Conventional tests of cardiac parasympathetic nervous system (PNS) activity were compared with the high frequency component of the heart rate power spectrum in a heterogenous group of normal subjects encompassing a wide age range. The data suggest that the linear associations between the various conventional tests of parasympathetic nervous activity even when statistically significant, were relatively modest, with r values ranging from 0.23 to 0.53. Three of the five conventional tests of parasympathetic nervous activity were significantly correlated (r = 0.33 to 0.46) with the absolute high frequency power of RR variability. However, these relationships were poorer and non-significant when the high frequency power spectrum was normalized for total power (r = 0.06 to 0.19, NS). An evaluation of the heart rate responses to cough and a single maximal hand grip indicated that the responses were repeatable but that the extent to which these manoeuvres induced vagally mediated cardioacceleration was significantly lesser than the other conventional tests of PNS activity. Taken together, the data suggest that despite the advent of heart rate variability measures, it is advisable to use multiple tests of parasympathetic nervous activity while evaluating autonomic dysfunction, since, despite the specificity of the tests, there is a variable correlation between them.