Meanwhile, the survival rate in T3 was higher than that in T1 and T2, and the statistical difference was significant (p < 0.05) (Table 2). Fish death at T1 occurred in the second week until the end of the experiment, the death in T2 occured in the second week, fourth week, and end of the experiment, meanwhile the death in T3 occured in the third week until the end of the experiment. Tilapia death that occured in this study (especially in T1) was allegedly caused by exposure of ammonia during the maintenance periods. The brief exposure levels of ammonia in this study only have an impact on growth, but the ammonia will cause lethal effects when exposed for a long time. Ammonia toxicity assumed due to NH3 molecules has the ability to diffuse in the cell membrane (Colt 2006). Normally NH3 is excreted by passive diffusion across the gill epithelium, from blood to water in response to an NH3 gas partial pressure gradient maintained by combination of NH3 with protons formed from the hydration of excreted CO2 in the epithelial boundary layer. If the partial pressure of NH3 in the epithelial boundary layer is greater than that in blood, excretion of NH3 is inhibited, and NH3 will diffuse from water into the blood (Hargreaves and Kucuk 2001).