Experimental Design
The trial was designed to evaluate the effect of feeding frequency on growth, survival, and feed conversion ratio (FCR) of the Pacific white shrimp, L. vannamei reared for eight weeks and fed commercial diets manufactured by two different processes, extrusion, and pelletizing. The design had a total of eight treatments and four replicates of each one, with ad libitum ration type.
Experimental System
The experimental system consisted of 36 fiberglass 60-L aquariums, which have a single seawater inlet filtered through a sand filter (70 µ), cartridge filter (10µ), and ultraviolet light. The water temperature in the tanks was controlled by 250-W submersible heaters. Aeration was provided by an external exhaust in each tank, which created a smooth water flow inside the tanks so as not to cause feed leaching. Water exchange of 75%/day was performed in all tanks once a day at 08:00 h. The experimental system illumination was controlled by a clock to maintain a constant photoperiod of 12 h light/12 h dark; the photophase was from 06:00 to 18:00 h.
Diets
Extruded and pellet diets were manufactured by commercial methods (Gisis S.A., Durán-Ecuador) and formulated to fulfill or exceed the National Research Council (NRC, 2011) requirements for shrimp.
The diets were equivalent with a 96% match of ingredients and were produced with the same grinding specifications. The pelletized feed diameter was 1.83 ± 0.08 mm and 4.02 ± 0.38 mm in length, whereas extruded had 1.83 ± 0.06 mm and 3.15 ± 0.31 mm in length. The dietary nutrient values were 35% of crude protein and 7% fat.
Management
Five thousand shrimp were collected from a grow-out pond (Rancho Arroyo Seco, Baja California Sur, Mexico), transferred to the Aquatic Nutrition Laboratory of the Centro de Investigaciones Biológicas del Noroeste (CIBNOR) in La Paz, Baja California Sur, Mexico, acclimated to laboratory conditions (temperature 27 ± 0.5 °C, salinity 38 ± 1 g/L and dissolved oxygen >5 mg/l), and held one week in several tanks supplied with filtered seawater (50% water replacement daily). Shrimp were fed twice a day with a commercial diet containing 35% protein until the organisms reached approximately 1 g. Just before stocking, 100 shrimp were randomly selected and individually weighed to the nearest 0.01g (Ohaus® and Scout® Pro Balance, Ohaus Corporation, NJ, USA) to estimate mean and standard deviations.
Growth trial
Shrimp were weighed individually, selected according to the required weight based on the experimental design and randomly stocked at a density of N = 8 organisms/tank. The global coefficient variation (CV) was under 10%. The mean ± standard error of the mean (SEM) across all tanks was 1.00 ± 0.01g. Four replicate tanks were randomly assigned for each treatment. Shrimp were maintained during the experiment in filtered seawater at 27 ± 0.1 ⁰C, 38 ± 0.2 g/L, 5.50 ± 0.2 mg/ml dissolved oxygen, and pH 7.6 ± 1.4.
The feed was administered slightly in excess of the expected satiety. Uneaten feed in all treatments was removed by siphoning and scored (counted) in each tank daily at 08:00 h, with the scoring used to estimate the amount of uneaten feed (number of pellets x average pellet weight) and adjusted to the following daily ration according to the feed intake score. The complete daily ration was weighed, recorded, and divided into similar amounts by volume. Daily feed amounts were divided into four feed frequencies for each type: (1) once a day (10:00 h);
(2) twice a day (10:00 and 12:00 h);
(3) four times a day (10:00, 12:00, 15:00, 18:00 h);
(4) six times a day (10:00, 12:00, 15:00, 18:00, 21:00, 24:00 h).
The first two rations were administered manually, while the following treatments were fed using Fish Mate F14 Automatic feeder (Pet Mate, Surrey, UK) for logistical reasons. At the beginning of the day, survival was recorded, and the automatic feeders were visually checked and any feed remaining recorded. Each tank was aerated with a single air diffuser, and dissolved oxygen, pH and salinity were monitored in each tank thrice weekly. The temperature was monitored every day in each tank. Feed was supplied to approximately 8% (Tacon, Jory, and Nunes, 2013) of the biomass in each tank, assuring a marginal excess. Water parameters, such as total ammonia nitrogen (TAN) and nitrite-nitrogen were analyzed every 20 days through a chemical colorimetric kit (Aquarium Pharmaceuticals, Mars Fishcare North America, Inc. Chalfont, PA, USA).
Calculations
Growth performance and survival of L. vannamei for all the groups were calculated with the following equations:
Weekly growth (g/week) = (Wf - Wi)/weeks
Survival (%) = final number of shrimp / initial number of shrimp × 100
Specific Growth Rate (SGR) = LN (Wf/Wi) / t
Daily Feed Intake (DFI) = Tf/Average biomass/t
Feed Conversion Ratio (FCR) = Tf / (Wf - Wi)
Where: Wf represents the final body weight (g), Wi is the initial body weight (g),
Tf is total feed consumption (g), t = time in days.
Statistical analyses
The statistical analyses were performed using Statgraphics® Centurion™ XVII (Copyright 1982-2014 Statpoint Technologies, Inc, Warrenton, VA, USA). Growth, survival and FCR were analyzed by one and two-way analysis of variance (ANOVA) to determine if significant differences (p < 0.05) existed among treatment means. Tukey’s honestly significant difference (HSD) test for mean separation was used to evaluate significant differences among treatments. The statistical model included the manufacturing process and feeding frequency as the main effects. Pairwise comparisons were performed separately where significant interactions within each level of the main effects were found using Tukey’s test.