Three studies were conducted in an attempt to replicate previous research concluding that semantic meaning is accessed in the absence of conscious awareness. A pattern mask was used to interrupt the processing of stimulus words after 30 milliseconds; at this duration subjects were not able to identify the stimulus words or even to determine whether a word was present. Experiments 1 and 2 used a forced-choice guessing procedure in which subjects were required to guess which member of a recognition pair was most similar in either visual or semantic characteristics to a masked stimulus word. Subjects' performance was somewhat higher than chance in Experiment 1, but Experiment 2, in which half of the trials for a given subject were preceded by a row of Xs instead of a word, indicated that this result could be attributed to guessing strategies. Experiment 3 used a semantic priming procedure that did not require a conscious response to the initial prime word. Results showed the usual semantic priming effect for unmasked primes; however, no priming occurred for masked primes. Thus, these studies did not support the claim that meaning is accessed prior to conscious awareness. (Author/HTH)