units, we added a figure to demonstrate the relation between both scales (Fig. F5A) and the effects on the appearance of frequency distribution plots given in millimeter and values (Fig. F5B, F5C).
Sediment samples were taken from Site 1095 cores using a 10- to 20-cm3 plugging device. After freeze-drying, the sample was split by wet sieving into the fine (<63 µm) and coarse (>63 µm) fractions. The coarse fraction was oven dried, balanced, and set aside for further analysis (the results of the coarse-fraction analysis are given in the data report by Wolf-Welling et al., Chap. 21, this volume).
The water content of the suspended fine-fraction sample was reduced in two steps, by settling and the subsequent removal of excess water (by vacuum pump). We added H2O2 to remove remaining organic complexes (Anderson, 1963; Matthews, 1991) and rinsed the fraction with distilled water before a final removal of excess water. Na2PO4 was used as a final dispersant to avoid fine-particle clogging (McCave, 1995a; McCave et al., 1986). The final preparation step included 24-hr shaking in a overturn shaker and 1 hr in an ultrasonic bath directly before sample analysis (Fig. F3).
The fine fraction was subsequently analyzed by laser diffraction. Measurements were carried out using the laser particle sizer "Analysette 22"-Economy (Fritsch GmbH Laborgerätebau, 1994) (Fig. F4A). The unit is equipped with a 632.8-nm wavelength and a 3-mW helium-neon laser and handles suspensions with a particle size between 0.1 and 600 µm. The laser analyzer consists of four components: a dispersion unit, the measuring cell and laser, a multielement detector, and a personal computer with software for recalculation and data display. The sample is fed into the dispersing and homogenization unit, which is equipped with an ultrasonic bath and a stirring compound. An interactive sample input, controlled by frequent absorption measurements, assures a certain grain-density range within the measuring cell. A centrifugal pump supports the transport of the sample to and from the measuring cell. After each measurement, the laser analyzer was set to rinse and clean the cell four times with distilled water.
The laser measurement itself is based on the sensory interpretation of Frauenhofer interference images produced by the scattering of monochromatic light on spherical grain boundaries (Von Bernuth, 1988). The size of the grains is related to wavelength and frequency (f), a geometry factor (k), and the radius of the first-order Frauenhofer interference ring (Ro) (Fritsch GmbH Laborgerätebau, 1994):
Particle sizes down to 0.02 µm can still be detected with sufficient accuracy using the Frauenhofer theorem (Von Bernuth, 1988). A special feature of this laser analyzing system is a fixed-focus optical setup with no focusing lenses behind the sampling cell, in contrast to other systems that use focusing lenses behind the sampling cell. In order to detect a wide range of particle sizes, the distance from sampling cell to the multimeter detector is variable (Fig. F4B, F4C). Shifting the sampling cell relative to the detector changes the diffraction angle and the size of the affected sensor area. A large distance between the sampling cell and detector therefore permits the measurement of larger particles with a smaller diffraction angle, and short distances are used for small particles with a wide diffraction angle (Fig. F4B, F4C) (McCave et al., 1986). Our fine-fraction data set consists of 31 size classes ranging from 0.425 to 63.314 µm. The size classes are given in Table T1. The size range of 0.4-63.3 µm was measured with a constant cell-to-detector spacing (single-range measurement). Since some of the plots displayed in this report are given in linear scale and others (e.g., the statistical moment plots) in units, we added a figure to demonstrate the relation between both scales (Fig. F5A) and the effects on the appearance of frequency distribution plots given in millimeter and values (Fig. F5B, F5C).
The age model and stratigraphy for Site 1095 is based on the magnetostratigraphic results published in "Magnetostratigraphy" in Shipboard Scientific Party (1999b).
We assigned ages for each sample assuming linear sedimentation rates between age dates. Even though depth below seafloor and recovery-corrected depth values do not differ significantly, we decided to work with meters composite depth (mcd). Because of the small overlap between Holes 1095A, 1095B, and 1095D, we do not expect important changes for our model compared to stratigraphically spliced depth scales. We excluded two samples in the overlapping part of Holes 1095A and 1095B (between 83.0 and 87.3 meters below seafloor) to avoid assigning overlapping ages to samples.
The hiatus at ~60 mcd discussed by Hillenbrand and Ehrmann (Chap. 8, this volume) and in "Magnetostratigraphy" and "Seismic Stratigraphy" in Shipboard Scientific Party (1999b) has not been considered in our age model. Even so, the data presented supports the ideas presented in Shipboard Scientific Party (1999b) and Hillenbrand and Fütterer (Chap. 23, this volume).
