Process and Quality Characteristics of Ocean Pout Surimi

··•·••• ACKNOWLEDGEMENT . TABLE OF CONTENTS • I • e •• e e • 9 a • I I I I I e e I I I I I I I I I I I I I I I I I I I I I I I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page

moisture and salt levels, ii) thermal gelation in a single as well as a two-stage heating process, iii) optimization of gel formimg properties and freeze-thaw stability of blended surimi~ iv) texture-modifying effect of different nonfish proteins in the ocean pout surimi and its applicability as a binder in formed products, and v) use of ocean pout in the turkey as an extender. Compressive force (cohesiveness) and penetration force (rigidity) were measured for the evaluation of gel forming properties of surimi, and expressible moisture for water binding ability as well as freeze-thaw stability of surimi gel. Sensory quality of ocean pout incorporated turkey roll was also evaluated.
The 74% moisture level produced the most cohesive gel while NaCl progressively increased cohesiveness and rigidity, accompanied by a general decrease in expressible moisture.   and NaCl levels, ii) thermal gelation in a single-and a two-sta~e heating process, iii) opt imi za t ion of gel forming properties and freeze-thaw stability of blended surimi, and iv) texture-modifying effect of different nonfish proteins in the ocean pout surimi and its applicability as a binder in formed products, and v) suitability of ocean pout surimi in a turkey roll as an extender. The sensory quality of the ocean pout surimi was evaluated alone and in combination with turkey meat as it was formed into a turkey roll.  (Lee, 1984).
u. S. seafood processors have since then expressed interests in sharing the the world surimi market, which has been dominated by the Japanese (King,19 8 5 ) .
The importance of fabricated seafoods in supplementing and substituting for conventional sources has been reflected 4 in the past and pre sent market growth.
Chronic shortages of raw materials, increasing costs and increased per capita consumption has forced seafood processors to focus on fabricated products such as shrimp, lobster, and fish portions to improve the market share. The process of fabrication and extrusion offers the process or an opportunity to utilize a variety of cheap supply of raw materials to produce profitable products.
Institutional acceptance of fabricated seafoods was \ first recognized in 1973 (Katz, 1974 to the success in the manufacture of finished products.
The gel forming ability of surimi is known to vary with the functional characteristics of these proteins which is 5 species-dependent Hashimoto, 1985;Shimizu, 1974 Initial preliminary studies indicated that the surimi made from ocean pout possessed unique textural and flavor characteristics and can be processed into a product that cannot be produced by red hake or Alaskan pollack which is conventionally being used. Presently, use of ocean pout is limited to fillets and nobody has explored its potential application for production of surimi and surimi-based products.
For any species to be used for producing a good quality surimi, one must understand the species-dependent thermal gelation behavior of myofibrillar proteins, the relationship of moisture to freeze-thaw stability, and the effect of ingredients incorporated in formulation.
With respect to the supply of surimi, a continued decline in the stock of Alaskan pollack has been noted 6 (Lee, 1986).
Blending of surimi from different sources is expected to reduce the production costs as well as improve the supply problem (Lanier, 1985). It is therefore proposed in this study to evaluate ocean pout for its suitability for surimi product ion, and to determine how effectively ocean pout can be used in blended products as well as its applicability as a binding material in the production of formed meat products. processing temperature of paste below 10 C in order to prevent any possible changes in gel forming properties of myofibrillar proteins as a result of increase in temperature (Lee, 1984). Surimi gels were prepared by stuffing the paste into 25mm cellulose casings using an extruder. The extruder worked on the principle of volumetric displacement from a hopper to an appropriate shaping spout by the action of a water powered piston. Volume and speed were adjusted to produce uniformity of the extrudate. The extrudate in casings was then immersed into a water bath at 90 C and heat set for 40 min for further gel testing or steam cooked for 20 min for sensory evaluation, removed, cooled in running tap water for 5 min and placed at room temperature for 24 hr before ,testing. Following the procedures for gel testing proposed by Lee (1984a), compressive force (cohesiveness), penetration force (rig id i ty) and expressible mo is tu re were measured using an Instron testing machine.

