SENSORY DETERM‘NATION OF THE AMOUNT OF FLAVOR CHANGE CAUSED BY GAMMA iRRADlATION EN SELECTED ANIMAL PROTElN FOODS Thesis fer the Degree of M. S. MiGHlGAN STATE ONi‘iERSifi' BLAME? SOOARMAEL 19?} IHIHWI'IHllHIl/H)“1“!!!"leMINIMUM!!! L 301032 6613 SENSORY DETERMINATION OF THE AMOUNT OF FLAVOR CHANGE CAUSED BY GAMMA IRRADIATION IN SELECTED ANIMAL PROTEIN FOODS ,BY Slamet Sudarmadji AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1971 ABSTRACT SENSORY DETERMINATION OF THE AMOUNT OF FLAVOR CHANGE CAUSED BY GAMMA IRRADIATION IN SELECTED ANIMAL PROTEIN FOODS BY Slamet Sudarmadji Irradiation of foods can cause a characteristic flavor change. In animal protein foods such as meat, there has been reported a species variation in this flavor development. The objective of this research was to measure the response of animal protein foods derived from twenty species of animals of different biological classifications. A qualified expert panel of judges scored these foods on a flavor intensity scale of five. The foods were irradiated with a number of doses of gamma radiation over a range of 0 to 5 Mrad. Statistical analyses of the dose- flavor relationship were made and conclusions drawn as to flavor threshold dose and species variation in sensitivity. Foods studied were: beef, lamb, pork, chicken, turkey, venison, bear, whale, sea turtle, hippopotamus, elephant, horse, rabbit, opossum, beaver, shrimp, lobster, trout, halibut and frog. SENSORY DETERMINATION OF THE AMOUNT OF FLAVOR CHANGE CAUSED BY GAMMA IRRADIATION IN SELECTED ANIMAL PROTEIN FOODS BY Slamet Sudarmadji A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1971 ACKNOWLEDGMENT The author very sincerely wishes to express his thanks to Dr. W. M. Urbain, his advisor, for encouragement, patient guidance in his graduate study program and for his valuable suggestions and assistance in the preparation of this thesis. 1 The same appreciation is extended to Dr. B. S. Schweigert, past the chairman of the Department of Food Science for his interest. Also to his guidance committee members, Dr. P. Markakis and Dr. W. T. Magee, his sincere thanks for their generous assistance, especially in statistical analysis of this thesis. Special thanks are given to George Giddings and Cheryl Grosbeck, research assistants, for their assistance 60Co radiation unit. in Operation of The author expresses gratitude to the Indonesian Government, also Gadjah Mada University, his alma mater, and the Midwest Universities Consortium for International Activities (MUCIA Inc.) for assistance and most of the financial support of this work. ii Because of the involvement of above mentioned persons and institutions, this thesis was made possible. iii TABLE OF CONTENTS LIST OF TABLES . . . . LIST OF FIGURES . . . . INTRODUCTION . . . . . LITERATURE REVIEW . . . The Critical Dose . . The Chemistry of Irradiated Flavor and Odor . . . . . The Role of Lipids . . The Role of Protein . . The Carbohydrates . . . Synthetic Odor Approach . Elimination and Prevention Odor and Flavor . . . MATERIALS AND METHODS . . Materials . . . . . Preparation of Sample . Irradiation Process . . Cooking Method . . . . Taste Testing Method . . The Taster and Panel Room The Statistical Methods . RESULTS AND DISCUSSION . . Irradia Approach for Determination of "Threshold Dose" for Each Animal Protein Food SUMMARY AND CONCLUSIONS . LIST OF REFERENCES . . iv Page vi 11 14 14 15 18 18 19 20 21 21 23 26 28 50 55 56 LIST OF TABLES Table Page 1. Irradiated flavor intensity score of 20 animal protein foods in ten irradiation doses by five trained expert judges . . . . 32 2. Judges' scores of irradiated flavor for samples evaluated . . . . . . . . . . 38 3. Degree of Freedom, Mean Square and F values for Analysis of Variance . . . . . . . 38 4. Interpolated value of SSR . . . . . . . . 39 5. Ranking of animal protein food samples according to their means of irradiated flavor for overall dose evaluation based on flavor intensity scale of one to five (none to very strong) . . . . . . . . 40 6. Irradiated flavor means of animal protein food samples and their standard error . . . 45 7. Irradiated flavor score means for all animal protein food samples at each dose 0 O I I O O O O I O O O O O 46 8. The "threshold dose" for each animal protein food investigated, determined at flavor intensity score value of 2 (slight irra- diation flavor) . . . . . . . . . . 53 LIST OF FIGURES Figure Page 1. The scission of glycerol stearate . . . . . 10 2. Significant or not significant differences among the animal protein food samples . . . 42 3. Curves showing irradiated flavor vs. dose for pork (most sensitive to radiation), elephant (least sensitive), bear (medium sensitive) and the average of 20 food samples . . . . . . . . . . . . . 43 4. Irradiated flavor score vs. irradiation dose for pork and bear meat . . . . . . . . 51 vi INTRODUCTION The application of ionizing radiation technology to food preservation has been the subject of intensive scien— tific research. This research appears promising and feasible for treating many foods. The values of radiation preservation of meat and poultry products are mainly the extension of product life at refrigeration temperatures with the pasteurizing doses, and indefinite preservation without refrigeration at a sterilizing dose. The irradiation process accomplishes these effects mainly by destruction of spoilage microorgan- isms. Beef, for example, irradiated to 100 Krad has its storage life increased two to five times (Hannan and Thorn- ley, 1957). For fishery products, 150-450 Krad is the Optimum radiation dose range and the refrigerated life of these foods can be expected to be doubled or tripled. The radiation preservation process of fish and shellfish has been reported to be feasible (Holston, et_al., 1967). Regardless of achievements in the technology of food irradiation or of demonstration of the wholesomeness and of government clearances of the products or of considerations of economic feasibility, the commerciali— zation of the irradiated products will be largely deter— mined by consumer reaction and acceptance of such products. Consumer acceptance mainly is determined by factors such as significant physical or chemical changes in the products caused by irradiation process. Changes in flavor, odor and color can greatly influence the acceptance. Undesirable flavor and odor can be developed by the irradiation process in protein foods. Although the flavor and odor are similar in character among the several kinds of meats, intensity of flavor and odor has been reported to be different. The intensity is reported to be greater in beef, for example, compared with chicken and pork (Hannan, gt_gl., 1957). Continued research on the development of the tech- nology of food irradiation both in the United States and elsewhere may be expected to lead to a number of appli— cations. Of special interest is in the use of low tempera- 'tures during irradiation to reduce off-flavor develOpment. While a great amount of work has been reported on the chemical nature of the irradiated flavor and odor, there are only scattered reports on the flavor response of different animal protein foods to irradiation. This present work was undertaken to measure by taste panels the flavor changes obtained by irradiating protein foods from twenty species of food animals. Five trained expert panel-members evaluated the irradiated flavor of protein food samples with a number of doses of gamma radiation over a range