is a long-term average recorded by a measuring device worn by a subject. Other measurements, such as the peak field or field exceeded 10% of the time, are used occasionally.
Transmission-line measurements correlate well with calculated fields, but they only reflect line currents during the measurement period, which might or might not be typical for the power line. Measurements are the only practical way to determine the field from complex sources, such as distribution lines, appliances, and ground currents. A major criticism of contemporary magnetic-field measurements is that they might not reflect conditions accurately that prevailed years before—during the period the disease developed.
Occupational studies typically rely on job title as an indicator of a subject's magnetic-field exposure or on magnetic fields measured at representative work locations (either personal exposure or spot measurements) and combined with each subject's work history. Both methods are likely to result in misclassifications because of the large overlapping range of fields measured for a given job title (e.g., see the measurements reported by Theriault et al. 1994).
Electric Fields The most commonly used exposure apparatus consists of parallel plates between which an alternating voltage (50 or 60 Hz, or other frequencies) is applied to produce the electric field. Typically, one of the plates (bottom) is grounded. When proper dimensions of the plates are selected (large area in comparison to their separation), a uniform field can be produced within a reasonably large volume between the plates. Distribution of the electric-field strength within a volume of interest can be calculated. The field uniformity deteriorates close to the plate edges.
Electric fields of up to 100 kV/m have been used in experiments with animals. This field strength for a rodent corresponds to about 10 kV/m for a human, as described in the following section on in vitro studies.
The original reasonably uniform field in an animal-exposure system can be significantly perturbed by two factors. One factor, which is unavoidable but controllable, is due to the presence of test animals and their cages. A considerable amount of information is available on proper spacing of animals to ensure the same exposure field for all test animals (Kaune 1981a) and to limit the mutual