INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 118
INORGANIC MERCURY
This report contains the collective views of an international group of
experts and does not necessarily represent the decisions or the stated
policy of the United Nations Environment Programme, the International
Labour Organisation, or the World Health Organization.
Published under the joint sponsorship of
the United Nations Environment Programme,
the International Labour Organisation,
and the World Health Organization
First draft prepared by Dr. L. Friberg,
Karolinska Institute, Sweden
World Health Orgnization
Geneva, 1991
The International Programme on Chemical Safety (IPCS) is a
joint venture of the United Nations Environment Programme, the
International Labour Organisation, and the World Health
Organization. The main objective of the IPCS is to carry out and
disseminate evaluations of the effects of chemicals on human health
and the quality of the environment. Supporting activities include
the development of epidemiological, experimental laboratory, and
risk-assessment methods that could produce internationally
comparable results, and the development of manpower in the field of
toxicology. Other activities carried out by the IPCS include the
development of know-how for coping with chemical accidents,
coordination of laboratory testing and epidemiological studies, and
promotion of research on the mechanisms of the biological action of
chemicals.
WHO Library Cataloguing in Publication Data
Inorganic mercury.
(Environmental health criteria ; 118)
1.Mercury poisoning 2.Environmental pollutants
I.Series
ISBN 92 4 157118 7 (NLM Classification: QV 293)
ISSN 0250-863X
The World Health Organization welcomes requests for permission
to reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made
to the text, plans for new editions, and reprints and translations
already available.
(c) World Health Organization 1991
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.
The designations employed and the presentation of the material
in this publication do not imply the expression of any opinion
whatsoever on the part of the Secretariat of the World Health
Organization concerning the legal status of any country, territory,
city or area or of its authorities, or concerning the delimitation
of its frontiers or boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar
nature that are not mentioned. Errors and omissions excepted, the
names of proprietary products are distinguished by initial capital
letters.
CONTENTS
ENVIRONMENTAL HEALTH CRITERIA FOR INORGANIC MERCURY
1. SUMMARY AND CONCLUSIONS
1.1. Identity
1.2. Physical and chemical properties
1.3. Analytical methods
1.3.1. Analysis, sampling, and storage of urine
1.3.2. Analysis and sampling of air
1.4. Sources of human and environmental exposure
1.4.1. Natural occurrence
1.4.2. Sources due to human activities
1.5. Uses
1.6. Environmental transport, distribution, and transformation
1.7. Human exposure
1.8. Kinetics and metabolism
1.8.1. Reference and normal values
1.9. Effects in humans
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, ANALYTICAL METHODS
2.1. Identity
2.2. Physical and chemical properties
2.3. Conversion factors
2.4. Analytical methods
2.4.1. Analysis, sampling, and storage of urine
2.4.2. Analysis and sampling of air
2.4.3. Quality control and quality assurance
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
3.1. Natural occurrence
3.2. Man-made sources
3.3. Uses
3.4. Dental amalgam in dentistry
3.5. Mercury-containing cream and soap
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
5.1. General population exposure
5.1.1. Exposure from dental amalgam
5.1.1.1 Human studies
5.1.1.2 Animal experiments
5.1.2. Skin-lightening soaps and creams
5.1.3. Mercury in paint
5.2. Occupational exposure during manufacture, formulation, and
use
6. KINETICS AND METABOLISM
6.1. Absorption
6.1.1. Absorption by inhalation
6.1.2. Absorption by ingestion
6.1.3. Absorption through skin
6.1.4. Absorption by axonal transport
6.2. Distribution
6.3. Metabolic transformation
6.4. Elimination and excretion
6.5. Retention and turnover
6.5.1. Biological half-time
6.5.2. Reference or normal values in indicator media
7. EFFECTS ON ORGANISMS IN THE ENVIRONMENT
7.1. Uptake, elimination, and accumulation in organisms
7.2. Toxicity to microorganisms
7.3. Toxicity to aquatic organisms
7.4. Toxicity to terrestrial organisms
7.5. Effects of mercury in the field
8. EFFECTS ON EXPERIMENTAL ANIMALS AND IN VITRO TEST SYSTEMS
8.1. Single and short-term exposure
8.2. Long-term exposure
8.2.1. General effects
8.2.2. Immunological effects
8.2.2.1 Auto-immunity
8.2.2.2 Genetics
8.2.2.3 Mechanisms of induction
8.2.2.4 Autoregulation
8.2.2.5 Immunosuppression
8.2.2.6 Conclusions
8.3. Reproduction, embryotoxicity, and teratogenicity
8.3.1. Males
8.3.2. Females
8.4. Mutagenicity and related end-points
8.5. Carcinogenicity
8.6. Factors modifying toxicity
8.7. Mechanisms of toxicity - mode of action
9. EFFECTS ON HUMANS
9.1. Acute toxicity
9.2. Effects on the nervous system
9.2.1. Relations between mercury in central nervous system
and effects/response
9.2.2. Relations between mercury in air, urine or blood
and effects/response
9.2.2.1 Occupational exposure
9.2.2.2 General population exposure
9.3. Effects on the kidney
9.3.1. Immunological effects
9.3.2. Relations between mercury in organs and effects/response
9.3.3. Relations between mercury in air, urine and/or blood and
effect/response
9.4. Skin reactions
9.4.1. Contact dermatitis
9.4.2. Pink disease and other skin manifestations
9.5. Carcinogenicity
9.6. Mutagenicity and related end-points
9.7. Dental amalgam and general health
9.8. Reproduction, embryotoxicity, and teratogenicity
9.8.1. Occupational exposure
9.8.1.1 In males
9.8.1.2 In females
10. EVALUATION OF HUMAN HEALTH RISKS
10.1. Exposure levels and routes
10.1.1. Mercury vapour
10.1.2. Inorganic mercury compounds
10.2. Toxic effects
10.2.1. Mercury vapour
10.2.2. Inorganic mercury compounds
10.3. Dose-response relationships
10.3.1. Mercury vapour
10.3.2. Inorganic mercury compounds
11. RECOMMENDATIONS FOR FURTHER RESEARCH
12. PREVIOUS EVALUATIONS BY INTERNATIONAL BODIES
REFERENCES
RESUME ET CONCLUSIONS
RESUMEN Y CONCLUSIONES
WHO TASK GROUP ON ENVIRONMENTAL HEALTH CRITERIA FOR INORGANIC
MERCURY
Members
Professor M. Berlin, Institute of Environmental Medicine,
University of Lund, Lund, Sweden (Chairman)
Professor P. Druet, Broussais Hospital, Paris, France
Professor V. Foà, Institute of Occupational Health, Uni-
versity of Milan, Milan, Italy
Professor L. Friberg, Karolinska Institute, Department of
Environmental Hygiene, Stockholm, Sweden
Professor P. Glantz, Prosthetic Dentistry, Faculty of
Odontology, University of Lund, Tandlakarhogskolan,
Malmö, Sweden
Professor C.A. Gotelli, Centre for Toxicological Research,
Buenos Aires, Argentina
Professor G. Kazantzis, Institute of Occupational Health,
London School of Hygiene and Tropical Medicine, London,
United Kingdom (Rapporteur)
Dr L. Magos, Toxicological Unit, Medical Research Council,
Carshalton, Surrey, United Kingdom
Dr W.B. Peirano, Environmental Criteria and Assessment
Office, Office of Research and Development, US Environ-
mental Protection Agency, Cincinnati, USA
Professor B.S. Sridhara Rama Rao, Department of Neurochem-
istry, National Institute of Mental Health and Neuro-
sciences, Bangalore, India
Professor M. Riolfatti, Institute of Hygiene, Faculty of
Pharmaceutical Science, Padova, Italy
Dr M.J. Vimy, Health Science Centre, Department of Medi-
cine and Medical Physiology, Faculty of Medicine, Uni-
versity of Calgary, Calgary, Alberta, Canada
Observers
Dr M. Ancora, Centro Italiano Studi e Indagini, Rome,
Italy
Professor K.S. Larsson, Institute for Odontological Toxi-
cology, Faculty of Dentistry, Karolinska Institute,
Huddinge, Sweden
Observers (contd.)
Professor C. Maltoni, Institute of Oncology, Bologna,
Italy
Dr A. Mochi, Centro Italiano Studi e Indagini, Rome, Italy
Professor A.A.G. Tomlinson, Centro Italiano Studi e
Indagini, Rome, Italy
Secretariat
Dr D. Kello, Toxicology and Food Safety, World Health
Organization Regional Office for Europe, Copenhagen,
Denmark
Dr T. Kjellström, Prevention of Environmental Pollution,
Division of Environmental Health, World Health Organiz-
ation, Geneva, Switzerland (Secretary)
NOTE TO READERS OF THE CRITERIA MONOGRAPHS
Every effort has been made to present information in
the criteria monographs as accurately as possible without
unduly delaying their publication. In the interest of all
users of the environmental health criteria monographs,
readers are kindly requested to communicate any errors
that may have occurred to the Manager of the International
Programme on Chemical Safety, World Health Organization,
Geneva, Switzerland, in order that they may be included in
corrigenda, which will appear in subsequent volumes.
* * *
A detailed data profile and a legal file can be
obtained from the International Register of Potentially
Toxic Chemicals, Palais des Nations, 1211 Geneva 10,
Switzerland (Telephone No. 7988400 or 7985850).
ENVIRONMENTAL HEALTH CRITERIA FOR INORGANIC MERCURY
A WHO Task Group on Environmental Health Criteria for
Inorganic Mercury met in Bologna, Italy, at the County
Council Headquarters (Provincia) from 25 to 30 September
1989. The meeting was sponsored by the Italian Ministry
of the Environment and organized locally by the Institute
of Oncology and Environmental Sciences with the assistance
of the County Council. Professor C. Maltoni, Director of
the Bologna Institute of Oncology, opened the meeting and
welcomed the participants on behalf of the host insti-
tution. Mr A. Vecchi, Dr M. Moruzzi, and Dr A. Lolli, wel-
comed the participants on behalf of the local authorities.
Dr A. Mochi, Centro Italiano Studi e Indagini, greeted the
participants on behalf of the Ministry of the Environment,
and Dr D. Kello, WHO Regional Office for Europe, addressed
the meeting on behalf of the cooperating organizations of
the IPCS (ILO/UNEP/WHO).
The Task Group reviewed and revised the draft document
and made an evaluation of the human health risks from
exposure to inorganic mercury.
The draft of this report was prepared by Dr L.
Friberg, Karolinska Institute, Stockholm, Sweden. Dr T.
Kjellström, WHO, Geneva, was responsible for the overall
scientific content and Dr P.G. Jenkins, WHO, Geneva, for
the technical editing.
* * *
Partial financial support for the publication of this
report was kindly provided by the National Institute of
Environmental Medicine, Stockholm, Sweden, and the Minis-
try of the Environment of Italy. The Centro Italiano Studi
e Indagini and the Institute of Oncology, Bologna, con-
tributed to the organization and provision of meeting
facilities.
ABBREVIATIONS
AAS atomic absorption spectrophotometry
CNS central nervous system
CVAA cold vapour atomic absorption
EEC European Economic Community
EEG electroencephalogram
GBM glomerular basement membrane
GC gas chromatography
GEMS Global Environment Monitoring System
GLC gas-liquid chromatography
LOAEL lowest-observed-adverse-effect level
MGP membranous glomerulopathy
NOAEL no-observed-adverse-effect level
SD standard deviation
SMR standardized mortality ratio
TWA time-weighted average
1. SUMMARY AND CONCLUSIONS
This monograph concentrates primarily on the risk to
human health from inorganic mercury, and examines research
reports that have appeared since the publication of
Environmental Health Criteria 1: Mercury (WHO, 1976). In
the period since 1976, new research data has become avail-
able for two main health concerns related to inorganic
mercury, i.e. mercury in dental amalgam and in skin-
lightening soaps. The emphasis in this monograph is on
exposure from these two sources, but the basic kinetics
and toxicology are reviewed with all aspects of inorganic
mercury in mind.
Human health concerns related to the global transport,
bioaccumulation, and transformation of inorganic mercury
almost exclusively arise from the conversion of mercury
compounds into methylmercury and exposure to methylmercury
in sea-food and other food. The global environmental and
ecological aspects of inorganic mercury have been summar-
ized in this monograph. More detailed descriptions may be
found in Environmental Health Criteria 86: Mercury -
Environmental Aspects (WHO, 1989) and Environmental Health
Criteria 101: Methylmercury (WHO, 1990).
