Figure 2. A. Hand-contoured 100-m contour bathymetric map of Site 735 showing the location of Hole 735B (modified from Dick et al., 1991a). SeaBeam tracks hand-shifted by eye to eliminate conflicts in the data. Solid lines indicate actual data, while hatched lines show inferred contours. Contour interval is 250 m. Small solid dots and arrows indicate the starting point and approximate track of dredge hauls. Large solid dots show the location of Sites 735 and 732 (just north of the contoured area on the crest of the median tectonic ridge). Filled circles indicate the approximate proportions of rock types recovered in each dredge: + = gabbro, white = basalt and diabase, light stipples = greenstone, and heavy stipples = serpentinized peridotite. B. Hand contoured bathymetric map of the eastern rift mountains north of the SW Indian Ridge axis showing crust of the same age as Site 735 and the conjugate position of Hole 735B (735B') on the counter-lithospheric flow line, based on magnetic anomalies and plate reconstruction. This conjugate site is the location of SWIR 6, the final backup site for Leg 176, where the volcanic carapace originally overlying Hole 735B is preserved intact.
Figure 3. Bathymetric map of the Atlantis II Fracture Zone modified from Dick et al. (1991b). Locations of Hole 735B and the conjugate Site 735B' (SWIR 6) shown. SWIR 6 is located on the counter lithospheric flow line on crust of the same age and position relative to the paleotransform as Hole 735B. Active southern and northern rift valleys are at 33°40'S and 31°50'S, respectively.
Figure 4. Magnetic anomalies over Site 735 based on the survey of Dick et al. (1991a). Bathymetry contoured at 200 m intervals. Crustal magnetization is shown shaded, with normal polarity crustal magnetization shown as gray and reverse polarity shown as white. Dark gray areas have crustal magnetization greater than 1 A/m. Polarity identifications and numbering modified from Dick et al. (1991a) by M. Tivey (pers. comm., 1997) based on the time scale of Cande and Kent, 1992.
Figure 5. Temporal cross sections across the Southwest Indian Ridge rift valley drawn parallel to the spreading direction (not across the fracture zone, but parallel to it), showing the postulated tectonic evolution of the transverse ridge and Hole 735B (Dick et al., 1991a). The sequential sections are drawn at about 18 km from the transform fault. Crust spreading to the right passes into the transverse ridge and spreads parallel to the transform valley. Crust spreading to the left spreads into the rift mountains of the Southwest Indian Ridge parallel to the inactive extension of the Atlantis II Fracture Zone. Dense stipple = mantle, filled diamonds = gabbro, inverted "v" = basalt. A. Initial symmetric spreading, possibly at the end of a magmatic pulse. Late magmatic brittle-ductile deformation occurs because of lithospheric necking above (and in the vicinity of whatever passes for a magma chamber at these spreading rates). Hydrothermal alteration at high temperatures accompanies necking and ductile flow in subsolidus regions. B. At some point, the shallow crust is welded to the old, cold lithosphere to which the ridge axis abuts, causing formation of a detachment fault and nodal basin, initiation of low-angle faulting, continued brittle-ductile faulting, and amphibolite-facies alteration of rocks drilled at Hole 735B. C. and D. Block uplift of the rift mountains at the ridge/transform corner forms a transverse ridge enhanced by regional isostatic compensation of the local negative mass anomaly at the nodal basin. Initiation of the block uplift terminates the extension driving cracking, and drastically reduces permeability in the Hole 735B rocks, effectively terminating most circulation of seawater and alteration. Greenschist-facies retrograde alteration continues along the faults on which the block is uplifted to account for the greenschist-facies alteration that predominates in dredged gabbros.
Figure 6. Possible north-south geologic cross sections of the Atlantis II Bank through Hole 735B consistent with existing geological and geophysical data, gravity, and magnetics. Except at Hole 735B, fault locations and geometries are uncertain. The presence of a dike-gabbro transition as shown on the right of all the models is also hypothetical. A. Crustal stratigraphy around Hole 735B as suggested by Cannat (1993) and Swift and Stephen (1992), consistent with earlier inferences of Hess (1962). B. Crustal stratigraphy assuming Hole 735B is representative of the lower crust down to the mantle (Dick et al., 1991a). C. Crustal stratigraphy based on the layered intrusion model for ophiolites as proposed for Oman (after Pallister and Hopson, 1981; Smewing, 1981).
Figure 7. Seismic velocity structure from Muller et al. (1997). A. P-wave seismic velocity model on the north-south seismic line CAM101 of Muller et al. (1997). The velocity contour interval is 0.3 km/s. OBS positions are shown on the seafloor. The position of ODP Hole 735B has been projected from 1 km west of the line. The MOHO is indicated as a thicker line where its depth is constrained by wide-angle reflections. B. Resolution contours for the seismic model with Layers 2 and 3 and Upper/Lower Layer 3 boundaries from (A) superimposed. Numbered OBS positions are shown on the seafloor. The resolution of each velocity node is given by the diagonal of the inversion resolution matrix, a number between 0.0 and 1.0, affected by the ray coverage sampling each node. Values >0.5 (stippled area) are considered well resolved and reliable.