7 showed the increase in proteins was largely due to an increase in the gammaglobulin fraction. The reason for this is not apparent. Numerous chronic infections may be an expianation. Sodium levels in- the urine and food indicated about the same consumption of NaCl as in Americans. The generally lower incidence of hypertension in the Marshallese might be related to the fact that the former native diet was probably lower in salt content than the present, more westernized diet. It will be interesting to see whether the incidence of hypertension will later increase. Serum cholesterol levels (1957, 1959) were some- whatlowerin the exposed population than in the comparison or Utirik populations, but were in the low normal range. No abnormally low readings were noted. Serum creatimne levels (1957) were in the normal range with no abnormal levels noted. Serum vttamin B,, concentrations (1958, 1959) were generally significantly higher than American Immunoelectrophoretic analysis showed neither a Pparaproteinemia nora typical picrure of antibodydeficiency-syndrome, but a high frequencyofincreases of some of the immunoglobulins was noted. Blood volume studies with Cr*'-labeled sodium chromate showeda significant reduction in red cell mass and/or plasma volumein 15 of 23 Marshallese. DIFFICULTIES ASSOCIATED WITH THE EXAMINATIONS As mentioned in previous reports, several diffi- culties were associated with carrying out the ex- aminations as well as interpreting the findings. 1. The language barrier made examinations difficult, since very little English is spoken by the Marshallese. However, there were sufficient Eng- lish-speaking Marshailese to assist the medical team in most instances. 2. The lack of vital statistics or demographic levels. The possibility of contamination of the sam- data on the Marshallese imposed a serious diffi- considered, since myeloproliferative and liver dis- cal data. Records of births, deaths, etc., have been ples with bacteria producing vitamin B,, must be eases were not seen. culty in interpretation and evaluation of the medi- Serum protein boundiodine levels (1957, 1959, 1962) made bythe health aides or magistrates of the villages and supposedly forwarded to the district ad- roid dysfunction was not apparentin the people. complete or jost in most instances, and vital statis- were generaily slightly elevated. Evidence for thy- ministrator; however, such records have been in- people (1 exposed and 7 unexposed). An increased tics are therefore inadequate. Trust Territory officials are now attempting to assemble such data. 73% of the people to be infected with various culties in interpreting some of the studies to be Glucosunia and elevated blood sugar were found in 8 incidence of diabetes is prevalent in the Marshallese people. A survey for intestinal parasites (1958) showed types.'' For the three major pathogens found, the over-all infection rates were, for Entamoeba histo- fytica, 18.2%; for hookworm, 5.5%; and for Trichurts trichiura, 34.3%. Eosinophilia >5% has consistentiy been noted in about haif the people. The fact that half the cases with eosinophilia showed no helminthic infections at all suggests that other factors besides parasitic infections must be responsible. The eosinophilia may be related to chronic fungus and otherinfections, particularly ofthe skin. Complementfixation studies for parainfluenza1, 2, and 3, respiratory syncitial, psittacosis, and Q fe- ver showed antibodies to all groupsof viruses except that for Asian influenza, which probably had not yet seriously involved the people of the Marshall Islands. The antibody titers appeared to be somewhat lower in the exposed people. 3. There is uncertainty on the part of some of the Marshallese as to their exact ages, particularly among the older group. This imposes certain diffi- outlined. COMPARISON POPULATIONS During the first 2 years, two separate groups of Marshallese people were used for comparison, each of comparable size to the exposed Rongelap group and matched for age and sex. However,this population was found to be unstable, with a large attrition rate over the 2 years, which madeit unsatisfactory. At the timeof the 3-year survey, it was found that during the preceding 12 months the Rongelap population at Majuro Atoll had doubled because of the influx of relatives who had come back from otherislandsto live with them. These people had been awayfrom Rongelap Atoll at the timeof the accidental exposure. This group matched reasonably well for age and sex and was