(Supplement Table1) Abundances of major elements from KH93-3_DR10 at the Southwest Indian Ridge near the Rodriguez Triple Junction

Petrographic and geochemical analyses of basaltic rocks dredged from the first segment of the Southwest Indian Ridge near the Rodriguez Triple Junction have been completed in order to investigate water-rock interaction processes during mid-ocean ridge (MOR) hydrothermal alteration in the Indian Ocean. In the study area, we have successfully recovered a serial section of upper oceanic crust exposed along a steep rift valley wall which was uplifted and emplaced along a low angle normal fault. On the basis of microscopic observation, dredged samples are classified into three types: fresh lavas, low-temperature altered rocks, and high-temperature altered rocks. The fresh lavas have essentially the same chemical composition as typical N-MORB, although LILE and Nb are slightly enriched and depleted, respectively. Low temperature alteration brought about the enrichment of K2O, Rb, and U due to the presence of K-rich celadonite and U-adsorption onto Fe-oxyhydroxide and clay minerals. On the other hand, chloritization, albitization, and addition of base metals by high temperature hydrothermal alteration result in enrichments of MnO, MgO, Na2O, Cu, and Zn and depletions of CaO, K2O, Cr, Co, Ni, Rb, Sr, and Ba. In addition, U-enrichment is also observable in the high temperature altered rocks probably due to the decrease of uranite solubility in the reducing high-temperature hydrothermal solution. These petrological and geochemical features are comparable to those of the volcanic zone to transition zone rocks in the DSDP/ODP Hole 504B, indicating that our samples were recovered from the upper ~1000 m section of the oceanic crust. Only the alteration minerals related to off-axis alteration are absent in our samples dredged from near the spreading axis. The similarity of alteration between our samples from the Indian Ocean and the Hole 504B rocks from the Pacific Ocean suggests that MOR hydrothermal systems are probably similar across all world oceans.

Physical properties and neutron texture measuerments of deep-sea carbonates of DSDP Holes 62-463 and 62-465A

In weakly indurated, nannofossil-rich, deep-sea carbonates compressional wave velocity is up to twice as fast parallel to bedding than normal to it. It has been suggested that this anisotropy is due to alignment of calcite c-axes perpendicular to the shields of coccoliths and shield deposition parallel to bedding. This hypothesis was tested by measuring the preferred orientation (fabric) of calcite c-axes in acoustic anisotropic, calcareous DSDP sediment samples by X-ray goniometry, and it was found that the maximum c-axis concentrations are by far too low to explain the anisotropies. The X-ray method is subject to a number of uncertainties due to preparatory and technical shortcomings in weakly indurated rocks. The most serious weaknesses are: sample preparation, volume of measured sample (fraction of a mm3), beam defocusing and background intensity corrections, combination of incomplete pole figures, and necessity of recalculation of the c-axis orientations from other crystallographic directions. Goniometry using thermal neutrons overcomes most of these difficulties, but it is time consuming. We test the interferences made about velocity anisotropy by X-ray studies about the concentration of c-axes in deep-sea carbonates by employing neutron texture goniometry to eight DSDP samples comprising mostly nannofossil material. Fabric and sonic velocity were determined directly on the core specimens, thus from the same rock volume and requiring no preparation. The c-axis orientation is obtained directly from the [0006] calcite diffraction peak without corrections. The fabrics are clearly defined, but weak (1.1 to 1.86 times uniform) with the maximum about normal to bedding. They have crudely orthorhombic symmetry, but are not axisymmetric around the bedding normal. The observed c-axis intensities, although higher than determined by the X-ray method on other samples, are by far too low to explain the observed acoustic anisotropies.