EFFECT OF COMBINED WAXY :MAIZB AND REGULAR CORN .STARCH» ON VISCOSITY AND COLD PASTE: FLOW OF SIMPLE PUDDINGSI Thesis For the Degree 65 M. S. MICHIGAN STATE UNIVERSITY Betty Jo Sullivan. I956 ..... e -..‘"‘-‘.M~'ét ......... IIIIIIWIIHlllfllilillWlHlIlHIIHIUHIIIHIIJUIII 3 1293 10668 3224 RETURNING MATERIALS: I}V1ESI_J PIace in book drop to remove this checkout from w your record. FINES wil] be charged if book is returned after the date stamped below. EFFECT OF COVPINED WAXY “AIZE AND ”PG”LAR COD” ”TADS? UV VIQCOSITV A?“ COLD PAST? $IOW O? SIF?L? PVPDINGS 9y Betty Jo Sullivan Submitted to the Dean of the College of Fome Economics of Nichiaen Eton: “EiVQPCity of Aerlculture end Anolled Science in martial fulfillment of the requirements for the degree of WASTE“ OF SCILNCE Dannrtment of Institution Admlristretion THESIS (1' ‘1’3 '0? C‘ flrnlv‘vv I‘VKK‘ JI‘o‘LJH ;\J-Ja' :A‘T The writer Viehes to exoreee her sincere gratitude to Dr. Peerl J. Aldrich fir her encouregement end evidence during the oreoeretion of this theqie. She also wiehee to thank Pr. William Eaten for essigte anoe in stetietioel analysie of the data, Professor Katherine Hart for her intereet in the nroject and assistnnoe in read— ing the menuecriot, Dr. Elizebeth Osman for her interest, ‘ . 'heloful survestione, on: for the Uee of the Corn Infinqtries viscometer. Aooreciotion is extended to T“r. Mergeret Ohlson for oerzissiou to use laborfitory 92039 in the Nutrition Deonrtment. Acknowledgenents are also mece to the American Veize— Product Comoeny for enoolyinv the wnxy maize starches used in this study and to Robert L. Lloyd for making nveilable information concerning there products. 11 INTRODUCTION . TABLE OF CONTENTS REVIEW or LITERATURE . . . . . . . .‘. . . . . Starch Fractions . . . . . . . . . . . . . . . Structure of Starch Granules . . . . . . . . . . . . Swelling of Sterch Granules . . . . . . . Viscosity Gel Formation . . . . . . . . . . . . . . . . . . . . Factors Affectina BeheVior of Cooked Starch Psstes. . Source: inherent characteristics . . . . . . . Effect of treatment during manufacture . Effect of electrolytes . . . . . . . . . . . . . Effect of starch concentration . . . . Effect of time and temoereture . . . . Effect of fatty acids . . . . . Effect of sucrose . . . . . . . . . . Effect of mechanical agitation . . . . . . . . . The Waxy Starches . . . . . . . . . . . . . . . . . . Objective Tests for Cooked Starch Pastes . . . . . . Viscosity measurements . . . . . . . . . Scott test for hot osste viscosity . . Stormer viscosimeter . . . . . . . . NscMichsel viecosimeter . . . . Hoepoler viscosimeter . . . . Caesar consistometer . . . . . . . . 28> 3o 31 39 '49 '39 no b1 #1 b1 TAFLE or couTsNTs (contd.) Brnhender smyloersch . . . . . . . . . . . . b2 Corn Industries viscometer . . . . . . . . . b3 Line—screed tests . . . . . . . . . . . . . . . . Uh Grswemeyer and Pfunfl test . . . . . . . . . 1“ Miller test for cold osste flow . . . . . . L5 lel strength tests . . . . . . . . . . . . . . . 46 Tarr-Esker Jelly tester . . . . . . . . . . L7 Exchange Ridgelimeter . . . . . . . . . . . L'7 Sssre €190 methOd . . . . . . . . . . . . . Q8 Fuchs tenetrometer . . . . . . . . . . . . . b9 Recorfiine eel tester . . . . . . . . . . . . SO PROCEDVRQ . . . . . . . . . . . . . . . . . . . . . . . . . 51 Selection of Formula . . . . . . . . . . . . . . . . . 51 Starch concentration . . . . . . .'. . . . . . . 91 Sucrose concentretion . . . . . . . . . . . . . . 52 Liquid mefiium . . . . . . . . . . . . . . . . . . S2 Sslt concentration . . . . . . . . . . . . . . . 57 Basic Formula . . . . . . . . . . . . . . . . . . . . 53 Veristions . . . . . . . . . . . . . . . . . . . . . . 5h Preosration of Formula . . . . . . . . . . . . . . . . 56 Ingredients . . . . . . . . . . . . . . . . . . . 56 Mixing orocedure . . . . . . . . . . . . . . . . 57 Cookinv procedure . . . . . . . . . . . . . . . . 69 Testing Equinment . . . . . . . . . . . . . . . . . . 63 iv TABLE OF CONTENTS (contd.) Testing Procedure . . . . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . Viscosity Tests . . . . . . . . . . . . . . . . . . Maximum viscositv values . Prooortion of starch . . . . . . . . . . . . . . Cookine time . . . Kind of starch . . . . . . . . . . . . . . . . Cold Paste Flow . . . Proportion of waxv maize starch Cooking time Holding time . Kind of starch . . . . . . . . . SUNMARY . . . . . . . . CONCLUSIONS . LITERATURE CIT? . . . . . . . . . . . . . . . . . . . . . an 67 67 67 71 75 79 79 8h 86 d7 91 93 LIST OF TABLES Starch comeosition for Series A,‘ . . . . . . . . Starch comoosition for Series B . . . . . . . . . Range and average maximum viscosity values in three reolications of Series A and B . . . . . . Analysis of variance of maximum viscosity values for Sprieg A. e I e e e e o e e O O O 0 0 Analysis of variance of maximum viscosity values for Series E. . . . . . . . . . . . . . . . Cookins times and temoeratures to reach maximum viscosity, and temoeratures of initial viscosity droo in all reolications of Series A . . . . Cookina times and temoeratures to reach maximum viscosity, and temperatures of initial viscosity droo in all reolications of Series B . . . . . . Ranse in flow values of three reolications of each series . . . . . . . . . . . . . . . . . . Summary of average flow readinys of three reolications for each series at l-hour and 3-hour h>lding oeriods . . . . . . . . . . . . Analysis of variance of flow readings in Series A Analysis of variance of flow readings in Series B vi 55 67 69 71 73 79 80 80 81 Figure LIST OF VlGVRES Sectional firawins of the Corr Tnfiustries viscometer . . . . . . . . . . . . . . . . . . . . . 61 Effect of oercentase of waxy maize starch w-11 on the initial viscosity rise, maximum viscosity, ans final viscosity of nudflinss in series A . . . . 63 Effect of nercentase of waxy maize starch Amaizo "£90" on the initial viscosity rise, maximum viscosity, and final viscosity of ouddinrs in series B . . . . . . . . . . . . . . . . I Effect of oercentage of waxy maize starch on average maximum viscosity values for series A and B as shown by lines of regression . . . . . . . 77 Effect of kind of waxv maize starch at corresponding levels of total starch concentration on viscosity curves of nuddinss . . . . . . . . . . . . . . . . . / Effect of cooking time on line—soread readinss for both l—hour and W-hour samules in series A and B I o o o o o o . o o . a o o a o o o o o o o o 8.- vii ' INTRODUCTION Deve10ping and maintaining desirable consistency in cream puddings and pie fillings are often problems in quantity food preparation. Ordinary cornstarch is the thickening agent most frequently employed in making these products. At normal starch concentrations, cornstarch puddings are fluid when hot but undergo the process of gelation when allowed to cool. The freshly cooled product has a delicate, gel-like character with a slightly firm body and creamy consistency. 'These _characteristics are very desirable in cream puddings and pie fillings. .The longer holding periods often required in institution food service, however, frequently result in excessive stiffness, cracking, and some liquid separation in these products. The use of cornstarch in combination with a non-gelling type of thickening agent has been suggested as a possible way to eliminate these undesirable characteristics and still retain the-fairly firm body which is desirable in cream puddings and pie fillings. Tapioca, which does not retrograde on cooling, has been used to some extent in this role. The war period saw the deveIOpmsnt of the commercial production of waxy cereal starches which make pastes possessing the non—gelling properties of tapioca to an even greater degree. The starch from waxy maize has been processed in the largest quantities. Industry has developed methods for chemical treatment to modify the raw waxy maize starch; and, as a result, some of its less desirable prooerties have been overcome. Several of these modified starches are now available for use in institutions. The purpose of this study was to investigate the effects produced upon hot and cold paste characteristics by modified waxy maize starches used in combination with ordinary corn starch in preparation of cream-type puddings. Two waxy maize starches, representing different levels of modification, were used in this investigation. It is hoped that the results of this study may reveal some basic information concerning the behavior of these modified, waxy starches when they are used as part of the total starch concentration in cream fillings and puddings. Such data could point the way to future explor— ation of possibilities for the use of these waxy starches in similar products in large quantity food production. ~15) REVIEW 0? LITERATURE Prautlecht (1*) described starch as that material formed in all green Dlants Which is subseouentlv storm1 as micro— scopic granules to be used as a future food supply for the germinating seed. Starch is commercially obtainable from cereal grains, tubers, roots, and the pith of certain palms. The Egyptians, prior to 1000 P.C., obtained the first starch from wheat and utilized it for purposes of food and for bind~ ing fibers to make papyrus. Whistler and Smart (69) reported that since that time starches have been made in increasinu quantities. During the fourteenth century the industrial preparation of starches first became prevalent in Europe. Today the manufacture of starch is a world—wide industry of great importance. Schoch (61), in a review of starch research, stated that only since the late 1910's has real prOgress been made in exolaining the basic structure, composition, and behavior of starch. This advancement has paralleled the devel- opment of modern colloid chemistry. Starch Fractions According to Frankel (2‘), starch belongs to that group of organic substances which, besides carbon, contain hydrOgen and oxygen in the same proportions found in water. Among other substances belonging to this classification are the celluloses, dextrins, and sugars. Schoch (<1) stated that starch is a polymer of glucose units which, on complete hydrol- ysis with acid, yields 100 per cent of this hexose sugar. As early as 181k the behavior of a cooked starch suspension was interpreted as evidence that starch contained two or more carbohydrate substances (51). In l92b Alsberg and Rash (2) reported that starches of certain varieties of cereals gave a red color with iodine and those of other varieties showed a blue color. Pastes made from the starches giving the red color with iodine were more viscous than those that colored blue. In a discussion of the physical properties of starch, French (27) stated that with the perfection of fractionating procedures two extremes of starch types are recognized. These fractions are thought to be homogeneous with regard to chemical type but heterogeneous with regard to molecular size and the degree and order of branching. Current definitions are made on the basis of chemical structure, although earlier defini— tions were based on such properties as solubility, enzyme digestibility, and iodine color. Whistler and Smart (59) refer to the linear molecule as amylose and to the branched structure as amylouectin. Some investigators prefer to call these the A—fraction and the B—frsction, respectively. Schoch and Elder (52) described the linear fraction as being made up of a long chain of glucose units. The intro— duction of other glucose units along these linear chains re- sults in a branched structure and, ultimately, in a tree-like molecule. Ordinary corn, wheat, potato, and tapioca starches contain both the linear and branched molecules. The ratio of these components is fairly constant for any given plant species. The amount of linear material ranges from 17 per cent in tapioca starch to 28 per cent in ordinary corn starch. Starches designated as waxy starches contain only the branched fraction. On the other hand, starches of wrinkled-seeded garden peas and certain varieties of sweet corn contain a larger proportion of the linear fraction. With regard to molecular size, Kerr (#0) stated that there is considerable variation among the molecules of each structural type, even in one variety of starch. This investigator reported that Beckmann and Landis estimate the average molecular size of the amylose fraction as 300 to £00 glucose units and the amylopectin fraction as 1000 times larger. Hadley (“9) reported that, according to several investi— gators, the presence of phosphoric acid in the amlepectin molecule and its resistance to enzyme action distinguish this fraction from the amylose fraction. Schoch (51) discussed some of the differences between the A and B fractions. Five to ten per cent solutions of the linear A—fraction show exaggerated tendencies to retrograde and to set up to an irreversible gel on cooling. This behavior is attributed to the strong associative forces between the long, linear molecules. The bush-like structure of the B—fraction prevents any such orderly parallel alignment of molecules. Only in instances where gel formation is a desirable characteristic is the A~fraction alone technologically useful. The B~fraction is the component of starch which makes it useful as a thickening agent, protective colloid, or as a sizing material for paper and textiles. Although the presence of the A-fraction makes a starch unsuitable for certain ourOOses, the undesirable characteristics can be reduced to a minimum by chemical modi— fication. Many investieetors have studied the nature of the iodine reaction with starch. Bates and co-workers (6) used a potentiometric method for quantitative determination of amylose components of starch. His preliminary results showed that_ affinity for iodine varied inversely with the degree of branching of the starch chains but directly with the length of the starch chain. Structure of Starch Granulese Whistler and Smart (59) stated that starches typically occur in nature as discrete granules. The appearance and size of starch granules are characteristic of the plant type from which they originate. Starches from various sources can be differentiated by careful physical examination of the granules under a microscooe. According to Schoch and Elder (52), eranules of potato starch are relatively large, ranging from 15 to 100 microns in diameter. The smallest of the common starch eranules are those from rice with a diameter of 5 to 6 microns. Corn starch granules are similar in appearance to rice starch granules but are considerably larger. Brautlecht (1h) described the starch granule as a minute structure built of layers of molecules arranged concentrically. According to his investigations, shapes ranged from small polygons to large spherical granules-with a dark spot, the hilum, eccentrically located in each. In addition to this concentric organization, Schoch and Elder (52) observed the presence of a radial structure which resulted in the typical radial direction of cleavage. SJostrom (5‘) supported this latter view. - Hadley (h?) concluded that the inner portion of the starch granule consisted of soluble amylose and the outer layer was composed of amylooectin which save the pastins character— istics to a cooked starch suspension. Alsberg (l) and Katz (39) supported this theory. In 1955 Badenhuizen (5) found no such decrease in content of the linear fraction from center to periphery. This investigator concluded that the linear molecules are distributed uniformly throughout the layers of the starch granules in most starches. Swelling of Starch Granules Starch granules are insoluble in cold water. In 192k Alsberg and Rask (2) observed that when a suspension of starch granules in water was heated, the granules first swelled, then burst, and finally lost their anistropy. The resultant viscous product they termed a starch paste. These investi— gators reported that this swelling was not a sharp transition point, as previously postulated, but a phenomenon occurring over a range of temperatures specific for each of the various starches. Using a viscometric technique, they determined that increases in viscosity were gradual and extended over a range of 25° to 300 C. They defined the temperature of gelatinization as the point at which the initial stage in the process begins. Whistler and Smart (59) reported that the gelatinization ranges may be used as an aid in distinguishing starches from various sources. Durine the process of swelling, the starch granules absorb water slowly at first and the viacositv of the suspension is not preceptably altered. At the critical temperature the granules undergo sudden rapid swelling to several hundred times their original size, and the viscosity of the suspension is greatly increased. Crossland and Favor (22) reported that some starches exhibit more than one stage of gelatinization. In 1926 Alsberg (1) made the observation that swollen starch granules did not burst even after many hours of boil— ing. He stated that boiled starch suspensions were not true solutions unless the cooking period had been unduly prolonged or other treatment had been applied to disintegrate the sus- pended particlea. This investigator interpreted gelatiniza— tion to be dependent upon the softening of the rigid structure of the granules by moist