Total and Methyl Mercury Concentrations in Seabird Feathers and Eggs

Total and Methyl Mercury Concentrations in Seabird Feathers and Eggs
Arch Environ Contam Toxicol (2009) 56:286–291
DOI 10.1007/s00244-008-9185-7
Total and Methyl Mercury Concentrations in Seabird Feathers
and Eggs
Alexander L. Bond Æ Antony W. Diamond
Received: 22 April 2008 / Accepted: 27 May 2008 / Published online: 26 August 2008
Ó Springer Science+Business Media, LLC 2008
Abstract Seabirds are used frequently as indicators of
mercury contamination in marine ecosystems, but few
studies have examined the forms of mercury found in
seabird tissues. Here we compare concentrations of total
and organic mercury in feathers (n = 5) of six sympatric
nesting seabirds and in egg components of Leach’s stormpetrels from Machias Seal Island, New Brunswick, Canada,
during the 2006 breeding season. Essentially all (82–133%)
mercury found in seabird feathers and egg components was
methyl mercury, with no interspecific differences in percentage methyl mercury. This pattern is consistent with the
hypothesis that feather-molt and egg production eliminate
toxic methyl mercury, while inorganic forms from
demethylation in the liver remain in internal tissues.
Additional studies across more species, and comparisons
with percentage methyl mercury in internal tissues, are
required to validate this theory.
Mercury is a pervasive anthropogenic environmental contaminant that is transported atmospherically around the
world (Nriagu and Pacyna 1988; Nriagu 1989; Mason and
Sheu 2002) and is particularly elevated in the Gulf of
Maine region (Evers and Clair 2005). Seabirds have been
A. L. Bond A. W. Diamond
Atlantic Cooperative Wildlife Ecology Research Network &
Department of Biology, University of New Brunswick, PO Box
45111, Fredericton, New Brunswick, Canada E3B 6E1
Present Address:
A. L. Bond (&)
Department of Biology, Memorial University of Newfoundland,
St. John’s, Newfoundland & Labrador, Canada A1B 3X9
used frequently as indicators of mercury contamination in
the marine environment (Burger 1993; Monteiro and
Furness 1995; Furness and Camphuysen 1997; Kim et al.
1998; Burger and Gochfeld 2004; Goodale et al. 2008), yet
few studies have examined the forms of mercury present in
seabird tissues (Thompson and Furness 1989a, b; Kim
et al. 1996; Burger and Gochfeld 2002). Methyl mercury
represents usually less than 1% of the mercury in marine
and freshwaters, yet since methyl mercury is the form that
bioaccumulates in the food web, top predators contain
methyl mercury levels of at least 95% of the total mercury
(Weiner et al. 2003). Methyl mercury affects the nervous,
circulatory, and endocrine systems in birds (Fimreite and
Karstad 1971; Spalding et al. 2000; UNEP 2002), and it is
considered to be mercury’s most toxic form (UNEP 2002;
Weiner et al. 2003). Effects of mercury contamination in
birds are species-specific, as are effect levels (Burger 1993;
Burger and Gochfeld 1997).
Some seabirds apparently tolerate higher total mercury
concentrations than other birds, but this is based on
assessments of total mercury in tissues such as feathers and
whole eggs (e.g., Monteiro et al. 1999; Burger and Gochfeld 2000), and determining effect levels in seabirds is
challenging, as many studies have focused on other families (e.g., quails and pheasants [Burger and Gochfeld
1997]). Although controlled dosing studies have been done
in some seabirds (Monteiro and Furness 2001), these did
not confirm effect levels, which are species-specific
(Burger 1993). Only two previous studies (Thompson and
Furness 1989b; Kim et al. 1996) have explicitly compared
total and methyl mercury levels in adult seabirds. Feathers
are easily collected, and representative of internal total
mercury burden (Agusa et al. 2005).
Here we present comparisons of total and methyl mercury from body contour feathers of six seabird species from
Arch Environ Contam Toxicol (2009) 56:286–291
three families—Leach’s storm-petrels Oceanodroma leucorhoa (Hydrobatidae), arctic Sterna paradisaea and
common terns S. hirundo (Laridae), and Atlantic puffins
Fratercula arctica, common murres Uria aalge, and razorbills Alca torda (Alcidae)—as well as from storm-petrel
eggs, from New Brunswick, Canada.
