The analysis of the 450 discrete samples indicates various behaviors during demagnetization. Orthogonal vector projections of the direction of remanent magnetization during demagnetization are shown in Figure F2 (Zijderveld, 1967).
Samples with normal and reversed polarity without an overprint or with a small overprint were generally found (Fig. F2A). These overprints were destroyed at a demagnetization level of 5-10 mT. Hence, the overprint of these samples was easily destroyed during standard demagnetization at 20 mT of the archive halves aboard the JOIDES Resolution, and the shore-based analysis of the discrete samples confirmed the polarities that were found on board. In Figure F2B and F2C, an example of another type of demagnetization behavior is shown. The normal overprint of the reversed characteristic remanence is completely removed only at higher fields. Figure F2C shows the Zijderveld plot of a sediment sample with normal characteristic remanence and reversed overprint. Only after demagnetization to 45 mT has the overprint been removed. The standard shipboard demagnetization at 20 mT was not enough to obtain the accurate direction of the characteristic magnetization. Hence, in the parts of the site where that behavior was found, the magnetic polarity record had to be corrected after analysis of discrete samples. This occurred in Cores 183-1138A-17R, 18R, 37R, 48R, 51R and 53R.
Data from the continuous and discrete measurements on sediments from Site 1138 are shown in Figure F3 together with the polarity record determined from the shipboard inclination data and the characteristic inclinations of single sample cubes.
For sites in the Southern Hemisphere, negative inclinations correspond to normal polarity of the Earth's magnetic field, and positive inclinations correspond to reversed polarity. The inclinations obtained from the discrete samples confirm, in most cases, the results obtained on board during Leg 183 from archive-half cores. An interpretation of the inclination record in terms of normal and reversed polarity is shown by black and white bars, respectively.
The AF demagnetization experiments on ~100 single samples from Hole 1140A confirmed the results obtained aboard the JOIDES Resolution. No additional information could be obtained for the polarity record from discrete sample measurements at this site. In Figure F4, representative examples of AF demagnetization and decay curves of samples from Site 1140 are shown. Some sediments carry a single component without overprint (Fig. F4A). Most of the sediments have a partial secondary magnetic component, which, however, was easily destroyed at 20 mT (Fig. F4B, F4C).