UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Quality enhancement of canned late-run chum salmon (Oncorhynchus keta) Collins, Lindley Simeon 1989

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

Item Metadata

Download

Media
831-UBC_1990_A6_7 C64.pdf [ 5.86MB ]
Metadata
JSON: 831-1.0098099.json
JSON-LD: 831-1.0098099-ld.json
RDF/XML (Pretty): 831-1.0098099-rdf.xml
RDF/JSON: 831-1.0098099-rdf.json
Turtle: 831-1.0098099-turtle.txt
N-Triples: 831-1.0098099-rdf-ntriples.txt
Original Record: 831-1.0098099-source.json
Full Text
831-1.0098099-fulltext.txt
Citation
831-1.0098099.ris

Full Text

QUALITY ENHANCEMENT OF CAMMED LATE-RUN CHUM SALMON {ONCORHYNCHUS KETA) by LINDLEY SIMEON COLLINS Licienciate degree (Food Science), University of Havana, 1985 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF FOOD SCIENCE We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA November, 1989 (c) LINDLEY SIMEON COLLINS, 1989 ^ 0 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Food Science The University of British Columbia Vancouver, Canada D a t e 22/11/89 DE-6 (2/88) ABSTRACT In this study, a number of experiments were undertaken to investigate possible methods for effective improvement of the texture and flavour of canned late-run chum salmon. These included removal of the skin and bones from the f i s h , processing of the boneless-skinless steaks in r e t o r t pouches, brine treatment using two washes with an 8% s a l t solution for one hour each time, and a precanning treatment in which the boneless-skinless steaks were soaked in a solution of 10% tripolyphosphate and 2% brine for two minutes. Only fi s h of advanced sexual maturity were used. The canned salmon was steam processed at 120°C for 6 5 minutes in an FMC laboratory r e t o r t . This was based on a known commercial process for 307 x 115 cans. Heat penetration studies were carried out to design the process schedules for the pouched samples. It was found that the pouched product required 4 8% less thermal processing time than the canned product to achieve similar l e t h a l i t y . Sensory re s u l t s showed that the removal of the skin and bones did not produce any significant improvement in the flavour and acceptability of the fish. There was no significant difference between the polyphosphate/brine samples and the untreated (control) samples for a l l attributes tested. The brine i i treatment also did not improve the texture of the samples. However, there was less detection of late-run flavour in the brine treated samples when compared to the control. Comments offered by panelists described these samples as having a salty/briny flavour. Pouched samples had a firmer, drier and more fibrous texture than the canned product. They also scored better in terms of late-run flavour. Acceptance of the fi s h however was only moderate. As a consequence, although this study demonstrated an improvement in the texture and flavour of the pouched late-run chum in comparison to the canned product, i t was concluded that a more acceptable pouched product could probably be obtained by using late-run salmon of less advanced sexual maturity. Results of linear regression analysis showed that significant relationships were obtained between sensory firmness, fibrousness and chewiness and instrumental hardness, maximum slope and chewiness. However, none of the sensory parameters were well predicted by the instrumental results. i i i TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES vi LIST OF FIGURES x LIST OF APPENDICES xii ACKNOWLEDGEMENT xiii INTRODUCTION 1 LITERATURE REVIEW 3 A. HISTORICAL BACKGROUND 3 (i) Cans 3 (ii) Retort pouches 4 B. THERMAL PROCESSING IN CANS AND RETORT POUCHES 6 (i) Heating media 6 (ii) Heat penetration tests 7 (iii) Methods of data analysis 9 C. QUALITY OF THERMAL PROCESS PRODUCT 11 (i) Retention of quality attributes 11 (ii) Quality of canned late-run chum salmon 12 (iii) Quality of pouched product 14 (iv) Quality of canned fish treated with salts 16 D. MEASUREMENT OF TEXTURAL PROPERTIES 18 MATERIALS AND METHODS 24 A. FISH SAMPLE PREPARATION AND SOURCE 24 B. PROCESS DETERMINATION BY HEAT PENETRATION 24 (i) Retort system used 27 (ii) Processing conditions 28 (iii) Heat penetration 30 (iv) Process calculations 32 C. PROCESSING FOR INSTRUMENTAL AND SENSORY ANALYSIS 33 D. SAMPLING AND PRODUCT ANALYSIS 38 (i) Sampling procedure 38 (ii) Sensory Analysis 40 (a) Selection, training and establishing rating scales 40 Texture 40 Flavour 4 4 (b) Sensory testing 44 (iii) Instrumental technique 47 (iv) Data Analysis 49 RESULTS AND DISCUSSION 50 iv A. PROCESS DETERMINATION 50 B. SENSORY DATA 56 C. INSTRUMENTAL ANALYSIS 69 (i) General Observations 69 (ii) Instrumental measurement of texture 77 D. SUBJECTIVE-INSTRUMENTAL INTERRELATIONS 97 E. VOLUME OF COOK-OUT LIQUID 100 CONCLUSIONS 102 bibliography 10 5 APPENDICES 112 v LIST OF TABLES Table 1 Classification of textural characteristics, based on the General Foods Texture Profile Method 23 Table 2 Maturity of chum salmon 25 Table 3 Processing conditions for cans and retort pouches used during process determination work 29 Table 4 Definitions used in sensory profiling of salmon 41 Table 5 Scales used in sensory analysis 43 Table 6 Test conditions at which the instron was operated during instrumental analysis of chum salmon 48 Table 7 Heating and cooling parameters for chum salmon obtained during process determination work 53 Table 8 Process times used in the product analysis phase of the work and their corresponding lethalities calculated by Stumbo's method 55 Table 9 Mean panel ratings for sensory attributes of the boneless-skinless salmon steaks and the steaks with skin and bone processed at 248°F 57 Table 10 Mean panel ratings for sensory attributes of brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F 58 Table 11 Analysis of variance data for sensory studies of the boneless-skinless steaks and the steaks with skin and bone processed at 248°F 59 Table 12 vi Analysis of variance data for sensory studies of brine, polyphosphate/brine, pouch and untreated samples of chum salmon processed at 248°F 60 Table 13 Multiple comparison hypothesis test comparing sensory firmness of the control, brine, polyphosphate/brine and pouched salmon processed at 248°F 62 Table 14 Multiple comparison hypothesis test comparing sensory dryness of the control, brine, polyphosphate/brine and pouched salmon processed at 248°F 63 Table 15 Multiple comparison hypothesis test comparing sensory fibrousness of the control, brine, polyphosphate/brine and pouched salmon processed at 248°F 64 Table 16 Multiple comparison hypothesis test comparing sensory chewiness of the control, brine, polyphosphate/brine and pouched samples of salmon processed at 248°F 65 Table 17 Multiple comparison hypothesis test comparing late-run flavour of the control, brine, polyphosphate/brine and pouched salmon processed at 248°F 66 Table 18 Multiple comparison hypothesis test comparing overall acceptability of the control, brine, polyphosphate and pouched salmon processed at 248°F 67 Table 19 Means of results obtained by instrumental analysis and their corresponding standard deviations for boneless-skinless salmon steaks and steaks with skin and bones processed at 248°F 80 Table 20 Means of results obtained by instrumental analysis and their corresponding standard deviations for the brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F 81 Table 21 Analysis of variance data derived from texture profile vii analysis results of boneless-skinless steaks and steaks with skin and bones processed at 248°F 82 Table 22 Analysis of variance data for texture profile analysis of brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F 83 Table 23 Multiple comparison hypothesis test comparing instrumental results of hardness 1 for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon 8 4 Table 24 Multiple comparison hypothesis test comparing instrumental results for hardness 2 of the brine, polyphosphate/brine, pouch and control (untreated) samples of the processed chum salmon 85 Table 25 Multiple comparison hypothesis test comparing instrumental results for maximum slope 1 of the brine, polyphate/brine, pouch and untreated (control) samples of the processed chum salmon . 86 Table 26 Multiple comparison hypothesis test comparing instrumental results for maximum slope 2 of the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon 87 Table 27 Multiple comparison hypothesis test comparing instrumental results for the gumminess of the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon 88 Table 28 Multiple comparison hypothesis test comparing instrumental results for the chewiness of the brine, polyphosphate/brine, pouch and control (untreated) samples of the processed chum salmon 89 Table 29 Coefficient of determination (R2) between panel scores and instrumental texture scores 98 viii Table 30 Volume of cook-out liquid obtained following processing of brine, polyphosphate/brine, pouch and control samples of chum salmon at 24 8°F 101 ix LIST OF FIGURES Figure 1 External appearance of Grade 3 late-run chum salmon 34 Figure 2 Flow diagram of the procedure used for the salmon canning process in Experiment 1 36 Figure 3 External appearance of the Grade 4 late-run chum salmon 37 Figure 4 Flow diagram of the procedure used for the salmon canning process in Experiments 2, 3 and 4 39 Figure 5 Typical heating and cooling curve obtained for canned salmon processed at 248°F 51 Figure 6 Typical heating and cooling curve obtained for the pouched salmon samples 52 Figure 7 Bar diagram showing mean panel ratings for firmness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 70 Figure 8 Bar diagram showing mean panel ratings for dryness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 71 Figure 9 Bar diagram showing mean panel ratings for fibrousness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 72 Figure 10 Bar diagram showing mean panel ratings for chewiness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 73 Figure 11 Bar diagram showing mean panel ratings for late-run flavour of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 74 x Figure 12 Bar diagram showing mean panel ratings for overall acceptability of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 75 Figure 13 Typical Force-Time curves of the processed Flaked salmon samples 78 Figure 14 Bar diagram showing instrumental measurement of hardness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 91 Figure 15 Bar diagram showing instrumental measurement of maximum slopes for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 92 Figure 16 Bar diagram showing instrumental measurement of cohesiveness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 93 Figure 17 Bar diagram showing instrumental measurement of springiness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 94 Figure 18 Bar diagram showing instrumental measurement of gumminess for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 95 Figure 19 Bar diagram showing instrumental measurement of chewiness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon 96 xi LIST OF APPENDICES Page Appendix 1 Definitons of terms and symbols 112 Appendix 11 Estimated process lethality calculations for cans 113 Appendix 111 Estimated process time calculations for pouches 114 Appendix IV Estimated process lethality calculations for pouches 115 Appendix V Questionnaire for sensory analysis 116 xii ACKNOWLEDGEMENT I will like to express my sincere gratitude to all those who have been instrumental in helping me complete this thesis. Special mention must be made to Dr. T. Durance under whose supervision this work was done. His patient assistance, advice and suggestions throughout all the stages of the work are greatly appreciated. I am also grateful to the members of my supervisory committee, Dr. W. Powrie, Dr. B. Skura and Dr. D. Kitts, for their helpful comments and for reviewing this manuscript. A further word of thanks goes to Ms. L. Robinson for her advice on the operation of the retorts and to the Canadian Commonwealth Scholarship and Fellowship Plan for the financial assistance provided throughout my study. xiii INTRODUCTION Chum salmon (.Oncorhynchus keta) belongs to the Pacific salmon family, which is a very important seafood not only in British Columbia but in many countries of the northern hemisphere. Nutritionally, they are rich in proteins and vitamin D and are an exceptionally rich source of many of the minerals and nutrients necessary to life. Commercially, the chum is a low value item of reduced consumer acceptance because of its poor quality. This is especially true in the latter part of their spawning migration when their flesh becomes soft and mushy and a late-run odour and flavour may develop together with an undesirable dark grey colour. Such undesirable changes have been associated with reduced protein and fat content, increased water content (Howgate, 197 7), autoxidation of lipids (Josephson et al., 1984), presence of odorous constituents in the skin (Huynh, 1988) and a drop in the muscle pigment responsible for the red colour (Huynh, 1988). Overprocessing (Sikorski et al., 1984) and poor handling conditions may also contribute to poor quality. These changes are recognized as contributing to a major economic loss in the value of this important commercial product in British Columbia. Consequently, this study was undertaken to develop processing methods or conditions which would improve the texture and flavour of canned late-run chum. Studies have been done on new product development 1 utilizing chum salmon flesh as the main ingredient e.g. a canned smoked product using artif i c i a l liquid smoke was prepared (Hyunh, 1989). Nevertheless late-run chum salmon is s t i l l mainly available as a traditional canned product despite its poor acceptance. Very l i t t l e work, on improving the quality of the raw material through changes in process parameters, has been reported in the literature. It was therefore the purpose of this investigation to study the effects of process parameters on the texture and flavour of canned late-run chum salmon. The following process changes were investigated: (i) processing of boneless-skinless steaks. This was aimed at removing flavour and odour precursors associated with the skin of the fish. In addition, i t was hoped that the removal of the skin and bone would result in a more homogeneous product for which texture might be more precisely measured. (ii) processing of steaks in retort pouches. This would allow a reduced cook which was expected to result in a firmer texture and possibly better flavour. (iii) addition of salts to the boneless-skinless steaks. It was hoped that this would alter the water holding capacity and, through mediation of the Malliard reaction, might alter flavour. 