|
<br />DeForest, Brix, amI Adams
<br />
<br />Debate/Commentary
<br />
<br />Hilton, Hodson, and Slinger (1980). Rainbow trouL wcre fed six differcnL
<br />dieLs conLaining measured seleniLe concenLraLions of 0.2, 0.4, 1.3, 3.7, .\lId
<br />13.1 mg/kg dw. Trout fed the 13,1 rng/kg dw dicL had a significantly (jJ < 0,05)
<br />lower body weighL and higher mortality rate than fish in the oLher Lreatments,
<br />resulLing in a dieLary chronic value of 6.9 mg/kg dw (Table 2). For compari-
<br />son, Lcmly (l993a) noted that this study suggests thaL mortality results aL
<br />dietary concenLrations greater than 3 mg/kg dw. HilLon elal, (1980), however,
<br />report thcre were no mortalities in fish fed a diet cOIILainillg 3.7 mg/kg dw
<br />selenium. This study appears to be the basis of Lemly's recommended dietary
<br />Lhreshold of 3 mg/kg dw.
<br />Lemly (1993a) also reports from this study thaL a whole hody selenium
<br />concentration of 5 mg/kg dw resulted in fish mortality (TalJle 3). Ililton el al.
<br />(1980) did not measure wholc body seleniulll concentrations, but conccntra-
<br />tions in the carcass, kidney, and liver were approximaLely 5, 50, and
<br />100 mg/kg dw. Given the relatively high selcnium concentrations in the kid-
<br />ney and liver, iL is uncertain how welI the carcass concentration represenLs the
<br />whole body concenLraLion. Comparison to the interpretaLion ofJuvinen and
<br />Ankley (1999) is noL possible because they did not report whole hody residues
<br />from Lhis sLlldy.
<br />Hicks, Jlilton, and Fergu501l (1984). .Juvenile rainbow LrouL were /Cd dieLs
<br />containing selenite concentrations 01'0,6,6.6, and 11.4 mg/kg dw 1(11' J Ij weeks
<br />in the laboratory. Trout fed the 11.4 mg/kg dw diet had significantly (jJ < (),05)
<br />reduced growLh and inneasecl mortaliLy relative LOLhe controls. No cllccts
<br />were obselved in fish fed the 6.t; mg/kg dw dieL, resulting in a chronic value
<br />of 8.7 mg/kg dw. These I'esults indicate that I'aillbow Lrout mortality docs not
<br />occur at a dietary concentration of 3 mg/kg dw. Lcmly (1993a) did 1I0t review
<br />this study.
<br />HUlm, Hamilton, and Buckler (1987). Rainbow trout were exposed to aqueous
<br />concentrations of selenite in the labol'aLOI)' for 90 days. The test was initiated
<br />using sac fry. Percent survival was significantly (p < 0.05) reduced relative to
<br />the control for fish wilh a mean whole body selenium concelllratioll of
<br />4.3 mg/kl{ dw. No signilicanL elIects on mortality were ohselved in fish with a
<br />mean whole body selenium concelllraLion of 2.6 mg/kg dw, rcsulting in a
<br />chronic value for this endpoiut of 3.3 mg/kg dw (Table 3). l.emly (I !19:la)
<br />similarly reported Lhat mortality was associated with a whole body residne of
<br />4 mg/kg dw..Jatvinen and Ankley (1999) did not cite this stndy in their review.
