Pigeons given bilateral hippocampal lesions, together with sham-operated and normal birds, were trained on a Konorski pair comparison (recognition memory) task. In this paradigm, the subject must indicate whether or not two successively presented stimuli were identical, or dissimilar; a delay may be interposed between the presentations. All subjects learned the basic (simultaneous, or no delay pair comparison) task to similar levels of performance, indicating that neither perceptual nor general cognitive abilities were selectively impaired. However, the hippocampal group was impaired when delays were interposed, i.e. when (labile) memory was required. These subjects tolerated only short delays, and lost information('forgot') at faster rates. A second experiment explored (a) the importance of acquisitional variables by limiting exposure to the to-be-remembered stimulus (S1), and (b) the role of proactive interference, by varying the inter-trial-interval (ITI). Neither manipulation selectively disrupted the hippocampal group, although reductions in S1 and ITI duration progressively impaired performance in all groups. It is concluded that the avian and mammalian hippocampii are functionally similar, at least with respect to recognition memory processing. The results are discussed in terms of the hypothesis proposed by Olton et al., that hippocampal lesions impair working, but not reference, memory probably by disrupting a process analogous to storage.