Extremely low frequency (ELF) fields (EHC 35, 1984)

INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY

ENVIRONMENTAL HEALTH CRITERIA 35

EXTREMELY LOW FREQUENCY (ELF) FIELDS

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, World Health Organization, and the
International Radiation Protection Association

World Health Orgnization
Geneva, 1984

Reprinted 1992

ISBN 92 4 154095 8

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 1984

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 EXTREMELY LOW FREQUENCY (ELF) FIELDS

1. SUMMARY AND RECOMMENDATIONS

1.1. Purpose and scope
1.2. Sources of exposure
1.3. Clinical applications
1.4. Field measurement and dosimetry
1.5. Characteristics of biophysical interactions
1.6. In vitro studies
1.7. Experimental animal studies
1.8. Effects on man
1.9. Exposure standards
1.10. Conclusions and recommendations

2. PHYSICAL CHARACTERISTICS, MEASUREMENT, AND DOSIMETRY

2.1. Quantities and units
2.2. Computational methods and measurements of ELF
electric fields
2.3. Field polarization and homogeneity
2.4. Energy carried by the field
2.5. Determination of ELF field exposure
2.6. The physical interaction of man and laboratory
animals with electric fields
2.6.1. Surface fields and internal current
density
2.6.2. Capacitive coupling of the electric
field to man and laboratory animals
2.6.3. Shock currents
2.7. Dosimetry and scaling between laboratory animals
and man
2.8. Magnetic induction of electric fields

3. NATURAL BACKGROUND AND MAN-MADE ELF FIELDS

3.1. Natural electric fields
3.2. Natural magnetic fields
3.3. Man-made sources of ELF
3.3.1. High-voltage transmission lines
3.3.2. Electric fields near transmission lines and substations
3.3.3. Magnetic fields near transmission lines
3.3.4. Man-made ELF fields in the home, workplace,
and public premises
3.4. Corona and noise effects of transmission lines
3.5. Electric shock
3.6. Interference of ELF fields with implanted
cardiac pacemakers

4. MECHANISMS OF INTERACTION

4.1. Biophysical mechanisms of electric field
interactions
4.2. Biophysical mechanisms of magnetic field
interactions

5. BIOLOGICAL EFFECTS IN CELLS AND ANIMALS

5.1. Cellular and membrane studies
5.2. Neurophysiological studies in animals
and animals tissues
5.3. Behavioural studies
5.4. Sensory phenomena
5.5. Effects on the haematopoietic system in animals
5.6. Cardiovascular effects
5.7. Effects on endocrinology and blood chemistry
5.8. Effects on the immune system
5.9. Growth and development studies
5.10. Reproduction and fertility
5.11. Mutagenesis
5.12. Circadian rhythms in animals
5.13. Bone growth and repair
5.14. The problems of extrapolating animal exposure
data to human beings

6. HUMAN STUDIES

6.1. Sources of information
6.2. Study design
6.3. Health status of occupationally-exposed
human beings
6.4. Studies on the general population
6.4.1. Studies on inhabitants of areas in the
vicinity of HV-lines
6.5. Studies on human volunteers
6.6. Summary

7. HEALTH RISK EVALUATION

8. STANDARDS AND THEIR RATIONALES

9. PROTECTIVE MEASURES

9.1. Goals
9.2. Groups to be protected
9.3. Protection rationale

GLOSSARY OF TERMS USED IN THE DOCUMENT

REFERENCES

APPENDIX I

APPENDIX I REFERENCES

NOTE TO READERS OF THE CRITERIA DOCUMENTS

While every effort has been made to present information in
the criteria documents as accurately as possible without
unduly delaying their publication, mistakes might have
occurred and are likely to occur in the future. In the
interest of all users of the environmental health criteria
documents, readers are kindly requested to communicate any
errors found 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.

In addition, experts in any particular field dealt with in
the criteria documents are kindly requested to make available
to the WHO Secretariat any important published information
that may have inadvertently been omitted and which may change
the evaluation of health risks from exposure to the
environmental agent under examination, so that the information
may be considered in the event of updating and re-evaluation
of the conclusions contained in the criteria documents.

WHO/IRPA TASK GROUP ON EXTREMELY LOW FREQUENCY (ELF) FIELDS

Members

Dr J. Bonnell, Central Electricity Generating Board, London,
England

Dr B. Bosnjakovic, Ministry of Housing, Physical Planning, and
Environment, Rijswijk, The Netherlands^a

Dr J. Cabanes, Medical Committee, Electricité de France - Gaz
de France, Paris, France

Dr M. Grandolfo, Laboratory of Radiation, Institute of Public
Health, Rome, Italy

Dr B. Knave, Research Department, National Board of
Occupational Safety and Health, Solna, Sweden

Dr J. Kupfer, Occupational Hygiene Standardization, Central
Institute of Occupational Medicine, Berlin, German
Democratic Republic (Vice-Chairman)

Dr R. Phillips, Biology Department, Pacific Northwest
Laboratory, Richland, Washington, USA

Dr A. Portela, Institute of Biophysical Research, National
Council of Scientific and Technical Research (CONICET),
Buenos Aires, Argentin^a

Dr M. Repacholi, Royal Adelaide Hospital, Adelaide, South
Australia (Chairman)^a

Dr A. Sheppard, J.L. Pettis Memorial Hospital, Loma Linda,
California, USA (Rapporteur)

IRPA Secretariat

Mrs A. Duchêne, Commissariat à l'Energie Atomique, Déparement
de Protection Sanitaire, Fontenay-aux-Roses, France^b

WHO Secretariat

Mr G. Ozolins, Manager, Environmental Hazards and Food
Protection, Division of Environmental Health, WHO, Geneva,
Switzerland (Secretary)

Dr M. Shore, National Center for Devices and Radiological
Health, Food and Drug Administration, Rockville, Maryland,
USA (Temporary Adviser)

---------------------------------------------------------------------------
^a Members of the International Non-Ionizing Radiation
Committee of IRPA.
^b Scientific Secretary of the International Non-Ionizing
Radiation Committee of IRPA.

Electric and magnetic field quantities and units in the SI system
-------------------------------------------------------------------
Quantity Symbol Unit
-------------------------------------------------------------------
Frequency f hertz (Hz)

Electric field strength E volt per metre (V/m)

Electric flux density D coulomb per square metre (C/m²)

Capacitance C farad (F)

Current I ampere (A)

Current density J ampere per square metre (A/m²)

Electric charge Q coulomb (C = A x s)

Impedance Z ohm (omega)

Volume charge density p coulomb per cubic metre (C/m³)

Magnetic field strength H ampere per metre (A/m)

Magnetic flux density B tesla^a (1 T = 1 Wb/m²)
(weber per square metre)

Permittivity epsilon^b farad per metre (F/m)

Permittivity of vacuum epsilon_o epsilon_o = 8.854 x 10^-12 F/m

Permeability µ henry per metre (H/m)

Permeability of vacuum µ_o µ_o = 12.57 x 10^-7 H/m

Time t seconds (s)
-------------------------------------------------------------------
^a 1 T = 10⁴ Gauss (G), a unit in the CGS unit system.
^b Designates a complex number.

ENVIRONMENTAL HEALTH CRITERIA FOR EXTREMELY LOW FREQUENCY (ELF)
FIELDS

Following the recommendations of the United Nations Conference
on the Human Environment held in Stockholm in 1972, and in response
to a number of World Health Resolutions (WHA23.60, WHA24.47,
WHA25.58, WHA26.68), and the recommendation of the Governing
Council of the United Nations Environment Programme, (UNEP/GC/10, 3
July 1973), a programme on the integrated assessment of the health
effects of environmental pollution was initiated in 1973. The
programme, known as the WHO Environmental Health Criteria
Programme, has been implemented with the support of the Environment
Fund of the United Nations Environment Programme. In 1980, the
Environmental Health Criteria Programme was incorporated into the
International Programme on Chemical Safety (IPCS). The result of
the Environmental Health Criteria Programme is a series of criteria
documents.

A joint WHO/IRPA Task Group on Environmental Health Criteria
for Extremely Low Frequency Fields met in Geneva from 5 to 9 March
1984. Mr. G. Ozolins, Manager, Environmental Hazards and Food
Protection, opened the meeting on behalf of the Director-General.
The Task Group reviewed and revised the draft criteria document,
made an evaluation of the health risks of exposure to extremely low
frequency electromagnetic fields, and considered rationales for the
development of human exposure limits.

The International Radiation Protection Association (IRPA)
initiated activities concerned with non-ionizing radiation by
forming a Working Group on Non-Ionizing Radiation in 1974. This
Working Group later became the International Non-Ionizing Radiation
Committee (IRPA/INIRC) at the IRPA meeting in Paris in 1977. The
IRPA/INIRC reviews the scientific literature on non-ionizing
radiation and makes assessments of the health risks of human
exposure to such radiation. Based on the Environmental Health
Criteria documents developed in conjunction with WHO, the
IRPA/INIRC recommends guidelines on exposure limits, drafts codes
of safe practice, and works in conjunction with other international
organizations to promote safety and standardization in the non-
ionizing radiation field.

This document is a combination of drafts prepared by Dr A.
Sheppard and Dr W.R. Adey (J.L. Pettis Memorial Veterans
Administration Hospital, Loma Linda, California), Dr M.G. Shandala,
Dr V. Akimenko and colleagues (A.N. Marzeev Institute of General
and Community Hygiene, Kiev, USSR), and Dr P. Czerski and Mr J.C.
Villforth (National Center for Devices and Radiological Health, US
Department of Health and Human Services, Rockville, Maryland). The
drafts were integrated at working group meetings in Grenoble
(1980), and Paris (1982). A subsequent draft of the document was
prepared by Dr P. Czerski, Dr B. Bosnjakovic, Dr M. Repacholi, Dr
V. Akimenko, Dr M. Grandolfo, Dr J. Cabanes, and Mrs A. S. Duchêne
at the WHO/IRPA working group in Paris in March 1983. A final
draft, incorporating the comments of reviewers from WHO National
Focal Points and many international experts, was prepared by Dr M.

Repacholi and Dr A. Sheppard in Geneva in December 1983.
Scientific editing of the draft, approved by the WHO/IRPA Task
Group in March 1984, was completed by Dr M. Repacholi and Dr A.
Sheppard. The efforts of all who helped in the preparation and
finalization of the document are gratefully acknowledged.

Subjects briefly reviewed, with particular reference to power
frequency (50 and 60 Hz) electric fields, include: the physical
characteristics of ELF fields; measurement techniques and
dosimetry; sources and applications of ELF; levels of exposure from
devices in common use; mechanisms of interaction; biological
effects in animals and animal tissues; human studies; health risk
evaluation and guidance on the development of protective measures
such as regulations or safe-use guidelines.

Although the emphasis of this document is on the effects of ELF
electric fields at 50 and 60 Hz, effects of ELF magnetic fields are
briefly mentioned since they always exist when electric current
flows. However, the specific problems related to static and time-
varying magnetic fields will be the subject of a separate
environmental health criteria document.

Health agencies and regulatory authorities are encouraged to
set up and develop programmes to ensure that the lowest exposure
occurs with the maximum benefit. It is hoped that this criteria
document will provide useful information for the development of
national protection measures against ELF fields.

