The paleomagnetic results of the peridotites from the two holes are summarized in Table 1 and Table 2. The most common features are outlined below.
The NRM intensities of the peridotites (Table 1, Table 2) range from 14 to 924 mA/m in Hole 897D and from 290 to 2455 mA/m in Hole 899B. In both holes, the intensities become stronger in the lower part of the peridotite section (772.18-819.48 m below seafloor [mbsf] in Hole 897D and 475.10-511.56 mbsf in Hole 899B). Although the intensity in Hole 899B is scattered (Fig. 4), the peaks and lows in the downhole profile appear to coincide with petrographical boundaries in the section. In Hole 897D, the remanence successively shows a steplike decrease at first and then increases with depth. In detail, it follows three trends: (1) from 694.49 through 754.86 mbsf, the NRM intensities are scattered with a mean of 296 mA/m; (2) from 755.81 through 763.24 mbsf, the intensities are much weaker (mean 33 mA/ m); and (3) from 772.69 through 828.40 mbsf, the intensities are stronger (mean = 394 mA/m) and more scattered (Fig. 4).
One of the major experimental requirements in paleomagnetic research is to isolate the characteristic remanent magnetization by selective removal of secondary magnetization. As shown in Figure 5, thermal demagnetization on Samples 149-899B-23R-1, 12-14 cm, and 149-897D-16R-2, 72-74 cm (both from the fresher lower part of the section) removed a "soft" component, probably of viscous origin, at low to intermediate temperatures (200°-400°C). Up to 585°-620°C demagnetization, the magnetization revealed the stable component of magnetization. In the altered upper part, however, most samples show a single component of magnetization during demagnetization (Fig. 6). In addition, it appears that there is no significant difference in demagnetization behavior of peridotite samples throughout this part of the section whether samples are from the matrix, clasts, or even veins. Examples of this behavior are shown in Figure 6: Sample 149-899B-20R-2, 112-114 cm, which was specifically chosen from a heavily veined piece of core, exhibited magnetic properties identical to veinless Sample 149-899B-21R-2, 75-77 cm.
Although in general a stable component of magnetization can be identified by both thermal and AF demagnetization techniques (Fig. 5, Fig. 6), the removal of secondary magnetization was better accomplished through thermal demagnetization than through AF demagnetization. The dominant magnetic mineral in these peridotites appears to be magnetite as indicated by the unblocking temperatures and coercivities. The magnetically cleaned inclinations from both holes are systematically shallower than the expected inclinations at the drilling sites, assuming that Iberia has been part of stable Europe since the Cretaceous. The discrepancy in inclination (about 25°) may indicate that Iberia had a microplate nature at the time of acquisition of the magnetization. As discussed later, Iberia is indeed believed to have been independent plate in the Cretaceous.
The magnetic susceptibility (K) is controlled by the volume concentration of ferromagnetic minerals as well as by grain size and other parameters such as stress. The magnetic susceptibility of Hole 899B (mean 2.959 × 10-2 SI units) is about 1.4 times higher than that of Hole 897D (mean 2.067 × 10-2 SI units). Although the scatter in susceptibility is less than that of the NRM intensities, the susceptibility values show a pattern similar to the NRM intensities (Fig. 7).
The Koenigsberger ratio (Q ratio) is defined as the ratio in a rock of remanent magnetization to the induced magnetization in the Earth's field:
Q = NRM (A/m) / (K (SI) • H (A/m), (1)
where K is the magnetic susceptibility in SI units and H is the local geomagnetic field (the International Geomagnetic Reference Field value at the Leg 149 sites [45,000 nT = 35.83 A/m] was used for calculating Q). In general, the Koenigsberger ratio is used as a measure of stability to indicate a rock's capability of maintaining a stable remanence.
The distribution of the Koenigsberger ratios in both holes (Fig. 8) resembles that of the NRM. Although the mean susceptibility values of the two holes are comparable, Hole 899B has a higher mean intensity of the remanence, and consequently, the Koenigsberger ratio of Hole 897D (mean 0.423) is lower than that of the Hole 899B (mean 2.049). Within the altered upper part section in Hole 899B, a sharp increase in the Koenigsberger ratio occurs at 421.93 mbsf (Sample 149-899B-21R-4, 97-99 cm). A similar spike is also present in Hole 897D (Sample 149-897D-13R-1, 62-64 cm). The Koenigsberger ratio peaks in both holes may serve as one of the stratigraphic markers for these two sites (the significance of this peak is discussed in more detail later and is summarized in Table 3). The Q ratio is less than 1.0 in all but two of the samples in Hole 897D, whereas in Hole 899B it is normally close to or slightly greater than 1.0 (with only four exceptions). The low values of the Q ratio demonstrate that induced magnetization would either dominate or be comparable to that of the remanent magnetization if present in an external magnetic field. This underscores the importance of measuring these rocks in a field-free room at the shore-based laboratory.
