In the past, the nature of the oceanic basement has been deduced mainly from studies of ophiolite complexes (Gass, 1968; Church and Stevens, 1971; Moores and Vine, 1971; Christensen and Salisbury, 1975; Karson et al., 1984), remote geophysical surveys, and studies made from dredged or cored samples from the ocean floor. During the last three decades, our knowledge of the oceanic seismic structure has improved dramatically from the simple three-layer model of Raitt (1956, 1963) that was based on first arrival information. Advances in shooting, recording, and wireline logging technology have provided the means to recognize subdivisions within oceanic Layers 2 and 3, which were first interpreted as horizontal layers of constant velocities (Hussong et al., 1969; Shor et al., 1969; Sutton et al., 1971; Fox et al., 1973; Lister and Lewis, 1974; Peterson et al., 1974; Houtz and Ewing, 1976). Furthermore, results from synthetic seismograms and inversion techniques suggested that the velocity structure of the oceanic crust may be more accurately described in terms of velocity gradients instead of constant velocity layers (Helmberger and Morris, 1969; Kennett and Orcutt, 1976; Orcutt et al., 1977; Spudich and Orcutt, 1980).
Largely because of these studies, the igneous oceanic crust is typically thought of as comprising an upper crust (Layer 2) characterized by a rapid increase in seismic velocity with depth and a thicker lower crust (Layer 3) that is distinguished from Layer 2 by both a higher compressional wave velocity and a much smaller vertical velocity gradient. However, a direct correlation between the seismic layering and the in situ lithologic and physical properties of the oceanic crust has never been established. The transition between seismic Layers 2 and 3 has been interpreted as a change in igneous rock textures from doleritic (diabase) sheeted dikes to gabbro, an increase in metamorphic grade from greenschist to amphibolite facies metamorphism (Christensen, 1970; Fox et al., 1973; Christensen and Salisbury, 1975; Salisbury and Christensen, 1978; Spudich and Orcutt, 1980), or a change in crustal bulk porosity with depth (Lort and Matthews, 1972; Spudich and Orcutt, 1980; Becker, 1985). Recent interpretations of seismic refraction studies on the southern flank of the Costa Rica Rift have proposed that the seismic Layer 2/3 boundary is located within the sheeted dike complex (Detrick et al., 1994). This interpretation correlates with gradual downward changes in crustal porosity and alteration, with no apparent association with the lithologic transition from the sheeted dikes to gabbro.
Seismic reflection studies (Mutter et al., 1985; McCarthy et al., 1988; Géli and Renard, 1994) have shown that the structure of the North Atlantic oceanic crust is more complex than the previously mentioned layered model. Studies from ocean basins reveal a close interplay between brittle faulting in the upper crust and magmatism and intrusion at depth. These reflection profiles show basement-cutting faults that in some instances can be traced downdip into packages of laminated high-amplitude reflections confined within 1 s of the Moho reflection. In some cases, these deeper reflection packages dip as much as 20° to 40° and thicken downward before terminating abruptly at the subhorizontal Moho reflection, whereas in other instances they penetrate through what appears to be the upper mantle. It has been suggested (Mutter et al., 1985; McCarthy et al., 1988) that although the lower crustal events can be interpreted as downdip equivalents of basement-cutting faults, their high amplitudes and laminated character may be a result of constructive interference through a package of interlayered mafic-ultramafic cumulates crystallized along the walls of the magma chamber at the time of crustal accretion. Furthermore, the apparent continuation of upper crustal faults into these layered sets of mafic-ultramafic material suggests a spatial correlation between faulting and magmatism.
In the proximity of Hole 735B, seismic refraction measurements (Muller et al., 1997) found an overall crustal thickness north and south of Atlantis Bank of 4 km and Layer 2 having a normal thickness of ~2 km (Fig. F3). However, a projection from a seismic line 1 km to the west of the Atlantis Bank shows that beneath this area, the depth to the Moho is 5 ± 1 km below seafloor (Fig. F3). In contrast, dredged samples (Dick et al., 1991) and inversion of rare element concentrations in basalts dredged from the conjugate site to the north of the Atlantis Bank (Muller et al., 1997) suggest a crustal thickness of 3 ± 1 km.
The results from Hole 735B provide the opportunity to test these hypotheses and models using downhole and physical properties measurements. Although several attempts to drill and log deep into the oceanic crust have been previously made (Kirkpatrick, 1979; Salisbury et al., 1980; Salisbury, 1983; Cann and Von Herzen, 1983; Salisbury et al., 1985), this potential was more fully realized in results from Hole 735B, DSDP Hole 504B in the Costa Rica Rift, and the findings from the Hess Deep area during Leg 147.