A study was conducted in Argentina, to investigate the period of protection of a single injection of doramectin administered subcutaneously (s.c.) at 200 micrograms kg-1 (1 ml/50 kg) compared with single treatments of ivermectin (200 micrograms kg-1 s.c.) and fenbendazole (5 mg kg-1 p.o.), against field infections of gastrointestinal parasites of cattle. Eighty-three animals were selected and ranked on the basis of serial fecal egg counts (e.p.g.'s). From this group, three animals were slaughtered before treatment and their lungs, abomasum, small and large intestines, were processed for parasite counts and identification. The remaining 80 animals were allocated in ranked groups of four to a control or one of three treated groups. Animals of the four groups were grazed together in the same pasture for the duration of the study. Treatments were administered on Day 0. Individual fecal samples were collected at weekly intervals for the first 49 days post-treatment and twice a week from Day 52 to Day 84 (end of study). At each collection day fecal samples were pooled for coprocultures. On Day 28 and 56, two animals from each group, previously identified on Day 0, were killed and their parasite burdens determined. The duration of protection of a single injection of doramectin was longer than ivermectin or fenbendazole treatment. On Day 56, the total number of parasites found in doramectin-treated animals was significantly (P < 0.05) lower than parasite burdens found in either ivermectin- or fenbendazole-treated animals. The longer persistent activity of doramectin was expressed by the lower number of adults and L4 stages of Ostertagia ostertagi. Data from this experiment demonstrated the limitations of using fecal egg counts to evaluate the persistent efficacy of anthelmintics. The duration of activity of doramectin was demonstrated more accurately by parasite counts in cattle from each group since decreasing e.p.g.'s were seen in non-medicated animals without changes in total parasite burdens.