2 LITERATURE REVIEW A. HISTORICAL BACKGROUND (i) Cans The f i r s t major stage i n the development of thermal processing of food i n cans was the s c i e n t i f i c d i s c o v e r i e s of Appert i n the e a r l y 19th century. At t h a t time the process was not q u i t e understood (Goldblith, 1971). L a t e r , P a s t e u r , between 1857 to 1862, provided an explanation f o r Appert's accomplishment i.e., t h a t Appert's heat process had k i l l e d the microrganisms present, while the hermetic c o n t a i n e r prevented p o s t p r o c e s s i n g contamination (Pflug, 1987). F u r t h e r c o n t r i b u t i o n s by Underwood and P r e s c o t t i n the l a t e 19th ce n t u r y e s t a b l i s h e d the r e l a t i o n s h i p between thermophilic b a c t e r i a and the spoilage of foods. They showed t h a t b a c t e r i a were responsible f o r spoilage and heating t o the temperature g r e a t e r than the boilin g point was n e c e s s a r y to achieve s t e r i l i z a t i o n (Goldblith, 1972). Around 186 4 the f i r s t salmon cannery was e s t a b l i s h e d by Hume and Hapgood. Although there was nothing c e r t a i n about r e s u l t s i n t h i s cannery and cans o f t e n exploded, Hume and Hapgood were able t o r e a l i z e t h a t the temperature of boiling water was not s u f f i c i e n t to p r e s e r v e f i s h (Crandall, 1946). Today, the p r i n c i p l e used by the e a r l y canners i s s t i l l used with the 3 help of modern equipment. Machinery has been developed to replace operations that were once done by hand. In addition, major advancements in thermobacteriology by Bigelow and co-workers (1920), and process technology by Ball (1923) and Stumbo (1971), among others, have led to s c i e n t i f i c a l l y based methods for calculation of minimum safe s t e r i l i z a t i o n processes for canned food s t e r i l i z a t i o n . (ii) Retort pouches Development of the r e t o r t pouch concept began in the United States in the early 1950's. Most of the work was done by the U.S. Army Natick Development Center, Reynolds Metal Co. and Continental Group Inc. (Lampi, 1977; Mermelstein, 1978). For example, in 1968 a r e l i a b i l i t y study of the pouch was undertaken by the U. S. Natick Development Center. In comparing seal integrity, s t e r i l i t y and overall defects i t was found that r e t o r t pouches could be produced that were of equal quality to metal cans. In 1974, the U. S. Department of Agriculture (USDA) gave approval for use of r e t o r t pouches with a l l meat products. In 197 5, the U. S. Food and Drug Administration asked the USDA to withdraw this approval because FDA was concerned that the polyester and epoxy components of the laminating adhesive used in the pouch could migrate into the food (Lopez, 1987). 4 In 1977, the FDA issued a regulation for high temperature laminates which specified the materials acceptable for manufacture of the pouch and set extraction limits for these materials. Since then work on the retort pouches have progressed from exploratory investigations to reliability studies required to make the pouch a commercial reality (Lopez, 1987). The structure of the retort pouch in general use today consists of three layers starting from the outside: polyester, aluminium fo i l and an inner layer of polypropylene or polyolefin. The outer layer of polyester film provides printability and strength. Aluminium foil serves as a moisture, light and gas barrier. The inner layer is heat sealable and provides an inert food-contact surface (Mermelstein, 1978; Heintz, 1980). Among the advantages of the retort pouch over the cylindrical can is that its thin profile provides greater surface area. This permits more rapid heat transfer and thus reduces process times. With the reduced heat exposure, improved food quality, both sensory and nutritional, is expected as compared to similar products packaged in conventional cans. In addition, the pouch can be easily torn or cut open and both empty and full retort pouches take less space and weigh less than metal cans. The individual container cost to the canner is less than the conventional rigid containers presently in use. Also, preparation of products which need to be heated to serving 5 temperature can be accomplished in three to five minutes by immersing the pouch in boiling water, or placing the plastic container in a microwave oven (Williams et al., 198 2; Lopez, 1987). The main disadvantage is that filling and sealing equipment for the pouches operate at relatively slow speed (Williams et al., 1983). There is also a need for proper inspection of the pouch seals; this requires special equipment which are not presently available. These factors, together with the need for new filling and closing equipment have slowed introduction of the retort pouch on a large scale (Davis, 1981). Nevertheless, in Japan and Europe a variety of low acid foods has been succcessfully sterilized in retort pouches for more than 20 years (Lampi, 1980). In recent years several companies in the United States and Canada have introduced foods packed in retort pouches to retail markets (Heintz, 1980). B. THERMAL PROCESSING IN CANS AND RETORT POUCHES (i) Heating media Steam, steam/air mixtures and water are the potential conventional heating media available for thermal processing in cans and retort pouches. Salmon in cans are traditionally processed in steam. Pure steam has the advantage of providing a high heat transfer rate to the container with rapid retort 6 response (Wilson, 1980). Retort pouches are generally processed in steam/air mixtures or water immersion/overpressure systems in order to provide overriding air pressure. Pouch seals are weakened by heating and are unable to withstand the same pressure differential that may be sustained by metal cans (Lampi, 1977). Both temperature and pressure of the media must be controlled when processing with steam/air mixtures because heat transfer rate decreases with increasing air content of the mixture (Ramaswamy, 1983). With water immersion/overpressure systems, both medium temperature and pressure must also be controlled. With water, heat transfer rates are lower than for pure steam and good circulation must be maintained to prevent temperature stratification (Lampi, 1977). According to Pflug et al. (1963) if conventional retorts are used for the water process extra time is required for heating water at the start of the cook. In a FMC retort, however, process water can be preheated in a separate reservoir, thus reducing come up time (Lampi, 1977). Processing pouches with water also has the disadvantage in that i t may soil pouches and cause scale build up on separation plates (Pflug, 1963). (ii) Heat penetration tests To know whether or not a method of sterilizing a packaged 7 food by application of heat will be successful and safe, the processor must f i r s t conduct a heat penetration test on that food and then use the data obtained to calculate process time (Herndon et al., 1968). For this, temperature measurements are made at the slowest heating point (coldspot) in the filled container using thermocouples. In cans, Ecklund Type T non-projecting thermocouples are the main type used to obtain heat penetration data. Appropriate tests should be made to determine the effect of errors introduced by conduction of heat along the thermocouple wires and fittings required to hold the thermocouple in place (Lopez, 1987). Ecklund in 1956 published these correction factors for 307 x 409 and smaller cans for his thermocouples. Several methods have been proposed for the insertion of thermocouple leads into pouches. The thermocouple wires can be inserted through a small hole in the pouch wall which is then sealed with a silicone sealant. This technique requires overnight drying and curing and the seal is susceptible to leaks (Spinak and Wiley, 198 2). A special packing gland which has a compression fixture to achieve a seal around the thermocouple can also be used (Pflug et al., 196 3). This gland may however interfere with the normal geometry of the pouch (Berry, 1979). Receptacles obtained from 0. F. Ecklund Inc., which allow the positioning of a 8 rigid thermocouple in a manner similar to those used in cans can also be employed. There i s l i t t l e information available to compare the various methods and therefore the choice of method used depends on the experiences of the investigator. (iii) Methods of data analysis The res u l t s of a heat penetration t e s t are experimentally derived heating and cooling parameters (Lopez, 198 7). Various calculation methods have been developed for process time determinations using these parameters. The General Method or Accumulative Lethality Method, in which the time-temperature data from a heat penetration t e s t i s analyzed for determining process l e t h a l i t y , i s the most accurate method possible (Stumbo, 1973). However, this method provides l i t t l e f l e x i b i l i t y in allowing mathematical determination of process changes when variations in conditions occur. In contrast, the formula methods allow increased f l e x i b i l i t y and reduce the amount of calculation required to determine process time. Ball in 1923 developed a formula method which i s s t i l l one of the most commonly used method in the food industry for process calculation (Spinak and Wiley, 1982). This method of thermal process time calculations has, however, drawn criticisms from researchers working on the subject. For example, Ball assumed that the cooling index (f =) equals the heating index (fn). 9 This usually results in underestimates in approximating lethality accumulated during the cooling period of the process (Spinak and Wiley, 1 9 8 2 ) . A second assumption by Ball is that the cooling lag factor ( j c r c ) is equal to 1 . 4 1 . Hayakawa ( 1 9 7 8 ) commented that this assumption overestimates the sterilizing value of heat process when j « , < 1 . 4 1 and underestimates i t when j e e > 1 . 4 1 . Stumbo and Longley ( 1 9 6 6 ) published tables for process evaluations taking into account the variability of j c== values. The values in these tables were obtained through planimeter measurements of hand-drawn temperature histories plotted on lethal rate paper, and subsequent interpolation of graphs. Revised tables were developed through use of computer integration of thermal histories generated from heat transfer equations, using finite difference simulations. The assumption of f w = f c , however, remained. Tung and Garland ( 1 9 7 8 ) use Stumbo's method in calculation and mentioned that when fc=>fn the method will provide an additional small margin of safety in the estimated process lethality and this margin of safety will be reversed when fc-<f*. Nevertheless, Smith and Tung ( 1 9 8 2 ) found that Stumbo's method gave the best estimates of process lethality when compared with other formula methods. 10 C. QUALITY OF THERMAL PROCESS PRODUCT (i) Retention of quality a t t r i b u t e s Food i s thermally t r e a t e d to eliminate or reduce the a c t i v i t y of microorganisms and/or enzymes t h a t would, upon st o r a g e , r e s u l t i n d e t e r i o r a t i o n of the food or would endanger the h e a l t h of the consumer. However, together with the d e s t r u c t i o n of microorganisms and enzymes, degradation of sensory and n u t r i t i o n a l p r o p e r t i e s of the food occurs (Lund, 1977). Among the q u a l i t y a t t r i b u t e s t h a t are e f f e c t e d are the t e x t u r e and f l a v o u r . In the case of f i s h , although i t has been s t a t e d t h a t the a c c e p t a b i l i t y of cooked f i s h i s more dependent on i t s f l a v o u r than on i t s t e x t u r e (Rasekh e t a l . , 1970; Connell and Howgate, 1971; Howgate, 1977), t e x t u r e has a l s o been shown to be an important determinant of p r e f e r e n c e , e s p e c i a l l y f o r f i s h with mild f l a v o u r c h a r a c t e r (Wesson e t a l . , 1979). I t i s t h e r e f o r e the d e s i r e of the thermal p r o c e s s o r to provide the consumer with a s a f e product, while a t the same time exhibiting good r e t e n t i o n of q u a l i t y a t t r i b u t e s such as t e x t u r e and f l a v o u r (Teixeria e t a l . , 1975; Lund, 1977). P r o c e s s e s which are equivalent in terms of accomplished l e t h a l i t y do not n e c e s s a r i l y r e s u l t i n the same r e t e n t i o n of q u a l i t y a t t r i b u t e s (Lund, 1977). In addit i o n , the temperature s e n s i t i v i t i e s of thermal d e s t r u c t i o n r a t e s f o r q u a l i t y f a c t o r s are l e s s than f o r spores as ind i c a t e d 11 by greater Z values for quality factors (Lund, 1977). As a consequence, i t is possible to improve the quality of a thermally processed food by taking advantage of this fact and by the use of container geometry which allows more rapid heat penetration compared to conventional cans, while at the same time ensuring a safe product (Teixeira et al, 1975). (ii) Quality of canned late-run chum salmon It is well accepted that the quality of canned fish is dependent on the quality of the raw material and an acceptable product can be obtained only if good handling practices are observed during processing and freezing. However, in the case of late-run chum salmon, the canned product can be of poor quality due to a number of other factors. For instance, Love (1980) explained that the energy that the salmon use to travel back to their spawning grounds is obtained by drawing from the protein of their white muscle tissues, as well as lipids in their liver and muscle. In fatty fish such as salmon, this reduction in lipid is accompanied by an increase in water content of the muscle. As a result, late in their spawning migration, the chum are in poor condition. The breakdown of the protein and the reduction of lipid content in the fish prior harvest cause the cooked flesh to become soft and sloppy with a poor mouthfeel (Howgate, 1977). In addition, characteristic late-run odour and flavour may 12 develop i n t h e c a n n e < j p r o d u c t . Josephson et al. (198 4) found the eight and n i n e carbon alcohol and carbonyl volatiles derived from autoxidation of lipids to be associated with the development of this odour. Huynh et al. (1988) found a detectable amount of putrescine and caderverine in the skin of late-run samples. These diamines (putrescine and caderverine) are produced by the decarboxylation of the free amino acids within the fish tissues and are odorous constituents (Staruszkiewez and Bond, 1981). The quality of the fish product can also be affected by the heat transfer rate at the can surface. Connell (1980) found that for a steam/air mixture the heat transfer rate is less efficient than using pure steam under pressure and this can affect the quality of the final product. Van Den Broeck (1965) suggested several possible methods for increasing heat penetration during processing to get a better quality product. The form and size of the can should be designed so that at least one of the dimensions is as small as possible. Another method is to accelerate the cooling process by dipping or spraying the cans with cold potable water. Quality can also be affected by processing time. The retort time used must be sufficient not only to commercially sterilize the product, but also to ensure proper tenderization of the muscular and bony parts of the fish (Sikorski et al., 198 4). Because of the need to soften the bones, canned fish is often 13 heated at a far greater temperature and a much longer time than is required for sterilization and this can impair the eating quality of the product. A higher quality may therefore be achieved if a shorter processing time is taken advantage of whilst s t i l l ensuring that a safe margin of safety is maintained. The research efforts of several workers (Pflug et al., 1963; Goldfarb, 1970; Tung et al., 1975; Lyon and Klose, 1981) have proven the technical feasibility of using the retortable pouches for that purpose. (iii) Quality of pouched product A number of researchers have carried out processing experiments to compare retort pouches and cans for quality attribute retention. Chen and George (1981) in evaluating the quality of green beans using home-processing conditions, found that beans processed in cans contained slightly more ascorbic acid, but had poorer texture and lower overall acceptance than those processed in pouches. They felt that the texture difference may have been related to a shorter process time and the presence of lower amount of brine in the pouched product compared to the canned product. Abou-Fedel and Miller (198 3) found that green beans packed in cans retained less thiamine and ascorbic acid and were lighter in colour, more yellow and softer than those 14 processed in retort pouches of the same volume. Cherries processed in retort pouches were found to contain significantly more ascorbic acid, and were more intensely red and firmer than those processed in cans. Comparisons of meat and fish in retort pouches and in cans have been conducted. Shrimp in retortable pouches were superior in flavour and colour to the canned product (Dymit, 198 3). Lyon and Klose (1981) studied the effect of heat on sensory properties