Effects of moisture levels
After a storage period of two weeks, surimi was thawed overnight at 4 C and chopped for 10 min into paste by adding 2. 0% NaCl and enough water to bring mo is tu re level to 7 2%, 75%, 76%, 78% and 80%, respectively. The paste was stuffed into 25mm cellulose casings, cooked at 90 C for 40 min, cooled in running tap water for 5 min, and left at room temperature for 24 hr before testing for gel properties.
Results were used to determine the moisture-dependent gel setting behavior of ocean pout surimi.

Effects of two stage heating process
Heat-induced gel setting behavior was determined by subjecting the extruded paste to a single-stage as well as two-stage heating. The half-thawed surimi was chopped with 2.0% NaCl on a surimi weight basis and enough water to adjust moisture level to 78%.
and treated as follows.
The paste was divided into two lots The first lot was subjected to a single-stage cooking at 90 C for 30, 40, 50 and 60 min to ' determine the cooking time dependency of gel set ting. The second lot was heat-set in a two-stage heating process for 20 min at 40 C, 50 C and 60 C, followed by cooking for 30 min at 90 C to determine the temperature-dependency of gel setting.
The resulting gels were tested as described previously.

Effects of NaCl concentration
Ocean pout surimi were thawed overnight at 4 C and chopped with varying amounts of NaCl and enough water to adjust moisture level to 78%. NaCl was added at 1%, 1.5%, 2.0%, 2.5% and 3% on a surimi weight basis. The paste was extruded into 25mm cellulose casings and cooked at 90 C for 40 min. After equilibrating to room temperature overnight, the gels were tested for the textural properties. surimi gels wer e prepared in the usual manner and were \ divided into two groups. Group one was immediately subjected to Instron testing following a 24 hr period of equilibration at room temperature. Group two was subjected to three freeze-thaw cycles before testing in order to evaluate freeze-thaw stability in terms of changes in the expressible moisture and physical parameters of the gels (Lee, 1984).
For each parameter, a set of duplicate gels were tested before and after freeze-thaw storage. Each cycle consisted of three days in the freezer at -20 C and one day thawing at 4 C. The final thawing was accomplished at room temperature in order to equilibrate gels to room temperature. The scores used by the panel is ts were 1 = low and 9 = high for all sensory parameters. Analysis of variance were used to analyze the effect of treatment according to SAS (1982).   The solubilized proteins undergo changes upon heat treatment to form a gel-like mass whose visco-elastic characteristics become pronounced as NaCl is increas~d, yielding products with increasing gel strength.

r---------------------------------r
The manner in which NaCl affects proteins has been a subject of numerous studies. (Siegel and Schmidt, l 979a;Yasui et al., 1980;MacFarlene et al., 1986). It is concluded through these studies that when myosin and actomyosin are heated in high ionic strength salt solutions, the proteins formed a coherent three-dimensional network of fibers which was necessary for the proteins to bind (Siegel and Schmidt, 1979a). At a low level of NaCl, a weakened gel accompanied with the production of spongy-like mass was evident (Siegel and Schmidt, 1979a). Addition of NaCl produced marked responses in all of the observed textural characteristics of ocean pout surimi gel. Increasing the level of NaCl led to significant ( P<O. O 5) decreases in the expressible mo is tu re (Fig. 3). Ambient NaCl levels of 2% and 2.5% were not sign if ican tly different in their ef feet on the moisture of the paste, expressible moisture, cohesiveness and rigidity.
An inverse relationship was obtained between expressible moisture and the NaCl level. The cohesiveness of ocean pout surimi was markedly affected by salt concentration (Fig. 4).
