Interstitial water was squeezed from 645 whole-round cores immediately after core retrieval using hydraulic pressure and standard ODP titanium-stainless steel squeezers (Manheim and Sayles, 1974). Pore waters were filtered (0.45 µm) and subsequently analyzed for concentrations of major and minor dissolved components, including sulfate, chloride, sulfide, alkalinity, and nutrients (ammonium, phosphate, and nitrate) on board the JOIDES Resolution (D'Hondt, Jørgensen, Miller, et al., 2003) using standard ODP methods (Gieskes et al., 1991). From 112 of the 645 samples, ~10 mL of filtered pore water was immediately fixed with zinc acetate to precipitate zinc sulfide and to prevent artifactual oxidation of sulfide to sulfate. Following gravity settling of zinc sulfide over several days, ~5 mL of the supernatant containing dissolved sulfate was transferred to high-density polyethylene containers and stored at 4°C for subsequent shore-based analyses of oxygen (this study) and sulfur (Böttcher et al., this volume) isotopic composition of dissolved sulfate.
Dissolved sulfate was precipitated quantitatively as barium sulfate according to standard gravimetric procedures (Clesceri et al., 1989). Briefly, 10% barium chloride solution was added to diluted pore water samples that were acidified to ~pH 4 to preclude formation of barium carbonate. Precipitated barium sulfate was filtered (0.2 µm), washed, and dried. Dried barium sulfate samples were thoroughly homogenized and ~5–10 mg of sample was loaded into a quartz vessel and baked at 650°–700°C for 1 hr to remove residual water and organic matter from the barium sulfate crystals. Next, 150 µg of sample was loaded into a 3.5 mm x 5 mm pressed silver foil capsule and placed in a Costech Zero-Blank autosampler for oxygen isotope analysis using a Finnigan DeltaplusXP with TC/EA module (High Temperature Conversion) and Conflo III interface operating in continuous-flow mode (Earth System Center for Stable Isotope Studies of the Yale Institute for Biospheric Studies, USA). Barium sulfate samples are reacted at 1450°C in a graphite reactor crucible to release oxygen, which in turn reacts with the graphite to produce carbon monoxide (CO) (Kornexl et al., 1999). The CO is entrained in He carrier gas, passed through a gas chromatograph, and introduced into the mass spectrometer. Oxygen isotope ratios were determined by integrating the areas under CO peaks of masses m/z 28, 29, and 30. All oxygen isotope data are reported as 
-values in permil and referenced to the V-SMOW standard. Two internal laboratory BaSO4 standards with 18O values of +0.1
and +16.9 and the International Atomic Energy Agency-National Bureau of Standards 127 BaSO4 standard (18O = +9.3) were used for data calibration and correction. The precision of the method based on replicate measurement of standards is ±0.5.