1.1. Identity
Mercury exists in three states: Hg0 (metallic);
Hg2++ (mercurous); and Hg++ (mercuric). It can form
organometallic compounds, some of which have found
industrial and agricultural uses.
1.2. Physical and chemical properties
Elemental mercury has a very high vapour pressure.
The saturated atmosphere at 20 °C has a concentration over
200 times greater than the currently accepted concen-
tration for occupational exposure.
Solubility in water increases in the order: elemental
mercury < mercurous chloride < methylmercury chloride <
mercuric chloride. Elemental mercury and the halide com-
pounds of alkylmercurials are soluble in non-polar
solvents.
Mercury vapour is more soluble in plasma, whole blood,
and haemoglobin than in distilled water, where it dis-
solves only slightly. The organometallic compounds are
stable, although some are readily broken down by living
organisms.
1.3. Analytical methods
The most commonly used analytical methods for the
quantification of total and inorganic mercury compounds
are atomic absorption of cold vapour (CVAA) and neutron
activation. Detailed information relating to analytical
methods are given in Environmental Health Criteria 1:
Mercury (WHO, 1976) and Environmental Health Criteria 101:
Methylmercury (WHO, 1990).
All analytical procedures for mercury require careful
quality control and quality assurance.
1.3.1. Analysis, sampling, and storage of urine
Flameless atomic absorption spectrophotometry is used
in routine analysis for various media. Particular care
must be taken when choosing the anticoagulant used for
blood sampling in order to avoid contamination by mercury
compounds. Special care must also be taken in the sampling
and storage of urine, since bacterial growth can change
the concentration of the numerous forms of mercury that
may be present. Addition of hydrochloric acid or bacteri-
cidal substances and freezing the sample are the best
methods to prevent alteration of urine samples. Correc-
tion of concentration by reference to urine density or
creatinine content are recommended.
1.3.2. Analysis and sampling of air
Analytical methods for mercury in air may be divided
into instant reading methods and methods with separate
sampling and analysis stages. Instant reading methods can
be used for the quantification of elemental mercury
vapour. Sampling in acid-oxidizing media or on hopcalite
is used for the quantification of total mercury.
The cold vapour atomic absorption (CVAA) technique is
the most frequently used analytical method.
1.4. Sources of human and environmental exposure
1.4.1. Natural occurrence
The major natural sources of mercury are degassing of
the earth's crust, emissions from volcanoes, and evapor-
ation from natural bodies of water.
The natural emissions are of the order of 2700-6000
tonnes per year.
1.4.2. Sources due to human activities
The world-wide mining of mercury is estimated to yield
about 10 000 tonnes/year. These activities lead to some
losses of mercury and direct discharges to the atmos-
phere. Other important sources are fossil fuel combustion,
metal sulfide ore smelting, gold refining, cement pro-
duction, refuse incineration, and industrial applications
of metals.
The specific normal emission from a chloralkali plant
is about 450 g of mercury per ton of caustic soda
produced.
The total global amount and release of mercury, due to
human activities, to the atmosphere has been estimated to
be up to 3000 tonnes/year.
1.5. Uses
A major use of mercury is as a cathode in the elec-
trolysis of sodium chloride. Since the resultant chemicals
are contaminated with mercury, their use in other indus-
trial activities leads to a contamination of other
products. Mercury is used in the electrical industry, in
control instruments in the home and industry, and in lab-
oratory and medical instruments. Some therapeutic agents
contain inorganic mercury. A very large amount of mercury
is used for the extraction of gold.
Dental silver amalgam for tooth filling contains large
amounts of mercury, mixed (in the proportion of 1:1) with
alloy powder (silver, tin, copper, zinc). Copper amalgam,
used mostly in paediatric dentistry, contains up to 70%
mercury and up to 30% copper. These uses can cause ex-
posure of the dentist, dental assistants, and also of the
patients.
Some dark-skinned people use mercury-containing creams
and soap to achieve a lighter skin tone. The distribution
of these products is now banned in the EEC, in North
America, and in many African countries, but mercury-
containing soap is still manufactured in several European
countries. The soaps contain up to 3% of mercuric iodine
and the creams contain ammoniated mercury (up to 10%).
1.6. Environmental transport, distribution, and transformation
Emitted mercury vapour is converted to soluble forms
and deposited by rain onto soil and water. The atmospheric
residence time for mercury vapour is up to 3 years,
whereas soluble forms have a residence time of only a few
weeks.
The change in speciation of mercury from inorganic to
methylated forms is the first step in the aquatic bioac-
cumulation process. This can occur non-enzymically or
through microbial action. Methylmercury enters the food-
chain of predatory species where biomagnification occurs.
1.7. Human exposure
The general population is primarily exposed to mercury
through the diet and dental amalgam. Depending on the con-
centrations in air and water, significant contributions to
the daily intake of total mercury can occur. Fish is a
dominant source of human exposure to methylmercury.
Recent experimental studies have shown that mercury is
released from amalgam restorations in the mouth as vapour.
The release rate of this mercury vapour is increased, for
example, by chewing. Several studies have correlated the
number of dental amalgam fillings or amalgam surfaces with
the mercury content in tissues from human autopsy, as well
as in samples of blood, urine, and plasma. Both the pre-
dicted mercury uptake from amalgam and the observed ac-
cumulation of mercury show substantial individual vari-
ation. It is, therefore, difficult to make accurate
quantitative estimations of the mercury release and uptake
by the human body from dental amalgam tooth restorations.
Experimental studies in sheep have examined in greater
detail the distribution of mercury released from amalgam
restorations.
Use of skin-lightening soap and creams can give rise
to substantial mercury exposure.
Occupational exposure to inorganic mercury has been
investigated in chloralkali plants, mercury mines, ther-
mometer factories, refineries, and in dental clinics.
High mercury levels have been reported for all these
occupational exposure situations, although levels vary
according to work environment conditions.
1.8. Kinetics and metabolism
Results of both human and animal studies indicate that
about 80% of inhaled metallic mercury vapour is retained
by the body, whereas liquid metallic mercury is poorly
absorbed via the gastrointestinal tract (less than 1%).
Inhaled inorganic mercury aerosols are deposited in the
respiratory tract and absorbed, the rate depending on
particle size. Inorganic mercury compounds are probably
absorbed from the human gastrointestinal tract to a level
of less than 10% on average, but there is considerable
individual variation. Absorption is much higher in newborn
rats.
The kidney is the main depository of mercury after the
administration of elemental mercury vapour or inorganic
mercury compounds (50-90% of the body burden of animals).
Significantly more mercury is transported to the brain
of mice and monkeys after the inhalation of elemental
mercury than after the intravenous injection of equivalent
doses of the mercuric form. The red blood cell to plasma
ratio in humans is higher (> 1) after administration of
elemental mercury than mercuric mercury and more mercury
crosses the placental barrier. Only a small fraction of
the administered divalent mercury enters the rat fetus.
Several forms of metabolic transformation can occur:
* oxidation of metallic mercury to divalent mercury;
* reduction of divalent mercury to metallic mercury;
* methylation of inorganic mercury;
* conversion of methylmercury to divalent inorganic
mercury.
The oxidation of metallic mercury vapour to divalent
ionic mercury (section 6.1.1) is not fast enough to pre-
vent the passage of elemental mercury through the blood-
brain barrier, the placenta, and other tissues. Oxidation
in these tissues serves as a trap to hold the mercury and
leads to accumulation in brain and fetal tissues.
The reduction of divalent mercury to Hg0 has been
demonstrated both in animals (mice and rats) and humans.
The decomposition of organomercurials, including methyl-
mercury, is also a source of mercuric mercury.
The faecal and urinary routes are the main pathways
for the elimination of inorganic mercury in humans,
although some elemental mercury is exhaled. One form of
depletion is the transfer of maternal mercury to the
fetus.
The biological half-time, which only lasts a few days
or weeks for most of the absorbed mercury, is very long,
probably years, for a fraction of the mercury. Such long
half-times have been observed in animal experiments as
well as in humans. A complicated interplay exists between
mercury and some other elements, including selenium. The
formation of a selenium complex may be responsible for the
long half-time of a fraction of the mercury.
1.8.1. Reference and normal values
Limited information from deceased miners shows mercury
concentrations in the brain, years after cessation of
exposure, of several mg/kg, with still higher values in
some parts of the brain. However, lack of quality control
of the analysis makes these data uncertain. Among a small
number of deceased dentists, without known symptoms of
mercury intoxication, mercury levels varied from very low
concentrations up to a few hundred µg/kg in the occipital
lobe cortex and from about 100 µg/kg to a few mg/kg in
the pituitary gland.
From autopsies on subjects not occupationally exposed
but with a varying number of amalgam fillings, it seems
that a moderate number (about 25) of amalgam surfaces may
on average increase the brain mercury concentration by
about 10 µg/kg. The corresponding increase in the kid-
neys, based on a very limited number of analyses, is
probably 300-400 µg/kg. However, the individual vari-
ation is considerable.
Mercury levels in urine and blood can be used as indi-
cators of exposure provided that the exposure is recent
and relatively constant, is long-term, and is evaluated on
a group basis. Recent exposure data are more reliable
than those quoted in Environmental Health Criteria 1:
Mercury (WHO, 1976). Urinary levels of about 50 µg per g
creatinine are seen after occupational exposure to about
40 µg mercury/m3 of air. This relationship (5:4) between
urine and air levels is much lower that the 3:1 estimated
by WHO (1976). The difference may in part be explained by
different sampling technique for evaluating air exposure.
An exposure of 40 µg mercury/m3 of air will correspond
to about 15-20 µg mercury/litre of blood. However, inter-
ference from methylmercury exposure can make it difficult
to evaluate exposure to low concentrations of inorganic
mercury by means of blood analysis. A way to overcome the
problems is to analyse mercury in plasma or analyse both
inorganic mercury and methylmercury. The problem of inter-
ference from methylmercury is much smaller when analysing
urine, as methylmercury is excreted in the urine to only a
very limited extent.
1.9. Effects in humans
Acute inhalation exposure to mercury vapour may be
followed by chest pains, dyspnoea, coughing, haemoptysis,
and sometimes interstitial pneumonitis leading to death.
The ingestion of mercuric compounds, in particular
mercuric chloride, has caused ulcerative gastroenteritis
and acute tubular necrosis causing death from anuria where
dialysis was not available.
The central nervous system is the critical organ for
mercury vapour exposure. Subacute exposure has given rise
to psychotic reactions characterized by delirium, halluci-
nations, and suicidal tendency. Occupational exposure has
resulted in erethism as the principal feature of a broad
ranging functional disturbance. With continuing exposure a
fine tremor develops, initially involving the hands. In
the milder cases erethism and tremor regress slowly over a
period of years following removal from exposure. De-
creased nerve conduction velocity has been demonstrated in
mercury-exposed workers. Long-term, low-level exposure
has been associated with less pronounced symptoms of
erethism.
There is very little information available on brain
mercury levels in cases of mercury poisoning, and nothing
that makes it possible to estimate a no-observed-effect
level or a dose-response curve.
At a urinary mercury excretion level of 100 µg per g
creatinine, the probability of developing the classical
neurological signs of mercurial intoxication (tremor,
erethism) and proteinuria is high. An exposure correspond-
ing to 30 to 100 µg mercury/g creatinine increases the
incidence of some less severe toxic effects that do not
lead to overt clinical impairment. In a few studies
tremor, recorded electrophysiologically, has been observed
at low urine concentrations (down to 25-35 µg/g creati-
nine). Other studies did not show such an effect. Some of
the exposed people develop proteinuria (proteins of low
relative molecular mass and microalbuminuria). Appropriate
epidemiological data covering exposure levels correspond-
ing to less than 30-50 µg mercury/g creatinine are not
available.
The exposure of the general population is generally
low, but may occasionally be raised to the level of occu-
pational exposure and can even be toxic. Thus, the
mishandling of liquid mercury has resulted in severe
intoxication.
The kidney is the critical organ following the
ingestion of inorganic divalent mercury salts. Occu-
pational exposure to metallic mercury has long been
associated with the development of proteinuria, both in
workers with other evidence of mercury poisoning and in
those without such evidence. Less commonly, occupational
exposure has been followed by the nephrotic syndrome,
which has also occurred after the use of skin-lightening
creams containing inorganic mercury, and even after acci-
dental exposure. The current evidence suggests that this
nephrotic syndrome results from an immunotoxic response.
