Tandem pore potassium channels (K2P) set the resting membrane potential of many neuronal types due to the generation of a potassium leak conductance. The leak conductance resulting from K2P expression also affects neuronal excitability by setting the input conductance. Based on in situ hybridization studies, the fifteen paralogous K2P genes have different, but often overlapping expression patterns. For example, adult cerebellar granule neurons (CGNs) express five K2P channel genes, TWIK-1, TASK-1, TASK-3, the TREK-2c splice variant and THIK-2 at high levels. However, identifying which K2P channels actually produce the leak conductance in any particular cell type is challenging, primarily because decisive pharmacological reagents are lacking. Thus, we believe that for the K2P family, gene knockouts will prove essential for identifying the K2P channel composition found in vivo. In this study we are primarily concerned with two K2P members: TASK-1 & TASK-3. The TASK-1 and –3 subunits are believed to form voltage independent non-inactivating K+ channels that are inhibited by acidic pH in the physiological range. In recombinant systems and in vivo they can assemble heteromerically if the cell type expresses both genes at suitable levels; alternatively, they can function as homomers if either TASK-1 or TASK-3 dominates. Using a TASK-1 knockout approach we have recently demonstrated that in adult CGNs, the majority of channels contributing to the pH-sensitive component of the standing outward potassium current (IK(SO)) are heterodimers of TASK-1 and TASK-3. The TASK-1 KO strain has also recently been used to examine the subunit composition of TASK-like K+ channel populations found in other neuronal populations; e.g. thalamic relay neurons of the lateral geniculate nucleus and dorsal vagal neurons. All these neuronal populations express TASK-1 and TASK-3 mRNA. However, we see little functional evidence of functional TASK-1 containing K2P channels in the thalamus and the TASK-like conductance recorded from vagal neurons appears to be more consistent with a TASK-1 homodimeric population. This illustrates a complexity to TASK channel functional expression that makes it difficult to simply correlate mRNA patterns with expression of specific K2P types. The ability of neurons to fire at high frequencies during sustained depolarisation is generally explained by the presence of voltage-gated ion channels. K2P channels are considered to be purely voltage independent channels with instantaneous activation/inactivation. However, CGNs lacking the TASK-3 type K2P channel exhibit marked accommodation of action potential firing in response to sustained depolarisation. We have examined the functional significance of TASK-3 channel expression in CGNs and, as expected, the magnitude of the standing outward leak conductance was significantly reduced in TASK-3 knockout mice. However, we also observed a reduction in the magnitude of a slowly inactivating conductance, not previously associated with K2P channel function. Examination of recombinant TASK-3 channel properties revealed that this K2P channel population does exhibit a kinetic component to its inactivation; a feature not shared by the closely related TASK-1 channel. We propose that the faster membrane time constant resulting from this TASK-3-mediated conductance enables CGNs to fire at high frequencies in response to sustained depolarization. This highlights a previously unappreciated consequence of K2P expression that will influence neuronal firing patterns in the CNS.