Materials and Methods
Sample Collection and Preparation
All samples were collected on Machias Seal Island, Bay of
Fundy, Canada (44°30’N, 67°06’W), during the 2006
breeding season. Breast feathers were collected from five
individuals of unknown sex from each species during
routine banding, or from freshly dead carcasses of birds
that had collided with structures on the island. Feathers
were stored at -18°C in individual sterile polythene bags
until analysis. Immediately prior to analysis, feathers were
washed for 1 min each in a 0.25 M NaOH solution and
three deionized water baths to remove external contamination. Mercury in feathers is unlikely affected by such
treatments (Appelquist et al. 1984).
Five fresh storm-petrel eggs were collected soon after
laying, and freshness was determined by flotation. Eggs
were wrapped individually in cellophane and frozen at 18°C for transportation. Albumin and yolk were separated
manually and weighed (±0.1 g) before drying in a Virtis
Benchtop freeze-dryer for 36–48 h.
Mercury Analysis
Total mercury concentration of the samples was determined by high-temperature combustion SP-3D analyzer
(Nippon Instruments Corp., Japan). One whole feather was
analyzed for both total and methyl mercury. After combustion at 850°C, mercury was converted catalytically to
elemental mercury. Following dual gold amalgamation the
quantity of mercury was measured by the cold vapor
atomic absorption (CVAA) method at a wavelength of
253.7 nm. The samples required no chemical treatment
prior to the analysis. Up to 10.0 mg of dried yolk and
albumin samples from eggs and between 10 and 50 mg of
feather samples were placed on a layer of an additive
(mixture of sodium carbonate and calcium hydroxide;
EMD Chemicals) in a ceramic boat as suggested by the
supplier. The sample was then covered with a layer of the
same additive. A layer of aluminum oxide (Al2O3; EM
Science) was placed over the sodium carbonate-calcium
hydroxide layer. Another layer of the latter covered this
aluminum oxide layer and the boat was transferred manually into the ceramic thermal decomposition chamber.
Reproducibility and accuracy of the method were assessed
every five samples using a standard of 5.0 ppb. Dogfish
muscle (DORM-2, NRC) was the standard reference
material, with certified values of 4.64 ± 0.26 mg/kg dry
weight. Recovery of DORM-2 was 98% (n = 4; values—
4.51, 4.60, 4.56, and 4.47 mg/kg).
Determination of methyl mercury concentration was
carried out by capillary gas chromatography coupled with
atomic fluorescence spectrometry (GC-AFS). About 0.05 g
of dried yolk and albumin samples and between 0.01 and
0.02 g of feather samples were placed in 20-ml scintillation
vials with Teflon PTFE caps. Then 2.0 ml of deionized
water and 2.0 ml of 6 M potassium hydroxide solution
were added and the samples were shaken for 4 h at
300 rpm. After that, 2.0 ml of 6 M hydrochloric acid was
added and the pH was checked to be \3.0 (high acidic
medium). Four millimeters of an acidic (5% H2SO4, v/v)
potassium bromide/1.0 M copper sulfate mixture was
added. To extract MeHg into the organic phase, 5.0 ml of
methylene chloride was added and the vials were shaken
overnight at 300 rpm. The next day the samples were
centrifuged for 10 min at 3500 rpm. A 2.0-ml aliquot of
the methylene chloride was transferred to 7-ml glass tubes,
and 1.0 ml of 0.01 M sodium thiosulfate added to each
sample. The samples were shaken for 20 min, mixed on a
vortex mixer, and centrifuged for 5 min at 3500 rpm. A
volume of 0.4 ml of the aqueous top layer (sodium thiosulfate) was placed in polyethylene microcentrifuge vials,
and 0.3 ml of an acidic potassium bromide/1.0 M copper
sulfate mixture (3:1) and 0.3 ml of dichloromethane were
added. Again, the vials were shaken for 15 min, mixed, and
centrifuged. Finally, the lower phase (dichloromethane
containing the extracted MeHg) was extracted carefully
and transferred through a small layer of anhydrous sodium
sulfate (packed in a pipette tip) to a 2-ml amber glass vial
with a 200-ll glass insert. Methyl and total mercury were
determined for the same feather samples simultaneously.