heat. From this he concluded that the amylopectin fraction, located in the outer layer, pre- serves the suspensoid character of the boiled starch. Katz and co—workers (39) reported that the swollen starch granules were surrounded by vesicle walls which enve10ped the soluble starch material. Whistler and Smart (59), however, stated that starch granules had no definite membranes and that the apparent sacs were artifacts produced in the process of swelling. Bechtel and Fischer (10) reported that starch granules from the potato and those from highly modified starches were easily ruptured. Granules of native or unmodified corn starch they found to be strongly resistant to rupture even at high temperatures and with vigorous agitation. Hadley (#9) described a starch paste as a mixture of granules at various stages of swelling, some soluble material, and debris from the outer portion of the burst granules. According to this investigator, the larger granules appeared to swell at a lower temperature than did the smaller granules. He pointed out that this difference in swelling behavior of large and small granules may be due to a difference in the ratio of the amylose and amylopectin fractions. French (27) reported that swelling of starch granules could be induced at room temperature by many alkalies and metallic salts. Viscosity Hadley (”9) stated that starch granules occupied a much larger volume in the system after swelling and gelati— nizing than before and that viscosity altered more or less with the swelling. Kerr (10) states that starch nastes nossessed an anomalous viscosity instead of viscosity in the same sense as that observed in the more nearly perfect fluid bodies. The effect measured appeared to this investigator to be due to a combination of nronerties. Bechtel and Fischer (10) referred to this phenomenon in starch pastes as apparent viscosity and attributed it to the presence of swollen granules, granule fragments, and colloidal oarticles. Anker and Geddes (5) stated that apparent viscosity demanded on the extent of aggregation of the granules, of swelling, and of granule disintegration. Schoch (51) attributed viscosity to be the result of the mechanical jostling of the swollen starch granules. He re~ oorted that these swollen masses could be disintegrated to nroduce a oaste of greatly recuced viscosity by autoclaving. Katz and associates (19) stated that viscosity was due to the friction of the swollen starch granules against one another and to the inhibition of the flow of water by the suspension of starch vesicles. Whistler and Smart (59) attributed the great difference in viscosities of various starch nastes to the degree of swelling and the ease with which granules were runtured. Gel Formation According to Brimhall and Hixon (l6), viscosity measuree ENHfits shonlfi be made only on hot pastes. When a starch paste 11 is cooled, portions of the dissersed material tend to revert to a more insoluble state by a process called retrogradation. The workers believed that the setting-up of starch gels was caused by the interlacing of crystalline starch fragments between and within the aggregated starch granules. Therefore, they explained the rigidity of cooked starch pastes as the result of crystallization as well as an increase in the degree of structural viscosity. Kerr (“0) attributed retrogradation of cooked starch pastes to the amylose molecules with the straight chains. Schoch and Elder (52) stated that the branched starch fraction had less tendency to retrograde and also had a moderating influence on retrogradation of the linear fraction. They found that only extremely high concentrations of the branched fraction would harden to a gel on standing at room temperature. Retrograded linear fraction could not be reversed by autoclav- ing, whereas associated branched fractions returned to their original state upon heating to a temperature of 50° to 60° C. Anker and Geddes (h) observed from their studies that starch pastes had gel prOperties when quiescent, became more fluid on the application of shearing force, and again formed gels after the disturbing force had been removed. Brimhall and Hixon (15) studied the rigidity of starch pastes and observed that viscosity and gel strength measured two differ- ent pronerties in starch. Woodruff and Machasters (60) studied corn and wheat starches from various sources and 12 concluded that viscosity differences in the starches studied were very small, whereas differences in gel strength were more pronounced. Knowles and Harris (uZ) conducted a similar study and found no correlation between viscosity and gel strength. Schoch (50) stated that the speed of retrogradation was dependent on the starch concentration and varied directly with it. He concluded that the associative force between the starch molecules themselves was more powerful than that between the starch and water. Kerr (no) pointed out that retrogradation could be hastened by freezing. Factors Affecting Behavior of Cooked Starch Pastes Some of the factors which have been reported to affect prOperties of starch pastes are inherent starch characteristics, treatment of starch during manufacture, starch concentration, presence of electrolytes and non-electrolytes, mechanical agitation, and the tine and temperature of cooking. Source: inherent characteristics Whistler and Smart (59) stated that, although starches may be almost identical in chemical composition, those from the various sources produce pastes of very different character. This difference in behavior they attributed to the complexity of the colloidal prOperties of the starch pastes. Beckford and Sandstedt (11) studied gelatinization of two types of wheat starch by means of a light transmission 1? device. These investigators observed that large and small granule wheat starch preparations gave similar swelling curves and concluded that these two types of wheat starch granules were similar in gelatinization characteristics. Morgan (b6) employed a similar photoelectric method to follow changes in tuber and grain starch pastes during heating. The tuber type of starch showed steeper pasting curves than did grain starches. Waxy corn starch pasted completely over a narrow temperature range below 80° C. This was also charac~ teristic of the behavior of sweet potato and tapioca starches. Tanner and Englis (57) studied starches from different varieties of corn and found distinct differences between hard and soft corn starches. Granules of hard starches appeared to be smaller and less rounded than soft granules which con- tained more nonecarborydrate material. Hard starches formed more viscous pastes than did the soft starches of approximately the same granule size. Mangels and Bailey (DL) compared the viscosities of four types of wheat starches treated with cold palatinizing reagents. These different wheat starches showed great variations in swelling capacity; the extent of swelling appeared to be deter— mined by the type of gelatinizing agent employed. These investigators concluded that variations in properties of starches were due in part to morphological differences in the starch granules as well as to certain complex chemical differences. 11; Harris and Jesperson (33) made a study of the effects of various factors on the swelling of a variety of cereal starches. These investigators noted that all the starches studied varied among themselves in swelling capacity at all temperatures; barley starch showed the lowest and corn starch the highest values. Woodruff and Nicoli (61) observed the behavior of cooked starch pastes made from various cereal and root starches. The gelatinization temperatures varied with each individual type of starch and ranged from 69° C for potato starch to 87° C for wheat starch. Hughes and couworkers (37) compared temperature and time periods for gelatlnization of starches and cereals of Wheat and corn. These investigators observed that the starches were more rapidly and completely gelatinized at any given temper- ature or time period of heating than the cereals containing those starches. Harris and Jesperson (1b) found that swelling power de- creased as wheat matured, irrespective of variety and environ— mental factors. At the same time, gel strength and viscosity tended to increase with degree of ripeness. These changes with maturity were associated with changes in other factors such as moisture content and density. Little difference in viscos~ ities was noted between wheat, rice, and barley starches. Potato starch, however, exhibited a much higher viscosity and swell than did the cereal starches. 15 Woodruff and MacMasters (60) studied gelatinization and retrogradation changes in corn and wheat starches. These in- vestigators noted that gel strength fluctuated widely with the variety of corn and with the growing cOnditions under which it' was produced. Knowles and Harris (h2)'made Observations on the behavior of starch gels from different classes and varieties of wheat. Environmental and climatic conditions under which wheats were grown showed definite effects on gel strength. Differences were noted between the same varieties grown in two different years. Tests also revealed an increase in gel strength with increasing maturity of the-wheat. Bechtel (9) stated that certain native starches, such as tapioca, potato, and waxy maize, showed less tendency to form firm gels than did the ordinary cereal starches such as corn and wheat. Effect g£_treatment during manufacture Tanner and Englis (57) studied effects of grinding hard and soft corn starches. Grinding corn 62h hours was sufficient to disintegrate hard starches but did not completely disrupt soft corn starches. French (2?) stated that the application of physical stress to the starch granule disarranged the molecules rather than merely breaking the crystals apart. He concluded that bonding between crystals was stronger than the forces be- tween the polysaccharide chains within the crystals. According to Radley (#9), starch granules injured by grinding formed pastes of lower viscosity than those formed with uninjured granules. 16 Kerr (30) stated that after the native starch has been prepared, it is frequently modified to make it more suitable for snecific uses. These various treatments during manufacture may produce very minor to very significant changes in the physical and, in some cases, the chemical structure of the starch granule. Treatments have been developed to produce starches of very high viscosities at specific temperatures, starches of reduced viscosities or thinmbolling starches, and starches which produce a paste with cold water. Schoch and Elder (52) described some of the types of modifications used in commercial starch manufacture. The simplest type of manufactured starch is the thin~boiling starch, which 18 prepared by suspending ungelatinized starch in warm, dilute acid. The acid hydrolyzes a few of the gluco- sidic bonds and weakens the intermicellar structure within the granule without altering the outward apoearance. Such starches produce pastes of relatively low viscosity. Enzymes can be applied in a similar way to weaken the intermicellar structure. Thick-cooking products are prepared by chemically cross—bonding these intermicellar areas with such agents as formaldehyde. These granules swell more slowly and show an increased resistance to mechanical disintegration, producing a paste of high and stable viscosity. Heating dry starch with traces of acid brings “about partial hydrolysis to produce dextrins. Ordinary starches can be oxidized to reduce the influence of the linear fraction so that it no longer associates or retrogrades. 17 Harris and Jesperson (3?) found the swelling curve to be greatly altered in pastes made from starches oretreated with sodium hydroxide. These divergences in pasting charac— teristics increased with rising temperature. According to Sjostrom (55), granules of thin-boiling starches increased in volume less than granules from ordinary starch when heated in water. The room occupied by these swollen granules was less in relation to total volume and the pastes were correspondingly thinner. Morgan (U6) found that tapioca starch, with increasing degrees of oxidative treatment, produced pastes at a lower temperature and with greater clarity. Other modified starches showed similar prOperties. Bisno (13) stated that the neutral and alkali starches tended to show greater resistance to acid, somewhat greater clarity, but had weaker gels than did pastes made from chlori— nated starches. Felton and Schoomeyer (2b) described a method of producing starches with granules that were tough and resistant to disintegration when they were heated with water. These starches were prepared by treatment with an acid chloride in the presence of large amounts of water at a pH of 8 to 12. When a suspension of this modified starch is cooked, swelling occurs at a much slower rate than it does in a similar sus— pension of the raw starch. Viscosity also tends to increase over longer periods of cooking. 18 Bechtel (9) studied properties of gels made from regular commercial grades of unmodified and modified corn starches. This investigator found the effect of cooking temperature on the properties of starch gels varied with the degree of modification. With increasing modification of the starch, the viscosity of the hot pastes decreased and the breaking strength of their gels decreased to a lesser degree. Hadley (“9) stated that the drying treatment to which starch is subjected during manufacture may affect the viscosity of starch pastes. A starch dried slowly at very low temper- atures shows greater viscosity when made into a paste than one dried quickly at higher temperatures. Effect 3: electrolytes French (27) stated that starch granules which have been hydrolyzed by the action of acids no longer swell in hot water but disintegrate and pass into solution. The acid breaks down starch to smaller molecules which are incapable of forming the giant networks characteristic of swollen starch granules. Bechtel (8) discussed the effect of differences in pH on the viscosity of starch pastes. This investigator stated that differences in pH altered the temperature of initial viscosity rise and maximum viscosity and also changed the maximum viscosity values and the rate of paste breakdown. 19 Harris and Banaaik (7?) studied the effect of the elec~ trolytes on swell of corn and wheat starch pastes. Treatment with 0.125 N sodium hydroxide during preparation in the lab— oratory increased the pH and greatly increased the swell. Preparations treated with 0.12? N sulfuric acid lowered the pH and increased swell only slightly. Starch pastes cooked in the presence of hydrochloric acid and sodium hydroxide also showed an increased swell. These investigators suggested that H+ and OH- ions were more effective in increasing swell than other ions studied. Sodium chloride and magnesium sulfate decreased swelling to some extent. Cooking starches, previously prepared with sodium hydroxide, showed marked decreases in swell in the presence of acid. These writers concluded that their results emphasized the complexities involved in tie prohlem of the effect of electrolytes on the swelling power of starch. Anker and Geddes (h) reported from gelatinization studies on commercial wheat starch suspensions that maximum paste viscosity decreased in linear fashion With an increase in pH from 5.2 to 6.8. Schoch (50) stated that retrOgradation of starch pastes is favored by a low pH. Raine (29) reported the effect of 0.1 N citric acid at levels of U0, hU.5, h9.5 per cent of the water content of the formula on the viscosity of sweetened and unsweetened starch pastes. Acid decreased viscosity of pastes made with no sugar and with 20 per cent sugar but had no appreciahle effect on the pastes containing an and 60 per cent sugar. The gel strength of the starch pastes at all sugar levels increased when they were cooked with citric acid. Morse and co—workers (h?) added sodium chloride to pastes made with flour, water, and nonfat dry milk solids. Sodium chloride increased the viscosity to some degree in the thin oastes but had no effect on gel strength of the thick castes tested. Alteration of the OP from 6.9 to 6.0 in the thick pastes resulted in a decrease in gel strength; a rise in 0H from 6.9 to 7.7 showed no consistent effect. Thin oastes were curdled by lowering the UH from 6.9 to 6.0 and at s 0H of 8.0 there was a slight but inconsistent rise in viscosity above that observed at 6.9 pH. Hadley (b9) asserted that certain salts and alkalies lowered specific gel ooints of starch. He found that the Dresence of the very small amount of electrolytes in tan water resulted in lowering of the viscosity of starch pastes. Bisno (1?) stated that the solids present in tap water were sufficient to affect significantly the consistencies of starch castes. Differences in flow rate were noted when distilled water and tan water were comnared as liquid medium for starch pastes. The effect of OH variation uoon viscosity did not aooear to lfie the same for starches fron different sources. However, Bisno oointed out that much of the data from various investi— ewltors was not comoarable because of differences in methods of Study. Ferree (25) found that sweetened and unsweetened 21 cornstarch