and forms it by its relaxation, and each time it does so it gets a shock. It thus dances to its own music, as you now see and hear.
We have hitherto studied muscular movements caused either by a direct electrical stimulus or by the action of a nerve. There are, however, rhythmic movements of muscular substance. You see here a frog's heart attached to an apparatus by which it can be fed with blood, and you see it beating with great regularity. The heart is caused to work a little manometer, and on the mercury in one limb of the manometer I have placed a little glass rod bearing a flag, so that you may see the flag moving up and down with each beat of the heart. In this way a heart can be kept alive for hours, and we can estimate the amount of work it is able to perform. This also illustrates a method by which physiologists can examine the influence of substances on the heart. Thus we might feed the heart with blood free from any poison and note how it worked. Then we might feed it from this other tube with blood containing a small percentage of the substance to be examined, and again note the effect. By comparative