1. SUMMARY AND RECOMMENDATIONS

1.1. Purpose and Scope

This document comprises a review of data on the effects of ELF
fields, predominantly ELF electric fields at 50 and 60 Hz, on
biological systems pertinent to the evaluation of health risks for
man. The purpose of the document is to provide information for
health authorities and regulatory agencies on the possible effects
of ELF field exposure on human health, and to give guidance on the
assessment of risks from occupational and general population
exposure. Areas in which uncertainties exist and further research
is needed are also indicated.

The document includes a review of the data on the biological
effects on human beings and animals of exposure to low frequency
electric and magnetic fields in the frequency range of zero to 300
hertz (Hz) (ELF).^a Data on the biological effects of exposure to
sinusoidally varying fields are mainly concerned with effects in
the range of 5 - 20 Hz or at 50 and 60 Hz, while limited data are
available on effects scattered throughout the ELF spectrum. Data
on studies with non-sinusoidal waveforms in this range have also
been considered. Effects of electrostatic and magnetostatic fields
are not included.

As the document mainly concerns effects directly attributed to
ELF electric fields, the effects of co-generated ozone, noise,
ultraviolet radiation (UVR) and X-rays from corona discharges,
induced short-circuit currents, etc., which may be important
factors in the overall transmission line environment, are discussed
only briefly.

In general, the effects of contact currents have not been
considered in detail since restriction of leakage currents from,
for example, household appliances and electromedical devices, is
already treated by national and international standards.

1.2. Sources of Exposure

Natural electric fields at extremely low frequencies are very
weak, while those of man-made origin are much stronger. The
strongest of the man-made electric fields are those surrounding
high voltage transmission lines ^b at 50 or 60 Hz, distribution
lines, and traction (transportation) systems that may operate at
16.67, 25, or 30 Hz. Within the home, the proximity of appliances

---------------------------------------------------------------------------
^a According to generally accepted usage in Europe, the region
from 30 Hz to 300 Hz is designated as extremely low frequency;
the region below this ELF band is unnamed. In the USA, the
ELF region is sometimes designated as 0 - 100 Hz (Polk, 1974).

^b These fields range up to about 10 kV/m within transmission
line corridors, and decrease to a background level of 10^-4 V/m
at approximately 10³ m.

and low voltage wiring produces ambient electric fields of 10^-1 -
10² volt/metre (V/m), depending on the mains voltage and the
distance.

The natural 60 Hz magnetic field is approximately 10^-9
millitesla (mT), which is low compared with the average fields (up
to 0.01 mT) found in private homes. Under the centre line at the
midspan of 1100 kV transmission lines, the 60 Hz magnetic field at
1 m above the ground is less than 0.035 mT. This is weaker than the
magnetic fields of up to 1 mT that occur close to common household
appliances.

It has become common practice to specify fields in terms of
their electric and magnetic field strength (E and H). The electric
field strength is specified in units of volts per metre (V/m). The
magnetic field is given by the field strength H in ampere/metre
(A/m) or the magnetic flux density B in weber/square metre (Wb/m²),
where 1 Wb/m² = 1 tesla.

1.3. Clinical Applications

The growth of bone tissue can be stimulated by electric
currents, and pulsed ELF fields are being used successfully in
clinical applications with patients suffering from intractable bone
disease or fractures. In the latter technique, electric currents
at ELF and higher frequencies are induced by pulsed magnetic
fields.

1.4. Field Measurement and Dosimetry

To characterize ELF fields, the strength, frequency, and
orientation of the electric and magnetic fields have to be
determined. Under power lines, the electric field has its major
component oriented vertically (perpendicular to the Earth's
surface), while the main magnetic field component is horizontal
(parallel to the Earth's surface). Principles of calculation and
measurements of these fields are outlined in section 2 of this
document.

A human or animal body located in an ELF electric field causes
perturbation of the field, resulting in an uneven distribution of
the field around the body. Both the electric and magnetic fields
induce electric currents in the exposed body. The electric fields
at the body surface and currents induced in man (a biped) and
quadruped animals are quite different, even at the same unperturbed
field strengths. The factors affecting the magnitude and
distribution of fields at the surface of the body and currents
induced inside the body are discussed below.

1.5. Characteristics of Biophysical Interactions

In regions of strong alternating electric fields, three
interactions occur:

(a) large surface fields exist, particularly at highly
curved regions, and may stimulate surface body
receptors, producing sensations;

(b) small currents flow within the body due to the large
surface fields; their magnitude is very small in
comparison with the currents that flow when contact
is made with charged conductors. The associated
internal electric field is some 10⁵ - 10⁷ times
smaller than the applied external electric field;

In most experimental situations with whole animals or in human
studies, the complex interrelationship between surface field
effects and possible internal electric field effects makes it
impossible to reach a clear conclusion on the importance of each
factor.

Although the non-magnetic nature of most biological materials
strongly suggests exclusion of magnetic field interactions,
alternating magnetic fields can induce electric currents similar to
the type of currents induced by coupling to electric fields. However,
in the transmission line context, the magnetically-coupled electric
currents are generally smaller, but within an order of magnitude of
the electric field-coupled currents. For exposures of prolonged
duration, currents produced by pulsed magnetic fields (peak
intensities of the order of 1 mT) are effective in modifying cell
functions (e.g., in the repair of bone fractures in human beings).
In laboratory studies with these same fields, changes have been
reported in bone growth, amphibian nucleated erythrocyte
dedifferentiation, nerve regeneration, and initiation and
alteration of DNA transcription, at current densities in fluid
bathing body cells of about 1 - 10 µA/cm² and electric gradients of
the order of 0.1 - 1 V/m.

The mechanisms by which a weak ELF field may interact with
biomolecular systems and tissues are incompletely understood.
However, from in vitro studies, there is now evidence of
field-induced interactions, including the phenomena of ionic
interactions with membrane surface macromolecules, which appear to
involve coupling of the cell interior to signals from neurotransmitters,
hormones, and antibodies.

1.6. In vitro Studies

In vitro studies are conducted for two main reasons:

(a) to elucidate mechanisms of interaction of ELF fields
with biological materials; and

(b) to provide information on end-points to search for in vivo.

These studies have included examination of interactions with
excised and cultured tissues, cell biochemistry, neurophysiology,
and growth of bone tissue. Electric fields were reported to affect
endocrine gland secretion, response to hormonal stimulation, brain
calcium ion exchange, immunoreactivity of lymphocytes, electrical
excitability of neuronal tissue, and tissue growth rates.

Some of these studies have revealed ELF field effects occurring
within certain "windows" in frequency and amplitude.

1.7. Experimental Animal Studies

The majority of ELF research has focused on effects directly or
indirectly involved with the central nervous system including
physiological, ultrastructural, and biochemical alterations, changes
in blood composition, behaviour, reproduction, and development.
Studies have been conducted almost exclusively on small laboratory
animals, except for a few studies carried out with miniature swine
and non-human primates.

Although some experimental data exist, one of the most serious
shortcomings of the studies on small animals results from an
inability to make extrapolations to human beings because of
uncertainty about applying the mechanisms proposed for the effects
seen so far. In particular, it is difficult to cite equivalent
human exposure because of vast differences in the distribution of
surface electric field strengths and internal current densities
between human beings and animals, and because there are no data on
the species dependency of effects.

Studies with small animals exposed to electric fields up to
100 kV/m have revealed effects on components of the nervous system,
including synaptic transmission, on circadian rhythms, and on the
biochemical properties of brain tissue. Results of behavioural
studies suggest that the nervous system may be affected by an ELF
electric field that is far too weak to stimulate synaptic function
or cell firing, although in vivo studies often do not exclude the
possible role of tactile sensory phenomena.

Field effects on peripheral blood composition and biochemistry
have been studied by numerous investigators with inconsistent
results. Generally, the changes in blood picture involve small
deviations from individual norms, but the values usually remain
within physiological norms. Results of studies on the influence of
ELF fields on immunocompetence in whole animals appear to be
negative.

Studies on swine exposed to 30 kV/m and rodents exposed to 65
kV/m for up to 18 months have revealed evidence of teratological
effects. These data are not conclusive and do not prove the
teratogenic potential of ELF fields in general.

Many studies on genetic effects and effects on cardiovascular
function have been reviewed and the conclusion reached that such
effects have not been convincingly demonstrated.

1.8. Effects on Man

Existing surveys of the state of health of high voltage (HV)
substation workers and HV line maintenance crews have been based on
small populations and have produced conflicting results. Soviet
authors noted an increased incidence of subjective complaints
attributable to effects on the nervous system and shifts in blood
biochemistry, but other authors have not reported such observations.
Differences in method often make comparison difficult, if not
impossible. Field strengths to which personnel were exposed were
only estimated, and only approximate data on the duration of
exposure to fields in a given strength range were available.

Some studies on volunteers exposed to electric fields up to 20
kV/m for short periods (days), under laboratory conditions,
confirmed the existence of slight changes (within the normal
physiological range) in populations of peripheral blood cells and
biochemistry, similar to those observed in experimental animal
studies.

Several recent epidemiological reports have presented
preliminary data suggesting an increase in the incidence of cancer
among children and adults exposed to magnetic fields through living
close to various types of electrical power lines or devices (e.g.,
power lines coming into the home, transformers, or other electrical
wiring configurations), and among workers in electrotechnical
occupations.

Slight increases in genetic defects or abnormal pregnancies
have been reported in one study. Epidemiological studies have been
performed on linemen and switch-yard workers, the groups considered
to be subjected to the highest electric-field exposure levels.
However, the exposure levels to which these people are subjected
have been found to be remarkably low. The preliminary nature of
the epidemiological findings, the low levels of exposure, and the
relatively small increment in the reported incidence of any
effects, suggest that, though the epidemiological data cannot be

dismissed, there must be considerable study before they can serve
as useful inputs for risk assessment.

No pathological effects resulting from ELF field exposure have
been established. However, thresholds for perception, startle, let-
go, respiratory tetany, and fibrillation due to contact currents
(electric shocks) have been quantified.

1.9. Exposure Standards

The few instances where countries have developed standards
limiting human occupational or environmental exposure to ELF fields
are discussed and compared in section 9 of the document.

1.10. Conclusions and Recommendations

1. In order to relate biological findings from in vitro and in
vivo studies on experimental animals to human beings, it is
recommended that dosimetry studies should be continued to
measure and relate external electric field strengths and
internal current density distributions in the whole body of
both animals and human beings.

2. From studies on man and animals, observed sensitivities are
consistent with two proposed models, one on the basis of
stimulation of peripheral sensory receptors in strong local
electric fields at the body surface, and the other on the basis
of current densities induced in the extracellular fluid. It is
recommended that models be devised that correlate exposure and
biological effects in terms of physical factors, such as
surface electric field, tissue current density, spark
discharges, and waveform.

3. The continuation of basic research on electric and magnetic
field interaction mechanisms is strongly recommended.
Investigations should be conducted on the possible synergism or
antagonism of field influences with physical and chemical
agents, since such data are not available.

4. In some studies, restriction of ELF effects to certain "windows"
in frequency and amplitude has been reported. It is recommended
that the window concept be further investigated to determine the
applicability of data obtained with various frequencies and
waveforms, and to relate the findings to potential health
detriment in human beings.