Until recently, effects of elemental mercury vapour on the
kidney had been reported only at doses higher than those
associated with the onset of signs and symptoms from the
central nervous system. New studies have, however, re-
ported kidney effects at lower exposure levels. Experi-
mental studies on animals have shown that inorganic
mercury may induce auto-immune glomerulonephritis in all
species tested, but not in all strains, indicating a
genetic predisposition. A consequence of an immunological
etiology is that, in the absence of dose-response studies
for groups of immunologically sensitive individuals, it is
not scientifically possible to set a level for mercury
(e.g., in blood or urine) below which (in individual
cases) mercury-related symptoms will not occur.
Both metallic mercury vapour and mercury compounds
have given rise to contact dermatitis. Mercurial pharma-
ceuticals have been responsible for Pink disease in
children, and mercury vapour exposure may be a cause of
"Kawasaki" disease. In some studies, but not in others,
effects on the menstrual cycle and/or fetal development
have been reported. The standard of published epidemio-
logical studies is such that it remains an open question
whether mercury vapour can adversely affect the menstrual
cycle or fetal development in the absence of the well-
known signs of mercury intoxication.
Recently, there has been an intense debate on the
safety of dental amalgams and claims have been made that
mercury from amalgam may cause severe health hazards.
Reports describing different types of symptoms and signs
and the results of the few epidemiological studies
produced are inconclusive.
2. IDENTITY, PHYSICAL AND CHEMICAL PROPERTIES, ANALYTICAL METHODS
2.1. Identity
This monograph focuses on the risk to human health
from the compounds of inorganic mercury. Other forms of
mercury are discussed where they are relevant to the full
evaluation of human health risks, e.g., the metabolic
transformation of methylmercury to inorganic mercury.
Elemental mercury has the CAS registry number
7439-97-6 and a relative atomic mass of 200.59. There
are three states of inorganic mercury: Hg0 (metallic),
Hg2++ (mercurous), and Hg++ (mercuric) mercury. The
mercurous and mercuric states form numerous inorganic and
organic chemical compounds. Organic forms are those in
which mercury is attached covalently to at least one
carbon atom.
2.2. Physical and chemical properties
In its elemental form, mercury is a heavy silvery
liquid at room temperature. At 20 °C the specific gravity
of the metal is 13.456 and the vapour pressure is 0.16 Pa
(0.0012 mmHg). Thus, a saturated atmosphere at 20 °C con-
tains approximately 15 mg/m3. This concentration is
300 times greater than the recommended health-based occu-
pational exposure limit of 0.05 mg/m3 (WHO, 1980).
Mercurials differ greatly in their solubilities.
Solubility values in water are: elemental mercury (30 °C),
2 µg/litre; mercurous chloride (25 °C), 2 mg/litre; mer-
curic chloride (20 °C), 69 g/litre (Linke, 1958; CRC,
1972). The solubility of methylmercury chloride in water
is higher than that of mercurous chloride by about three
orders of magnitude, this being related to the very high
solubility of the methylmercury cation in water (Linke,
1958; Clarkson et al., 1988b). Certain species of mercury
are soluble in non-polar solvents. These include elemental
mercury and the halide compounds of alkylmercurials
(Clarkson et al., 1988b).
From the biochemical point of view the most important
chemical property of mercuric mercury and alkylmercurials
is their high affinity for sulfhydryl groups.
Hursh (1985) showed that mercury vapour is more sol-
uble in plasma, whole blood, and haemoglobin than in dis-
tilled water or isotonic saline.
The following speciation among mercury compounds has
been proposed by Lindqvist et al. (1984), where V indi-
cates volatile species, R water-soluble particle-borne
reactive species, and NR non-reactive species:
V: Hg0 (elemental mercury), (CH3)2Hg
R: Hg2+, HgX2, HgX3-, and HgX42- (where X = OH-,
Cl-, or Br-), Hg0 on aerosol particles, Hg2+ com-
plexes with organic acids.
NR: CH3Hg+, CH3HgCl, CH3HgOH, and other organomer-
curic compounds, Hg(CN)2, HgS, and Hg2+ bound to
sulfur in fragments of humic matter.
The main volatile form in air is elemental mercury,
but dimethylmercury may also occur (Slemr et al., 1981).
Uncharged complexes, such as HgCl2 and CH3HgOH, oc-
cur in the gaseous phase, but are also relatively stable
in fresh water (snow and rain as well as standing or flow-
ing water). HgCl42- is the dominant form in sea water.
2.3. Conversion factors
1 ppm = 1 mg/kg = 5 µmol/kg
1 mol creatinine = 113.1 g creatinine
2.4. Analytical methods
Detailed information relating to analytical methods
was given in Environmental Health Criteria 1: Mercury
(WHO, 1976) and in Environmental Health Criteria 101:
Methylmercury (WHO, 1990). This monograph contains further
information concerning the sampling and analysis of urine
and air, the most frequently studied media for evaluation
of exposure to inorganic mercury. A summary of the com-
monly used analytical methods is given in Table 1. More
advanced methods, such as inductively coupled plasma
atomic emission spectrometry and spark source mass spec-
trometry, are described in Kneip & Friberg (1986).
2.4.1. Analysis, sampling, and storage of urine
For routine analysis, various forms of flameless
atomic absorption spectrophotometry (AAS) are used. The
"Magos" selective atomic absorption method determines
both total and inorganic mercury and, by difference,
organic mercury. The neutron activation procedure is
regarded as the most accurate and sensitive procedure and
is usually used as the reference method.
Table 1. Analytical methods for the determination of mercury
--------------------------------------------------------------------------------------------------------------------
Media Speciation Analytical Detection Comments References
method limit
(ng Hg/g)
--------------------------------------------------------------------------------------------------------------------
Food, tissues total mercury atomic 2.0 method has many adaptations Hatch & Ott (1968)
absorption (see Peter & Strunc, 1984)
Blood, urine total mercury atomic 0.5 also estimates organic mercury Magos (1971); Magos &
inorganic mercury absorption as difference between total Clarkson (1972)
and inorganic
Blood, urine total mercury atomic 2.5 automated form of the method Farant et al. (1981)
hair, tissues inorganic mercury absorption of Magos (1971)
Blood, urine total mercury atomic 4.0 automated form of the method Coyle & Hartley (1981)
hair, tissues inorganic mercury absorption of Magos (1971)
All media total mercury neutron 0.1 reference method (review) WHO (1976)
activation
--------------------------------------------------------------------------------------------------------------------
Blood samples are best collected in "vacutainers"
containing heparin (without mercury compounds as preserv-
ative) (WHO, 1980) and stored at 4 °C prior to analysis.
This method of collection is especially important if mer-
cury levels in plasma and red blood cells are to be
measured. Blood samples can usually be stored for one or
two days before haemolysis becomes significant (Clarkson
et al., 1988c).
The sampling and storage of urine have been discussed
in detail by Clarkson et al. (1988c). It is important to
avoid contamination of urine samples; special cleaning
procedures and the use of metal-free polyethylene con-
tainers have been recommended.
As a rule, urine is saturated with several inorganic
salts. Precipitates are sometimes seen in freshly voided
samples and are normally present in urine samples that
have been stored at low temperature (1-4 °C). To lessen
problems of precipitates, urine samples should be homogen-
ized by shaking before analysis. Alternatively, a strong
acid, preferably hydrochloric acid, can be added to the
urine sample to lower pH and increase the solubility of
the salts.
Bacterial growth is rapid in urine at room tempera-
ture. Even urine samples from healthy people become over-
grown with bacteria after only a few hours. If urine
samples are frozen (to below -20 °C), bacterial growth is
reduced substantially. Bacteria may reduce some mercury
compounds to elemental mercury, which might give rise to
significant losses of mercury by volatilization (WHO,
1976). Bactericidal substances, such as sodium azide, may
be added to urine samples. However, sodium azide is a
strong reducing agent and may form Hg0 from Hg2+. The
addition of 1 g sulfamic acid and 0.5 ml of a detergent
(Triton X-100) to 500 ml of urine produces stable urine
samples at room temperature for at least one month (Skare,
1972).
Even when the rate of metal excretion is constant,
metal concentration in urine varies according to the urine
flow rate (Diamond, 1988). It is therefore necessary to
adjust the measured concentrations of metals in spot urine
samples for variations in the urine flow rate. This can be
done by correcting for urine relative density or osmo-
lality or by dividing by the concentration of creatinine
in the urine sample. Another alternative is the use of
timed urine specimens (e.g., 4 h or 8 h). If the concen-
tration of a substance is standardized to a constant rela-
tive density (usually 1.018 or 1.024), the basis of cor-
rection chosen profoundly changes the figures obtained.
Correction to 1.024 gives values 33% higher than correc-
tion to 1.018 (Aitio, 1988). Furthermore, many chemicals,
including mercury, exhibit diurnal variation in concen-
tration (Piotrowski et al., 1975). Correction using cre-
atinine values has the advantage that the mercury concen-
tration will be independent of hydration status.
2.4.2. Analysis and sampling of air
Analytical methods for mercury in air may be divided
into instant reading methods and methods with separate
sampling and analysis stages (WHO, 1976).
One instant reading method is based on the "cold
vapour atomic absorption" (CVAA) technique, which
measures the absorption of mercury vapour by ultra-violet
light using a wave length of 253.7 nm. Most of the AAS
procedures have a detection limit in the range of 2 to
5 µg mercury/m3.
Another instant reading method that has been used
increasingly in recent years is a special type of gold
amalgamation technique. This method has been used in a
number of studies for evaluating the release of elemental
mercury vapour in the oral cavity from amalgam fillings
(Svare et al., 1981; Vimy & Lorscheider, 1985a,b).
McNerney et al. (1972) gave a detailed description of the
method, which is based on an increase in the electrical
resistance of a thin gold film after adsorption of mercury
vapour. The detection limit is 0.05 ng mercury. Within the
range of 0.5 to 25 ng, the relative standard deviation was
found to vary between 3 and 10% when 15 samples from each
of 6 mercury vapour standards were examined. At higher
mercury concentrations, the films become saturated with
mercury and precision decreases. It is possible to correct
for this saturation with a calibration curve. However,
there are no data on the accuracy of the method when used
in actual field studies, such as the ones by Svare et al.
(1981) or Vimy & Lorscheider (1985a,b).
In an analytical method based on separate sampling and
analysis, the air is sampled in two bubblers in series,
containing sulfuric acid and potassium permanganate (WHO,
1976). The mercury is subsequently determined by CVAA.
With this method the total mercury in the air is measured,
not just mercury vapour. Another sampling technique uses
solid absorbants. Different amalgamation techniques using
gold have been shown to have good collection efficiency
for mercury vapour (McCammon et al., 1980; Dumarey et al.,
1985; Skare & Engqvist, 1986). Roels et al. (1987) used a
filter with two layers of hopcalite (a mixture of metal
oxides that can absorb metals) to collect the mercury.
After solubilization, the mercury was analysed by a CVAA
technique. It was necessary also to measure blanks of hop-
calite and scrubbing solution. Large variations were found
for background mercury contamination of hopcalite from
batch to batch (6-93 ng mercury per 200 mg hopcalite).
Sampling of air for mercury analysis can be made by
static samplers or by personal monitoring. Personal
samplers are recommended. A study by Roels et al. (1987)
compared results obtained with the use of static samplers
with results from personal samplers. In most of the
workplaces, personal samplers yielded higher exposure
levels (time-weighted averages) than did static samplers
(see section 6.5.2).
2.4.3. Quality control and quality assurance
General considerations of quality control and quality
assurance have been recommended by WHO (UNEP/WHO, 1984;
WHO, 1986; Aitio, 1988). At a recent conference on "Bio-
logical Monitoring of Toxic Metals" (Friberg, 1988), a
WHO approach based on a GEMS programme (Vahter, 1982) was
described in detail. Specific quality control programmes
for mercury in hair using the GEMS approach have been
described (Lind et al., 1988). Roels et al. (1987) suc-
cessfully used another regression method when analysing
mercury in urine.
In almost any quality control programme, there is a
need for reference materials containing the metal in con-
centrations covering the expected working range of moni-
toring samples. Several reference materials are commer-
cially available for both environmental samples and for
urine and blood (Muramatsu & Parr, 1985; Parr et al.,
1987; Rasberry, 1987; Parr et al., 1988; Okamoto, 1988).
