TANK CULTURE OF TILAPIA 



35 



Table 7. — Fry produced in tank 3, per female and month, 

 at the pilot plant, December 1956-July 1958 



1 Because of mortalities and transfer of fish, the number of females used to 

 calculate the production per female is based on the number of females that 

 were present in the tank for more than 2 weeks. 



A second experiment in which the effect of tem- 

 perature on production of young was examined 

 was conducted at the Kewalo plant from Janu- 

 ary to August 1959. 



For tihipia to be most useful as a supplementary 

 skipjack bait, there must be a stock of bait-size 

 fish on liand in May or June, at the beginning of 

 the main fishing season in Hawaiian waters. 

 To achieve this, heavy fry production must be 

 under way by late winter. Brock and Takata 

 (1955 : p. 24) reported that tilapia spawn through- 

 out the year in Hawaiian waters, but that the 

 spawning is less intense during the winter months. 

 Consequently, an experiment was conducted at 

 the Kewalo plant to determine if raising water 

 temperatures would induce tilapia to spawn at a 

 high rate during the winter months. Three red- 

 wood tanks (tanks 13, 14, and 15) were arranged 

 as follows for the experiment : 



1. Tank 13 was not modified in any way and 

 served as the control. 



2. The water in tank 14 was artificially heated 

 with a 60-foot, lead-sheatlied heating cable rated 

 3.63A-115V, tliat produced 400 watts, or 6.7 watts 

 per foot. A thermostat with a capillary tube was 

 placed in the tank to control tlie temperature. A 

 cover made of sisal-glaze, a clear, longlasting 

 plastic, was placed over the tank (fig. 8) to pre- 

 vent excessive heat loss, especially at night. 



Figure 8. — Plastic cover over artificially heated tank 14. 



3. The water in tank 15 was also artificially 

 heated with the same type of heating cable used 

 in tank 14, but the tank was not covered to pre- 

 vent heat loss. 



Each tank was stocked with 32 males and 64 

 females and fed trout feed. Emerging iry were 

 collected and counted daily. The daily variation 

 in water temperature was recorded by thermo- 

 graphs and minimum and maximum thermom- 

 eters. 



The production per female per month and the 

 average and range of minimum and maximum 

 temperatures for each of the three tanks are re- 

 corded in table 8 and shown graphically in figure 

 9. In January, tank 14, which was modified with 

 heating cable and cover, had relatively better pro- 

 duction than the other two tanks. In February, 

 artificial heating was started in tank 15, and re- 

 sulted in a marked increase in production compa- 

 rable to that in tank 14. Production was still low 

 in the control tank. In March, with rising air 

 temperatures the water warmed in all three tanks 

 and production increased in all; however, the in- 

 crease in the slightly warmed tank 15 greatly ex- 

 ceeded that in the other tanks. Temperatures re- 

 mained alxiut the same in April, but production 

 dropped, particularly in tanks 14 and 15. With 

 liiglier temperatures in May, tlie control tank 

 maintained its sligiit lead over tank 15, and dur- 

 ing the last 3 months of the experiment, it out- 

 produced the two artificially heated tanks by a 

 significant margin. Over the coui-se of the experi- 

 ment, total fry production in tank 13 (control) 

 was about twice that in the other two tanks. 



