Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
Food Additives and Contaminants, September 2007; 24(9): 969–975
Total and organic mercury concentrations in the white muscles of
swordfish (Xiphias gladius) from the Indian and Atlantic oceans
M.-H. CHEN1, C.-Y. CHEN2, S.-K. CHANG3, & S.-W. HUANG1
1
Department of Marine Biotechnology and Resources, National Sun Yat-sen University, 70 Lianhai Road, Gushan,
Kaohsiung 804, Taiwan, 2Department of Marine Environmental Engineering, National Kaohsiung Marine University,
142 Haichuan Road, Nantzn, Kaohsiung 811, Taiwan, and 3Deepseas Fisheries Research and Development Center,
Fisheries Agency, Council of Agriculture, 2 Chaojou Street, Taipei 100, Taiwan
(Received 29 June 2006; revised 26 January 2007; accepted 26 February 2007)
Abstract
A total of 226 swordfish samples collected from Taiwanese fishing vessels in the Indian and Atlantic oceans were examined
for total mercury (THg) and organic Hg (OHg). Analysis of 56 pooled white muscle samples showed that THg and OHg
concentrations ranged from 0.056 to 3.97 (1.3 0.97) and from 0.043 to 3.92 (1.01 0.82) mg g1 flesh mass, respectively.
These values were similar to those from various previous studies during the past three decades. THg and OHg were
significantly linearly correlated with fork length (FL, cm) of the fish from Indian and Atlantic oceans; however, there was no
significant OHg%–FL relationship. OHg and THg also were significantly correlated. Fishes with FL 140 cm met the
methyl Hg (meHg) regulatory standard set by the European Commission Decision (meHg 1.0); and fish with
FL 211 cm met the Taiwanese Food and Hygiene Standard (meHg 2.0). Weekly swordfish consumption rates and
amounts are recommended accordingly.
Keywords: Heavy metals, regression line, marine fish, regulatory standards, food safety
Introduction
As a top predator in marine pelagic ecosystems,
swordfish (Xiphias gladius) bioaccumulate high levels
of mercury (Hg) that from time to time exceed
1 mg g1 in their muscle (Sun & Chang 1972,
Freeman et al. 1978, Monteiro & Lopes 1990).
They have been reported as having higher Hg
contents than other pelagic predatory fish such as
shark, tuna and marlin (Dabeka et al. 2004, Forsyth
et al. 2004). The high levels of Hg in these
piscivorous fish become even higher as fish become
larger (Freeman et al. 1978, Monteiro & Lopes
1990). This is thought to be a result of natural Hg
biogeochemical processes from globally natural and
anthropogenic emissions of Hg through biomagnification in the marine food chain (Riisgård & Hansen
1990, Bargagli et al. 1998). Thus, any elevated Hg
level in the species may be indicative of the health of
oceanic ecosystems on temporal and spatial scales.
Correspondence: M.-H. Chen. E-mail: [email protected]
ISSN 0265–203X print/ISSN 1464–5122 online ß 2007 Taylor & Francis
DOI: 10.1080/02652030701305470
It is well known that fish consumption is the major
source for human exposure to Hg (MacIntosh et al.
1996, Nakagawa et al. 1997, Ysart et al. 2000).
Hg in the edible portions of marine creatures has
attracted considerable attention from scientists
around the world (e.g. Andersen & Depledge 1997,
Adams & Onorato 2005). The World Health
Organization (WHO)/The Joint FAO/WHO Expert
Committee on Food Additives (JECFA), and many
countries (e.g. the USA and Canada) have established regulatory guidelines and consumer advisory
panels for safe levels of Hg in fish and fish products
(e.g. Canadian Food Inspection Agency 2002, WHO
2003, Center of Food Safety and Applied Nutrition
2004). For the safe consumption of tuna, shark, and
swordfish, the maximum levels of Hg in the form of
methyl Hg (meHg) in the meat was set at less than
1.0 mg g1 flesh mass by the European Commission
(2005); however, in light of the small amounts
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
970
M.-H. Chen et al.
consumed, the standard was set at 2.0 mg g1 flesh
mass by the Department of Health in Taiwan
(Department of Health 1992). Since more than
50% of total Hg in the muscle of fishes, except
planktivores, is meHg (Cappon & Smith 1982,
Barghigiani et al. 1989, Watras & Bloom 1992,
Holsbeek et al. 1997, Kannan et al. 1998), organic
Hg concentrations in highly migratory sea fishes,
such as swordfish, are very important when estimating the risk of meHg to the general public through
dietary exposure to Hg.