of zero to five Mrad. Twenty selected food animals covering a wide dis- tribution over the animal kingdom were used in this experi- ment. They were: beef, swine, lamb, chicken, turkey, deer, bear, whale, sea-turtle, hippopotamus, elephant, horse, rabbit, opossum, beaver, shrimp, lobster, trout, halibut and frog. LITERATURE REVIEW For describing the degree of flavor and odor changes produced by the ionizing radiation process in the animal protein foods, no exact methods of chemical or physical measurement have been worked out to date. Despite the inaccuracy of the subjective evaluation method, it has been used widely. Because of the low concentrations of odor and flavor constituents, variabilities of the samples and differences of human responses, it is not unusual for disagreement to occur among the results from various laboratories in odor and flavor evaluation of irradiated foods. I It is not surprising also that there is some dis- agreement on the description of irradiated odor and flavor. Various terms or combination of those terms have been used to describe the human response to the irradiated odor and flavor, such as burnt, metallic, bitter, cured meat, remi- niscent of cress, cheesy, goaty, wet dog, wet grain, acrylic, or unappetizing (Huber, et_al., 1953; Mehrlich, 1966; Batzer, et_al., 1959). Regardless of the different terms used to describe the irradiation odor and flavor, Merritt, et al. (1967) concluded that the irradiation odor in raw meat is a characteristic property, is the same for beef, pork, lamb and the other meats, and does not vary in type but only in intensity. In trying to clarify the mechanism and the back- ground of irradiated flavor and odor development, some research workers have evaluated the intensity of irradiated flavor and odor and ranked the meats according to their sensitivity to radiation. Hannan and Thornley (1957) reported their finding that beef is more sensitive than chicken, but pork is less sensitive. Another report ranks meats in order of decreasing sensitivity as follows: beef, lamb, veal and pork (Huber, gt_al,, 1953). A statement by Coleby (1959) agrees with Huber's that beef and lamb are more sensitive than pork, chicken, turkey and bacon. No ranking of irradiated odor and flavor sensi- tivity of other meats is available. The Critical Dose No off-odor could be detected in beef up to 93 Krad but at 470 Krad or higher it could be detected readily (Anon., 1962). Hannan and Thornley (1957) suggested the critical dose for beef is about 100 Krad, below which dose no distinctive flavor change from the unirradiated control could be detected. They reported the figure for lean pork is 250 Krad. And only a slight detectable flavor change developed in commercial pork sausage, as high as the dose of two Mrad. For the British bacon the critical dose is similar to that of the lean pork. Huber, et_al. (1953) believed that irradiated beef at 100 Krep has a better flavor than the unirradiated con- trol. The critical dose appears to be 250 Krad for irra- diated chicken, stored anaerobically. The irradiated flavor at that dose is just detectable if the chicken is cooked by steaming (Coleby, 1959 and Thornley, 1957). Holston, et_al. (1967) reported the optimum dose range for irradiation of fish and shellfish is 150—450 Krad. No irradiated flavor could be detected in trout (Salvelinus namaycush) irradiated up to 0.5 Mrad, cooked by baking without seasoning at 350° F for ten minutes (Graikoski, et_§l., 1967a). Groninger and coworkers (1956) examined tuna irra— diated at two Mrep. This dose produced a desirable pink color but an undesirable burned odor. Graikoski, gt_gl. (1967b) reported that Whitefish was acceptable if irradi- ated up to 0.3 Mrad but not at 0.4 Mrad. No perceptible odor change in cured ham was reported up to two Mrep. Bacon and corned beef were free from detectable irradiated flavor at 1.5 Mrep after cooking (Groninger, et al., 1956). The Chemistry of Irradiated Flavor and Odor Batzer, et_al. (1959) described the odor produced in beef by irradiation up to two Mrad as "sulfide- mercaptan"-like odor, up to four Mrad as "wet grain" odor and up to eight Mrad as "slight burnt and wet grain" odor type. These sensory descriptions give us a vague picture of the chemical changes quantitatively as well as quali- tatively in beef with increasing irradiation dose. Batzer and coworkers (1959) further found out that hydrogen sulfide, methyl mercaptan, acid-soluble carbonyl compounds and pH were increased in irradiated beef; on the other hand glutathione and glycogen disappeared. The possible precursors of the undesirable odor components formed during gamma irradiation of beef are water soluble and probably contain nitrogen and/or sulfur (Huber, gt_al., 1953; Drake, gg_al., 1957). Huber, gt_al. listed the possible source of the irradiated odor as: 1. Formation of hydrogen sulfide and other sulfur compounds. 2. Formation of isovaleraldehyde from free leucine. 3. Various soluble protein and other amino acids. 4. Cabbage odor from irradiated methionine solution. 5. Indole from tryphthophane. 6. Geranium odor from phenylalanine solution. 7. Reaction involving lipid and lipid soluble compounds. The Role of Lipids It was reported by Huber, et_al. (1953) that in beef with higher fat content less destruction of gluta— thione and less hydrogen sulfide occurred, but there was no significant difference of irradiated flavor between lean and fat beef. Groninger, e£_§l. (1956) observed the increasing peroxide concentration in beef and pork during irradiation in the presence of oxygen. Rancidity was believed to have a linear correlation with peroxide content and in this way peroxides contribute to the overall irradiation odor and flavor. They made an observation on the correlation between peroxide value and unsaturated fatty acid content. Due to the higher unsaturated fatty acid content of pork than beef, the peroxide value in pork is two and a half times that of beef at 30 Mrep. Mitchell (1957) supported Groninger's statement by his finding that the source of irradiated flavor and odor appears to be the lipid com- ponents. Although peroxides seem to be involved in the for— mation of irradiated flavor and odor, peroxides themselves are odorless. They play a role in oxidation process by producing secondary products such as aldehydes and ketones. Through the work by Champagne and Nawar (1967), hydrocarbon series of n-alkanes, l-alkenes, internally unsaturated alkenes and alkadienes were identified in irra- diated beef and pork fat. They determined that a typical off-odor was detectable in beef and pork fat irradiated at two Mrad and was more intense at higher doses. It has been observed that the unsaturated hydrocarbons are much more odorous than the saturated ones. Champagne and Nawar also determined that the quantities of l-heptene, and l- octene in beef and pork fat irradiated at six Mrad exceeded their odor concentration threshold in mineral oil. At two Mrad beef and pork fat exhibit some off odor. At this dose, however, the concentrations of all hydrocarbons were found to be below the threshold levels for a detectable odor. These workers suggested the possibility, therefore, of additive or synergistic effects. Merritt, §E_al. (1966) believed that the hydrocar- bons found in irradiated meats can come only from the lipid. They also noted that the irradiated butter fat had a characteristic fat irradiation odor, unlike the typical rancid odor produced in an oxidized