30 mT, we used the "double-demagnetization" technique (Tauxe et al., 1995) because samples tend to acquire a spurious magnetization in the direction of the alternating field. A few high-coercivity samples were demagnetized up to 180 mT.
Discrete samples were taken from the working half of the cores in standard 8-cm3 cubes. A few samples were taken with small glass tubes, 3 cm long by 1 cm in diameter and etched with a fiducial line. These samples are indicated by a "t" in front of the sample name in Table AT1 of the "Appendix." Coarse sampling (one sample per core section) of Hole A from each site was accomplished shipboard, while sampling of critical intervals at denser spacing (as much as 10 cm) was carried out postcruise. These critical intervals included the upper Paleocene to lower Eocene interval containing the C24r/C24n boundary at Sites 1262, 1266, and 1267. We also densely sampled several other reversal boundaries, predominantly in the Paleocene and Cretaceous, in the hope of clarifying or refining the reversal position. It should be noted that many of the cores sampled postcruise were characterized by numerous small cracks on the split surface. These were not desiccation cracks, as the cores were still quite moist, and we were unable to determine the origin of these cracks.
All of the postcruise samples, as well as all shipboard samples from Site 1267, were either alternating-field (AF) or thermally demagnetized. Based on the results of these samples, only some of the remaining coarsely spaced shipboard samples were demagnetized. Most samples were AF demagnetized in steps up to 40 or 60 mT. For AF fields 30 mT, we used the "double-demagnetization" technique (Tauxe et al., 1995) because samples tend to acquire a spurious magnetization in the direction of the alternating field. A few high-coercivity samples were demagnetized up to 180 mT.
Twelve samples from Hole 1267A (across the hypothesized C24r/C24n boundary) were instead subjected to thermal demagnetization. Samples were dried inside a magnetically shielded room, removed from their plastic cubes, and wrapped in aluminum foil. Natural remanent magnetization (NRM) directions for these dried samples are well grouped and statistically indistinguishable from samples measured wet from the same interval. Dried samples were heated in 50°C steps up to 550°C and then additionally at 575°C. Bulk susceptibility was measured between heating steps in an attempt to monitor sample alteration.
One sample from each core from Hole 1267A was subjected to further analysis in the form of the Lowrie three-axis isothermal remanent magnetization (IRM) experiment (Lowrie, 1990). A saturation IRM was acquired in steps, with a maximum applied field of 2.0 T. One exception was Sample 208-1267A-8H-4, 88 cm (0712), where the maximum field was 2.5 T. This saturation IRM, followed by IRMs of 0.3 and 0.1 T, were applied in three orthogonal directions in order to separate the high-, intermediate-, and low-coercivity spectra. The samples were then thermally demagnetized to provide information on the blocking temperature spectra of the various coercivity fractions. A small amount of material from these same samples was reserved for hysteresis measurements, which were carried out on a Princeton Measurements Corp. MicroMag Model 2900 Alternating Gradient Force magnetometer with a maximum applied field of 1 T.