Until the present study, Hg data for swordfish
were mostly reported from the Atlantic Ocean
(Beckett & Freeman 1974, Freeman et al. 1978,
Monteiro & Lopes 1990), with few data from
elsewhere, such as fish from the Indian and Pacific
oceans (Sun & Chang 1972), from the
Mediterranean Sea (Storelli & Macrotrigiano
2001), from the Fiji Islands, Pacific Ocean (Kumar
et al. 2004), and from the south-western Indian
Ocean (Kojadinovic et al. 2006). As one of the
worlds most important exploratory swordfish-fishing
fisheries (Chang & Hsu 2002, Chang 2004),
Taiwanese fishing vessels that are deployed all over
the world provide excellent opportunities to collect
samples for global Hg biomonitoring. This also
provides information concerning Hg concentrations
in safe quality fish products for international trade.
Hence, in 2004, for the first time, fishery observers
collected swordfish meat samples to investigate total
Hg and organic Hg concentrations in swordfish and
to compare samples between two oceans.
Materials and methods
Fifty-five samples from the Indian Ocean and 171
samples from the Atlantic Ocean were collected
from June to November 2004 and from May
to December 2004, respectively, by Taiwanese
scientific observers aboard fishing vessels. For each
sample the species was carefully identified, the fork
length (FL) was measured from the tip of the lower
jaw to the fork of the caudal fin, and the code of the
fishing area and the names of the fishing vessel were
recorded.
A punching drill was inserted into the caudal
peduncle of the swordfish muscles by fisheries
inspectors to obtain a small amount of tissue such
that the value of the fish product was not reduced.
The samples with sufficient tissue (40.3 g) were
analysed individually; however, for tissue 50.3 g,
tissue from similarly sized fish from the same ocean
were combined. Thus, two to five and three to nine
individual samples were combined for some Indian
and Atlantic samples, respectively. In total, 21
samples from the Indian (a total of 55 fish) and 35
samples from the Atlantic Ocean (a total of 171 fish)
were analysed. The amount of these samples was
only enough for the THg and OHg analyses, but not
for fat analysis.
Analysis of total Hg concentrations (THg, in
mg g1 flesh mass) and organic Hg concentrations
(OHg, in mg g1 flesh mass) followed the methods of
Chen & Chou (2000) and Chen et al. (2002),
respectively. Briefly, for the THg analysis, 0.2–0.5 g
of homogenized muscle tissue was weighed in a
75-ml graduated test tube. A total of 1 ml of
concentrated nitric acid (HNO3), 4 ml of concentrated sulfuric acid (H2SO4), and 15 ml of 5%
potassium permanganate (KMnO4) were added to
wet-digest the tissue, resulting in a final volume of
25 ml. With 5% of tin (II) chloride dihydrate (SnCl4)
as the reductant, the Hg concentration was measured using a cold vapour atomic absorption spectrophotometer (CVAAS, Hitachi Z-8200 and HFS-2).
For analysis of OHg, acetone was used first to
remove lipid cover the surface of tissue. Then, 5 ml
of 3 M potassium bromide (KBr) and 10 ml of 0.1 M
copper sulfate (CuSO4) as the extracting agent were
added to 0.3–0.5 g of homogenized muscle in a
40-ml conical graduated centrifuge tube. This
extractant was extracted again with toluene, and
the upper organic phase taken out and further
extracted back to 1 ml of 0.005 M sodium persulfate
(Na2S3O3). Finally, this 1-ml Na2S3O3 extractant
was transferred into a 75-ml test tube for the Hg
digestion procedure by described above, following
the CVAAS method.
Regent blanks were inserted as every 20th sample
to detect any alien contaminants. In addition, the
duplicates of the certified reference materials —
DORM-2 (dogfish muscle) and DOLT-2 (dogfish
liver), purchased from the National Research
Council of Canada — were analysed simultaneously
in each digesting process. All chemical reagents used
in this study were GR grade from Merck Co,
Germany. The instrumental detection limits of
THg and OHg were calculated based on 3 standard
deviations (SDs) of the blank after a series of
analyses. They were 1.0 and 0.5 ng ml1, respectively. For QA and QC, the analytical results of
four replicates of each certified reference material
are presented as mean standard deviation for
DORM-2 (THg ¼ 4.42 0.12 and OHg ¼ 3.66 0.26 mg g1 dry mass), and for DOLT-2 (THg ¼
2.20 0.12 and OHg ¼ 0.74 0.06 mg g1 dry mass).
