Materials and methods

Eggs and larval rearing

PBT fertilized eggs were obtained by spontaneous spawning of cultivated brood stock fish in a sea net-cage at the FLKU. Eggs of the late embryonic stage were placed in cylindrical fibreglass tanks (1.0 kl; 135 cm in internal diameter and 70 cm in depth) at a density of 6000 eggs per tank. They had a normal hatching rate of 98.7, 88.3 and 93.8% in Experiment 1, 2 and 3 respectively. Larvae were subsequently subjected to each rearing experiment, detailed below, in the same tanks. The dph was defined to hatching day as 0 dph.

As in study of Section 3.1 in this chapter, incubation temperature of eggs and hatched larvae before feeding was set at 23 and 25°C, respectively, and larval rearing temperature until the end of the each experiment after feeding was set at 26.5°C.

To prevent larval surface death in prelarvae rearing, 0.3 ml per tank of feed oil added as drops every day onto the rearing water surface to form a surface film from hatching to 2 dph as in the previous report and section (Munday et al. 2003; Section 3.1 in this chapter).

The larvae were fed with S-type rotifers Brachionus plicatilis sp. complex (Hagiwara et al. 2007), enriched with Nannochloropsis oculata of 1.5 × 10⁷ cells/ml, a commercial product (Marine Glos EX; Nisshin Marinetech, Yokohama, Japan) and taurine of 0.4 g/l (Japan Nutrition, Tokyo, Japan), from 2 dph onwards at a food density

68

of 5–15 ind./ml. Water exchange in each tank was provided by flow through sand filtered sea water at 1000 l/dayfrom feeding to the end of the each experiment. Natural sunlight was attenuated using a plastic sheet, which covered the experimental facilities.

In addition, subsidiary fluorescent lighting was used within the natural light period, which includes fade-in and fade-out of light intensity after the start and before the end of light period respectively. In this study, dark period and light period were defined as the period when the light intensity was 0 μmols/m², and the period other than the dark period respectively. Light intensity was measured using alight photon meter (MDS-MkV/L; Alec Electronics, Kobe, Japan) every 10 min during light period for experimental period in Experiment 3 (refer to Fig. 3.2.3). Light period was 14 h (natural light period; 05:00–19:00). N. oculata was added every day to the rearing water from 2 dph to the end of experiment.

Other rearing conditions were as follows: rearing water temperature, 26.5 ± 0.1 ºC; salinity of 31.2 ± 0.3; dissolved oxygen, 107.4 ± 6.3 %; and pH, 7.88 ± 0.36 in Experiment 1; temperature, 26.5 ± 0.1ºC; salinity, 31.6 ± 0.3; dissolved oxygen, 101.0 ± 2.2 %; and pH, 8.13 ± 0.03 in Experiment 2; temperature, 26.5 ± 0.1 ºC; salinity, 31.6 ± 0.3; dissolved oxygen, 100.3 ± 1.9 %; and pH, 8.17 ± 0.10 in Experiment 3.

Experimental design

A series of three rearing experiments (Experiment1, 2, 3) was conducted to verify the optimal timing of the day to promote ISI using different batches of fertilized eggs, which were spawned on different days. The examined timings in the experiments were narrowed down based on the results of preceding experiment (Fig. 3.2.1). The rearing experiments with SFR were carried out from 3 to 9 dph as in our previous study.

69

Fig. 3.2.1. Timing of surface film removal to promote ISI in each experiment. Black areas and grey areas represent the timing of surface film removal and the dark period respectively.

Initially, in Experiment 1, larval swimbladder inflation frequency was compared among the following different SFR timings of the day: from 05:00 to 19:00 (i.e. during the light period: S.5–19), from 19:00 to 05:00 (i.e. during the dark period:

S.19–5), from 08:00 to19:00 (the general running period of surface skimmer in hatchery: S.8–19) and the entire day (i.e. during 24 h: S.24).

