Evaluation of Five Commercially Prepared

Diets for Striped Bass


U.S. Fish and Wildlife Service, Southeastern Fish Cultural Laboratory

Route 3, Box 86, Marion, Alabama 36756, USA

'Present address: U.S. Fish and Wildlife Service Marquette Biological Station, 446 East Crescent Street Marquette, Michigan 49855, USA.

2Present address: Texas Cooperative Fish and Wildlife Research Unit, Department of Range and Wildlife Sciences, Texas Tech University, Lubbock, Texas 79409-2125,



Abstract.—Fingerling striped bass (Morone saxatilis) were held in tanks at 20C and fed five commercially prepared diets over an 8-week period. There were no significant differences in weight gain among fish fed a commercial trout diet (daily gain, 1.49% of body weight), the U.S. Fish and Wildlife Service open-formula trout diet GR-6-30 (1.76%/d), or the Service's Atlantic salmon diet ASD-2-30 (1.72%/d). Striped bass fed a modified catfish diet had significantly lower growth (0.74%/d) than those fed the trout or salmon diets. Striped bass fed a commercial diet consisting of all fish by-products gained only 0.09% body weight/d after losing weight during the first 2 weeks. Fingerling and yearling striped bass reared in ponds grew significantly faster when fed a commercial trout diet than when fed a modified catfish diet.


Hatchery propagation of striped bass (Morone saxatilis) and striped bass x white bass (M. chrysops) hybrids is important for the establishment of new sport fisheries for these taxa and for the restoration of declining populations (Stevens 1984). Commercial landings of striped bass declined from 1973 through the early 1980s (Rothschild et al. 1981); in 1984 commercial and recreational fishing along the Atlantic coast of the USA was restricted in an attempt to protect the species. Major restoration efforts now are under way in coastal areas including Chesapeake Bay (Parker and Miller 1987).

Although feeding is an important aspect of artificial propagation, relatively little is known about specific dietary requirements of striped bass (Kerby et al. l 983). Striped bass fry are typically reared for 4-6 weeks (phase I) in ponds fertilized to enhance natural food production (Bonn et al. 1976). Some culturists supplement the natural food with manufactured trout or salmon diets after the fry are 21 d old or older (Fitzmayer et al. 1986). Intensive monoculture beyond the phase-I stage to phase II (fingerlings, 2-6 months old) and phase III (yearlings, 6-18 months) requires a nutritionally complete diet. Commercially available trout or salmon feeds are generally used for striped bass culture (Bonn et al. 1976; Kerby et al. 1983; Zeigler et al. 1984) on the assumption that striped bass are physiologically similar to trout and salmon (Zeigler et al. 1984). It is not known, however, if trout and salmon feeds yield optimum growth and survival.

Intensively produced striped bass prefer to eat floating or slowly sinking pellets and seldom (if ever) eat feed that reaches the bottom. The feeding behavior of striped bass fed sinking trout pellets is difficult to observe in turbid ponds, but it is believed to be the same as that observed with floating or semifloating formulations in tanks. If a sinking feed is delivered too fast, a portion sinks to the bottom and is unlikely to be eaten. A desirable alternative is to offer only the floating feeds to fish in ponds. This would provide a visual check on the health and feeding behavior of the fish.

The objectives of our study were to (1) test four commercially available salmonid diets and one modified extruded catfish diet on fingerling striped bass reared in tanks, (2) compare growth of fingerling striped bass fed the modified catfish diet or a commercial trout diet in ponds, and (3) compare the growth of yearling striped bass fed the catfish diet or the trout diet in ponds.



Striped bass fry were obtained from the Marion (Alabama) State Fish Hatchery 1 d after hatching, stocked into indoor tanks supplied with 20C aerated well water, and fed Artemia salina. After 14 d, they were stocked into 0.04-hectare ponds and reared for an additional 6 weeks. In this pondculture phase, the fry were offered a commercial trout diet ad libitum from days 21 to 56. The fish were then divided into two groups; one group was restocked into the ponds and fed fingerling test diets and the other was placed into 1,600-L fiberglass tanks supplied with 20C aerated well water. The group in the tanks was fed a commercial trout diet for 5 months before the fish were restocked into 100-L tanks and fed the test diets.

