7817 - Laserfiche WebLink

Home Browse Search

. a" , , 198 COMMUNICATIONS drained by a surface standpipe. Water exchange in breeder baskets was through the basket bottom and sides. Each of the test diets was fed to triplicate lots of fish for 40 d (until the fish were 45 d old). The five diets were AP-I00 (Zeigler Brothers, Inc., Gardners, Pennsylvania), UV (Farm and Wildlife Products, Inc., Omaha, Nebraska), 4200 (Bio-Ma- rine, Hawthorne, California), and A-250 and B-250 (BioKyowa, Inc., SL Louis, Missouri). All diets were subjected to proximate analysis according to standard methods (AOAC 1984). In addition, we determined relative particle sizes by sieving 1.0- g subsamples through three sizes of Nit ex. nylon bolting cloth: 102, 153, and 243 ILm mesh (Wild- life Supply Company, Saginaw, Michigan). Diets were sieved three times by shaking, except LIV, for which a small brush was used to preak up clumps of this fine-grained food. Fish were hand-fed at 5% of their body weight per day. Food was divided into eight feedings per day Monday-Friday and six feedings per day Sat- urday and Sunday. A subsample of fish was weighed with an electronic balance (sensitivity, zO.OI g) at the initiation ofthe study to determine daily feeding rate. Fish were not weighed during trials for feeding rate adjustment, because addi- tional handling of the small fish could have in- creased mortality and thus confounded treatment effects. Tanks and breeder baskets were cleaned by siphoning, and fish deaths were recorded daily. All fish were removed from each basket after 40 d and weighed as a group. The largest and smallest individuals in each replicate were mea- sured (TL) to determine size range. The stated size of fish for each treatment was obtained by sum- ming the lengths of the large and small individuals for the three replicates and using the resultant av- erage. The size range for each treatment was the absolute range among replicates. Survival and av- erage weights were compared by one-way analysis of variance (ANOV A) and Duncan's multiple- range test. Analysis of covariance (ANCOV A) was used to compare weight gain with survival within diets (Sokal and Rohlf 1981). Tests of significance were not performed on total length data. Results and Discussion Growth and Survival Diets A-250 and 4200 produced the highest weight gain (0.092 and 0.066 g, respectively) but the lowest survival (32 and 20%, respectively; Ta- ble 1), and diets UV and B-250 produced the ",. '. highest survival (78%) but the lowest weight gains (0.026 and 0.031 g, respectively; Table 1). Diet AP-loo was intermediate for both weight gain (0.036 g) and survival (59%; Table 1). Fish density (survival to 45 d of age) had no discernible effect on average weight gain within diets (ANCOV A, P = 0.36). However, fish density had a weak in- verse relationship with weight gain when all diets were combined (r = -0.67, P < 0.01). Papoulias (1988) found that razorback sucker larvae held at 150C experienced a 10% mortality rate if not fed by 11 d after hatch, and this rate increased to 93% if the fish were not fed until 31 d after hatch. We assumed that fish in our study that survived 40 d were accepting feed, and that the diff'erences in survival were due to the diets. Fish size at completion of the study (range, 11- 27 mm TL; average, 18.3 mm) waseomparable to that obtained with live nauplii of brine shrimp (Artemia salina). Papoulias (1988) reported sur- vivors that were 11.8-28.8 mm TL (six trials; av- erage, 18.8 mm TL) for larvae fed brine shrimp for 50 d. In trials with abundant food (50-1,311 naupllilfish daily), fish survival was high (80-93%) and average fish size was 21 mm TL (Papoulias 1988). We presume that our ration was limiting and that better growth would have resulted had more food been available to the larvae. Although the initial ration was 5% of body weight per day, the average ration at completion of the study ranged from 0.9 to 2% of body weight per day for the different diets. Proximate Composition and Particle Size The five closed-formula diets varied in proxi- mate analysis and particle size (Table 2). All diets had similar protein content (range 49.8-57.0%) but varied in fat content (range 6.4-30.2%). The two diets producing the highest survival (LIV and B-250) diff'ered in fat content and dry particle size. The UV diet had a higher fat content (30.2%) and smaller particles (average size, 50 ILm;. all < 102 ILm), and B-250 had lower fat (14.6%) and larger particles (average, 240 ILm; 87% > 153 ILm). How- ever, both diets had a ftourlike consistency. Diets A-250 and 4200, which produced the largest weight gains but lowest survivals, had different amounts of fat (20.1 and 6.4%, respectively); both diets consisted of larger particles (averages, 284 and 185 ILID, respectively). We hypothesized that fish growth and survival could be associated with diet particle size. How- ever, a regression of dry particle size with fish survival and growth (r = -0.36, P = 0.19 for