Kronecker, Hall, Schäfer, Horsley, and others have found, when stimulating the spinal cord with rapid induced currents (ten, twenty, thirty, forty times per second), that the muscles always responded by giving a curve showing oscillations of one invariable period, no matter what the period of the stimulation may have been. With rapid stimulation, the muscles in the hands of the first named observers gave oscillations of twenty a second, in the hands of the latter observers of ten a second. These results are to be explained in the following way. When the central nervous system is stimulated, the muscles contract but never smoothly, for local fascicular movement, as the author has elsewhere shown, always occurring. These cause the registering apparatus to oscillate at its own period, just as any swinging body may be kept in motion by occasional disturbances. We can thus explain how each observer obtained the same number of oscillations every time he stimulated the nervous system; it was because he used the same recording apparatus each time. We can also see that the different periods observed by different observers are due to the fact that they each used a different recording apparatus. The author finds that on changing the recording apparatus (lever or tambour) the tracing obtained can be changed at will, and is practically a tracing of the oscillating period of the instrument used.