Experimental diets (Tables 1, 2) consisted of commercial trout pellets (diet 1), a modified catfish pellet formulation (diet 2), a commercial feed advertised as consisting of 100% fish by-products (diet 3), the U.S. Fish and Wildlife Service open formula ASD-2-30 Atlantic salmon (Salmo salar) diet (diet 4), and the Service's open formula GR6-30 trout diet (diet 5). Proximate analysis of samples of feeds used in the study was determined by the U.S. Fish and Wildlife Service's Tunison Laboratory of Fish Nutrition, Cortland, New York.



TABLE 1.—Composition of the modified catfish diet (diet 2) fed to striped bass.

table1.gif (3037 bytes)


TABLE 2.—Proximate analysis of five diets fed to striped bass.

table2.gif (4894 bytes)

Tank experiments.—Fingerling striped bass were stocked into 100-L tanks supplied with 20C aerated well water at a flow rate of 3 L/min. The five diets (five tanks per diet, 25 fish per tank) were offered 2 d after the fish were stocked. Feed was offered three times daily, at 2% of the body weight at 0700 hours and 1100 hours and to satiation at 1500 hours. At the time of stocking and biweekly thereafter for 8 weeks, a random sample of 10 fish from each replicate tank was measured (total length) and weighed. Fish were not fed the day before, the day of, or the day after sampling. Feed allowances were adjusted at the beginning of each 2-week test period on the basis of group weight in each tank. All fish were returned to the tank after they were measured. At the end of the 8-week test, a 1-g muscle sample was removed from just below the dorsal fin of two fish from each replicate and tissue lipid concentrations were determined by the method of Bligh and Dyer (1959). Weight gain and muscle lipid content of fish on each diet were compared by single-classification analysis of variance. Two more fish from each replicate were selected at random for determination of visceral fat weight. All viscera were excised from each fish and stripped free of all fat, which was then collected and weighed. Analysis of covariance was used to detect differences in visceral fat (body weight was the covariant). Tukey's highest significant difference (HSD) was used in multiple mean comparisons. Treatment means were considered significant at the 0.05 level of probability.

Pond experiments.—Fingerling striped bass were randomly stocked into 0.02- and 0.04-hectare ponds; 10 ponds of each size were used and treatments were assigned to ponds at random. Each pond was equipped with a continuously operating airlift pump (10-cm diameter), which moved water at a rate of about 200 L/min when injected with air at a rate of 85 L/min (Parker and Suttle 1987). Aquazine was applied to control vegetation. Groups of five ponds each were stocked on 27 June 1985 at one of four rates: 5,000, 10,000, 20,000, and 40,000 fish/hectare. The required number of fish were hand-counted into each pond. Within each density, fish in two ponds were fed diet 1 and those in three ponds were fed diet 2. Feed was offered to the fish six times daily from automatic feeders. Feed amounts were projected on the basis of a survival decline from 100 to 80% over the course of the study and a food conversion value (food weight fed/fish weight gained) of 1.5. These feed amounts were considered optimistic and in excess of fish requirements.

The fish were harvested after 13 weeks, and the total numbers and weights of fish in all ponds were determined. Individual lengths and weights of a 20-fish subsample from each pond were determined at the same time. Data were analyzed by a two-way unbalanced factorial (diet x density) analysis of variance. Tukey's HSD was used to detect differences in treatment means.