Compared with the certified values of DORM-2
(THg ¼ 4.64 0.26,
OHg ¼4.47 0.32)
and
DOLT-2
(THg ¼ 2.14 0.28,
OHg ¼ 0.693 0.053), the mean values were all within the 95%
confidence interval of the certified values.
Statistical analyses were performed using SAS
software, and consisted of a Student’s t-test to detect
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
Hg of swordfish in the Indian and Atlantic Oceans
971
Table I. Sample numbers and fork lengths of whole and pooled samples of swordfish, Xiphias gladius, used in the study and the results of
total Hg (THg, mg g1 flesh mass) and organic Hg (OHg, mg g1 flesh mass) concentrations, as well as the percentage of organic Hg (OHg)
in the white muscle of the fish. The last column shows the results from Student’s t-tests on the differences between the Indian and Atlantic
oceans.
Indian Ocean
Atlantic Ocean
t-test
Whole
Sample number
Fork length (cm)
55
156 37 (73–232)
171
132 39 (59–255)
p50.0001
Pooled
Sample number
Fork length (cm)
THg
OHg
OHg (%)
21
156 38 (98–232)
1.47 0.63 (0.26–2.54)
1.10 0.43 (0.14–1.93)
76.4 14.1 (44.8–95.4)
35
132 49 (63–255)
1.20 1.12 (0.06–3.97)
0.96 0.99 (0.04–3.92)
80.3 15.4 (53.9–100)
n.s.
n.s.
n.s.
n.s.
the differences in Hg concentrations between the
two oceans (p50.05) and linear regressions to
determine correlations between THg and OHg to
FL (Statistical Analysis Software Institute 1998).
The percentage of OHg (OHg%) was calculated
from OHg divided by THg and multiplied by 100.
Results and discussion
The average fork length (FL) of swordfish from the
Indian Ocean was larger than that from the Atlantic
Ocean, which was statistically significant for whole
samples (p50.0001; Table I), but not for the pooled
samples, probably because of the smaller sample
sizes (p40.05; Table I).
No significant differences in the THg and OHg
concentrations or in OHg% were found between the
Indian and the Atlantic oceans (p40.05; Table I);
however, the Hg measurements from Indian Ocean
fish were slightly higher than those from the Atlantic
Ocean, which probably is due to the larger size of the
fish collected from the Indian Ocean (Sun & Chang
1972, Beckett & Freeman 1974, Monteiro & Lopes
1990). The total THg and OHg concentrations in
56 pooled samples of swordfish ranged from 0.06 to
3.97 and 0.04 to 3.92 mg g1 flesh mass, respectively.
Concentrations averaged (mean and standard
deviation) 1.30 0.97 THg mg g1 flesh mass and
1.01 0.82 OHg mg g1 flesh mass, and OHg% was
78.8% 14.9%.
The results are very consistent with earlier
measurements of mercury in swordfish (Figure 1a).
Importantly, the mercury contents of swordfish have
been very stable over three decades (Table II).
Variations among these different reports may be
ascribed to differences in the analyses with respect to
the laboratories and methods used and the size and
gender of the fish, as well as differences in sampling
locations. These results suggest that on a global scale
with geographical variation, the Hg concentrations
of swordfish in the two oceans have remained fairly
constant.
THg and OHg concentrations increased with size,
but OHg% remained fairly constant. THg and OHg
were linearly correlated with FL (Figure 1), but
OHg% was independent of FL (p40.05; Figure 1).
The regression equations were as follows:
THg ¼ 0.0170
FL–1.1158
(R2 ¼ 0.6600,
p50.0001), and OHg ¼ 0.0141 FL – 0.9812 (R2 ¼
0.6372, p50.0001). This pattern was similar to the
Type III Hg accumulation pattern described by
Holsbeek et al. (1997). This pattern recently has
been found for various Bangladesh’s freshwater and
seawater fishes, such as one of bottom dwelling
freshwater fish, Puntinus sarana (Holsbeek et al.
1997), sardine, Sardinella aurita (Joiris et al. 1999),
and a marine fish, Stromateus cinereus (Joiris et al.