5. Studies have been performed on workers with long-term exposure
to electric and magnetic fields, but no adverse health effects
have been identified. However, these studies were not designed
to evaluate effects on reproductive functions, or long-term
carcinogenic risks. In two of the studies, electric field
exposure was carefully evaluated, and it was found that average
exposures in the occupationally-exposed groups were remarkably
low.

A suggestion of increased cancer incidence has been reported by
a number of investigators who have examined occupational and
general population groups exposed to electric and magnetic
fields. The studies performed have serious deficiencies in
epidemiological design and do not adequately characterize
levels and duration of exposure.

The limited knowledge of the potential human health risk
associated with exposure to electric and magnetic fields makes
it imperative that well-designed epidemiological studies should
continue to be undertaken to provide a firmer basis for risk
assessment.

6. Occupational exposure to strong electric fields is generally
intermittent and of short duration; exposed populations have
been identified, and there are some limited data based on
practical experience. At field strengths where spark
discharges are prevalent, prolonged exposures may impair
performance. Such exposures should be avoided, where possible.

7. Linemen working on energized extra- or ultra-high-voltage
conductors experience extreme electric field conditions, and
the use of appropriate protective clothing or devices is
desirable.

8. Whilst it would be prudent in the present state of scientific
knowledge not to make unqualified statements about the safety
of intermittent exposure to electric fields, there is no need
to limit access to regions where the field strength is below
about 10 kV/m. Even at this field strength, some individuals
may experience uncomfortable secondary physical phenomena such
as spark discharge, shocks, or stimulation of the tactile
sense.

9. It is not possible from present knowledge to make a definitive
statement about the safety or hazard associated with long-term
exposure to sinusoidal electric fields in the range of 1 - 10
kV/m. In the absence of specific evidence of particular risk
or disease syndromes associated with such exposure, and in view
of experimental findings on the biological effects of exposure,
it is recommended that efforts be made to limit exposure,
particularly for members of the general population, to levels
as low as can be reasonably achieved.

10. In principle, electric and magnetic field interference with
implanted cardiac pacemakers can lead to reversion to a fixed
rate, but cessation of stimulation is possible. Direct
interference has not been reported in fields below 2.5 kV/m.
Although body currents produced by contact with a vehicle in a
weaker field may cause interference, the risk of pacemaker
reversion is believed to be slight.

It is recommended that pacemaker designers and manufacturers of
other similar electronic equipment ensure that their devices
are resistant to failures caused by electric or magnetic field-
induced currents.

2. PHYSICAL CHARACTERISTICS, MEASUREMENT, AND DOSIMETRY

2.1. Quantities and Units

The electric (E) and magnetic (H) fields that exist near sources
of ELF electromagnetic fields must be considered separately, because
the very long wavelengths (thousands of kilometres) characteristic
of extremely low frequencies means that measurements are made in the
non-radiating near field. The E and H fields do not have the same
constant relationship that exists in the far field of a radiating
source.

In the vicinity of high-voltage transmission lines, the E and H
fields are typically of the order of a few kilovolts per metre
(kV/m) and a few tens of amperes per metre (A/m), respectively.
Persons standing under HV-transmission lines or in the vicinity of
charged conductors will have currents induced in their bodies as a
result of the interactions of these fields with the body tissues.
To fully assess the health implications of these ELF fields on man,
the fields must be measured accurately and interactions with the
body described quantitatively. The human body has a complex
geometrical structure making it difficult to obtain precise
theoretical or experimental descriptions of these interactions.

The quantities and units used in describing ELF electric and
magnetic fields are given on page 8.

The electric and magnetic fields are each described by a vector
defined by space components along three orthogonal axes. For
steady-state sinusoidal fields, each space component can be
represented as a phasor, i.e., a complex number having a magnitude
and phase. The magnitude is expressed as the root mean square
(rms) value of the field strength in volts per metre or amperes per
metre, respectively. The electric field strength is defined in
terms of a force exerted by the electric field on a unit charge and
the magnetic field by the force on a unit current in a unit length
of wire.

The flow of charge is the electric current measured in amperes
(A). The electric charge is the integral of electric current over
time and is expressed in ampere-seconds (A x s) or coulombs (C).
Derived quantities are surface charge density (A x s/m²) and volume
charge density (A x s/m³). The current density is defined as the
current flowing through a unit area perpendicular to the current
direction and is expressed in amperes per square metre (A/m²).

The electric flux density D is a vector quantity, the divergence
of which is equal to the volume charge density. The unit of the
electric flux density is A x s/m², and it is related to the
electric field strength by the equation D = epsilonE, epsilon being
the permittivity. Permittivity of the vacuum is epsilon_o = 8.854
x 10^-12 A x s/Vm. The relative permittivity, epsilon_r, a
dimensionless quantity, is defined as epsilon_r = epsilon/epsilon_o.
In free space, epsilon_r = 1, but in tissues, the values of epsilon_r
are significantly greater than 1. The capacitance between two

objects, measured in farads (1 F = 1 A x s/V), is defined as the
charge acquired by an object divided by the potential difference
existing between them.

The magnetic field can be described by the magnetic field
strength H and by the magnetic flux density B, where B = µH, µ
being the permeability. In free space, µ_o = 12.566 x 10^-7
V x s/Am. The relative permeability µr, a dimensionless quantity, is defined
as µ_r = µ/µ_o, µ_r = 1 in air, by definition, and also, for all practical
purposes, in biological tissues as well. The magnetic field is an axial
vector quantity, the curl (rotation) of which is equal to the total current
density, including the displacement current. Magnetic flux density, sometimes
called magnetic induction, is expressed in tesla, where 1 T = 1 V x s/m² = 1
Wb/m².

When describing exposure conditions, the electric and magnetic
field strengths and orientations should be indicated together with
the frequency. In the case of AC transmission lines, electric and
magnetic field components have a fundamental frequency equal to 60
Hz in North America and 50 Hz elsewhere.

Harmonic content is due to the distortion of sinusoidal wave-
form of the fundamental frequencies by waveforms of other frequencies,
and can be characterized by a Fourier series. The harmonic content
may be of importance at points near large industrial loads or in
laboratory installations.

Three-phase transmission lines generate fields, the space
components of which are not in phase. The field at any point close
to line current conductors can be described by the field ellipse,
i.e., the field vector describes an ellipse in any full cycle. At
distances of about 15 m or more away from the outer conductor, the
electric field of transmission lines can be considered practically
a single phase field.

The vertical component of the electric field under a
transmission line is the rms value of the component of the electric
field along the line perpendicular to the ground and passing through
the point of measurement. This quantity is often used to characterize
induction effects in objects close to ground level.

The space potential of a point is a phasor representing the
voltage difference between the point and the ground. The space
potential is perturbed by the introduction of an object into the
field. The "unperturbed space potential" that would exist if the
object were removed is often used in describing the field. This is
the induction field potential.

It is important to consider the action of an electric field on
the human body. Within the body, the low frequency electric field
is attenuated by about 10⁵ - 10⁷ from the value of the external
field. This is in contrast with a magnetic field. Because of the
high conductivity of the human body, the electric field is distorted
and localized at the surface of the body. In an alternating field,
a current is produced within the body, that has the same frequency
as the external field.

The human body acts as a conductor at ground potential, when a
person is in good electrical contact with earth, as when wearing
conducting shoes. A person wearing well-insulated shoes assumes a
free or floating potential above ground. However, the resistance
to earth of a person wearing shoes with leather soles is about
15 k ohms and with plastic shoes about 100 M ohms (Deno, 1977).
When the body is earthed, a current flows through the body to
ground. This current is approximately 14 µA (for 50 Hz) and 17 µA
(for 60 Hz) for each 1 kV/m of undisturbed field strength
(depending on the body size and shape). Of this current, about
one-third flows into the head (Hauf, 1982).

Kaune & Gillis (1981) formalized a number of concepts that
simplify the description of the interaction between an animal and
an ELF electric field. These authors showed that the electric-
field intensity at the surface of the body and induced currents
passing through various segments of the body are determined by:
(a) the characteristics of the applied electric field, i.e., field
strength, spatial structure, and frequency; (b) the shape of the
body; (c) the location of the body relative to ground and other
conductors; and (d) any conduction currents from the body to ground
or other conductors. Because these quantities do not depend on the
internal structures of the body, they can be measured using
conducting models, which may be hollow. The authors showed that
the electric field outside the body and the induced charge density
on the surface of the body are independent of frequency in the ELF
range for both grounded and ungrounded exposure conditions. They
also showed that the electric field outside and inside a body will
be unchanged by a scaled change in the size of the body. Finally,
these authors proved that the electric charge induced inside the
body of an exposed human being or animal is small compared with
that induced on the surface of the body.

The magnetic field is not perturbed by objects that are free of
magnetic materials. Magnetic field induction in objects causes two
types of electric currents (Zaffanella & Deno, 1978):

- a circulating current inside the object (eddy
current) induced by the magnetic flux density; and

- a current entering and leaving the object which may
be induced by the magnetic flux density through some
large loop external to, but including, the object.

2.2. Computational Methods and Measurements of ELF Electric Fields

Computational methods for the determination of electric and
magnetic fields are presented in standard textbooks on engineering
and physics. Detailed data on computational methods for in case of
HV transmission lines are presented in The Transmission Line
Reference Book (1975) and Zaffanella & Deno (1978).

There are basically two different approaches to the measurement
of 50 or 60 Hz E fields:

(a) free-body probes that measure fields at points remote
from the ground (Transmission Line Reference Book,
1975; Bracken 1976); and

(b) ground-reference instruments that measure the current
to ground that is collected by a metallic surface
(Miller, 1967).

The principles of operation of both types of instruments are
closely related. A free-body instrument consists of a hollow
metallic shell that is cut in half and the two halves insulated
from each other. The displacement current intercepted by a half-
shell is the time derivative of the surface unit charge, and for a
sinusoidal field:

I = dQ/dt = k omega epsilon_oE cos omega t Equation (1)

where E is the E-field strength, Q is the charge induced on one of
the half-shells, omega is the angular frequency, epsilon_o is the
permitivity of free space, and k is a constant.

The theory of operation for the ground-reference instrument is
quite similar to the above. A flat reference plate is placed on
the ground in electrical contact with the ground. A second plate
is placed a small distance above the reference plate and insulated
from it. The displacement current is again given by Equation (1).

The free-body approach is recommended for outdoor measurements
near power lines, since it does not require a known ground reference
for measurements anywhere above ground. The ground-reference probes
can be used only under special conditions as discussed in the
Transmission Line Reference Book (1975).

The electric field is perturbed (in some circumstances
significantly) by the presence of human beings, vegetation, or
other structures. Data presented in the literature show that, as
a general rule, in the presence of each perturbing influence, the
measured values are somewhat less than the unperturbed ones.

Three main procedures are used for the calibration of E-field
instruments:

(a) parallel plate techniques (usually with guard rings);

(b) single ground plate;

These three techniques have been reviewed by Kotter & Misakian
(1977). The first method is the best as it provides an accuracy of
1% or better for calibration of the field.

The electric field strength meter should be calibrated
periodically at intervals determined by the stability of the meter.
The instrument on a long (at least 2.5 m) handle is held between
the plates at the centre of the structure to take the appropriate
measurements while the plates are at a known voltage.

For some instruments, a correction for temperature and humidity
may be required. Therefore, these parameters should always be
recorded at the time of calibration and at the time measurements
are made.

