Previous studies have suggested that there are two characteristic storage lesions that develop during storage of PC at 22 degrees C with constant agitation in containers constructed from PVC plastic. One occurs only in PC with high platelet count in which there is a fall in pH due to hypoxic conditions within the container. This storage lesion is characterized by an increase in apparent mean volume as determined by Coulter Counter and a change in platelet shape from disc to sphere. The other is observed in PC with low platelet counts; in these, a major pH fall does not occur. Under these conditions, there is a decrease in the apparent mean volume and an increase in the dispersion of the size distribution determined by Coulter Counter. The cause of this storage lesion is uncertain. However, it appears to be more pronounced with some forms of agitation than with others. In the current studies, we have attempted to clarify further the nature of these two storage lesions. We now show that during fall in pH of PC to levels below 6.8, the increase in apparent platelet volume corresponds well with a decrease in platelet density and refractive index, both consistent with cellular swelling due to influx of extracellular fluid. We were previously unable to determine whether this lesion was due to pH fall per se or to the associated hypoxic conditions. In the current study, platelets stored under a nitrogen atmosphere for as long as 7 days show no swelling or shape change when the pH is kept above 6.8, suggesting that the damage is due to pH fall per se. In previous studies of the second storage lesion, which occurs without pH fall, we have found that the increase of the dispersion of the Coulter volume distribution correlates with loss of in vivo viability measured as percent recovery after labelling with chromium-51. In this study, we show that the increase in dispersion results from the presence of a population of abnormally light platelets. Since this light population labels normally with chromium-51, we hypothesize that it contains the cells which fail to circulate after in vivo infusion. In three infusions of thrombocytopenic patients, we support this hypothesis by showing that the abnormal light population does not circulate in vivo and that the dispersion of the apparent volume distribution of the circulating cells is normal.