T HE estimated probabilities of fetal mortality reported in this paper and the discussion of problems of obtaining and analyzing data on fetal losses in early pregnancy are based on a followup study of pregnancies on the island of Kauai during the 4-year period 1953-56 (1, 2). While this paper is concerned with measuring the risk of fetal mortality, the followup study of which it is a part uses a broad epidemiologic approach to the investigation of many conditions that may be associated with the qualitative as well as the quantitative outcome of pregnancy. It is hoped that our analyses, which are still in progress, and the studies of others engaged in field and laboratory investigations will throw additional light on the nature and relationships of the many factors which determine whether the outcome of pregnancy will be death of the conceptus, congenital defect, or healthy development. When estimating the risk of loss for the successive stages of pregnancy, it is important to keep in mind that different methods of investigation are required for different periods of gestation. For the latter part of pregnancy, customary fetal death ratios, based on registered fetal deaths and live births, can be used with a good deal of confidence. For example, it was estimated that in 1956 in the Scandinavian countries 1,000 women whose pregnancies had progressed to the 28th week of gestation would have 17 fetal losses (3). For women in Neiw York City reaching the same stage of pregnancy in 1958 the chances of fetal loss were 14 for the white and 19 for the nonwhite population, excluding fetal deaths of unknown length of ge,station, according to a special tabulation prepared for this study by the U.S. National Office of Vital Statistics. The farther we push back toward the beginning of antenatal life, however, the more limited knowledge becomes. Studies of loss and damage up to the first missed menstrual period are in the domain of embryologists using highly specialized laboratory methods. (This ovular period corresponds to the first 2 weeks of fetal life, from fertilization to about a week after implantation.) During the month after the first missed menses, the 4-7 week period of gestation, pregnancies can be recognized clinically. But it is unrealistic to assume that many women in a community will seek medical care this early in pregnancy. However, since some women will recognize a fetal loss during these early weeks, field studies such as the Kauai Pregnancy Study can add important information. From the eighth week of gestation on, more and more fetal losses come to the attention of physicians. Nevertheless, efforts to obtain reports prior to the 28th week of pregnancy have so far met with little success except in a few areas. The Kauai Pregnancy Study (KPS) was designed to provide information about early pregnancies, from 4 weeks' gestation on, which is Dr. French is a biostatistician and Dr. Bierman is professor of maternal and child health, School of Public Health, University of California, Berkeley. The study was financed through a grant from the Children's Bureau and Public Health Service National Institutes of Health grant No. 7734. Data for the study were collected through the cooperative efforts of the School of Public Health, University of California, and the Hawaii Department of Health.