Initial methods for successful cryopreservation of spermatozoa (Spz) from cultured teleosts have been reported for a large number of species. Nevertheless, the main obstacle to the formation of large-scale 'sperm banks' lies in the efficient upscaling of these techniques. In the present paper, attempts to develop reliable, simple and efficient methods for operating a large-scale cryobank for the Japanese ornamental carp (Cyprinus carpio L.) cultured in Israel, are reported, First investigated was the individual and season variability. Significant differences were found in the number of fertilized eggs obtained after crossing Spz and eggs from four males and four females. The use of pooled samples of the Spz and eggs from these individuals reduced this variability. The overall average number of sperm cells found was similar to 15 x 10(9) ml(-1), with no significant differences during the culture period, lasting from January-August in 1993 and 1994. Similarly, no significant differences were found during these periods in the percentage of fertilized eggs obtained from the test system, Significant damage was incurred by the type of extender solution and the cryoprotective agents used for suspending sperm cells, While similar to 5000 sperm cells were required to fertilize one egg when suspended in Extender 2 solution, similar to 300,000 sperm cells were needed if they were suspended in a solution containing 15% dimethylsulfoxide (DMSO). Most of the damage occurred within 10 min of adding the DMSO to the sperm suspension. Post-thaw viability of frozen Spz was tested using trypan blue, to reveal the proportion of damaged cells, No significant correlation was found between the percentage of dyed damaged sperm cells and the percentage of fertilized eggs that were obtained in parallel experiments from replicate unstained samples. The percentage of fertilized eggs obtained with thawed cryopreserved Spz samples was tested in three systems: (a) petri dishes holding similar to 200 eggs, (b) petri dishesholding 1000 eggs, and (c) 100 l funnels holding similar to 100,000 eggs. In several trials, 50% (similar to 75% of control) or more of the eggs were fertilized with 0.2 ml of thawed Spz samples which were tested in systems (a) and (b). On several occasions, large numbers (10,000-60,000) of fertilized eggs were produced in system (c), but no clear correlation was found between results obtained in petri dish trials and those found in 100 l containers. It was impossible to increase the number of fertilized eggs by using larger amounts of thawed Spz, indicating an interference of the dead cells during fertilization. Future work will have to concentrate on developing a small-scale assay system that would adequately reflect large-scale trials. The methods that were developed for the cryopreservation of Spz samples pooled from several fish were also found to be suitable for preserving samples of selected individual fish. The results of this work led to the formation of an operating 'sperm bank' for C. carpio L. in Israel. (C) 1997 Elsevier Science B.V.