Preservation of Marine Products by Salting and Drying
PRESERVATION OF MARINE PRODUCTS BY SALTING AND DRYING
by
NORYATI ISMAIL (M.Sc., University of Nottingham)
A thesis submitted in fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY UNIVERSITY OF NEW SOUTH WALES
March 1990 UNIVERSITY OF N.S.W.
2 1 MAk 1991
LIBRARY ABSTRACT
Dried, salted morwong, shark and sardine salted in saturated brine at 30°C and dried at different temperatures under ambient
RH were prepared. Dried squid was prepared without salting.
Salt uptake was very rapid and highest in shark followed by morwong and sardine. Moisture loss was highest in sardine
followed by shark and morwong.
Squid attained the highest drying rates amongst all the species
and at 50°C its rate was the highest followed by shark, morwong
and sardine. A drying temperature of 50°C gave a compromise between product quality and drying rates.
Lightly salted products were significantly (p < 0.01) prefe^ed in
all species. Products dried at lower temperatures were
significantly more acceptable (p < 0.01) for shark and sardine but not for morwong and squid. Products salted for 8h (morwong
and sardine) 4h (shark) and non-salted (squid) and dried at 50°C were used for storage studies. Storage at 5°C was superior to that at 25°C or 37°C in terms of product appearance, browning, rancidity, moisture loss, product texture and rehydration behaviour.
The effects of salting, drying and storage on the protein properties were demonstrated by decreased protein solubility in
KCl and SDS + B mercoptoethanol, disappearance and lost intensity of some bands in the IEF pattern of water soluble proteins in all the species. However, changes in in vitro protein digestibility and amino acid contents were not significant. pH declined during salting in all the species. TVB and TMA contents in shark decreased but increased initially in sardine and morwong with subsequent decline during further salting.
During drying TVB and TMA rose increasingly (p < 0.01) with increasing temperature of drying, increases in TVB and TMA were also observed on storage (p < 0.01).
Shark, morwong and sardine had two endothermic peaks in their DSC thermograms, at 146°-50°C (myosin) and 72°-80°C (actin). Squid had three endothermic peaks at 37°, 43° and 80°C. The first peak disappeared after 8-12h salting and Tmax of peak II (actin) in shark, morwong and sardine decreased as did peak area (i.e. A HQ)
During drying, peak areas decreased and peaks broadened, to a greater degree with increasing drying temperature. Drying effects were more acute in shark and squid. These effects reflect denaturation of protein due to salt and higher temperature.
SEM examination of the fish tissues showed the effects of disruption due to salting and the reduction in compactness due to drying.
The composition of these products was comparable with that of some commercial products from South East Asia. Dried salted sardines were also produced from sardines stored for 0, 1 and 2 weeks at 5°C. Products from the stored sardines were found to be of inferior quality to those made from fresh fish. ACKNOWLEDGEMENTS
I am greatly indebted to my supervisor, Prof. M. Wootton,
Senior Lecturer for his advice, suggestions and guidance throughout the research period. Dr. Wootton has unselfishly given a lot of his time and energy especially during the preparation of the manuscript.
I wish to express my gratitude to the Australian
Development Assistance Bureau (ADAB) for the award of a scholarship and financial support during my stay in Australia. I thank Prof. R. A. Edwards, Head of School of Food Technology,
University of New South Wales for allowing me the use of facilities and instruments.
Lastly, this thesis is dedicated to my family for their support and encouragement.
iv This thesis contains no material which has been accepted or submitted for the award of any degree or diploma in any university.
Furthermore, this thesis is original and contains no material, published or written by another person, except where due reference is made in the text.
(NORTATI ISMAIL)
March 1990
(v) CONTENTS
Page
Abstract i
Acknowledgements iv
Declaration vi
Table of Contents vii
List of Tables viii
List of Figures x
List of Plates xxi
List of Appendices xxiii
CHAPTER 1
INTRODUCTION 1
1.1 Fish in Civilisation 1
1.2 World Production of Dried Fish 5
1.3 Wastage 8
1.4 The Project 10
CHAPTER 2
LITERATURE REVIEW 12
2.1 General Methods of Fish Preservation 18
2.2 Salting 18
2.2.1 The Ingredient 18 2.2.2 The Preservative Action 2.2.3 Methods of Salting 20 2.2.4 Salt Penetration 22 2.2.5 Salting Equilibria 23 2.2.6 Innovative Salting Techniques 24
vi Page
2.3 Drying Processes 24
2.3.1 Preservative Role of Drying 25 2.3.2 The Theory of Drying 27 2.3.3 Factors Affecting the Drying 31 Rate of Salted Fish 2.3.4 Innovative Drying Techniques 35
2.4 Chemical Properties of Salted Dried Fish 38
2.4.1 Volatile Bases 38 2.4.2 Moisture and Salt Content 45 2.4.3 Water Activity 49 2.4.4 Proteins 58 2.4.5 Lipids 70
2.4.5.1 Rancidity in Salted Dried Fish 72
2.5 Physical Attributes of Salted Dried Fish 75
2.5.1 Colour 75
2.5.1.1 Colour definition 76 2.5.1.2 Application to fish products 77
2.5.2 Microscopic studies 82 2.5.3 Thermal studies of fish 84 2.5.4 Reconstitution properties 90
2.6 Organoleptic Property of Salted Dried Fish 95
2.7 Packaging and Storage of Salted Dried Fish 100
CHAPTER 3
EXPERIMENTAL 106
3.1 Processing 106
3.1.1 Fish samples 106 3.1.2 Salting 106 3.1.3 Drying 107
3.2 Chemical Analysis 107
3.2.1 Moisture content 107 3.2.2 Salt (Sodium chloride) 107 3.2.3 Protein content 108 3.2.4 Fat content 108 3.2.5 Water activity (Aw) 109
vii Page
3.3 Solubility of Proteins 110 3.4 Digestibility of Protein by Pepsin 110 3.5 Isoelectric Focussing 112
3.5.1 Materials 112 3.5.2 Reagents 112 3.5.3 Sample preparation 113 3.5.4 Methods 114
3.5.4.1 Casting the gel 114 3.5.4.2 Moulding 114 3.5.4.3 Running the gels 114 3.5.4.4 Fixing and staining 115
3.6 Total Amino Acid Analysis 116
3.6.1 Materials 116 3.6.2 Reagents 116 3.6.3 Reagent preparation 117
3.6.3.1 Constant boiling HCl 117 3.6.3.2 Ninhydrin 117 3.6.3.3 Buffer A 118 3.6.3.4 Buffer B 118
3.6.4 Methods 118
3.6.4.1 Acid hydrolysis 118 3.6.4.2 Preparation for chromatography 119 3.6.4.3 Chromatographic conditions 119
3.7 Rancidity 120
3.8 Scanning Electron Microscopy 121
3.9 Colour 121
3-io Differential scanning colorimetry 121 3.10.1 Equipment and materials 121 3.10.2 Procedure 122
3.11 Reconstitution Properties 123
3.12 Sensory Evaluation 123
3.13 Storage Studies 124
3.14 Statistical Analyses 124
viii Page
CHAPTER 4
RESULTS AND DISCUSSION 125
4.1 Processing of Fish 125
4.1.1 Morwong 125
4.1.1.1 Brining 128 4.1.1.2 Drying of morwong fillets 131 4.1.1.3 General remarks on the dried 134 products 4.1.1.4 Sensory evaluation of morwong 136 4.1.1.5 Chemical properties of salted 139 dried morwong
4.1.1.5.1 Effects of salting 139 4.1.1.5.2 Effect of drying on 149 morwong
4.1.1.6 Storage studies on morwong 157
4.1.1.6.1 Evaluation of stored 157 morwong
4.1.1.7 Chemical properties of stored 171 morwong
4.1.2 Shark 185
4.1.2.1 Chemical composition of shark 185 4.1.2.2 Salting of shark 187 4.1.2.3 Drying studies 188 4.1.2.4 Product quality 190 4.1.2.4 Sensory evaluation of the 191 dried product 4.1.2.6 Chemical changes of ^elf-ed 193 dried shark
4.1.2.6.1 Effects of salting 193 4.1.2.6.2 Changes to salted shark 199 during drying
4.1.2.7 Storage Studies on Shark 206 4.1.2.7 Evaluation of the product 207
4.1.2.7.2 Chemical properties of 213 stored dried shark
ix Page
4.1.3 Sardine 223
4.1.3.1 Salting of sardines 225 4.1.3.2 Drying of sardines 226 4.1.3.3 Product quality 227 4.1.3.4 Sensory evaluation of dried sardines 229 4.1.3.5 Chemical properties of salted 231 dried sardine
4.1.3.5.1. Effects of salting 231 4.1.3.5.2. Effects of drying on 238 sardine
4.1.3.6 Storage studies 244 4.1.3.7 Chemical Properties of stored 267 sardines
4.1.4 Squid 267
4.1.4.1 Chemical composition of fresh squid 268 4.1.4.2 Drying of squid and product quality 268 4.1.4.3 Sensory evaluation: Acceptability 272 vs drying temperature 4.1.4.4 Chemical properties in dried squid 273 4.1.4.4 Storage studies on dried squid 281 4.1.4.5 Chemical properties of stored 287 dried squid
4.1.5 Salting and drying of aged fish 296
4.1.5.1 Effect of fish freshness on 297 salt uptake 4.1.5.2 Drying of the aged fish 301 4.1.5.3 Chemical properties of salted 301 aged fish 4.1.5.4 Product quality 318
4.2 Differential Scanning Calorimetry 322
4.2.1 Effect of salting on the thermogram of fish 327 4.2.2 Effects of drying on DSC thermogram of fish 335
4.3 Analyses of Commercial Dried Products 342
CHAPTER 5
CONCLUSION 345
REFERENCES 346
APPENDICES
x LIST OF TABLES
Page
Table 1.1: Total world fish production and 6 disposition.
Table 1.2 Total regional production of dried, salted 7 or smoked fish.
Table 1.3 Major fishing nations arranged by 1984 9 landings.
Table 2.1 Rough guide to the good quality storage 28 life of dried fish with different water and salt contents.
Table 2.2 Effect of humidity on final water content 33 of elan fish.
Table 2.3 Expansion of guide to shelf-life of cured 49 fishery products with differnt moisture and salt contents.
Table 4.1: Composition of amino acids in fresh shark, 127 morwong, sardine and squid (g/16gN)
Table 4.2: In vitro digestibility of fish species. 128
Table 4.3: Aw and moisture contents in dried morwong. 137
Table 4.4: Mean scores for saltiness in dried 138 products.
Table 4.5: Mean scores for salted dried products. 141
Table 4.6: In vitro digestibility of morwong 149
Table 4.7: Total amino acid content in salted morwong. 151
Table 4.8: In vitro digestibility of morwong. 156
Table 4.9: Total amino acid composition in dried 159 morwong.
Table 4.10: Hunter L, a, b for morwong stored at 5°, 160 25° and 37°C.
Table 4.11: Rehydration ratio of morwong salted for 171 different times and dried at 50°C.
xi Page
Table 4.12: Rehydration ratio for morwong salted for 8h 171 and dried at different temperatures.
Table 4.13: Rehydration ratio for stored morwong. 171
Table 4.14: TBA no. in stored fish. 176
Table 4.21: Digestibility of stored morwong. 183
Table 4.22: Total amino acid contents in stored 184 samples.
Table 5.1: Chemical composition of shark. 188
Table 5.2: A w and moisture contents of dried shark. 190
Table 5.3: Mean scores for dried salted shark. 192
Table 5.4 In vitro digestibility of shark on salting. 196
Table 5.5 Total amino acid in shark. 198
Table 5.6: In vitro digestibility of dried shark. 205
Table 5.8. Amino acid content in dried shark. 206
Table 5.9: Hunter L, a, b, for shark stored at 5°C, 208 25° and 37°.
Table 5.10: Rehydration ratio of shark salted for 212 different times and dried at 50°C.
Table 5.11: Rehydration ratio for shark salted for 4h 212 and dried at different temperature.
Table 5.12: Rehydration ratio for stored shark 213 products.
Table 5.13: Thiobarbituric acid number. 216
Table 5.14: In vitro digestibility of stored shark. 222
Table 5.15: Amino acid content shark stored at 5°, 25° 223 and 37°C for 16 wk.
Table 6.1: A w and moisture content in dried sardine. 229
Table 6.2: Mean scores of d./^ed salted sardines. 231
Table 6.4: In vitro digestibiligy of salted sardines. 237
Table 6.5: Total amino acid content in salted sardines. 238
xii Page
Table 6.6: In vitro digestibility of sardine. 244
Table 6.7: Hunter L, a, b for sardine stored at 5°, 247 25° and 37°*
Table 6.8: Rehydration ratio of sardine salted for 253 different times and dried at 50°C
Table 6.9: Rehydration ratio for sardine salted for 8h 254 and dried at different temperatures.
Table 6.10: TBA no. in stored dried sardines. 257
Table 6.11: In vitro digestibility of dried sardine on 265 storage at different temperatures
Table 6.12: Total amino acid contents in stored sardine 266
Table 7.1: A w and moisture content in dried squid. 272
Table 7.2: Mean scores for dried squid. 273
Table 7.3: Total volatile bases and trimethylamine 276 contents in squid dried products.
Table 7.4: Digestibility of dried squid. 279
Table 7.5: Composition of amino acids in dried squid. 280
Table 7.6: Hunter L, a, b for squid stored at 5°, 25° 283 and 37°C.
Table 7.7: Rehydration ratio for dried squid. 287
Table 7.8: Thiobarbituric acid number in stored squid. 289
Table 7.9: Digestibility in stored squid products. 295
Table 7.10: Total amino acids in stored squid products. 296
Table 8.1: Composition of amino acids in aged 316 sardines (jg/16gN‘.
Table 8.2: Total amino acids in salted aged sardines 317 (g/16gN) .
Table 8.3: Analyses of fresh and aged fish brined for 320 8h and dried at 40°C.
Table 9.1: Changes in aHd during salting. 334
xiii Page
Table 9.2: Changes in ethalpy of denaturation (AHD) 341 during drying
Table 10.1: Analyses of commercial and prepared dried 343 fish.
xiv LIST OF FIGURES
Page
Figure 2.1: Growth range of microorganisms with 26 respect to water activity (FAO, 1981).
Figure 2.2: Times for visible colonies of dun mould 52 (Wallemia sebi) at constant Aw with Aw of salted fish calculated from measured salt, moisture and fat contents and the isohalic soprtion isotherms for salted cod (Poulter, Doe and Olley, 1982).
Figure 2.3: Diagramatic representation of the 59 influence of water activity on chemical, enzymatic and microbial changes and on overall stability and moisture sorption properties of food products (Rockland and Nishi, 1980).
Figure 2.4: Summary of influence of heating on the 62 muscle components and properties of fish meat.
Figure 2.5: UV absorbance spectra for the heat stable 66 proteins of barramundi muscle (Keenan and Shaklee, 1985).
Figure 2.6: Schematic description of deterioration in 80 products kept at constant time and temperature (Heiss and Eichner, 1971).
Figure 2.7: Thermogram of rabbit myosin (Wright et 86 ai., 1977).
Figure 4.1: Salt uptake by different species of fish 129 during brining.
Figure 4.2: Changes in moisture content during 130 brining.
Figure 4.3: Per cent moisture loss in fish during 132 brining.
Figure 4.4: Effects of temperature on drying rates of 133 salted morwong.
Figure 4.5: Effect of species on the drying rates of 135 fish being dried at 50°C.
xv Page
Figure 4.6: pH in fish during brining. 140
Figure 4.7: TVB content in fish during brining. 143
Figure 4.8: TMA content in fish during brining. 144
Figure 4.9: Solubility of morwong proteins in 146 different media on brining.
Figure 4.10: IEF patterns of water soluble proteins of 148 morwong fillets after brining for 0, 2, 4, 8, 24, 36 and 72h .
Figure 4.11: TVB and TMA contents in morwong dried at 150 different temperature.
Figure 4.12: Solubility of morwong proteins dried at 153 different temperature.
Figure 4.13: IEF patterns of water soluble proteins of 155 morwong fillets brined for 8h and dried at 30°, 40°, 50° and 60°C.
Figure 4.14: Changes in moisture content in morwong 172 during storage at 5°, 25° and 37°C.
Figure 4.15: Changes in Aw in morwong during storage at 174 5°, 25° and 37°C.
Figure 4.16: Changes in TVB content during storage at 177 5°, 25° and 37°C.
Figure 4.17: Solubility of morwong proteins in K£l at 180 77°C during storage at 5°, 25° and 37°C.
Figure 4.18: Solubility of morwong proteins in SDS + 181 8 mercaptoethanol during storage at 5°, 25° and 37°C.
Figure 4.19: IEF patterns of morwong fillets brined for 182 8h, dried at 50°C and stored at 5°, 25° and 37°C each for 4, 12, and 24 weeks.
Figure 5.1 : Effect of temperature on the drying rates 189 of shark
Figure 5.2: Solubility of shark on salting. 195
Figure 5.3: IEF patterns of water soluble proteins of 197 shark fillers after brining for 0,2, 4, 8, 12, 18, 24, 36 and 48h.
xvi Page
Figure 5.4: Effect of drying temperature on TVB and 200 TMA contents in shark.
Figure 5.5: Effect of drying temperature on the 202 solubility of shark proteins.
Figure 5.6: IEF pattern of water soluble proteins of 204 shark fillets brined for 4 h and dried at 30°, 40°, 50°, 60° and 70°C.
Figure 5.7: Moisture changes in dried shark stored at 214 5°, 25° and 37°C.
Figure 5.8: A changes in dried shark stored at 5°, 215 25° and 37°C.
Figure 5.9: TVB changes in shark stored at 5°, 25° and 217 37°C.
Figure 5.10: Solubility of shark proteins in KCl at 219 77°C during storage at 5°, 25° and 37°C.
Figure 5.11: Solubility of shark proteins in SDS + 220 B mercaptoethanol during storage at different temperature.
Figure 5.12: IEF patterns of water soluble proteins of 221 dried shark stored at 5°, 25° and 37°C 4 weeks.
Figure 6.1: Effect of temperature on the drying rates 228 of salted sardines.
Figure 6.2: Effect of salting time on protein 234 solubility of sardine.
Figure 6.3: IEF patterns of water soluble proteins of 236 sardine after brining for 0, 2, 4, 8, 12, 24, 36, 48 and 72h.
Figure 6.4: TVB and TMA contents in sardine 239
Figure 6.5: Effect of drying temperature on the 241 solubility of proteins of sardine in KCl at 25° and 77°C and in SDS + B mercaptoethanol.
Figure 6.6: IEF patterns of water soluble proteins of 243 sardine brined for 8h and dried at 30°C, 40°, 50°, 60° and 70°C.
xvii Page
Figure 6.7: Effect of storage temperature and time on 255 moisture content in dried sardines.
Figure 6.8: Changes in Aw of dried sardines during 256 storage at different temperatures.
Figure 6.9: Changes in TVB content in dried sardines 259 on storage at different temperature.
Figure 6.10: Solubility of proteins of dired sardines 260 on storage at different temperatures in KC1 at 77°C.
Figure 6.11: Effect of storage time and temperature on 261 protein solubility of dried sardines in SDS + B mercaptoethanol.
Figure 6.12: IEF patterns of water soluble proteins of 263 sardine brined for 8h, dried at 50°C and stored at 5°, 25° and 37°C for up to 24 weeks.
Figure 7.1: Effect of temperature on drying rates of 269 squid.
Figure 7.2: Effect of drying temperature on the 275 solubility of proteins of squid.
Figure 7.3: IEF patterns of water soluble proteins of 278 squid stored at different temperatures.
Figure 7.4: Changes in moisture content in dried squid 288 during drying at 5°, 25° and 37°C.
Figure 7.5: Changes in Aw in dried squid during 288 storage at 5°, 25° and 37°C.
Figure 7.6: Changes in TVB content in dried squid 290 during storage at different temperatures.
Figure 7.7: Changes in TMA content in dried squid 291 during storages at different temperature.
Figure 7.8: Solubility of proteins of dried squid 293 stored at different temperature in SDS + B mercaptoethanol and in KC1 at 77°C.
Figure 7.9: IEF patterns of water soluble proteins in 294 squid dried at 30°, 40°,50°, 60° and 70°C (b, c, d, e, f respectively; a being fresh squid).
xviii . Page
Figure 8.1: Salt uptake by sardine of different degree 298 of freshness during brining.
Figure 8.2: Moisture content in sardine of different 299 degree of freshness during brining.
Figure 8.3: Effect of temperature on drying rates of salted aged sardine (1 week ).
Figure 8.4: Effects of temperature on drying rates of 303 salted aged sardine (2 weeks).
Figure 8.5: Effect of freshness of sardine on drying rates at 40°C.
Figure 8.6: pH of fish of different freshness during 305 brining.
Figure 8.7: TVB content in fish of different degree of 307 freshness during brining.
Figure 8.8: TMA content in fish of different freshness 308 during brining.
Figure 8.9: Effect of brining time and fish freshness 309 on the soluble proteins of sardine in KC1 at 25°C.
Figure 8.10: Effect of brining time and fish freshness 310 on the soluble proteins of sardines in KC1 at 77°C.
Figure 8.11: Effect of brining time and fish freshness 311 on the soluble proteins of sardine in SDS + 8 mercaptoethanol.
Figure 8.12: IEF patterns of water soluble proteins of 314 1 wk aged fish on being brined for 0, 2, 4, 8, 12/ 18, 24, 36 and 48h.
Figure 8.13: IEF patterns of water soluble proteins of 315 2 week aged fish on being brined for 0, 2, 4, 8, 12, 18, 24, 36 and 48h.
Figure 8.14: IEF patterns of water soluble proteins of 321 lwk aged fish, brined for 8h and dried at 30°, 40°, 50° and 60°C.
Figure 8.15: IEF patterns of water soluble proteins of 323 2 wk aged fish, brined for 8h and dried at 30°, 40°, 50° and 60°C.
xix Page
Figure 9.1: DSC thermogram of salted and unsalted 324 shark whole muscle.
Figure 9.2: DSC thermogram of salted and unsalted 326 morwong whole muscle.
Figure 9.3: DSC thermogram of salted and unsalted 328 sardine whole muscle.
Figure 9.4: DSC thermogram of 24h salted morwong. 330
Figure 9.5: DSC thermogram of 36h salted morwong. 331
Figure 9.6: DSC thermogram of 24h and 48h salted 332 sardine.
Figure 9.7: DSC thermogram of dried salted morwong 336 previously dried at different temperatures.
Figure 9.8: DSC thermogram of dried salted sardine 337 previously dried at different temperatures.
Figure 9.9: DSC thermogram of dried shark, previously 338 dried at different temperatures.
Figure 9.10: DSC thermogram of dried sguid previously 339 dried at different temperatures.
xx LIST OF PLATEX
Page
Plate 1: Dried salted morwong stored at 5°, 25° and 158 37°C.
Plate 2: SEM micrographs (150x) of the surface layers 163 of fish brined for 0, 2, 4, 8h.
Plate 3: SEM micrographs (150x) of the surface layers 164 of fish brined for 24 and 48h.
Plate 4: SEM micrographs of deep tissue of fish brined 165 for 0, 2, 4 and 8h.
Plate 5: SEM micrographs of deep tissue of fish brined 166 for 24 and 48h.
Plate 6: SEM micrographs (65x) of the surface layers 167 of fish dried at 30°, 40°, 50° and 60°C.
Plate 7: SEM micrographs (125x) of the deep tissue of 168 fish dried at 30°, 40°, 50° and 60°C.
Plate 8: SEM micrographs (150x) the shark tissues 210 brined for 0, 4, 12 and 24h.
Plate 9: SEM micrographs of shark fillets brined for 211 4h and dried at 30°, 40°, 50° and 60°C.
Plate 10: Appearance of dried sardines stored up to 245 24wk at 5°, 25° and 37°C.
Plate 11: Micrographs of sardine fresh salted for 0, 2, 249 4 and 8h.
Plate 12: Micrographs of sardine fresh salted for 24, 250 36 and 48h.
Plate 13: Micrographs showing sardine fresh brined for 252 8h and dried at 40°, 50°, 60° and 70°C.
Plate 14: Dried squid stored at 5°, 25° and 37°C for 282 4 wk 8, 16 and 24 wk.
Plate 15: SEM of squid tissue dried at different 285 temperatures fresh, 40° and 70°C.
xxi LIST OF APPENDICES
Appendix Is Standard curve for TBA no.
Appendix 2: (a) Calibration for g^latinization temperatures and enthalpy. 2 (b) The relationship between area (cm ) and mg biphenyl used for standardization in gelatinization of wheat starch observed by DSC.
Appendix 3: Sensory evaluation
xxii CHAPTER 1
INTRODUCTION
1.1 Fish in Civilisation
The role of fish in civilisation is undeniably important. The earliest evidence of this role was in the form of fish bones discovered in the refuse heaps of Late Stone Age dwellings in the caves in Dordogne, France dating back to 40,000 B.C. Artistic manifestations to this effect also appeared on the walls of caves such as Altamira and other places in Spain and France (60,000 to
10,000 B.C.) and on the walls of fjords in Norway (5,000 to 1,000
B.C.). Cutting (1955; 1962) and Kruezer (1974) wrote excellent accounts of the development of fish utilisation throughout civilisation.
The transition from nomadic to agricultural life necessitated the use of salt in man's diet during the Bronze Age. The great urban civilisations of Ancient Egypt, Mesopotamia and the Indus Valley were based on agriculture alongside of which fishing must have played an important role. Techniques for preserving fish by drying and salting were developed. Salt fish known as 'ukas' in
Egypt, Sumeria and Mesopotamia became an important supplement to the staple diet of cereals. Later when the Ptolemies held salt monopolies, guilds of fish-salters also existed. Trading in dried fish developed as sea faring and colonising activities
1 prospered. Many famous cities in the Mediterranean owed their origin to fishing settlements, including Sidon (i.e., 'the fisher's town'), Gades (Cadiz), Malaga (i.e., 'the salting place'), Sinope, on the Black Sea and even Byzantium, later
Constantinople.
During the Iron Age (commencing about 1000 B.C.) Rome was kept supplied with 'salsamentum', from its colonies, of a variety of types, made from numerous classes of fish, including tuna, swordfish, sturgeon, mackerel, pike, mullet, bream, eel, oysters, and even whales. Seals and sharks were also taken in the
Atlantic. Tuna seems to have been the most important sea fish all over the classical world, and salt fish was usually synonymous with salt tuna. Vinegar was also used extensively for shorter-term preservation. Mullet roe, when salted, was a favourite dish, but there is no mention of caviar, made from the roe of sturgeon, until the ninth century. Fresh fish could not be transported far inland on account of poor transport.
Elaborate fresh and salt-water ponds were built to maintain supplies of fresh fish. Mortality was considerable but eel, because of its durability became popular. By any standards fresh fish were a luxury. Most people, if they tasted fish at all, did so only if it were salted, brined, or pickled in vinegar.
During the Middle Ages, the geographic centre of the fishing industry shifted from the Mediterranean to the more prolific
North Sea. Right up until the railway era in about 1840, salting, drying and smoking remained the basic methods of long-
2 term preservation, with 'potting' in vinegar used for shorter preservation periods. By the end of the Middle Ages, it was possible to distinguish empirically certain sequences and combinations of treatments advantageous to particular geographical, technological and social environments, which became synonymous with the region. Thus, in the North Atlantic countries, methods of salting and drying, whether combined (as in dried salted cod, etc.), or with smoking (as in 'red herring') or alone (as in salt pickled herring and wind-dried cod
('stockfish') have persisted practically unchanged for centuries down to the present day.
Fish must have also featured quite considerably in the East.
Kruezer (1974) wrote that about 600 B.C. fish appeared in Persian records and that fish processing was recorded as early as 1500
B.C. in China. The earliest Chinese handbook on fish culture was written by Fan Li about 500 B.C. Literature on early records of fishery in South East Asia is rare. Nevertheless, one 'Malay
Fisherman: Their Peasant Economy' by Firth (1946), provided some insight into the state of fisheries in the Malayan and Indonesian regions whilst van Veen (1953) described some preserved fish products from the entire South East Asian region. Fresh fish trade existed only to a small extent due to rapid perishability.
To counter this, cooked fish was traded although it too had limited life. Far more important to most areas were cured and dried fish which were traded into the hinterland. Other products include fish sauces, pastes and crackers.
3 As early as the Bronze Age, fatty fish such as mackerel, herring and salmon, due to their potential for oxidation, were recognised as poor salters. They were found to last better pickled in brine, either in earthenware as in earlier times or in tight wooden barrels in later times.
Famine, malnourishment and misery racked mediaeval Europe. Fish were the predominant protein source for rich and poor during times of famine and shortages of meat, which often occurred in spring, Bacalao, stockfish or salted herring or cod known also as 'beef of the sea', formed the staple food of explorers on their way to new continents and of the sailors on warships fighting the wars brought on by colonising activities. Shifts in the spawning migration pattern of herrings and the desire to control the lucrative salt fish trade also encouraged the growth of such products. However, Cutting (1962) wrote that even though
'there were religious injunctions in 16th and 17th century,
England, backed by political expediency, to enforce the regular eating of fish, meaning hard smoked salted and dried products for most people, there were literary allusions that made it obvious that these were consumed with resignation rather than enthusiasm.*
Consequently, the history of the development of the fishing industry in modern times has been a succession of attempts to prevent fish from spoilage between catching and consumption.
From the impracticality of bringing fish alive from the sea, the development of icing on board evolved into freezing on board the
4 trawler. Simultaneously the advent of the railway network made it possible to distribute fish in the fresh, unpreserved condition farther inland. Thus the stage was set for the diminished importance of preserved fish and development of lightly cured specialty products such as the Gaspe cure which persists today. Preserved fish was held in contempt for its heavy salt and hard texture as exemplified by a quote from
Cutting (1955) that they 'are left in a condition that demands the greatest efforts by the stomach to extract the food value from the fibrous masses'.
Salting and drying, either individually, in combination, or along with heavy smoking, although superceded in the most advanced countries, still promise long storage in parts of the world where infrastructures of transport and handling are poor and climatic conditions unfavourable for long storage of fresh fish. In the
Third World these conditions are common.
1.2 World Production of Dried Fish
The disposition of world catch (Table 1.1) demonstrates that total catch is ever on the increase. This increase is necessitated by the demand for protein from the growing world's population whose per caput supply stood at 12.0kg in 1982, somewhat lower than the 1975-1977 average of 12.2kg per caput
(James, 1984). By the year 2000 an anticipated world population of 6.1 billion will require an additional 19 million tons of fish to maintain the present consumption levels (Clucas, 1981; James,
1984). Much of this requirement will be in the developing
5 countries which will represent about 74% of the world's population and will account for 90% of the increase. These countries, particularly South and the South East Asia will require an additional 5 million tonnes of fish to maintain present levels of consumption (James, 1984). Whilst the first demand is by humans, the second demand for protein is from the pig and poultry industries in the developed nations. These industries use high protein artificial diets in the form of fish meal (Pitcher and Hart, 1982). In fact, in Denmark two-thirds of the fish catch is used for reduction to fish meal (OECD, 1979).
Table 1.1
TOTAL WORLD FISH PRODUCTION AND DISPOSITION
Million tonnes
Disposition 1980 1981 1982 1983 1984
Total World Catch 72.0 74.9 76.6 76.8 82.8
Total Human Consumption 52.9 55.4 55.9 56.7 59.9
Fresh 15.7 17.0 16.4 16.5 17.3
Frozen 15.9 16.6 17.9 18.2 19.8
Cured 11.1 11.3 11.5 11.7 12.2
Canned 10.3 10.5 10.1 10.2 10.5
Total Animal Feed 18.4 18.7 19.9 19.4 22.0
Source: FAO, 1984
More fish is being cured each year but at a consistent proportion of about 15% of total catch (15.4% being the highest in 1980 and
6 14.8% the lowest in 1984). Whether this signals a downward trend for cured fish production remains to be seen, but it is noteworthy that the disposition for frozen products steadily increased from 22.0% in 1980 to 23.9% in 1984.
Table 1.2
TOTAL REGIONAL PRODUCTION OF DRIED, SALTED OR SMOKED FISH
100 ,000 tonnes
Region 1980 1981 1982 1983 1984
Africa 2.56 2.59 2.58 2.60 2.62
America, North 1.03 1.00 1.09 0.85 0.68
America, South 0.79 0.65 0.74 0.68 0.82
Asia 28.91 29.40 29.66 31.40 33.03
Europe 5.20 5.27 4.90 4.42 4.70
USSR 6.83 7.11 7.73 7.93 7.94
Source: FAO 1984
Asia is by far the largest producer of cured fish. Most is for internal consumption and intra-region trading (FAO, 1984, Moen,
1983) . Amongst the highest producers in Asia are China, Japan,
Indonesia, Philippines, Thailand and Korea, in that order (FAO,
1984) . In Indonesia about half the total catch is preserved by salting and drying (FAO, 1983). The total protein intake for people of Asia is among the lowest in the world being only 59.3g per caput per day (Africa takes 58.lg per caput per day), whilst the Northern Hemisphere countries and Oceania consume well above
7 90g per caput per day. Fish forms about 29.3% of the animal protein and about 5.8% of the total protein intake in Asia
(James, 1984). Fishery industries are generally labour
intensive, in Indonesia involving over one million people and in the Philippines, over seven hundred thousand (Kruezer, 1974 and
Clucas, 1981). Intensification of the fishing industry in Asia, especially the South and the South East, offers one possibility of elevating the animal protein intake in the region (van Veen,
1953).
1.3 Wastage
Currently, about two-thirds of total fish landings are in developing countries. This is further illustrated in Table 1.3, which shows that several developing countries are among the leading fishing nations of the world. Complete utilisation of the total catch is not possible even under the best of circumstances since losses occur as a result of spoilage, caused principally by lack of chilling coupled with poor storage, distribution and marketing facilities. Estimation of these losses is particularly difficult as much fish which is already spoiled is still consumed and, in addition, stale fish is often processed by drying. Even so, 10% is considered to be a conservative estimate of spoilage loss of fresh fish (James,
1984), representing about 1.7 million tonnes (for 1984) on a world wide basis.
The situation with regard to losses of cured fish is much more severe. FAO's (1981) review of the subject of losses of cured
8 fish points out that while nutritional losses as a result of the process can be significant, the major losses in quantity and quality occur as a result of blow-fly infestation during drying and attack by beetles on the dried product.
Table 1.3
MAJOR FISHING NATIONS ARRANGED BY 1984 LANDINGS
Million tonnes
1. Japan 10.6 2. USSR 10.1 3. China 6.9 4. Peru 4.3 5. Norway 3.4 6. United States 3.0 7. Korea Rp 2.4 8. India 2.4 9. Denmark 1.9 10. Thailand 1.6 11. Spain 1.5 12. Indonesia 1.4 13. Philippines 1.4 14. Chile 1.3 15. Canada 1.1 16. Vietnam 1.0
Source: FAO, 1984
Physical deterioration of dried products by crumbling and disintegration can also lead to significant losses. Accurate quantification of such losses is difficult due to the variety of products, methods of processing, packaging, local conditions and customs. In addition, much of the cured fish production is in isolated areas where data collection is difficult. A loss of 25%
9 was estimated by James in NAS (1978) on the basis of his observation and experience in the global fisheries. FAO (1981) confirms 25% as a realistic overall estimate, though emphasising the need for more precise information. At 25% the actual loss would have been 3.0 million tonnes for 1984.
1.4 The Project
Salted and dried marine products have created immense interest for food scientists and economists. Being the food that has descended the origins of civilisation it has aroused avid curiosity in the minds of scientists searching for reasons for its evolution and for means of improving the product. Society today is dedicated to the cause of eliminating hunger and malnutrition as is said in the 'Universal Declaration on the
Eradication of Hunger and Malnutrition' (FAO, 1975). Since millions of people still rely on salted and dried fish as their primary source of animal protein, it is paramount that the quality of this product be researched and improved in order to develop its value as a food. Observations and scientific research have so far indicated that much is left to be desired in terms of the whole process of preparing, storing and retailing the product.
This thesis was undertaken with an overall purpose of studying the quality of fish products which were salted and dried under controlled conditions in the laboratory. The studies would serve to alleviate some problems that are inherent in salted dried products. Four marine species were converted into salted dried
10 products by brining and drying at different temperatures.
Various quality appraisals were applied and the products judged to be the most acceptable were subjected to further tests involving storage behaviour. Recommendations on quality appraisal and processing factors which influenced the quality of the products will be made.
11 CHAPTER 2
LITERATURE REVIEW
2.1 General Methods of Fish Preservation
Fish, upon death, inevitably spoil, limiting its distribution to places away from the point of catching and restricting storage life, causing wastage in times of glut. Spoilage is brought upon by autolytic, microbial and chemical action. Autolysis is a whole series of complicated changes administered by endogenous enzymes, of which flavour changes occurring in the first few days of storage are perhaps the most significant. Microbial activity, affected by the microflora whether intrinsic or extrinsic, results in a well-defined sequence of changes in odoriferous and flavour compounds initially followed by gradual softening of the tissues and foul smell. In addition to the above, chemical changes involving oxygen from the air and the fat in the fish may result in rancid odour and flavour.
Preservation of fish may be achieved in several different ways.
Temperature control, one of the easiest factors over which control is possible, is affected by chilling in ice or refrigerated liquid media (plain or sea water) and/or freezing.
A great deal has been published on the storage of cold and temperate zone fish in ice (Shaw and Shewan, 1968; Hansen, 1972;
Disney et al. , 1974; Shewan, 1977) and tropical fish in ice
(Watanabe, 1965-1966; Disney et al., 1971; dos Santos, 1981;
12 Poulter et al., 1981). The quality of landed fish will depend on the way it is handled on board, including prompt, effective icing and proper sanitation and stowage. FAO/UN published 'Standards and requirements for the handling, processing, distribution and quality control of fish' in 1963 which stipulated various countries' requirements for their fishing industry. In general terms, for cold/temperate water fish, flat shaped fish keep far longer than round shaped fish, red fleshed fish longer than white fleshed fish, low fat fish longer than high fat fish and teleosts longer than elasmobranch (Bramsnaes, 1965). Tropical fish keeps much longer in ice (Disney et al., 1974; Nair and Dani, 1975;
Disney, 1976; Shewan, 1977; Sumner, et al., 1984) though dos
Santos (1981) questions this notion. The main explanation given for this phenomenon centres on differences in the mesophilic or mesotrophic microflora of tropical species whereas bacterial spoilage of fish in ice is attributed to psychrophilic or psychrotropic organisms. Other factors include antibacterial compounds in the surface slime (Liguori et al., 1963; Shewan,
1977), and flesh (Watanabe, 1965-1966) and qualitative differences in the microflora of the surface slime (Shewan, 1977;
Liston, 1980).
Reduction in temperature delays the resolution of rigor mortis, thereby retarding enzymic reactions and the onset of bacterial attack. The latter is believed to be because the pH during rigor* is less conducive to bacterial growth (Reay and Shewan,
1949). Enzymic reactions soften the flesh and render the desirable sweetish, meaty characteristic fish flavour more neutral and insipid with the eventual accumulation of the nucleotide breakdown product, hypoxanthine, contributing to bitter taste. The visceral enzymes responsible for the digestion of food during life are powerfully proteolytic on the organs themselves and the surrounding tissues upon death (Connell,
1980). The influence of temperature on the growth of fish spoiling bacteria is considerable and the growth is markedly curtailed by small decreases in the range: -1°C to 5°C
(Bramsnaes, 1965). The microbial load can also be significantly reduced by washing the slime off the fish and gutting and heading before chilling providing correct sanitation and handling practices are applied.
Maximisation of the shelf-life of chill-stored fish and other seafoods using modified atmosphere storage has been extensively reviewed by Sta, .tham (1984). Many gasses have been tried but carbon dioxide has been found to be the most effective for preserving fresh foods. The mechanism of inhibition of bacteria has been given various explanations, ranging from pH lowering effects of CC>2 to its effects on enzyme systems (King and Nagel,
1975; Kritzman, Chet and Henis, 1977) and on the cell membrane
(Sears and Eisenberg, 1961). Extension of the shelf-life by 6-10 days in brown shrimp (Lannelongue et al. , 1982a), swordfish
(Lannelongue et al., 1982b), finfish (Lannelongue et al., 1982c), freshwater crayfish (Wang and Brown, 1983) and perch, sea trout, croaker and blue fish (Gray, Hoover and Muir, 1983) has been reported. In other cases no growth occurred during storage of fish for 14 days (Parkin et al., 1981), 26 days (Fey, 1980) and
14 21 days (Barnett et al., 1982). There are three methods by which modified atmosphere packing maybe employed for transportation and distribution (Bell, 1982). The first involves bulk transportation in refrigerated seavans, railcars or trailers whereby the container is loaded with pre-cooled material and atmosphere injected and sealed in. This method has been successfully employed for the transportation of Pacific and
Alaskan salmon (Veranth and Robe, 1979; Bell, 1980). The second method, the master pack concept, employs permeable over-wrap packs, of the type used in supermarkets, placed in a large impermeable master pouch flushed with CC>2. This system has been successful for a Gulf-coast area processor in the USA (Bannar,
1979). The third type of package is the individual consumer pack intended for direct retail sale. In Britain, this is being used for prepackaged cod, haddock, plaice, mackerel, dover and lemon sole, crab, lobster, scampi (Tiffney and Mills, 1982) and smoked fish lines (Anon, 1983).
Records of the early commercial distribution of naturally frozen fish show that fish quality suffered drastically even if bacterial spoilage is prevented (Hansen, 1980). Similar records also come from attempts last century to use mechanical refrigeration to freeze fish for long-term preservation. Certain studies (Connell, 1968) show that deterioration during freezing and thawing IS caused by irreversible changes in the colloidal state of proteins induced by the increased concentration of solutes in the unfrozen fluid in the fish tissue. Bacterial growth on the fish will break down the structure of the tissues
15 and cause the formation of trimethylamine and other volatile substances (including dimethylamine and formaldehyde in some gadoid fish). This detracts from the texture (Sikorski, 1980) and affects the smell and taste of the fish (Regenstein et al. ,
1982). Potter (1978) showed that bacterial growth is reduced at temperatures below 0°C and is usually arrested at -10°C. Enzymic processes which are active within the fish muscle itself, including the formation of hypoxanthine, can impart bitter flavour to the fish (Jones, 1965). These processes are not prevented until the storage temperature is below -20°C (Fennema et al., 1973). Slavin (1968) showed that oxidation of fish lipids would result in rancidity even at temperatures as low as
-29°C. The time for which Indian sardines can be stored frozen before becoming unacceptable is inversely proportional to the oil content (Kaimal, 1969). Providing a barrier between the fish tissues and atmospheric oxygen by glazing the fish increases the frozen storage life (Stansby, 1982). The use of antioxidants has been studied (Pawar and Mag^r, 1966; Stansby, 1982) but this seems to have little effect. So far, there are indications that tropical fish keep longer in frozen storage (Poulter, 1978; King and Poulter, 1985*) than temperate fish.
In canning, temperature control is affected on the fish by raising it as high as possible to inactivate enzymes and microorganisms. The hermetically sealed container (can or bottle) protects the product from subsequent reinfection, atmospheric oxygen and also against damage and contamination.
16 Control of quality once again depends on the use of good raw material and effective processing (Connell, 1980; van den Broek,
1965).
Chilling, freezing and canning represent the three principle ways by which temperature is used to preserve food commodities.
However, preservation can also be afforded by the application of salt and/or wood smoke, and also with or without the complete or partial removal of water from the fish. This is the most common and established means of making products that keep for some time at ordinary temperatures. Unlike the former three methods of preservation, the latter requires very little in terms of equipment and technology and is easily executed. In most of the
Third World it is still the most commonly adopted method for long-term fish preservation as any other means would prove to be economically nonviable. FAO Fisheries Reports 160 and 279
(Watermann, 1976 and James, 1983) reveal that much of the procedure for salting and drying still remains more of an art than a technology. The technique has survived the test of time and when, or if ever, it will be developed on a sound technological basis is a matter for the technologists, economists, social scientists and politicians to consider. One characteristic of this method of preservation is that it has remained very much a subsistence activity. Thus, there is very little opportunity for improvement of the technology involved.
Sripathy (1983) noted that despite research developments in terms of processing and product quality, application of these advances is not apparent. Salted dried fish production itself has always
17 been an individual or a small group venture rather than a major industry, a situation which is likely to persist.
2.2 Salting
2.2.1 The Ingredient
Salting is both a method of preserving fish and a preliminary operation to some smoking, drying and marinading processes. It is a combination of operations aimed at preserving fish in salt, beginning with washing and gutting and ending with packing the salted fish in containers. The salt is common salt or sodium chloride, which is obtained either as solar, mineral or chemical salt. The quality and type of salt is a matter of some importance. It should be of reasonably small grain size to facilitate close contact with the fish and rapid dissolution, but not so fine as to impede drainage of expelled juices. Salt containing excessive iron or copper gives rise to unsightly yellowish or brownish colour in finished products and should be avoided. On the other hand about 0.5% calcium plus magnesium (as sulphates) in the salt impart a desirable whiteness and rigidity although higher concentrations cause excessive bitterness and brittleness (Beatty and Fougere, 1957). White fish cured with pure sodium chloride tend to be flexible and amber in colour.
2.2.2 The Preservative Action
The preservative action of salt was first thought to lie in its capacity to exert a high osmotic pressure, giving rise to plasmolysis in bacterial cells. Today it is realised that common
18 salt not only causes this plasmolysis but alters the state of the proteins and enzymes, in such a way that proteins become impervious to the action of enzymes and lose their efficacy
(Zaitsev et al. , 1969). Common salt has a bacteriostatic and a bactericidal action. However, salting alone may not stop the spoilage of fish completely no matter how high a concentration of salt is used. Microbiological changes occur mainly during the early stages of salting and drying while the water activity (Aw) level still permits this (Aw > 0.90) (Dusseault, 1958).
Conditions in wet or pickle curing are so anaerobic that growth of aerobic organisms is rarely supported. Therefore microbial deterioration is associated mainly with dry or stack curing of white fish. Fish spoilage bacteria are quite active in salt concentrations up to 6% (wet basis), but cannot survive long at salt contents above 6 to 8%. The slime-forming bacteria can grow in 6-12% salt and spoil the fish (Beatty and Fougere, 1957) while at salt concentrations between 10-15%, spoilage is often caused by the growth of 'dun' mould Wallemia sebi (syn. Sporendonema episzoum) which appears as chocolate-coloured or fawn spots on the cut surface of the fish, associated with localised decay
(Frank and Hess, 1941). Spoilage of salted fish also occurs at salt concentration > 13% due to growth of 'halophiles' or 'salt- loving' bacteria. Species of the genus Halobacterium and
Halococcus attack salted fish and produce pink discolourations
(Liston, 1980). These bacteria often are found in solar salt which is used commonly for fish processing, and are most troublesome with salted and dried-salted fish when the A w is
19 above 0.75 (FAO, 1981). Mould growth spreads over the fish and causes an increase in surface moisture which enables microorganisms such as halophilic bacteria to attack and spoil the fish. Cured fish deteriorates rapidly after the initial mould attack. Because of this, the onset of mould attack has been taken as the end of 'good quality' storage life by Poulter
(1980). Poulter, Doe and Olley (1982) used isohalic sorption isotherms for cod (Doe et al., 1982) to calculate the A^ of dried salted fish under tropical conditions, and together with known germination times for the dun mould (W. sebi) at different Aw, they estimated a minimum period free from mould growth. Observed mould free-storage lives of several tropical fish species were in agreement with the predictions.
2.2.3 Methods of Salting
Success in producing salt fish of consistent quality depends upon closely following traditional methods paying particular attention to the following points:
The quality of fish prior to salting,
The ratios of salt to fish,
The method of splitting and gutting,
The method of stacking and restacking of white fish or
packing in barrels of fatty fish;
The ambient temperature and
The method and degree of drying after salting
20 There are three methods by which fish can be salted (FAO, 1981):
1. Kench salting: This is a method whereby dry salt crystals
are rubbed into the flesh after which the fish is then
stacked. While the salt penetrates the flesh, the extracted
moisture drains away. Kench salting is one of two methods
of dry salting: it is suitable for lean fish such as cod,
but it cannot be used successfully for fatty fish such as
sardines and anchovy (Waterman, 1976).
2. Pickling: This is the other dry salting method in which the
moisture extracted whilst the salt is penetrating is not
drained away, thus the fish subsequently is immersed in a
salt pickle of extracted fluids.
3. Brining: This is a wet salting method in which the fish is
soaked in a concentrated salt solution. This particular
method is most advantageous for fatty fish such as sardines.
Being immersed in the brine, fat is protected from
atmospheric oxygen, thus preventing or hindering rancidity.
The salt content of the final product can be regulated by
altering the duration and temperature of brining
(FAO, 1981). The original brine concentration decreases
rapidly as moisture is withdrawn from the fish. Additional
salt must be added to the fish before covering with brine to
ensure saturation throughout the brining procedure
(Voskresensky, 1965).
21 2.2.4 Salt Penetration
Salt penetration into the fish flesh ends when the salt concentration in the aqueous phase of the tissue becomes equal to the concentration of salt in the surrounding solution.
Concurrently, water will move either to or from the brine depending on concentration and temperature of the brine. Hamm
(1960), Crean (1961) and Del Valle and Nickerson (1967) noted that at higher salt concentrations both meat and fish muscle lose water due to the salt denaturation of the muscle proteins. Thus, there is a critical salt concentration at which the initial uptake of both salt and water changes to a process of water removal accompanied by continued salt uptake. Duerr and Dyer
(1952) and Crean (1961) found this critical value to be about 8 to 10% salt content (wet basis), above which the protein is denatured rapidly and salt gain is accompanied by water loss.
Temperatures between 0° and 20°C had no effect on the critical salt level (Duerr and Dyer, 1952) although this level is reached faster at higher temperatures.
Water loss is an inverse function of salt uptake in fish muscle.
Crean (1961) pointed to this as evidence that salt and water exchange in fish muscle occurs in a narrowly defined region. The process may be envisioned as the front at which denaturation occurs as salt move into the muscle. When a critical salt concentration is reached, the front gradually moves inward through the muscle as the water is extracted from the fish by osmotic diffusion. As the front penetrates deeper, the rate of water diffusion is reduced. Finally, an equilibrium is reached,
22 when the movement of water from the fish ceases entirely
(Voskresensky, 1965). Salt penetration also depends on the conditions of the fish; those in rigo r take longer to salt than those in the first stages of autolysis. The process here is doubtless affected by such factors as change of tissue structure and viscosity of tissue fluid (Zaitsev et al., 1969).
2.2.5 Salting Equilibria
The salting equilibrium has been described by Crean (1961) as the point at which concentration of sodium chloride in the aqueous phase of the fish muscle and in the brine is the same. Del Valle and Nickerson (1967), after having ascribed an equation to express the equilibrium salt distribution between the fish muscle and brine, and Zugarramurdi and Lupin (1980) after having made a few assumptions found the equilibrium constant (Ke) for cut fish salted in saturated brine to be equal to one. The Kg value for salting of whole fish were found to be approximately 0.60. The latter being lower is due to the osmotic effect exerted by the soluble organic compounds inside the whole fish which are prevented from diffusing freely to the outside. The equilibrium is reached when the pressures on both sides of the skin is equal, but the concentration of sodium chloride inside the fish is obviously lower than that in the surrounding brine (Zugarramurdi and Lupin, 1980). Kg for whole fish is also found to decrease with an increase of protein content in the fish muscle
(Zugarramurdi and Lupin, 1980).
23 2.2.6 Innovative Salting Techniques
Many researchers have studied conventional salting techniques and have advocated more efficient and more adaptable (for the different variety of fish) procedures (Ingram and Kitchell, 1967;
Mandelsohn, 1974). A rapid salting process has been described in which fish is ground and mixed with salt, pressed to release fluid and formed into cakes which are then air dried (Del Valle and Nickerson, 1968; Del valle and Gonzalez-Inigo, 1968), oven dried (Poulter and Disney, 1977; Young et al. , 1979; Bligh and
Duclos, 1981; Wood, 1982) or drum dried (Anderson and Mendelsohn,
1972). This product was prepared for use in fish cakes by mixing reconstituted salt fish with rehydrated potato flakes and herbs and spices. Bagged salted fish was developed in Britain (Orr,
1967) by placing fish in plastic bags with salt and water after which the bag is evacuated and sealed. Each bag is packed in a carton which bears instructions for use only after a certain date by which complete salting has been assumed. A dehydration- injection method has also been described in which frozen fish fillet is freeze dried and the water replaced by saturated brine
(Carver, 1969). Another method involved the propulsion of salt crystals into fish fillets as pellet-projectiles in which innate water of the fish or fillet serves as the vehicle for salt distribution (Lee, 1969). Brining/salting and subsequently pressing the fish gave products with some promise (King et al.,
1985) as was also observed in the salting and ripening of anchovies (Filsinger, 1987).
24 2.3 Drying Processes
2.3.1 Preservative Role of Drying
Salting alone does not stabilise fish products sufficiently to allow long-term preservation. Thus, after salting, most products are dried or dehydrated and sometimes smoked. The prime reason for drying fish is to reduce its moisture content to such a level that insufficient water remains to support microbial growth and promote degradative biochemical activities. It has long been realised that it is the water activity (Aw) of a food which mainly influences its susceptibility to microbial attack (Scott,
1957; Troller and Christian, 1978). Aw is a measure of the free or available water in the food which is able to react chemically or to support the growth of microorganisms during spoilage
(Waterman, 1976). By reducing water levels and/or partially binding it with salt will lower the Aw and prevent or retard spoilage. Aw of pure water is assigned the value of 1, and the water activity in food is expressed as a fraction relative to pure water. Microorganisms have specific A^ requirements for growth. Most spoilage bacteria will cease to grow in foods whose
Aw is below 0.9, the growth of mould is inhibited below 0.8 and halophilic bacteria do not grow below 0.75 (FAO, 1981). Almost all microorganisms are inhibited below 0.6 (Fig. 2.1). Salwin
(1959) and Labuza (196^) have shown that most deteriorative reactions in food systems have the lowest rate at the Brunauer-
Emmet-Teller (BET) monolayer (Brunauer et al., 1938) which usually corresponds to the 0.2-0.4 Aw range or moisture content
25 l PURE WATER
Figure .0 MOULDS YEASTS BACTERIA
2.1:
0.9 Growth to 0.91 HALOPHILIC
water
range
activity 0SM0PHILIC XEROPHILIC
0.8 BACTERIA of
26 m i c r o o r g a n i s m s
(FA0, 0.75
YEASTS MOULDS
1981) 0.7 i
with 0.65
respect 0.6 ~r 1 0.6 I O L CJ CC O CQ CD o oc 3 “ l . ------WATER 1 ------
ACTIVITY^) 0.5 1 ------in the region of 5-10% (Labuza, 1968). This Aw range is probably unrealistically low for practical reasons in foods. The moisture content generally aimed for during drying of fish was reported as between 15-20% by Connell (1980) although moisture contents as high as 40% have been described (Sachithananthan, 1976; Zain &
Yusof, 1983). Poulter (1980) published a rough guide to good quality storage life of dried fish with different water and salt contents (Table 2.1) whilst Doe et al. (1982) published methods for calculation of A^ from compositional data.
