Automaticity training, aimed at decreasing latencies to read a specific set of words, was given to an experimental group of third-graders. A control group of third-graders was not given this training. Following training, the cloze method was used to test the reading comprehension of texts composed from the training words. No evidence was found to support the hypothesized positive transfer of automaticity training to reading comprehension. Possible interpretations of this result are discussed.