Toxic Thresholds
29
The authors recommended that gravid ovarian tissue be used to monitor selenium residues and potential effects on bluegill populations,
since this tissue delivers the toxic "dose" to the developing fry. They
concluded that ovarian selenium levels in excess of 13 Ilg Se/g may
result in reproductive impairment. This number agrees very well with
the findings of Gillespie and Baumann (1986), who observed that feral
bluegill with ovarian selenium levels of 12 Ilg Se/g or greater failed to
produce viable offspring. It also agrees with Hermanutz et al. (1992),
who dosed outdoor experimental streams with sodium selenite, allowed natural cycling and bioaccumulation to occur in the food chain
and in adult bluegill, and then measured spawning success. Ovarian
selenium residues of 10 to 28 Ilg Se/g (skeletal muscle, 10-24 Ilg Se/g;
liver, 22-85 Ilg Se/g; whole body, 12-35 Ilg Se/g) were associated
with a 15 % decrease in survival of adult fish and almost complete
reproductive failure, with larvae exhibiting edema, lordosis (spinal
deformities), and hemorrhaging.
Lemly (1993) studied the teratogenic effects of selenium in natural
populations of centrarchids and other warm-water fish species. He determined the prevalence of abnormalities and associated tissue selenium concentrations in a contaminated lake and 2 reference lakes
over a period of 17 years. Whole-body selenium concentrations of 15
Ilg Se/g were associated with a 10-fold higher incidence of teratogenic
defects in centrarchid populations in the contaminated lake (tissue selenium was 1-3 Ilg Se/g in the reference lakes). The relationship between tissue selenium residues and the prevalence of malformations
approximated an exponential function over the range of 1 to 80 Ilg
Se/g and 0 to 70% deformities (R2 = 0.881, P< 0.01). Lemly concluded
that this relationship could be used to predict the impact of teratogenic
defects in warm-water fish populations suspected of having seleniumrelated reproductive failure (see Chapter 5).
Depending on the specific tissue (skeletal muscle, ovary, liver, whole
body, etc.), concentrations of selenium in fish from control test groups
or habitats with low ambient selenium levels usually range from about
1 to 8 Ilg Se/g (Baumann and May 1984; Lemly 1985a; Gillespie and
Baumann 1986; Hermanutz et al. 1992; Coyle et al. 1993). However,
tissue damage in major organs, reproductive impairment, and mortality begin to occur when levels reach 4 to 16 Ilg Se/g (Table 2.2). This
extremely narrow margin between "normal" and toxic levels in tissues, along with the propensity of selenium to bioaccumulate in aquatic
food chains, underscores the biological importance of even slight increases in environmental selenium.
I recommend that the following tissue residues be used as toxic effects thresholds for the overall health and reproductive vigor of freshwater and anadromous fish: whole body, 4 Ilg Se/g; skeletal muscle,
8 Ilg Se/g; liver, 12 Ilg Se/g; ovary and eggs, 10 Ilg Se/g. Laboratory
29
The authors recommended that gravid ovarian tissue be used to monitor selenium residues and potential effects on bluegill populations,
since this tissue delivers the toxic "dose" to the developing fry. They
concluded that ovarian selenium levels in excess of 13 Ilg Se/g may
result in reproductive impairment. This number agrees very well with
the findings of Gillespie and Baumann (1986), who observed that feral
bluegill with ovarian selenium levels of 12 Ilg Se/g or greater failed to
produce viable offspring. It also agrees with Hermanutz et al. (1992),
who dosed outdoor experimental streams with sodium selenite, allowed natural cycling and bioaccumulation to occur in the food chain
and in adult bluegill, and then measured spawning success. Ovarian
selenium residues of 10 to 28 Ilg Se/g (skeletal muscle, 10-24 Ilg Se/g;
liver, 22-85 Ilg Se/g; whole body, 12-35 Ilg Se/g) were associated
with a 15 % decrease in survival of adult fish and almost complete
reproductive failure, with larvae exhibiting edema, lordosis (spinal
deformities), and hemorrhaging.
Lemly (1993) studied the teratogenic effects of selenium in natural
populations of centrarchids and other warm-water fish species. He determined the prevalence of abnormalities and associated tissue selenium concentrations in a contaminated lake and 2 reference lakes
over a period of 17 years. Whole-body selenium concentrations of 15
Ilg Se/g were associated with a 10-fold higher incidence of teratogenic
defects in centrarchid populations in the contaminated lake (tissue selenium was 1-3 Ilg Se/g in the reference lakes). The relationship between tissue selenium residues and the prevalence of malformations
approximated an exponential function over the range of 1 to 80 Ilg
Se/g and 0 to 70% deformities (R2 = 0.881, P< 0.01). Lemly concluded
that this relationship could be used to predict the impact of teratogenic
defects in warm-water fish populations suspected of having seleniumrelated reproductive failure (see Chapter 5).
Depending on the specific tissue (skeletal muscle, ovary, liver, whole
body, etc.), concentrations of selenium in fish from control test groups
or habitats with low ambient selenium levels usually range from about
1 to 8 Ilg Se/g (Baumann and May 1984; Lemly 1985a; Gillespie and
Baumann 1986; Hermanutz et al. 1992; Coyle et al. 1993). However,
tissue damage in major organs, reproductive impairment, and mortality begin to occur when levels reach 4 to 16 Ilg Se/g (Table 2.2). This
extremely narrow margin between "normal" and toxic levels in tissues, along with the propensity of selenium to bioaccumulate in aquatic
food chains, underscores the biological importance of even slight increases in environmental selenium.
I recommend that the following tissue residues be used as toxic effects thresholds for the overall health and reproductive vigor of freshwater and anadromous fish: whole body, 4 Ilg Se/g; skeletal muscle,
8 Ilg Se/g; liver, 12 Ilg Se/g; ovary and eggs, 10 Ilg Se/g. Laboratory