2.3.2 The Theory of Drying
In order to separate water from fish, the forces which bind water molecules both to the non-aqueous substrate and to other water molecules must first be overcome (Jason, 1980). Once separated they will return and recombine unless removed from the immediate environment. Energy is therefore required to overcome these molecular forces and to exhaust the vapour into the ambient atmosphere. This energy can be provided in many ways including solar heating, microwave radiation, radio-frequency dielectric heating, or ultrasonic heating. The relative importance of the various mechanisms involved depends upon the nature of the material being dried, the internal and external conditions, the means of supplying thermal energy and the water content (Jason,
1965).
The oldest and best known method for drying of salted fish is air-drying, either natural or artificial. Natural drying, which is totally weather dependent, involves placing the product on the
27 Table 2.1
ROUGH GUIDE TO THE GOOD QUALITY STORAGE LIFE OF DRIED FISH WITH
DIFFERENT WATER AND SALT CONTENTS
Water content Salt content Minimum 'good quality' (% wet weight) (% wet weight) storage life
40 5 h week 40 10 1 week 40 15 3 weeks
40 20 lh months
35 5 h week 35 10 2 week2
35 15 2 months
35 20 lh months
30 5 h week
30 10 lh months 30 15 2 months
30 20 2 months
25 5 1 week
25 10 2h months 25 15 2 months
25 20 2 months
20 5 3 weeks 20 10 4 months 20 15 4 months 20 20 3Js months 15 5 1 year + 15 10 1 year + 15 15 1 year + 15 20 1 year +
Source: Poulter, 1980
28 ground, on mats or on racks in the open sun and relying on air circulation around the product to evaporate the excess moisture.
To be efficient, outdoor drying requires a dry atmosphere, sunlight, and also a slight breeze.
Artificial drying methods consist of placing the product in an enclosed drier. Air entering the drier is controlled to the desired velocity, humidity and temperature. Artificial drying has been the subject of investigations of optimum conditions of drying. Linton and Wood (1945) reported that optimum drying conditions for heavily salted cod were an air temperature of
26°C, relative humidity (RH) of 45-55%, and air velocity of about
1.25m/s. For lightly salted cod, Legendre (1955) found that temperature of 27°C, air velocity of 1.5-2.0m/s and RH of 50-55% were optimum whilst Damograi and Demyanov (1980) reported that optimal drying for salted fish occurred at 30°C, RH of 50% and air velocity at 1.5m/s. Aitken et al. (1967) applied the
Torry/Yarrow method of accelerated mechanical drying on unsalted and salted cod up to a temperature of 115°C and still obtained good products. High temperatures have also been used for drying tropical fish. Thus, 43°C, 50-55% RH and 1.3-2.5m/s air velocity were optimum for Cambodian fish (Legendre, 1961); and 45°-50°C,
60-65% and 2m/s air velocity for Indian species (Chakraborty,
1981). Yu et al. (1982) obtained good products by drying salted tropical species at 45°C, air-speed of 2m/s and environmental RH of 95.9%.
In principle, the air drying process can be divided into a
'constant-rate' period and one or two 'falling-rate' periods.
29 During the constant-rate period, moisture loss within the drying food is sufficiently rapid to maintain water-saturated air at the surface and the rate of drying is controlled by the rate of heat transfer to the evaporating surface (McCormick, 1973). The rate of mass transfer balances the rate of heat transfer and the temperature of the saturated surface remains constant at the wet- bulb temperature. The resistance to heat and mass transfer is located solely in the air stream, therefore the rate of drying does not change with time provided that conditions of the air are constant. Jason (1958) showed that drying rate is considerably increased locally by air turbulence generated downstream from the leading edges of the fish. He also concluded that in the initial stages of drying the surface of cod muscle behaves as though it were saturated with water and the rate of evaporation is identical with that of free water although Leninger (1959) pointed out that such conclusion is not completely admissible on theoretical grounds. In practice, the constant-rate period is terminated when the rate of diffusion of water from the interior of the muscle to the surface is too slow to sustain the initial maximum rate of evaporation.
The 'falling-rate' period is characterised by a rapid decline in the drying rate. The rate of drying in this period is governed by the transfer of water within the fish, largely by diffusion.
Jason (1958) and Del Valle and Nickerson (1968) showed that the falling-rate period for fish and salted fish occurs in two distinct phases, each characterised by two diffusion coefficients. The coefficient for the first phase is greater
30 than that for the second phase. Most moisture is removed from the salted fish during the falling-rate period.
2.3.3 Factors Affecting the Drying Rate of Salted Fish
(a) Thickness: Thickness of the fish influences its drying
rate (Jason, 1958) and Saravacos and Charm (1962) reported
that moisture removal rate during the falling-rate period
was inversely proportional to the square of the thickness.
Hence thicker products require longer drying times. Thinner
fish dries faster because the diffusion path to the surface
is shorter (Watermann, 1976).
(b) Temperature: Temperature affects the heat transfer and the
diffusion coefficient. During the constant-rate period, the
rate of drying is proportional to the difference in
temperature between the air and surface; where the surface
temperature would also be the wet-bulb temperature. This
therefore means that the rate of drying is proportional to
the wet-bulb depression for any value of air temperature
(Jason, 1958). The effect of temperature on the diffusion
coefficient is to increase the rate of diffusion as
temperature increases. It is always lower for the second
phase because the activation energy at this stage is higher
(Jason, 1980). Therefore, it takes longer time to dry the
same amount of water in the second phase than in the first
phase, e.g. at 100°C, six times as long is needed (Jason,
1980).
31 (c) Air Velocity: Increased air velocity results in an
increased drying rate during the constant-rate period
because the water molecules diffuse through a stagnant layer
of air adjacent to the surface and are carried away during
evaporation by the turbulent air stream beyond. The rate of
diffusion is inversely proportional to the thickness of the
stagnant layer which is inversely proportional to the air
velocity. Since the rate of surface evaporation is
determined by the rate of diffusion, higher air velocity
results in a higher evaporation rate. Increased air
velocity is more costly and practical experience has shown
that air velocities in mechanical dryers in the range of 1
to 2m/s are satisfactory in providing a reasonable drying
rate (Wheaton and Lawson, 1985). In the falling rate period
the rate of drying depends mainly on the temperature and
thickness of fish, hence air velocity does not have a
critical influence (Jason, 1980) as was seen in the drying
of cod fillets (Jason, 1958).
(d) Relative Humidity: In the initial stage of drying, the
surface of wet fish behaves as though it were saturated with
water and the rate of evaporation is proportional to the
difference in water vapour pressure between the fish surface
and the air. Therefore, at constant air velocity, the
drier the air, the faster the rate of drying.
Excessively rapid drying or drying at too low an RH results
in a poorer quality product and a lower drying rate during
the falling rate period. If the air is very dry a crust may
32 toughen the fish surface and retard the diffusion of water
from deeper layers to the surface. Thus, relative
humidities between 45% to 60% are generally recommended for
drying of salted fish (Wheaton and Lawson, 1985).
During the falling rate period, drying rate is a function of
the difference between initial and equilibrium moisture
contents and therefore depends on the RH of the drying air.
The equilibrium moisture content varies with RH. However,
changes in equilibrium moisture content are very small in the RH range from 20% to 70%. Beyond about 80% RH, equilibrium moisture content increases rapidly with increase
in RH (Jason, 1980). Table 2.2 shows the minimum water content obtainable in lean fish at different atmospheric RH
(Waterman, 1976).
Table 2.2
EFFECT OF HUMIDITY ON FINAL WATER CONTENT OF LEAN FISH
Relative humidity of Minimum water content the air obtainable in fish <%) (%)
20 7
30 8
40 10
50 12
60 15
70 18
80 24
Source: Waterman, 1976
33 (e) Fat Content: Fat is hydrophobic and therefore inhibits
diffusion of water and thus decreases the diffusion
coefficients. Jason (1965) studied the effects of fat
content on the diffusion coefficients in fish muscle and
confirmed this statement.
(f) Salt Content: No diffusion coefficients were observed
during the falling rate period in the drying of salted
swordfish and cod muscles (Del Valle and Nickerson, 1968;
Peters, 1971). According to Peters (1971) water exists in
two forms. The first is the water in excess of that
necessary to dissolve the salt present and the other is the
amount of water available to diffuse as a saturated salt
solution. He concluded that for both salted and unsalted
fish, entry into the second diffusion stage was initiated
when removal of the water monolayer begins. The presence of
salt in solution causes this to happen at the point of
aqueous phase saturation.
Diffusion coefficients associated with the first and second
forms differ considerably, being greater for the former than
the latter. Peters (1971) found that the second diffusion
coefficient is invariant in the temperature range 15° to
30°C while the first coefficient decreased with temperature
increases in this range. The negative effect of temperature
rise was ascribed to collision or associations between
inward moving ions (Na+ and Cl”) and outwardly diffusing
water molecules, thus retarding diffusion rate.
34 Del Valle and Nickerson (1968) reported that the first
diffusion coefficient depends on the salt content of the
fish. The diffusion initially increases, reaches a maximum
and then decreases as salt content increases. They found
the maximum diffusion coefficient to occur at a salt
concentration of about 1.0-1.5 moles/L.
It has been found that drying rate decreases with decreasing
RH both in heavily salted fish (Linton and Wood, 1945) and
in lightly salted fish (Legendre, 1955) due to the
progressive formation of a thick crust of salt and protein
at the surface. Interrupted drying, immediately followed by
press-piling, allows water in the fish to migrate to the
surface, rewetting the salt layer, which reduces crust
resistance to vapour flow. Therefore, the drying rate is
higher after press-piling until the crust dries.
2.3.4 Innovative Drying Techniques
Natural drying in the open air is subject to the vagaries of the weather. In the tropics where humidity is high it is often difficult and sometimes impossible to achieve drying to the required level. Salting will remove some of the water, so that less water has to be removed by subsequent drying, but the presence of salt in the flesh in turn reduces the rate at which water diffuses through the fish, and so increases the drying time. Some improved drying procedures have been implemented.
Raising fish off the ground on racks (as has always been done with the stockfish in the Netherlands) exposes them to more wind,
35 increases surface area of the fish in contact with the air and hence improves the drying rate (Waterman, 1976). Sloping racks, such as the inverted V-shaped racks used in Norway for the stockfish (Nesje, 1986), will allow excess surface moisture to drain away and prevent damp pockets forming in the gut and gill cavities of fish (Waterman, 1976). Press-piling at night will aid drying by enabling water to diffuse to the surface of the fish. Protection of the fish from rain is of course desirable either by removal to a dry place or covering with a waterproof sheet.
Solar driers have been investigated as an alternative to traditional sundrying in tropical developing countries (Doe et al. , 1977; Chakraborty, 1978; Ismail, 1980; Doe, 1982). Solar driers employ some means of collecting and concentrating solar radiation to achieve elevated temperatures and reduced relative humidities during drying. This results in increased drying rates, lower final moisture contents and higher quality products.
Solar dryers are less susceptible to variations in the weather, and they provide shelter from rain. Pests are discouraged by the high internal temperatures and the physical barrier of the drier itself.
Solar driers fall into two categories based on the mode of air flow through the dryer; natural convection or forced convection.
Doe (1977) and Doe (1982) pioneered the use of a simple tent of polythene sheeting supported on bamboo frame. Meynell
(1978) tried three types of solar drier in Malawi which were satisfactory during the dry season but like Ismail (1980) found
36 them unsatisfactory during the wet season. Thus, alternative heat sources such as fuels based on agricultural waste were suggested (Ismail, 1980).
Curran and Trim (1983) successfully produced low moisture, stable dried fish in the Galapagos IslanclS by using three types of natural convection solar driers. Like Doe et al. (1977) and
Ismail (1980) they found that the initial drying rates between the solar and sun drying on racks were similar and proposed that sun drying be used initially, followed by solar drying for a greater drying rate at the end of the process. Chakraborty
(1978) has described a fish drier which used a black painted galvanished iron heat collector coupled to a separate fish holding section. An electrically powered fan was used to force the heated air from the heat collector over the drying fish. The problem with the solar drier is its initial drying rate is no faster than a conventional one. Without the use of a fan it is difficult to see how solar driers can outperform the natural drying process during the all-important initial drying phase when blow-fly attack can be a particular problem. Alternatively, solar driers have been suggested for the later drying stages and for heating to deinfest dried fish. It was found that in an enclosed drier, the hot air had to be expelled before the cooler night temperature causes condensation inside the tent (FAO,
1981).
In the Philippines, several low cost agro-waste fish driers were developed and tested. There were the chimney type, the open
37 type, the diverted type and the diverted type with modification
(Villadsen and Flores, 1983). A solar agro-waste fish drier/smoker (Orejana and Embuscado, 1983) gave good dried products using both coconut husk or rice hulls as the source of energy.
Other artificial drying methods have been investigated. Disney
(1974) used a diesel oil burning heater which proved effective under wet weather conditions; Chakraborty (1978) used electrically and steam-heated tunnel driers and concluded that these may be viable for sufficiently large operations. In
Norway, USA and Nigeria the revolutionary heat pump drying plants
(Strommen, 1982) have been successfully employed for indoor drying of stock or clipfish (Nesje, 1986). The advantage of the heat pump drier is that its drying conditions can be adjusted to it suit the needs of the moment and^ is operational all year round regardless of the weather. Despite the sophistication of drying equipment that hauS been developed, there is still doubt as to whether artificial drying will prove to be economically viable in developing countries (FAO, 1981).
2.4 Chemical Properties of Salted Dried Fish
2.4.1 Volatile Bases
These compounds are of particular interest because of their use as indicators of fish quality. The question of quantifying the freshness of fish has been addressed in many ways. There are two main approaches, namely, sensory methods and objective methods.
38 Non-sensory methods are objective in nature and are of three kinds depending upon whether they are related to ;(i) some sensory attribute of quality, such as eating quality; (ii) safety and wholesomeness; (iii) composition or identity. Eber (1891) proposed a simple chemical test for putrefaction based on volatile amines, and since then numerous chemical, biochemical, physical and microbiological methods have been investigated. Of these the determination of total volatile basic nitrogen (TVBN) is perhaps the oldest (Clark and Almy, 1917; Lucke and Gerdel,
1935; Botta et al., 1984). This method is of widespread use and correlates reasonably well with sensory changes during spoilage or deterioration. This method has the advantage of simplicity, cheapness and relative rapidity. Its conditions of volatilisation must be specified exactly. The validity of TVBN as indicators of quality in dried salted fish has also been questioned (Connell, 1980, Hebard et al. , 1982 and Ryder et al.,
1984).
TVBN in fish is comprised of ammonia and amines such as mono-, di- and trimethylamine. Trimethylamine (TMA) and dimethylamine
(DMA) are derived principally from trimethylamine oxide (TMAO), an osmoregulatory compound present in most marine fishes. It is reduced partly by intrinsic enzymes to give DMA (Beatty, 1938;
Dyer and Mounsey, 1945) and formaldehyde (FA) (Amano et al.,
1963a, b) during frozen storage. This reaction was shown to be of very small extent by Beatty and Gibbons ( 1937 ) but is certainly due to the action of bacteria which use TMAO as the terminal electron acceptor (Yamamoto and Ishimoto, 1977). TMAO,
39 first reported by Hoppe-Seyler (1933), exists in many fish and
shellfish, generally in the largest amount in the elasmobranchs
and negligible in freshwater fish (Dyer, 1952; Harada, 1975) with the exception of pike (Shewan and Jones, 1957). Yamada (1967) postulated that freshwater fish did not contain significant
levels of TMAO because they excrete TMA promptly without having to store its detoxified form in their muscle, hence, its non- appearance. White-fleshed fish (demersal) generally contain larger quantities of TMAO than red-fleshed (pelagic) fish (Reay and Shewan, 1949). TMAO is also found at levels up to 200 mg percent in the mantle of squid (Simidu et al., 1953; Endo et al.,
1962; Takagi et al., 1967) and at lower levels in octopus (Sato,
1960; Harada, 1975). Thus, variations in the amount of TMAO exist between fish species although intraspecies differences may be greater than interspecies differences (Hebard et al., 1982).
Incidentally, it was noted by Shewan (1962) that 'the lower the animal is on the evolutionary scale the richer it is in muscle extractives'.
Only about 0.05% to 43% (Reay and Shewan, 1949 and Adams et al.,
1964) of the total bacterial population in the fish is involved in the reduction of TMAO. Psychrotrophic gram negative and oxidase positive (Watson, 1939; Tarr, 1940) bacteria of the types
Aclnetobacter, Pseudomonas (Lerke et al. , 1965 ) and
Flavobacterium are involved primarily (Castell and Mapplebeck,
1952). Secondary involvement of Enterobacterlaceae with the exception of Shigella and Erwinia is also found (Wood and Baird,
1943). Clark and Amy (1917) were the first to report that the
40 TVBN content increased during the storage of shucked oysters and
fish. Since then quantitive measurement of volatile bases has been suggested as an indicator of the onset of spoilage in fresh
fish. Tillmans and Otto (1924) suggested an upper limit of
30mgN/100g for acceptability, as did Tanikawa (1935). Kiamura and Kiamakura (1934), working with salmon, recommended
lOmgN/lOOg or less for fresh fish, 20-30mgN/100g for onset of spoilage, and the first stages of decomposition at >30mg percent
TVBN. Lucke and Gerdel (1935) found values of 20mg percent TVBN in all types of fish and claimed that fish is acceptable up to
30mg percent TVBN. However, Beatty and Gibbsons (1937) reported that TVBN in cod fish muscle increased between the pre-rigour period and the first appearance of odour to approximately
6mg/100g of tissue, further increasing during spoilage 15-20 times the original value. This paralleled the increase in fish odour. Sato (1960) noted that the TVBN was different between bottom and surface fish, the latter always being lower. The white muscle of some fish spoils faster than even the dark muscle of others and, of course, faster than the white muscle of the latter (Sakaguchi et al. , 1984). Farber (1965), cited many reports of variable levels of TVBN including those of Tanikawa
(1938 a,b) for carp, salmon, sardines, crabs and oysters; Shewan and Ehrenberg (1957) for cod; Labarre and Fougere (1942) for salted dried cod and Tarr (1942) for canned herring. He noted that 'there probably are more conflicting results for this test than for any other' and he concluded, like Ryder et al. (1984), that 'in general the content of total volatile bases is a rather unsatisfactory indicator of spoilage'. Farber (1965) attributed
41 this variation in results to differences in composition bacterial flora and handling methods between commodities. James and Olley (1971) demonstrated the inverse relationship between storage life and keeping temperature of shark indicating ammonia content as the primary limiting factor. Connell and Shewan
(1980) who noted the same inconsistency in results attributed it to biological variations in the concentrations of precursors and leaching of volatile amines during iced storage (Karnop, 1976).
Botta et al. (1984) noted that the quantitive difference between methods could be very large and attributed such differences to incomplete recovery of volatiles and/or breakdown of protein during analysis, giving high values. Davidovich and Giannini
(1984) when correlating the TVBN results by the distillation and the Conway methods noticed that the predicted and the experimental results agree for low TVBN contents but diverge increasingly as volatile base contents increase.
The method has been further explored by many workers (Pearson and
Musslemuddin, 1969; Guardia and Haas, 1969; Sidhu et al. , 1974;
Botta et la., 1984; Malle, Eb and Tailliez, 1986 inte^alia) .
Overall TVBN proved to be a poor indicator of early changes in quality but was good for advanced spoilage (Beatty and Gibbons,
1937; Connell and Shewan, 1980; Botta et al., 1984). Botta et al. (1984) explained that although the actual values of TVBN are affected by the methods of determination, this in turn did not affect the overall relationship between TVBN and sensory assessment. Thus, 30-40mg TVBN/lOOg are usually regarded as the limits beyond which round, whole chilled fish are too spoiled for
42 most uses. 30mg TVBN/lOOg has been specified as maximum for frozen tuna and sword fish and not greater than 20mg/100g for the raw material used in various canned products. 100-200mg
TVBN/lOOg are maximum limits for a variety of salted dried fish
(Connell, 1980). Consistently high TVBN values were also observed in storage studies on salted sunjdried Nigerian catfish (Faturoti and Aransiola, 1984; Faturoti, 1984).
Poller and Linneweh (1926) reported that during fish spoilage
TMAO was reduced by bacteria to trimethylarmine (TMA). Boury and
Schvinte (1932, 1935) and Boury (1936) suggested the potential of
TMA as a spoilage index, but it was Beatty and Gibbons (1937) who first used it and presented the rather simple microdiffusion technique for its determination. Later, its probable origin and the mechanism of its formation from TMAO, were elucidated (Beatty
1938, 1939). Many studies have been reported since, including methodology (Dyer, 1945; Hashimoto and Akaichi, 1957; Miller et al. , 1972; Ritskes, 1975, Tokunaga et al., 1977; Gill and
Thompson, 1984; Storey et al., 1984; Lundstrom and Racicot,
1985). Results have been conflicting (Farber, 1965), partly because of the use of different species of fish with varying compositions and the employment of different storage conditions.
However, TMA has been used extensively as an index of bacterial spoilage more specific than TVBN.
Like TVBN, upper limits of TMA contents have been suggested as a guide to fish freshness. For instance, 1.5mg TMAN/lOOg product has been recommended for very good quality cod for prepackaging
43 and 10-15mg TMAN/lOOg for good quality of round whole chilled fish (Connell, 1980). However, the data on the amines in cured or smoked and salted fish are scant and inconclusive. Dyer and
Mounsey (1945) found higher TMA content in smoked and salted fish than in the fresh product. Light salt cures, which naturally deteriorate faster than heavy salt cures, were characterised by high concentrations of TMA and total volatile acids (Cardin et al. , 1961). It has been shown, however, that when fish is salted, TMA formation is greatly suppressed and does not correlate with spoilage (Labrie and Gibbons, 1937). Bilinski and
Fougere (1959) found that TMA formation in cod muscle was inhibited by 10% NaCl for at least 15 days at 15°C or 25°C. Yu and Cruess (1951) reported that, in smoked fish, TMA formation paralleled spoilage as evidence by organoleptic examination.
Spoilage was quite evident when the TMA reached 50mg/100g flesh.
However, Velanker (1952), who used salted sundried fish reported lower TMA content in red discoloured fish than that of samples in good condition. Fujii et al. (1974) found the amount of TMA did not increase when moulds grew abundantly on the surface of salted fish muscle homogenates and that a Penicillium isolated from the homogenates utilises TMA rapidly. This was confirmed by Ishida et al. (1976) who also found that TMA was still quite low when salted fish was spoiled and moulds and yeasts were present.
However, TVBN was observed to be high in mouldy salted and unsalted dried fish stored under high humidities (85%)
(Muslemuddin et al., 1984). However, both Fujii et al., (1977) and Nozawa et al. (1979) observed that TMA-producing bacteria are
44 suppressed by 5% salt and even more so by the combination of salt and low temperature.
2.4.2 Moisture and Salt Content
The properties and stability of any salted or salted dried fish depends very much on the final moisture and salt content. Many processors disregarded these criteria, thus products with high water activity result which are liable to microbial attack. Some processors aim for greater yield at the expense of high moisture content in the product by drying for lesser times (Zain and
Yusof, 1983; FAO, 1981), an approach which will only shorten the life of the product.
Each country has its own standard as to the amount of salt and moisture (Tapiador and Carroz, 1965) desirable in their products.
Levels of salt in the products vary enormously. For example, salted herring may range in salt content from 16-35% depending on the method of salting, the ratio of salt to fish, the condition of the fish at the time of harvest and the chemical composition of the salt used (Voskresensky, 1965). In Asian countries where most of the processing and trade in salted dried fish takes place the problem of incorrect moisture and salt content is widespread and accounts for heavy losses of the products. Despite standards, Sripathy (1983) reported that Goonawardene (1978) found that majority of market samples do not confirm to any of the specifications, as was also observed by Gopakumar and
Devadesan (1983).
45 The proposed Malaysian Standard for salted dried threadfin bream, stipulates moisture content to be <40% and the salt content <25%
(Mat, 1983). Zain and Yusof (1983). Surveying the quality of dried fish in Malaysia, found moisture content to be 33-40% and salt content 20-24% for large fish (>600g); 24-42% and 14-23% respectively for medium sized fish (50-600g) and 13-35% and 3-8% for small fish (<50g) and concluded that these were within the specifications. However, they found that salt crystals appeared on the surface of some samples, others appeared wet, rancid odours were evident and many products were dry and brittle.
Curran (1984), on the other hand, found some Malaysian samples such as dried salted queenfish and red snapper to have moisture contents as high as 55%. Malaysians generally are said to prefer salted dried fish which are slightly wet (Latif et al., 1983).
Indonesian products are characterised by high salt (22-38%) and high moisture (44-52%) contents (Sumardi et al., 1983). Ah-Weng et al. (1985) examined salted dried fish from various countries in the Asian region and also observed high moisture (30-60%), but reasonable salt contents (6-21%) for large salted dried samples.
Dried anchovies (Stolephorus spp.) from these countries also exhibited variable moisture 15-35% and salt (1.6-30.0%) contents.
Interestingly these authors also found that the flesh just underneath the skin and that further inside next to the backbone differed in moisture and salt contents and that there are also differences in the moisture and salt contents of the right and left belly flaps of the fish. One reason for the high moisture contents in these products is the sunprying procedure which has a
46 reputation for wetter products. Yu (1985) and Yu, Siaw and Idrus
(1982) reported that sun-dried fish contained higher final moisture content than oven-dried ones similarly salted. Sidaway and Balasingham (1971) and Poulter (1980) noted that much sun- dried fish had moisture contents which were too high to ensure anything other than very short shelf lives.
Reabsorption of moisture on storage is a typical phenomenon of salted dried fish. Daniel and Etoh (1983) observed the change in stored skate and Rubbi et al. (1983) noted that when the relative humidity is over 85% reabsorption occurs, as did Sumardi et al.
(1983). Sripathy (1983) reported that high moisture content because of incomplete drying and/or atmospheric moisture absorption due to inadequate conditions of storage makes the product susceptible to microbial attack and consequent deterioration of quality. Suryanarayana Rao et al. (1962) found that a relative humidity of over 70% was conducive to mould attack, and in Malaysia, for example relative humidities of about
90% are common (Sidaway and Balasingham, 1971). Curran (1984) noticed that most samples including some large fish and fatty fish, except for squid and octopus, reabsorb substantial moisture after 10 days storage at room temperature in Malaysia and that most samples become unacceptable. Sachithananthan (1976) noted that losses were apparently highest in areas where relative humidities were greater than 65%. Rainy conditions caused even higher losses than humid climates (FAO, 1981).
Redrying of the products is a standard practice under such conditions. This process extends the storage life, even more so
47 if the product is then packaged (Goonewardeen and Etoh, 1980).
However, Curran (1984) concluded that redrying does not extend the shelf life as long as if the same A^ or moisture content had been achieved during the original drying process. Microbial spoilage that occurs between drying and redrying, or during redrying, is not reversible. Thus, although redrying retards further spoilage, it cannot overcome damage.
Table 2.1 produced by Poulter (1980) to estimate the mould-free life of salted dried fish, was expanded to include products with higher moisture content, Table 2.3, by Curran (1984).
Theoretically, at higher moisture (45-55%) and salt (20%) contents, longer shelf lives than those indicated in Table 2.3 would be expected, but studies by Curran (1984) suggested that these figures are likely to be more accurate predictions.
Obviously, too much salt causes excessive salty flavour.
Consumer preference listed 'not too salty' as the first criterion for good salted fish in a survey by Latif et al. (1983). This characteristic i3 also thought to have contributed to the decline of the popularity of heavily cured fish on the Western market.
Another reason is that, society is wary of salt intake for health reasons. High salt content also gives products an unflattering gritty appearance due to salt crystals on the surface. Grade 1 quality salted dried fish should not exhibit crystals on the surface (Dagbjartsson, 1983). Too much salt also hinders proper drying by appearing on the surface and forming a barrier to moisture diffusion. However, it has been observed that the
48 presence of salt hinders or reduces infestation by blow flies
(FAO, 1981).
Table 2.3
EXPANSION OF GUIDE TO SHELF-LIFE OF CURED FISHERY PRODUCTS WITH
DIFFERENT MOISTURE AND SALT CONTENTS
% Moisture % Salt Minimum acceptable storage life
5 Less than Jj week
10 Less than h week
55 15 1 week
20 h week
5 Less than Jj week
10 Less than % week
50 15 lh weeks
20 1 week
5 Less than % week
10 Less than \ week
45 15 weeks
20 1 week
Source: Curran (1984)
2.4.3 Water Activity
The importance of water activity in the microbial stability of foods has long been established (Scott, 1957; Troller and
49 Christian, 1978). A recent review (FAO, 1981) highlighted its role in cured products. Hence, the control of moisture and salt contents in salted dried fish is inevitably pivotal in determining the shelf life of the product. The 80's was a remarkable time for concerted research efforts into processing and upgrading quality of salted dried fish, generally considered to be the food of the poor. Thus, the pioneering work of Doe et al. (1982) and Poulter, Doe and Olley (1982) into water activity prediction and, consequently, shelf life prediction for such products represents a major advance in understanding the properties of dried salted products.
Water activity measurements can be made from moisture sorption studies or by the use of instrumentation. However, Ross (1975) proposed that the Aw of intermediate moisture foods can be found from the product of the A^ of its individual components. Doe et al. (1982) applied this principle to dried salted fish by assuming that fish has only four components: water, salt, fat and fish muscle (protein). Since fat is hydrophobic, it plays no part in the calculation provided that water and salt contents are expressed on a fat-free, dry matter basis (Leistner, 1976;
Inglesias and Chirife, 1977). Aw of the salted dried fish was then calculated from the equation:
Awn * V> (1) where A^n is the water activity of the salt-water component of the fish and Awq is the water activity of the fish muscle. The
A^ of the salt in the fish is taken to be equal to that of a pure
50 sodium chloride solution at a molality, m, calculated from the measured water and salt contents using the expression:
m = 17.09 (Mg/Mb) (Mb/Mw) ...... (2) where M is the mass of salt, M is the mass of water and M, is s w d the fat-free, salt-free dry mass. The relationship between m and
Aw can be found from published vapour pressure data for sodium chloride (Robinson and Stokes, 1959). Values for AwQ at different water contents are found by applying equation (1) to the sorption isotherm for unsalted cod (Doe et al. , 1982). To simplify this further, Doe et al. (1983) presented a table giving the Aw values corresponding to different water (Mw/Mb) and salt
(Ms/Mb) contents calculated using equation (1). They also found that there is a good agreement between the measured Aw values and those calculated from the water and salt contents of the dried fish products.
The growth curve for the dun mould, Willemia sebi, adapted by
Poulter, Doe and Olley (1982) from data of Pitt and Hocking
(1977) can be used to predict the mould-free shelf life of dried fish (Fig. 2.2) (Poulter, Doe and Olley, 1982). Curran and
Poulter (1983) examined the isotherms for dried salted tropical fish and found that the sorption characteristics of the muscle from tropical fish species are very similar to that of cod and that the salt concentration and drying temperature experienced commercially in the tropics do not affect the results. However,
Curran (1984) reported that the isohalic sorption isotherms for octopus, squid, shrimp and anchovy were all different as would be
51 Water a c tiv ity Figure
2.2:
moisture WaHernia sorption Times Doe fish
and
calculated for
01
and Sebi isotherms
visible ley
52 fat ,
at
1
982). from
contetns constant
colonies for
measured
salted
A and
of
of dun cod the salt,
stored
mould isohalic (Poulter,
65
Storage time
(week) expected from the different classification and salt contents. A comparison between the water sorption isotherms (where the influence of salt is removed) shows that of these four, the data for anchovy alone fits the isotherm for cod. Therefore, their
Aws cannot be read from the table provided by Doe et al. (1983) and must be calculated by using the moisture and salt contents together with the relevant sorption isotherm and the method of
Doe et al. (1982).
As A^ is lowered, microbial growth is suppressed during all phases of growth. The overall effect is seldom death, but rather a delayed and reduced rate of growth, and usually some reduction in one or more of microbial metabolites. In a food, the microbe must contend with a variety of growth influencing elements, such as pH and salt. If conditions are not ideal, growth inhibition through A^ reduction is increased (Troller and Christian, 1978).
Sporulation and spore germination are believed to occur at A^ levels appreciably below those permitting growth and the rate of spore formation is influenced by the Aw level and the solute employed to adjust the Aw.
There is some disagreement about the effect of A^ on the shape of microorganisms. Troller and Christian (1978) described striking morphological differences between a strain of Staphylococcus aureus grown at 0.997 Aw and the same organism grown at 0.90 Aw.
However, Troller (1980) suggested that cell distortion by water loss is the most common morphological effect on bacteria exposed to osmotic stress, although this may not occur in all organisms.
53 Motility is expected to be impaired with reduced Aw as membrane involvement is commonly observed at low Aw. At Aw 0.6, or below, all microbial growth is inhibited but Aw above this may retard growth so that microorganisms cannot damage the product before it is consumed.
In an osmotically hostile environment, microbial cells must cope either by excluding solutes or by developing the capability to accomplish its metabolic tasks in the presence of solutes. Thus, osmotic stress almost invariably requires that a physiologically benign yet osmotically 'active' compound accumulates intracellularly to counter the osmotic imbalance across the cell membrane in low Aw systems. These compounds have been termed
'compatible solutes' and include potassium ions in halophilic bacteria, polyols in yeasts, proline in osmotolerant bacteria, aminobutyric acid in moderately tolerant bacteria, and glutamic acid in the less tolerant bacteria (Troller, 1980). Some osmotolerant bacteria can cause spoilage below Aw of 0.95, but those responsible for spoilage of fresh fish require Aw of about
0.98 for rapid growth. At A^ levels below 0.85 development of bacteria and yeasts is normally arrested and halophilic bacteria, xerophilic moulds and osmophilic yeasts predominate. It is these microorganisms which may attack cured products.
Moulds are more resistant to dry conditions than bacteria and some are capable of survival in products with Aw as low as 0.60, although 0.70 is the minimum Aw which will sustain growth of most storage moulds. In cured fish the 'dun', Wallemia sebi, the most
54 common type of mould growth has an optimum Aw of about 0.75
(Frank and Hess, 1941). Other moulds include the Aspergillus
spp. Wheeler et al. (1986) reported the most commonly encountered species in 74 samples of Indonesian dried fish (with
Aw varying from 0.65-0.79) which are possibly important in spoilage. These include Polypaecllum plsce, a newly described species in dried fish which appears in extreme cases as conspicuous white growth (Pitt and Hocking, 1985), Eurotium chevallerl, E. rubrum and Aspergillus wentii were also isolated from visibly mouldy fish. Aspergillus penicilloides, A. flavus,
A. niger and Penicillium citrinum were also found while less common species were only isolated by impression plating. The
'dun' was rarely present on Indonesian dried fish and this was thought to be due to its inability to grow at high temperatures.
Eurotium was found to be the dominant fungus on marine products from Japan and South East Asia (Ichinoe et al., 1977; Ogasawara et al., 1978; Okafor, 1968; Wu and Salunkhe, 1978).
Dun does not cause objectionable flavour or textural changes in the fish although visible discoloration or growth can make the product unacceptable or reduce its economic value (Sen and
Lihiry, 1964). Moulds may also produce mycotoxins having a variety of harmful effects. The metabolism of microorganisms in general involves the release of water, causing a localised rise in Aw around the affected parts allowing growth of some organisms previously inhibited. Poulter (1980) observed that cured fish is often initially attacked by moulds on the tail. The mould growth spreads over the fish, increasing surface moisture. This enables
55 other microorganisms such as halophilic bacteria to attack and
the fish. Cured fish deteriorates rapidly after the
initial mould attack, so much so that Poulter (1980) took the onset of mould attack as the end of 'good quality' storage life.
All mould growth is temperature-sensitive and 30°C a common ambient temperature in tropical regions, is optimal for several mould species (Christensen and Kaufmann, 1974).
Bacteria, on the other hand, do not generally grow in products with Aw <0.88, but certain haloduric and halophilic bacteria will grow in cured fish at Aw down to 0.75. Halophilic bacteria usually have a specific requirement for 10% or more salt and hence are termed obligate halophiles. Species of the genus
Halobacterlum and Halococcus attack cured fish and produce a pink discolouration. They have a strong proteolytic action causing the flesh breakdown, off-flavours and odours. They are aerobic and cannot multiply at temperatures below 10°C. These bacteria are very often found in large numbers in solar salt commonly used for fish processing, and hence are a particular problem with salted fish. Onishi et al. (1980) reported that 75 out of 168 strains bacteria from salted and dried fish belong to the type
IIIA which is moderately halophilic, requiring NaCl for growth but not KC1. The aerobic nature of these bacteria means that their growth is stopped by full immersion of the fish in saturated brine, but adequate drying is necessary to prevent subsequent attack during storage.
Most yeasts can grow at Aw down to 0.85, and halophilic species will grow at salt concentrations of 10-20%. If the moisture and
56 salt levels are relatively high, as occasionally found in heavily
salted but poorly dried fish, conditions may favour the growth of
halophilic yeasts. The principal halophilic yeasts which can grow in a medium containing 10-20% salt are Saccharomyces rouxii,
Debryomyces hansenil, Pichia ohmeri and H ansenula anomala
(Davenport, 1975). Debaryomyces hansenii is a common yeast in ocean waters, and may well survive in solar salt and this way contaminate salted fish. Yeasts have not, however, been shown to cause appreciable losses of traditional cured fish products, and it is not clear whether growth of Debryomyces hansenii on cured fish constitutes a significant problem.
Some moulds capable of growing on cured fish may also produce mycotoxins. Certain toxigenic strains of A. ochraceus, whose minimal Aw for growth is about 0.76, can sometimes produce penicillic acid and Ochratoxin A in foodstuffs with Aw of 0.80-
0.85, (Bacon et al., 1973). Dangerously high levels of aflatoxin
(600-700 ppm) have been found in dried fish (Okonkwo, Umerah and
Nwoko; 1977). This toxin is produced by the mould A. flavus and the closely related A. parasiticus, the latter being commonly isolated from cured fish in Vietnam (Townsend et al., 1971).
Wheeler et al . (1986) isolated A. flavus frequently from
Indonesian dried fish but found no aflatoxin present. Similar observations were made by Wu and Salunkhe (1978) and Sim et al.
(1985) on dried shrimp. The lack of reports of toxicological problems with consumers, may simply reflect absent of data in this area (FAO, 1981).
57 Apart from its influence on microbial stability, Aw has some bearing on chemical and enzymatic changes as well as the overall
stability and moisture sorption properties of food products.
Rockland and Nishi (1980) presented a diagrammatic representation
(Fig. 2.3) of the relationship between Aw and certain changes affecting food quality and stability. These include autoxidation
(Martin, 1958), lipid hydrolysis and oxidation (Chou and Breen,
1972; Gustafson and Cooke, 1952), enzyme reactions (Acker, 1969;
Skujins and McLaren, 1967), non-enzymatic browning (Burall et al,
1978; Labuza and Saltmarch, 1980) microbial proliferation
(Troller and Christian, 1978; Troller, 1980), and other phenomena which have been reviewed by Labuza (1980). It suffices to say that the properties playing a role in the above reactions can be divided into two categories, namely: (a) water as a solvent in which reactants dissolve, are transported and react; (b) water as a participant in specific reactions (Labuza, Tannenbaum and
Karel, 1970). Water was shown to act as a solvent even at low Aw by Duckworth et al. (1963) while water as a reactant was investigated in a model system by Schobel et al. (1969).
2.4.4 Proteins
Proteins content of fish, averaging about 19%, may vary from 6 to
28% (Stansby, 1962). Protein content of fish is considered low if less than 15%, high if between 15-20% and very high if above
20%.
Fish meat consists of three different types of proteins: sarcoplasmic, myofibrillar and stroma. Sarcoplasmic protein
58 K DC
DC > g? < U at <
-t-> CD <-> 03 O T- OC = E C -M o 03 03 CD 03 S- =3 E ©+-> CDi— Cl 03 4- N TD S- £. •r~ C C C O °- Q •!— CD 03 00 "—
CO
CNJ
CD ZJ CD
59 contains water soluble proteins called myogen which are extractable in low ionic strength salt solution. Its content varies with fish species, but is generally higher in pelagic fish such as sardine and mackerel and lower in dermersal fish like snapper (Hansen, 1970). Myofibrillar protein is the contractile protein consisting of myosin, actin and regulating proteins such as tropomyosin, troponin and actinin. It represents 66-77% of the total protein in fish meat (Suzuki, 1981). Fish meat contains proportionately more myofibrillar protein than mammalian skeletal muscle (Suzuki, 1981).
Stroma is the protein which forms connective tissue. It is not extracted with water, acid, alkali or neutral salt solution of
0.01-0.1M concentration. The components of stroma are either collagen, elastin or both. Stroma is the protein on the outer side of the muscle cell. Dark meat contains more stroma protein than white meat (Soevik and Braekkan, 1979).
To food scientists and technologists the special feature of the main fish muscle proteins, including enzymes, is their instability compared to their meat counterparts.
Aitken and ConneU (1979) reported that Aitken and Campbell (1969) observed three distinct changes of opacity in fish flesh when it is heated. The initial increase in translucence is followed by two successive increases in opacity. The second increase in opacity is due to precipitation of the thermally denatured sarcoplasmic proteins which begins at 45°C. The initial
60 increased translucence and the first opaque band begin at lower temperatures and are not unequivocally explained.
The amount of denaturation and coagulation that occurs when the fish is heated is measured by the amount of protein extractable by, or remaining soluble, in salt solution. Protein extractability however, only gives an indirect measure of changes and the results must be interpreted with caution. Connell (1964) pointed out that two main factors govern the removal of proteins from a muscle into a solvent. These are the intrinsic solubility of the protein and the ability of the solubilised protein to diffuse out of the muscle. According to Aitken and Connell
(1979) Lobanov and Bykova (1938) reported that fish protein extractability decreased at 30-35°C and was 60% and 10% at 40°C and 60-65°C respectively. The solubility of the myofibrillar and sarcoplasmic proteins decreased even when the fish were dried at
5-10°C (Suzuki, 1981). Sarcoplasmic proteins, based on their extractability, were more stable than the myofibrillar group when the two proteins were subjected to the same heat treatment
(Howgate and Ahmed, 1972). Connell (1961) also observed that the myosin of freezed dried cod became inextractable much more rapidly than the sarcoplasmic proteins. Thus, Howgate and Ahmed
(1972) postulated that changes in the salt solubility of myofibrillar proteins of cod heated at 30°C were in fact only due to changes in the myosin components; the actin remaining unchanged but unable to be extracted without the myosin.
However, Simidu and Simidu (1960) observed that about 10% of the total myofibrillar protein does not become insoluble even after
61 FRESH FISH MEAT _V 13 jo X • £-< x
2 to e C o C C t-H o a. S Ut 03 £ o> /2 /2 /2 h n
o X 03 o o J NN C/3 UH&« 0222 03 O z WWy &££ is J g^H p ^ H c/3 | a: J NN OCJ OP Q ho W as .