2000). With increasing length and age, fish with this
type of Hg accumulation pattern simultaneously
increase THg and OHg, and keep the OHg% within
a constant range, showing that the fish have little
ability to demethylate Hg in their body (Joiris et al.
2000).
Concentrations of OHg and THg in the white
muscle of swordfish were linearly correlated according to the equation: OHg ¼ 0.8100 THg – 0.0296
(R ¼ 0.9041, p50.0001; Figure 2). This is the first
time that such a relationship in swordfish has been
established. Furthermore, the wide range of OHg%
(45–100%) was not related to the size of the
swordfish (Figure 1). Such wide variations were
also found by Forsyth et al. (2004), who reported
43–76% methylmercury (meHg) in swordfish collected from both the Pacific and Atlantic oceans.
A swordfish fillet purchased at a supermarket in
New York contained 81% meHg (Cappon & Smith
1982), and frozen swordfish steaks sold in the USA
contained up to 100% organic Hg (Kamps et al.
1972). Such wide variations in OHg% in the white
muscle of swordfish may be related to their
physiological condition, age, sampling location, and
M.-H. Chen et al.
THg (mg kg−1 flesh mass)
4
Atlantic Indian
Sun & Chang,1972
Freeman et al., 1978
Monteiro & Lopes, 1990
Kojadinovic et al., 2006
3
2
y = 0.0170x −1.1158
R2 = 0.6600
(n = 56, p<0.0001)
1
0
0
50
100
150
200
250
300
OHg (mg kg−1 flesh mass)
4
3
y = 0.0141x -0.9812
R2 = 0.6372
(n = 54, p<0.0001)
2
1
0
0
50
100
150
200
250
300
0
50
100
150
200
250
300
120
90
OHg %
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
972
60
30
0
FL (cm)
Figure 1. Linear relationships between total mercury concentration (THg, mg g–1 flesh mass), organic mercury concentration (OHg, mg g1
flesh mass), and the percentage of organic mercury (OHg%) to fork length (FL, cm) of swordfish, Xiphias gladius, in the Indian and Atlantic
þ Kojadinovic et al.
oceans. Published data from , Sun & Chang (1972); þ, Freeman et al. (1978); , Monteiro & Lopes (1990); and ,
(2006) are shown for comparison.
dietary sources, which may affect their ability to
demethylate Hg.
The maximum permissible recommended level
of meHg in oceanic migratory fishes set by
the Commission of the European Communities
is 1.0 mg g1 flesh weight (European Commission
2005). By use of the OHg-FL regression to backcalculate the size of fish containing acceptable
OHg levels, it was determined that fish
shorter than 140 cm FL would meet the standards
of meHg 1.0. In Taiwan, the maximum
permissible recommended level of meHg in
oceanic migratory fishes has been set at 2.0 mg g1
flesh weight (Department of Health 1992);
back calculation by the regression showed that fish
shorter than 211 cm FL would meet the accepted
standard.
The Joint FAO/WHO Expert Committee on Food
Additives (JECFA) set the provisional tolerable
weekly intake (PTWI) level for meHg at
1.6 mg kg1 body weight week1 (WHO 2003). If
one assumes that the OHg in this study was all
meHg and use the mean OHg concentration of
1.0 mg g1 in the white muscle of swordfish, the
allowable weekly intake of swordfish would be 104
and 88 g for the average adult male (65 kg) and
female (55 kg), respectively.
In 2004, Health Canada lowered the PTWI level
of meHg for children and women of child-bearing
age (18–34 years old) to 1.4 mg kg1 body weight
week1, but kept the PTWI level at 3.3 mg kg1 body
weight week1 for the rest of the population (Health
Canada Mercury Issues Task Group 2004). Thus,
the PTWI level of OHg for children (20 kg) would be
0.82
1.05 0.52
1.15
1.32 0.99
M: 1.30 0.17, F: 0.93 0.07
0.49 0.26
1.82
1.81 0.82
0.38 0.26
1.24 0.83
1.30 0.97
0.08–5.20
0.48–2.30
0.05–4.90
0.03–4.40
0.06–4.91, 0.06–4.31
0.15–1.05
0.40–3.85
0.99–2.81
–
–
0.06–3.97
THg (mean SD; range)
73–189
–
74–247
77–224
83–178, 80–241
5 100 kg
–
–
75–191
90–187
63–255
FL (cm)
Location
Indian, South Pacific, and Atlantic oceans
Markets, USA and Sweden
Western Atlantic Ocean
North-western Atlantic Ocean
Azores, mid-Atlantic Ocean
Ionian Sea, Mediterranean
Markets, Halifax, Vancouver and Toronto, Canada
Fiji Islands, Pacific Ocean
Mozambique Channel, Indian Ocean
Reunion Island, Indian Ocean
Indian and Atlantic oceans
M, males; F, females; FL ¼ fork length (cm); other values are body weight in kg;- indicate data are unavailable.