In a companion test, another group of striped bass fry obtained from the Marion State Hatchery was stocked into a 0.2-hectare pond and fed a commercial trout feed to produce fingerlings. Management of the pond was similar to that of the other ponds in this study. After 5 months, fingerlings were harvested, and 220 of them were stocked into each of six 0.04-hectare ponds and reared to the yearling stage. These fish were handfed to satiation twice daily. Fish in three ponds were fed diet 1, and those in three other ponds were fed diet 2. Each pond was equipped with a continuously operating airlift pump to reduce stratification and to aid in the control of aquatic vegetation. Aquazine and emergency aeration were used as required.

Individual lengths and weights were determined for a sample of 20 fish per pond every 56 d and at harvest, after 11 months of feeding. Fish in each pond were hand-counted to accurately determine the percentage surviving at harvest. Data at each 56-d sample time and at harvest were analyzed by single-classification analysis of variance.


Protein content of the five diets tested ranged from 39.3% for the commercial bout diet to 58.1% for the Atlantic salmon diet (Table 2). Lipid content ranged from 3.4% for the modified catfish diet to 18.1 % for the commercial fish by-products diet.

Tank Experiments

There were no statistical differences in weight gain and condition factor (105 x weight/length3) of the striped bass fed diets 1, 4, or 5 over the 8-week feeding period in tanks (Table 3). Striped bass fed diet 3 had significantly lower condition factors than those fed diets 1, 2, 4, or 5 and, after an initial weight loss, maintained about the same weight over the remainder of the study. Fish fed diet 2 were intermediate in growth and condition factor between those fed diet 3 and diets 1, 4, or 5. After 8 weeks, striped bass fed diet 4 had significantly higher abdominal fat content than fish fed other diets and a significantly higher muscle lipid concentration than those fed diets 2, 3, or 5.


TABLE 3.—Mean fish weights, abdominal fat weights (adjusted by covariance for fish weight), muscle fat contents, total food conversions (weight of feed given divided by weight gain), specific growth rates, and mean condition factors of striped bass fingerlings fed five diets for 8 weeks. Parenthetic numbers are SDs. Values with the same letter within a column are not significantly different (P > 0.05)

table3.gif (10779 bytes)

Pond Experiments

In the pond study of diet x density, all fish in three ponds died because of low dissolved oxygen during cloudy weather—two ponds in which the fish were fed diet 1 (density of 20,000 and 40,000 fish/hectare) and one pond in which the fish were fed diet 2 (density 20,000/hectare). Statistical analysis of data for the remaining ponds showed there was no significant difference in survival (Table 4). However, there were significant differences in weight and condition factor between fish fed diets 1 and 2 at most stocking densities, and there were significant interactions between diet and density for fish weight and condition factor at harvest. Fish fed diet 1 were significantly larger and had higher condition factors than fish fed diet 2. Within diet 1, fish stocked at 40,000/hectare were significantly smaller than fish stocked at 20,000/ hectare. The significant interaction appears to result because within diet 1, fish stocked at 20,000/ hectare were significantly larger than those stocked at 10,000 or 40,000/hectare, and within the stocking density of 10,000 fish/hectare, there was no significant difference between diets.


TABLE 4.—Mean weights, condition factors, and survivals of striped bass stocked into ponds at four densities and fed two diets. Parenthetic numbers are SDs. The diet x density interaction was significant (P 0.05) for fish weight and condition factor. Values with the same lowercase letter are not significantly different between diets (compare down within the same density); those with the same uppercase letter are not significantly different between densities (compare across; P > 0.05).

table4.gif (10358 bytes)

Growth of striped bass stocked into ponds and fed diet 1 during winter to produce yearling foodsize fish was similar to growth of fish fed diet 2 during the same time (Table 5). However, fish fed diet 1 began increasing in weight in May faster than those fed diet 2 and were significantly larger in July and October than those fed diet 2. Condition factors were significantly higher in July and October in fish fed diet 2 than in fish fed diet 1. Mean survival of fish fed diet 1 was 42.7 2.7% (SD), which was significantly higher than the 27.5 5.4% for fish fed diet 2.