<br />Because the study only evalnated the water exposure pathway it was not
<br />considered in our development of tissue thresholds,
<br />
<br />mosquitofish from San Luis Drain used in a similar sLudy contained elevated
<br />concentrations of boron, chromium, and strontium relative to controls
<br />(Hamilton el ai" 1990). Moreover, pesticide concentrations have not been
<br />measured, but could be relatively high. For example, ODD, DOE, and DOT
<br />have been deLected ill tissuc collected ill Mud Slough, which runs parallel to
<br />Lhe San Luis Drain, at concenLratiollS ranging up to 79.3 jlg/kg ww (samples
<br />collecLed in 1992) (Brown, 1997). Hamilton el al. (1990) (summarized below)
<br />conducted additional Lests wilh chinook salmon, including a test where salmon
<br />were exposed to a reference diet fortified with organoselenium. Given Lhat
<br />these clata are available, and due to the uncertainties in linking efTects to
<br />selenium residues in fish that may contain elevated levels of several chemicals,
<br />thc clata in Hamilton el ai, (198li) were noL considered appropriate for devel-
<br />oping selenium guidelines,
<br />Hamiltoll el ai, (1990). Chinook salmon were fed diets containing eiLher
<br />high-seleniulll Illosl)uitollsh collected from San Luis Drain or mosquitofish
<br />collected from a reference site LhaL were subsequently fortilled with seleno-oL-
<br />meLhionine, HamilLon el al. (1990) suggest LhaL the increased Loxicity ob-
<br />selved in the San Luis Drain dieL may be at least panially explained by Lhe
<br />presence ofoLher contaminants in the mosquitofish. Given that results are also
<br />reported based on selenium-forLilied lish from a reference siLe, we disregarded
<br />the dala based on fish collected in the San Luis Drain. In the first of two
<br />sLudies, swim-up larvae were exposed Lo selenium-fortified food for 90 days.
<br />The hllvae were reared in reconsLituLed water thaL simulated a 1:37 dilution
<br />li'OIn the San l.uis Drain (minus trace elements). Selenium concenLrations in
<br />lhe water were always below the deLection limits of 1.5 to 3.1 jlg/L. Dietary
<br />selenium concenLrations were 3,2, 5.3, 9.6, 18.2, and 35.4 mg/kg dw. In Lhe
<br />second study, fingerlings were exposed to the same dietary selenium concen-
<br />trations fur 120 days in brackish water.
<br />Upon termination of the first study aL 90 days, survival and growth in
<br />chinouk salmon fed a diet c011laining 18,2 mg/kg dw selenium were signifi-
<br />cantly (I' < (),(J5) reduced relative tu cOlltrol fish. No significant (P < 0.05)
<br />effects on slJlvival 01' growth were observed in fish fed a dietary selenium
<br />concentration of 9,Ii mg/kg dw, resulting in a clietary chronic value of
<br />13 mg/kg dw for these endpoints (Table 2). The reliability of Lhese results is
<br />I)ueslionahle, however, given Lhat control survival decreased from 99% on day
<br />liO to 67% on day 90, IL is unusual lor high controlmonaliLy in an early life-
<br />st<lge LesL to OCClll' this late in the test and is greater than the acceptable level
<br />of wlllrolmortality (30%) in salmon early life-sL.'lge Lests (ASTM, 1998). In
<br />addition, slllvival in the treaLmenL groups was greatly reduced beLween days 60
<br />and !JO despite whole body selenium residues noL substanLially changing over
<br />this time period. The high monality rate in thc control fish suggests thaL some
<br />of the mOllilliLy ohselved in the treatmenL groups may be due to factors other
<br />than selenium. Given this uncertainty, we feel this tesL should be repeated Lo
<br />confinn the results and recommend thaL the 90-day results of this tesL are not
<br />nsed in developing selenium,Lflresholds aL this time,
<br />
<br />Chinook Salmon (Oncorhynchus tshawytscha)
<br />
<br />Hamilton el al. (1986). Chinook salmon parr were feci a dieL conlaillilll{
<br />selenium-contaminated mosquitolish ({;mnbu5ia aJJilli!i) collected from San
<br />Luis Drain and Kesterson Wildlile Refuge ill California lor 6 weeks. Use of
<br />mosquitofish from these locations confounds interpretatiol\ due to the pres-
<br />ence of chemicals in addition to selenium ill, the tissue. For example,
<br />
<br />1111111, Ecol. Risk Assess, Vol. 5, No.6, 1999
<br />
<br />1201
<br />
|