The following are suppliers of reference materials: NIST
(Office of Standard Reference Materials, National Insti-
tute of Standards and Technology, Rm. B311, Chemistry
Bldg., Gaithersburg, MD 20899, USA), IAEA (International
Atomic Energy Agency, Analytical Quality Control Services,
Laboratory Seibersdorf, A-1400 Vienna), BCR (Community
Bureau of Reference, Commission of the European Communi-
ties, 200 Rue de la Loi, B-1049 Brussels, Belgium); NIES
(National Institute for Environmental Studies, Japan
Environment Agency, P.O. Yatabe, Tsukuba Ibaraki 300-21,
Japan), NRCC (National Research Council Canada, Division
of Chemistry, Ottawa, K1A OR6, Canada), Nycomed AS Diag-
nostics (P.O. Box 4220, Torshov, 0401 Oslo 4, Norway),
Behring Institute (P.O. Box 1140, D-3550 Marburg 1,
Germany), Kaulson Laboratories Inc. (691 Bloomfield
Avenue, Caldwell, New Jersey 07006, USA). However, the
available reference materials do not cover the demand for
different mercury species, biological media or for differ-
ent concentrations. Only NRCC has a reference material
(fish) for total mercury and for methylmercury.
3. SOURCES OF HUMAN AND ENVIRONMENTAL EXPOSURE
3.1. Natural occurrence
The major natural sources of mercury are the degassing
of the earth's crust, emissions from volcanoes, and evap-
oration from natural bodies of water (National Academy of
Sciences, 1978; Nriagu, 1979; Lindqvist et al., 1984). The
most recent estimates indicate that natural emissions are
of the order of 2700-6000 tonnes per year (Lindberg et
al., 1987).
The earth's crust is also an important source of mer-
cury for bodies of natural water. Some of this mercury is
undoubtedly of natural origin, but some may have been
deposited from the atmosphere and may ultimately have been
generated by human activities (Lindqvist et al., 1984).
Thus, it is difficult to assess quantitatively the rela-
tive contributions of natural and anthropogenic mercury to
run-off from land to natural bodies of water. Data con-
cerning mercury in the general environment and in food
have been reviewed in Environmental Health Criteria 101:
Methylmercury (WHO, 1990).
3.2. Man-made sources
The worldwide mining of mercury is estimated to yield
about 10 000 tonnes/year. Mining activities result in
losses of mercury through the dumping of mine tailings and
direct discharges to the atmosphere. The Almaden mercury
mine in Spain, which accounts for 90% of the total output
of the European Community, was expected to produce 1380
tonnes in 1987 (Seco, 1987). Other important sources are
the combustion of fossil fuel, the smelting of metal sul-
fide ores, the refining of gold (sometimes under very
primitive conditions), the production of cement, refuse
incineration, and industrial metal applications. The
emissions of mercury to the atmosphere in Sweden in 1984
were estimated to be as follows (in kg/year): incineration
of household waste (3300), smelting (900), chloralkali
industry (400), crematories (300), mining (200), combus-
tion of coal and peat (200), other sources (200) (Swedish
Environmental Protection Board, 1986). Analogous data for
the estimated atmospheric emissions of mercury in the
United Kingdom were (in kg/year): fossil fuel combustion
(25 500), production and use of articles containing mer-
cury (10 100), municipal waste incineration (5900), non-
ferrous metal production (5000), cement manufacture
(2500), iron and steel production (1800), sewage sludge
incineration (500) (Dean & Suess, 1985). In developing
countries the emissions from industry and mining may be
much greater. For example, the emission to water from one
single chloralkali plant in Nicaragua in 1980 was 24 kg
per day (9 tonnes/year) (Velasquez et al., 1980). It was
estimated that 450 g of mercury was emitted per tonne of
soda produced in six chloroalkali plants in Argentina, and
the quantity of mercury released in the environment was
about 86 tonnes/year (Gotelli, 1989).
The total global release of mercury to the atmosphere
due to human activities has been estimated to be of the
order of 2000-3000 tonnes/year (Lindberg et al., 1987;
Pacyna, 1987). It should be stressed that there are con-
siderable uncertainties in the estimated fluxes of mercury
in the environment and in its speciation. Concentrations
in the unpolluted atmosphere and in natural bodies of
water are so low that they are near the limit of detection
of current analytical methods, even for the determination
of total mercury.
Although amounts of mercury resulting from human ac-
tivities may be quite small relative to global emissions,
the anthropogenic release of elemental metal mercury into
confined areas was the source of the poisoning outbreaks
in Minamata and Niigata (WHO, 1976).
3.3. Uses
A major use of mercury is as a cathode in the elec-
trolysis of sodium chloride solution to produce caustic
soda and chlorine gas, which has important uses in the
paper-pulp industry. It should be noted that all the elec-
trolytic products (hydrogen, chlorine, sodium hydroxide,
sodium hypochlorite, and hydrochloric acid) are contami-
nated with mercury (Gotelli, 1989). These substances are
important in the economy of other industrial activities
and the presence of mercury can contaminate other prod-
ucts. About 50 tonnes of liquid metal are used in each
manufacturing plant. In most industrialized countries,
stringent procedures have been taken to reduce losses of
mercury. Mercury is widely used in the electrical industry
(lamps, arc rectifiers, and mercury battery cells), in
control instruments in the home and industry (switches,
thermostats, barometers), and in other laboratory and
medical instruments. It is also widely used in the dental
profession for tooth amalgam fillings. Other therapeutic
agents, such as teething powders, ointments, and laxa-
tives, contain inorganic mercury (ATSDR, 1989), as do some
antihistaminic preparations sold in Italy (EDIMED, 1989).
Organic mercury compounds continue to be used in anti-
fouling and mildew-proofing latex paints and to control
fungus infections of seeds, bulb plants, and vegetation.
The World Health Organization has warned against the use
of alkylmercury compounds in seed dressing (WHO, 1976).
One of the uses of liquid metallic mercury that may
have a serious impact on health is the extraction of gold
from ore concentrates or from recycled gold articles.
Reports from China (Wu et al., 1989) indicate high ex-
posure in the vicinity of "cottage industry" operations
of this type, and Villaluz (1988) reported that 50 000
people may be exposed around small scale gold mining oper-
ations in Indonesia, Kampuchea, the Philippines, and
Viet Nam. The same problem also occurs in Brazil and
Colombia. The release of elemental mercury from these
activities is about 120 tonnes/year in Brazil (Gotelli,
1989).
3.4. Dental amalgam in dentistry
WHO (1976) estimated that in industrial countries
about 3% of the total consumption of mercury was used for
dental amalgam. Amalgam has been used extensively as a
tooth-filling material for more than 150 years and
accounts for 75-80% of all single tooth restorations
(Bauer & First, 1982; Wolff et al., 1983). It has been
estimated that each American dentist in private practice
uses on average 0.9-1.4 kg of amalgam per year (Naleway et
al., 1985).
Most conventional silver amalgams consist of a 1:1
mixture of metallic mercury and an alloy powder consisting
of silver (about 70% by weight), tin (about 25%), and
smaller amounts of copper (1-6%) and zinc (0-2%). A modern
type of silver amalgam is also available, containing
higher amounts of copper (up to about 25%). At the time of
trituration (mixing), the amalgam generally contains simi-
lar weights of alloy powders and mercury. Excess mercury
(< 5%) is removed immediately before or at the conden-
sation of the plastic amalgam mix in the prepared tooth
cavity. The amalgam begins to set within minutes of inser-
tion and therefore needs to be carved to satisfactory
anatomic form within this period of time. Finishing (e.g.,
polishing) with rotating instruments can take place after
setting for 24 h, but continuing hardening of amalgam
restorations takes place over many months (ADA, 1985;
Enwonwu, 1987; SOS, 1987).
Previously, amalgam was usually prepared with mortar
and pestle. The amalgam mixture was thereafter placed on
a cloth filter and squeezed to expel excess mercury. This
method of handling amalgam easily vapourizes mercury and
there is also a risk of spillage. The technique is still
in use in some countries (section 9.5.2.2). The modern,
safer method for the preparation of amalgam involves
mixing the alloy with mercury in a sealed capsule. This
decreases the occupational exposure substantially (Harris
et al., 1978; Skuba, 1984).
A second type of dental amalgam is the so-called
"copper amalgam" used mostly in paediatric dentistry
until a few decades ago. This material contained 60-70%
mercury and 30-40% copper, and was prepared by open
heating in the dental surgery. This process naturally gave
rise to considerable occupational mercury vapour exposure.
Copper amalgams were easier to retain in dental cavities
because of their higher initial plasticity than silver
amalgams. Contrary to silver amalgam fillings, copper
amalgam undergoes easily detectable dissolution with time.
This solubilization was, for some time, actually con-
sidered an advantage because of the associated bacteri-
cidal effects (SOS, 1987).
A source of mercury loss to the atmosphere is the
release of metallic mercury vapour during the cremation of
cadavers. Crematories are often located in densely popu-
lated areas and do not have high chimneys. All the mercury
from amalgam fillings vapourizes during the cremation, as
the temperature is above 800 °C. In a Swedish study, it
has been estimated that 170-180 kg of metallic mercury is
released annually from a total of about 50 000 cremations
per year (Mörner & Nilsson, 1986). The use of amalgam in
Sweden is estimated to be 5-7.5 tonnes per year (SOS,
1987), compared with 90-100 tonnes in the USA (Wolff et
al., 1983; Naleway et al., 1985). It is difficult to esti-
mate the global release of mercury vapour from cremation
due to uncertainties about dental status at the time of
death in relation to frequency of cremations.
3.5. Mercury-containing cream and soap
Mercury-containing cream and soap has for a long time
been used by dark-skinned people to obtain a lighter skin
tone, probably due to inhibition of pigment formation.
There are mainly two types of products distributed for
this purpose: skin-lightening creams and skin-lightening
soaps. This subject has recently been reviewed by Berlin
(personal communication to the IPCS by M. Berlin).
The distribution of the two products is now banned in
the European Economic Community, in North America, and in
many African states. Mercury-containing soap is, however,
manufactured in several European countries and sold as
germicidal soap to the Third World, and it has frequently
been found in European cities with a substantial black
population, such as London and Brussels. This implies that
the mercury-containing soap manufactured in Europe has
been re-imported illegally from African countries.
English community health authorities (Lambeth, 1988)
have identified several brands of soap containing mercury.
The soaps have been analysed and contain typically 1-3% of
mercuric iodide. There are also skin-lightening creams
containing ammoniated mercury from 1-5% (Marzulli & Brown,
1972) or 5-10% (Barr et al., 1973). Both the soap and the
cream are applied on the skin, allowed to dry on the skin
surface, and left overnight.
4. ENVIRONMENTAL TRANSPORT, DISTRIBUTION, AND TRANSFORMATION
There is a well-recognized global cycle for mercury,
whereby emitted mercury vapour is converted to soluble
forms (e.g., Hg++) and deposited by rain onto soil and
water. Mercury vapour has an atmospheric residence time of
between 0.4 and 3 years, whereas soluble forms have resi-
dence times of a few weeks. Transport in soil and water is
thus limited and deposition within a short distance is
highly likely.
The change in mercury speciation from inorganic to
methylated forms is the first step in the aquatic bioac-
cumulation process. Methylation can occur non-enzymati-
cally or through microbial action. Once methylmercury is
released, it enters the food chain by rapid diffusion and
tight binding to proteins. It attains its highest levels,
through food-chain biomagnification, in the tissues of
such predatory species as freshwater trout, pike, and bass
and marine tuna, swordfish, and shark. The ratio of the
methylmercury concentration in fish tissue to the concen-
tration of inorganic mercury in water is usually between
10 000 and 100 000 to one. Levels of selenium in the
water may affect the availability of mercury for uptake
into aquatic biota. Reports from Sweden and Canada point
to the likelihood of increased methylmercury concentration
in fish after the construction of artificial water reser-
voirs (WHO, 1990).
5. ENVIRONMENTAL LEVELS AND HUMAN EXPOSURE
The general population is primarily exposed to mercury
from dental amalgam and the diet. However, depending upon
the level of contamination, air and water can contribute
significantly to the daily intake of total mercury. In
most foodstuffs, mercury is usually in the inorganic form
and below the limit of detection (20 ng mercury/g fresh
weight). The exceptions are fish and fish products, which
are the main source of methylmercury in the diet. Levels
greater than 1.2 mg/kg are often found in the edible
portion of shark, swordfish, and Mediterranean tuna. Simi-
lar levels in pike, walleye, and bass taken from polluted
fresh water have been identified. Table 2 indicates the
average daily intake and retention of total mercury and
mercury compounds in the general population not occu-
pationally exposed to mercury.
The level of mercury in fish, even for humans consum-
ing small amounts (10-30 g of fish/day), can markedly
affect the intake of methylmercury and, thus, of total
mercury. The weekly consumption of 200 g of fish having
500 µg mercury/kg will result in the intake of 100 µg
mercury (predominantly methylmercury). This amount is one-
half of the tolerable recommended weekly intake (WHO,
1989).
