Utilization of fermented skipjack tuna viscera as a dietary protein source replacing fish meal or soybean meal for juvenile abalone Haliotis discus hannai.

ABSTRACT This study was carried out to evaluate the usefulness of fermented skipjack tuna viscera (FSTV) in the diet for juvenile abalone Haliotis discus hannai. Lactobacillus bulgaricus was used for fermentation of skipjack tuna viscera. Eight isonitrogenous (about 30% crude protein) diets were formulated to include different levels (0%, 10%, 20%, and 30%) of FSTV as a replacement for either fish meal or soybean meal in diets. Three replicate groups of abalone were fed the experimental diets for 7 wk. The inclusion of FSTV up to 30% in fish meal-based diet had no significant effect on survival, body weight, shell growth, and proximate composition of abalone (P > 0.05). Weight gain of abalone fed the diet substituting 10% FSTV for soybean meal, was not significantly different to that of abalone fed the control diet, however this value decreased in abalone fed the 20% and 30% FSTV (P < 0.05). The contents of crude protein and lipid of soft body in abalone fed soybean meal-based diets were significantly affected by dietary FSTV level (P < 0.05). The results of this study indicate that FSTV can be used as a partial substitute protein source for fish meal or soybean meal in the formulated diet for juvenile abalone.

KEY WORDS: fermented skipjack tuna viscera, Haliotis discus hannai, protein source

INTRODUCTION

Abalone has a high market demand in Korea. The culture of abalone relies heavily on marine macroalgae such as Undaria or Laminaria, however, this is a problem for several reasons: unstable supply of macroalgae and increased storage cost. In addition, it has been reported that growth of abalone fed the macroalgae was lower than that of abalone fed the formulated diets (Hahn 1989, Viana et al. 1993). The better growth of abalone fed the formulated diets was attributed to the good protein quality of artificial feeds (Viana et al. 1993). In feed formulation, protein source among the dietary ingredients is very important because it is more expensive than other energy sources. Furthermore, the growth of the animal could be easily influenced by availability of dietary protein sources. Thus, high-quality proteins are necessary for a good abalone feed. Fish meal, soybean meal, and casein are the most commonly used for protein sources in abalone diets (Fleming et al. 1996). Uki et al. (1985) reported that casein was the best protein source for the diet of Haliotis discus hannai. But the best quality protein sources including casein are expensive. Therefore, low-cost practical alternate protein sources in diets could be of considerable economic advantage. Recently, Lee et al. (1998) suggested that the cost of H. discus hannai feed production could be reduced by using low cost dietary protein sources such as fish meal and soybean meal.

The utilization of waste by-products from commercial fish processing as a feed ingredient for aquaculture will be able to decrease production cost and minimize environmental pollution. Silage has proved to be a good dietary protein source for fish (Raa & Gildberg 1982). Fish silage is generally a product of high value presenting practically the similar composition as the original raw material. Previous study (Lee et al. 1998) reported that each of casein, fish meal, and soybean meal was equally good protein source for H. discus hannai. This study is carried out to evaluate the utilization of fermented skipjack tuna viscera as a protein source in the diets for juvenile H. discus hannai.

MATERIALS AND METHODS

Preparation of Fermented Silage

The viscera of skipjack tuna (Katsuwonus pelamis), which was obtained from Dongwon Industry Co. (Changwon Korea), was ground and mixed thoroughly with sugar beet molasses (100 g/kg) and Lactobacillus bulgaricus KCTC 3188 (50 g/kg) as starter and then fermented at 35[degrees]C for 10 days in a sealed 50-L plastic bucket. After 10 days, the fermented sample was filtered by using gauzes and used for the dietary ingredient.