pastes made with tap water were more unstable on continued cooking after maximum viscosity was reached than pastes made with distilled water. Effect 2; starch concentration Bechtel and Fischer (10) observed that the initial vis- cosity rise occurred at successively lower temperatures as starch concentration was increased. The rate of rise of the curve was more rapid, the maximum viscosity occurred at a lower temperature, and the decrease in apparent viscosity on continued cooking was more rapid. Anker and Geddes (h) supported these findings. Bisno (12) stated that with increase in dry starch concentration a relatively greater viscosity drop is observed with continued heating and agitation. .This investigator attributed this phenomenon to the increased internal stress developed from the numerous contacts among the granules. Katz and associates (39) reported that there was no proportionality between viscosity of pastes and concentration of starch. According to Harris and Jesperson (31) the lowest starch concentrations, on a dry basis, gave the highest swelling power. ~ Grassland and Favor (22) studied starch gelatinization in 11 viscous, water-binding, dispersion medium of low solids content. These investigators observed that the temperature of initial swelling was more sharply defined as the level of starch was ifuereased. The temperature of initial swelling remained 2? fairly constant at all starch levels in contrast to results Obtained when water alone was used as the dispersion medium. Varied levels of starch produced a greater influence on the contours of the curves than did variations in concentration of the medium. Schoch (50) reported speed of retrogradation to be dependent on the concentration of the starch in the paste or solution. An autoclaved 1 per cent starch solution remained fairly clear for several days, whereas a 2 per cent solution flocculated overnight. French (27) stated that dilute starch pastes, thinned by mechanical treatment or boiling, showed less tendency to set rapidly to strong gels than did thick starch pastes similarly treated. Trempel (58) concluded that types and proportions of other ingredients included in a formula may influence the concentration of starch required to thicken a given amount cf liquid to produce a pie filling of desired consistency. Effeg£_9£_timfi_and temperature Alshers (1) concluded from his studies on starch that there was no such physical constant as palatinization temper» ature. Results of his work indicated that there was only a rmnPP of temperatures over which initial swelling of starch granules occurred. This investigator stated that most starches remained swollen but intact even after boiling for an hour. Alsaherg and Rask (2) observed that the viscosity of a suspension 23 of wheat starch in water increased gradually over a range of 25° to 30° C during gelatinization. These workers stated that there was probably no definite temoerature of gelatin— ization. If the temoerature of gelatinization is defined as that ooint where anisotrooy disapnears, then it must be regarded as the temnerature of the beginning of the gelatinization process. Gelatini7ation temoerature, they concluded, could not be viewed as a sharo transition point, soecific for the various starches as the melting ooint is for crystalline substances. Schoch and Elder (52) stated that little change occurred until a certain critical temoerature was reached during the heating of a starch susoension. Gelatinization began suddenly and because some granules swelled more slowly than others, continued over a range of several degrees. After initial raoid swelling, the individual granule continued to expand more slowly with a continued rise in temperature. Unless stirred vigorously, swollen granules were not disrupted but behaved as individual gelled bodies, coherent and elastic. Autoclavinp a starch naste at 18 to 20 pounds steam pressure for an hour or two was required for solution of the starch granules. Hughes and co—workers (37) studied the effect of various temperatures and time periods on the gelatinization of cereals and starches from wheat and corn. Comnlete gelatinization of corn and wheat starch was obtained at 100° C after the solutions were boiled two minutes. A shorter period of heating was required when a higher temperature was used. Granular cereals were not completely galatinized when subjected to similar ‘ heating treatment. Beckford and Sandstedt (11) studied starch gelatiniza- tion with a spectrOphotometer. The rate of heating was con- trolled by a variable transformer. These investigators found that rates of heating, ranging from 2.’° to 0.150 0 per minute, produced identical pasting with no significant changes in gelatinization temperatures of wheat and corn starches. Harris and Jesperson (33) stated that the viscosity of any particular starch was influenced by the pasting temperature and the duration of heating. These investigators, finding very significant differences between the swelling power at the different temperatures, reported that the swelling power increased greatly with the gelatinization temperature. Anker and Geddes (b) studied gelatinization of wheat starches with the amylograph. In this instrument the thermo- regulator allowed the temperature to rise at a constant rate of approximately 1.50 C per minute. Wheat starch suspensions, gelatinized from initial temperatures above “5° 0, gave markedly higher peak viscosities than did corresponding sus- pensions gelatinized from lower initial temperatures. Bechtel (8) investigated paste characteristics of varioud starches with the Corn Industries viscometer. Viscosity curves for a starch slurry prepared at 25° C with the water bath preheated to 920 C were compared with those of a similar starch slurry prepared at 25° C with a bath of the same 25° C initial temper— ature.v Viscosity curves on unmodified cornstarch indicated that rapid heating lowered the temperature of initial viscosity rise and decreased the time reguired to attain maximum viscosity after the paste reached 90° C. With rapid heating, 2 minutes were required to reach maximum viscosity after the paste temperature had reached 900 C. With slow heating, 7 minutes were required for naximum viscosity to be reached after the paste temperature was 90° C. Rapid heating increased the maximum viscosity value and resulted in a correspondingly higher curve for the remainder of the test. In all unmodified corn and wheat starches and in the modified corn starches tested, higher temperatures favored more rapid paste break- down; small temperature differences also resulted in very different viscosity characteristics. Effect of fatty acids Whistler and Smart (59) reported that cereal starches usually contained 0.5 to 1.0 per cent fatty acids. According to these workers, these acids are not combined directly with the carbohydrate but adsorbed on the carbohydrate and may be completely removed by extraction with aqueous methanol. Waxy corn starch contains only 0.06 per cent fatty material. Bechtel (8) found that defatted corn starch differed markedly from normal corn starch throughout its viscosity 26 curve. Pastes made from defatted corn starch were found to show initial viscosity rise and maximum viscosity at lower temperatures than pastes made with normal corn starch. These pastes made with defatted corn starch also had lower maximum viscosity values than pastes made with ordinary corn starch. The rate of breakdown of pastes made from defatted corn starch was usually greater than that of pastes from the corresponding commercial starch. Bisno (1?) stated that vegetable shortenings were fre- quently added to pie fillings to increase tenderness and to retard quick~drying of the starch—thickened mixture. He reported that addition of monoglycerol stearate and poly- oxyethylene stearate tended to delay retrogradation. Accord~ ing to Schoch and Elder (52), the linear fraction of starch forms insoluble complexes with these polar organic substances. They found that the monoglycerides or the higher fatty acids precipitated the linear fraction from a cooked starch paste and thus produced a paste of increased opacity, which showed short, thick consistency and almost a complete loss of gel strength. Starch pastes cooked with polar fatty adjuncts had a higher pasting temperature, retarded swelling, and no break» down or dissolution of the granule. Commercial corn starch ordinarily contains about 0.65 per cent free fatty acids bound by the linear fraction. If corn starch is defatted by alcohol extraction, the resulting starch gives a paste curve with lower temperature of gelatinization and much lower 27 viscosity. The curve can be restored by introducing lipid material into the defatted starch. Jordan and co-workers (18) studied the effect of homoge- nized milk upon the viscosity of cornstarch puddings. The Brookfield viscosimeter was used for measurement of pasting characteristics. When a cornstarch concentration of three grams or more per 100 grams of milk was used, the puddings made with homogenized milk were more viscous when hot and firmer when cold than similar puddings made with non- homogenized milk. The degree of difference was dependent upon the level of cornstarch. These investigators attributed these differences in consistency to the differences in the surface area of the fat globules in the two kinds of milk tested. Morse and associates (“7) studied the effect of nonfat dry milk solids on viscosity of thin pastes and gel strength of thick pastes made from flour and water or fluid milk. The Stormer viscosimeter was used to measure viscosity of thin pastes and the Bloom gelometer to measure gel strength of heavier pastes. The nonfat dry milk solids increased the viscosity of the thin pastes and gel strength of the thick pastes in proportion to the amount of dry milk used. The addition of fat to the same pastes decreased the gel strength in the heavier pastes but did not affect the viscosity of the thin pastes. 28 Effect f sucrose In 1911 Woodruff and Nicoli (61) studied gels of 5 per cent cereal and root starch pastes. Sucrose was added at levels of O, 10, 30, 50, and 60 grams per 100 grams of starch paste. Differences in gel characteristics were described and photographs served to record differences in physical appearance and outline of the gel. In all starches studied, each increment in sugar concentration up to 50 per cent produced a gel of more tender and transparent character. No gel was formed with the addition of 60 per cent sucrose. Hains (29) reported viscosity measurements of 12 per cent cornstarch puddings with sucrose added at levels of O, 20, ho, and 60 per cent by weight of the liquid. A sucrose con- centration of 20 per cent produced a pudding with a higher viscosity than noted in the control without suCrose. Puddings containing this level of sucrose required only a slightly longer cooking time than did the control. Additions of no and 60 per cent sucrose, however, produced pastes of progrese sively decreasing maximum viscosity values which required higher temperatures and much longer cooking periods. Ferree (25) studied the effect of the proportion of sucrose on the viscosity of corn starch pastes using different liquid mediums. Sucrose concentrations of O, 15, 21, and 27 per cent were used in this study. The liquid mediums were distilled water, tap water, nonfat dry milk solids reconstituted 29 in distilled water, and whole dry milk solids reconstituted in distilled water. Starch slurries containing 27 per cent sucrose made with all liquid mediums required longer cooking times and higher temperatures to reach maximum viscosity than did the control pastes made without addition of sucrose. Lower sucrose concentrations did not consistently alter the cooking time or the temperature required by the control to reach maximum viscosity. All sucrose concentrations appeared to decrease the rate of breakdown of the starch paste with increase in cooking time. From a comparison of the maximum viscosity values of sweetened and unsweetened pastes, viscosity appeared to depend more on the particular liquid medium than on the sucrose concentration. In general, the results did not show a consistent decrease in viscosity with an increase in sucrose concentration. . Trempel (58) stated that the addition of certain solids, such as sugar, raised the pasting point of starch. He found that, under normal cooking conditions, a pie filling contain— ing more than 3% times as much sugar as starch (by weight) produced a thin gel on cooling which had a raw cereal taste. This characteristic he attributed to incomplete cooking of the starch. In preparing fillings with relatively high sugar concentrations, Bisno (13) suggested that 2/3 of the sugar be added after the starch paste became transparent and thick. This writer stated that sugar competes with starch for water; at high levels of sugar concentration the amount of available 30 water is reduced to the point at which complete gelatiniza- tion of the starch cannot occur. [£21393 2: mechanical agitation Bechtel and Fischer (10) concluded from their experiments that the rate of pasting was affected by differences in agitetion. The rate of pasting in the double boiler was found to be approximately double that in the Corn Industries viscometer. Differences in agitation, however, apparently did not alter the nature of the pasting process. Corn, wheat, and unmodified starches exhibited a somewhat smaller chance than did potato and acid—modified corn starches in apparent viscosity with variations in the rate of shear. Schoch and Elder (52) stated that when starch paste was subjected to violent shear, as in the Waring Blendor or by high-pressure homogenization, the opposing stresses might actually break glucosidic bonds rather than destroy the individual micelle. French (27) stated that dilute starch pastes may be thinned by boiling or by other mechanical treatment such as passing them through a homogenizer. Pastes which have been treated in this manner have less tendency to set rapidly into strong gels. At higher concentration, however, gels are readily formed even from thoroughly dispersed starch pastes. 71 The Waxy Starches A relatively new kind of starch extracted from waxy and glutinous varieties of certain cereals is being manufactured in increasing quantities in the United States. These starches nossess many unique characteristics and are gaining acceptance for specific uses in food oreoaration. In 1909 Collins (19) reported on a new tyoe of Indian corn from China. He stated that in Varch, 1908, the office of Foreign Seed and Plant Introduction received a small sample of shelled corn from a missionary in Shanghai, China. Plants were grown from this seed the followina season and proved to be quite unlike any of the cultivated varieties known in the United States or Tropical America. This new type of corn plant showed a resistance to drying out of the silks by dry, hot winds. The endosoerm of the grain was completely opaque as onoosed to the translucent appearance of the common varieties known in the United States. The texture of the endosperm suaeested that of hardest wax, and from this resemblance the term cereous or waxy endosoerm was suggested. Because this waxy endosoerm was completely recessive to the horny and starchy endosoerm of our common varieties of corn, its appearance in all kernels of the original seed indicated that the seed was grown in a section of China where varieties of corn with a horny endosoerm were completely absent. In 1913 Collins and Kemoton (20) stated that maize varieties were based on the character of the seed. The 72 varieties with horny, soft, and sweet endosperm represented the only forms known to exist until the discovery of the waxy endo- sperm in a variety of maize introduced from China in 1908. Brink (I?) explained the difference in the character of reserve starch in waxy maize from the endosoerm and pollen of non-waxy maize. He concluded that both kinds of starch are composed of sugar units but that the intermediate compounds between the polysaccharides and the sugars differed in molecular structure or in degree of association of their elementary molecule. He also pointed out that waxy starch contained only about one— twelfth as much combined organic phosphorus as did common maize starch. In 19b? Sprayue and associates (56) presented evidence that the waxy gene in corn was not completely recessive in its influences as had been previously believed. These investigators asserted that starches of intermediate character— istics between waxy and starchy endosperm could be produced. Schoch (51) stated that plant geneticists have discovered waxy strains of sorghum, rye, barley, and rice. No waxy variety of wheat has yet been found. It is believed that the waxy strain represents a primitive variety which has been completely bred out of wheat during its long history of cultivation. Waxy strains of the root and tuber starches have not been reported. Hixon and Spracue (35) stated that starches from waxy maize and other waxy cereals stain red—brown rather than blue with iodine. Corn exhibitins these characteristics was termed ”waxy"; rice, millet, and sorghum with similar properties were termed “glutinous". These starches possess many other unique properties. Granules from waxy and ordinary corn starch appear alike with respect to size and shape, crystalline x-ray patterns, and diaestion with B—amylsse. The essential differ- ences in the properties of waxy corn starch and ordinary corn starch appear to lie in the molecular structure. It is thought that these structural differences may be responsible for differences in behavior of pastes prepared from these starches and pastes prepared from common corn starch. Gelatinization of ordinary corn starch begins at about 6&0 C and continues over a range of 30°. Pastes of waxy starches show initial rise in viscosity at about 70° C and progress uniformly and completely within a narrow ranpe of 8°. The shape of the latter aelatinization curve resembles that of tapioca rather than corn starch. The hot viscosity of dilute waxy corn starch pastes was reported to be greater than that of tapioca from 75° to 90° C but less resistant to breakdown on continued cooking (76). Morgan (h6) reported similar findings concerning the pasting curves of waxy corn starch pastes. Schopmeyer and associates (5b) stated that waxy corn starches are almost completely lacking in the linear fraction, whereas potato, tapioca, and the common corn starches have been shown to contain up to 22 per cent of the straight— chained glucose polymer. They explained that the absence of .J.