In 1978, IEEE presented a technique using parallel plates to
calibrate power-line field survey meters. Two parallel, square
metallic plates separated by a distance d are supplied by an
alternating voltage source. The electric field strength E at the
midplane of the setting is given by:

E = V/d

where V is the voltage difference existing between the plates.
Fringing field effects at the periphery of the plates tend to
modify the field that would be expected to occur at the centre
point of the plates. It was found that for a pair of parallel
plates 1 m² each and spaced 0.5 m apart, the variation in field
magnitude was less than 1% from the simply computed value at the
centre in the midplane. This system constitutes a simple method of
evaluating survey meters.

For additional data, see the IEEE Standard for Recommended
Practices for the Measurements of Electric and Magnetic Fields from
Power Lines (IEEE 1978, 1979) and Tell (1983).

2.3. Field Polarization and Homogeneity

At ground level, beneath the transmission line, the electric
field is essentially a vertical homogeneous field with a horizontal
component that is about 20% of the vertical component (Poznaniak et
al., 1979). At distances of more than 15 m from the outer conductor,
this horizontal field drops to less than 10% of that of the vertical
field (Zaffanella & Deno, 1978).

Most experimental arrangements for the exposure of animals
involve a pair of horizontal parallel electrodes to produce a
vertical electric field that is quite homogeneous, if the electrode
spacing is adequate. Calculations (Ware, 1975; Shih & DiPlacido,
1980) indicate that the unperturbed electric field strength between
parallel plates is quite uniform in both the horizontal and vertical
directions, when the horizontal dimensions are two or more times
greater than the distance between the plates.

2.4. Energy Carried by the Field

A 10 kV/m electric field has an energy density of 4.42 x 10^-4
J/m³ producing an average power density in the body of man of about
10 µW/m³, which is about 10^-8 times the metabolic rate of the human
body (Sheppard & Eisenbud, 1977). Thus heating of a body by an ELF
field is completely negligible.

2.5. Determination of ELF Field Exposure

There are no universally accepted and clearly defined concepts
relating ELF field "dose" to biological effects, comparable with
ionizing or radiofrequency dosimetry in terms of exposure and
absorbed doses. Deno (1977), for example, suggests that exposure
to the electric field can be expressed as the product of electric
field strength and the duration of exposure. The current dose
(charge) on the various body surfaces and inside the body is a
constant ratio in the unperturbed field, if a person stays erect.
An "equivalent" E field is the vertical field at ground level (0.5
m), which would cause the same induction space potential at each
body position. The fields induced inside the body depend only on
the charges at the body surface (Deno 1979).

A "dose monitor" to measure the electric field exposure in
terms of the time integral of the unperturbed field in the ranges
of 0 - 5 kV/m, 5 - 10 kV/m, and above 10 kV/m has been constructed.
A separate device discharges the integrator monitors and gives a
reading of the doses after the exposure period. The actual
exposures were shown to be lower than those obtained by multiplying
the electric field strength in the area of work by the total time
spent in these areas. A similar device, constructed by Lövstrand
et al. (1979), was used to measure the exposure of workers in 50 Hz
EHV-substations. Lövstrand et al. (1979) stressed, however, that
the relationship between the unperturbed electric field strength
and the biological effect was by no means clearly established.
They maintained that further work was needed to develop dosimetric
concepts and to establish the relative significance of surface
charge densities, internal electric field strength, current, and
current densities for the "dose"-biological effect relationship.

2.6. The Physical Interaction of Man and Laboratory Animals with
Electric Fields

The exposure of intact organisms to ELF electric fields is
conventionally specified in terms of the unperturbed field
strength, in V/m or kV/m, which is measured or calculated before
the subject enters the field. The unperturbed field is not,
however, the field that acts directly on an exposed subject. The
fields to which a subject is actually exposed can be categorized as
follows:

(a) Electric fields acting on the outer surface of the
body. These fields can cause hairs to vibrate and
can thereby be perceived; they may also be able to
stimulate other sensory receptors in the skin.

(b) Electric fields induced inside the body. These
fields act at the level of the living cell, and their
presence is accompanied by electric currents because
of the conductivity of living tissues.

It has been shown that electric fields at the surface of a
conducting object are enhanced relative to the unperturbed field,
while induced fields inside the body are attenuated by about 10⁶
(Barnes et al., 1967; Deno, 1977; Kaune & Phillips, 1980).

2.6.1 Surface fields and internal current density

The electric field lines (the directions along which a charge
is moved by the force imposed by the field) are perpendicular to
the surface of the body. A greater concentration of electric field
lines (i.e., higher field strength) exists at a curved surface,
such as the human head, than on less curved surfaces of the body.
For this reason, it is useful to specify the surface electric field
that exists on various parts of the body.

A conducting object placed in an electric field carries a
current that is directly related to these surface fields. Thus,
the internal currents are greatest at the areas of the most intense
surface electric field. The current carried within the body (or a
portion thereof) can be calculated from the capacitance of the
body, a quantity that takes into account the size and shape of the
body and its proximity to other conducting objects such as the
ground and high voltage electrodes or wires, or perhaps other
animals, fences, trees, etc. (Deno, 1974, 1975, 1976, 1977;
Bracken, 1976; Zaffanella & Deno, 1978; Kaune & Phillips, 1980).

Within the body, the two quantities of interest are the current
and the current density. The total current is more easily measured
or calculated, but the current density is more directly relevant in
discussion of electric field effects in a particular tissue or
organ. The electrical complexity of the interior of the human
body, due to the presence of insulating membranes and tissues of
various impedances, has so far frustrated confident analysis of
precise interior current densities (Kaune & Phillips, 1980;
Spiegel, 1981).

2.6.2 Capacitive coupling of the electric field to man and
laboratory animals

A body is coupled to an electric field in proportion to its
capacitance such that the greater the capacitance the greater the
current flow in the body. For example, the capacitance of a rat is
about 5 picofarad (pF), while human beings have capacitances of
about 125 pF, when in close proximity with ground (Deno, 1974; Deno
& Zaffanella, 1975).

In many laboratory exposures of small animals, the distance
between the animal and the lower electrode is small or nonexistent
so that the animal's capacitance to this lower (usually earthed)
electrode represents a substantial portion of the total
capacitance.

By definition, in capacitive coupling, the body, according to
its capacitance C, "acquires" a certain amount of surface charge Q
and attains a potential V = Q/C. This view finds formal expression
in models that express any arbitrarily complex body as an equivalent
plate at an equivalent height, such that the total current collected
by the plate is the same as that for the actual body (Deno, 1974).
The capacitance, and thus the induced current, decrease for a body
separated from the ground and not close to the energized electrode.
The capacitance is dependent on the size, especially on the surface
area, shape, and orientation of the body, so that internal currents
will differ between fat and thin persons, between persons standing
and reclining, and between persons walking barefoot and those
wearing thick rubber-soled shoes or standing on a platform. It
would be useful, in all cases, to define the conditions under which
the capacitance has been measured.

A short-circuit current I_sc flows in a body placed in an
electric field and connected to the ground through a low resistance
path (paws, bare feet, a hand grasping an earthed pole). This
current is the sum of all the displacement currents collected over
the surface of the body. The only place on the body where a
current of the magnitude of the short-circuit current flows is
where there is connection with the ground.

2.6.3 Shock currents

In contrast with capacitive coupling to the field, a person
touching a conductor carries a "shock current", the magnitude of
which is determined by the total circuit impedance including the
electrical impedance of the skin and body. Exposure to an
extremely strong electric field would be needed to produce
displacement currents of several milliamperes, which would
represent a hazard similar to that of touching a live wire (Schwan,
1977).

2.7. Dosimetry and Scaling Between Laboratory Animals and Man

The surface and induced fields to which quadrupeds (e.g.,
laboratory animals) and bipeds (e.g., human beings) are exposed are
markedly different at the same unperturbed field strength. Hence,
it is necessary to scale exposures across species to compare
biological responses.

At present, there are several ways in which electric-field
exposure effects found in animals might be scaled to possible
effects in human beings. One way is to scale on the basis of
equivalent surface electric fields. Alternatively, the induced
currents or electric fields in corresponding tissues and organs
could be determined.

When scaling of exposure is made on the basis of equivalent
surface electric fields, it is assumed that the mechanism by which
biological effects are produced involves stimulation of receptors
on the surface of the body or currents at the surface of the body.
Stimulation of peripheral somatosensory receptors has been
demonstrated by Jaffe (1982, in press). Also, electric fields at
the surface of the body can produce oscillation of hairs on the
surface with a resultant stimulatory effect.

Classical neurophysiology suggests that induced current
densities could produce changes in cell physiology when
transmembrane current densities are of the order of 0.1 mA/cm²
(Schwan, 1982b). Large current densities are not normally possible
because such high E fields would be needed that electrical break-
down of the air would occur long before these current densities
could be induced. Novel mechanisms of interaction of the E field
with various biological systems would be needed to explain any
effects at the current densities of 10^-6 - 10^-9 A/cm² that may
be caused by fields found in the environment.

An erect grounded human being (biped) couples more strongly to
an ELF electric field than laboratory animals (quadrupeds).
Surface electric fields and axial current densities have been
measured in models of man, pig, and rat by Kaune & Phillips (1980)
(Fig. 1). At the tops of the bodies, surface electric fields are
enhanced over the unperturbed field strength present before the
subjects entered the field by factors of 18, 6.7, and 3.7 for human
beings, swine, and rats, respectively. For an unperturbed field
strength of 10 kV/m, average induced axial current densities in the
neck, chest, abdomen, and lower part of the legs are, respectively:
550, 190, 250, and 2000 nA/cm² for human beings; 40, 13, 20, and
1100 nA/cm² for swine; and 28, 16, 2, and 1400 nA/cm² for rats.

Recently an attempt has been made to determine human exposure
conditions simulated by animal exposures to 60-Hz electric fields
(Guy et al., 1982). A thermographic method for determining the
specific absorption rate (SAR) was used to quantify the electric
current distributions in homogeneous models of animals and human
beings exposed to uniform 60-Hz electric fields by exposing models
(of scaled size and conductivity) to 57.3 MHz fields. Although the

values of maximum current density predicted in the ankles of models
of human beings exposed to 60-Hz fields at 1 kV/m, 200 nA/cm,
agreed with independent measurements on full-scale models, the
simulation of the 60-Hz field with a 57.3-MHz field may not be
exact, when determining corresponding current densities in animals
and man.

Theoretical models of biological effects of electric fields
must distinguish between the importance of the microscopic electric
field versus the microscopic electric current, although these two
quantities will always be interrelated. If the model for inter-
action depends on transport of a certain quantity of charge, then
the microscopic electric field is not the quantity of interest, and
various experimental results should be scaled according to the
current density in the tissue. On the other hand, if the tissue is
sensitive to the electric field strength, independent of the charge
transported by that field strength, the model would require scaling
on the basis of internal electric field strengths. In general terms,
two such models would characterize the ultimate biophysical inter-
action measurement as either "voltmeter-like" or "ampmeter-like".

Whether an experimental result should be scaled according to
the current or the electric field can be very important because the
tissue conductivity that relates these two quantities varies

signficantly among various tissues and, even more widely, among the
tissue or subcellular components (e.g., plasma membranes) of
different species (section 5).

