Gabbroic samples recovered during Leg 176 preserve a complex record of high-temperature metamorphism, brittle failure, and hydrothermal alteration that began at near-solidus temperatures and continued down to zeolite facies metamorphic conditions. Secondary mineral assemblages in the Leg 176 core are broadly similar to those higher up in the sequence (Dick et al., 1991b; Robinson et al., 1991; Stakes et al., 1991). Rocks deeper in the section, however, are remarkably fresh over large sections of the core, with extensive intervals (>300 m) marked by less than 10% total background alteration. This is in striking contrast to plutonic samples recovered by both drilling and submersible studies of other fracture-zone environments (Gallinati, 1984; Mével et al., 1991; Gillis et al., 1993b; Früh-Green et al., 1996; Kelley, 1996). In Hole 735B, there is a remarkable decrease in the abundance of amphibole veins downsection, and a corresponding decrease in high-temperature background static alteration, reflecting only very localized penetration of high-temperature hydrothermal fluids along the vein networks. Cessation of fluid flow in the plutonic section is marked by the development of late smectite, carbonate, and zeolite ± prehnite veins that are most likely associated with attenuation and uplift of the massif during formation of the transverse ridge. For details of these metamorphic effects, see "Macroscopic Description of Secondary Mineralogy" (i.e., background metamorphic variations at the hand specimen and larger scales); "Microscopic Description" (thin-section characterization); and "Magmatic and Hydrothermal Veins".
Rocks recovered during Leg 176 generally vary from fresh to 40% altered, although there are many small intervals of more altered rocks (40%-70% recrystallized), and locally rocks are 80%-90% recrystallized (Figs. F33, F34 ). Several general trends were observed: (1) The most intensely altered portion of the Leg 176 core occurs between 500 and 600 mbsf (Figs. F34, F35, F36; Cores 176-735B-89R to 103R), where the core is, on average, 10% to 40% altered. Calcite veins are common in this interval and are associated with low-temperature oxidation of the rocks. (2) A second zone of intense recrystallization occurs between 800 and 1030 mbsf (Cores 176-735B-130R to 156R), where many of the rocks exhibit high-temperature plastic deformation (see "Structural Geology") and veins are rare. (3) Two less-altered zones are located in an interval of abundant smectite veins at 700-800 mbsf (Cores 176-735B-119R to 130R), and in relatively uniform olivine gabbro with common smectite veins at 1300-1500 mbsf (Cores 176-735B-187R to 210R). At depths greater than 1030 m, the intensity of alteration is typically much less than 10%.
In many places, the intensity of alteration in the gabbroic rocks from Hole 735B is strongly related to brittle deformation and the distribution of veins (see "Magmatic and Hydrothermal Veins"). The upper limit of alteration grade is represented by recrystallized oxide gabbros, which commonly exhibit aggregates of equant grains with 120º grain-boundary triple junctions, and the development of neoblastic olivine and pyroxene that are typical of granulite facies metamorphic conditions (see "Igneous Petrology" and "Structural Geology"). The lower limit of alteration conditions is represented by zones of orange-red smectite and oxyhydroxide minerals that are related to the formation of late carbonate veins in a low-temperature, oxidative environment. The secondary minerals can be divided into three main groups (Fig. F33; Table T8): (1) a high-temperature assemblage that reflects formation under granulite to amphibolite facies conditions, (2) lower temperature mineral assemblages that are typical of formation under greenschist to zeolite facies metamorphic conditions, and (3) a very low-temperature mineral assemblage that is mainly represented by carbonate and clay minerals belonging to the smectite group.
Olivine is more easily altered than the other silicates in Hole 735B gabbros, and, as a result, the alteration profile for olivine roughly correlates with that of total rock (Fig. F34). Olivine, however, exhibits more extreme variations in alteration intensity over small intervals of the core than total rock alteration. Because plagioclase and clinopyroxene are, on average, the dominant primary minerals, variations in rock alteration correlate well with the distribution of secondary plagioclase and altered pyroxene as recorded in the hard rock visual core descriptions (VCDs). Comparison of the thin-section and VCD data confirm that it is difficult to distinguish fine-grained aggregates of pyroxene or olivine neoblasts that rim porphyroclasts in the gabbroic shear zones from high-temperature hydrothermal minerals (Figs. F37, F38, F39).
Secondary diopside is not abundant in the gabbroic rocks, but where it does occur, it is generally associated with veins (see "Magmatic and Hydrothermal Veins"). It was observed in thin sections between 500 and 1100 mbsf. Diopside commonly replaces augite and is associated with green and brown amphibole and, in some sections, with oxide minerals.
Dark green (Ca) amphibole is abundant (up to 20%-25%) and widespread above 700 mbsf (Cores 176-735B-89R to 118R); it is much less common and sparsely distributed below 800 mbsf (Fig. F33). Dark green amphibole forms in the halos of monomineralic amphibole veins and as reaction rims around olivine and pyroxene. In thin section, the amphibole forms coronas that are characterized by fine-grained laths elongated perpendicular to the bounding primary ferromagnesian minerals. Dark green amphibole is also developed in the fine-grained matrix of highly deformed rocks; however, in these zones it is difficult to distinguish amphibole in hand samples from pyroxene and olivine neoblasts.
