40 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



cordino^ to subsequent findings of the P^xtension 

 Service in Indonesia, good growth is limited by a 

 salinity of 4 percent and spawning by one of 

 3 percent." Brock (1954) reported tilapia spawn- 

 ing in sea water of a chlorinity of 19.'29 "Ao 

 (equal to a salinity of about 34.85 %o) and a pH 

 of 7.95. 



Four tanks were used in this experiment: two 

 test tanks contained brackish water and two con- 

 trols contained fresh water. The experiment was 

 started in August 1959 and terminated during the 

 latter part of September when heavy mortalities 

 among the adults caused by either asphyxiation 

 or hydrogen sulfide occurred in both brackish- 

 water tanks. Each tank was stocked with 36 

 females and 12 males, that is, with a 3 : 1 sex ratio, 

 and at a concentration of 1 square foot per 

 individual. 



The yield of fry for the 2 months of the experi- 

 ment is given in table 13. The total production 

 was almost three times as great from the treatment 

 tanks (brackish water) as from the controls. 

 Analysis of variance indicated a significant dif- 

 ference (F=472.65, f <0.05) between fry produc- 

 tion with respect to brackish- and fresh-water 

 methods of culture, but no significant difference 

 between replicates. 



Exactly what influence the saline environment 

 exerts in bringing about this increased fry pro- 

 duction is not known, but the results of this ex- 

 periment emphasize the desirability and possibili- 

 ties of tilapia culture in brackish water. 



FACTORS AFFECTING GROWTH 

 OF YOUNG 



The effects of environmental factors on growth 

 of young tilapia were studied experimentally at 

 the Kewalo plant for 12 weeks, beginning in April 

 1959. The effects of si)ace or density of the fish, 

 diet, and salinity were examined during the 

 experiment. 



"We tried to vary one factor at a time, keeping 

 the others constant, so that the single factor under 

 observation could be evaluated with some degree 

 of precision. We did not attempt to control the 

 temperature, since differences among the tanks 

 were not significant. Other factors, such as 

 volume of water in the tank and the rate of water 

 flow, were held as uniform as possible. In each 

 experiment, the fish were fed at the same rate per 

 fish regardless of lot size. An excellent quality 

 trout feed was fed to the fish twice a day, except 

 weekends, at the regular feeding times. 



Once each week, length and weight measure- 

 ments were made on a randomly collected sub- 

 sample from each tank under observation. All of 

 the fish in the subsample were returned to their 

 res])ective tanks after measurements were com- 

 pleted. The number of fish in a subsample varied 

 with each experiment. 



We realize that, ideally, all phases of the ex- 

 periments should have been conducted simultane- 

 ously. This was not possible, however, because of 

 a lack of sufficient quantities of fry. As a result, 

 different phases of the experiments were com- 

 menced as fr}' became available in adequate 

 amounts to stock the tanks. 



CONCENTRATION OF FISH AND GROWTH 



A series of tests to determine the rate of growth 

 of young tilapia in relation to their concentration, 

 or the amount of space available for growth, was 

 started at the Kewalo plant in April 1959. Five 

 lots of ivy, ranging in number from 1,000 to 0,000. 

 were placed in the assault-boat tanks; however, all 

 tanks were not stocked simultaneously. 



The fish were fed trout feed (starter, fry, and 

 small fingerling grades) for 12 weeks. Tlie par- 

 ticle size and amount of feed was gradually in- 

 creased as growth progressed. The fisli were fed 

 twice daily except on weekends, when they were 

 fed once a day. 