Two analytical blanks and a certified reference material
(DORM-2, NRC, 4.47 ± 0.32 mg kg-1 methyl mercury)
were run every five samples to evaluate reproducibility and
accuracy of analysis method. Methyl mercury recovery
from DORM-2 was 94% (n = 4; values—4.46, 4.15, 4.03,
and 4.21 mg/kg) and all feather samples were run in
duplicate, with mean standard deviations within samples of
\0.05 ppm.
Statistical Methods
Statistical tests were performed in SPSS v.11 using a significance level of 0.05 for all tests. We used the GamesHowell post hoc test (GH text [Games and Howell 1976])
to make pairwise comparisons among species following
analysis of variance (ANOVA) testing. We chose the GH
Arch Environ Contam Toxicol (2009) 56:286–291
test since it is relatively robust to comparisons involving
fewer than eight groups, with unequal variances, but with
n C 5 in each group, and because we were not necessarily
interested in the level of significance but desired greater
power than Dunnett’s T3 or C procedures (Day and Quinn
Outliers in linear regression models were detected using
Cook’s distance (Di), with points deemed outliers if
Di [ F0.50 with n and n – k degrees of freedom, where n is
the sample size and k is the number of coefficients in the
regression analysis (Cook 1977, 1979). Regression models
of methyl and total mercury were done to determine the
percentage methyl mercury as indicated by the slope.
Mercury concentrations are presented below as
mean ± SD parts per billion (ppb, or lg kg-1) fresh weight
for feathers and mean ± SD parts per billion dry weight
for egg components.
There was no difference among species in the mean proportion of methyl mercury in seabird feathers (ANOVA,
F5,24 = 0.92, p = 0.49). Methyl mercury levels were significantly correlated with total mercury concentrations
among species (r = 0.912, p \ 0.001). The slope of the
regression of total against methyl mercury, representing the
percentage methyl mercury (b), was 1.06 (Fig. 1a). Arctic
tern (r = 0.96, p = 0.01, b = 0.81) and Leach’s stormpetrels (r = 0.92, p = 0.03, b = 1.10; Fig. 1b, c) had a
significant relationship when species were examined individually. No outliers were detected (all Di’s \ 0.38).
The proportion of methyl mercury ranged from 82% in
razorbills to 133% in common murres, averaging
104.0% ± 41.2% across all species (Table 1; note that
values [100% arise from differences in mercury concentrations within the tissue). Since there is spatial
heterogeneity in the tissues, duplicate samples do vary, and
this is not due to machine variability. There were significant differences in the amount of methyl mercury among
species (ANOVA, F5, 24 = 5.41, p = 0.002) but post hoc
tests could not detect this difference (GH test, all
p’s [ 0.20). Variation in feather methyl mercury concentration between the two samples from each individual was
small, as measured by the coefficient of variation (range,
0.34–11.34%; overall mean, 3.58% ± 2.92%) and was
similar across species (ANOVA, F5, 24 = 0.26, p = 0.93).
There was no relationship between the total mercury
concentration and the proportion of methyl mercury
(r = -0.021, p = 0.91).
Fig. 1 Linear regression of feather mercury and methyl mercury
concentrations with 95% confidence intervals from (a) six seabird
species (n = 30), (b) arctic tern (n = 5), and (c) Leach’s storm-petrel
(n = 5) from Machias Seal Island in 2006
Albumin methyl mercury concentrations ranged from 3836
to 5178 ppb (mean, 4569 ± 579 ppb). Total mercury ranged from 3849 to 7994 ppm (mean, 5502 ± 1535 ppb), of
which 51.5%–107.2% (mean, 87.6% ± 21.7%) was methyl
mercury. Cook’s distance identified two outliers from the
Arch Environ Contam Toxicol (2009) 56:286–291
Table 1 Total and organic
mercury concentrations (parts
per billion, fresh weight) in
seabird breast feathers from
Machias Seal Island, NB, in
Note: Methyl mercury values
are means of duplicate samples
Mean total mercury,
ppb ± SD
Mean methyl mercury,
ppb ± SD
Mean methyl mercury,
% of total ± SD
Arctic tern
891 ± 507
791 ± 284
95 ± 15
Common murre
987 ± 361
1249 ± 349
133 ± 33
Common tern
1380 ± 991
1619 ± 1814
114 ± 65
1404 ± 559
1073 ± 383
82 ± 35
Atlantic puffin
1805 ± 668
1634 ± 664
100 ± 50
Leach’s storm-petrel
4855 ± 2791
5330 ± 3423
99 ± 40
albumin data set (Di = 1.38, 5.93; Table 2), but even when
these are removed, the regression model is not significant
(r = 0.85, F1,1 = 2.64, p = 0.35), and the sample size was
severely reduced, so they were retained in the data set.