C/5 w g 5 C/3 H t £ W H
cm Q .2 i i|K CD X •= •S < *■3 PP x .o PP i lisscagef 5 l X 4-1 — U C « V o o c a 0 o - C3 t- E o C Og_i ^<2 03 c C aSgg-sl §51 s jaO 8 h x
i
5
.='€ -6 — -2 X X >2 © X © ^ « « 3 & C 2 2 g > 62 J x c/3 X (N X X ^ w a O 3 - 2 o C w >_ OS " g .2 -:= ^ £ ’ -X . X X SgiSgf 2 oa= £§?§! u 6 fl o-oowc^ 03 C a J O.X c3 3 o Sb 2-1 - - < C/3 ^ ^ X ‘ -’ i- 13 m ra 3 0 0 o N 3 o t- js'o O 3 §'§ = g>'g 2:1 -.2 X X 5" 5 ^3 x , XT* X < m 4 SCSI'S O 3 *^.2 w O n NoitivzniHais .S •a ■- “ O c US S'= o o 03 o O o § <».c VS hi i o3 .a .2 X Q iCfjgl a §^-5 Q. 03 O g cq O g.sI o _T o c „ X - .52 .3 hn 5 a C 5.2 t_ , /2 <-> saoNVHD - X .a 4 2.2 c o o a C 0X5 CO 3 c -H -:r ’ -' '5 u &s c co u, s Cj 0-2 O 2 02 H K C/3 pp S ^ < 2 H- H X X x -X X H-J S x CO ° ^ O O o o 3 3 c a J-H Figure 2.4 shows the summary of the influence of heating on the muscle components and quite prolonged heating up to 100°C. According to Odense et al. (1969) the major component of this heat stable fraction is tropomyosin. Saiki et al., (1959) observed that the heat coagulability of fish sarcoplasmic proteins also varied between species, the pelagic was more liable than the dermasal species. Even the pH, given the same time and temperature of heating, at which half of the protein of different species of fish became insoluble also varied with species (Simidu and Simidu, 1960). Howgate and Ahmed (1972) found that the myofibrillar proteins of cod were less extractable compared to tropical fish hilsa when given identical heat treatment. This was attributed to the differences in protein stability of the cold water and warm water species. Connell (1964) observed that fish myosins, either when isolated or when in the intact tissue, denature much more rapidly than beef or chicken myosin under the same conditions. Connell (1961) therefore concluded that myosin of different species had adapted to the different body temperature and that there was a connection between this phenomenon and the resistance of a muscle towards changes caused by processes such as freezing, frozen storage dehydration. The thermal shrinkage or denaturation temperatures of fish connective tissue (myocommata) collagen are much lower than those of the corresponding beef protein. Menashi et al. (1976) observed that connective tissue proteins (collagens) of various species increase in stability with increase in their In vivo body temperature. Poulter et al. 63 varied between species. The more stable proteins were those from fish found in the waters of higher ambient temperature. The stability of the sarcoplasmic proteins also exhibited some specie dependence but the actins present in the three species were of similar thermal stability. Davies et al. (1988) demonstrated that the myosin of cod, at pH 6 and ionic strength 0.06M, exhibited a thermal melting transition 10°C lower than the myosin of snapper, a warm water species, reflecting a comparable differential in transitions attributed to myosin in intact muscle from the two species at 42°C and 52°C respectively. However, with increasing pH and ionic strength, conditions favourable to disaggregaion and dissolution of myosin in vitro, the temperature of snapper myosin transition decreased to that of cod myosin, which remained essentially unaffected by the conditions suggesting that differences in the aggregation characteristics of fish myosin may explain in part the different thermal melting properties in vitro. Properties of fish proteins are also examined by their ability to separate under given conditions. The proteins of the sarcoplasm of fish muscle are compounds of relatively low molecular weight, many of them enzymes. They can be extracted with distilled water and separated from insoluble proteins by centrifugation (Bremner and Vail, 1983). The insoluble myofibrillar proteins are often solubilised by urea or sodium dodecyl sulphate (SDS) solutions. These extracts are then separated into protein bands either electrophoretically or by other techniques such as isoelectric focussing (IEF) either by agarose or polyacrylamide gels. The 64 extracts must be run on gels containing urea or SDS to keep the proteins in suspension (Bremner and Vial, 1983). Such techniques are widely employed in intra- and inter-specie studies and can also be utilised for product identification and inspection (Keenan and Shaklee, 1985). Cooked products (but not heat sterilised canned fish) can be differentiated as 6M urea extracts using IEF gels containing 6M urea. With canned fish, the more severe form of heating renders the method unsuitable and instead, differentiation is based on the separation patterns of the peptides released from the flesh on treatment with cyanogen bromide (Laird et al. , 1982; Mackie, 1980). Samples extracted with 10M urea gave the darkest banding patterns after staining. This was more successful as an extracting solution than 2% Triton X-100, 1% and 10% sodium dodecyl sulphate, 10% glycine or 20% diethylene glycol (Wiggin and Krzynowek, 1983). In cooked products a group of highly anodal proteins (about four major bands) were found to be heat stable. Because of the marked heat tolerance, these proteins proved to be extremely good markers for species identification of cooked fillets. These proteins were highly negatively charged i.e. had very low isoelectric points, had low molecular weight (10,000-12,000 daltons), were monomers and had distinctive UV absorbance profile. These characteristics, particularly the absorbance spectra, identify these proteins as parvalbumins (Pechere et al., 1974; Sullivan et al., 1975). 65 Optical Density 0.25 0.75 1.25 Figure 2.5: and barramundi heat UV absorbance Shaklee, stable 66 muscle proteins 1965). spectra (Keenan of for the Wavelength (nm) In fish that have received severe drying treatments such as fish meals, concerns are abound about their nutritive value. Meals known to have received severe heat treatments are lower in available lysine (Ford, 1973) and in addition deficiencies in methionine and cysteine/cystine may arise. However, Aitken and Connell (1979) reported that Pieniezak and Rakowska (1974) observed total amounts of methionine and cystine were not affected by heating at 115°C or 126°C but the 'available' amounts of these same amino acids were significantly reduced. Miller (1965) also concluded that heating cod muscle at 116°C resulted in greater decrease of protein quality than the decrease in total amino acids content. Other reports (Geiger and Borgstrom, 1962; Cutting, 1962) also reported that normal heat treatment such as in drying, salting and smoking and severe treatment as in canning did not affect the amino acids of several species of fish. This was true of amino acids as heat sensitive as methionine and such heat treatments were also found not to reduce the biological value as measured by animal feeding tests. On the other hand Clifford et al. (1980) has confirmed that lysine is the most sensitive of the essential amino acids during smoking (a 25% destruction of lysine is accompanied by a 7% fall in other nutritionally critical amino acids including histidine, arginine and N-terminal residues collectively). Losses were found to be much greater in the outermost 5-10mm layer of the fish and declined rapidly towards the centre. Bhuiyan et al. (1986) also reported losses in both available and total lysine (5.1 and 6.6% respectively) in mackerel fillets due 67 to smoking at 30°C for 1 hour, then at 50°C for 45min, and at 50- 80°C for the rest of the time. The outermost layer showed the highest loss. Hoffman et al. (1977) reported a loss of 11% when they smoked fresh water tropical fish tilapia at 75°C for 38h, higher temperature and longer process than that of Bhuiyan et al. (1986). The loss of lysine and other essential amino acids is essentially proportional to the time and temperature of processing and may exceed 55% in traditional hot-smoke tropical operations (Caurie et al., 1974). ScOG-SI et al. (1980) noted that the main decrease in available lysine concentration occurred during the first 90min. of smoking and that available lysine and digestibility decreased with increasing smoking time. Lysine became unavailable via its reaction with the amide groups of asparagine and glutamine to give new peptides which are not hydrolysable or are sterically prevented from being hydrolysed by digestive proteases (Ford, 1973; Bjarnason and Carpenter, 1970). Heating fish at 115°C for 20 min. in a can after the initial temperature has been obtained was found to cause a loss in cystine and cysteine (Opstveat et al., 1984). The amino acid lanthionine has also been found in severely heated dried cod but not in unheated samples (Connell, 1958). Lanthionine contains a thioether link formed from disulphide bonds present in the original protein. Other protein reactions known to occur include conversion of methionine to methionine sulphoxide and Maillard reactions with sugars such as ribose and galactose (Dworshak, 1980). 68 Nutritional values of many foods are increased by heat processing through the destruction of protease inhibitors and toxic substances along with the opening of the protein structure through denaturation. Processing can also decrease protein digestibility via non-enzymatic browning and thermal cross- linking reactions (Tannenbaum, 1974). The damaging effects of high temperature drying on fish protein have been reported earlier and offer explanation of the large variation in protein quality of commercial fish meals used in animal feeds (Miller, 1956, Carpenter et al., 1957). However, de Groot (1963) found no significant decrease in protein quality as a result of dehydration in most of the food products examined. Small differences pointed to the overall tendency for greater damage in air dried than in spray dried or freeze dried samples. Sheik and Shah (1974) observed that digestibility of roller dried fish samples was adversely affected at 80° and 100°C. Opstvedt et al. (1984) confirmed this in drum dried fish and explained that the formation of S-S bonds from -SH group reduced the protein and amino acid digestibility of these samples as compared with the raw fish protein. On the other hand Udarbe et al. (1985) did not find any significant difference between the protein efficiency ratio (PER) of casein and processed fish, although the PER of fresh fish was higher than these two samples. Surjdrying was also found to cause a severe drop in digestibility of squid (Tanikawa and Suno, 1952) and pollock meat (Ryu and Lee, 1985). 69 Steaming gave squid products of higher digestibility than boiling (Sawant and Magar, 1961; Ryu and Lee, 1985). Microwave treatment of food also increased digestibility (Hung et al., 1984). 2.4.5 Lipids Lipid content is highly variable in fish ranging from under 0.6% in cod (Addison et al., 1968) to a reported 25.5% in mackerel (Ackman and Eaton, 1971). In addition to species variability, lipid content varies with anatomical position, sex, season and diet. Most of this variation is reflected in triglyceride content as phospholipid levels are much more constant. The range of phospholipid concentrations ranges from 0.19% in rock bass to 0.87% in rainbow trout (Kinsella et al., 1977a). Red muscle contains a higher concentration of total lipid and phospholipid than white muscle (Love, 1970; Mai and Kinsella, 1979). The classes of lipids most commonly found in marine animals are hydrocarbons, wax esters, triglycerides, phospholipids, diacyl glyceryl ethers and neutral plasmalogens. Sterols, steryl esters, carotenoids and vitamins are usually present as minor components of marine oils (Malins and Wekell, 1970). Fatty acid profiles are available for a number of fish species (Ackman, 1974; Bonnet et al., 1974; Kinsella et al., 1977b; Viswanathan Nair and Gopakamar, 1978). Fatty acid composition of red and white muscle tissue have been compared in mackerel (Ackman and Eaton, 1971) and in white sucker (Mai and Kinsella, 1979). Phospholipid and triglyceride levels have been measured 70 in cod (Addison et al. , 1968), mackerel (Ackman and Eaton, 1971; Hardy and Keay, 1972), menhaden (Ackman et al., 1976) sardine (Hayashi and Tagaki, 1977 a, b) and white sucker (Mai and Kinsella, 1979). The degree of saturation of fatty acids is highly variable between species and also depends on environment (Bonnet et al. 1974 and Kinsella et al. 1977b). In mackerel and white sucker, while fatty acid composition of triglycerides is similar in red S’ and white muscle, white muscle phospholipid fatty acids are more unsaturated than those in red muscle phospholipids. Phospholipid fatty acids are more unsaturated than those of triglycerides. In fish muscle, phosphatidylcholine (PC) and phosphatidylethanolamine (PE) are the prominent phospholipids. PC is the predominant phospholipid but the ratio of PC to PE is lower in red muscle than white muscle (Shewfelt, 1981). Other compounds reported in the phospholipid fraction include sphingomyelin, phosphatidylinositol, phosphatidylserine, cardiolipin, phosphatiaic acid and lysophosphatidylcholine. The phospholipid pattern may be influenced by environmental conditions. Hydrolysis of phospholipids has been studied more than hydrolysis of triglycerides (Shewfelt, 1981). Phospholipase activity has been noted during the frozen storage of numerous species including cod (Olley and Lovern, 1960), lemon sole, halibut (Olley et al. , 1962), trout (Jones and Bilinski, 1967 ), herring (Bosund and Ganrot, 1969), freshwater whitefish (Awad et al., 71 1969), salmon (Botta et al. , 1973), carp and red sea bream (Toyomizu et al., 1977). It has been established that it is readily inactivated upon heating (Olley and Lovern, 1960; Lovern and Olley, 1962). Lipases from lingcod (Wood, 1959 a, b) and mackerel (George, 1962) capable of hydrolysing short chain triglycerides have been described while long-chain triglyceride lipases have been reported in the red muscle of rainbow trout (Geromel and Montgomery, 1980). The short-chain lipases of mackerel were found to be more active in red than white muscle (George, 1962; Morshita and Takahashi, 1969). Shewfelt (1981) in his review concluded that triglyceride hydrolysis leads to increased lipid oxidation, whilst phospholipid hydrolysis leads to decreased oxidation. 2.4.5.1 Rancidity in Salted Dried Fish Fish lipids are very susceptible to oxidation by atmospheric oxygen leading to rancid flavours (Tsuchiya, 1961). These flavours may decrease the acceptability of the product but, as salted and dried products are often used as flavour-giving condiments, rancid fish is not necessarily unacceptable (Waterman, 1976) and some degree of rancidity is considered normal (Watanabe and Dzekedzeke, 1970; Watanabe and Choongo, 1971). In its table of recommendation for grading dried fish, FAO (1983) listed moderately rancid as being normal of Grade A product, significantly rancid as Grade B and excessively rancid as Grade C. 72 The most common method of measurement of rancidity is the ■r<£5<^Z'/0/' between thiobarbituric acid the oxidation products of unsaturated fatty acids to give a red pigment (Tarladgis et al., 1960) which is then measured spectrophotometrically. By using 1,1,3,3-tetra-ethoxypropane as a standard which yields malonaldehyde on acid hydrolysis, it was possible to express the results as 'TBA number' defined as mg of malonaldehyde per 1000 g sample. Improvement of the method has been along the line of direct determination, without the distillation procedure, (Vyncke, 1975) or improved distillation-spectrophotometric method (Ke et al., 1984). Salt and heme protein appear to have a prooxidant effect (Castell et al., 1965; Koizumi et al., 1980). A combination of heme and fat peroxides can work as active components enhancing rancidity. Fat peroxides react with chloride ions producing free chlorine which brings about further oxidation of fat. However, Nambudiry (1980) observed that sodium chloride acts as a prooxidant in fatty fish when present at lower concentration but inhibits oxidation as its concentration increases. The inhibitive effect of high concentration of sodium chloride may be due to its interaction with the catalysts of lipid oxidation. Metal ions, especially non-heme iron were also found to be major prooxidants in heated muscle systems. The order of the activity was Fe^+ > Cu^+ > Fe~*+ > Mb > Met Mb. The order of oxidation of lipid fraction observed in the heated muscle was total lipids > phospholipids > triglyceride (Trichivangana and Morrissey, 1984). 73 Phenolic substances in wood smoke used for smoking fish have some antioxidant activity (Simpson, 1965) however, Woolfe (1975) did not see any evidence of this. In dried fish the range of moisture content affects fat oxidation. It has been shown that in the low moisture range, that is below the BET (Brunaeur-Emmett-Teller) monolayer value, water acts as antioxidant. As A w increases its oxidative rate decreases in this range. In the intermediate moisture range, that for most dried fish, water acts as a prooxidant and hence lipid oxidation increases with increasing moisture content in this range (Woolfe, 1975). Koizumi et al. (1980) confirmed this by using laboratory model systems of freeze dried flesh from various species. Various parts of fish also exhibited different degrees of lipid ojr oxidation which was 8 times faster in the skin^mackerel than in either the white or dark muscles (Nambudiry, 1980). Lipid oxidation produces hydroperoxides whose decomposition results in formation of aldehydes, ketones, alcohols, hydrocarbons and other products. Hydroperoxides may react with oxygen to form secondary products such as epoxyhydroperoxides which decompose to form volatile breakdown products which include hydroperoxide precursors. In addition, hydroperoxides and their products may react with proteins, enzymes and membranes (Frankel, 1984). Eventually, a threshold is attained whereby the product is no longer acceptable to the customers either from the aspect of overall appearance of the product or its aroma. 74 The nutritional value of oxidised fish oil is appreciably lower than that of unoxidised fat. There is also some evidence that lipid peroxides, an intermediate product of oxidation, are toxic (Toyama and Kaneda, 1962). Arai and Kinumaki (1980) found no evidence that the levels of rancid fat in cured products cause any toxicological problems when fed to rats, although vitamins A and E can be destroyed by lipid peroxides (Olcott, 1962) and dietary rancid fat may aggravate vitamin E deficiency (March, 1962). Chen and Pern (1982) confirmed that lipid oxidation occurred faster at higher temperatures as reported by Lea (1962) and Labuza (1971). By drying fatty and non-fatty fish in a cold forced air drought they demonstrated that lipid oxidation is not necessarily faster in the former as commonly thought. 2.5 Physical Attributes of Salted Dried Fish 2.5.1 Colour Colour is an important visual characteristic of foods. Other aspects of quality are often related to colour, for example ripening of fruit or colour changes accompanying deterioration and spoilage. Appearance is the combination of the visually perceived information contained in the light reflected, transmitted or scattered by the object. Translucency and opacity may be as important in describing appearance as colour and the directional reflectance from the surface. 75 Colour and other aspects of appearance influence food appreciation and quality since man has subjective standards for the acceptable range and preferred optima for these for almost every food. Schutz (1954) noted that 'spurious conclusions about food preferences may be reached by considering colour independently of flavour factors, and colours can be experimentally manipulated to serve as standards of good quality'. Little is known about the significance of colour perception in food acceptance. Observers use certain colours to judge acceptance, indifference or rejection. The human eye is not a quantitative instrument and hence precise colour measurement requires modern instrumentation. 2.5.1.1 Colour definition The visible spectrum lies in the range of 400-700nm (violet-red) and the eye is most sensitive to differences in colour in the green-yellow region (520-580nm). The C.I.E. system (Commission Internationale de l'Eclairage) depends upon three primary colours: red, blue and yellow. Matching an unknown colour against a proper mixture of these gives a satisfactory colour match. A set of stimuli called XYZ were chosen to represent them for mathematical convenience. Although it is not quite true, for ease of remembering, the X value maybe considered as degree of redness, Z blueness and Y greenness. Colour mixing is not a mixture of sensations but a mixing of a stimuli to give a unitary sensation. With certain pairs of complementary colours, however, 76 intermediate hues are not obtained but rather some shade of one or the other of the pair, or a new colour (Amerine et al. , 1965). In 1976, the CIE introduced a useful method for quantifying the appearance of a surface colour. Three new parameters, L*, a* and b* were defined in terms of X, Y and Z which are the tristimulus values for the material and Xn, Yn and Zn are the tristimulus values of a white object and these correspond to the normalised tristimulus values of the illuminant. The HUNTER L, a, b opponent-colours scale was designed to give measurements of colour in units of approximate visual uniformity throughout the colour solid. Thus, L measures lightness and varies from 100 for perfect white to zero for black, approximately as the eye would evaluate it. The chromaticity dimensions (a and b) give understandable designations of colour as follows: a measures redness when plus, grey when zero and greenness when minus; b measures yellowness when plus, grey when zero and blueness when minus There is a mathematical relationship between the Hunter L, a, b values and the CIE X, Y, Z values which is described elsewhere (King, 1980). 2.5.1.2 Application to fish products Most of the work on colour measurement has been with minced fish (Young and Whittle, 1985; Whittle et al., 1980; King and Ryan, 77 1977). This was mainly used as an index of whiteness and to detect the presence of greyness from melanin pigments of the skin, black blemishes from pieces of skin or black belly membrane, or red and red/brown discolourisation due to the presence of blood or dark muscle. Hincks and Stanley (1985) attempted measuring colour changes of the surface of squid by taking optical measurements at 420 and 520 nm which represented the absorption maximal of the oxidised and reduced state of the ommachraome pigment responsible for the squid integumental colouration. Their results showed the decline in colour over storage time and the squid stored under non- contact icing conditions were superior to those stored in iced sea water for maintenance of bright reddish-brown skin colour. ca« No reports exist on the use of^Tristimulus colour grading system to follow or order colour development in dried fish. Salted dried fish undergo browning on storage. Browning in salted dried fish is non-enzymic and of two origins: (1) the Maillard browning and (2) the interaction of oxidising lipids and protein. In the first, the reactions are principally between the amino acids and sugars to form Amadori compounds (Dworschak, 1980). The major contributors to browning and the loss of free sugar from the extractives of fish are anserine and taurine, as in the squid (Haard and Arcila, 1985). Secondly, some browning results from the reaction of fat oxidation products with the basic nitrogenous constituents of the flesh (Dworshak, 1980, Khayat and Schwall, 1983; Frankel, 1984) 78 forming high molecular weight brown materials known as melanoidins. Browning can also occur as a result of browning of the extractives (Jones, 1962). Browning which lowers the quality of properly dried salted white fish is derived primarily from sugar-amino reactions. A pronounced activation of the discolouration by Cu++ indicates that the oxidation of fat plays little part. Ca++ does not potentiate the browning of such material but activates the discolouration of poorly prepared salt fish by halophilic bacteria (Jones, 1962). However, Lee et al. (1960) demonstrated that much of the loss of biologically available lysine from heated herring meal results from reaction between fat oxidation products and protein rather than the sugar related reactions common in non-fatty fish. Browning rate increases with increasing moisture and is maximum in the intermediate moisture range (Rockland and Nishi, 1980) (Fig. 2.3). However, Heiss and Eichner (1971) suggested that interdependency of spoilage reactions is such that browning was noted in freeze-dried salmon which had a moisture content low enough for fat oxidation to be a serious problem. Browning occurred despite the fact that the Maillard reaction was inhibited and was attributed to the reaction of carbonyls (formed during the fat oxidation) with the amino groups of proteins. Thus, Heiss and Eichner (1971) drew a line of reactivity that sums up the two reactions of lipid oxidation and Millard reaction (Fig. 2.6). 79 CD E r- C- 4~> • ^ O 4-> c ro -t-> O x <4- C-r- •«- 034-* 03 • • +-> U -O storage) co 03 C C CD •<- •» O 5- c c_> ^ o 'D-—T- 4-> i~ -r- +j ro ro EE O >— ro >> CD Qjr- i- Cl prolonged CD ro T3 ■> / •“ -T—1 co • • (O (O ‘f- i— O'-----C r— a fte r 13 O -r- 73 i. ZZ 03 b Or- ZZ X rO • • D +-> • • C -C 03 and cr-r- cn — • f— •i— ro -X CC CD COr— 1-4, C-r- CD O O E • JC ■r- •—' CL E 4-> CD -Q co 13 fOr— O co D- O X> E oc C -M •• •r- 5 ro ro T3 C. CD X3 ro C\> ~ +-> CD jq Qj -r* 73 i- O CD O •r- C *4- E ro O •• =3 T3 -*-> co gj ro cn cz *♦— £ O CD •• 0 4- •r- _|l O +-> o 4-> co i- CL - E • i— • cd C3C 3 c S- oo C\J CD L- ZJ A}l L^nb sso[ CD Li- 80 Dworschak (1980) reviewed the destruction of a number of amino acids, including lysine during heat treatment. The loss of available lysine is greater during storage under fluctuating, than at constant temperature; (Lee et al. , 1982). Incorporation of antioxidants in the browning system improved the retention of available lysine (Yen and Lai, 1987). Browning in these products has also been attributed to the oxidation of myoglobin (Williams, 1976). Mathew and Parpia (1971) have also reported that flavanoid-like polyphenols take part in some non-enzymic browning reactions. Caramelisation can occur concurrently with non-enzymic browning. Non-enzymic browning is often measured by digestion with trypsin (Labuza, 1973; Choi et al., 1949) or pronase (Palombo et al. , 1984) followed by measurement of the absorbance of the clear aqueous extract at 420nm. Petriella et al. (1985) and Buera et al. (1987 a,b,c) measured colour development in non-enzymic browning reactions by % V spectrophotometric tristmulus colour measurement. The metric > saturation Suv, chroma C uv and the metric chroma C ab were chosen as the most suitable functions. Both groups observed that all the three colour functions show similar behaviour. An initial lag phase or induction period of short duration is followed by one of rapid colour increase and finally the rate of colour development tends to diminish. The induction period refers to the time during which browning is slow and colourless intermediates are formed (Clegg and Morton, 1965). Temperature and pH exerted a strong influence (Petriella et al. , 1985). 81 Generally, the presence of water enhances the Maillard reaction, but according to known data, reactivity is greater in anhydrous media (Dworschak, 1980). However water contents of 8 to 10% were optimal for the formation of fructoseglycine (Reynolds, 1965) whilst an Aw of 0.65 to 0.70 is maximal for the decomposition of lysine (Jokinen and Reineccius, 1976). Kunimoto et al. (1985) observed that the browning of carp occurred more rapidly in high than low humidity environments and that loss of available lysine was proportional to the browning of protein moiety. 2.5.2 Microscopic studies Light and electron microscopy has been used to elucidate the structure of fish. Jarenback and Liljemark (1975) have published photographs of ultrastructure of cod myofibrils and Partmann (1967) has reviewed the fine structure of vertebrate fish muscle. Howgate (1979) also reviewed the structure of white and red muscle of fish and the effects of processing on the structure of fish muscle. Bremner and Hallett (1985, 1986) demonstrated degradation in muscle fibre-connective tissue junctions in iced fish. Connell (1957) examined the microscopic structure of dried cod muscle. The air-dried sample had a compact, coherent structure with few spaces between the fibres; the freeze-dried samples had an open cellular structure with the spaces originally occupied by ice-crystals still retained. Although the freeze-dried samples almost regained their original weight on rehydration, the water was not fully resorbed into the fibrillar tissue. Similar 82 results were obtained by Kanna et al. (1971) using sea bass. Drying temperature is a factor controlling the resorption of water by air dried fish. Samples dried at 0°C regained virtually all their original volume, whereas those in the sample dried at 18-20°C remain partially shrunk. This is confirmed by the water contents of the reconstituted products which were 51% and 72% for the samples dried at 18-20°C and 0°C respectively, compared with 76% in the original. Connell (1957) and Kanna et al. (1971) also observed that freeze-dried fish does not fully resorb water into the tissue and voids remain where ice crystals have been. Bromlei (1949) studied the microscopic changes in both hot and cold smoked fish. There was no gelatinisation of connective tissue during cold smoking but there were signs of desiccation with diminution of extracellular space. During hot-smoking there was complete gelatinisation of the connective tissue of the sarcolemma and myocomata. Charley and Goertz (1958) studied the effect of cooking of salmon at several temperatures on the microscopic structure of the flesh. Granular material was precipitated on the fibres during cooking and the connective tissue was gelatinized. There was also precipitation of granular material in the connective tissue areas. The effect of heating trout muscle at 60°C and 97°C was studied by Schaller and Powrie (1972) using scanning electron microscopy. The structure of the fibres was retained though there was disintegration of the filaments in the A-band and H- zone. Transmission and scanning electron microscopy revealed that all cooking methods resulted in the development of 83 supercontraction bands alternating with areas showing tissue fragmentation and tearing. Microwave cooking produced smaller and less dense supercontraction nodes in pre-rigor muscle with less tearing and fragmentation but more fibre separation (Hsieh et al., 1980). Rodger et al. (1984) detected little disintegration of myofilaments but extensive break-up of Z-lines in salted and acidified herring muscle. Microscopic studies are therefore useful techniques in elucidating changes influencing the charateristics of the fish muscle, such as texture which are not visible to the eyes. 2.5.3 Thermal studies of fish The texture and appearance of fish change during heat treatment. The most important heat-induced process is the thermal denaturation of muscle proteins which initiates various other processes such as muscle aggregation. To understand and measure its extent a technique termed as the differential scanning calorimetry (DSC) has been applied (Wright, 1982). It is one of the three most commonly encountered thermoanalytical techniques, the other two being thermogravimetric analysis (TGA) and differential thermal analysis (DTA). DSC is a technique for monitoring changes in physical or chemical properties of materials as a function of temperature by detecting thermal changes associated with such processes. In DSC the measuring principle is to compare the rate of heat flow to the sample and to an inert material which are heated or cooled at the same rate. Changes in the sample that are associated with absorption or 84 evolution of heat cause a change in the differential heat flow which is then recorded as a peak. The area under the peak is directly proportional to the enthalpic change and its direction indicates whether the thermal event is endothermic or exothermic. The peak temperature is often referred to as Td, T , and Tmax. The heating curves or the plots of the differential heat input against temperature are termed thermograms. DSC has been applied to thermostability studies of proteins. In attempting to relate the quality of cooked meat with the denaturation of actin and myosin many investigators have studied the thermostability of isolated proteins and extrapolated their findings to the whole muscle (Hamm, 1977). The behaviour of the proteins may differ between the intact tissue and in the isolated preparations. DSC has the advantage of being able to study the thermal properties of proteins in the intact tissue and in the isolated preparations. DSC has the advantage of being able to study the thermal properties of proteins in the intact tissue with no need for solubilisation (Biliaderis, 1983). The technique has allowed the thermally induced transitions of rabbit and beef actin, myosin and sarcoplasmic proteins to be studied as a function of pH and ionic strength (Wright et al., 1977; Stabursvik et al. , 198if.) . The effects of process conditions on beef and vegetable protein denaturation have also been investigated (Quinn et al. , 1980; Arntfield and Murray, 198J) . Fish proteins are more unstable than mammalian proteins to physical processes such as freezing and frozen storage (Shepherd, 1960). DSC analysis would help to establish whether this 85 D ifferential h e a tin p u t 300 Figure 0. 310 2.7: 02mcal/s Thermogram 1977). 320 Temperature of 330 86 rabbit K myosin 340 (Wright 350 et al., 360 instability exists during other food manufacturing processes. It also affords and an opportunity of establishing whether the thermal denaturation characteristics of fish muscle proteins are dependent on the environmental conditions since fish species live in temperature of -2° to 28°C with no control of their body temperature. Hastings et al. (1985) found cod muscle to have the same basic DSC profile as rabbit muscle, made up from a pattern of myosin, sarcoplasmic proteins and actin (Wright et al., 1977). The main difference was that the major myosin transition in cod occurred at about 10°C below that in rabbit. However, Findlay and Stanley (1984) do not agree that the three major endothermic transitions of beef muscle are the discrete events attributed to myosin, collagen and actin by Stabursvik and Martens (1980) but are in fact a net response of the muscle proteins reflecting their association and environment. Connell (1961) found that myosins of fish species were much less stable than those of ox, rabbit or chicken muscle. He concluded that myosins of different species had adapted to different body temperatures and that there was a connection between this phenomenon and resistance of a muscle towards changes caused by processes such as freezing and frozen storage or dehydration. DSC revealed that freezing cod muscle at -10°C for 2 weeks effected some denaturation on the cod myosin with little subsequent effect on the myosin transition (Hastings et al. , 1985). The Tmax value for the first peak decreased as did its relative size whilst actin and collagen are largely unaffected by frozen storage (Hastings et al., 1985 and Poulter et al. , 1985). 87 Quinn et al. (1980) and Hastings et al. (1985) observed that treatment with salt depressed both transition temperatures and peak areas of fish muscle. Studies of Wright et al. (1977) indicated that actin would be more destabilised by salt than myosin or sarcoplasmic protein. It was made more sensitive to heating by that which decrease its denaturation temperature from 82°C to 72°C. Quinn et al. (1980) and Stabursvik and Martens (1980) studied beef muscle and reported that Ca 2+ and Cl - ions 3_ had a strong destabilising effect on actin whilst PO^ may have a stabilising effect. Various proteins are affected differently by neutral salts (Joly, 1965). The ions K+, Na+ and Cl~ had little effect on the stability of the native conformation of several proteins (von Hippie and Wong, 1964). Quinn et al. (1980) noted that a degree of reversibility in the salt effect was evident in the thermograms of beef which was salt treated and later dialysed. DSC thermograms for cod muscle dried for 2h at various temperatures showed that for drying at 30°C the transitions were close to the raw state but at 45°C onwards the major myosin transitions were lost and the actin transition temperatures were increased and its peak area diminished. After drying at 90°C, there were no discernible peaks. Parsons and Patterson (1986) reported a promising correlation between maximum treatment temperature and the onset temperature of denaturation. The duration of heat treatment was reflected in the area of the thermogram. As the moisture content of the muscle decreased, the transition temperatures increased. No thermal transitions 88 remained after overnight drying of cod at 30° or 45°C/ after 4h at 60°C or lh at 90°C (Hastings et al., 1985). This study highlighted the lower stability of fish myosin compared to fish actin. Hagerdal and Martens (1976) suggested that proteins become more stable to heat as the moisture content fell, this could explain the relative stability of actin. As for myosin, loss of moisture may initiate irreversible aggregation reactions as in frozen storage. DSC profiles of temperate (cod) and tropical (tilapia) fish revealed that proteins of the latter species are more thermostable. The transition temperatures for myosin of tilapia are over 10°C higher than cod myosin (Hastings et al., 1985 and Poulter et al., 1985). The cod myosin shows a doublet peak whilst that of tilapia shows a singlet. Isolated myosin may display three transitions depending upon pH and ionic strength (Wright, 1978; Wright and Wilding, 1984) although only one is seen in the fresh muscle of beef. The second peak in cod is quite pronounced at Tmax of 53°C whereas there was none in tilapia. The third peak which by analogy to meat protein is the actin transition was centred at 73°C for both fish. It is also well established that the connective tissue proteins (collagens), of various species increase in stability with increase in In vivo body temperatures (Menashi et al. , 1976). An onset temperature (Tq) for the collagen peak was 31°C for insoluble cod skin by Finch and Ledward (1972; 1973) and Menashi et al. (1976) but Poulter et al. (1985) observed mean values of 34.0°C and 40.3°C for two sets of cod samples, presumably reflecting the 89 differences in the water temperatures in which they were caught. The corresponding value for tilapia was 57°C. DSC therefore represents a useful technique for studying changes to protein occurring during salting and drying of fish. 2.5.4 Reconstitution Properties Shrinkage during the drying of biological samples is often looked upon as a volume shrinkage, with no distinction between different dimensions. Balaban and Pigott (1986) however found the shrinkage to be highest along the width and thickness. Dried fish, whether prepared in the open air, over a fire, or in a well-controlled air tunnel, can never recover the properties of the fresh from which it was made, no matter how efficient the process or how good the storage. In general, the faster the drying process, the easier it is to put the water back in the fish when preparing it for eating. Freeze dried fish, from which the water has been removed very quickly, looks much like fresh fish when reconstituted although differences in taste and texture are still discernible (Connell, 1957 and Kanna et al., 1971). Thus, dried fish should always be regarded as a completely different product with its own particular flavour and texture (Waterman, 1976). Reconstitution, however, is not an absolute criterion for preparing dried fish dishes. Most product in South East Asia is cooked and served in the dried state. Reconstitution practices seems more important with minced salted dried products. In the 90 preparation of 'kheema', salted dried minced fish brickets are reconstituted. In a study of this product Sudhakharan and Sudhakhara (1985) reported that an oven-dried sample which was dried for the shortest time had a higher rehydration ratio than a sun-dried product. Salted dried minced fish of 15% salt content had the greatest water-binding capacity which was retained in the cooked product (Bligh and Duclos, 1981). Rehydration of capelin mince showed that those dried at higher temperatures had lower rehydration indices (Rustad and Nesse, 1983) which corresponded with the least extractable protein. Changes in the sedimentation pattern and viscosity of myofibrillar protein occurs faster than the decrease of the solubility of myofibrillar protein (Kanna et al., 1966). Dried fish has a reputation of inferior texture when reconstituted. The same has been found with meat. Deficiencies in texture have been attributed to the loss of water-holding capacity by the muscle proteins and protein denaturation during drying (Suzuki, 1981). Air-dried product is highly compact and feels tougher in the mouth than the less compact freeze-dried product. On the other hand samples of freeze dried fish felt more fibrous (and dry) in the mouth than fresh fish, presumably because individual fibres of the dried product, although easily separated from one another, were less soft and less completely hydrated than those of fresh fish. The very large differences in dehydration rates which exist between the different products can also be well explained by reference to 'micro-structural' 91 differences even allowing for different ratios of volume to surface area of the samples. Freeze dried samples allow water to be carried deep into the piece by the capillarity of an extensively ramifying system of spaces permeating the whole (Wang et al., 1954). With the compact tissue of air dried fish, water penetrates to the centre mostly by diffusion through the protein of the fibre itself. With further work on dried samples of previously rapid frozen fish Connell (1957) was able to report that there was an optimum pore-size for rapid reconstitution. Although differences in toughness and fibrousness between these dehyrated products maybe explained, partly at least, on the basis of differences in microscopical structure there remain differences, between the products and fresh fish, which must have their basis in the condition of the fibre substance itself. It has already been established that product on storage can become tougher and drier in texture without alteration to its micro structure (Cutting et al. , 1956^ establishing that this change resides in the fibre itself. Such changes can be ascribed to cross-linking processes between previously denatured protein molecules. The linkages described could be characterised as (i) electrovalent or salt-linkages (ii) covalent linkages (iii) hydrogen bonds and (iv) van der Waals attractions (Connell, 1957). With salt linkages, presumably the divalent cations form inter chain bridges by means of the carboxyl groups of adjacent peptide chains. High salt concentration within the 'salting-in' range of 92 a water-salt protein system may dissociate salt-links between neighbouring peptide chains (Connell, 1957). A wide variety of covalent linkages between peptide chains can be envisaged, but the most likely to occur are amide linkages formed from carboxyl and amino groups and disulphide bridges formed by oxidation of sulphydryl groups. Other interactions are possible between sulphydryl groups and carboxyl- or amino-groups and between carboxyl and hydroxyl side-groups, for example. Of these only the disulphide bridge is generally accepted as normally occurring between adjacent peptide chains of the same protein molecule or between peptide chains of two adjacent molecules. With the presence of some considerable amounts of potentially reactive non-protein substances, other types of covalent bonding cannot be disregarded (Connell, 1957). The formation of covalent cross-linkages would lead to a diminution in the total numbers of acidic and/or basic groups in the protein. In dried fish the contents of acidic groups are higher than those of fresh fish where some of the loss of basic groups could be accounted for by 'browning' reactions involving amino-groups of lysine. Increases in acidic groups and decreases in basic groups have been observed in heated pure dry proteins (Connell, 1957). The formation of disulphide bridges or such linkages as -S-CO- result in a loss of total sulphydryl groups. The sulphydryl groups of most native proteins have varying reactivity and only become fully available when the protein is denatured. 93 In a concentrated solution of urea the solubility of the protein is increased and this additional solubilisation is accompanied by marked swelling of the dried products (Bremner and Vail, 1983). Such behaviour is characteristic of insoluble protein systems that are stabilised by intermolecular hydrogen bonds (Connell, 1957). Apart from indicating their denatured state, the low solubility of dried products, indicates a stability imparted by cross-linkages. Physical analysis of dried fish muscle also demonstrated that denaturation of myosin is irreversible irrespective of the drying temperature and actin recovery was more complete at the lower drying temperature of 30°C (Hastings et al. , 1985; Connell, 1957). Apart from indicating their denatured state, the low solubility of dried products, indicates a stability imparted by cross-linkages. Physical analysis of dried fish muscle also demonstrated that denaturation of myosin is irreversible irrespective of the drying temperature and actin recovery was more complete at the lower drying temperature of 30°C (Hastings et al., 1985). 94 2.6 Organoleptic Property of Salted Dried Fish There exists the view that as products of simple, cheap processing methods, salted and dried fish are of low quality. Infested, strong smelling cured fish as a consequence of uncontrolled processing and poor storage are not unusual and it is believed the local consumers accept this situation as a matter of necessity. However, the person who has an intimate knowledge of the fish markets and who is acquainted with local production, marketing and consumption habits and problems and has studied these products, knows that the consumers (albeit with some exceptions) are able to distinguish between the different qualities of preserved fish (FAO, 1981). Sripathy (1983) reported that there have been occasions when regular consumers of dry fish have been offered an ideally produced laboratory product and complained of bland taste and flavour. Although conditioned to the available commercial product, consumers can be conditioned to accept better products. It seems however, that there is a need to study and codify the perceived quality of dry fish by different cross-sections of consumers. What is off-flavour to research scientists could be desirable characteristic to the consumers. Quality, in simple terms, can be defined as those organoleptic characteristics or attributes which make foodstuffs acceptable. Generally, the consumer will pay more for fish considered to be of higher 'quality' and will continue to buy it as long as the quality remains constant (Azeza, 1982). However, FAO (1981) 95 noted that high quality products do not attract a correspondingly higher market price. Consumer preferences vary from country to country; from region to region, between individuals and may well change over the years. In order to control quality, it is therefore, essential to know what the consumer is seeking when buying fish. Quality is more important in South East Asia with dried anchovy, dried small shrimp and dried squid. In the more quality conscious countries such as Thailand, Singapore and Hong Kong, these varieties can be separated into as many as eight categories according to quality, grade and size. Hong Kong is the most selective market in terms of product type and quality (Infofish, 1983). Some quality characteristics are inherent in the fish on capture (intrinsic quality); others are associated with the post-harvest fate, i.e. quality deterioration or spoilage, and loss of quality during handling and processing (extrinsic quality). The more important factors that determine quality are: species presence or absence of bones appearance of fish and flesh absence of parasites ease of preparation freedom from pathogenic microorganisms odour condition flavour composition freshness packaging si ze The quality factors that effect the sensory judgement of the food may perhaps be grouped according to various senses being used: 96 (a) Sight and Touch The selection of species and size, basic to any fishery, falls into this category. Hand sorting needs little training. Visible signs of deterioration are detectable by trained or untrained people. In almost all cases visual detection of deterioration and defects is accomplished efficiently and rapidly. For the assessment of gloss on articles like smoked fish the practised eye has so far proved better than any instrument (Connell, 1980). In assessing textural attributes (firmness, softness, mushiness, rubberiness, woodiness, mealiness, succulence, dryness) the sense of touch in fingers or mouth are used as occasion demands. If a taste test is being conducted it is sometimes convenient to include an assessment of texture as well as of odour and flavour. For the most part there is no substitute for sensory assessment of fish texture, though some instruments are available for measuring degree of firmness. Colour matching can be done very effectively by eye. Grading of salmon, tuna and fried breaded products is aided by comparing their colour with a set of tiles or cards carrying different permanent shades. The use of instruments to measure the colour of fish and fish products is very limited because of the problems involved (Connell, 1980). In the proposed Malaysian Standard for grading of ikan bilis (dried anchovy), whitish/yellowish colour has been suggested for grade A, darker for grade B and dark greyish for grade 97 C. Texture is dry and firm for grade A, dry and less firm for grade B and moist for grade C. Breakage is less than 5% for grade A, 5-10% for grade B and greater than 10% for grade C (Mat, 1983). Dried fish should harbour no insects, larvae or mould; the skin and flesh is not, or only slightly damaged and no intestines, blood or foreign matter should be visible (Dagbjartsson, 1983). Texture is perceived as hard, that is it should not be compressible between two fingers. The Codex Alimentarius Commission (1985) has proposed a draft standard for dried salted fish of the gadidae fish family which stipulates standards similar to the Malaysian ones above. With regards to sensory properties it says that 'dried salted fish shall have organoleptic properties which are characteristic of the product and shall be free from any objectionable odour'. (b) Odour and Flavour Responses in the mouth are limited to the basic tastes of saltiness, sweetness, bitterness and sourness. Overall flavour includes much of what is experienced on smelling. Thus, odour and flavour must be taken together. Taste and smell are powerful tools in assessing quality. The smell of fresh fish can easily be distinguished from bad fish. With some practice the whole pattern of changes in odour can be differentiated easily and rapidly, enabling degree of freshness to be accurately determined. Similarly, off- odours, rancidity, taints and unusual intrinsic odours are readily detected and their intensity judged reproducibly. 98 Although the quantitative measurement of certain chemical constituents can be used in some circumstances, tasting is the most reproducible method taking all types of product into account and is capable of assessing far more satisfactorily the changes in character of the flavour (FAO, 1981) . Different levels of tastes can be distinguished easily so long as comparisons are possible but the measurement of degree of saltiness or acidity with accuracy is difficult. Furthermore, the reaction to saltiness depends to some degree on how much fat is present (Connell, 1980). Odour is a difficult attribute to describe in relation to dried fish. It is completely different from that of fresh fish but is very strong and characteristic of specific products. Any change from this characteristic odour is noticeable. Rancidity is one of the major causes of deterioration in dried fish and it brings a rather pungent smell to the product. Flavour is also influenced by odour but it encompasses the sensations of smell, taste and mouth feel. Salted dried fish is characteristically salty and with a slightly tangy taste. The fibres feel tougher than fresh fish in the mouth but it must be seen as a different product to fresh fish. Toughness in the fibres is tolerated only to a certain degree. Very tough texture is abhorred and so is a very salty product. One of the ways of trying to reduce both the toughness and saltiness is by rehydration (FAO, 1981) 99 2.7 Packaging and Storage of Salted Dried Fish Packaging is one of the major means of reducing losses of foodstuffs during handling, storage and distribution. The basic functions of any package may be defined as: (a) containment, enabling a specified quantity of foodstuff to be handled conveniently as a single unit (b) protection of the foodstuff against the various hazards of distribution (c) providing information about the product. An effective package will, while fulfilling these functions, assist in minimising the total unit costs of the distribution system. Cured fish, of both European and tropical origin, is distributed almost exclusively in bulk packs of 10 kg upwards. Traditional packages involved hessian sacks, jute sacks, palm leaf cartons, uncovered cane baskets as well as wooden, plastic and cardboard boxes (FAO, 1981). In India, alkathene-lined gunny bags for bulk storage and polythene covers for retail packing are being popularised (Gopakumar and Devadasan, 1983). Dried fish is brought into Malaysia and Singapore from Thailand or Burma by motor carrier. In Indonesia, dried fish is often transported between islands by small motor vessels of 200-300 t cargo capacity. In the Philippines, inter-island shipments travel by commercial cargo vessels and containerised cargo is 100 coming into use from major out-ports. Intra-regional and foreign shipments move by commercial shipping lines. Storage facilities vary enormously between regions. Warehouses in Africa are often open, mud or mat walled and roofless (Osuju, 1976). In these, the fish is unpacked, sometimes redried and repacked for storage and distribution. In other areas roofed warehouses are used and the fish may be stacked loosely on the floor where products are liable to rodent attack and insect infestation. In Hong Kong,Malaysia and Singapore, the products are often kept under cold storage although in retail shops or market stalls fish are generally kept loose in baskets or on mats. Large ones are even hung from the ceiling and pieces cut to consumer requirements. In these regions few products are retailed as packaged items, one exception is perhaps the thread bream in Malaysia. Traditional packages provide little other than containment. Wooden or cardboard boxes and some baskets may provide some protection against mechanical hazards. Any closed package provides some resistance to rain. Of the packages indicated, probably only a well-made closed wooden box provides any control of moisture exchange with the atmosphere. Little work has been reported on improving the packaging of cured fish products. Laboratory investigations, such as those of Sen et al. (1961) and Kandoran and Valsan (1974) have been concerned primarily with the effects of moisture-barrier wrapping materials and were not followed up by full-scale commercial trials. 