A retention deficit attributable to prior learning was obtained in pigeon short-term memory by preceding a delayed matching-to-sample (DMTS) trial (Trial 2) by a prior, conflicting DMTS trial (Trial 1). The retention deficit was obtained when the matching and nonmatching colors on Trial 1 appeared in reversed roles on Trial 2, but was not obtained when the Trial 1 colors did not appear on Trial 2. A prior complete DMTS trial was found to provide a more potent source of interference than various components of a DMTS trial. The magnitude of the retention deficit attributable to prior learning in pigeon short-term memory was also found to be sensitive to the length of time separating the interfering and test trials. These findings were discussed in terms of the Grant and Roberts independence and competition model of trace interaction.