"Funny" channels are responsible for generation of pacemaker activity and autonomic modulation of cardiac rate. Because of their low conductance (Ϸ1 pS), single f-channels were elusive, and the first (cell-attached) recording published in 1986 1 was achieved thanks to a modification of the patch-clamp technique with 2 pipettes on the same cell, one to apply voltages and one to record single channels. With this arrangement there was no need to subtract capacitive transients in the patch pipette because there were none, which greatly increased the recording resolution. Cell-free inside-out f-channel recording was published later 2 also with a modification (voltage steps delivered via the bath pellet) for improved resolution.Because our laboratory has been the only one to publish single f-channel data, I read with interest the recent article by Michels and colleagues on single-HCN (hyperpolarization-activated, cyclic nucleotide-gated; the molecular correlates of native pacemaker channels) and human f-channel properties. 3 However, single channels recorded by Michels and coworkers do not appear to be the same as those published previously. One difference is in the conductance, which for native f-channels is nearly 20-fold higher 3 than previously reported. 1 Conductances reported for HCN isoforms are also much higher (Ϸ13-to 35-fold). These differences are not compatible with normal variability.Another substantial difference concerns the current time course. At or around Ϫ90 mV, time constants of activation of I f , HCN1, -2, and -4 have values close to 1200, 120, 1500 and 3000 ms, respectively. 4,5 Ensemble average records shown by Michels and collaborators 3 for HCN isoforms and I f are, however, flat and do not reveal any time dependence, reflecting an "instantaneous" rather than a time-dependent behavior. The similarity between kinetics of whole-cell and ensembleaverage patch current recorded simultaneously from the same cell was one of the distinctive properties of single-channel recordings previously reported, 1 which, together with the direct cAMP-induced activation demonstrated in cell-free mode 2 are characteristic signatures of f-channels.Another unexplained property, as noted by the authors, 3 is the much more positive activation threshold of single-channel than whole-cell currents; finally, mean first latencies on the order of tens of milliseconds 3 are short compared with the several hundred millisecond firstlatency values previously reported. 2 It is not clear why the single-channel recordings of Michels and collaborators differ so substantially from previous data. Single-channel time-dependent kinetics mimicking whole-cell kinetics and cAMP sensitivity during direct inside-out patch perfusion should be demonstrated to verify how these recordings relate to the pacemaker current.