101 Exporters from Europe have introduced some changes, in particular the use of polyethylene film shrink-wrap around a corrugated board case for salt cod, and woven plastic fabric instead of jute for baling stockfish. Adoption of appropriate improved packaging methods could give useful reductions in losses in cured fish. Cured fish products present several packaging problems. The first of these is its geometry is irregular. Very often it has sharp protrusions which may puncture the packaging materials. Dried fish may be brittle and any attempts to force it into a regular shaped package may cause damage to the extent of fragmentation. The second problem with packaging is moisture. For safe storage without microbial spoilage the product must have an Aw below 0.6. In an atmosphere of humidity greater than 60%, the dry fish will pick up moisture, with consequent risk of spoilage. FAO (1981) suggested using vapour proof packaging but at the same time pointed out that in cases of 'case-hardened' fish, moisture would eventually diffuse from the centre and the humidity within the package would be raised and mould growth could occur. Such risk is even higher when the package is exposed to the sun. Recommendations on packaging made to World Food Program (Murray and Jobber, 1968) include the use of paper wrap in immediate contact with the fish to absorb any condensed water, with the maximum moisture content limited to 28% (salt free basis). Substantial contribution of packaging to product protection depends on the adoption of bulk packages which reduce breakage, 102 lower infestation levels and prevent moisture uptake which could lead to damage by microorganisms. To reduce breakage it is necessary to choose a package which will be sufficiently robust to retain its integrity throughout marketing and which will distribute stresses of impact and stacking. Movement of individual pieces of fish should be prevented, and it should be possible to fill the pack by hand without damaging the product. Rigid packages would be preferable, perhaps with some internal padding or intermediate wraps. The use of bales and sacks should be avoided for fragile products, especially smoked fish, since breakage is possible during filling and compression as well as in transit. Baskets are of variable performance, often giving no protection against stacking loads. Containers of wood or corrugated fibre-board could give adequate mechanical protection against stacking loads. Where these are in use already they may require simple changes in design or in quality of materials or construction to reduce damage. Reduction in package size to 15 kg would be helpful. To combine mechanical protection with moisture resistance, various types of box might be considered for evaluation including waxed corrugated fiberboard, polyethylene- coated solid fiberboard and polypropylene profile board. For the package to act as a physical barrier to insects it must be well sealed, with no apertures greater than 0.2 mm diameter, and of a material which is not readily penetrated. Boxes such as those of fiberboard and polypropylene could be made reasonably insect-resistant by sealing all openings with adhesive tape. Wooden boxes could be sealed with a flexible line r and sacks of 103 suitable multiwall paper construction could resist insects. Woven sacks will never be completely insect proof but other materials of potential either as sacks, baling materials, box- liners or components of a multiwall sack, include woven plastic fabrics (coated and uncoated), cross-laminated plastic films, microporous polypropylene film and spun-bonded polyolefins. However, these will be useful only when the products have been treated to eliminate infestation. Paper treated with pyrethrin formulations as has been used in sacks for cereals is worthy of consideration (FAO, 1981). The use of other insecticides has been investigated (Golob et al., 1987; Taylor and Evans, 1982). Cured fish may be stored at the processing site, at the point of marketing and distribution, and at intermediate collecting centres. There are various reasons for storage, including the need to accumulate an economical load for transport, evening out seasonal variations in supply, waiting to find a buyer or for transport, and as a commercial strategy. The duration of storage at each point in the chain may vary substantially. While the attention given to storage should be greater where the period is long, severe damage could result from even a short period of bad storage. Under the very primitive conditions found in small- scale processing operations, fish will inevitably become infested. Reprocessing maybe useful in disinfesting and redrying fish, and hence reducing losses when storage facilities are very crude. Storage facilities should provide at least security, a roof to shade the stored fish from rain and sun, and protection from rising damp. A good store should be rodent proof, bird 104 proof, insect proof, have controllable ventilation, and be suitable for fumigation. Factors determining the storage life of cured fish include Aw and pH of the product, temperature of storage, degree of smoking and microbial load. For traditional dried products the Aw is the major factor in determining if microorganisms will grow. The inhibitory effect of reduced Aw which is not sufficiently low to ensure that all microbial growth is prevented, coupled with the fact that it takes a finite time for microorganisms to cause spoilage, means that the Aw need only be reduced to ensure that the product is sufficiently well preserved to prevent spoilage during storage. There is no necessary advantage in reducing the Aw to levels which will ensure long-term preservation, indeed this may not be possible for many products. However, Aw should be matched as closely as possible to that required maximum storage life. 105 CHAPTER 3 EXPERIMENTAL 3.1 Processing 3.1.1 Fish samples Four species of fish were bought from the Sydney fish market. Morwong (Nemadactylus macropterus) were bought whole but were filleted in the market. Sardines (Sardinops neopilchardus) were bought whole, ungutted. Shark {Notogaleus rhinophanes) were bought in fillet forms while squid (Nototodarus gouldi) were bought whole. When brought back to the laboratory the fish were cleaned, sardine was left intact whilst squid were cut open, ventrally and eviscerated, the tentacles were left on and the skin pulled off. 3.1.2 Salting Saturated brine was prepared and primed at 30°C. Fish were placed in a bucket containing brine in a fish/brine ratio of 1:2 w/v and the surface of the fish were weighted down. The brining was conducted at 30°C. Fish were removed at intervals. One lot was taken for drying, the other for analysis. Any lot which was not immediately processed was kept at -18°C. The squid was not salted. 106 3.1.3 Drying Drying was conducted at 30°,40°, 50°, 60° and 70°C. No attempt was made to control the relative humidity. Drying was carried out in a cross draught dehydrator till sufficient moisture hae| been driven off to a moisture content below 30%. The wet bulb temperature was recorded. Weight losses were monitored by weighing at suitable intervals. 3.2 Chemical Analysis 3.2.1 Moisture content Minced fish samples were weighed into tared lidded aluminium drying dish. The dish was placed in a vacuum oven at 70°C under a pressure of 70-80 pKa. Drying was carried out to constant weight (24h). Percent moisture was calculated as follows: loss in weight of sample after drying % mature = ------weight of wet sample 3.2.2 Salt (Sodium chloride) Salt content was determined using silver nitrate titration with potassium chromate as indicator (Pearson, 1970). Approximately 2g sample was accurately weighed and macerated for about 2min. in distilled water. The extract was transferred quantitatively into a 250ml volumetric flask and made up to volume. Aliquots (25ml) of this extract were titrated against 0.1N silver nitrate solution using potassium chromate (5% w/v solution in water) as indicator. 107 3.2.3 Protein content About lg homogenised ground sample was accurately weighed onto a tared nitrogen-free paper. Total nitrogen was determined using a Kjel-Foss automatic protein analyser 16210-A/SN which gave readings based on N x 6.25. Read-out x 1000 % protein = ------Weight of sample (mg) 3.2.4 Fat content Fat content was determined using an acid hydrolysis method (Wills and Greenfield, 1982). Approximately 2g of wet homogenised fish was accurately weighed into a 50ml beaker. The sample was moistened with 2ml alcohol, and 10ml 7N HC1 was added while stirring with a watch glass on top, the beaker was placed on the steam bath for 40min. with frequent stirring. At the end of the heating, 10ml of alcohol was added. When cooled, the mixture was transfered to a Mojonnier fat-extraction flask in a fume cupboard. The beaker was rinsed with 25ml diethyl ether added in 3 portions and consequently transferred to the Mojonnier flask eh which was stopped and shaken vigorously for lmin. Petroleum ether (25ml) was added, the shaking repeated and the sample was then left to stand. The clear ether-fat supernatant was decanted and filtered into a previously weighed Erlenmeyer flask through a cotton plug. The liquid remaining in the Mojonnier flask was re extracted twice, each time with 15ml of each ether, shaken well and decanted. The funnel was washed thoroughly with a few ml of 108 the mixture of the two ethers in equal volume. The solvent was evaporated slowly on a steam bath and the flask was dried at 100°C for 90min. The flask was removed and left to stand for exactly 30min. in a desiccator before reweighing. This weight was corrected by a blank determination. 3.2.5 Water activity (Aw) Ground samples were placed in small closed plastic containers and allowed to equilibrate for at least 24h. Saturated salt solutions covering the range of the samples were prepared in similar closed plastic containers and equilibrated similarly. A Novasina AG Hygrosensor (Model en ZFBA-3(4)ePP) was used throughout this study. The sensor operates through adjustable plug-in units (EZFB-3(4) within a range of 0.10-1.00 Aw. Aw measurements were made by placing a small, plastic container (40mm diam x 12mm deep) containing a sample in a special holder. The sensor, mounted in a gasketed and enclosed fitting was then used to cap the sample holder. Readings were taken directly from the 10-100% equilibrium relative humidity (ERH) meter scale and converted to A^ by division by 100. Readings were made to two decimal places. A standard curve was made using the Aw for the corresponding salt solutions prior to the measurement of the samples. 109 3.3 Solubility of Proteins 1. Solubility in 0.1M potassium chloride Homogeneously ground sample was suspended in 10ml 0.1M KCl at 25°C and was shaken mechanically for 30min. at lOOrpm. The sample was filtered and the total nitrogen content of the filtrate was measured on the Kjel-Foss automatic protein analyser (Obanu et al., 1975). 2. Solubility in sodium dodecvl sulphate + B mercaptoethanol Approximately lg ground sample was accurately weighed and suspended in 50ml 3% SDS + 1% B mercaptoethanol for 30min., heated in a boiling water bath for 30min. and centrifuged at 6000 x g for 20min. The nitrogen content of the clear supernatant were determined on the Kjel- Foss protein analyser (Obanu et al., 1975). 3.4 Digestibility of Protein by Pepsin Pepsin solution of 0.2% (activity 1:10,000) in 0.075N HCl was prepared and warmed up to 42-45°C. 150ml of the prewarmed freshly prepared pepsin solution was added to approximately lg ground sample which was accurately weighed in stoppered erlenmeyers and incubated with constant agitation for 16h at 45°C. After incubation the flasks were removed and the residue allowed to settle. A preweighed dried filter paper (filter paper dried at 110°C in moisture dish and weighed, Wl) was placed in a 110 buchner funnel, suction was applied and the filter paper moistened with water. Pepsin hydrolysate was filtered rapidly and the remaining residue in the flask was removed by washing with 15ml acetone. Washing was repeated thrice if necessary. The filter was then transferred to its moisture dish, dried at 70°C in a vacuum oven, cooled and weighed as before (W2). Percent indigestible residue was calculated as follows: (W2 - Wl) % Indigestible residue = ------X 100 g sample Indigestible protein was determined using the Kjel-Foss automatic protein analyser. Percent protein was calculated as follows: read out % Protein =------X 1000 actual sample weight (mg) This result represented the weight of indigestible protein in the sample and was converted to percent indigestible crude protein as follows: % indigestible protein in sample Protein Indigestible ------X 100 % total crude protein in sample (AOAC, 1984: Method No. 7.053 - 7.059). Ill 3.5 Isoelectric Focussing 3.5.1 Materials Electrophoresis unit (Pharmacia Flat Bed Apparatus FBE-3000) Constant power supply (Pharmacia ECPS 2000/300) Agarose IEF (Pharmacia) Pharmalyte carrier ampholyte (Pharmacia) Gel casting frame Agarose IEF Accessory kit (Pharmacia) containing: gelbond (114 x 225mm) sample applicators electrode strips (6 x 10mm) filter paper Levelling table Spirit level Spring clips Scalpel Hair dryer 3.5.2 Reagents Gel composition: 0.3g Agarose IEF 3.6g Sorbitol 27ml distilled water 1.9ml Pharmalyte (pH 3-10) 112 Electrode solution: Cathode (-) 1M NaOH Anode ( + ) 0.05M H2SC>4 Fixing solution 5% Sulphosalicyclic acid and 10% Trichloroacetic acid in distilled water Destaining solution: 35% Ethanol and 10% Acetic acid in distilled water Staining solution: 0.2% Coomasie Blue R250 in destaining solution 3.5.3 Sample preparation lOg samples were macerated in an electric blender in 50ml of distilled water. The homogenised sample was transferred quantitatively to a centrifuge tube and was centrifuged under refrigeration at 8000 x g for 30min. The supernatant was poured into dialysis tubing quantitatively and dialysed against distilled water at 4°C with 8 changes of water for 48h. The dialysed solution was reduced in the freeze drier and the final volume was made up to 25ml in a volumetric flask. 113 3.5.4 Methods 3.5.4.1 Casting the gel The agarose IEF and sorbitol were dissolved by simmering in water. While the agarose was being dissolved, the levelling table was made horizontal using a spirit level. About 2ml of distilled water was poured onto the leveling table and a sheet of gelbond with the hydrophilic surface on top was placed onto the levelling table. The film was rolled flat using a glass tube to remove excess moisture and air bubbles beneath it. The plastic film and surroundings were dried with tissue paper. The gel casting frame was placed in position over the plastic film and fastened down with spring clips. The levelling table and plastic film were prewarmed to approximately 45°C using a hair dryer. 3.5.4.2 Moulding After all the agarose IEF has been dissolved, the mixture was allowed to cool to around 75°C. Pharmalyte was added and mixed thoroughly before pouring onto the mould. The gel was allowed to set for 15min. and a scalpel was run around the edge of the casting frame and the frame and gel were removed from the levelling table. The prepared gel was left for lh at 4°C or overnight at room temperature to harden fully before use. 3.5.4.3 Running the gels About 2ml of distilled water was poured onto the cooling plate and the gel was put over it, ensuring that the water spread to a 114 thin film under the gel plate. The excess was removed with paper towel. Electrode strips were soaked in appropriate solutions and blotted on filter paper for about 2min. to remove excess liquid. They were then carefully placed on top of the long edge of the gel on the corresponding side. 25ul samples were applied using the paper sample applicators. The constant power supply was set to deliver a maximum of 15W and 1500V with unlimited current. Gels were run for 90min. with a coolant temperature of 8-10°C. The sample applicators were removed after 45min. 3.5.4.4 Fixing and staining The gel was fixed immediately in the fixing solution for 30min. and washed in two lots of destaining solution first for 30min, then for lh. It was then dried by placing three layers of filter paper on top of it, followed by a glass plate and a weight of about 1kg. After 15min, they were removed and the gel was dried with a hair-dryer. The gels were then stained in a stained solution for 15min. and destained overnight. Finally, the film was dried again with a hair dryer. The methods adopted were from Agarose IEF leaflet by Pharmacia Fine Chemicals (1984). 115 3.6 Total Amino Acid Analysis 3.6.1 Materials The Waters HPLC Amino Acid Analyser consisted of: Ninhydrin AAA system consisting of 2M510 pumps Absorbance detector M440, dual channel Sample processor M710B Automated gradient controller M680 Data module M730 Column heater Post-column reaction system Reactor and heater (specially designed) Filter, Millipore Durapore, HVL P04700 3.6.2 Reagents Constant boiling HC1 Ninhydrin Ninhydrin, Pierce 21003 ph Dimethylsul^oxide (DMSO) Pierce 20687 Hydrindantin Pierce 24000 Lithium acetate buffer pH 5.2, Pierce 27203 Buffer A: Trisodium citrate dihydrate, Pierce 27209 Nitric acid, Merck Buffer preservative, Pierce 27207 116 Buffer B: Trisodium citrate dihydrate, Pierce 27209 Nitric acid, Merck Sodium chloride, AR, Univar Amino acid standards, Pierce 3.6.3 Reagent preparation 3.6.3.1 Constant boiling HCl 500ml AR grade HCl was mixed with 500ml H20 in a 2L round bottom flask and 200mg SnCl2 added. The solution was mixed gently and heated slowly to boiling and distilled. The first 200ml of distillate was rejected and the next 500ml was collected. The solution was used for acid hydrolysis of samples while the residue was disc 3.6.3.2 Ninhydrin 750ml of dimethylsulfoxide (DMSO) was poured into a 2L amber glass bottle and sparged for lOmin. with high purity, ammonia- free nitrogen. 18g of ninhydrin was added and sparging continued until all materials were dissolved. lg hydrindantin was added and sparging continued. 250ml 4M lithium acetate buffer pH 5.2 was added and sparging continued for 15min. The reagent was stored under nitrogen for 12h prior to use. 117 3.6.3.3 Buffer A 800ml of distilled water purified by Milli Q system to remove ammonia was added to trisodium citrate dihydrate (19.6g). After dissolution, 3 drops of buffer preservative was added and pH adjusted to 3.08 + 0.02 with concentrated nitric acid. This was made up to 1L with water and pH readjusted to 3.08 if necessary. The buffer was then filtered through Millipore Durapore filter, sparged for 20min. with high purity nitrogen and blanketed with the same until required. 3.6.3.4 Buffer B 800ml of distilled water purified by the Milli Q system to remove ammonia was added of trisodium citrate dihydrate (19.6g) and of sodium chloride (58.4g). After dissolution, 3 drops of buffer preservative was added and the solution made up to 1L. Concentrated nitric acid was added dropwise to adjust pH to 6.45. The buffer was filtered, sparged for 20min. with high purity nitrogen then blanketed with the same until required for use. (Adapted from Waters Associates). 3.6.4 Methods 3.6.4.1 Acid hydrolysis 0.5g freeze-dried sample (equivalent to 50mg protein) was placed in a screw-capped test-tube with 2ml of 6N HC1. It was then flushed with nitrogen for a minute and frozen in acetone-dry ice. The tube was tightly sealed and hydrolysis was carried out 118 in an air oven at 110°C for 24h. After hydrolysis the solution was dried on a rotary film evaporator to remove the HC1. The residue was resuspended in trisodium citrate dihydrate buffer (Buffer A) and the hyrolysate filtered through Whatman No. 1 filter paper to remove humin. The sample was transferred to 25ml volumetric flask and made up to volume with Buffer A. 3.6.4.2 Preparation for chromatography A Sep-pak C18 cartridge (Waters Assoc.) was activated with two 10ml volumes of methanol and then washed with two 10ml volumes of double distilled water. This was further washed with 10ml water: methanol (80:20). 1ml sample was mixed with 2ml water:methanol (70:30) and passed through the Sep-pak cartridge. The first 1ml was discarded and the next 2ml were collected. 20ul of this sample was injected into the amino acid analyser. An amino acid standard mixture (0.5um/ml) was injected in a similar manner as the samples. Amino acid concentration was expressed as g amino acid/16gN. 3.6.4.3 Chromatographic conditions Column: Waters AAA, 4.6 x 250mm Part No. 80002 Column temperature: 55°C Reactor temperature: 110°C Eluents: A 0.066M Trisodium citrate, pH 3.08 B 0.066M Trisodium citrate and 1.0M sodium chloride, pH 6.45 119 Reagent: Ninhydrin Wavelength: 546nm Elution time: lOOmin Gradient employed: Time Flow % A % B Curve o 0 IT) 100 0 45 0.5 20 80 8 60 0.5 0 100 8 84 0.5 0 100 6 86 0.5 100 0 6 3.7 Rancidity Determination of thiobarbituric acid (TBA) value in a trichloracetic acid extract of fish (Vyncke, 1975) was adopted. 20g of ground fish sample was homogenised with 100ml of 7.5% trichloracetic acid (TCA) solution and 0.1% (w/v) of both propyl gallate (PG) and ethylenediamine tetra acetic acid disodium salt (EDTA) for lmin. in a blender and filtered. 5ml of TBA reagent (0.02M 2-thiobarbituric acid in distilled water) were added to 5ml of filtrate in test tubes with screw caps which were placed. In a boiling water for 40min. After cooling absorbance was read at 538nm on a Varian-Tecjrpm UV-visible spectrophotometer model 635. The original TCA extract was used as a blank. A standard curve was constructed (Appendix I). Recovery tests were carried out with 1,1,3,3-tetraethoxy propane (TEP) which is hydrolysed to malonaldehyde. 120 3.8 Scanning Electron Microscopy Samples were freeze-dried and kept in a d££j Representative areas were photographed. 3.9 Colour The colour of the stored samples w 3.|Q Differential Scanning Colorimetry 3.10.1 Equipment and materials Differential scanning colorimeter, Du Pont Instruments Series 190 121 Sample pan, Du Pont 900796.901 Sample lid, Du Pont 900790.901 Chart paper, Du Pont 990528 Biphenyl, AR Grade, BDH 3.10.2 Procedure Approximately lOmg samples weighed with an accuracy of ± O.Olmg were sealed in volatile sample pans. The instrument was programmed as follows: Heating rate: 10°C/min. Range: 10°C/min. Shift: + 20cm Range (sensitivity): 2mv/cm Initial temperature: 5°C Final temperature: 120°C The DSC head was precooled using liquid nitrogen. The sample pan was placed on the sample holder with an empty pan as the reference. One minute was allowed for equilibration of the sample chamber prior to starting the program. A stream of dry nitrogen was passed through the DSC head at 30ml/min throughout the procedure. Endotherms were recorded and denaturation temperatures observed. For each sample the mean of 3-5 determinations was obtained. Temperature and enthalpy calculations were carried out using biphenyl as a standard for which various weights (0.4, 0.8, 1.2 and 1.6mg) were examined. The endothermic areas were measured by 122 using the method as described by Pope and Judd (1977). A linear 2 regression equation was obtained by plotting area (cm ) and mg biphenyl (Appendix 2). The denaturation energy was then calculated by using the formula as follows: Slope x A^ x Si x Hc Hi - x q So A^ = Area of sample, cm gS^ = Sensitivity, cal/sec/cm = Mass of sample, g S = Sensitivity of standard, cal/sec/cm M. Hq = Enthalpy of standard, cal/g 3.11 Reconstitution Properties 20g of dried fish were placed in 1L distilled water in a stainless steel container at 30°C. The product was allowed to rehydrate for lOmin. At the end of the rehydration period, samples were removed and drained through a filter paper for 2min. and the product removed from the paper and weighed. The rehydration ratio was calculated by dividing the maximum rehydrated weight by the initial weight. 3.12 Sensory Evaluation All dried products prepared were subjected to sensory evaluation. The products were rehydrated in tap water for lOmin., drained and ^ried in oil at 200°C for 3min. before serving to a panel of 25 assessors who were familiar with salted fish products. 123 Evaluation was carried out using Hedonic scale ranging from a score of 10 for the extremely liked to 1 for extremely disliked sample. The panelists were asked to assess the saltiness, aroma, colour, texture and overall acceptability (Appendix 3). All samples were identified by random numbers and the results were analysed by the Least Significant Difference Method. 3.13 Storage Studies Products which were judged the most acceptable were used in the storage studies. Products were packed in low density polyethylene bags and kept at 5°C, 25°C and 37°C for six months at ambient relative humidity. Products were evaluated fortnightly for visual mould, colour, moisture uptake, water activity, soluble protein and digestibility. 3.14 Statistical Analyses Analysis of variance (ANOVA) was carried out on the SPSS package of University of New South Wales Computer System. 124 CHAPTER 4 RESULTS AND DISCUSSION Products were prepared by salting and drying under various controlled conditions, based on Morwong (Nemadactylus macropterus), sardines (Sardinops neopilchardus), shark (Notogaleus rhinophanes) and squid (Nototodarus gouldi). 4.1 Processing of Fish Procedures used were chosen to simulate those commonly utilised in the tropics. Brining with saturated salt solution is widely used in South East Asia and offers the advantage over dry salting that the brine may be reused. Morwong was selected for this study because its size and shape resembled that of Mergui, a common salted, dried fish in S.E. Asia. Squid is expensive and popular as a dried product. Sardine on the other hand is well liked for its slightly tangy taste whilst shark though not well established as a salted dried product has potential for development as one. 4.1.1 Morwong Fresh Morwong contains 78.7 + 0.43% moisture, 18.05 + 0.29%, protein 2.1 + 0.6% fat and 1.8 + 0.14% ash. These values compare well with those of sea bream reported by Iwaski and Harada (1985). The pH of the fresh fish was 6.3 and the total volatile base (TVB) content was 16.4mgN/100g and trimethylamine (TMA) content was 14.3mgN/100g. The latter is below the 30mgN/100g 125 acceptable limit (Connell, 1980). The pH, TVB and TMA contents indicate satisfactory freshness in the fish. The amino acid composition of the morwong is given in Table 4.1. The data for morwong compare well with those of sea bream although serine, histidine, proline and arginine (4.11, 6.79, 4.18 and 9.39g/16gN respectively) were higher than those reported by Iwasaki and Harada (1985; 2.88, 3.08, 2.97 and 5.71g/16gN respectively). Tryptophan was not determined because of loss during acid hydrolysis of the samples. The ratio of the essential to non-essential amino acids (E/NE) based on 10 essential amino acids as previously calculated (FAO 1970) is 0.75 compared to 0.77 as reported by Iwasaki and Harada (1985). The E/NE ratio for morwong based on 12 essential amino acids (Rechcigl, 1983) is 1.11. This means that morwong is an excellent source of dietary amino acids. The in vitro protein digestibilities of various fish species are summarised in Table 4.2. The digestibility for morwong is 98.4%. With its high protein content, good amino acid profile and high digestibility, morwong provides a good source of protein in the diet. 126 Table 4.1: Composition of amino acids in fresh shark, morwong, sardine and squid (g/16gN) Amino acid Shark Morwong Sardine Squid Aspartic acid 9.66 9.91 9.00 9.79 Threonine 4.88 4.80 4.54 4.64 Serine 4.40 4.11 4.14 4.43 Glutamic acid 15.08 12.76 11.62 14.56 Proline 3.87 4.18 9.27 5.83 Glycine 4.46 3.99 6.24 5.71 Alanine 7.35 5.96 8.57 5.87 Cystine 0.90 0.50 0.87 0.93 Valine 5.49 5.04 6.13 5.11 Methionine 3.56 3.33 3.13 3.76 Isoleucine 5.36 4.28 4.66 5.29 Leucine 8.17 6.77 7.84 8.11 Tyrosine 3.58 4.50 3.02 3.29 Phenylalanine 2.69 4.21 2.88 3.30 Lysine 10.83 9.47 8.11 8.39 Histidine 2.39 6.79 4.22 2.36 Arginine 7.34 9.39 6.13 8.68 Ratio = Essential 0.8 0.75 0.62 0.69 Non-essential (based on 10 amino acids) Ratio = Essential 1.11 1.44 1.06 1.17 Non-essential (based on 12 amino acids) 127 Table 4.2: In vitro digestibility of fish species Sample % protein digested Shark 99.3 + 0.52a Morwong 98.4 + 0.81a Sardine 99.5 + 0.43a Squid 88.0 + 0.56b Means with the same superscript a or b are not significantly different at 99 percent confidence level 4.1.1.1 Brining During brining, salt penetrates the fish flesh with accompanying loss of moisture. Under ideal conditions salt uptake will continue until salt concentration in the aqueous phase of the tissue becomes equal to that in the brine. Factors such as brine concentration, time in the brine, temperature and size of fish will influence salt penetration (Graham et al., 1986). Figs. 4.1 and 4.2 demonstrate the inverse relationship between salt uptake and water loss during brining (Crean, 1961). Morwong fillets (length 19.7 ± 3.6cm, max thickness 1.4 + 0.4cm) presented a large surface area for salt uptake, although this was restricted by skin on one side of the fillet. Crean (1961) reported that surface area is the biggest factor affecting the uniformity of salt uptake. Fish thickness would also influence salt uptake since the thicker the fish, the more slowly the salt diffuses into the flesh. Fillets of 2.5cm thickness reached 10% salt 128 L sd at \% = 0.1880 ( qp ) ^ juajuoD 129 tp?s “ CM S H Figure 4.1: Salt uptake by different species of fish during brining. —° Morwong - o Shark ( ) % juajuoa 130 airus|o^ o I © CM 1 CQ & H * 1-1 p 05 £ « Figure 4.2: Changes in moisture content during brining. content after 24h, but fish of 5cm takes 3 days to reach 10% salt content (Waterman 1976). Fig. 4.2 shows that moisture loss is rapid in the first 12h, decreasing from 80.8% initially to 68.5% at 12h and levelling off to 65.5% after 48h salting. Moisture loss was the least in morwong (Fig. 4.3). This probably reflects the water holding capacity of the different fish. Fig. 4.1 shows that the salt uptake for morwong reached equilibrium after 8h when it contained 12% (wb) salt and 69% moisture. Whether the salt distribution was uniform throughout the flesh was uncertain as comminuted samples were analysed. Nevertheless since salt equilibrium was reached in 8h, this time was chosen for salting morwong prior to drying at different temperatures. 4.1.1.2 Drying of morwong fillets Fig. 4.4 shows that drying rate for morwong was significantly faster (P < 0.01) at 70°C than at other temperatures. The next fastest rate was at 60°C followed by 50°C, and lastly 30°C and 40°C which did not differ significantly from each other. Drying rate decreased with drying times. The drying curve did not exhibit the constant rate period. Wuttijumnong (1987) observed similarly when drying salted sardines. Thus the rate of water evaporation depends on the diffusion of water from inside the flesh to the outside (Jason, 1980). Table 4.3 shows the moisture content and Aw of the products. Drying at 50°C and above gave Aws smaller than 0.75 with concomitant lower moisture contents and higher salt contents. When comparing the rates of drying at 50°C with the other 2 fish 131 - -o Shark 132 CQ £ •§ i- J 05 Figure 4.3: Percent moisture loss in fish during brining, o------o30°c o------o 40°C ------o 50 °C e----- « 60°C *_---- « 70°C Lsd at 1% = 0.0855 Drying time ( h ) Figure 4.4: Effect of temperature on drying rates of salted morwong. 133 species (Fig. 4.5) the rate of drying for morwong is the second slowest although significantly faster than sardine. Although morwong presented a large surface area for evaporation, the drying is restricted because of the presence of skin on one side of the fillet, preventing drying from occurring from all sides. Furthermore salt content in morwong is high (40%, db) and as reported by Waterman (1976), would slow down drying rate. 4.1.1.3 General remarks on the dried products Rates of drying are also influenced by the relative humidities (RH). RH was determined by ambient conditions and drying temperatures since the dryer did not possess RH control mechanisms. However it appeared from a wide range of experiments that variation in RH at a given temperature had no apparent effect on product quality in terms of appearance, reconstitution properties, SEM examination and IEF patterns. Generally rates are highest when drying at 70°C and 60°C but such products were often of burnt appearance, brittle texture with salt crystals on the surface. Although a drying temperature of 43°C was reported to be suitable with an inlet RH of about 45-55% giving a good product (Waterman, 1976) in tropical countries, 50°C is considered the optimum temperature for all the four species in this study after considering both product quality and rates of drying. The stability of any dehydrated foodstuff is closely related to Aw rather than the total moisture content. A dry foodstuff will absorb or desorb moisture depending on ambient RH and also its 134 D rying r a te s( g H20 / g Dm-h ) Figure Lsd 4.5: • o o o ------* ------ at o o o Sardine Shark Squid Effect Morwong 1*. 0.0545 of 135 Drying species time on ( h drying ) rates at 50°C. intrinsic properties (Muslemuddin et al., 1984). Drying at 30°C and 40°C often produced products with rather high moisture content and Aw of approximately 0.80 (Table 4.3). In the case of morwong a much longer drying time of 30°C and 40°C would reduce the final moisture content to a more acceptable level. However for all the other drying temperatures the products had reasonable moisture content and Aw is below 0.80. Growth of mould is inhibited at 0.80 and below whilst halophilic bacteria do not grow at 0.75 and below (FAO, 1981). Waterman (1976) suggested that moisture content be below 25% to prevent bacterial spoilage, but also mentioned that a higher moisture content may be acceptable when some salt is present. On storage, the products may gain or lose moisture depending on RH and hence keeping quality will be affected. To minimise such changes, products should be protected by adequate packaging. In Table 4.3 it can be seen that the predicted shelf-life of morwong dried at 30°C is only 1 week. The predicted shelf lives of products dried at 40°, 50° and 60°C are 1^-2 months while that dried at 70°C was one year. Although microbial attack will be retarded at such low Aw browning reaction and lipid oxidation may be accelerated (Labuza et al., 1970). 4.1.1.4 Sensory evaluation of morwong Irrespective of chemical or physical properties and storage life, organoleptic properties of a product determine its acceptability to the consumer. The production of dried marine products in this 136 Table 4.3: Aw and moisture contents in dried morwong Drying Temp °C Aw Moisture Salt Predicted measured content(%) (% wb) shelf life+ 30 0.83 47.2 21.3 1 week 40 0.79 37.2 25.3 1*5 months 50 0.74 29.6 28.3 2 months 60 0.71 24.7 30.3 2 months 70 0.69 11.9 35.5 1 year +predicted from Poulter (1980) and Curran (1984). study involves a period of time in saturated brine (except for squid) and drying at various temperatures. Thus brining time and drying temperatures were individually examined for their effects on acceptability of the dried products. (a) Effect of brining time Morwong brined for various times were dried at 50°C to approximately 30% moisture content and Aw 0.75. The samples were then presented to a taste panel after 10 min. reconstitution in water at 30°C and frying in oil at 200°C for about 7 min. Saltiness scores are presented in Table 4.4. Salting for 2 and 4 hours gave products which were significantly preferred (p < 0.01) over samples brined for longer times because of lower saltiness. 137 Table 4.4: Mean scores for saltiness in dried products. Products Salting Time (h) 2 4 8 12 18 (15.6)+ (20.8) (26.4) (29.1) (30.2) Sardines 7.97a 7.67a 5.57bc 5.49c 3.69d 2 4 8 18 12 (23.3)+ (28.0) (43.5) (44.4) (42.4) Morwong 7.95a 6.95ab 6.76b0 5.32c 3.95d 4 2 12 8 18 (43.6)+ (43.2) (45.6) (48.3) (46.4) Shark o to 7.21a 6.82a 6.41a 4.50b u> +Figures in parentheses denote salt content (% db) in the samples. Different superscripts along horizontal column indicate significant difference at 99% confidence level. Even though fish with lower salt contents (i.e. short brining times) scored highest, selection of brining time for preparation and storage studies was based on achievement of equilibrium salt content and hence stability (Section 4.1.1.1). For morwong an 8h salting time was chosen. (b) Effect of drying temperature Morwong brined for 8 hours were dried at 30°, 40°, 50°, 60°, and 70°C. Products were reconstituted and presented to the panel for evaluation of aroma, colour, texture and overall 138 acceptability. In Table 4.g it is seen that aroma did not vary significantly with drying temperature. Colour varied significantly with drying temperatures. The lightest coloured products were obtained by drying at 30°C and 40°C and these were preferred over the other products. Texture scores did not vary significantly with drying temperature. Finally, overall acceptability did not differ significantly between drying temperatures. Based on colour preferences drying rate and overall product guality appraisal, 50°C was chosen as the drying temperature for morwong. 4.1.1.5 Chemical properties of salted dried morwong 4.1.1.5.1 Effects of salting (a) pH Fig. 4.6 shows changes in morwong pH during brining in saturated salt. Initial pH of 6.3 decreased to 6.0 during 4h salting after which it rose significantly and remained constant up to 48h. The initial depression could possibly be due to the removal of water soluble amines as moisture is being withdrawn by the salt. A slight rise in pH after the fourth hour may indicate development of bases in the fish, possibly due to microbial degradation. Any such degradation would be limited by the salt. Beatty and Fougere (1957) reported that salt concentration higher than 12% would retard microbials. This concentration was achieved after 139 CJ £ H I € QQ brining. during fish in pH 4,6: Figure H<* 140 Table 4.5: Mean scores for salted dried products+ Attributes Drying Temp°C 70 60 40 30 50 Aroma 7.31a 7.23a 6.76a 6.26a 5.98a 30 40 50 60 70 Colour 7.2a 6.92ab 6.85b 6.80b 3.43C 60 50 60 40 30 Texture 7.33a 6.87a 6.85a 6.13a 5.60a 70 40 60 40 30 Overall acceptibility 7.22a 7.07a 6.92a 6.30a 5.82a +Different superscripts along horizontal column indicate significant different at 99% confidence level only 8h salting. Malle, Eb and Taillez (1986) found high pH in a culture medium used to study the level of contamination in fish muscle with NaCl in the medium. Fish proteins are denatured in saturated brine with 10% salt in the aqueous phase of the fish being sufficient to start the denaturation (Beatty and Fougere 1957). One manifestation of protein denaturation is loss of water holding ability. Hamm (1960) accounted for the salting out effect by saying that the electrolyte ions attract the water molecules more strongly than do the protein molecules, therefore the former 141 'rob' the latter of their hydration water, thus reducing their water holding capacity. Kida and Tamoto (1969) also noted that in general pH decreases with decreased water holding capacity of muscle tissue. Kolodziejska and Sikorski (1979) also observed that protein solubility decreases in fish containing NaCl or KC1 with changing pH, thus reducing its water holding capacity. Regenstein et al. (1984) also demonstrated decreased water binding potential in fish samples on pH decrease. The positive charge of the protein is neutralised by Cl- which decreases its water holding capacity especially at pH around its isoelectric points and below. This could therefore also contribute to decreased pH during salting. (b) Total volatile bases (TVB) and trimethylamine (TMA). Both TMA and TVB increased in the first 4h of salting (Figs. 4.7 and 4.8) after which they declined gradually. The initial rise may be due to microbial degradation before the fish could acquire the necessary salt concentration to retard spoilage. More than 5% salt was observed to retard the formation of TMA in mackerel homogenates (Ishida et al. 1976). TVB and TMA production ceased as salting progressed. This is consistent with Fujii et al. (1977), Nozawa et al. (1979) and Bilinski and Fougere (1959) who reported salt depressed the growth of TMA producing organisms. Labrie and Gibbons (1937) and Cardin et al. (1961) noted that TMA contents were much higher in lightly salted products than in heavily 142 o------o Shark •-----'•Sardine Lsd at 1% ■ 0.3647 M S !J B 001/N 143 Bui Figure 4,7: TYB content in fish during brining. brining, during fish in content o o TMA 4.8: Figure q*H B OOl/N Buj 144 salted ones, consistent with decreasing volatile base production as brining continued. (c) Protein solubility Protein solubility decreased significantly (p < 0.01) during brining. Solubility of salted morwong is very poor in KC1 at 25°C being 22% at the beginning and decreasing to 10.1% at the end of brining. At 77°C the solubility of proteins in KCl slightly improved (Fig. 4.9). Meinke et al. (1972) observed similarly. The solubility in KCl at 77°C is 32% at the beginning and 18% at the conclusion of brining. Poor solubility was expected as Duerr and Dyer (1952) observed that when fish muscle was immersed in concentrated brine, the total myofibrillar protein rapidly became inextractable. Water holding capacity also declined. When the average salt content reached 10%, the concentration of the electrolytes reaches 2M, there is a decrease in bound water and a change in hydration which may result in precipitation (Kinsella^1982). However, this will not happen throughout the fish tissue until salt penetration is complete i.e. at about 10% (wb) salt content (Crean, 1961). Complete inhibition of the fish muscle proteolysis occurs at 12% NaCl (Bilinski and Fougere, 1959). When proteins have denatured, extraction by solubilising agents such as the ionic detergent, SDS, is necessary to dissolve the proteins. 145 salting. on media different in r>» cm proteins morwong of Solubility 4.9: Figure U|a)Ojd a|qn|os % 146 Solubility of morwong in SDS + J3 mercaptoethanol was 72% at the start of brining, declining gradually to 41.5% after 48h. Generally the solubility in SDS + B mercapthoethanol is higher than in the other two extracting media. This observation concurs with that of Rehbein and Karl (1985). Although Dyer and Snow (1950) demonstrated that maximum extraction of soluble proteins from unsalted fish is obtained using 3 and 7% NaCl in the extracting medium, the salted fish tissues in this study contain more salt than this and therefore maximum protein extraction would not occur. (d) Isoelectric focussing Shown in Fig. 4.10 are IEF patterns of the water-soluble proteins extracted from morwong fillets after brining for various times. Six major bands, designated A - F were observed. The disappearance of band C after brining was accompanied by enhancement of band D and the appearance of band B. As brining time increased, the intensity of most protein bands decreased, although the two most anodic bands E and F seemed unaffected. These changes provide evidence for protein denaturation caused by salt which increases with brining time. (e) In vitro digestibility Digestibility remained high throughout the salting period (Table 4.6) and changes in the digestibility over the 147 ok ik b I 4k C d I I ) } n mi i I 11 Mil i f 3tk 5 ) •KV \r\ A 3 c 0 E F Figure 4.10: IEF patterns of water soluble proteins of morwong fillets after brining for 0, 2, 4, 8, 24, 36 and 72h (a, b, c, d, e, f, g, h respectively). 148 salting period are negligible. Digestibility therefore was not negatively affected by salting. Table 4.6: In vitro digestibility of morwong Salting time (h) 0 4 8 12 18 24 36 48 % Digestible 98.4a 99.6a 99.6a 99.5a 99.4a 99.5a 99.5a 99.7 protein (f) Total amino acid in salted morwong Table 4.7 shows the total amino acid contents in morwong salted for different times. Amounts of each amino acid varied but no trends were discernible and these changes were not significant except for histidine (p < 0.01). This could well be due to its reduction to histamine by some microorganisms. Lysine seemed unaffected. This observation concurs with that of Takama et al. (1985) who also found no effect on the amino acids during salting and pickling of masu salmon. 4.1.1.5.2 Effect of drying on morwong (a) Total volatile bases (TVB) and trimethylamine (TMA) On drying the TVBN increased from 19.8mgN/100g to 29.6mg% at 30°C, 52.5mg% at 50°C and 64.6mg% at 70°C (Fig. 4.11). Increases in TMA were from 18.2mg% to 21.0mg% at 30°C/ 39.0mg% at 50°C and 49.5mg% at 70°C (Fig. 4.11). These results concur with development of amines in dried products 149 Lsd at IX. 4 .9 2 6 150 Figure 4.11: TVB and TMA contents in morwong dried at different temperature, Table 4.7: Total amino acid content in salted morwong Amino acid Salting time (h) g/16 gN 0 12 24 48 Aspartic acid 9.91 10.63 10.07 ].0.59 Threonine 4.80 4.88 3.37 3.58 Serine 4.11 5.98 5.41 4.21 Glutamic acid 12.76 18.01 14.34 ].6.48 Proline 4.18 4.84 4.47 4.05 Glycine 3.99 5.41 5.72 5.91 Alanine 5.96 7.05 6.68 6.17 Cystine 0.50 0.56 0.85 0.53 Valine 5.04 5.70 5.95 5.44 Methionine 3.33 2.86 3.14 2.68 Isoleucine 4.28 4.30 5.23 5.39 Leucine 6.77 3.18 8.48 6.59 Tyrosine 4.50 2.51 4.11 3.78 Phenylalanine 4.21 2.67 3.47 3.48 Lysine 9.47 9.89 8.86 9.19 Histidine* 6.79a 3.46b 2.43C 1.82 d Arginine 9.39 8.42 7.42 8.56 'fc Amino acid showing a decreasing trend, significant at 99% confidence level. observed previously (Hebard et al ., 1982). This is attributed to the instability of TMAO to heat (Sigurdson, 1947) but Tokunaga (1975) reported that the rate of thermal decomposition of TMAO to TMA iand DMA varied with species. Connell (1980) reported values of 100- 200mgN/100g of TVB in dried fish. Drying rate also affects rate of TMA and DMA formation in Alaskan pollock with faster drying resulting in 151 lower TMA and DMA formation (Kida and Tamoto, 1976). Drum- dried fish contained more amines than freeze-dried fish (Spinelli and Koury, 1979) emphasising the influence of processing conditions. Preheated whiting (40-60°C) also contained more amines than unheated muscle (Spinelli and Koury, 1981). (b) Protein solubility Howgate and Ahmed (1972) observed that the effects of heating and drying on the extractability of fish proteins differed between species. In KC1 at 25°C solubilities of morwong dried at 30°C decreased from 18.5% to 16.5%, to 16% when dried at 50°C and 12.2% when dried at 70°C. In KC1 at 77°C corresponding decreases in solubility were from 23.8% to 22% (30°C, drying) to 19.5% (50°C drying) and 16% (70°C drying). In SDS the solubility increased slightly in the product dried at 30°C from 61.2% to 63.8% and then decreased to 58.7% in the 50°C dried product and to 50.6% in the 70°C dried product. Decrease in solubility was observed by Migita et al. (1960) in fish dried at 5-10 °C. These workers reported that solubility of myofibrillar protein was lowered while denaturation of sarcoplasmic protein took place slowly and its solubility was lowered slightly. Suzuki (1981) reported that the heat coagulative sarcoplasmic protein adhered to myofibrillar protein when fish is heated. This leads to insolubilisation of the latter. Actin is also 152 % Soluble protein 10 30 50- Figure - - 4.12: o o — — — -*KC1 oSDS oKCl — Drying t e m p e r adifferent t u r e s Solubility 30 L -a 0 at at ------ 77°C temperature 25°C 153 media of , morwong L-x 50° on ( ------°C) drying proteins at different 70° in 1 „ soluble in KCl solution and is most probably not changed by low heating. However actin cannot be extracted out if myosin becomes inextractable (Howgate and Ahmed, 1972). Parsons and Patterson (1986) and Poulter et al. (1985) also observed decreases in protein solubility in heated fish samples. The denaturing effects of both salting and drying processes reduced solubility of the dried products perhaps by changes in the number and distribution of SH groups, formation of cross-linking S-S bands, aggregations, partial loss of hydration and interaction with other components (jSikorski 1980). All these could contribute to the loss of protein solubility. (c) Isoelectric focussing From Fig. 4.13 it can be seen that on drying morwong at 30°C band F appears to be unchanged with bands B,C,D and E disappearing. New bands B' (slightly more cathodic than B) and E' (sightly more anodic than E) are visible in this sample. At higher drying temperatures only E' and F are visible. These are anodic bands and this observation contrasts with earlier findings that cathodic beef proteins are more thermostable (Lee and Grau, 1966). Protein bands disappear after both salting and heating, reflecting the effects of high salt levels and heat on fish proteins. 154 k ^ pM 1- lo 6 WpL •TP 1)11 I 1)11 I ' M |.il i i | _ I I M I I I ^ I! | « I , ™ I I 3o #c I I itfc Cv ) I ) I So*c 4 A n c D ee'f ♦ «/M# Figure 4.13: IEF patterns of water soluble proteins of morwong fillets brined for 8h and dried at 30°, 40°, 50° and 60°C (b, c, d and e respectively) (a = 8h brined undried sample). 155 (d) In vitro digestibility Table 4.8 illustrates the effect of drying temperature on in vitro digestibility of morwong. The digestibility ranged between 98.4% in the fresh sample to 99.8% in the 70°C dried product. Sheikh and Shah (1974) also noted that digestibility was raised when fish were dried at 60°C and suggested that heating at these temperatures possibly produces simple peptides which are more susceptible to enzyme attack and therefore became more digestible. These results concur with Adachi et al. (1958) who detected no impairment of protein digestibility in haddock due to dehydration and Udarbe et al. (1985) who also observed high digestibilities (91-97%) in dried fish. Table 4.8: In vitro digestibility of morwong Drying temperature °C 0 0 0 0 o o o o in r- Fresh fish 30°C U Digestible % protein 98.4a 99.6a 99.8a 99.7a 99.8a 99.8a "fc Different superscript denote significance at 99% confidence level (e) Total amino acid content in dried fish Table 4.9 shows the amino acid composition of the dried fish. Changes are seen in individual amino acids such as 156 lysine, tyrosine, phenylalanine, arginine and histidine (p < 0.01). However, no trends were apparent for amino acid profile in relation to drying temperatures. Previous reports stated that changes in total amino acids were only discernible at temperatures >100°C (Miller et al., 1965 and Yanez, 1970). However, Clifford et al. (1980) reported lysine becoming unavailable during the smoking of fish. 4.1.1.6 Storage studies on morwonq Morwong, brined for 8h and dried at 50°C till the moisture content approximated 30% were packed in low density polyethylene bags and stored at 5°, 25° and 37oC at ambient RH. The products were intermittently examined for mould growth, general appearance and chemical composition. 4.1.1.6.1 Evaluation of stored morwonq (a) Product appearance The appearance of dried morwong during storage at the three temperatures can be seen in Plate 1. The products had good appearance at 5°C up to 8 weeks but after 24 weeks browning had developed. Products kept at 25°C became mouldy by 16 weeks. Products kept at 37°C were shrivelled, browned, brittle and were unacceptable after 4 weeks. The surfaces of the products kept at 5° and 25°C were covered by salt, possibly due to migration of salt onto the surfaces as moisture was lost during storage. The products kept at 5°C were still acceptable by the 24th week, despite browning and 157 Plate 1: Dried salted morwongs stored at 5°, 25° and 37°C; (a) for 4 weeks (b) for up to 24 weeks 158 Table 4.9: Total amino acid composition in dried morwong Amino acid Drying temperature °C Fresh fish 30°C 50°C 70°C Aspartic acid 9.91 10.65 10.36 9.12 Threonine 4.80 5.26 3.93 4.10 Serine 4.11 4.76 3.58 3.88 Glutamic acid 12.76 13.97 12.32 12.37 Proline 4.18 4.66 2.89 2.87 Glycine 3.99 7.11 4.78 6.40 Alanine 5.96 8.71 6.95 7.39 Cystine 0.5 0.78 0.57 0.58 Valine 5.04 5.95 5.29 4.75 Methionine 3.33 2.89 3.07 2.63 Isoleucine 4.28 3.94 4.73 4.13 Leucine 6.77 6.39 8.10 7.56 Tyrosine* 4.50a 3.13b 2.57C 2.65C Phenylalanine* 4.21a 2.61b 2.28b 2.49b Lysine* 9.53a 9.47a 9.48a 9.14b Histidine* 6.79a 2.22b 2.58bC 2.89c Arginine* 9.39a 7.43b 6.54c 5.05d Amino acid showing decreasing trend. Different superscripts denote significance at 99% level the products kept at 25°C were rejected by the 16th week. The products kept at 37°C were unacceptable by the fourth week of storage. (b) Product colour Browning during storage was followed using the Hunter Lab. colour meter. L, a, b, values are given in Table 4.10. L 159 Table 4.10: Hunter L, a, b for morwong stored at 5°, 25° and o * 37°C Storage time (week) Temperature °C 0 4 8 12 16 20 24 0 IT) 7 5.18a 73.22aC 70.77a 66.40b° 67.40b° 63.80b 66.45b° a - 0.90a - 0.50b - 0.40b - 0.45b 0.95c 1.50d 1.65d b 11.90a 11.85a 12.70a 13.10b 13.40b 19.60° 21.15d 25° L 75.18a 66.30b 69.05a 66.40b 63.95b 58.51° 60.90b a 0.9a 0.45b 0.85° 1.45d 1.65de 1.85e 2.00e b 11.90a 15.65b 16.70° 16.15° 19.35d 17.60e 19.75f 37o L 75.18a 66.75b 65.85b 60.93b 60.02b 56.15° 57.38° a 0.9a 2.00b 3.25° 3.30° 3.45° 3.95d 5.40e b 11.90a 18.35b 20.30° 22.30d 22.75d 23.85e 24.40e *Different superscripts along the horizontal columns indicate significance at 99% confidence level. L: Lsd at 1% = 7.1100 a: Lsd at 1% = 0.2532 b: Lsd at 1% = 0.9559 values decreased s.ignificantly (p < 0.01) as the product became darker during storage at all temperatures. The reduction in L values was quite rapid in the 25° and 37°C stored samples. The a values increased in all samples, more rapidly at higher storage temperature, indicating increased redness while b values also increased indicating increases in yellowness during storage. This was greatest in the 37°C stored sample. These increases in L, a, and b values help to 160 quantitate browning in the product. This browning could be due to the Maillard reaction, rapid oxidation and further reaction involving protein. These reactions may be reflected in the decrease in lysine during drying and storage. Lysine is usually lost more rapidly than other amino acids in stored products (Labuza and Saltmarch, 1981). Browning due to lipid oxidation-protein interaction is also possible as lipid oxidation occurs in the stored products (Section 4.1.1.7c) albeit less than reported values (Nambudiry, 1980). Such reactions are very important in the storage of dried salted fish as they also affect the organoleptic and nutritional quality of the product. (c) Product texture Toughness and dryness, problems associated with dried food products, intensify during storage (Villota et al. , 1980). Loss of moisture which accompanies cell structure damage results in a general decrease of water holding capacity which will affect the rehydrability of the product. Textural characteristics of dried foods are also greatly affected by storage conditions (Heldman et al. 1973). In this study products kept at higher temperatures were harder, rigid and brittle to touch, especially those stored at 37°C which lost a considerable amount of moisture upon storage. The 5°C samples were pliable throughout the storage period. Upon rehydration, the products stored at 37°C were very pliable. Niwa (1976) observed a conformational change in the 161 structure of the proteins of dehydrated fish which accounted for texture deterioration. Scanning electron microscopy (SEM) was used to examine the appearance of the products during brining and drying. Plates 2 a, b, c and d demonstrate the effects of brining on the structure of morwong flesh. As brining time increases the surface is disrupted increasingly and more salt is evident. This trend continues with longer brining (Plates 3 a, b, c and d) . Corresponding SEM micrographs of the deep tissue are shown in Plate 4. The fibres became more disrupted and greater quantities of salt are visible as brining time was increased (Plate 5). The disruption of the fibres could possibly be caused by withdrawal of water to the outside of the fibres, thus increasing salt concentration. This in turn could decrease water holding capacity and allow shrinkage of neighbouring fibres away from each other. Rodger et al. (19841)however detected extensive break up of the Z-lines in salted and acidified herring although little disintegration of the myofilaments was perceived. The micrographs of the surface and deeper tissues of fillets dried at 30°, 40°, 50° and 60°C are presented in Plates 6 and 7 respectively. Both series of micrographs indicate a decrease in compactness of the tissue structure as drying temperature is increased. This is reflected ultimately by the fragmentation of products dried at 60° and 70°C during 162 fish of layers respectively). d surface c, the b, of (a, 8h 4, (150x) , 2 , 0 for micrographs SEM brined 2: Plate 163 fish of . layers surface respectively) the b of (a, 48h (150x) and 24 for micrographs SEM brined 3: Plate 164 for brined fish of tissue respectively). d deep c, of b, (a, 8h micrographs and SEM 4, 4: Plate 165 24 for brined fish of . tissue deep respectively) of b and (a micrographs 48h SEM and 5: Plate 166 d fish c, of b, (a, layers 60°C surface and the 50° of 40°, (65x) . 