A review of the relevant literature reveals that reading reversals, whether in sequence or orientation, comprise a very small proportion of the total errors made by even poor readers. Young children tend to make more reversals, yet this tendency generally disappears with age. Top-down theorists feel that the reversal tendency of young children is due to unfamiliarity with the stimuli, and that maturation will alleviate the problem. While bottom-up theorists feel that perceptual training will correct the tendency to reverse, it has been demonstrated that this type of training is generally unsuccessful, particularly with children beyond age seven. Interactive theorists state that if reversals were a perceptual problem, persistent reversers would exhibit the same problem in other activities, a phenomenon not yet documented. The tendency to reverse has been considered an indicator of dyslexia, but studies with dyslexics and normal disabled readers have demonstrated that the two groups do not differ significantly in the number and type of reversal errors made. Assessing reversals has been fraught with methodological flaws, concatenating perceptual ability with verbal ability, motor ability, and memorial processes. While only a few studies have assessed reversals in connected text, such studies have demonstrated that the proportion of reversals is small compared to the other types of errors made, and that connected text can in fact help a child in constraining the type and number of reversals made. (FL)