The possibility, heard during a conference session while I was a graduate student, of measuring vagal tone from heart rate changes during respiration, hooked me into heart rate variability research. Vagal tone seemed to be a sort of Holy Grail. Anything that promised to increase it would be life sustaining, and that claim would be testable in an objective way. Loss of vagal tone could serve as an early warning signal and catastrophe could be prevented. The truth has proved to be far more complex, but it is my impression, as a reviewer of numerous manuscripts in the field that the essence of this myth persists. Where then does a more nuanced view of this field take us?The first issue is the confusion between vagal tone and what is really measurable from heart rate variability (HRV) that is vagal modulation of heart rate. Indeed, I am embarrassed to admit that one of the first review articles I wrote in this field (in 1993) was entitled "Heart rate variability: A measure of cardiac autonomic tone." 1 Malik and Camm clarified this distinction in the same year in their article "Components of heart rate variability-What they really mean and what we really measure." 2 Understanding the difference between vagal tone, as directly measured by vagus nerve activity and vagal modulation as measured by HRV is critical. Although direct measurement of peripheral (although, importantly not cardiac) sympathetic nerve traffic is now feasible, direct measurement of vagal activity is too invasive to be practical. Thus, we are limited to two measurements. Basal vagal tone can be estimated by giving atropine and measuring heart rate in the absence of all parasympathetic activity. This, however, does not provide any information about vagal tone under dynamic or stress conditions. Alternatively, vagal modulation of heart rate can be estimated, under controlled laboratory conditions or in an ambulatory setting, using appropriate manipulations of the beat-to-beat changes in heart rate and, as demonstrated in the current article by Kuo et al., 3 under certain circumstances, reflect vagus nerve traffic. However, the relationship between vagal activity and HRV does not appear to be linear, and a lack of vagally modulated HRV could indicate either a nonresponsive or a saturated system. In human studies, for example, Goldberger et al., 4 using phenylephrine and nitroprusside to stimulate and reduce vagally modulated baroreflex activity in subjects under beta-blockade, found that the relationship between HRV and parasympathetic effect was far from linear. They observed an ascending limb where HRV increased with increasing parasympathetic effect, followed by a plateau level. After that, HRV decreased with increasing parasympathetic effect. Moreover, this relationship was not uniform across subjects. A useful analogy might be to imagine that we can measure the height of the wave (modulation) but not the depth of the ocean (tone) using HRV.A second problem is equally important in clinical research. The various HRV indices that reflect vagal modula...