1970s
1970s
1970s
1970s
1980s
1990s
2000s
2000s
2000s
2000s
2000s
Decade
Sun & Chang (1972)
Kamps et al. (1972)
Beckett & Freeman (1974)
Freeman et al. (1978)
Monteiro & Lopes (1990)
Storelli & Macrotrigiano (2001)
Dabeka et al. (2004), Forsyth et al. (2004)
Kumar et al. (2004)
Kojadinovic et al. (2006)
Kojadinovic et al. (2006)
This study
References
Table II. Means and ranges of total mercury (THg) concentrations (mg g1 flesh mass) in the white muscle of swordfish collected between the early 1970s and the early 2000s.
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
Hg of swordfish in the Indian and Atlantic Oceans
973
M.-H. Chen et al.
4
Atlantic
OHg (mg kg−1 flesh mass)
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
974
Indian
3
y = 0.8100x – 0.0296
R2 = 0.9041
(n = 54, p<0.0001)
2
1
0
0
1
2
THg (mg
kg−1
3
4
flesh mass)
Figure 2. The Linear regression line between organic mercury concentration (OHg, mg g1 flesh mass) and total mercury concentration
(THg, mg g1 flesh mass) in swordfish, Xiphias gladius, in the Indian and Atlantic oceans.
28 mg week1, and 70 mg week1 for children and a
woman of child-bearing age (50 kg), equivalent to 28
and 70 g of swordfish meat, respectively. For adult
males (65 kg) and females (55 kg), the PTWI level
can be increased to 215 and 182 mg week1 per
person, equivalent to 215 and 182 g of swordfish
meat week1, which doubles the amount to the
recommendation of the JECFA.
According to the 1993–96 Nutrition and Health
Survey in Taiwan (NAHSIT), the average daily sea
fish consumption amount for the general public in
Taiwan was approximately 34.4 and 21.1 g for adult
males and females, respectively (Wu et al. 1999),
equivalent to 241 and 148 g per week for each male
and female, respectively. Therefore, if swordfish
were to become the main substitute of the daily sea
fish diet for consumers in Taiwan, the health
regulation of the PTWI for meHg would likely be
commonly exceeded, except general females who did
not in the child-bearing age and took the Health
Canada regulation. Therefore, owing to human
health concerns and the necessity of reducing the
risk of for Hg exposure, it is suggested that the
average adult in Taiwan consume no more than
three servings (35 g per serving) week1, or 100 g
week1 of swordfish. For pregnant women and
women of child-bearing age (18–34 years old), it is
recommended that they eat only two servings per
week, or to less than 70 g week1. It is also
recommended that children consume no more than
one serving of less than 28 g week1, or three
servings of 35 g month1.
Conclusions
The swordfish caught by Taiwanese fishing vessels
in the Indian and Atlantic oceans contained
means (ranges) of THg and OHg concentrations
(mg g1 flesh mass) of 1.3 (0.056–3.97) and 1.0
(0.043–3.92), respectively, which are similar to those
reported over the past three decades. OHg was
approximately 81% of THg and remained fairly
constant regardless of fish size, showing a Type III
Hg bioaccumulation pattern (Holsbeek et al. 1997).
Swordfish with fork lengths of less than 140 and
211 cm would not exceed the allowable maximum
Hg levels set by both the Commission Regulation of
European Communities (meHg51.0 mg g1 flesh
mass) and the Taiwanese Standards of Food and
Hygiene (meHg52.0 mg g1 flesh mass), respectively. Based on the PTWI set by the Joint FAO/
WHO Expert Committee on Food Additives
(JECFA) and Health Canada, no more than three
servings per week (35 g per serving) or 100 g week1
of swordfish should be consumed by the general
public, but only two-thirds and one-quarter of that
amount are recommended for pregnant women and
for women of child-bearing age (18–34 years old)
and children, respectively.