sample. Likewise they also observed that a large amount of hydrocarbons but only a small quantity of carbonyl compounds were produced in irradiated butter fat, whereas a large amount of carbonyl compounds and small amount of hydrocarbons were produced in oxidized butter fat. Their conclusion was that the lO mechanism for irradiation production of volatiles is different from the mechanism for oxidation. Another work by Merritt, et_31. (1966) using gas chromatography and mass spectrometry determined the vola- tile components of irradiated meats. It was concluded that the radiation changes in lipids appear to be the result of radiation-induced direct bond cleavage of the lipid mole- cules. The scission of glycerol stearate at all points of the chain, will produce alkyl free radicals, and with recombination or hydrogen termination, all the n-alkanes from methane to heptadecane could be formed. CH CHZCHZCH2CH2CH2CH2CH2CH2CH2CH2CHZCHZCHZCHZCHZCHZCOOCH 3 I “J ‘ -coo - CH (:2 1 -COO - CH2 2 C3 C4 Cn - - --— - (From Merritt, et a1.: Irradiation Damage in Lipids, 1966.) Figure l. The scission of glycerol stearate. Their hypothesis was supported by their finding that the alkanes from methane to pentadecane occur in irra- diated meat. The homologous series of alkenes, which were also detected in moderate quantities, they theorized resulted from secondary collisions, in which a second electron was 11 extracted from the radicals. Alcohol production was explained as a reaction of alkyl free radicals with water molecules. They concluded that their data support the hypothesis that radiation products are primarily the result of direct bond cleavage, while the carbonyl compounds are produced by an indirect route through the absorption of oxygen by arIalkyl free radical followed by decomposition. The Role of Protein Many researchers believed that water soluble pro- teins are a major source of the irradiated flavor and odor (Drake, e£_31,, 1957; Huber, e£_al., 1953; Batzer, et_gl., 1955; and Harlan, gt_al., 1967). Merritt, et_§1. (1966) stated that amino acids were responsible for sulfur and aromatic compounds production in irradiated meats. Drake, gt_al. (1957a) made an extensive study of irradiation of protein in dry and wet state. They found that the irradiated odor was more pronounced from a wet sample, whereas almost odorless products were derived from the dry protein sample. It was also noted that odor increased when water was added to protein which had been irradiated in the dry state. That individual amino acids react differently to irradiation was proved by Drake, et_al. (1957b) by their measurement of amino acid survival in l per cent insulin solution irradiated up to 40 Mrep. Cystine appeared to be 12 the most sensitive to irradiation followed by tyrosine, phenylalanine, proline, histidine and glycine. Besides cystine, tyrosine, phenylalanine, proline and histidine, Drake, et_al. (1957a) believed that methio— nine, cysteine and tryphtophane are likely the sources of irradiated odor and flavor in protein. Schweigert (1959) also reported an undesirable odor was produced from irra- diated leucine. Drake, gt_al. (1957a) reported on the specific odor and flavor produced by irradiation of amino acids. Methio- nine for example, if electron irradiated will produce a strong odor described as similar to cooked garlic and cab- bage. A rose-like flowery odor was reported by them from irradiated phenylalanine, polyphenylalanine and n-acetyl- phenylalanine solutions. The role of protein in irradiated odor and flavor develOpment is unclear. Again Drake, gt_al. (1957a) reported the levels of free amino acid, for example, in highly acceptable protein foods such as pork or chicken are not dissimilar from those in a less acceptable food such as beef. They showed that the addition of certain amino acids to the sample prior to irradiation does not significantly lower its acceptability. These workers pointed out also that the odors from irradiated protein are not directly related to amino acid composition, but more closely to the availability of functional groups. 13 Using gas chromatography and mass spectrometry techniques, Merritt, et;gl, (1966) found the presence of components from irradiated meats which they concluded arose from the direct bond cleavage of amino acids. They believed that sulfides, disulfides and mercaptans are derived directly from cystine or methionine. Aromatic com- pounds such as benzene and toluene can be produced from phenylalanine and phenol and p-cresol from tyrosine. Some compounds such as hexyl mercaptan or ethyl-buthyl-disulfide, they explained as probably originating from free radicals derived from lipid and protein portions of meat. Very similar to the finding by Merritt, Ronsivalli, gt_§l. (1967) using Time of Flight Mass Spectrometer (TOFMS) detected at least 32 compounds in the volatile fraction of clam meat (Mya arenaria). They found that dimethyl sulfide was the dominant component and the source of the typical clam odor. They identified components which are present in irradiated meat such as sulfides, esters, alcohols, amines, carbonyls, carboxylic acids and hydro- carbons, can also be postulated as arising from the radi- ation effects on the protein and lipid portions of sea foods. In work somewhat contrary to the postulated impor- tant role of fat and protein in producing irradiated odor and flavor, Volkova, et_§l. (1955) reported that lipid- protein complexes decrease some of the effect of irradi- ation. 14 The Carbohydrates Very little information is available on the possible role, if any, of carbohydrates in the formation of irradiated flavor and odor. Long, et_al (1957) reported the depolymerization of carbohydrates by radiation. For- maldehyde was produced by gamma-irradiation of carbohy- drates. They made an observation on irradiation of a d- glucose solution resulted in decomposition products such as acid and a reducing sugar different from d-glucose. Because of the relatively small amount of carbo- hydrates present in animal protein foods, they cannot be considered to be a major source of irradiation flavor and odor. Synthetic Odor Approach The role of lipids and protein in the contribution of irradiated flavor and odor has been previously reviewed. Many researchers believed that the irradiation flavor and odor in animal protein foods are caused by the volatile chemical compounds produced by radiation impact on the protein and lipids molecules (Huber, §E_§l., 1953; Drake, g£;al,, 1957a; Wick, et_gl., 1967; Merritt, et_al., 1967). Using gas chromatography techniques it is possible to trace very small concentration of volatile components in animal protein foods. With this method many odor com— ponents can be determined to a degree approaching the human sensitivity. 