Records of the Cobb Mountain Subchron from the Bermuda Rise (ODP LEG 172)

Duplicate records of the Cobb Mountain Subchron were obtained from two holes (Holes 1063B at 33°41.204′N, 57°36′W and 1063C at 33°41.181′N, 57°36′W) drilled in a sediment drift at Ocean Drilling Program (ODP) Site 1063 on the northeast Bermuda Rise. We sub-sampled the cores from both holes using U-channels (up to 1.5 m long samples with 2×2 cm cross-section) focusing on continuous sampling through the late Matuyama interval containing the Cobb Mountain Subchron. Results of standard progressive alternating field demagnetization experiments reveal that these sediments generally have a stable characteristic component after removal of a low-coercivity drilling overprint. Additional discrete samples (∼7 cm3 oriented cubes) used in progressive thermal demagnetization studies display similar behavior. By taking duplicate sets of U-channels, we were able to identify and remove several measurement artifacts, such as edge effects that affect measurements made near the ends of the U-channel samples. The upper normal-to-reverse polarity transition was not obtained because of a coring gap or an interval of coring deformation. After removing measurement artifacts, the records from the two holes were stacked to obtain a composite record. Virtual geomagnetic poles (VGPs) computed from the composite record follow a path through the central Pacific during the lower polarity transition, very similar to VGP paths from Cobb Mountain Subchron records from other sites around the world. Furthermore, the VGPs cluster near southern Africa early in the reversal process and then group in a nearly antipodal patch in the north Pacific late in the reversal. Taken together, these records indicate that the transitional field remained nearly dipolar through the polarity reversal, or at least during the beginning and end of the reversal when the field was dominated by an axial dipole tilted about 20–40° to the spin axis. If a tilted dipole is an inherent part of the initiation and ending of a reversal, it may help explain why many sediment recorders, which are thought to smooth the field, record antipodal VGP paths through many reversals.