Thirty-six nonreading children aged four and five years old participated in a study comparing the ease of learning words represented in traditional orthography (T.O.) and in three logographic systems: Carrier-Peak, Bliss, and Rebus. (The Carrier-Peak symbols, normally moveable abstract shapes, were converted to graphic form for the study.) The subjects, randomly selected from four day care centers in a midwestern metropolitan area, were assigned to one of four experimental conditions: Carrier-Peak, Bliss, Rebus, or T.O. Fifteen words, randomly selected from the pool of 26 words common to all four graphic systems, were presented to subjects in individual 15-minute sessions in which they were shown the words on stimulus pages, told what they were, and then asked to say them as the examiner pointed to them on response pages. Analysis of the results indicated that the three logographic systems were significantly easier to learn than T.O., that the partially iconic logographies (Bliss and Rebus) were significantly easier to learn than the noniconic logography (Carrier-Peak), and that the Rebus system was significantly easier to learn than the Bliss system. (The treatment pages used with all four groups are reproduced.) (GW)