Six healthy lactating women ranging in age from 28 to 35 years were selected to participate in the study. The women had been nursing their infants from 3 1/2 weeks up to 17 weeks. On the morning of the study each participant received 100 mg dose of caffeine. The oral dose of caffeine was taken with a glass of water on an empty stomach. Blood samples were collected from zero time to 24 h after caffeine administration. Each subject aseptically collected breast milk from the right and left breasts by manual expression or using a mechanical breast pump, from zero time to 24 h after caffeine administration. Caffeine in plasma and breast milk was determined by gas liquid chromatography using nitrogen phosphorous detection. Caffeine was rapidly absorbed producing plasma time to peaks ranging from 0.50 to 1.00 h, with values for peak concentrations between 3.60 and 6.15 micrograms ml-1 in plasma. In breast milk, time to peak ranged from 0.75 to 2.00 h and concentration values between 1.98 and 4.30 micrograms ml-1. No significant differences were found between attainment of the time of the peak in plasma and breast milk, p greater than 0.05. In addition, peak concentration for the right and left breast were not statistically different, p greater than 0.05. However, significant differences, p less than 0.05, between plasma and both breasts in regards to the peak concentrations were found. The overall breast milk/plasma ratio obtained from the respective area under the curves were 0.815 +/- 0.0521 and 0.809 +/- 0.202 for the right and left breast, respectively, and no statistical differences were observed between the right and left breasts, p greater than 0.05. In summary, caffeine is rapidly transferred to breast milk. Multiple dose studies are needed to provide more information about caffeine concentration in breast milk of women who receive multiple doses of caffeine or who chronically consume caffeine.