Teleost fish demonstrate the neurophysiologic capacity to experience pain and analgesia. A common model for assessing analgesic effect is the reduction of minimum anesthetic concentration (MAC). The present study adapted the model of MAC depression to evaluate the analgesic effects of morphine, butorphanol, medetomidine, and ketoprofen in goldfish (Carassius auratus). MAC was determined by an up–down method of sequential population sampling, anesthetizing fish with tricaine methanesulfonate (MS-222) in concentration increments of 10 parts per million (ppm), and using intramuscular needle insertion as a supramaximal noxious stimulus. Baseline MAC was determined in triplicate at the beginning (MAC_i) and conclusion (MAC_f) of the experiment (approximately 60 days). For drug trials, MAC was redetermined 1 hr after administration of morphine (10, 20, 40 mg/kg i.m.), butorphanol (0.1, 0.2, 0.4 mg/kg i.m.), medetomidine (0.01, 0.015, 0.025 mg/kg i.m.), ketoprofen (0.5, 1.0, 2.0 mg/kg i.m.), or saline control. Each drug/dose was tested in random order with a >6-day washout period. MAC_i and MAC_f were 163 and 182 ppm, respectively, and were significantly different from each other (P = 0.02). All doses of morphine and ketoprofen decreased MAC below MAC_i. The highest dose of medetomidine decreased MAC below MAC_i. The lowest dose of butorphanol decreased MAC below MAC_i, but higher doses increased MAC above MAC_f. The authors conclude that MAC determination in fish using MS-222 was feasible and reproducible in the short term. The fact that MAC increased over time and/or exposure may limit the usefulness of MS-222 in MAC depression studies. Morphine and ketoprofen decrease anesthetic needs in goldfish and may provide analgesia.