Experimental Diets

Two sets of experimental diets were formulated using fermented skipjack tuna viscera (FSTV) at 0%, 10%, 20%, and 30% levels replacing mainly fish meal in experiment 1 (Table 1) and soybean meal in experiment 2 (Table 2), respectively. Wheat flour and a-potato starch were used as carbohydrate, and squid liver oil was used as lipid source. Procedures for feed preparation were adapted from the method of Mai et al. (1995a). Experimental diets were dried at room temperature and stored at -25[degrees]C until used. All diets were similar in contents of crude protein (28.4% to 30.0%) and lipid (4.3% to 6.8%), which are considered to be sufficient to maintain optimum growth for H. discus hannai (Mai et al. 1995a, Mai et al. 1995b).

Experimental Design

Two feeding trials with completely randomized block designs were conducted to evaluate the usefulness of FSTV in the diet for juvenile H. discus hannai in flow-out system. Juvenile abalone were acclimated to the experimental condition for 2 wk while being fed a commercial abalone diet containing 30% protein and 5% lipid. They were randomly stocked into twenty-four 20-L tanks at a density of 100 abalone in each tank. Three replicate groups of abalone (mean weight 848 mg and 821 mg in experimental 1 and 2, respectively) were fed ad libitum at intervals of 2 days for 7 wk. Before feeding, uneaten diets in each tank were cleaned by siphoning off. Filtrated seawater was supplied at a flow rate of approximately 3 L/min in each tank. Photoperiod was left at the natural condition, and mean temperature was 22 [+ or -] 0.9[degrees]C during the feeding trial. Abalone in each tank were collectively weighed on the days of initiation and termination at the feeding trial after being fasted for 24 h. Two-hundred abalone samples at the beginning and all abalone at the end of feeding trial were sacrificed and stored at -70[degrees]C for chemical analysis.

Chemical Analysis

Crude protein content was determined by Kjeldahl method using Auto Kjeldahl System (Buchi B-324/435/412, Switzerland), crude lipid content by ether-extraction method, moisture content by a dry oven (105[degrees]C for 12 h), crude fiber content by an automatic analyzer (Fibertec, Tecator, Hoganas, Sweden), and ash content by a furnace muffler (550[degrees]C for 4 h). Amino acids content was determined using automatic amino acid analyzer (Hitachi 835, Japan). Tryptophan was determined according to the method of Hugli and Moore (1972). Volatile basic nitrogen was measured using Conway's microdiffusion method (Conway 1950). Lipid for fatty acids analyses was extracted by mixture of chloroform and methanol (2:1, v/v) according to the method of Folch et al. (1957). Fatty acid methyl esters were prepared by transesterification with 14% B[F.sub.3]-MeOH (Sigma, St. Louis, USA), and were analyzed by using a gas chromatography (HP 5890, Hewlett-Packard, USA) with flame ionization detector, equipped with HP20M capillary column (0.25[micro]m x 30 m). Injector and detector temperatures were 250[degrees]C and 270[degrees]C, respectively. The column temperature was programmed from 150[degrees]C to 230[degrees]C at a rate of 2[degrees]C/min. Helium was used as the carrier gas. Fatty acids were identified by comparison with known standards.

Statistical Analysis

The data were subjected to one-way analysis of variance (ANOVA) using the SPSS program version 7.5 (SPSS Inc., Michigan Avenue, Chicago, Illinois, USA). Significant differences (P < 0.05) among mean were determined by Duncan's multiple range test (Duncan 1955).

RESULTS AND DISCUSSION

Proximate composition and volatile basic nitrogen content of the raw skipjack tuna viscera (RSTV) and fermented skipjack tuna viscera (FSTV) are shown in Table 3. The FSTV resulted in increase of carbohydrate and volatile basic nitrogen, and decrease in the contents of protein and lipid. Increase of carbohydrate in the FSTV probably originated from molasses used. The total amino acids composition (Table 4) of FSTV was similar to that of RSTV. The contents of Glu, Leu, Asp, Gly, and Ala were high among amino acids in both RSTV and FSTV. The contents of free amino acid in the FSTV were higher than those of RSTV. Considerably increased free amino acids content in the FSTV may be the result from that of the protein of RSTV, which is broken down into smaller units during storage for fermenting. Batty acid compositions of the RSTV and FSTV are shown in Table 5, 16:0, 18:ln-9, and 22:6n-3 were most abundant fatty acids in RSTV and FSTV. The polyenes in FSTV were higher than those of RSTV. Considering compositions of amino acid and fatty acid (Mai et al. 1995b, Mai et al. 1996) in the FSTV, this fermented silage has a good nutritional condition for dietary ingredient of H. discus hannai.

