Notes: An increase in the accuracy and age resolution of the apparent polar wander path of the Pacific plate could be important for testing reconstructions that relate the motion of Pacific basin plates to other plates, for testing if hotspots in different ocean basins are stationary relative to one another, and for estimating the motion of hotspots relative to the spin axis. With these goals in mind, herein we investigate how accurately a palaeomagnetic pole can be estimated from skewness analysis of many crossings of a single magnetic anomaly on the Pacific plate. Apparent effective remanent inclinations of the sea-floor magnetization were estimated from the skewnesses of 132 useful (out of 149 total) crossings of anomaly 25r (56.5–57.8 Ma) distributed over a distance of more than 11000 km across the Pacific plate. These estimates were inverted to obtain a best-fitting palaeomagnetic pole latitude, pole longitude, and anomalous skewness for this single reversed-polarity chron. The best-fitting model gives a pole of 78.2°N, 4.8°E with a 95 per cent confidence ellipse having a 6.4° major semi-axis oriented 93° clockwise of north and a 4.1° minor semi-axis; anomalous skewness is 16.2°± 4.6° (95 per cent confidence limits). We also investigated the effect of the dependence of anomalous skewness on spreading rate by correcting our data using an empirical model. The pole obtained from the inversion of this alternative data set lies a statistically insignificant 0.6° from the pole obtained using no correction. That a pole with usefully compact confidence limits and a narrowly resolved, precisely estimated age can be so determined suggests that an accurate apparent polar wander path with a fine-age resolution can be determined for the Pacific plate by applying the same approach to the shapes of other marine magnetic anomalies.Comparison of our chron 25r pole with other Pacific palaeomagnetic and palaeoequatorial sediment facies data indicates that the Pacific plate remained nearly stationary relative to the spin axis during the Eocene (-0.05°Myr−1± 0.28° Myr−1), but probably moved rapidly northward during the Paleocene (0.83° Myr−1± 0.46° Myr−1). Comparison of these data to latitudes of dated volcanic edifices along the Hawaiian-Emperor chain indicates that the Hawaiian hotspot drifted southward by 10.2°± 3.4° (95 per cent confidence limits) since 57 Ma, but only by 1.7°± 1.9° since 39 Ma, which gives a southward displacement of 8.5°± 3.9° (95 per cent confidence limits) between 57 and 39 Ma, corresponding to a rate of southward motion of 52°24mm yr−1. Incorporation of realistic uncertainties of volcano ages would increase these uncertainties considerably, however. We also examined the distance between the crossings of anomalies 25 and 27 on all the profiles we analysed; along the palaeo-Pacific-Farallon boundary these distances are inconsistent with the joint hypotheses of symmetric spreading and single Pacific and Farallon plates between 62 and 56 Ma, indicating that the evidence for a single Pacific plate in early Tertiary time is not as compelling as it had previously seemed.

Notes: We present a method for simultaneously determining palaeomagnetic poles and anomalous skewness from the observed skewness of marine magnetic anomalies from a single plate. Skewness is esimated by visual comparison of phase-shifted observed anomaly profiles with an ideal zero-phase synthetic anomaly calculated for sea-floor formed and observed in a vertical magnetic field. We assume that each crossing of the same anomaly or anomalies is anomalously skewed by an identical amount. This assumption is supported by the great reduction in squared error (i.e.,x2) obtained by adding just this one adjustable parameter. Our inversion procedure determines the palaeomagnetic pole and anomalous skewness that give a maximum likelihood best fit in a least-squares sense to the observed effective remanent inclinations, which are inferred by simple calculation from the observed phase shift. Our method determines 95 per cent confidence limits both from a constant-chi-square boundary and by linear propagation of errors and also estimates the information distribution of the data. The anomalous skewness values we determine agree with values estimated from comparison of anomalies across a spreading centre. These encouraging results suggest that further skewness studies may permit variations in anomalous skewness with age to be quantified, provide constraints for models that describe the processes affecting the magnetic source layer during its creation and evolution, and permit oceanic plate apparent polar wander paths to be determined with a fine age resolution.