30°, at micrographs SEM respectively) dried 6: Plate 167 dried fish of respectively). d and tissue c deep b, (a, the of 60°C and (125x) 50° 40°, micrographs 30°, at SEM 7: Plate 16b reconstitution. Decreased protein solubilities indicate some damage to the proteins in the dried products and this is further echoed in the rehydration behaviour of the dried fish. Irreversible changes such as the denaturation of the contractile proteins, heat shrinkage of fibres and heat coagulation cause a decrease in myofibrillar water holding capacity. This allows a large part of the muscle water to be released from the tissue. At temperatures >60°C collagen shrinkage occurs and more water is squeezed out with concomitant increase in toughness (Tarrant, 1982). This coagulation probably results from random associations between unfolded peptide chains (Hamm, 1977). All these effects possibly combine to give a tough chewy texture to dried fish. (d) Reconstitution Reconstitution of the dried fillets gave satisfactory products for samples brined up to 24h. There was some tendency, however for the flesh to break into segments. Brining for longer than 24h resulted in dried products with a greater propensity for fragmentation and with surface regions that dispersed as small particles. Variable susceptibility of dried fish to fragmentation has been found between species (Mills, 1977) and it appears that brining time also affects this behaviour. These defects are attributed to salt-induced protein denaturation, which increases with brining time. Rehydration ratio of the 169 products brined for different times and dried at 50°C is shown in Table 4.11. It can be seen that the ratio decreased as brining time increased. Generally the products did not reconstitute to the original state. The reconstituted products remained fibrous and moisture could easily be pressed from them. Increasing drying temperature also decreased rehydration ratio (Table 4.12). Reconstitution properties worsen with increasing drying temperature. At 30°, 40°, and 50°C the products were acceptable. However at 60° and 70°C the products were fibrous and tough. The initial physical nature of the 30° and 40°C are still quite dense whereas the 60° and 70°C dried samples were hard and brittle in texture. Temperature was shown to be the most important factor causing protein denaturation during drying and in addition, salt can denature the protein and alter its pH (Connell, 1958; Howgate and Ahmed, 1972). The rehydration ratios (Table 4.13) for the stored products decreased as storage time lengthens and more so at higher storage temperature. At 37°C the products rehydrated very poorly. Rustad and Nesse (1983) reasoned that when moisture content is reduced, the distances between the protein chains diminish. This therefore increases the formation of cross- linkages between them leading to a higher network of proteins with decreased water holding capacity and solubility. The products stored at 37°C were too damaged to be satisfactorily reconstituted. 170 Table 4.11: Rehydration ratio of morwong salted for different times and dried at 50 o C * Brining time (h) 4 8 12 24 36 48 Rehydration ratio 1.47a 1.19bc 1.21b 1.17bc 1.12c 1.02d *Different superscripts indicate significance (p < 0.01) Table 4.12: Rehydration ratio for morwong salted for 8h and dried at different temperatures Drying temperature 70 Rehydration ratio 1.43‘ 1.2 1.21 1.08 1.04b Different superscripts indicate significance (p < 0.01) ★ Table 4.13: Rehydration ratio for stored morwong Temperature Storage time (week) of ______storage °C 0 8 16 20 24 5°C 1.23a 1.21b 1.16° 1.09cde 1.04d 2 5°C 1.23a 1.17e 1.03fh l.llfe 0.98gd 37°C 1.23a 1.09ie 0.96gh 0.93g 0.94g •ff Different superscripts indicate significance (p < 0.01) 4.1.1.7 Chemical properties of stored morwong (a) Moisture content The moisture content in the products (Fig. 4.14) over storage time decreased significantly (p < 0.01) except at 171 o co -ac H3 O Ln CVI o 5 at storage during JC * CM ~ £ im in v «oj ot* content moisture in Changes 4.14: Figure 172 5°C where it fluctuated. At 5°C condensation was noted in the bag which account for the fluctuations. Moisture loss was higher at 37°C than at 25°C. This moisture loss reduced the Aw of the product but ensuing problems of browning and texture hardening as seen in Plate 1 occurred. Moisture losses through protein cross-linking are also common in products of intermediate moisture stored at tropical temperatures which would affect product texture (Obanu et al., 1975). Thus moisture content is a very important parameter in the stability of dried foods as suggested by Tannenbaum (1976). (b) Aw Aw changes (Fig. 4.15) reflect changes in moisture content. As the ambient RH was only 50%, moisture pick up was not a problem. Aw change from 0.76 to 0.72 in the product at 5°C, from 0.76 to 0.70 in products stored at 25°C and from 0.76 to 0.59 in products stored at 37°C over the 24 week storage time. Thus products kept at 5° and 25°C are more susceptible to mould growth since the minimum Aw at which mould growth is limited is 0.70 (FAO, 1981). However, storage at 5°C will repress mould growth which was found on product stored at 25°C after 16 weeks (Plate 1). The major problems in product stored at 37°C are browning and lipid oxidation as the Aw of any product approaches the intermediate moisture range browning rate increases (Labuza et al., 1970). 173 o---o 5 °C 0.8r o---o25°C o--o 3 7 °C Lsd at 1%*1.986 174 my - - CN 00 Figure 4,15: Changes in in morwong during storage at 5°, 25° and 37° u (c) Rancidity Rancidity was measured as thiobarbituric acid number (TBA no. ) . Although the TBA number changed significantly over the storage period at the three temperatures (Table 4.14) values remained small. Morwong is not classified as a fatty fish and thus less oxidation will occur. The TBA number monitors levels of hydroperoxides which form and at the same time are used up in further reactions such as with proteins (Obanu, 1987) to form melanoidins (brown pigments) or with amino acids to form acyl esters. TBA numbers increased with storage time (Table 4.14) and were highest in the samples stored at 25°C. Obanu (1987) observed higher TBA numbers in dried fatty fish stored at 38°C. On the other hand Nambudiry (1980) observed that lower salt concentrations accelerate TBA number increase whilst higher salt concentrations retard it. Rancid odour was only slight in the products during the storage period. 175 * Table 4.14: TBA no. in stored fish TBA no. (mg malonaldehyde/lOOg) Storage time (week) Temperature of storage °C 5°C 25°C 37°C 0 0.09a 0.09a 0.09a 4 0.16b 0.18b 0.09a 8 0.18b - 0.09a 12 0.16b 0.35d 0.14C 24 0.29e 0.34d 0.15C ★ Different superscripts indicate significant difference (p < 0.01) (d) Total volatile bases (TVB) in stored products Total volatile bases increased substantially on storage of the products (Fig. 4.16). This observation concurs with other reports (Connell, 1980; Hebard et al., 1980). Velankar (1952) observed increases in total volatile bases as ranging from 90-214 mgN/lOOg in a variety of cured fish stored for 2 months. Valsan et al. (1961) also reported increases of 151.2mg% TVBN after 3 months storage of salted dried mackerel. These increases are probably due to microbial action as most products contained over 30% moisture. However in this study, the bases were higher in products kept at 37°C than at 5° or 25°C where the moisture content is below 176 mg Figure 4.16: -o5 °C Changes at 5 5 , 25° in TVB and content 37 177 u c. during / storage 30% and Aw <0.70. Thermal decomposition of TMAO may explain the increase in TVB. Spinelli and Koury (1981) and Hebard et al. (1982) reported thermal breakdown of TMAO although at higher temperatures. Faturoti (1984) also observed increases in TVB during storage of sun-dried fish with moisture contents of less than 20% as did Hamed and Alley (1974) in dried shrimps. Haaland and Njaa (1983) also reported increases of 13-79mg N/lOOg of TVB in freeze-dried capelin stored at 20°C and an increase of 13-30mgN/100g at 2°C. Development of TVB in these products is probably not due to microbial spoilage as moisture content and Aw are too low for microbial growth. Tarr (1945) also noted that total volatile bases and trimethylamine increased in dried fish stored at RH too low for growth of bacteria or mould. Spinelli and Koury (1979) reported maximum formation of DMA in drum-dried and freeze- dried fish at RH 44%. These changes are more likely due to chemical and enzymatic reactions. Vaisey (1956) reported that TMA.O could be reduced in the presence of cysteine and either ferrous ion (Fe 2 + ) or haemoglobin as a catalyst. Several reduced ionic compound such as Fe 2+ , Sn2 + and S02 can also induce, in vitro, the degradation of TMAO to DMA (Spinelli and Koury, 1979). Breakdown products of cystein were found to degrade TMAO in vitro (Spinelli and Koury, 1981) and cystein breakdown products might be formed enzymatically during the heating cycle. This host of 178 possibilities may explain the observed increase in TVB during storage of dried products. (e) Protein solubility Protein solubility decreased during storage and was very poor in KC1 (Fig. 4.17). However salt soluble proteins are most sensitive to heat (Suzuki, 1980). Protein solubility worsens in products stored at 37°C. However there is no significant difference between samples stored at 5° and 25°C except at the 4th week in KC1 (Fig. 4.17). Solubility is fair in SDS + B mercaptoethanol, however storage time and temperature also affected the proteins solubilised in SDS (Fig. 4.18). This observation is in line with Obanu et al. (1975c) who worked with intermediate meat products. (f) IEF patterns of stored morwong The IEF patterns of stored morwong revealed that the same two bands E' and F seen in Fig. 4.19 are also observed in the stored samples. In sample c in Fig. 4.19 (4 weeks at 37°C) damage occurred in the proteins and isoelectric points are uncertain. In samples f and i (both kept at 37°C for 12 and 24 weeks respectively) both bands E' and F have disappeared. This provides evidence for protein damage during storage period, although this is only for water soluble proteins. Devadasan and Nair (1971) also observed less protein bands in fish after 15 days in ice as did Wiggin and Krzynoweck (1983) in IEF patterns of cooked products. 179 o------O 5 °C o------o 25 °C o------o3 7 °C Lsd at 1 % * 1.463 u|d)Oid a|qn|0§ % 180 o ■I 4- o r-> X5 •i +-> CL O 5- aj c l/> sz 03 — CJ Z3 i_ c cn — Figure 4,17: Solubility of morwong CO (•". > <_> * O storage at 5U, 25° and o------o370c Lsd at 1* = 1.217 ujaiojcl 181 a|qn|05 5; rj- Figure 4.18: S o lu b ility of morwong proteins in SOS + 6 mercaptoethanol during storage at 5°, 25° and 37°c. yr&4#0 •**** p*l-AO g af M -*|j- «• >» <*ll? C. *is d| I • M I I I » I * t sM * *■ '%• ‘*lu f- X J J « *iu l Figure 4.19: IEF patterns of morwong fillets brined for 8h, dried at 50°C and stored at 5°, 25° and 37°C each for 4, 12, and 24 weeks (a, b, c is 4 weeks at 5°, 25° and 37°C; d, e, f is 12 weeks at 5°, 25° and 37° and g, h, i is 24 weeks at 5°, 25° and 37°C). 182 (f) Digestibility in stored products In Table 4.21 it is seen that storage at 5°C did not affect in vitro protein digestibility and slight decreases occurred at 25°C. At 37°C the digestibility decreased with storage time. The decrease in digestibility results from changes due to salting, heat treatment and prolonged storage. Maillard reactions on proteins also decrease protein digestibility (Hegarty, 1982). Table 4.21: Digestibility of stored morwong* % digestibile protein Storage temperature (°C) Storage duration (week) 0 8 16 24 5 99.18a 99.14a 99.09a 99.07a 25 99.18a 99.12a 98.37a 97.56a 37 99.18a 98.23a 94.61b 89.96C * Different superscripts indicate significance (p < 0.01) (h) Total amino acids in stored product Table 4.22 shows the total amino acid content in dried salted morwong during storage at different temperatures. There is variation in some amino acids although no trend is discernible. Certain amino acids are clearly lost. These include proline, leucine, isoleucine, tyrosine and 183 phenylalanine. Lysine follows a decreasing trend only in the samples stored at 37°C. This Table 4.22: Total amino acid contents in stored samples Storage temperature °C Amino 5° 25° 37° acids ______g/16gN Storage time (week) 4 12 24 4 12 24 4 12 24 Aspartic acid 9.16 9.97 10.99 9.17 9.81 10.56 9.34 10.31 10.56 Threonine 4.08 4.81 5.28 4.14 4.75 4.33 4.12 5.11 5.19 Serine 3.89 4.46 4.65 4.26 4.47 4.50 4.42 4.68 4.7 Glutamic acid 14.50 13.47 14.94 13.74 14.32 19.31 13.75 14.97 14.4 Proline *5.82 5.18 4.53 +7.34a 4.37» 4.15b *5.27a 4.92ab 4.76 Glycine *7.46a 6.99a 5.25b +7.40a 7.23a 5.30b 6.17 6.90 6.61 Alanine 6.67 8.42 8.16 +7.89a 7.18a 6.00b 7.58 7.67 8.26 Cystine 0.65 0.78 0.88 0.78 0.73 0.91 0.80 0.91 0.94 Valine 4.58 5.50 5.94 4.69 5.72 5.23 5.77 5.55 5.59 Methionine 3.55 3.69 3.64 3.31 3.81 3.32 *3.82a 3.52ab 3.23b Iso leucine *5.23a 5.02a 4.53b 4.52 5.13 4.33 *4.94a 4.45ab 4.25b Leucine *9.89a 7.72b 6.92c 7.91 7.98 6.73 *7.79a 6.80b 6.4713 Tyrosine *3.63a 3.13ab 2.73b 3.13 3.25 2.92 *3.37a 2.80b 2.64b Phenylalanine *3.00a 2.31a 1.44b +2.34a 2.24ab 2.19b *2.78a 1.78b 1.90b Lysine 9.71 9.05 10.08 8.85 9.12 9.44 *9.89a 9.49ab 9.31b Histidine 1.76 2.34 2.56 1.83 2.41 2.49 2.29 2.48 2.51 Arginine 7.68 7.18 7.48 +8.70a 7.39 7.44b 7.50 7.66 7.89 ★ Different superscripts along horizontal column denote significantly difference at 99% confidence level. is due to Maillard reactions as these samples developed browning by the fourth week. Other amino acids such as methionine, cystine, tryptophan may also be lost during the 184 formation of the browning pigments (Hegarty, 1982) and browning is enhanced at higher temperatures (Dworshak, 1980). Most amino acid losses are probably due to browning although lipid oxidation-protein reactions (Hegarty, 1982) may also be involved. Decreases in some amino acids are also seen during storage at 25° and 5°C. Thermally reduced reactions to proteins leading to the destruction of amino acids, formation of amino acid complexes inter- and intramolecular cross-linking of proteins also may contribute to amino acid losses. Such reactions are a function of time, temperature, moisture content, reducing substances and pH (Hegarty, 1982). 4.1.2 Shark Shark has been an under utilised species as a salted and dried product. Notogaleus rhinophanes, school shark, was used in this experiment. 4.1.2.1 Chemical composition of shark The chemical composition of the shark is shown in Table 5.1. These values are comparable to published data (Watabe et al. , 1983). Crude protein of shark was 20.2% although shark is reported to contain as much as 3330 mg% non-protein nitrogen (Gordievskaya, 1973). This component is comprised of volatile bases, trimethylamine (TMA), trimethylamine oxide (TMAO), urea, creatinin and free amino acids. Urea, however, accounts for most of the non-protein nitrogen varying between 1570-2330 mg%. The pH 185 of the shark meat was 7.2 indicating that there could be a build up of ammonia originating from urea breakdown. Waller (1980a) indicated that as the ammonia level rises during spoilage, there is a concomitant increase in flesh pH, particularly on the surface. The pH of freshly caught shark is approximately 6 (Waller, 1980a). The TVB content was 73.7 mgN/lOOg exceeding the limit of 30 mgN/lOOg suggested by James and Olley (1971) although the fillets were judged acceptable on the basis of odour and appearance, TMA (368 mgN/lOOg) was also high compared to levels for cod and haddock of 0-1 mgN/lOOg for grade I fish, 1-5 mgN/lOOg for grade II and over 5 mgN/lOOg for grade III. A limit of 60 mgN/lOOg for ammonia has also been proposed by Vyncke (1968) for dogfish. However inconsistensies regarding such limits with respect to acceptability of fish have been reported (Hebard et al. 1982; Le Blanc and Gill, 1984). The amino acid profile (Table 4.1) of shark was similar to that reported by Mashelkar and Sohonie (1958). Simidu (1977) reported values of 63.6g/16gN for histidine; 0.2g/16gN for lysine; 29.9g/16gN for cy3tine; 12.1g/16gN for threonine; 14.0g/16gN for methionine and 2.7g/16gN for glutamic acid compared with 2.39, 10.83. 0.90, 4.88, 3.56 and 15.08g/16gN respectively in this study. However, different species were used. Tryptophan was not determined due to loss during hydrolysis of the samples. In Table 4.1 the E/NE ratio for shark was 0.85. This questions the belief that shark meat is lower in protein value and contains less of some essential amino acids than teleosts (Ronsivalli, 1978). 186 The in vitro protein digestibilities of various fish species are summarised in Table 4.2 (p 128 )• The digestibility for shark was 99.3% much higher than a reported value of 72.8% (Mashelkar and Sohonie, 1958). 4.1.2.2 Salting of shark Figs. 4.1 and 4.2 (p 129 and 130 ) show the salt uptake and moisture loss by shark fillets during brining. Shark attained the highest salt content amongst the three species salted with salt uptake very rapid for the first 4h (13.5%) and then gradually slowing down, reaching 19.5% (wb) after 48h. This rapid uptake is attributable to its coarser fibres (Ronsivalli, 1978). Shark meat was also reported to 'absorb' salt very quickly by Gordievskaya (1973). This rapid uptake could also be due to the large surface area (Crean 1961) as shark fillet was skinless. Thickness (Waterman 1976), ratio of brine to fish, length of brining time, physical conditions of the fish (Weckel and Wosje, 1966) and brine temperature (Zaitsev et al., 1969) also affect ratio of salt uptake. Between 2-20°C salt uptake is more dependent on curing time than on temperature (Rodger et al., 1984). On the other hand Del Valle and Nickerson (1967) observed that salting at 5°C gave a higher distribution coefficient than at 25°C and salting at 37°C resulted in lower salt content than either at 5°C or 25°C. Moisture loss (Fig. 4.2) is fairly rapid in the first 12h with moisture content decreasing from 79.4% to 64.6% at 12h and further decreasing to 61% at 48h salting. Salt content was 187 inversely related with moisture content. Even though shark reached higher salt content (Fig. 4.1, p 129) than sardine and morwong, its moisture loss (Fig. 4.3, p 132 ) was intermediate between morwong and sardine. Moisture loss was rapid initially with 15.2% loss after 12h salting but slowed subsequently reaching 19.2% after 48h. Thus the ability to lose moisture seems to be specie specific and dependent on the condition of the fish. Since the salt content of shark was almost maximum after 4h salting, there was no benefit in prolonging it and 4h was chosen as the time of salting for drying studies at different temperatures. Chemical composition of i3hark Moisture (%) 77.4 + 0.47 Protein (%, wb) 20.2 ± 0.52 Fat (%, wb) 2.06 ± 0.44 Ash (%, wb) 0.64 + 0.11 pH 7.2 TVBmgN/lOOg 73.7 TMAmgN/lOOg 36.8 4.1.2.3 Drying studies Fig.^.,5 compares the rate of drying at 50°C of morwong, sardine, squid and shark. Shark had second fastest rate, drying significantly faster (p < 0.01) than sardine and morwong. The large surface area and the fibrous nature of shark meat encouraged fast drying. Drying rates for shark at other temperatures are seen in Fig. 5.1. The drying rates at 60° and 188 © o------ ( q.uipB/o2HB ) Buj/Uq 189 70°C were similar, differing significantly (p < 0.01) only at the 2nd, 4th and 6th hour of drying at 60°C. The rates at 30° and 40°C were similar, differing (p < 0.01) only in the first few hours of drying. The products were dried to moisture contents and Aw shown in Table 5.2 4.1.2.4 Product quality The moisture contents of the dried products varied between 26.8% (dried at 60°C) and 44.5% (dried at 30°C). Aw ranged from 0.65 - 0.72. The product was creamy white which was most attractive in products dried at 30° and 40°C. However at 30°C the moisture content was high (44.5%). Drying at 50°C gave slightly yellowish products which were still quite attractive. Products dried at 60°C had a slight brownish tinge while at 70°C, the products became brown. Bose et al. (1958) observed similar colour changes. The products dried at 50°C were chosen for storage studies due to rapid drying rate and attractive product colour. Table 5.2 shows the predicted shelf-life for the shark products to range from 1 Table 5. 2: Aw and moisture contents of dried shark Drying Aw Moisture Salt Predicted* measured content % content % (wb) shelf life 30P° 0.72 44.5 24.6 1 week 0 o 0.72 29.8 31.1 2 months 50° 0.71 34.1 29.2 l h months 60° 0.65 26.8 32.4 2 months 0 o r- 0.70 37.7 27.6 1 h months h from shelf life table of Poulter (1980) and Curran (1984). 190 week for the 30°C sample to longer than 2 months for the 60°C sample. On storage, the 50°C sample lasted over 6 months (Section 4.1.2.7). Commercial products of moisture content 30-35% and salt content of 16-17% were reported by Jinadatharaya and Vernekar (1979) to last less than 6 months. 4.1.2.4 Sensory evaluation of the dried product Shark was salted for various times, dried at 50°C and subjected to sensory evaluation. The product was rehydrated for 10 min. drained and fried in oil at 200°C for 7 min. and was served to a panel familiar with salted fish products. (a) Saltiness vs. brining time Table 4.5 shows the mean scores for saltiness in the products. For shark the saltiness for the 4, 2 and 8h products did not differ significantly, although the mean scores were highest for the 4h product. From these results, 4h is confirmed as the optimum time for brining the product. (b) Aceptibility vs. drying temperature From Table 5.3 it is seen that samples dried at 70° and 60°C are significantly different in aroma to those dried at 50°, 40° and 30°C which are not significantly different to each other. For colour, samples dried at 30° and 40°C had the highest scores and more significantly preferred (p < 0.01) over other samples. The 40°C sample scored the highest for texture, was not significantly different to the 50°, 30° and 60°C samples, but differed significantly from the 70°C 191 sample. For overall acceptibility, the 30°, 50° and 40°C samples scored the highest and did not differ significantly. They are however significantly preferred (p < 0.01) over the 60° and 70°C samples which were significantly different (p < 0.01) from each other. Based on acceptibility only shark could be dried either at 30°, 40° or 50°C. The last was chosen because of its drying rates. Table 5.3: Mean scores for dried salted shark Sensory attributes Drying Temperature °C 50° 40° 30° 70° 60° Aroma 7.66a 7.60a 7.21a 6.17b 6.14b 30° 40° 50° 60° 70° Colour 7.49a 7.80ab 6.65b 5.55C 5.2 ld 40° 50° 30° 60° 70° Texture 6.56a 6.45a 6.41a 6.09a 5.09b 30° 50° 40° 60° 70° Overall acceptibility 7.32a 7.22a 6.90a 5.84b 5.70a 192 4.1.2.6 Chemical changes of salted dried shark 4.1.2.6.1 Effects of salting (a) pH Fig. 4.6 (p140) shows changes in shark pH during brining in saturated salt. Initial pH of 7.2 decreased rapidly to 6.9 after 4h, further decreased to 5.7 after 8h and remained relatively constant for the remainder of the salting period. The initial decrease is attributed to extraction of water soluble bases as moisture was withdrawn by the salt. During salting 71% of the urea was leached into the curing brine (Yang et al., 1981). 40% of the urea was removed from white tip shark with 12h pickling and 24h soaking and 89.9% after pickling to 3 days and soaking for up to 8h (Gordievskaya, 1973). This may also contribute to the drop in pH. (b) TVB and TMA Figs. 4.7 and 4.8 show that TVB increased from 73mgN/100g to 26mgN/100g after 24h salting and remained constant up to 48h salting. TMA also decreased, from an initial value of 37mgN/100g to 8mg/N/100g at 18h salting. These decreases are attributed to removal of amines into the brine (Gordievskaya, 1973). The initial pH of 7.2 reflects the presence of TVB and TMA in the shark prior to salting and decreases in reponse to removal of these from the shark flesh during brining. Urea, which breaks down by urease to ammonia is also removed into the brine on salting (Zaitsev 193 et al., 1969). The TVB and TMA in shark did not rise during the course of salting possibly because of the rapid salt uptake by shark in the first few hours thereby retarding microbial degradation and enzymatic breakdown of urea. This observation concurs with Yang et al. (1981) who did not observe increased amines during salting of gray-fish. (c) Protein solubility Although the maximum level of extractable protein in NaCl or KCl in good quality shark has been reported to be 50% (Waller, 1980b) the solubility observed in this study is much lower. Protein solubility in KCl at 25°C was initially 25% but had decreased to about 5% by the end of salting (Fig. 5.2). In KCl at 77% the initial solubility was slightly better, i.e. 29% but gradually decreased to about 7.5% by the end of salting. Waller (1980b) using the same medium also obtained low extractable protein, between 22- 35%. Extractability could be influenced by the starting quality of fish, pH of the muscle and method of extraction (Sikorski et al., 1976). Solubility in SDS is much higher than in KCl. 67% of shark protein initially dissolved in SDS but solubility decreases gradually to 42.6% during salting up to 48h. Decreased protein solubility during salting is a consistent finding in this study. 194 2 £ cn (0 bm +■>o c/> salting. on shark of o o Solubility 5.2: s-a) ai •i— u;a)Oid a|qn|o$ % 195 (d) Isoelectric focussing The water soluble proteins of shark are anodic (Fig. 5.3). The bands streaked and were not clearly defined. However about 6 bands, A, B, C, D, E and F are apparent in sample a. In the rest of the samples other bands such as A', B' and C' appeared in more cathodic positions than A, B and C respectively. These new bands are visible in all salted shark samples however their intensity decreases at higher salting times. Bands E and F were the most stable, remaining visible at longer salting time. In samples d, e and h band E was more clearly defined than band F although the reverse applied in samples c, f, g and i. Disappearance of the bands in samples d, e, f, g, h and i are apparent. This provides evidence of denaturation of water soluble proteins during salting in line with the observed decreased protein solubility. (e) In vitro digestibility Digestibility of shark remained high throughout the salting time (Table 5.4). Very small changes are seen in the digestibility which are of no nutritional significance. Table 5.4 In vitro digestibility of shark on salting Salting time (h) 0 4 8 12 18 24 36 48 % Digestible 99.4a 99.6a 97.7a 97.4a 99.4a 99.3a 99.la 98.2a protein ±0.3 ±0.3 ±0.2 ±0.4 ±0.3 ±0.6 ±0.2 ±0.2 196 4 /(f/il «»»«* U»uu ) ■)! ok 0. it; I IK t> I * c » Utt 9 * 11 *H c| » m |) * M> M *• f* r f r * f ** / • V aSb'ccP£ f Figure 5.3: IEF patterns of water soluble proteins of shark fillets after brining for 0,2, 4, 8, 12, 18, 24, 36 and 48h (a, b, c, d, e, f, g, h and i respectively). 197 (f) Amino acids in salted shark The amino acid composition in shark during salting (Table 5.5) showed variations in the amounts of individual amino acids. Trends are difficult to establish but cystine and histidine decreased over the salting period. The decrease in Table 5.5 Total amino acid in shark Amino Salting time (h) acids g/16g N 0 4 12 24 48 Aspartic acid 9.66 9.61 9.63 9.75 9.58 Threonine 4.88 4.96 4.86 5.02 4.87 Serine 4.40 3.97 4.39 4.34 4.45 Glutamic acid 15.08 12.26 14.69 15.45 15.28 Proline 3.87 5.24 4.47 4.49 4.69 Glycine 4.46 3.94 5.09 4.66 4.56 Alanine 7.35 9.62 8.03 7.46 7.41 . * Cystine 0.90a 0.55b 0.62c 0.7 ld 0.71d Valine 5.49 5.30 5.14 5.52 5.23 Methionine 3.56 3.61 3.42 3.52 3.57 Isoleucine 5.36 5.50 4.95 5.07 4.86 Leucine 8.17 8.93 7.62 7.77 7.64 Tyrosine 3.58 4.05 3.55 3.57 3.43 Phenylalanine 2.69 3.42 2.94 2.72 2.44 Lysine 10.83 9.45 10.71 10.15 11.16 Histidine* 2.39a 2.34a 2.35a 2.33a 2.36a Arginine 7.34 6.24 7.53 7.47 7.50 Amino acid showing decreasing trend, significant at 95% confidence level. 198 histidine is not significant. However in cystine the decrease is significant at p < 0.05. Although solubility is reduced by salting, the digestibility and total amino acid content are not similarly affected. Salting up to 48h therefore does not incur significant nutritional damage to the proteins of salted fish. 4.1.2.6.2 Changes to salted shark during drying (a) Total volatile bases (TVB) and trimethylamine (TMA) TVB and TMA increased significantly (p < 0.01) in shark during drying, more so at higher temperatures (Fig. 5.4). From an initial value of 52mg% for TVB and 30mgN for TMA in shark salted for 4h, the TVB increased to 56.8mg% at 30°C, 74.2mg% at 50°C and 81.6mg% at 70°C. These results concur with previous observations (Babbitt 1977; Spinelli and Koury 1978). The rate of TMA and DMA formation was proportional to raw material freshness with fresher materials leading to less amines in dried products (Kids and Tamoto, 1974) Spinelli and Koury (1979) related amine formation to the water activity of the drum dried and freeze dried hake while Nakamura et al. (1985a) observed thermal decomposition of urea and TMAO when hake meat was heated at 90° and 115°C. They also observed that when fillet was dried at 30°C for 12h and press piling was applied for 20h during further drying, the increase in NH^ and TMA was negligible (Nakamura et al. , 1985b). Noda et al. (1978) observed high TVB (74- 199 mg N /lO O g Figure o o ------ 5.4: oTVB oTMA Effect and Drying TMA of temperature contents drying 200 temperature in (^C shark. ) on TVB 243mgN/100g) and TMA (1.1-111.8mgN/100g) in a variety of dried and seasoned shark meats prepared and on the market. (b) Protein solubility Solubility of dried shark protein in both KC1 at 77° and SDS + B mercaptoethanol (Fig. 5.5) is lower than that of the salted only products (Fig. 5.2). Solubility in KCl at 77°C decreased gradually with increasing drying temperature from 23% in the 4h salted shark to 13% in the sample, dried at 70°C. In SDS + B mercaptoethanol the 4h salted sample solubility was 56% which reduced to 45% in the 70°C dried sample. Poulter et al. (1985) also reported a loss in protein nitrogen extractable in 1M KCl when fish were heated for lh at temperatures from 25°-100°C. This large change in protein extractability was attributed to changes in the myofibrillar fraction. Sarcoplasmic protein exhibited similar solubility loss although somewhat delayed compared to the myofibrillar fraction. The effect of salting and drying is seen in the IEF patterns in Section 4.1.1.3.6.2c. Madovi (1980) also observed that fresh muscle tissues are more soluble in SDS than cooked or processed meat. Solubility reduces even further on storage (Obanu et al., 1976). Freeze drying also reduces solubility of hake from 70-80% to about 15% (Spinelli et al., 1972). Solubility therefore is affected by processing and decreases with increasing temperature. 201 o o 2 6 0 the ) on (®C ° . 50 proteins. temperature temperature shark ° of drying 4 0 Drying of ^ cr> f" a Effect solubility ° u 3 0 / 202 (c) Isoelectric focussing In fresh shark (sample a, Fig. 5.6) there are seen six bands A, B, C, D, E and F, two of which (E and F) are very distinct. In sample c,d and e which were dried at 30°, 40° and 50°C respectively, only bands A, C, E and F are present. In samples f and g, dried at 60° and 70°C respectively, band F has become slightly cathodic, and only bands A, C and E are seen in f and only C and E in g. These changes are manifestations of salt and heat damage to the sarcoplasmic proteins, the water soluble fraction. Sarcoplasmic proteins coagulate between 40° and 60°C although in some preparations heat denaturation of sarcoplasmic proteins is not completed below about 90°C (Hegarty, 1982). Thus some bands persist in the isoelectric pattern of sample g, dried at 70°C. (d) In vitro digestibility Table 5.6 shows the in vitro digestibility of dried shark at different temperatures. There was no significant effect of drying temperature on the in vitro digestibility, despite the changes to protein noted above. This observation was also made by Faturoti (1984) who reported that processing had no adverse effect on the digestibility of dried fish. However Aitken and Connell (1979) described conflicting reports in the literature regarding the effects of heat treatment on digestibility. 203 c SH A & vc pH 3-10 <» f cc :8til 8K (, ''V ! SolH4 1 III M* ) STp 1 III 1II 1 rnXXm 50c e to*c * IS 3ft •i j«# H|l 7ot 3 i* 9* 11111 I A 6 C HE f wl>*|n Figure 5.6: IEF pattern of water soluble proteins of shark fillets brined for 4 h and dried at 30°, 40°, 50°, 60° and 70°C (c, d, e, f and g; a = fresh fish, b = 8h salted fillet but not dried). 204 Table 5.6: In vitro digestibility of dried shark Drying Temperature °C 0 0 0 0 ID o o o Fresh 30° o in % Digestible 99.3a 99.9a 99.7a 99.5a 99.6a 99.5a protein * Same superscripts denote non-significance (e) Total amino acids Drying did not adversely affect the total amino acids in shark (Table 5.8). This concurs with most previous reports (Tarr, 1962; Anon, 1973, Aitken and Connell, 1980). Yang et al. (1970) found no effects of drying at 105°C but at 170°C available lysine fell by 20%. Specific trends are not discernible in this study, however decreases are seen in cystine and lysine, both of which are known to be heat sensitive. 205 Table 5.8. Amino acid content in dried shark Amino Drying temperature °C acids g/16g N Fresh 30° 50° 70° Aspartic acid 9.66 9.61 9.63 9.75 Threonine 4.88 4.96 4.86 5.02 Serine 4.40 3.97 4.39 4.34 Glutamic acid 15.08 12.26 14.69 15.45 Proline 3.87 5.24 4.47 4.49 Glycine 4.46 3.94 5.09 4.66 Alanine 7.35 9.62 8.03 7.46 . * Cystine 0.90a 0.55b 0.62bc 0.71aC Valine 5.49 5.30 5.14 5.52 Methionine 3.56 3.61 3.42 3.52 Isoleucine 5.36 5.50 4.95 5.07 Leucine 8.17 8.93 7.62 7.77 Tyrosine 3.58 4.05 3.55 3.57 Phenylalanine 2.69 3.42 2.94 2.72 Lysine* 10.83a 9.45c 10.71ab 10.15b Histidine 2.39 2.34 2.35 2.33 Arginine 7.34 6.24 7.53 7.47 *Amino acid showing decreasing trend during drying, significant at p < 0.01. 4.1.2.7. Storage Studies on Shark Shark, salted for 4h in saturated brined and dried at 50°C a moisture content of about 30% was stored at 5°, 25° and 37°C at ambient RH of about 50% for 16 weeks. The products were packed in low density polyethylene bags and were intermittently examined for mould growth, product appearance and chemical composition. 206 4.1.2.7. Evaluation of the product (a) A description of product appearance No photographs are available to show the appearance of the stored products which were creamy yellow at the start of storage. During storage at 5°C the product kept very well and did not change colour appreciably. The products at 25°C lost moisture and consequently developed a harder texture. Their colour changed very slightly developing a tinge of light brown at the edges of the product. These changes became quite obvious from the 12th week of storage. The products stored at 37°C were harder in texture than those at 25°C and had developed brown colour by the 8th week. Dried, salted shark did not brown as extensively as other fish products during storage. It did not shrivel as did morwong; it kept its shape and hardened and did not develop mould growth during storage. (b) Colour The L values (Table 5.9) did not change significantly during storage at all temperatures except at 12 and 16 weeks at 37°C confirming visual observations that the product did not darken on storage. The a values increased significantly during storage at all the temperatures, indicating development of redness in the product. The b values however 207 Ua>ei Table 5.9: Hunter L, a, bA for shark stored at 5°, 25 ° and 37°C Temperature Storage time (wk) >°C 0 4 8 12 16 5 L 75.40a 73.00a 72.63a 72.96a 69.17a a -3.00a -2.36b -1.50C 0.30d 2.23e b 19.78a 21.872 21.90a 22.67a 22.77a 25 L 75.40a 72.33a 69.23a 69.17a 68.13a a -3.00a 1.57b 2.05C 2.27d 2.87e b 19.78a 22.2a 21.90a 24.80b 24.83b 37 L 7 5.40a 71.23a 69.10a 67.50b 65.57b a -3.00a -0.7b 1.30C 2.50d 2.73d b 19.78a 24.65b 24.67b 25.43b 25.5b L : Lsd at 1% = 7.1100 a : Lsd at 1% = 0.2532 b : Ld at 1% = 3.169 did not change significantly during storage at 5°C. However, in the 12th and 16th week samples at 25°C and the fourth week at 37°C there were significant changes. These readings are consistent with the observation that dried shark did not develop browning as extensively as morwong or sardine during storage. This is perhaps an inherent characteristic of dried shark meat. No reference to carbohydrates in shark was found in the literature. This may indicate low reducing sugar levels and hence restricted 208 Maillard browning in stored shark although browning can also be due to lipid-protein reactions. Rancidity development has also been noticed in smoked dogfish (Shiau and Chai, 1985) although shark tissue exhibited lower TBA values compared to mackerel on storage (Younathan et al., 1983). On the whole, colour development was not a limiting factor to the quality of stored dried shark in this study. (c) Flesh structure Plate 8 demonstrates the effects of brining on the structure of shark flesh. As brining time increases the surface is disrupted increasingly and more salt is evident. Disruption of the fibres may be due to salting out which reduces the water holding capacity of the fibres thus causing shrinkage. The micrographs of the fillets dried at 30°, 40°, 50° and 60°C are shown in Plate 9. The compactness of the tissue structure decreases as drying temperature is increased. At lower temperatures the fibres appear to gel whilst at 60°C the fibres seemed to 'separate'. The micrographs, as with the other fish, portray the effects of salting and drying on the fibres which are also reflected by the solubility, IEF and rehydration studies. (e) Reconstitution Reconstitution properties of dried shark are given in Tables 5.10, 5.11, and 5.12. Generally, poor reconstitution is 209 Plate 8: SEM micrographs (150x) of shark tissues brined for 0, 4, 12 and 24h (a, b, c and d respectively). 210 a b Plate 9: SEM micrographs of shark fillets brined for 4h and dried at 30°, 40°, 50° and 60°C (a, b, c and d respectively). 211 observed. The sample salted for the least time showed highest rehydration ratio (Table 5.10) as was observed by Bligh and Duclos (1981). Increasing drying temperature reduces the rehydration ratio (Table 5.11) as was also observed by Rustad and Nesse (1983). Longer exposure to heat also reduces the rehydration ratio (Sudhakharan and Sudhakaran, 1985). On storage (Table 5.12), the rehydration ratio decreases and to a larger extent for products stored at higher temperatures. Poor rehydration ratios reflect irreversible changes occurring in the fibres which may be ascribed to cross-linking processes between denatured proteins (Connell, 1957). Table 5.10: Rehydration ratio of shark salted for different times and dried at 50°C* Brining time (h) 4 12 24 36 48 Rehydration ratio 1.28a 1.24a 1.09b 1.04b 1.03b Different superscipts indicate significance (p < 0.01) Table 5.11: Rehydration ratio for shark salted for 4h and dried at different temperature Drying temperature °C 30 50 70 Rehydration ratio 1.32a 1.28a 1.07b *Different superscripts indicate significance (p < 0.01) 212 Table 5.12: Rehydration ratio for stored shark products* Storage time (week) Temperature of Storage °C 0 4 8 12 16 5° 1.28a 1.26a 1.23a 1.19b 1.14b 25° 1.28a 1.24a 1.17b i.nbc 1.06C 37° 1.28a 1.12b 1.09b 0.98C 0.93C Different superscripts indicate significance (P < 0.01) 4.1.2.7.2. Chemical properties of stored dried shark (a) Moisture content and water activity (Aw) Moisture content of dried shark decreased on storage, more so at higher temperatures (Fig. 5.7) due to the prevailing storage RH. This occurs in stored products if packaging does not prevent moisture loss. However it must be mentioned that moisture vapour build up could occur if the packaging is not permeable and the moist environment would allow microbial growth. In this study moisture is lost from the products although packing was used. Aw (Fig. 5.8) followed the same trends as moisture control. Although Aw was sufficiently low to limit microbial growth, problems with non enzymatic browning could arise. Troller and Christian (1978) have shown that Maillard type reaction in cod and anchovies show a maximum rate at relatively low Aw levels with significant browning at A^ > 0.75%. Extremely 213 5 at stored shark dried in changes Moisture o o M 5.7: Figure }ua;uo3 ajmjsfow % 214 shark dried in changes 5.8: Figure at stored c rt3 ro x> <_> LD LD CM o o O i e tim torage S ) wk ( 00 - ii S ■o CM 00 © ^ © ? <1 CO ? 4 CM UU o in I '00*8 low moisture contents would be required (5-10%) to prevent Maillard reaction in dried fish, lower than normally achieved in dried salted products. Kunimoto et. al (1985) also observed rapid incidence of browning in freeze dried fish in higher RH environments than in lower RH environments. (b) Rancidity On storage TBA no. increased in the 5°C stored samples but decreased in these at 25° and 37°C (Table 5.13) although TBA no. remained small on storage. Shark contained only about 2% lipid and the product did not smell rancid throughout the storage period. Table 5.13: Thiobarbituric acid number* Storage temperature °C Storage period (wk) 5 o 25° 37° 0 0.25a 0.25a 0.25a 4 0.22a 0.25a 0.24a 12 0.33b 0.19b 0.22a 16 0.39b 0.18b 0.21a ♦Different superscripts in vertical columns indicate significance p < 0.01. (c) Total volatile bases (TVB) Significant increases (p > 0.01) in TVB are seen in Fig. 5.9 during storage of dried shark. The increase is greater at higher storage temperatures. These increase gave final TVB levels of 105mgN/100g at 5°C, 118mgN/100g at 25°C and 216 k —o 2 5 °C —o37°C 217 Storage time ( wk ) Figure 5.9: TVB changes in shark stored at 5 135mgN/100g at 37°C. These changes are more likely due to chemical and enzymatic reactions rather than microbial effect (Section 4.1.1.7d) since these products are high in salt and low in Aw . (d) Solubility Solubility was low for products stored at the three temperatures in KC1 at 77°C (Fig. 5.10). The initial solubility of 18.5% decreased gradually at 5° and 25° but more rapidly at 37°C to 16, 14 and 10.1% respectively after 16 weeks. Solubility in SDS + 8 mercaptoethanol also decreased (Fig. 5.11) from an initial value of 49.5%, to 47.5, 45.5 and 43.5% at 5°, 25° and 37°C after 16 weeks. This again shows greater effect of higher storage temperature and longer storage time on the properties of dried products. (e) Isoelectric focussing (IEF) In the IEF patterns of stored shark (Fig. 5.12) only three bands are clearly defined. These are B, E and F. Other bands appear as diffused streaks with minor bands appearing between the major areas. These are possibly the denatured proteins and peptides due to protein breakdown during storage. It appears that band F has disappeared in samples stored at 37°C. In samples e, f, g and h decreasing intensity and integrity of the bands is apparent. 218 c:cn •r— ■o o o CCO C_) •r-Oo> -*-> ro os- -O CL fOC s-o fO LD .JCco CVJ er\ M-O O LD >1 +J 4-> n3 J=> 13 O oo storage o LD £ CD •r— Ll_ ujajoid 3[qnj05% 219 —o25°C ujajoid »|qn|os % 220 -C 4 +-> 03 c o 03 C during storage at d iffe re ntemperature t STo«ei> sHA*X pH 3-10 SJLCmI a*/lo/g7 Figure 5.12: IEF patterns of water soluble proteins of dried shark stored at 5°, 25° and 37°C for 4wk (a, b, and c respectively) . 12wk at 5° and 37°C (d and e) and 24wk at 5°, 25° and 37°C (f, g and h respectively). 221 (f) In vitro digestibility In vitro digestibility of stored dried shark (Table 5.14) remained high on storage. Digestibility was not affected either by storage time or temperature. This observation concurs with with Kunimoto et. al (1985) who observed no adverse effects on digestibility during storage of freeze- dried carp interacted with fish oil which had already undergone browning. Faturoti (1984) made similar observations during storage of dried salted fish. The digestibility of the products are above the minimum limits (92% pepsin digestibility and 6.5% available lysine) proposed for dried fish products (Heen and Kruezer, 1962). Table 5.14: In vitro digestibility of stored shark % Digestibility Storage Temperature °C Storage duration (wk) 0 4 8 12 16 5 99.31a 99.80a 99.84bc 99.26a 99.74a 25 99.31a 99.63a 99.60ac 99.68a 99.62a 37 99.31a 99.61a 99.80a 99.79a 99.84b * Same superscripts indicate non-significance (g) Total amino acid Table 5.15 shows total amino acids in dried shark stored at 5°, 25° and 37°C for 16 weeks. Although variations are 222 seen, general trends are not apparent. Decreasing tendencies are however perceived in serine, glycine, alanine, tyrosine, lysine and arginine. There are significant changes (p < 0.05) over the same period of storage at different temperatures . These amino acids are presumably involved in non-enzymic browning and lipid-protein reactions. Table 5.15: Amino acid content in shark stored at 5°, 25° and 37 for 16 wk. Storage temperature °C Amino acid g/16gN Day 1 5° 25° 37° Aspartic acid 10.18 10.79 10.07 10.31 Threonineq 4.56 5.23 5.29 5.47 Serine 5.04a 4.40b 4.39b 4.61b Glutamic acid 12.30 15.79 14.73 14.82 Proline 4.15 5.01 4.61 4.41 Glycine 6.63a 5.01b 5.16b 5.13b Alanine 9.49a 7.48b 7.45b 7.58b Cystrne 0.79 0.77 0.92 0.80 Valine 5.25 5.81 5.71 5.65 Methionine 3.32 3.11 3.69 3.54 Isoleucine 4.85 5.04 5.36 4.95 Leucine 7.23 7.48 7.67 7.40 Tyrosine 3.26a 2.28b 2.72C 2.90c Lysine 9.37 9.51 9.59 9.59 Histidine 2.60 2.70 2.67 2.82 Arginine 8.07a 7.26b 7.51C 7.83a * Different superscripts along horizontal column denote significance at P < 0.05. 4.1.3. Sardine Sardine (Sardinop neopilchardus) measuring 120-150 mm were found to contain 75% moisture, 20% protein, 3.7% fat and 1.2% ash on wet weight basis. The protein content is higher than a reported 223 value of 18% (Iwasaki and Harada, 1985). However a range of 17.2 - 21.6% has been reported (Kizevetter, 1973). Fat content was much less than the 10% reported by Jimenez- Colmenero et al. (1988). Fat content in sardine varies greatly with values of 1-24% for Sardinella longiceps (Mulyanto, 1982); 9-20% for Sardinella pilchardus (King et al., 1984) and 11-21% for Sardinella melanosticta (Kizevetter, 1973) being reported. Fat content varies depending on spawning, maturity, migration and season. An inverse correlation between the moisture and fat contents also occurs (Sen and Chaluvaiah, 1968; Kizevetter, 1973) . The pH of the sample was 6.1 indicating freshness. However pH of sardine also varies with season. Jimenez-Colmenero et al. (1988) observed the highest pH in February when the fish is seen in a more starved condition. The TVB and TMA were 34.7mg/N/100g and 16.4mgN/100g respectively, both exceeding suggested upper limits for acceptable quality (Connell, 1980). Although the fish were in sound condition, Malle et al. (1986) remarked that studies on TVB and TMA have not established a definite correlation between the levels of these substances and the quality of fish sample. Generally fish with dark meat like sardine contain more TMAO than white flesh species (Suzuki, 1981). Brown musculature is also reported to contain more fat and nitrogenous extractives, particularly histidine although it is lower in lysine-N 224 (Kizevette, 1973). The amino acid content in sardine is given in Table 4.1 (p 127 )• Differences in reported and determined values are also discernible. Proline content was 9.27g/16gN as compared to 2.89g/16gN, alanine 8.57g/16gN compared with 5.76gN/16gN and lysine 8.11g/16gN compared with 10.62g/16gN. Latter values were reported by Iwasaki and Harada (1985). The data for the essential amino acids however concur with those reported by Geiger and Borgstrom (1962). The ratio of essential to non-essential amino acids (E/NE) was 0.62 lower than the reported value of 0.74 (Iwazaki and Harada, 1985). This stems from the lower values for methionine, isoleucine, leucine, phenylalanine and lysine (3.13, 4.66, 7.84, 2.88 and 8.11g/16gN respectively) whilst the reported values are 3.33, 4.89, 8.47, 4.41 and 10.62g/16gN respectively. However the ratio 0.62 is above the limit of 0.60 set for good quality protein (FAO, 1970). Fresh sardine was found to be very digestible, 99.5% (Table 4.2, p 128 ) • During maceration, autolysis could have occurred leading to protein breakdown and increased digestibility. This effect was observed by Meinke and Mattil (1973). 4.1.3.1. Salting of sardines In Fig. 4.1 (p 129 )' it can be seen that both rates of salt uptake and total salt uptake for sardine were smaller than far morwong and shark. The two observations could perhaps be explained by the following: (a) sardine was salted whole, therefore its skin would act as a barrier between the muscle and the brine thus slowing down the salt front (Lupin, 1981), (b) 225 sardine is a fatty fish. Observations were made by Burgess et al (1965) that fat content in herrings acts as a barrier to both the entry of salt and the withdrawal of moisture and that both salt uptake and moisture loss become progressively slower with increasing fat content, (c) sardine protein is characterised by its low pH after death, when very fresh, 20-30% of its myofibrillar protein is already denatured, losing some of its water (Suzuki, 1981). Thus being in this state, it may not be able to absorb as much salt. These factors therefore combine to result in slower rate and smaller uptake of salt. Sardine contained the lowest initial moisture content (74.5%) amongst all the three species salted (Fig. 4.2, p 130). This may be due to its fat content which has an inverse relationship with moisture content. During salting moisture content fell steadily to 62% after 12h and eventually to 57% after 48h. Fig. 4.2 (p 130 ) shows the loss of moisture during brining. Sardine experienced the higher loss of moisture than shark and morwong although its salt uptake (Fig. 4.1 p 129 ) was the least amongst the three species. This is perhaps a 3pecies specific characteristic, which is related to its water holding capacity which in turn is dependent as its post rigo jr\. conditioning. 8h salting was chosen for drying studies since the salt content in the fish neared maximum at this time (Fig. 4.1, p 129 )• 4.1.3.2. Drying of sardines It can be seen in Fig. 4.4 (p 133 ) that sardine dries more slowly than the other species. It is significantly different (p 226 < 0.01) from shark at all drying times although different from morwong only in the first 4h. The presence of skin and its fat content could contribute to the slow drying rates. Drying is significantly faster (P < 0.01) at 50° and 70°C compared to the other temperatures in the first lOh (Fig. 6.1). Drying at 70°C resulted in products of brown appearance. 4.1.3.3. Product quality Sardines dried at 30° and 40°C had an attractive appearance and did not suffer from severe shrinkage although some browning was apparent around the belly region. Products dried at 50°C were slightly more brown than those at 30° and 40°C. The browning was apparent along the belly and tail region on the ventral side. Browning was more severe in the 60° and 70°C dried samples, (under the skin the meat also appeared brown) shrinkage was apparent and products had hard texture. Table 6.1 shows the moisture content and A w of sardines dried at different temperatures. The moisture ranged from 22.2% for samples dried at 70°C to 47.9% for those dried at 30°C. The Aw in the 30° and 40°C are > 0.75. This means that the products required longer drying time to achieve microbial stability. Being a fatty fish the products are also liable to fat oxidation leading to rancidity and browning from lipid-protein reactions. Products dried at 50°, 60° and 70°C had Awg of 0.67-0.70 which were indicative of better microbial stability although still subject to rancidity problems. Predicted shelf lives of the products are shown in Table 6.1. The products dried at 50°, 60° and 70°C were 227 sardines. salted of rates drying the on temperature of Effect 6.1: Figure M-WaB/0CHB) 8ul*Ja 228 predicted to have shelf lives of between 2-4 months and those dried at 30° and 40°C, because of the high moisture and low salt content, are expected to last only for a week. Table 6.1.: A and moisture content in dried sardine Drying Aw Moisture Salt % Predicted# Temperature measured Content (wb) shelf life °C (%) 30 0.86 47.9 9.14 1 week 40 0.77 38.4 11.7 1 week 50 0.67 22.7 13.7 2 months + 60 0.70 29.2 12.3 2 months + 70 0.70 22.2 12.0 4 months #Predicted shelf life is read from Poulter (1980) and Curran (1984). 4.1.3.4. Sensory evaluation of dried sardines (a) Effect of brining time Sardines brined for various times and dried at 50°C were presented to a taste panel after reconstitution in water and frying. The results of the mean scores for saltiness over brining time are presented in Table 4.5 (pl41). Salting for 2 and 4h gave products which were significantly preferred (p < 0.01) over the rest of the products on the basis of saltiness. The 8h product was ranked third and was significantly less preferred to the 2 and 4h products. The 8 and 12h products were not significantly differentiated. 229 Eventhough the 2 and 4h products were preferred, the 8h products were still recommended since they had higher salt content for the same moisture content leading to lower Aw. Salt content may be reduced by rehydrating the products for longer time. (b) Effect of drying temperature Sardines brined for 8h were dried at 30°, 40°, 50°, 60° and 70°C to a moisture content of approximately 30%. Products were prepared as above and presented to the panel for evaluation of aroma, colour, texture and overall acceptability. In Table 6.2 it is seen that aroma did not vary significantly amongst the samples. The colour of the 40°, 50° and 30°C products did not differ significantly but 40°C differed significantly (P < 0.01) from the 60° and 70°C products. 50°, 30°, 60° and 70°C products were not significantly differed. Texture did not vary significantly amongst the products but for overall acceptability the 40°, 30° and 50°C products were significantly preferred (p < 0.01), although the 30°, 50° and 60°C products were not significantly different. Based on this and the rapid drying rate achieved at 50°C, this temperature was chosen as the drying temperature for storage studies. 230 Table 6.2: Mean scores of dried salted sardines* • Attributes Drying Temperature °C 60° 50° 70° 30° 40° Aroma 7.21a 7.13a 6.91a 6.60a 40° 50° 30° 60° 70° Colour 7.43a 6.67ab 6.42ab 6.27b 5.63b 60° 50° 40° 30° 70° Texture 7.13a 6.93a 6.90a 5.85a 5.46a 40° 50° 30° 60° 70° Overall 7.73a 6.87ab 6.81ab 6.12ab 4.65C ★ Different superscripts indicate significance at 99% confidence level. 4.1.3.5. Chemical properties of salted dried sardine 4.1.3.5.1. Effects of salting (a) pH The initial pH of fresh sardine was 6.1 which decreased to 5.7 after 4h brining rising to 5.85 after 12h and then remaining relatively constant (Fig. 4.6, p 140) • The initial decrease in pH could be accounted for by removal of amines as water was removed from the tissue and subsequent small rises by further production of amines by microbial action. Purnomo (1985) also observed the initial pH decline on 231 salting sardines and accounted the buffering capacity of the salt as the primary factor for the decline. Further production of amines is possible as the fish was salted while and bacterial degradation could still occur as salting proceeded. Fuji et al. (1977) found the Vibrio aeromonas group of bacteria in fish homogenate which were responsible for the production of TMA. This group of bacteria were detected in the digestive tracts of sea bream by Sera and Ishida (1972) although they disappeared after seven days' storage with 15% salt. (b) TVB and TMA TVB nitrogen in sardine increased during the first 4h of brining, from an initial level of 35mg% to 48mg%, and then gradually decreased to 18.5mg% after 48h brining (Fig. 4.7), TMA followed a similar pattern, from an initial value of 16.5mg% to 44mg% after 4h, then gradually decreasing to 17.5mg% after 48h. The rises are attributed to bacterial action during early stages of brining as was also observed by Fujii et al. (1977). The rise in both TVB and TMA during salting were higher than in either morwong or shark. Suzuki (1981) reported that dark-fleshed fish contains more TMA than white-fleshed fish. Kawabata (1953) have also discovered that dark-fleshed fish such as sardine contain enzymic activity to reduce TMAO to TMA. These observations explain the higher rise in volatile base levels in sardine as compared to shark and morwong at early stages of brining. 232 At longer brining time, high salt concentration in the flesh would inhibit the microbial and enzymic activity responsible for production of the amines (Beatty and Fougere, 1959). (c) Protein solubility Protein solubility decreased significantly (P < 0.01) during brining. The solubility is low in KC1 at 25°C, being initially 21% and decreasing gradually to 7.5% after 48h. The solubility in KCl at 77°C was similar to that in KCl at 25°C, except prior to brining when it was significantly different (Fig. 6.2). The solubility in SDS + 8 mercaptoethanol is much higher. However the salting out effect reduced the solubility as brining time increased (Fig. 6.2). From an initial 61.5% solubility decreased gradually to 34.5% after 48h salting. The observed effects of salting on solubility of sardine protein follows the pattern found with other fish used in this study. The salt soluble fraction consists mostly of myofibrillar protein and some sarcoplasmic protein (Poulter et al., 1985). The myofibrillar fraction is highly sensitive to changing conditions (Aitken and Connell, 1979) About 20-30% of the myofibrillar proteins in unsalted sardine is inextractable due to physiological changes (Suzuki, 1981). Thus further changes such as salting out cause reduction in water holding capacity of the proteins which in turn become increasingly insoluble (Kinsella, 233 SDS ■+■ P mercaptoethanol KC1 at 25 °C KC1 at 77 OC 234 oo Tf Figure 6.2: Effect of salting time on protein solubility of sardine. 1982). The solubility of the soluble fraction decreased further with increasing salting time. (d) Isoelectric focussing Shown in Fig. 6.3 are the IEF patterns of the water-soluble proteins extracted from sardine after brining for various times. Seven major bands A, B, C, D, E, F and H were observed in the non-salted sample, s. However on brining some bands are seen to intensify and others become faint. In b for instance band C intensified whilst band D became diffused. Band F disappeared but band E intensified. Band G also intensified. On increasing brining time, band A decreased, band B remained intact, but C has intensified and C' generated more bands than originally present. Band D have remained intact in samples f and g and i and were faint in other samples. Bands E and G generally lost intensity as brining time increased, but band H remained intact. Generally fewer bands are seen with increasing brining time. Only the very anodic and some cathodic bands are seen in the IEF patterns. (e) In vitro digestibility The digestibility remained high during salting with small changes only. Salting therefore does not affect digestibility of the product negatively. 