Growth performance of H. discus hannai fed the diets replacing fish meal with different FSTV level is presented in Table 6 (experiment 1). In the present study, the inclusion of FSTV up to 30% of the fish meal-based diet had no significant effect on survival, body weight, and shell growth of abalone (P > 0.05). When H. discus hannai fed the diets replacing soybean meal with different FSTV level (Table 6, experiment 2), weight gain of abalone fed the diet containing 10% FSTV was not significantly different to that of abalone fed the control diet, however this value decreased in abalone fed the 20% and 30% FSTV (P < 0.05).

The feeding value of the dietary protein source depends not only on the quantity of nutrient such as protein and essential amino acids, but also on the ability of the fish to use that nutrient (National Research Council 1993). Additionally, various characteristics of the feedstuff including nutritional values, cost, stable supply, and palatability should be considered to commercially use as the protein source in practical diet. Viana et al. (1993) reported that fish meal was a good dietary protein source for the growth of H. fulgens. The development of commercial feeds for aquaculture has been traditionally based on fHHish meal as the main protein source due to its high protein content and balanced essential amino acid profile. However, because of high price of good quality proteins including fish meal, many studies have been focused on alternative practical protein sources which have economic advantages for aquaculture (Viana et al. 1996, Lee et al. 1998, Mukhopadhyay & Ray 1999, Giri et al. 2000, Middleton et al. 2001). Fish silage has a good potential as a protein source for some fish feed (Hardy et al. 1983, Espe et al. 1992, Fagbenro et al. 1994, Fagbenro & Jauncey 1995). The results obtained from this study suggest that H. discus hannai maybe use the protein from FSTV and fish meal with similar efficiency. This is agreement with other studies (Fagbenro 1994, Vidotti et al. 2002). Fagbenro et al. (1994) reported that up to 75% of fish meal protein was successfully replaced by blended lactic acid fermented fish silage:soybean meal (1:1) incorporated in diets fed to Nile tilapia fingerlings.

Weight gains (406-575 mg/abalone) of H. discus hannai fed the soybean meal-based diets (experiment 2) were higher than those (279-323 rag/abalone) of abalone fed the fish meal-based diets (experiment 1), regardless of different FSTV levels. This result suggests that soybean meal is very good protein source in diet for growth of H. discus hannai. However, Uki et al. (1985) reported that growth of H. discus hannai fed the casein diet was higher than that of abalone fed the diet containing fish meal or soybean meal. On the other hand, Lee et al. (1998) reported that each of casein, fish meal, and soybean meal was equally good protein source for H. discus hannai. The different growth responses to dietary protein sources among studies are probably due to feeding and abalone conditions, such as water temperature, abalone size, and dietary composition used in their studies.

There were no significant differences (P > 0.05) in moisture, protein, lipid, and ash contents of soft body in H. discus hannai fed the diets replacing fish meal with different FSTV level (Table 7, experiment 1). When H. discus hannai fed the diets replacing soybean meal with different FSTV level (Table 7, experiment 2), crude protein content of soft body was higher in abalone led control diet, while crude lipid content in abalone fed the diet containing 30% FSTV was lower than that of in the others (P < 0.05).

The results obtained from this study indicate that FSTV has a good potential as a substitute protein source, especially for fish meal in the formulated diet for juvenile H. discus hannai. The use of FSTV as an ingredient in diet for H. discus hannai could result in cheaper artificial feeds, making its culture more economically viable, and use of tuna viscera is potentially a profitable solution to a waste problem of the canned tuna industries.

