The delayed‐matching‐to‐sample procedure was modified to study pigeons' spatial memory. Nine pecking keys, arranged as a three‐by‐three matrix, served as the spatial cues. Trials began with a brief “ready” stimulus (dimming of the houselight). Then a randomly chosen key was lit briefly as a sample. After a short delay the sample key was lit again along with one of the other eight keys. A peck at the key that had served as the sample produced grain reinforcement, whereas a peck to the other key produced only the intertrial interval. After delayed matching of key location was learned, the effects of sample and delay duration, number of keys illuminated as sample and comparisons, and organization of three‐key samples were studied. Matching accuracy decreased as sample duration decreased, delay increased, the number of locations serving as samples increased, the number and proximity of comparisons increased, and when the three‐key samples were “discontinuous