In order to determine the effect of grapheme-phoneme correspondence on beginning reading, word acquisition as related to the degree of regularity of the word was used as a measurement. Twenty children enrolled in a preschool project for culturally disadvantaged children at Peabody College were randomly assigned to two groups. One group used the three preprimers of the Scott Foresman basal reading series printed in traditional orthography, and the experimental edition of the same series printed in the Initial Teaching Alphabet was used for the other group. In all other ways the books and procedures were alike. A criterion test of 40 words was individually administered to each subject at the end of the instructional phase of the procedure and analyzed in a Winer Three-Factor Experiment with Repeated Measures design. Results of these tests revealed that the type of orthography used in instruction had no significant effect on the child's ability to recognize words. A bibliography is included. (DH)