Summary. Ejaculates (164) were obtained from 17 donors serving on an artificial insemination by donor panel. Semen analysis was performed before and after freezing by an integrated microcomputerized system employing the multiple-exposure photography (MEP) method. Sperm count, motility, velocity, motility index (MI; product of the sperm velocity and percentage of motile spermatozoa) and motile density (MD) were determined for each ejaculate. After the initial evaluation the ejaculates were frozen in liquid nitrogen, thawed 24 h later, and assessed for post-thaw motility, velocity, MI and MD. The mean ± s.e. sperm count and volume for this group of donors was 148 ± 4 × 106/ml and 3·1 ± 0·1 ml, respectively. Mean ± s.e. values obtained from the prefreeze analysis were: motility = 64 ± 1%, velocity = 30 ± 0·4μm/sec, MI = 19 ± 0·5 μm/sec and MD = 94 ± 3 × 106/ml. Post-thaw analysis revealed a significant reduction (P < 0·01 in all values measured. Motility was reduced to 27 ± 1%, MI was reduced to 5 ± 0·3 μm/sec, and MD was reduced to 33 ± 1 × 106/ml Velocity was the least affected by cryopreservation, being reduced to 21 ± 0·5 μm/sec (P < 0·01). Cryopreservation resulted in a marked shift in the frequency distribution of sperm motility and motility index towards subnormal values while in the majority of ejaculates velocity and motile density were maintained in the normal range. Significant differences were noted amongst donors in the percentage change of the various semen measures as a result of cryopreservation. When within-subject coefficients of variation were calculated, velocity was the least variable parameter. These results indicate that, while cryopreservation results in significant reductions in the number of motile spermatozoa in the ejaculate, the velocity is only marginally reduced.