Individual differences in reading comprehension may reflect differences in working memory capacity, specifically in the trade-off between its processing and storage functions. A poor reader's processes may be inefficient, so that they lessen the amount of additional information that can be maintained in working memory. A test with heavy processing and storage demands was devised to measure this trade-off in 2 experiments with 41 undergraduates. Ss read aloud a series of sentences and then recalled the final word of each sentence. The reading span, the number of final words recalled, varied from 2 to 5. This span correlated with 3 reading comprehension measures, including verbal SAT and tests involving fact retrieval and pronominal reference. Similar correlations were obtained with a listening span task, showing that the correlation is not specific to reading. Results are contrasted with traditional digit span and word span measures that do not correlate with comprehension.