Differences were seen in the ability of 2 strains of C. parvum to augment cytotoxicity attributable to NK cells towards a rat lymphoma. Furthermore, 2 batches of the same strain of C. parvum prepared by different methods also differed in their ability to augment cytotoxicity. Other factors influencing cytotoxicity were dose, route of inoculation and time after injection at which the assay was performed. Although all preparations of C. parvum augmented the cytotoxicity of peritoneal-exudate cells when injected i.p., only the most stimulatory preparation consistently augmented splenic cytotoxicity when given by this route. I.v. administration of 1 mg of C. parvum produced peak levels of splenic cytotoxicity 2-3 days later, but this response was strictly dose-dependent, since 1 microgram depressed splenic cytotoxicity. This dose-dependent effect also extended to ADCC, since 1 mg stimulated cytotoxicity towards antibody-coated P815 cells, whilst 1 microgram depressed it in a manner similar to its effect on natural cytotoxicity. Whilst the cytotoxic cells of stimulated rats closely resembled the NK cells of normal rats, BN rats responded differently to C. parvum from W/Fu or WAG rats, in that marked lysis of P815 or RBL-5 cells was observed, though these targets are usually resistant to lysis by rat NK cells in short-term assays.