Brown amphibole commonly occurs in small amounts (usually <1%) along the cleavage planes of clinopyroxene, as small blebs within pyroxene, and as rims around pyroxene and oxide crystals; less commonly it rims olivine. The modal abundance of brown amphibole generally increases near felsic veins and near some plagioclase + amphibole veins. In these zones it may completely replace clinopyroxene. Where the brown amphibole is igneous or hydrothermal is not yet clearly established (see "Sulfide and Oxide Minerals in Thin Section"). In deformed rocks, brown amphibole commonly rims clinopyroxene augen and is developed in recrystallized tails together with neoblastic pyroxene.
(Mg-Fe)-amphibole (e.g., cummingtonite and anthophyllite) was not observed in hand sample, but was observed in thin sections of the gabbroic rocks; modally it is a minor constituent, but it is common as an alteration phase after olivine and orthopyroxene.
Talc is ubiquitous in all gabbros containing olivine, which represent the most abundant lithology in the core (e.g., Fig. F40). The abundance of talc is typically low (<1%), but it increases in olivine-rich gabbros and is particularly high in the upper units (e.g., Sections 176-735B-102R-2 and 102R-3), where the average alteration is highest. The amount of talc is probably underestimated in the VCDs, because it occurs in close association with iron oxide minerals (magnetite) in the cracks of olivine and some pyroxenes (mainly orthopyroxene) and as rims on olivine and orthopyroxene, where it is commonly intergrown with amphibole and more rarely chlorite. The talc abundance decreases with depth (Fig. F36). However, in the lowermost units (Cores 176-735B-205R through 209R) olivine has coronas of talc, amphibole, and oxide minerals surrounding fresh cores. In some samples, these cores are replaced by chlorite-smectite.
Secondary plagioclase, either hydrothermal or as a recrystallized phase, is the most abundant metamorphic mineral (Fig. F33). The abundance of milky plagioclase is generally very low in undeformed rocks, and it is sparse away from felsic, amphibole + plagioclase, and amphibole veins; however, it is a common phase in vein halos. Relatively fresh and undeformed rocks, particularly from the lower part of the hole, generally do not exhibit secondary plagioclase in thin section. In these rocks, plagioclase is incipiently altered along cracks and grain boundaries to actinolitic amphibole. Not all the secondary plagioclase is hydrothermal in origin; extensive intervals show evidence of dynamic recrystallization of all the primary phases, and there is a low degree of alteration from 800 to 1100 mbsf.
Magnetite commonly is associated with alteration of ferromagnesian minerals to talc and with amphibole (Fig. F36).
Green amphibole after brown hornblende locally constitutes 5%-10% of both the gabbroic and felsic rocks (Figs. F40, F41). This amphibole is strongly pleochroic in thin section, and it varies from brownish green to greenish blue. It commonly forms monomineralic veinlets between primary crystals. Based on similar descriptions of amphibole in other locales, the green amphibole may contain a significant hornblende component and may reflect formation under amphibolite facies conditions (Robinson et al., 1991; Vanko and Stakes, 1991). These amphiboles are distinct from pale green acicular amphibole that forms halos associated with veins of amphibole and chlorite ± amphibole and patches in alteration areas. The pale green acicular amphibole is likely actinolite and is related to lower temperature conditions of alteration (greenschist or transitional actinolite facies).
Chlorite is rare and is largely restricted to that part of the core above 750 mbsf (Figs. F33, F36). It generally forms <1 modal% of the rock, but there is a slight increase in abundance (up to 5%) near 512 mbsf (Core 176-735B-90R), and locally at depths near 920, 1242, 1452, and 1493 mbsf. Chlorite is commonly associated with actinolite, chlorite-actinolite, and chlorite veins, and it is present in vein halos. It also forms in patches associated with some amphibole needles and with white secondary minerals that are difficult to identify clearly in hand specimen. In thin section, the white patches are composed of quartz, albite, and zeolite.
Quartz appears to be mostly magmatic in origin and occurs in segregation patches and late veins. However, small amounts of secondary quartz (Fig. F36) occur sparsely between 1000 and 1200 mbsf, where it is associated with other greenschist metamorphic minerals, replacing plagioclase.
Sodic feldspar is difficult to distinguish from more calcic secondary plagioclase and from magmatic plagioclase related to felsic veins. In some sections, milky feldspar is described as albite in the VCDs, and cloudy albite crystals are identified in thin sections of altered gabbroic rocks from the upper cores (Sections 176-735B-90R-3 and 90R-8). Based on its milky white color, some plagioclase in more felsic rocks is probably also replaced by albite. This is particularly evident in some felsic veins in which the cores of the veins appear to be highly altered to albite and chlorite. In some felsic veins, plagioclase is almost completely replaced by white clay.