‘ this gel—formin: constituent contrihutec to the fluid char— acter of oastes made from both unnofiified and modified waxy corn starch. According to Whistler anfl Smart (59), starches from waxy varieties of corn, sorghum, rice, barley, and millet may contain only smylonectin or, in certain varieties, uo to 6 ner cent amylose. Schonmeyer and courorkers (Sb) stated that the yield of the orisinal Chinese waxy corn was too low to be considered a commercial source of waxy corn starch. Through a coooerative breeding nrogram carried out by the Iowa Agricultural Experi~ ment Station and the United States Deoartment of Agriculture an imoroved variety of waxy corn was develooed in 19b1. Part of this seed was nlsnted in Iowa and oart in Indiana. Another quantity of ooen-ocllinated waxy corn obtained from the University of Nebraska was olanted in Nebraska for the National Starch Products Comoahy. The croo harvested from all these lots was combined anc processed by American Maize—Products Comnany. The starch thus obtained represented the first commercial nroduction of waxy maize starch in the United States. Schonmeyer (53) stated that the oroduction of amioca (a generic term adootpd by all meshers of the starch industry for waxy corn starch) has flevelooed at a ranid rate since its hepirnins in l9h3. Chemical analysis shows that waxy corn has a consistently higher content of oil and of water soluble material than does common corn. The amioca granule is composed entirely of amylooectin, and its properties are very similar I to those of amylopectin fractionated from common starch. Although unswollen amioca granules annear very much like those .of common corn starch, the two can be differentiated readily by their color reaction with iodine. ASChOnmeyer carried out a study of the effect of various agents on the viscosity of amioca to determine the usefulness of the starch. At low concentrations, the hot viscosity of unmodified amioca was much higher than that of tanioca and similar to that of ootato starch. With an increase in concentration of the starch and length of cookins time, the viscosity of amioca was only slightly higher than that of tanioca and lower than that of notato starch. Mixtures of amines with common starch shoved a higher viscosity than either amioca or common corn starch alone in the same concentration. Amioca, taoioca, and corn starch nastes, adjusted to varying UH values with citric acid and allowed to stand overnirht, were comnared for viscosity changes. Both amioca and tanioca showed a marked decrease in viscosity at pH values of 5.5 or higher and were comparatively stable in the range of 3.5 to 5.0. Both thinned out at pH values below 3.5. Ordinary corn starch did not show sensitivity to OH changes. Liquifying enzymes were found to breakdown amioca and taoioca at about the same rate. Common corn starch_oroved more resistant to the action of these enzymes. The addition of amioca greatly retarded the gelling of common starch, even at high concentrations of the latter. Although concentrated oastes of acid—modified amioca remained 36 soft and fluid on cooling, ordinary corn starch with similar modification formed a brittle gel when cooled. Schoch and Elder (52) stated that tapioca and waxy maize starches produced pastes of "long“, ochesive stringinese. They found that these starches required less cooking and yielded clearer pastes than corn starch but gave very high peak vis— cosities and thinned out quickly on continued cooking. The internal cross—bonding within the waxy starch granules by treatment with small amounts of acid chloride or other agents eliminated the stringy paste character and stabilized the viscosity. Schoomeyer and associates (5“) pointed out that modification or waxy starches by treatment with acid or oxidizing agents improved the product for specific uses. These starches are characterized by remarkable clarity, high fluidity, and lack of gel formation on cooling. Bisno (12) reported that waxy corn starch has gradually begun to replace tapioca for use in conjunction with corn starch in the preparation of fruit pie fillings to produce desirable transparency and flow characteristics. If used alone in pie fillings, starches with the amylopectin charac- teristics cause excessive flow. For this reason, thickeners of the tapioca or waxy maize type are usually blended with a gel-forming starch. 1 According to Schoch and Elder (52), a 5 per cent paste of waxy maize starch remained clear and fluid for long periods of time; however, a 30 per cent paste hardened to a gel on 37 standing. In a 5 per cent waxy starch paste, which was frozen and thawed, the branched material was converted to an insoluble state and x—ray diffraction patterns indicated crystalline association of linear chains. Noznick and co—workers (L8) studied the effect of waxy maize starch on the stalinp of bread in which it was an ingredient. Results showed that the waxy maize starch had a very detrimental effect on dough and bread quality and was of no value in decreasing the rate of steline. These results suggest that hread stalinr may be associated with the amylo- pectin component of the starch molecule and cannot be completely attrihuted to the retrogradation tendencies of the amylose fraction of wheat starch. Hanson and co—workers (10) made extensive studies to determine possible factors contributing to liquid separation in sauces and gravies subjected to freezing temperatures. When waxy starches and flours were used as thickening agents for these frozen sauces and gravies, liquid separation upon thawing was almost completely eliminated even after several months of frozen storage. Waxy rice flour was found to be he beat thickening agent, but the other waxy flours and starches were markedly superior to ordinary flour. These investigators pointed out that all waxy starches are not the same. The derree of branching of waxy rice starch lies between that of ordinary amylopectin and glycoeen. The results of these studies may reflect a greater amount of branching in molecules of waxy cereals than in molecules of common cereals. The results further suggest variation in the extent of branching and molecular size of the amylooectin present in the different waxy cereal starches. The instability of the thawed sauces aoneared to be associated with the amylose fraction of the starch molecule and, to a lesser extent, with the molecular size and degree of branching of the amylooeotin fraction. Hanson and associates (31) resorted the effect on the stability of frozen nuddings of substituting waxy rice flour for corn starch and for part of the egg in frozen custards of both soft and baked varieties. A definite improve- ment was observed in the stability of the thawed ouddings and 50ft custards containing waxy rice flour as all or part of the thickening agent. Those ouddinas and custards containing waxy rice flour stored at ~10° F showed greater stability after long storage oeriOCs than did nuddings and custardq stored at 0° F. Liquid separation in baked custards subjected to freezing was not prevented by using waxy rice flour as part of the thickening agent. Davis and co-workers (2?) oreoared oastes and thickened carrot sauna with ordinary and waxy cereal starches and flours as thickening agents. ‘These products were canned, autoclaved, and stored at room temperature. The waxy starches and flours appeared to orevent retrogradation and liquid seoaration in these canned food products. The results of these investigators 38 branching in molecules of waxy cereals than in molecules of common cereals. The results further suggest variation in the extent of branching and molecular size of the amylopectin present in the different waxy cereal starches. The instability of the thawed sauces appeared to be associated with the amylose fraction of the starch molecule and, to a lesser extent, with the molecular size and degree of branching of the amylonectin fraction. Hanson and associates (31) reported the effect on the stability of frozen puddings of substituting waxy rice flour for corn starch and for part of the egg in frozen custards of both soft and baked varieties. A definite improve- ment was observed in the stability of the thawed puddings and soft custards containing waxy rice flour as all or part of the thickening agent. Those puddings and custards containing waxy rice flour stored at «10° F showed greater stability after long storage periods than did puddings and custards stored at 0° F. Liquid separation in baked custards subjected to freezing was not prevented by using waxy rice flour as part of the thickening agent. Davis and co—workers (2?) prepared oastes and thickened carrot soups with ordinary and waxy cereal starches and flours as thickening agents. 'These products were canned, autoclaved, and stored at room temperature. The waxy starches and flours appeared to prevent retrogradation and liquid separation in these canned food products. The results of these investigators branching in molecules of waxy cereals than in molecules of common cereals. The results further suggest variation in the extent of branching and molecular size of the amylopeotin present in the different waxy cereal starches. The instability of the thawed sauces appeared to be associated with the amylose fraction of the starch molecule and, to a lesser extent, with the molecular size and degree of branching of the amylonectin fraction. Hanson and associates (31) reported the effect on the stability of frozen puddings of substituting waxy rice flour for corn starch and for part of the egg in frozen custards of both soft and baked varieties. A definite improve- ment was observed in the stability of the thawed puddings and soft custards containing waxy rice flour as all or part of the thickening agent. Those puddings and custards containing waxy rice flour stored at «100 F showed greater stability after long storage periods than did puddings and custards stored at 00 F. Liquid separation in baked custards subjected to freezing was not prevented by using waxy rice flour as part of the thickening agent. Davis and co—workers (23) prepared pastes and thickened carrot soups with ordinary and waxy cereal starches and flours as thickening agents. 'These products were canned, autoclaved, and stored at room temperature. The waxy starches and flours appeared to prevent retrogradation and liquid separation in these canned food products. The results of these investigators 39 suggests that the use of starches from-waxy cereals may find increasing application in food preparation. Objective Tests for Cooked Starch Pastes Various objective tests are applied in industry for evaluation of starches for specific uses. These tests involve the measurement of colloidal paste prOperties such as viscosity, plasticity, and gel strength. V1 8 enemy ._-.___..m e e e ‘ili’ii’i’i‘i Industry has developed several methods for measurement of viscosity in both hot and cold starch pastes. Several of these procedures are described on the following pages. Scott test for hot paste viscosity (DO). This test is extensively used to determine hot paste viscosity of starch pastes. A quantity of starch at known oH is stirred to a slurry with 280 cubic centimeters of distilled water in a German~silver beaker. The starch is gelatinized by placing the beaker and contents in a boiling water bath. The paste is stirred mechanically and heated for 15 minutes. At some definite period of time prior to 15 minutes, 200 cubic centimeters of the paste are transferred to a Scott viscosity cup, also in the boiling water bath. At the end of the total heating time of 15 minutes, the plunger valve which closes the orifice on the bottom of the cup is raised and the time, in LO seconds, is noted for a given volume of paste to fall into a graduated cylinder. The specific Scott test viscosity of the starch under consideration is measured in terms of the time reouired for a specified amount of oaste to flow from the test cup. The measurement is one of relative viscosity in the sense that standard starches are used to set no permissible limits of variation of the starches to be tested. Storms: viscosimeter (ha). his test may be used to good advantage in evaluating cold paste body. The starch is gelatinized by heating for 15 minutes in a water bath. The paste is then placed in a constant temperature water bath, preferably 250 C, in a closed container. At the end of a definite eeinr period, any surface skin is carefully removed and discarded. The remaining paste is very gently stirred with a spatula for several seconds and then transferred to the cup of the Stormer viscosimeter. The Stormer viscosimeter is comprised of a cylinder immersed in the test paste which is contained in a metal cup surrounded by a water bath. The immersed cylinder is rotated by a free-falling weight acting through a gear-and-oulley system. The time, in seconds, necessary for a given weisht to produce a certain number of revolutions is taken as a measure of cold paste body. The revolution counter is a part of the instrument. bl MacMichael viscosimeter (L0). This instrument may be used for measurement of cold caste body. The aooaratus consists of a cylindrical cuo, in which the starch oaste is Olaced, surrounded by a bath with an electric.heating device to regulate the bath temnerature. The cuo and bath rotate at constant soeed. A second cylinder is susoended in the oaste by means of a torsion wire. The twist that develoos in the torsion wire is read by noting the oosition of a fixed oointer in relation to a movins scale mounted on the torsion wire. The instrument is suoolied with several interchangeable torsion wires to oermit measurement of a wide ranee of viscosities. Hoeooler viscosimet§£_ (DO). This instrument involves the use of a orinciple different from that of the Stormer and the MacMichael viscosimeters. Measurements are made of the time required for a given weight to fall, in vertical motion, through a measured column of paste. The column is surrounded by a constant-temoerature bath so that, with only slight modification, the instrument can be adaoted to the measurement of hot paste viscosity as well as cold caste body. Caesar consigtometer (b0). This aooaratue records a con- tinuous history of casting characteristics of the starch over the entire period from the time gelatinization starts until the paste is cooled to a given temperature. The essentials of the method described by Caesar are to susoend a standard #2 paddle, connected by a shaft to an electric motor, 1“,? beaker containing the starch slurry. The latter is surrounded by a heating bath, and the temperature of the bath is raised at a regular rate. The paddle and the motor move at constant speeds. Changes in the viscosity or paste body are indicated by the differences in the electrical input to the motor necessary to maintain a constant speed of agitatiOn. After the starch is cooked, cold water may be introduced into the bath to cool the paste. Caesar and Moore (18) reported that this apparatus is only sensitive to starch concentrations of 20 per cent or greater. The degree of degeneration of a starch is revealed by the form of the curve, on which paste temperature is. nlotted against the net power in watts required to maintain a constant stirring speed. Brabender amylograph (k). This instrument is a torsion viscosimeter which automatically records the resistance to shear offered by a flour or starch suspension as the temper- ature of the suspension is increased at a constant rate. The cylindrical, tinned-brass bowl in which the suspension is placed contains eight fixed, vertical pins and is rotated at a constant rate in an electrically heated air bath by means of . a synchronous motor. This motor also Operates the kymograph and the device for controlling rate of temperature increase. The measuring device consists of a circular, metal disc to which are attached seven metal pins. The frictional resistance exerted L3 as the bowl rotates around the disc causes the free—moving pins to rotate on their central axis. The torque is trans- mitted through the central shaft and coiled Wire spring to the recording torsion balance. Corn Industries viscometer (ul, 7). This instrument was used by Pechtel in testing various corr starches for the corn industry. It provides for a continuous record of viscosity changes during cooking of starch pastes. The starch slurry to he tested is placed in a stainless steel beaker immersed in a thermostatically controlled liquid bath. A synchronous electric motor drives the stirring mechanism at a constant speed. The stirrer is constructed with scrapers to remove pasted starch from the walls of the container. This mechanism Keeps the paste smooth and of uniforn temperature throughout. A propeller is separately mounted within the hollow shaft of the scraper. The resistance which the propeller meets as it turns in the paste is the basis for noting viscosity changes and is independent of any internal friction encountered by the scraper blades coming in contact with the walls of the cooking vessel. The force thus encountered is transmitted throush the gear differential to the drum and the dynamometer and recorder. A pen attached to the dynamometer traces the record of viscosity changes on a strip—type recorder. The recording paper moves at a speed of one~half inch per minute, giving a continuous record during the cooking period. The L'li weight arm of the dynamometer alone has a torque at full scale of 225 gram centimeters above that of water. Additional weights are provided to extend the range of torque values which can be measured. A thermometer is inserted through the condenser cover into the paste to obtain temperature readings. Lips-spread tests .