It is clear from these data that exposures in studies with
laboratory animals must be scaled to compare biological effects
from such studies to possible effects in man. Although
considerable progress has been made in the dosimetry of ELF
electric fields during the past several years, additional research
is needed before data from experimental animal studies can be
extrapolated to man. Information is needed on:

(a) interaction mechanisms;

(b) critical sites in the body that produce any effects;

(d) physiological differences between species.

2.8. Magnetic Induction of Electric Fields

An animal or human body does not appreciably affect a magnetic
field, but the magnetic field induces currents within the body.
The magnitude of these internal currents is determined by the
radius of the current path, the frequency of the magnetic field,
and its intensity at the location within the body. Unlike the
electric field, for which the internal field strength is many times
less than the external field strength, the magnetic field strength
is virtually the same outside the body as within. The magnetically-
induced electric currents are greatest at the periphery of the body
where the conducting paths are largest, whereas microscopic current
loops anywhere within the body would have extremely small current
densities. The magnitude of the current density is also influenced
by the conductivity of the tissues, and the exact paths of the
current flow depend in a complicated way on the conducting
properties of tissues.

The induced current density and power absorbed by a prolate
spheroid model of a man exposed to the magnetic-field component of
a transmission line have been calculated by Spiegel (1977).

3. NATURAL BACKGROUND AND MAN-MADE ELF FIELDS

3.1. Natural Electric Fields

The electric and magnetic fields of the Earth consist of a
static component, which is dominant, and a time-varying component,
which is smaller than the static component by several orders of
magnitude in the 50 - 60 Hz frequency range (Polk, 1974). The
fields are characterized by vertical components E_z and H_z for the
electric and the magnetic fields, respectively, as well as by two
horizontal components E_x,y and H_x,y.

The most important sources of man-made fields in the ELF range
operate at the power frequencies of 50 Hz or 60 Hz. The natural
electric field strength at the power frequencies of 50 Hz or 60 Hz
is about 10^-4 V/m, which means that fields in the close vicinity of
HV transmission lines are 10⁸ times stronger, and the fields
introduced into homes by wiring or appliances are still about 10³ -
10⁶ times stronger than the natural background.

The natural electric field near the Earth's surface is a static
field of about 130 V/m (Dolezalek, 1979). This is due to a
separation of electric charge between the atmosphere and the
ground, so that the Earth resembles a spherical capacitor and the
ground and upper atmosphere represent conducting surfaces. Daily
changes in the natural electric field are attributed to factors,
such as thunderstorms, that affect the rate of charge transfer
between the ground and the upper atmosphere. According to Chalmers
(1967), thunderstorms have electric fields of 3 - 20 kV/m.

The alternating fields at low frequency are related to
thunderstorm activity and magnetic pulsations that produce currents
within the Earth (telluric currents). The strength of the Earth's
electric field varies in time and over the frequency range 0.001 -
5 Hz (Krasnogorskaja & Remizov, 1975). Local variations occur
depending on atmospheric conditions and variations in the magnetic
field. The main characteristic of the Earth's electric field are
presented in Table 1.

3.2. Natural Magnetic Fields

The natural magnetic field is composed of an internal field,
due to the Earth acting as a permanent magnet, and to an external
magnetic field in the environment from such components as solar
activity, telluric currents, atmospheric activity, etc.

The internal magnetic field of the Earth originates from the
electric current in the upper layer of the Earth's core. There are
significant local differences in the strength of this field,
varying from about 50 A/m at the poles to about 23 A/m at the
equator (Presman, 1971; Benkova, 1975). These field strengths also
vary with time.

Table 1. Characteristics of the Earth's electric field in the ELF range
---------------------------------------------------------------------------
Frequency Nature of the field Field strength Reference
range (Hz) (V/m)
---------------------------------------------------------------------------
0.001 - 5 Short duration pulses 0.2 - 1000 Krasnogorskaja &
(magnetohydrodynamic for E_z Remizov (1975);
origin) Vanjan (1975)

7.5 - 8.4 3 - 6 quasisinusoidal On the average, Beresnev et al.
and pulses of undetermined (0.15 - 0.6) (1976)
26 - 27 origin during an 10^-6 for
interval of 0.04 - 1 s E_x,y with a
maximum of 10^-6

5 - 1000 Related to atmospheric 10^-4 - 0.5 Aleksandrov et al.
changes (atmospherics) for E_z, and (1972);
present all the time one order of Presman (1971);
magnitude lower Kleinmenova (1963)
for E_x,y. The
amplitude
decreases with
increasing
frequency
---------------------------------------------------------------------------
The external magnetic field consists of many components
differing in spectral and energy characteristics (Aleksandrov et
al., 1972; Polk, 1974; Benkova, 1975). The variations in the
magnetic fields are related to solar activity, particularly with
respect to the ELF components, which change over 11-year and 27-day
periods and also exhibit circadian variations. Other causes of
variations in the natural magnetic fields are thunderstorms,
atmospheric changes, and air ionization. About 2000 thunderstorms
are occurring simultaneously over the globe, and lightning is
striking the Earth's surface about 160 times per second; the
currents involved may reach 2 x 10⁵ A at the level of the Earth
(Kleimenova, 1963). Electromagnetic fields having a very broad
frequency range (from a few Hz up to a few MHz), originate the
moment lightning strikes and propagate over long distances,
influencing the magnitude of magnetic fields.

The characteristics of the Earth's magnetic field can be
summarized as follows:

(a) The amplitudes from 4 x 10^-2 to 8 x 10^-2 A/m are at
pulsation frequencies ranging from 0.002 to 0.1 Hz.

(b) The geomagnetic pulsations up to 5 Hz are of short
duration, lasting from a few min to a few h.

(d) At 50 or 60 Hz, the natural magnetic field is
approximately 10^-9 mT (Polk, 1974).

The geomagnetic field exhibits temporal and spatial variations
related predominantly to solar activity and local magnetic
aberrations.

3.3. Man-Made Sources of ELF

3.3.1 High-voltage transmission lines

The principal man-made sources of ELF are HV transmission
lines, and all devices containing current-carrying wire, including
equipment and appliances in industry and in the home operating at
power frequencies of 50 Hz in most countries and at 60 Hz in North
America.

Electrical energy is transmitted from the power plant, where it
is generated, along conductive, metallic transmission connections
(overhead power lines or underground cables) to substations and
finally to energy consumers.

A typical overhead line (Fig. 2) consists of supporting
structures (transmission towers or pylons) from which the live
conductors are suspended by sets of insulators. Each set of
insulators supports a single conductor or a bundle of two or more
conductors, which carries one electrical phase of the power supply.
The conductors of each phase are suspended far enough away from the
other conductors and the transmission tower to prevent flashover or
short-circuiting between one phase and another, or between the
phases and earth (via the supporting structure). In overhead
lines, the conductors consist of bare metal cables. Thus, any
approach to a live conductor presents a lethal danger due to
flashover and a resulting electric current flow that would precede
actual contact with a conductor.

High voltage lines are operated at standard voltages up to 750
or 765 kV and a line at 1100 kV is operating in the USSR. The
construction of 1000 - 1200 kV or 1500 kV lines is in progress or
at various stages in planning.

Most widely used are alternating current (AC) 3-phase HV lines.
One circuit of the 3-phase line comprises 3 single or 3 sets of
conductors under high-voltage and 1 or 2 grounded conductors that
protect the live conductors against lightning.

Typically, the unperturbed electric field at the height of an
average man, standing at the location of the maximum field (just
outside the outer conductor) of a high-voltage transmission line of
750 kV, is of the order of 10 kV/m. A lower value of about 1 kV/m
exists where the line is highest from the ground (20 m) and about
12 kV/m where it is lowest (13 m) (Zaffanella & Deno, 1978). The
electric field strength is a function of the lateral distance from
the centre of the HV line as shown in Fig. 3.

Occupational exposures that occur near high voltage transmission
lines depend on the worker's location either on the ground, or at
the conductor during live-line work at high potential. When working
under live-line conditions, protective clothing may be used to
reduce the electric field strength and current density in the body
to values similar to those that would occur for work on the ground.
Protective clothing does not weaken the influence of the magnetic
field.

3.3.2. Electric fields near transmission lines and sub-stations

At ground level, beneath high-voltage transmission lines, the
electric fields created have the same frequencies as those carried
by the power lines. The characteristics of these fields depend on
the line voltage, and on the geometrical dimensions and positions
of the conductors of the transmission line. The field intensity
selected for reference or comparison purposes is the undisturbed
ground level electric field strength. To avoid the effects of
vegetation or irregularities in the terrain, the unperturbed field
strength is usually computed or measured at a given height above
ground level (0.5, 1, 1.5, or 1.8 m).

There are several primary influences on the electric field
strength beneath an overhead transmission line. These include:

(a) the height of the conductors above ground (which is
influenced considerably by the ambient temperature
and heating caused by the current passing through the
conductor);

(b) the geometric configuration of conductors and
earthing wires on the towers, and in the case of two
circuits in proximity, the relative phase sequencing;

(d) the proximity of other tall objects (trees, fences,
etc.);

(e) the lateral distance from the centre line of the
transmission line;

(f) the height above ground at the point of measurement;
and

(g) the actual (rather than the nominal) voltage on the
line.

Inside buildings near HV transmission lines, the field
strengths are typically lower than the unperturbed field by a
factor of about 10 - 100, depending on the structure of the
building and the type of materials (Manders & van Nielen, 1981).

Conductor height, geometric configuration, lateral distance
from the line, and the voltage of the transmission line are by far
the most significant factors in considering the maximum electric
field strength at ground level. At lateral distances of about
twice the line height, electric field strength decreases with
distance in an approximately linear fashion. Reference to typical
measured or calculated field contours in the vicinity of the line
(Zaffanella & Deno, 1978) indicates that, for a 525 kV transmission
line (height about 10 m), the field is always less than 1 kV/m at
distances of more than 40 m from the outer conductor, while for a
1050 kV line, which has much higher conductors, the 1 kV/m field
occurs at a distance of about 100 m from the outer conductor.
Typically, where a right-of-way (RoW) is used for a transmission
line of 500 kV or more, it varies from 35 to 70 m, so that electric
fields at the edge of the RoW are of the order of 1 kV/m.

The electric field strengths at and above ground level from
various transmission lines are shown in Fig. 4 (Gary, 1976). The
electric field distribution within various voltage substations is
given in CIGRE (1980).

3.3.3. Magnetic fields near transmission lines

Just as an electric field is always linked with the presence of
charges, a magnetic field always appears when electric current
flows. A static magnetic field is formed in the case of direct

current, whereas time-varying electric and magnetic fields are
induced in the vicinity of alternating current power transmission
systems.

The magnetic field beneath high-voltage overhead transmission
lines is directed mainly transversely to the line axis. The
maximum flux density at ground level may be either on the route
centre line or approximately under the outer conductors, depending
on the phase relationship between the conductors.

Apart from the geometry of the conductor, the maximum magnetic
field strength is determined only by the magnitude of the current.
The maximum magnetic flux density at ground level for the most
common overhead transmission line systems is approximately 0.1
mT/kA (Hylten-Cavalius, 1975).