The subject of human mercury dietary exposure has been
discussed in previous Environmental Health Criteria mono-
graphs (WHO, 1976, 1990). This section emphasizes human
exposure to inorganic mercury from dental amalgam and
skin-lightening creams and soaps among the general popu-
lation, and occupational exposure due to the use of
amalgam in dentistry. Industrial exposure was described in
detail in WHO (1976); more recent information is discussed
in section 9.
5.1. General population exposure
5.1.1. Exposure from dental amalgam
5.1.1.1 Human studies
The release of mercury vapour from dental amalgam
fillings has been known for a very long time (Stock,
1939). The next major contribution to this field was that
of Frykholm (1957). Using a radioactive mercury tracer, he
showed that the insertion of amalgam in both humans and
dogs resulted in significant concentrations of mercury in
urine and faeces. In humans, the concentration of urinary
mercury increased during a 5-day period following the
insertion of 4-5 small occlusal fillings. A new higher
peak occurred a couple of days after removal of these
fillings. Faecal elimination showed a similar pattern,
appearing on the second day after amalgam insertion.
Another maximum appeared 1-2 days after amalgam removal.
Frykholm (1957) also measured the concentration of mercury
in the oral cavity during amalgam placement in teeth.
Recently, concern over amalgam usage has been revived by
the publication of a number of experimental studies
showing that, among other elements, inorganic mercury is
released from amalgam in vitro (Brune, 1981; Brune & Evje,
1985). More importantly, mercury vapour released in the
mouth in vivo leads to an increased uptake of mercury in
body tissues (Gay et al., 1979; Svare et al., 1981;
Abraham et al., 1984; Ott et al., 1984; Patterson et al.,
1985; Vimy & Lorscheider, 1985a,b; Vimy et al., 1986;
Langworth et al., 1988; Nylander et al., 1987, 1989;
Berglund et al., 1988; Aronsson et al., 1989). Vimy &
Lorscheider (1985b) showed that the release rate of mer-
cury vapour increases dramatically when the amalgam is
stimulated by continuous chewing, reaching a plateau
within 10 min. After the cessation of chewing, it takes
approximately 90 min for the mercury release rate to
decline to the basal pre-chewing value (Fig. 1). A con-
firmatory study has recently been published by Aronsson et
al. (1989), who also made daily dose estimates.
Table 2. Estimated average daily intake and retention (µg/day) of
total mercury and mercury compounds in the general population not
occupationally exposed to mercurya
---------------------------------------------------------------------
Exposure Elemental Inorganic mercury Methylmercury
mercury vapour compounds
---------------------------------------------------------------------
Air 0.030 (0.024) 0.002 (0.001) 0.008 (0.0064)
Food
Fish 0 0.600 (0.042) 2.4 (2.3)
Non-fish 0 3.6 (0.25) 0
Drinking-water 0 0.050 (0.0035) 0
Dental amalgams 3.8-21 (3-17) 0 0
Total 3.9-21 (3.1-17) 4.3 (0.3) 2.41 (2.31)
---------------------------------------------------------------------
a From: Environmental Health Criteria 101: Methylmercury (WHO, 1990).
Values given are the estimated average daily intake; the figures
in parentheses represent the estimated amount retained in the body
of an adult.
Values are quoted to 2 significant figures.
Critical reviews have been made of published infor-
mation on mercury release and exposure from amalgam
(Enwonwu, 1987; Friberg & Nylander, 1987; Langan et al.,
1987; Mackert, 1987; Olsson & Bergman, 1987; Clarkson et
al., 1988a). From these reviews it can be concluded that
it is difficult to make accurate quantitative estimations
of the mercury release from amalgam and the uptake of
mercury by the human body. Problems include uncertainty
about analytical quality control, differences in sampling
methodology, breathing pattern, dilution with inhaled air,
and uncertainty about time since previous meals. Due to
these factors, some studies may have overestimated and
others underestimated the daily dose of mercury, while
others may have underestimated or overestimated the mer-
cury uptake.
Several studies have correlated the number of dental
amalgam fillings or amalgam surfaces with the mercury
content in brain and kidney tissue from human autopsy.
Subjects with no dental amalgam had a mean mercury level
of 6.7 ng/g (2.4-12.2) in the occipital cortex; whereas,
subjects with amalgams had a mean level of 12.3 ng/g
(4.8-28.7) (Friberg & Nylander, 1987; Nylander et al.,
1987). Amalgam-free subjects had a mean mercury level in
kidneys of 49 ng/g (21-105), whereas subjects with amalgam
fillings had a corresponding level of 433 ng/g (48-810).
In a similar investigation, Eggleston & Nylander (1987)
showed mean mercury levels of 6.7 ng/g (1.9-22.1) and 3.8
ng/g (1.4-7.1) in grey and white brain matter, respect-
ively, in subjects with no amalgam fillings. In subjects
with amalgam fillings, mercury levels were 15.2 ng/g
(3.0-121.4) and 11.2 ng/g (1.7-110.1) for grey and white
matter, respectively. In a more recent extensive study,
Schiele (1988) showed a mean brain occipital mercury con-
centration of 10 ng/g for 44 subjects with an average of
14 amalgam surfaces each. Kidneys from the same subjects
showed a sex difference in the mercury concentrations,
mean values being 484 ng/g for the 16 females and 263 for
the 28 males. Amalgam-free subjects were not included in
this study.
Using published experimental data (Svare et al., 1981;
Abraham et al., 1984; Patterson et al., 1985; Vimy &
Lorsheider, 1985b), the amalgam mercury release rate,
average daily mercury uptake, and its steady-state contri-
bution to blood, urine, brain, and kidney were estimated
by Clarkson et al. (1988a). These estimations gave brain,
kidney, and urine values that are similar to data reported
from human studies (brain and kidney autopsy samples:
Friberg et al., 1986; Nylander et al., 1987; Schiele,
1988; urine: Nilsson & Nilsson, 1986b; Olstad et al.,
1987; Langworth, 1987). A representative illustration of
the type of relationship found is given in Fig. 2. Esti-
mates of daily dosages of mercury attributed to amalgam
have also been reported by Mackert (1987) and Olsson &
Bergman (1987), although they are somewhat lower than
those of Clarkson et al. (1988a).
Snapp et al. (1989) studied the blood mercury level
before and 18 weeks after the removal of amalgam fillings.
After the removal, nine of the ten subjects examined
exhibited a statistically significant mean decrease of
1.13 ng (± 0.6) mercury/ml in the blood mercury level.
Recently, Molin et al. (1990) studied mercury concen-
trations in human plasma, erythrocytes, and urine before
and up to 12 months after removal of amalgam fillings and
replacements with gold alloy restorations. They noted an
initial increase in all recorded mercury levels after
amalgam removal. About three months thereafter, plasma and
erythrocyte levels decreased markedly. A continuous
reduction in urine mercury levels took place, reaching a
plateau of approximately 25% of the pre-removal mercury
level within 9 months.
It is important to note that, in the studies cited,
both the predicted mercury uptake from amalgam and the
observed accumulation of mercury in the body are average
values. It is also clear from the original reports that
substantial individual variations exist.
5.1.1.2 Animal experiments
Frykholm (1957), using radioactive mercury in amalgam,
studied the release and uptake of mercury in dogs and
monkeys. He concluded that the mercury exposure from
amalgam was essentially limited to the immediate placement
procedures. This is in contrast to more recent studies
that examined the disposition of radioactive mercury
released from amalgam restorations in sheep (Hahn et al.,
1989; Vimy et al., 1990a).
Hahn et al. (1989) demonstrated by whole-body image
scan that amalgam mercury could be readily visualized in
the kidney, liver, jawbone, and gastrointestinal tract
after only 29 days of chewing with amalgam. Vimy et al.
(1990a) demonstrated that the mercury levels in maternal
blood, fetal blood, and amniotic fluid reached a peak
within 48 h after amalgam placement and remained at that
level for the duration of the studies (140 days). Mercury
levels of 4 ng/g in maternal blood and amniotic fluid and
of 10 ng/g in fetal blood were found. The erythro-
cyte/plasma ratios of mercury from amalgam in both the
ewe and fetal lamb were less than unity. The maternal
urine mercury concentration ranged from 1-10 ng/g during a
16-day period. Approximately 7.7 mg of mercury could be
eliminated per day in the faeces.
All tissues examined displayed mercury accumulation.
By 29 days, kidney mercury levels rose to approximately
9000 ng/g, and these levels were maintained throughout the
duration of the study. A similar pattern was observed in
the liver, but the levels remained at approximately 1000
ng/g. The fetal kidney contained mercury levels of 10-14
ng/g, whereas fetal liver had levels of 100-130 ng/g.
The maternal brain (cerebrum, occipital lobe, and
thalamus) showed a mercury accumulation ranging from 3-13
ng/g. In the pituitary, thyroid, and adrenal glands, con-
centrations ranged from approximately 10-100 ng/g. In the
fetal cerebrum, occipital cortex, and thalamus the highest
levels were approximately 10 ng/g. The fetal pituitary
gland had mercury concentrations of more than 100 ng/g,
whereas the thyroid and adrenal glands contained less than
10 ng/g.
Milk obtained at lamb parturition or within several
days following birth (25-41 days after amalgam placement)
contained levels of mercury from dental amalgam that
reached as high as 60 ng/g.
Other recent reports indicate that both kidney func-
tion (Vimy et al., 1990b) and intestinal bacterial popu-
lation (Summers et al., 1990) may be affected when animals
are exposed to dental amalgam mercury.
5.1.2. Skin-lightening soaps and creams
Elemental mercury and soluble inorganic mercury com-
pounds can penetrate the human skin. Mercury-containing
skin-lightening soaps and creams are left on the skin
overnight. Therefore, the possibility of substantial mer-
cury exposure exists both via the skin and through inha-
lation. There are no empirical data showing the relative
importance of the different exposure routes, but the evi-
dence indicates that the total exposure to mercury is sub-
stantial from these sources. Barr et al. (1973) reported
that in a group of 60 African women using skin-lightening
creams (5-10% ammoniated mercury), the mean urinary mer-
cury excretion was 109 µg/litre (range: 0-220 µg per
litre). A subgroup of 26 women with a nephrotic syndrome
had a mean urinary mercury level of 150 µg/litre (range:
90-250 µg/litre). Marzulli & Brown (1972) reported uri-
nary mercury levels from 28 to 600 µg/litre among a group
of 6 women who had used skin-lightening cream containing
1-3% ammoniated mercury for two years.
Lauwerys et al. (1987) reported the case of a woman
who had recently given birth and who had used during preg-
nancy and lactation a soap containing 1% mercury as mer-
curic iodide and a mercury-containing cream. The urinary
mercury content of the mother was 784 µg/g creatinine
4 months after the birth at a time when she was still
using the soap and cream. Although no mercury-containing
cream or soap was used on her baby's skin and the lac-
tation period lasted only one month, the baby's blood (at
the age of three months) contained 19 µg/litre and the
urine 274 µg/g creatinine.
5.1.3. Mercury in paint
Mercury compounds are added to water-based latex
paints to inhibit the growth of bacteria and mould.
Several reports have highlighted that mercury vapour can
be released from the paint on interior house walls
(Hirschman et al., 1963; Jacobs & Goldwater, 1965; Foote,
1972; Sibbett et al., 1972).
A recent study by Agocs et al. (in press) compared
homes recently coated with a paint containing a median
concentration of 754 mg mercury/litre with homes not
coated with a mercury-containing paint to determine
whether the recent application of such a paint is associ-
ated with elevated concentrations of mercury in air and
urine. Air samples from the 19 homes of exposed people
contained a median level of 2 µg/m3 (range, undetectable
to 10 µg/m3), while concentrations of mercury in air
from 9 homes of unexposed people were below the detection
limit of 0.1 µg/m3 (p < 0.001). The median urine mercury
concentration was higher for the 65 exposed people
(8.4 µg/g creatinine; range, 2.5-118) than for the 28
unexposed people (1.9 µg/g creatinine; range, 0.04-7)
(p < 0.001).
5.2. Occupational exposure during manufacture, formulation,
and use
Occupational exposure to mercury in chloralkali plants
and in mercury mining was reviewed in WHO (1976). In more
recent studies, average urine mercury levels of 50-100 µg
per litre have been reported (see sections 9.1.2 and
9.2.2).