When included, the regression model is not significant
(r = 0.11, F1,3 = 0.38, p = 0.86; Fig. 2).
Yolk methyl mercury concentrations were significantly
lower than those in albumin (paired-sample t-test,
t4 = 16.29, p \ 0.001), and concentrations between yolk
and albumin were not correlated (r = -0.15, p = 0.81), even
when outliers were removed (r = 0.41, p = 0.59). Methyl
mercury values from yolk ranged from 128 to 349 ppb
(mean, 227 ± 83 ppb), and total mercury ranged from 192 to
404 ppb (mean, 298 ± 81). Methyl mercury represented
67–86% (mean, 75% ± 7.3%) of the total mercury concentration (Table 2). Cook’s distance identified one outlier
(Di = 2.37; Table 2), but the regression model of percentage
methyl mercury and total mercury was significant whether
this was excluded (r = 0.99, F1,2 = 476.57, p = 0.002) or
included (r = 0.99, F1,3 = 149.98, p = 0.001; Fig. 2). The
proportion of methyl mercury was significantly higher in
albumin than yolk when outliers were removed (pairedsample t-test, t3 = 9.04, p = 0.003; Table 2).
Fig. 2 Linear regression of (a) albumin, (b) yolk, and (c) whole-egg
methyl and total mercury concentrations from Leach’s storm-petrels
on Machias Seal Island in 2006 (n = 5)
Table 2 Total and methyl mercury concentrations (parts per billion, dry weight) in albumin, yolk, and whole eggs of Leach’s storm-petrels on
Machias Seal Island, NB, in 2006
Sample No.
THg (ppb)
MeHg (ppb)
Mean ± SD
5501 ± 1535
4569 ± 579
88 ± 22
298 ± 81
227 ± 83
75 ± 7
1170 ± 297
1008 ± 291
89 ± 20
% MeHg
THg (ppb)
Whole egg
MeHg (ppb)
% MeHg
THg (ppb)
MeHg (ppb)
% MeHg
Outlier as determined by Cook’s distance (see text for explanation)
Whole Eggs
Whole-egg total mercury concentrations (1170 ± 297 ppb)
were 89% ± 20% methyl mercury (1008 ± 261 ppb). The
regression model of total against methyl mercury was not
significant (r = 0.41, F1,3 = 0.59, p = 0.50), although
much of the variation in percentage methyl mercury was
driven by one sample (56%; all others, 85–107%), and no
outliers were detected in whole-egg total or methyl mercury concentrations (all Di’s \ 0.52). When the one
consistent outlier for albumin and yolk is removed, the
model is almost significant (r = 0.94, F1,2 = 15.49,
p = 0.06; Fig. 2c).