of chicken meat processed in retortable pouches and cans. They concluded that the retort pouch process offered a method for improving the texture of processed chicken meat from spent hen by adequately cooking to tenderize the meat but not overcooking i t to the extent that the meat chunks were reduced to fibrous, shredded or stringy pieces. Pflug (1964) found that beef processed in retort pouches was preferred over that processed in cans. Chia et al. (198 3) who compared the properties of rainbow trout, pollock and shrimp processed at equal lethalities in cans and retort pouches found that the pouched product had a firmer texture than the canned product. In sensory evaluations, the pouched products were scored higher in most cases for colour, flavour and overall acceptability. Adams et al. (1983) used large retort pouches with a thickness of one inch to process red snapper, Spanish mackeral, blue crab, shrimp and flaked yellow fin 15 tuna. The product processed in retort pouches were compared to the product processed in 307 x 113 cans. All the product processed in pouches were in general judged, by sensory panelists, to be as good or superior to the products processed in much smaller cans. A change in container geometry therefore offers the possibility of improving retention of quality attributes. (iv) Quality of canned fish treated with salts Salts, through their interactions with some of the constituents of food systems, have very useful and important functions in fish processing. For example, the texture and flavour of sea food products have been reported to be improved by treatments with various phosphates. Wekell and Teeny (1988) found that dipping or soaking salmon steaks in a solution of 10-15% polyphosphate and 2% salt for periods ranging from a very brief "in and out" to two minutes produced a significant improvement in the texture and flavour of the salmon. English et al. (1988) found that dipping mullet in solutions containing low concentrations of tripolyphosphates, prior to canning, significantly improved the texture and reduced curd formation. They felt that the ability of the phosphate to bind water could account for the textural differences in the canned product. Other investigators claimed treatment of fish with a combination 16 of 12% polyphosphate and 4% salt effectively inhibited development of rancidity in the fish and, therefore, improved flavour and odour (Anon., 1962). MacCallen et al. (1962) reported that addition of tripolyphosphates to dips for cod fillets significantly improved the texture and increased tenderness of the treated fillets over controls. The effects of polyphosphates on improving the flavour of seafood products are, however, disputed (Ellinger, 1972). Claims for improved flavour may be due to the dipping effect masking some of the symptons of deterioration of poor quality sea products and may only be an impression that the quality is better. Spinelli et al. (1968) detected no differences in flavour or odour when they compared fish fillets or steaks dipped in polyphosphate solutions with undipped controls. However, Tarr et al. (1969) reported that an enzyme that hydrolyzes the 51-nucleotides is strongly inhibited by pyrophosphates. Since the 5'-nucleotides are good flavour intensifiers, the inhibition of the enzyme that would inactivate them may offer some explanation for flavour improvement due to the presence of pyrophosphate. Brining may also improve the texture and flavour of the fish since i t provokes an increase in water holding capacity of the muscle. As muscle hydration increases, the amount of liquid released decreases. Tarr (1971) described a "drip preventing 17 function" of the salt in which light brining of the fish muscle (0.5-1%) largely eliminated the free liquid exudate. Bilinski et al. (1977) treated herring for 40 hours in 5 to 100% saturated salt solutions (1.5 g/L-30.3 g/L). After brining, the herring was gibbed and water, instead of 3% NaCl, was added to cans before sealing. They found that brining of the herring caused up to a four fold increase in the firmness of the canned flesh. The firming up effect was increased further by raising the strength of the brine. Studies by Huynh et al. (1989) found that late-run chum salmon fillets soaked and washed in an 8% salt solution for two hours and then desalted for one hour in a 1% salt solution before canning showed a significantly reduced amount of late-run odour and flavour intensity and a significant improvement in firmness. They recommended that additional brine changes were necessary for fish of more advanced maturity. D. MEASUREMENT OF TEXTURAL PROPERTIES Development of methods for the instrumental measurement of the textural properties of food began at the turn of the century when devices for measuring the tenderness of meat were described. Since that early start, numerous methods and instruments have been developed to evaluate texture-related properties of a large variety of foods. The subject has been extensively reviewed and discussed in recent years by Szczesniak 18 (196 3) and Voisey (1971). Szczesniak (1963) divided these methods into three categories: fundamental, empirical and imitative. Fundamental tests measure fundamental rheological properties such as viscosities and elastic moduli (Szczesniak, 1963). They correlate very poorly with sensory evaluation. Empirical tests measure parameters, often poorly defined, that practical experience indicate to be related to textural quality. These include puncture, shear and extrusion tests and tend to correlate better with sensory evaluation. For example, high correlations have been obtained in studies on a wide variety of food including meats (Segars et al., 1981) and fish (Love et al., 1974) using the punch and die test. Imitative methods of measurement imitate the conditions to which the material is subjected in the mouth or on the palate (Szczesniak, 1963; Szczesniak, 1975). A typical example is the Texture Profile Analysis which describes the procedure of compressing bite-size pieces of food by some mechanical device and analyzing the force-time curve resulting from this simulated mastication (Friedman et al., 1963). Imitative tests may be performed using instruments such as the Instron Universal Testing Machine, the Brabender Farinograph, Amylograph and Alveograph (Szczesniak, 1963). Despite the various methods available for measuring texture, the objective study of fish texture is not easy due to 19 nonuniformity of fish muscle structure and difficulties in the test specimen preparation. Briefly, the body is divided into segments called myotomes by transverse sheets of connective tissue called myocommata. The size of these myotomes varies along the length of the body. Parallel to the long axis of the body there are muscle cells or fibers. The orientation of these fibers varies throughout the myotome. Their length and diameter differ within individuals according to their location in the body. The muscle can be divided into two types, dark and white muscles, whose proportion varies along the body length. Dark muscle, which lies along the side of the body under the skin, has more connective tissue around the muscle cells and i t also contains less protein and more lipid than the white muscle (Howgate, 1979; Suzuki, 1981). ' Dunajski (1979), reviewing the use of instrumental methods to assess the texture of fish, commented that those devices applicable to testing red meat are not suitable for cooked fish. He explained that fish muscle has a relatively low content of connective tissue which disintegrates during heating and the muscle falls apart easily. Thorough flaking of fish muscles, however, can improve homogeneity of the test sample and increase the randomness of fiber orientation with respect to the shearing blade (Bilinski et al., 1977; Borderias et al., 1983). Bilinski et al. (1977) who broke up the muscle of whole herring 20 (free from bones and skin) into flakes (about 3-5mm) using two forks obtained good reproducibility using the Ottawa Texture Measuring System for firmness of canned herring. Borderias et al. (1983) obtained higher coefficients of variations for raw and cooked fillets than for the minced fish when different kinds of cell attachments were used with an Instron tester. He felt that this occured because application of a compression method to an item with a layered structure, like fish f i l l e t s , cause the myotome layers to slide away from the force of compression. This phenomenon makes i t impossible to reproduce the results of an analysis in a different test. He did not find significant correlations between results of instrumental analyses and that of sensory evaluation for cooked and raw fish fillets but did however find good correlations for the fish minces. An ideal sensory test which facilitates a correlation between instrumental and sensory evaluation of texture is the General Foods Sensory Profiling Technique (Brandt et al., 196 3; Szczesniak et al., 1963; Civille and Szczesniak, 1973). It is based on the classification of textural characteristics into three main types: mechanical properties, geometrical properties and those relating to fat and moisture content (Table 1). The mechanical properties are those characteristics that are related to the responses of food material to applied force; the geometrical properties are those characteristics that are related to the 21 geometrical arrangement of the c o n s t i t u e n t s of food (e.g. s i z e , shape and o r i e n t a t i o n of p a r t i c l e s ) ; the c h a r a c t e r i s t i c s r e l a t e d to f a t and moisture content are those c h a r a c t e r i s t i c s t h a t are a s s o c i a t e d with the water and f a t content of the food (Brandt e t a l . , 196 3). This t e x t u r e p r o f i l e a n a l y s i s r e q u i r e s a panel of judges with p r i o r knowledge of the t e x t u r e c l a s s i f i c a t i o n system, use of standard r a t i n g s c a l e s , and proper panel procedures with regard to the mechanics of t e s t i n g and sample c o n t r o l (Brandt e t a l . , 1963). 22 Table 1: Classification of textural characteristics, based on the General Foods Texture Profile Method (Cardello and Mailer, 1987). A. Mechanical hardness fracturability cohesiveness adhesiveness viscosity chewiness springiness gumminess B. Geometrical size and shape powdery, chalky, gritty, beady, grainy, coarse, lumpy orientation flaky, fibrous, pulpy, cellular, aerated, puffy, crystalline. C. M o i s t u r e / f a t moistness oiliness greasiness 23 MATERIALS AMD METHODS A. FISH SAMPLE PREPARATION AND SOURCE The chum salmon used in this investigation were obtained from two sources: one sample was obtained in late July from commercial fishing boats; the other was obtained in October from the Chilliwack Hatchery. They were all returning salmon and were graded relative to their degree of maturity using the description shown in Table 2. For the purpose of this study, only fish of advanced sexual maturity (Grades 3 and 4) were used. Approximately two hours after leaving the hatcheries, all fish were cleaned, dressed, frozen at -30°C, and stored at that temperature until tested. Prior to testing, all frozen fish were brought out from the freezer and thawed at room temperature to bring them from a totally frozen state to a temperature of approximately 0°C. B. PROCESS DETERMINATION BY HEAT PENETRATION Recommended process times have been outlined for canned salmon processed with pure saturated steam and being cooled afterwards by water (National Canners Association, 1976; Lopez, 1987). However, the rate of heating and cooling of food products in containers is not only a function of the physical properties of the food, but also the geometry and heat transfer 24 Table 2: Maturity 1 of chum salmon No. Rank Description 1 Silver-semi S i l v e r y skin colour; some s l i g h t bright colour change (colour bars) may be present on the v e r t i c a l and dorsal surface but very f a i n t . Flesh colour ranges from pinkish red to orange red. 2 Intermediate Complete loss of uniform s i l v e r bright appearance. D i s t i n c t colour change with bars ranging in colour from l i g h t red to dark green or black with traces of purple. Flesh is l i g h t pinkish to orange. 3 Dark Intensive breeding colour. Many d i s t i n c t c o l o u r f u l bars of dark green or black on the skin. Skin is thick and is covered by a heavy slime layer. B e l l y flaps are t h i n ; f l e s h colour i s pale with l i t t l e pinkish tone remaining. Some late odour can be noted from the f l e s h . 4 Spawning River-caught f i s h of spawning condition. Body colour is green or very dark, skin i s very thick and has a very heavy slime layer. Very th i n b e l l y f l a p s ; f l e s h colour is greyish white. A very strong late odour i s evident from both skin and f l e s h . ^Source: Huynh (1988) with a few modifications. 25 characteristics of the container (Pflug and Esselen, 1980). Heat penetration studies were therefore necessary to design the process schedules for the pouched samples since there was no information available on heat processing times for salmon in pouches. To do this i t was necessary to determine the lethality of the canned salmon process using commercially acceptable conditions and then use this lethality as the target lethality for the pouches. The traditional approach of basing the process on the slowest heating container was used to estimate the lethality of the canned process. Then by means of a computer program (Smith, 1987), i t was possible to estimate the process time required to achieve a similar lethality for the pouched samples. For both the cans and pouches, heat penetration data of three process runs were used in the estimation (Appendix 11 and 111). The heating and cooling curves for each test container were simultaneously displayed graphically on the screen of the computer monitor for analysis. Straight lines were drawn graphically on the monitor screen through the data points on the linear portion of the curve. Linear regression coefficients were applied to determine the best fitting straight line. Simultaneously, the heating and cooling parameters were calculated. 26 (i) Retort system used A l l processing was carried out in an FMC 500W laboratory r e t o r t (FMC Corporation, Santa Clara, CA) which could be operated with steam or water as the heating medium. Construction details for this r e t o r t was reported by Morello (1987). A steam cook was used for the can process and a water cook for the pouch. For the water cook, the reservior tank was two-thirds f i l l e d with water and preheated to about 10C° above the target r e t o r t temperature to reduce come up time. The reservoir tank was then pressurized to 206.8 kPa (30 psig) to aid the transfer of water to the ret o r t . Water was then transferred from the reservoir to the r e t o r t at the beginning of the process. At the end of hot water transfer, the r e t o r t vent was closed, the air supply was opened and also the steam by-pass valve. When the r e t o r t mercury thermometer reached within 2C° of the target temperature, the steam by pass valve was closed. The r e t o r t was operated at 248°F and 172.37 kPa (25 psig) air pressure. At the end of the process, the reservoir was prepared for the cooling cycle by releasing a i r pressure. The main steam valve was closed at the s t a r t of cooling and the hot process water was then transferred back to the reservoir tank. Cooling was then achieved by simultaneously adding cold water from the main water supply and draining the equivalent flow from the r e t o r t , while circulating that portion of the t o t a l flow that was not being 27 drained. Air overpressure was maintained during the cooling cycle. For the steam cook, all valves connected to the reservoir tank were closed off. Controlled steam was then allowed to enter via a manifold at the bottom of the retort. The retort was operated at 15 psig steam pressure and 248°F. Its drain valve was opened slightly for condensate removal and to provide steam flow past the controller sensing bulb. Following steam processing, cooling was achieved by the addition of cold water. Air flow was started and maintained to allow for pressure cooling. All valves were regulated by a Taylor Fulscope Recording Temperature and Pressure Controller (Taylor Instruments Ltd., Toronto, ON). (ii) Processing conditions The processing conditions and container dimensions used in process determination work are outlined in Table 3. The 307 x 115 three piece cans (Wells Can Company Limited, Burnaby, B. C.) and the retort pouches (Fijitoku Ltd., Japan) with outside dimensions of 176 mm by 250 mm and inside dimensions of 146 mm by 240 mm were packed with the same amoumt of product (213g plus an extra 10% overweight to similate the worst possible condition at which the retort could be operated). 