Abstract
A new sweat capsule was developed, which can accurately detect the local perspiration from a human body with a high response sufficient to follow the rapid and unsteady change. By a contrivance to diminish vapor stagnation in the capsule, the time lag in detection could be decreased to about 0.73 seconds. The experiment was done on two famale subjects in a climate-regulated chamber. The time variation of the perspiration was analysed by the Fast Fourier Transform (FFT). The results suggest that it behaves like as 1/f-fluctuation for lower frequencies but as 1/f2-fluctuation for higher ones, and that sweating phenomenon can be classified into different kinds by FFT analysis of the data of local perspirating rate measured by the present capsule. The apparatus was also equipped with a microscope and a video camera, by which the sweat droplets on the skin surface were directly observed.