15 
about half, although in some cases the slo\\ung is more progressive 
and drops by degrees to a certain point or in certain cases rapidly to 
zero. The rate in both auricle and ventricle usually remains the 
same until late in the poisoning, when the auricles are observed to 
be less affected than the ventricle and continue to beat for some 
time after the latter has ceased to contract. 
Another rather peculiar feature of the poisoning is that after the 
heart has been exposed to the action of the drug for some time the 
heart muscle seems to gain some tolerance for the poison, for al- 
though never regaining its normal rhythm, the toxic effect appears 
to become less and often the rate and output are secondarily aug- 
mented. The same thing is shown in other instances by the stop- 
page of the heart following the first introduction of the drug. After 
a short time it is allowed to regain its original rhythm by perfusing 
it with Ringer’s solution, whereupon a second introduction of the 
drug, although slovdng the rate, shows no tendency to check it 
entirely. Calm and Hepp ® in perfusing the hearts of frogs (species 
not recorded) reported some slovdng but no change in the energy, 
but their work may be questioned because of an insufficient range in 
the per cent of acetanilide used. TVeil ^ noted a primary increase in 
both rate and energy, but this was followed by a secondary decrease 
in both. Lepine reported that acetanihde increased the energy but 
decreased the heart rate. My own observation was that the energy 
was sometimes considerabl}^ augmented, but this effect was always 
of very short duration. Less often there appeared also a momentary 
increase in heart rate, but in the case of all my experiments this was 
quickly followed by a very marked slovdng. 
The next problem was to determine what strength of caffeine cit- 
rate could be used to the greatest advantage in perfusing the heart 
of Rana pipiens. It is well known that rigor is produced in the 
skeletal muscles and the heart of frogs, although the latter is not so 
readily affected, and small doses cause definite stimulation, both rate 
and output being increased. It is obvious, of course, that an amount 
of the drug should be used too small to produce rigor and yet suffi- 
ciently large to cause stimulation. By experiment this was found to 
be between one-fifteen hundredth per cent and one-five thousandth 
per cent. Below one-fifteen hundredth per cent the heart rate was 
most often slowed, and even at this dilution the output was very 
materially lessened. In stronger solutions both rate and output rap- 
idly fell to zero. Examples of the effects produced are shown in 
Table II. 
«Cahn and Hepp, Berl. kliu. Wchnschr., 1887, XXIV, 27. 
& Weil, Thesis, Paris, De 1’ Acetanilide, 1887, 47 pp. 
c Lepine, Revue de med.. Par., 1887, VII, 310. 