TABLE 5.—Mean weights and condition factors of striped bass reared to the yearling stage in ponds and fed two diets. Parenthetic numbers are SDs. Values with the same letter within each month are not significantly different between diets.

table5.gif (6373 bytes)



Millikin (1983) reported that a diet containing 47% protein and 12% lipid produced maximum growth by 7-g striped bass. In our study with larger striped bass in tanks, a diet (GR-6-30) with 44% protein and 13.2% lipid gave the highest specific growth rate followed by ASD-2-30 (58.1% protein and 15.6% lipid) and commercial trout grower (39.3% protein and 8.3% lipid). This is consistent with the physiological principle that the protein and lipid requirements may decrease as fish become larger. The Atlantic salmon diet (ASD-2-30) produced fatter, but not larger, fish when fed to satiation. Increases in body lipids of fish fed a diet high in protein has been previously reported (Cowey 1979). Feeding a rather small quantity of the Atlantic salmon diet (per body weight of fish) may result in similar weight gain due to the high energy content of the feed (5,040 kcal/kg) compared to that of fish fed larger amounts of lower energy trout feeds.

Striped bass fed the commercial fish by-products diet lost weight during the 8-week test period in tanks, even though the diet had a rather high energy content (4,530 kcal/kg). We do not know the cause of the poor growth, but perhaps the diet had inadequate amounts of essential vitamins or amino acids.

Growth of striped bass fed the modified catfish diet in tanks and ponds was less than that of fish fed the trout feeds. The catfish diet was modified in an attempt to make an extruded floating feed suitable for striped bass, but only about 20% of the pellets floated, perhaps because of the additional fish meal and fat. However, the remaining 80% of the pellets sank more slowly than those of the trout or salmon diets. The modified catfish diet seemed to provide adequate nutrition during winter, when water temperatures were below 10C and energy requirements were low. The slower sinking modified catfish diet may allow the fish extra time to feed in cold water and thereby attain growth similar to that in fish fed higher energy trout food. The economics of winter feeding is an important aspect of commercial striped bass culture that needs further study.

The diet x density interaction was difficult to measure because the population sizes during the culture period were not known; only the survival at the end of the study was available. Although there was no significant difference between diets 1 and 2 in growth of fish stocked at 10,000/hectare (Table 4), the average survival of diet-2 fish stocked at 10,000/hectare was 15.8%, leaving fewer fish than those stocked at 5,000/hectare with a survival of 60.8%. If most of the mortality occurred early in the culture period, the remaining fish in the 10,000/hectare treatment may have obtained sufficient supplemental nutrition from natural food in the pond to increase their growth to near that of fish fed diet 1. The best survival and largest fish fed diet 1 were harvested from the pond stocked at 20,000/hectare (Table 4). Striped bass swim and feed in schools in the wild and (based on observations at this station) in ponds as well. These data suggest that the stocking density of striped bass in ponds may be an important factor in fish weight gain because of the feeding competition that occurs at the higher stocking rates.

Of the feeds available, the U.S. Fish and Wildlife Service trout grower ration GR-6-30 or a commercial equivalent is suggested for production of fingerling and yearling striped bass. Further studies are needed to quantify specific vitamin, amino acid, and lipid requirements for all life stages of striped bass.



We thank C. J. Turner, Marion State Fish Hatchery, Marion, Alabama, for supplying the striped bass used in this study; G. A. Rumsey, U.S. Fish and Wildlife Service, Cortland, New York, for providing the proximate analysis of the diets; H. K. Dupree, U.S. Fish and Wildlife Service, Stuttgart, Arkansas, for assisting with formulation of the modified catfish diet; and G. A. Rumsey, H. K. Dupree, G. Ketola, and H. Poston, U.S. Fish and Wildlife Service, Cortland, New York, for reviewing the manuscript.



Cowey, C. B. 1979. Protein and amino acid requirements of finfish. Pages 3-16 in J. E. Halver and K Tiews, editors. Finfish nutrition and fish feed tech

nology, volume 1. Heenemann, Berlin.