Summary. Accumulation of neutral amino acids by isolated chick epithelial cells has been studied and the results discussed in terms of the ion gradient model, and a model invoking a direct input of metabolic energy. The cells establish four-to eightfold concentration gradients of amino acids at an extracellular concentration of 1 mM. The accumulation is sodium-dependent, inhibited by high extracellular potassium concentrations, and is sensitive to a variety of metabolic inhibitors. Also, amino acid uptake is depressed by actively transported sugars, and certain other amino acids, and is stimulated by phloridzin.Cells equilibrated with valine and loaded with 30 to 40 mM intracellular sodium begin immediately to actively accumulate valine when suddenly introduced to media containing 20 mM sodium. The cells establish a threefold gradient of amino acid during the interval when intracellular sodium is higher than extracellular sodium.Amino acid accumulation and 22Na efflux were monitored simultaneously in cells treated with phloridzin. While phloridzin causes a 30% stimulation of amino acid uptake, no variation in the rate of 22Na efflux or the steady-state level of 2ZNa maintained by the cells can be detected. Similarly, either 2.5 mM glucose or 2.5 mM 3-O-methylglucose cause approximately a 50 % inhibition of 1 mM valine uptake, but no detectable change in steady-state cellular 22Na content. Several aspects of the data seem inconsistent with concepts embodied in the ion gradient hypothesis, and it is suggested that a directly energized transport mechanism can better accommodate all of the data.The mechanism by which the cell membrane of intestinal epithelial cells transduces metabolic energy and utilizes it for active accumulation of sugars and amino acids has been the subject of a wide variety of independent investigations. Basically, two different views have been presented with regard to the nature of the energy input for intestinal nonelectrolyte transport events. Models of the type first proposed by Riggs, Walker and Christensen [22] and Crane and his colleagues [3-5] envision a mechanism in which 1 J. Membrane Biol. 12