Abstract An experiment was performed on the maternity wards of three public hospitals in Cebu City, Philippines to determine whether the distribution of free samples of infant formula reduced the likelihood that mothers would breast feed or caused mothers to terminate nursing early. Samples were given or withheld alternately for 2 week intervals to mothers as they left maternity wards. They were followed for 8 months in the first experiment ( N = 273) and for 2 months in a replication ( N = 284). We found that there were no statistically significant differences between those who received samples and those who did not in initiation or maintenance of breast feeding. Mothers in both groups frequently turned to mixed schedules, but these varied from day to day depending on money to buy other forms of milk, or on the mother's health, or her plan to be away from the baby for one or more feeding periods. After the baby reached an age of 2–3 months, mothers, with few exceptions, used diluted sweetened condensed milk as a supplement and/or substitute for their own milk. It was found that, while mothers recognize the nutritional, economic and health benefits of breast feeding, they may terminate early on the basis of folk beliefs. Receiving formula samples, however, had no measured effect on their breast feeding practices.