To improve the cryopreservation protocol for mouse sperm, we attempted to estimate the type and extent of cryoinjury at various steps of the process. First, we demonstrated that mouse sperm are sensitive to chilling at -15 C and that the sensitivity is dependent on the length of exposure. To estimate cryoinjuries, sperm suspensions were ice-seeded at -5 or -15 C, frozen with liquid nitrogen (LN(2)) gas and then frozen in LN(2). In one experiment, sperm seeded at -5 C were cooled slowly to -15 C before deep freezing. At various steps of the cryopreservation process, the sperm were warmed and their viability was assessed based on motility and the integrities of the plasma membrane and acrosome. The motility of frozen-thawed sperm was higher on seeding at -5 C (28%) than at -15 C (9%). The motility did not decrease when the sample was transferred from LN(2) gas to LN(2). To estimate cryoinjury of sperm, we presumed the viability of frozen sperm to be decreased by chilling, hypertonic stress and formation of intracellular ice. When the sperm suspension was cooled and seeded at -5 C, the motility decreased by 25% due to hypertonic stress. When the sperm were cooled in LN(2) gas, the motility decreased by 17% with the formation of intracellular ice. When the sperm were cooled to -15 C, the motility decreased by 51% from chilling. After seeding, the motility decreased by 18% due to formation of intracellular ice and by 7% due to hypertonic stress. Considering the results, it would be preferable to seed samples at a higher temperature to prevent intracellular ice from forming and to cool seeded samples rapidly enough to minimize chilling injury and hypertonic stress, but not too rapidly to allow intracellular ice to form.