From June 1953 to January 1954, the water temperature and amount of flow of a spring called IITanaka-no-Yu", Misasa Hot Springs, Tottori Prefecture, were observed continuously. It issues 10 metres north of the riverside of the River Misasa and a stream runs 3 metres south of it. The water level of the River and the stream were also observed. The water level of the stream have close connection with the water volume of rice-field which surround that spring, because it is a watercourse of irrigation to rice-field. It was found that when the amount of flow increased, the water temperature rose, namely there was a positive correlation between them, and the correlation coefficient was 0.952 (highly significant). The analysis of variance showed that the amount of flow would be affected by the water level of the River but would not be severely affected by that of the stream. The correlation between the amount of flow and the water level of the River was also positive and the correlation coefficient was 0.731 (highly significant). As a result of the analysis of variance, the computation of confidence limit and the analysis of covariance, in summer the averages of the amount of flow and the water temperature would be higher than those in autumn
and winter. This difference was seemed to be induced by the water level of the stream. The end of the observation, a dam had been constructed 20 metres down (west) the River and she rised, but the effect of it must be observed after this.