Abstract: Using results from the World Fertility Survey (WFS) for 28 countries, socioeconomic differences in neonatal, postneonatal, and child mortality were examined. To maintain some degree of comparability and to make presentation of the results feasible, focus was on 5 variables which are available for each survey. It can be argued that each of the 5 socioeconomic variables considered here--mother's education, mother's work status since marriage, current or most recent husband of mother's occupation and education, and current type of place of residence of mother--affects infant and child mortality, although often as surrogates for other variables which were usually not directly available. For over 24 countries, the neonatal mortality rate varied from 84 in Nepal to 15 in Malaysia. In Nepal the rate for children of the skilled and unskilled was high (124) but where the husband had received 7 or more years of education the rate of 54 was low. At the other extreme, rates in Malaysia varied from 5 when mother's had 7 or more years of education to 23 for offspring of the least educated husbands. The highest overall postneonatal rate of 89 was again found in Nepal and the lowest national rate in Trinidad and Tobago at 13. In 9 out of 24 countries the high values were over 3 times as great as the low values and the absolute difference exceeded 30/1000 in 13 countries. Differences on child mortality are substantial, reflecting the greater influence of socioeconomic factors on mortality in early childhood. Nationally, the values ranged from 186 in Senegal to a low of 8 in Trinidad and Tobago. In only Haiti, Guyana, and Pakistan did the ratio of the maximum to the minimum rates for sizeable groups fall below 2. At the other extreme, in 5 countries the ratio exceeded 10 and in a further 6 was above 4. Differences between the high and low groups within countries exceeded 30 in 18 out of 28 countries and were over 50 in 10 of these. In 9 countries the highest rates occurred among mothers with no education and in a further 6 among husbands with no education. Education of mother, followed by education of her husband and his occupation were generally the strongest explanatory variables. The work status of the mother was not likely to be an important explanatory variable in these analyses. Results of a multivariate analysis suggested intriguing differences in the relative roles of different socioeconomic variables. Mother's education seemed to play an important role in determining children's chances of surviving in several Latin American and South East Asian countries. In no country did husband's level of education appear in all 3 models. The occupation of the husband was possibly the purest indicator of socioeconomic status, and this factor appeared in the models for all 3 segments of infant and child mortality. Mother's work status appeared least often.