235 Figure 6.3: IEF patterns of water soluble proteins of sardine after brining for 0, 2, 4, 8, 12, 24, 36, 48 and 72h (a, b, c, d, e, f, g,h. i respectively). 236 Table 6.4: In vitro digestibiligy of salted sardines. Salting time (h) 0 4 8 12 24 36 48 % Digestible 99.5a 99.3a 99.2a 99.4a 98.6a 98.5a 98.la protein (f) Total amino acid Table 6.5 shows the total amino acid contents in sardine salted for different times. Variations occurred but no specific trends could be detected. Decreases were observed in glycine and histidine, the latter could be utilised by microorganisms as indicated by the TVB rise during salting. Histamine has been observed to form during curing of pickled sardine (Wado and Koizumi, 1986). Lysine was not affected at all and it appears that salting does not affect total amino acid content negatively. 237 Table 6.5: Total amino acid content in salted sardines3 Salting time (h) Amino acid g/16gN 0 12 24 48 Aspartic acid 9.00 10.74 10.46 11.48 Threonine 4.54 5.06 5.07 5.31 Serine 4.14 4.50 4.50 4.51 Glutamic acid 11.62 13.29 12.93 11.81 Proline 9.27 5.92 5.84 8.26 * Glycine 6.24a 6.23a 5.39b 5.06b Alanine 8.57 8.60 7.98 7.91 Cystine 0.87 0.82 0.82 0.82 Valine 6.13 6.91 6.75 6.56 Methionine 3.13 3.37 3.29 2.89 Isoleucine 4.66 5.01 5.69 5.12 Leucine 7.84 8.10 8.59 8.26 Tyrosine 3.02 2.63 2.92 3.07 Phenylalanine 2.88 2.59 2.53 2.98 Lysine 8.11 9.11 9.35 7.99 Histidine* 4.22a 1.81b 2.16b 3.04C Arginine 6.13 6.95 7.11 5.72 Amino acids showing decreasing trend. Different superscripts along horizontal columns indicate significance at 99% confidence level. 4.1.3.5.2. Effects of drying on sardine (a) Total volatile bases (TVB) and trimethylamine (TMA) Fig. 6.4 shows that TVB and TMA in sardine increased significantly (p < 0.01) on drying, more so at higher drying temperature. The TVB increased from 44.8 mg% to 49.2% at 238 80 o TVB • TMA __ L______1______j 30° 50° 70° Drying temperature °C Figure 6.4; TVB TMA contents in sardines dried at different temperature. 239 30°, 64.5 mg% at 50°C and 83.4mg% at 70°C. The TMA increased from 42 mg% to 46.5mg% at 30°C, to 59.4 mg% at 50°C and to 70 mg% at 70°C. Such increases have been reported previously (Sigurdson, 1947; Kida and Tamoto, 1976 a, b). Velanker (1952) also observed high volatile nitrogen in cured sardines and Tokunaga (1975) reported that decomposition of TMAO to TMA and DMA was generally high for red-fleshed fish and low for white-fleshed fish. (b) Protein solubility Solubility of sardine decreased with increasing drying temperature (Fig. 6.5). The solubility in both KC1 at 25° and 77°C were very poor being < 20%. In SDS + B mercaptoethanol the solubility is fair, but decreased from 47.5% in the 8h salted sardine used as a starting material to 40.5% in the sample dried at 70°C. Suzuki (1981) reported that proteins of pelagic fish such as sardine and mackerel are very heat coagulable. Water-extracted proteins from pelagic fish were almost completely congulated by heating for 10 min. at 90°C, whilst only 65-75% of proteins of demersal fish were coagulated under these conditions. The T50, or temperature at which 50% of myofibrillar protein becomes denatured, has been identified for sardine at 33°C at pH 7.0 and 21°C at pH 5.5 (Suzuki, 1981). The effect of salting and drying on the heat sensitive proteins in sardine, therefore leads to decreased solubility of the proteins. 240 Soluble protein Figure o Lsd — 6.5: -o -• at KCl SDS 1%- mercaptoethanol Effect solubility KCl at + 4.996 77°C Drying at P mercaptoethanol of 25 u temperature^ drying 241 and of proteins 77 , temperature u c and of in sardine SDS on + the $ in (c) Isoelectric focussing Fig. 6.6 clearly demonstrates the effects of salting and drying on the IEF of water-soluble proteins of sardine. It can be seen that most bands start disappearing even at a drying temperature of 30°C (sample c) . At higher drying temperatures almost all the bands disappear except for bands B and E which is still faintly visible in the 60°C sample (sample f). Band E is more cathodic than the very anodic bands surviving drying in the IEF patterns of shark, morwong and squid. In the 70°C dried samples, all bands had disappeared. (d) In vitro digestibiligy Digestibility of salted sardines dried at different temperatures is shown in Table 6.6. Differences are very small and the digestibility was high at all drying temperatures. Aitken et. al (1967) also counteracted the belief that the nutritional value of fish is lowered during salting and drying by finding no significant effects as a result of salting and drying processes on net protein utilisation (NPU). Yanez et al (1970) also reported that digestibility was not affected in hake dried at 105°C although it declined by 25% at 180°C. Lee and Ryu (1987) on the other hand observed increases in the digestibility of salted yellow corrina and boiled anchovy on drying in hot air at 55°C and cited structural changes in the proteins forming enzyme susceptible bonds as causes of the increase. 242 d. PH 3" 10 G iCMl !»i i 1 51P Ml ! 11 1 ) II* F 1 IK. III ts II* F ^ 1 IK: II- tat ; Sfc t 1 IS 1 8 soil* \\m i im i 30*c C im 1 * «41 i 1 1 . iH 1 I Sol e 1 *#•«. f 70*o J | " ill l IN 1 > 1 Mil 1 * ) 1 A 8 cD t p cH - 1 HI I 1 II 4 Figure 6.6: IEF patterns of water soluble proteins of sardine brined for 8h and dried at 30°C, 40°, 50°, 60° and 70°C (c, d, e, f and g respectively; a = fresh sample, b = 8h salted sample). 243 Although Ryu and Lee (1965) observed that fat content in seafood could affect the digestion of proteins this factor was not apparent in this study. Although sardines are regarded as a fatty fish, their fat content was only 3.6% in this study which may explain the lack of effect of fat content observed. Table 6.6.: In vitro digestibiligy of sardine* Drying temperature °C 0 Fresh fish 30° o 50° 60° 70° % Digestible 99.5a 98.5a 97.5a 97.0a 97.4a 94.0b protein *Different superscripts indicate significant at 95% confidence level 4.1.3.6. Storage studies Sardines were salted for 8h and dried at 50°C to a moisture content of approximately 30% and stored at 5°, 25° and 37°C at 50% RH. The products, packed in low density polyethylene bags, were examined periodically for mould growth, general product appearance and chemical composition. (a) Product appearance The appearance of dried sardine during storage at each temperature can be seen in Plate 10. Products stored at 5°C developed yellowish light brown colour around the belly and 244 Plate 10: Appearance of dried sardines stored for (a) 4 wk and (b) for up to 24wk at 5°, 25° and 37°C. 245 the head region. This colour presumably originated from the fats in the fish. The products stored at 5°C appeared to be acceptable until the 16th week of storage. The product kept for 24 weeks at 5°C had developed rupturing of the belly area and browning in the flesh. Products stored at 25°C were completely unacceptable by the 4th week due to severe browning, loss of skin, fat deposition on the external body surface and considerable physical breakdown. Products stored at 37°C were also unacceptable by the 4th week (products not shown in the picture). In the products kept at 37°C, liquid fat which was brown and rancid was visible in the package and the flesh was disrupted and broken. At 5°C the products smelled rancid by the 16th week but at 25° and 37°C they were rancid after 4 weeks. The texture of the product stored at 5°C remained hard but pliable. At 25°C texture remained hard except at 24 weeks where flesh disintegrated. At 37°C disintegration occurred after only 8 weeks. (b) Product colour The changes in the L, a, and b values of sardine on storage are given in Table 6.7. The L values decreased significantly with storage time (p < 0.01), slightly more 246 Table 6.7: Hunter L, a, b for sardine stored at 5°, 25° and 37°C * Storage Time (wk) Temperature °C 0 4 12 24 5° L 32.95a 31.00a 29.75a 2 7.05b a 4.40a 4.63a 4.70a 5.10a b 10.00a 10.60ab 12.70b 12.75b 25° L 32.95a 30.50a 26.80b 27.00b a 4.40a 4.83a 4.77a 5.50b b 10.00a 10.55a 12.70a 12.85b 37° L 32.95a 29.05b 24.05C 25.75C a 4.40a 4.83a 4.80a 5.10b b 10.00a 11.00cb 12.7 5bc 13.10C *Different superscripts along horizontal column indicate significance at 99% confidence level L: Lsd 1% = 3.67 a: Lsd 1% = 1.01 b: Lsd 1% = 2.30 rapidly at 37°C. The decrease in L values indicate progressive darkening. The a values increased slightly on storage indicating some development of redness in the products. The b values are also seen to increase slightly over the storage period and with increasing temperature of storage. These increases quantify development of browning through Maillard reaction or lipid-protein interaction. The 247 browning problem is very severe in the products stored at both 25° and 37°C which were rejectable after 4 weeks. Fujimoto et al. (1971) stressed the importance of reaction of lipid oxidation products in browning processes, particularly in fatty fish (Obanu, 1983). Considerable browning was found in salted dried mackerel (Rao and Bandyophyay, 1983). Autoxidation of lipids with subsequent browning was observed during freeze drying and storage of sardine (Kunimoto et al., 1985), more so in higher humidity environment. Sundhakaran and Sundhakara (1985) also observed development of high TBA no., with high levels of browning in dried sardine products during storage. Maillard browning could also occur through the interaction of amino acids and sugars in the fish. Ono and Nagayamo (1959) also observed that the degree of browning grew in parallel with the length of storage and that free reducing sugars increased during storage. Regardless of the main cause of discolouration, browning was very severe during storage of sardine at 2 5° and 37°C and such products should be rejected at 4 week storage. (c) Flesh structure Plates 11 and 12 show SEM micrographs of sardine tissue at different stages of brining. As brining time increases the tissue appear more disrupted due to the effects of salt. As the water is withdrawn, the proteins lose water holding 248 8h and 4 2, 0, for brined . sardines of respectively^ d and c micrographs (®/ SEM 111 Plate 249 48h and 36 24, for brined sardines of respectively). c and micrographs b (a, SEM 12: Plate 250 capacity, resulting in fibre shrinkage and distortion leading to disrupted appearance which increases with brining time. In Plate 13 the effects of drying temperature on tissue structure of sardine brined for 8h are demonstrated. In micrograph a in Plate 13 it is seen that on drying at 40°C the tissues appear to be undistorted and cohesive in structure. Kanna et al. (1971) observed similarly and described the cohesion of muscle fibres when drying fish between 18-20°C. At higher temperature, fibre shrinkage and some degree of fragmentation occurs. These appear to be disruption of the connective tissue between the fibres at higher temperatures. Myofibrillar protein water holding capacity decreases between 40° and 50°C and collagen starts shrinking at 60°C (Tarrant, 1982). Changes in structure during drying are attributed to these effects. The texture of 40°C dried samples was still quite pliable, but at 50°C and above it became hard and inflexible. This behaviour parallels structural changes shown by SEM. (d) Reconstitution Reconstitution of the dried fish gave varied results (Tables 6.8 and 6.9). While lower salted samples rehydrated very well, the flesh of higher salted samples tended to fragment on draining. Salt denaturation probably increased the tendency for the flesh to break along the length of the 251 . dried and 8h for respectively) d c, brined b, (a, sardines 70°C of and 60° 50°, micrographs SEM 40°, 13: Plate 252 fish. The skin of the highly salted samples was softer and dislodged from the surface. Generally poor products are obtained on reconstituting more highly salted fish. Products dried at 30°C rehydrated rapidly that the flesh became soft very fast. Products dried at higher temperatures rehydrated to give more manageable products. Although the rehydration ratios appeared satisfactory, salted sardines need not be rehydrated before being cooked. Rehydration made the products soft and during frying the products broke up. Unrehydrated products gave better cooked products. There are already some species of dried fish which are customarily cooked without rehydration such as anchovies (Stolephorus sp. ) white bait (Anchioviella sp. ) and scad (Caranx sp) . Rehydration of stored products was not attempted because of their poor quality. Table 6.8: Rehydration ratio of sardine salted for different times and dried at 50°C Brining time (h) 4 8 24 48 Rehydration ratio 1.98a 1.73b 1.57b 1.62b * Different superscripts denote significance at p < 0.05 253 Table 6.9: Rehydration ratio for sardine salted for 8h and dried at different temperatures Drying temperature °C 30° 50° 70° Rehydration ratio 2.05a 1.73b 1.43c Different superscripts denote significance at p < 0.05 4.1.3.7 Chemical Properties of stored sardines (a) Moisture content and A Moisture contents of dried sardine stored at 5°C increased (Fig. 6.7) from 28% to 39% after 4 weeks and then fluctuated reaching 34% after 24 weeks. At 25°C moisture content increased slightly to 29.5% at 4 weeks and then gradually decreased to 26.5% at 24 weeks. At 37°C, the moisture decreased slightly to 27% after 4 weeks storage and decreased further to 23.5% at 24 weeks. The Aw (Fig. 6.8) reflected changes in moisture content. The Aw in the 5°C product increased initially from 0.65 to 0.74, fluctuating, and finally decreasing over the last three months to 0.72. At 25°C Aw increased very slightly to 0.66, fluctuated and gradually decreased to 0.64 after 24 weeks. At 37°C, the Aw remained the same for 8 weeks and then decreased gradually to 0.64 at 24 weeks. No mould growth was seen in the samples. But rancidity was strongly evident. Moisture loss in sardine during storage is much less than was observed for morwong or shark. Obanu (1987) attributed this to the water retention effect of muscle fat since water diffusion is slower when the fat content is higher. 254 +-> dried to 4- in O +JO QJ 4— 4- LU content CD CU DS- cn •r- U_ }U»}B03 »in)9|0^ % 255 0.801- 256 © so o Ifl N to l ? o VO M 6 9 2 CO SO TJ a — I ^N. <0 O) v o E >- *> m Figure 6,8; Changes in A of dried sardines during storage w at different temperatures. (b) Rancidity Table 6.10 shows that TBA no. increased with storage at the three temperatures. The TBA no. at 5°C peaked at 16 weeks and then decreased. At 2 5°C the TBA no. peaked at 8 weeks Table 6.10: TBA no. in stored dried sardines Storage Storage time (week) Temperature °C 0 4 8 12 16 24 5° 1.20a 3.20b 4.56c 4.62C 5.43de 4.98cd 25° 1.20a 4.44C 5.82d 5.40de 4.90ce 3.81f 37° 1.20a 4.98C 6.70g 6.153d 5.43d 3.20f Different superscripts denote significance at 99% confidence level and at 37°C at 8 weeks. Since TBA value is a measure of malonaldehyde type compounds, this trend indicates generation of carbonyl compounds which are involved in further changes such as lipid-protein interaction (Obanu, 1987). The increase in TBA no. is accompanied by rancid smell. According to Lupin (1981) TBA no. of 1-1.5 accompanies detectable rancid odours. The products stored at 5°C had the tangy odour typical of dried salted fatty fish but those stored at 25°C and 37°C were already rancid by week 4 and were obnoxious during the later part of the storage. Sudhakaran and Sudhakaran (1985) observed 257 development of rancidity in salted dried sardine products even during drying. On storage the TBA no. increased and peaked at 2 months (4.3) and decreased to 2.4 at 3 months. Nambudiry (1980) also observed increase in TBA no. in salted sardines but the increase is less in sardines brined in saturated salt than in lower salt concentration. Rao and Bandyopadhayay (1983) also reported high TBA no. in cured mackerel while Obanu (1983) observed higher rancidity in fatty fish than in similarly preserved lean fish. (c) Total volatile bases (TVB) TVB increased substantially during storage of the products (Fig. 6.9) and to a greater degree as storage temperature increased. TVB rose from 65mgN/100g at the start of storage to 124mgN/100g in the 37°C product at 24 weeks. In the product stored at 25°C, TVB fluctuated, attaining a maximum level of 94.5mgN/100g at 16 weeks which decreased slightly to 93mgN/100g at 24wks. At 5°C, the TVB rose to 79mgN/100g by the end of storage. Aw remained about 0.65 in the stored products stored at 25° and 37°C and <0.75 in the products kept at 37°C, it is concluded here that the TVB rise is probably due to the chemical or enzymatic effects rather than microbial changes. (d) Protein solubility Protein solubility of dried sardine in KC1 at 77°C and in SDS + B mercaptoethanol decreased over storage time. This 258 uo ~o 37 o b OOI/NB ui 259 Figure 6,9: Changes in TVB content in dried sardines on storage at different temperatures, KC1 iri sardine dried of temperature proteins different of at storage Solubility IQ: 6. Figure ujajojd 3{qn|o§ % 260 o c (O sz -M protein mercaptoe on 6 + SDS in temperature and sardines time dried of storage of U o o Effect solubility 6.11: Figure u|»)Oid »|qn|o$ % 261 decrease was greater at higher storage temperatures (Figs. 6.10 and 6.11). The decrease in the solubility in KCl for products stored at 5°C is small, from an initial level of 14.5% to 13.5% after 24 weeks. Corresponding final solubilities were 9.8% at 25°C and 9.2% at 37°C. Decreased solubility in SDS + B mercaptoethanol paralleled that in KCl. Thus, from an initial level of 42% it decreased to 41.7% at 5°C, 35% at 25°C and 34.8% at 37°C after 24 weeks. Decreased solubility of proteins reflect the effects of denaturation due to brining, drying and storage. Lipid oxidation could well play an important part in this insolubilisation as carbonyls formed are very effective cross-linking agents (Chio and Tappel, 1969). Howgate and Ahmed (1972) observed similar losses in solubilities. Although Toyomizu et al. (1963) concluded that the effects of oxidised lipids on the extractibility of myofibrillar protein is small, it may not be the case in this study as lipid oxidation was extensive. (e) Isoelectric focussing Fig. 6.12 shows the IEF patterns of water soluble proteins of stored sardines. Most of the bands became more diffused and lost intensity during storage. The IEF patterns of products stored at 5°C stained lightly darker than those products stored at the other temperatures. The same bands E and F as in the IEF patterns of dried products (Fig. 6.6, P243) survive storage. These bands are more cathodic than 262 sre*£f &AAD1MS pU %-\o ® Hut as/(o/g 7 &-1 I m/s' a H(tf b i|n c sw 0 nf» e t>fo f 9 *4* h »* i E F Figure 6.12: IEF patterns of water soluble proteins of sardine brined for 8h, dried at 50°C and stored at 5°, 25° and 37°C for up to 24wks. (a, b, c = 4wk at 5°, 25° and 37°C respectively) (d, e, f = 12wk at 5°, 25° and 37°C respectively) (g, h, i = 24wk at 5°, 25° and 37°C respectively). 263 those observed in the IEF patterns of other stored products and are perhaps specific to the sardine species. However, the effects of drying and storage conditions are well demonstrated in the IEF patterns. (f) In vitro digestibility Digestibility of dried sardines were found to be high on storage (Table 6.11) at all the three temperatures. Despite the extensive browning observed in the products, they were higher than the 92% limit proposed for dried salted fish (Heen and Kruezer, 1962) and higher than the 88% digestibility of casein. Some amino acids were found to decrease significantly during storage at 25° and 37°C. These may have been involved in oxidised lipid-browning reactions. Lee and Rhu (1987) noted the difficulties in studying the digestibility of seafoods and proposed that the methods of Jewell et al. (1980) for predicting the quality of protein solely from amino acid composition data be used. Nevertheless in this study it appears that storage did not affect digestibility of the products to a significant degree in a nutritional sense. 264 Table 6.11: In vitro digestibility of dried sardine on storage at different temperatures Drying Storage time (week) Temperature °C 0 4 8 12 16 20 24 5° 99.8a 99.0a 99.2a 98.3a 99.2a 98.0a 97.2a 25° 99.8a 98.7a 97.2a 98.6a 98.9a 97.8a 96.4a 37° 99.8a 99.0a 98. la 98.3a 96.2a 94.9b *Different superscripts denote significance at 95% confidence level (g) Total amino acids The total amino acids (Table 6.12) vary during storage at different temperatures. However no trends can be observed except for some amino acids which show a decreasing trend. At 5°C, the amino acids which show a decreasing trend (p < 0.01) were threonine, serine, alanine and valine. At 25°C these amino acids were proline, alanine, methionine, isoleucine, leucine, tyrosine, lysine and histidine. At 37°C the amino acids which decreased were isoleucine, leucine, tyrosine, lysine and arginine. These amino acids were probably involved in both Maillard (Dworshak, 1980) and oxidised lipid-protein reactions (Frankel, 1984). Kunimoto et al. (1985) observed that loss of available lysine was proportional to browning of protein moiety in fish. The decline in some amino acids in intermediate moisture food on storage was also attributed to cross-linking reactions 265 Table 6.12: Total amino acid contents in stored sardine Storage temperature °C Amino acid 5° 25° 37° g/16gN ______ Storage time (wk) 4 12 24 4 12 24 4 12 24 Aspartic acid 1.14 9.55 9.98 12.14 9.92 10.98 10.64 10.06 11.00 Threonine 5.73a 5.16b 5.10b 6.32 5.04 5.30 5.26 4.99 5.39 Serine *4.87a 4.42a 4.51a 5.27 4.34 4.64 4.47 4.53 4.13 Glutamic acid 10.16 11.77 14.10 14.94 11.85 15.16 12.47 14.19 14.86 Proline 4.40 3.80 4.29 *5.39a 4.73b 4.65b 4.71 4.85 5.28 ★ Glycine 5.79 4.60 5.39 6.00 5.01 6.12 5.67 5.72 6.67 Alanine *11.62a 8.14b 6.98C 10.12a 8.71b 6.64C 9.19 7.16 8.34 Cystine 0.20 0.82 0.62 0.53 0.68 0.72 0.79 0.68 0.74 Valine 6.68a 6.57a 6.03b 8.06 6.26 6.63 6.53 6.02 6.51 Methionine 3.09 4.01 3.35 *4.89a 3.67b 3.54b 3.29 3.19 3.40 I so leucine 4.52 5.32 4.71 *6.38a 5.18b 4.51C *5.21a 4.62ab 4.22* Leucine 7.41 8.03 7.44 *9.75a 8.19b 6.45C *8.00a 7.25ab 6.51* Tyrosine 2.56 3.77 3.07 *4.32a 3.29a 2.83b *3.56a 2.96a 2.66' * Phenylalanine 2.53 3.47 3.58 *3.90a 2.96ab 1.88b 2.93 3.61 1.54 Lysine 8.82 9.23 9.16 *11.67** 9.10b 8.85b 10.203 9.19b 8.60* Histidine 4.09 4.32 4.69 *5.65a 4.42b 4.30b *4.64a 4.13 4.57 Arginine 6.43 7.01 7.01 8.40 6.63 6.79 *7.06a 6.86a 5.59* ★ Amino acids showing decreasing trend. Different superscripts along horizontal columns indicate significance at 99% confidence level. between carbonyls (as measured by TBA) and amino acids (Obanu et al. 1976). Devadesan et al. (1985) reported that oxidation of fish oil in the initial stage affects only the sulphur amino acids without pronounced deterioration of 266 sulphur amino acids without pronounced deterioration of protein quality, However further oxidation of the oils seems to affect other essential amino acids such as lysine causing substantial reductions in the protein quality. 4.1.4 Squid 4.1.4.1 Chemical composition of fresh squid Squid of species Nototodarus gouldi was used to prepare dried squid products. Squid as purchased contained 80.8% moisture, 17.1% protein, 1.01% fat and 1.78% ash. These values compared well with published data (Vlieg, 1984; Jhaveri et al. , 1984 and Sikorski and Kolodziejska, 1966). The pH of the fresh squid was 6.4, its total volatile bases (TVB) content was 18.9mgN/100g and trimethylamine content (TMA) was 5.6mgN/100g. The pH and TVB values indicated freshness as TVB content of 30mgN/100g mantle has been proposed as a measure of grade A quality, 30-45mgN/100g for grade B and >45mgN/100g as unacceptable (Ke et al. , 1984). However, the TMA value in this product is higher than 3 which is advocated by Ke et al. (1984) as maximum for grade A quality. Their recommended guidelines were TMA values of <3 for grade A, 3-10 for grade B (acceptable) and >10 as being unacceptable. On the whole, the extractable nor* protein nitrogen is high, at 750- 900 mg% compared to 300-400 mg% in other fishes (Ishikawa and Nakamura, 1975) and 27.2% of this is accounted for by trimethylamine oxide nitrogen (Konosu et al., 1958). The total amino acid composition of the fresh squid is given in Table 4.1 (page 127) and is comparable to that reported by 267 Hayashi and Takagi (1979). The ratio of the essential to the nonessential (E/NE) was 0.69 compared to 0.62 as reported by Iwasaki and Harada (1985). Squid contains large amounts of free amino acids such as glycine, alanine, proline and betaine (Ke et al., 1979) which are responsible for the sweet taste of squid meat. The in vitro protein digestibility is given in Table 4.2 and is significantly lower (P < 0.01) then other fishes. This observation concurs with those of Tanikawa and Suno (1952) and JTiaveri et al. (1984). Squid protein was reported to have a PER value inferior to other fishes although still higher than casein (Jhaveri et al., 1984). The low digestibility of raw squid protein could be related to its musculature which has been described by Ward and Wainwright (1972) as consisting of alternating bonds of circular and radial obliquely striated fibres interdispersed with connective tissue fibres. Squid contains 1.5% (Stanley and Hultin, 1982) and 2.5 - 3.0% (Kolakowski and Gajowiecke, 1973) of collagen which is much more than fish muscle and about three times that found in beef L. dorsi muscle (Lawrie, 1966). This could contribute to lower digestibility of squid protein compared to other fish. 4.1.4.2 Drying of squid and product quality The effect of temperature on the rate of drying of the squid products is shown in Fig. 7.1 Compared to other species studied drying rates are high at all temperatures with that at 50°C 268 o------o 30*^0 o------0 40°C o------c 50°C o------a 60°C •------• 7d°C Lsd at 1%*0.0538 X 1.4 Drying time ( h ) Figure 7,1: Effect of temperature on drying rates of sguid. 269 significantly higher (P < 0.01) than at other temperatures. However, it is only significantly higher (P < 0.01) than the rate at 70°C in the first few hours of drying. At 50°C and above much of the moisture was removed in the first few hours of drying, thus the constant rate period is approached rapidly. When compared with other fish (Fig. 4.4, page ^33 ) squid dried significantly faster at 50°C (P < 0.01). Since the squid were cut open and laid flat exposing a large surface area for a small volume (the mantle thickness ranged from 0.4 - 0.8 cm), high rates of drying resulted at all temperatures. Drying at 50° and 60°C gave products which were rather brown in colour. Drying at 70°C on the other hand gives product which is pale yellow. This is attributed to the drying time being very short. Thus, the product is not exposed long enough to heat for browning to occur. Browning in squid has been attributed to nonenzymic action between reducing sugars and amino acids. Squid also contains betaine derivatives which accelerate Maillard browning (Haard and Arcilla, 1985). The problem with this irreversible change is the loss of acceptable colour, development of off flavour and loss of nutritive value. Colour is important from aesthetic point of view. In the Chinese National Standard for Dried Squid (1981) it is specified that colour for grade A product should be yellowish white or yellowish brown whilst that for grade B is dark grey or dark yellow in colour. The ivory white colour is also specified 270 in Bangladesh Standards (Krishnan, 1982). However, for the proposed Canadian Dried Squid Standard, amber colour is chosen (Ke et al., 1984). At 30° and 40°C there is no browning problem. However, all products develop browning on prolonged storage. Thus, from the aspect of colour 30°C and 40°C could be considered as suitable drying temperatures. Drying at higher temperature gives product of brittle texture as opposed to pliable texture that is obtained at lower temperatures. The texture is also important from the packaging point of view. Brittle products will break very easily whilst pliable products will be able to withstand stresses and pressures better. This suggests that drying at temperatures of 30° or 40°C are the best choice but 40°C is preferred because of the slightly faster rate of drying. Table 7.1 shows the moisture content and Aw in squid products dried at different temperatures. The moisture is < 15% for all the samples and the Aw ranges from 0.51-0.65. The products therefore will not have microbial problems unless they subsequently absorb moisture. The product will be prone to other d.egradative problems such as non-enzymatic browning and lipid oxidation (Karel, 1973) although squid only contain low amounts of lipids. The moisture content in sun dried products is, however, often quoted to be between 16-22% (Buisson et al., 1985) or even 15% as produced in Bangladesh (Krishnan, 1982). In Table 7.1, it is seen that the predicted shelf life for squid with such moisture content and Aw is more than a year. 271 content in dried squid. Table 7.1: A w and moisture Drying Moisture Salt % Predicted temperature o C measuredAw content (wb) shelf life+ 30° 0.43 13.5 8.0 1 year + 0 o 0.38 12.1 8.2 1 year + 0 o in 0.31 10.7 8.3 1 year + 60° 0.41 14.9 7.9 1 year + 0 o f'- 0.44 15.0 7.9 1 year + +Predicted shelf life is read from Pulter (1980) and Curran (1964) 4.1.4.3 Sensorv evaluation: Acceptabilitv vs drvinq temperature Squid dried at different temperatures were prepared for sensory evaluation. The product was fried in oil at 200°C for 3 min. and served to a panel of untrained assessors. Table 7.2 shows the mean scores for each attribute evaluated. There was no significant difference among the treatments for aroma, texture and overall acceptability. However, the 30°C scored significantly less (p < 0.01) for colour. Eventhough the 50°C sample scored the highest for colour, it was not significantly different from the others except the 30°C sample. Hence drying temperature for squid could be any of those examined, based on the colour, except 30°C, 50°C is recommended as drying temperature on the basis of high drying rate. 272 Table 7.2: Mean scores for dried squid+ Sensory Drying temperature °C attributes 70°C 60°C 40°C 50°C 30°C Aroma 6.83a 6.69a 6.69a 5.65a 5.56a 50°C 70°C 40°C 60°C 30°C Colour 6.69a 6.46a 6.41a 6.29a 3.45a 60°C 50°C 70°C 40°C 30°C Texture 6.62a 6.38a 5.99a 5.57a 4.95a 70°C 40°C 60°C 50°C 30°C Overall acceptability 6.53a 6.51a 6.01a 5.84a 5.32a +Different subscripts denote significance at 99% confidence level. 4.1.4.4 Chemical properties in dried squid (a) Total volatile bajes (TVB) and trimethylajfTiJ'n^ (TMA) Table 7.3 shows the TVB and TMA contents in squid dried at different temperatures. The values for the 30° and 50°C dried samples were less than others. This is possibly due to faster rate of drying at 50°C and the lower temperature at 30°C. The development of TVB and TMA in squid on drying is less than that of the other species (Figs. 4.11, 5.4, 6.4). Although squid contains high amount of TMAO, the faster rate of drying led to smaller thermal decomposition 273 of TMAO which results in smaller development of TVB and TMA. TVB and TMA contents are also dependent on the initial quality of the raw material and the processing treatment. Fresh raw material of TVB below 10 mgN/lOOg and rapid drying with given dried products of good quality whilst low quality raw material (TVB 10-30 mgN/lOOg) and slow drying will result in dried products of inferior quality (Ke et al. , 1984). Kawamura et al. (1971) found great accumulation of DMA (dimethylamine) in squid and some other fish species after drying. This was confirmed by Kawabata et al. (1982). Nitisewojo and Hultin (1986) observed that the nonenzymic breakdown of TMAO is influenced by temperature. Other drying studies in squid found that when drying at 100°C, TMA increased progressively from 50.3 mg% to 82.5 mg% in 90 min and then remained caster. At 200°C, TMA rose rapidly from 50.3 to 107.7 mg% in 60 min. and fell to 53.9 mg% by 120 min (Lin and Hurng, 1985). (b) Solubility of dried squid protein Solubility of dried squid protein (Fig. 7.3) is low in KCl at 25°C and slightly higher at 77°C. However, in SDS + J3 274 •---- • SDS P mercaptoethanol o-----o KC1 at 25°C °-----"oKCl at 77°C Ud at 1% = 0.9132 Fresh Drying temperature °C Figure 7.2: Effect of drying temperature on the solubility of proteins of squid. 275 Table 7.3: Total volatile bases and trimethylamine contents in squid dried products. Drying conditions TVB mgN/100 g TMA mgN/100 g Fresh 13.7 5.6 30°C 23.2 12.6 40°C 30.2 17.9 0 o m O 27.9 11.2 60°C 54.4 27.14 70°C 62.9 49.6 + Japanese dried squid 89.4 50.3 + New Zealand squid 62.7 50.6 + Formosan squid 73.2 33.6 ++ Canadian dried squid <140 (proposed standard) + Lin and Hurng, 1985 ++ Ke et al., 1984. mercaptoethanol, solubilisation of proteins (mostly native and denature) is greater. Squid protein is more soluble in SDS + 8 mercaptoethanol than that of the other species. This is probably because of its high drying rate and the absence of a salting step. Solubility is SDS + B mercaptoethanol and KCl at 25° or 77°C decreased significantly (P < 0.01) on drying up to 70°C. However, the solubility of products dried at the different temperatures are not significantly different in KCl at a either 25° or 77°C. Solubility in SDS + B mercaptoethanol 276 of samples dried at 40°, 50° and 60°C does not differ significantly. Squid has been reported to lose about 30% of its weight during the first 5 min. of cooking (Kolodziejska et al. 1987). It is anticipated that some loss of protein along with the aqueous phase may take place during drying. However, the decrease in solubility is probably only due to heat denaturation. (c) Isoelectric focussing Water soluble squid proteins congregate in the anodic region (Fig. 7.3) In the fresh sample, three major bands A, B and C, which is less distinct, are present. There are also other minor bands around bands A and B. As drying temperature increases these bands lose intensity. Band A is still, clearly seen in the 40°C dried sample, however, at 50°C it almost disappeared with band B the only major one seen in the latter sample. In the 60° and 70°C samples, the bands are faint and diffuse, with some streaking. Streaking is seen even in the fresh sample. Streaks are mainly due to the presence of peptides and denatured proteins of uncertain pis. Squid is also known to autoproteolyse (Rodger et al. , 1984^ which could contribute to the streaking (Kolodziejska et al., 1987). Thus IEF demonstrates the effects of heating on the water soluble proteins of squid. (d) Digestibility of dried products Table 7.4 demonstrates that digestibility improves significantly (P < 0.01) on drying. Digestibility is lowest 277 ABC n Figure 7.4: IEF patterns of water soluble proteins in squid dried at 30°, 40°, 50°, 60° and 70°C) (b, c, d, e, f respectively; a being fresh squid). in the fresh meat, improves to 98.3% when dried at 30°C and decreases slightly to 97.8% at 70°C. This is in contrast with the findings of Tanikawa and Suno (1952) who observed that dried squid meat decreases in digestibility from 80.2% on the second day of drying to 60.6% on the 6th day of drying. Flat fish, treated similarly, had higher digestibility of 87.2% on the second day of drying and decreased to 75.5% on the 6th day. The digestibility of the raw material (88%) in this experiment concurs with that obtained by Tanikawa and Suno (1952). One possible reason for the improved digestibility is that the prepared squid meat was held at 3-5°C while awaiting drying. Self proteolysis may have occurred (Sikorski and Kolodziejska, 1986). This proteolysis leading to protein solubilities could have occurred. Squid muscle contain proteinases, active at slightly alkaline pH, which have maximum activity at 60°C (Rodger et al., 1984). All these could have helped to improve the digestibility. Table 7.4: Digestibility of dried squid*. Drying temperature °C % Digestibility Fresh squid 88.05a 30° 98.25b 0 o 96.65b 50° 97.95b 0 cr» o 97.05b 70° 97.75b 279 (e) Total amino acids in dried squid Variations in the total amino acids in the dried squid are seen in Table 7.5. Specific trends are not discernible except for proline, valine and methionine and histidine whose decrease is significant at P < 0.01. It can be concluded that drying did not affect negatively the total amino acids in squid. Table 7.5: Composition of amino acids in dried squid . Drying Temperature °C Amino acids Fresh 30°C 40°C 50°C 60°C 70°C Aspartic acid 9.79 10.75 10.19 10.50 8.53 11.15 Threonine 4.64 5.06 4.48 4.86 4.04 5.04 Serine 4.43 4.87 4.10 4.52 3.73 4.94 Glutamic acid 14.50 15.49 15.23 15.60 15.44 17.20 . * Proline 5.83a 4.86C 3.74d 3.53d 5.12b 4.95C Glycine 5.71 6.76 5.66 5.46 6.30 6.03 Alanine 5.87 6.81 6.84 6.81 6.87 6.57 Cystine 0.93 0.75 0.53 0.96 0.48 0.82 Valine* 5. lla 4.75° 4.70C 4.82b 4.70C 4.89b Methionine 3.76a 13.34C 3.59b 3.65ab 3.41C 3.20d Isoleucine 5.29 3.97 5.20 5.51 4.69 4.08 Leucine 8.11 6.44 8.92 8.91 8.13 7.26 Tyrosine 3.29 3.14 3.76 3.69 3.37 3.05 Phenylalanine 3.30 2.25 3.17 3.52 3.17 2.44 Lysine 8.39 9.12 8.61 7.71 6.64 8.35 Histidine* 2.36b 2.34b 2.50a 2.10c 1.77d 2.02c Arginine 8.68 8.91 8.77 7.85 6.57 8.02 Different superscripts along horizontal column denote significance difference at 99% confidence level. 280 4.1.4.4 Storage studies on dried squid (a) Product appearance Plate 14 shows the appearance of the products stored at various temperatures for 24 weeks. Products which were originally pliable with a creamish white colour became brown, hard and brittle during storage. The products kept at 5°C remained pale until the 16th week. Browning developed in all products kept at other temperatures almost immediately. The tendency for dried squid, dried and rolled squid and shredded squid to undergo non-enzymatic browning has been noted before (Haard, 1982). Haard and Arcilla (1985) identified precursors of Maillard browning in squid: reducing sugars are glucose, ribose, glucose-6-phosphate and Ribose-5-phosphate and the major amino acids: proline and taurine. Homarine, a compound previously shown to accelarate browning in irradiated squid was also identified. The dried products had Aw in the range of 0.20-0.55 which would favour browning more so in the higher range which is nearing the intermediate moisture region. Browning still occurred in the products kept at 25° and 37°C although their Aws are in the lower end of the A^ range favouring browning. Loss of moisture occurs over the storage period RH 50%) causing the texture to become brittle, hard and with the tendency to break transversely. Storage at lower temperatures such as 5°C is therefore preferable for dried squid. 281 Plate 14: Dried squids stored at 5°, 25° and 37°C for (a) 4 wk and (b) for up to 24wk. 282 (b) Colour The development of browning (Table 7.6) in squid was followed by measuring the colour with Hunter Lab colour meter. The L values, measuring lightness, decrease while the a and b values, measuring redness and yellowness respectively increase significantly (p < 0.01) with storage time. Squid gradually change in colour from very light creamish colour to dark brown during storage (Fig. 7.2) and this change was replicated in the Hunter L, a, b readings. Table 7.6: Hunter L, a, b for squid stored at 5°, 25° and 37°C Storage time (wk) Storage temperature °C 0 4 8 12 16 20 24 a ab b b b 5° L 80.45 78.00 74.00 73.05 71.20 73.65 69.90C l lo d a - 0.78^ 0.58 0.42b 0.42b 0.40 1.60° 3.13 CL lo c 4 b 7.85 13.60 15.00 17.05 14.35^ 18.50e 18.05e 0 CL ab b CN o ab aio b L 80.45 78.00 73.05 74.10 73.20 71.65b 69.90 . _ _d a - 0.78* 1.53b 1.36*° 1.96° 2.25^ 2.60d 2.45 a. be. c cd b 7.85 17.65 18.55 19.05 19.00C 19.70 18.30C b b b'c. c < d 37° L 80.45 73.05 70.55 71.25 73.40 69.90 67.35 a - 0.78° 1.37^ 2.01° 2.03° 2.73^ 2.97e 4.17T a be C(* d e b 7.85 17.85^ 18.65 19.55 19.05 20.35 19.90^€ Different superscripts along horizontal column denote significance difference at 99% confidence level. L: Lsd at 1% = 7.1100; a: Lsd at 1% = 0.2532; b: Lsd at 1% = 0.9559 283 (c) Flesh structure An examination of the structure of squid dried at different temperatures was carried out using scanning electron microscopy (SEM). Fresh squid have 'loosely' bound fibres and which were more bound or fused together at higher drying temperatures (Plate 15)• Stanley and Smith (1984) reported that the structural changes when squid meat was heated included closer packing of fibres, fusing of myofibrils and the presence of (presumably) heat denatured protein in the shrunken cytoplasmic core. The tissues of squid dried at higher temperatures were also more difficult to prise apart compared to fresh tissue. Otwell and Hamann (1979) observed five layers of different tissues in squid mantle. Each layer would lose its fibre integrity at different cooking temperature e.g. gelatinization of the inner tunic occurred at 100°C, whilst the outer tunic layer experiences interfibril melting at 70°C which at 100°C would gelatinise into a solid sheet. The degradation of the myofibril unit started at 50°C with a slight loss of myofibril distinction. Degradation of myofibril unit is more obvious at 60°C and the sarcoplasmic proteins appear coagulated in the central cores. The sarcolemma is completely disintegrated at 70°C, the central cores have begun to shrink and dehydration of myofibrils are evident along the outside surface. This conformational and spatial changes in myofibrills would result in loss of moisture from muscle tissue (Buttkus, 1974) leaving a gelled mass of tissue proteins. Changes in 284 Plate fQ : SEM micrographs of fresh squid tissue and those dried at 40° and 70°C (a, b, c respectively). 2 85 the squid mantle during drying at different temperatures can be explained on this basis. (d) Rehydration Rehydratability of squid decreases with increasing drying temperature (Table 1.1). The ratios for samples dried at 40° and 50°C are not significantly different although they decreases at 60° and 70°C. Rehydration ratio is related to freshness, drying conditions and the final degree of dryness. It is determined by the extent of cell damage that has occurred (Villota et ai. , 1980). On rehydration, the elastic texture of the squid was regained. Ke et al. (1984) have proposed minimum limits for rehydration carried out in a 2% sodium bicarbonate solution overnight. Grade A should contain at least 50% moisture, Grade B 40-50% moisture and Grade F less than 40% moisture. Based on this recommendation, samples dried at 30°, 40° and 50°C would pass as Grade A as their weight increases were 62, 53 and 56% respectively. Takahashi and Takei (1956) observed that water absorbed by dried squid showing a high degree of texture recovery upon reconstitution is less available for evaporation and less expressible than the water taken up by samples of dried squid which do not reconstitute well. Thus, these findings suggest that the inter- or -intra-linkages in protein molecules or larger aggregates are related to the water- reversibility of the dried meat (Takahashi, 1965). 286 ★ Table 7. 7: Rehydration ratio for dried squid Drying temperature °C 30° 40° 50° 60° 70° Rehydration ratio 1.62a 1.53b 1.56ab l.llc 1.14c Different superscripts indicate significance (P < 0.01) 4.1.4.5 Chemical properties of stored dried squid (a)' ------Moisture content and Aw As the moisture content and Aw of the stored products were low, 10-20% and 0.20-0.55 respectively, no microbial spoilage should occur. Figures 7.4 and 7.5 respectively show the moisture and Aw changes in the products during storage. The moisture content and Aw of the product stored at 5°C increased slightly over the storage period whilst products stored at 25°C and 37°C lost moisture. Therefore, packaged dried squid of similar moisture and Aw could be kept well over a period of 6 months at all temperatures with no microbial spoilage. However, at 25° and 37°C rejection could be made on the basis of colour and texture. (b) Thiobarbituric acid number (TBA no.) stored squid Although squid is very low in lipid, it is important not only for its nutritional value but also because of its role in taste and aroma. During storage TBA no. increased significantly (p < 0.01) more so at higher temperature (Table 7.8). El-Dashlouty et al. (1984) have reported that 287 M oisture co n ten (%) t 20. Or- Figure Figure 7.4: 7.5: 8 and squid Changes 288 storage Changes 12 37°C. during 16 in at in moisture A 20 storage in 5°, squid 24 25° contegt at and during 5°, 37°C in 25° TBA no. increased on processing of squid (boiling and frying) and have ascribed the increase to partial breakdown of lipoprotein complexes. Despite the increase, products in this work did not become rancid during storage. * Table 7. 8: Thiobarbituric acid number in stored squid . Storage (wk) Temperature °C 0 4 12 24 5° 0.46a 0.66b 1.13c 0.60ab 25° 0.46a 0.62b 0.63b 1.02e 37° 0.46a 0.63b 0.99f 1.15g Different superscripts denote significance at 99% confidence level (c) Total volatile base (TVB) and trimethylamine (TMA) contents in stored squid TVB and TMA contents of dried squid increased on storage, faster at higher temperature (Fig. 7.6 and 7.7). Ke et al. (1984) have proposed a standard for dried squid as follows: Grade A less than 140mg% TVB, Grade B 140-200mg% TVB and 200mg% TVB are unacceptable. Thus, judging by this standard the sample stored at 37°C is no longer acceptable at the 12th week. Although the TVB and TMA values were determined to be high, product quality did not correlate to this. The products would still be acceptable even after 24 weeks of storage if the consumer is prepared to discount the brown colour (Plate 14). 289 at storage on squid (wk) dried time in . Storage content yyg temperatures in different Changes 7.6: Figure a|dui»t 800I/NBui 290 a storage on (wk) squid tim e dried in S to ra g e content TMA temperatures. in Changes different 7.7: igure a|dui«t &00X/N&UJ 291 (d) Solubility of stored squid Storage temperature and duration did not affect solubility of stored squid in KCl at 77°C significantly although a decrease in solubility from 8% to below 5% at 24 weeks was seen (Fig. 7.8). The solubility in SDS + 8 mercaptoethanol was only significantly different (p < 0.01) between the start and the 24th week of storage. Differences in solubility at other intervals were not significantly different. Differences in solubilities in SDS + 8 mercaptoethanol between 5° and 37°C, and 25° and 37°C were significant (p < 0.01) at the 16th, 20th and 24th week of storage. However, the solubilities in SDS + 8 mercaptoethanol are not significantly different between 5° and 25°C at any storage interval. (e) IEF patterns of stored squid Fig. 7.9 demonstrates the effect of storage temperature and time on the electrophoretic separation of water soluble squid proteins. The protein bands are diffuse, more so at higher stored temperature and longer storage time. The intensity of the streaks are also decreased at higher storage temperature and longer storage duration. Extensive changes have occurred to the water soluble fractions which do not separate well electrophoretically. 292 in SDS + P mercaptoethanol Lsd at 15K-7.119 in KCI at 77°C Storage time (wk) Figure 7.8: Solubility of proteins of dried squid stored at different temperatures in SDS :+ 6 mercaptoethanol and in KCI at 77°C. 293 SQUID tTOftEO 3-iD @ A£*jL 3/11/87 r >M M# > 11 Figure 7.9: IEF patterns of water soluble proteins in squid dried at 30°, 40°, 50°, 60° and 70°C (b, c, d, e , f respectively; a being fresh squid). 294 (f) In vitro digestibility The digestibility of the products were high on storage (Table 7.9) and there were no significant changes. Thus, storage time and temperature did not affect the digestibility of the products. * Table 7.9: Digestibility in stored squid products . Storage (wk) Temperature 0°C 0 4 8 12 16 20 24 5° 97.95a 98.60a 97.74a 98.90a 98.60a 98.03a 98.14a 0 CM in 97.95a 97.72a 98.96a 99.27a 97.83a 97.83a 98.90a 37° 97.95a 98.13a 97.95a 98.69a 97.34a 97.41a 97.90a * Same superscripts denote nonsignificance (g) Total amino acids in stored products Variation in the total amino acids occur in the stored products (Table 7.10). No apparent trends were apparent in the effects of storage time and temperature on amino acid levels. However some decreases in threonine, lysine and histidine occur at 37°C. At 25°C, decreases in threonine, alanine, valine, isoleucine and tyrosine are apparent while valine, methionine, isoleucine, leucine and tyrosine decrease at 5°C. Decreased amino acid levels may have arise due to browning reaction during storage. 295 Table 7.10: Total amino acids in stored squid products Storage Temperature °C Amino acids 5 25 37 Storage Time (wk) 4 12 24 4 12 24 4 24 Aspartic acid 9.29 9.77 10.24 9.79 11.08 10.77 10.47 9.82 Threonine 4.22 4.08 4.31 *4.63a 4.63a 4.27b *4.85a 3.93b Serine 4.11 4.63 4.78 4.41 4.97 4.73 4.47 4.37 Glutamic acid 13.76 15.27 16.48 14.35 17.45 16.01 15.71 14.17 Proline 7.23 5.71 7.83 8.61 9.14 6.93 4.02 6.34 Glycine 7.89 9.61 10.93 10.02 10.94 9.09 4.65 7.80 Alanine 7.80 8.03 7.64 *8.51a 7.56b 7.62b 6.62 7.02 Cystine 0.74 0.97 0.82 0.75 0.87 0.98 0.56 0.82 Valine *4.59a 4.34a 3.98b 5.10 3.87 5.08 *5.38a 4.75b Methionine *3.49a 2.92b 2.56b 3.52 2.35 3.62 3.29 3.34 Iso leucine *4.80a 2.93b 2.13c 4.67 1.77 4.62 4.59 4.80 Leucine *7.62a 4.91b 3.54c 8.08 2.90 8.09 7.94 8.27 Tyrosine *3.01a 1.63b 1.25b 3.28 2.40 3.22 3.09 3.34 Phenylalanine 2.30 0.64 2.55 2.54 0.16 2.59 2.96 2.94 Lysine 8.72 8.71 9.44 9.13 9.86 9.00 *8.52a 8.09b Histidine 2.01 5.77 2.19 2.03 2.20 2.07 *7.68a 1.91b Arginine 8.42 10.06 9.31 9.57 9.88 10.29 7.24 8.39 ★ Different superscripts along horizontal column denote significant at p < 0.05 4.1.5 Salting and drying of aged fish It is common practice in many regions of the world to use lower quality fish for drying. Some opinion favours this practice on the basis of product quality. Hence, sardines aged for 1 week and 2 weeks at 5°C were brined and the products examined. 296 4.1.5.1 Effect of fish freshness on salt uptake Salt penetration depends on the condition of the fish and is doubtless affected by such factors as change of tissue structure and viscosity of tissue fluid (Zaitsev et al. , 1969). The salting of aged sardines was undertaken to study this aspect. The 1 week aged fish had red eyes, occasional burst bellies, soft texture and there was some exudate at the bottom of the container. The 2 week's sample was putrid, slimy with burst bellies, soft texture and presence of exudate was noted. The pH at the end of the aging time was 6.4 and 6.9 for the 1 week and 2 weeks aged fish compared with 6.1 for fresh fish. Changes in salt level are shown in Fig. 8.1 and moisture content in Fig. 8.2. It can be seen that moisture loss and salt uptake increased with aging (P < 0.01). Waterman (1976) reported that the staler the fish the quicker the uptake of salt and the greater the weight loss and this is confirmed by data in Figs. 8.1 and 8.2. Roberts (1982) noted that it is more difficult to salt very fresh fish where the flesh is still firm while the flesh of stale fish will be softer and show less resistance to water loss and salt penetration. During aging enzymic hydrolysis and bacterial action increase the pH of its flesh. This has the effect of attracting the negative chloride ions and therefore increases the salt uptake in the fish. The high salt uptake may cause more salt-induced denaturation which further reduces the water holding capacity compared to fresh fish. This would result in more fluid being lost. 297 o A ged sa rd in e(1 (qp) % juajuoD 298 Figure 8.1: Salt uptake by sardine of different degree of freshness during brining. F resh sa rd in e Aged sardine (lwk) )u»)uo3 o »jnjs|ojtf .o 299 % Salting time (h) Figure 8.2: Moisture content in sardine of different degree of freshness during brining. after 8h and 5.9 after 48h salting. pH for the 1 week and the 2 week fish differed significantly (p < 0.05) only at the beginning of salting and was not significantly different during the rest of the salting time. The pH for the fresh sardine also decreased during the first 8h of salting and then remained relatively constant for the remainder of the process. The pH of the fresh sardine was significantly different (p < 0.01) to the 1 week and the 2 week aged sardines during the first 8h of salting and at the 36th hour. The pH decrease in the aged fish is attributed to leaching of bases into the brine. For the fresh fish the pH decrease is not reflected in a fall in amine contents (Figs. 8.7 and 8.8). On the contrary the amines increased up to 12h salting after which they began to decrease again. However, any alteration to the state of protein caused by the salt would affect its water holding capacity and its pH. (b) Total volatile bases and trimethylamine content The TVB and TMA contents of both the aged fish were much higher than in the fresh fish at the start of the salting procedure. The TVB and TMA for the 1 week aged sardine were 69 and 65mgN/100g respectively and for the 2 week aged fish were 104 and 94mgN/100g respectively compared with 35 and 17mgN/100g respectively for fresh sardine. Haaland and Njaa (1988) observed higher development of TVB and TMA, 116 and 53mg/g total N respectively, in capelin stored at 5-7°C. However, on brining the TVB and TMA contents of both aged 300 4.1.5.2 Drying of the aged fish The 1 week and 2 week aged fish dried most rapidly at 40°C (Figs. 8.3 and 8.4). The rates at 40°C in both cases are significantly higher (p < 0.01) than at all other temperatures. The drying rates at 40°C for both the aged fish are also significantly higher than for the fresh fish (Fig. 8.5). At higher temperatures, the faster diffusion of water brings the salt to the surface of the fish forming a barrier to evaporation of water thus slowing down the rate of drying. At 40°C there must have existed some differential between the migration of salt to the surface and the removal of water vapour, thus creating a favourable rate of drying. After salting, the aged fish contained less moisture than the fresh fish (Fig. 6.2) and the proteins of the aged fish have lesser water holding capacity due to enzymic and microbial action allowing higher rate of drying in the aged fish. 4.1.5.3 Chemical properties of salted aged fish (a) pH After 1 week and 2 weeks at 5°C in the refrigerator the pH of sardines was 6.4 and 6.9 respectively (Fig. 8.6) increasing from 6.1 in the fresh fish. The products of autolysis or microbial degradation contributed to this rise in the pH. These products include the amines which form in the aged fish. During salting the pH for the aged fish decreased. The pH for the 1 week fish was lowered to 6.1 301 d 00 o- —0 40 —o 50 302 © t - - © o eo o Figure 8.3: Effect of temperature on drying rates of salted aged (1 wk) sardine 0 .8 o co o u r- © ? o v ? o o uu a o ?? o 'O ! o v£> o ° © in *o * ^ - II (M-uipB/o o kO 3 HB) 303 d *»»»J 8 co d u |A jq cvi o o •mm *>> £ OJ c Figure 8.4: Effects of temperature on drying rates of salted aged (2 wk) sardines.. F resh sa rd in e D rying tim e (h) Figure 8.5. Effect of freshness of sardine on drying rates at 40°C. CD C •r— e •r— JO cn c •r— E 13 -o CO DO CO (U c C -C CO 4— 4-> c O) i- O) 4- 4— 4- O .c CO 4- O rc Q. CD 00 305 fish followed similar patterns with rapid decrease during the first 8h with subsequent slower decline to the end of the salting period (Figs. 8.7 and 8.8). Changes in TMA and TVB for fresh sardine have been discussed in Section 4.13.5.1(b). It is apparent that TMA and TVB levels in the aged fish were substantially reduced during brining. Production of TVB and TMA may still occur but leaching of the bases into the brine more than compensates for this. The growth of TMA producing organisms was depressed by salting, more so at the lower temperatures (Nozawa et al. , 1979). In this work the salting was conducted at 30°C and hence suppression of TVB and TMA production may not have been complete although such suppression was noted by Bilinski and Fougere (1959) and Cardin et al. (1961). (c) Solubility of proteins in aged fish Figs. 8.9, 8.10 and 8.11 show that protein solubility decreases with the aging of sardines. In KC1 at 2 5°C protein solubility at the initial stage of salting was higher in the aged fish (28 and 29% in the 1 week and the 2 week respectively) than in the fresh fish (21%). Protein solubility in this solvent was similar for both the 1 week and the 2 week aged fish but that for the fresh fish was significantly different (p < 0.01). After 4h salting the solubility for the 1 week and the 2 week sardines (14.5 and 13.5% respectively) was lower than that for fresh fish 306 brining. during freshness different of fish in u. < < content TVB 8.7: Figure 9|duivs Booi/lSjBiu 307 brining. during V oo g CM JS ■«-* 00 freshness c s ICQ different of fish in content TMA 8.8: t 13 cn jlj" 308 F resh 1 wk ujajojd a|qn|OS 309 % Figure 8.9: Effect of brining time and fish freshness on the soluble protein Fresh sardi A ged sardir 310 -.00 Figure 8.10. Effect of brining time and fish freshness on the soluble proteins of sardines in KC1 at 77°C. soluble . the on (h) tim e freshness mercaptoethanol 3 & fish Brining and SDS in time sardine brining of of Effect proteins 8.11: Figure u|»)Oid »|qn|OS % 311 (18.5%). This was so until the 48h salting time. After 12h salting the solubility of the 1 and 2 weeks aged fish was not significantly different but was significantly lower than for the fresh fish. On the whole protein solubility decreased with salting time but for both the aged fish the solubility stabilised after 24h and remained constant up to 48h salting. In KC1 at 77°C, fresh sardine protein was most soluble during salting. Initially the 1 week aged fish was more soluble (23%) than the 2 weeks aged fish (15.5%) but was not significantly different during salting. Protein solubility decreased during slating, rapidly in the first 4h and gradually afterwards, regardless of the freshness of fish. In SDS + B mercaptoethanol the initial solubility of proteins did not differ significantly between the fresh and aged fish. The solubility of proteins of the fresh fish fell rapidly in the first 4h from 62.5 to 49.5% and then gradually decreased to 34.5 after 48h salting. The solubility of proteins for the 1 week aged fish remained about the same for the first 4h of salting (61 to 60.5% ) and then fell rather rapidly till 12h salting time (43%) and afterwards gradually to 37% at the end of salting. The 2 week fish fell rapidly till the 24 h salting from 60 to 37% and remained relatively constant to the end of salting. 