Page 2

cable with available information and study resources, identify particular chemical substances, geographic areas, contaminant sources, human subpopulations, and fish and wildlife populations of special concern with respect to HAAs; and (3) if possible and warranted, suggest general approaches for identifying and mitigating toxicologic problems.

The charge to the committee did not include, nor did the committee attempt to evaluate, risk-management policy options. Due to constraints of time and resources, the committee also did not consider approaches for mitigating toxicologic problems.

The Committee's Approach and Difficulties Encountered

To evaluate the endocrine-disruptor hypothesis, the committee's approach involved ( ) identification of chemicals with hormonal or antihormonal activity: (2) evaluation of scientific literature on the effects in both the adult and the developing organism associated with those chemicals in vertebrates—humans. laboratory animals, and wildlife; and (3) consideration of whether the effects can be attributed to the hormonal properties of the chemicals and environmental exposure to them.

The committee focused its attention on compounds that have been reported to induce reproductive changes, developmental defects, neurobehavioral abnormalities, immunologic deficits, carcinogenesis, and ecologic effects.

It became clear as the work of the committee progressed that limitations and uncertainties in the data could lead to different judgments among committee members with regard to interpreting the general hypothesis, determining appropriate sources of information, evaluating the evidence, defining the agents of concern, and evaluating environmental and biologic variables.

Some of the differences reflect areas where additional research would help; others reflect differing judgments about the significance of the existing information. The differences are not confined to this committee but are reflected in the scientific community at large. Some differences appear to stem from different views of the value of different kinds of evidence obtained by experiments, observations, weight-of-evidence approaches, and extrapolation of results from one compound or organism to others, as well as allowable sources of information and criteria for arriving at meaningful conclusions and recommendations.

Other difficulties arise from questions about the observed effects: for example, is human sperm concentration really in decline? In other cases, the effect is clear (e.g., developmental abnormalities in some wildlife species), but the cause is in question. Often, organisms are exposed to many environmental chemicals as well as other environmental and ecologic perturbations. In addition, one must recognize that under multiple exposures, there is the potential for interaction among agents. These factors make assigning cause to specific chemicalscontinue