A NIOSH survey in 1983 of 84 workers in a thermometer
factory showed that five workers had urinary mercury
levels above 150 µg/g creatinine and three workers had
levels above 300 µg/g creatinine. Personal air sampling
showed exposure levels of 26-271 µg/m3 (Ehrenberg et
al., 1986). Other studies of instrument and thermometer
factories in the USA yielded similar results (Price &
Wisseman, 1977; Wallingford, 1982; Lee, 1984). In gold and
silver refineries in the USA, the mean urinary mercury
concentration was 108 µg/litre for four regularly exposed
workers (Handke & Pryor, 1981).
Recently, particular interest has focused on occu-
pational exposure to mercury in dentistry (see also
section 3.2). Several studies made during the period 1960-
1980 have reported average levels of mercury vapour in
dental clinics ranging between 20 and 30 µg/m3 air,
and certain clinics have been found to have levels of 150-
170 µg/m3 (Joselow et al., 1968; Gronka et al., 1970;
Buchwald, 1972; Schneider, 1974). Some of these studies
also reported the urine mercury levels of dental person-
nel. Joselow et al. (1968) found an average urinary
mercury concentration of 40 µg/litre among 50 dentists,
some values exceeding 100 µg/litre. These levels are
similar to the urinary mercury concentrations reported by
Gronka et al. (1970) and Buchwald (1972).
Kelman (1978) reported statistically significantly
higher urine mercury levels among dental assistants
(38 µg/litre) than among dentists (22 µg/litre). On the
other hand, Nixon et al. (1981) found only small differ-
ences between dentists and dental assistants. The average
environmental mercury exposure in 200 clinics studied was
11 µg/m3 (with a range from 0 to 82 µg/m3), while
the mean urine mercury concentration was 26 µg/litre
(2-149 µg/litre).
In a nationwide American study by Naleway et al.
(1985), the average mercury level in urine sampled between
1975 and 1983 from 4272 dentists was 14.2 µg/litre (SD
± 25.4 µg/litre; the frequency distribution did not re-
semble a normal distribution), the range being 0-556 µg
per litre. In 4.9% of the samples, levels were above
50 µg/litre, and above 100 µg/litre in 1.3% of
samples. The wide range of values was probably due to the
sampling techniques, methodological problems, and vari-
ations in occupational exposures to amalgam.
In a similar Norwegian study, Jokstad (1987) reported
that 2% of a group of 672 dentists had urine mercury
levels greater than 20 µg/litre. The highest recorded
value in this group was 50 µg/litre.
Recently Nilsson & Nilsson (1986a,b) reported a com-
paratively low mercury level (4 µg/m3) in the air of
private dental clinics. The median urine mercury concen-
tration was 6 µg/litre (range: 1-21 µg/litre) for den-
tists and 7 µg/litre (range: 1-70 µg/litre) for dental
assistants. In a Belgian study of dentists by Huberlant
et al. (1983), the mean urine mercury concentration was
also relatively low (11.5 µg/g creatinine).
Dentists and dental assistants may be momentarily
exposed to high local peaks of mercury vapour during
insertion, polishing, and removal of amalgam fillings,
especially if adequate protective measures are not taken
(Frykholm, 1957; Buchwald, 1972; Cooley & Barkmeier, 1978;
Reinhardt et al., 1983; Richards & Warren, 1985). Richards
& Warren (1985) reported mercury vapour concentrations
approaching 1000 µg/m3 in the breathing zone of dentists
not using coolants or adequate aspiration techniques
during operative procedures. The corresponding concen-
trations when proper measures were used were approxi-
matively ten times lower (110 µg/m3).
When Battistone et al. (1976) analysed the blood
mercury level of 1389 American dentists, the mean value
was 9.8 µg/litre (18 dentists having levels above 30 µg
per litre). In a study of 380 American dentists, Brady et
al. (1980) reported a mean concentration of 8.5 µg per
litre, 7.4% of the participants having blood mercury
levels greater than 15 µg/litre. These levels were found
to decrease within 16 h after termination of exposure.
This finding agrees with the documented short biological
half-time in blood for the majority of the mercury (see
section 6.5).
These studies suffered from variations in the sampling
techniques, the analytical techniques, and the occu-
pational exposure of the participants. Although the extent
of occupational exposure could be evaluated from mercury
concentrations found in critical organs, few data are
available in the literature. Kosta et al. (1975) reported
levels of mercury in the central nervous system and the
kidneys of deceased mercury miners several years after
cessation of exposure. Average levels of 700 µg/kg wet
weight of brain (SD ± 640 µg/kg) were, for example, re-
ported in six cases. In the same group plus an additional
miner, pituitary mercury levels were reported to be as
high as 27 100 µg/kg (SD ± 14 900 µg/kg). Non-exposed
controls showed mean brain levels of 4.2 µg per kg (SD
± 2.6 µg/kg, n = 5), mean pituitary levels of 40 µg per
kg (SD ± 26 µg/kg, n = 6), and mean kidney levels of
140 µg/kg (SD ± 160 µg/kg, n = 7) (see also sections
9.1.1 and 9.2.1).
A Swedish study of seven former dentists and one
dental nurse reported elevated concentrations of mercury
in the pituitary gland and occipital lobe cortex (Nylander
et al., 1989). Values of up to 4000 µg/kg wet weight
were observed in the pituitary gland, and of up to
300 µg/kg in the occipital lobe cortex. Two of the sub-
jects were 80 years old and had been retired for several
years. High mercury levels were also noted in the kidneys
and thyroid. In one subject, the thyroid concentration was
28 000 µg/kg despite several years retirement.
6. KINETICS AND METABOLISM
There are major differences in the kinetics and metab-
olism of the various mercury species. Metallic mercury is
rapidly oxidized to inorganic mercury compounds in the
body. However, its kinetics and membrane permeability are
different from those of mercuric mercury. Also methylmer-
cury can be converted to inorganic mercury in vivo (WHO,
1990). Thus, the ultimate fate of absorbed mercury com-
pounds will depend on their chemical transformation in the
body as well as the kinetics. The details of the kinetics
and metabolism of methylmercury have been described in WHO
(1990).
6.1. Absorption
6.1.1. Absorption by inhalation
Inhalation of mercury vapour is the most important
route of uptake for elemental mercury. Approximately 80%
of inhaled mercury vapour is retained. The retention
occurs almost entirely in the alveoli, where it is almost
100%. The retained amount is the same whether inhalation
takes place through the nose or the mouth (WHO, 1976;
Hursh et al., 1976).
The uptake of metallic mercury vapour from inspired
air into the blood depends on the dissolution of mercury
vapour in the blood as it passes through the pulmonary
circulation. The dissolved vapour is then very soon oxi-
dized to Hg++, partly in the red blood cells and partly
after diffusion into other tissues. This oxidation occurs
under the influence of the enzyme catalase. The oxidation,
and in consequence the absorption, of mercury vapour in
humans can be reduced considerably by alcohol or the
herbicide aminotriazole (WHO, 1976; Halbach & Clarkson,
1978; Magos et al., 1978; Hursh et al., 1980).
WHO (1976) concluded that information on pulmonary
retention of inorganic mercury compounds was lacking.
Deposition should follow the physical laws governing depo-
sition of aerosols in the respiratory system. Particulates
with a high probability of deposition in the upper respir-
atory tract should be cleared quickly. For particulates
deposited in the lower respiratory tract, a longer reten-
tion period would be expected, the length depending on
solubility, among other factors. In experiments on dogs,
approximately 45% of a radioactive mercury(II) oxide aero-
sol, with a median droplet diameter of 0.16 (± 0.06) µm,
was cleared in less than 24 h and the remainder with a
half-time of 33 days (Morrow et al., 1964). Radioactivity
was detected in blood as well as in urine. The concen-
tration in blood followed the curve of its disappearance
from the lungs. The in vivo solubility of the particles
was found to be of great importance for the clearance
during the slow phase. Recent evidence has shown that lung
macrophages are able to increase the solubility of only
slightly soluble metals (Lundborg et al., 1984; Marafante
et al., 1987) and that this is due to a low pH in the
phagolysosomes (Nilsen et al., 1988).
Although there are still no data to allow a quantitat-
ive evaluation of the absorption of different inorganic
mercury compounds, significant absorption must take place
directly from the lung and, probably, to some extent from
the gastrointestinal tract after mucociliary clearance of
non-absorbed mercury.
6.1.2. Absorption by ingestion
Liquid metallic mercury is poorly absorbed. Some data
indicate an absorption of less than 0.01% in rats. How-
ever, humans who accidently ingested several grams of
metallic mercury showed increased blood levels of mercury
(WHO, 1976). Metallic mercury has been incorporated into
tissues after accidental breakage of intestinal tubes,
containers, and thermometers. This has sometimes caused
local tissue reactions with or without signs of systemic
poisoning (Geller, 1976). The reason for the different
types of reactions is not known.
The absorption in humans of inorganic mercuric mercury
compounds from foods was estimated by WHO (1976) to be
about 7% on average and by Elinder et al. (1988) to be
less than 10% (probably about 5%). The data were mainly
obtained from tracer studies on human volunteers (Rahola
et al., 1973), who received single oral doses of protein-
bound inorganic mercuric mercury. Although individual
variation was considerable, the proportion of the dose
excreted in the faeces during the first 4-5 days was 75-92%.
Absorption in young children may be considerably
greater. Kostial et al. (1978, 1983) observed an average
absorption in newborn rats of 38% six days after an oral
dose of mercuric chloride. The absorption in older animals
was only about 1%. As breast milk may contain significant
amounts of inorganic as well as organic mercury, this
route of exposure should not be overlooked (section 6.4).
The low solubility of mercurous chloride limits absorp-
tion. However, after prolonged intake the accumulation of
mercury in tissues, urinary mercury excretion, and adverse
effects indicate that some absorption takes place.
6.1.3. Absorption through skin
Little information was available on skin absorption
when WHO (1976) was published, although some animal exper-
iments revealed a certain degree of skin penetration (a
few per cent of an aqueous solution of mercuric salts
during the first hours of skin application) (Friberg et
al., 1961; Skog & Wahlberg, 1964; Wahlberg, 1965). Recent
studies on human volunteers (Hursh et al., 1989) indicate
that uptake via the skin of metallic mercury vapour is
only about 1% of uptake by inhalation. However, it is
obvious that the use of skin-lightening creams containing
inorganic mercury salts causes substantial absorption and
accumulation into the body (section 5.1.2), although there
is no information on how much of the mercury is absorbed
through the skin and how much is absorbed via other
routes.
6.1.4. Absorption by axonal transport
Arvidson (1987) reported an accumulation of mercury
from a tracer dose of 203HgCl2 in the hypoglossal nuclei
of the brain stem of rats after a single injection into
the tongue. A similar accumulation was not seen in con-
trols after a similar injection into the gluteus maximus
muscle. The author concluded that the results provided
evidence of retrograde axonal transport of mercury in the
hypoglossal nerve.
6.2. Distribution
From studies on animals and humans (WHO, 1976; Khayat
& Dencker, 1983a, 1984; see also sections 8 and 9), it is
known that mercury has an affinity for ectodermal and
endodermal epithelial cells and glands. It accumulates in,
for instance, the thyroid, pituitary, brain, kidney,
liver, pancreas, testes, ovaries, and prostate. Within
the organs the distribution is not uniform. This explains
why biological half-times may differ not only between
organs but also within an organ. The kidney is the chief
depository of mercury after the administration of elemen-
tal mercury vapour or inorganic salts. Based on animal
data, 50-90% of the body burden is found in the kidneys.
Significant amounts were transported to the brain after
exposure of mice and monkeys to elemental mercury vapour.
The brain mercury levels were ten times higher than after
equal doses of mercuric mercury given intravenously
(Berlin & Johansson, 1964; Berlin et al., 1969; WHO,
1976). In rats given daily subcutaneous doses of mercuric
chloride for six weeks, only 0.01% of the total dose of
mercury was found in the brain, while about 3% of the dose
was retained in the kidneys (Friberg, 1956).
The red cell to plasma ratio in humans was approxi-
mately 1.0 after exposure to Hg0 vapour, but was 0.4
after exposure to inorganic mercury salts (WHO, 1976). The
ratio may vary, however. Suzuki et al. (1976) observed a
red cell to plasma ratio of about 1.5-2 for workers
exposed only to mercury vapour, while the corresponding
ratio for 6 chloralkali workers (where the exposure may
have been to both vapour and inorganic salts) averaged
only 0.02. The reason for this extremely low ratio is
unknown. In a report by Cherian et al. (1978), a ratio of
about 2 was observed during the first few days after
exposure of volunteers to metallic mercury vapour.