Essentially all mercury in seabird feathers is organic
methyl mercury, and deviations from this are likely caused
by heterogeneity within the tissue. Leach’s storm-petrels
have higher mercury concentrations in feathers, eggs, and
blood than other seabirds in the Gulf of Maine (Bond 2007;
Goodale et al. 2008), and this could be related to either
phylogeny, lengthy molt cycles (Bond 2007), or high
mercury concentration in myctophid fish, their main prey
items (Martins et al. 2006). The levels observed in feathers
from this study are below those thought to cause sublethal
effects (Burger and Gochfeld 2000), but this level has not
been confirmed in laboratory studies of the species we
Feather molt and growth is the main mercury excretion
pathway for seabirds (Braune and Gaskin 1987; Monteiro
and Furness 2001), and it has been hypothesized that some
seabirds have the ability to demethylate organic mercury,
creating the less toxic inorganic form (Thompson and
Furness 1989a; Kim et al. 1996). Inorganic mercury is
immobile and remains in the liver, while the toxic form is
excreted during feather molt. Thus, mercury concentrations
in feathers could act as an indicator of the efficiency of
demethylation, assuming that individuals share a common
mercury intake. It would then follow that species that have
a poor demethylation ability would have higher mercury
concentrations in feathers as an alternative method for
eliminating methyl mercury. While examining mercury
concentrations in internal tissues would be ideal for
determining efficiency of demethylation (Monteiro and
Furness 2001), this type of destructive sampling may not be
appropriate in all cases.
When this was tested using the seven seabirds examined
by Kim et al. (1996)—royal Diomedea epomophora,
black-footed Phoebastria nigripes and Laysan albatrosses
P. immutabilis, white-chinned petrels Procellaria aequinoctialis, northern fulmars Fulmarus glacialis, brown
Arch Environ Contam Toxicol (2009) 56:286–291
boobies Sula leucogaster, herring gulls Larus argentatus,
and arctic terns—the relationship between mean liver
percentage methyl mercury and mean feather total mercury
was almost significant (F1,6 = 4.36, p = 0.08); however,
this study compared only seven species. Further investigation of the link between liver and feather mercury types
is needed before firm conclusions can be drawn.
Albumin total mercury concentrations were higher than
those in yolk, as mercury is bound to protein rather than
lipid and egg protein is concentrated in albumin (Magat
and Sell 1979). This is the first study to document the
differences in mercury types in egg components. Methyl
mercury made up the majority of the mercury concentration in albumin from Leach’s storm-petrels, and
significantly less so in yolk. Again, this is likely due to the
increased protein content of albumin over yolk egg fractions. Examinations of additional species are required. As
with feathers, it appears that the mercury forms are
excreted differentially according to the inherent chemical
properties of each compound, although we did not
explicitly test internal organs for body mercury burden.
The vast majority of mercury present in seabird feathers
and eggs is in the methyl form. This appears to be the result
of differential excretion of toxic forms of mercury, while
less toxic forms are sequestered in internal tissues. This
pattern requires further corroboration across more species
before generalizations regarding seabird mercury toxicology can be made.
Acknowledgments We thank G. Bouchard, T. Clarke, and M.-P.
Godin for assistance in the field, A. Patterson (Bold Coast Charter
Co., Cutler, Maine) for providing safe transportation to the island, E.
Yumvihoze for performing lab analyses, and T. Jardine and D. Lean
for many stimulating conversations on this topic and assistance
throughout the process. Comments from J. Lavers, R. Lavoie, D.
Lean, and two anonymous reviewers improved the manuscript. This is
ACWERN publication number UNB-71.
Agusa T, Matsumoto T, Ikemoto T, Anan Y, Kubota R, Yasunaga G,
Kunito T, Tanabe S, Ogi H, Shibata Y (2005) Body distribution
of trace elements in black-tailed gulls from Rishiri Island, Japan:
age-dependent accumulation and transfer to feathers and eggs.
