|
<br /> -;
<br />, 1
<br /> . ,
<br /> ,
<br /> i
<br /> ,
<br /> ,;
<br /> 'I
<br /> i:
<br /> :1
<br /> :,
<br /> '.
<br /> 'I
<br /> ,!
<br /> I
<br /> f
<br /> .'I
<br /> If
<br /> \
<br /> I
<br /> r
<br /> I:
<br /> tl
<br /> II
<br /> r
<br /> I
<br /> ~
<br /> '.
<br /> .,
<br /> :i
<br />
<br />DeForest, 81'ix, and Adams
<br />
<br />survival, growlh, or reproduclion, which is curious, because lhc ovary concen-
<br />lraUons presenled in Coyle el ai. (1993) are eXIJ1'csscd on a dry wciglll basis
<br />(Coyle, pcrsonal communicalion).
<br />Based on lhe whole body selenium conce11lralions in adulls on lhe day
<br />spawning was inilialed, a chronic value of 10.(i mg/kg dw for whole body lissue
<br />is eSlimaled fOl'lan'al sUlVival over 39 days based on an LOEC of 16 mg/kg dw
<br />and an NOEC of 7 mg/kg dw (Table 3). Lemly ()!193a) Cilcd lhe LOEC of
<br />16 mg/kg dw as lhe cffecllevcl for reprOllul:livc failurc (Table 3). As in lheir
<br />inlerprelalion of ovary COnCelllralions in lhis slUdy, il appcars lhal Jalvinen
<br />and Ankley (1999) may have misilllcrpreled lhe whole body concclllralions
<br />reponed in lhis sludy as lhey associ ale a whole uody cOllccnlralion of 19,0 mg/
<br />kg ww wilh no cffecls on sUlVival, growlh, or reproduclion.
<br />Skorupa el ai. (1996) reviewed lhis sludy and noted lhal, duc to the poor
<br />lalVal sUlVival of the conlrols (20 lo 25%) OIL day 30, lhis lest ouly had lhe
<br />stalislicaI power to delecl "cataslrophic reproduclive impairmenl". Coyle et ai,
<br />(1993) suggcst thal high Ialvalmorlalily ouselved ueginning at day 7 was likely
<br />related lo slalVation rcsulting from the ullsucccssfullransition betwcen cndog-
<br />enous and exogenous fecding (yolk sac ausorplion occurrcd UClwcen days 5
<br />and 6). Given that lalval sUlVival was high in controls (>90%) during cndog-
<br />enous fecding OIL day 5, and lalval sUlvival fi'OIn females Wilh an ovary selcnium
<br />concenlration of 35 mg/kg dw was eXlremely low (<10%), a c1eal. selenium-
<br />rclated elTect can uc obselVed. Ahhough the high conlrolmOl'lality aftcr swim-
<br />up imparls some uncerlainly to lhe results, use of lalVal sUlvival dala on day
<br />5 docs not requirc that calaslrophic reproductive impainncnl must occur lo
<br />obselVe significanl rcproduclive cHccts I'clalivc lo conlrols.
<br />Lemiy (199Jb). Based on ccntrarchid data for juvcniles and adults collccted
<br />from Belews Lake, NOrlh Carolina, l.emly (1993u) uscd a polynomial regrcs-
<br />sion lo fit a cuuic model lo the relalionship uetween whole uody selenium
<br />concentrations and lhe percenlage of dcfoHned juvcnile and adult fish (Fig-
<br />ure 1). Based on his model, a whole body selenium conccntralion of 15 mg/
<br />kg dw would lranslale lo deformities in approximalely 4 % of lhc fish. He slales
<br />lhallhe inflection poilll for lhis model occurred UClwecll 40 lo 50 mg/kg dw,
<br />corresponding lo it frequcncy of deformiLics of auoUl20 lo 30%. No informa-
<br />lion was provided as to how the CUlve was constructed, but our scatter plOl of
<br />lhc raw data reporled in his paper closely resemules lhe plOl in Figure 9 of his
<br />paper. Using polynomial regression, we fit a cuuic model (Y;= /Jo + lJ,x + LJ2".2
<br />+ BJX1) lo the raw dala using the slalistical COlllpuler program 5J>SS (SJ>55,
<br />1998) (Figure 1). Our CUlves are similar up to an abuormality level of approxi-
<br />mately 40%, at which p,oinllhe Lcmly (I 993u) Clllve becomes much steeper.