— 6 Two investigators have described procedures for the determination of flow prooerties of hot and cold starch pastes. These tests involve the measurement of spread of a given amount of material when released from a mold to flow on a flat surface. Grawemeyer and Pfund test (28). For this test a flat glass clate is Dlaced on a level surface; beneath the olate is a diagram of concentric circles soaced 1/9 inch anart and numbered in order of size. The smallest circle with a diameter of two inches is not numhered. A hollow cylinder, the exact diameter of the smallest circle, is placed on the glass plate directly above this innermost circle. The cylinder is filled with the material to be tested and leveled with a spatula. The cylinder is then carefully lifted and naterial allowed to scread for exactly two minutes. At the end of this period, readings are taken at four widely seoarated points that mark the limits reached by the substance. The average of the four readings is recorded as the line~spread of the sample. Nil}§3_test for cold East: flow (#5). This technique was developed for objective measurement of flow of a cut sample of pie filling. For this test a glass plate 12x12x.25 inches, from which a circle 9.175 inches in diameter had been removed, was superimposed upon a second glass plate of similar dimen— sions but without the circle removed. The two plates were bound torether by four rubber bands. A metal mold, nine inches in diameter, was devised from a standard baking pan with a circle seven inches in diameter removed from the bottom. The pan was inverted and placed within the cut-out area of the upper glass plate. A sample cutter was constructed with a diameter of 3.125 inches and a height of 1 inch. An outer band of metal .375 inches wide and .0625 inches thick was welded one~fourth inch from the top edge of the cutter to form a base for the expansion band. A round, stainless steel expansion band, constructed with a diameter of 3.25 inches and a height of .75 inch, was mounted in place so that it rested on the thickness of the outer band of the sample cutter, Three pints of hot filling were poured into the large metal mold adjusted on the double glass plates. The opening was covered with a double layer of absorbent paper and an outer layer of Saran Wrap. The poured molds were allowed to cool at room temperature for a specified time period. At the end of the cooling period, the rubber bands were severed and the coverings removed. The cutting edge of the sample cutter was lubricated with salad oil, and one sample was cut from the center of the mold of filline. The upper glass plate and the metal mold were then removed. With the cutter still in position, the excess filline was removed from around the edges and the glass plate wiped clean. The expansion hand was then removed from the cutter and the sample leveled to a constant depth of 1 inch, the height of the cutter without the expansion band. The cutter was then removed and the sample released to flow. Each sample was permitted to flow 90 minutes before measurements were taken. The image of the sample was recorded by exposing the sample on the glass plate, placed over a sheet of light- sensitive paper, to a photoflood lamp placed in position 12 inches above the sample. Imaees were then deVeloped on the paper with ammonia fumes, and their areas were determined by measurement with the compensating polar planimeter. The amount of flow was determined by subtracting the area of the cutter from the area of flow of the cut sample. Gel strength tests Kerr (30) stated that gel strength of starch pastes has been measured by a variety of instruments based on different principles and involving measurement of different character— istics in a gel. Some of the instruments employed in practice are the rigidometer, penetrometer, and Terr—Baker Jelly tester. These instruments may measure rigidity, elasticity, resistance to cutting action, or a combination of these characteristics. 1+7 Tarr—Baker Jelly tester (b0). The principle of this test for gel strength is the gradual anplication of pressure to a plunger, of known area, resting upon the surface of the gel. The sample to be tested is cooked in a boiling water bath for 30 minutes. The paste is cooled quickly and sufficient water is added to compensate for moisture evaporation. The paste is poured into aluminum moisture dishes, covered with a light film of mineral oil, and placed in a water bath at 20° C for 1 hour. At the end of this aging period, the oil is drained off and the dish is placed under the plunger of the tester. With 500 cubic centimeters of water in the Tarr—Baker Jelly testing Jar at the start, the flow of water is adJusted so that the manometer column rises 60 centimeters per minute. The manometer is read the instant the starch Jelly breaks. Tests are made in triplicate and averaged. The gel strength is reported as the height, in centimeters, reached by the liquid in the manometer at the breaking point of the gel. ~Exchangg_Ridgelimeter (21). For this test, paste batches are made using identical formulas and cookins periods. Jelly glasses are provided with paper-strip extensions acting as sideboards. Containers are filled to capacity, covered, and set aside for 18 to Zn hours. At the end of the aging period, the paper strips are removed and the exposed gel is sliced level with the rims of the glasses with a tightly stretched wire. The gel in the glass is then unmolded in an inverted 58 oosition on a circular glass plate. The olate is placed on the base of the tester and the gel centered under the point of the micrometer. The ooint is made to tonch the gel surface by carefully turning the micrometer head and watching for the first sign of contact. At this point, the sag is read to the nearest 0.1 oer cent. This test measures rigidity of the 991 in terms of percentage of sag. Saare disc method (DC). This test was among the first orooosed for the evaluation of a cold oaste. Kerr described a modification of this method which he has used in research. The starch oaste is cooked in a manner similar to that de- scribed for the Scott test. The paste is noured into a glass vessel to a definite level. A circular metal disc of known diameter is susoended in the paste by means of a metal rod connected to the center of the unoer surface of the disc. The metal rod, which is crooked at the too, is hung over a bar which rests on the top edge of the vessel holding the paste. The length of the rod permits immersion of the disc in the paste to a deoth of three centimeters. A thin film of light oil is olaced on too of the paste and the vessel and contents olaced in a constant temperature bath of 25° C and held for 2L hours. At the end of this time the vessel is removed and placed on a bridge over one pan of a large but sensitive beam balance. A hook susoended from the beam of the balance engages the crock of the rod. Small size shot are added at a fixed MO I rate to the other nan, and the weight of shot is noted at the time the disc fractures the gel. The weieht of the shot measured, less the weisht of disc and all connections to the beam balance, divided by the exact area of the lower surface of the disc is taken as a°l strensth. The result is exoressed in grams oer square centimeter for a given concentration of starch. Fuchs penetrometer (LC). According to Kerr, most of the various oenetrometers which have been nroposed for use in measuring eel prooerties are not sufficiently sensitive to distinguish between the various grades of acid—modified industrial starches. This writer states, however, that the instrument constructed by F.W. Fuchs is sensitive and accurate. For this test, the oaste is cooked as in the Scott test and olaced immediately in a closed, wide-mouth obntainer for storage in a constant temoerature bath. At the end of the aging oeriod a i-inch layer 0f the gel is cut off and dis— carded. The elunser, Which consists of a sharnened, highly oolished, metal tuhe, is adjusted to rest on the surface 0f the oreoared eel. A weight is imoosed on the slunser in a receotacle on too of the rod attached to the too of the olunser. The olunger is released and the time, in seconds, required "5 or the fllnnser to cut into the eel up to certain deoth is noted. SO Recording gel tester. HJermstad (36) described a gel tester -designed to utilize the embedded disk method proposed by Snare and Martens for measurement of gel strength. The system is composed of a device which lowers, at'a constant rate, a gel containing an embedded disk which is attached by a cord directly to a dynamometer cepahle of registering the desired range of _.force. As the gel is slowly lowered, the dynamometer resets upward with a continuously increasing force on the disk until the yield point of the gel is reached. The motion of the dynampmeter is transmitted to s chart-recorder mechanism to oroduce a continuous curve defining the apolied force, the .0 deformation of the gel, and yield-point of the gel. S1 PROCEDVRE Selection of Formula A formula which contained the ingrediente commonly found in cimnle, nueoinp—tyoe desserts was selected for this study. The urcoortionq of ingredients were defined after a oreliminery examinetion of etnnflerfi recioee for cornstarch tyne ouddines and Die filiinve. Pronortione of the individuel ingredients were hngafl nn gtnnfinrfl formulae hut afijusted to yield a finished ornfiuct which had hot onete viscositieq within the range of aensitivitv of the cooking and recorfiinp instrument used. Vanv of the recioee reviewed inclufiefl epgg es an imnortnnt incredient. In nfifiition to contributing to flavor and nutritive value of n orofluct, egg oroteins oroduce a definite thickening effect when coeguleted by heat. Because the thickening effect of the combination of starches wnq the major fnctcr under consideration in this investigation, eggs were omitted from the formula need in this stufiy. A blenc mange type of Dudding, very satisfactory in consiqtency was obtained without them. Starch crwunllgrgtion, b '«C a. C‘m Several different starch concentrations were used in pre— liminary cookinv trials. The orooortion of starch in eimnle blanc mnnpe was founc to vary from 5 to 11 oer cent of the weight of the liquifl. Lower urooortions of stqrch were used in most ouddtng recioeq which included eggs. During the oreliminerv seriod a S oer cent eterch concen— tretion wes tested in the cooking instrument. These 5 per cent starch nestes were somewhet thinner then desirehle for nudding mixtures and registered in the lowest limits of the renee detectable bv the recording soosretus. Starch concentretions were increesed until 2 desirable consistency was obtained. A total starch concentretion of 6.7 oer cent mes finally selected for this study. croeg.congentretio - M“ w - a. -~-c- The concentration of sugar in the pudding and nie filling recioes varied from b to 27 oer cent of the weight of the liquid. A sucrose concentretion of 10 oer cent wes srbitrerilv selected for this study. This amount wes lower then found in many nrevious studies but anceared to be sufficient to yield 9 simole cudding of ecceotehle sweetness. Liquifl_medium h Most nuddiny recipes require some form of milk as the liquid ingredient. At the oresent time dry milk solids are recommended es en economicel and orscticel renlscement for fluid milk. The slirht variations in the comoosition of fresh fluid milk, even when obteined from the same source of suocly, could introduce uncontrolled variables during the course of an extended study. Dry milk solids of known comnosition, obtained from the same processing run, cen be stored for use throughout the study. Whole dry milk solids were selected for use in the formula for this study. These solids were reconstituted in distilled water in the orooortion of 12.01 per cent of the weight of the water in the formula. This percentage was consistent with the compo— sition of solids in whole fluid milk. ‘Distilled water was used because of the variation in the mineral content of tap water. The presence of electrolytes in tap water has been shown to have a definite effect on the Velatinization of starch pastes. _ Salt concentration The proportion of salt used was based on a preliminary check of formulas, and the level was arbitrarily set at 0.217 per cent of the weight of the liquid. Basic Formula The percentages of all dry ingredients in this study were based on the weight of the distilled water. Weights of all in— gredients in the formula, based on predetermined percentage relationships, were calculated to yield batches of approximately one liter, the capacity of the beaker in the cooking instrument. The weights and percentages of all ingredients, except the starches, remained the same in the preparation of all starch slurries. The total percentage of starches remained constant in all slurries. However, the prooortion of the starches consti— tuting this total varied in progressive steps. The kind and proportion of starch or starches used to make up the total amount were the sources of variation in the study. The basic formula was as follows: Ingredients Weicht Per_g:nt_1£_lig1id * (srams) Distilled water 870.0 100.00 Whole dry milk solids 101.§ 12.01 Sugar 37.0 7 10.00 Salt 1.9 .917 Total starch 58.29 6.70 Variations The ouroose of this study was to determine the effect of using a combination of cornstarch and a modified waxy maize starch on the hot paste viscosity and cold paste flow of ouddinswtype desserts. The thickening agents used included ordinary cornstarch and two waxy maize starches representing different degrees of modification. The modified waxy maize starches were designated by the trade names W—13 and Ameizo “£00“. Both of the modified starches undergo a certain degree of treatment during processing from the original raw waxy maize starch (in), During? manufacture the raw waxy maize starch is treated with an acid chloride with which it is very reactive in the presence of water. This treatment reduces the lone nastiness character of the raw starch. The W—13 starch undergoes slightly less treatment than the Amaizo "boo". Fach of these modified waxy maize starches was used in combination with cornstarch at specified oercentages of the total starch concentration. Series A consisted of varying percentages of cornstarch and W—l3 and Series B consisted of similar variations of cornstarch and Amaizo ”#00". One variation using cornstarch alone was prepared as a control. The amounts and percentages of the starches makinr uo the combinations in the two series are shown in the following tables. Table 1. Starch comoosition for Series A. Cornstarch W—lj Variation Gm. % of total starch Gm. % of total starch ' 1 58.29 100 0 O 2 GS.