The electrocardiogram of frogs were obtained in winter (January), spring (April), summer (July) and autumn (September and November). Electrocardiograms were recorded applying electrodes to the atria, ventricle and apex of the heart by unipolar or bipolar leads. V wave was recorded prior to P wave, for the presence of the sinus venosus which controls the automaticity of the frog heart, in four seasons. Regardless of the leads or the position of the electrodes P wave was diphasic and wide. According to the rise of temperature the rate of heart beat was increased, and V-P and P-R interval were shortened. Two regression line between R-R interval and both V-P interval and P-R interval were drawn. These were calculated as V-P interval=1 0.276R-R interva1+0.067±0.15 (sec.) and P-R interval=0.179R-R interva1+0.155±0.1 (sec). From these calculation the larger gradient of V-P interval than P-R interval was suggestive that the heart rate is more dependent on the changes of V-P interval than that of P-R interval. Changes of the heart rate were also measured in four seasons and artificial temperatures. Two regression lines between the heart rate (H.H.) and both seasonal temperature (T) and artificial temperature, were drawn. These two lines were calculated as H.R.=20+3.71 (T-10) and H.R.=32+1.425 T respectively. From two gradients of the above equations it is considered that the changes of the heart rate in artificial temperature were milder than that in seasonal temperature. The number of RBC and WBC of frogs were measured in four seasons and a tendency of the changes was observed according to the seasonal variation.