SUMMARY: Demand‐feeding behavior of juvenile yellowtails was examined under two different light conditions: in an experimental room and in the outdoors. In the indoor experiment, five groups of 20 juvenile yellowtails (mean bodyweight, 16.0 g) were allocated individually to five 200 L cylindrical tanks set up in the experimental room. A rod‐type tactile switch was used for activating the feeding device which delivered 5–10 pellets per actuation. Each tank was illuminated by an overhead fluorescent lamp with a LD 12:12 photoperiod regime. In the outdoor experiment, 69 juvenile yellowtails (mean bodyweight, 79.8 g) were allocated to a 7000 L cylindrical tank. A switch with a string that ends in a rubber, pellet‐like knob was used. The feeding device delivered 90–120 pellets by activation of the switch. In both experiments, yellowtail learned demand‐feeding within 1–3 days. The number of feeder actuations gradually increased during the experimental period in both experiments, presumably relating both to the rate of learning and to growth of the fish. The yellowtails in the indoor system showed a clear diurnal feeding pattern synchronizing to the given photoperiod, whereas those in the outdoor system showed a clear nocturnal feeding pattern also synchronizing to the outdoor natural photoperiod. Thus, the results showed that yellowtail has an ability for demand‐feeding both in light and dark phases. These exact opposite results may have been caused by the difference in intensity of the luminance between the inside of the experimental room and the outdoors.

No abstract

The effect of reward level (amount of food received in response to one trigger actuation) on the self‐feeding activity of juvenile yellowtails was examined. Three groups of 10 juvenile yellowtails (mean bodyweight, 65.6 g) were allocated randomly to three 200 L cylindrical tanks set in the experiment room. Each tank was illuminated by overhead fluorescent lamps with a light : dark photoperiod regimen of 12 h : 12 h (light 06.00–18.00 hours). A rod‐type tactile switch was used to activate the self‐feeding device. Three levels of reward were set up by installing a different number of identical feeders for each group: one feeder for the low‐reward group, three feeders for the medium‐reward group and five feeders for the high‐reward group. Each feeder delivered 3–7 pellets per actuation. Yellowtails in all groups showed a clearly diurnal feeding pattern that was synchronized to the given photoperiod. Average number of feeder actuations per day was highest (219/day) in the low‐reward group. Second highest was the medium‐reward group (116/day), and last was the high‐reward group (67/day). The resultant average feed intake per day was 27.1 g/day in the low‐reward group, 31.5 g/day in the medium‐reward group and 28.4 g/day in the high‐reward group. Thus, yellowtails appear to have the ability to regulate the amount of feed delivered daily by increasing the number of food demands when the reward is low and decreasing the number of food demands when the reward is high.

Self-feeding of grouped ayu Plecoglossus altivelis altivelis was examined under an artificial light-dark (LD) cycle and natural day length using a trigger with a photosensitive sensor. In Experiment 1, approximately 15 juvenile fish (mean body weight, 0.6 g) were kept in 60-L glass tanks (four tanks) under LD 12:12 (lighting period 06:00-18:00 hours) and self-feeding and locomotor activities were simultaneously recorded. Self-feeding was detected within one day and feeding activities were strongly synchronized to the LD cycle, with almost strictly diurnal feeding (98.2%). Although locomotor activities in the upper part of the tank were also greatest during the light phase (76.8%), the activity profiles were different from those of self-feeding, suggesting the validity of the photosensitive sensor. In Experiment 2, 25 fish (mean body weight, 15 g) were kept in outdoor 1000-L tanks (four tanks) under natural day length and water temperatures in May, and self-feeding was recorded from mid-August. Stable self-feeding was observed from early September in all the tanks. Nearly all food demands were diurnal, and usually crepuscular. Somatic growth was seen in all tanks. These results demonstrate that the self-feeding system using a photosensitive sensor is applicable for grouped juvenile ayu as small as 0.6 g and adult fish reared under natural conditions.

No abstract