15 By using the results of chromatography and by trial and error methods, Wick, §E_gl. (1967) succeeded in pre- paring a mixture of substances which had an odor resembling the irradiated odor of beef. This odor was produced by mixing just three components (which are also present in non-irradiated and irradiated beef), in the right concen- tration and proportion, i.e., methional, l-nonanal, and phenylacetaldehyde at the concentration of 5.0 ppm, 0.5 ppm and 0.25 ppm respectivelyor proportion of 20:2:1. .They believed that the three substances are not completely responsible for irradiated odor and flavor, however, they are the most important contributors. Elimination and Prevention of Irradiated Odor and Flavor There are two kinds of molecular damage caused by irradiation. The first is a direct effect of radiation, in which a molecule is struct directly by an incoming par— ticle or photon and split into fragments. The direct effect is very important in dry food materials or in very concentrated solutions (Frigerio, 1967; Drake, et_al., 1957). The second is the indirect effect in which a molecule, such as water, is split into reactive fragments or radicals which then react with other molecules. With indirect action the diffusion of fragments or radicals may be sufficiently slow, that chemical protective agents may 16 be placed in their path, and sacrificed to protect more critical molecules (Frigerio, 1967). If the irradiation is done at very low temperature, active fragments and radicals will not be able to move freely from their point of origin and their indirect action thereby reduced. Harlan, et_al. (1967) showed that beef steak irra- diated with six Mrad at -l96° C (-320° F) had a flavor com- parable with the non-irradiated control. This finding apparently supports the belief that the irradiation flavor is caused largely by indirect effect of radiation. As noted previously, besides the irradiation at low temperature, some chemicals have a preventive effect on the development of the irradiated odor and flavor. Some vitamins, such as a mixture of alpha tocopherol, ascorbic acid and vitamin A can cause less irradiation flavor in beef (Anonymous, 1962). The effect of formation of lipid-protein complexes in prevention of irradiation effect in meat has been mentioned earlier. Mitchell (1957) further suggested that the elimi- nation of oxidative changes related to the radiation fla— vor, can be achieved by removal of oxygen, exclusion of light, use of low temperature or a combination of those. Cooking methods may help in reducing the irradi— ation flavor. Thornley (1957) reported that chicken 17 irradiated at 250 Krad cooked by steaming could be distinguished from control, but by roasting, chicken irra- diated at 375 Krad could not be distinguished from the con- trol. Waters, et_al. (1969) conducted taste tests, and concluded that irradiated salmon steaks with doses up to 4.5 Mrad at -30° C (-22° F) were more acceptable when heated in oil rather than in an open air oven. Furthermore they found that tuna and salmon were highly acceptable when breaded and cooked, or when using hickory smoke flavor for masking the irradiated flavor. Finally, Urbain (1965, 1971) suggested the follow- ing possible ways to reduce the irradiated flavor and odor by: 1. -Exclusion of oxygen during and after irradiation. 2. Use substances as free radical acceptors. 3. Use low temperatures during irradiation. 4. Absorbents and flavoring for masking (spices, tomato). 5. Selection of foods insensitive to irradiation. 6. Using a small dose. MATERIALS AND METHODS Because the main tool employed in this investi- gation of irradiated flavor is a subjective method using human sensory perception of flavor, broad representation of samples is needed. Only expert trained tasters were employed. Materials The animal protein foods were selected to cover animal species as varied as possible, but still keeping in mind the possibility of the taster's acceptance as foods. The animals used as sources of the foods can be placed into the phylla of Chordata (subphyllum Vertebrata) and Arthropoda (class Crustacea). The systematic classification of the samples is as follows (Gray, 1965; Thomson, 1964; Walker, 1968): Phyllum: Arthropoda Class: Crustacea --Shrimp —-Lobster Phyllum: Chordata, subphyllum Vertebrata Class: Mammalia Order: Carnivora Order: Proboscidea --Bear --Elephant 18 19 Class: Mammalia (cont.) Order: Lagomorpha Order: Artiodactyla --Rabbit --Beef --Deer Order: Cetacea --Lamb --Whale "SW1ne -—Hippopotamus Order: Rodentia Order: Marsupialia --Beaver --Opossum Order: Perissodactyla --Horse Class: Aves --Chicken --Turkey Class: Reptilia --Turt1e Class: Amphibia --Frog Class: Pisces -—Halibut --Trout The samples to be tested were supplied by Michigan State University Food Store in East Lansing or Czimer Foods, Inc. in Chicago. All the samples were kept frozen until shortly before use. The histories of the foods prior to receipt were unknown. Preparation of Sample The samples were cut into steaks (except shrimp, lobster, and frog leg) in frozen state, about one inch thick and about 0.5-0.75 lb for each cut. Altogether 2O twelve sample steaks were needed for each food, ten for the ten dose levels of irradiation plus two non-irradiated con- trols. Each cut or reasonable division of shrimp, whole lobster or frog legs was packed and vacuum sealed in Inter- national Kenfield's I.K.D. Super Vacuum packaging pouch (All-Vak # 13 F.B.R.). This gas-impermeable pouch consists of Mylar polyester base with a thin coat of polyvinylidine chloride (Saran) applied to the outer surface of a heavier coat of polyethylene as a sealant. The samples then sealed with the Kenfield Vacuum Sealer (Model C14AN). The samples were thawed prior to irradiation. Irradiation Process 6OCo irradiator in The source of radiation was the Food Science Building at Michigan State University, East Lansing. The main source consists of 24 BNL MK-l radio- active Co strips, doubly encapsulated in stainless steel sheaths, arranged to form a cylinder or a well. The dose rate in the center well was about two Mrad/hr at the height of 15.2 centimeters. The temperature in the irra- diation chamber was kept at 40-50° F using the refriger- ation facilities. The sample temperature prior to irradi- ation was about 40-50° F and after removal from the source was about 60° F. Samples were irradiated in I.K.D. plastic bags. One sample of each food was irradiated at one of the 21 following doses: 0, 10, 50, 100, 500, 1000, 2000, 3000, 4000, and 5000 Krad. Two non-irradiated samples were used as control. All the samples were irradiated in the center well of the source. Cooking Method It is known that the cooking method has an effect on taster's ability to detect the irradiation flavor (Hannan, gt_al., 1959). Chicken samples irradiated at two Mrad, and cooked by pressure cooking method and by steam- ing, had an irradiated flavor easily detected by panel. If the samples were grilled lightly, or stewed in water, or deep fat fried, the ability of panel to detect the irra- diated flavor was decreased. In this present study, the samples in I.K.D. plastic bags which were evacuated and sealed were cooked in boiling water for 30 minutes. This method assured a well—done degree of cook, no burning and no leaching by the cook water. The samples were cut into pieces suitable for the panel (about ten grams), and kept warm over warm water during the time of sample presentation to the panel (as long as 30 minutes). Taste Testing Method Taste testing methods and procedures were dis- cussed in some publications (Anonymous, 1963; Anonymous, 1964; Kramer, 1966; Hirsh, 1970). 