Jugo (1976) compared the retention of mercuric chlor-
ide after a single injection in adult and 2-week-old suck-
ling rats. The whole-body retention 6 days after treatment
was significantly higher in the suckling animals, and the
accumulation of mercury was 13- and 19-fold higher in the
brain and liver, respectively, compared to adult rats. On
the other hand, the mercury concentrations in the kidneys
were markedly higher in the adult group.
In two pregnant women who had been accidentally
exposed to metallic mercury vapour, the concentration of
mercury in the infant blood was similar to that in the
maternal blood at the time of delivery (Clarkson &
Kilpper, 1978). There are no other data on the transfer of
inhaled mercury vapour to the fetus in humans.
Based on studies in rodents, elemental mercury vapour
easily penetrates the placental barrier and, after oxi-
dation, accumulates in the fetal tissue. Only a fraction
of divalent mercury enters the fetus, but it can accumu-
late in the placenta. Clarkson et al. (1972) found that
mercury levels in the fetuses of rats exposed to mercury
vapour were 10-40 times higher than in animals exposed to
equivalent doses of mercuric chloride. Differences in the
penetration of the placental barrier have been confirmed
in mice by Khayat & Dencker (1982), who found a 4-fold
higher fetal mercury concentration after exposure to met-
allic mercury vapour than after exposure to mercuric
chloride. The uptake of mercury vapour increased with
gestational age. Only traces of radioactive mercury were
found in embryos at 8 and 10 days of gestation. A distinct
accumulation of mercury was seen in the fetal tissue from
day 12 of gestation with a pronounced uptake in the fetal
liver and heart. The mercury concentration in the CNS was
rather low in early and mid gestation but increased just
prior to birth (Ogata & Meguro, 1986).
Yoshida et al. (1986, 1987) studied the uptake and
distribution of mercury in the fetus of guinea-pigs during
late gestation after repeated exposure to 200-300 µg mer-
cury vapour/m3 2 h/day and after a single exposure for
150 min to 8-11 mg/m3. Mercury concentrations in fetal
brain, lungs, heart, kidneys, and blood were much lower
than those in maternal tissues, the concentrations dif-
fering by a factor of about 5 in the brain and a factor of
up to 100 in the kidneys. Mercury concentrations in fetal
liver were up to two times higher than those found in
maternal liver. In the fetal liver, more than 50% of the
mercury was bound to a metallothionein-like protein with a
relative molecular mass of about 10 000 to 12 000. The
bulk of the eluted mercury in the maternal liver was
associated with a protein of high relative molecular mass.
The authors suggested that the fetal metallothionein-like
protein plays a role in preventing further distribution of
mercury from the liver after in utero exposure to mercury
vapour.
Mercury distribution in the neonate differs from that
in the fetus (Yoshida et al., 1989). A significantly
increased level was found in kidney, lung, and brain in
neonate guinea-pigs, compared with fetuses, and there was
a progressive decrease in liver concentration, with dimin-
ishing hepatic metallothionein levels, in the neonates.
These results suggest a redistribution of mercury to other
tissues in the neonate.
The oxidation of elemental mercury vapour in the body
(section 6.1.1) can be reduced considerably (to about 50%
of normal values) by moderate amounts of alcohol. In an
in vivo study, the uptake of labelled mercury into human
red cells was reduced by almost a factor of ten by etha-
nol, while there was an increase in liver mercury concen-
trations (Hursh et al., 1980). Observations on rats, mice,
and monkeys confirm these results (Khayat & Dencker,
1983a,b, 1984). They also show a marked decrease in mer-
cury concentrations in several organs, including the
brain. However, somewhat higher concentrations of mercury
were observed in the brain and liver of pregnant mice with
a congenital catalase deficiency that were exposed for 1 h
to metallic mercury vapour during day 18 of gestation
(Ogata & Meguro, 1986). The blood mercury concentration
in the catalase-deficient mice was only about half of that
in the control mice. The uptake in the fetus was 2% of the
dose compared to 1.2% for the controls.
Lower mercury levels have been observed in the brain
tissue of humans classified as chronic alcohol abusers
than in controls (Fig. 3).
6.3. Metabolic transformation
Several forms of metabolic transformation occur:
* oxidation of metallic mercury vapour to divalent mer-
cury;
* reduction of divalent mercury to metallic mercury;
* methylation of inorganic mercury;
* conversion of methylmercury to divalent inorganic mer-
cury.
The oxidation of metallic mercury vapour to divalent
ionic mercury (section 6.1.1) takes place very soon after
absorption, but some elemental mercury remains dissolved
in the blood long enough (a few minutes) for it to be
carried to the blood-brain barrier and the placenta (WHO,
1976). Recent in vitro studies on the oxidation of mercury
by the blood (Hursh et al., 1988) indicate that because of
the short transit time from the lung to the brain almost
all the mercury vapour (97%) arrives at the brain unoxi-
dized. Its lipid solubility and high diffusibility allow
rapid transit across these barriers. Oxidation of the
mercury vapour in brain and fetal tissues converts it to
the ionic form, which is much less likely to cross the
blood-brain and placental barriers. Thus, oxidation in
these tissues serves as a trap to hold the mercury and
leads to accumulation in brain and fetal tissues (WHO,
1976).
The reduction of divalent mercury to Hg0 has been
demonstrated both in animals (mice and rats) and humans
(WHO, 1976; Dunn et al., 1978, 1981a,b; Sugata & Clarkson,
1979). A small amount of exhaled mercury vapour is the
result of this reduction. It is increased in catalase-
deficient mice (Ogata et al., 1987) and by alcohol (both
in vitro and in vivo ) in both mice and humans (Dunn et.
al., 1981a,b). The increased exhalation of mercury vapour
in the latter case may be explained by assuming that the
oxidation by catalase is less than normal.
It was stated in WHO (1976) that there is no evidence
in the literature for the synthesis of organomercury com-
pounds in human or mammalian tissues. Minor methylation
may occur in vitro by intestinal or oral bacteria (Rowland
et al., 1975; Heintze et al., 1983). A slight increase in
the concentration of methylmercury in blood and/or urine
has been reported among dentists and workers in the
chloralkali industry (Cross et al., 1978; Pan et al.,
1980; Aitio et al., 1983). These data cannot be taken as
evidence of methylation, however, due to lack of analyti-
cal quality control and possible confounding by exposure
to methylmercury. Chang et al. (1987) did not observe any
methylation in a study of dentists.
The conversion of methylmercury to inorganic mercury
is considered a key step in the process of excretion of
mercury after exposure to methylmercury (WHO, 1990). If
the intact molecule of an organomercurial in an organ is
more rapidly excreted than inorganic mercury, biotrans-
formation will decrease the overall excretion rate, and
the ratio of inorganic to organic mercury in that particu-
lar organ will increase with time. The fraction of total
mercury present as Hg++ will depend on the duration of
exposure to methylmercury and/or the time elapsed since
cessation of exposure. Even if the demethylation rate is
very slow, this process may in the long run give rise to
considerable accumulation of inorganic mercury. The ratio
of methylmercury to inorganic mercury depends on the rate
of demethylation and the clearance half-times of methyl-
mercury and inorganic mercury.
After short-term exposure of experimental animals to
methylmercury the kidneys usually contain the highest
fraction of Hg++ in relation to total mercury, while the
relative concentration in the brain is low (WHO, 1976).
In studies on squirrel monkeys (Berlin et al., 1975), the
short-term biotransformation to inorganic mercury was as
follows: of the total mercury, about 20% was inorganic in
the liver; 50% in the kidney; 30%-85% in the bile; and
less than 5% in the brain.
More recent data from long-term studies on monkeys
show a different pattern. Mottet & Burbacher (1988) sum-
marized a long series of studies on the metabolism and
toxicity of methylmercury in monkeys (Macaca fascicu-
laris). The monkeys had been orally exposed to high levels
of methylmercury for a period of years and sacrificed
during the ongoing exposure. At the end of the exposure
period, 10-33% of the mercury in the brain was present in
the inorganic form (Lind et al., 1988). In monkeys that
had been without mercury exposure for 6 months to almost
two years after the same treatment, the relative concen-
tration of inorganic mercury was much higher, i.e. about
90%. Exact half-times for the different compounds could
not be established in the absence of data on the concen-
trations of inorganic and organic mercury in the brain at
different time intervals during the accumulation and
clearance phases. Recent data by Rice (1989) also demon-
strate demethylation in the brain. Female monkeys (Macaca
fascicularis) were dosed for at least 1.7 years with mer-
cury as methylmercury chloride (10-50 µg/kg per day).
After dosing ceased, the blood mercury half-time was about
14 days. Approximately 230 days after cessation of dosing,
the monkeys were sacrificed and brain total mercury levels
determined. These levels were considered to be at least
three orders of magnitude higher than those predicted by
assuming the half-time in brain to be the same as that in
blood. The author considered the most likely explanation
to be demethylation of methylmercury and subsequent bind-
ing of inorganic mercury to tissue.
Similar results were recently reported by Hansen et
al. (1989) who fed fish contaminated with methylmercury to
one Alsatian dog for 7 years. The dog was examined after
its death at the age of 12 years, 4 years after the
exposure to methylmercury had ceased. Two dogs of the same
age and breed served as controls. In the CNS, the mercury
was fairly uniformly distributed and 93% was in the inor-
ganic state, whereas the skeletal muscles contained
approximately 30% inorganic mercury. The authors concluded
that the results demonstrated time-dependent demethylation
and suggested a variation in the rate from one type of
tissue to another. High levels of mercury were demon-
strated by a histochemical method in the liver, thyroid
gland, and kidney, whereas practically no mercury was
found in any of the organs examined in the control dogs.
The distribution of inorganic mercury was determined by a
histochemical method for locating mercury in tissue
sections. Total mercury was analysed by flameless atomic
absorption and organic mercury by GC.
A considerable fraction of the mercury in human
brains is reported to be in the form of inorganic mercury.
Kitamura et al. (1976) analysed autopsy material from 20
Japanese subjects for total mercury using flameless atomic
absorption and for methylmercury using GC. The median
concentration of total mercury in the cerebrum was 0.097
mg/kg wet weight and of methylmercury 0.012 mg/kg wet
weight. The values for the cerebellum were similar. No
analytical quality control data were reported.
In a Swedish autopsy study covering six cases (Friberg
et al., 1986; Nylander et al., 1987), about 80% of the
mercury in the occipital lobe cortex was inorganic. The
concentration of inorganic mercury varied between 3 and
22 µg/kg wet weight. Both total mercury and inorganic
mercury were determined by the method of Magos (Magos,
1971; Magos & Clarkson, 1972). For quality control pur-
poses total mercury was also analysed by neutron acti-
vation analysis. In this study, however, the concen-
trations of mercury in the brain were considerably lower
than in the Japanese study. As has been discussed in sec-
tion 5.1.1, an association between the number of amalgam
fillings and total mercury concentration in the occipital
lobe has been found. Exposure to inorganic mercury from
dental fillings could explain the high proportion of inor-
ganic mercury in the Swedish study but not in the Japanese
study, as it seems reasonable to assume that the mercury
exposure from amalgam should be approximately the same in
the two countries. The exposure to methylmercury could,
however, easily differ considerably.
Takizawa (1986) reported the total mercury and methyl-
mercury brain concentrations in about 30 humans who had
died from 20 days to 18 years after the onset of symptoms
of methylmercury poisoning. The total mercury content was
measured by flameless atomic absorption spectrophotometry,
while methylmercury was analysed by electron capture GLC
(Minagawa et al., 1979; Takizawa, 1986). The total mercury
content in "acute" cases (autopsy < 100 days after onset
of symptoms) was 8.8-21.4 mg/kg and the concentration of
methylmercury was 1.85-8.42 mg mercury/kg. The concen-
trations for the "chronic" cases were 0.35-5.29 mg/kg
for total mercury and 0.31-1.02 mg mercury/g for methyl-
mercury. On average, only 28% of the mercury was present
as methylmercury in the acute cases and 17% in the chronic
cases. Takizawa (1986) also presented data for residents
near Minamata Bay and for a non-polluted area. The best
estimate from these data is that only 16% and 12%,
respectively, of the total mercury was present as methyl-
mercury. Unfortunately, in these reports quality control
data were not presented. The authors measured total mer-
cury and methylmercury and assumed that the difference
between these analyses was due to inorganic mercury. It
could in principle, in whole or in part, also have been
methylmercury that was not extracted in the gas chromato-
graphic procedure. Ideally, analyses should be carried out
using, for instance, the method by Magos (1971), which
measures total mercury and inorganic mercury.
