ON THE TIME-SENSE. By LEWIS T. STEVENS. The experiments which form the basis of this article have been in progress during the past two years : the greater portion of them were performed, under the supervision of Professor G. Stanley Hall, in the Psychophysical Laboratory of the Johns Hopkins University, Baltimore ; the results there obtained have received confirmation from additional experiments made in the Physio- logical Laboratory of Professor Henry P. Bowditch, at the Har- vard Medical School, Boston. The method of experimenting was somewhat similar to that of Vierordt, 1 and consisted, in a general way, in impressing upon the mind intervals of time by means of a metronome, and in reproducing the same after the metronome had been stopped. The apparatus employed in the research was a horizontal- drum kymographion of Marey, with horizontal-screw attach- ment, by means of w r hich was obtained a continuous spiral tracing. A delicate time-marker, writing upon the sooted-paper coating of the drum, was attached to the vertical support seated upon the horizontal screw, and by the revolution of the latter was made to proceed slowly from one end of the drum to the other. A tuning-fork, or a vibrating rod, 2 and a resistance- coil were placed in the same electrical circuit with this time- marker. In a shorter circuit with the tuning-fork were placed a mercury-cup and a compound lever; this latter being so constructed that by depressing one end with the finger the other extremity (hook-shaped) was made to dip into the cup of mercury and thus close the circuit. When the hooked extremity was out of the mercury-cup, the time-marker registered the oscillations of the tuning-fork, but by tapping the lever the fork was shifted into the shorter circuit, and its vibrations failed for a moment to be recorded. By tapping the lever, therefore, at the beginning and the end of an interval of time, the interval could be recorded upon the sooted paper, and its length be subsequently determined by counting the number of vibrations between two successive inter- ruptions on the tracing. This end was attained at one stage of the investigation by inserting a Morse's key into the same circuit with the tuning-fork and the time-marker, and thus directly break- ing the circuit ; but the key possessed the disadvantages of requir- ing a considerable amount of muscular force to press it, and of 1 See Eeferences at the end. 2 In five out of the seven series of experiments to be reported, a tuning- fork vibrating 50 times per sec. was used ; in the remaining two series, in order to render the task of counting out the tracings less tedious, a rod making 10 vibrations per sec. was employed.