Reproducibility of aerobic power was investigated on the basis of 382 measurements of aerobic power for 54 boys and 45 girls aged 5 and 6 years. Each subject ran about 360m on the ground with warming up and maximal effort for last one minute. Expired gas was collected into a Douglas bag during the last one minute of running. Each Douglas bag was carried by one of the investigators who ran along the subject. Expired gas was analyzed using Respilyzer BM l0 (Fukuda Sangyo Co.) which was calibrated frequently by a micro-Scholander apparatus. Heart rate was recorded continuously using telemetry system during running. Two series of measurement of aerobic power were carried out in October, 1981 (Experiment l) and in March,1982 (Experiment 2) on the same subjects. Subjects ran twice with an interval of one or two days in each experiment. The following results were obtained in the present study. 1. The mean values of aerobic power in Exp. 1 were 0.888 1/min and 0. 925 1/min for boys of 5.95 years old, and 0.890 1/min and 0.948 for girls of 5.94 years old. The values in Exp. 2were 0.997 1/min and 0.985 1/min for boys of 6.43 years old, and 0.971 1/min and 0.948 1/min for girls of 6.48 years old. Correlation coefficients between first and second measurements in aerobic power were 0.705 (p<0.001) and 0.582 (p<0.001) for boys and girs in Exp.1,and 0.694 (p<0.001) and 0.781 (p<0.001) for boys and girls in Exp.2, respectively. Mean va1ues of individual difference between first and second measurements in aerobic power were 11.3% for boys and 12.0% for girls in Exp. l and 8.9% for boys and 7.8% for girls in Exp. 2,respectively. It might be concluded from these values that reproducibility of aerobic power were in an acceptable level of confidence. 2. The corelations between first and second measurements in ventilation ranged from 0.708 to 0.749 for boys and girls. Those in maximum heart rate were 0.778 for boys and 0.807 for girls .3. There was no significant reproducibility in respiratory exchange ratio (R.). It seems that respiratory exchange ratio is not suitable for a criterion of aerobic power.