A positive response was observed by increasing NaCl level with the most cohesive gels produced at the 3% level on a surimi weight basis (Fig. 4). Rigidity (Fig. 5) as a measure of gel strength also increased with increased NaCl, peaking at 3%, though the changes were less sensitive to Instron textural analysis as was obtained with expressible moisture and coh~siveness.
There was no statistical significance in rigidity at salt concentrations of 1.5%, 2% and 2.5%. This suggests that penetration test is not as discriminative as compression test and is in agreement with the previous report by Lee and Chung (1988).
The changes in cohesiveness, rig id i ty, and expressible moisture of ocean pout surimi after salt treatment were observed to differ in the mode of action between red hake and Alaskan pollock (Douglas-Schwarz and Lee, 1988). The cohesiveness of ocean pout gel linearly increased and peaked at 3%; whereas red hake and Alaskan pol lock showed maximum cohesiveness at 2% and 1. 5%, respectively, and declined as the NaCl level increased. Their findings contrasted with those of MacFarlene et a 1.
(1980), and Siegel and Schmidt, (1979b) as to the ability of each individual component to form stronger bonds. Yasui et al., (1980) observed that actin, alone, could not form a stronger gel, but when myosin was added, the gel produced was much stfonger. Along with the difference in gel forming properties between ocean pout and red hake/Alaskan pollock surimi (Douglas-Schwarz and Lee, 1988), observations of these isolated model systems may lead us to theorize that one of the factors accounting for the differences may probably be found in the relative amounts of individual protein fractions of the myofibrils and their effectiveness to bind or interact with other proteins in the system, besides the species specificity. Okada (1973) reporte d that commercial surimi gel from Atlantic pollock is produced at a salt level betwe~n 2.5%-3.5% above which textural characteristics are drastically reduced. This level is much higher than 1. 5% that was reported as an optimal level by Douglas-Schwarz and Lee ( 19 8 8 ) .
The loss in texture at a high salt level attributed to the salting out phenomena (Lehninger, 1970) which is accompanied by increasing salt concentration. C for 20 min followed by cooking for an additional 30 min at 90 C for a two-stage heating.

23
There wer e no s igni f ican t differences ( P<O. 0 5) in the cohesiveness, rigidity and e xpressible moisture of surimi at gels cooked stage heating. di f ferent times when subjected to a single Thermally heat-set and cooked gels were significantly d ifferent from each other and from the gels which were subjected to a single stage heating process. Table 1 show that the strength of ocean pout surimi was marginally affected by cooking time, being at a maximum at 40 min which corresponded very well to the ' reported s tandard cooking time (Lee, 1984).

Results in
The Japanese Ministry of Welfare recommended that an internal cooking temperature of 75 C should be reached (Suzuki, 1981).
Tournas (1984)  This study is in contrast with the previous report in which a two-stage heating resulted in formation of a stronger gel (Lanier et al., 1983). Such a discrepancy could be due to lack of gel-network forming ability of ocean pout during a partial heat-setting process. The latter showed  , while losses in sulfhydryl groups are thought to occur at temperatures above this level.

Formulation Optimization in Blended Surimi
The optimization of formulation with respect to the production of surimi-based products has been examined (Lee et al., 1986). Subsequent to the establishment of a target specification, it is recommended that functionality testing 26 be carried out to find the degree of flexibility from the target functionality at which we can deviate without sacrificing product quality.
In order to assess the changes in gel forming properties due to blending of ocean pout and red hake surimi, the textural properties of surimi gel and subsequent freeze-thaw stability were compared at the following formulations, 100:0, difference in the initial gel strength was shown between 100:0 and 0: 100 formulations. The expressible mo is tu re was higher for ocean pout surimi than red hake surimi -and progressively decreased as the ocean pout content was reduced in the formulations. This trend was also evident after three freeze-thaw eye les. During frozen storage, Connell ( 1959), Buttkus (1970Buttkus ( , 1971, and Matsumoto. (1977)  giving rise to ice crystal formation.
