The aim of this work was to study the effects of dilution and centrifugation (i.e., two methods of reducing the influence of the seminal plasma) on the survival of spermatozoa and the structure of motile sperm cell subpopulations in refrigerated Catalonian donkey (Equus asinus) semen. Fifty ejaculates from nine Catalonian jackasses were collected. Gel-free semen was diluted 1:1, 1:5 or 1:10 with Kenney extender. Another sample of semen was diluted 1:5, centrifuged, and then resuspended with Kenney extender until a final dilution of 25x10(6) sperm/ml was achieved (C). After 24 h, 48 h or 72 h of refrigerated storage at 5 degrees C, aliquots of these semen samples were incubated at 37 degrees C for 5 min. The percentage of viable sperm was determined by staining with eosin-nigrosin. The motility characteristics of the spermatozoa were examined using the CASA system (Microptic, Barcelona, Spain). At 24h, more surviving spermatozoa were seen in the more diluted and in the centrifuged semen samples (1:1 48.71%; 1:5 56.58%, 1:10 62.65%; C 72.40%). These differences were maintained at 48 h (1:1 34.31%, 1:5 40.56%, 1:10 48.52%, C 66.30%). After 72 h, only the C samples showed a survival rate of above 25%. The four known donkey motile sperm subpopulations were maintained by refrigeration. However, the percentage of motile sperms in each subpopulation changed with dilution. Only the centrifuged samples, and only at 24h, showed exactly the same motile sperm subpopulation proportions as recorded for fresh sperm. However, the 1:10 dilutions at 24 and 48 h, and the centrifuged semen at 48 h, showed few variations compared to fresh sperm. These results show that the elimination of seminal plasma increases the survival of spermatozoa and the maintenance of motility patterns. The initial sperm concentration had a significant (P<0.05) influence on centrifugation efficacy, but did not influence the number of spermatozoa damaged by centrifugation. In contrast, the percentage of live spermatozoa in the fresh semen significantly influenced the number of spermatozoa damaged by centrifugation, but not centrifugation efficacy.