22 In this present work, the series of twelve samples for each food were divided into two groups, one for each of two panel sessions. The first session employed samples with doses of 0, 50, 500, 2000, and 4000 Krad with one non—irradiated control. The samples for the second session had doses of 10, 100, 1000, 3000, and 5000 Krad, with one non-irradiated control. The type of test used was the rank order test (Committee on Sensory Evaluation of the Institute of Food Technologists, Anonymous, 1964) or ranking test (Anonymous, 1963). To harmonize the result of the flavor evaluation in each session the taste testers discussed and agreed to the value of the irradiated flavor at a particular dose. For this purpose, the first session determined the value of irradiated flavor of the 500 Krad sample; in the second session the 3000 Krad sample was used to serve this purpose. To help the tasters, one identified non-irradiated control was also served. After the tasters agreed upon the value of the score of irradiation flavor at that particular dose, the other samples were presented in individual booths, disguised under coded numbers, and concealed from color differences as much as possible by using blue fluo- rescent light. The irradiated flavor intensity was recorded by number scores on score sheets; the score range is one to five, representing: 1 = no irradiation flavor detected 23 2 = slight irradiation flavor 3 = moderate irradiation flavor 4 = strong irradiation flavor 5 = very strong irradiation flavor The score sheets employed appear on the next two pages. Score sheet A was used in determining the value of the score of the irradiated flavor of the 500 and 3000 Krad samples. Score sheet B was intended for scoring of other samples. The Taster and Panel Room According to the Committee on Sensory Evaluation of the Institute of Food Technologists (Anonymous, 1964) for the rank order type test, three to ten trained panel- ists are needed, and two to seven samples per test can be served. In this present work, the panelists were obtained from the staff of the Department of Food Science of Michi- gan State University in East Lansing. Prior to the actual taste testing, several training sessions to familiarize the panel with the irradiated flavor of beef, pork and chicken were conducted. In addition, triangle tests were given in order to select panel members who had good sensi— tivity to the irradiated flavor of animal protein foods. Five permanent members were selected, with five other members as alternates. 24 Score Sheet: A CALIBRATING PROCEDURE FOR RANKING IRRADIATION FLAVOR: Name: Material: This is a ranking test with two samples. One untreated sample, Coded R, is your reference. Please taste the samples, discuss with other panel members, and decide in agreement one of the following categories: Irradiation flavor: Sample number: 1. None 2. Slight 3. Moderate 4. Strong 5. Very strong 25 Score Sheet: B FLAVOR INTENSITY SCORE Judge: Date: Material: Indicate the intensitygf irradiated flavor, if any, by checking the appropriate box opposite the term which des- cribes the degree of irradiated flavor. Use as much time as you need. Please use as reference: Sample No.: Score: Term: SAMPLE NUMBER Score/Term 1. None 2. Slight 3. Moderate 4. Strong 5. Very strong Comments: Thank you. 26 The panel room used for testing was equipped with individual boots each with blue fluorescent light. The room was pressurized to avoid contamination with odors from the kitchen and was fully air conditioned. The sessions were conducted in either morning or afternoon. No more than one session was held on a given day. The Statistical Methods The statistical methods used in this analysis are described in "Principles and Procedures of Statistics" by Steel and Torrie (1960). The procedure for the Analysis of Variance of Split-plot designs is summarized as follows: Step one: Find the Correction Term and Total Sum of Squares. X 2 Correction Term = r.a.b = C X = grand total of observation r = the number of blocks or judges a = the number of animal foods or whole units per judge b = the number of doses or subunits per whole unit Total Sum of Squares (Total SS) of Subunits = 2 *) 15k xijk C )X denote the score value in the ith block (or judge) fromltfie subunit for the jth level of factor A (ani- mal food samples) and the kth level of factor B (dose of irradiation). 27 Step two: Complete the Whole—unit Analysis. Z. x2 . Whole Unit 53 = il—B—iJ;-- c i X21 Judge SS = a.b - C 2x2. SS (A = Animal Food Samples) = l—E—é14-- C Error (a) SS Whole Unit SS - Judge SS - SS (A) Step three: Complete the Subunit Analysis. EXZ 55 (B = dose) = ——E—é;5 - c A x2 . 55 (AB) = 3k r°3k - c - SS(A) — 35(3) Error (b) = Total SS (Subunit) - Whole Unit SS - SS(B) - SS(AB) The F-test Calculation SS Mean square = degree of freedom Mean Square of Animal Foods F animal fOOd samples = Mean Square of Error Ia) Compare F calculated with F value from Table S (Rohlf and Sokal, 1969). If F calculated > F table, there are significant differences among the animal protein food samples. This F test method can be extended to test the difference among the treatments (radiation dose). 28 Duncan's New Multiple-range Test If there are significant differences among the animal protein food samples or among the treatments, we may proceed to the Duncan's multiple range test to see its individual difference. Procedure: rror mean square r 1. Determine S; = \[e See Table A.7 (Steel and Torrie, 1960) for Significant Studentized Ranges (SSR). Find the Least Significant Ranges (LSR) by multipli- cation of SSR by Si' 2. Rank the means. Arrange the means of the ani- mal protein food samples in order from the smallest to the highest value. 3. Test the differences: largest minus smallest, largest minus second smallest, . . . , largest minus second largest, then second largest minus smallest, second largest minus second smallest and so on. Each difference is declared significant if it exceeds the corresponding LSR. RESULTS AND DISCUSSION Statistical Analysis. Analysis of Variance of Split-plot design (Steel and Torrie, 1960) was applied The data secured with in Table 1. in treating the data. the taste panels are given Calculation: 2 2 - _ _ X... _ 2404 _ Correction term — C - r.a.b - 5 x 20 x 10 — 5779.2 Total SS = ,2 X2.. - C 12 + l2 + ... + 52 - C ijk 13k = 7178 - 5779.2 = 1398.8 1' x21. 292 + 282 + + 232 Whole Unit SS = ' - C = °'° b 10 - 5779.2 = 5881.2 - 5779.2 = 102 z 2 . X .. 2 2 2 _ l _ _ 468 + 481 + ... + 469 Judges SS - _—ETB— C — 20 x 10 - 2779.2 = 4.7 2x2. SS A (Animal Food Samples) = l—E—fil; - C = 1412 + 1192 + ... + 1222 5 x 10 - 5779.2 = 60.7 29 30 Error (a) SS = Whole Unit SS - Judge SS - Animal Food Sample SS (A) = 102 - 4.7 - 60.7 = 36.6 Z x2 2 2 2 _ k ..k _ _ 130 + 122 + ... + 403 SS B (Dose) — r.a C - 5 x 20 - 5779.2 = 1058 Zk x2 .k SS A-B (Animal Food Sample - Dose) = l——E—42—-- C 2 2 2 _ SS(A) _ SS(B) = 5 + 5 +5... + 23 - 5779.2 - 60.7 - 1058 = 141.9 Error (b) SS = Total SS - Whole Unit SS — SS (B) - SS(AB) = 96.9 F Test Calculation F calculated for animal food samples = Mean Square of food samples = 3.195 Mean Square of Error—TaIF- 0.482 = 6.63 Interpolated value of F (Table SzRohlf and Sokal, 1969): F (0.05, v1, v2) = F (0.05, 19, 76) = 1.73 F calculated > F table; there is a highly significant difference of irradiated flavor among the animal protein food samples or there is a significant difference in flavor sensitivity of animal protein food samples to gamma irra- diation. Mean Square of Dose (B) F calculated for dose = Mean Square of Error (b) _ 117.555 _ — —OTI§4— — 877.27 31 Table S: F(0.05, v1, v2) F(0.05, 9, 720) = F(0.05, 9,”) = 1.88 There is a very significant effect of irradiation dose on the irradiated flavor of animal protein food samples. F calculated for interaction of animal protein food . _ Mean Square AB samples Wlth dose (AB) — Mean Square of Error (b) _ 0. 