Tectonic tilt is a possible explanation for the discordant Cretaceous paleomagnetic poles obtained from intrusions in the Coast Belt of the Canadian Cordillera. If tilting of these intrusions has taken place, post-Cretaceous intrusions may have been similarly tilted. Paleomagnetism of the Middle Miocene Rogers Creek pluton (16 Ma) in the central-southern Coast Belt was utilized to test the tilt hypothesis. Well-grouped stable primary magnetizations have been isolated in 11 sites with both normal and reverse polarities. The mean remanence component, directed at D = 000.7°, I = 67.9° (α95 = 3.9°, paleolatitude 50.9 ± 9°N, paleopole 88.3°N, 257.5°E, A95 = 5.7°), was likely acquired upon emplacement at 16 Ma. This direction agrees with the Middle Miocene reference direction (D = 357°, I = 68°) calculated from the updated Miocene (10–20 Ma) cratonic reference pole for North America. This concordant result, when combined with concordant data from other mid-Tertiary intrusions to the south, defines a southern Coast Belt domain that shows no evidence of post-Middle Miocene tectonic tilt. This result does not preclude tilting of the mid-Cretaceous Spuzzum batholith within this domain; however, if tilting took place it must have occurred prior to the Middle Miocene. The concordant data from the Rogers Creek pluton provide a Middle Miocene paleohorizontal datum for future work investigating crustal tilts in the Coast Belt.A comparison of Tertiary Coast Belt paleopoles demonstrates that they are systematically rotated clockwise and are near-sided relative to the cratonic reference poles. This implies that the Coast Belt has been displaced 300 km southward and rotated 8° clockwise, or regionally tilted downward to the northwest, or some combination of these two end members, since the Late Miocene. These scenarios appear untenable. Evidence is given that the directional bias of the paleopoles is caused by unremoved present Earth's magnetic field (PEMF) overprint magnetizations, which comprise 10–25% of the total remanence in most cases. The implication is that demagnetization and analysis techniques used in previous Tertiary Coast Belt studies, to varying degrees, may have insufficiently removed the PEMF overprints. The case is made for the importance of thorough alternating field and thermal step demagnetization, of all specimens, in order to maximize magnetic component data from which the principal magnetic components can be best isolated and PEMF components removed.