Morphine is a potent analgesic, exerting its effects through the G protein-coupled μ-opioid receptor (MOR). Since the initial cloning of the MOR, many splice variants, differing only in the intracellular C-terminal tail, have been described in the human, rat and mouse but the literature describing their physiological role, distribution, or even existence, is confused. To clarify this, we used PCR to look for previously described splice variants throughout rat CNS. MOR1 and MOR1A were readily detectable and present in similar amounts in all tissues tested. MOR1B had a similar distribution but was only detectable using nested PCR and many more cycles. However, we could only detect MOR1C using nested PCR with 30 cycles and 26 cycles in each round respectively. Even so, its detection was erratic even in the same brain region. We did not detect MOR1D. Hence, we conclude that MOR1 and MOR1A are present in similar amounts in the CNS; MOR1B is present in lesser amounts but MOR1C is unlikely to have any physiological role and it is likely that, in this case, we are amplifying sequences from incompletely spliced intermediates. Given that MOR1, MOR1A and MOR1B differ only in the C terminal region we then studied possible functional differences between them at the level of agonist-induced desensitisation, using whole-cell patch clamp recordings from HEK-293 cells stably transfected with the G protein-coupled inwardly rectifying K⁺ (GIRK) channel subunits Kir3.1 and Kir3.2A and transiently transfected with receptor constructs¹. Receptor saturating concentrations of either morphine or DAMGO (a peptide MOR agonist) were applied to ensure comparable levels of receptor occupancy. For each splice variant/opioid combination, the evoked GIRK current reached a peak and then declined (desensitised) in the continued presence of the drug. MOR1 and MOR1A desensitised to both morphine and DAMGO to similar extents and with similar kinetics (MOR1, t½ 1.8 min; MOR1A, t½ 1.1 min). Although the rate of morphine-induced desensitisation of MOR1B was the same (t½ 1.7 min) as of MOR1, DAMGO-induced desensitisation (t½ 4.0 min) was much slower. Interestingly, the rate of DAMGO-induced MOR1B desensitisation was increased when receptor internalisation was blocked by a dominant negative mutant of dynamin, suggesting that DAMGO causes apparently slower desensitisation of MOR1B because the splice variant is able to internalise and recycle rapidly in a resensitised state². Morphine-induced desensitisation does not seem to be influenced by internalisation. In conclusion, we have shown that MOR1, MOR1A and MOR1B splice variants are present throughout the CNS and that the different C-terminal sequences can affect functional properties of the receptor.