Summary:The efficiency of five different cryopreservation protocols (our original controlled-rate and noncontrolled-rate protocols) was evaluated on the basis of the recovery after thawing of very primitive pluripotent hemopoietic stem cells (MRA CFU-GM ), pluripotent progenitors (CFUSd12) and committed granulocyte-monocyte progenitors (CFU-GM) in mouse bone marrow. Although the nucleated cell recovery and viability determined immediately after the thawing and washing of the cells were found to be similar, whether controlled-rate or noncontrolled-rate cryopreservation protocols were used, the recovery of MRA CFU-GM , CFU-Sd12 and CFU-GM varied depending on the type of protocol and the cryoprotector (DMSO) concentrations used. It was shown that the controlled-rate protocol was more efficient, enabling better MRA CFU-GM , CFU-Sd12 and CFU-GM recovery from frozen samples. The most efficient was the controlled-rate protocol of cryopreservation designed to compensate for the release of fusion heat, which enabled a better survival of CFU-Sd12 and CFU-GM when combined with a lower (5%) DMSO concentration. On the contrary, a satisfactory survival rate of very primitive stem cells (MRA CFU-GM ) was achieved only when 10% DMSO was included with a fivestep protocol of cryopreservation. These results point to adequately used controlled-rate freezing as essential for a highly efficient cryopreservation of some of the categories of hematopoietic stem and progenitor cells. At the same time, it was obvious that a higher DMSO concentration was necessary for the cryopreservation of very primitive stem cells, but not, however, for more mature progenitor cells (CFU-S, CFU-GM). These results imply the existence of a mechanism that decreases the intracellular concentration of DMSO in primitive MRA cells, which is not the case for less primitive progenitors.