More recently, all available pathologic specimens, including microscopic slides and paraffin embedded tissues, and the pertinent records were assembled into a central archive at Brookhaven National Laboratory. This allowed reevaluation of these materials according to prescribed histopathologic criteria and in context with pre- and postoperative clinical data, findings at surgery, and gross pathologic descriptions and drawings. An ad hoc committee of pathologists,* most of whom had examined some material previously, and the surgeon** who had performed virtually all the thyroid operations since 1969, simultaneously reviewed sixty of these cases. Variable numbers of slides were available for review, depending spon the thyroid protocols used by the various surgical pathology groups. In some cases the small number of slides (as few as one to three) increased the possibility of missing occult carcinomas simply through sampling error. Microscopic slides were unavailable for review in three cases (Nos. 23, 54, 3074). Patients were identified by name and identification number but not by exposure group. Pathologists did not have access to their previous diagnoses, if any, until after conclusions were reached in each individual case. A slight modification (see Section IX.C.l.d, Histopathology) of the World Health Organization classification of thyroid tumors was adopted by the committee beforehand to ensure standardization of nomenclature. In addition, uniform morphologic criteria for distinguishing among categories were reviewed and approved. 3. Thyroid Function Evidence of mild hypothyroidism, designated "biochemical thyroid hypofunction," can often be foind in patients with thyroid enlargement by sensi- tive assays for serum TSH evels (111). Such patients may have normal serum T, concentrations (albeit i the lower range of normal) and yet have mild elevations in serum TSH and exa:gerated TSH responses to thyrotropin releasing hormone (TRH) infusion. Usually they are not hypothyroid clinically, since the thyroid dysfunction may be so mild that elevated TSH can maintain serum thyroid hormone concentrations in the normal or near normal range. Accord- ingly, the most sensitive test for evaluation of possible thyroid hypofunction is determination of serum TSH concentration. In some cases this may be supplementec by measuring the TSH response to TRH, though there is excellent correlation between basal and stimulated TSH (1,36). Prophylactic thyroxine supplement was discontinued in the Marshallese being tested at least 2 to 3 months and in some cases 6 months prior to testing for thyroid function, but strict adherence to this regimen could not be verified in all instances, an important consideration when evaluating the results given in Appendix IV, Table 2. Serum was obtained for measurement of TSH, Ty» and T; charcoal uptake | (thyroxine-binding globulin index, TBGI). This last test measures the fraction of tracer T, bound to charcoal after a 30-minute incubation, a value which is compared with that for a quality-control pool assayed simultaneously *Drs. L.V. Ackerman, W.A. Meissner, D.E. Paglia (University of California, Los Angeles), J.D. Reid, A.L. Vickery, and L.B. Woolner. *#Dr., B.M. Dobyns (Cleveland Metropolitan General Hospital). . - 56 =