The "Trond" protocol of nerve excitability tests has been used widely to assess axonal function in peripheral nerve. In this study, the routine Trond protocol was expanded to refine assessment of cAMP-dependent, hyperpolarization-activated current (I(h)) activity. I(h) activity is generated by hyperpolarization-activated, cyclic nucleotide-modulated (HCN) channels in response to hyperpolarization. It limits activity-dependent hyperpolarization, contributes to neuronal automaticity, and is implicated in chronic pain states. Published data regarding I(h) activity in motor nerve are scant. We used additional strong, prolonged hyperpolarizing conditioning stimuli in the threshold electrotonus component of the Trond protocol to demonstrate the time-course of activation of I(h) in motor axons. Fifteen healthy volunteers were tested on four occasions during 1 week. I(h) action was revealed in the threshold electrotonus by the limiting and often reversal, after about 100 ms, of the threshold increase caused by strong hyperpolarizing currents. Statistical analysis by repeated-measures analysis of variance enabled confidence limits to be established for variation between subjects and within subjects. The results demonstrate that, of all the excitability parameters, those dependent on I(h) were the most characteristic of an individual, because variance between subjects was more than four times the variance within subjects. This study demonstrates a reliable method for in vivo assessment of I(h,) and also serves to document the normal variability in nerve excitability properties within subjects.