In thin section, titanite occurs sporadically as an alteration product of oxide minerals and is locally related to chlorite patches. Some larger titanite crystals (<1 mm) were identified in hand specimen (Core 176-735B-202R). Rarely, titanite forms after clinopyroxene.
Prehnite and epidote are the only calc-silicates recognized as vein-forming minerals in hand specimen. Epidote was found in only two veins, but it occurs sporadically as a background alteration phase at various intervals of the core. Prehnite was identified in thin sections of the core recovered from the deepest part of the hole (about 1500 mbsf) and by shipboard X-ray diffraction (XRD) on hand-picked vein fragments from Core 176-735B-188R (Fig. F42). A single peak, possibly corresponding to pumpellyite, appears on the same XRD profile.
Some zeolites were identified in thin sections, but they are rare as background alteration products. The highest occurrence of zeolite in thin section is from Core 176-735B-181R, where it forms veins in association with chlorite-smectite and minor carbonate. Zeolites in Cores 176-735B-188R and 199R are provisionally identified as natrolite or possibly scolecite by shipboard XRD analyses (Fig. F43).
Most sulfide minerals form dispersed angular or globular aggregates or crystals in interstitial areas and are probably primary in origin. However, secondary pyrrhotite, pyrite, and, less commonly, chalcopyrite form in association with greenschist facies mineral assemblages (Fig. F36). Pyrrhotite occurs with magnetite, talc, and amphibole in alteration coronas around olivine. Other occurrences of secondary sulfide minerals are related to the formation of late smectite veins and are described in the following section on low-temperature mineralogy.
Smectite is the third most abundant alteration phase after dark green amphibole and secondary plagioclase (Fig. F33). Above 600 mbsf, orange to reddish patches of smectite and Fe-oxyhydroxide minerals partially or completely replace olivine and are related to the presence of carbonate and smectite veins (Cores 176-735B-89R to 94R, and 102R). A second type of smectite (dark green to pale bluish green) is spatially related to smectite veins and appears deeper in the core between 720 and 1500 mbsf. There is a gap in the distribution of green smectite veins between 800 and 1100 mbsf (see "Magmatic and Hydrothermal Veins"), corresponding to a zone of highly deformed rocks.
In some cores, dark brownish green smectite almost completely replaces olivine and, where alteration intensity is high, pyroxene as well. Olivine is commonly altered to smectite near smectite veins or at some distance if the veins cut felsic material (Fig. F44). This type of smectite is commonly associated with sulfide minerals in olivine pseudomorphs, and it is believed to form under low-temperature, anoxic conditions. Pale green to white smectite after plagioclase occurs mainly close to veins and in or near felsic areas. However, it is not clear if there is any systematic relationship between the presence of felsic material and the development of pale green smectite veins. Green smectite may also replace amphibole in both the felsic material and adjacent gabbroic rocks.
Carbonates are mainly found as vein-forming minerals, but they also are intergrown with orange smectite in background alteration near veins (Fig. F36). These phases are mainly Ca-carbonate minerals, most probably calcite. Calcite replaces olivine in the oxidative zone (500-600 mbsf) and is found locally deeper in the core near smectite + calcite + pyrite veins. Calcite also forms after plagioclase along some smectite + calcite + pyrite veins (e.g., Sample 176-735B-132R-1 [Piece 11A]). Deeper in the core, calcite locally accompanies zeolite in veins. Fine-grained translucent needles in Section 176-735B-181R-3 may be aragonite.
Sulfide never exceeds a few percent of the altered rocks and generally represents less than 1% of the core. Sulfide minerals commonly form after olivine in association with smectite, particularly near smectite veins. In many cases this sulfide appears in hand specimen to be pyrite. However, in Cores 176-735B-205R and 207R a reddish brown sulfide is abundant as an alteration phase in the cores of olivine. This could be marcasite or pyrrhotite, both of which have been identified in thin section.
Hematite and orange to brown oxyhydroxide phases are present mainly between 500 and 600 mbsf, in weathered zones related to carbonate veins and in association with orange smectite. In these zones, the minerals form iddingsite-like products after olivine and some pyroxene. Some very limited oxidative alteration occurs further downhole, between 1340 and 1355 mbsf (Cores 176-735B-191R and 192R).
Narrow zones of crystal-plastic deformation are common in the upper parts of the core. In these zones, plagioclase and, to a lesser extent, pyroxene are granulated and recrystallized (Fig. F45). In many cases, these zones are also impregnated with magnetite and ilmenite. Most of these zones underwent extensive chemical and mineralogical modification. Recrystallized feldspar is typically more sodic than the original plagioclase, recrystallized clinopyroxene is more diopsidic, and orthopyroxene is common. Olivine is rare or absent and, if present, is almost completely altered. Clinopyroxene is typically altered to brown hornblende and green actinolite, particularly in the recrystallized zones. Oxides replace much of the recrystallized feldspar and pyroxene; zircon, titanite, and apatite are abundant.
