A methodology for the quantification of vitamin A in human milk was developed and validated. Vitamin A levels were assessed in 223 samples corresponding to the 5th, 6th and 7th postpartum months, obtained in the province of Santiago del Estero, Argentina. The samples (500 microL) were saponified with potassium hydroxide/ethanol, extracted with hexane, evaporated to dryness and reconstituted with methanol. A column RP-C18, a mobile phase methanol/water (91:9 v/v) and a fluorescence detector (lambda excitation 330 nm and lambda emition 470 nm) were used for the separation and quantification of vitamin A. The analytical parameters of linearity (r2: 0.9995), detection (0.010 microg/mL) and quantification (0.025 microg/mL) limits, precision of the method (relative standard deviation, RSD = 9.0% within a day and RSD = 8.9% among days) and accuracy (recovery = 83.8%) demonstrate that the developed method allows the quantification of vitamin A in an efficient way. The mean values + standard deviation (SD) obtained for the analyzed samples were 0.60 +/- 0.32; 0.65 +/- 0.33 and 0.61 +/- 0.26 microg/ mL for the 5th, 6th and 7th postpartum months, respectively. There were no significant differences among the three months studied and the values found were similar to those in the literature. Considering the whole population under study, 19.3% showed vitamin A levels less than 0.40 microg/mL, which represents a risk to the children in this group since at least 0.50 microg/mL are necessary to meet the infant daily needs.