This study assesses the operational feasibility and validity of a unique telephone-administered 24-h dietary recall methodology. Dietary interviews were conducted by telephone with 204 low-income elderly subjects. Food portion size estimation was done using a two-dimensional food portion visual aid the that was mailed to respondents' homes. The mean intakes and distributions of nine nutrients among males and females in the research sample were compared with those reported in the 1971 to 1974 Health and Nutrition Examination Survey. After correcting for time trends in nutrient intake, the mean intake values found in the survey were very similar to those reported by the Health and Nutrition Examination Survey. In testing the frequency distributions of intakes, only two nutrients for women and one for men had statistically significant differences. The telephone-administered dietary methodology thus produced acceptable estimates of the means and distributions of nutrient intakes among groups of individuals and has the potential of markedly reducing the cost, time, logistical, and personnel constraints associated with nutrition surveys.