°8 95 2.91 5 3 52.26 90 5.93 10 b 1‘9.55 85 8.7b 15 5 b5.63 80 11.66 20 6 “3.72 75 13.57 25 7 no.8o 7O 17.b9 , 30 Table 2. Starch composition for Series B. Cornstarch Amaigo ”#00” Variation Gm. g of total starch Gm. % of total starch 1 58.29 100 ' o o 2 55.38 95 2.91 5 3 52.h6 90 5.93 10 b h9.55 85 8.7b 15 5 u6.63 80 11.66 20 6 u3.72 75 1u.57 25 7 “0.80 70 17.u9 3o Preparation of Formula Ingredients The cornstarch was obtained from a single lOO-Oound has from the University Food Stores. It was stored in a covered container at room temoarature. The waxy maize starches, desig— nated by the trade names W-l? and Amaizo "500", were supplied by the American Maize Products Comoany and were stored at room temperature in the lined bags in which they were shiooed. These bags were placed in covered containers. All remaining dry inpredients were obtained from the Food Stores. The granulated sugar was placed in a covered container and kept at rOnm temner— ature. One hox of iodized salt was designated for use in this study and kept at room temoereture. Whole dry milk solids', man— ufactured by the snray dryinz process by the Borden Milk Company, were obtained in five—pound sealed cans. The comnosition of the whole dry milk solids was as follows: Butterfat ~—- 28.0% Milk sugar —— 37.7% Protein ————— 26.5% Minerals ~ee~ 5.8% Moisture —-—- 2.0% Freshly distilled water was used as the liquid medium for reconstituting the milk solids in all reolications. The distilled water was kept in a atoooered glass Jug and stored at room temperature. All dry ingredients were weighed out in advance. A triple-beam balance, accurate to 0.01 gram, was used for *Parlac weighing both the cornstarch and,the waxy maize starches. The starches were weighed on souares of stiff white paper from which they were easily removed to the storage containers. All other dry ingredients were weighed on a Torsion balance on squares of wax paper. A rubber scraper was used to aid in complete removal of All material from the wax paper. Sugar, salt, and starches for the individual variations were combined and placed in plastic, moisture proof storage containers and covered with tightly fitting lids. Individual weighines of whole dry milk solids were stored separately in similar storage containers and the lids were sealed on with freezer tape. Distilled water was weighed out in a l—liter glass beaker on a Harvard Trio single-beam balance at the time of each experiment. Wiring procedure Care was taken throughout the experiment to standardize all procedures. From preliminary trials the required time for combininv and pouring slurries was determined. A time schedule for these procedures was outlined and followed throughout the study. Three replications were prepared of each variation. The distilled water was drawn and 870 grams weighed into a l—liter beaker. This beaker was then immersed in a bath of warm tap water where it remained until the temperature of the distilled water reached L10 C. When this temperature was attained, approximately l/1 of the distilled water was poured into a 250 cc glass beaker. The whole dry milk solids were sprinkled on the surface of the remaining 2/? of distilled water which had been removed from the warm water bat}. The mixture was stirred with a narrow rubber scraper to remove lumps and allowed to stand for 15 minutes, stirring occasionally, to dissolve the milk solids completely. During this period the mixture of starches, sugar, and salt was transferred to a l~liter graduated glass beaker. After the milk solids were dissolved the final step in combinm ing ingredients was begun. Only enough of the reconstituted milk solids were added to moisten the dry ingredients. The mixture was stirred vigorously with a narrow rubber scraper to produce a smooth paste. When the slurry was free from lumps the remaininp milk was added and the mixture blended well. Half of the distilled water set aside was then used to rinse the beaker from which the reconstituted solids were poured. The temperature of the mixture at this point was 31.50! 10 C. This temperature was achieved by the previous adjustment of the temperature of the distilled water and the maintenance of a time schedule for combining ingredients. The motor of the stirring mechanism in the viscometer was then turned on, and the slurry was immediately poured into the preheated cooking container in the Corn Industries viscometer. The beaker was rinsed with the remaining distilled water and these rinsinps added to the.mixture in the cooking beaker. The entire mixing and pouring process required a period of four minutes. The condenser cover was then replaced and the motor of the recording device switched on. Cookingrprocedure The Corn Industries viscometer, described in the review of literature, was used for cookine the starch puddings and for recording viscosity measurements. Figure 1 shows a cross section diagram of the viscometer. Bechtel (g5 recommended that standard speed of the stirring mechanism be used during the gelatinization of starch. Standard stirring speed was used throughout this investigation. His . recommendation for a chart speed of one-half inch per minute was also followed. During the preliminary trials in this study it was found necessary to extend the range of the weight arm on the dyna— mometer by adding the specified weight to increase the torque range to 900 gram-centimeters. This additional weisht was required for the entire series. The recording chart used in this study was scaled up to 225 pram-centimeters making it possible to read toroue values directly. The additional weieht used in this study required that the torque readings be multiplied by four to obtain the correct values. A temperature endpoint of 95° C was selected arbitrarily for the pastes in this study. This assured attaining a record of the maximum viscosity of the paste. To achieve this paste temperature, a bath temperature of 1000 3 10 C was necessary. Figure 1. Sectional 5rnwinr of tho C0“n Infiwqtrins viqcom 10. decoruer and fiynamwmctPr (Dynnmflmotor not shown) C8519 from viscametcr to récorfier Cable firum 5, 6 7. Gears of sun and slnnvt fiffprcntial Worm, turnpd by synchronous motor (not shown) W0?m wonr Saving 0149 for holding Contcr sinft Conflinu to pttqch stirrer Confionsor cover Linnifi bnth Ovonflow Drain Dock Starch ankPr Elpctric hontor, thormnstntic”lly contrfillpd Scr309r blgdfls Prooellor Thormvmetor in waste Thérmomoter in bath 61 HHIzmImI] T / ¢/////////////? ///2 _| J HHNH‘NNH 13 m R f \V’fl § V \ \\\N\\\ 7 H h flJLflHHHH N <§>— —-> lllllll . \\ 70 fr r VA, 1V4! r / . H~L L N k N , m ,/ .7 Fl . N 2 r 1 . c I x. G \\ “\111 \ 3_\"; X \ \x \\4\1\\1\\\\\\\\\\\ L (III Soctional Drawing of the Corn Industries Viscometer. Figure 1. 62 Because a bath of distilled water alone evaporates excessively at this temoerature, a glycerol-tyne of permanent antifreeze was combined with an equal amount of water for the bath. The thermostat was adjusted to maintain a constant temperature of the bath. With this tyoe of bath some evaooration still occurs, and it was necessary at intervals during the investigation to add more glycerol-water mixture to the bath. These commonents were combined using soecific gravity determinations to obtain the desired concentration of each. Previous to each cooking oeriod, the stainless steel beaker, stirrer and condenser cover were olaced in nosition. The viscometer was then olueged in to beein heating. A temper- ature of 70° C inside the cooking container and a bath temner— ature of 100° C were reached before the slurry was poured and cookinfi was begun. The temoerature of the pudding was recorded at l-minute intervals throughout the cooking oeriod from the thermometer inserted into the paste throurh a rubber stoooer in the conden- ser cover. As soon as the temperature of the cooked paste reached 950 C, the mechanical stirrer and recording mechanism were turned off and the beaker containing the cooked pudding was removed at once. Four samoles were poured immediately into the snecially designed metal molds adjusted on double glass plates. These molds, which hold anoroximately 7/8 cup each,were filled about level with the too. A period of 1% minutes was required to remove the beaker from the bath and 63 pour the {our samples. The pudding samples were Covered with a circle of absorbent paper in addition to an overall covering of Saran wrap which was placed over the top of the absorbent paper and the edges were pressed firmly to the upper glass plate. The glass plates were numbered with a red wax pencil in the order in which the molds were poured. Testing Equipment The testing equipment used for measurement of cold paste flow was a modification of the method used by Miller (“5). A glass plate 12“ x 12" x &" from which a circle of 3.5" in diameter had been removed was superimposed upon a second glass plate of the same dimensions without the circle removed. These two plates were held together by the means of four rubber bands stretched over each of the four edges. The mold was a circular metal band of stainless steel, 1.5" in diameter and 1:“ deep, with one edge relied. This mold was placed with the rolled edge up within the cut~out area of the upper glass plate. A round metal sample cutter was constructed with a diameter of 1 1/8' and a height of 1". An outer hand of metal 1/8“ wide and 1/16” thick was welded i" from the top edge of the cutter to form a base for the expansion band. A round metal expansion band was constructed with a diameter of 1%" and a height of fi/b”. When mounted in Dlace, the expansion band rested upon the 1/16'I ridge formed by the thickness of the outer band of the sample cutter. 6h Testing Procedure After cooling at room temperature for a period of one hour for the first 2 samples and three hours for the last 2 samples from each batch, the samples were cut and line spread measurements made. The samples were handled consecutively in the order in which they were poured initially. The investi— gator prepared the first sample and released it to flow; during the period of flow for the first sample, the second sample was prepared for release. Each complete testing proce- dure recuired eleven minutes; but with the overlap of time in handlinp the samples, the two tests were completed within a fifteen-minute period. The second sample was ready for flow measurements exactly four minutes after the first sample. To prepare the samples for testing, the paper coverings were removed gently from the top of the mold. Care was taken to prevent any moisture condensed between the two layers from droppine back on the sample. The cutting edge of the sample cutter was lubricated with salad oil and wiped free of excess. This procedure was necessary to prevent material from sticking to the sides of the cutter. The cutter was then pressed firmly down into the center of the molded pudding. The cutter was held firmly in place and the mold was removed. The rubber bands were snipped with scissors, and the upper glass plate gently withdrawn. Excess material was removed from around the edge of the cutter and the lower glass plate was wiped clean. The unmer exnnnelon benfi was then detached and the enmnle was leveled with A etrnieht~edge knife. The glese plate wns epaln vined clean. The cutter wee then lifted and the snmnle was allowed to flow freely for five minutes. A phntnflood lame vlth 3 No. 2 bulb was mountefi on a ringetend l2 inchve ebove the unoer eurfece of the glass plate hnlfilnp thD eamnle. After the 5—minute flav nrrlod, a sheet of Tecnifex Dlezo blue line newer, Which is llght~eensitlve, veg Uleoed flirectly unéer the 91988 Wlnte nnfi the light switched on. The area eround the nemwle was exyosefi to the light and the image of the eemnle was recorded 99 the unexnosed area on the Tecnifex D1823 blue line paner. The exposure time for all enmelee wan erbitrerily eet far 2 minutes. The exnoeed Deoer van immefilntelv “laced in a dark colored box. After teeting was coneluded the imnpee of all four eemoles were develooefl with emmnnln fumes. About 9 cnnfnl of household Ammonip wne nlnned 1n n ateirleee eteel howl mounted on a rlneetend over 9 lighted Bunsen turner. When the ammonia wee eufflciently hentefi to volatilize, the eXfioeed pener was removed from the etornge box nnfl held over the ammonia fumes. The unexnoqefl area an the Deber develfihefi to a eharely defined blue outline in a few Reconfie. The ereee of the nrlnts were measurefi with the compensating molar nlanlmeter. The area of flow for each eemnle wnq fieter~ mined by subtracting the area of the cutter, which V98 the size of the original semnle, from the area of the image taken after 66 the sample was allowed to flow. As the cutters veried slightly in size, all cutters were coded and their indivieunl erees determined. 67 RESULTS AND DISCUSSION‘ Viscosity Tests Maximum viscositx_vslues The range and average torque values at maximum viscosity of the three replications of each variation in Series A and B is shown in Table 3. Table 3. Range and average maximum viscosity values in three replications of Series A and B. EeVEl of waxy starch, in terms Series A, V—l3 starch Series B, Amaizo "boo" of percentage of ‘ starch. total starch Range Average Range Average 0 (control’) h08-h28 blB” h08~428 #18 5 h12-h52 #35 h12-b32 h21 10 #36—n60 hug ugz-h36 435 15 h56-h68 #63 4 u.uus 4h? ’20 h80-h88 “Sh he6-hh8 _. “bl 25 500-512 505 4 8-h92 #71 30 508—56h 5h0 “ #76-496 , .h89 *fisgular cornstarch comprised entire starch content. Proportion g; starch The effect of the percentage of uaxy maize starch W-13 on the viscosity of cornstarch puddings is shown in Figure 2. Analysis of variance for the puddings made with varying levels of waxy maize starch W—13, Table h, showed highly significant differences in the average maximum viscosity values Within the series. 68 Torque 600 500 u. 450 _ 400 __ 35C _. 300 __ 250 _. 200 __ 150 _. 100 .— 50 _. o 1 I I I 1 1 I Time 0 5 10 15 20 25 3o 35 40 in . Minutes Figure 2. E’fect of percentage of waxy maize starch W-lB on the initial viscosity rise, maximum viscositv, and final viscosity of puddings p in series a. 69 Table b. Anelveie of variance of maximum viscosity valuee for Series A. —.--—~-.—-—‘..-—O—-...- ”A —-.-“---P- — Source of verinnee D. F. . N. S. F. Totnl 20 Between treatment avereree 6 5&2? 21.78** Within treatments lb 228 “1"8‘1 prn 1 r15}Lt”5’£‘iif"ié‘62‘i‘"6?"BREEZE T13?" “M" ”M" "‘m Maximum viecoeitv veluee ehowed nroereeqive increases with increase in fiercentnwe cf waxy maize starch W—13. Differences between Avernpee were tested by Studentized rnnpee. At the 5 Der cent level of nrohebility the following eignificnnt differences were noted in Seriee A. 10 oer cent éifferefi from all lower levele. 2§ per cent differed from 15 oer cent one all lower levels. 20 oer cent differed from 10 ner cent and all lower levels. 15 oer cent differed from the control. 10 per cent éifferod from the control. The effect of the nercenteee of waxy maize starch Ameizo "bOO" on the viecoeitv of cornstnrch oufifiinee is shown in Figure 1. Anelyeie of verience for the nudéinge mnfie with varyinp levole of wnxv maize stench Amnizo "DOD”, Table 8, showed highly ainnificnnt differences in Dvorege maximum viecoaity values within the seriee. Torque 600 550 _. 500 _ 45c__ 400_. 350.. 200__ lOO__ 50__ O 70 I l l l l l l J Time 0 in Minutes 5 10 15 20 25 3o 35 40 Figure 3. Eifect of percentnfie of waxy maize starch Arnizo "4“?" on the initial viscositv rise, maximum viscositr, and final visoositv of puddinws in series B. 71 Table 5. Analysis of variance of maximum viscosity values for Series B. Source of variance D. F. . M. S. F. Total 20 Between treatment averages 6 2013 16.395*“ Within treatments in 12b fi*51gnificant at lgfilevel of probability Maximum viscosity values showed progressive increases With increase in oercentape of waxy maize starch Amaizo "boo" with the exception of the 20 oer cent level which fell slightly below the 15 oer cent level. Differences between averages were tested by Studentized ranges. At the 5 per cent level of probability the following significant differences were noted in Series B. 10 oer cent differed from 20 oer cent and all lower levels. 25 per cent differed from 20 oer cent and all lower levels. 20 Der cent differed from the control. 15 per cent differed from 5 oer cent and the control. Cooking time Temoeratures of initial viscosity rise ranged from 77° to 780 C in Series A and from 77.250 to 780 C in Series B. No progressive changes in temperatures of initial viscosity rise occurred with an increase in the level of waxy maize starch in either Series A or B. Five to seven minutes were required in both Series A and B to reach the point of initial viscosity rise. The time required for initial rise in viscosity was not related to the level of waxy maize starch in either Series A or B. In this study all puddings were Cooked to an endooint of 95° C. The length of time required to reach this temDereture varied from 2? to 18 minutes in Series A and from 22 to 30 minutes in Series B. The time required to reach maximum viscosity varied from 18 to 2b minutes in both Series A and B. The temp— erature at maximum viscosity ransed from 92° to 9&0 C in both Series A and B. Tables 6 and 7 show data on time and tempera- tures reouired to reach maximum viscosity and temneratures of initial viscosity droo in Series A and B. These differences in cooking time did not aonear to be associated directly with the level of waxy maize starch but appeared to be more related to a variation in the rate of heatine. No logical exclanation could be given for this behavior as the temoerature of the cooking bath was apparently well controlled. The differences in rate of heating were not sufficient to overcome the influence on maximum viscosity exerted by the level of waxy maize starch when the three rep- lications were averaged. Table 6. of Series A. WIT-f-..