FISHERY BULLETIN: VOL. 84, NO. 4 



to Galveston (Yang et al., 1980b; 1983a, b). Only 

 early stage eggs were shipped and cultured (never 

 beyond stage 19, Arnold 1965). The eggs were ac- 

 climated gradually to the temperature and salinity 

 of the culture tank water; incubation temperature 

 was maintained around 15°C while salinity ranged 

 between 34 and 36%o. Bundles of a few capsules 

 each were suspended from a rack at the water sur- 

 face to ensure oxygenation and uniform develop- 

 ment of eggs. Styrofoam panels covered the rear- 

 ing tank and the illumination level was kept below 

 1 lux to prevent the growth of benthic diatoms on 

 egg capsules. 



A circular tank (CT) system consisting of two cir- 

 cular tanks (each 1,300 L) was used for incubation 

 and early rearing of hatchlings and juvenile squid. 

 Water circulation was modified in L.O. 1982 when 

 compared with earlier culture experiments (Yang 

 et al. 1980b: fig. 1, 1983a: fig 1). Prior to L.O. 1982, 

 a laboratory-constructed particle/carbon filter was 

 used with circulation first passing through an ultra- 

 violet (UV) sterilizer. L.O. 1982 used modular type 

 particle and carbon filters, with the UV sterilizer 

 in the last position in the water conditioning pro- 

 cess. The raceway (RW) system (RW culture tank 

 volume-10,970 L in L.O. 1981, and 13,180 L in L.O. 

 1982) was used for final grow-out after transferring 

 the squid from the CT culture tanks. The transfer 

 was necessary to give the squid greater horizontal 

 swimming space. The initial RW system in experi- 

 ment L.O. 1981 had been modified from previous 

 experiments (Yang et al. 1980b, 1983a) to improve 

 water quality by 1) adding a rectangular, 960 L 

 capacity water conditioning tank (0.46 x 1.22 x 

 1.83 m, water depth 0.43 m) with water circulation 

 of 54 L/minute, 2) adding another cooling unit, 

 3) adding three protein skimmers, 4) adding three 

 UV light sterilizers (each 30 W, total 90 W), 5) modi- 

 fying the water uptake system in the RW with a 

 float near the center to remove near-surface water 

 without sucking up squid or food organisms and to 

 increase the lateral swimming space for the squid, 

 6) painting an irregular mottled pattern on the sides 

 of the RW to make the walls more visible to the 

 squid, and 7) most importantly, by increasing RW 

 water depth gradually from 24 cm initially to 40 cm 

 (average depth 38.8 cm) to provide swimming space 

 for the squid and to increase the average culture 

 water volume in the RW from 5,990 to 8,610 L. 



A further improved RW system (Fig. 1) was used 

 in experiment L.O. 1982. It consisted of two bio- 

 logical filter tanks (A, C) with oyster shell subgravel 

 filters and airlifts for water circulation, a tank for 

 growing macroalgae (B), the RW where the squid 



were cultured (D), and a separate tank where pro- 

 tein skimmers were operated continuously (E). The 

 surface water was taken from the RW through pipes 

 suspended in a screened floating core. Water within 

 the system was recirculated by three routes. First, 

 water was pumped to filter tank A that contained 

 approximately 0.15 m 3 of oyster shell over a false 

 bottom. Water passed through the filter bed, then 

 flowed through a constant-level siphon to tank B 

 where algae were illuminated by two 400-W metal 

 halide lamps. Water flowed by gravity into the sec- 

 ond filter tank C that contained 0.18 m 3 of oyster 

 shell substrate and two 1-hp cooling units, and final- 

 ly returned by gravity to the RW proper. Second, 

 water was pumped through two sets of six modular 

 filters: four modules containing pleated 20 ^m fiber 

 particle filters and two containing activated carbon. 

 From the modular filters, water either flowed direct- 

 ly into the RW or through a 60 W UV sterilizer 

 before returning to the RW. Third, water was 

 pumped at 36 L/minute to a tank that contained five 

 protein skimmers and then flowed back into the RW. 

 The outflow of the three recirculating routes created 

 a clockwise water flow in the RW proper. This mo- 

 tion accumulated dead squid and food organisms in 

 one place on the bottom. The bottom was painted 

 solid black with nontoxic Thixochlor 2 paint and the 

 sides were painted with an irregular mottled pat- 

 tern. Three 11 x 28 cm windows were mounted in 

 one side of the RW for observing the squid's feeding 

 and behavior. The tanks were insulated with poly- 

 styrene sheeting and 2.3 cm thick polystyrene 

 covers. 



To ensure activation of the biological filter for both 

 CT and RW systems, filter beds were inoculated 2 

 to 3 wk beforehand with nitrifying bacteria on oyster 

 shell from other systems. Fish and shrimp were 

 placed in the water conditioning tank to build up the 

 bacterial population. Thus the filter beds were estab- 

 lished by organic conditioning methods (Moe 1982) 

 instead of by directly adding ammonia source 

 chemicals. 



A set of black silk nets was used to transfer squid 

 from the CT system into the RW system. A tri- 

 angular lift net was laid on the bottom of the tank 

 while two rectangular net curtains were slowly 

 drawn from the left and right sides of the tank to 

 concentrate the small squid above the lift net. The 

 lift net was gradually raised, a wash tub placed 

 underneath, and both were moved to the RW tank 

 where the squid were gently released into the tank. 



2 Reference to trade names does not imply endorsement by the 

 National Marine Fisheries Service, NOAA. 



772 