Environ Toxicol Chem 24:2107–2120
Appelquist H, Asbirk S, Drabæk I (1984) Mercury monitoring:
mercury stability in bird feathers. Mar Pollut Bull 15:22–24
Bond AL (2007) Patterns of mercury burden in the seabird
community of Machias Seal Island, New Brunswick. M.Sc.,
University of New Brunswick
Braune BM, Gaskin DE (1987) A mercury budget for the Bonaparte’s
gull during autumn moult. Ornis Scand 18:244–250
Arch Environ Contam Toxicol (2009) 56:286–291
Burger J (1993) Metals in avian feathers: bioindicators of environmental pollution. Rev Environ Toxicol 5:203–311
Burger J, Gochfeld M (1997) Risk, mercury levels, and birds: relating
adverse laboratory effects to field biomonitoring. Environ Res
Burger J, Gochfeld M (2000) Metal levels in feathers of 12 species of
seabirds from Midway Atoll in the northern Pacific Ocean. Sci
Total Environ 257:37–52
Burger J, Gochfeld M (2002) Effects of chemicals and pollution on
seabirds. In: Schreiber EA, Burger J (eds) Biology of marine
birds. CRC Press, New York, pp 485–525
Burger J, Gochfeld M (2004) Metal levels in eggs of common terns
(Sterna hirundo) in New Jersey: temporal trends from 1971 to
2002. Environ Res 94:336–343
Cook RD (1977) Detection of influential observation in linear
regression. Technometrics 19:15–19
Cook RD (1979) Influential observations in linear regression. J Am
Stat Assoc 74:169–174
Day RW, Quinn GP (1989) Comparisons of treatments after an
analysis of variance in ecology. Ecol Monogr 59:433–463
Evers DC, Clair TA (2005) Mercury in northeastern North America: a
synthesis of existing databases. Ecotoxicology 14:7–14
Fimreite N, Karstad L (1971) Effect of dietary methylmercury on redtailed hawks. J Wildl Manage 35:293–300
Furness RW, Camphuysen K (1997) Seabirds as monitors of the
marine environment. ICES J Mar Sci 54:726–737
Games PA, Howell JF (1976) Pairwise multiple comparison procedures with unequal n’s and/or variances: a Monte Carlo study. J
Educ Stat 1:113–125
Goodale MW, Evers DC, Meirzykowski SE, Bond AL, Burgess NM,
Otorowski CI, Welch LJ, Hall CS, Ellis JC, Allen RB, Diamond
AW, Kress SW, Taylor R (2008) Marine foraging birds as
bioindicators of mercury in the Gulf of Maine. EcoHealth (in
Kim EY, Murakami T, Saeki K, Tatsukawa R (1996) Mercury levels
and its chemical form in tissues and organs of seabirds. Arch
Environ Contam Toxicol 30:259–266
Kim EY, Goto R, Tanabe S, Tanaka H, Tatsukawa R (1998)
Distribution of 14 elements in tissues and organs of oceanic
seabirds. Arch Environ Contam Toxicol 35:638–645
Magat W, Sell JL (1979) Distribution of mercury and selenium in egg
components and egg-white proteins. Proc Soc Exp Biol Med
Martins I, Costa V, Portiero FM, Santos RS (2006) Temporal and
spatial changes in mercury concentrations in the North Atlantic
as indicated by museum specimens of glacier lanternfish
Benthosema glaciale (Pisces: Myctophidae). Environ Toxicol
Mason RP, Sheu G-R (2002) Role of the ocean in the global mercury
cycle. Global Biogeochem Cycles 16:1–14
Monteiro LR, Furness RW (1995) Seabirds as monitors of mercury in
the marine environment. Water Air Soil Pollut 80:851–870
Monteiro LR, Furness RW (2001) Kinetics, dose-response, and
excretion of methylmercury in free-living adult Cory’s shearwaters. Environ Sci Technol 35:739–746
Nriagu JO (1989) A global assessment of natural sources of
atmospheric trace metals. Nature 338:47–49
Nriagu JO, Pacyna JM (1988) Quantitative assessment of worldwide
contamination of air, water and soils by trace metals. Nature
Spalding MG, Frederick PC, McGill HC, Bouton SN, Richey LJ,
Schumacher IM, Blackmore CGM, Harrison J (2000) Histologic,
nerologic and immunologic effects of methylmercury in captive
Great Egrets. J Wildl Dis 36:423–435
Thompson DR, Furness RW (1989a) Comparison of the levels of total
and organic mercury in seabird feathers. Mar Pollut Bull 20:577–
Thompson DR, Furness RW (1989b) The chemical form of mercury
stored in south Atlantic seabirds. Environ Pollut 60:305–317
UNEP (2002) Global mercury assessment. United Nations Environmental Programme—Chemicals, Geneva, Switzerland
Weiner JG, Krabbemhoft DP, Heinz GH, Scheuhammer AM (2003)
Ecotoxicology of mercury. In: Hoffman DJ, Rattner BA, Burton
GA Jr, Cairns J Jr (eds) Handbook of ecotoxicology. 2nd edn.
CRC Press, New York, pp 409–463
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF