Although samples from Leg 168 represent an incomplete record (partly because of the very low recovery rates at some sites; Davis, Fisher, Firth, et al., 1997), many indexes suggest that the alteration intensity of pillow basalts from Leg 168 systematically increases with the distance from the present-day ridge. This indicates that the aging of the igneous crust (i.e., the time available for secondary mineral growth) and the measured increase of temperatures at the sediment/basement interface, from the younger HT to the older RB Sites (from 15° to 64°C; Davis, Fisher, Firth, et al., 1997), play a very important role in the overall alteration processes. The observation of increased alteration with crustal age has also been reported for a 1.6-Ma transect through pillow basalts of the Blanco Fracture Zone located at a different latitude on the eastern flank of the JdFR (Juteau et al., 1995).
Other important factors controlling the alteration intensity are the degree of fracturing of the pillow units and subordinately the local and regional variations in lithology and primary porosity (e.g., vesicularity, degree of crystallinity). Particularly, in both younger and older pillow units, the style of the alteration processes is mainly controlled by the presence of concentric and radial fractures, which crosscut all pillows from the outer glassy zone to the holocrystalline interior. Fractures are the most important pathways that allow significant penetration of fluids into the rock, controlling the alteration of the glassy rim and the early oxidizing stage of pillow alteration. In fact, most of the glassy margins of the pillow basalts recovered during ODP Leg 168 are unaltered or weakly altered except along and around concentric and radial cracks or veins.
The alteration of glass occurs in several different steps: (1) penetration of the fluids in the glass along the newly formed cracks allows the formation of a first generation of palagonite; (2) palagonite then evolves from amorphous brown patches to brownish yellow incipiently crystalline (sometimes fibrous) symmetric areas; and (3) palagonite is progressively replaced by dioctahedral clays (celadonite), Fe-oxyhydroxides, and trioctahedral smectites. More evolved glass alteration occurs in the oldest Sites 1026 and 1027, where a new veining stage is superimposed on the primary veins through repeated opening and crack filling processes along the median line of the original crack. Also in these cases the infilling secondary minerals are commonly symmetrically arranged and occur in the following sequence: (1) fibrous trioctahedral smectites, (2) fibrous radiating zeolites, and (3) fibrous to blocky anhedral carbonates. Most of the elements involved in the glass breakdown are used in, or immediately around, the vein to form secondary authigenic minerals. The strong enrichment in K2O from the circulating seawater allows the growth of dioctahedral clays (celadonite type) in the first step of alteration, whereas smectites, zeolites, and carbonates act as a sink for a significant portion of the removed Mg, Ca, Na, Al, and Si. Part of the leached iron is immediately used to form thin layers of Fe-oxyhydroxides often associated with K- and Fe-bearing clay, whereas the remaining Fe and most of the Mn are removed from the glassy zone by the circulating fluids.
The hypocrystalline and holocrystalline portion of pillow basalts are characterized by two different styles of alteration that occur under oxidizing or reducing conditions. These alteration styles agree well with many studies performed on the alteration of other young pillow basalt sequences (Andrews, 1980; Böhlke et al., 1980, 1981; Alt, Kinoshita, Stokking, et al., 1993; Laverne et al., 1996). The most common mineral paragenesis characterizing the two style of alteration are well defined: (1) assemblages under oxidizing conditions: Fe-oxyhydroxides ± iddingsite ± celadonite ± Mg-rich saponite; and (2) assemblages under reducing conditions: Fe-rich saponite ± carbonates ± sulfides ± talc.
All the observed mineralogical and chemical variations occurring during the early stage of the alteration are interpreted as the result of rock interaction with "normal," alkaline, and oxidizing seawater along preferential pathways represented by the concentric and radial crack systems. The chemical composition of the fluid progressively evolved while moving into the basalt leading to a reducing alteration stage, which is responsible for the precipitation of Fe-rich saponite and minor sulfides. The widespread aragonite/calcite formation represents the last stage of alteration that occurs as a consequence of the progressive saturation of the fluid with respect to carbonates. This stage probably occurred when the eastern flank of JdFR was completely buried by turbiditic sediments that hydrologically sealed the igneous crust and inhibited open seawater circulation within the igneous basement.