TABLE 1.

Ingredient and nutrient contents (%, DM basis) of the experimental
diets replacing white fish meal with silage (experiment 1).

                                          Silage Level (%)

                                      0      10     2      30
Ingredients
  White fish meal (1)                35.0   29.0   23.0   17.0
  Wheat flour                        24.0   21.2   18.4   15.6
  Fermented skipjack tuna viscera     --    10.0   20.0   30.0
  Undaria powder                      5.0    5.0    5.0    5.0
  [alpha]-potato starch               5.0    5.0    5.0    0.5
  Squid liver oil (2)                 4.0    2.8    1.6    0.4
  Vitamin premix (3)                  2.5    2.5    2.5    2.5
  Mineral premix (4)                  4.0    4.0    4.0    4.0
  Sodium alginate                    20.0   20.0   20.0   20.0
  Choline salt                        0.5    0.5    0.5    0.5
Nutrient contents
  Crude protein                      28.4   28.7   29.1   29.4
  Crude lipid                         6.8    6.0    5.2    4.4
  NFE (5)                            48.2   48.9   49.4   50.0
  Crude fiber                         1.5    1.5    1.5    1.5
  Ash                                15.1   14.8   14.8   14.6
  n-3HUFA (6)                         1.6    1.5    1.3    1.2

(1) Imported from Rusia, contained 0.015% ethoxyquin.

(2) Provided by E-wha Oil & Fat Ind. Co.. Busan, Korea.

(3) Vitamin mix contained the following amount which were diluted in
cellulose (g/kg mix): L-ascordic acid, 200: DL-[alpha]-tocopheryl
acetate. 20: thiamin hydrochloride, 5; riboflavin, 8: pyridoxin
hydrochloride, 2: niacin, 40: Ca-D-pantothenate, 12; myo-inositol,
200; D-biotin, 0.4; folic acid. 1.5: p-aminobenzoic acid, 20,
menadione, 4; retinyl acetate, 1.5: cholecalciferol, 0.003:
cyanocobalamin, 0.003.

(4) Mineral mix contained the following ingredients (g/kg mix):
NaCl, 10, MgS[O.sub.4] x 7[H.sub.2]0, 150; Na[H.sub.2]P[O.sub.4] x
2[H.sub.2]O, 250; K[H.sub.2]P[O.sub.4], 320: Ca[H.sub.4]
[(P[O.sub.4]).sub.2] x [H.sub.2]O, 200; Ferric citrate, 25;
ZnS[O.sub.4] x 7[H.sub.2]O, 4: Ca-lactate, 38.5; CuCl, 0.3:
Al[Cl.sub.3] x 6[H.sub.2]O, 0.15; KI[O.sub.3], 0.03;
[Na.sub.2][Se.sub.2][O.sub.3], 0.01; MoS[O.sub.4] x [H.sub.2]O,
2: Co[Cl.sub.2] x 6[H.sub.2]O, 0.1.

(5) N-free extract calculated by difference: 100 - (crude protein
+ crude lipid + crude fiber + ash).

(6) Highly unsaturated fatty acids (C [greater than or equal to] 20).

TABLE 2.

Ingredient and nutrient contents (%, DM basis) of the experimental
diets replacing soybean meal with silage (experiment 2).

                                          Silage Level (%)

                                      0      10     20     30
Ingredients
  White fish meal (1)                 5.0    5.0    5.0    5.0
  Soybean mea1 (2)                   45.0   35.0   25.0   15.0
  Wheat flour                        12.9   14.3   15.8   17.2
  Fermented skipjack tuna viscera     --    10.0   20.0   30.0
  Undaria powder                      5.0    5.0    5.0    5.0
  Squid liver oil (1)                 5.1    3.7    2.2    0.8
  Vitamin premix (1)                  2.5    2.5    2.5    2.5
  Mineral premix (1)                  4.0    4.0    4.0    4.0
  Sodium alginate                    20.0   20.0   20.0   20.0
  Choline salt                        0.5    0.5    0.5    0.5
Nutrient contents
  Crude protein                      30.0   29.7   29.5   29.2
  Crude lipid                         6.4    5.7    4.9    4.3
  NFE (1)                            48.4   49.8   51.2   52.4
  Crude fiber                         3.6    3.2    2.7    2.2
  Ash                                11.5   11.6   11.7   12.0
  n-3HUFA (1)                         1.2    1.2    1.1    1.0