In contrast to an electric field, a magnetic field is more
penetrating and very difficult to shield. It easily penetrates
human beings and, in the case of an alternating or rotating field,
induces circulating or eddy currents that are not conducted to
ground. The internal voltage differences induced within the body
by a magnetic field from power lines may be as high as 1 mV, if the
magnetic flux density reaches approximately 0.028 mT (Hauf, 1982).

The maximum ground level magnetic field strengths associated
with overhead transmission lines are of the order of 0.01 - 0.05 mT
and are also related to line height. Unlike the electric field,
they are also directly affected by the current carried by the line.
The magnetic flux density decreases in an approximately linear
fashion with distance from the conductor (Lambdin, 1978; Zaffanella
& Deno, 1978).

In principle, these magnetic fields can induce electric
currents in the body and could induce effects via the same
mechanisms as electric field-produced currents. However, for
exposures near a HV transmission line, the smaller magnitude of
these magnetically-induced currents (generally no more than 25% of
the electric field-induced currents) has resulted in little
emphasis on their contribution. The largest current densities
occur at the periphery of the body and they are lower inside. Fig.
5 and 6 show the magnetic field distribution near a HV-transmission
line carrying only about 10% of the typical rated load current for
such lines.

3.3.4. Man-Made ELF Fields in the Home, Workplace, and Public
Premises

In the home or workplace, ELF electric field sources occur at
electric wiring, appliances, and light fixtures, or industrial
electrical machines. Measurements of electric fields in a typical
American home (115 V circuits) ranged from less than 1 V/m to about
10 V/m, while fields measured at 30 cm from some appliances varied
from 2 to 5 V/m near a light bulb to several hundred volts per
metre near an electric broiler (Miller, 1974; Zaffanella & Deno,
1978). As a rule, values appear to be greater than 10 V/m near
appliances, and will vary with the nominal voltage.

Typical values of electric field strengths and magnetic flux
densities in the vicinity of home appliances and the potential
leakage currents through the body if contact is made with these
appliances is given in Appendix I.

3.4. Corona and Noise Effects of Transmission Lines

A high-voltage electrode can create ozone (O₃) by means of
ionization of air near the high-voltage conductors. In the vicinity
of corona-free, high-voltage, overhead transmission lines (fair
weather conditions), no ozone is created. Under certain weather
conditions causing corona discharges in the vicinity of HV
transmission lines, formation of ozone occurs. However, since
ozone is a very unstable gas, it rapidly decomposes into harmless
oxygen compounds in the open air and biological effects should
not be expected. Measurement and calculations of ozone near
transmission lines show that local increments in levels of the gas
are insignificant (Frydman et al., 1972; Roach et al., 1973).

Noise is of concern in regions beneath or near power lines, and
in switch-yards. Techniques are available to reduce corona-induced
noise beneath power lines to acceptable levels or to standards set
by law. The switch-yard acoustic environment is special and can
differ considerably from that near the overhead lines. Both the
frequency spectrum of the noise and intensity in different spectral
regions should be taken into account. Effects of noise in terms of
annoyance, sleep disturbance, and community reaction in the case of
HV transmission lines in the USA are discussed in detail by
Pearsons et al. (1979). Results of their study indicate that
transformer and transmission line noise may create problems,
particularly in densely-populated urban and suburban areas.

3.5. Electric Shock

In regions of high electric field strength, objects, if
insulated from ground, can assume large potential differences.
If the human body becomes the pathway for currents between such
objects, substantial electric currents (of the order of 1 mA) can
flow. Investigation of these effects has revealed two exemplary
cases in which significant shock currents exist:

(a) a long unearthed metal fence running parallel to the
line; and

(b) a large vehicle beneath the conductors and
effectively insulated from ground.

It is necessary to distinguish between the transient short-
circuit current and the steady-state short-circuit current.
Zaffanella & Deno (1978) presented data, obtained under various
circumstances, that indicated that peak currents of up to 20 A can
flow for a few microseconds when a person draws a spark discharge
from an object with a hand-held metal key. The peak currents are
an order of magnitude smaller if the finger is used for contact.
The energy content of spark discharges obtained from a carpet are

found to be similar to those in a 10 kV/m electric field with the
important exception that the AC field continually recharges the
electrified body so that repeated frequent sparks are possible,
whereas several steps must be taken on a carpet to recharge the
body. The steady-state short-circuit current that flows when the
charged object is earthed depends on the capacitance to earth of
the object, and the open-circuit voltage to which the object is
charged, when disconnected from ground, according to the relation:
I_sc = omegaV_oc x C, where I_sc is the short circuit current,
omega is the angular frequency of the electric field, V_oc is the
open circuit voltage, and C the capacitance to earth of the object
(Deno, 1974).

For human beings standing on the ground with arms at the side
of the body in an electric field of frequency f, the short circuit
current I_sc in amperes is given approximately by the empirical
formula (Deno, 1974):

I_sc = 9.0 x 10^-11 h² E x f

where f is 50 or 60 Hz, h is the person's height in metres (m), and
E the electric field strength in volts per metre (V/m). Thus, in a
10 kV/m, 60 Hz electric field, a person 1.7 m tall carries a short
circuit of about 160 µA.

Typical capacitances for objects range from 700 pF for a small
vehicle to several thousand pF for buses and large trucks and about
1000 pF for a 150 m fence (Deno, 1974). Thus, the short-circuit
current for a 150 m fence could be as great as 2.2 mA, if the fence
were located in a field of 5 kV/m. Zaffanella & Deno (1978)
measured the short-circuit currents of a farm tractor, jeep wagon,
and a school bus. In a 10 kV/m electric field, these vehicles
conducted 0.6, 1.1, and 3.9 mA of current to earth, respectively.
Although the shock currents are of appreciable magnitude, they
should not present a hazard if appropriate safety procedures are
followed. Good engineering practice to reduce the risk of shocks
includes the carefully earthing of fences, gutters, and other long
metallic objects in a strong electric field.

Data (Zaffanella & Deno, 1978) concerning human beings exposed
to spark discharges of various intensities showed that 50% of the
population perceived spark discharges in a field of 2.7 kV/m and
that 50% of the population found the spark discharges annoying at 7
kV/m. To obtain these data, persons standing in an electric field
touched a metallic post with a finger; it is assumed that their
capacitance was of the order of 170 pF.

The sensations that result from microshocks do not appear
hazardous (except insofar as they may produce a startle reflex that
could result in an accident), but they may be highly significant in
the evaluation of effects attributed to the fields. Although the
scope of this document does not include the possible health effects
of such microshocks or transient spark discharges, more than
cursory mention is given to these effects because of their
importance.

For the human responses, it is useful to define 3 thresholds:

(1) Perception: the minimum current for perception by touch
is about 0.4 µA;

(2) the Let-Go Current: the maximum current for which a person
can release the involuntary muscular contraction (Fig. 7)
(Dalziel & Lee, 1968);

(3) the Fibrillation Threshold: the minimum body-current to
cause ventricular fibrillation is especially dependent on
the pathway of the current in the body and the duration
(Fig. 8) (Kupfer, 1979; Kupfer et al. 1981). If the
current is directly applied to the heart, the fibrillation
threshold is about 5 x 10² times lower (Kupfer, 1982;
Weirich et al., 1983).

3.6. Interference of ELF Fields with Implanted Cardiac Pacemakers

An implanted pacemaker is an electromedical device that
artificially stimulates the heart, thus making it possible for
persons with certain heart diseases to lead relatively normal
lives. Although pacemakers may be susceptible to some forms of
electrical interference, hazardous situations resulting from
ambient electromagnetic fields have not been reported. Results of
a research programme reported by Bridges & Frazier (1979), who
carried out bench studies and studies with implants in animals,
showed a wide range in interference sensitivity among various
devices and for different arrangements of the implanted leads.

Pacemaker reversion can be brought about via the following three
mechanisms (Bridges & Frazier, 1979):

(a) direct coupling to an ambient electric field (typical
threshold range, 3 - 600 kV/m);

(b) transient coupling to an ambient electric field
through vehicle leakage current ("microshock")
(typical threshold range, 60 V/m - 60 kV/m);

Butrous et al. (1983) studied 35 patients (fitted with 16
different pacemaker models from 6 manufacturers) who were exposed
to 50-Hz electric fields up to a maximum of 20 kV/m. Current flow
measures varied between 15 and 300 µA, depending on the field
strength and the position of the patient in the field. Four
different response patterns were encountered: (a) normal sensing
and pacing in all conditions (one manufacturer); (b) reversion to
the fixed (interference) rate; (c) slow and irregular pacing; and
(d) mixed behaviour over a critical range of field strengths and
then reversion to a fixed rate. Their responses depended on the
pacemaker units. The field strengths required to induce such
behaviour varied with unit and model. Generally, the interference

threshold depended on the magnitude and distribution of induced
body current relative to the pacemaker, as well as field strength,
and thus varied with patient height, build, and posture.

4. MECHANISMS OF INTERACTION

Several mechanisms have been proposed to explain the reported
effects of ELF electric fields on laboratory animals, and in
tissues and cells in vitro including:

a) stimulation of peripheral receptors in the skin;

b) induced electric fields and currents inside the body
acting at the level of cells; and

c) non-specific stress.

4.1. Biophysical Mechanisms of Electric Field Interactions

Electric field coupling occurs through capacitive and
conductive modes. Energy is transferred to the object from the E
field and an electric charge in the object is put into motion. The
amount of charge involved depends on the size and location of the
object with respect to the E field. When a path to ground is
provided, the charge movement results in a current flow. If the
object is insulated from the ground, a potential develops with
respect to the ground, the magnitude of which depends on the
capacitance to ground.

The penetration depth of the field lines into the body is very
shallow at low frequencies. There is evidence that the field
induces direct effects on skin sensors of the cat paw at a
threshold local field strength of over 200 kV/m (Jaffe, in press).
Some reported behavioural effects in chickens, mice, rats, and pigs
exposed to unperturbed fields of 30 - 100 kV/m may be related to
sensory stimulation. These effects are presented in reports on
field perception, arousal, avoidance, transitory activity changes,
and transitory increases in cortecosterone levels (Moos, 1964;
Graves et. al., 1978; Hjeresen et. al., 1980; Sagan et. al., 1981;
Rosenberg et. al., 1983; Stern et. al., 1983).

With large field strengths, discharges may be detected. Small
currents flow within the body due to capacitive coupling to the
fields. In principle, an electric field of sufficient magnitude
could have a direct effect on biological tissues by acting directly
on the free ions in the extracellular milieu, on the charged
portions of the biomolecules, or by interaction with electric
moments of molecular electronic structure. However, the very small
internal electric fields that result from capacitive or magnetic
coupling (Barnes et al., 1967; Sheppard & Eisenbud, 1977) could not
affect covalent molecular structures or the electrostatic bonds
between molecules, nor could there be direct effects on steric
structure.

In his consideration of ELF electric field interactions with
neural cells, Schwan (1977) stated that, under a wide range of
assumptions for cellular shape and cellular electrical properties,
it was impossible that the largest electric fields in air could

significantly affect neural membrane potentials by the passage of
transmembrane currents. Schwan added, however, that the anomalous
properties at frequencies below 100 Hz, though still poorly
understood, "provide for more possibilities of subtle effects if
there are any at all".

