A new inhalant anaesthetic, sevoflurane, was used to maintain anaesthesia in 40 animals (2 mules and 38 horses of 9 breeds) presented for various surgical procedures. Eighteen mares, 11 stallions and 11 geldings underwent 6 orthopaedic and 34 soft tissue operations. Induction of anaesthesia was achieved with combinations of xylazine (0.5–1.1 mg/kg), diazepam (0.03–0.1 mg/kg), butorphanol (0.02 mg/kg), guaifenesin (50–84 mg/kg) and ketamine (1.1 mg/kg). Following tracheal intubation, a surgical plane of anaesthesia was maintained with sevoflurane in oxygen delivered from a precision vaporiser. Temperature, ECG, arterial blood pressure and expired gas composition were monitored. Mechanical ventilation was used in most animals (n=37) because of hypoventilation (P a CO 2 > 7.31 kPa [55 mmHg]). Following surgery, horses were moved to a recovery room and allowed to recover alone (n=36) or with assistance (n=4). Time to sternal recumbency, standing, the time when satisfactory coordination was present (after standing) and the number of attempts to stand were recorded. The quality of recovery was scored on a 1 (best) to 6 (worst) scale. Mean blood pressures at 30, 60, 90, 120 and 150 min of anaesthesia were 72, 73, 74, 75 and 72 mmHg, respectively. Systolic and diastolic pressures at 30, 60, 90, 120 and 150 min of anaesthesia were 97, 97, 94, 96, 93 and 59, 63, 64, 68, 67 mmHg, respectively. Dobutamine was used in 23 horses to maintain mean arterial blood pressure > 60 mmHg. Mean heart and respiratory rates at 30, 60, 90, 120 and 150 min of anaesthesia were 36, 38, 39, 38 and 38 beats/min, and 9, 8, 8, 8 and 8 breaths/min. Mean duration of anaesthesia was 121 rnin (sd: 56 min), mean time to sternal recumbency was 27 min (sd: 13 min), average time to standing (all horses) was 33 min (sd: 12 min) and time to satisfactory coordination was 44 min (sd: 13 min). Most horses (n=37) received xylazine during recovery (mean dose 0.18 mg/kg iv). The median number of attempts to sternal recumbency and standing were 1.0 (range; 1–7) and 2.0 (range; 1–20), respectively, while the median recovery score was 1.5 (range; 14). The ‘depth’ of anaesthesia was easy to control and recoveries were generally very satisfactory.