‘ “Era x i me centape viscosity to reach maxi— Of value w—12 (torque) 0 L28 18 a una 18 o hi6 19 S b62 19 5 LLo 19 5 hlZ 20 10 L62 19 10 D60 19 10 A16 21 15 bee 18 1:) 1’ 6L; 1 9 15 E68 21 20 uau 20 2O b80 19 20 EBB 22 25 512 22 25 500 22 25 Got 22 3o 5&8 19 so 56b 21 10 508 211 Cooking times and tenderatures to reach temoeratures of ‘3 kid maximum viscosity, initial viscosity drco in all Cochin? time mum viscosity (minutes) Temnerature (0 Centirrade) - vfi‘“ — replications Temperature uoon reach— of initial ins maximum viscosity viscosity droo (0 Centigrade) ~»--vq—---—-. ->---m -*H~-~~. M~-—.— 9b.0 9h.0 9h.25 911.0 91“.?) QB.25 9h.5 9h.o 9b.25 9b.0 95.25 9b.25 9b.25 911.5 93.0 93.0' 93.25 93.5 91‘107‘5 9b.o 91.25 . ------ -—. - —’v~O-——_ Table 6. Cook no times and tenderaturee to reach maximum viacoeity, ano temoeraturea of initial viacosity firoo in all replications of Series A. FEET." "V a? 1' r5}? 635151711: "f GE; " ‘ “17535575555117ng "1' 631' pe ra t a. re centape viscoeity to reach mexi— uoon reach— of initial of value mum viscosity inp maximum viscosity W—l? (torque) (minutes) viscosity drop (0 Centirrade) (0 Centigrade) 0 L28 18 92.5 9b.o O #08 13 99.9 9b.0 o u16 19 91.0 9b.25 s bsz 19 91.25 9b.o 5 Lbo 19 91.25 9L.5 5 “12 20 91.5 Qu.25 10 £62 19 93.25 9N.5 19 E60 19 91.5 9b.O 1o U16 21 9n.o 9b.25 16 b66 18 92.5 9h.0 15 uéu 19 91.25 9L.25 16 U68 21 91.25 9b.25 20 uau 20 97.€ 9b.25 20 U80 19 93.25 05.5 20 L88 2? 92.0 93.0 25 612 22 92.5 93.0 25 900 22 92.5 93.29 25 50b 22 92.26 93.5 30 5&8 19 99.0 9“.75 10 56b 21 93.25 9h.0 30 508 2‘l 92.25 99.25 ‘J \.A: ' ---‘——~-—. - -- v~-.-v Table 6, mw~-~ h. Per~ “SKETbum 063215} time "femoerature Tgmperature centape viscosity to reach maxi- upon reach— of initial of value mum viscosity ing maximum viscosity W—l‘ (torque) (minutes) viscosity dgoo (0 Centisrade) ( Centigrade) 0 L28 18 92.5 99.0 0 908 19 99.0 99.0 0 L16 19 91.o 99.25 5 L52 19 99.25 9L.o 5 LLo 19 99.25 9L.5 5 912 20 99.5 99.25 10 L52 19 93.25 99.5 10 960 19 91.5 99.0 in L16 21 99.0 99.25 16 L66 18 92.5 9L.0 5 969 19 9? 2S 9L.25 15 968 21 99.25 99.25 2o LRL 20 91.5 9L.25 2o L80 19 93.2! 99.5 20 L88 2? 92.0 93.0 25 512 22 92.5 93.0' 25 590 22 92.5 93.25 25 Gob 22 92.25 93.5 30 598 19 99.0 99.75 10 56b 21 93.25 99.0 90 508 29 92.25 97.25 —. «'vm-vr ‘J ‘4.) Cooking times and temoeratures to reach maximum viscosity, and temoeratures of initial viscosity flroo in all replications of Series A. -‘O‘N-UV‘~- .--- C -- ~-~-—'-ow-- O -- _.--.-._. -. --~'~T Table 7. Cooking times and tsmoeratures to reach maximum viscosity, and temperatures of initial viscosity droo in all replications of Series B. .0..- F:?— Maximum Cooking time Temperature 'Temperature centage viscosity to reach maxi- uoon reach- of initial or value mum viscosity ing maximum viscosity Amaizo .(torque) . (minutes) viscosity oroo (#00" ‘ (0 Centigrade) (0 Centigrade) O h28 18 92.5 99.0 0 908 18 99.0 9L.O O L‘16 , 19 93.9 92.25 5 “32 19 91.5 9L.o 5 L120 19 99.25 99.0 5 #12 21 93.0- 99.25 10 h36 20 99.75 99.0 10 L36 19 99.25 9L.25 10 L32 22 93.75 9L,o 15 LLB 20 93.5 99.5 15 LLB 23 93.5 99.0 15 LLL 22 99.75 99.0 20 LLO 21 93.0 93.5 20 L36 22 93.75 99.5 20 LLQ 22 91.0 99.25 25 h72 20 99.0 95.0 25 992 20 93.75 99.5 25 hh8 29 92.0 92.5 30 L96 21 99.5 9L.o 30 L96 21 91.5 9L,25 3O 976 21 99.25 99.25 75 Because of these differences in cooking time an attempt was made to find a relationshio between cooking time to reach maximum viscosity and the maximum viscosity value of the indi- vidual reolications. These factors did not show any straight correlation when the entire series was considered because the level of waxy maize starch anoeared to have much greater influence on maximum viscosity than cooking time. A partial correlation, holding the level of starch constant, however, showed a negative correlation between maximum viscosity and cooking time which was highly significant at the 1 per cent level of orobability in Series 8. However, a significant correlation could not be shown between cooking time and maxi— mum viscosity in Series A. These results indicate that the longer cooking time to reach maximum viscosity may have reduced the maximum viscosity value in Series B. These findings are not in agreement with Harris and Jesperson (92) who reported that swelling oower was not_ significantly affected by rate of heating. Bechtel (8), however, found that an increase in rate of heating increased the maximum viscosity to a marked degree and that the difference in viscosity was evident throughout the remainder of the cook— ing period. Kind of starch The height of the casting curves and the maximum viscosity values were greater in puddings made with all levels of both 76 waxy maize starches in combination with ordinary corn starch than in those nuddinps made with only ordinary corn starch as the thickening agent. Chemists for American Maize—Products Comnany (7) have found that the raw waxy maize starch (amioca) Fives much higher viscosity values than does ordinary corn starch at similar concentration in a caste. A straight line was fitted to the maximum viscosity values in both Series A and P. The slopes of the lines were tested statistically and results showed the two starches to be differ- ent at he 1 per cent level of orohability. Figure b shove granhically the average viscoaity values at each level of waxy maize starch and the regression lines fitted to these values of both Series A and B. The rate of increase in maximum viscosity of ouddincs with each increnent of waxy mai7e starch W—l? was sisnificantly greater than the rate of increase in viscosity observed in those made with ccrresoondins levels of waxy maize starch Amaizo "L00". Figure G shows a series of graOhG cnmoar~ ins viscosity curves in Series A an"3 F at each level of waxy maize starch. With each increment in the level of waxy maize starch the difference between the heiehts of the viscosity curves of Series A and B became preater. Tiese findings are in agreement with the statements of the manufacturers of these two modified waxy maize starches. They assert that with increased treatment the waxy maize starch ex— hibits characteristics anoroachin? more nearly those of the ordinary common corn starch (Ll). Waxy maize starch Araizo "hon” had undergone more chemical treatment than had waxy maize starch W-17 Torou 55 525 500 9 0 E75 :- ‘J\ DZ 40 % waxy starch C) 5 O O 77 * W Senes A / (w-13)\/ / ’ / / l/ / J ’ / / / 1 / . / o / u h / '0 / I / / / / / / f. "‘ y- , ._ ,o\ Sci-RS B u .. ,, ..v' ‘°'"--n5 (Amnlzo 4+00) / H / / I d >- / , / . (U... 1 A l 1 L O 5 10 15 20 25 30 Figure b. Effect of parcentagp of wnxy mqu? starch on average maximum v1901slty values for sorips A and B as shown by lines of regression. Y4 3: 24 78 73r7u¢ 75’7ut Co ' 72r7v< o.- 400{ ‘00[ 5-00 P 5.00 b 500 _ v. (”-13 w-I3 / 400 >- 00 >- ~\ . u I u ‘/ nmn/zo 'qoo 510‘} Amalia ’qaa 300 b 300- 300}. i 1 1°. ,_ 200,, 2001- loo P /00 _ /00 ,_ n 1 . 7.-..-.J o L__..L_-_- ..H___1_ V _, _ . ,, . ,1 , ,. .1 7, _ 1 . _._. O 1 L J 1 \ _l 35‘ ¢o 0 1:.“ Jo 3f 40 a s; (a ,5 2,0 15— 3° 35— #o 77:8 m 77’": M Ware M Inc/725 fll/7?5 a (Z Level 7’ [CV‘L ”pawns A” /, 1.: ch 15“] ‘00 .. ‘00, LOOr €00». /Lu_(3 5‘00?- / flmmzo 9400‘. ”400‘ ‘~ 400" 400 t- 300 r- 3005.- l i 2 I 200' l ; ¥ ) No i.- l /00 , I ‘ l L , I 1 _A o 1 J L _ 1 I 4 L A_J d A I A L L A 1 1 3o Jr 40 o 5' /o I3 20 2.5 30 3.5” 54° 0 5‘ lo (5‘ z a z 5‘ 3 c 35‘ «a fun: ,4 - 7mm m , .20 0/0 level 4/00/é5 ‘45 % Level Mun/7e: 3° A Level Halo/{e 5“, E/-/:T:"c7” 0F AVA/.0 0F W/{X/ /‘7/;’/zz:‘ STfl/rcfl' H7 CoKfit‘SPOMa/A/CJ- [En/[AS 0F 72>an SfH/f‘C/f COMCE/VTKAT/oxv o/v V/SCOS/T/ CURVES 0F Poop/Mes 79 Cold Paste Flow Progortion of wexv rmize sterch wavy-“u. The ssmcles of succing used for flow tests were held at room temhersture. The first tests were made on 2 samples efter s 1—honr holding period, and the tests on the remaining ‘1 2 ssmnles were made efter e 3—hour holding Deriod. The measure— ment of flow for samples, expresseé in squsre centimeters, were determined by messurirg the prints of the ssmrles with the comnensstihe holer olsnimeter. The flow readings for the two semnles tested at eech holéing neriod were averaged. The range of the sverege velues'for the three replicsticns for each hold~ ing pericc are shown in Table 8 for both Series A and B. Table 8. Range in flow velues of three renlicetions of each series. Tyne o?“hhld~ Wexy starch level in terfis cfrnercentege Series wsxy ing of total starch. sterch Elfi2.. O 5 10 15 2O 25 10‘__ A W—l? 1 hr. ql.6~ 31.0— 31.b- 11.6— 12.#- 36.1- 33.1— ?3.5 33.7 3b.8 32.6 37.5 36.9 38.1 B Amsizo "L100" 1 hr. 11.6— 2909“ 11.1.... 3208'” ?500‘ 2906'. 3202'- ??.S 3b.? 33.9 13.9 bO.2 b0.0 35.1 A w—11 1 hrs. 22.6- 22.2- 21.8- 20.7- 21.2- 23.2~ 23.3- 22.9 2L.O 2B.l 22.1 2h.5 25.1 2h.5 B Ameizo "#00" ? hrs. 22.6- 21.2- 21.6— 21.7- 2?.h— 21.8- 21.6— 22.9_ 2?.“ 2b.l 2?.h 2U.0 26.0 22.2__ The svereee‘resdinss for the three reolicstions for each holding nerind for both Series A and B are shown in Table 9. Tnhle 9. Summary of average flow readings of three reolicetions for each series at 1—hour and 3-hour holding periods. Tyne of Hold— Waxy starch level in terms of nercentage Series waxy in; of total starch. starch time 0 5 10 15 20 25 39 A w_13 1 hr. 32.1 e2.h 12.8 31.9 3b.? 36.? 3b.9 B Amaizo "boo" 1 hr. 32.? 12.0 32.2 33.3 38.0 33.u 33.8 A w-11 B Amai7o . "boo" 3 hrs. 22.7 22.6 21.0 22.1 23.8 23.3 22.3 co.) hrs. 22.7 23.3 21.1 21.3 2b.o 2b.u 23.8 Analysis of variance of flow readings showed no significant difference between the mean squares of between reolications within treatments and within replications within treatments. Therefore, a common error term was found by combining the sums of souares for these two sources of variance. Analysis of variance of flow readings in Series A, Table 10 showed highly significant differences between levels of waxy maize starch W—ll within each holdins neriod. Table 10. Analysis of variance of flow readings in Series A. After l-hour After 3—hour Source of variance D. F. holding holding period period _ N. S. F. M. S. F. Total 51 Treatments 6 18.h b.00** 5.96 4.89** Between reolications within treatments lb 6.2 1.82 l.bO 1.15 Within reolicetions within treatments 21 3.h 1.10 Common error' 15 “.6 1.22 **Significant at 1% level of probability 81 Analysis of variance of flow readings in Series B, Table 11, showed highly significant differences between levels of waxy maize starch Amaizo "boo" for the 1-hour holding period but no significant differences for the-Q—hour holding period. Table 11. Analysis of variance of flow readines in Series B. After l-hour After 3—hour Source of variance D. F. holding holding period period it. 8. F‘. N. S. 1“. Total , bl Treatments 6 25.58 3.h7** 1.86 .93 Between replications within treatments In 1b.29 1.9” | 2.83 l.h2 Within replications within treatments 21 2.77 1.2m Common error 35 7.38 2.0 *“Signiricant at lflevel of probability These differences in flow readings did not apnear to be related to the level of waxy maize starch. There was no progression in flow values within each series with increase in the level of waxy maize starch. At the concentrations used in this study, the waxy maize starches had no apoarent effect on gel-forming tendencies of cooled cornstarch nuddines. These findings are not in agreement with Schopmeyer (53) who stated that the addition of amioca greatly retarded the selling tendency of common starch, even when common corn starch consti— tuted the major pronortion of the total starch. 82 Cookingftime Since the significant differences in flow measurements between treatments could not be attributed to the level of waxy maize starch, this variable was disregarded and an attempt made to find another cause for the differences in flow measure- ments within each series. As cooking time appeared to be a factor influencing maximum viscosity values, it appeared that this factor might also losically affect end viscosity and perhaps in some way affect flow of the cooled sample. Analyses were made to correlate cooking time with end viscosity as well as cooking time and flow measurements of the cut samples. A partial correlation carried out between cooking time and end viscosity, holding the level of waxy starch constant, showed a negative correlation significant at the l per cent level of probability in both Series A and B. These results indicate that as cooking time was increased, the and viscosity values were correspondingly decreased when the influence of the level of waxy maize starch was eliminated. It is generally agreed that continued cooking of a starch paste after maximum viscosity is reached will result in paste breakdown. After maximum viscosity has been reached the extent of decrease in viscosity of a given paste depends on the length of the cooking period. There was no straight correlation between end viscosity and the flow of cut samples in either series. A partial correlation carried out between end viscosity and the flow of 83 cut samples of puddinss in Series A, with the level of waxy maize starch held constant, showed a negative correlation at the l-hour holding period, significant at the l per cent level of probability. The partial correlation between and viscosity and the flow of the cut samples at the 3—hour holding period was not significant in Series A. A similar correlation was found to be hishly significant at the l per cent level of probability for both l—hour and 3—hour holding periods in Series B. In both series the correlation between end viscosity and the flow of cut samples was greater at the l-hour holding period than at the 1—hour holding period. These results in- dicate that as and viscosity decreased, the flow of the out samples increased when adjustments were made to eliminate the influence of the level of waxy maize starch. The correlation appeared to diminish gradually with increased holding time. As the level of waxy maize starch did not appear to influ— ence flow of the cut samples a straight correlation was carried . out between flow measurements and cooking time. A positive correlation between cooking time and flow of the l—hour samples in Series A was found to be hiahly sisnificant at the 1 per cent level of probability. A similar correlation was found between cooking time and flow of the cut samples at the 3—hour holding period sisnificant at the 5 per cent level of probability. The correlation between cookinr time and flow of cut samples of onddinps in Series B for both l—hour and 3—hour holding periods was significant at the l per cent level of probability. 84 The correlation was greater, however, after the shorter holding period in both series. Figure 6 hows values for flow plotted against cooking time for both l-hour and 3~hour holding periods in both series. These results indicate that as cooking time was increased and end viscosity decreased, the flow of the cooled sample of pudding was increased. The fact that the correlation is less pronounced for the samples held ? hours would indicate that the end viscosity value had less influence on consistency of these samples than on samples held only 1 hour. Many investigators have found no relationship between hot paste viscosity and cold paste characteristics. In this study it appeared that within the limits of a 3—hour holding period complete gelation had not yet occurred. Therefore, it seemed that the effect of hot paste viscosity, as related to extent of paste breakdown, had not entirely disappeared. Holding time An easily discernable difference was observed between the -flow of the cut samples of puddins after the 1—hour holding period and after the 1-hour holding period. Increase in holding time decreased the flow of the cut samples in both Series A and B. Z/A/t' {, _. ‘5); ‘W'L‘kiu < 97- 67m \ [I 31 . 3!». Jop 21.. 13,, o l O GdOIfi/fi 7),)76 //V 07/279723 27 - 2L - A/A/E 2? ” vW'A’Ffip 24 (5f-C'm. '3 13 P u. _ llr 20" 0 G CZ’C‘If/Iv7 ”01C /'V ”7/4Vf65 “‘ 24 F‘r/es A) W-—/3 D! 1 1 I 1 L L 1 23' 2o 27 z: z? 30 3/ 32 33 / flour- Jflflm/Ugcb 1 .1 ,l -_L VAL .. 25' 2‘ 27 27 2.7 30 3/ 32. 33 3 bCU" 5407/2165 /?GUNE A. l k... (.V._-.|‘_.r .. U 85 3 V13} 34 h (5904/ ”J, ’/-m<; / /A/ ”7/ '2‘ 4"65 Se r /65 0) fl/fifi/Lo 2400‘ \ 3 hour 5'407/016’5 577ch Of: Coo/(7'77 77/»6 0N Z/nc S/Ofeflc/ /<)c’,4a//,775 FOR £075 //?/‘- ’7/70/ 3/7n SHIN/018$ /n 5€rv€5 fi/‘Mc/ 8 ‘ . 1 , ‘ 3 d r ' i I _ J73 l 4 T l - 37 y 4' ; 30' .. ‘ 1 l .. 1 / Vt” wit, . ‘ ‘, I 3—,» 73 ‘c‘ ‘v'q : 17, i ' U .— _ / , P . (5 ~_ 3m \ f l ' Z i . .U ' A .. r- ' _ ML . 4. \ ' y I O .. 1 g. 0 fl: 5 a 3"." __ .4 ‘ z A I] f— o l \ ‘ L“ A A L 4 L l J L L I l 1 1 ‘ L _ 35’ .37 $40 ‘- 4;“ 49* 2.4 27 45K 47 J‘n .j/ .5). If I“ 3 " M J/ JX j / 9m {pf ‘ I \r’ \‘r 7:.«4 ( 1 a I , ,/ 7‘ , .;/mnng / wovr éfinvw£> 28 1 F 27 r- -{ J 24» _. q I i .. .45 _ .4 A /N b D I .— O o 5 67-6») ° . ; 5+ ‘15 — , . CO ' " ”LL r- o “ V o a ' .4 LI - A I £0 .. - L__L_____L J l l ”-1— 1 l _L 1 L l I 1 1 I L 1 - MLW__ a _. H .. _ 5 ‘ é 37 3? 