OBJECTIVE: To determine the time course and certain cardiopulmonary effects of trunk-breathing elephants immobilized with thiafentanil-azaperone. STUDY DESIGN: Prospective descriptive study. ANIMALS: A convenience sample of 10 free-ranging African elephant bulls (estimated weight range: 3000-6000 kg). METHODS: Elephants were immobilized using thiafentanil (15-18 mg) and azaperone (75-90 mg) administered by dart. Once recumbent, the respiratory rate, minute ventilation (V˙e), end-tidal carbon dioxide (Pe'CO2), arterial blood pressure and heart rate were recorded immediately after instrumentation and at 5 minute intervals until 20 minutes. Arterial blood gases were analysed at the time of initial instrumentation and at 20 minutes. On completion of data collection, thiafentanil was antagonized using naltrexone (10 mg mg-1 thiafentanil; administered intravenously). A stopwatch was used to record time to recumbency (dart placement to recumbency) and time to recovery (administration of antagonist to standing). Data were compared using a one-way anova. Data are presented as mean ± standard deviation. RESULTS: All elephants were successfully immobilized, and there were no significant changes in cardiopulmonary variables over the monitoring period. Average time to recumbency was 12.5 (± 3.9) minutes. The measured V˙e was 103 (± 30) L minute-1. The average heart and respiratory rates over the 20 minute immobilization were steady at 49 (± 6) beats minute-1 and 5 (± 1) breaths minute-1, respectively. The mean arterial blood pressure was 153 (± 31) mmHg. The elephants were acidaemic (pH: 7.18 ± 0.06), mildly hypoxaemic (PaO2: 68 ± 15 mmHg; 9.1 ± 2.0 kPa) and hypercapnic (PaCO2: 52 ± 7 mmHg; 6.9 ± 0.9 kPa). Average time to recovery was 2.2 ± 0.5 minutes. CONCLUSION AND CLINICAL RELEVANCE: African elephant bulls can be successfully immobilized using thiafentanil-azaperone. Recumbency was rapid, the cardiopulmonary variables were stable over time, and recovery was rapid and complete. Mild hypoxaemia and hypercapnia were evident.