<br />The basis for lhis discrepancy in the models is unknown, but probably incon-
<br />scquential givenlhal the approximale ECIO, ECw, and ECso values froml.emly's
<br />(l993b) model were approximalcly 31,47, and 72 mg/kg dw, COlli pared Wilh
<br />29, 47, and 73 mg/kg dw Ii'olll our cuuic model. IL is only al concelllralions
<br />abovc the ECso whcre our lwo models uegin to suuslanlially diverge. This
<br />divergence would not inllucncc the idcnlilicalion of a whole body Ihreshold
<br />for lhis endpoilll since the lhreshold wiII ue IIIl1ch lower lhan lhc 50% cllcCl.
<br />level.
<br />
<br />Debale/ Commentary
<br />
<br /> .
<br />u
<br />:c r- ....
<br />ClO\
<br />u.... 'o:J' r-
<br />-
<br />'be r- 0\ co
<br />j.... 'o:J' r-
<br />-
<br />:c
<br />e~ r- ....
<br />'o:J' r-
<br />ll.
<br /> II II II
<br /> .. l'l 51
<br /> U U U
<br /> ~ ~ ~
<br />
<br />co
<br />co
<br />...
<br />
<br />co
<br />0\
<br />
<br />co
<br />00
<br />
<br />co
<br />r-
<br />
<br />co
<br />\C
<br />
<br />co
<br />II'l
<br />
<br />
<br />IBWJOUqV %
<br />
<br />HIIIll. Ecol. Risk Assess, Vol. 5, No, 6, ]999
<br />
<br />-
<br />ii:
<br />:Q
<br />....
<br />8:
<br />e.
<br />- >. u
<br />:c j B :c
<br />e tl CI
<br />ll. o-l u
<br />
<br />I I I I
<br />
<br />co
<br />'o:J'
<br />
<br />co co
<br />.... ....
<br />
<br />co
<br />...
<br />
<br />co
<br />\C
<br />...
<br />~
<br />...
<br />co
<br />'o:J'
<br />...
<br />co
<br />....
<br />...
<br />co
<br />t'I
<br />...
<br />co
<br />...
<br />...
<br />g
<br /> i'
<br /> ."
<br />a: ~
<br /> ~
<br />~ ..
<br />II)
<br /> >.
<br /> '8
<br /> co =
<br /> r- ..
<br /> '0
<br /> ~ ~
<br /> ~
<br /> co
<br /> .,.
<br /> co
<br /> ....
<br /> co
<br /> ....
<br /> co
<br /> ...
<br />
<br />co
<br />
<br />co
<br />
<br />':D
<br />C()
<br />Ol
<br />Ol
<br />.....
<br />?.....
<br />E
<br />j
<br />8
<br />o
<br />~
<br />13
<br />rl
<br />~
<br />
<br />-d
<br />:a
<br />~
<br />
<br />c:
<br />u
<br />u
<br />"tl
<br />u
<br />ti
<br />u
<br />:::l
<br />o
<br />y
<br />"Cl
<br />1i
<br />lC
<br />,5
<br />
<br />VI
<br />.~
<br />t:
<br />u
<br />~
<br />~
<br />~
<br />"tl
<br />~
<br />8
<br />:l
<br />'2
<br />u
<br />]
<br />oS
<br />o
<br />.D
<br />u
<br />'0
<br />~
<br />
<br />...;
<br />
<br />~
<br />
<br />1209
<br />
|