hydrogen sulfide up to 0.008 mg/1, there was no appreciable difference be- 

 tween treatments and controls, except at 0.002 mg/1 where there was a 

 significant increase in growth. Reproduction (number of eggs deposited 

 per gram of female) in the second experiment was significantly reduced at 

 0.001 mg/1 H 2 S and completely inhibited at 0.003 and 0.008 mg/1 H 2 S (Table 

 5). In the experiment started with young-of-the-year no spawning occurred 

 after 826 days at concentrations of 0.002 mg/1 and higher. The failure of 

 egg deposition appeared to be caused by the inhibition or absence of normal 

 spawning behavior, since apparently average numbers of viable eggs for fish 

 of comparable size were found in ovaries of non-spawning fish. 



A two-generation, long-term experiment on fathead minnows (three repli- 

 cations) was also run to determine the "no-effect" levels of hydrogen sul- 

 fide. The test was started with eggs which were hatched and carried 

 through to spawning adults. The second generation was continued with eggs 

 from females reared in the same hydrogen sulfide test levels as the first 

 generation. The first generation was continued for 297 days and the second 

 for 404 days. Growth measurements were all made prior to the start of 

 spawning in both generations. The day length was reduced and lengthened 

 during the second generation to induce spawning, which resulted in a 

 lengthening of the total exposure period. Hydrogen sulfide ranges in the 

 first generation were 0.0004-0.0061 mg/1, 0.0007-0.0069 mg/1 during the 

 first 112 days of the second generation, and 0.007-0.0069 mg/1 during the 

 remainder of the cycle. For both cycles, mean temperature was 23 °C, pH 

 7.7, and 6.4-7.3 mg/1 (Table 6). 



Growth in weight after 112 days in the first generation was less at 

 all test levels. In the second generation, after 252-day exposure, there 

 was an apparent growth stimulation at 0.0007 mg/ H2S. Growth inhibition 

 in the second generation occurred in the early periods, but was less in 

 later periods, suggesting that early effects of exposure are greatest, and 

 tend to lessen as growth proceeds. These results may be influenced to 

 some degree by greater mortality of smaller fish in the hydrogen sulfide 

 treatments. Survival was much lower than in the control at the highest 

 treatment in both generations. In lower treatments, survival was also 

 lower than the control, but not markedly so. 



The success of spawning, measured by the number of eggs produced per 

 female, did not appear to be affected in the first generation at any level 

 of hydrogen sulfide treatment, although at the highest level, female sur- 

 vival was substantially lower than in the control. In the second genera- 

 tion, there was an apparent stimulation of egg production at 0.0007 mg/1, 

 but a reduction at higher levels. The number at the highest level may 

 have been increased by mortality of smaller females. 



DISCUSSION 



Both toxic and long-term toxic concentrations of hydrogen sulfide have 

 been shown to be lower than levels commonly found in natural and polluted 

 waters. Because sites of sustained hydrogen sulfide concentrations in the 

 ecosystem are frequently overlooked, or low levels are not identified, the 



120 



