This study discusses the nature of rhythm as it may apply to speech and language, reviews some of the literature on the development of rhythm, and presents some thoughts relating these findings to specific examples of children's speech. There is evidence to support the view that one need not look at the exact rhythm of any utterance, but only for sequential organizing principles that children must recognize are at work in the language and then see how they go about learning these principles. Vanderslice and Ladefoged's study of English prosody leads to the notion that the single most important rhythmic phenomenon in English is the distinction between stressed and unstressed vowels. While the small amount of research on the development of stress, rhythm and segment-timing in children's speech reveals some inconsistencies, it leads to some tentative conclusions. Children appear to produce durational and other prosodic differences as early as 18 to 24 months, but their productions are characteristically variable until much later, stabilizing to adult norms around age ten. In conclusion, examples of children's utterances illustrate some rhythmic phenomena of interest, namely, examples of the correct or incorrect reduction of light syllables, examples of deletion of whole syllables, and samples of doubling or extreme lengthening of phrase-final syllables. (AMH)