Adey (1980) suggested that it is important to take into
account the possibility that one cell may influence another in
brain and other tissues through modulation of their shared electro-
chemical environment. The same author (Adey, 1981) proposed that
amplification of the weak initial stimulus occurs by a cascade of
intracellular processes taking place at receptor sites on the cell
membrane surface. This model may be supported by data on the
coupling of the parathyroid hormone receptors to the cyclic
adenylase and cyclic AMP responses in bone cells (Luben et al.,
1982).

It was suggested by Cain (1981) that voltage-sensitive ion
channels play a role at sufficiently large field strengths. He
proposed that an alternating potential across the cell membrane may
change membrane conductance by interacting with the charged groups
of the protein macromolecules that gate voltage-sensitive ion
channels.

Pilla (1980) developed a model for electrochemical information
transfer at membrane surfaces that involves a minimal electrostatic
perturbation of the molecular structure. The essence of the model
is that specific surface adsorption is expected to exhibit a
significantly longer relaxation time than dielectric or electro-
static interactions, due to the number of aqueous and membrane
steps involved, so that the characteristic time for adsorption may
be about 10 ms. This is in agreement with data obtained from toad
bladder membrane (Pilla & Margules, 1977). This mechanism would
work in parallel with the charge transfer processes already known
to occur, and could mediate enzymatic reactions to have significant
effects on cellular chemistry.

Recently, Schwan (1982a,b) discussed the possible role of
alternating field-induced ponderomotoric forces, i.e., forces
exerted by electric fields on nonpolar particles. The theory
developed on this basis can be used to explain dielectrophoresis
(Pohl, 1978), rotation, deformation, destruction of cells (Schwan,
1982a), and electrical cell fusion (Pilwat et al., 1981; Richter et
al., 1981) in cases where electric field strength greatly exceeds
that which could be produced in tissue by an environmental ELF
field.

4.2. Biophysical Mechanisms of Magnetic Field Interactions

The eddy currents created by magnetic ELF fields in the human
body cannot be measured directly, but they can be calculated and
confirmed by measurements on phantom models. The biological effects
of such induced electric currents are discussed above, but any
direct magnetic field effects are not well understood at present.

5. BIOLOGICAL EFFECTS IN CELLS AND ANIMALS

Since human volunteers cannot be used for studies that could
potentially cause harmful effects, biological investigations are
normally conducted using various other animal species. Studies
have been performed, mainly using rats and mice, but a wide variety
of other subjects, including insects, birds, dogs, swine, and non-
human primates, have also been used. A broad range of exposure
levels have been employed, and an equally large number of
biological end-points have been examined for evidence of possible
electric-field effects. Since all animal studies cannot be
discussed, this review will be limited to studies having some
bearing on health risk assessment. Experiments not discussed will
be summarized in the tables. Some studies showed effects from
exposure, and others showed no effects. There is general consensus
among scientists that exposure to electric fields produces
biological effects; however, more data are still needed to
determine whether these effects constitute a hazard.

Many studies have been performed based on the explicit or
implied hypothesis that because electrochemical processes are
involved in nervous system functioning, there might be an
interaction of the electric field with the nervous system. Such
hypotheses became of greater interest when initial reports on
linemen and switch-yard workers (Asanova & Rakov, 1966; Korobkova
et al., 1972) suggested the occurrence of a generalized alteration
in central nervous system function. Other studies were based on
generalized physiological hypotheses, such as the expectation that
electric field exposure, continued over a long period of time,
might induce a stress response, alter cardiovascular function,
affect immune responses, or alter various biochemical and
physiological variables, especially blood chemistry and blood cell
populations. Study areas briefly reviewed below also include
growth and development, reproduction, fertility, and behaviour.

Studies on the effects of electric and magnetic fields on the
ecosystems involving plants, invertebrates (including insects),
birds, fish, and mammals have been summarized (Lee et al., 1979,
1982). As these studies do not have a direct relevance for human
health risk assessment they are not discussed further.

5.1. Cellular and Membrane Studies

The effects of electric fields on in vitro systems have
been studied in a few laboratories. With these studies, it is
possible to use large sample sizes and to have a high degree
of control over experimental variables. Such studies also
provide a more direct investigation of the possible mechanisms
of interaction between a biological system and an electric
field. However, the most serious problems with in vitro
experiments are those of dosimetry and extrapolation. The
dosimetric relationship between exposure in cellular systems
and in whole animals is unclear, and extrapolation of results
from less complicated systems to human beings is extremely
uncertain.

Preliminary experiments using cultured Chinese hamster ovary
(CHO) cells exposed to 3.7 V/m showed no effects on cell survival,
growth, or mutation rate (Frazier et al., 1982). Cell-plating
efficiency, however (reflecting a possible alteration in the cell
membrane), was reduced in cells exposed to 60-Hz fields at
strengths greater than 0.7 V/m. At the same field strength (0.7
V/m) (Marron et al., 1975; Goodman et al., 1976, 1979), after
several months of exposure, slime mold showed frequency-dependent
effects on mitotic rate, cell respiration, and protoplasmic
streaming. These effects were observed with both electric fields
and magnetic fields, alone or in combination.

Studies using a variety of models (Greenebaum et al., 1979a,b;
Miller et al., 1979) have given contradictory results. Effects on
cell division, growth, and metabolism may appear at field strengths
of the order of tenths of a V/m or tenths of a mT in the medium.
On the other hand, electrical cell rotation and fusion (Pohl, 1978)
appear in the range of 10 - 100 kV/m.

Experimental findings suggest that the principal site of
interaction between ELF fields and the interior of living systems
is the cell membrane (Adey, 1975, 1977, 1980, 1981; Bawin et al.,
1975, 1978; Sheppard & Adey, 1979; Adey et al., 1981). These
include a 10 - 20% alteration in the calcium exchange from chick or
cat brain tissues exposed to ELF electric fields, either amplitude-
modulated radiofrequency (RF) carrier waves of 50, 147, or 450 MHz,
or ELF sine wave fields (Bawin et al., 1975, 1978; Blackman et al.,
1979, 1980, 1982). The calcium effect is windowed in frequency,
where maximal effects occur for 16 Hz modulation, and in the case
of direct ELF exposures, Blackman et al. reported several windows
at 15 Hz and its harmonics up to 105 Hz, in fields of less than 100
V/m in air. A similar narrow amplitude window limits the range in
field strength (Bawin et al., 1978; Blackman et al., 1979, 1982).
Bawin et al. (1978) found a relationship between the observed
effect and the ionic composition of the bathing medium.

In the case of the ELF modulation of a RF field, the magnitude
of the effective ELF field (obtained by demodulation of the RF
field envelope) that acts on the calcium-binding sites depends on
an unknown efficiency for a demodulation process occurring at an
unidentified site. Assuming complete demodulation, the effective
ELF field would correspond to an ELF-only field in air of the order
of 100 kV/m (Adey, 1981), though by use of the RF carrier there is
no significant heating of tissue (Tenforde, 1980) and no known
artifact (such as spark discharges).

A calcium efflux effect is also reported for in vivo studies
on the cat (Adey, 1980). Possible underlying biophysical mechanisms
and a relationship to the electric properties of the brain (electro-
encephalograph waves or EEG waves) are discussed by Grodsky (1976).
However, the physiological implication of the calcium efflux
phenomenon is not known.

Electric field effects on synaptic transmission and peripheral
nerve function in rats exposed for 30 days to a 60 Hz field of
effective strength 65 kV/m were studied in replicate (Jaffe et al.,
1980, 1981). The exposure apparatus was designed to eliminate the
confounding influence of electric shock currents. Neurons of the
superior cervical ganglion showed significantly increased
excitability compared with the control group, as determined from
tests in which the amplitudes of paired compound action potentials
were measured (conditioning test response or C-T response). None
of several other indices of neural function was altered to a
significant extent. The authors interpreted the data as evidence
of an effect on pre- or post-synaptic mechanisms, possibly indicating
enhanced excitability, and as evidence against a significant effect
on nerve conduction mechanisms.

An investigation (Wachtel, 1979) in which invertebrate
neurons from the sea hare Aplysia were exposed in vitro to a low-
frequency electric field indicated a strong frequency dependence in
response to extracellular currents that included synchronization
with the applied field. The neuron was most sensitive at frequencies
below 1 Hz, close to the natural firing rate of Aplysia neurons,
and for a particular neuronal orientation with respect to the field.
Other data were reported by Sheppard et al. (1980) concerning the
ELF field exposure of Aplysia neurons, including transient changes
in the firing rate and increased variability during exposure to
an electric field of 0.25 V/m rms. Episodic synchronization between
the neuron and the applied field was reported at 1.4 x 10^-4 A/cm²
(rms).

In a study by Bawin et al. (in press) on rat brain tissue
slices exposed to either 5- or 60-Hz electric fields at field
strengths in the range of the EEG, 1 - 10 V/m, evidence was
presented of long-lasting changes in neuronal excitability that
differed with field frequency and exposure duration. While 5-Hz
fields were generally excitatory, brief 60-Hz fields either
potentiated or depressed the tissue response following field
exposure, and prolonged 60-Hz fields depressed the response.
Although potentiations (believed to be due to an effect on synaptic
mechanisms) can last indefinitely (observations have lasted for as
long as 7 h), the depressed response after 60-Hz exposures was
transient, lasting about 10 min.

In summary, the results of in vitro studies suggest that
time-varying ELF electric fields may change the properties of
cell membranes and modify cell function. Several theoretical
explanations have been proposed (section 4), and it seems
conceivable that several parallel mechanisms exist. No
comprehensive and experimentally confirmed theory has been

proposed. Some of the effects observed on cells and tissues in
vitro can be detected in vivo.

5.2. Neurophysiological Studies in Animals and Animal Tissues

Blanchi et al. (1973) reported changes in the electroencephalo-
graph (EEG) patterns of guinea-pigs exposed for 30 min to a 100-
kV/m, 50-Hz electric field. Gavalas et al. (1970) noted EEG
spectral power peaks in the hippocampus, and less frequently in the
amygdala and centrum medianum, in all three monkeys exposed in 7-
and 10-Hz electric fields (7 V/m peak to peak). Others failed to
see any EEG alterations in chicks exposed at 40 kV/m (Bankoske et
al., 1976), and cats exposed at 80 kV/m (Silney, 1979). EEG
effects have not been reported in other studies.

Hansson (1981a,b) reported that Purkinje cells of the cerebella
of rabbits exposed to the 14-kV/m (50-Hz) field of an outdoor
substation or exposed in the laboratory showed pathological changes
in the cellular cytoskeleton and alterations in the concentrations
of two glial cell proteins (S-100, GFA). When young rabbits were
exposed to a 50-kV/m electric field for 6 months, no ultrastructural
changes were found in cerebellar cells, nor changes in several
plasma hormones (Portet & et al., 1984).

Jaffe et al. (1981) found a significant effect of field
exposure (30 days, 65 kV/m) on neuromuscular physiology for one
type of muscle (slow-twitch soleus), but not for another (fast-
twitch soleus).

The data from neurophysiological tests in vivo and in vitro
indicate that electric fields may have effects on tissues, especially
components of the nervous system. The physiological significance
for human beings exposed to environmental fields has not been
determined. Information is needed on the relationships between
biophysical and biological effects. In some in vitro studies,
the fields or current densities clearly exceed the values estimated
for internal fields or current densities in human beings exposed to
environmental fields.

