Abstract This paper describes some of the findings from a comparative study to investigate infant feeding practices and their determinants in four Third World urban areas: Bangkok, Thailand; Bogota, Colombia; Nairobi, Kenya; and Semarang, Indonesia. The information about developing country urban woman provided by these data allows examination of the interaction of feeding practices with socio-economic and biomedical variables. Through the use of descriptive, bivariate, and multivariate analytic techniques, it is possible to explore some of the questions which have been debated regarding infant feeding practices. Data addressing five major questions are described in this paper: 1. (1) Is breast feeding declining? 2. (2) Is bottle feeding making women breast feed less? 3. (3) Why do women use bottles? 4. (4) How do mothers get the idea of using bottles? 5. (5) How does paid employment affect infant feeding practices and the use of baby bottles? The study documents changes in infant feeding that can be expected to have detrimental effects for child health and for child spacing. Bottle use appears to interfere with breast feeding in all cultures, but more dramatically in more ‘modernized’ societies. Mothers resort to bottle use for a variety of reasons, but not usually as an attempt to wean. The health care system often provides the first contact between mothers and bottle use, and health care providers frequently encourage the use of artificial feeding. Women who work away from home early in their infants' lives must often use bottle feeding, but the percent of women affected is very small. Many more women use bottles and wean early than work away from home, and most artificially-fed babies do not have working mothers.