These experiments were conducted to investigate the relationship between reading comprehension level, digit span, and short-term memory for Morse code-like temporal patterns. Consistent with previous research on children, Experiment 1 demonstrated that college students performed better when the first pattern was auditory than when it was visual or tactual. Experiments 2 and 3 found no relationship for either college or fifth-grade students between digit span and accuracy in comparing patterns of tones presented a few seconds apart. However, both tasks discriminated between children with normal and poor reading comprehension scores on a standardized test. It appears that these two tasks index fundamental processes that underlie reading comprehension. Digit span seems to assess an individual's ability to rapidly develop meaningful codes in memory for incoming verbal stimuli. The auditory pattern comparison procedure appears to measure ability to maintain information in short-term memory.Differences in short-term memory capabilities have been shown to distinguish reading-disabled children from normal readers of the same age (e.g., Elkins & Suitman, 1981). However, Torgesen and Houck (1980) have pointed out that the specific deficiencies in learning-disabled children are not well understood. Birch and Belmont (1964) asserted that the difficulties of retarded readers derive from their inability to form connections between visually and auditorially coded information in short-term memory. Clearly, reading instruction concentrates on the translation of visually presented materials into sounds and vice versa. However, Rubinstein and Gruenberg (1971) pointed out that the experimental task employed by Belmont (1964, 1965