28 Table 3: Processing conditions for cans and r e t o r t pouches used during process determination work. Container Amount of product Process Temp. Heating medium 307 x 115 234 g 248°F steam can 176 x 250mm 234 g 248°F water pouch 29 Cans were steam processed at 248°F for 65 minutes. This was based on a known commercial process for 307 x 115 cans. According to Lopez (1987), 62 minutes at 245°F or 54 minutes at 250°F are safe process for 307 x 113 cans. Pouches were processed in the same FMC retort using a water cook with superimposed air pressure and retort temperature similar to the cans. Process time for the pouches was determined from the heat penetration data to be 34 minutes. All process times were operator's process times (Pt) i.e., the length of heating period following come up time. (iii) Heat penetration Thermocouples were inserted at the centre of six cans and six pouches in each run. Four retort thermocouples were also used in each run. Three replicates runs were carried out for both the cans and pouches. All thermocouples were calibrated against the retort thermometer of a small vertical conventional retort in flowing steam at 248°F which in turn was calibrated against a ASTM certified mercury in glass thermometer. Thermocouples were type T/C copper/constantan wire (Omega Engineering Inc., Stamford, CT) with soldered ends. Retort thermocouples were used to monitor the environment temperature in the retort and were placed near the containers. Ecklund needle-type rigid thermocouples with stainless steel bodies were 30 used to monitor temperature in the canned product. Locking receptacles and connectors from 0. F. Ecklund (Omega Engineering, Inc., Stamford, CT) were used to hold the thermocouples in the cans. The thermocouple tips were located at the center of the cans so as to be positioned in the slowest heating zone. The cans were filled with the desired weight of chum steaks, one percent salt was added and they were then sealed with an automatic can sealer (Rooney's Machine Co., Bellingham, WA), under vacuum. For the pouches, thermocouples wires were imbedded into a piece of fish cutlet so as to be located at the centre of each pouch. The wires were sealed into the pouches by inserting them through Ecklund packing glands. Approximately 60 to 70mm of thermocouple wire was left inside each pouch. The retort pouches were then filled with desired weight of steaks, one percent salt was added and they were vacuum sealed using a Multivac vacuum sealer (Sepp Haggenmuller KG, Allgau, West Germany). The filled containers were held at 1°C overnight and were processed the following day. The retort was loaded as quickly as possible. Pouch were oriented in the horizontal direction and restrained to 0.75 inch (19 mm) thickness by the position of the retort shelves. The temperatures for all thermocouples during processing were recorded at one minute intervals using a Datataker (Boronia, Australia). Following processing, retort 31 pouches were opened and checked to ensure that the end of the thermocouple or thermocouple wire had not shifted from the geometric center of the container. (iv) Process calculations Calculations were performed using a computer program (Pro Calc Associates, Surrey, B.C). The program permitted direct transfer of data from files containing data-logger type output to the thermal processing files. Regression analysis was used to determine the parameters f n and jeh from the heating curve and f c and J w from the cooling curve for each container. In these determinations, cooling water temperature was taken as the mean of the constant reading of retort thermocouples at the end of the cooling period. The initial temperature of the product was taken as the average of all the test containers at the start of the process (Bee and Park, 197 8). Commercially, the initial temperature of the product should not be less than 35°F. Come up time for the pure steam process was six minutes and for the water cook seven minutes. Appendices 11, 111 and IV give examples of the heating and cooling parameters obtained from the computer output. Process lethalities were calculated for each can using the determined heating and cooling parameters. In these calculations, the j values for the 307 x 115 cans were corrected using the 32 appropriate factor presented by Ecklund (1956) to account for the metal nut of the packing gland protruding within the can. The lethality of the slowest heating can (F o = 6.50 min) (see Appendix 11) was then used in calculations involving the pouched process. The process times for pouches determined for use in the final phase of the work was the mean plus three standard deviations (Appendix 111). This procedure assumes a very small possibility that any container would receive less than the target lethality of 6.50 minutes at the cold spot, C. PROCESSING FOR INSTRUMENTAL AND SENSORY ANALYSIS Four pilot experiments were conducted to determine the effect of treatments and processing methods on the texture and flavour of the late-run chum. The f i r s t experiment was undertaken to determine whether the removal of the skin and bone had any effect on the flavour of late-run salmon. Late-run chum salmon obtained from commercial fishing boats, in late July 1988 was used for this study. The fish were of advanced sexual maturity (Grade 3, Figure 1). Two fish were chosen for this experiment. Following overnight thawing at about 10 0C, each fish was washed with cold water and divided into steaks I V 2 inches (38 mm) in thickness. Steaks from each fish were randomly assigned to one of two lots; the skin and bones from one were removed while the other 33 Figure 1: External appearance of Grade 3 late-run chum salmon SMPLE L Tu-v 2JLW* 34 lot was processed with skin and bones intact (Figure 2). The second pilot experiment was designed to evaluate the effect of a change in container geometry on the texture and flavour of the raw material. Late-run chum salmon steaks l V z Inches in diameter were processed in retort pouches. The third pilot experiment was undertaken to evaluate a precanning treatment with polyphosphate/brine on the texture and flavour of the raw material. Steaks were soaked in a solution of 10% polyphosphate and 2% brine for two minutes. At the end of the dip time, the samples were drained of excess liquid and packed into cans. Salt was added to the level of 1% by weight to these cans. The fourth experiment was undertaken to evaluate the effect of a brining treatment on the texture and flavour of the canned product. Steaks received two washes with 8% salt solution for one hour each time at the ratio of two parts fish to one part brine. This was followed by a 1 hour desalting in a 1% salt solution to remove the salty taste. Fish for the second, third and fourth study were obtained in late October 1988, from the chum spawning run at the Chilliwack Hatchery. These fish were of advanced sexual maturity (Grade 4, Figure 3). Four fish were chosen. They were prepared by removing the skin and bones and then dividing each fish into steaks I V 2 in diameter. The steaks from each fish were then 35 Figure 2: Flow diagram of the procedure used for the salmon canning process in Experiment 1 HAND BUTCHER water,blood, head, I viscera, t a i l , f i n s , etc. WASHING water 1 SEALED IN POLYETHYLENE PLASTIC BAGS i STORAGE AT -30°C i THAWING I CUTTING IN STEAKS 38mm RANDOM DIVISION OF STEAKS 2 parts per f i s h REMOVAL OF SKIN AND BONES — HAND-FILLED— WEIGH AND PATCH SEAM RETORT i >0L I OC I COO  STORAGE AT ROOM TEMPERATURE WITH SKIN AND BONES PRODUCT ANALYSIS 36 Figure 3: External appearance of the Grade 4 late-run chum salmon 37 randomly subdivided i n t o f o u r p a r t s : one p a r t was packed i n t o cans (307 x 115) each containing 213g and 1% s a l t added. This s e r v e d as the c o n t r o l . Another q u a r t e r was h a n d - f i l l e d i n t o r e t o r t a b l e pouches (213g to each pouch) and 1% s a l t added; the t h i r d q u a r t e r was t r e a t e d with polyphosphate/brine s o l u t i o n and the f i n a l p a r t r e c e i v e d the brining treatment (Figure 4). A l l c o n t a i n e r s were vacuum sealed and processed a t conditions p r e v i o u s l y d e s c r i b e d (Table 3). Thermocouples were i n s e r t e d i n s i x c o n t a i n e r s per run to measure the accomplished l e t h a l i t y of the batch. A f t e r p r o c e s s i n g , the c o n t a i n e r s were checked f o r leakage and p h y s i c a l damage. The canned and pouched samples were then s t o r e d a t room temperature f o r a t l e a s t one month p r i o r to product a n a l y s i s t o allow time f o r the s a l t to e q u i l i b r a t e and the c h a r a c t e r i s t i c canned f i s h f l a v o u r to develop. D. SAMPLING AND PRODUCT ANALYSIS (i) Sampling procedure Two c o n t a i n e r s from each treatment f o r each f i s h were chosen f o r product a n a l y s i s . A n a l y s i s was performed about 2-3 months a f t e r processing. The cook-out l i q u i d was drained from the can/pouch i n t o a graduated c y l i n d e r and measured. Samples 38 Figure 4: Flow diagram of the procedure used for the salmon canning process in Experiments 2, 3 and 4. HAND BUTCHER water / b l o o d , head, | v i s c e r a , t a i l , f i n s , 4> e t c . WASHING water SEALED IN POLYETHYLENE PLASTIC BAGS STORAGE AT -30°C 1 THAWING CUTTING |N STEAKS 38mm RANDOM DIVISION OF STEAKS 4 p a r t s per f i s h 1 SOAKING..10% phosphate—UNTREATED POUCH" 2% s a l t 2 min DRAIN —HAND-FILLED 1 WEIGH AND PATCH SEAM RETORT i COOL STORAGE AT ROOM TEMPERATURE fWO WASHES..8% s a l t 1 hr each f i s h : b r i n e ( 2 : l ) DESALTING..1% s a l t 1 hr f i s h : b r i n e ( 1 : 1 ) PRODUCT ANALYSIS 39 were then removed from their containers and placed on individual plastic plates. The samples were broken into flakes (about 3-5 mm), using two forks. Samples from each can were divided into two parts: one for Instron measurements and the other for sensory assessment. (ii) Sensory Analysis (a) Selection, training and establishing rating scales Texture The method used was modeled a f t e r the General Foods Texture Profile Method (Brandt et al., 19 6 3; Szczesniak et al., 196 3) and the Quantitative Descriptive Analysis (Stone et al., 197 4) with a few modifications. These modifications included using only characteristics appropriate to the fi s h samples and using only f i s h and meat samples of various tenderness for training judges, among others. Definitions of the attributes used are given in Table 4. Nine panelists were selected to tr a i n for sensory evaluation of the chum salmon product. They were selected on the basis of inte r e s t , co-operative attitude, willingness and availability. Six of these panelists had prior training and experience in sensory evaluation of fi s h and fish products. The 40 Table 4: Definitions used i n sensory p r o f i l i n g of salmon A t t r i b u t e s D e f i n i t i o n s 1. F i r m n e s s The f o r c e r e q u i r e t o compress t h e m a t e r i a l between t h e molars or between the tongue and p a l a t e . 2. Dryness The amount of f r e e f l u i d s i n t h e o r a l c a v i t y d u r i n g m a s t i c a t i o n . 3. F i b r o u s n e s s The p e r c e i v e d degree of f i b e r s e v i d e n t d u r i n g m a s t i c a t i o n . 4. Chewiness The t o t a l e f f o r t r e q u i r e d t o p r e p a r e the samples t o a s t a t e r e a d y f o r s w a l l o w i n g . 5. L a t e - r u n The presence of a t a s t e s e n s a t i o n not F l a v o u r t y p i c a l of canned salmon which produces an u n d e s i r a b l e b u r n t / s o u r f l a v o u r i n the mouth. 6. O v e r a l l O v e r a l l i m p r e s s i o n of the t e x t u r e and A c c e p t a b i l i t y f l a v o u r of t h e sample. 41 panelists were informed of the general purpose of the study. The selected panelists were then trained in a series of three sessions during which they were given samples covering a wide range of texture and asked to rate them on a scale subsequently used in the experimemt. This scale consisted of a line with descriptive anchor points at the ends and at the centre (Stone et al., 1974) (Table 5). Each panelist marked his/her response with a vertical line across the horizontal scale at the point representing his/her judgement of the specific attribute. All panelists were in agreement with the scales chosen for firmness, dryness and fibrousness. However, there was a semantic problem with the terms used for chewiness - too mushy, too rubbery. This was finally changed to: "little effort needed to swallow, much effort needed to swallow." Preliminary experiments were designed to test the performance of the panelists and the reliability of the scale. Analysis of data from the preliminary experiments indicated that while most panelists were discriminating differences among samples for specific characteristics, none of the panelists discrimated a ll the required characteristics consistently. Training period was halted and the pertinent scales were retained for actual testing of experimental products. 