BACKGROUND: A small, but statistically significant, improvement has previously been demonstrated when retesting visual acuity within 24 hours. Because letter chart testing procedures are familiar to most adult patients, this improvement may be due to memory of the chart letter subset, rather than increased familiarity with the test procedure. Chart letter subset memory does not appear to have previously been measured directly. METHODS: A letter recognition task was used to measure chart letter subset memory for groups of routine clinical patients. Comparisons were made between assessments after a single acuity measurement and assessments at increasing intervals after a routine examination involving several references to the same letter chart for both eyes. RESULTS: The recognition score for the 'one-minute-after a single acuity measurement group' was an average of 2.5 letters (range 0.7 to 4.7). Low levels of chart letter subset memory were found to persist for 10 days. CONCLUSION: The chart letter subset memory found is consistent with enhanced acuity scores that occur on retesting. Scores appear likely to have a greater chance of enhancement for a second eye measurement when it follows immediately after the first. Similarly, scores at the end of a full examination may be enhanced by memory that is accumulated during multiple measurements for both eyes using the same chart. Chart memory may be acquired when acuity is measured frequently in monitoring the progress of disease. Chart memory, or its loss over time, may contribute to measurement imprecision and associated expansion of the confidence limits for significant change in acuity.