ALBERT

Description: ODP Legs 106 and 109 drilled zero-age crust at Site 648 on the axis of the Mid-Atlantic Ridge and recovered sparsely to moderately phyric but chemically homogeneous basalt. Plagioclase phenocrysts in Site 648 basalts exhibit a variety of zoning patterns and compositions including: (1) tabular (An84) cores mantled by more sodic zones (An80 and An75), (2) tabular (An80) cores mantled by narrow (An84 and An75) zones, (3) skeletal cores (An80-78) enclosing sodic (An75) patches, and (4) rounded cores (An75) surrounded by skeletal (An80-78) mantles. Backscattered electron images have been used to define zone morphologies in 28 crystals; major element compositions of the zones have been determined by electron microprobe; and Mg, K, Ti, Fe, and Sr concentrations have been determined by ion microprobe. Equilibrium partitioning of minor and trace elements between crystals and liquids occurred during stable growth of tabular cores and inner rims. Disequilibrium partitioning occurred during rapid growth of crystal rims, resulting in enrichment of incompatible trace and minor elements, and during growth of calcic zones with skeletal and acicular morphology, resulting in enrichment of Fe and Sr and depletion of Mg, K, and Ti. Crystal/liquid partition coefficients estimated from the composition of inner rims and glass are DMg = 0.03, DK = 0.15, DTi = 0.028, DFe = 0.05, and DSr = 1.82. Using these partition coefficients, we have calculated the composition of primitive liquid in equilibrium with phenocryst cores. The estimated liquid composition is similar to primitive basalt compositions from the Kane Fracture Zone except for a much higher Ca/(Ca + Na) ratio which resembles values for MAR plume basalts. The diversity in plagioclase morphologies and compositions is explained by mixing of melts at different stages of evolution and stored in separated but interconnected reservoirs, as previously suggested by Kuo and Kirkpatrick (1982, doi:10.1007/BF00376957).

Description: SeaBeam echo sounding, seismic reflection, magnetics, and gravity profiles were run along closely spaced tracks (5 km) parallel to the Atlantis II Fracture Zone on the Southwest Indian Ridge, giving 80% bathymetric coverage of a 30- * 170-nmi strip centered over the fracture zone. The southern and northern rift valleys of the ridge were clearly defined and offset north-south by 199 km. The rift valleys are typical of those found elsewhere on the Southwest Indian Ridge, with relief of more than 2200 m and widths from 22 to 38 km. The ridge-transform intersections are marked by deep nodal basins lying on the transform side of the neovolcanic zone that defines the present-day spreading axis. The walls of the transform generally are steep (25°-40°), although locally, they can be more subdued. The deepest point in the transform is 6480 m in the southern nodal basin, and the shallowest is an uplifted wave-cut terrace that exposes plutonic rocks from the deepest layer of the ocean crust at 700 m. The transform valley is bisected by a 1.5-km-high median tectonic ridge that extends from the northern ridge-transform intersection to the midpoint of the active transform. The seismic survey showed that the floor of the transform contains up to 0.5 km of sediment. Piston-coring at two locations on the transform floor recovered more than 1 m of sand and gravel, which appears to be turbidites shed from the walls of the fracture zone. Extensive dredging showed that more than two-thirds of the crust exposed in the transform valley and its walls were plutonic rocks, principally gabbros and residual mantle peridotites. In contrast, based on dredging and seafloor morphology, only relatively undisrupted pillow basalt flows have been exposed on crust of the same age spreading away from the transform. Magnetic anomalies are well defined out to 11 m.y. over the flanking transverse ridges and transform valley, even where layer 2 appears to be absent. The total opening rate is 1.6 cm/yr, but the arrangement of the anomalies indicates that the spreading for each ridge is asymmetric, with the ridge flanks facing the transform spreading at a rate of 1.0 cm/yr. Such an asymmetric spreading pattern requires that both the northern and southern ridges migrate away from each other at 0.2 cm/yr, thus lengthening the transform at 0.4 cm/yr for the last 11 m.y. To the north, the fracture zone valley is oriented differently from the present-day transform, indicating a paleospreading direction change at 17 m.y. from N10°E to due north-south. This change placed the transform into extension for the 11-m.y. period required for simple orthogonal ridge-transform geometry to be reestablished and produced a large transtensional basin within the transform valley. This basin was split by continued transform slip after 11 m.y., with the larger half moving to the north with the African Plate.

