TECHNIQUES

Interstitial water samples were obtained using conventional methods (Manheim and Sayles, 1974). Chloride concentrations were measured on the ship (Eberli, Swart, Malone, et al., 1997) using methods described by Gieskes (1973, 1974). Oxygen and hydrogen isotopic measurements of the pore waters were made in the Division of Marine Geology and Geophysics at the University of Miami. Hydrogen isotopic data were only analyzed on one hole as a result of operational problems. Oxygen isotopic data were made in duplicate on most samples from Sites 1003 through 1007. As the methods for the measurement of both isotopes are different from the conventional techniques (Epstein and Mayeda, 1953; Friedman, 1957), they will be elaborated here. Both measurements were made using a water equilibration system (WEST) attached to a Europa GEO. In the WEST, the oxygen isotopic composition is determined on CO₂ that has been injected into serum bottles at slightly above atmospheric pressure containing 1 cm³ of sample. The samples are subsequently equilibrated at 40°C for 4 hr without shaking. The process is entirely automated: the CO₂ is injected into the sample bottles and retrieved using an autosampler, and the gas is transferred to a dual-inlet mass spectrometer through a cryogenic trap (-70°C) to remove water. The precision of this method for oxygen, determined by measuring 59 samples of our internal standard, is ±0.07 (Fig. 4). The relatively large standard deviation is a result of a 0.3°C temperature variation in the equilibration bath. The hydrogen isotopic composition is determined using the same device as employed for CO₂. Equilibration with hydrogen gas takes place in the presence of a platinum catalyst (Hokko Beads) at 40°C (Coplen et al., 1991). As a result of the above-mentioned stability problems with the temperature bath, precision using this method is ±1.5 (Fig. 4). All data are calibrated using Vienna Standard Mean Ocean Water (SMOW) and are reported in parts per thousand () according to the conventional notation.