36. Aspinall, G. 0., 1959. Structural chemistry of the hemicellu- loses. Advan. Carbohydrate Chem. 1A, A29. Baldi, V., L. Little, and E. E. Hester, 1956. Effect of the kind and proportion of flour components and of sucrose level on cake structure. Cereal Chem. A2, A62. Bhattacharjee, S. S. and T. E. Timell, 1965. A study of the pectin present in the bark of amabilis fir (Abies amabilis). Can. J. Chem. A3, 758. Bowes, A. and C. F. Church, 1970. Food Values of Portions Commonly Used. J. B. Lipponcott Co., Philadelphia, Pa. Burkitt, D. P., 1971. Epidemiology of cancer of the colon and rectum. Cancer. 28, Burkitt, D. P., 1971. The aetiology of appendicitis. Br. J. Sug. 9, 695. Burkitt, D. P., 1973. Some diseases characteristic of modern western civilization. Br. Med. J. 1, 27A. Burkitt, D. P., A. R. P. Walker, and N. S. Painter, 197A. Dietary fiber and disease. JAMA. 229:8, 1068. Cleave, T. L., G. D. Campbell, and N. S. Painter, 1969. Diabetes, Coronary Thrombosis and the Saccharine Disease. 2nd ed. John Wright and Sons, Ltd. Bristol, England. Collins, 0. D., G. L. Sunderlin, 19A0. Cake batter viscosity as related to batter structure. J. Home Econ. 32, 556. Cowgill, G. R. and W. E. Anderson, 1932. Laxative effects of wheat bran and washed bran in healthy men. A compara- tive study. JAMA. 98, 1866. 81 82 Cummings, J. N., 1973. Progress report: dietary fiber. Gut. 1A, 69. de Groot, A. P., R. Luyken, and N. A. Pikaar, 1963. Choles- terol-lowering effect of rolled oats. Lancet. 2, 303. Donelson, D. H. and J. T. Wilson, 1960. Effect of the relative quantity of flour fractions on cake quality. Cereal Chem. 32, 26. Donelson, D. H. and W. T. Yamazaki, 1972. Soft wheat flour particle-size analysis by integrated sieve and Coulter Counter procedures. Cereal Chem. A9, 6A1. Donelson, J. R., 197A. A new method for wet-fractionating cookie flour. 36th Annual Report of the Soft Wheat Quality Laboratory, Agricultural Research Service, 130-136. Duncan, D. B., 1957. Multiple range test for correlated and heteroscidastic means. Biometrics. 13, 16A. Ellinger, R. H. and F. J. Shappeck, 1963. The relation of batter specific gravity to cake quality. Baker's Dig. 37:6, 52. Frasar, J. R. and D. C. Holmes, 1959. Proximate analysis of wheat flour carbohydrate. IV. Analysis of whole- meal flour and some of its fractions. J. Sci. Fd. Agr. 10, 506. Friend, B., 1967. Nutrients in United States food supply: A review of trends, 1909-1913 to 1965. Am. J. Clin. Nutr. 20, 907. Gilles, K. A., 1960. The present status of the role of pen— tosans in wheat flour quality. Baker's Dig. 3A:5, A7. Hill, M. J., J. S. Crowther, B. S. Grasar, G. Hawksworth, V. Aries, and R. E. 0. Williams, 1971. Bacteria and aetiology of cancer of large bowel. Lancet. 1, 95. Howard, N. B., 1972. The role of some essential ingredients in the formation of layer cake structures. Baker's Dig. A6z5, 28. Howard, N. B., D. H. Hughes, and R. G. K. Strobel, 1968. Function of the starch granule in the formation of layer cake structure. Cereal Chem. A5, 329. Hunter, M. B., A. M. Briant, and C. J. Personius, 1950. Cake quality and batter structure. Cornell University Agr. Exp. Sta. Bulletin 860. 