312 Expected page number is not in original print copy sard in (ft C lu>ea>k) 2-lo CfiAuTtfP 0"l><) 5T0 ofv O. 3lk *o «k 4 tfk £ H 3 I 34k ^ Hk V/ A SC J) g p ^ w Figure 8.12; IEF patterns of water soluble proteins of lwk aged fish on being brined for 0, 2, 4, 8, 12, 18, 24, 36 and 48h (a, b, c,d, e, f, g, h and i respectively). 314 AaSD £AfU)iN&S C% U/CtJLs) pi* 2-10 CiA.r-D MVLY) B cc b e t Gr Figure 8.13: IEF patterns of water soluble proteins of 2wk aged fish on being brined for 0, 2, 4, 8, 12, 18, 24, 36 and 48h (a, b, c, d, e, f, g, h and i respectively). 315 Table 8.1: Composi+tion of amino acids in aged sardines g/16gN. Amino acids Fresh 1 week 2 week sardine aged sardine aged sardine Aspartic acid 9.00 7.68 9.40 Threonine 4.54 4.43 5.25 Serine 4.14 3.52 4.10 Glutamic acid 1 1.62 12.24 12.73 Proline 9.27 8.33 5.80 Glycine <6.24 5.33 6.30 Alanine <3.57 9.01 7.97 Cystine t0.87 1.06 0.76 Valine 6.13 6.51 6.36 Methionine 3.13 3.26 2.54 Isoleucine 4.66 5.31 4.75 ★ Leucine 7.84a 7.68a 7.26b * Tyrosine 3.02a 2.34b 2.36b Phenyl alanine 2.88 2.67 3.00 * Lysine 8.113 7.68b 7.52b Histidine+ 4.22a 3.93b 1 .35C Arginine+ 6.13a 8.98b 12.78c ★ Amino acids showing decreasing trend; Different superscripts indicate significance at 99% confidence level. for 2 weeks at 5°C caused extensive changes in the total amino acid content in sardine. Table 8.2 shows the amino acid content in aged sardine salted for different times. Levels of the amino acids fluctuated during brining and there appears to be no pattern of amino acid content vs. salting time. However, in the 1 week aged fish, aspartic acid and serine increase 316 significantly (p < .0.01). In the 2 week aged fish, serine, alanine, valine and lysine are seen to increase significantly (p < 0.01) whilst aspartic acid, threonine, histidine and arginine declined (p < 0.01) during salting. Table 8.2: Total amino acids in salted aged sardines (g/16 gN)* 1 week aged sardine 2 weeks aged sardine Amino acids Salting Time (h) 0 24 48 0 24 48 Aspartic acid 7.68 8.22 10.33 *9.40a 8.12b 3.84C Threonine 4.43 5.24 5.01 *5.25a 5.07a 5.04b Serine+ 3.52a 4.13b 4.32b 4.10 4.24 4.79 Glutamic acid 12.24a 10.26b 10.10b 12.73 8.81 9.93 * Proline 8.33a 3.91b 3.03C 5.8 4.95 5.59 Glycine 5.33 4.65 4.75 6.30 5.01 5.70 Alanine 9.01 9.83 7.27 7.97 10.05 11.05 Cystine 1.06 0.66 0.84 0.76 0.61 0.89 Valine 6.51 6.63 5.86 6.36 6.66 7.84 Methionine 3.26 3.57 3.06 2.54 3.70 1.85 Isoleucine 5.31 6.18 4.40 4.75 6.41 6.28 Leucine 7.68 3.33 7.08 7.26 8.79 9.04 Tyrosine 2.34 3.71 2.94 2.36 3.76 2.69 Phenylalanine 2.67 3.47 2.49 3.00 3.81 3.50 Lysine 7.68 9.04 8.75 +7.52a 8.66b 10.62c Histidine 3.93 3.25 3.35 *1.35a 0.91b 0.56C Arginine* 8.98a 8.93a 6.37b * 12.78a 10.40b 10.62b * observed decreasing trend: Different superscripts indicate significance at 99% confidence level. 317 4.1.5.4 Product quality (a) Product comparison The 1 week aged fish dried at 40°C gave a product comparable to that from the fresh one. It dried very well, the colour was acceptable but the appearance had flaws in that some of its skin had flaked off and its belly was open. Products dried at higher temperatures developed strong yellowish brown colour around the head, belly and along the ventral side to the tail region. Products dried at higher temperatures had more salt appearing on the head and tail surfaces. The 1 week aged sardine dried at 40°C was acceptable on the basis of taste and aroma as it possessed that tangy taste and aroma typical of dried fatty fish. The dried fish prepared from fresh fish lacked this taste immediately after drying although its flavour developed on storage. The dried products prepared from the 2 week aged fish was not acceptable. The product appearance was poor with burst bellies, blotchy skin, salt deposits and yellowish brown colour. The 2 week aged fish was rancid at the outset. It possessed a slightly bitter taste, was quite wet in texture and when pressed between two fingers resembled a paste. Table 8.3 summarises the analyses of the dried products of the fresh and aged fish. The desired moisture content could be achieved by controlling the duration of drying, although 318 it also depends on other factors such as air speed and the relative humidity. However, Aw achieved was 0.77 for the dried products originating from fresh fish, 0.69 for 1 week aged fish and 0.74 for the 2 week aged fish. Thus, the 1 week aged fish would keep best except for incipient rancidity. The dried products from the fresh fish could be dried further to reduce its moisture content and thus A...w The total volatile bases and trimethylamine are higher in the aged fish than the fresh one. The product from fresh fish did not smell rancid whereas the 1 week aged fish already harboured some very slight tangy aroma associated with rancidity and 2 week aged samples were totally rancid. The TBA number for the fresh fish was 0.9, the 1 week aged fish, was 1.7 and the 2 week aged fish was 3.1. The protein solubility of the three products were compared. Products from fresh fish were found to be the most soluble in all the three solubilising media. The solubility of the aged fish are as given in Table 8.3. All the three products were found to have high digestibility. The digestibility of the 2 week product was the lowest of the three, being 96.4% as compared to 97.9% for the 1 week aged fish and 97.5 for the fresh one. Thus, aging of fish did not substantially affect its digestibility. 319 Table 8.3: Analyses of fresh and aged fish brined for 8 h and dried at 40°C. Analyses Fresh 1 week 2 week fish aged fish aged fish Moisture content (%) 38.4 22.3 39.7 Salt content (% db) 19 26.9 31.1 Water activity Aw 0.77 0.69 0.74 Total volatile bases 57.0 67.7 70.3 (mgN/lOOg) Trimethylamine 52.9 50.9 59.2 (mgN/lOOg) Thiofcxjrbituric 9.0 1.7 3.1 (acid number) Solubility in KCl at 16.5 11.7 10.9 (% soluble protein) Solubility in KCl at 77°C 18.6 12.5 13.2 (% soluble protein) Solubility in SDS + 47.5 41.2 45.5 B mercoptoethanol (% soluble protein) % digestibility (in vitro) 97.5 97.9 96.4 (b) Isoelectric focussing In Fig. 8.14 it is seen that some protein bands are still discernible in the IEF patterns of the aged fish on drying. Bands B, D, E, F, G and H are still quite distinct in the patterns of the 30° and 40°C dried samples. With increasing drying temperature the intensity of each band decreases. 320 I CL b d Figure 8.14: IEF patterns of water soluble proteins of lwk aged fish, brined for 8h and dried at 30°, 40°, 50° and 60°C (c, d, e and f respectively; a = lwk aged fish unsalted, b = lwk aged fish brined for 8h). 321 Only bands D, F and G are faintly detectable after drying at 50°C and above. The IEF bands of the 2 weeks aged fish are less distinguishable than those of the 1 week aged fish (Fig. 8.15). The bands are more diffuse at all drying temperatures and lose intensity with increasing drying temperature. Only anodic bands survived drying. 4.2 Differential Scanning Calorimetry (DSC) The effects of salting and drying on the thermal properties of fish protein were studied by using DSC. Fig. 9.1 shows the thermograms of salted and unsalted shark muscle tissue. Shark has two distinct endothermic peaks in its thermogram, the first and the smaller of the two displays a peak maximum (Tmax) at 50° + 2°C and the second and larger peak has a Tmax at 72° + 2°C. These represent thermal denaturation of the proteins. Peak maximum temperature was recorded because it is a more distinct part of the endothermic peaks than either onset or conclusion temperatures of the thermal transition and can be measured more precisely. The thermogram lacks the basic three peak profile previously observed for fresh fish (Poulter et al. , 1985) and mammalian meat (Wright et al., 1977). The three peaks seen on a typical muscle thermogram are due primarily to transitions involving myosin, actin and possibly sarcoplasmic proteins. The first peak is due to the transitions at Tmax 60° involving myosin and may also be related to the less stable sarcoplasmic proteins. The second maybe due to the sarcoplasmic protein transition at Tmax 67°C and 322 b c <* e £ Figure 8.15: IEF patterns of water soluble proteins of 2 wk aged fish, brined for 8h and dried at 30°, 40°, 50° and 60°C (c, d, e, and f respectively, a = 2wk aged fish; b = 2wk aged fish brined for 8h). 323 h e a t Fi gure 9.1: 20 ° muscle DSC 40 ■ * thermogram ______ . 60 I ^ ______ Temperature of 80 i __ salted ^ ______324 100 (®C) and i Fresh ______° unsalted , unsalted 120 i ° shark 140 i ° whole 160 » ° the third is primarily due to actin transition at T" 80°C (Wright et al. , 1977). There are fish species such as tilapia which display only two distinct peaks (Poulter et al., 1985). Davies et al. (1988) working on intact muscle of cod and snapper also noted that Peak II on the typical thermogram was not always present. Thus in the thermogram of shark in Fig. 9.1 both myosin and sarcoplasmic proteins are assumed to cause the peak at 50°C and actin to cause the peak at 72°C. Poulter et al. (1985) and Davies et al. (1988) observed the latter peak to have Tmax at about 73°C and the myosin peak at 41°C in cod. As the samples were frozen before analysis, the freezing and frozen storage could have affected the thermograms of these samples. Hastings et al. (1985) observed that myosin suffered partial denaturation on freezing and found a 30% decrease in peak area of the low temperature (myosin) peak. A decrease in the Tmax of myosin on freezing was observed by Poulter et al. (1985). Frozen storage however, does not seem to affect collagen and actin in the fish (Hastings et al., 1985). The thermogram of unsalted morwong is shown in Fig. 9.2. Morwong also exhibits two peaks, the first has a transition temperature T_at 46° + 2°C and the second, larger, peak has a T _ at 73° + 1°C. Sardine on the other hand shows a very distinct two peak thermogram 50° + 1°C and 80° + 1°C but there is an inflection at 32°C (Fig. 9.3) which may represent the transition of collagenous material as the sardine sample was macerated whole. Previous reports (Finch and Ledword, 1972; 1973; Menashi et al., 1976; Poulter et al., 1985) have noted that the onset temperature (TQ) 325 Endothermic h ea tflow Figure 9.2: whole DSC thermogram muscle. Temperature of 326 salted (°C) and unsalted Fresh,unsalted morwong of the collagen denaturation peak ranges from 31-40.3°C. Species differences have led to differences in the transition temperatures especially the myosin fraction. Attempts were made to relate these differences to habitat temperature (Poulter et al., 1985; Davies et al. , 1988). Heating rate, protein concentration and sample size are all known to lead to variation in Tmax (Wright and Wilding, 1984). Keeping these variables as constant as possible, the values obtained should be mutually consistent but not necessarily comparable with other published data. In the three species examined in this study, the first peak transition (myosin) is about 10°C below that in rabbit (Wright et al., 1977). This 10° difference was also noted by Hastings et al. (1985). However, the second transition temperature for sardine, 80°C (Fig. 9.3) was higher than for the other fish and published data (about 71-73°C). 4.2.1 Effect of salting on the thermogram of fish The effects of salting on thermogram of fish proteins are also shown in Figs. 9.1 - 9.3. In shark (Fig. 9.1) transition temperature of the second, larger, peak (actin) decreased with increasing salting time from 72°C in the unsalted sample to 64° + 4°C in the 8h salted sample and 45° + 2°C in the 36h salted samples. Secondly, at longer salting times, peak broadening was noted. The peak temperature of the first transition (myosin) also decreased to 46°C and in the longer salted samples it disappeared altogether. The salt content in the unsalted, 8h and 36h samples were 3.1, 48.3 and 56.1% (db) and this no doubt has a major role in changing peaks. Quinn et al. (1980), working with 327 Endothermic h e a t Figure 9.3: whole DSC thermogram muscle. of 328 salted and Fresh,unsalted unsalted sardine beef and rabbit muscles, observed similarly and proposed that ionic strength effects caused the changes observed. Stabursvik and Marten (1980) also showed that salt depressed transition temperatures in beef muscle suggesting that the muscle protein structure was destabilised. Cl appear to be a particularly efficient destabilising agent on actin molecules (Weinberg et al., 1984). In Fig. 9.2 the effects of salting time on morwong protein are illustrated. Transition temperatures in both peaks are decreased (from 73° to 70°C for the first peak and 46° to 42°C) and peak broadening is also noted in the 4 and 12h samples. However, thermograms of the longer salted samples of morwong were poorly reproducible (Figs. 9.4 and 9.5) in terms of peak areas and temperatures. This is attributed to inhomogeneity of samples in terms of both moisture and salt contents. Since the sample pans for DSC hold only about 10 mg, sample inhomogeneity represents a major potential source of error. Hargerdal and Martens (1976) have demonstrated that thermal transitions of protein are dependent on moisture content. Non-uniformity of the tissue and non-uniform salt distribution through the sample could contribute to the variations observed. Similar problems are encountered with longer brined sardines (Fig. 9.6). After brining sardines were macerated whole. The Tmax t*ie major peaks were 80° and 65°C in the unsalted and 12h salted samples (Fig. 9.3) respectively decreased substantially ranging from 41° - 55°C in the 24 - 48h samples (Fig. 9.6). 329 Endothermic heat flow Figure 9.4: DSC thermogram Temperature 330 of 24 (®C) h 100° salted 120° morwong. 140° 16 (P Endothermic h e a t ------Figure 20 9.5: ^ DSC 40 ^ thermogram Temperature 60 ® of 331 80 36 ® (®C) h salted 100 ® morwong. 120 ® 140 ° Endothermic heat Figure 20 9.6: I 5 TOO DSC i ______thermogram 60 i Temperature 5 of 80® I 24 332 ______ h (°C) and 100° t 24h 48 ______ h 120° salted i 140° sardine. i 160? i However, poor reproducibility of the thermograms make these temperatures suspect. The peaks in the sardine thermograms were broader in the 12h sample than in the unsalted one (Fig. 9.3). Weinberg and Regenstein (1984) observed that peak broadening and a decrease in the total area on treating fish muscle with salts was related to the enthalpy of the transition. Table 9.1 shows changes in enthalpy of denaturation ( A HD) in fish on salting. A HD for shark was found to be 18.85 + 1.88kJ/g, 13.17 ± 1.17kJ/g for morwong and 20.31 + 0.88kJ/g for sardine. These values are probably species specific and depend on moisture and protein content and the cons^istuent proteins. Generally, decreasing enthalpy is observed with increasing salting time. The A HD for samples brined >12h were not calculated for morwong and sardine due to poor reproducibility of the thermograms. However, in shark samples A HD reduced substantially with increasing salting time. The decrease probably indicates the extent of salt denaturation. By analogy, Karmas and Dimarco (1970) examined salt extracted protein by DSC. They found no peaks in the thermogram and concluded that denaturation had taken place. The observations in this DSC study is consistent with the decline in protein solubility and decreases in intensity and number of protein bands in the IEF patterns with increasing salting time. 333 Table 9.1: Changes in A hd during salting Shark Morwong Salting Salt Total Salt Total Salt Total time content A HD content A HD content A hd (h) (% db) kJ/g (% db) kJ/g (% db) kJ/g 0 3.1 18.85 5.7 13.17 4.7 20.31 4 43.6 9.66 28.1 9.32 20.8 15.72 8 48.3 4.81 43.5 8.28 26.4 8.49 12 45.6 8.32 42.4 7.02 29.1 8.15 24 47.1 0.21 36 56.1 0.75 48 50.2 no peak The nature of the denaturation was not examined. However, according to Quinn et al. (1980) actin is the protein in beef and rabbit most destabilised by the ionic strength effect. Although they noted no marked change in the first peak (myosin) in this study the myosin peak altered slightly in Tmax and decreased in area, eventually disappearing with increasing salting time. Davies et al. (1988) also observed a decrease in shift in T and area of the myosin peak in snapper and cod due to ionic effects. Myosin of fish was found to be less stable than that of mammalian species (Connell, 1961). Myosin is also more labile than the other proteins (Howgate and Ahmed, 1972). In the three species examined the first peak disappeared after 8-12h salting. The salting concentration at this time varied between 26.4% in sardine to 48.3% in shark. The rapid fall in A during the 334 first 4h salting of shark is probably due to rapid salt uptake in that species. However, for the species at similar salt content Hd differed in accord with Connell (1961) who observed that the degree of protein aggregation differs greatly amongst fish species. 4.2.2 Effects of drying on DSC thermogram of fish Figs. 9.7-9.10 show the thermograms of morwong, sardine, shark and squid after drying. In both morwong and sardine (Figs. 9.7 and 9.8) the major peak or peak II observed in the unsalted fish at Tmax 73°C for morwong and at 80°C for sardine persist after drying at temperatures from 30°-70°C. The first peaks in unsalted morwong thermogram with Tmax 46°C and 50°C in sardine persist in both samples dried at 30°C but is absent in the thermogram of fish dried at higher temperatures. The remaining actin peak in the thermogram of the morwong and sardine decrease in size with increasing drying temperature. In Fig. 9.9, the thermogram of shark shows that peak I completely disappeared on drying. Peak II remains and is clearly seen in the sample dried at 50°C. At drying temperature above 50°C, the peak broadens and decreases in area. The squid thermograms show similar behaviour (Fig. 9.10). Fresh squid thermogram shows two small peaks at 37° and 43°C and a major one at 80°C. The first two peaks are presumably myosin and sarcoplasmic protein followed by actin. Stanley and Hutlin 335 Endothermic h ea th ea t Figure 20^ i 9.7: Jo® DSC dried I ______ thermogram at 6 o i different O Temperature of 80 i 336 dried s temperatures (®C) 100° salted i Fresh ______ morwong, 120<> . t 140° previously * 160° * Endothermic h eat Figure 20 9.8: ° DSC dried 40* 5 thermogram at 60 different Temperature ® of 80 337 dried ® temperatures (®C) 100 salted ® Fresh sardine, 120 . ° 140 previously ° I6(f Endothermic h eat flo w Figure 9.9: DSC at different thermogram Temperature temperatures of 338 dried (°C) shark, . Fresh previously dried h ea tflo w Figure 9.10: 40 dried DSC ° thermogram Temperature at 60 different 339 5 80 of (®C) ° dried temperatures. squid previously Fresh reported the denaturation temperature for actin to be between 76- 7 8°C. Matsumoto et al. (1981) reported a denaturation temperature of 80° - 82°C for actin. The first two peaks disappear at all drying temperatures leaving only the actin peak which broadens and reduces in area as temperature of drying increases. Parsons and Patterson (1986) observed similar behaviour for previously heated beef samples. Hamm (1966) states that changes in myofibrillar proteins start at approximately 50°C whereas denaturation of the sarcoplasmic proteins starts at lower temperatures. Coagulation starts at 65°C while collagen shrinks at temperatures around 63°C and gelatinised at higher temperatures. The second transition temperature is mainly due to a change in the water structures of the water-protein system (Karmas and DiMarco, 1970). Judging by the shape of the thermograms at the same drying temperature for all the species examined in this study, shark and squid seem to have experienced more severe effects of heat denaturation than either morwong or sardine. This is confirmed by the enthalpy (A. HD) data in Table 9.2. Even though the samples contained different moisture contents decreasing HD reflects the energy changes associated with protein denaturation during drying. Although squid dried faster at 50°C than the other fish species, the effect of heat treatment on squid was severe than in the other species. Squid textural qualities change rapidly with increasing cooking temperatures up to 60°C (Otwell and Hamann, 1979b) . This was associated with sarcoplasmic protein coagulation, destruction of 340 myofibrils (Otwell and Hamann, 1979a) and thermal gelatinisation of the large amount of connective tissue in squid (Stanley and Smith, 1984) which could contribute to the components in the muscle may also be important in leading to different levels of stability. Lipids present in the fish muscle which could bind to and consequently stabilise actomyosin (Hamada et al. (1982). Parsons and Patterson (1986) noticed a correlation between maximum heating temperature and the onset of denaturation and that length of heat treatment is reflected in peak areas of the thermogram. However, in this study it seems that heating temperature is a major influence on the thermogram. Drying at higher temperatures gave smaller peak areas even though drying rate is faster at higher temperature and was conducted for shorter times. Duration of heating must affect the thermograms although this aspect was not investigated. Table 9.2: Changes in ethalpy of denaturation ( HD) during drying Morwong Sardine Shark Squid Drying Moisture HD Moisture HD Moisture HD Moisture HD temperature content kJ/g content kJ/g content kJ/g content kJ/g (°C) (X ) (% ) (% ) (% ) Fresh 80.7 13.17 74.4 19.19 79.4 18.85 80.8 16 30° 47.2 10.32 47.9 11.54 44.5 3.47 13.5 4.05 50° 29.6 9.74 22.7 8.74 34.1 1.09 10.7 2.17 70° 11.9 6.90 22.2 5.94 37.7 0.21 14.9 0.25 341 4.3 Analyses of Commercial Dried Products In Table 10.1, the composition of morwong, shark, sardine and squid products were compared with those of a range of commercial South East Asian dried fish products. Protein, fat and salt levels are given on DMB with other data on an as is basis. The anchovy and barracuda, apparently not brined, were lowest in salt, moisture content and water activity (Aw)• Amongst the prepared products squid was the lowest in moisture and salt content and A w . Shark was highest in salt and moisture content followed by morwong and sardine. Amongst the commercial samples yellow croaker and milkfish contained the highest moisture and salt contents. Salt content in shark and morwong were higher than in the commercial samples. Aw in the commercial samples ranged from 0.55 to 0.75 while the Aw for the prepared samples range from 0.31 - 0.74. Fat content in the commercial samples ranged from 2.0% in yellow croaker to 18.5% in barracuda whilst it ranged from 5.0% in squid to 12.2% in sardine. TBA numbers were relatively low for all samples except for yellow croaker (9.4). TBA numbers did not correlate with fat content. Despite this both mergui and yellow croaker had noticeable rancid odours, the prepared samples did not exhibit any rancid smell at all. 342 Table 10.1: Analyses of Commercial and Prepared Dried Fish. Fish A Moisture Protein Fat Salt TVB TBA (species) w content content content (mgN/ number (mg <%) (%DMB) (%DMB) (%DMB) 100 g malonaldehyde fish) per 100 g fish) Mergin' 0.67 33.0 78.3 2.3 21.2 328.0 2.6 (Polynemus indicus) Yellow croaker 0.75 42.4 71.9 2.0 23.7 162.8 9.4 (Pseudosciaena crocea) Milkfish 0.71 42.4 62.5 13.7 25.2 325.0 1.8 (Chanos chanos) Anchovy 0.63 14.2 82.2 5.8 4.0 427.0 0.7 (Stolephorus spp.) Barracuda 0.55 9.4 73.3 18.5 5.0 141.3 1.7 (Spyraena spp.) Morwong+ 0.74 29.6 62.1 7.6 40.2 52.5 0.3 (Nemadactylus macropterus) Sardine+ 0.67 22.7 58.7 12.2 17.7 45.7 1.2 (Sardinella pi Ichardus) Squid+ 0.31 10.7 79.4 5.0 9.3 39.0 0.46 (Nototodaarus gouldi) . + Shark 0.71 34.1 64.0 8.9 44.3 89.4 0.25 (Notogaleus rhinophares) o w + this denote that these products were prepared: brined at o and dried at 50°C . TVB nitrogen ranged from 141 - 430mg/100g fish in commercial samples compared with 39-90mg/100g in the prepared samples. The latter products were prepared from fresh fish and were analysed soon after drying. However, the processing and storage history 343 of the commercial products is unknown. It should also be noted that lower quality fish is often used for drying in South East Asia whereas better raw materials are used for canning and boiled fish production. Despite their high TVB levels the commercial products were regarded as acceptable. Levels of 100-200mg/100g fish of TVB nitrogen have been reported previously in a number of salted and dried fish products (Connell, 1975). Because of its wide range and high levels in acceptable products of this type, TVB is of little value as a quality indicator in dried fish. Overall, products from morwong, shark, sardine and squid produced in the current study compare favourably with the commercial imported products in terms of analytical parameters. However, their comparative sensory acceptability, a vital criterion of quality, was not evaluated. 344 CHAPTER 5 CONCLUSION Different species demonstrated different rates and maxima of salt uptake during bringing. However, no benefits were observed in lengthening the salting time beyond that required for maximum uptake, namely 8h for morwong and sardine and 4h for shark. The disadvantages of longer brining were demonstrated clearly in the poor appearance of salt encrusted products, evidenced visually as well as by SEM studies and poor rehydration properties. Drying temperatures between 30° - 70°C were applied to products salted for their optimum time. Squid was also dried without salting. A drying temperature of 50°C was found to be the best compromise between drying rate and product quality. Overall effects of salting included a decrease in soluble proteins, a decrease in intensity and number of IEF bands and pronounced changes to the flesh structure as shown by SEM. Salting did not adversely affect either in vitro protein digestibility or amino acid levels. Drying had similar effects on the above parameters. Protein denaturation during both salting and drying was also exemplified in DSC studies of thermograms of the fish samples. During salting there were also changes in pH and volatile bases while during drying increase in volatile bases was also observed. 345 Salting and drying of sardines which had been aged showed that although salt uptake was faster the quality of final products was inferior to those from fresh sardines. The products obtained from the above salting and drying procedures had sufficiently low Aw to ensure microbial stability, however, deterioration via browning reaction and oxidative rancidity limited storage life. Uptake of water during storage would render the products susceptible to microbial spoilage. Obviously storage life was enhanced at lower temperatures. The acceptability of freshly prepared products was limited by higher salt levels, brittle texture and browning reaction products. In the products obtained using minimum brining times to reach maximum salt level and a drying temperature of 50°C none of these factors were found to be a problem. Thus, on the basis of acceptability and nutritional considerations the above salting and drying conditions are recommended for production of salted dried fish products. 346 REFERENCES 347 Acker, L. (1969). Water activity and enzyme activity. Food Technol. 23(10) :27-40 . Ackman, R.G. (1974). Marine lipids and fatty acids in human nutrition. In: Fishery Products Kreuzer R. ( ed.). Fishing News (Books) Ltd., London. Ackman, R.G. and Eaton, C.A. (1971). Mackerel lipids and fatty acids. Can. Inst. Food Technol. J. 4:169-174. Ackman, R.G., Eaton, C.A. and Hingley, J.H. (1976). Menhaden body lipids: details of fatty acids in lipids from an untapped food resource. J. Sci. Food Agric. 27:1132-1136. Adachi, R.R., Sheffner, L. and Spector, H. (1958). The in vitro digestibility and nutritional quality of dehydrated beef, fish and beans. Food Research 23:401-406. Adams, R., Farber, L. and Lerke, P. (1964). Bacteriology of spoilage of fish muscle. II. Incidence of spoilers during spoilage. Appl. Microbiol. 12:277-279. Addison, R.F., Ackman, R.G. and Hingley, J. (1968). Distribution of fatty acids in cod flesh lipid. J. Fish. Res. Bd. Can. 25:2083- 2090. Ah-Weng, P., Hanson, S.W. and McGuire, K.J. (1985). Water activity data in relation to quality loss for Southeast Asian cured fish. Spoilage of tropical fish and development. Proc. of a Symposium Held in Conjunction With the Sixth Session of the Indo-Pacific fishery Commission Working Party on Fish Technology and Marketing. Royal Melbourne Institute of Technology, Melbourne, Australia, 23-26 October, 1984. FAO Fish. Rep., 317 suppl. :306-315. Aitken, A., Jason, A.C., Olley, J. and Payne, P.R. (1967). Effects of drying, salting and high temperatures on the nutritive value of dried cod. Fishing News International 6 (9):42-43. Aitken, A.D. and Connell, J.J. (1979). Fish. In Effects of heating on foodstuffs, Priestley R.J. (ed.). Applied Science Publishers Ltd. London : 219-254. Amano, K., Yamada, K. and Bito, M. (1963a). Detection and identification of formaldehyde in gadoid fish. Bull. Jpn. Soc. Sci. Fish. 29 (7):695-701. Amano, K., Yamada, K. and Bito, M. (1963b). Contents of formaldehyde and volatile amines in different tissues of gadoid fish. Bull. Jpn. Soc. Sci. Fish. 29 (9):860-864. Amerine, M.A., Pangborn, R.M. and Roessler, E.B. (1965). Principles of Sensory Evaluation of Food. Academic Press Inc., New York. Anderson, M.L. and Mendelsohn, J.M. (1972). A research note: A rapid salt-curing technique. J. Fd. Sci. 37:627-628. Anon. (1973). Expanding the utilisation of marine fishery resources for human consumption. FAO Fisheries Report, No. 175. Anon. (1983). CAP is the fashion to follow. Fish. News Int. 22 (2) :58 . 348 Arai, K. and Kinumaki, T. (1980). The retardation of growth by taking oxidised lipids together with BHA decomposition products. Bull. Tokai Reg. Fish. Res. Lab. 102:25-29. Arntfield, S.D. and Murray, E.D. (1985). The influence of processing parameters on food protein functionality. II. Factors affecting thermal properties as analysed by differential scanning calorimetry. Can. Inst. Food Sci. Technol. J. 18:158-162. Awad, A., Powrie, W.D. and Fennema, 0. (1969). Deterioration of fresh-water whitefish muscle during frozen storage at -10°C. J. Food Sci. 34:1-9. Azeza, N.I. (1982). Quality assessment in processed fish from Lake Chad destined for Southern markets. Proceedings of the FAO Expert Consultation on Fish Technology in Africa, Morocco, 7-11 June, 1982. FAO Fisheries Report No. 268. Bacon, C.W., Sweeney, J.G., Robbins, J.D. and Burdick, D. (1973). Production of penicillic acid and ochratoxin A on poultry feed by Aspergillus ochraceus: Temperature and moisture requirements. Appl. Microbiol. 26:155-160. Balaban, M. and Piggot, G.M. (1986) . Shrinkage in fish muscle during drying. J. Food Sci. 51 (2):510-511. Bannar, R. (1979). Vacuum-packaging for fresh fish - scaling the first hurdle. Food Eng. Int. 4:56-58. Barnett, H.J., Stone, F.E., Roberts, G.C., Hunter, P.J., Nelson, R.W. and Kwok, J.A. (1982) . A study in the use of a high concentration of CC>2 in a modified atmosphere to preserve fresh salmon. Mar. Fish. Rev. 44 (3):7-ll. Beatty, S.A. (1938). Studies of fish spoilage, II. Origin of trimethylamine produced during spoilage of cod muscle press juice. J. Fish. Res. Bd. Can. 4:63-68. Beatty, S.A. (1939). Studies of fish spoilage, III. The trimethylamine oxide content in the muscles of Nova Scotia fish. J. Fish. Res. Bd. Can. 4:229-232. Beatty, S.A. and Fougere, H. (1957). The processing of dried salted fish. Bulletin no. 112. Fisheries Research Board of Canada. Halifax, N.S. Beatty, S.A. and Gibbons, N.E. (1937). The measurement of spoilage in fish. J. Biol. Bd. Can. 3 (1):77-91. Bell, L. (1980). Keep fresh foods fresh. Package Eng. 25:72-75. Bell, L.D. (1982). Commercial perspectives on various modes of modified and controlled atmosphere applications for the seafood industry. Martin, R.E. (ed). Proc. of the First National Conference on Modified and Controlled Atmosphere Packaging of Seafood Products, San Antonio, 1981. Washington, DC: National Fisheries Institute, as cited in Stantham, J.A. (1984) . Bhuiyan, A.K.M., Aminullah, Ackman, R.G. and Lall, S.P. (1986); Effects of smoking on protein quality of Atlantic mackerel (Scomber scombrus). J. Food Proc. and Pres. 10(2): 115-126. Biliaderis, C.G. (1983). Differential scanning calorimetry in food349 research - a review. Food Chem. 10:239-265. Bilinski, E. and Fougere, H. (1959). The effect of sodium chloride on proteolysis and on the fate of amino acids present in muscle of codfish (Gadus Callarias). J. Fish. Res. Bd Can. 16 (5):747-754. Bjarnason, J. and Carpenter, K.J. (1970). Mechaniams of heat damage to proteins. 2. Chemical changes in pure proteins. Brit. J. Nutr. 24:313-329. Bligh, E.G. and Duclos, R. (1981). Salting of minced fish. In Fish By-Catch bonus from the sea. FAO and IDRC Report of a Technical Consultation on Shrimp By-Catch Utilisation, Georgetown, Guyana, 27- 30th Oct. 1981 :81-83. Bonnet, J.C., Sidwell, V.D. and Zook, E.G. (1974). Chemical and nutritive values of several fresh and canned finfish, crustaceans, and molluscs. II. Fatty Acid Composition. Marine Fisheries Review 36(2):8-14. Bose, A.N., Das Gupta, S.K. and Srimani, B.N. (1958). Studies on fish of the Bay of Bangladesh. J. Sci. Ind. Res. XXVIII (part III). 163-169. Bosund, I. and Ganrot, E. (1969). Lipid hydrolysis in frozen Baltic herring. J. Food Sci. 34:13-18. Botta, J.R., Lauder, J.T. and Jewer, M.A. (1984) . Effect of methodology on total volatile basic nitrogen determination as an index of quality of fresh Atlantic cod (Gadus morhua). J. Food Sci. 43:734-736, 750. Botta, J.R., Richards, J.F. and Tomlinson, N. (1973). Thiobarbituric acid value, total long-chain free fatty acids, and flavour of Pacific halibut (Hippoglossus stenolepis) and Chinook salmon (Oncorhynchus tshawytscha) frozen at sea. J.Fish.Res.Bd.Can. 30:63-69. Boury, M. (1936). Recherches sur l'alteration du poisson. II. Rev.Trav.Office Sci.Tech.Peches Maritimes. 3:401-419 (cited in Farber, 1965). Bramnaes, F. (1965). Handling of fresh fish. In Fish as Food, Vol.IV Borgstrom, G. (ed.). New York: Academic Press. Bremner, H.A. and Hallet, I.C. (1986). Degradation in muscle fibre- connective tissue junctions in the spotted trevalla (Seriolella punctata) examined by scanning electron microscopy. J.Sci.Food Agric. 37:1011-1018. Bremner, H.A. and Hallett, I.C. (1985). Muscle fibre-connective tissue junction in the fish blue grenadier (Macruronuc novaezelandiae) . A scanning electron microscopy study. J. Food Sci. 50:975-980. Bremner, H.A. and Vail, A.M.A. (1983) . Electrophoretic identification of fish species or salmon on Friday but burra(on)mundi. Food Technol.Aust. 35:322-326. Bromlei, G.F. (1949) as cited in Howgate (1979). Brunauer, S., Emmet, P.H. and Teller, E. (1938). Adsorption of gases in multimolecular layers. J.Am.Chem.Soc. 63:309-319. Buera, M.D.P., Chirife, J., Resnik, S.L., and Lozano, R.D. (1987a).350 Non-enzymatic browning in liquid model systems of high water activity. Kinetics of colour changes due to caramelisation of various single sugars. J. Food Sci. 52:1059-1062. Buera, M.D.P., Chirife, J., Resnik, S.L. and Wetzler, G. (1987b). Non-enzymatic browning in liquid model systems of high water activity: Kinetics of colour changes due to Maillard's reaction between different single sugars and glycine and comparison with caramelization browning. J. Food Sci. 52:1063-1067. Buera, M.D.P., Chirife, J., Resnik, S.L. and Lozano, R.D. (1987c). Non-enzymatic browning in liquid model systems of high water activity: Kinetics of colour changes due to reaction between glucose and glycine peptides. J. Food Sci. 52:1068-1070. Buisson, D.H., O'Donnell, D.K., Scott, D.N. and Ting, S.C. (1985). Squid processing options for New Zealand. Fish Processing Bull. 6. Fish Processing Research Section Div. of Horticulture and Processing, DSIR, Private Bag, Auckland, New Zealand. Burvall, A., Asp, N.G., Bosson, A., San Jose, C., and Dahlqvist, A. (1978). Storage of lactose-hydrolysed dried milk: Effect of water activity on the protein nutritional value. J.Dairy Sci. 45:381-389. Buttkus, H. (1974). On the nature of the chemical and physical bonds which contribute to some structural properties of protein foods: a hypothesis. J. Food Sci. 39:484-489. Cardin, A., Bilinski, E. Maltais, F. Bordeleau, M.A. and La Framboise, A. (1961). Chemical characteristics of salted cod. J. Fish.Res.Bd.Can. 18:851-858. Carpenter, K.J., Ellingher, G.M., Munro, M.I. and Rolfe, E.J. (1957). Fish products as protein supplements to cereals. Brit.J.Nutr. 11:162-173. Carver, J.H. (1969) as cited by Mendelson, J.M. (1974). Castell, C.H. and Giles, J.G. (1961). Spoilage of fish in the vessels at sea: Further studies on seasonal variations in the landed quality of gutted, trawler-caught Atlantic cod and haddock. J.Fish.Res.Bd.Can. 13:295-302. Castell, C.H. and Mapplebeck, E.C. (1952) The importance of Flavobacterium in fish spoilage. J. Fish. Res. Bd. Can. 9:148- 156. Caurie, M., Lee, T-C., Salmon, M. and Chichester, C.O. (1974). Hot smoke fish curing. J. Nat.Sci.Coun.Sri Lanka. 2:77-86. Chakraborty, P.K. (1978). Technological development of artificial and solar drying of fish in India. Proc.IPFC 18(3):322-329. Chakraborty, P.K. (1981). Dehydration of fish and its scope of commercial exploitation in India. Seafood Exp.J. 13:9-19. Charley, H. and Goertz, G.E. (1958). Effects of oven temperature on certain characteristics of baked salmon. Fd. Res. 23:17-24. Chen, R.H., Perng, Y.C., Weih, Y.C., Tsai, K.H. and Wang, T.T. (1982). Effect of lipid content and drying temperature-time on quality, drying rate, and energy consumption in fish drying. Proc. Third International Drying Symposium 2:266-280. Chinese National Standard for Dried Squids No. CN2298, N5054 351 (1981). National Bureau of Standards of China. Chio, K.S. and Tappel, A. L. (1969). Synthesis and characterisation of fluorescent products derived from malonaldehyde and amino acids. Biochem. 8:2821-2827. Choi, R.P. Koncus, A.F.; O-Malley, C.M., Fairbanks, B.W. (1949). A proposed method for the determination of colour of dry products of milk. J.Dairy Sci. 32:580-586. Chou, H. and Breen, W. (1972). Oxidative deterioration of B- carotene in low-moisture model systems. J.Food Sci. 37:66-68. Christensen, C.M. and Kaufmann, H.H. (1974). In: Storage of Cereal Grains and Their Products, Christensen, C.M. (ed.) Monogr.Ser.Am.Assoc.Cereal Chem., V. :158-192. Clark, E.D. and Almy, L.H. (1917). Preliminary studies on chemical methods of detecting deterioration in fish flesh. J.Ass.Off.Ag.Chem. 2:229. Clegg, K.M. and Morton, A.D. (1965). Carbonyl compounds and the non-enzymatic browning of lemon juice. J.Sci.Fd.Agric. 16:191-198. Clifford, M.N., Tang, S.L. and Eyo, A.A. (1980). The development of analytical methods for investigating chemical changes during fish smoking. In: Advances in Fish Science and Technology; Connell, J.J. (ed.). Fishing News Books Ltd. England. Clucas, I.J. (1981). Fish handling, preservation and processing in the tropics: Part 1: Report of the Tropical Products Institute, G144 viii + 144. Codex Alimentarius Commission (1985). FAO/WHO Food Standard Programme, Codex Alimentarius Commission, Sixteenth Session, Geneva, 1-12 July. Connell, J.J. (1957). Some aspects of the texture of dehydrated fish. J.Sci.Fd.Agric. 8:526-537. Connell, J.J. (1961). The relative stabilities of the skeletal muscle myosins of some animals. Biochem.J. 80:503-509. Connell, J.J. (1964). Fish muscle proteins and some effects on them of processing. In: Proteins and Their Reactions, Schultz, H.W. and Anglemeirt, A.F. (eds.), AVI Publ.Co., Westport, Connecticut. Connell, J.J. (1968). The effect of freezing and frozen storage on the proteins of fish muscle. In: Recent Advances in Food Science Vol.4, Hawthorn J. and Rolfe, E.J. (eds.). Pergamon Press, Oxford. Connell, J.J. (1980). Control of Fish Quality. 2nd ed. Fishing News Books Ltd, Surrey, England. Connell, J.J. and Shewan, J.M. (1980). Sensory and non-sensory assessment of fish. In: Advances in Fish Science and Technology; Papers presented at the Jubilee Conference of the Torry Research Station, Aberdeen, Scotland. 23-27 July 1979. Connell, J.J. (ed.). Fishing News Books Ltd., Surrey. Crean, P.B. (1961). The light pickle salting of cod. J.Fish Res.Bd.Can. 18:833-844. Curran, C.A. (1984). Storage of cured fisheries products in352 Malaysia. Report of the Tropical Development and Research Institute, L71 viii. Curran, C.A. and Poulter, R.G. (1983). Isohalic sorption isotherms. III. Application to a dried salted tropical fish (Xenomugil thoburni). J.Fd.Technol. 18(b):739-746. Curran, C.A. and Trim, D.S. (1983). Comparative study of solar and sundrying of fish. In: The Production and Storage of Dried Fish. Proc. of the Workshop on the Production and Storage of Dried Fish. Universiti Pertanian Malaysia, Serdang, 2-5 November 1982. FAO Fish.Rep., (279) . Suppl.:63-69 . Cutting, C.L. (1955). Fish Saving: A History of Fish Processing from Ancient to Modern Times. Leonard Hill, London. Cutting, C.L. (1962). Historical aspects of fish. In: Fish as Food Vol.2 Nutrition, Sanitation and Utilisation. Borgstrom, G. (ed.), Academic Press, New York. Cutting, C.L., Reay, G.A. and Shewan, J.M. (1956). Spec. Rep. Fd Invest. Bd, London, No. 62, HMSO. Dagbjartsson, B. (1983) . Production of salted and dried fish in the Maldives. The Production and Storage of Dried Fish. Proc. of Workshop on the Production and Storage of Dried Fish. Universiti Pertanian Malaysia, Serdang, Malaysia, 205 Nov. 1982. FAO Fish.Rep. 279 Suppl. James. D. (ed.) 58-63. Damograi, A.N. and Demyanov, A.N. (1980) . Optimisation of the process of air drying of salt fish. Rybnoe Khozyaistvo No.8: 64-66. Davenport, R.R. (1975). The distribution of yeasts and yeast-like organisms in an English vineyard. PhD Thesis, University of Bristol, England. Davies, J.R., Bardsley, R.G., Ledward, D.A. and Poulter, R.G. (1988). Myosin thermal stability in fish muscle. J.Sci.Food Agric. 45:61-68. de Groot, A.P. (1963). The influence of dehydration of foods on the digestibility and the biological value of the protein. Food Technol. March:103-107. De Valle, F.R. and Gonzalez-Inigo, J.L. (1968). A quick-salting process for fish. 2. Behaviour of different species of fish with respect to the process. Food Technol. 22:1135-1138. De Valle, F.R. and Nickerson, J.T.R. (1967). Studies on salting and drying fish. 1. Equilibrium considerations in salting. J.Food Sci. 32:173-179. De Valle, F.R. and Nickerson, J.T.R. (1968a). Salting and drying fish. III. Diffusion of water. J.Food.Sci. 33:499-503. Devadesand, K., Vishwanatahan Nair, P.G. and Antony, P.D. (1985). Effect of oxidation of dietary fish lipids on the quality of proteins in diet. Fishery Technol. 22:70-73. Disney, J.G. (1976). The spoilage of fish in the tropics. Proc. First Ann.Trop.Subtrop.Fish.Technol.Conf. of the Americas: 23-39. College Station, USA: Texas A & M University. Disney, J.G., Carmeron, J.D., Hoffman, J.D. and Jones, N.R. (1971).353 Quality assessment in Tilapia species. In Fish Inspection and Quality Control : 71-72, Kruezer, R. (ed.). London: Fishing News Books Ltd. Disney, J.G., Cole, R.C. and Jones, N.R. (1974). Considerations in the use of tropical fish species. In Fishery products: 329-337, Kruezer, R. (ed.). Surrey: Fishing News Books Ltd. Doe, P.E., Ahmed, M., Muslemuddin, M. and Sachithananthan, K. (1977). A polythene tent drier for improved sun-drying of fish. Food Technol.Aust. 29:437-441. Doe, P.E., Curran, C.A. and Poulter, R.G.(1983). Determination of the water activity and shelf life of dried fish products. Proc. of a Workshop on the Production and Storage of Dried Fish. Universiti Pertanian Malaysia, Serdang, Malaysia, 2-5 November 1982. FAO Fish Rep., 279 Suppl., James, D. (ed). :202-209. Doe, P.E., Hash Mi, R., Poulter, R.G. and Olley, J. (1982). Isohalic sorption isotherms. 1. Determination for dried salted cod (Gadus morrhua). Fd.Technol. 17:125-134. Dos Santos, L. (1981) . The storage of tropical fish in ice - A review. Trop.Sci. 23(2):97-127. Duckworth, R.B. and Smith, G.M. (1963). Diffusion of solutes at low moisture levels. In Recent Advances in Food Science Vol.III. J.M. Leitch and D.N. Rhodes, (eds.). Butterworth, London. Duerr, J.D. and Dyer, W.J. (1952). Proteins in fish muscle, 4. Denaturation by salt. J.Fish.Res.Bd.Can. 9:325-31. Dusseault, H.P. (1958). The salt tolerance of bacteria from lightly salted fish. In: The Microbiology of Fish and Meat Curing Brines. Eddy, B.P. (ed.), HMSO, London: 61-66. Dworschak, E. (1980). Non-enzymic browning and its effect on protein nutriton. CRC Critical Reviews in Food Science and Nutrition. 13(l):l-40. Dyer, W.J. (1952). Amines in fish muscle VI. Trimethylamine oxide content of fish and marine invertebrates. J.Fish Res.Bd.Can. 8(5):314-324. Dyer, W.J. and Mounsey, Y.A. (1945). Amines in fish muscle II. Development of trimethylamine and other amines. J.Fish.Res.Bd.Can. 8(5):359-367. Dyer, W.J., French, H.V. and Snow, J.M. (1950). Proteins in fish muscle I: Extraction of protein fractions in fresh fish. J.Fish Res.Bd.Can. 7(10):585-593. Eber, W. (1891). A chemical measure of putrefaction. Zeit. Fleish und Milchhygiene 1:11-19. Eichner, K. and Ciner-Doruk, M. (1980) . Formation of browning intermediates and visible sugar-amine browning reactions. In Properties of Water Related to Food Quality and Stability. L.B. Rockland and Stewart, G.F. (eds.). Academic Press, San Francisco. Eitenmiller, R.R. and DeSouza, S.C. (1984). Enzymatic mechanisms for amine formation in fish. Seafood Toxin. American Chemical Society :36-3 7. El-Dashlouty, A.A., El-Gedaily, A., El-Baahay, A.M. and Zahran, N.A.354 (1984). Fatty acid composition and lipid changes of squid as influenced by storage and processing. Egypt J.Food Sci., Vol. 12 (1-2):51-59. Endo, K., Fujita, M. and Shimizu, W. (1962). Muscle of aquatic animals. XXXIII Seasonal variations of nitrogenous extractives in squid muscle. Bull. Jap. Soc. Sci. Fish. 28:1099-1103. FAO (1970). Amino acid content of foods and biological data on proteins. FAO Nutritional Studies No.24. FAO Nutrition Division, Rome, Italy. FAO (1975). Food and Nutrition. (FAO) 1:41-43. FAO (1981). The prevention of Losses in Cured Fish. FAO Fish.Tech.Pap. 219. FAO (1983). The production and Storage of Dried Fish. FAO Fish Rep. 279. Rome. FAO (1984). Yearbook of fishery statistics. Vol.59. Farber, L. (1965). Freshness tests. In: Fish as Food. Vol.4, 65- 126, Borgstrom, G. (ed.). Academic Press, New York. Faturoti, B.O. (1984). Biological utilisation of sun-dried smoke African catfish (Chrysicthys nigrodigitatus). Nutrition Reports International 30(6):1395-1342. Faturoti, E.O. and Aransiola, M.O. (1984). Biochemical evaluation of the nutritive quality of differently processed fish (Heteroti niloticus). Nutr.Rep.Internat. 30(5):1221-1229. Fennema, O.R., Powrie, W.D. and Marth, E..H. (1973). Characteristics of food systems and their behaviour during freeze drying. In Low Temperature Preservation of Foods and Living Matter. Food Science Series of Monographs, Marcel Dekker, New York. Fey, M.S. (1980). Extending the shelf life of fresh fish by potassium sorbate and modified atomospheres at 0 - 1 degree C. Ithaca, NY: Cornell University: PhD thesis. Filsinger, B.E. (1987). Effect of pressure on the salting and ripening process of anchovies (Engraulis anchoita) . J. Food Sci. 52:919-921, 927. Finch, A. and Ledward, D.A. (1972). Shrinkage of collagen fibres : a differential scanning calorimetric study. Bioch. Biophys. Acta 278:433-439. Finch, A. and Ledward, D.A. (1973). Differential scanning calorimetric study of collagen fibres swollen in aqueous neutral salt solutions. Bioch. Biophys. Acta 295:296-300. Findlay, C.J. and Stanley, D.W. (1984). Differential scanning calorimetry of beef muscle: influence of postmortem conditioning. J. Food Sci. 49:1513-1516. Firth, R. (1946). Malay Fisherman: Their Peasant Economy. Kegan, Paul, Trench, Trubner & Co., Ltd., London. Ford, J.E. (1964). A microbiological method for assessing the355 nutritional value of proteins. 3. Further studies on the measurement of available amino acids. Brit. J. Nutr. 15:449-460. Ford, J.E. (1973). In Proteins in Human Nutrition, eds. Porter, J.W.G. and Rolls, B.A. (eds.). Academic Press, London and New York. Frank, M. and Hess, E. (1941). Studies on salt fish. VI. Halophilic brown moulds of the genus Sporendonema emend. (Ciferr and Redaelli). J. Fish. Res. Bd Can., 5:287-292. Frankel, E.N. (1984). Recent Advances in the Chemistry of Rancidity of Fats. Spec. Publ. R. Soc. Chem. 47:87-118. Fujii, T., Ishida, Y. and Kadota, H. (1974). Utilisation of trimethyalmine by a Penicillium strain isolated from salted fish. Bull. Jpn. Soc. Sci. Fish. 40(12):1309. Fujii, T., Ishida, Y. and Kadota, H. (1977). Changes in microflora during storage at low temperature of salted fish. Bull. Jpn. Soc. Sci. Fish. 43(10) : 1241-1247. Fujimoto, K. Abe, I. and Kaneda, T. (1971). Brown discolouration of fish products, Nippon Suisan Gakkaishi 37: 40-43. Geiger, E. and Borgstrom, G. (1962). Fish protein-nutritive aspects. In Fish as Food, ed. Borgstrom, G. Vol. 2:29-65. Academic Press,New York. George, J.C. (1962) . A histophysiological study of the red and white muscles of the mackerel. Am. Midi. Nat. 55:487-494. Geromel, E.J. and Montgomery, M.W. (1980). Lipase release from lysosomes of rainbow trout (Salmo gairdneri) muscle subjected to low temperatures. J. Food Sci. 45:412-415, 419. Gill, T.A. and Thompson, J.W. (1984). Rapid, automated analysis of amines in seafood by ion-moderated partition HPLC. J. Food Sci. 45:603-606. Golob, P., Co x, J.R. and Kilminster, K. (1987). Evaluation of insecticide dips as protectants of stored infestation. J. Stored Prod. Res. 23 (1) :47 — 56. Goonawardeen, I.S.R. (1978). A preliminary study on quality of dried salted fish. Paper presented at the FAO/Danida Workshop on Fish Technology and Inspection, Colombo, Sri Lanka, 16 October - 24 November, 1978. Goonewardeen, I.S.R. and Etoh, S. (1980). Keeping quality of imported dried salted fish. Bull. Fish. Res. Stn., Sri Lanka, 30 (1-2):63-74. Gopakumar, K. and Devadesan, K. (1983) . The fish curing industry in India. The Production and Storage of Dried Fish. Proc. of the Workshop on the Production and Storage of Dried Fish. University Pertanian Malaysia, Serdang,Malaysia, 2-5 November, 1982. FAO Fish. Rep., 279 Suppl. James, D. (ed.). Gordievskaya, V.S. (1973). Shark flesh in the food industry, translated by H.M. Mills, National Marine Fisheries Services, National Oceanic and Atmospheric Administration, U.S. Dept. of Commerce and the National Science Foundation, Washington, DC. Graham, P.P., Hamilton, R.S. and Pierson, M. (1986). Influence of356 brining procedures on salt content and distribution in smoked whitefish chubs. J. Food Proc. Pres. 10(4):295-309. Gray, R.J.H., Hoover, D.G. and Muir, A.M. (1983). Attenuation of microbial growth on modified atmosphere-packaged fish. J. Food Prot. 46:610-613. Gustafson, F.G. and Cooke, A.R. (1952). Oxidation of ascorbic acid to dehydroascorbic acid at low temperatures. Science 116:234. Haaland, H. and Njaa, L.R. (1988) . Ammonia and total volatile nitrogen in preserved and unpreserved stored, whole fish. J. Sci. Food Agric. 44:335-342. Haard, N.F. (1982). Utilisation of squid in Canada. In: Proceedings of International Squid Symposium; Learson, R. (ed.) Unipub, New York. Haard, N.F. and Arcilla, R. (1985). Precursors of Maillard browning in Atlantic short finned squid. Can. Inst. Food Sci. Technol. J. 18(4):326-331. Hagerdal, B. and Marens, H. (1976). Influence of water content on the stability of myoglobin to heat treatment. J. Food Sci., 41:933- 937. Hamada, I., Yabuno, F., Furutdatsu, K. and Niwa, E. (1982) . The effect of lipid on the heat denaturation of actomyosin. Bull. Jap. Soc. Sci. Fish. 45:189-193. Hamed, M.G.E. and Adly, Z.S. (1974). A contribution to the study of free tyrosine, trimethylamine, and ammonia as measures of the degree of deterioration of dehydrated shrimp. Egypt. J. Food Sci. 2(l):79-86. Hamm, R. (1960). Biochemistry of meat hydration. Advances in Food Research, 10:355-463. Hamm, R. (1966). Heating of muscle system. In Physiology and Biochemistry of muscle as food. Briskey, E.J., Cassens, R.G. and Trautman, J.C. (eds.). Univ. of Wisconsin Press, Maddison, WI. Hamm, R. (1977). Changes of muscle proteins during the heating of meat. In Physical, chemical and biological changes in food caused by thermal processing. Hoyem, T. and Kvale, 0. (eds.), Applied Science, London. Hansen, P. (1972). Storage life of prepacked wet fish at 0°C. II. Trout and herring. J. Food Technol. 7:21-26. Hansen, P. (1980). Fish preservation methods. In Advances in fish science and technology, papers presented at the Jubilee Conference of the Torry Research Station, Aberdeen, Scotland, 23-27, July, 1979, Fishing News Books Ltd., Surrey. Hansen, R.P. (1970). Isoprenoid fatty acids in Antarctic krill (Euphasia superba). J. Sci. Food Agric., 21:203-206. Harada, K. (1975). Studies on enzymes catalyzing the formation of formaldehyde and dimethylamine in tissues of fishes and shells. J. Shimonoseki Univ. Fish. 23(3):163-241. Hashimoto, Y. and Akaichi, T. (1957). On the determination of357 trimethylamine and trimethylamine oxide. Modification of the Dye method. Bull. Jpn. Soc. Sci. Fish. 23:269-273. Hastings, R.J., Rodger, G.W., Park, R., Matthews, A.D. and Anderson, E.M. (1985). Differential scanning calorimetry of fish muscle: the effect of processing and species variation. J. Food Sci., 50:503- 506, 510. Hayashi, K. and Tagaki, T. (1977a). Seasonal variation in lipids and fatty acids of sardine, Sardinops melanostitca. Bull. Fac. Fish. Hokkaido Univ. 28 (2) :83—94. Hayashi, K. and Tagaki, T. (1977b). On the fatty acid composition of fish affected by excessive stress. Bull. Jap. Soc. Sci. Fish. 43:1189-1194. Hayashi, K. and Takagi, T. (1979). Browning of dried squid products I. On the chemical constituents for amino acids for squid mantles. Hokkaido Univ. Fac. Fish. Bull. 30(4):288-293. Hebard, C.E., Flick, G.J. and Martin, R.E. (1982). In Chemistry and Biochemistry of Marine Food Products. Martin, R.E., Flick, G.J. Hebard, C.E.; Ward, D.R. (eds) AVI Publishing Co. Westport. Heen, E. and Kruezer, R. (1962). Fish in Nutrition. Fishing News Books Ltd, London. Hegarty, P.V.J. (1982). Influence of processing on nutritive value of proteins. In Food Proteins. Fox, P.F. and Condon, J.J. (eds) Applied Sci. Pub., London, New York. Heiss, R. and Eichner, E. (1971). Moisture content and shelf life. Food Manufacture, June: 37-42. Heldman, D.R., Reidy, G.A. and Palnitkar, M.P. (1973). Texture stability during storage of freeze dried beef at low intermediate moisture contents. J. Food Sci. 33:282-285. Hincks, M.J. and Stanley, D.W. (1985). Colour measurement of the squid Illex illecebrosus and its relationship to quality and chromatophore ultrastructure. Hoffman, A., Barranco, A., Francis, B.J. and Disney, J.G. (1977). The effect of processing and storage upon the nutritive value of smoked fish from Africa. Trop. Sci. 19: 41-53. Hoppe-Seyler, F.A. (1933). Trimethylamine oxide and other nitrogen bases in cod muscles. Z. Physiol. Chem. 221:45-50. Howgate, P. (1979). Fish. In Food Microscopy. Vaughan, J.G. (ed) Academic Press, London. Howgate, P.F. and Ahmed, S.F.(1972). Chemical and bacteriological changes in fish muscle during heating and drying at 30°C. J. Sci. Food Agric. 