[1] The ongoing work to resolve outstanding questions regarding the Cretaceous paleogeography of the North American Cordillera has spawned a healthy, and sometimes heated, examination of the geologic and paleomagnetic data on which conflicting interpretations of the amount of largescale northward translation of terranes are based. Butler et al. [2002] (hereinafter referred to as BEA) interpret paleomagnetic data from the granitic Ecstall pluton, located near Prince Rupert, British Columbia, as evidence that the body was deformed via large-scale folding after acquiring its remanent magnetization in the late Mesozoic. This inference is then used by BEA to conclude that, after accounting for this folding, the Ecstall pluton records a unique magnetic inclination that can be used to rule out large-scale translation hypotheses such as ââBaja BCââ [i.e., Irving et al., 1996]. In this comment, we recognize the value of the BEA study, which points out some very real complications in the magnetization history of this pluton. We wish to discuss the evidence for the fold postulated by BEA, to point out that there is little evidence to support its existence, and that the pattern of dispersion of the paleomagnetic data presented by BEA offers an alternate interpretation that provides a superior fit to the data, and also is consistent with the Baja BC hypothesis. [2] The paleomagnetism of the Ecstall pluton was first investigated by Symons [1974], who found that it recorded an inclination of remanent magnetization that was too shallow relative to expected directions for its age and location based on apparent polar wander of North America. This observation, which conformed to a general orogenwide pattern, was used by Beck [1980], Irving et al. [1996], and others, as evidence that portions of the western edge of the North American Cordillera had been transported relatively northward with respect to the interior of the continent, through distances exceeding several thousand kilometers, during latest Cretaceous and early Tertiary time. However, BEA reject this interpretation and instead regard their results as evidence that the pluton formed essentially in situ and acquired its suspiciously low inclination through local tectonics. The correct interpretation of the Ecstall shallow inclination has important tectonic implications. [3] Although paleomagnetic results for the Ecstall pluton are only a small (and far from straightforward) piece of the Cordilleran puzzle, they assume much greater consequence in light of their subsequent interpretation. For instance, evidence of folding of the Ecstall pluton is used by Butler et al. [2001] to propose that a miscellany of unrelated causes (local deformation, compaction shallowing in sedimentary rocks, remagnetization, etc.) rather than northward translation is the correct explanation for all of the consistently shallow inclinations in Cretaceous rocks from the western Cordillera. In turn, Butler et al. [2001] is cited by Dickinson [2003] to characterize northward transport as a ââgeomythââ and later, in a single sentence, to dismiss the results of 40 years of Cordilleran paleomagnetism [Dickinson, 2004]. Other examples of the tectonic importance of the Ecstall results exist. [4] In this discussion we take another look at the Ecstall paleomagnetic results. In agreement with BEAwe conclude that the paleomagnetic evidence requires folding of the pluton about a NW trending axis sometime after it acquired its remanent magnetization. However, we show that this is fully compatible with an origin far to the south, and that in fact the latter interpretation fits the evidence much better than the tectonic scenario advanced by BEA. We do agree with BEA that because of internal folding it is not possible to use the Ecstall plutonâs paleomagnetism to uniquely determine its latitude of origin. [5] Field and laboratory methods are well described in BEA. We agree that Ecstall granitic rock retains a welldefined, ancient direction of magnetization. The Ecstall results are unusually scattered (Figure 1), but they record two polarities which are nearly antiparallel, and the entire data set is clearly nonrandom. Figure 1 includes two reference poles: the mid-Cretaceous pole of Housen et al. [2003] (essentially identical to that used by BEA) and the late Cretaceousâearly Tertiary pole of Beck and Housen [2003]. Because data comprising the mid-Cretaceous referJOURNAL OF GEOPHYSICAL RESEARCH, VOL. 110, B01101, doi:10.1029/2004JB003346, 2005