5.3. Behavioural Studies

Among the most sensitive measures of insult to a biological
system are tests that determine modifications in the behavioural
patterns of animals. This sensitivity is especially valuable in
studying environmental agents of relatively low toxicity (Anderson
& Phillips, 1984). Behavioural studies in several species provide
evidence of field perception and the possibility that the fields
may directly alter behaviour. In rats, the threshold of detection
varies from subject to subject in the range of 4 - 10 kV/m with an
average level at about 8 kV/m (Sagan et al., 1981; Stern et al.,
1983). In mice, responses to a 35-kV/m field were reported
(Rosenberg et al., 1983); perception was seen in pigeons at
approximately 30 - 35 kV/m (Graves, 1977), and in pigs at 30 - 35
kV/m (Kaune et al., 1978).

Hjeresen et al. (1980) reported on field avoidance among rats
exposed at 75 - 100 kV/m (60 Hz). Preference for shielded areas at
night was found among pigs exposed at 30 kV/m (Hjeresen et al.,
1982). However, at 25 kV/m, rats preferred the field region during
the inactive phase (Hjeresen et al., 1980). Tests of aversion in
rats exposed to fields of 32 - 130 kV/m produced a complex pattern
of null effects in some cases (Creim et al., 1980) or positive
effects in others (Lovely, 1982), depending on the behavioural
test.

Alterations in rat activity were noted at 1.2 kV/m by Moos
(1964). Other studies on activity indicated transitory
increased response on initial exposure of rats or mice at 25 -
35 kV/m (Hjeresen et al., 1980; Rosenberg et al., 1983),
depressed activity in chickens exposed at 26 - 40 kV/m
(Bankoske et al., 1976; Graves et al., 1978), and increased
activity among bees exposed at 4.2 kV/m (Greenberg & Bindokas,
1981).

Tests with monkeys at 7 - 100 V/m exposed to frequencies
typical of the EEG (1 - 32 Hz) showed altered behavioural
reponses in an operant conditioning task (Gavalas et al., 1970;
Gavalas-Medici & Day-Magdaleno, 1976), while in other tests
involving exposure to magnetic and electric fields, behaviour
was unaffected (DeLorge, 1972, 1973). Feldstone et al. (1980)
observed minor changes in behaviour among baboons exposed to 30
kV/m (60 Hz).

Tests on the behaviour of cats exposed to ELF-modulated radio-
frequency signals were reported to show evidence of long-lasting,
frequency-specific changes in brain rhythms (EEG), and studies of
brain rhythms in rabbits exposed to ELF-modulated radiofrequencies
were also reported to show specific changes in the EEG (Takashima
et al. 1979).

Behavioural tests which most frequently showed an effect of
exposure were those relating to detection of the field or to
activity. Most other behavioural tests did not change with
electric-field exposure at field strengths up to 100 kV/m. Table 2
includes a summary of experimental results from nervous system and
behavioural studies in animals.

5.4. Sensory Phenomena

Strong electric fields cause hairs to oscillate. The movement
of hairs on the ear tips of swine was detected photographically in
60-Hz electric fields at 50 kV/m (Kaune et al., 1980); rat
vibrissae movement was observed in a 50 Hz, 50 kV/m field by
Cabanes & Gary (1981). Stern et al. (1983) attempted to examine
field sensitivity thresholds in nude or shaved rats, but saw little
difference from results with fur-bearing subjects.

Jaffe (in press) observed a direct field effect on mechanoreceptors
of the cat paw above a threshold local electric field strength of
220 kV/m.

Extraordinarily sensitive electroreceptive capabilities exist
in some species (e.g., Elasmobranch fish), particularly where
there has been evolutionary adaptation to refine sensory organs
(Kalmijn, 1966; Bullock, 1973).

Cues, including magnetic field direction, seem important in
birds (Walcott, 1974) and in several species ranging from bacteria
and bees (where ferromagnetic materials have been found) (Gould et
al., 1978) to dolphins and man (Blakemore et al., 1979), although
the data in man are disputed. These findings highlight the fact
that extrapolation of the results of experimental amimal studies to
man is quite complex. Allowances must be made for differences in
species sensitivities to ELF fields.

5.5. Effects on the Haematopoietic System in Animals

Numerous studies on animals (Blanchi et al., 1973; Cerretelli &
Malaguti, 1976; LeBars & Andre, 1976; Graves, 1977; Graves et al.,
1979; Marino & Becker, 1977; Cerretelli et al., 1979; Phillips et
al., 1979; Conti et al., 1981; Ragan et al., 1983) concern field-
related variations in blood cell populations. There is evidence in
these studies of a prompt effect on neutrophilic cells and possibly
an effect on thrombocytes and reticulocytes. The data do not permit
determination of possible mechanisms that may involve either an
effect of the internal fields directly on haematopoietic tissues,
or an effect on tissues affected via the central nervous system as
a result of peripheral sensory stimuli. In all cases, the changes
in peripheral leukocyte counts have been within the range of
physiological norms. A summary of studies on the haematopoietic
system in animals is presented in Table 3.

Table 2. Nervous system and behavioural studies in animals
--------------------------------------------------------------------------------------------------------
Exposure Frequency Subject Effects examined Reference
(kV/m) (Hz)
--------------------------------------------------------------------------------------------------------
0.0074 60 monkey no effect on operant behaviour deLorge (1973)

up to 45, 60, monkey altered behaviour (frequency specific) Gavalas-Medici &
0.056 75 Day-Magdaleno (1976)

0.01 - 7, 10 monkey changes in interresponse time, dose- Gavalas et al. (1970)
0.056 dependent, EEG entrainment at field frequency

0.1 60 rat no effect in preference behaviour or in deLorge & Marr (1974)
temporal discrimination

up to 0.1 45 mouse no effect on brain and serum sertotonin Krueger & Reed (1975)

up to 0.1 45 rat altered brain acetyl transferase Noval et al. (1976)

0.8 - 1.2 60 mouse more active in dark periods Moos (1964)

4.2 60 bees increased activity during exposure Greenberg & Bindokas
(1981)

up to 25 60 rat initial startle reaction Stern et al. (1980)

2 - 10 60 rat detection threshold approximately 8 kV/m Stern et al. (1983)

25, 50 60 mouse initial sterile reaction Graves (1977)

25, 50 60 rat preference for area of exposure Hjeresen et al. (1980)

26 60 chick peck suppression, 28% decrease in motor Graves et al. (1978)
activity

30 60 swine perception of field, prefer shielded area Hjeresen et al. (1982)
at night

30 60 rat no effect in taste aversion Creim et al. (1980)

30 60 baboon small behavioural changes Feldstone et al. (1980)
--------------------------------------------------------------------------------------------------------

Table 2. (contd.)
--------------------------------------------------------------------------------------------------------
Exposure Frequency Subject Effects examined Reference
(kV/m) (Hz)
--------------------------------------------------------------------------------------------------------
32 60 pigeon perception of field altered in exposed Graves et al. (1978)
animals

10 - 75 60 mouse transient hyperactivity in inactive phase, Ehret et al. (1980b)
35 kV/m average threshold Rosenberg et al. (1983)

40 60 chicks decreased activity in exposed animals Bankoske et al. (1976)
Graves et al. (1978)

50 50 mouse, no effect on behaviour Le Bars et al. (1983)
rat,
guinea-
pig

10 - 75 60 mouse hyperactivity with intermittent exposure Rosenberg et al. (1983)
(commencing at 50 kV/m)

67 60 chicken no effect on activity or gross behaviour Bankoske et al. (1976)

67 60 mouse hyperactivity with intermittent exposure Ehret et al. (1980a,b)

75, 90, 60 rat rats spend more time out of field Hjeresen et al. (1980)
100

80 60 cat EEG changes Silney (1979)

65^a 60 rat increased excitability of sympathetic Jaffe et al. (1980)
ganglion

65^a 60 rat no effect on peripheral nerve function Jaffe et al. (1980)

65^a 60 rat excitatory changes in neuromuscular Jaffe et al. (1981)
function; slower recovery from fatigue

100 60 rat aversion behaviour Lovely (1982)
--------------------------------------------------------------------------------------------------------
^a Effective field strength.
5.6. Cardiovascular Effects

Cardiovascular function can be assessed by measuring blood
pressure and heart rate and by performing ECGs. So far, reported
evidence of changes in cardiovascular function has been limited and
contradictory. In early studies, a decrease was reported in the
heart rate and cardiac output of dogs exposed to 15 kV/m (Gann,
1976), and an increase in heart rate in chickens exposed to 80 kV/m
(Carter & Graves, 1975). Comprehensive studies in rats showed no
effects from exposure to 100 kV/m (Hilton & Phillips, 1980).
Transient increases in blood pressure in dogs exposed to field
strengths greater than 10 kV/m have been reported (Cerretelli &
Malaguti, 1976).

5.7. Effects on Endocrinology and Blood Chemistry

A large body of data has been collected under different
exposure conditions on measurements of different blood plasma
proteins, enzymes, etc. Some of these data are summarized in Table
4. No consistent picture of physiological or pathological changes
is evident.

Many of the major hormones have been examined for the effects
of electric-field exposure, particularly in rats and mice (Phillips
et al., 1979). Possible effects have been observed in only three:
corticosterone, testosterone, and melatonin. Because corticosterone
is produced by the body in response to stress, blood levels of the
hormone are extremely sensitive to the method used in obtaining
samples. Perhaps because of this sensitivity (rather than the
effects of electric-field exposure), a number of laboratories have
reported conflicting results.

5.8. Effects on the Immune System

In considering the pattern of effects on white cell populations,
it is of special importance to evaluate the immunocompetence of
electric-field-exposed animals. Schneider & Kaune (1981) did not
find any effects on the response to infection in chicks exposed to
2 kV/m. Morris & Phillips (1982, 1983) did not find any effects on
cell-mediated or humoral immune response in rats or mice exposed to
fields of 0.2 kV/m. No effect was observed from electric-field
exposure on infectivity by a leukemogenic virus in chickens (Phillips
et al., 1981). Lyle et al. (1983), however, observed significant
decrements in the cytolytic capacity of lymphocytes exposed to radio-
frequency fields modulated at 60 Hz. In an extensive study, Le Bars
et al. (1983) found no significant effects on immune response of
rats, mice, or guinea-pigs exposed to 50 kV/m, 50 Hz electric fields
for 8, 14, or 18 h/day over periods varying from 1 to 6 months.

Table 3. Haematopoietic studies in animals
----------------------------------------------------------------------------------------------------
Exposure Frequency Subject Effects Reference
(kV/m) (Hz)
----------------------------------------------------------------------------------------------------
0.01 50 mouse altered leukocyte distribution Blanchi et al. (1973)

0.01 50 rat altered leukocyte distribution Blanchi et al. (1973)

0.01 45, 60 rat all effects within normal range Mathewson et al. (1977)

5 60 mouse decrease in RBC concentrations Marino & Becker (1977)

10 50 dog no effect on haematology Cerretelli et al. (1979)

10 50 dog no effect on haematology Conti et al. (1981)

25 60 mouse higher WBC count Graves et al. (1979)

25 50 dog altered leukocyte distribution, Cerretelli & Malaguti
RBC count and haemoglobin (1976)

50 50 rabbit altered total leukocytes and RBC LeBars & Andre (1976)

50 50 rat no effect on haematology LeBars & Andre (1976)

50 50 rat, mouse, no effect on haematology