Just as a speech-language pathologist uses knowledge of normal language development to make a diagnosis of language delay, normative data on fluency development can be used to make a differential diagnosis of normal disfluency from fluency disorders. Data at each discrete age level on normal disf luency is an important aspect in understanding the development of fluency.The purpose of this study was to compare the frequency of specific disf luencies in 3 year old and 5 year old normal 2 male children in terms of part-word repetitions, word repetitions, phrase repetitions, interjections, revisionincomplete phrases, disrhythmic phonations and tense pauses.The disfluencies were observed while each child spontaneously interacted with an investigator in a clinical room. Two questions were addressed:1. Do three-year-old male children exhibit a higher overall frequency of disfluencies than five-yearold male children? 2. Do three-year-old male children exhibit a greater frequency of certain disf luencies than five-yearold male children? Twenty normal preschool male children comprised the subjects of this study. All children were randomly selected from the greater Portland area and passed the selection criteria. The children were of two groups:10 three-yearold male children (x=36 months) and 10 five-year-old children (x=60 months). After spontaneous speech samples were obtained and recorded from each subject, the investigator transcribed the text verbatim and categorized seven types of disfluencies. The Mann-Whitney U Test (Siegel, 1956) was used to analyze the data. The results indicated that a statistically significant difference on the total disfluency did not exist between the two age groups.Results indicated that at the 0.05 level of confidence, a statistical significance did exist for interjections. Male children in the five-year-old age group evidenced a higher number of interjections than the three-year-old children.No other statistically significant differences were found.The results yielded the following conclusions: 3 1. The three-year-old children did not exhibit a higher overall frequency of disfluencies than the five-year-old children. Although no statistical difference was found, the data indicated that the five-year-old children exhibited a higher overall frequency than the three-year-old children.2. There was no statistically significant difference in the frequency of occurrence of part-word repetitions, word repetitions, and phrase repetitions exhibited by the two age groups of normal male children.3. There was no statistically significant difference in the frequency of occurrence of revisionincomplete phrases, tense pauses, and disrhythmic phonations.