30 _ ‘ 0.134 ' 6'19 Table S: F(0.05, v1, v2) 00 = F(0.05, 171, 720) = F(0.05,~ , ~) = 1.00 F calculated > F table. There is significant inter- action between samples and doses. Have been previously proved statistically that there is a significant difference of irradiated flavor or sensitivity of animal protein food sample to irradiation among each other. To determine the significance of dif- ferences between each sample, Duncan's new multiple—range test was applied. ..-.—..¢—. ,e...n. H ‘7 \ I55. I‘J 1 protein foods in ten irra- ima tensity score of 20 an diation doses by five trained expert judges. 1n TABLE 1.-Irradiated flavor Krad D088 Animal 1000 2000 3000 4000 5000 Total Mean 500 100 10 "judge ll Food HO VLF) H 29 28 28 Beef 82 29 27 17 16 20 25 25 141 15 32 ONO HLD H 23 27 23 Venison 2.38 24 22 10 16 15 22 18 119 10 10 KOO Hm r-1 23 21 24 Lamb 2.32 26 22 15 20 19 116 16 10 LOO vm H 30 28 29 28 Pork 90 30 145 21 101 10 15 15 19 20 23 23 10 10 13 10 16 17 10 Hippo- potamus 33 HO Hm H 21 21 21 Horse 2.22 24 24 111 23 23 5 21 14 20 15 10 OO NLn ...; 24 26 Bear 2.40 moderate flavor. 24 ion 21 22 120 iation flavor; 3 irradiat 20 irrad very strong 17 slight 5 10 2 flavor; 10 ion flavor; iat ion irrad t no irradia = strong 4 1 flavor; Legend irradiation CECE ~M...C._. CCC: COO? 000m» OOON OOOH DOD, OOfi 00 OH. O I I .lll '1 II1 : UHWWVJ h... NJOvNJKN. «iv-H NV! HIV-dd alvAln Nu an-ufi -< TABLE 1.-Cont. Krad Dose Animal Food "judge II 3000 4000 5000 Total Mean 2000 500 1000 50 100 10 0 19 Elephant 20 10 13 10 18 18 100 10 34 II N ll v r40 . r~o NLn N oLn H H r~uwocoox r4 (VOJH NCVGWNrfi N oach H omumaomumthnu.v mqm 0.35 (table): Significant 41 Pork - Hippopotamus = 2.90 - 2.02 = 0.88 > 0.35: Significant Pork - Chicken = 2.90 — 2.68 = 0.22 < 0.31: not sig- nificant Determination of significant or not significant difference between two samples was continued until covering all twenty food samples. The results of differences among the animal protein food samples are summarized on the fol- lowing Figure 2. Any sample means not covered by the same vertical line are significantly different. Otherwise, any means covered by the same vertical line are not signifi- cantly different. It has been proved statistically that there was a significant interaction between kind of food and dose. If this interaction is large enough to be of biological impor- tance, the effect of radiation on irradiated flavor pro- duced in the foods should be discussed for each level of radiation, not on the basis of differences among the kinds of food averaged over all doses. Despite the evidence obtained from the statistical analysis of the interaction between kinds of food and dose, an inspection of the data reveals that this interaction is not of great biological significance. In general a food showing greater develop- ment than another displayed this greater sensitivity at all doses (Figure 3), and as a consequence, one can refer to this food as more sensitive to radiation than the other. Pork Beef Turkey 0 Chicken .5 Trout. Rabbit. .90 Pi.70 -2.60 —2.50 Shrimp/Lobster Frog/Whale Bear Halibut/Venison Lamb. Turtle. Horsefi Beaver. Opossum Hippopotamus- Elephant! .40 -2.20 —2.10 ‘2.00 42 High Irradiation Flavor A AL Low Irradiation Flavor The numbers are overall means of samples for whole dose irradiated flavor scores. Figure 2. Significant or not significant differences among the animal protein food samples. arranged as they are on Table 5. The samples are 43 .mmHmem poem om mo omnuo>m on» man A0>HuwmmMm Edwpoav numb .Am>auHmcmm ammoHv ucmnmmHo .AcoHumHomuuH ou w>HuHmcom umOEv xuo new mmoo m> uo>mHm pmumemuuH mcH3onm mo>usu .m musmwm Umux once A JL ooom oowv oowm 0&ON ooom owm 00H om 0H ucmndem a ..... momuw>¢ L mcouum hum> u m mcouum u v mumuoooe u m Sam \ £33m u ~ ”COG M H xuom uo>mHm pouanmuuH I roast; pauerpsxzxw 44 To illustrate the way each animal protein food responded to irradiation, Figure 3 shows the relationship between irradiated flavor intensity vs. irradiation dose (Krad). Table 6 gives the mean score of all doses of each animal protein food, mean of all foods and the standard error of the means. Table 7 shows the means flavor intensity scores for all foods at each dose studied, and clearly shows increas- ing flavor intensity with increasing dose. The data shown in Figure 2, indicate there is no evidence of a relationship between flavor sensitivity to irradiation of a food and the biological classification of the animal from which it is derived. For example, beef, deer, lamb, swine and hippOpotamus belong to the same order of Artiodactyla in the class of Mammalia, but the flesh foods obtained from these animals were scored in a scat- tered manner over the flavor sensitivity range. On the other hand, all the fish and sea foods investigated and the amphibian (shrimp, lobster, halibut, trout, whale and frog) have no significant difference in irradiated flavor intensity with respect to each other. As a group they fall in the middle of the flavor sensitivity scale. Arranging the animal food samples in order from the lowest to the highest mean values of the irradiated flavor 45 TABLE 6.--Irradiated flavor means of animal protein food samples and their standard error. Food Mean Score of All Doses Elephant 2.00 Hippopotamus 2.02 Opossum 2.04 Beaver 2.14 Horse 2.22 Turtle 2.26 Lamb 2.32 Halibut 2.38 Venison 2.38 Bear ' 2.40 Frog 2.42 Whale 2.42 Shrimp 2.44 Lobster 2.44 Rabbit 2.50 Trout 2.58 Chicken 2.68 Turkey 2.72 Beef 2.82 Pork 2.90 Mean of all foods 2.40 Standard Error of each food mean 0.0981 46 TABLE 7.--Irradiated flavor score means for all animal protein food samples at each dose. Means of Dose, Krad Irradiated Flavor a O 1.30 10 1.22 50 1.36 100 1.51 500 2.15 1000 2.27 2000 3.03 3000 3.30 4000 43.87 5000 4.03 aIrradiated flavor intensity score range from one to five: 1 = none; 2 = slight; 3 = moderate; 4 = strong; 5 = very strong. 47 scores, or in other words from the least sensitive to the most sensitive in flavor change due to irradiation, as shown in Figure 2, reveals that the most intensively domes- ticated animals such as swine, beef, turkey and chicken yield foods that are the most sensitive to irradiation. Less domesticated or unusual animals such as elephant, hip- popotamus and opossum give foods that are among the least sensitive of all observed. Possible reasons for this finding may be: 1. Animals, such as elephant and hippopotamus ordi- narily are wild (game) animals and the particular foods derived from them used in this study probably had such an origin. This is in contrast with the highly domesticated pOpular animals such as swine, beef, turkey and chicken. It is possible that the mode of living, environmental liv- ing condition, physical activity, variety of feeds, or even psychological condition of the animal, will have some effects on the animal sensitivity of flavor change due to irradiation. This is supported by the fact that all aquatic animals (sea foods and amphibian frog) have no sig- nificant difference in sensitivity to irradiation among each other. The effects of those factors to the physical or chemical nature of the animal foods which will contribute to the development of irradiated flavor are not exactly known. 