Multiple Oral Rereading (MOR), which involves repeated reading of the same instructional unit, has been found effective in remedial reading instruction. In this study, which was designed to provide basic information about the dynamics of such repetition, 32 first-grade children were selected as subjects on the basis of their ability to read, out of context, a list of words taken from their reading textbooks. Four 20-word paragraphs were constructed from the words, and four corresponding random word lists were created by scrambling the words in each paragraph. Performances in reading paragraphs and word lists were compared for reading speed and errors and for changes over four successive readings of the same selection. It was found that the reading of words organized into paragraphs was consistently faster than the reading of the same words in a random list. Paragraphs were read faster on the first trial than were the comparable word lists on the fourth trial. Mean reading speed increased on each of four successive readings of paragraphs and word lists. However, the incidence of errors was not significantly different for paragraphs and word lists, and there was no significant decrease in errors over four successive readings of the same paragraph or word list. (Author/GW)