In order to develop a long-term anesthesia for flighty antelope species in field situations, two different protocols for induction and maintenance with an intravenous infusion were evaluated in wild-caught impala (Aepyceros melampus). Ten adult female impala were induced with two induction protocols: one consisted of 0.2 mg/kg medetomidine, 4 mg/kg ketamine, and 0.15 mg/kg butorphanol (MKB) and one consisted of 0.375 mg/kg etorphine, 0.2 mg/kg medetomidine, and 0.2 mg/kg midazolam (EMM). In both treatments, anesthesia was maintained with a continuous intravenous infusion (CII) at an initial dose rate of 1.2 μg/kg per hr medetomidine, 2.4 mg/kg per hr ketaminen and 36 μg/kg per hr midazolam. Partial reversal was achieved with naltrexone (2 : 1 mg butorphanol; 20 : 1 mg etorphine) and atipamezole (5 : 1 mg medetomidine). Evaluation of anesthesia included respiratory rate, heart rate, rectal temperature, arterial blood pressure, oxygen saturation, end tidal carbon dioxide tension, and tidal volume at 5-min intervals, palpebral reflex and response to painful stimuli at 15-min intervals, and arterial blood gases at 30-min intervals. Plasma cortisol concentration was determined after induction and before reversal. Duration and quality of induction and recovery were evaluated. EMM caused a faster induction of 9.5 ± 2.9 min compared to 11.0 ± 6.4 min in MKB. Recovery was also quicker in EMM (EMM: 6.3 ± 5.4 min; MKB: 9.8 ± 6.0 min). However, EMM also produced more cardiopulmonary side effects, including hypoxemia and hypercapnia, and calculated oxygenation indices (PaCO₂-PETCO₂) were worse than in MKB. One animal died after induction with EMM. The CII provided surgical anesthesia in 7 of 10 animals in MKB and in 9 of 9 animals in EMM for 120 min. In conclusion, the MKB induction protocol had advantages for prolonged anesthesia in impala with significantly less cardiopulmonary depression compared to EMM. The comparably decreased anesthetic depth could easily be adjusted by an increase of the CII.