42 Table 5: Scales used in sensory analysis. FIRMNESS so f t : low resistance to breakdown hard: high resistance to breakdown soft hard DRYNESS wet: presence of f l u i d in sample mass dry: absence of f l u i d in sample mass wet dry FIBROUSNESS few: presence of few fibers many: presence of many fibers few many CHEWINESS l i t t l e : l i t t l e e f f o r t needed to swallow much: much e f f o r t needed to swalow l i t t l e much LATE-RUN FLAVOUR none: absence of late-run flavour strong: presence of late-run flavour none strong OVERALL ACCEPTABILITY unacceptable acceptable unacceptable acceptable 43 Flavour In order to establish descriptive terminology for the flavour study, the panel members were asked to individually examine fish with a high degree of late-run flavour and write down words which best described the impressions recorded when the substance was taken in the mouth i.e., the basic taste factors, odour, mouth sensations and after-effects. Most panelists used more than one word. In round table discussions that followed, all the panelists felt none of these words accurately described the principal aroma and flavour of the sample perceived. Since i t was very difficult to agree on a word that best described the flavour of the samples, a common terminology - late-run flavour was adopted. Training sessions were then conducted during which panelists were given fish covering a wide range of late-run flavour and asked to rate them on a scale to be used in the experiment. Training was completed when most of the panelists gave similar ratings to the samples. (b) Sensory testing The training period helped prepare panelists for this phase of the sensory evaluation. It was conducted in a sensory panel room which was equipped with individual booths and fluorescent lighting. Water, unsalted soda and crackers were available for 44 cleansing the mouth of residual flavour and other particles after tasting each sample. Two different studies were conducted. In the f i r s t study panelists were presented with two coded samples from the f i r s t pilot experiment- one represented the boneless-skinless steaks, the other the steaks with skin and bone. These samples came from the same fish and were divided into standard-size pieces of flakes about 3-5 mm thick using two forks. Two replications were done for each fish and two fish were used. A total of 36 judgements were therefore made for each treatment. In study 2, four coded samples from the same fish were presented to each panelist. They represented the treated samples from experiments 2, 3 and 4 plus the control. Four fish were used with two replications done for each fish. Nine judges were present in three sessions and eight in the fourth. A total of 70 judgements were made for each treatment. Panelists were asked to evaluate the fish samples for firmness, dryness, fibrousness, chewiness, late-run flavour and overall acceptability. The order of the tasting of the samples were randomized throughout the study to cancel out any sequential effect on panelist judgement. All the characteristics were rated on a horizontal line scale 6 in (15 cm) long with anchor points 0.5 in (1.3 cm) from each end (Appendix V). Each anchor point was labeled with a word or expression. A separate 45 line was used for each sensory property evaluated. Each judge recorded his/her evaluation by marking a vertical line across the horizontal line at the point which best reflected his perception of the magnitude of that property. Panelists were asked to compare the texture of the salmon samples to two reference samples: flaked ham and canned tuna. They were required to use the flaked ham as the f i r s t anchor point (0.5) and the canned tuna as the next (5.5). The flaked ham represented a product with a soft, juicy flesh, very few fibers present and l i t t l e effort needed to chew and swallow. Such textural characterictics were considered unacceptable for a fish product. The canned tuna, on the other hand, represented a fish product which was firm, dry, fibrous, and required considerable chewing effort before swallowing. For flavour, the scales were reversed: a score of 0.5 represented a product with no detectable late-run flavour while 5.5 represented a product with very strong late-run flavour. As to overall acceptability, a score of 5.5 represented a very acceptable product and 0.5 represented an unacceptable product. Panelists' marks were then converted to numerical scores from 0 to 6 by measuring the distance of the judges' marks from the lef t end of the line in units of 1 in (2.5 cm). 46 (iii) Instrumental technique A l l samples used in the i n s t r u m e n t a l a n a l y s i s were taken from the same can/pouch as t h a t used f o r sensory. A f t e r a b r i e f mixing, 10 g each of the flaked samples was packed to a c o n s t a n t height between two nylon disks by applying a gentle p r e s s u r e i n a 50 c c . syringe the end of which was severed a t the zero point. This formed a c y l i n d e r measuring 2.5 cm in diameter by 2 cm in height. Samples were then compressed using a t e f l o n coated compression p l a t e 5.5 cm i n diameter, attached to the I n s t r o n U n i v e r s a l T e s t i n g Machine (Model 1122, I n s t r o n Corporation, Canton, MA). T e s t conditions (Table 6) such as crosshead and c h a r t speeds were s e l e c t e d a f t e r preliminary t r i a l s t o ensure t h a t a complete h i s t o r y of the changes i n the measured v a r i a b l e s could be g r a p h i c a l l y recorded. A t y p i c a l r e c o r d c o n s i s t s of a t e x t u r e p r o f i l e curve with two peaks from which the hardness, firmness, cohesiveness, chewiness and gumminess of the samples could be determined. Taking the data from the s t r i p - c h a r t r e c o r d e r provided with the I n s t r o n i s time-consuming since i t r e q u i r e s measurements of the v a r i o u s d i s t a n c e s and a r e a s with a r u l e r and planimeter. In addition, since slopes are of i n t e r e s t , determining l i n e a r l i m i t s and a c c u r a t e l y measuring the slopes of s h o r t s e c t i o n s of the curve are d i f f i c u l t . As a r e s u l t , the I n s t r o n was i n t e r f a c e d with a microcomputer and the readings were taken in m i l l i v o l t s . Since 47 Table 6: Test conditions at which the instron was operated during instrumental analysis of chum salmon. Analys i s Crosshead speed Chart speed # of c y c l e s of crosshead TPA lOOmm/min lOOmm/min 2 48 the basis of data collection and crosshead movement is time, the computer was programmed to take readings of the load cell signal every 0.02 seconds. The Instron was f i r s t calibrated by measuring the difference in load cell output with no weight and with a known weight. (iv) Data Analysis All statistical calculations involving sensory and instumental data were performed using a program package SYSTAT (Wilkinson, 1988). The sensory evualuation data was analyzed as a randomized block design with judges as blocks. Instrumental results were treated by one way analysis of variance to test for significance between treatments. If significance was found (p<0.05), an hypothesis test based upon Bonferroni multiple comparison test was applied. This hypothesis test defines a single degree of freedom hypothesis and permits the comparison between any two treatments giving an F value which can be used to determine the level of significance, and is available as part of the SYSTAT package. Simple linear regression was used to find the correlation between instrumental and sensory results. 49 RESULTS AND DISCUSSION A. PROCESS DETERMINATION Figures 5 and 6 show typical heating and cooling curves obtained for the canned and pouched process respectively. The plot of the heating curve shows the logarithmic value of the difference (g) between retort and product centre temperatures as a function of heating time on a linear scale starting with steam on, as zero time. The plot of the cooling curve shows the logarithmic value of the differences between product centre and cooling water temperature also as a function of heating time in minutes. The heating curve was characterized by an initial lag of the product centre temperature followed by a linear relationship between log g and heating time. In Figure 6 (pouch process) there was very l i t t l e lag at the start of the heating curve, resulting in a Jch value of less than one (Table 7). Also, in Figure 6, as in the majority of cases for the pouched process, the data points deviated from the linear relationship as the product centre temperature approached the retort temperature at the end of the heating. This was because in determining process times the pouches were processed until the temperature remained almost constant. Table 7 gives the mean heating and cooling parameters determined for the cans and pouches during the process 50 Figure 5: Typi c a l heating and cooling curve obtained for canned salmon processed at 248°F-119.0r Channel 4 - Run i - s a l can 116,8. 110,0. / / / / / 1.4" 20,0T«>< -196L-LJ 1 1 L J I -326.2-1 110,0-41,6 20,0 13,2-J I 0 20 40 60 100 120 140 Time, min Note: the di s t a n c e between the arrows represents the slope of the heating curve 51 F i g u r e 6: T y p i c a l heating and c o o l i n g curve obtained f o r the pouched salmon samples-Channel 2 - Run 6 - salpouch / 8/ J I L J 1 1 L -325.2-1 189,8. 48.6 19.8 12.2 J 1 L 28 48 68 188 128 148 Time, min 52 Table 7: Heating and cooling parameters for chum salmon obt-ained during process determination work. Process Mean (standard deviation) f h r m i n f a / m i n jaw j a e cans 31.91(1.6) 36.54(3.6) 2.35(0.47) 1.55(0.7) pouches 19.12(1.2) 16.40(4.0) 0.52(0.05) 1.37(0.32) 53 determination work. Comparing the parameters among the two types of c o n t a i n e r s , i t was obvious t h a t heat p e n e t r a t i o n was more ra p i d i n the pouches than the cans. This was indi c a t e d by smaller heating and cooling r a t e indices (Table 7). The t h i n p r o f i l e and the i n c r e a s e d s u r f a c e area of the r e t o r t pouches are responsible f o r r e d u c t i o n i n the heating time. For the cans, the mean f= was g r e a t e r than f n but f o r the pouches the e f f e c t was reversed. Stumbo's method does not consider the a c t u a l f<= in the c a l c u l a t i o n procedure, but i n s t e a d assumes fc==f»». Since f e was g r e a t e r than f»» f o r the cans th e r e would be an a d d i t i o n a l margin of s a f e t y i n the pr o c e s s times used while f o r the pouches t h i s s a f e t y margin w i l l be revers e d . Table 8 shows the proc e s s times used f o r the cans and pouches during p r o c e s s i n g f o r product a n a l y s i s and the mean d e l i v e r e d l e t h a l i t y of these processes. Compared to the canned product, the r e t o r t pouches allowed f o r a 47.8 % re d u c t i o n i n the processing time. No other r e p o r t s of pro c e s s i n g time f o r salmon i n r e t o r t pouches were found i n the l i t e r a t u r e . However, redu c t i o n of heating time using pouches has been r e p o r t e d by a number of workers (Tung e t a l . , 1975; R i v z i and Ac t i o n , 1982; Chia e t a l . , 1983). R i z v i and A c t i o n (1982) s t a t e d t h a t p r o c e s s times r e q u i r e d t o achieve equivalent l e t h a l i t i e s with s t i l l p r o c e s s ing was on e - t h i r d t o one half l e s s f o r r e t o r t pouches compared to cans containing the same volume of product. Tung e t a l . , (197 5) 54 Table 8: Process times used i n the product analysis phase of the work and th e i r corresponding l e t h a l i t i e s c a l c -ulated by Stumbo's method. Container type Process time, min Mean F D, min* cans 65 7.61(0.72) pouches 34 8.71(0.69) * Data expressed as mean (standard d e v i a t i o n ) 55 found that cream style corn processed in cans and retort pouches required 75 and 33 minutes respectively in a 121°C s t i l l retort to achieve a minimun lethality of Fo = 5 minutes. Chia et al. (1983) using a modified pressure cooker to process pouched rainbow trout, pollock and shrimp found that they required approximately two thirds the processing time.,--of the canned products which were processed using a laboratory autoclave. For each container type, the mean calculated process lethality was greater than the target lethality of 6.50, ensuring the safety of both processes (Table 8). The mean F« for the pouches was greater than that for the cans. This could be accounted for by the higher variation in the heating and cooling parameters for the pouches (Table 7). B. SENSORY DATA The mean panel ratings for the various characteristics of the processed late-run chum salmon are shown in Tables 9 and 10. Data for individual judgements were treated by two way analysis of variance to test for significance between judges as blocks and between treatments. The analysis of variance data for all sensory attributes are given in Tables 11 and 12. There were significant differences among judges for almost all attributes tested. Judge differences are not unusual since the portion of the scale used for scoring sensory attributes may differ among 56 Table 9: Mean panel r a t i n g s 3 for sensory att r i b u t e s of the boneless-skinless salmon steaks and the steaks with skin and bone processed at 248°F. Treatments Parameters Boneless-skinless With skin and steaks bones Firmness 3.2(0.9)- 2.4(0.9) b Dryness 2.7(0.9)- 2.2(0.9) fc> Fibrousness 3.5(0.9)- 2.5(0.9) b Chewiness 3.6(0.7)- 2.8(0.9) b Late-run 3.1(1.3)- 2.9(1.4)-Flavour Overall 3.2(1.4)- 3.5(1.3)-Accep t a b i l i t y 3Data expressed as mean (standard deviation), n=36. -toMeans within the same row with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t (p>0.05), as tested by the SYSTAT hypothesis t e s t . 57 Table 10: Mean panel ratings* for sensory att r i b u t e s of brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F. Parameters Treatments Brine Phosphates Control Pouch Firmness 1.9(1.2)* 2.0(1.0)* Dryness 2 . 1 ( l . l ) t o 2 . 1 ( l . l ) b Fibrousness 2.2(1.2)* 2.4(1.0)* Chewiness 2.0(1.4)* 2.3(1.1)* 2.1(1.0)* 3.5(0.9)' 2.2(1.0)* 3.9(0.9) 2.5(0.9)*= 2.4(1.0)* 3.9(1.0)' 3.8(0.9)' Late-run Flavour Overall A c c e p t a b i l i t y 3.0(1.4)-= 3.5(1.3)*= 3.7(1.4)* 2.6(1.3)' 3.0(1.5)-= 2.5(1.4)*= 2.3(1.4)* 3.3(1.5)' * Data expressed as mean (standard deviation), n=70. -*= Means within the same row with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t (p>0.05), as tested by the SYSTAT hypothesis t e s t . 58 Table 11: Analysis of variance data for sensory studies of the boneless-skinless steaks and the steaks with skin and bone processed at 248°F. Parameters Source DF Mean square F value Firmness Judge 8 1. 810 2. 787* Treatment 1 12. 251 18. 864** Error 62 0. 649 Dryness Judge 8 0. 515 0. 575"-Treatment 1 5. 780 6. 451* Error 62 0. 896 Fibrousness Judge 8 2. 289 3. 513** Treatment 1 19 . 740 30. 297** Error 62 0. 652 Chewiness Judge 8 1. 623 2. 720* Treatment 1 12. 751 21. 368** Error 62 0. 597 Flavour Judge 8 6 . 003 4. 667** Treatment 1 0. 432 0. 336"-Error 60 1. 286 Overall Judge 8 4. 526 3. 168** Acceptabi1ity Treatment 1 1. 500 1. 050"-Error 58 1. 428 * * s i g n i f i c a n t (p<0.01) • s i g n i f i c a n t (p<0.05) "-not s i g n i f i c a n t 59 Table 12: Analysis of variance data for sensory studies of brine, polyphosphate/brine, pouch and untreated samples of chum salmon processed at 248°F. Test Source DF Mean square F value Firmness judge 8 4. .165 4. .338** treatment 3 37. .522 39. .098** error 268 0. .960 Dryness judge 8 2. .267 2, .141* treatment 3 59. .507 56. ,203** error 268 1, .059 Fibrousness judge 8 2. .400 2. ,260* treatment 3 42, .533 40, .042** error 268 1. .062 Chewiness judge 8 3, .184 3. ,048** treatment 3 47. .170 45. ,153** error 268 1, .045 Flavour judge 8 8. .655 5. , 464** treatment 3 14 , .548 9, , 184** error 259 1. .584 Overall judge 8 8, .082 4 , .313** Ac c e p t a b i l i t y treatment 3 12. .493 6. ,666** error 240 1. . 874 • s i g n i f i c a n t (p<0.05) * * s i g n i f i c a n t (p<0.01). 60 individual judges. In experiments 2, 3 and 4, whenever significant differences were found between treatments a multiple comparison hypothesis test was employed to determine which treatments differed significantly (Tables 13-18). In the f i r s t pilot experiment, the boneless-skinless salmon steaks had higher overall mean scores for all attributes (except overall acceptability) than the steaks with skin and bone (Table 9). Analysis of variance results however indicate no significant improvement in flavour and acceptability between the two samples (Table 11). Significant differences (p<0.05) were found between the samples for firmness, dryness, fibrousness and chewiness. In the second pilot experiment, pouched samples showed higher mean scores compared to the control (untreated) samples for all textural attributes tested (Table 10). Significant differences (p<0.05) were found between the pouched and untreated samples (Tables 13-18) with the pouched samples being judged firmer, drier, more fibrous and less chewy than the untreated samples. These differences in texture may have been in part due to the shorter processing time used as well as to the physical pressing of the pouched samples during processing. Less presence of late-run flavour was detected in the pouched samples than the control. However, comments offered by the majority of panelists indicated that both samples exhibited a 61 Table 13: Multiple comparison hypothesis test comparing sensory firmness of the co n t r o l , brine, poly-phosphate/brine and pouched salmon processed at 248°F. Comparison F P Pouch vs Brine 95. 005"" 0 .000 Pouch vs Phosphate 75. 000*" 0 .000 Pouch vs Control 62. 703*" 0 .000 Brine vs Control 2. 800"- 0 .095 Brine vs Phosphate 0. 805"- 0 .370 Control vs Phosphate 0. 603"- 0 .438 * * s i g n i f i c a n t (alpha<0.01)) "-not s i g n i f i c a n t Note: For s i g n i f i c a n c e P<= alpha k where alpha i s the chosen significance l e v e l , k i s the number of possible comparisons = 6. 