Control tank. — Total mortality of chinook and 

 coho salmon in the control tank was 2.2 and 4.1%, 

 respectively, during the 72 days of the test period (3 

 April to 13 June). These mortalities are not more than 

 would normally be expected from the effects of con- 

 fining and holding a population of fish for 72 days. Ex- 

 ternal symptoms of gas-bubble disease did not 

 develop on the fish nor were symptoms found on any 

 of the fish that died. Mortality data by date for tank 3 

 are compiled in Appendix Table 3. 



Deep tank. — Mortalities of chinook and coho 

 salmon in the deep tank were 8.7 and 4.2%, respec- 

 tively, during the 72-day test period (13 April to 23 

 June). Mortality by date is shown in Figure 5 and 

 Appendix Table 1. The first dead fish in this tank 

 with external symptoms of gas-bubble disease was a 

 chinook salmon that died on 8 June. On 10 June two 

 coho salmon died, one of which showed symptoms. 

 Prior to this, nine fish had died in this tank, none of 

 which showed external symptoms. 



Beginning on 26 May, nitrogen concentrations (at 

 the surface) in the deep tank remained above 120% 

 until the end of the test. It was during this time that 

 external symptoms of gas-bubble disease first became 

 evident on the fish in this tank (Table 4). Nitrogen 

 concentrations at the surface of 120% and 130% cor- 

 respond to 100% of saturation at 2.13-m (7-foot) and 

 3-m (10-foot) depths, respectively; they also corres- 

 pond to 110% of saturation at 1-m (3-foot) and 2-m (6- 

 foot) depths, respectively. 



25 - 10"- 



i MO 



z 



Accumulative % mortality 



_l tJ_ 



CalTo 



5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 



APRIL MAY JUNE 



Figure 5.— Nitrogen concentrations and accumulative mortal- 

 ity offish in deep test tank (river water, tank 2.5 m deep) during 

 test. 



Shallow tank. — Mortalities of chinook and coho 

 salmon in the shallow test tank were 98.2 and 80.1% 

 respectively. Mortality by date is shown in Figure 6 

 and Appendix Table 2. The first mortality in the 

 chinook salmon group occurred 5 days after the test 

 began on 3 April. In the coho salmon, the first mortali- 

 ty occurred on test day 8. Until that date, the concen- 

 trations of nitrogen gas in the river and the shallow 

 tank had been below 113% of saturation. Fifty percent 

 mortality was recorded by test day 50 for the chinook 



Table 4. — Incidence (percentage) of symptoms of gas-bubble disease on samples (20 fish each) taken from the 

 holding tanks on the dates Indicated. Included for comparison is the accumulative mortality (percentage) 

 from the shallow test tank. 



'Tank supplied with unaltered Columbia River water. 



^ank supplied with Columbia River water; nitrogen gas content of water had been reduced to about 1009 

 of saturation. 



6 



