Adrenal corticoid production can be influenced by EMFs, and the dynamics of the effect depend on many factors: field strength, frequency, duration of exposure, whether the exposure is continuous or intermittent, the ratio of the exposure to the nonexposure period in intermittent exposures, and the organism's predisposition. Since the adrenal-cortical response to EMFs is the same as that caused by known stressor agents (2), it follows that EMFs can also be biological stressors. Other endocrine organs that can be triggered by EMFs include the thyroid, pancreatic islets, and the adrenal medulla.

There are many important but unanswered questions. Where within the organism does the EMF-tissue interaction occur? What is the level of the interaction-organ, cellular, or molecular? What is the temporal sequence of events and the factors which influence it? Are the thyroid, adrenal, and pancreas particularly sensitive to certain types of EMFs, or are the changes in these organs reflective of an EMF interaction with more central structures-or both? Suppose, for example, that the thyroid is sensitive to a particular EMF: an EMF-induced change in thyroxine production would alter pituitary production of TSH, but measurements of thyroxine and TSH would not, in themselves, tell us either the location, level, or sequence of the interaction. Indeed, given the pervasive changes that can be induced by EMFs in the nervous system and the endocrine system-and in view of the intimate interconnection and synchronization of the two-there is a serious question concerning whether it is methodologically possible to demonstrate a specific causal sequence in many instances. The diversity of the reported effects suggests that EMF-induced changes in the endocrine system are mediated by the CNS. However, until now, most investigations have focused on the need to demonstrate an EMF impact on the endocrine system, and thereby to lay the foundation for more in-depth studies. Only Udinstev has even approached what might be called a systematic study of a particular EMF (200 gauss, 50 Hz). When other EMFs are studied systematically, perhaps it will be possible to delineate the sites and the level of the interaction (see chapter 9).

Most of the endocrine system effects seemed to be compensatory rather than pathological (see table 6.2 for example). But even though the homeostatic mechanism generally brought the corticoid level back to normal, it does not follow that the animal became physiologically equivalent to what it would have been at that point in time if it had not been exposed to the EMF. Animals that have been exposed to one stressor are known to have a diminished capacity to deal with a second simultaneous or contemporary stressor. Thus, animals that have accommodated to an EMF would, in general, be more susceptible to a second stress, compared to animals that experienced only the second stress.

There is, of course, a difference between the existence of an EMF-induced biological effect, and its detection in a given experiment. In our study, for example, the lack of a consistent statistically significant difference between the exposed and the sham-irradiated rats in each experiment sugggested that uncontrolled variables were present in the study. Possibilities include zoonoses, and genetic predispositions. This can cause individual animals, in an apparently homogeneous population, to react in completely opposite ways to the same EMF. In such cases there is no average response of the group to the EMF, despite the occurrence of individual responses. The most sensitive experimental paradigms for EMF research, therefore, do not rely on the comparison of group averages for the assessment of an effect.

Despite the difficulties with experimental design and interpretation, the evidence clearly indicates that exposure to EMFs can resuit in an activation the neuro-endocrine axis that is expressed in a general way as the stress syndrome.

Chapter 6 Index