AMS is a physical property of rocks that is used for petrofabric and structural studies. It measures the sum of the anisotropies of the individual minerals in a rock. AMS results can be described by an ellipsoid of magnetic susceptibility, with dimensions defined by the magnitudes of the principal susceptibilities. These lie along the three orthogonal axes of the ellipsoid and are designated the maximum, intermediate, and minimum susceptibilities (K1, K2, and K3, respectively). These quantities are combined in various ways to describe different features of the ellipsoid and of the magnetic fabric it represents (Hrouda, 1982). AMS in peridotites most likely arises from the preferred orientation of anisotropic magnetic minerals (MacDonald and Ellwood, 1988). It is probable that net preferred alignment can be enhanced in these rocks by massive flow during emplacement. Thus, AMS is possibly relevant to distinguishing between emplacement modes. The ratio K1/K3 (P in Table 1, Table 2) is commonly used as a measure of the degree of anisotropy. In the studied samples it is very low, ranging from 1.010 to 1.261 in Hole 899B and from 1.030 to 1.267 in Hole 897D (with five exceptions in the lowest part of the section, see Fig. 9), which suggests that there is very little anisotropy in the samples. The values of the anisotropy factor P at both holes show a pattern similar to the magnetic susceptibility. Under the assumption that cooling of these peridotites was contemporaneous with the emplacement, the AMS results would then suggest that the emplacement process of these rocks was probably not under a strong stress field.
In this study, polarity sense can be assigned with reasonable assurance on the basis of the inclination determined from discrete sample. The most exciting though unexpected result generated from this study is the identification of magnetic polarity zones in the peridotites. As shown in Figure 10, the inclinations of characteristic magnetization in the fresher lower part of the peridotite section are stable and show a consistent polarity pattern in a depth zone of about 21m. In Hole 897D, this zone starts at 743.15 mbsf (Sample 149-897D-16R-2, 72-74 cm), which corresponds to the onset depth of the fresher lower part of the section, and ends at 763.24 mbsf (Sample 149-897D-18R-1, 16-18 cm). In Hole 899B, it ranges from 435.72 mbsf (Sample 149-899B-23R-1, 12-14 cm) through 457.81 mbsf (Sample 149-899B-25R-3, 34-36 cm) and also coincides with the first appearance of greenish, fresher peridotites in the lower part of the section. At both holes, the inclinations of samples are predominantly negative (reversed) within this zone. In contrast, the inclinations of samples from the more "oxidized" upper part are almost all positive (normal).
There is also a difference in mean inclination between the two holes. In Hole 897D, the mean inclination for the upper part is 30.2° (N = 5, a95 = 20.5°, k = 19.4) and the mean inclination in the lower part is -31.4° (N = 12, a95 = 10.8°, k = 16.5). In Hole 899B, the mean inclination in the upper part is 22.6° (N = 20, a95 = 9.5°, k = 11.3) and for the lower part the mean inclination is -20.1° (N = 8, a95 = 8.3°, k = 49.2). These mean inclinations are significantly shallower than both that of present-day field (~64°) and those calculated from Cretaceous to Tertiary reference paleopoles for Europe, Africa, and North America, which range from 40.3° to 51.7° (see Van der Voo, 1993, p. 136, for the reference paleopoles).
Two observations of the paleomagnetic data from Site 897 are worth mentioning:
1. Several samples from the top of the altered upper part display negative inclinations (hence, reversed polarity?), suggesting that the normal polarity magnetization in the altered upper part was probably not a Holocene overprint. The most diagnostic examples are shown in Figure 11: a light-colored clayey limestone Sample 149-897C-66R-4, 15-17 cm, which was assigned a Late Cretaceous age based on the preliminary shipboard biostratigraphic ages, as well as a peridotite Sample 149-897D-11R-4, 71-73 cm, are both reversely magnetized. Similarly, indications for this reversed signal are present in peridotite Samples 149-897D-11R-2, 10-12 cm, and 149-897D-12R-1, 44-46 cm, at 695.37 and 703.94 mbsf, respectively (Table 2).
2. There is a possible narrow normal polarity interval at 747.2 mbsf in Hole 897D in which no obvious physical disturbance is present. It appears that this polarity transition is represented simultaneously in both inclination and declination, although the declination shows some additional shifts owing to the lack of internal orientation of these cores upon coring. It is also interesting to note that below this transition cores have the weakest NRM intensities within the basement section (see Table 2). These features, if true, would increase our confidence in the polarity determinations for both the upper and lower parts of the peridotite sections.