When the system is thawed, some of the water leaves the tissue as drip (Noguchi, 1974), and the protein-water affinity is reduced. Lee and \ Kim ( 1985) noted thctt higher expressible moisture was manifested by freeze-thaw instability. Sone et al. ( 1983) noted that some relationship existed between water holding capacity and protein-water interactions, whereby a strong protein-water interaction would endow a more elastic nature to the gel. From the textural property data obtained in this experiment, usage of ocean pout .surimi in blended products would be recommended at the 20% level, though much work is needed to establish the proper processing factors. varying levels ranging from 0% to 8%, gels showed a maximum strength at 2% and increasing above this level produced linear decreases in the cohesiveness and rigidity (Fig. 18).
Overall, egg albumin outperformed the rest in all measured textural characteristics. This is in agreement with the findings of Lee and Kim (1985). Slight increases in the 29 texture at an additive level of 5% were attributed to the loss of water from the network possibly due to the absorbtive effect of the protein.
!so et al. ( 1984) and Bugarella et al. (1985a, b) observed that the additive protein absorbed water and became swollen and merely filled the interstitial spaces of the protein network rather than contributing to the network structure itself. Decreases in texture above the 2% level, as in the case with egg albumin, indicates that a proper amount of the solid content between the surimi and the \ additive protein must be reached for the filler effect to be optimal. The negative interaction observed, when the level of the protein additive is increased, has been reported (Bugarella et al., 1985;Acton and Dick, 1984;Lanier et al., 1982;Lee, 1986;Chang, 1982;Peng and Nielsen, 1986). It has also been noted that myosin-albumin mixtures proceed with formation of stronger gels at temperatures above 80 C (Foegeding et al., 1986), thus thermal conditions necessary for the albumin to interact with myosin is an important consideration.

Surimi as a binder
Restructured and formed meat systems share one basic but fundamental characteristic, that is, the ability to bind their constituent meat pieces together into a cohesive product that simulates intact muscle. These products often 30 contain from 10% to 15% fine ground meat or mechanically deboned meats (Field,197  These results are comparable with those Erom experiments comparing beef-turkey patties ( Novakofski et al., 1987) in which there were no significant differences in textural scores.
Positive correlations between sensory firmness (r = 0.81), sensory cohesivenss (r = 0.69), appearance (r = 0.81), and overall acceptability scores (r = 0.83) with increasing turkey suggests that the product increased binding as a ' result of increased turkey in the formula ti on.
The product became less moist as the level of turkey ( 7 5%) increased.
Considered that the moisture content of turkey meat and ocean pout sur imi ( 7 6%) were not significantly different, such a change reflect the poor water binding ability of ocean pout surimi.
A reduction in the turkey meat as a result of increasing surimi in the formula contributed to a reduction in the muscle fiber content of the matrix. Theno et al. , (1978)  interfere with the formation of a rigid matrix (Lee, 1984) which is believed to be due to retardation of cross-linkage formation of actomyosin (Okada, 1964;Shimizu and Nishioka, 1974 In comparison of a single-stage and a two-stage heating, the best results were obtained in a single-stage heating at 9 0 C for 40 min, while a two stage heating showed a steady loss of textural firmness reaching its lowest at 60 C. This 35 study is in contrast with the previous ~eport in which a two-stage heating resulted in formation of a stronger gel ( Lanier et a 1. , 1 9 8 3 ) . Such a discrepancy could be due to the lack of gel network forming ability of ocean pout during a partial heat-setting process. While the effects of low       In "Advances in Protein Chemistry". 1985. Application of functionality measurements to the least-cost linear programming of surimi-based product formulations.
Paper #24, presented at the International Symposium on Engineered Seafood including surirni. Washington.
In proceedings of International Symposium on Engineered Seafood including surimi. Lee, c.M. 1986. Surimi manufacturing and fabrication of surimi-based products.
