We performed two sets of experiments in which observers were instructed to make saccades to an odd colored target embedded in an array of distractors. First, we found that when the colors of the target and distractors switched unpredictably from trial to trial (the mixed condition), saccadic latencies decreased with increasing numbers of distractors. In contrast, saccadic latencies were independent of the number of distractors when the color of the target and distractors remained the same on each trial (the blocked condition). This pattern of results mirrors visual search tasks in which focal rather than distributed attention is required (Bravo, M.J., Nakayama, K. (1992). The role of attention in different visual search tasks. Perception and Psychophysics, 51, 465-472.). Second, we found that saccades to an odd target were made more quickly and accurately when the target was the same color as on previous trials than when it changed color. This priming of the target color accumulates across five to seven trials over a period of approximately 30 s. A similar priming effect has been previously shown for the deployment of focal attention (Maljkovic, V., Nakayama, K. (1994). Priming of popout: III. Role of features. Memory and Cognition, 22(6), 657-672.). Thus, we show a close congruence between the pattern of saccadic eye movement latencies and the deployment of focal attention. This supports the view that (1) the execution of saccades requires focal as opposed to distributed attention and that (2) this focal attention is guided by a short term memory system which facilitates the rapid refixation of gaze to recently foveated targets.