Coelomocytes were extruded from three earthworm species: Lumbricus terrestris, Eisenia fetida and Octolasion tyrtaeum. Featuring a simple low-vacuum holding device, the proposed methodology allows the recovery of cells with;minimum risk of contamination by faecal material. The viability of O. tyrtaeum coelomocytes was highly reproducible (average 93%), with an average yield of 0.92 x 10(6) viable cells per earthworm. Cell. viability for L. terrestris and E. fetida averaged similar to 68% but the cell yields were higher (respectively 1.67 x 10(6) and 1.28 x 10(6)). Large inter-individual differences in cell yields were observed with L, terrestris. Flow cytometric analyses indicated species to species differences in cell populations. Coelomocytes from E, fetida were the smallest with similar to 57% of the total, viable cells recovered being monitored between 2 and 10 mu m. Large granulated cells (greater than or equal to 20 mu m) were detected in fairly large proportions in L, terrestris and O, tyrtaeum (similar to 52 and similar to 96%, respectively) while they were less abundant in E, fetida (similar to 9%). Using the vital dye neutral red to assess functional integrity, average cellular uptakes were significantly higher for L, terrestris and O, tyrtaeum than for E, fetida (2.94, 2.66 and 0.64 mu g/2 x 10(5) cells, respectively). In summary, the extrusion methodology herein described is applicable for the recovery of coelomocytes from a wide range of earthworm sizes and species. Moreover, this study strengthens the fact that extruded coelomocytes could be used for the evaluation of cell dysfunction and/or cell death following an in vitro and/or in vivo treatment.