Paleomagnetic studies on middle and early Tertiary rocks from the Coast Range consistently find inclinations that are a few degrees shallower than predicted by reference poles for the craton. Rocks of very similar age and lithology in the Cascade Range and elsewhere in Washington and Oregon do not show this effect. This suggests that the Coast Range has moved northward several hundred kilometres, relative to stable North America. This movement could have been accomplished in a few million years prior to accretion of the Coast Range to the continent, by passive transport aboard the Kula and/or Farallon plates. Alternatively, northward displacement could reflect dextral strike-slip faulting in the North American borderland. Probably both mechanisms contributed.