48 2. Another possibility is the subjectivity of the sensory panel procedure. It is likely that lack of famili- arity with the unusual animal foods may have affected the judgements on the irradiated flavor. Somewhat contrary to the reports (Hannan and Thornley, 1957; Huber, e£_al,, 1953; Coleby, 1959), in this present work no significant evidence observed in sensitiv- ity difference among beef, chicken, turkey and pork. As shown in Figure 2, pork was the most sensitive to irra- diation among all the animal foods observed. There is the possibility that this result which differs from that of other investigators may be due to differences in quality of the meats or part of the animal used as sample. More work is indicated to clear up this disagreement, such as the application of procedures and statistical analyses which will be more sensitive in determination of the flavor change due to irradiation among beef, pork, chicken and turkey. For example the evaluation of irradiated flavor intensity among these animal foods at the same dose at one time can be suggested. Although there was no significant difference in sensitivity among the common kinds of meats, from Figure 2 in this present work it can be observed that beef is more sensitive than chicken and turkey as Hannan and Thornley (1957) and Coleby (1959) have reported. The contribution of animal fat to the irradiated flavor development seems not to be justified. From a 49 visual observation, opossum and beaver were among the fattest by the fact that thick layer of fat surrounded small piece of lean meat, but they belong to the least sen- sitive among other animal protein foods on the irradiated flavor scale. A report by Huber, gt_al. (1953) apparently supports this present observation. They stated that there was no significant difference of irradiated flavor between lean and fat beef. A report by Groninger, et_al. (1953) stated that beef was more senSitive to irradiation than pork, regardless of the belief that peroxides were produced in higher quantities in pork due to the higher unsaturated fatty acid content which could contribute more in overall irradiated flavor deveIOpment (Hannan and Thornley, 1957; Huber, g£_gl., 1953; Coleby, 1959). Once again in this present study, the role of fat and lipids as important con- tributors to the development of irradiated flavor and odor is in question. It is indeed, true that the chemical nature and the role of lipids in the development of irra- diated flavor in animal protein foods is not quite under- stood yet. Because of the amino acid composition of the animal foods in this work were not known, we have no background to explain the distribution of the sensitivity of animal pro— tein foods to radiation in relation to their amino acid composition. 50 Approach for Determination of "Threshold Dose" for Each Animal Protein Food The term "threshold dose" occurs in the literature. This is considered to be the dose at which an irradiated flavor can be just detected. The data obtained in this study indicate a correlation of flavor intensity with dose. The intensity score scale of one to five which was used is arbitrary and not necessarily linear. There are some dif- ficulties of interpretation of the data in defining a threshold dose for each food. Since there is interest, both in the threshold dose level itself for a given food and in the comparative value of threshold doses, there may be justification in the following effort. The score value of "one indicates absence of any irradiated flavor or odor. The score value of "two" indi— cates a level of irradiated flavor or odor just detected. One may consider a score just less than two to be the upper limit of absence of irradiated flavor. Above this value a flavor or odor is detectable (by definition). By plotting for each food flavor intensity scores vs. dose, and noting the dose corresponding to a 2.0 score, one can assign a dose value as the threshold dose. This is illus- trated in Figure 4 for two meats, pork and bear. These estimates are more area estimates than precise point esti— mates since the volume of the data did not justify deter- mining precise response curves. Appropriate response curves were drawn and threshold values made from these curves. 51 .N no muoom NuwmcmucH uo>mHu um owcHEumumo mm; amoo oHonmoucB .mmoo coHumHomuuH MN muoom uo>oHu pmumemuuH .v muomHm omux .mmoo coHumHtmuuH 83 I 83 . 8m 2: o k . P n d . 1 H _ n u _ u u _ _ u . u . . _ _ . I1 _ _ 1'. ........... Av IIIIIIIII -.d21IIr~ d d In A A? ion 4 ummm ... 1 axons zoneta pausrpezzx 52 Hence the threshold values given in Table 8 should be recognized as only approximate. The threshold levels so determined for all protein foods investigated are listed in Table 8. The data shown in Table 8 show the threshold dose range from 150 Krad (for turkey) to 875 Krad (for bear). In overall picture, the threshold dose is higher for the less sensitive animal food. But the correlation between threshold dose and the sensitivity is not necessarily linear, since the threshold doses were derived from plotting the score values of irradiated flavor of lower doses only (0 to 2000 Krad), and the sensitivity was determined from all dose treatments from 0 to 5000 Krad. In 1962 it was reported that no off odor could be detected in beef up to 93 Krad but at 470 Krad it could be detected readily (Anonymous, 1962). Hannan and Thornley (1957) suggested the threshold dose of beef was about 100 Krad. The figure in this present work (Table 8) of 250 Krad for threshold dose for beef probably is greatly dif- ferent from the dose range of the previous investigators. The deviation probably is due to sampling methods, statis- tical or cooking methods especially in picking the score value of taste threshold. For lean pork Hannan and Thornley (1957) reported 250 Krad as its threshold dose. That figure apparently is fairly close to the figure of 175 Krad obtained in this study. 53 TABLE 8.—-The "threshold dose" for each animal protein food investigated, determined at flavor intensity score value of 2 (slight irradiation flavor). Animal Food Threshold Dose Krad Pork 175 Beef 250 Turkey 150 Chicken 250 Trout 450 Rabbit 350 Shrimp 250 Lobster 250 Frog 400 Whale 400 Bear 875 Halibut 500 Venison 625 Lamb 625 Turtle 450 Horse 650 Beaver 550 Opossum 500 Hippopotamus 525 Elephant 650 54 Coleby (1959) and Thornley (1957) both reported the same threshold dose of 250 Krad for chicken cooked by steam- ing, the same as obtained in this present work. In 1967 Holston, ep_al. reported the optimum dose range for irradiation of fish and shellfish is 150-450 Krad. The data in Table 8 of this present study indicate a threshold dose range of 250-500 Krad for fish and sea- foods (including whale, sea turtle and frog). No data were available for other animal foods for comparison with this present study of threshold dose of irradiated flavor in animal protein foods. Clearly more work is needed to obtain an accurate scientific explanation of the irradiated flavor and odor develOpment in animal protein foods. SUMMARY AND CONCLUS IONS The effect Of gamma irradiation on changes of flavor in twenty selected animal foods was studied. The animals selected as the source of protein foods were beef, deer, lamb, swine, hippopotamus, bear, elephant, rabbit, whale, beaver, Opossum, horse, chicken, turkey, sea turtle, frog, halibut, trout, shrimp and lobster. In general, the most intensively domesticated ani- mals such as swine, beef, turkey and chicken yield foods that are the most sensitive to irradiation. Less domesti- cated or unusual animals such as elephant, hippopotamus and Opossum give foods that are among the least sensitive of all foods observed. The importance of the contribution of animal fat to irradiated flavor development seems doubtful. Opossum and beaver meats were among the fattest, but were among the least