62 Table 1 4 : Multiple comparison hypothesis test comparing sensory dryness of the control, brine, poly-phosphate/brine and pouched salmon processed at 2 4 8 ° F . Compar ison F P Pouch vs Brine 119.155"" 0.000 Pouch vs Phosphate 114.362"" 0.000 Pouch vs Control 102.729"" 0.000 Brine vs Control 0.609"- 0.436 Brine vs Phosphate 0.049"- 0.825 Control vs Phosphate 0.312"- 0.577 * * s i g n i f i c a n t (alpha<0.01)) "-not s i g n i f i c a n t Note: For sign i f i c a n c e P<= alpha k where alpha i s the chosen significance l e v e l , k i s the number of comparisons possible = 6. 63 Table 15: Multiple comparison hypothesis t e s t comparing sensory fibrousness of the co n t r o l , brine, poly-phosphate/brine and pouched salmon processed at 2 4 8 ° P . Comparison F P Pouch vs B r i n e 96 .834*- 0. 000 Pouch vs Phosphate 76 .415** 0. 000 Pouch vs C o n t r o l 60 .689-- 0. 000 B r i n e vs C o n t r o l 4 .203"- 0. 041 B r i n e vs Phosphate 1 .207"- 0. 273 C o n t r o l vs Phosphate 0 .905"- 0. 342 * * s i g n i f i c a n t (alpha<0.01)) "-not s i g n i f i c a n t Note: For s i g n i f i c a n c e P<= a l p h a k where a l p h a i s t h e chosen s i g n i f i c a n c e l e v e l , k i s t h e number of comparisons p o s s i b l e = 6. 64 Table 16: Multiple comparison hypothesis t e s t comparing sensory chewiness of the c o n t r o l , brine, poly-phosphate/brine and pouched salmon processed at 248°F. Compar i s o n F P Pouch vs B r i n e 113. 076"" 0 .000 Pouch vs Phosphate 80. 342"" 0 .000 Pouch vs C o n t r o l 68. 784*" 0 .000 B r i n e vs C o n t r o l 5. 476"- 0 .020 B r i n e vs Phosphate 2. 790"- 0 .096 C o n t r o l vs Phosphate 0. 449"- 0 .504 * * s i g n i f i c a n t (alpha<0.01) ) "-not s i g n i f i c a n t Note: For s i g n i f i c a n c e P<= a l p h a k where a l p h a i s t h e chosen s i g n i f i c a n c e l e v e l , k i s t h e number of comparisons p o s s i b l e = 6. 65 Table 17: Multiple comparison hypothesis t e s t comparing late-run flavour of the co n t r o l , brine, poly-phosphate/brine and pouched salmon processed at 248°F. Comparison F P Pouch vs Control 21.977-* 0 .000 Pouch vs Phosphate 13.920** 0 .000 Brine vs Control 10.919** 0 .001 Brine vs Phosphate 5.493"- 0 .020 Brine vs Pouch 1.949"- 0 .164 Control vs Phosphate 0.923"- 0 .338 * * s i g n i f i c a n t (alpha<0 "-not s i g n i f i c a n t .01) ) Note: For si g n i f i c a n c e P<= alpha k where alpha i s the chosen significance l e v e l , k i s the number of comparisons = 6. 66 Table 18: Multiple comparison hypothesis t e s t comparing o v e r a l l a c c c e p t a b i l i t y of the con t r o l , brine, polyphosphate/brine and pouched salmon processed at 248°F. Compar ison F P Pouch vs Control 16 .491"" 0. 000 Brine vs Control 9 .359"" 0. 002 Pouch vs Phosphate 8 199-- 0. 005 Brine vs Phosphate 3 .488"- 0. 063 Control vs Phosphate 1 .434"- 0. 232 Brine vs Pouch 0 .991"- 0. 320 * * s i g n i f i c a n t (alpha<0.01)) "-not s i g n i f i c a n t Note: For sign i f i c a n c e P<= alpha k where alpha i s the chosen si g n i f i c a n c e l e v e l , k i s the number of comparisons = 6. 67 burnt aftertaste which was stronger and lingered longer for the untreated samples. Overall the pouched samples were judged higher in terms of acceptance, although the mean score indicated only moderate acceptability. This seems to have been due to the advanced sexual maturity of the fish. In the third pilot experiment there were no significant differences between the polyphospate/brine samples and the untreated samples for all attributes tested (Tables 13-18). In fact, the treatment of the boneless-skinless steaks with polyphosphate prior to canning resulted in slightly lower scores for firmness and a slightly more juicy and less fibrous sample. The flavour data indicated that the samples had a greater than moderate intensity of late run flavour with the samples being described in the majority of times as being burnt, sour or fishy. Although the combination of tripolyphosphate and sodium chloride has been demonstrated to improve preference ratings for flavour and acceptability in many food products (Schults et al., 1976; Burgin et al., 1988), the late-run flavour and overall acceptability of the chum as viewed by the panelists were not improved by this treatment. In the fourth pilot study, though the late-run flavour of the fish was significantly reduced after two washes with brine, the texture was not improved. The textural properties were not affected significantly although the brine treated samples were 68 scored lower than the untreated samples. This was in contrast with studies conducted by Huynh (1988) who found that both firmness and late-run flavour of the late-run chum salmon were significantly improved by the brine treatment. It was possible that the poorer quality fish used in this study was responsible for these differences. Comments offered by most panelists described samples as having a salty/briny flavour, while the samples from other treatments were not similarly criticized. It is possible that this salty taste might have masked some of the attributes of late-run flavour present in the samples, causing panelists to be l e f t with the impression that the quality was better. The mean scores for the textural properties of the four treatments as judged by the panelists in decreasing order were pouch > control > phosphate > brine (Figures 7, 8, 9 and 10). As to the presence of late-run flavour, the order was control > phosphate > brine > pouch (Figure 11). The pouch samples were the most acceptable and the control samples the least (Figure 12) C. INSTRUMENTAL ANALYSIS (i) General Observations Bilinski et al. (197 7) noted that reproducibility of objective texture measurements can be greatly affected by the sampling 69 Figure 7: Bar diagram showing mean panel ratings for firmness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 70 Figure 8: Bar diagram showing mean panel ratings for dryness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 5 C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 71 Figure 9: Bar diagram showing mean panel ratings for fibrousness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 5 72 Figure 10: Bar diagram showing mean panel ratings for chewiness of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 5 73 Figure 11: Bar diagram showing mean panel ratings for l a t e -run flavour of brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 4 -Lt D o > < LL z D Lt I 111 I-< 2 -1 " 0 C O N T R O L P O U C H P H O S P H T R E A T M E N T S BRINE 74 Figure 12: Bar diagram showing mean panel ratings for ove r a l l a c c e p t a b i l i t y of brine, polyphosphate/ brine, pouch and untreated (control) samples of processed chum salmon. C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 75 procedure. He advised that pooling flesh from several fish must be avoided during texture measurement because pronounced differences could exist between individual fish subjected to identical treatments before canning. Since prolonged mixing of flesh before testing is not desirable, i t is usually difficult to obtain a sufficiently uniform sample when a pooled material is used. As a result, samples compared for textural differences in this study all came from the same fish. Bilinski et al. (1977) also mentioned that the presence in the flesh of other fish constituents such as bones can introduce serious variations. In this study, the bones and skin of the fish were removed in all cases except when i t was necessary to compare the boneless-skinless steaks to the steaks with skin and bone. This should be taken into consideration in interpreting results of textural comparisons between the boneless-skinless steaks and the steaks with skin and bone. Furthermore, the above mentioned authors suggested that incomplete draining of the can content or drying of the fish can cause variations in the determination of firmness. Can contents were properly drained and efforts were made to avoid drying of samples as much as possible. However, since samples from both the instrumental and sensory analysis came from the same can, instrumental measurements of texture were only possible after sensory analysis was completed. This slight delay could have resulted in some drying of the fish flesh 76 and may have produced higher standard deviations than expected. This should be noted in intrepreting the results of texture measurement. Both Bilinski et al. (1977) and Borderias et al. (1983) found that thorough flaking of the fish samples was necessary to improve the homogeneity. Samples were therefore broken up into flakes using two forks as proposed by Bilinski et al. (1977). (ii) Instrumental measurement of texture Figure 13 shows a typical texture profile curve obtained for the flaked cooked salmon in this study. The definitions of the various textural parameters used were based on those of Szczesniak et al. (1963), Bourne (1968) and Henry and Katz (1969). The force at the instant of maximum compression is defined as hardness. In Figure 13 this is listed as hardness 1 in contrast to hardness 2 which is the force at maximum compression during the second bite. The slopes of the curves obtained during the fi r s t and second compression cycles were calculated at intervals of 0.4 seconds and the maximum value obtained per cycle were named maximum slope 1 and 2 respectively. They represented the firmness of the samples during the f i r s t and second bite. The area under the curve up to the point of maximum compression (that is, to the left of the vertical dotted line) represents the work done to compress and crush the food during the f i r s t bite 77 F i g u r e 13: T y p i c a l F o r c e - t i m e c u r v e s of t h e p r o c e s s e d f l a k e d salmon samples. 70 Hardness 1 60 50 -j 30 -i 20 -f 104 Hardness 2 -10 10 20 30 AO TIME (SECONDS) 7 8 and is designated as area 1. During the second bite the force at the point of maximum compression is defined as hardness 2. The area under the compression portions of the second bite is area 2 and represents the work done on the food by the machine during the second compression. Springiness is defined as the distance the food recovers its height between the end of the fi r s t rise in force above zero in the second bite and the point of maximum compression. The property of cohesiveness was calculated as the ratio of area 2 to area 1. The property of gumminess was calculated as the product of hardness x cohesiveness; and the property of chewiness was calculated as the product of hardness x cohesiveness x springiness i.e., gumminess x springiness. Tables 19 and 20 show the means of the results obtained by the instrumental analysis together with their corrresponding standard deviations for the various characteristics of the processed salmon product. The data for instrumental analysis was treated by one way analysis of variance to determine if there were differences between treatments. The results appear in Tables 21 and 22. In Tables 23-28 appears the results of the multiple comparison hypothesis test used to determine which treatments differ significantly. For the f i r s t pilot experiment, there were no significant differences in cohesiveness and springiness between the 79 Table 19: Means of re s u l t s obtained by instrumental analysis and th e i r corresponding standard deviations for boneless-skinless salmon steaks and steaks with skin and bones processed at 248°F. Parameters Treatments Boneless-skinless With skin and bone Steaks Hardness 1 70. .30(9 . 2 ) - 58, .07(10 . 3 ) * Hardness 2 50. .70(8. D - 38. ,79(9. 1 ) b Cohesiveness 0, .34(0. 0 5 ) - 0, .33(0. 0 3 ) -Springiness 4 . 38(0. 4 ) - 4. ,35(0. 7 ) -Maximum slope 1 28. .03(4. 2 ) - 23. .47(5. 3 ) b Maximum slope 2 36. ,02(9. 8 ) - 26. ,75(6. 8 ) b Gumminess 24. .19(4. 8 ) - 19 . 01(4. D h Chewiness 106. .21(27 • 2 ) - 83 . 52(26 • l ) b Units Hardness 1 and 2 : Newtons (N) Springiness : millimetres (mm) Maximum slope 1 and 2 : N/mm Gumminess : Newtons (N) Chewiness : Nmm ""Means within the same row with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t (p>0.05). 80 Table 20: Means of r e s u l t s obtained by instrumental analysis and the i r corresponding standard deviations for the brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F. Parameters Treatments Brine Phosphates/ Control Pouch brine Hardness 1 33 . 8 7 ( 1 3 • 2 ) b 32 . 0 2 ( 1 0 . 2 ) * 36 . 8 3 ( 1 0 . 8 ) " 48 . 4 1 ( 1 2 . 9 ) -Hardness 2 22 . 8 3 ( 8 . 8 ) b 21 . 4 1 ( 7 . 2 ) * 25 . 4 2 ( 7 . l ) b 32 . 1 7 ( 8 . 6 ) -Cohesiveness 0 . 3 6 ( 0 . 0 4 ) - 0 . 3 6 ( 0 . 0 4 ) - 0 . 3 8 ( 0 . 0 3 ) - 0 . 3 5 ( 0 . 07 ) -Springiness 4 . 5 4 ( 0 . 8 ) « 4 . 4 2 ( 0 . 8 ) - 4 . 6 2 ( 0 . 7 ) - 4 . 5 5 ( 0 . 7 ) -Max.slope 1 10 . 8 4 ( 4 . 7 ) b 11 . 3 4 ( 4 . 5 ) b 12 . 8 7 ( 5 . 4 )"•= 15 . 8 4 ( 5 . 0 ) -«= Max.slope 2 12 . 2 5 ( 5 . 6 ) b 11 . 9 9 ( 4 . 7 ) b 13 . 6 5 ( 5 . D b 18 . 0 1 ( 6 . 3 ) -Gumminess 11 . 9 7 ( 4 . 6 ) b 11 . 4 5 ( 3 . 5 ) b 13 . 9 6 ( 4 . 16 . 9 5 ( 5 . 5 ) - = Chewiness 54 . 7 ( 2 4 . D b 5 0 . 7 9 ( 1 9 . 6 ) b 6 5 . 6 0 ( 2 6 . 7 ) b ° 77 . 9 3 ( 2 9 . 1 ) - = Units Hardness 1 and 2 : Newtons (N) Springiness : millimetres (mm) Maximum slope 1 and 2 : N/mm Gumminess : Newtons (N) Chewiness : Nmm -tocMeans within the same row with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t (p>0.05), as tested by the hypothesis t e s t . 81 Table 21: Analysis of variance data derived from Texture P r o f i l e Analysis r e s u l t s of boneless-skinless salmon steaks and steaks with skin and bones processed at 248°P. Test Source DF Mean square F-values Hardness 1 Treatment Error 1 30 1196. 95. 227 245 12. 560-" Hardness 2 Treatment Error 1 30 1136. 75. 553 059 15. 142--Cohesiveness Treatment Error 1 30 2670. 1357. 627 885 1. 967"-Spr inginess Treatment Error 1 30 0. 0. 003 358 0. 010"-Maximum slope 1 Treatment Error 1 30 166. 22. 592 767 7. 317"-Maximum slope 2 Treatment Error 1 30 687. 71. 724 498 9 . 619"-Gumminess Treatment Error 1 30 214. 19 . 689 799 10. 843*-Chewiness Treatment Error 1 30 4117. 709 . 772 654 5. 803" - - s i g n i f i c a n t (p<0.01) - s i g n i f i c a n t (p<0.05) "-not s i g n i f i c a n t (p>0.05) 82 Table 22: Analysis of variance data for Texture P r o f i l e Analysis of brine, polyphosphate/brine, pouch and untreated (control) samples of chum salmon processed at 248°F. Test Source DF Mean square F-value Hardness 1 Treatment Error 3 124 1732 140 .317 .456 12.333"-Hardness 2 Treatment Error 3 124 728 63 .463 .127 11.540--Cohesiveness Treatment Error 3 124 3633 2348 .883 .172 1.548"-Springiness Treatment Error 3 124 0 0 .220 .599 0.367"-Maximum slope 1 Treatment Error 3 124 162 24 .023 .169 6.704--Maximum slope 2 Treatment Error 3 124 247 29 .963 .527 8.398--Gumminess Treatment Error 3 124 198 20 .450 .520 9.671"" Chewiness Treatment Error 3 124 4752 632 .087 .493 7.513--'-s i g n i f i c a n t (p<0.01) '-not s i g n i f i c a n t (p>0.05) 83 Table 23: Multiple comparison hypothesis t e s t comparing instrumental r e s u l t s of hardness 1 for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Compar ison Pouch vs Phosphate Pouch vs Brine Pouch vs Control Control vs Phosphate Brine vs Phosphate Brine vs Control F P 30.615"- 0.000 24.079"" 0.000 15.285-- 0.000 2.635"- 0.107 0.392"- 0.532 0.995"- 0.321 - - s i g n i f i c a n t (p<0.01) "-not s i g n i f i c a n t p>alpha k where alpha i s the chosen si g n i f i c a n c e l e v e l = 0.05, k i s the number of comparisons possible = 6. 84 Table 24: Multiple comparison hypothesis t e s t comparing instrumental r e s u l t s of hardness 2 for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Comparison F P Pouch vs Phosphate 29 .320-" 0. 000 Pouch vs Brine 22 .098-* 0. 000 Pouch vs Con t r o l 11 .529-" 0. 001 C o n t r o l vs Phosphate 4 .078"- 0. 046 Brine vs Phosphate 0 .510"- 0. 477 Brine vs Co n t r o l 1 .704"- 0. 194 ' - s i g n i f i c a n t (p<0.01) '•not s i g n i f i c a n t p> alpha k where alpha = the chosen s i g n i f i c a n c e l e v e l = 0.05, k = the number of comparisons p o s s i b l e = 6. 