As discussed above, the process by which normal senescent red cells are selected for removal from the circulation is the subject of much ongoing research and is not yet well understood. This in turn creates a problem for studies on the enhanced removal that occurs in xenobiotic-induced hemolytic states; specifically, whether the enhanced removal should be considered as an increase in rate of the normal sequestration mechanism or as an unrelated process, in part or in whole. This difficulty bears directly on the interpretation of much of the mechanistic hemolytic literature. Because of its dual in vivo and in vitro hemolytic capability, and because of its capacity to induce frank lysis in the incubation mixture, phenylhydrazine has been used extensively as a model compound for mechanistic studies. These data have contributed heavily to our current concepts of how chemicals induce damage in the red cell. The comparison studies presented above cast doubt on the relevance of many of these phenylhydrazine studies for the in vivo hemolytic response. Phenylhydrazine, like divicine and DDS-NOH, shows an overwhelming predominance of uptake into the spleen, as distinct from removal by the RES system in general, as evidenced by relatively low liver uptake. This suggests strongly that damaged cells are removed intact by the spleen and do not lyse or fragment in the general circulation, at least to any significant extent. The studies with DDS-NOH indicate that neither Heinz body formation nor lipid peroxidation per se are essential steps in the process by which damaged red cells are removed from the circulation in the rat. It is not yet clear whether this lack of obligatory involvement of Heinz bodies and lipid peroxidation is peculiar to the arylhydroxylamine-induced hemolytic state or whether it will prove to be of general applicability. On the other hand, cysteamine failed to reverse the hemolytic damage caused by phenylhydrazine. Since cysteamine "rescued" DDS-NOH treated cells under the same experimental conditions, this observation raises the possibility that protein-thiol oxidation per se is also not an obligatory step in the sequence of events leading to premature sequestration. Clearly, the ratio of lipid to protein oxidation is markedly different in these three examples of hemotoxic compounds. DDS-NOH showed high protein oxidation with no discernible lipid oxidation, divicine showed both high protein and high lipid oxidation, and phenylhydrazine showed high lipid and low protein oxidation. While the significance of these markedly different patterns of injury is far from clear, it seems reasonable to conclude that there is more than one way by which chemicals damage the red cell. It is intriguing that these apparently different chemical insults within the red cell result in a common "message" on the outside of the cell, such that the cell appears as "prematurely" aged. Although the pattern of injury inside the cell may be significantly different, the process by which the three hemotoxic compounds enhance uptake by splenic macrophages may remain the same. That is, there may be a variety of insults sustained within the red cell that lead by different pathways to similar "recognition-specific" changes on the external surface of the red cell. Clearly, comparison of the effects of the three hemotoxic compounds will shed light on both the hemolytic process and on normal red cell sequestration mechanisms.