Food Tech. 40 (3) MacFarlene, J.J., Turner, R.H. and Jones, P.N. 1986. Binding of meat pieces: Influence of some processing factors on binding strength and cooking losses. J. Food Sci. 5 1\ 1) 740-41. Nakayama,T and Sato,Y. 197la. Relationship between binding quality of meat and myofibrillar proteins. Part EE. The contribution of native tropomyosin and actin to the binding quality of meat.
In 1975, the U.S. Accounting Office suggested that one of the most promising strategies to revitalize and strengthen the fishing industry is through the development of underutilized fishery resources (Constantinides, et al, 1977 (Lee, 1984;Douglas, 1986).
Ocean pout ( Macrozoarces americanus) is abundant from the Gulf of St. Lawrence to Delaware and is also available in ample supplies in Southern New England waters (Constantinides et al., 1979;Orach-Meza, 1975).
In 1980, 23 0 metric tons of ocean pout were landed in the states of Maine, Massachusetts, and Rhode Island, and increased to 408 metric tons in 1983 (Constantinides et al., 1985).
Known to belong to the Zoarc idae family, Macrozoarces americanus, is characterized by such names as ocean pout, ellpout, catfish, yellow eel, muttonfish, conger eel and ling.
They are known to be present in ample supplies in southern New England during winter and spring and migrate  (Sheely et al., 1977Alton, 1978). From a nutritional point of view, ocean pout is a very lean fish whose fillets contain an average of 81.9% moisture, 16.64% protein, 0.91% lipid and 1.13% ash (Constantindes et al., 1977), which is favorably compared to red hake 82.2% moisture, 16.5% protein, 0.75% lipid and 0.86% ash (Kelleher et al., 1981). Institutional acceptance of fabricated seafood was first recognized in 1973, (Katz, 1974) and a surmounting evidence of retail acceptability is seen from the already growing number of new products appearing on supermarket frozen food shelves.
Surimi, the functional "essence" in fabricated seafood products, is primarily composed of 76% water, 16% protein, 4% sucrose, 3.5% sorbitol, 0.3% polyphosphate, 0.2% fat and 0.0038% calcium (Anonymous, 1985). Myof ibrillar proteins, The difference in solubility has been attributed to the differences in their content of alpha helix, where HMM contains lower alpha helix than LMM (Lowey et al., 1961;Matsumoto, 1980). At physiological pH, myosin is expected to be negatively charged due to its content of a large amount of aspartic, glutamic acid residues and a fair amount of 70 histidine, lys ina and arginine basic residues, bes ides its ATPare activity.  (Noguchi, 1977;Buttkus, 1970Buttkus, , 1971Connell, 1968;Noguchi et al., 1970;Matsumoto et al., 1971;Oguni, et al., 1975), significantly when heat~d to about 70 C. Elastin which contains desmosine and isodesmosine amino acids is not decomposed by heat and is resistant to acid and heat treatment (Bendall, 1964). Because of its amino acid content, it is thought to be highly involved in crosslinking of polypeptide to give its elastic characteristics.  (Bard, 1965;Gillet et al., 1977). These researchers found that maximum extraction temperature to be 7.2 C with a marked decrease in extractability at 0 c. Solomon and Schmidt (1980) found that the extraction of crude myosin increased linearly with time , and was compounded by use of prerigor meat than postrigor meat which is in agreement with Saffle and Galbreath (1964) and Acton and sa ff 1 e ( 19 6 9 ) • Solubilized meat proteins bind to the insoluble components in a protein matrix forming a stable, coherent combination and much research has been done to augment th is concept. With increase in protein extraction, ' concurrent increases in binding strength and the relationship between them has been found to be significant and highly correlated but there appears to be a maximum messaging time by which extraction reaches a constant (Theno et al., 1978) beyond which binding strength levels off which is explained by the fact that muscle fiber is disrupted and weakening of the bind strength is evident.
