The in vitro incorporation of inorganic (32)P into erythrocyte phospholipids has been studied in normal subjects and in splenectomized patients with hereditary spherocytosis (HS). Phosphatidic acid (PA) was the only lipid measurably labeled in both kinds of cells. The actual turnover rate of PA phosphate was determined by simultaneously isolating inorganic phosphate (P(i)) and adenosine triphosphate (ATP) and determining their specific activities. This turnover is very small: 1.3 mumoles P/liter of erythrocytes per hr in normal cells and 4.0 mumoles P in HS erythrocytes when either ATP or cellular P(i) is considered the immediate precursor. This value represents less than 0.1% of the total membrane lipid phosphate. Incorporation of added (32)P(i) into the other phosphatides, including phosphatidyl serine, was essentially zero in both kinds of cells. The effects of stimulation and inhibition of active cation transport, metabolic depletion, and extracellular phosphate concentration on both the degree of labeling and the actual turnover of PA phosphate were studied. In any given experiment, the degree of labeling of PA depended on the specific activities of the other intracellular phosphates (P(i) and ATP). The actual turnover rate of PA phosphate, however, did not vary with active transport or metabolic depletion. The greater turnover of PA phosphate in HS erythrocytes may be due to the somewhat younger age of these cells. The results suggest that the very low turnover of PA phosphate in erythrocytes is mediated by nonspecific enzyme reactions, and that it is quantitatively insignificant in both normal and HS erythrocytes. The results also emphasize the importance of measuring intracellular phosphate precursors in any study evaluating cellular phospholipid turnover from added (32)P(i).