During Leg 176, 253 thin sections were prepared from core recovered from Hole 735B; 10 of these were from the lower 100 m of core recovered during Leg 118, and the remainder from new core recovered during Leg 176. The thin sections were all described by a single observer to provide consistency of observation. All secondary minerals were identified optically, and modes are visual estimates.
The average thin-section sample density down the core is about one per 4 m. Thin sections were typically made for all samples analyzed by XRF and for samples from interesting igneous, metamorphic, or structural intervals. Thus, the amount and character of alteration described in thin section correlates roughly with that determined by visual core description.
In the thin-section descriptions, recrystallized grains, particularly of plagioclase and clinopyroxene, were classified as secondary and included in the estimates of alteration. A study of Leg 118 core showed that recrystallized plagioclase is commonly more sodic than the original igneous grains and that recrystallized pyroxene is more diopsidic (Stakes et al., 1991; Robinson et al., 1991). Sheared and recrystallized zones are commonly impregnated with iron titanium oxide minerals, and they typically have much higher percentages of orthopyroxene than the core as a whole.
Olivine generally occurs as irregular grains of various size, commonly rimmed with orthopyroxene. Alteration ranges from about 1%-2% to 100% (Fig. F37). Because of the irregular shapes of the grains, completely altered olivine may be difficult to distinguish from altered orthopyroxene. However, secondary minerals after orthopyroxene typically preserve some of the original pyroxene cleavage.
Even the freshest olivine is cut by a network of irregular cracks or fractures lined with dark, opaque material (Fig. F46A, F46C). This material appears to be a mixture of smectite, possibly talc, and very finely divided magnetite, although individual magnetite grains rarely can be identified under the microscope (Fig. F27B). Small irregular grains of sulfide minerals are commonly associated with the smectite. As alteration increases, the rims of the grains are replaced by fine-grained mixtures of talc, magnesian amphibole, and finely divided magnetite or pyrite. Small grains may be completely replaced, whereas larger grains, even in the same sample, exhibit only marginal alteration. Irregular patches of talc, amphibole, and magnetite, mark completely altered olivine grains (Fig. F46B). In a few cases, the cores of olivine grains are replaced by mixtures of deep red to orange hematite and smectite, which in turn are surrounded by rims of talc, amphibole, and magnetite. This late-stage oxidative alteration occurs primarily in a zone from Sections 176-735B-90R-4 to 94R-2 (512 to 539 mbsf), where it is commonly associated with carbonate veinlets.
Orthopyroxene is relatively rare in samples recovered during Leg 176. In olivine gabbro, it typically occurs as narrow rims or bands on olivine grains, particularly where olivine is in contact with plagioclase. In more evolved, oxide-rich gabbros, orthopyroxene occurs as discrete crystals either as relatively large, subhedral grains or small, anhedral grains. The latter are most common in deformed zones where the primary minerals was granulated and recrystallized and the rock was impregnated with oxide-rich liquids.
Orthopyroxene is typically much less susceptible to alteration than olivine. In some cases where it rims olivine, the olivine shows marginal alteration, whereas the orthopyroxene is completely fresh or only slightly replaced by pale green amphibole. Where alteration is more intense, the orthopyroxene is partly replaced by green amphibole, brown smectite, and disseminated magnetite or pyrite. In a few cases, the cores of orthopyroxene grains are also replaced by mixtures of reddish orange hematite and smectite. Again, this type of alteration is most common in the interval between Sections 176-735B-90R-4 and 94R-2 (512 to 539 mbsf).
Clinopyroxene is the most abundant ferromagnesian silicate in the rocks from Hole 735B, occurring in every thin section examined. It typically occurs in large, subhedral grains intergrown in complex relationships. It is commonly associated with olivine and, in a few cases, is completely surrounded by it (Fig. F45C). Alteration of clinopyroxene is highly variable, ranging from less than 1% to a maximum of 90%, but in any thin section it is always considerably less than that of olivine. Throughout most of the core, it is less than 10% altered, and more intense alteration is almost always adjacent to veins or in narrow shear zones.
Alteration of clinopyroxene is primarily to brown hornblende and green actinolite. The brown hornblende occurs chiefly in oxide gabbros where the rocks have been sheared, recrystallized, and impregnated with fluids, but it occurs elsewhere as well. The hornblende occurs as small blebs within the pyroxene grains, some of which are aligned along cleavage planes. In other cases, brown hornblende occurs along the margins of pyroxene grains, particularly where they are in contact with oxide minerals. Commonly, the brown hornblende is accompanied by small amounts of ilmenite. Only rarely does brown amphibole exceed 2 modal%.
Green actinolite, on the other hand, may compose up to 50 modal% of the rock, depending on the percentage of original clinopyroxene. The actinolite occurs almost exclusively along the margins of pyroxene grains and penetrates short distances into the crystals along cleavage planes. It also extends into adjacent plagioclase grains along narrow cracks. More extensive alteration of clinopyroxene occurs only adjacent to major veins. The actinolite may be accompanied by small amounts of light green to colorless chlorite, and, in rare cases, chlorite is the dominant secondary phase.
Locally, the clinopyroxene is weakly sheared, granulated, and recrystallized. Where this occurs, the recrystallized grains are typically lighter colored and may be more diopsidic than the original grains. These grains are also partly to completely replaced by brown hornblende or light green actinolite, which may or may not be accompanied by minor oxides.
In all specimens examined, plagioclase is the most stable phase. Over large intervals of the core, alteration of plagioclase is limited to 1% or less and consists of minor amounts of actinolite or smectite along cracks and grain boundaries. Where plagioclase grains are in contact with altered clinopyroxene, the actinolite on the margins of pyroxene grains typically extends outward into the plagioclase for short distances. Alteration is commonly more intense adjacent to veins, where chlorite, carbonate, prehnite, or epidote may occur in addition to amphibole or smectite.
The most extensive alteration of plagioclase occurs in narrow shear zones where the original grains are granulated, recrystallized, and partly replaced by more sodic feldspar. Within these zones the plagioclase may also be replaced by minor epidote and/or chlorite and by more abundant oxides. The plagioclase may be completely recrystallized and replaced, but the zones are typically quite narrow, on the order of a few millimeters to a few centimeters. Elsewhere, alteration of plagioclase rarely exceeds 10% of the amount present.
Based on macroscopic description, 21 vein assemblages (Table T9 ) were recognized in core recovered during Leg 176, and 2792 veins were described and measured. Vein distribution shows wide variation, depending on the mineral paragenesis, but many vein types (e.g., felsic, diopside, and amphibole-bearing veins) show a striking decrease in abundance downsection. Total vein abundance averaged over the core is less than 1%, with felsic veins the dominant vein type by volume; smectite veins are the dominant type by number (Figs. F47, F48). Veins generally were described based on hand samples, as relatively few were observed in thin section. In thin section, most veins are narrow, discrete features with relatively little wall-rock alteration. The common veins in thin section are filled with carbonate, smectite, felsic material, amphibole, diopside ± feldspar, and zeolite in various combinations.
The origin of some felsic veins in gabbroic rocks from Leg 176 is equivocal because of a strong hydrothermal overprint (Fig. F49); therefore, the veins were logged by both the igneous and metamorphic working groups. Where there is no strong hydrothermal overprint, mineral assemblages reflect dioritic to trondhjemitic and granitic compositions, with 1 to 2 modal% of accessory minerals (e.g, apatite, zircon, titanite, and magnetite). By number, the felsic veins make up only 4.08% (N = 114) of the veins measured, but volumetrically they are the dominant type, constituting 45% of the total (Fig. F48; Table T9). Their average length and width is 11.69 cm (±10.79) and 8.75 mm (± 8.24), respectively, and they have sharp boundaries with the host rock. The felsic veins do not show a strong preferred orientation within the core (see "Structural Geology"). The felsic veins are typically, but not always, associated with oxide-bearing gabbros, and they decrease markedly in abundance downsection; below ~1260 mbsf they are rare (Fig. F47B).
Rare felsic veins have cores rich in amphibole ± hydrothermal clinopyroxene (Fig. F50) that are rimmed by plagioclase, but generally mineral phases are irregularly distributed within the veins (Fig. F49). The veins typically contain variable amounts of highly zoned plagioclase, rare potassium feldspar, and myrmekite (Fig. F51C), well-crystallized brown and green amphibole, hydrothermal clinopyroxene, and trace amounts of quartz and biotite (Fig. F51B). The primary vein minerals exhibit negligible to high alteration intensities, and centimeter-wide alteration halos are common.
Commonly, plagioclase forms subhedral to anhedral crystals up to 5 mm across. These typically show strong concentric zoning and patchy or irregular replacement by sodic feldspar and irregular veinlets of secondary plagioclase; in these zones vapor-dominated fluid inclusions are common. In some cases, the plagioclase is extensively replaced by chlorite, which may be accompanied by minor epidote or carbonate. In some veins, plagioclase contains numerous small inclusions of amphibole, chlorite, and dark, very fine-grained material that give the feldspar a pitted or "dirty" appearance. Microveinlets of actinolite and chlorite cut some samples. In more highly altered zones within the veins, prehnite, calcite, smectite ± zeolite, and rare epidote form irregular pods and patches after plagioclase.
Quartz in these veins commonly is rounded to irregular in habit, and the grains typically contain liquid-dominated inclusions with halite daughter (±opaque) minerals. Quartz in some of these veins is partly replaced by brown smectite. In some veins, there is a high concentration of oxide minerals within the vein margins and in the adjacent host rock. Titanite is also common in some of these veins, and it typically occurs as irregular grains along the vein margins or adjacent to mafic minerals like diopside.
Alteration halos associated with these veins reflect moderate to intense alteration of wall-rock plagioclase to secondary plagioclase, and clinopyroxene to hydrothermal clinopyroxene, amphibole, and oxide minerals. Olivine is replaced by amphibole, traces of chlorite, and oxide minerals. In a few cases, the felsic veins have narrow, late-stage cracks filled with carbonate, zeolite, or smectite.
Plagioclase + amphibole and monomineralic plagioclase veins are commonly associated with the felsic veins. In some sections they may represent small veinlets that have splayed off of the larger dioritic to trondhjemitic veins (Fig. F52 ). Like the felsic veins, they strongly decrease in abundance downsection (Fig. F47B) and only occur sporadically below 1250 mbsf. Numerically they constitute 9.6% and 4.9% of the veins measured, respectively, and 19.3% and 4.9% by volume (Fig. F48). They typically form narrow, 2- to 4-mm-wide veinlets that lack a strong preferred orientation (Fig. F53; see "Structural Geology"). The plagioclase + amphibole veins contain highly variable amounts of these two mineral phases, although most veins are dominated by plagioclase. Some of these veins are highly zoned with amphibole-rich cores and plagioclase-rich rims (Fig. F54). In areas of intense alteration, the plagioclase is altered to secondary feldspar, prehnite?, zeolite, carbonate, and smectite. Green amphibole is altered to pale-green to colorless amphibole, variable amounts of chlorite, and fine-grained oxide minerals. Typically, these veins do not have strong alteration halos.
Clinopyroxene-bearing veins are rare in gabbroic rocks recovered during this leg, occurring predominantly in two intervals at 600-650 mbsf and 700-750 mbsf (Figs. F47C, F55. They comprise only 3% of the veins by number and 4% by volume and occur in zones where gabbros and gabbronorites are common. They do not exhibit a strong preferred orientation, but in some zones they are intensely deformed into mylonites (see "Structural Geology"). They are typically narrow veins, 2 to 4 mm in width. The plagioclase + diopside veins commonly have diopside-rich cores that are rimmed by plagioclase (Fig. F55). Diopside crystals in the veins are euhedral to subhedral, up to 2 mm across, and commonly zoned. These crystals commonly contain small, dark inclusions and may be rimmed by amphibole. Like the other felsic veins, these commonly contain low-grade minerals, such as prehnite, carbonate, zeolite, and smectite in the groundmass. Alteration of these veins is highly variable, and some veins exhibit moderately developed alteration halos (Fig. F55). In more intensely altered zones, plagioclase may be altered to secondary plagioclase, and hydrothermal clinopyroxene is common.
Amphibole veins form the second most abundant vein assemblage, constituting 17% of the veins by number, but they are only 5% of the veins by volume (Figs. F47D, F48; Table T9). They are commonly < 0.5 mm wide, dark green (Fig. F55), and in rare cases they form anastomosing fine vein nets. They are generally subvertical (see "Structural Geology"), and in places they cut zones of intense deformation (Figs. F56, F57, F58A). Although in the upper 500 m of the core, the amphibole veins are strongy correlated with deformed regions (Dick et al., 1991a), this relationship is not observed in core recovered during Leg 176 (see "Structural Geology"). Amphibole vein abundance strikingly decreases downsection, and amphibole veins are rare below 1100 mbsf. In rare zones, they are associated with intense, centimeter-wide alteration halos (Fig. F59). In such areas, wall-rock plagioclase is strongly altered to secondary plagioclase, and clinopyroxene is altered to amphibole and less commonly chlorite. Amphibole and chlorite microveinlets cutting plagioclase are common in these zones. Amphibole veins are typically narrow cracks filled with well-crystallized, green amphibole. Where these occur in deformed zones or amphibole gneisses, they typically cut the foliation and, in some samples, they offset bands in the host rock. Most of these veins consist entirely of amphibole, but the largest ones also contain feldspar.
We logged 293 carbonate-bearing veins in the Leg 176 core (Figs. F47F, F48, F60). Carbonate veins compose 9.1% of the total number and 1.8% of the total volume of veins in the core (Fig. F48; Table T9). Carbonate veins are concentrated in the zone between 500 and 600 m, where the rocks exhibit low-temperature oxidation effects and some samples are intensely calcitized. A major fault is present at 560 mbsf within this zone (see "Structural Geology"), and fracturing related to faulting provided pathways for ingress of seawater solutions to form carbonate veins. Carbonate veins also occur in small amounts locally at greater depths. X-ray diffraction of several samples indicates the presence of calcite (Fig. F61), but prismatic crystals in a 0.2-mm-wide vein at 1245.6 mbsf (Core 176-735B-181R) were tentatively logged as aragonite. Iron oxyhydroxide occurs with calcite in four veins in rocks exhibiting low-temperature oxidation at 522 mbsf, and variable proportions of smectite are present with calcite in 35 veins locally throughout the core at greater depths. Calcite veins range from 0.1 to 11 mm wide and from 3 to 39 cm long but average 0.5 mm wide and 7 cm in length. In thin section, carbonate veins are most common in the intervals where olivine and orthopyroxene are replaced by hematite and smectite.
