occurred at the control site in Racetrack Creek. Mortality 

 in Silver Bow Creek and in the Clark Fork at Warm Springs, 

 Deer Lodge, and Bearmouth was statistically higher than at 

 the control site. Copper and zinc concentrations present 

 during the test are summarized in Table 3-30. 



Additional bioassays were conducted during May and June 

 of 1987 (Phillips and Hill, unpublished data) , including 

 tests of both rainbow and trout fingerling and swim-up-stage 

 fry. Test sites were the same as those used during 1986. 

 Tests began on May 4 and were completed by July 1. The test 

 vessels were eguipped with an automatic feeding system that 

 provided hatchery food to the fry four times per day. Fry 

 were less tolerant than fingerling during these tests. Rates 

 of mortality for fry were: Warm Springs (18 percent) , Gold 

 Creek (36 percent) , Bearmouth (55 percent) , and Silver Bow 

 Creek (92 percent) . Mortality at the latter three sites was 

 significantly higher than the 8 percent observed at the 

 control site and the 10 percent observed at Clinton (down- 

 stream of Rock Creek) . A pair-wise multiple comparison 

 technique was employed using Bonferroni adjusted confidence 

 intervals. Rates of mortality for fingerling were: Warm 

 Springs (7 percent) , Bearmouth (12 percent) , Gold Creek (24 

 percent) , and Silver Bow Creek (88 percent) . Mortality at 

 the control site was only 2 percent. Mortality at both Gold 

 Creek and Silver Bow Creek was statistically higher than the 

 control . 



The Warm Springs bioassay location is on the east side 

 of the river and is in the plume of Pond 2 discharge water. 

 During both the 1986 and 1987 bioassays, maximum and average 

 zinc concentrations were higher at this site than at 

 downstream sites. High metals concentrations below the ponds 

 occurred during periods of high winds .-that stirred up 

 particulate materials in the Warm Springs Ponds. Unlike 

 resident fish in this vicinity of the river, the bioassay 

 fish were unable to seek refuge from the higher metals 

 concentrations by moving into water originating from either 

 Warm Springs Creek or the Mill-Willow Bypass. Such movements 

 may allow resident fish to escape high concentrations of 

 metals. 



In summary, the instream bioassays indicate that early 

 life stages of rainbow trout are adversely affected in Silver 

 Bow Creek and in the mainstem Clark Fork. Statistically 

 significant mortality has been documented from Warm Springs 

 to near Bearmouth. This occurred even during years when 

 metals concentrations were relatively low because of modest 

 runoff. Tributaries that contribute good-quality water to 

 the river may provide potential refuges from high metals 

 concentrations, but the extent to which these are utilized by 

 resident fish has not been documented. Use of refuges may 



3-116 



