The purpose of the present study was to examine how the reading materials of the Japanese language and the delay time of delayed auditory feedback(DAF)influened reading rate and disfluency. Method The experiments cocsisted of two parts: the first experiment using familiar sentences and nonsense syllables as reading materials was made under 6 delay conditions of .00, .11, .15, .20, .25, and .30 sec : the second experiment using familiar sentences, nonsense syllables, and familiar words was done under 10 delay conditions of .00, .11, .15, .20, .25, .30, .35, .40, .45, and 50 sec.. Seven male undergraduate students(18-24 years in age)served as subjects of the first experment, and fifteen male undergraduate students(19-23 years in age)did as those of the second. The reading materials were placed at eye level immediately before the subject's head. The subjects were instructed to read the materials aloud at a usual reading and speaking rate. The apparatus producing DAF was a Sony taperecorder modified by the authors, and capable pf producing a wide variety of speech delays. The apparatus returned DAF channel speech of the readers to their ears with various delay times. The recorded speech under a normal condition and DAF conditions was caluculated, and analyzed by reading time and disfluency. Results (1) The greatest decrease of reading rate in the first experiment was found at the delay time of about .20 sec. in familiar sentences and nonsense syllables. (2) In the first experiment the reading rate of familiar sentences was remarkably faster than nonsense syllables, while the similarity of the pattern of reading rate over delay time was observed between sentences and nonsense syllables. (3) Reading rate under DAF condition in the second experiment was closely related to reading materials; sentences had much faster rate than familiar words or nonsense syllables. The effect of familiarity of reading materials on reading rate, however, could not be found. (4) The pattern of reading rate changes over delay time in nonsense syllables was much the same as the sentences in the first experiment, and that of the familiar words in the second, dependent upon the size(numbers of letters)of nonsense syllables. (5) The reading rate of sentences tended to be faster above delay time of .30, sec., while words typically were slower. (6) Disfluency of reading under DAF was most evident in nonsense syllables, somewhat more in familiar words. (7) The most outstanding effect of DAF upon disfluency in sentences and familiar words was obtained at delay time of .25 sec.. (8) Above delay time of .25 sec., sentence and familiar words produced an obvious decrease in disfluency, but nonsense syllables did not.