The tissues in the studies by Takizawa (1986) were
stored for long periods after fixation with a 10% neutral
formalin solution. Miyama & Suzuki (1971) found that the
ratio of inorganic to total mercury in the cerebral cortex
increased from about 35% (tissues stored frozen) to about
50% after storage in 10% formalin for one year. However,
there was no loss of inorganic mercury. Eto et al. (1988)
compared results from a small number of analyses of
formalin-fixed tissues with results from analyses of
frozen tissues. There was no systematic loss related to
storage in formaldehyde.
The concentrations of inorganic mercury in the brain,
reported in overt cases of methylmercury poisoning, are
very high, similar to those observed after toxic exposure
to metallic mercury vapour. Whether or not an accumulation
of inorganic mercury actually contributed to the toxic
effects is not known, but seems unlikely. Even assuming
no analytical problems, it should be borne in mind that
the methylmercury poisoning usually occurred after rela-
tively short exposure to methylmercury when no significant
biotransformation should yet have taken place. However, a
comparison of the toxicology of methylmercury with that of
ethylmercury, which decomposes significantly more quickly,
indicated that cerebellar damage could not be related to
inorganic mercury. The higher concentration of inorganic
mercury in the brain of ethylmercury-treated rats, com-
pared with methylmercury-treated rats, was associated with
less cerebellar damage (Magos et al., 1985). It is more
difficult to evaluate the possible long-term effects of
inorganic mercury, which slowly accumulates in the brain.
The distribution of ionic mercury in the brain will
depend on whether Hg++ enters the brain in the ionic form
or as a result of in situ biotransformation following
penetration of the brain barrier by elemental mercury or
methylmercury. The toxicological aspects of such possible
differences in distribution are not known.
6.4. Elimination and excretion
A small portion of absorbed inorganic mercury is ex-
haled as metallic mercury vapour, formed by the reduction
of Hg++ in the tissues (Dunn et al., 1978), but urine and
faeces are the principal routes of elimination (WHO,
1976). The urinary route dominates when exposure is high.
After exposure to metallic mercury vapour, a small frac-
tion of the mercury in the urine may be present as elemen-
tal mercury (Stopford et al., 1978; Yoshida & Yamamura,
1982). One form of depletion is the transfer of maternal
mercury to the fetal unit. Thus, inorganic mercury was
detected in the amniotic fluid in all but two out of 57
Japanese pregnant women, while organic mercury was found
in only 30 women (Suzuki et al., 1977). In a study by
Skerfving (1988), it was reported that the concentrations
of total mercury in breast milk and in the blood plasma of
breast-fed infants were similar to those in the maternal
plasma of Swedish fishermen's wives. Although the women
were exposed to methylmercury, 80% of mercury excreted in
breast milk was in the inorganic form. No formal analyti-
cal quality control procedures were applied in the studies
where mercury was speciated.
6.5. Retention and turnover
6.5.1. Biological half-time
Only very limited data were available on the biologi-
cal half-time of inorganic mercury when WHO (1976) was
published. Studies on a small number of volunteers had
shown that the elimination of mercury, after a single
exposure to metallic mercury vapour, followed a single
exponential process with an average half-time of 58 days
during the first few months after the exposure. Similar
data were available from studies involving oral exposure
to mercuric mercury. It was pointed out that there had
been a few reports of high brain mercury concentrations
in workers several years after cessation of exposure to
mercury vapour. This indicated that the half-time in
brain is longer than that in other organs, although no
quantitative estimations were made.
As a result of tracer studies on human volunteers
(Nakaaki et al., 1975, 1978; Cherian et al., 1978; Newton
& Fry, 1978; Hursh et al., 1980) and animals (Berlin et
al., 1975), more data are now available on the kinetics
during the first few months after exposure. The elimin-
ation of inorganic mercury follows a complicated pattern
with biological half-times that differ according to the
tissue and the time after exposure. The best estimate is
that after short-term exposure to mercury vapour, the
first phase of elimination from blood has a half-time of
approximately 2-4 days and accounts for about 90% of the
mercury. This is followed by a second phase with a half-
time of 15-30 days.
In tracer studies on nine human volunteers (Hursh et
al., 1976, 1980; Clarkson et al., 1988a) the half-time for
most of the mercury in the brain was 19 (± 1.7) days
during the first 35 to 45 days. Newton & Fry (1978) found
half-times of 23 and 26 days in the head of two subjects
accidentally exposed to radioactive mercuric oxide. In a
study by Berlin et al. (1975), a steady state was not
reached in the brains of squirrel monkeys exposed for two
months to mercury vapour. In one study on monkeys (Macaca
fascicularis) (section 6.3) lasting several years where
inorganic mercury accumulated in the brain (probably as a
result of demethylation of methylmercury), there was still
considerable inorganic mercury in the brain 1-2 years
after cessation of exposure (Lind et al., 1988). These
results indicate a very long half-time for a fraction of
the inorganic mercury in the brain. This is in accordance
with data from deceased miners and dentists (section
5.2).
The half-time in the kidneys for inorganic mercury in
the studies by Hursh et al. (1976, 1980) was 64 days,
about the same as that for the body as a whole. As in the
case of the brain, a fraction of the mercury probably has
a long biological half-time (section 5.2).
A few attempts to perform a quantitative evaluation of
the half-time for inorganic mercury have been made using
multicompartment models (Sugita, 1978; Bernard & Purdue,
1984). According to the recommendation of ICRP (1980), the
four-compartment model of Bernard & Purdue (1984) included
one compartment with a half-time of 27 years. As the basic
assumptions are uncertain, the models are uncertain, but
may be of value for a possible "worst case" estimation
of the retention of inorganic mercury in the brain. The
model of Bernard & Purdue (1984) has been used by Vimy et
al. (1986) for calculating mercury accumulation in the
brain from amalgam fillings (section 5.1). The form of
mercury that is responsible for the long biological half-
time may be biochemically inactive mercury selenide.
6.5.2. Reference or normal values in indicator media
A considerable amount of information is given in
Environmental Health Criteria 101: Methylmercury (WHO,
1990). The mean concentration of total mercury in whole
blood (in the absence of consumption of fish with high
concentrations of methylmercury) is probably of the order
of 5-10 µg/litre, and in hair about 1-2 mg/kg. The aver-
age mercury concentration in urine is about 4 µg per
litre and in the placenta about 10 mg/kg wet weight,
although the individual variation is substantial. One
source of the variation in urine levels seems to be
exposure from dental amalgam (Fig. 4), while for blood and
hair levels fish consumption is the major source of ex-
posure. Increased hair levels may also be due to external
contamination.
There are at present no suitable indicator media that
will reflect concentrations of inorganic mercury in the
critical organs, the brain or kidney, under different
exposure situations. This is to be expected in view of the
complicated pattern of metabolism for different mercury
compounds. One important consequence is that concen-
trations of mercury in urine or blood may be low quite
soon after exposure has ceased, despite the fact that
concentrations in the critical organs may still be high.
There is some information, obtained from subjects not
occupationally exposed and with only a moderate fish
consumption, on the relationship between exposure to
metallic mercury vapour and concentrations of mercury in
urine and brain tissue. This relationship (section 5.1)
indicates that ongoing long-term exposure to elemental
mercury vapour, leading to a mercury absorption of
5-10 µg/day, will result in a mercury excretion in urine
of about 5 µg/litre and average mercury concentrations
in the occipital lobe cortex and kidney of approximately
10 µg/kg and 500 µg/kg, respectively.
The distribution between blood and hair is well known
for different exposure levels of methylmercury, which
forms the basis for the use of hair as an indicator media
for this compound. There is no corresponding information
for inorganic mercury. When high levels of total mercury
in hair have been reported, for instance, among dentists
exposed to metallic mercury vapour (see e.g. Sinclair et
al., 1980; Pritchard et al., 1982; Sikorski et al., 1987),
it was not known how much was due to external contami-
nation. In a report on biological monitoring of toxic
metals, Elinder et al. (1988) concluded that hair is not a
suitable indicator medium for monitoring exposure to inor-
ganic mercury.
There is good epidemiological evidence from occu-
pational exposure that, on a group basis, recent exposure
is reflected in the mercury levels in blood and urine.
When exposure is low (e.g., from amalgam), it is difficult
to find an association between exposure levels and blood
concentrations due to confounding exposures to methylmer-
cury in fish. A way to overcome the problem may be to
analyse mercury in plasma or speciate the analysis for
inorganic mercury (Elinder et al., 1988). The problem of
confounding exposures is not so important when analysing
urine, as only a very small fraction of absorbed methyl-
mercury is excreted in urine.
Data amassed by Smith et al. (1970) from the chlorine
industry were used by WHO (1976) to evaluate the relation-
ship between concentrations of metallic mercury vapour in
air and concentrations of mercury in blood and urine.
Long-term time-weighted occupational exposure to an aver-
age air mercury concentration of 50 µg/m3 was considered
to be associated, on a group basis, with blood mercury
levels of approximately 35 µg/litre, and with urinary
concentrations of 150 µg/litre. The ratio of urine to air
concentrations was re-evaluated by WHO (1980) to be closer
to 2.0-2.5 instead of 3.0. The mercury concentrations in
air were measured with static samplers. Results from a
number of more recent studies have been reported where
both static samplers and personal samplers have been used
(Ishihara et al., 1977; Lindstedt et al., 1979; Müller et
al., 1980; Mattiussi et al., 1982; Roels et al., 1987).
Where personal samplers have been used, the ratio between
urinary mercury (µg/litre or per g creatinine) and
mercury in air (µg/m3) has as a rule been 1-2. When
blood values were reported they were either similar to
those given in WHO (1976) or somewhat lower.
In the study by Roels et al. (1987), personal monitor-
ing was used, detailed quality control procedures were
implemented and reported, and the examined subjects had
been exposed to defined concentrations for at least one
year. A good relationship could be established between the
daily time-weighted exposure to mercury vapour and the
daily level of mercury in blood and urine (Fig. 5A and B).
Urinary levels of about 50 µg/g creatinine were seen after
occupational exposure to about 40 µg/m3 of air. Such an
exposure would correspond to about 17 µg/litre of blood.
Several studies have reported a correlation between
mercury in blood and urine. The results vary considerably
and it is not known whether the ratio between concen-
trations in urine and blood is constant at different
exposure levels. At low exposure levels the possibilities
of a significant confounding effect on blood levels should
always be borne in mind.
On the basis of studies by Smith et al. (1970) and
Lindstedt et al. (1979), Skerfving & Berlin (1985)
suggested that a urine mercury level of 50 µg/g creati-
nine is associated with a blood mercury level of 20 µg
per litre. Roels et al. (1987) reported a regression
equation, where a urine mercury level of 50 µg/g creati-
nine leads to a blood mercury level of 16 µg/litre.
7. EFFECTS ON ORGANISMS IN THE ENVIRONMENT
This chapter is extracted from the summary of Environ-
mental Health Criteria 86: Mercury - Environmental Aspects
(WHO, 1989).
7.1. Uptake, elimination, and accumulation in organisms
Mercuric salts, and, to a much greater extent, organic
mercury, are readily taken up by organisms in water.
Aquatic invertebrates, and most particularly aquatic
insects, accumulate mercury to high concentrations. Fish
also take up the metal and retain it in tissues, princi-
pally as methylmercury, although most of the environmental
mercury to which they are exposed is inorganic. The source
of the methylation is uncertain, but there is strong indi-
cation that bacterial action leads to methylation in
aquatic systems. Environmental levels of methylmercury
depend upon the balance between bacterial methylation and
demethylation. The indications are that methylmercury in
fish arises from this bacterial methylation of inorganic
mercury, either in the environment or in bacteria associ-
ated with fish gills, surface, or gut. There is little
indication that fish themselves either methylate or
demethylate mercury. Elimination of methylmercury is slow
from fish (with half times in the order of months or
years) and from other aquatic organisms. Loss of inorganic
mercury is more rapid and so most of the mercury in fish
is retained in the form of methylmercury. Terrestrial
organisms are also contaminated by mercury, with birds
being the best studied. Sea birds and those feeding in
estuaries are most contaminated. The form of retained
mercury in birds is more variable and depends on species,
organ, and geographical site.
7.2. Toxicity to microorganisms
The metal is toxic to microorganisms. Inorganic mer-
cury has been reported to have effects at concentrations
of the metal in the culture medium of 5 µg/litre, and
organomercury compounds at concentrations at least 10
times lower tha