An experiment was carried out in the pigeon to assess the effects of sodium amylobarbitone (SAB) on memory, using a delayed pair comparison task. Performance was disrupted at all delays by doses of 10 and 20 mg/kg SAB and it was concluded that this effect was not due to perceptual impairment or stereotyped responding, which the drug might have caused, since performance on a visual discrimination task was unaffected by the same doses of SAB. Two further experiments assessed proactive and retroactive effects by varying the inter-trial interval and by introducing an irrelevant orange stimulus at various stages of the delay interval. Although these manipulations disrupted performance, there was no evidence to suggest that the drug potentiated their effects in a specific manner. The results are discussed in terms of an encoding hypothesis, and some reference is made to the possible neuropharmacological basis of action of this drug.