The author reported the value of stroke-volume and momentum of cardiac ejection of normal subjects obtained by ballistocardiogram, and at the same time these methods of measurement were critically examined. As ballistocardiogram showes, as previously reported, the impulse wave of circulation which is mediated by human body, the values obtained by it inevitably contain some degree of inexactness. In quantitative investigation by ballistocardiogram if we want to get the minimum error of it, it is better to use the value of momentum of cardiac ejection, where survey of the relation of it to other physiologic functions is sacrificed. In order to supplement this weak point, one needs deducing stroke-volume from ballistocardiogram, though its value is not so exact. In the latter it is advisable to investigate the relative changes in same subject.