We recorded 1016 smectite veins in the Leg 176 section. These veins make up 47% of the total number and 15.2% of the total volume of veins in the core (Figs. F44, F47F, F48; Table T9). Variable proportions of smectite also occur in 35 smectite + calcite veins, 22 smectite + zeolite veins, and 2 smectite + amphibole veins, where later smectite fills former open space in earlier amphibole veins. Smectite commonly forms small (<1 mm) veins at the center of felsic veins throughout the core.
The presence of smectite was confirmed by X-ray diffraction of several air-dried samples (Fig. F61), but observations in thin section suggest that mixed-layer smectite/chlorite may be present in some samples. Dark green smectite veins occur with local small amounts of pyrite in Cores 176-735B-100R through 131R (575 to 813 mbsf), whereas all smectite veins below this depth consist of pale green to white smectite. Smectite veins range from 0.1 to 7 mm wide and 1 to 130 cm long but average 0.6 (+0.6) mm wide and 7 (+5.7) cm long (Fig. F58B).
Smectite veins occur in two general zones where they are by far the dominant vein type, from 575 to 833 mbsf (Cores 176-735B-99R to 134R) and from 1054 to 1500 mbsf (Cores 176-735B-159R to 210R; Fig. F47F). The upper zone of smectite veins overlaps slightly with the interval of abundant calcite veins (Fig. F47F). Fracturing related to the fault at 560 mbsf and another fault marked by cataclasis at 690-700 mbsf (see "Structural Geology") may have provided pathways for access of seawater solutions to form smectite within this upper zone. Only very rare smectite veins are present from 833 to 1054 mbsf (Cores 176-735B-135R through 158R; Fig. F47F), corresponding to a zone of high-temperature shearing and intense crystal-plastic deformation (see "Structural Geology"). Below this interval, smectite veins are abundant from 1054 to 1508 mbsf; they are particularly abundant from 1230 to 1330 mbsf. Local barren zones occur within the lower smectite vein zone, especially from 1340 to 1380 mbsf (Cores 176-735B-192R to 195R). In contrast to the calcite and upper smectite vein zones, the lower smectite vein zone does not correspond to a fault. Fractures in this zone are tensional (see "Structural Geology"), and they occur within a sequence of relatively uniform and undeformed olivine gabbros (see "Igneous Petrology"). The host rocks for up to 1-2 cm away from smectite veins are variably altered to smectite: olivine is intensely altered to smectite + magnetite + sulfide (pyrite, pyrrhotite), and plagioclase and pyroxene are slightly altered to smectite.
Zeolite and zeolite + smectite veins occur within the zone of smectite veins near the base of the core, from 1130 to 1490 mbsf (Cores 176-735B-168R to 209R), but are concentrated in the interval 1385-1453 mbsf (Cores 176-735B-197R to 205R; Fig. F47E). These veins range from 0.1 mm to 5 mm wide and from 11.2 cm to 144 cm long, but they average 0.9 mm wide and 11.2 cm long. They typically exhibit a broad range of dips (see "Structural Geology"), but vertical zeolite + smectite veins occur in Cores 176-735B-201R through 205R (1418-1455 mbsf). A few individual veins extend through more than one section and were thus greater than 144 cm long before curation.
X-ray diffraction of several veins indicates the presence of natrolite, prehnite, and possibly pumpellyite (Fig. F42). Prehnite was confirmed in observations of thin sections, but the tentative identification of pumpellyite is based on the presence of one peak in a single diffractogram, and must be confirmed by further analyses. Prehnite and zeolite were identified in Cores 176-735B-205R to 210R, and their presence was confirmed by X-ray diffraction (Figs. F42, F43). Prehnite veins are commonly zoned with a narrow band of light brown smectite along the outer edge, followed by a band of prehnite. The prehnite in some cases fills the entire central part of the vein, and in other cases occurs with discontinuous patches of carbonate. In a few cases, the prehnite is intimately intermixed with smectite.
Chlorite veins are present in small amounts locally in the core (Fig. F47E): five veins are present in Core 176-735B-121R at 721 mbsf; one vein each in Cores 176-735B-200R and 202R at 1403 and 1430 mbsf; and 15 subvertical veins of chlorite, chlorite + amphibole, and chlorite + zeolite (natrolite and prehnite) in Cores 176-735B-204R through 209R (1446-1493 mbsf; Fig. F62). The chlorite veins in Cores 176-735B-121R, 200R, and 202R have narrow (millimeter wide) chloritic alteration halos. Along the subvertical chlorite-bearing veins in Cores 176-735B-204R through 209R, plagioclase is altered to a white mineral (zeolite, prehnite, or albite/K-feldspar) in millimeter- to centimeter-wide chloritic alteration halos along the veins (Fig. F62).
Quartz veins are rare, with only six identified locally from 1264 to 1505 mbsf (Cores 176-735B-183R through 210R). Amorphous silica was also identified in two veins within this interval. Smectite is present in some of the quartz veins, which are generally less than 2 mm wide.
A single large (12 mm wide) epidote vein occurs in Section 176-735B-114R-3 (Fig. F63). Wall rock for as much as 2 cm away from the vein is recrystallized to a chloritic assemblage.