3 $0 . A ( o 5/ J; 5.5 J S J r7 3: 37 4o 0 294 z; 16 .4/ X z, 3 (,4 86 Kind 2: starch No significant diffawonceg could be found between the flow ménsurnmpnts of out snmpleg of Puddings in Series A and B at corresnonding levels of waxy maize starch concentra— tions and at corresponding holfiinp neriods. If any differences were oresent, thpy wnrp comnlptoly concealed by the more obvi- ous Pffacts ornduced by vnrintinnn in conking time. 87 SUMMARY The maximum viscosities of cudcinps mace with waxy maize starch W~l1 st levels of 0 5 10, 15, 20 25 and 30 Der cent of the total starch concentration shored hiehly siznificent differences. stimum viscosity values showefi ercqressive in- creases with increase in the level of waxy maize starch W-l}. Puddinss mace with wsxy maize stnrch Amsizfi "“00" at similar levels of the tctnl starch concentration showed h :hly signifi— cant differences in maximum viscosity values. A Drogressive increase in maximum visccsities with increase in the level of wsxy maize starch Amnizc “b0?" was observed with the excection cf the 20 ner cent level which fell slightly below the viscosity of the 15 oer cent level of this wsxy starch. Initinl viscosity rise cccurefi within e very narrow range of time and temcereture for nucdings in botk.Series A and B, enfi slicht verieticns 316 not ecceer to be related to the level or kind of waxy maize starch. ‘ The cooking time required to reech maximum viscosity as well as to reach the endnnint of 950 C showed a wide variation for cudfiings in both Series A and B. These differences in cooking time did not ancesr to be related to the level or kind of waxy maize starch but to an unexulninsble vsriaticn in the rate of heating. A highly signiflcent correlation was found in Series B between time required to reach maximum viscosity and the 88 maximum viscosity value when the level of waxy maize starch Amaizo "LOO" was held constant. As cookine time necessary to reach maximum viscosity increased, the maxinum viscosity value was corresnondingly decreased.. A similar correlation was not fourd to be significant in Series A. Puddings made with all levels of hoth waxy maize starches pave higher viscosity readinss than the control in which ordinary corn starch constituted the entire starch concentra- tion. Maximum viscosity values of puddings at all levels of waxy maize starch W—lB were hisher than maximum viscosity values of puddinrs at correspondins levels of Amsizo "DOG". The rate of increase in maximum viscosities of puddings with each increment in the level of waxy maize starch W-l1, Series A,.was sisnificnntiv creater than that mafia with correspond- ing increases in the level of waxy maize starch Amaizo “bOO", Series 8. Flow readinvs for cut samples of puddings in Series A showed significant differences between treatrents for both the l-hour and l-hour holding periods. Significant differ» ences were found between flow readings of puddings after the l—hour holding neriod in Series B but differences after th 3—hour holding period were not significant. The differences in flow readines for puddings in both Series A and 8 did not appear to be related to the level of waxy maize starch. d 9 There was no progression in values for flow measurements within each series with increase in the level of waxy maize starch. The correlation between time required for the paste to reach 950 C and the and viscosity of puddings, with the level of waxy starch held constant, was found to he hiohly sisnifi— cent in both series. As the cooking time required for the paste to reach the temperature endooint of 950 C increased, the viscosity decreased. The correlation between end viscositv and the flow of cut samples of puddings, with the level of waxy maize starch held constant, was highly significant after the l-hour holding reriod in both Series A and B and after the 1—hour holding nericd in Series B. Flow of cut samnles of puddinp increased with a decrease in the end viscositv. In both Series A and B this correlation was less with an increase in holding time. A significant correlation was found hetween cooking time and the flow readings for puddings after hath lehour and S—hour holdina periods in both Series A and B. As cookinr time was increased, the flow of the cooled puddinr sample was also in~ creased. This correlation was areater at the l—hour holding period than at the 1~hour holding nnriod in both Series A and B. A pronounced difference was observed in th flow of cut samples of puddings in both Series A and B after the l—honr holdinw period and after the 1—hour hpldinr period. Increase holdine time decreased the flow of the cooled sam'les of iv i.) Dudéing. 90 No signifiC9nt differences werp found in flow measurémonts of nufidinrn mnfc with thP two waxy 59129 starchog at any of th corrnqccnfilng levels of the total starch concentration. 91 C ONCLVSI ONS Waxy maize starch used in combinetion with commdn cornstarch had significant effects on the hot neste viscosities of simole cornstarch nuddinps. Each increment in the level of waxy mnize stsrch produced puddings with urogressively higher maximum viscosity vslnes and correspondingly higher nesting curves than did the control in which ordinary corn- sterch made 110 the totel sterch concentration. The waxy maize starch Which had undergone less chemical treatment increesed hot paste viscosities to a greeter degree than did the waxy meize sterch which hsd received more treatment. From this limited study it snoeers that concentrations of wsxy maize sterch no to 30 eer cent of the total starch do not alter snorecishly the gellinp tendencies of ordinsry corn— starch in simnle nuddings. In soite of the wide varietion in hot nsete viscosities attributeble to the kind end the level of waxy maize starch, no differences were observed between the flow meesurements of the cut nudding semnles which could be releted to these same factors. Differences in cookinr time as related to the rete of heating and the extent of osste breakdown ennesred to effect the flow of the cooled ssmoles of pudding. An increase in . cooking time produced en increase in flov. This relationshin seemed to diminish with the length of holding time. It 92 aooeared from this study that cold caste flow cannot be Dre— dicted directly from hot paste viscosi y, although it did seem to he somewhat related to the extent of caste breakdown and cooking time. As this study was confined to the nroduction of ouddinps in small nuantity, no definite conclusions can be reached concerning the behavior of nuddings made from similar formulas cooked in large quantity. However, When waxy maize starches are used as part of the total starch concentration, it seems feasible to exoect an increase in hot oaste viscosity above that obtained with ordinary cornstarch. 11. 12. \L) -. .d IITERATVPE CITED I. Strliesaizoon stermfit. Ind. Ehnr. Grins. 18: 1 l Alsherr, C.L. “DC 399k. 0- }:eai of when? and ‘39 he relatinisetion by h. Cereal Chem. 1: American Maize—Products Company. AmioCs A new domestic starch. New York. \ Anker, C.A. and Geddes, W.F. Gelatinization studies unon wheat and other starches with the amylosraoh. Cereal Chem. 21: 735~°62. 193“. Badenhuizen, N.P. Observations on the distribution of the linear fraction in starch granules. Cereal Chem. 32: 287—295. 1955. Bates, ?.L., French, D., and Rundle, R.E. Amylase and amvlooectin content of starches. J. Am. Chem. Soc. 65 (oart l)f lL2—lh8. 19U3. Bechtel, W.G. Instruction manual for Corn Industries viscometer Progress Report No. 12 of Corn Industries fellowshio. Sent. 1954 — March 19b7. Bechtel, W.G. A study of some oaste characteristics of starches with the Corn Industries viscometer. Cereal Shem. 2h: 200~21U. 19h7. Bechtel, W.G. Measurement of properties of cornstarch gels. J. of Colloid Sci. 5: 260~270. 1950. Bechtel, W.G. and Fischer, E.K. The measurement of starch caste viscosity. J. of Colloid Sci. h: 265~282. 191:9. Beckford, 0.0. and Sandstedt, R.N.. Starch gelatinization studies. I. Simnlified equinment for the study of starch gelatinization by means of light transmission. Cereal Chem. 2L: 250~258. 19b7. Bisno, L. Thickeninss used in oie fillings. Part I. Bakers Digest 25 (no. 1): 26~29. 1951. Bisno, L. Thickenings used in nie fillings. Part II. Bakers Digest 25 (no. 2): 29—1“. 1951. 1h. 15. 16. 17. 18. 19. 20. 21. 22. 23. 2b. 25. 26. 9b Brautlecht, C.A. Starch. Its sources, oroduction, and uses. New York. Reinhold Pub. Corp. 1953. Brimhall, B. and Hixon, R.N. The rigidity of starch castes. Ind. Eng. Chem. (anal. ed.) 11: Brimhall, B. and Hixon, R.M. Interoretation of viscosity measurements on starch pastes. Cereal Chem. 19: uzs-uuo. 19b2, Brink, R.A. Dynamics of the waxy gene in maize. II. Nature of the waxy starch. J. Biochem. 22: Caesar, G.V. and Moore, E.E. Consistency chances in starch castes. Ind. Ens. Che m. 27: 11U47—1b91. 11975. Collins, C.N. A new type of Indian corn from China. ".3. Deot. of Apr. Bur. Plant Ind. Bul. 161. 1909. Collins, G.N. and Kempton, J.H. Inheritance of waxy endosperm in hybrids with sweet corn. V.S.Dept. of Agr. Bur. Plant Ind. Circ. 120: 21~27. 1913. Cox, R.E. and Higby, R.H. A better way to determine the Jelling power of pectins. Food Ind. 16: bulwbhz; 505—507. 19bb. Crossland, L.B. and Favor, H.H. Starch gelatinization studies. II. A method for showing the SLPH9S in swelling of starch during heating in the amylograoh. Cereal Chem. 25: 213-219. 19U8. Davis, J.G., Anderson, J.H., and Hanson, H.L. Starch cereal thickening agents for canned food products. Food Tech. 9: 13-17. 1955. Felton, G.E. and Schoomeyer, H.P. Thickebodied starch and method of making. United States Patent 751,947. Sept. 7, 191110 Ferree, M.J. The effect of the prooortion of sucrose on the viscosity of cornstarch pastes made with different liquid mediums. Unouh. M.S. Thesis. East Lansing, Michigan, Michigan State College Library; 1955. ' Frankel, J.A. A practical treatise on the manufacture of starch, glucose, starch-sugar and dextrine. D. 1-10. Philadelphia. Henry Cary Baird Co. 1881. 28. 29. 30. 11. 7.2, 18. French, D. Physical nronerties of starch. In Kerr, R.W., Editor. Chemistry and industry of starch. 2nd Ed. New York. Academic Press Inc. 1950. Grawemeyer, E.A. and Pfund, M.C. Line-spread as an objective test for consistency. Food Res. 8: Eains, A.J. The effect of sucrose, citric acid, and temoerature on viscosity and gel strength of cornstarch pastes. Unoub. M.S. Thesis. East Lansing, Michigan, Michigan State College Library. 1953. Hanson, H.L., Camobell, A., and Lineweaver, H. Preparation of stable frozen sauces and gravies. Food Tech. 5: h32-L39. 1951. Hanson, F.L., Nishita, K.D., and Lineweever, H. Preparation of stable frozen ouddinps. Food Tech. 7: h62-h65. 1951. Harris, R.H. and Benasik, O. A further inquiry regarding effect of electrolytes on the swelling of cereal starches. Food Res. 1?: 70~81. 19h8. Harris, R.H. and Jesoerson, E. A study of the effect of various factors on the swelling of certain cereal starches. J. of Colloid Sci. 1: “79-h93. 19h6. Harrie, R.H. and Jesnerson, E. Factors affectina some ohysiochemicel nroperties of starch. Food Res. 11: .21'4“228o lgi'éo Hixon, R.V. and Soracue, G.T. Waxy starch of maize and other cereals. Ind. Eng. Chem. 3h: 959—962. 19b2. Hjermstad, E.T. A recordinr eel tester. Cereal Chem. 32: ZOO—20?. 1955. Hughes, 0., Green, E., and Camohell, L. The effect of various temperatures and time periods on the percentage of gelatination of wheat and corn starch and of cereals containing those starches. Cereal Chem. 15: 795-800. 1938. Jordan, 8., Wegner, 5.8. and Hollander, E.A. The effect of homogenized milk noon the viscosity of cornstarch oudding. Food Res. 18: 639—658. 1953. 19. hfl b1. L2, [11, LR. \O O\ Katz, J.R., Desar, i.C., and Seiberlich, J. The change in the viscosity of starch pastes exolained by the changes in microscopical structure of the starch granules of the caste. The Faraday Soc. Trans. ?“ (cart 2): 1258~1268. 1938. ‘ Kerr, P.W., Editor. Chemistry and industry of starch. 2nd Ed. New York. Academic Press Inc. 1950. Kesler, C.C. and Bechtel, W.G. Recording viscometer for starches. Anal. Chem. 19 (no. 1): 16-21. 19h7. Knowles, D. and Farris, R.H. Observations on the behavior of starch eels from different classes and varieties of wheat. F‘oorq Res. 8: “09-b19. 19L3. Lloyd, R.L., American Vaize—Products Comoany, New York. Information on modified waxy maize starches. (Private communication.)\ 1956. (- Manaels, C.E. and Pailev, C.P. Relative viscosities of wheat starches. Ini. Ens. Chem. 25: LS6—L40. .1973. Miller, G.A. The effect of batch size noon the flow of chocolate nie filling. Unoub. M.S. Thesis. East Lansing, Michiran, Michigan State Collere Library. 1955- Moryan, W.L. Pastinr and identification of starches. Ind. Eng. Chem. (anal. ed.) 12; 311—317. 19L“. Morse, L.M., Davis, D.S., and Jack, E.L. Use and Drooerties of nonfat dry milk solids in food preparation. Part 1. Food Res. 15: 200—215. 1080. Noznick, P.P., Merrett, P.P., and Geddes, W.F. Staling studies on bread containing vaxv maize starch. Cereal Chem. 23' 297—105. 19h6. I ( Hadley, J.A. Starch and its derivatives. New York. D. Van Nostrand Co. Inc. 19km. Schoch, T.J. PhysiCA1 asnects of starch behavior. Cereal Chem. 13: 121-128. 19L1. Schoch, T.J. A decade of starch research. B"K°“R Dinfit 2 (no. 1): 17s2o, 2?. 19h7. 52. 530 5b. 55. 56. 57. 53. S9. 60. 61. 97 Schoch, T.J. and Elder, A.L. Starches in the food industry. Am. Chem. Soc. Conference. Los Angeles. 1953. Schoomeyer, H.H. Amioca — the starch from waxy corn. ~Food Ind. 17: lb7o-lb78. 19b5. SchOpmeyer, H.H., Felton, G.E., and Ford, C.L. Waxy cornstarch as a replacement for tapioca. Ind. Eng. Chem. 15: 1168~1172. 19b1, SJostrom, O.A. Microecooy of starches and their modi~ fications. Ind. Eng. Chem. 28: 63—7b. 1936. Sorague, G.F., Brimhall, B., and Hixon, R.M. Some effects of the waxy gene in corn on prooerties of the endo— sperm starch. Am. Soc. Agron. J. 35: 817-822. 19h3. Tanner, L.P. and Englis, D.T. A study of starch from different varieties and tynes of corn. Food Res. 5: 563—581. 19h0. ' Tremoel, L. The use of starch in oie fillings. Bakers Digest 20 (no. 6): 2?~25, 28. 19h6. Whistler, R.L. and Smart, C.L. Polysaccharide chemistry. New York. Academic Press Inc. 0. 229—275. 1957. Woodruff, S. and MacMasters, M.h. Gelatinization and retrogradation changes in corn and wheat starches shown by ohotomicrograohs. 111. Apr. Exot. Ste. Bul. has. 1918. Woodruff, S. and Nicoli, L. Starch gels. Cereal Chem. 8: 2u7~251. 1991. EFFECT OF COMBINED WAXY MAIZE AND RFG"LAR CORN STAPCP ON VISCOSITY AND COLD PAST? FLOW OF IVPLE P"DDISGS By Betty Jo Sullivan AN ABSTRACT Submitted to the Desn of the College of Home Economics of Michigan State University of Agriculture and Applied Science in partial fulfillment of the requirements for the degree of MAST?R OF SCIENCE Department of Institution Administration 1956 Approved by ,,2.¢adbj/7 ABSTRACT The primary objective of this study was to determine the effect of combined waxy maize and regular corn starch on the viscosity and cold paste flow of simple puddings. The formula consisted of constant prOportions of sugar, distilled water, Whole dry milk solids, and salt. The total starch concentration of 6.7 per cent, based on the weight of the liquid, remained at a constant level throughout the study. The prooortions of waxy maize starch and ordinary corn starch making up the total starch were varied in progressive steps. Two waxy maize starches representing different levels of modification were used separately in combination with regular corn starch. The waxy maize starches constituted O, 5, 10, 15, 20, 25, and 10 per cent of the total starch concentration. The Corn Industries viscometer was used for cooking the puddings and for measuring the viscosity of the hot pastes. The beginning temperature of the uncooked puddings was 33.50 C. All puddings were cooked to a temperature endpoint of 950 C in a liquid bath thermostatically controlled at 1000 C t 10 C. At the end of the cooking period b samples of the hot pudding were poured immediately into specially designed metal molds adjusted on double glass plates. These samples were covered and held at room temperature for l—hour and l-hour periods. Flow measurements were made on 2 samples at the end of each of these holding periods. Results of viscosity tests showed that maximum viscosity values and the height of the pasting curves were proeressively increased with an increase in the level of waxy maize starch. The waxy maize starch which had undergone less chemical modi— fication had a greater effect on increasing viscosity than did the more highly modified waxy maize starch. Despite the wide variation in hot paste viscosities attributable to the kind and the level of waxy maize starch, no differences were observed between the flow measurements Of the cut pudding samOles which could be related to these same factors. From this limited study it appears that concentrations of waxy maize starch up to 10 per cent of the total starch do not alter appreciably the gelline tendencies of ordinary corn starch in simple puddings. Differences in cooking time required to reach 950 C were not related to the kind or level of waxy maize starch but appeared to be associated with some unexplainable differ— ences in the rate of heatinm in the cookinp instrument. Differences ih flow of the cooled samples were shown to be related to the length of cooking time and the degree of paste breakdown. An increase in cooking time produced an increase in flow of the cooled sample. The significance of this rela— tionship appeared to diminish with an increase in the length of the holding period. ' r w». ,, ' ' r .v 1“ .’-- . ‘.-. ,(‘t ;.H gaigffi . $32. 3 v. n.’ 3' .‘ ‘if'V' at } 3;. I .' ‘1'. f? 1" Latf I.‘ 1 fl {:4‘4’; y 'U 09% if; ’57 JUL I 3 '58 he at O (in' ...No Tia m. ‘3’: .4 HICHIGRN STQTE UNIV. LIBRQRIES llll 3 ||| llllllillllllllll llll ||||| ”Ill“ 9 1 8 2 12306632 “WI 4