85 Table 25: Multiple comparison hypothesis t e s t comparing instrumental r e s u l t s of maximum slope 1 for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Comparison P P Pouch vs Brine 16. 536"* 0 .000 Pouch vs Phosphate 13. 409-" 0 .000 Pouch vs Control 5. 831"- 0 .017 Control vs Brine 2. 728"- 0 .101 Control vs Phosphate 1. 555"- 0 .215 Brine vs Phosphate 0. 164"- 0 .686 - " s i g n i f i c a n t (p<0.01) "-not s i g n i f i c a n t p>alpha k where alpha = the chosen sig n i f i c a n c e l e v e l = 0.05, k = the number of comparisons possible = 6. 86 Table 26: Multiple comparison hypothesis test comparing instrumental r e s u l t s of maximum slope 2 for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Comparison F P Pouch vs Phosphate 19 . 589** 0 .000 Pouch vs Brine 17 .936** 0 .000 Pouch vs Control 10 .276* 0 .002 Control vs Phosphate 1 .490"- 0 .225 Control vs Brine 1 .061"- 0 .305 Brine vs Phosphate 0 .036"- 0 .849 - - s i g n i f i c a n t (p<0.01) - s i g n i f i c a n t (p<0.05) "-not s i g n i f i c a n t p>alpha k where alpha = the chosen significance level = 0.05, k = the number of comparisons possible = 6. 87 Table 27: Multiple comparison hypothesis t e s t comparing instrumental r e s u l t s of gumminess for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Comparison F P Pouch vs Phosphate 23 .582** 0. 000 Pouch vs Brine 19 .274"- 0. 000 Pouch vs Control 6 .950"- 0. 009 Control vs Phosphate 4 .927"- 0. 025 Control vs Brine 3 .076"- 0. 082 Brine vs Phosphate 0 .217"- 0. 642 - - s i g n i f i c a n t (p<0.01) "-not s i g n i f i c a n t p> alpha k where alpha = the chosen sig n i f i c a n c e l e v e l = 0.05, k = the number of comparisons possible = 6. 88 Table 28: Multiple comparison hypothesis t e s t comparing instrumental r e s u l t s of chewiness for the brine, polyphosphate/brine, pouch and untreated (control) samples of the processed chum salmon. Compar ison Pouch vs Phosphate 18. 637"" 0 .000 Pouch vs Brine 13. 652** 0 .000 Control vs Phosphate 5. 551"- 0 .020 Control vs Pouch 3. 864"- 0 .051 Control vs Brine 3. 006"- 0 .085 Brine vs Phosphate 0. 387"- 0 .535 '"s i g n i f i c a n t (p<0.01) '-not s i g n i f i c a n t p> alpha k where alpha = the chosen sig n i f i c a n c e l e v e l = 0.05, k = the number of comparisons possible = 6 89 boneless-skinless steaks and the steaks with skin and bone. Significant differences (p<0.05) were however obtained for hardness, maximum slope, gumminess and chewlness. For the second pilot experiment, the pouched samples had higher mean scores for hardness, maximum slope, gumminess and chewiness than the control (Table 20). Analysis of variance in Table 22 indicated that container type had a significant effect on the textural properties with F values significant at p < 0.01 (Tables 23-28) for hardness 1 and 2 and p < 0.05 for maximum slope 2. There was no significant differences (p>0.05) between the samples for cohesiveness, springiness, gumminess and chewiness. For the third and fourth pilot experiments, the objective textural properties were not affected significantly (p>0.05) by the polyphosphate/brine and brine treatments respectively (Tables 23-28). These treatments however resulted in lower mean scores for all the textural parameters measured when compared to the control (Table 20). The mean results obtained by instrumental analysis of the fish in decreasing order were pouch > control > brine > phosphate for all attributes except maximum slope 1, springiness and cohesiveness (Figures 14, 15, 16, 17, 18 and 19). 90 Figure 14: Bar diagram showing instrumental measurement of hardness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 9 1 Figure 15: Bar diagram showing instrumental measurement of maximum slopes for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 9 2 Figure 16: Bar diagram showing instrumental measurement of cohesiveness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 09 UJ z UJ > CO UJ I o o 1.00 0.90 -0.10 -0.00 C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 93 Figure 17: Bar diagram showing instrumental measurement of springiness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 6 94 Figure 18: Bar diagram showing instrumental measurement of gumminess for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 2 0 C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 95 Figure 19: Bar diagram showing instrumental measurement of chewiness for brine, polyphosphate/brine, pouch and untreated (control) samples of processed chum salmon. 9 0 8 0 C O N T R O L P O U C H P H O S P H BRINE T R E A T M E N T S 9 6 D. SUBJECTIVE-INSTRUMENTAL INTERRELATIONS To determine how well panel r e s u l t s could be predicted by combining instrumental r e s u l t s , simple linear regression analyses were done between the sensory attributes and the physical parameters of texture measured with the Instron. Since i t was apparent that the various sensory panel notes were not independent of each other the mean of the panel scores for each fis h was taken as the dependent variable and the mean of the corresponding instrumental values as the independent variable. The coefficient of determination (R2) obtained for the mean panel scores and the mean instrumental scores are shown in Table 29. Based on these r e s u l t s , none of the sensory parameters were well predicted by the instrumental results. The best association appeared between sensory firmness and instrumental hardness. There were also significant relationships between (i) sensory firmness and instrumental maximum slope and chewiness (ii) sensory fibrousness and instrumental hardness, maximum slope and chewiness (iii) sensory chewiness and instrumental hardness, maximum slope and chewiness. Sensory dryness could not be correlated to any of the instrumental measurements. This lack of correlation was probably due to the fact that the instrumental t e s t measured the resistance to the applied forces i.e., mechanical properties of texture and not the expressed fluid. 97 Table 29: C o e f f i c i e n t of determination (R 2) between panel scores and instrumental texture scores. 3 Instrumentation Firmness Panel Dryness Scores Fibrousness Chewiness Hardness 1 0.459"- 0.128"- 0.291-- 0.371"" Hardness 2 0.441"- 0.104"- 0.281-- 0.373"" Cohesiveness 0.002"- 0.000"- 0.000"- 0.000"-Springiness 0.010"- 0.005"- 0.035"- 0.045"-Maximum Slope 1 0.361"" 0.033"- 0.221- 0.337--Maximum slope 2 0.340-- 0.029"- 0.209* 0.305--Chewiness 0.321"" 0.026"- 0.192- 0.283--°n=20. - " s i g n i f i c a n t (p<0.01) - s i g n i f i c a n t (p<0.05) "-not s i g n i f i c a n t (p>0.05) 98 Instrumental cohesiveness and springiness also did not correlate well with any of the sensory parameters. The mean scores for the textural properties of the four treatments as judged by the panelists in decreasing order were pouch > control > phosphate > brine. For instrumental analysis of texture, the mean re s u l t s in decreasing order for a l l textural paremeters except maximum slope 1, springiness and cohesiveness were pouch > control > brine > phosphate. The pouched and control samples were therefore ranked in the same order by both testing methods while the order was reversed for the brine and phosphate samples. However, i t should be noted that there were no significant differences between the brine and phosphate samples for any of the textural parameters measured instrumentally or by sensory analysis. This lack of differences between the samples can perhaps account for the reversed ranking order. In general, i t appeared that the Instron TPA data supported the conclusions of the sensory scores for texture. This confirmation can be an important reason for using objective methods together wth sensory methods. Nevertheless, i t i s often d i f f i c u l t to relate the physical measurement obtained by means of an instrument to any one or more of the factors influencing the general sensory impression of texture, because other properties which cannot be evaluated by instruments, such as smell and taste, can affect a person's judgement at the time of 99 tasting. The sensory evaluation was therefore a more sensitive method for evaluating texture. E. VOLUME OF COOK-OUT LIQUID Table 30 shows the volume of cook-out liquid obtained following the processing of the brine, polyphosphate/brine, pouched and untreated salmon steaks. This liquid was expressed from the fish during processing. According to Wilson (1980), i t can r e s u l t in heat transfer being a combination of conduction and convection. However, since the f i s h steaks did not move inside the can during processing, the main mechanism of heat transfer was by conduction. For the brine, polyphosphate/brine and untreated samples the volume of cook-out liquid did not change very much. The pouch process, however, resulted in a greater volume of drip loss than the others. This difference in volume was not expected and was probably due to physical pressing of the pouched samples during processing. This in part could have led to the pouch samples having a firmer, drier texture than the control. 100 Table 30: Volume of cook-out l i q u i d obtained following processing of brine, polyphosphate/brine, pouch and control samples of chum salmon at 248°F. Treatment Volume 8 of cook-out l i q u i d mL Brine 23.25(2.12) Polyphosphate/brine 23.75(2.55) Control 26.00(2.93) Pouch 43.25(3.96) 5data expressed as mean(standard deviation), n=8 101 CONCLUSIONS Enhancement of late-run chum salmon quality by the processing of boneless-skinless steaks in cans and r e t o r t pouches as well as by polyphosphate/brine dips and the treatment of steaks with an 8% s a l t solution was investigated in this study. The processed salmon was shown to exibit simple straight line semi-logarithmic heating and cooling curves during retorting. The process time for the pouches was 47.8% shorter compared to the same amount of product processed in cylindrical cans at the same temperature. This resulted in the pouch processed salmon being judged by a sensory panel to be firmer, drier, more fibrous and less chewy than the canned product and having higher mean scores for hardness, maximum slopes, gumminess and chewiness when measured instrumentally. The differences in texture might also have been in part due to physical pressing of the pouched samples during processing. This phenomenon was probably demonstrated by a greater volume of cook-out liquid being released in the pouches. Less intense late-run flavour was detected in the pouched samples than the control. However, mean panel scores indicated only moderate acceptability of the pouched product. This i s possibly due to the advanced sexual maturity of the fish used. Removal of the skin and bone did not show any significant 102 improvement i n the late-run flavour and acceptability of the salmon. The texture, late-run flavour and acceptability of the polyphosphate/brine treated samples as perceived by the panelists were not improved. Instrumental measurements also showed no improvement in the texture of these samples. Washes with an 8% brine solution did not improve the texture of the fish. This might have been due to the advanced sexual maturity of the fish used. The late-run flavour of the fish samples, however, appeared significantly reduced. Based on comments offered by panelists, i t was possible that the strong briny/salty flavour perceived might have masked some of the attributes of the late-run flavour causing panelists to be left with the impression that the quality was better. Results of linear regression analysis showed that significant relationships were obtained between sensory firmness, fibrousness and chewiness and instrumental hardness, maximum slope and chewiness. However, none of the sensory parameters, were well predicted by the instrumental results. In the majority of cases, the Instron TPA data appeared to support the conclusions of the sensory scores for texture. In summary, the most important conclusion that can be gained from this study is that there was a reduction in the thermal processing time required to achieve a similar lethality 103 for salmon in retort pouches when compared to canned salmon. This shortened process time produced significant improvement in the texture, flavour and acceptability of the pouched product over the canned samples. Nevertheless, acceptance of the canned fish was only moderate. Future studies should therefore be done using fish of less advanced sexual maturity with the hope of obtaining higher consumer acceptance of the product. Changes in flesh colour is also a major factor contributing to downgrading the value of the fish. As a result, i t is recommended that colour measurements should be done on the pouched product to determine if there is significant improvement in the colour. 104 BIBLIOGRAPHY Abou-Fadel, 0. S. and Miller, L. T. 198 3. Vitamin retention, colour and texture in thermally processed green beans and Royal Ann cherries packed in pouches and cans. J. Food Sci. 4 8: 920. Adams, J. P., Peterson, W. R. and Otwell, W. S. 1983. Processing of seafood in institutional-sized retort pouches. Food Technol. 6(5): 185 Anon. 1962. Process improves frozen fish. Food Eng. 34(9): 106. Ball, C. 0. 1923. Thermal process time for canned food. Bulletin 37. National Research Council, Washington, D.C. Bee, G. R. and Park, D. K. 1978. Heat penetration measurement for thermal process design. Food Technol. 32(6): 56. Berry, R. Jr. 1979. The sterilization of food in pouches- critical parameters for s t i l l processing. Proc. of the Conference Using the Retort Pouch World Wide- Focus on the Present with a Look to the Future. Sponsored by the Food Sciences Inst., Purdue Univ., March 14-15, p. 7. Bilinski, E., Lau, Y. C. and Jonas, R. E. E. 1977. Objective measurement and control of the firmness of canned herring. Fisheries and Marine Science Technical Report No. 727. Vancouver, B. C. Bigelow, W. D., Bohart, G. S., Richardson, A. C. and Ball, C. O. 1920. Heat penetration in processing canned foods. NCA Bull. 16-L. Borderias, A. J., Lamua, M. and Tejado, M. 1983. Texture analysis of fish fillets and minced fish by both sensory and instrumental methods, J. Food Technol. 18: 85. Bourne, M. C. 1968. Texture profile of ripening pears. J. Food Sci. 33: 223-226. Brandt, M., Skinner, E. and Coleman, J. 1963. Texture profile method. J. Food Sci. 28: 404. 105 Burgin, J. M., Rogers, R. W., and Ammerman, G. R. 198 5. Effects of salt and tripolyphosphate on some quality characteristics of breaded patties made from catfish trimmings. J. Food Sci. 50: 1598. Cardello, A. W. and Mailer, 0. 1987. Sensory texture analysis: an integrated aproach to food engineering. In 11 Food Texture: instrumental and sensory measurement." Moskowitz, H. R. (ed.) Marcel Dekker Inc., New York, New York, p. 177-215. Chen, T. S. and George, W. L. 1981. Ascorbic acid retention in retort pouch green beans. J. Food Sci. 46: 462. Chia, S. S., Baker, R. C. and Hotchkiss, J. H. 198 3. Quality comparison of thermoprocessed fishery products in cans and retortable pouches. J. Food Sci. 48: 1521. Civille, G. V. and Szczesniak, A. S. 1973. Guidelines to training a texture profile panel. J. Texture Studies 4: 204-233. Connell, J. J. and Howgate, P. F. 1971. In " Fish inspection and quality control," Kreuzer, R. (ed.) Fishery News Books Ltd., London. Connell, J. J. 1980. Advances in Fish Science and Technology. Fishery News Books Ltd., Farnham, Surrey, England. Crandall, J. V. 1946. The story of the pacific salmon. Benford and Mort Publishers, Portland, Oregon, p. 27-29. Davis, J. T. 1981. The retortable pouch - a product in search of a home. Activities Rep. of R & D Associates for Military Food and Packaging systems, Inc. 33(1): 11 Dunajski, E. 1979. Texture of fish muscle. J. Text. Stud. 10: 301. Dymit, J. M. 1983. Today's food service system. Food Pack. 25(2): 56. Ecklund O. F. 1956. Correction factors for heat penetration thermocouples. Food Technol. 10(1): 4 3. Ellinger, R. N. 19 7 2. " Phosphates as Food Ingredients." CRC Press, Cleveland, OH. pl31-14 7. English, P. M., Gerdes, D. L., Finerty, M. W. and Grodner, R. M. 198 8. Effects of tripolyphosphates dips on the quality of thermally processed mullets {magil cephalus). J. Food Sci. 53: 1319. 