A prerequisite to effective binding is the mechanical treatment applied to most meat systems. Chopping, mixing, messaging, tumbling and mechanical tenderization are among the most common mechanical treatments employed in meat industries to increase the binding strength.
The importance of mechanical treatment has been explained by Koo (1980) and Therno et al. (1978b) who found that cell disruption and the amount of myof ibrillar protein solubilized increased. Using scanning electron microscopy, Theno et al. (1978b) showed that myofibrils and muscle fibers which are normally tightly packed separate after messaging causing the solubilized proteins of the exudate to be worked on the loose fiber structure allowing a more cohesive bond to form between the protein matrix and meat surface. Extended mechanical treatment excessively disrupted the muscle fibers and the protein matrix system looses its integrity. A certain degree of mechanical treatment is therefore recommended with respect to the product being manufactured.
The two most widely used salts are sodium chloride and  (Turner et al., 1979;Ishioroshi et al., 1979). Salt, when added from 0 to 5% produced increasing gel strength up to 3%, after which salt concentration had little effect. By using scanning electron microscopy, Siegel and Schmidt ( 1979) concluded that myosin and actomosin, when heated in high ionic strength salt Solutions, formed a coherent, three dimensional network of fibers which was necessary to produce a satisfactory bind.
The absence of salt led to the format ion of a spongy structure with little strength.
In addition, Swift and Ellis (1957) found that addition of 0.5% polyphosphate used in conjunction with 2% salt greatly improved the binding strength of bologna and was considerably greater than when 2% salt was used alone.
Binding of meat is a heat-initiated reaction and it does not occur in the raw state (Schnell et al., 1970). The previously \ dissolved proteins, when heated, undergo rearrangement in order to facilitate an interaction with the insoluble components on the meat surface to form a coherent structure. This process is thought to be initiated at temperatures around 45 C and has been described by Hamm and Deatherage (1960) as a noncovalent process based on the fact that even heating to 70 C did not cause any observable formation of intermolecular bonds leading to the conclusion that hydrogen and ionic interactions were the main forces behind stabilized bonds when heat is applied. The increase in gel strength of fish myofibrils was found to start at 30 C and continued until a temperature of 80 C (Quinn et al., 1980). The above conclusions and the work of Wright et al. (1977) have indicated that denaturation temperatures of different protein components was a characteristic of the species of animal from which the protein came, pH and ionic strength which explains why results by researchers in this area differ.

Roles of specific meat proteins in binding
Of the three groups of proteins found in muscle, myofibrillar, sarcoplasmic and stromal, myofibrillar is the one that has been implicated as being the most important in  Fukazawa et al. (196la,b); Samejima et al. ( 1969) and Nakayama and Sato ( 197 la,b) and through these studies it has generally been concluded that myosin and actomyosin were the proteins that produced the greatest gel strengths and so were the most important in binding. In certain cases, actomyosin was found to be a more effective binding agent than myos in. Yasui et a 1. ( 1980) produced a ' much stronger gel when myosin was added to actomyosin than either myos in or ac tomyos in alone. The binding quality of heat set gel increased when F-actin was present in an appropriate ratio to myosin and this increased further when tropomyosin was added to the myosin/F-actin model system.
The presence of tropomyosin was found to be effective in increasing the water holding capacity.
Products made without stromal proteins are known to be soft, jelly-like and lack cohesion (Schut, 1976), but model systems with high connective tissue have yielded products which were poor in binding and so it is thought that a certain level of this protein must be present to produce acceptable binds. Very little work has therefore been undertaken in finding out proper processing levels of stromal proteins that can produce optimal binding quality. Up to now, the most common approach has been to limit the connective tissue to an economic minimum. Researches that 77 have been carried out have not been conclusive. Randall and voisey ( 1977) showed that stromal and myofibrillar proteins increased binding in a meat emulsion when added separately, but the effect was synergistic when added together, which explains why isolated protein systems cannot always be extrapolated to more complex systems, such as meat products, due to possible interactions. Gelatin is formed on heat processing and can liquefy if the product is reheated above a certain critical tempera tu re which, in the case of beef, is ' 49 c.
