By P. H. Davies, J. P. Goettl Jr., J. R. Sinley
Water Research, Volume 12, Issue 2, 1978, Pages 113-117, ISSN 0043-1354, DOI: 10.1016/0043-1354(78)90014-3.

The mean 96-h LC50's of silver with rainbow trout were 6.5 [mu]g l-1 and 13.0 [mu]g l-1 in soft water (approximately 26 mg l-1 hardness as CaCO3) and hard water (350 mg l-1 hardness as CaCO3), respectively. The long-term, 'no effect' concentration for silver, added to the water as silver nitrate, was between 0.09 and 0.17 [mu]g l-1 after 18 months exposure in soft water. The 'no effect' concentration is that concentration range which defines no observed effect. Based on mortalities different from the control, no mortalities attributable to silver occurred at 0.09 [mu]g Ag l-1, whereas 17.2% mortality occurred to fish exposed to 0.17 [mu]g ll-1. The 'no effect' concentration does not reflect possible effects of silver on spawning behavior or reproduction, since female rainbow trout will not generally reach sexual maturity before 3 yr. At silver concentrations of 0.17 [mu]g l-1 or greater, silver caused premature hatching of eggs and reduced growth rate in fry. In one experiment, the eggs were completely hatched within 10 days of exposure; whereas, control eggs completed hatching after 42 days. The prematurely erupted fry were not well developed and frequently died. The growth rate of surviving fry was greatly reduced.

More details...