106 Friedman, H. H., Whitney, J. E. and Szczesniak, A. S. 196 3. The texturometer - a new instrument for objective texture measurement. J. Food Sci. 28: 390. Goldblith, S. A. 1971. A condensed history of the science and technology of thermal processing - part 1. Food Technol. 25(12): 44. Goldblith, S. A. 1972. A condensed history of the science and technology of thermal processing - part 2. Food Technol. 26(1): 64. Goldfarb, P. L. 1970. Pouch for low acid foods. Part 1. Mod. Pack. 43(12): 70. Hayakawa, K. I. 1978. A critical review of mathematical procedures for determining proper heat sterilization process. Food Technol. 32(3): 59 Heintz, D. A. 1980. Marketing opportunities for the retort pouch. Food Technol. 34(9): 32. Henry, W. F. and Katz, M. H. 1969. New dimensions relating to the textural quality of semi-solid foods and ingredient systems. Food Technol. 23(6): 822-825. Herndon, D. H., Griffin, R. C, Ball, C. O. 1968. Use of computer-derived tables to calculate sterilizing processes for packaged foods. Food Technol. 22(4): 474. Howgate, P. 1977. Aspects of fish texture, in "Sensory properties  of food," Birch, G. G., Brennan J. G., Parker, K. J. (eds.), Applied science Publishers Ltd., London, p. 249. Howgate, P. 1979. Fish In "Food Microscopy," Vaughan, J. G. (ed.), Academic press, London, p. 34 3-392. Huynh, M. D. 1988. Study on utilization of late-run chum salmon. Progress Report. Division of Applied Biology, British Columbia Research Corporation, Vancouver, B. C. Huynh, M. D. 1989. Utilization of late-run chum salmon. A potential food resource. Technical Report No. 22. Fisheries and Food Products Program. Applied Biology Division. British Columbia Research Corporation, Vancouver, B. C. 107 Josephson, D. B., Lindsay, R. C. and Stuiber, D. A. 1984. Variations in the occcurences of enzymically-derived volatile aroma compounds in salt and freshwater fish. J. Agric. Pood Sci. 32: 1344-1347. Lampi, R. A. 1977. Flexible packaging for thermoprocessed foods. Adv. Food Res. 23: 305. Lampi, R. A. 1980. Retort pouch. The development of a basic packaging concept in today's high technology era. J. Food Process Eng. 4: 1. Larmond, E. 1977. "Laboratory methods for sensory evaluation of foods." Publication 1637/E. Food Res. Inst. Canada. Dept. of Agriculture, Ottawa. Lopez, A. 1987. A complete course in canning. 12th ed. The Canning Trade, Baltimore, MD. Love, R, M. 1980. The chemical biology of fishes, vol. 2. Academic Press, London. Love, R. M., Robertson, J., Smith, G. L. and Whiltle, K. J. 1974. The texture of cod muscle. J. Text. Stud. 5: 201-212. Lund, D. B. 1977. Design of thermal process for maximizing nutrient retention. Food Technol. 31 (2): 71. Lyon, B. G. and Klose, A. A. 1981. Effect of heat processing in cans and retort pouches on sensory properties of fowl meat. J. Food Sci. 46: 227. MacCallum, W. A., Chalker, D. A., Lauder, J.T., Odense, P. H., Idler, O. R. 196 4. Polyphosphate treatment of frozen cod. 3. Taste panel evaluation, chemical assessment and thaw drip in once frozen Newfoundland trap-caught cod. J. Fish. Res. Board Can., 21(6): 1397. Mermelstein, N. H. 1978. Retort pouch earns 1978 IFT Food Technology Industrial Achievement Award. Food Technol. 32(6): 22. Morello. G. F. 1987. Temperature and heat transfer studies in a water immersion retort. M. Sc. Thesis, University of British Columbia, Vancouver, B. C. 108 National Canners Association. 1976. Processes for low-acid canned foods in metal containers. Bulletin 26-L (eleventh edition). National Canners Association, Washington, D. C. Pflug, I. J. 196 4. Evaluation of heating media for producing shelf stable food in flexible packages. Phase 1. Final Report DA -19 - 129 - AMC - 14 5(N). U. S. Army Natick Laboratories, Natick, MA. Pflug, I. J. 1987. Factors important in determining the heat process value, F T, for low acid canned foods. J. Food Prot. 50: 528-533. Pflug, I. J. and Esselen, W. B. 1980. Heat sterilization of canned food. In " Fundamentals of food canning technology." Jackson, J. M. and Shinr, B. M. (eds.), AVI publishing company Inc., Westport, CT, p. 10-94. Pflug, I. J., Bock, J. H. and Long, F. E. 1963. Sterilization of foods in flexible packages. Food Technol. 7(9): 87. Ramaswamy, H. S. 1983. Heat transfer studies of steam/air mixtures for processing in retort pouches. Ph. D. thesis, University of British Columbia, Vancouver, B. C. Rasekh, J., Kramer, A., and Finch, R. 1970. Objective evaluation of canned tuna sensory quality. J. Food Sci. 35: 417. Rivzi, S. H. and Action, J. C. 1982. Nutrient enhancement of the thermostabilized food in retort pouches. Food Technol. 36(4): 105. Schults, G. W., Howker, J. J., and Wierbicki, E. 1976. Effects of salts and sodium tripolyphosphate on texture, organic volatiles and sensory characteristics of irradiated and nonirradiated pork. J. Food Sci. 41: 1096. Segars, R. A., Johnson, E. A., Kapsalis, J. G. and Peley, M. 1981. Some tensile characteristics of raw fish flesh. J. Text. Stud., 12: 375-387. Sikorski, Z. E., Scott, N. E. and Bulsson, D. H. 1984. The role of collagen in the quality and processing of fish. Crit. Rev. Food Sci. Nutr. 20: 301 - 34 3. Smith, T. and Tung, M. A. 198 2. Comparison of formula methods for calculating thermal process lethality. J. Food Sci. 47: 626. 109 Smith, T. 1987. Thermal processing analysis system version 4 lb. Program serial no. 41005, Surrey, B. C. Spinak, S. H. and Wiley, R. C. 1982. Comparision of the General and Ball formula methods for retort pouch process calculations. J. Food Sci. 47: 880. Spinelli, J., Pelroy, G. and Miyauchi, D. 1968. Effects of polyphosphate dips on fish quality. Fish. Ind. Res., 4: 37. Staruszkiewicz, W. F. and Bond, J. F. 1981. Gas chromatographic determination of caderverine, putrescine and histamine in foods. J. A. 0. A. C. 64(3): 584-591. Steffe, J. F., Williams, J. R., Chhinnan, M. S. and Black, J. R. 1980. Energy requirements and costs of retort pouch vs can packaging systems. Food Technol. 3 4(9): 39. Stone, H., Sidel, J., Oliver, S., Woolsey, A. and Singleton, R. C. 197 4. Sensory evaluation by quantitative descriptive analysis. Food Technol. 28(11): 2 4. Stumbo, C. R. 197 3. "Thermobacteriology in Food Processing," 2nd ed. Academic Press, New York, NY. Stumbo, C. R. and Longley, R. E. 1966. New parameters for process calculation. Food Technol. 20(3): 343. Suzuki, T. 1981. Fish and krill protein processing technology. Applied Science Publishers Ltd., London. Szczesniak, A. S. 1963. Classification of textural characteristics. J. Food Sci. 28: 385. Szczesniak, A. S. 1975. General Foods Texture Profile revised -ten years perspective. J. Text. Stud. 6: 5-17. Tarr, H. L. A., Gardener, L. J., and Ingram, P. 1969. Pacific cod muscle 5'-nucleotidase. J. Food Sci. 34: 637. Tarr, H. L. A. 1971. Effects of processing on fish products. In "Nutritional evaluation of food processing." Jarvis, R. S. and Von Coescke, J. (eds.), AVI Publishing co., Westport, Connecticut, p. 28 3-301. 110 Teixeira, A. A., Zinsmeister, G. E. and Zahradnik, J. W. 1975. Computer simulation of variable retort control and container geometry as a possible means of improving thiamine retention in thermally processed foods. J. Food Sci. 40: 656. Tung, M. A., Garland, M. R. and Campbell, W. E. 1975. Quality comparison of cream style corn processed in rigid and flexible containers. J. Inst. Can. Sci. Technol. Aliment. 8: 211. Tung, M. A. and Garland, T. D. 1978. Computer calculation of thermal processes. J. Food Sci. 4 3: 36 5. Van Den Broek, C. J. H. 1965. Fish canning. In "Fish as Food." Vol.IV. Processing: part 2. Borgstom, G. (ed.), Academic Press Ltd., New York, NY, p. 127-205. Voisey, P. W. 1971. Modernization of texture instrumentation. J. Text. Stud. 2: 129-195. Wekell, J. C. and Teeny, M. F. 1988. Canned salmon curd reduced by polyphosphate. J. Food Sci. 53: 1009. Wesson, , J. B., Lindsay, R. C. and Stuiber, D. A. 1979. Discrimination of fish and seafood quality by consumer population. J. Food Sci. 4 4: 878. Wilkinson, L. 1988. SYSTAT: the system for statistics. Evanston, IL: SYSTAT, Inc. Williams, J. R., Steffe, J. K. and Black, J. R. 1982. Economic comparison of canning and retort pouch systems. J. Food Sci. 47: 284. Williams, J. R., Steffe, J. F. and Black, J. R. 1983. Sensitivity of selected factors on costs of retort pouch packaging systems. Food Technol. 37(4): 92. Wilson, D. C. 1980. Theoretical problems in pouch processing. Paper presented at 1980 Winter Meeting Amer. Soc. Agri. Eng., Chicago, IL, Dec. 2-5. I l l APPENDICES APPENDIX 1: Definitions of terms and symbols. fn - Heating rate index. The number of minutes required for the straight line portion of the heating curve plotted on semi-logarithmic paper to pass through one log cycle. It is the negative reciprocal slope of the heating curve. f«= - Cooling rate index. The number of minutes required for the straight line portion of the semi-logarithmic cooling curve to tranverse one log cycle. It is the negative reciprocal slope of the cooling curve. j - A number representing the curved section before the semi-logarithmic heating or cooling curve assumes straight line characteristics. JOK - Heating lag factor. The j of the semi-logarithmic heating curve. j e a - Cooling lag factor. The j of the semi-logarithmic cooling curve. Fo - Process lethality. The equivalent, in terms of minutes at 121.1°C (250°F), of all lethal heat received by the cold spot in a container. z - The number of degrees for the thermal destruction curve to tranverse one log cycle. More information on the derivation and use of these terms can be found in and Stumbo (197 3) and Lopez (1987). 112 APPENDIX IL Example: Estimated Process Lethality Calculations for cans Fo determined for 307 x 115 cans containing chum salmon processsed at 248°F during process determination work. CALCULATION OF PROCESS LETHALITIES M u l t i p l y fh by : 1.00 M u l t i p l y jch by : 1.50 2 : 18.00 F Retort Temperature : 248.00 F I n i t i a l Product Temperature : 50.00 F Process Time : 65.00 min t Identl fh Jch fc j c c B a l l ' s Stumbo's (corr) 1 R l - C l s a l c a n 29.02 2.913 34. 64 1 .792 7. 58 8.90 2 R1-C2 salcan 30.94 2.605 33. 38 1 .747 6. 66 7.90 3 R1-C3 sal c a n 4 R1-C4 sal c a n 31.97 2.659 34. 47 1 .613 5. 72 6.63 5 R1-C5 sal c a n 30.90 2.950 36. 67 1 .762 5. 90 7.05 6 R1-C6 salcan 30.74 3.246 37. 93 1 .788 5. 43 6.58 7 R1-C7 sal c a n 29.84 3.163 37. 83 1 .675 6. 33 7.36 8 R1-C8 salcan 9 R1-C9 sal c a n 10 R2-C1 sal c a n 28.93 2.784 29. 68 1 .360 7. 96 8.46 11 R2-C2 salc a n 30.67 2.175 32. 05 1 .298 8. 11 8.51 12 R2-C3 sa l c a n 30.85 2.262 31. 70 1 .367 7. 69 8.23 13 R2-C4 salc a n 14 R2-C5 salcan 32.72 2.299 36. 59 1 .308 6. 11 6.50 15 R2-C6 sal c a n 33.20 1.894 33. 86 1 .718 7. 08 8.32 16 R2-C7 sa l c a n 32.92 1.945 43. 95 1 .292 7. 10 7.45 17 R2-C8 salcan 33.33 2.001 36. 33 1 .527 6. 61 7.41 18 R2-C9 sal c a n 19 R2-C10 sal c a n 20 R3-C1 sal c a n 32.65 2.016 36. 90 1 .553 7. 05 7.95 21 R3-C2 sa l c a n 33.40 1.779 40. 17 1 .513 7. 37 8.22 22 R3-C3 sal c a n 23 R3-C4 salc a n 32.53 2.188 37. 26 1 .612 6. 59 7.55 24 R3-C5 salcan 34.15 1.918 41. 78 1 . 560 6 . 30 7.18 25 R3-C6 sa l c a n 33.23 1.962 38. 08 1 .485 6. 82 7.55 26 R3-C7 salc a n 34.37 1.917 41. 01 1 .432 6. 16 6.80 27 R3-C8 salc a n 28 R3-C9 sal c a n MINIMUM 28.93 1.779 29. 68 1 .292 5. 43 6.50 MAXIMUM 34.37 3.246 43. 95 1 .792 8. 11 8.90 MEAN 31.91 2.352 36. 54 1 .547 6. 77 7.61 STAND. DEV. 1.65 0.472 3. 64 0 . 171 0. 76 0.72 113 APPENDIX 111. Example: Estimated Process Time Calculations for Pouches. Process times required to achieve Fo=6.50 min for pouches containing chum salmon processed at 2 4 8°F during process determination work. CALCULATION OF PROCESS TIMES M u l t i p l y fh by : 1.00 M u l t i p l y j c h by : 1.00 Z : 18.00 F R e t o r t Temperature : 248.00 F I n i t i a l P r o d u c t Temperature : 50.00 F T a r g e t L e t h a l i t y : 6.50 min » I d e n t l fh j c h f c j c c B a l l ' s Stumb< 1 R4-C1 s a l p o u c h 2 R4-C2 s a l p o u c h 19. 56 0 .449 9 . 68 1 .346 30. 48 29 .85 3 R4-C3 s a l p o u c h 16. 78 0 .580 15. 50 1 . 102 29. 57 29 .51 4 R4-C4 s a l p o u c h 20. 07 0 .406 14. 45 0 .910 30. 10 30 .68 5 R4-C5 s a l p o u c h 19. 86 0 .535 14 . 10 1 .195 32. 28 32 .05 6 R4-C6 s a l p o u c h 19. 59 0 .555 14 . 39 1 .233 32. 31 31 .97 7 R4-C7 s a l p o u c h 8 R4-C8 s a l p o u c h 9 R4-C9 s a l p o u c h 10 R5-C1 s a l p o u c h 18. 48 0 .565 15. 91 1 .103 31. 26 31 .25 11 R5-C2 s a l p o u c h 20. 26 0 .486 24. 02 1 .133 31. 86 31 .80 12 R5-C3 s a l p o u c h 20. 32 0 .497 18. 93 1 .501 32. 11 31 .07 13 R5-C4 s a l p o u c h 14 R5-C5 s a l p o u c h 19. 26 0 .480 15. 54 1 .172 30. 74 30 .57 15 R5-C6 s a l p o u c h 17. 81 0 .583 17. 03 1 .186 30. 77 30 .53 16 R5-C7 s a l p o u c h 17 R5-C8 s a l p o u c h 18 R6-C1 s a l p o u c h 18. 65 0 .542 13. 91 1 .637 31. 11 29 .79 19 R6-C2 s a l p o u c h 17. 74 0 .559 16. 79 1 .834 30. 37 28 .66 20 R6-C3 s a l p o u c h 21. 02 0 .494 17 . 56 1 .897 32. 76 30 .73 21 R6-C4 s a l p o u c h 17. 67 0 .538 25. 20 1 .348 30. 00 29 . 36 22 R6-C5 s a l p o u c h 19. 76 0 .560 13. 03 1 .919 32. 57 30 .57 23 R6-C6 s a l p o u c h 24 R6-C7 s a l p o u c h MINIMUM 16. 78 0 .406 9 . 68 0 .910 29. 57 28 .66 MAXIMUM 21. 02 0 . 583 25. 20 1 .919 32 . 76 32 .05 MEAN 19. 12 0 .522 16 . 40 1 .368 31. 22 30 .56 STAND. DEV. 1. 21 0 .051 3 . 98 0 .318 1. 03 0 .99 MEAN t 3STDEV 34. 32 33 .54 114 APPENDIX IV. Example: Estimated Process Lethality Calculations for Pouches Predicted Fo values obtained if the recommended process was used for pouches containing chum salmon processed at 248°F. CALCULATION OF PROCESS LETHALITIES Mu l t i p l y fh by : 1.00 Multiply jch by : 1.00 Z : 18.00 F Retort Temperature : 248.00 F I n i t i a l Product Temperature : 50.00 F Process Time : 34.00 min * Identl fh jch 1 R4 -Cl salpouch 2 R4 -C2 salpouch 19. 56 0. 449 3 R4 -C3 salpouch 16. 78 0. 580 4 R4 -C4 salpouch 20. 07 0. 406 5 R4 -C5 salpouch 19. 86 0. 535 6 R4 -C6 salpouch 19. 59 0. 555 7 R4 -C7 salpouch 8 R4 -C8 salpouch 9 R4 -C9 salpouch 10 R5 -Cl salpouch 18. 48 0. 565 11 R5 -C2 salpouch 20. 26 0. 486 12 R5 -C3 salpouch 20. 32 0. 497 13 R5 -C4 salpouch 14 R5 -C5 salpouch 19. 26 0. 480 15 R5 -C6 salpouch 17. 81 0. 583 16 R5 -C7 salpouch 17 R5 -C8 salpouch 18 R6 -Cl salpouch 18. 65 0. 542 19 R6 -C2 salpouch 17. 74 0. 559 20 R6 -C3 salpouch 21. 02 0. 494 21 R6 -C4 salpouch 17. 67 0. 538 22 R6 -C5 salpouch 19. 76 0. 560 23 R6 -C6 salpouch 24 R6--C7 salpouch MINIMUM 16. 78 0. 406 MAXIMUM 21. 02 0. 583 MEAN 19. 12 0. 522 STAND. DEV. 1. 21 0. 051 115 fc jcc Ball's Stumbo's 9 .68 1 .346 8.64 9 .15 15 .50 1 .102 9.38 9 .49 14 .45 0 .910 8.86 8 .54 14 .10 1 .195 7.51 7 .72 14 .39 1 .233 7.50 7 .77 15 .91 1 .103 8.18 8 .25 24 .02 1 .133 7.76 7 .86 18 .93 1 .501 7.60 8 .33 15 .54 1 .172 8.48 8 .67 17 .03 1 .186 8.53 8 .75 13 .91 1 .637 8.27 9 .24 16 .79 1 .834 8.79 10 .08 17 .56 1 .897 7.20 8 .56 25 .20 1 . 348 9.05 9 .57 13 .03 1 .919 7. 34 8 .69 9 .68 0 .910 7.20 7 .72 25 .20 1 .919 9.38 10 .08 16 .40 1 .368 8.20 8 .71 3 .98 0 . 318 0.68 0 .69 APPENDIX V. Example. Questionnaire for Sensory Analysis EVALUATION OF TEXTURE ATTRIBUTES Name: Date: Please evaluate the firmness, dryness, chewiness and fibrousness of these samples by marking a v e r t i c a l line on the horizontal line for each sample to indicate your ratings. Label each v e r t i c a l line with the code number of the sample i t represents: FIRMNESS s o f t hard DRYNESS wet dry FIBROUSNESS few many CHEWINESS l i t t l e e f f o r t much e f f o r t COMMENTS: 116 FLAVOUR-COMBINED ODOUR TASTE IMPRESSION Please evaluate the samples given for the presence of late-run flavour by placing a slash on the horizontal line to indicate your ratings. Comment on the character of the flavour perceived i.e., b i t t e r , burnt, sour, fishy, etc. LATE-RUN FLAVOUR none moderate extra strong OVERALL ACCEPTABILITY unacceptable acceptable COMMENTS: 117 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098099/manifest

Comment

Related Items