After getting the results of Experiment 1, Experiment 2 was conducted to refine the necessary period of SFR during the light period, where the effect of SFR on larval swimbladder inflation frequency was compared among the following different SFR timings of the day in: from 08:00 to 19:00 (same as the treatment of S.8–19 in Experiment 1:S.8–19-E2), from 08:00 to 13:00 (S.8–13) and from 13:00 to 19:00 (S.13–

19).

Based on the results of Experiment 2, Experiment 3 was conducted to refine the necessary period of SFR in the afternoon, where larval swimbladder inflation frequency

Experiment 1 S.24 S.5-19 S.19-5 S.8-19 Experiment 2

S.8-19-E2 S.8-13 S.13-19 Experiment 3

S.13-16 S.16-19 S.18-19

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Time (hour)

70

was compared among the following different SFR timings of the day: from 13:00 to 16:00 (S.13–16), from 16:00 to 19:00 (S.16–19) and from 18:00 to 19:00 (S.18–19).

In these experiments, the feed oil was added during the time period other than SFR timings from 3 to 9 dph to form the surface film to inhibit larval air gulping for ISI.

The surface skimmers for SFR used in this study were same as that used in Section 3.1 in this chapter.

There were three replicates per treatment for Experiment 1 and 3, two replicates per treatment in Experiment 2.

Aeration

Aeration was provided at the centre of tank bottom using an air stone (100 mm long, 23 mm diameter: Ibuki air stone #100; KING Stone, Tokoname, Japan) with air-flow rate of 130 ml/min in the light period. In the dark period, another air stone with air-flow rate of 1,300 ml/min was additionally placed at the centre of tank bottom from 2 to 9 dph to mitigate sinking death (Tanaka et al. 2009; Nakagawa et al.2011).

In S.5–19, S.19–5 and S.24 of Experiment 1, aeration was provided at the centre of tank bottom using an air stone with air-flow rate of 650 ml/min during 24 hours through the experimental period without additional aeration during the dark period to examine the effect of timing of SFR on swimbladder inflation frequency under a constant aeration condition.

Measurements and observations

Thirty larvae were examined to determine the swimbladder inflation frequency and 15 larvae were measured standard length (SL: length from the rostral tip to the end

71

of the notochord) for each tank at the end of the experiment (9 dph), and 6 dph when success of swimbladder inflation is likely to be important to prevent larval mortality caused by sinking death in PBT larviculture in mass production tanks (Chapter 2).

Larval sampling was conducted in nighttime (21:00–22:00) because PBT larvae deflate the swimbladder in daytime (Takashi et al. 2006) and this makes it difficult to determine the correct swimbladder inflation frequency. Larvae were anaesthetized in 0.06%

2-phenoxyethanol (Trotter et al. 2005) to observe swimbladder inflation under a stereomicroscope. SL was measured in digital images of samples taken by a digital camera (Moticam 2000, Shimadzu Rika, Tokyo, Japan) using software package for image analysis (MoticImages Plus 2.2s, Shimadzu Rika).

At the end of the experiment, all surviving larvae were collected and fixed in 5% formalin solution for counting the number to evaluate the survival. In the evaluation of survival, the initial number of larvae in each tank was estimated by multiplying the number of introduced eggs by the hatching rate as stated above.

To examine the diel change in occurrence frequency of surface death, larvae trapped at the rearing water surface and in the surface skimmer were counted as the surface death larvae every 2 h for 24 h on 3 dph when is the effective period to promote ISI and the number of surface death larvae reached the maximum level (Section 3.1 in this chapter), in the three tanks of S.24 treatment in Experiment 1 where the running of surface skimmer and air-flow rate of aeration were constant for 24 h (n = 3).

Statistical analysis

Significance was tested using Tukey-Kramer test in differences among the treatments on SL in each experiment, on survival and swimbladder inflation frequency

72

in Experiment 1 and 3, and among each time on the number of surface death larvae in