The liquification of gelatin above critical temperature can result in collapse of product structure and loss of moisture and fat (Poulanne and Ruusunen, 1981).
Difficulties arise as to the extraction and quantification of collagen (Sato et al., 1986). Very often, the denatured and insolubilized myofibrillar proteins are included during extraction of collagens. On the basis of hydroxyproline content of muscles, variations from species to species of fish is a major factor.
Significant information has been obtained as to the effect of using Alaskan pollack surimi in red hake meat and poultry gelation in ground meat products (AFDF, 1985). Of great importance is their conclusion that increased binding potentials of meat blends can easily be obtained by adding relatively low levels of lower-cost surimi.
In addition, the level of surimi added does not appear to contribute as significantly to the possible binding potential as does the actual inclusion of surimi into the formula.
protein gel and physical properties The potential of a protein to form heat-induced gel depends on its ability to interact with other proteins in the system. This interaction, which involves protein-water and protein-protein aggregations are important contributors to \ the functional properties of gel-type products, such as hot dogs, fabricated shellfish meat and kamaboko (Kinsella, 1976;Acton et al • , 198 3 } • In order to achieve the des ired function al properties in these systems, it is necessary to control the composition of the product from its initial processing history, formulation through thermal treatment. Myof ibrillar proteins which are more responsive for the functionality of comminuted meat products (Okada, 1963;Li-Chan et al., 1984;Hashimoto et al., 1983Hashimoto et al., & 1985Kinney et al., 1986;Hickson et al., 1982;Jiang et al., 1986} undergo an initial heat denaturation (unfold} and aggregate when heated to form a three dimensional, crosslinked protein network in which polymer-polymer, polymer solvents interact ions occur. The network formed is responsible for the texture and waterbinding characteristics of the finished Products (Kijowski et al., 1978;Acton et al., 1983;Ziegler and Acton, 1984) and may vary among animal species Makinodan et al., 1971;Shimizu, 1974Shimizu, & 1981oouglas, 1986;Katoh et al., 1986).
The effects of temperature, pH, salt, and protein concentration on gelation of myof ibrillar proteins have previously received wide attention (Ishioroshi et al., 1979;Yasui et al., 1980Yasui et al., & 1982Wu et al., 1985;Foegeding et al., 1986;Montejano et al., 1983;Iso et al., 1984;Hickson et al., 1982;Seki et al., 1985) and it is apparent that rheological changes occurring during gelation ' correlate well to the texture and visco-elastic measurements of the products (Smith et al., 1988). Myosin and actomyosin are the major proteins involved in gelations. Factors accounting for the differences in gelation among fish species have been attributed to differences in hydrophobic interactions on the surface of the protein molecules (Niwa et al., 1971(Niwa et al., & 1982Liu et al., 1982), thermal stability of actomyosin (Shimuzu et al., 1983(Shimuzu et al., & 1981, protein-protein interaction (Acton et al., 1981). Therefore it is generally concluded that the level of functional actomyosin which is measured as extractable actomyosin or ATPase activity is a measure of gel-forming ability of surimi as determined by the water binding capacity and correlates well with the gel strength.
With an increase in the number of washing cycles the level of actomyosin increases, but is greatly affected by the freshness of the fish (Lee, 1984). Loss in protein functionality and in particular, the gel forming ability is due to freeze denaturation and aggregation of the myofibrillar proteins (Sikorsky et al., 1976;Matsumoto, 1980Matsumoto, & suzuki, 1981 which decreases the stability of actomyosin and increases the susceptibility for rapid denaturation during thawing and/or subsequent storage (Scott et al., 1988). In addition, the quality of surimi during frozen storage is also affected by storage temperature, storage period, the level of moisture and cryoprotectants used (Lee, 1984;Okada & Iwata, 1969;Shimuzu and Fujike, 1985). Thus, gel ' forming ability of surimi will be maintained significantly with extended storage time for up to one year at a constant temperature of -20 C (Iwata et al., 1968(Iwata et al., & 1971. Reports (Iwata et al. , 1971) also indicate gradual decrease in gel forming ability at a storage temperature of -10 C rendering the surimi useless after three months.  (Yoon, 1986).
