It is generally recognized that low birth weight can be caused by many factors. Because many questions remain, however, about which factors exert independent causal effects, as well as magnitude of these effects, a critical assessment and meta-analysis of the English and French language medical literature published from 1970 to 1984 were carried out. The assessment was restricted to singleton pregnacies of women who lived at sea level and who had no chronic illnesses. Extremely rare factors were also excluded, as were complications of pregnancy. In this way, 43 potential determinants were identified. A set of a priori methodological standards were established for each potential determinant. Studies that satisfactorily met (SM) or partially met (PM) these standards were used to assess the existence and magnitude of an independent causal effect on birth weight, gestational age, prematurity, and intrauterine growth retardation (IUGR). A total of 921 relevant publications were identified, of whihc 895 were successfully located and reviewed. Factors with well-established direct causal impacts on intrauterine growth include infant sex, racial/ethnic origin, maternal height, pre-pregnancy weight, paternal weight and height, maternal birth weight, parity, history or prior low-birth-weight infants, gestational weight gain and caloric intake, general morbidity and episodic illness, malaria, cigarette smoking, alcohol consumption, and tobacco chewing. In developing countries, the major determinants of IUGR are Black or Indian racial origin, poor gestational nutrition, low pre-pregnancy weight, short maternal stature, and malaria. In developed countries, the most important single factor, by far, is cigarette smoking, followed by poor gestational nutrition and low pre-pregnancy weight. For gestational duration, only pre-pregnancy weight, prior history of premature or spontaneous abortion, in utero exposure to diethylstilbestrol, and cigarette smoking have well-established causal effects, and the majority of prematurity occurring in both developing and developed country settings remains unexplained. Modifiable factors with large effects on intrauterine growth or gestational duration should be targeted for public health intervention in the two settings, with an emphasis on IUGR in developing countries and prematurity in developed countries. Future research should focus on factors of potential quantitative importance for which data are either unavailable or inconclusive. In developing countries, the most important of these for intrauterine growth are caloric expenditure (maternal work), antenatal care, and certain vitamins and trace elements. For prematurity, especially in developed countries, factors deserving further study include genital tract infection, antenatal care, maternal employment and physical activity, and stress and anxiety.