Acknowledgements
Research funding came from various sources,
including the National Science Council of Taiwan,
Grant Nos NSC93-2313B-110-006 and NSC932621-Z-110-003, and the Ministry of Education,
Taiwan, Aim for the Top University Plan to
Kuroshio Research Group Asia-Pacific Ocean
Research Center in NSYSU, Kaohsiung. Sincere
thanks to the two anoymous reviewers for their
constructive comments.
References
Adams DH, Onorato GV. 2005. Mercury concentrations in red
drum (Sciaenops ocellatus) from estuarine and offshore waters of
Florida. Marine Pollution Bulletin 50:291–300.
Downloaded By: [2007 National Sun Yat Sen University] At: 11:23 5 November 2007
Hg of swordfish in the Indian and Atlantic Oceans
Andersen TL, Depledge MH. 1997. A survey of total mercury
and methylmercury in edible fish and invertebrates from
Azorean waters. Marine Environmental Research 44:331–350.
Bargagli R, Monaci F, Sanchez-Hernandez JC, Cateni D. 1998.
Biomagnification of mercury in an Antarctic marine coastal
food web. Marine Ecological Progress Series 169:65–76.
Barghigiani C, Pellegrini D, Carpene E. 1989. Mercury binding
proteins in liver and muscle of flat fish from the northern
Tyrrhenian Sea. Comparative Biochemistry and Physiology
94C:309–312.
Beckett JS, Freeman HC. 1974. Mercury in swordfish and
other pelagic species from the western Atlantic Ocean.
US Department of Commerce, NOAA Technical Report No.
NMFS SSRF-675. p. 154–159.
Canadian Food Inspection Agency. 2002. Food safety facts on
mercury and fish consumption. Fact Sheet No. P0083E-1, May
2002. Available: http://www.inspection.gc.ca/english/corpaffr/
foodfacts/mercure.shtml. Accessed 12 October 2005.
Cappon CJ, Smith JC. 1982. Chemical form and distribution of
mercury and selenium in edible seafood. Journal of Analytical
Toxicology 6:10–20.
Center of Food Safety and Applied Nutrition. 2004. Consumer
advisory, an important message for pregnant women and
women of childbearing age who may become pregnant about
the risk of mercury in fish. Washington, DC: US Food and
Drug Administration, March. Available: http://www.cfan.fda.
goc/dms/admehg.html). Accessed 4 October 2005.
Chang SK. 2004. Catch rate analysis of the Indian Ocean
swordfish from Taiwanese longline fishery using generalized
linear model. Journal of Taiwan Fisheries Research
31:115–126.
Chang SK, Hsu CC. 2002. Development of standardized catch
rate of South Atlantic swordfish for Taiwanese longline
fleet. ICCAT Collection Volume of Scientific Paper SCRS/
02/120.
Chen M-H, Chou CL. 2000. An instrumental correction for the
determination of mercury in biological and sediment samples
using cold vapor atomic absorption spectrophotometry. Journal
of the Chinese Chemistry Society 47:1145–1153.
Chen M-H, Shih C-C, Chou CL, Chou L-S. 2002. Mercury,
organic mercury and selenium in small cetaceans in Taiwanese
waters. Marine Pollution Bulletin 45:237–245.
Dabeka R, McKenzie AD, Forsyth DS, Conacher HBS. 2004.
Survey of total mercury in some edible fish and shell fish
species collected in Canada in 2002. Food Additives and
Contaminants 22:434–440.
Department of Health. 1992. Collection of food sanitation
standards — standards of food and hygiene for fish and
shellfishes. Department of Health, Executive Yuan, Taiwan,
1998. Taipei: Department of Health. p. 13-3-1.
European Commission.. 2005. Commission regulation (EC) No.
78/2005 of 19 January 2005 amending regulation (EC) No
466/2001 as regards heavy metals. Official Journal of the
European Union L16/45.
Forsyth DS, Casey V, Dabeka RW, McKenzie A. 2004.
Methylmercury levels in predatory fish species marketed in
Canada. Food Additives and Contaminants 21:849–856.
Freeman HC, Shum G, Uthe JF. 1978. The selenium content in
swordfish (Xiphias gladius) in relation to the total mercury
content. Journal of Environmental Science and Health
13:235–240.
