The hypothesis for this study assumed that subjects who received a diagram representing the structural relationships of a story would be more likely to solve a new and analogous problem than subjects who received two analogs without a diagram. The 151 graduate students who acted as subjects were randomly assigned to one of four treatments: one or two analogs, with or without a diagram. Learners receiving the diagram were asked to copy it, to think about it while reading the story, and to rate its usefulness. The learners' solutions to a new analogous problem were then judged as correct or incorrect. No significant differences were found between groups receiving one or two story analogs, although a significantly higher percentage of subjects receiving the diagram were able to solve the problem. It was concluded that interaction with a diagrammatic representation greatly facilitated the transfer of solutions to an analogous problem and that, although multiple similar problem-solving experiences may help learners solve new problems analogically, the key variable is not the number of experiences, but the manner in which they are presented and processed. A list of 26 references is provided. (MES)