99 Normal Hematologic Values Age, years 95% range’ (US, NHANESII) +2SDofmean? (Marshall Islands) 1-2 10,7-13.8 10.7~13.5 3-5 6-8 10.9-14.5 11.0-14.3 11.0-14.2 11.4-14.2 9-11 11.6-14.8 L1L.S-14.7 The US rangefor ages 1-2 years is for all races: for other age groupsthe values for white children are shown. 4 After exclusion of persons with abnormal values for transferrin saturation, high FEP or low MCV. ® After exclusion of persons with both an MCV > 5%from the mean and an Hb > 5%below the mean. bias resulting from the exclusion of some Hb levels above the mean in determining the normal range of Hb in children, the Marshallese data were analyzed with their inclusion. The analysis indicated the normal range (+2 SD from the mean) for Marshallese children wassimilar to the ranges established by Dallmanet al. [11] for children in the US (table IV). There are no previous reports on the hematologic status of Marshallese children. As indicated by a low MCVor high FEP,the prevalence of iron deficiency (approximately 3%) in the Marshallese children surveyed is similar to that derived from analysis of data collected on American children in the National Health and Nutrition Examination Survey II [NHANES IT; 11]. The prevalence of alow MCV or anemia and a higher age-specific mean FEP were greatest for ages 1-5 years, indicating that children in this age bracket are morelikely to be irondeficient. Although the study population represented nearly 90%of children residing on Rongelap and Utirik atolls and more than 3% of all Marshallese children, it is not known if the same prevalence of iron deficiency applies throughout the Republic of the Marshall Islands or Micronesia. The similarity of the Marshallese and US values (table IV) indicates that racial differences in norms for Hb for Marshallese and American children are probably notsignificant. These results differ from the small but well-documented differences in normal Hb levels of Blacks and Whites in all childhood age groups in the United States [12, 14, 15]. Although the 5012519 prevalence of anemia in Marshallese children in this study was not much different from that found in children living in the United States (table II], the distribution of age-specific Hb and MCVby island was not random. The children tested on Kwajalein atoll had both the highest mean Hb and lowest MCV. The significance of this finding is unknown. However,it could be explained if dietary folic acid deficiency were more commononthe otherthreeatolls. Only 5 of the 563 children (0.8%) had moderate to severe anemia (Hb < 10 g/dl). The values ranged from 8.8 to 9.8 g/dl. Four had a low MCV andof3 children tested all had an elevated FEP. Iron deficiency appears, therefore, to be the principal cause of the most severe as well as the most frequent anemias in Marshallese children. Acknowledgements The research forthis article has been carried out under the auspices of the US Department of Energy under contract No. DEACO2-76CH00016. The authors would like to thank Nanette Cheffers for the preparation of the manuscript, Capt. Keith Coberly and the crew of the Liktanur II, the Kwajalein-based Brookhaven National Laboratory medical team and John Reid and the technologists of the University of California Irvine Medical Center hematopathology laboratory for their cooperation. References Gilles, H.M.: Normal haematological values in tropical areas. Clin. Haematol. 10: 697-706 (1981). 2 US Department of Commerce, Bureau of the Census: Census of the population, 1980. General population characteristics: trust territory of the Pacific islands excluding the Northern Mariana Islands (1980). Craib, J.L.: Micronesian prehistory: an archeological overview. Science 219: 922-927 (1983). 4 Conard, R.A.; Paglia, D.E.; Larsen, P.R., Sutow, W.W.; Dobyns, B.M.; Robbins, J.; Krotosky, W.A.; Field, J.B.; Rall, J.E.: Wolff, J.: Review of medical findings in a Marshallese population twenty-six years after accidental exposure to radioactive lad Table IV. Comparison of normal ranges of Hb (gm/dl) for children in the US [NHANESII; 11] with the suggested normal ranges for the Republic of the Marshall Islands fallout. Brookhaven National Laboratory, Publication No. 51261 (1980), 5 Adams, W.H.; Harper, J.A.; Rittmaster, R.S.; Heotis, P.M.; Scott, W.A.: Medical status of Marshallese accidentally exposed to 1954 Bravo fallout radiation. January 1980 through December 1982. Brookhaven National Laboratory, Publication No. 51761 (1984). 6 BMDP statistical software manual: revised ed. (University of California Press, Berkeley 1983).