Health Canada Mercury Issues Task Group. 2004. Mercury: your
health and the environment — a resource tool. Health Canada,
1 October. Available: http://www.hc-sc.gc.ca/ewh-semt/
pus/contaminants/risk-risque/mercur/mercur-e.html. Accessed
7 October 2005.
975
Holsbeek L, Das HK, Joiris CR. 1997. Mercury speciation and
accumulation in Bangladesh freshwater and anadromous fish.
Science of the Total Environment 198:201–210.
Joiris CR, Das HK, Holsbeek L. 2000. Mercury accumulation
and speciation in marine fish from Bangladesh. Marine
Pollution Bulletin 40:454–457.
Joiris CR, Holsbeek L, Moatemri NL. 1999. Total and
methylmercury in sardines, Sardinella aurita and Sardina
pilchardue from Tunisia. Marine Pollution Bulletin
38:188–192.
Kamps LR, Carr R, Miller H. 1972. Total mercury-monomethylmercury content of several species of fish. Bulletin of
Environmental Contamination and Toxicology 8:273–279.
Kannan K, Smith RG Jr, Lee RF, Windom HL, Heitmuller PT,
Macauley JM, Summers JK. 1998. Distribution of total
mercury and methyl mercury in water, sediment and fish
from south Florida estuaries. Archives Environmental
Contamination and Toxicology 34:109–118.
Kojadinovic J, Potier M, Corre ML, Cosson RP, Bustamante P.
2006. Mercury content in commercial pelagic fish and its risk
assessment in the western Indian Ocean. Science of the Total
Environment 366:688–700.
Kumar M, Aalbersberg B, Mosley L. 2004. Mercury levels in
Fijian seafood and potential health implications. Report.
Geneva: World Health Organization. Available: http://www.
wpro.who.int/NR/rdonlyres/437775AO-671F-41AE-A068-628
7CC72DF3F/0/mercury_level_fijan_seafoods.pdf. Accessed
12 October 2005.
MacIntosh DL, Spengler JD, Ozkaynak H, Tsai L, Ryan B. 1996.
Dietary exposures to selected metal and pesticides.
Environmental Health Perspectives 104:202–209.
Monteiro LR, Lopes HD. 1990. Mercury content of swordfish,
Xiphias gladius, in relation to length, weight, age and sex.
Marine Pollution Bulletin 21:293–236.
Nakagawa R, Yumita Y, Hiromoto M. 1997. Total mercury
intake from fish and shellfish by Japanese people. Chemosphere
35:2909–2913.
Riisgård HU, Hansen S. 1990. Biomagnification of mercury in
a marine grazing food-chain: algal cells (Phaeodactylum
tricornutum), mussels (Mytilus edulis) and flounders
(Platichthys flesus) studied by means of a stepwise-reduction
CVAA method. Marine Ecology Progress Series 62:259–270.
Statistical Analysis Software Institute. 1998. SAS/STATÕ ‘User’s
guide’. Release 6.03. Cary, NC: Statistical Analysis Software
Institute, Inc.
Storelli MM, Macrotrigiano GO. 2001. Total mercury levels in
muscle tissues of swordfish (Xiphias gladius) and bluefin tuna
(Thunnus thynnus) from the Mediterranean Sea (Italy). Journal
of Food Protection 64:1058–1061.
Sun CT, Chang WH. 1972. A survey of mercury residue in fish
caught by Taiwan’s fishing vessels. Journal of Fisheries Society
of Taiwan 1:31–40.
Watras CJ, Bloom NS. 1992. Mercury and methylmercury in
individual zooplankton: Implications for bioaccumulation.
Limnology and Oceanography 37:1313–1318.
World Health Organization. 2003. Summary and conclusions of
the Sixty-first Meeting of the Joint FAO/WHO Expert
Committee on Food Additives (JECFA), Rome, Italy, 10–19
June. JECFA/61/SC. p. 18–22.
Wu SJ, Chang YH, Fang CW, Pan WH. 1999. Food sources of
weight, calories and three macro-nutrients — NAHSIT 1993–
1996. Nutritional Science Journal 24:41–58.
Ysart G, Miller P, Croasdale M, Crews H, Robb P, Baxter N,
L’Argy C, Harrison N. 2000. 1997 UK Total Diet Study —
dietary exposure to aluminum, arsenic, cadmium, chromium,
copper, lead, mercury, nickel, selenium, tin and zinc. Food
Additives and Contaminants 17:775–786.