83 Hutchinson, J. B. and H. F. Martin, 1970. Nutritive value of wheat bran. I. Effects of fine grinding upon bran, and of added bran upon the protein quality of white flour. J. Sci. Fd. Agr. 21, 1A8. Hutchinson, J. B., T. Moran, and J. Pace, 1962. High-protein breads and the growth of weanling rats. J. Sci. Fd. Agr. l3, 1. Jones, C. R., 1958. The essentials of the flour-milling process. Proc. Nutr. Soc. 17, 7. Kulp, K., 1972. Some effects of chlorine treatment of soft wheat flour. Baker's Dig. A6:3, 26. Leveille, G. A. and H. E. Sauberlich, 1966. Mechanism of the cholesterol-depressing effect of pectin in the choles- terol-fed rat. J. Nutr. 88, 209. Matz, S. A., 1960. Baking Technology and Engineering. AVI Pub. Co. Westport, Conn. Miller, B. S. and H. B. Trimbo, 1965. Gelatinization of starch and white layer cake quality. Fd. Tech. 19, 6A0. MacDonald, 1., 1968. The functional properties of surface active agents. Baker's Dig. A2:2, 2A. MacMasters, M. M. and G. E. Hilbert, 19AA. The composition of the "amylodextrin" fraction of wheat flour. Cereal Chem. 21, 5A8. McCance, R. A. and C. M. Walsham, 19A8. The digestibility and absorption of the calories, proteins, purines, fat, and calcium in wholemeal wheaten bread. Br. J. Nutr. 2, 26. McCance, R. A., K. M. Prior, and E. M. Widowson, 1953. A radiological study of the rate of passage of brown and white bread through the digestive tract of man. Br. J. Nutr. 7, 98. O'Brien, R. D., 1972. Production of cakes using latest emul- sifier systems. Baker's Dig. A6:2, 71. Painter, N. S. and D. P. Burkitt, 1971. Diverticular disease of the colon: A deficiency disease of western civili- zation. Br. Med. J. 2, A50. Painter, N. S., A. Z. Almeida, and K. W. Colebourne, 1972. Unprocessed bran in treatment of diverticular disease of the colon. Br. Med. J. 2, 137. 8A Platt, W. and P. D. Kratz, 1933. Measuring and recording some characteristics of test sponge cakes. Cereal Chem. 10, 73. Pomare, E. H. and K. W. Heaton, 1973. Alteration of bile salt metabolism by dietary fibre (bran). Br. Med. J. A, 262. Pomeranz, Y. (ed)., 196A. Wheat Chemistry and Technology. American Association of Cereal Chemists. St. Paul, Minn. Scala, J., 197A. Fiber - the forgotten nutreint. Fd. Tech. 1, 3A. Short, A. R., 1920. The causation of appendicitis. Br. J. Surg. 8, 171. Sloneker, J. H., 1971. Determination of cellulose and apparent hemicellulose in plant tissue by gas-liquid chromato- graphy. Analytical Biochem. A3, 539. Smith, F. and R. Montgomery, 1959. Chemistry of Plant Gums and Mucilages and Some Related Polysaccharides. Reinhold: New York. Sollars, W. F., 1958. Fractionation and reconstitution procedures for cake flours. Cereal Chem. 35, 85. Southgate, D. A. T., 1969. Determination of carbohydrates in foods. II. Unavailable carbohydrates. J. Sci. Fd. Agr. 20, 331. Sundaravalli, 0. E., K. S. Shurpalekar, and M. N. Rao, 1971. Effects of dietary cellulose supplements on the body composition and cholesterol metabolism of albino rats. J. Ag. Food Chem. 19, 116. Trowell, H. C., 1972. Crude fibre, dietary fibre, and atherosclerosis. Atherosclerosis. 16, 138. Van Soest, P. J. and R. H. Wine, 1967. Use of detergents in the analysis of fibrous feeds. IV. Determination of plant cell-wall constitutents. J.AOAC. l, 50. Van Zavelenburg, C., 190A. Obstruction and consequent disen- tion the cause of appendicitis. JAMA. A3, 820. Walker, A. R. P., B. D. Richardson, B. F. Walker, and A. Woolford, 1973. Appendicitis, fibre intake and bowel behaviour in ethnic groups in South Africa. Postgrad. Med. J. A9, 187. 85 Wheat Flour Institute, 196A. A kernel of wheat. Wheat Flour Institute, Chicago, Ill. Wheat Flour Institute, 1965. From Wheat to Flour. Wheat Flour Institute, Chicago, Ill. Whistler, R. L. and R. E. Pyler, 1968. Action of chlorine on wheat flour polysaccharides. Cereal Chem. A5, 183. Wilson, J. T. and D. H. Donelson, 1963. Studies on the dynamics of cake baking. I. The role of water in the formation of layer cake structure. Cereal Chem. A0, A66. Yamazaki, W. T., 1955. The concentration of a factor in soft wheat flours affecting cookie quality. Cereal Chem. 32, 26. Yamazaki, W. T., J. R. Donelson, and L. W. Briggle, 1968. Microtests for soft wheat quality evaluation. Crop Science. 8, 199. APPENDICES APPENDIX I PROXIMATE ANALYSIS OF BRAN, MIDDLINGS, AND FLOURS OBTAINED FROM MENNEL MILLS Moisture Protein Ash pH Identification % % % Bleached Patent 50% Extraction 13°6 8'85 0’36 “'75 Bleached Straight Grade 70% Extraction 13'3 9.A5 0’51 5'05 Bleached Clears 12.7 11.70 0.95 5.25 Wheat Midds Unbleached 12'3 15'55 “'80 '--- Wheat Midds Bleached 12.3 15.55 A.80 --—- Wheat Bran Unbleached 13'1 1A.55 6'60 "” 86 APPENDIX II AVERAGE PARTICLE SIZE1 OF FLOUR AND CYCLONE MILLED BRAN AND MIDDS PARTICLE 60% Flour 70% Flour Midds Midds SIZE Extraction Extraction Bran Bleached Unbleached in % Z % Z % .0232 0 0 0 0 0 .0165 0 0 0 0 0 .0116 55 6A 69 A8 56 .0058 22 27 3o 51 A3 .0029 21 8 l 1 1 .0017 2 1 0 0 0 1Average of 5 replications. 87 APPENDIX III TASTE PANEL EVALUATION OF WHITE LAYER CAKES CHARACTERISTICS Internal Factors (100 points) Score A. Cells (30 points) (Points) 1. Uniformity (10 points) a) Even (normal) 10 b) Slightly uneven 6 c) Uneven 2 2. Size (10 points) a) Dense (normal) 10 b) Close 8 C) Slightly open 6 d) Open A 3. Thickness of walls a) Thin (normal) 10 (10 points) b) Slightly thick 6 C) Thick 2 B. Grain (16 points) 1. Silky (normal) 16 2. Harsh 10 3. Coarse (corn bread) 8 C. Texture (3A points) 1. Moistness (10 points) a) Gummy 6 b) Moist (normal) 10 c) Slightly dry 8 d) Dry A 2. Tenderness (1A points) a) Very tender (normal) 1A b) Tender 12 0) Slightly tough 10 d) Tough A 3. Softness (10 points) a) Soft (normal) 10 b) Slightly firm 8 c) Firm A D. Crumb Color (10 points) 1. Bright white (normal) 10 2. White 8 3. Slightly dull 8 A. Slightly creamy 8 5. Creamy 6 6. Slightly dull and slightly creamy A E. Flavor (10 points) 1. Normal (no off-flavor due to flour) 10 2. Foreign _10 TOTAL 100 88 APPENDIX IV TASTE PANEL EVALUATION OF FLAVORED LAYER CAKE CHARACTERISTICS Quality Factors Scoring Scale 10 9 8 7 6 5A321 Color Very Moderately Slightly Very desirable desirable desirable undes. 10 9 8 7 6 5 A 3 2 1 Moistness Very Moderately Slightly Very desirable desirable desirable undes. 10 9 8 7 6 5 A 3 2 l Tenderness Very Moderately Slightly Very desirable desirable desirable undes. 10 9 8 -7 6 5 A 3 2 1 Texture Very Moderately Slightly Very desirable desirable desirable undes. 10 9 8 7 6 514321 Off-flavor Very Moderately Slightly Very desirable desirable desirable undes. 10 9 8 7 5 5 A 3 2‘ 1 iinzrgibilit Very Moderately Slightly Very C p y desirable desirable desirable undes. Comments: 89 gossamewep 90 w.eem.w :.oe=.w w.oH=.m m.oHo.me m.on.m e.ees.me m.oem.s o.ees.e e.eem.w mace: ewe + Cth RQH I I I I I I I emeommem «.mee.e m.oH:.w m.o+o.m m.o+w.me m.o+m.m s.o+e.me e.e+w.s m.e+e.s :.o+w.w move: “me + camem Rmd e.eee.w w.eem.w m.oem.m w.eem.me :.on.m w.eeo.we A.eem.w m.on.s m.oew.w oMMMMMemmw I I I I I I I I I ow m.e+w.w m.o+e.w w.o+m.m w.o+w.me s.o+m.m o.e+w.me s.o+w.w m.o+w.e w.e+o.w mwmmz mww e.eew.w s.oew.w s.oee.m w.eew.me w.ees.m w.oH=.me w.ees.s e.eee.s m.on.: seem ewe e.eeo.m m.oe=.w w.eew.w w.eeo.me m.on.m o.eee.me w.eew.s m.oew.s s.ouw.w swam em e.eew.m w.eem.s w.eem.w w.eew.ee e.ouo.m w.eew.:e m.on.s s.on.e w.eew.w capo we w.eee.m o.oem.m m.oHs.s w.wHw.oe w.eeo.m o.ee=.we e.oem.w m.oHA.w F.0Hm.w eospcoo poeoo mmoc mmoc mmo: moo: .INueE Ao>mem passe upwow Ismocme upmeoz cease Ixoeee muem Inches: mesmesm> eemz mmmmzmz qmzHQZH mo oszoom mboem ZOHBo xHszmm< UmnommeDCD 91 m.me.m w.eem.w e.oem.m o.eHw.we m.oHA.m w.eew.me e.oem.e w.eew.s e.eee.w eeeez ewe + seem ewe I I I I I eesoeeem w.eee.e e.wHw.w m.oew.we w.eew.m e.e+m.se e.e+w.m m.o+m.m o.e+:.e o.e+e.e eeeez ewe + Qweem RQH I I I I I I I I I w 060 C m.e+w.w w.o+w.w w.o+e.m e.o+o.we e.o+e.m w.o+o.se w.e+s.m m.o+m.e o.e+e.w semeezemww . I . . I . . I . . I . . I . . I . . I . . I . . I . eesoeeem w e+e w s 0+0 m w 0+w w m 0+0 we w 0+w m m 0+w me 0 e+w w 0 e+e e w e+m w ence: ewe e.eHo.e w.eew.w w.eem.m w.eew.we w.eew.m w.eHw.me e.wHe.e m.oHe.e w.eem.w seem ewe w.wew.e w.eem.w w.eew.m e.eHm.ee w.eee.m w.eew.me s.oHe.m m.oue.e s.oum.w seem em w.wHw.m m.on.e s.oem.m w.ouw.we e.oeo.m o.eHm.me m.on.m w.ees.e o.eHw.w seem es o.oeo.oe w.eem.m w.eee.m w.eew.ee w.eem.m e.oew.we e.wHe.m m.ouw.e w.eee.w eoessoo eoeoo mmoc mmoc mmo: mmoc mpee eoeeem seseo useom neessea neeeoz seeew Ixeese eeem neoeesm eeseeee> eeez mboqm zoeeo¢mexm «on mBHB Dmmdmmmm mmx¢o mmwHQZH mo wszoom H> xHszmm< 92 .memmn oespmeoE eze so meme ee< e sesoeeesss e.w e.w e.w e.w I--- I--- I--- I--- eeeez ewe + thm Rod eeseeeem e.w m.w w.s e.w I--- I--- I--- I--- moses ewe + seem ewe . . . . . . . . eesoeeeess w e e e e m o m w em 0 mm m em w mm esees ewe . . . . . . . . sesoeeem e e w e w e e e e em e em e em a em eeeez ewe m.w w.s w.w o.m e.w: e.em w.wm e.wm seem ewe w.: :.e w.ee m.m m.om e.om w.ww e.ww seem ew m.e m.e e.w w.ee m.ww w.ww m.mw w.ww seem e: e.: e.e e.w w.ee w.me m.we w.we m.me eoeesoo ewe .eow eoe eow eoe eow Ieoe eow eeseeee> e eeesseom eeeez e seesee e emseeeee e seeeem emmboem zoeeodmexm “on Qz< ow mom mmDQmoomm ZOHB¢ZOHBo NHszmmd APPENDIX VIII CRUDE FAT ANALYSES OF 60 AND 70% EXTRACTION FLOURSl Crude Fat - % Variable 860% 70% Control 1.50 1.88 A% Bran 1.93 2.26 8% Bran 2.31 2.70 16% Bran 2.80 2.9A 12% Midds Bleached 2'22 2'87 12% Midds Unbleached 2'59 2°“2 16% Bran + 12% Midds 3.03 3.98 Bleached 16% Bran + 12% Midds 3.A8 A.O7 Unbleached l 93 All data on 1A% moisture basis. HICHIGQN STQTE UNIV. LIBRQRIES 31293102239575