Summaryo1.Fifty cubic millimeters of citrated human blood containing various concentrations of glyeerol, DMSO, dextrose, sucrose, or PVP were frozen in capillary tubes which had an outside diameter of 1.5 mm and a wall thickness of 0.3 mm, by abrupt immersion into isopentane baths at temperatures ranging from −5 to −130°C.2.The frozen specimens were thawed rapidly by abrupt immersion of the capillary into a well stirred water bath at 35°C, and the extent of injury to the erythrocytes was measured by a colorimetric determination of the amount of hemoglobin released from the damaged cells.3.The results obtained indicate that the 5 protective agents investigated may be classified into three types: the type glycerol, the type dextrose and the type PVP.4.Protective agents of the type glycerol, which includes DMSO, become increasingly efficient in preventing lysis with increased concentration (up to a limit) at low cooling velocities and increasingly injurious at high cooling velocities.5.At low cooling velocities, at which hemolysis is minimum with protective agents of the type glycerol, the agents of the type dextrose, including sucrose, exert only a slight protection. At high cooling velocities, they become increasingly effective in reducing hemolysis.6.Protective agents of the type PVP give results intermediate between those of the type glycerol and those of the type dextrose. At low cooling velocities they give less protection than the additives of the type glycerol, but more than those of the type dextrose. Conversely, at high cooling velocities, they do not provide the protection afforded by sugars, nor do they exert the marked injurious effects of glycerol.7.The general shape of the hemolysis curves obtained with human blood frozen at various rates with different protective agents is very similar to those observed with bovine blood frozen under the same conditions. Fifty cubic millimeters of citrated human blood containing various concentrations of glyeerol, DMSO, dextrose, sucrose, or PVP were frozen in capillary tubes which had an outside diameter of 1.5 mm and a wall thickness of 0.3 mm, by abrupt immersion into isopentane baths at temperatures ranging from −5 to −130°C. The frozen specimens were thawed rapidly by abrupt immersion of the capillary into a well stirred water bath at 35°C, and the extent of injury to the erythrocytes was measured by a colorimetric determination of the amount of hemoglobin released from the damaged cells. The results obtained indicate that the 5 protective agents investigated may be classified into three types: the type glycerol, the type dextrose and the type PVP. Protective agents of the type glycerol, which includes DMSO, become increasingly efficient in preventing lysis with increased concentration (up to a limit) at low cooling velocities and increasingly injurious at high cooling velocities. At low cooling velocities, at which hemolysis is minimum with protective agents of the type glycerol, the agents of the type dextrose, including sucrose, exert only a slight protection. At high cooling velocities, they become increasingly effective in reducing hemolysis. Protective agents of the type PVP give results intermediate between those of the type glycerol and those of the type dextrose. At low cooling velocities they give less protection than the additives of the type glycerol, but more than those of the type dextrose. Conversely, at high cooling velocities, they do not provide the protection afforded by sugars, nor do they exert the marked injurious effects of glycerol. The general shape of the hemolysis curves obtained with human blood frozen at various rates with different protective agents is very similar to those observed with bovine blood frozen under the same conditions.