23:615-627. Hsieh, Y.P., Cornforth, D.P. and Pearson, A.M. (1980). Ultrastructural changes in pre- and post-rigour beef muscle caused by conventional and microwave cookery. Meat Sci. 4:299-311. Hung, N.D., Cseke, E., Vas, M. and Szabolcsi, G. (1984). Processed protein foods characterised by in vitro digestion rates. J. Fd. Sci. 49:1543-1551. Ichinoe, M., Suzuki, M. and Kurata, H. (1977). Microflora of358 commercial sliced dried fish including bonito. Bull. Nat. Inst. Hyg. Sci. 95:96-99. Iglesias, H.A. and Chirife, J. (1977) . Effect of fat content on the water sorption isotherm of air dried minced beef. Food Sci. Technol. 10:151-152. INFOFISH (1983) . Dried fish: An Asian staple food (A preliminary report on a joint study by the Asian Development Bank and FAO) . Proceedings of the Workshop on the Production and Storage of Dried Fish, Universiti Pertanian Malaysia, Malaysia, 2-5 Nov. 1982. FAO Fisheries Report No. 279 Supplement. Ingram, J. and Kitchell, A.G. (1967) . Salt as a preservative for foods. J. Food Technol. 2:1-15. Ishida, Y., Fujii, T. and Kadota, H. (1976). Microbiological studies on salted fish stored at low temperature I. Chemical changes of salted fish during storage. Bull. Jpn. Soc. Sci. Fish. 42(3):351-358. Ishikawa, N. and Nakamura, K. (1975) . On-board and on-land freshness preservation of squid. Prog. Rep . Squid Fish Survey World, No. 5: 197. Ismail, M.S. (1980). The use of solar energy and agricultural wastes (Biomass) in the drying of fish and fish products in Malaysia. Paper presented at the Conference on Solar Energy Utilisation, held University of Western Ontario, Canada. Iwasaki, M. and Harada, R. (1985) . Proximate and amino acid composition of the roe and muscle of selected marine species. J. Food Sci. 50:1585-1587. James, D. (1983) . The production and storage of dried fish. FAO Fish. Rep. 279, Rome. James, D. (1984). Fish for the future. In Food Science and Human Welfare, vol. 4 of the Proceedings of the 6th International Congress of Food Science and Technology, September, 1983 by J.V. McLoughlin and McKeena, B.M. (eds) Boole Press, Dublin. James, D.G. and Olley, J. (1971). Spoilage of shark. Australian Fisheries, April, 11-13. Jarenback, L. and Liljemark, A. (1975). Ultrastructural changes during frozen storage of cod (Gadus morhua L.) J. Fd. Technol. 10:229-239. Jason, A.C. (1958). A study of evaporation and diffusion processes in the drying of fish muscle. In Fundamental Aspects of the Dehydration of Foodstuffs. Society of Chemical Industry, London. Jason, A.C. (1965). Effects of fat content on diffusion of water in fish muscle. J. Sci. Food Agric. 16:281-188. Jason, A.C. (1980). General theory of drying of fish. Proceedings of the International Association of Fish Meal Manufacturers on Drying, Handling and Storage of Fish Meal. Athens, 6 October 1980. news Summ. Int. AFMM No. 49:5-25. Jewell, D.K., Kendrik, J.G. and Satterlee, L.D. (1988). The DC-PER359 assay: a method of predicting quality of protein solely from amino acid compositional data. Nutr. Reports Inti. 21(1):25-38. Jhaveri, S.N., Karakoltsidis, P.A., Montefalvo Jr., J. and Contaninides, S.M. (1984). Chemical composition and protein quality of some Southern New England marine species. J. Food Sci. 49:110-113. Jimenez-Colmenero, F., Tejada, M. and Borderias, A.J. (1988). Effect of seasonal variations on protein functional properties of fish during frozen storage. J. Food Biochem. 12:159-170. Jinadatharaya Sri, Y.A. and Vernekar Sri, V.S. (1979). Modern method for curing and drying shark fish. Seafood Export J. 11 (1) : 48-50. Jokinen, J.E. and Reineccius, G.A. (1976). Losses in available lysine during thermal processing of soy protein model systems, J. Food Sci. 41:816-819. Joly, M. (1965). A physico-chemical approach to the denaturation of proteins. Academic Press, New York. Jonas, R.E.E. and Bilinski, E. (1967). Phospholipase A activity in rainbow trout muscle. J. Fish. Res. Bd. Can. 24:2555-2562. Jones, N.R. (1962). Browning reactions in dried fish products. In Recent Advances in Food Science. 2. Processing. Hawthorn, J. and Leitch, J.M. (eds). Butterworths, London. Jones, N.R. (1965). Hypoxanthine and other purine-containing fractions in fish muscle as indices of freshness. In The Technology Of Fish Utilisation, Kruezer, R. (ed) . Fishing News Books Ltd., Surrey, for FAO. Jones, S.B., Carroll, R.J. and Cavanaugh, R. (1976). Muscle samples for electron microscopy : preparative techniques and general morphology. J. Fd Sci. 41:867-873. Kaimal, P.N.R. (1969). Freezing of oil sardines. Indian Seafood J. 7(2):24-40 (World Fish. Abstr., 1970, 21(3):29). Kalakowski, E. and Gajowiecke, L. (1973). Research on the chemical composition and thermal treatment of squid meat (Loligo vulgaris). Proceedings of Squid Symposium, Gydnia, Poland, 36-41. Kandoran, M.K. and Valsan, A.P. (1974). Packing of cured fish in India: existing methods and suggestions for improvements. Seafood Export J. 6(5):27-31. Kanna, K., Takahashi, T. and Tanaka, T. (1965). As cited in Suzuki (1981) . Kanna, K., Tanaka, T., Kakadu, K. and Shimizu, T. (1971) . Ascited in Suzuki (1981). Karel, M. (1973). Recent advances and developments in the field of low moisture and inter-media foods. CRC Crit. Rev Food Technol. 3:329-373. Karmas, E. and DiMarco, G.R. (1970). Denaturation thermoprofiles of some proteins. J. Food Sci. 35:725-727. Kawabata, T. (1953) . Studies on the TMAO reductase I. Reduction of360 TMAO in the dark muscle of pelagic migrating fish under aseptic condition. Bull. Jpn. Soc. Sci. Fish. 19:505-512. Kawabata, T., Matsui, M. Ishibahi, T., Hamano, M. and Ino, M. (1982). Formation of N-nitroso compounds during cooking of Japanese Food. IARC Scientific Publications. 41:287-297. Kawamura, T., Sakai, K., Miyazawa, F., Wada, H., Ito, Y. and Tanimura, A. (1971). Studies on nitrosamines in foods. V. Distribution of secondary amines in Foods (2) . Shokuhin Eiseigaku Zasshi. 12(5): 394-398. Ke, P.J., Burns, B.G. and Woyewoda, A. D. (1984). Recommended procedures and guidelines for quality evaluation of Atlantic short finned squid (Illex illecebrosus). Lebensm.-Wiss. u.-Technol., 17:276-281. Ke, P.J., Cervantes, E. and Robles-Martinez, C. (1984). Determination of thiobarbituric acid reactive substances (TBARS) in fish tissue by an improved distillation-spectrophotometric method. J. Sci. Food Agric. 35:1248-1254. Ke, P.J., Woyewoda, A., and Fierheller, M. (1979). Handling methods and quality evaluation of fresh Canadian Atlantic squid (Illex illecebrosus). Can. Fish. Tech. Report 898. Keenan, C.P. and Shaklee, J.B. (1985). Electrophoretic identification of raw and cooked fish fillets and other marine products. Food Technol. Aust. 37(3):117-128. Khayat, A. and Schwall, D. (1983). Lipid oxidation in seafood. Food Technol. July: 130-140. Kiamura, K. and Kiamakura, S. (1934). Detection of the onset of decomposition in fish meat as shown by the content of ammonia. Proc. 5th Pacific Sci. Congr. 5: 3709. Kida, K. and Tamoto, K. (1969) . Studies on the muscle of aquatic animals. IV. On the relation between various pH values and organic acids of Nagazuka muscle (Stichaeus grigorjewi Herzenstein). Scientific Report of the Hokkaido Fisheries Experimental Station 11:41-57. Kida, K. and Tamoto, K. (1974) . Studies on the development of amines in Alaska Pollock muscle by drying. Hokusuishi Geppo 31(12):16-26. Kida, K. and Tamoto, K. (1976) . Studies on the amines in Alaska Pollock muscle II. Development of trimethylamine and dimethylamine by drying. Hokusuishi Geppo 33(3):44-61. King, A.D. and Nagel, C.W. (1975). Influence of carbon dioxide upon the metabolism of Pseudomonas aeruginosa. J. Food Sci. 40:362-366. King, D. and Poulter, R.G. (1985). Frozen storage of Indian mackerel (Rastrelliger kanagurta) and Big eye (Priacanthus hamtur). Trop. Sci. 25:79-90. King, D., Kamara, V.A. and Wood, C.D. (1984). Salted and pressed sardine. Paper presented at IPFC Workshop on Fish Technology and Marketing, Melbourne, Oct. 22-26. King, D., Kamara, V.A. and Wood, C.D. (1985). Salted and pressed361 sardines. In: Proceedings of a Symposium "Spoilage of tropical fish and product development" held in conjunction with the Sixth session of the Indo-Pacific Fishery Commission Working Party on fish technology and marketing. Reilly, A. (ed) Royal Melbourne Institute of Technology, Melbourne, Australia, 23-26 October, 1984: 271-277. King, F.J. and Ryan, J.J. (1977). Development of a colour measuring system for minced fish blocks. Mar. Fish. Rev. Feb. 18- 23. King, R.D. (1980). The determination of food colours. In: Developments in Food Analytical Technique, vol. 2. ed. King, R.D. Applied Science Pub. Ltd, London. Kinsella, J.E. (1982). Relationships between structure and functional properties of food proteins. In Food Proteins; Fox, P.F. and Condon, J.J. (eds) App. Sci. Pub. London and New York: 51-104. Kinsella, J.F., Shimp, J.L., Mai, J. and Weirauch, J. (1977a). Sterol, phospholipid, mineral content and proximate composition of fillets of select freshwater fish species. J. Food Biochem. 1:131- 140 . Kinsella, J.F., Shimp, J.L., Mai, J. and Weirauch, J. (1977b). Fatty acid content and composition of fresh water finfish. J. Am. Oil Chem. Soc. 54:424-429. Kizevetter, I.V. (1973). Chemistry and Technology of Pacific Fish. Pacific Scientific Research Institute of Marine Fisheries and Oceanography (TNIRO), Dulizdat, Vladivostok, 1971, Israel Programme for Scientific Translation 1973. Koizumi, C. (1980). Lipid oxidation of salted freeze-dried fish meats at different equilibrium relative humidities. Bull. Jap. Soc. Sci. Fish. 46{1):871-877. Kolodziejska, I. and Sikorski, Z.E. (1979). Inorganic salts and functional properties of fresh and frozen muscle proteins. Proc. 25th European Meeting of Meat Research Workers, Budapest. Kolodziejska, I., Sikorski, Z.E. and Sadowska, M. (1987). Texture of cooked mantle of squid Illex argentinus as influenced by specimen characteristics and treatments. J. Food Sci. 52(4):932-935. Konosu, S., Katori, S., Ota, R., Eguchi, S. and Movi, T. (1958). Muscle extracts of aquatic aniimals. I. Amino acids, trimethylamine and trimethylamine oxide in the muscle extracts of a squid. Ommastrephes sloani pad ficus. Bull. Japan. Soc. Sci. Fisheries 23:561-564. Krishnan, G.S. (1982). Dried Squid Part 2: A note on processing, packing and storing. Seaf. Export J. 24(7) :9 — 13. Kritzman, G., Chet, I. and Henis, Y. (1977). Effect of carbon dioxide on growth and carbohydrate metabolism in Sclerotium rolfsii. J. Gen. Microbiol. 100:167-175. Kruezer, R. (1974). Fish and its place in culture. In: Fishery Products by Kruezer, R. (ed). Fishing News Ltd, London. Kunimoto, M., Matsumoto, K., Zama, K. (1985). The interaction of362 fish proteins and lipids during freeze-drying and storage. I. The interaction of carp myofibrils and fish oil. Bull. Fac. Fish., Hokkaido University [Hokkaido Daigaku Suisangakubu Kenkyu Iho] 36(1) :50-56 . Labarre, J. and Fougere, H. (1942) . Physical chemical changes in extracts of cod fillets with temperature and time. Trans. Roy. Soc. Can. 35:41-43. Labrie, A. and Gibbons, N.E. (1937). Studies on salt fish. II. The effect of salt concentration on preservation. J. Biol. Board Can. 3:439-449. Labuza, T.P. (1968). Sorption phenomenon in foods. Food Technol. 22(3):15-24. Labuza, T.P. (1970). Properties of water as related to the keeping quality of foods. Proc. 3rd Internat. Congress on Food Science and Technology. IFT. Labuza, T.P. (1971). Kinetics of lipid oxidation in foods. CRC Crit. Rev. Food Technol. 2:355-405. Labuza, T.P. (1973). Storage stability and improvement of intermediate moisture foods. Contract NAS 9-12560, NASA, Houston, Texas: 177-184. Labuza, T.P. and Saltmarch, M. (1981). The nonenzymatic browning reaction as affected by water in foods. In Water activity influences on food quality Rockland, L. (ed) . Academic Press, New York. Labuza, T.P., Tannenbaum, S.R. and Karel, M. (1970). Water content and stability of low-moisture and intermediate-moisture foods. Food Technology (May) 24:542-550. Laird, W.M., Mackie, I.M. and Ritchie, A.H. (1982). Differentiation of species by isoelectric focussing on agarose and polyacrylamide gels - a comparison. J. Assoc. Publ. Anal. 20:125-135. Lannelongue, M., Finne, G., Hanna, M.O., Nicklelson, R. and Vanderzant, C. (1982a). Storage characteristics of brown shrimp (Penaeus astecus) stored in retail packages containing C02~enriched atmosphere, J. Food Sci. 47:911-913, 923. Lannelongue, M., Finne, G., Hanna, M.O., Nicklelson, R. and Vanderzant, C. (1982b). Micorbiological and chemical changes during storage of swordfish (Xiphias gladius) steaks in retail packages containing CC>2-enriched atmospheres. J. Food Prot. 45:1197-1203. Lannelongue, M., Finne, G., Hanna, M.O., Nicklelson, R. and Vanderzant, C. (1982b). Storage characteristics of finfish fillets (Archosargus probatocephalus) packaged in modified gas atmosphere containing carbon dioxide. J. Food Prot. 45:440-444. Latif, K., Ismail, M.S., Yassin, Z. and Bardaie, Z. (1983). Production of dried fish in the East Coast of Penninsular Malaysia - A survey. The production and storage of dried fish. Proc. of the Workshop on the Production and Storage of Dried Fish. Universiti Pertanian Malaysia, Serdang, Malaysia, 2-5 Nov. 1982. FAO Fish Rep., 279 suppl. by James, D. (ed):44-57. Lawrie, R.A. (1966) . Meat Science, Pergamon Press, London,363 England. Lea, C.H. (1962). The oxidative deterioration of food lipids. In Symposium on Foods: Lipids and Their Oxidation. Schultz, H.W., Day, E.A. and Sinhuber, R.O. (eds). AVI Pub. Comp. Inc. Westport. LeBlanc, E.L. and Gill, T.A. (1982). Comparative studies of proteolysis in short-finned (Illex illecebrosus) and long finned (Loligo pealei) squid. Comp. Biochem. Physiol. 73B:201-210. Lee, F. A. and Grau, R. (1966). On the influence of temperature on the behaviour of soluble proteins of beef. Die. Flieischwirtschaft- Nr. 46:1239-1240 . Lee, J.M. (1969). Private communication. National Marine Fisheries Service, Atlantic Fishery Products Technology Centre, Emerson Avenue, Gloucester, MA 01930, cited by Mendelsohn, J.M. (1974 ) . Lee, K.H. and Ryu, H.S. (1987). Evaluation of seafood protein quality as predicted by C-PER assays. In Seafood Quality Determination. Kramer, D.E. and Liston, J. (eds) Developments in Food Science, Elsevier, Amsterdam, Oxford, New York, Tokyo. Lee, K.H., Song, D.S. You, B.J.and Kim, M.N. (1982). (Changes in available lysine and extractable N. and extent of browning during the storage of dried fish meat). Bull. Korean Fish. Soc. 15(4):271-282. Lee, Y.Z., Simpson, B.K. and Haard, N.F. (1982). Supplementation of squid fermentation with proteolytic enzymes. J. Food Biochem. 6(2):127-134. Legendre, R. (1955). The artificial drying of lightly salted codfish. J. Fish. Res. Bd. Can. 12:68-74. Legendre, R. (1961). Artificial drying of Cambodian fish. J. Fish. Res. Bd. Can. 15:147-162. Leisner, L. (1976). The stability of intermediate moisture foods with respect to microorganisms. In Intermediate moisture foods. Davis, R., Birch, G.G. and Parker, K.J. (eds) Applied Science Pub. Ltd. London. Leninger, H.A. (1959). Private communication as cited in Jason, A.C. (1965). Lerke, P., Adam, R. and Farber, L. (1965). Bacteriology of spoilage of fish muscle. III. Characterisation of spoilers. Appl. Microbiol. 13:625-639. Lewis, D.F. (1979). Meat Products. In Food Microscopy. Vaughan, J.G. (ed). Academic Press, London. Ligouri, V.R., Ruggieri, G.D., Baslow, M.H., Stempien, M.F. and Nigrelli, R.E. (1963). Antibiotic and toxicity activity of the mucous of the Pacific stripped bass Graiwnistes sexlineatus. Am. Zool. 31:546 (Abstract No. 302). Lin, J-K. and Hurng, D-C. (1985). Thermal conversion of trimethylamine-N-oxide to trimethylamine and dimethylamine in squids. Food Chem. Toxic. 23:579-583. Linton, E.P. and Wood, A.L. (1945). Drying of heavily salted fish.364 J. Fish. Res. Bd. Can. 6:380-391. Liston, J. (1980). Microbiology in fishery science. In Advances in Fish Science and Technology. Connell, J.J. and Torry Research Station Staff (eds). Fishing News Books Ltd. Surrey. Lobanov, D. and Bykova, S.V. (1938). Cited by Aitken and Connell (1979) . Love, R.M. (1970). Chemical Biology of Fishes. Academic Press, London. Lovern, J.A. and Olley, J. (1962) . Inhibition and promotion of post-mortem lipid hydrolysis in the flesh of fish. J. Food Sci. 27:551-55. Lundstrom, R.C. and Racicot, L.D. (1985) . Gas chromatographic determination of dimethylamine and trimethylamine in seafoods. J. Assoc. Off. Anal. Chem., 66(5):1158-1163. Lupin, H.M. (1981). Principles of Fish Salting, FAO, Rome. Mackie, I.M. (1980). A review of some recent applications of electrophoresis and isoelectric focussing in the identification of species of fish in fish and fish products. Mackie, I.M. and Ritchie, A.H. (1982). Differentiation of species of fish by isoelectric focussing on agarose and polyacrylamide gels - a comparison. J. Assoc. Publ. Anal. 26:125-135. Madovi, P.B. (1980). Changes in the free-NH2, free C02 and titratable acidity of meat proteins. J. Fd Technol. 15:311-318. Mai, J. and Kinsella, J.E. (1979). Lipid composition of dark and white muscle from white sucker (Catostomus commersoni) . J. Food Sci. 44:1101-1105, 1109. Mai, J. and Weirauch, J. • (1977) . Fatty acid content and composition of fresh water finfish. J. Am. Oil Chem. Soc. 54:424- 429. Malle, P., Eb, P. and Tailliez, R. (1986). Determination of the quality of fish by measuring trimethylamine oxide reduction. Internat. J. Food Micro. 3:225-235. Mallins, D. and Wekell, J.C. (1970). The lipid biochemistry of marine organisms. In: Progress in the Chemistry of Fats and Other Lipids. Vol. 10, R.T. Holman (ed) Pergamon, Oxford. March, B.E. (1962). Fish meal and condensed fish solubles in poultry and livestock feeding. In: Fish as Food. Borgstrom, G. (ed) Academic Press, New York. Martin, H.F. (1958) . Factors in the development of oxidative rancidity in ready to eat crisp oatflakes. J. Sci. Food Agric.: 817-824. Mashelkar, B.N. and Sohonie, K. (1958). Nutritive value of proteins from shark (Scoliodon sorrakowah) and skate (Rhynchobatus djiddensis) Part I. Amino acid make up and rate if release of essential amino acids during digestion in vitro. Ann. Biochem. Exp. Med. XVIII (5):135-140. Mat, A.A. (1983). Standards for Dried Fish. Proc. of the Workshop365 on the Production and Storage of Dried Fish. Universiti Pertania Malaysia, Serdang, Malaysia, 2-5 November 1982. FAO Fish Rep., 279 Suppl. James, D. (ed) :185-193. Mathew, A. G. and Parpia, H.A.B. (1971). Food browning as a polyphenol reaction. Adv. Food Res. 15:75-145. Matsumoto, J.J., Chihara, S., Akahane, T., Tsuchiya, T., Noguchi, S. and Ookami, H. (1981). Application of differential scanning calorimetry to food technological studies of meats of fish and rabbit. Rec. Adv. Food Sci. Technol. Vol. 11 S.M. Chang (ed) . Hua Shiang Tuan Publishing Co., Taipei, Taiwan. McCormick, P.Y. (1973). Gas-solid systems. Perry, R.H. and Chilton, C.H. (eds) Chemical Engineers' Handbook. 5th edn. McGraw-Hill, New York. Meinke, W.W. and Mattil, A.K.F. (1973) . Autolysis as a factor in the production of protein isolates from whole fish. J. Food Sci. 38:864-866. Meinke, W.W., Rahman, M.A. and Mattil, K.F. (1972). Some factors influencing the production of protein isolates from whole fish. J. Fd Sci. 37:195-198 . Menashi, S., Finch, A., Gardner, P.J. and Ledward, D.A. (1976). Enthalpy changes associated with the denaturation of collagen of different amino acid content, Biochim. Biophys. Acta, 444:623-627. Mendelsohn, J.M. (1974). Rapid techniques for salt-curing fish: A review. J. Fd Sci. 35:125-127. Meynell, P.J. (1978). Reducing blowfly spoilage during sun drying of fish in Malawi using pyrethrum. Proc. IPFC, 1893:347-353. Migita, M., Matsumoto, J.J. and Saishu, T. (1956). Denaturation of fish muscle proteins by dehydration. Bull. Jap. Soc. Sci. Fish., 22:433-439. Migita, M., Matsumoto, J.J. and Suzuki, T. (1960). Denaturation of fish muscle protein by dehydration. II. Changes in viscosity and streaming birefringence of the extractable proteins. Bull. Jap. Soc. Sci. Fish., 26:925-930. Miller, A., Scanlon, R.A., Lee, J.S. and Libbey, L.M. (1972). Quantitative and selective gas chromatography of dimethyl and trimethylamines in fish. J. Ag. Fd. Chem. 26:709-711. Miller, D.S. (1956). The nutritive value of fish proteins. J. Sci. Food Agric. 7:337-343. Miller, E.L., Hartley, A.W. and Thomas, D.C. (1965). Availability of sulphur amino acids in protein foods. 4. Effect of heat treatment upon the total amino acid content of cod muscle. Br. J. Nutr. 15:565-573. Mills, A. (1977) . Handling and processing of fish on Lake Chad, cited in FAO Fish. Tech. Pap. 219. Moen, E. (1983). Cured fish: market patterns and prospects. FAO Fisheries Technical Paper No. 233. Morshita, T. and Takahashi, T. (1969). Studies on the esterase and366 lipase of fish III. On the esterase and lipase in the fish musle. J. Fac. of Fish, Mie Prefect Univ. 5:41-51. Mulyanto, R. (1982) . Utilisation of fat in correlation with optimal utilisation of oil sardine. Nurhakim, S. (ed) Prosiding Perikanan Lemuru. Banyuwangi; 18-21 Jan. 1982, Jakarta. Badan Penelkitian dan Pengembangan Pertanian, Dept Pertanian. Murray, C.K. and Jobber, P.I. (1968). Manufacture and packaging of dried salted fish and stockfish of the World Food Programme: as cited in FAO (1981). Muslemuddin, M., Wahed, M.A. ad Khaleque, M.A. (1984). Influence of relative humidity on the moisture sorption characteristics of some dehydrated fish products. Bangladesh J. Sci. Ind. Res. XIX 1- 4:37-45. Nair, R.B. and Dani, N.P. (1975). Technology of utilisation of freshwater fish. Proc. Symp. Fish Processing Industry in India, 20- 21. Sharada Press, Mangalore, India. Nakamura, K., Iida, H., Nakamura, K. and Ishikawa, S. (1985a). Studies on utilisation of shark meat I: Changes in chemical composition of shark meat during storage and processing. Bull. Tokai. Reg. Fish. Res. Lab. 115(Feb):17-22. Nakamura, K., Iida, H., Nakamura, K. and Ishikawa, S. (1985b). Studies in utilisation of shart meat II:Qualities of products prepared from shark meat. Bull. Tokai. Reg. Fish. Res. Lab. 115(Feb):23-28. Nambudiry, D.D. (1980). Lipid oxidation in fatty fish: The effect Of salt content in the meat. J. Fd Sci. Technol. 17:176-178. NAS (1978). Post-harvest Food Losses in Developing Countries. National Research Council. National Academy of Sciences, Washington, DC. Nesje, R. (1986). Heat pump plants for fish drying. Scandinavian Refrigeration 15(4):187-190. Nitisewojo, P. and Hultin, H.O. (1986). Characteristics of TMAO degrading systems in Atlantic short finned squid (Illex illecebrosus). J. Food Biochem. 10(2):93-106. Niwa, E. (1976). X-ray and IR-studies on the dehydrated fish flesh. Agr. Biol. Chem. 40(4):807-808. Noda, H., Amano, H. and Horiguchi, Y. (1978). On the removal of volatile compounds contained in shark muscle. Bull. Mie. Univ. Sci. Fish. 5:165-171. Nozawa, E., Ishida, Y., Kadota, H. (1979). Combined effect of NaCl and temperature on TMA production by some bacteria isolated from chilled salted fish. Bull. Jap. Soc. Sci. Fish. 45(1) :1395-1399. Obanu, Z.A. (1983). Influence of muscle fat on the quality of intermediate fish. Fish Technol. 20:69-74. Obanu, Z.A. (1987). Use of simple fat analysis for evaluating quality changes in intermediate moisture fish stored at tropical temperature. Fishery Technol. 24:31-35. Obanu, Z.A., Ledward, D.A. and Lawrie, R.A. (1975a). The protein of367 intermediate moisture meat stored at tropical temperature. I. Changes in solubility and electrophoretic pattern. J. Fd Technol. 10:657-666. Obanu, Z.A., Ledward, D.A. and Lawrie, R.A. (1975b). The protein of intermediate moisture meat stored at tropical temperature. II. Effect of protein changes on some aspects of meat quality. J. Fd Technol. 10:667-674. Obanu, Z.A., Ledward, D.A. and Lawrie, R.A. (1976). The protein of intermediate moisture meat stored at tropical temperature. III. Differences between muscles. J. Fd Technol. 11:187-196. Odense, P.H., Leung, T.C., Green, W.A. and Dingle, J.R. (1969). Isolation of cod muscle tropomyosin by heat treatment. Biochim. Biophys. Acta, 188:124-131. OECD (1979). Review of Fisheries in OECD Member Countries 1978, Organisation for Economic Co-operation and Development, Paris. Ogasawara, K., Sekijo, I., Sunagawa, H. and Umemura, M. (1978). Studies on the toxigenic mould contamination of foodstuffs. 4. Mycoflora and frequency of toxigenic moulds in seasoned dried marine products. Rep. Hokkaido Inst. Pub. Health 25:26-31. Okafor, N. (1968) . Fungi associated with mouldy dried fish. Niger. J. Sci. 2:41-44. Okonkwo, P.O., Umerah, G. and Nwokolo, C. (1977). Procedures to reduce aflatoxin levels in common foods. W. Afr. J. Pharmocol. Drug Res. 4(1):621-623. Olcott, H.S. (1962). Marine products. In: Symposium on Foods: Lipids and Their Oxidation. Schultz, H.W., Day, E.A. and Sinhuber, R.O. (eds) AVI Pub. Inc. Westport. Olley, J., Pirie, R. and Watson, H. (1962). Lipase and phospholipase activity in fish skeletal muscle and its relationship to protein denaturation. J. Sci. Fd. Agric. 13:501-516. Olley, J.N. (1980). Structure and proteins of fish and shellfish. Part 1. In Advances in Fish Science and Technology, Connell, J.J. (ed). Fishing News Books Ltd, London. Olley, J.N. and Lovern, J.A. (1960). Phospholipid hydrolysis in cod flesh stored at various temperatures. J. Sci. Food Agric. 11:644-652. Onishi, H., Fuchi, H., Konomi, K., Hidaka, 0. and Kamekura, M. (1980). Isolation and distribution of a variety of halophilic bacteria and their classification by salt-response. Biol. Chem. 44(6) : 1253-1258. Ono, T. and Nagayama, F. (1959). Studies on the browning of fish flesh-I. Browning of white flesh by heat. Bull. Jap. Soc. Sci. Fish. 24(10):833-836. Opstvedt, J., Miller, R., Hardy, R.W. and Spinellik, J. (1984). Heat-induced changes in sulphydryl groups and disulphide bonds in fish protein and their effect on protein and amino acid digestibility in rainbow trout (Salmo gairdneri). J. Ag. Food Chem. 32:929-935. Orejana, F.M. and Embuscado, M.E. (1983). A new solar-agrowaste368 smoke-drier for fish and shellfish. In: Production and Storage of dried fish. Proc. of the workshop on the production and storage of dried fish. Kuala Lumpur, Nov. 1983: 133-147. Orr, A.T. (1967). Method and product for curing and packaging fish. British Patent 1.077.644. Osuju, F.N.C. (1976). The dried fish commerce in Nigeria: methods of processing, storage and marketing in relation to pest damage. Niger. Field 41(1):3-18. Otwell, W.S. and Hamann, D.D. (1979a). Textural characteristics of squid (Loligo pealei Lesuer): Scanning electron microscopy of cooked mantle. J. Food Sci. 44:1629-1635, 1634. Otwell, W.S. and Hamann, D.D. (1979b). Textural characterisation of squid (Loligo pealei_L.): Instrumental and panel evaluations. J. Food Sci. 44:1635-1643. Palombo, R., Gertler, A., and Saguy, I. (1984). A simplified method for determination of browning in dairy powders. J. Food Sci., 49:1609-1613. Parkin, K.L., Wells, M.J. and Brown, W.D. (1981). Modified atmosphere storage of rockfish fillets. J. Food Sci. 47:181-184. Parsons, S.E. and Patterson, R.L.S. (1986). Assessment of the previous heat treatment given to meat products in the temperature range 40°-90°C. Part 2: Differential scanning calorimetry, a preliminary study. J. Food Technol. 21:123-131. Partmann, W. (1967). Changes in proteins, nucleotides and carbohydrate during rigor mortis. Arch. Fishchwiss. 19:64-73. Pawar, S.S. and Magar, N.G. (1966). Chemical changes during frozen storage of pomfret, mackerel and sardines. J. Food Sci. 31:87-93. Pearson, D. and Muslemuddin, M. (1969). The accurate determination of total volatile nitrogen in meat and fish. J. Assoc. Publ. Anal. 7:73-82. Pechere, J.-F., Capony, J.A.P. and Demaille, J. (1974). Evolutionary aspects of the structure of muscular parvalbumins. Syst. Zool. 22:533-548. Peters, R.G. (1971). Diffusion in a medium containing a solvent and solutes, with particular reference to fish muscle. PhD thesis, Aberdeen Univ., Scotland. Petriella, C., Resnik, S.L., Lozano, R.D. and Chrife, J. (1985). Kinetics of deteriorative reactions in model food systems of high water activity:colour changes due to non-enzymatic browning. J. Food Sci. 50:622-626. Pieniezak, D. and Rakowska, M. (1974). Available methionine and cysteine in protein of processed food. Proc. 4th Int. Congr. Fd. Sci. Technol. 7a:64-66. Pitcher, T.J. and Hart, P.J.B. (1982). Fisheries Ecology. Croom Helm, London and Canberra, AVI Publishing Company Westport, Conn. Pitt, J.I. and Hocking, A. D. (1977). Influence of solute and369 hydrogen ion concentration on the water relations of some xerophilic fungi. J. Gen. Microbiol. 101:35-40. Pitt, J.I. and Hocking, A. D. (1985) . New species of fungi from Indonesian dried fish. Mycotoxon. 22:197-208. Pope, M.I. and Judd, M.D. (1977). Differential thermal analysis : a guide to the technique and its application. Heyden and Sons Ltd, London. Potter, N.H. (1978). Food Science. 3rd ed. AVI Publishing Co. Westport, Connecticut. Poulter, R.G. (1978). Quality changes in fish from the South China Sea. 2. Frozen storage of chub mackerel. Proc. IPFC, 18(3):330- 339. Poulter, R.G. (1980). Storage life of Sri Lankan cured fish. London, Tropical Products Institute, IFT/TPI report, cited by FAO, 1981. Poulter, R.G. and Disney, J.G. (1977). Development of novel products from tropical fish species. In Proceedings of the Second Annual Tropical and Sub-Tropical Fisheries Technological Conference of the Americas. College Station, Texas, Texas A&M University. TAMU-SG-78-101:43-58. Poulter, R.G., Doe, P.E. and Olley, J. (1982). Isohalic sorption isotherms. II. Use in the prediction of storage life of dried salted fish. J. Fd Technol. 17(2):201-210. Poulter, R.G., Ledward, D.A., Godber, S., Hall, G. and Rowlands, B. (1985). Heat stability of fish muscle proteins. J. Food Technol. 20:20 3-217. Poulter, R.G., Samaradivakera, B., Jayaweera, V., Samarasera, I.S.R. and Chinivasagm, N. (1981). Quality chagnes in three Sri Lankan fish species stored in ice. Trop. Sci. 23:155-168. Purnomo, A. (1985). Salting of Sardines, MSc Thesis, University of New South Wales, Australia. Quinn, J.R., Raymond, D.P. and Harwalker, V.R. (1980). Differential scanning calorimetry of meat proteins as affected by processing treatment. J. Food Sci. 45:1146-1149. Rao, B.Y.K. and Bandyopdhyay, C. (1983). Lipid composition of salted sun-dried Indian mackerel (Rastrelliger kanagurta) . J. Food Sci. Technol. (India) 20:62-64. Reay, G.A. and Shewan, J.M. (1949). The spoilage of fish and its preservation by chilling. Adv. Food Res. 2:343-398. Rechcigl Jr., M. (1983). CRC Handbook of Nutritional Supplements, vol. 1: Human Use. CRC Series in Nutrition and Food. CRC Press Inc., Boca Raton, Florida. Regenstein, J.M., Schlosser, M. A, Samson, A. and Fey, M. (1982). Chemical changes of trimethylamine oxide during fresh and frozen storage of fish. In Chemistry and Biochemistry of Marine Food Products, Martin, R.E., Flick, G.J.,Hebard, C.E. and Ward, D.R. (eds). AVI Publishing Co. Westport, Connecticut. Regenstein, J.M., Jauregui, C.A. and Baker, R.C. (1984). The370 effect of pH, polyphosphates and different salts on water retention properties of ground trout muscle. J. Food Biochem. 8(2) :123-131. Rehbein, H. and Karl, H. (1985). Solubilisation of fish muscle proteins with buffers containing sodium dodecyl sulfate. Z. Lebensm Unters Forsch. 180:313-318. Reynolds, T.M. (1965). Chemistry of nonenzymic browning II. Adv. Food Res. 14:167-283. Ritskes, T.M. (1975). Gas chromatographical determination of trimethylamine and dimethylamine in fish, fishery products and other foodstuffs. J. Fd. Technol. 10:221-228. Robinson, R.A. and Stokes, R.H. (1959). Electrolyte Solutions, 2nd ed. Butterworths, London. Rockland, L.B. and Nishi, S.K. (1980) . Influence of water activity on food product quality and stability. Food Technol. 34(4):42-51, 59. Rockland, L.B., Swarthout, D.M. and Johnson, R.A. (1961). Studies on English (Persian) walnuts (Juglans regia). Ill Stabilisation of kernels. Food Technol. 15:112-116. Rodger, G., Hastings, R., Cryne, C. and Bailey, J. (1984b). Diffusion properties of salt and acetic acid into herring and their subsequent effect on the muscle tissue. J. Food Sci. 45:714-720. Rodger, G., Weddle, R.B., Craig, P. and Hastings, R. (1984a). Effect of alkaline protease activity on some properties of comminuted squid. J. Food Sci. 45:117-123. Ronsivalli, L.J. (1978). Sharks and their utilisation. Marine Fish. Rev. 40(2):1-13. Ross, K.D. (1975). Estimation of water activity in intermediate moisture foods. Food Technol. 29:26-34. Rubbi, S.F., Md. Muslemuddin, Janan, S.S. and Begum, M. (1983). Storage of solar tent dried pomfret (Stomateus spp.). The production and storage of dried fish. Proc. of a Workshop on the production and storage of dried fish. Universiti Pertanian Malaysia, Serdang, Malaysia, 2-5 November 1982. FAO Fish. Rep. 279, suppl. by James, D. (ed), 93-101. Rustad, T. and Nesse, N. (1983). Heat treatment and drying of capelin mince. Effect on water binding and soluble protein. J. Fd Sci. 48:1320-1322, 1347. Ryder, J.M. (1985). Determination of adenosine triphosphate and its breakdown products in fish muscle by high-performance liquid chromatography. J. Agric. Food Chem. 33(40):678-680. Ryder, J.M. (1985). Determination of adenosine triphosphate and its breakdown products in fish muscle by high-performance liquid chromatography. J. Agric. Food Chem. 33(40):678-680. Ryder, J.M., Buisson, D.H., Scott, D.N. and Fletcher (1984). Storage of New Zealand Jack Mackerel in ice: chemical, microbiological and sensory assessment. J. Food Sci. 49(6):1435- 1477 . Ryu, H-S. and Lee, K-H (1985) . Effect of heat treatment on 371 the in vitro protein digestibility and trypsin indigestible substrate (TIS) contents in seafoods. J. Korean Soc. Food Nutr. 14(1):1-12. Sachithananthan, K. (1976). The dried fish trade in Sri Lanka. Rome, FAO, FII:CP/76/5. Saiki, M., Fang, S. and Mori, T. (1959) as cited by Suzuki (1981). Sakaguchi, M., Murata, M. and Kawai, A. (1984). Changes in free amino acid contents in juvenile mackerel Scomber japonicus muscle during ice storage. Bull. Jap. Sci. Sci. Fish. 50(2):323-329. Salwin, H. (1959) . Defining minimum moisture contents for dehydrated foods. Food Technol. 13:594-596. Sato, Y. (1960) . Estimation of the freshness of bottom fish. Bull. Jap. Soc. Sci. Fish. 26:312-316. Schaller, D.R. and Powrie, W.D. (1972). Scanning electron microscopy of heated beef, chicken and rainbow trout muscles. Can. Inst. Fd Sci. Technol. J. 5:184-190. Schobell, T., Tannenbaum, S.R. and Labuza, T.P. (1969). Reaction at limited water concentration. I. Sucrose hydrolysis. J. Food Sci. 34:324-329. Schutz, H.G. (1954). Colour in relation to food preference. In Colour in Foods, a Symposium. Farrell, K.T., Wagner, J.R., Peterson, M.S. and Mackinney, G. (eds) Natl. Acad. Sci., Natl. Res. Council, Washington, D.C. Scott, W.J. (1957). Water relations of food spoilage micro organisms. Adv. Food Res. 7:83-127. Sears, D.F. and Eisenberg, R.M. (1961). A model representing a physiological role of CO2 at the cell membrane. J. Gen. Physiol. 44:869-887. Sen, D.P. and Chaluvaian, G.L. (1968). Seasonal variation in the amount and characteristics of the oil of oil-sardine (Sardinella longiceps) fish. J. Food Sci. Technol. 5(2):117-122. Sen, D.P. and Lahiry, N.L. (1964). Studies on the production of better quality salt-cured and sundried mackerel (Rastrelliger kanagurta). Food Technol. 13:107-110. Sen, D.P., Arundaswamy, B.,Iyengar, N.V. and Lahiry, N.L. (1961). Studies on the storage characteristics and packaging of sundried salted mackerel (Rastrelliger canagurta__Cuv.) Food Sci. Mysore, 10:148-53. Senesi, E., Bertolo, G., Torregiani and Di Cesare (1980). The utilisation of mediteranean sardine by means of smoking. In: Advances in Fishery Technology, Connell, J.J. (ed) . Sera, H. and Ishida, Y. (1972) . Bull. Jap. Sco. Sci. Fish. 33:853- 858 . Shaw, B.G. and Shewanh, J.M. (1968). Psychrophilic spoilage bacteria of fish. J. Appl. Bact. 31:89-96. Sheikh Abdus-Salam and Shah, F.H. (1974). Effect of heat on the iri312 vitro digestibility of fish protein. Pakistan J. Sci. Ind. Res. 17(4-5):136-13. Shepherd, A.D. (1960) as cited in Hastings et al. (1985). Shewan, J.M. (1962). The bacteriology of fresh and spoiling fish and some related chemical changes. In Recent Advances in Food Science, vol.l Commodities:167-19. Shewan, J.M. (1977) . The bacteriology of fresh and spoiling fish and the biochemical changes induced by bacterial action. Proc. Conf. Handl. Process. Market. Trop. Fish, Tropical Products Institute, London. Shewan, J.M. (1949). Some bacteriological aspects of the handling, processing and distribution of fish. J. Royal Sanit. Inst., 59:394-421. Shewan, J.M. and Jones, N.R. (1957). Chemical changes occurring in cod muscle during chill storage and their possible use as objective indices of quality. J. Sci. Food Agric. 8(8) :491-498. Shewfelt, R.L. (1981). Fish muscle lipolysis - a review. J. Food Biochem. 5:79-100. Shiau, C.Y. and Chai, T. (1985). Smoked dogish processing and its refrigerated storage stability. J. Food Sci. 50:1348-1350. Sidaway, E.P. and Balaingham, M. (1971). Fish processing industry in W. Malaysia. Malaysia, Ministry of Agriculture and Fisheries. Sidhu, G.S., Montgomery, W.A. and Brown, M.A. (1974). Postmortem changes and spoilage in rock lobster muscle. II. J. Fd. Technol. 9: 371-380. Sigurdsson, G.J. (1947). Comparison of chemical tests of the quality of fish. Anal. Chem. 19:892-902. Sikorski, Z.E. (1980). Structure and proteins of fish and shellfish. Part 2. In Advances in Fish Science and Technology, Connell, J.J. (ed). Fishing News Books Ltd. London. Sikorski, Z.E. and Kolodziejska, I. (1986). The composition and properties of squid meat. Food Chem. 20:213-224. Sikorski, Z.E., Olley, J. and Kostuch, S. (1976). Protein changes in frozen fish. Crit. Rev. Food Sci. Nutr. 9:97-129. Sim, T.S., Teo, T. and Sim, T.F. (1985) . A note on the screening of dried shrimps, shrimp paste and raw groundnut kernels for aflatoxin-producing Aspergillus flavus. J. Appl. Bacteriol. 59:29-34. Simidu, U. and Simidu, W. (19 60) . Comparative study on coagulability of fish muscle protein. Bull. Jap. Soc. Sci. Fish., 26:1099-1106 . Simidu, W., Hibiki, S., Sibata, S., and Takeda, K. (1953). Studies on muscle of aquatic animals XVI. Distribution of extractive nitrogens in muscle of several kinds of gastropod. Bull. Jap. Soc. Sci. Fish. 19(8):871-876. Simpson, T.H. (1965). The chemistry of woodsmoke. In: Fish 373 as Food, Vol. 3. Borgstrom, G. (ed). Academic Press, New York. Skujins, J.J. and McLaren, A. D. (1967) Enzyme reaction rates at limited water activities. Science 158:1569-1570. Slavin, J.W. (1968). Frozen fish: characteristics and factors affecting quality during freezing and storage. In: The Freezing and Preservation of Foods. Tressler, D.K., Van Ardsel, W.B. and Copley, M. J. (eds) . AVI Publishing Co. Westport, Connecticut. Soevik, T. and Braekkan, O.R. (1979). as cited by Suzuki (1981) . Spinelli, J. and Koury, B. (1979). Nonenzymatic formation of dimethylamine in dried fishery products. J. Agric. Food Chem. 27:1104-1108 . Spinelli, J. and Koury, B.J. (1981). Some new observations on the pathways of formation of dimethylamine in fish muscle and liver. J. Agric. Food Chem. 25:327-331. Spinelli, J., Koury, B. and Miller, R. (1972). Approaching to utilization of fish for the preparation of protein isolates. J. Food Sci. 37:599-603. Sripathy, N.V. (1983). Some recent research on traditional fish products in the IPFC region. In: The Production and Storage of Dried Fish. Proc. of the Workshop on the Production and Storage of dried fish. Universiti Pertanian Malaysia, Serdang (Malaysia) 2-5 November, 1982. James, D. (ed) FAO Fish. Rep. (279) Suppl. Stabursvik, E. and Martens, H. (1980). Thermal denaturation of proteins in post rigor muscle tissue as studied by differential scanning calorimetry. J. Sci. Food Agric. 31:1034-1042. Stabursvik, E., Foystein, R. and Fretheim, K. (1984). Myosin denaturation in pale soft and exudative porcine muscle tissue as studied by differential scanning calorimetry. J. Sci. Food Agric. 35:240-244. Stanley, D.W. and Hultin, H.O. (1982). Quality factors in cooked North Atlantic Squid. Can. Inst. Food Sci. Technol. 15(4) : 277-282. Stanley, D.W. and Hultin, H.O. (1984). Amine and formaldehyde production in North American squids and their relation to quality. Can. Inst. Food Sci. Technol. 17(3):157-162. Stanley, D.W. and Smith, H.O. (1984). Proteolytic activity in North American squid and its relation to quality. Can. Inst. Food Sci. Technol. J. 17:163-167. Stansby, M.E. (1962). Proximate composition of fish. In Fish in Nutrition. Heen, E. and Kruezer, R. (eds) Fishing News Books. London. Stansby, M.E. (1982). Properties of fish oil and their 374 application to handling of fish and to nutritional and industrial uses. In: Chemistry and Biochemistry of Marine Food Products, Martin, R.E, Flick. G.J., Hebard, C.E. and Ward, D.R. (eds) AVI Publishing Co. Westport, Conn. Statham, J.A. (1984) . Modified atmosphere storage of fisheries products: the state of the art. Food Technol. Aust. 36:233-239. Storey, R.M., Davis, H.K., Owen, D. and Moore, L. (1984). Rapid approximate estimation of volatile amines in fish. J. Fd. Technol. 19:1-10. Strommen, I. (1982). New equipment in fish drying. Proc. of the Third International Drying Symposium (Norway), Vol. 1, 295-299. Sudhakaran, R. and Sudhakaran, N.S. (1985). Studies on the preparation of salted and dried minces from threadfin bream (Nemipterus japonicus) and Indian oil sardine (Sardinella longiceps). In Spoilage of Tropical Fish and Product Development. Proceedings of a Symposium held in conjunction with the 6th IPFC Working Party on fish technology and marketing. Reilly, A. (ed), RMIT, Melbourne, Australia. FAO Fish. Report 317:338-342. Sullivan, B., Bonaventura, J., Bonaventura, C., Pennel, L., Elliot, J., Boyam, Ryder, J.M., Buisson, D.H., Scott, D.N. and Fletcher (1984). The structure and evolution of parvalbumins. I. The amino acid compositional studies of parvalbumins from four perciform species. J. Mol. Evol. 5:103-116. Tagaki, M., Murayama, H. and Endo, S. (1967). Trimethylamine and trimethylamine oxide contents of fish and marine invertebrates. Hokkaido Daigaku Suisan Gakubu Kenkyu Iho. 18(3) :2 68-270. Takahashi, T. (1965). Squid meat and its processing. In: Fish and Food. Vol. 4. Borgstrom, G. (ed) Applied Science, New York and London. Takahashi, T. and Takei, M. (1956). Studies on processing of squid meat. Bull. Tokai Regional Fisheries Research Lab. 14: 31-90. Takama, K., Sugiura, S., Iseya, Z., Wakayama, E. and Furiu, T. (1985). Effect of salting and subsequent pickling on the quality of Masu Salmon fillets. Bull. Fac. Fish. Hokkaido Univ. 36(3):147-156. Tanikawa, E. (1935). Studies on measuring freshness of fish and shellfish meat. J. Soc. Ind. Fisheries (Japan) 3:267-296. Tanikawa, E. (1938a). Determination of volatile bases as a mean of detecting the beginning of decomposition of the flesh of fish and shellfish. Bull. Soc. Sci. Fisheries. 17:7-14. Tanikawa, E. (1938b). The significance of the quantitive determination of volatile base as a means of determination of the primary stages of decomposition of meat of fish and of shellfish. Mem. Alumni Soc. Hakodate College Fisheries, Hakodate, Japan. Tannenbaum, S. (1974). Industrial processing. In: Nutrients 375 in Processed Foods: Proteins. White, P.L. and Fletcher, D.C. (eds) Amer. Med. Assoc. Publ. Sci. Grp, Littleton, Mass. Tannenbaum, S.R. (1976). Interaction of chemical and biological changes in foods and their influence on quality. In Principles of Food Science, Part I, Food CHemistry. Fennema, O.R. (ed) Marcel Dekker, New York. Tapiador, D.D. and Carroz, J.E. (1963). Standards and requirements for fish handling, processing, distribution and quality control. FAO Fish. Rep. 9, FIT/R9. Tarladgis, B.G., Watts, B.M. and Younathan, M.T. (1960). A distillation method for the quantitative determination of malonaldehyde in rancid foods. J. Am. Oil Chem. Soc. 37:44- 48. Tarr, H.L.A. (1940). Specificity of triaminoxydase. J. Fish. Res. Bd. Can. 5:187-196. Tarr, H.L.A. (1942). Volatile base content of certain Pacific Coast fish prior to and subsequent to canning. Progr. Repts . Pacific Coast Stas. Fisheries Research Board Can. No. 52:24-26. Tarr, H.L.A. (1945). Water, microorganisms and volatile bases in dehydrated fish. J. Fish. Res. Bd. Can. 5:303-310. Tarr, H.L.A. (1962). In Fish as Food. Vol. 2, Borgstrom, G. (ed) Academic Press, London & New York. Tarrant, P.V. (1982). Muscle proteins in meat technology. In Food Proteins, Fox, P.F. and Condon, J.J. (eds) App. Sci. Pub. London and New York. Taylor, R.W.D. and Evans, N.J. (1982). Laboratory evaluation of four insecticides for controlling Dermestes maculatus Dedeer on smoke dried fish. Int. Pest Control, March/April. Tichivangana, J.Z. and Morrissey, P.A. (1984). Factors influencing lipid oxidation in heated fish muscle systems. Ir. J. Fd. Sci. Technol. 5:47-57. Tiffney, P. and Mills, A. (1982). Cooling the catch - a package system for chilled fish. Fish. News Int. 21(4): 20- 21. Tillmans, J. and Otto, R. (1924). Uber den Nachweis beginnender Fleisch-faulnis. Z. Untersuch. Nahr.-u. Genussm. 47:25-37. Tokunaga, T. (1975). Thermal decomposition of trimethylamine oxide in muscle of some marine animals. Bull. Jpn. Soc. Sci. Fish. 41 (5) :535-546. Tokunaga, T., Iida, H. and Miwa, K. (1977). G.C. Analysis of amines in fish. Bull. Jpn. Soc. Sci. Fish. 43:219-227. Townsend, J.F., Cox, J.K. and Sprouse, R.F. (1971). Fungal flora of South Vietnamese fish and rice. J. Trop. Med. Hyg., 74(4):98-100. Toyama, Y. and Kaneda, T. (1962). Nutritive aspects of fish 376 oils. In Fish as Food, vol. 2, Borgstrom, G. (ed) , Academic Press, New York. Toyomizu, M., Hanaoka, K., Satake, K. and Nakagawa, H. (1977). Effect of storage temperatures on accumulation of glycerylphosphorylcholine and decomposition of phosphatidylcholine in fish muscle during cold storage. Bull. Jap. Soc. Sci. Fish. 43:1181-1187. Toyomizu, M., Matsumura, Y. and Tomiyasu, Y. (1963). Lipid oxidation and protein denaturation in freeze-dried fish. Bull. Jap. Soc. Scient. Fish. 29:854-859. Troller, J.A. (1980) . Influence of water activity on microorganisms in foods. Food Technol. May:76-82. Troller, J.A. and Christian, J.H.B. (1978). Water activity and food. Academic Press, New York. Tsuchiya, T. (1961). Biochemistry of fish oils. In: Fish as Food, Vol. 1, Borgstrom, G. (ed). Academic Press, New York. Udarbe, M.A., Meercada, C.C., Santos, R.VC., Lozada, A. F. and Gonzales, J.M. (1985) . Protein quality evaluation of somke fresh and processed fish. Asean Food J. 1 (3):113-119. Vaisey, E.B. (1956). The nonenzymic reduction of trimethylamine oxide to trimethylamine, dimethylamine, and formaldehyde. Can. J. Biochem. Physiol. 34(6):1085-1090. Valsam, A.P., Nair, M.R. and Rao, S.V.S. (1961). Propionic acid as a preservative for cured fish products. J. Sci. Ind. Res. 20(9):315-317. Van den Broek, C.J.H. (1965). Fish Canning. In: Fish as Food, vol. 4. Borgstrom, G. (ed). Academic Press,New York. Van Veen, A.G. (1953). Fish preservation in Southeast Asia. Adv. Food Res. 4:209-229. Velankar, N.K. (1952) . Moisture, salt, trimethylamine and volatile nitrogen contents and bacterial counts of salt-cured marine fish. J. Sci. Ind. Res. Sect. A, 11:359-360. Veranth, M.F. and Robe, K. (1979). CC^-enriched atmosphere keeps fish fresh more than twice as long. Food Process. 40(4) :16-79. Villadsen, A. and Flores, F. (1983). Low-cost agrowaste fish drier development (Temperature profiles of different types of fish driers). Proceedings of the Workshop on the Production and Storage of Dried Fish, Universiti Pertanian Malaysia, Malaysia, 2-5 November 1982. FAO Fisheries Report No. 279 Supplement. Villota, R. and Saguy, I. and Karel, M. (1980). Storage stability of dehydrated food-evaluation of literature data. J. Food Qual. 3:123-212. Viswanathan Nair, P.G. and Gopakumar, K. (1978). Fatty acid compositions of 15 species of fish from tropical waters. J. Food Sci. 43:1162-1164 . Vlieg, P. (1984). Proximate composition of New Zealand squid 377 species. N.Z. J. Sci. 27:145-150. Von Hippie, P.A. and Wong, K.Y. (1964). Neutral salts: the generality of their effects on the stability of macromolecular conformations. Science 145:577-580. Voskresensky, N.A. (1965). Salting of herring. In: Fish as Food, vol. 3, Borgstrom, G. (ed). Academic Press, New York. Vyncke, W. (1968) . Determination of ammonia in dressed thornbackray as a quality test. J. Fd. Technol. 113:37-44. Vyncke, W. (1975) . Evaluation of the direct thiobarbituric acid extraction method for determining oxidative rancidity in mackerel (Scomber scombrus L.) Fette. Seifen Austrichm. 77:239-241. Wada, S. and Koizumi, C. (1986). Changes in histamine contents during processing of rice-bran pickles of sardine. Bull. Jap. Soc. Sci. Fish. 52(6):1035-1038. Waller, P.F. (1980b). Spoilage and spoilage indicators in frozen shark. Food Technol. Aust. 32(4):184-187. Waller, P.F. (1980a). Spoilage and spoilage indicators in shark held in ice. Food Technol. Aust. 32(3): 161-166. Wang, M.Y. and Brown, W.D. (1983). Effects of elevated C02 atmosphere on storage of freshwater crayfish (Pacifastacus leniusculus). J. Food Sci. 43:158-162. Ward, D. and Wainwright, S.A. (1972). Locomotory aspects of squid mantle structure. J. Zool. Lond. 167:437-449. Watabe, S., Ochiai, Y., Kanoh, S.and Hashimoto, K. (1983). Proximate and protein composition of requim shark muscle. Bull.. Jap. Soc. Sci. Fish. 49 (2): 265-268. Watanabe, K. (1965-66). Handling and keeping quality of iced kariba bream, Tilapia mortimeri, Trewavas (syn. T. mossambica, Peters). Fish. Res. Bull. Zambia, 4:59-64. Watanabe, K. and Choongo, S.M. (1971). Proposed Zambian standards for whole dried Lake Tanganyika sardine (Limnothrissa miodon and Stolothrissa tanganicae). Afr. J. Hydrobiol. Fish., 2(1):125-127. Watanabe, K. and Dzekedzeke, 0. (1970) . Recommended Zambian standards for grades for split, dried, smoked bream (Tilapia from Volta Lake, Ghana). Bull. Jap. Soc. Sic. Fish., 42(1):109-121. Waterman, J.J. (1976). The production of dried fish. FAO Fish. Tech. Pap. 160 Watson, D.W. (1939). Studies of fish spoilage. IV. The bacterial reduction of trimethylamine oxide. J. Fish. Res. Bd. Can. 4:252-280. Weckel, K.G. and Wosje, D. (1966) . Brine salting of Great Lakes chub (Leucichthys hoyi) for smoking. Res. Rep. 24, Extension Service, College of Agriculture, Univ. of Wisconsin. Weinber, Z., Regenstein, J.M. and Lilliford, P.J. (1984). The 378 effects of salts on thermal transition curves of cod muscle. J. Food Biochem. <9:335-339. Wheaton, F.W. and Lawson, T.B. (1985). Processing aquatic food products. John Wiley & Sons, New York. Wheeler, K.A., Hocking, A.D., Pitt, J.I. and Anggawati, A.M. (1986). Fungi associated with Indonesian dried fish. Food Micro. 3:351-357. Whittle, K.J., Borderiaas-Juarez, A.J., Howgate, P., Keay, J.N., Mills, A., Young, K.W. (1980). Some factors affecting the properties of minced fish. Third National Technical Seminar on Mechanical Recovery and Utilisation of Fish Flesh, Raleigh, N. Carolina, USA, Martin, R.E. (ed) National Fisheries Institute, Washington, DC. Wiggin, K. and Krzynowek, J. (1983). Identification of frozen, cooked shellfish species by agarose isoelectric focussing. J. Assoc. Off. Anal. Chem. 66(1):118-122. Williams, J.C. (1976). Chemical and non-enzymic changes in intermediate moisture foods. In Intermediate Moisture Foods. Davies, R., Birch, G.G., Parker, K.J. (eds). Applied Science Publishers, London. Wills, R.B.H. and Greenfield, H. (1982). Laboratory manual for food analysis. University of New South Wales, Australia. Wood, A. J. and Baird, E.A. (1943). Reduction of trimethylamine oxide by bacteria 1. The Enterobacteriaceae. J. Fish. Res. Bd. Can. 6(2) :194-201. Wood, C.D. (1982). Technical options for processing small fatty pelagic fish into cured products. In Proceedings of the FAO Expert Consultation on Fish Technology in Africa, Casablanca, Morocco, 7-11 June 1982; FAO, Rome 1982. Wood, J.D. (1959a). Lipase activity in lingcod muscle preparations. Can. J. Biochem. Physiol. 37:937-943. Wood, J.D. (1959b) . Distribution of a lipase enzyme in lingcod fillets and the effect of low temperature storage on its activity. J. Fish. Res. Bd. Canada, 16:155-151. Woolfe, M.L. (1975). The effect of smoking and drying on the lipids of West African herring (Sardinella spp.) J. Food Technol. 10:515-522. Wright, D.J. (1978). Differential scanning calorimetry - its application to the study of meat proteins. J. Sci. Food Agric. 25:1088. Wright, D.J. (1982). Application of scanning calorimetry to the study of protein behaviour in foods. In: Developments in Food Proteins, vol. 1. Wright, D.J. and Wilding, P. (1984). Differential scanning calorimetric study of muscle and its proteins: myosin and its subfragments. J. Sci. Food Agric. 35:357-372. Wright, D.J., Leach, I.B. and Wilding, P. (1977). 379 Differential scanning calorimetry studies of muscle and its constituent proteins. J. Sci. Food Agric. 25:557-564. Wu, M.T. and Salunkhe, D.K. (1978). Mycotoxin producing potential of fungi associated with dry shrimps. J. Appl. Bacteriol. 45:231-238. Wuttijumnong, P. (1987). Salting kinetics of sardines. PhD thesis, University of New South Wales, Australia. Yamada, K. (1967). Occurrence and origin of trimethylamine oxide in fishes and marine invertebrates. Bull. Jap. Soc. Sci. Fish. 3(6): 591-603. Yamamoto, I. and Ishimoto, M. (1977). Anaerobic growth of Escherichia coli on formate by reduction of nitrate, fumarate and trimethylamine N. oxide. Z. Allg. Mikrobiol. 17(3):235- 242 . Yanez, E., Ballester, D. and Donoso, G. (1970). Effect of drying temperature on quality of fish protein. J. Sci. Fd. Agric. 21:426-428. Yang, C-T., Jhaveri, S.N., Constantinides, S.M. (1981). Preservation of grayfish (Squalus acanthias) by salting. J. Food Sci. 46:1646-1649 . Yen, G-C. and Lai, Y-H. (1987). Influence of antioxidants on Maillard browning reaction in a casein-glucose model system: A research note. J. Food Sci. 52(4):1115-1116. Younathan, M.T., Oon, J.K. and Yusof, R.B.M. (1983). Control of heat induced oxidative rancidity in refrigerated shark and mackerel. J. Food Sci. 53:176-178. Young, K.W. and Whittle, K.J. (1985). Colour measurement of fish minces using Hunter Lab values. J. Sci. Food Agric. 36: 38 3-392 . Young, R.H. Coria, E., Cruz, E. and Baldry, J. (1979). Development and acceptability testing of a modified salt fish product prepared from shrimp by-catch. J. Food Technol. 14:509-519. Yu, C-T. and Cruess, W.V. (1951). A study of several factors in the salting and smoking of fish. Canner, 113(6):12-14, 16, 18. Yu, S.Y. (1985). Comparison of sun-dried and oven-dried salted fish. In: Spoilage of Tropical Fish and Product Development . FAO Fish. Rep. 317:328-331. Yu, S.Y., Siaw, C.L. and Idrus, A.Z. (1982). The application of tehcnology to the processing of dry-salted fish in peninsular Malaysia: comparison of sun-dried and oven-dried fish. J. Fd Technol. 17:211-218. Zain, A.M. and Yusof, S. (1983). Salt and moisture contents of Malaysian dried fish. In The Production and Storage of Dried Fish. Proc. of the Workshop on the Production and Storage of Dried Fish. Universiti Pertanian, Serdang, Malaysia. 2-5 November. FAO Fish. Rep., (279) Suppl.:216- 222 . Zaitsev, V., Kizevetter, I., Lagunov, L., Makarova, T., 380 Minder, L., Podsevalov, V. (1969). Salting and Marinading. In Fish Curing and Processing. Translated from the Russian by A. De Merindol, MIR publishers, Moscow. Zugarramurdi, A. and Lupin, H.M. (1980). A model to explain observed behaviour on fish salting. J. Fd Sci. 45:1305-1311. nm 538 at Absorbance Standard curve for TBA no. APPENDIX 2 Calibration for gglatinization tempratures and enthalpy. Table 1: Calibration data of Biphenyl ( A H = 120.6 J/g) Weight of Biphenyl (mg) Area (cm ) 0.40 1.68 0.80 2.99 1.20 4.67 1.60 6.11 nig of B iphenyl Appendix 2: Area observed and The in g e l a t i n i z a t i o n mg r e l a t i o n s h i p ( Biphenyl cm by DSC. 2 ) used between Y * of » » wheat for 0.99 0.27x area s t a n d a r d i z a t i o n starch - (cm 0.27 2 ) Appendix 3: Sensory evaluation Salted Fish Taste Test Name: You are presented with 5 samples of salted fish. Please indicate how much you like colour aroma saltiness and texture of each fish and then indicate how acceptable you find each fish by placing a mark on th^ appropriate scale for each sample. Remember to use the water palate cleanser between each sample and to wait 1 minute before you taste the next sample. Colour Sample code: like extremely dislike extremely Please give any comments you wish: Sample code: 359 ) ) 627 ) the colour is too dark ) 932 ) 576 light gray, glossy look good 748 the colour is dull Aroma Samples code: like esxtremely dislikce extremely ______ Please; give any comments you wish: Sample; code: 359 g 'Oo*{ 627 bland 932 good 576 ) ) fishy 748 ) Saltiness Sample code: like extremely dislike extremely ------ Please give any comments you wish: Sample code: 359 not so salty, just enough 627 too salty 932 very salty 576 too salty 748 too salty Texture Sample code: liJxe extremely dislike extremely______ Please give any comments you wish: Sample code: 359 good, not tough, not too soft 627 a bit tough 932 good 576 quite good 748 too tough Overall acceptability Sample code: like extremely dislike extremely Please give any comments you wish: Sample code: 359 like the taste, not too salty, also it has good texture 627 ) ) all right 932 ) 576 too salty, 748 the meat is broken