A masked, randomised, controlled clinical trial for the treatment of canine superficial pyoderma was undertaken. Dogs with a clinical diagnosis of superficial pyoderma, supported by bacterial culture were admitted to the trial and randomly assigned to treatment with either clindamycin hydrochloride at 5.5 mg/kg twice daily or clavulanate-amoxycillin at 12.5 mg/kg twice daily. After 21 days the animals were re-assessed, and therapy was continued for a further 21 days in the dogs with persistent lesions if bacterial culture demonstrated continued sensitivity. Twenty-nine dogs were treated with clindamycin hydrochloride and 27 with clavulanate-amoxycillin. Complete cure was obtained after three weeks in 17 (59 per cent) of the clindamycin-treated cases, but in only eight (30 per cent) of the clavulanate-amoxycillin treated group. Clindamycin was significantly more effective than clavulanate-amoxycillin for the treatment of superficial pyoderma in dogs.