(1) Refer to Table 1.

(2) Dehulled, solvent extracted.

TABLE 3.

Proximate analysis and volatile basic nitrogen (VBN) content of raw
and fermented skipjack tuna viscera.

                     Raw Viscera     Fermented Silage

Moisture (%)             73.6              69.7
Crude protein (%)        18.0              14.8
Crude lipid (%)           4.7               2.6
Carbohydrate (%)          0.7               9.7
Ash (%)                   3.0               3.2
VBN (mg/100 g)           46.7              86.7

TABLE 4.

Total and free amino acids contents (mg/100 g) of raw and
fermented skipjack tuna viscera.

          Total Amino Acids       Free Amino Acids

Amino      Raw     Fermented      Raw      Fermented
Acids    Viscera     silage     Viscera      Silage

Ala       1008        1070        262          640
Arg        779         665        194          507
Asp       1403        1265         39         1130
Cys        108         130         32           69
Glu       2165        1958        649         1494
Gly        982        1124        275          834
His        366         306        102          224
Ile        886         832        265          821
Leu       1466        1437        366         1433
Lys        301         247        167          241
Met        431         265        116          228
Phe        700         635        189          606
Pro        715         714        214          540
Ser        537         583        145          459
Tau        482         351        139          259
Trp        196         194         48          182
Thr        680         855        197          822
Tyr        289         158         69          126
Val        892         872        249          837
Total     14386      13661       3717        11452

TABLE 5.

Fatty acid compositions (% of total fatty acids) of raw and
fermented skipjack tuna viscera.

Fatty Acid    Raw Viscera    Fermented Silage

14:0              1.6               2.9
15:0              0.8               0.9
16:0             30.1              23.9
16:1n-7           4.2               7.5
18:0             10.1               7.1
18:1n-9          18.3              14.7
18:2n-6           0.3               3.9
18:3n-3           0.3               1.0
20:0              0.0               0.2
20:1n-9           0.6               0.6
20:2n-6           0.0               2.0
20:4n-6           2.9               0.6
20:5n-3           6.6               9.4
22:2n-6           0.0               1.7
22:3n-6           0.0               0.6
22:6n-3          24.2              22.1
24:1n-9           0.0               0.9
Saturates        42.6              35.0
Monoenes         23.1              23.7
Polyenes         34.3              41.3

TABLE 6.

Growth performance of juvenile abalone fed the diets replacing
fish meal (experiment 1) and soybean meal (experiment 2) with different
silage level for 7 weeks.

                                  Dietary Silage Level (%)

                                              0

                         Experiment 1

Initial mean wet wt. (mg)             890 [+ or -] 39.8
Final mean wet wt. (mg)              1129 [+ or -] 63.2
Weight gain (mg/abalone)              310 [+ or -] 35.1
Survival (%)                           97 [+ or -] 1.7
Final soft body wet wt. (mg)          663 [+ or -] 32.0
Final shell length (mm)              20.6 [+ or -] 0.24
Final shell width (mm)               14.8 [+ or -] 0.25
Final shell height (mm)              4.95 [+ or -] 0.058
Soft body wt./whole body wt.         0.63 [+ or -] 0.003

