A hypothesis is presented that explains the physiological reasons why the magnitude of respiratory sinus arrhythmia (RSA) appears to correlate with cardiac vagal tone. The hypothesis is that RSA is an intrinsic resting function of the cardiopulmonary system. Although RSA is mediated by respiratory modulation of cardiac vagal outflow and its magnitude is used as an index of cardiac vagal activity, RSA itself reflects cardiorespiratory interaction. RSA is universally observed among vertebrates throughout the evolution, suggesting that it may bear an intrinsic physiological role. Recent studies have shown that RSA improves pulmonary gas exchange efficiency by matching alveolar ventilation and capillary perfusion throughout respiration cycle. This suggests that in resting animals and humans, RSA could save cardiac and respiratory energy by suppressing unnecessary heartbeats during expiration and ineffective ventilation during waning phases of perfusion. Furthermore, evidence is accumulating for possible dissociation between the magnitude of RSA and vagal control of heart rate, suggesting separated and independent regulations for respiratory modulation of cardiac vagal outflow from those for cardiac vagal tone. By our hypothesis, the apparent associations between RSA and cardiac vagal tone are explained as indirect consequences; i.e., whenever the cardiac vagal tone changes in response to the resting level of the cardiopulmonary system, RSA appears to change parallel to it. Our hypothesis seems more consistent with both physiological and clinical evidence about RSA than that presuming RSA is an index of cardiac vagal activity.