Addition of cryoprotectants such as starch (for products which are to be cooked and frozen, and sorbitol for uncooked products to be frozen), has been suggested as the best approach to overcome freeze-thaw instability without reducing the myosin level.
The application of heat for cooking of further process products and addition of salt, however, are the two major factors in denaturation and gelation of muscle proteins. The solubized surimi paste gels rapidly upon heating at 80-90 C, \ but slowly at 40-50 C. Okada ( 1963) reported that slow setting of the gel at 40-50 C resulted in stronger gel than cooking without a slow set. This was not the case with ocean pout surimi and the phenomenon may be attributable to the species differences (Makinodan & Ikeda, 1971;Shimuzu & Nishioka, 1974;Shimuzu et al., 1981). Gel setting occurs at tempera tu res up to 50 C. Increasing temperature to 60-70 C causes a softening phenomena (Madori). In contrast, reports for surimi prepared from red hake (Urophycis chuss) and Alaskan pollock (Theragra chalcogramma) or the ocean pout appears to be affected by the softening phenomena. Makinodan and Ikeda (1971), Deng et al. (1979) and Lanier et al. (1982) have attributed this to the presence of alkaline protease enzyme that has maximum activity at 60-70 C. In the latter case, a theory of temperature-dependent gel setting was proposed. At a faster heating rate, a tight cohesive network with a large number of small aggregate is formed, whereas at 82 a slow heating rate a loose network with a small number of large aggregate is found.
Some loss of gel forming potential may be acceptable for products manufactured for western tastes where a less rubbery texture may be acceptable (Lanier, 1986). This would require a dilution of high quality with low quality surimi or use of other less functional ingredients.
In the manufacture of frozen products, both molded and fiberized, a combination of wheat gluten and wheat starch has a beneficial rubbery-' texture reducing abi 1 i ty and better freeze-thaw stab i 1 iz ing effect while a combination of wheat gluten and modified potato starch will produce a more acceptable texture in fresh products (Lee, 1987).
83 Figure 1 CONSENT FORM I, the undersigned, do hereby acknowledge that I am voluntarily participating in the below described sensory analysis of surimi produced at the surimi-making facility of the University of Rhode Island.
Further, I have read the statement below and understand what I am expected to do as a participant and the risks outlined below.
This sensory panel is being conducted to evaluate the desirable combination of ocean pout and red hake and their combination with sugar and sorbitol in surimi extruded products.
Surimi and extruded products are prepared using fresh ocean pout and red hake, and various processing techniques.
The newly extruded surimi and extruded products are steam, cooked and cooled before being presented to the panel.
Those which are obviously unacceptable after treatment are eliminated.
The desirability will be tested and rated on a numerical scale according to how desirable the taster finds them for compensation.
Any individual with allergy to surimi or fish should be removed from the study.
The information obtained in this panel will be used to determine optimum levels of ocean pout and red hake with desirable qualities. The names of the individual participants in the panel will not be associated in any way with the results of the panel, and wi 11 not appear in any publication which may result from the information gathered.
Finally, participation in the panel does not obligate the panelist in any way to participate in subsequent panels unless they wish to do so, and a panelist may remove himself from the survey at any time by simply returning the form to the individual conducting the panel and stating he or she does not wish to continue with the taste testing.         abed Means followed by the same letter in the same column are not significantly different (P<0.05).  a Average of 6 observations. abed Means in the same row followed by the same letter are not significantly different (P<0.05).