                            Experiment 2

Initial mean wet wt. (mg)             848 [+ or -] 34.5
Final mean wet wt. (mg)              1423 [+ or -] [52.1.sup.c]
Weight gain (mg/abalone)              575 [+ or -] [18.2.sup.c]
Survival (%)                           97 [+ or -] 0.7
Final soft body wet wt. (mg)          813 [+ or -] 37.1
Final shell length (mm)              21.8 [+ or -] 0.22
Final shell width (mm)               15.7 [+ or -] 0.20
Soft body wt./whole body wt.         0.64 [+ or -] 0.005

                                  Dietary Silage Level (%)

                                              10

                            Experiment 1

Initial mean wet wt. (mg)             812 [+ or -] 26.3
Final mean wet wt. (mg)              1097 [+ or -] 47.9
Weight gain (mg/abalone)              284 [+ or -] 14.1
Survival (%)                           98 [+ or -] 1.1
Final soft body wet wt. (mg)          633 [+ or -] 38.4
Final shell length (mm)              20.1 [+ or -] 0.31
Final shell width (mm)               14.5 [+ or -] 0.23
Final shell height (mm)              4.92 [+ or -] 0.180
Soft body wt./whole body wt.         0.64 [+ or -] 0.010

                            Experiment 2

Initial mean wet wt. (mg)             826 [+ or -] 37.3
Final mean wet wt. (mg)              1362 [+ or -] [18.6.sup.bc]
Weight gain (mg/abalone)              536 [+ or -] [43.3.sup.bc]
Survival (%)                           99 [+ or -] 0.7
Final soft body wet wt. (mg)          763 [+ or -] 52.0
Final shell length (mm)              21.3 [+ or -] 0.38
Final shell width (mm)               15.3 [+ or -] 0.24
Soft body wt./whole body wt.         0.63 [+ or -] 0.012

                                  Dietary Silage Level (%)

                                              20

                            Experiment 1

Initial mean wet wt. (mg)             854 [+ or -] 34.2
Final mean wet wt. (mg)              1134 [+ or -] 15.0
Weight gain (mg/abalone)              279 [+ or -] 46.9
Survival (%)                           98 [+ or -] 1.3
Final soft body wet wt. (mg)          643 [+ or -] 21.8
Final shell length (mm)              20.1 [+ or -] 0.07
Final shell width (mm)               14.3 [+ or -] 0.09
Final shell height (mm)              4.77 [+ or -] 0.056
Soft body wt./whole body wt.         0.62 [+ or -] 0.006

                            Experiment 2

Initial mean wet wt. (mg)             814 [+ or -] 32.4
Final mean wet wt. (mg)              1251 [+ or -] [66.7.sup.ab]
Weight gain (mg/abalone)              437 [+ or -] [44.7.sup.ab]
Survival (%)                           99 [+ or -] 0.7
Final soft body wet wt. (mg)          720 [+ or -] 72.1
Final shell length (mm)              21.2 [+ or -] 0.28
Final shell width (mm)               15.3 [+ or -] 0.20
Soft body wt./whole body wt.         0.63 [+ or -] 0.016

                                  Dietary Silage Level (%)

                                              30

                            Experiment 1

Initial mean wet wt. (mg)             834 [+ or -] 19.6
Final mean wet wt. (mg)              1156 [+ or -] 39.2
Weight gain (mg/abalone)              323 [+ or -] 27.0
Survival (%)                          100 [+ or -] 0.0
Final soft body wet wt. (mg)          640 [+ or -] 25.1
Final shell length (mm)              20.5 [+ or -] 0.14
Final shell width (mm)               14.7 [+ or -] 0.05
Final shell height (mm)              4.93 [+ or -] 0.104
Soft body wt./whole body wt.         0.60 [+ or -] 0.014

                            Experiment 2

Initial mean wet wt. (mg)             794 [+ or -] 17.9
Final mean wet wt. (mg)              1200 [+ or -] [23.6.sup.a]
Weight gain (mg/abalone)              406 [+ or -] [6.50.sup.a]
Survival (%)                           98 [+ or -] 1.1
Final soft body wet wt. (mg)          697 [+ or -] 13.3
Final shell length (mm)              20.8 [+ or -] 0.11
Final shell width (mm)               14.9 [+ or -] 0.16
Soft body wt./whole body wt.         0.63 [+ or -] 0.003

Values (mean [+ or -] S.E. of three replications) in the same row
not sharing a common superscript are significantly different
(P < 0.05).