Description: We redescribed the ~0.5-km gabbro section drilled in Hole 735B at the Ocean Drilling Program Gulf Coast Repository. Included in this work was a redivision and clarification of the location and nature of the major lithologic boundaries and a division of the major units into subunits. In all, we found 495 distinct lithologic intervals in the core. Most of the section consists of a single olivine gabbro body having only minor cryptic variations, which we think represents a small intrusion. At the top of the section, the olivine gabbro is intercalated with a medium- to coarse-grained gabbronorite, which we postulate was intruded by the olivine gabbro. The base of the olivine gabbro has been intruded by troctolites and troctolitic gabbros, which may be the precursors of a major troctolite intrusive body immediately below the base of the hole. This section is variously crosscut by small microgabbro bodies, which are the products of crystallization and wall-rock reaction of small magma bodies that migrated through the olivine gabbro prior to complete solidification. Overall, the plutonic section drilled in Hole 735B is unlike those found at layered intrusions as it lacks evidence for extensive magmatic sedimentation. Rather, it appears to represent a plutonic basement composed of small, relatively short-lived, rapidly crystallized intrusions. This is consistent with the ephemeral volcanism and low rates of magma supply postulated for very slow-spreading ocean ridges. This whole section underwent "syntectonic differentiation": a process in which deformation and compaction of a rigid, partially molten gabbro drove intercumulus melt out of the olivine gabbro into ductile shear zones. Chemical exchange, precipitation of oxides, and trapping of the migrating melt at the end of deformation altered the gabbro in the shear zones to ferrogabbro. These oxide-rich horizons have the potential to be major shallow-dipping seismic reflectors. The largest such zone is 103 m thick and consists of foliated disseminated oxide olivine and oxide olivine gabbros of lithologic Units III and IV. The last igneous event was back-intrusion of trondhjemite veins that formed either by fractional crystallization from the interstitial melt and/or by wall rock anatexis of intruded amphibolites. Alteration and relatively rapid cooling of the gabbro body occurred by penetration and circulation of seawater into the plutonic section caused by thermal contraction and cracking under tensile stress, much as envisaged by Lister (1970). Initially, this circulation was greatly enhanced tectonically by the tensile component provided by lithospheric necking and the formation of brittle-ductile faults beneath the median valley. This circulation was sufficiently pervasive to alter about 25% of all the matrix pyroxene in the body, mostly to amphibole, in the amphibolite facies. Alteration was heaviest in the vicinity of the brittle-ductile faults, where formation of crack networks, cataclasis, and granulation were ongoing processes continuously creating porosity and permeability during deformation. At the end of the brittle-ductile deformation phase, the brittle-ductile fault zones became the most impermeable horizons in the core and suffered little additional alteration. This was due to the extensive syntectonic recrystallization of the matrix mineralogy, which effectively reset the stored elastic thermal strain to zero. In the relatively undeformed horizons, where the stored elastic thermal strain remained substantial, cracking and alteration continued under static conditions as the gabbro cooled, though at lower rates of seawater circulation, following a similar pattern to layered intrusions such as the Skaergaard Complex (e.g., Bird, 1986). Alteration of the massif nearly stopped within the middle amphibolite facies with the cessation of brittle-ductile deformation. Significant lower amphibolite facies diopside-bearing vein networks occur only within the undeformed olivine gabbros in Unit V. Only minor amounts of greenschist and zeolite facies mineralization are found, primarily overprinting early higher-temperature vein and crack networks in the undeformed gabbros. The sharp decrease in alteration below middle amphibolite facies is thought to result from reduced circulation of seawater that accompanied a sharp drop in the available tensile stress for cracking. This probably reflected the transfer of the gabbro body out of the zone of brittle-ductile deformation and lithospheric necking by the formation of a new set of master faults in the median valley closer to the axis of volcanism. Following this, alteration continued under static conditions and accompanying lower rates of seawater circulation with initiation of block uplift of the gabbro massif into the transverse ridge of the Atlantis II Fracture Zone. The last alteration/tectonic event evident within the core is a set of vertically oriented, irregular cracks, frequently covered with smectite. These cracks probably formed during unloading of the gabbros by erosion to sea level after its initial uplift to form an island. They are largely absent from the brittle-ductile deformation zones, indicating that insufficient stored thermal strain was available there (even after cooling from near 500°C to ambient temperature) to overcome the internal strength of the rock under lithostatic load.