The highest temperature metamorphic effects in the plutonic sequence are transitional from magmatic processes; they most likely overlap both temporally and spatially, and distinguishing the effects of these two processes is difficult in places. High-temperature metamorphism (>800°-1000ºC) is clearly marked by localized, narrow zones of crystal-plastic deformation that cut igneous fabrics (see "Structural Geology"). These intervals are characterized by anastomosing bands of olivine and pyroxene neoblasts that are bounded by plagioclase-rich bands, reflecting formation under granulite-grade metamorphic conditions. In some places the high-temperature shear zones are associated with local impregnation of very iron-rich liquids into the rocks to form oxide gabbros, and in many cases these zones have abundant amphibole. In some of these intervals, porphyroclastic gabbros exhibit a well-developed foliation that is marked by olivine and pyroxene porphyroclasts wrapped by well-crystallized brown hornblende, reflecting formation under granulite to transitional amphibolite facies metamorphic conditions. Also present locally are small (centimeter sized) patches of late felsic material that commonly contain high concentrations of amphibole, ilmenite, magnetite, and minor amounts of sulfide, quartz, apatite, and zircon. Many of these rocks, veins, and shear zones may reflect the effects of late magmatic hydrous fluids, but these zones also acted as pathways for later hydrothermal fluids at various temperatures, resulting in the formation of hydrothermal amphiboles and clinopyroxene, chlorite, titanite, smectite, and zeolites. A zone of pervasive deformation and reverse shear from 800 to 1050 mbsf is associated with a high-temperature fault at about 960 mbsf. Brittle veining is rare in this interval. Other high-temperature effects probably resulting from late-stage magmatic activity include the formation of plagioclase + amphibole veins and diopside-rich veins, which in some intervals are progressively transposed into localized zones of high-temperature shear. These veins are most common from 500 to 650 mbsf, and they decrease sharply in abundance below this depth.
High-temperature static background alteration by hydrothermal fluids is patchy throughout the core, and it is commonly most strongly developed in alteration halos associated with felsic veins. This alteration is generally manifested by coronitic alteration halos around olivine grains and the common replacement of clinopyroxene by variable amounts of brown amphibole. In more evolved rocks, magnesian-amphibole ± talc typically replaces orthopyroxene. The secondary minerals most likely formed under low water-to-rock ratios over a range of temperature, from >600°-700ºC (talc, amphibole, magnetite, and hydrothermal clinopyroxene) down to much lower temperatures (e.g., where chlorite is present). Mineral paragenesis is commonly marked by disequilibrium textures, and low-temperature mineral phases typically overprint higher temperature assemblages.
Ingress of moderately high temperature fluids (400°-550ºC) into the plutonic rocks was facilitated by the development of subvertical amphibole veins that are probably related to cooling and cracking of the rocks in the axial environment. In the upper 500 m of the core, these veins are associated with intense zones of deformation (Dick et al., 1991a; Stakes et al., 1991), but this relationship was not observed at deeper intervals (see "Structural Geology"). The abundance of amphibole veins decreases markedly with depth in the Leg 176 section, and below 600 mbsf amphibole veins are rare. The absence of intense amphibole veining at depth is reflected in the low degree of background alteration. At a smaller scale, microcracks (<100 µm) filled with talc, magnetite, amphibole, sodic plagioclase, chlorite, and epidote are sporadically present throughout the core and represent smaller scale fracturing and fluid penetration at variable temperatures.
All of these processes reflect a wide temperature range, but temperatures were most likely high enough during this stage to be limited to the spreading axis or to the very near-axis environment. The upper 200 m of the Leg 118 section is more intensely veined and altered, has lost copper and sulfur, and contains fluids with salinities similar to hydrothermal vents; this section is interpreted to be the root zone of an axial hydrothermal system(s) (Alt and Anderson, 1991; Alt, 1995; Kelley, 1996, 1997). In contrast, some intervals of the Leg 176 section are strongly affected by low-temperature fluid penetration, as evidenced by abundant smectite ± carbonate veins. The abundant veins of calcite, smectite, and iron oxyhydroxide minerals and the associated intense alteration localized at 500-600 mbsf reflect low-temperature alteration by circulating seawater solutions. These veins are most likely related to the presence of a fault at 560 mbsf (see "Structural Geology"). Below this interval, veins of smectite + pyrite + calcite and associated smectitic alteration of surrounding wall rock reflect low-temperature hydrothermal reactions under more reducing conditions. These effects occur throughout much of the core, but the abundant smectite veins at 600-800 mbsf may be related to a second fault at 690 mbsf. The zone of smectite and/or chlorite/smectite, prehnite, and zeolite veins and associated host rock alteration in the lower part of the core (1300-1500 mbsf) requires further work to document and confirm mineral paragenesis, but represents subgreenschist alteration at variable temperatures and fluid compositions. This lower temperature set of veins formed in tensional fractures and is perhaps related to uplift and cooling of the block in an off-axis environment, outside the axial convective cell.