TABLE 7.

Proximate composition (%, wet basis) of the soft whole body in abalone
fed the diets replacing fish meal (experiment 1) and soybean meal
(experiment 2) with different silage level for 7 weeks.

                               Dietary Silage Level (%)

                           0                          10

                             Experiment 1

Moisture      75.8 [+ or -] 0.49             75.8 [+ or -] 0.65
Crude
  protein     16.0 [+ or -] 0.25             15.9 [+ or -] 0.62
Crude
  lipid       1.41 [+ or -] 0.052            1.53 [+ or -] 0.081
Ash           2.44 [+ or -] 0.032            2.35 [+ or -] 0.026

                             Experiment 2

Moisture     75.7 [+ or -] 0.83             76.0 [+ or -] 0.06
Crude
  protein    17.2 [+ or -] [0.27.sup.b]     16.0 [+ or -] [0.50.sup.a]
Crude
  lipid      1.72 [+ or -] [0.068.sup.ab]   2.01 [+ or -] [0.037.sup.c]
Ash          2.37 [+ or -] 0.052            2.38 [+ or -] 0.033

                         Dietary Silage Level (%)

                      20                           30

                             Experiment 1

Moisture    75.9 [+ or -] 0.46             75.5 [+ or -] 1.25
Crude
  protein   16.2 [+ or -] 0.72             16.0 [+ or -] 0.75
Crude
  lipid     1.56 [+ or -] 0.049            1.44 [+ or -] 0.086
Ash         2.37 [+ or -] 0.057            2.45 [+ or -] 0.133

                             Experiment 2

Moisture    74.6 [+ or -] 0.28             76.0 [+ or -] 0.06
Crude
  protein   15.3 [+ or -] [0.22.sup.a]     15.6 [+ or -] [0.20.sup.a]
Crude
  lipid     1.91 [+ or -] [0.043.sup.bc]   1.64 [+ or -] [0.100.sup.a]
Ash         2.34 [+ or -] 0.003            2.42 [+ or -] 0.065

Values (mean [+ or -] S.E. of three replications) in the same row not
sharing a common superscript are significantly different (P < 0.05).

LITERATURE CITED

Conway, E. J. 1950. Microdiffusion analysis and volumetric error. London, England: Cosby, Lochwood and Son.

Duncan, D. B. 1955. Multiple-range and multiple F tests. Biometrics 11: 1-42.

Espe, M., H. Haaland & L. R. Njaa. 1992. Autolysed fish silage as a feed ingredient for Atlantic salmon (Salmo salar). Comp. Bioehem. Physiol. 103A:369-372.

Fagbenro, O. A. & K. Jauncey. 1995. Growth and protein utilization by juvenile catfish (Clarias gariepinus) fed dry diets containing co-dried lactic-acid-fermented fish-silage and protein feedstuffs. Bioresouree Technol. 51:29-35.

Fagbenro, O. A. 1994. Dried fermented fish silage in diets for Oreochromis niloticus, Isr. J. Aqacult. Bamidgeh. 46:140-147.

Fagbenro, O. A., K. Jauncey & G. Haylork. 1994. Nutritive value of diets containing dried lactic acid fermented fish silage and soybean meal for juvenile Oreochromis niloticus and Clarias gariepinus. Aquat. Liv. Resour. 7:79-85.

Fleming, A. E., R. J. Van Barneveld & P. W. Hone. 1996. The development of artificial diets for abalone: A review and future directions. Aquaculture 140:5-53.

Folch, J., M. Lees & G. H. S. Stanley. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem. 226:496-509.

