Unbroken nannofossils are preserved throughout the sampled section (Fig. F1). Although the number of spine-shaped fragments of nannofossil is small (less than 5% at most), it tends to increase with depth. At 19.45 mbsf (Sample 174B-1074A-3H-1, 45-47 cm) (Fig. F1A), the relative abundance is trace (0%-0.1%), whereas it is present (>1%-5%) at 60.05 mbsf (Sample 174B-1074A-7H-3, 5-7 cm) (Fig. F1D).
The arrangement of nannofossils plays an important role in creating a void space in the unlithified sediments. Coccoliths are linked edge to edge and edge to face in all samples. In the case of clay minerals, such as kaolinite, such a fabric results in an open framework with very high porosity (Bennett and Hulbert, 1986). Even at 60.80 mbsf (Sample 174B-1074A-7H-3, 5.0-7.0 cm), both the edge-to-face and edge-to-edge linked coccoliths were preserved (Fig. F1D). This indicates that the void space is still maintained until this depth because the deformation of the sediments is weak.
Shape and volume of the void space are controlled by type of aggregation pattern and material features. In a nannofossil dominated sample, the edge-to-face or edge-to-edge linked coccoliths make a stairstep structure (O'Brien, 1971). Such void space is generally of random shape, with sizes ranging from a few to several tens of micrometers. On the other hand, a sample that contains well-preserved planktonic foraminifers (e.g., 174B-1074A-5H-4, 40-42 cm; 38.40 mbsf) (Fig. F1C) has a large void space (~100 µm in diameter) and seems to be less consolidated in that the foraminifers are few or absent. In addition, some stairstep structures of nannofossil aggregation develop in a larger void space and are surrounded by foraminifers. This structure might play the role of beam, supporting the wall of the larger void space from inside.
The SEM observation shows that nannofossils and planktonic foraminifers in the sediments are well preserved and that void space in the unlithified sediments is formed by an assemblage of microfossils consisting of a stairstep structure of nannofossils and aggregation of foraminifers.
Using AMS measurements, we could not detect any obvious deformation of the unlithified sediments. The vertical change of magnetic susceptibility measured on shore using discrete samples is consistent with that obtained by the MST measurements during the cruise. The magnetic susceptibility, which varies from ~30 × 10-6 SI to ~880 × 10-6 SI, tends to increase with depth and is sensitive to lithologic variations (Fig. F2). Particularly, the magnetic susceptibility is high in Subunit IB, which contains many small (<1 cm) basalt clasts.
The L and F values, which represent lineation and foliation of magnetic fabrics, are variable but tend to be weaker with depth. Both the L and F values decrease from 1.02 near the seafloor to <1.005 at the base of Unit I (63.5 mbsf). These results indicate that anisotropy of magnetic fabrics tends to be weaker with depth regardless of initial configuration.
The Flinn diagram of L vs. F shows that most plots are scattered around the left bottom of the diagram, ranging from ~1 to 1.025, along the slope of unit gradient (Fig. F3). This also indicates that the anisotropy is weak or almost neutral.