sensitive to irradiation. The "threshold dose" for each food was determined from the dose corresponding to a 2.0 flavor intensity score. As is to be expected, the threshold dose is higher for the less sensitive animal food. 55 LIST OF REFERENCES LIST OF REFERENCES Alexander, P., Fox, M., Stacen, K.A., Rosen, D. (1956). Direct and Indirect Effects Of Ionizing Radiation in Proteins. Nature 178:846. Anonymous (1962). The Effect of Sub-sterilizing Doses of Cathode Rays and Gamma Rays on the Keeping Quali- ties of Pork. Research and Engineering Command, Quartermaster Corps, U.S. Army, Chicago. Anonymous (1962). Merck Technical Bulletin: An Intro— duction to Taste Testing of Foods. Merck & CO., Inc. Anonymous (1964). Sensory Testing Guide for Panel Evalu- . ation of Foods and Beverages. Prepared by the Committee on Sensory Evaluation of the Institute of Food Technologists. Batzer, O.F., Sliwinski, R.A., Chang, L., Pih, K., Fox, J.B. Jr., Doty, D.M., Pearson, A.M., Spooner, M.E. (1959). Some Factors Influencing Radiation Induced Chemical Changes in Raw Beef. Food Tech— nology 9:501-508. Champagne, J.R. (1967). Investigation of Chemical Changes Occurring in Irradiated Animal Fats. Ph.D. Thesis, Department of Food Science and Technology, Univer- sity of Massachusetts, Amherst, Mass. Champagne, J.R., Nawar, W.W. (1969). The Volatile Compo- nents of Irradiated Beef and Pork Fats. J. Food Sc. 34:335-339. Coleby, B. (1959). The Effect of Irradiation on the Qual- ity Of Meat and Poultry. International Journal of Applied Radiation and Isotopes 6:115-121. Connor, T.J., Steinberg, M.A. (1966). Organoleptic Studies on Radiation Pasteurized Skinless Haddock Fillets. Food Technology 10:1357-1359. 56 57 Drake, M.P., Giffee, J.W. (1957a). Action of Ionizing Radiations on Amino Acids, Protein and Nucleopro— teins. In Radiation Preservation of Foods, U.S. Dept. of Commerce, Washington, D.C. Drake, M.P., Giffee, J.W., Johnson, D.J., and Koenig, V.L. (1957b). Chemical Effects of Ionizing Radiation on Protein. I. Effect of Gamma-radiation on the Amino Acid Content of Insulin. J. Amer. Chem. Soc. 79:1395. Frigerio, N. (1967). Your Body and Radiation. USAEC. Graikoski, J.T., Kazanas, N., Watz, J., DuCharme, S., Emerson, J.A., Seagran, H.L. (1967a). Irradiation Preservation of Freshwater Fish. Bureau of Com- mercial Fisheries, Ann Arbor, Mich. Graikoski, J.T., Kazanas, N., Watz, J., DuCharme, 8., Billy, T.J., Emerson, J.A., Seagran, H.L. (1967b). Irradiation Preservation of Freshwater Fish. Dept. of Commerce, Washington, D.C. Gray, P. (editor) (1965). The Encyclopedia Of the Biolog- ical Sciences. Reinhold Publishing Corporation, New York. Groninger, H.S., Tappel, A.L., Knapp, F.W. (1956). Some Chemical and Organoleptic Changes in Gamma Irradi- ated Meats. Food Research 5:555—564. Hannan, R.S., Thornley, M.J. (1957). Radiation Processing of Foods. A Commentary on Present Research. Food Manufacture 12:559-562. Hannan, R.S., Shepherd, H.J. (1959). The Treatment of Meats with Ionizing Radiation. 1. Changes in Odour, Flavour and Appearance of Chicken Meat. J. Sci. Food Agric. 10:288-293. Henry, M.C. (1957). Action of Ionizing Radiation on Hydro- carbons. In Radiation Preservation of Foods. The U.S. Army Quartermaster Corps, Washington, D.C. Hirsh, N.L. (1970). Attempts at Quantitating Flavor Dif- ferences. Food Product Development, Apri1:22-24. Holston, J.A., Ronsivalli, L.J., Ampola, V.G., King, F.J. (1967). Study of Irradiated-Pasteurized Fishery Products. Bureau of Commercial Fisheries, Tech- nological Laboratory, Gloucester, Mass. 58 Huber, W., Brasch, A., Waly, A. (1953). Effect Of Process— ing Conditions on Organoleptic Changes in Foodstuffs Sterilized with High Intensity Electrons. Food Technology 3:109-115. Kramer, A. (1966). Sensory Evaluation of Food Flavor. In Flavor Chemistry, Advances in Chemistry Series 56, American Chemical Society, Washington, D.C. Long, L., Lirot, S.J. (1957). Action of Ionizing Radiation on Carbohydrates and Polysaccharides. Radiation Preservation of Food. The U.S. Army Quartermaster Corps, Washington, D.C. Mehrlich, F.P. (1966). The United States Army Food Irra- diation Programme. Proceeding in Symposium in Karlruhe, I.A.E.C. Merritt, C. Jr., Angelini, P., Bazinet, M.L., McAdoo, D.J. (1966). Irradiation Damage in Lipids. Flavor Chemistry, Advances in Chemistry Series 56, Ameri- can Chemical Society, Washington, D.C. Merritt, C. Jr., Angelini, P., McAdOO, D.J. (1967). Vola- tile Compounds Induced by Irradiation in Basic Food Substances. Radiation Preservation Of Food. Advances in Chemistry Series 65, American Chemical Society, Washington, D.C. Mitchell, H.J. (1957). Action Of Ionizing Radiation on Fats, Oils and Related Compounds. Radiation Preservation of Food. The U.S. Army Quartermaster Corps, Washington, D.C. Miyauchi, D., Teeny, F., Pelroy, G. (1968). Application of Radiation-Pasteurization Process to Pacific Coast Fishery Products. Bureau of Commercial Fisheries, Technological Laboratory, Seattle, Wash. Novak, A.P., Liuzzo, J.A. (1967). Radiation Pasteurization of Gulf Shellfish. Louisiana State University, Baton Rouge. Novak, A.P., Liuzzo, J.A. (1968). Radiation Preservation of Gulf Shellfish. Louisiana State University, Baton Rouge. ROhlf, F.J., Sokal, R.R. (1969). Statistical Tables. W.H. Freeman and Company, San Francisco. 59 Ronsivalli, L.J., King, F.J., Mendelsohn, J.M., Gadbois, D.F., Brook, R.O., Holston, J.A. (1967). Chemistry of Radio—Pasteurized Seafoods. Bureau of Commercial Fisheries, Gloucester, Mass. Schweigert, B.S. (1959). The Effect of Radiation on Pro- teins. International Journal of Applied Radiation and Isotopes 6:76-77. Siu, R.G.H., Bailey, S.D. (1957). Radiolysis of Water. In Radiation Preservation Of Food. The U.S. Army Quartermaster Corps, Washington, D.C. Steel, R.G.D., Torrie, J.H. (1960). Principles and Proce- dures of Statistics. McGraw-Hill Book Company, Inc., New York, Toronto, London. Thomson, Sir A. L. (editor) (1964). A New Dictionary of Birds. The British Ornithologists Union. Thornley, M.J. (1957). Preservation of Chicken Meat: Observation on the Use of Cathode and Gamma Rays. International Institute of Refrigeration. Urbain, W.M. (1965). The Effect Of Ionizing Radiation on Fresh Meats and Poultry. Proceedings of an Inter- national Conference, Boston, Mass. Publication 1273 National Academy of Sciences--National Research Council, Washington, D.C. Urbain, W.M. (1966). Food Irradiation, Nuclear News, July: 34—35. Urbain, W.M. (1971). Food Irradiation . . . Benefits and Limitation. Proceedings of a Panel to Consider Technological Factors Involved in the Economical Application Of Food Irradiation. International Atomic Energy Agency, Vienna. In press. Volkova, M.S., Pasynskii, A.G. (1955). The Effect of Ultra— violet and Roentgen Irradiation on Solution of Pro- tein. Biokhimiya 20:470. In Chem. Abstr. 50:2704 (1956) . Walker, E.P. (editor) (1968). Mammals Of the World. The Johns Hopkins Press, Baltimore. Wasserman, A.E. (1971). Thermally Produced Components in Meat and Poultry Flavor. l62nd ACS National Meet- ing, Washington, D.C. 60 Waters, M.E., Thompson, M.H., Love, T.D., Farragut, R.N., Thompson, H.C. (1969). Development of Radiation- Sterilized Fish Product. U.S. Army Natick Labora- tories. Wick, E.L., Murray, E., Mizutani, J., Koshika, M. (1967). Radiation Preservation of Food. Advances in Chem- istry Series 65, American Chemical Society, Washington, D.C. (WW“Ml/(INNll (A I I. (I I I! I! I! IA l! I! I I I! I I. 3 6 6 2 3 0 3 9 2 3 I’M