A study was conducted to determine the influence of a holistic versus a phonic approach in beginning reading on children's comprehension scores at the end of grade two. Two groups, each of approximately 350 students, were studied. Group one began reading instruction in first grade using the "Early-to-Read i.t.a. Program" (initial teaching alphabet), followed by either Lippincott or Ginn basals in second grade. Group two began reading instruction using the Houghton Mifflin reading series, and continued with this series or used the Ginn basals in second grade. At the end of second grade each child was given the Iowa Test of Basic Skills to determine comprehension. Group one, which had the heavily phonic based i.t.a. program, scored significantly higher than group two, which had the more meaning-based approach to reading. Group two had a much higher percentage of students scoring at the extremes of the testing range, suggesting that meaning-based instruction produced greater differences in achievement and might be a handicap for less able students but an advantage for more able students. There was no evidence to indicate whether the significantly higher achievement in group one was short term or would persist into the higher grades. (HTH)