Studies show that breastfeeding is ideal for infant nutrition and is an important means of spacing births. Breast milk provides some immunological protection to the infant, protects against infection and malnutrition, satisfies the infant's nutritional needs, and costs less than other feeding substitutes. Breastfeeding also protects against pregnancy, although the length of the contraceptive effect cannot be predicted. Analysis of breastfeeding trends and patterns shows that although most women in developing countries initially breastfeed their children, the length of breastfeeding is declining, especially in urban areas. The practice of breastfeeding in contrast appears to be increasing in some developed countries after several decades of decline. Breastfeeding trends and patterns may be influenced by the following sociodemographic factors: 1) urban-rural residence, 2) parental education and socioeconomic status, 3) mother's age and parity, 4) support from family and friends, 5) mother's employment, and 6) contraceptive use. Biological scientists and demographers are unanimous in concluding that although breastfeeding substantially contributes to birth spacing and fertility control in many areas, it is an unreliable means of contraception. Breastfeeding women are therefore advised to seek alternative means of contraception to avoid pregnancy. The effects of hormonal contraception on breastfeeding have not been established but the following points can be made: 1) combined estrogen-progestin oral pills decrease milk volume in some cases, 2) progestin-only contraception does not affect milk volume or increase it, 3) small doses of hormones used for contraception do not appear to prevent initiation of lactation, and 4) minute amounts of hormones are transmitted to the infant in breast milk and although no serious effects have been observed, the long-term effects are unknown. Determining the appropriate contraception during breastfeeding requires consideration of the woman's personal preference, availability and convenience of various methods, impact on fertility of starting different methods at different times after delivery, and possible effects on lactation. Implications for family planning programs and research needs are also discussed.