Giri, S. S., S. K. Sahoo, A. K. Sahu & P. K. Mukhopadhyay. 2000. Nutrient digestibility and intestinal enzyme activity of Clarias batrachus (Linn.) juveniles fed on dried fish and chicken viscera incorporated diets. Bioresource Technol. 71:97-101.

Hahn, K. O. 1989. Nutrition and growth of abalone. In: K. Hahn, editor. Handbook of culture of abalone and other marine gastropods. Boca Raton, Florida: CRC Press. pp. 135-180.

Hardy, R. W., K. D. Shearer, F. E. Stone & D. H. Wieg. 1983. Fish silage in aquaculture diets. J. World Maricult. Soc. 14:695-703.

Hugli, T. E. & S. Moore. 1972. Determination of the tryptopban content of proteins by ion exchange chromatography of alkaline hydrolysates. J. Biol. Chem. 247:2828-2832.

Lee, S.-M., S.J. Yun & S. B. Hut. 1998. Evaluation of dietary protein sources for abalone (Haliotis discus hannai). J. Aquacult. 11:19-29.

Mai, K., J. P. Mercer & J. Donlon. 1995a. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L and Haliotis discus hannai Ino. II. Response of abalone to various levels of dietary lipid. Aquaculture 134:65-80.

Mai, K., J. P. Mercer & J. Donlon. 1995b. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L and Haliotis discus hannai Ino. IV. Optimum dietary protein level for growth. Aquaculture 136:165-180.

Mai, K., J.P. Mercer & J. Donlon. 1996. Comparative studies on the nutrition of two species of abalone, Haliotis tuberculata L and Haliotis discus hannai Ino. V. The role of polyunsaturated fatty acids of macroalgae in abalone nutrition. Aquaculture 139:77-89.

Middleton, T. F., P. R. Ferket, L.C. Boyd, H.V. Daniels & M. L. Gallagher. 2001. An evaluation of co-extruded poultry silage and culled jewel sweet potatoes as a feed ingredient for hybrid Tilapia (Oreochromis niloticus x O. mossambicus). Aquaculture 198:269-280.

Mukhopadhyay, N. & A.K. Ray. 1999. Effect of fermentation on the nutritive value of sesame seed meal in the diets for rohu, Labeo rohita (Hamilton), fingerlings. Aquacult. Nutr. 5:229-236.

National Research Council (NRC). 1993. Nutrient requirements of fish. Washington DC: National Academy Press. 114 pp.

Raa, J. & A. Gildberg. 1982. Fish silage: a review. CRC Crit. Rev. Food Sci. Nutr. 16:383-419.

Uki, N., A. Kemuyama & T. Watanabe. 1985. Nutritional evaluation of several protein sources in diets for abalone Haliotis discus hannai. Bullet. Jap. Soc. Sci. Fish. 51:1835-1839.

Viana, M. T., L. M. Lopez & A. Salas. 1993. Diet development for juvenile abalone Haliotis fulgens. Evaluation of two artificial diets and macroalgae. Aquaculture 117:149-156.

Viana, M. T., L. M. Lopez, Z. Garcia-Esquivel & E. Mendez. 1996. The use of silage made from fish and abalone viscera as an ingredient in abalone feed. Aquaculture 140:87-98.

Vidotti, R. M., D. J. Carneiro & E. M. M. Viegas. 2002. Growth rate of pacu, Piaractus mesopotamicus, fingerlings fed diets containing co-dried fish silage as replacement of fish meal. J. Appl. Aquacult. 12: 77-88.

SANG-MIN LEE, (1) * KYOUNG-DUCK KIM (2) AND TAE JIN KIM (3)

(1) Faculty of Marine Bioscience and Technology, Kangnung National University, Gangneung 210-702, Korea; (2) National Fisheries Research and Development Institute, Busan 619-900, Korea (3) Korean Association of Sliced Raw Fish, Busan 613-812, Korea

* Corresponding author. E-mail address: [e-mail]