18 Marshallese more nearly resemble the blood groupingsof the people of the Western Islandsof Indonesia than those of the Amerindians. Too great a generalization of population origin, however, cannot be drawn from sucha relatively small study group. Haptoglobin studies. Results of the haptoglobin studies are taken from a report of Blumberget al.” With the starch gel method of electrophoresis there are three distinct patterns of haptoglobins." In type 1 —1, there is one haptoglobin travelling near the beta globulin region; in type 2 —2, there are three haptoglobins travelling between the alpha and beta globulins; and in type 2—1, there are four haptoglobins. Oneof these appears to be the same as in type 1—1 and the other three migrate slightly further than the three haptoglobins of type 2—2. It has been shownthat these hapto- globin patterns are genetically determined, and they also appear to have a physiological function associated with hemoglobin transport.**-*9 Occasionally an individual is found who has no haptoglobin at all. It is not yet clear if this type is genetically or environmentally determined. In the Marshallese, one such individual was found, and he was excluded in computing the phenotype frequencies. The actual incidenceof the haptoglobin types wasas follows. Tyre No. % 1-1 2-1 2-—2 0-0 51 54 20 1 40.48 42.86 15.87 0.79 126 100.00 Total The Marshall Islanders have a very high incidence of the | —1 type of haptoglobin, and incidence of the H,' gene exceeded only by that of the Yorubas of Nigeria.*° The Alaskan Eskimos have a low incidenceof this type. There appears to be a primafacie relationship with latitude; populations from areas near the equator havea high incidence of | —1 and a low incidence of 2—2, personsin high latitudes the reverse, and populations from temperate climates intermediate values. Thereis, of course, no evidence that the relation- ship is more than coincidental. Anotherpossible correlation is diet; the Artic people live primarily on a protein diet (sea mammals, fish, and caribou), while the Nigerians have a diet with a high carbohydrate content. Further studies are required to confirm or reject this association. The hapto- globin system represents one of the best examples of genetic polymorphism yet described in humans. A studyofits incidence in various people may help in elucidating the selective factors that maintain its balance in the population. Hemoglobin types. In the 45 Marshallese blood samples analyzed, hemoglobin A, was found to be a little more diffuse than in the control samples. There was also an increased smearingofthe hemoglobin behind the A, zone. However, the hemoglobins were considered to be normal, and the abovefindings were believed to be due to possible denaturation which caused increased diffusion and smearing of the hemoglobin during electrophoresis. Serological and Urine Studies. Serum proteins. Asnotedin the past, the serum protein levels were elevated in both exposed and unexposed people. The meanlevel of the exposed group was 7.6, and of the unexposed group 7.8 g. Electrophoretic studies last year showed the gammaglobulin to be elevated. Tests of thyroid function. The following was reported by Dr. J.E. Rall. The level of serum protein-boundiodine in both groupswassignificantly above the normal range (see Table 9). The Normal PBI in man in the United States by this technique ranges from 3.5 to 7.5 ug%. There appeared to be no correlation of PBI with either age or sex. The ages of those examined ranged from 6 to 83. The total iodine of these sera averaged 1.0 ug% above the PBI, which is a normal value and anindication that contamination with iodine was not contributing to the elevated PBI values. To determine whether the PBI mightconsist of other organic but notthyroidaliodine, butanol extractable iodine tests were performedon6 sera. These values were at the upperlimit of normal (normal range 3.2 to 6.4 ug%). To delinate further these findings, thyroxine-binding capacity studies were performed on selected sera by methods previously described.’ In 12 cases the values averaged 0.26 pg thyroxine/ml serum and ranged from 0.186 to 0.32. The normal value is 0.20 pg/ml, but inadequate data are available to define precisely whether the values in these Marshallese individuals were significantly different from nor- mal. It can, however, be calculated that the level of thyroxine-bindingprotein is insufficient to cause an elevation of serum thyroxine (presumably to maintain a normallevelof free thyroxine) as seen in these subjects. Contamination ofthe sera with