SCIENTIFIC RESULTS

Expedition 309, Superfast Spreading Rate Crust 2, successfully deepened Hole 1256D by 503 m to a total depth of 1255.1 mbsf, or 1005.1 msb. At the end of Expedition 309, Hole 1256D penetrates a total of >750 m of extrusive lavas and proceeds a further 250 m into a region dominated by intrusive rocks. At 1255 mbsf, Hole 1256D is tantalizingly close to the predicted minimum estimated depth for the frozen axial magma chambers (1275 mbsf; Fig. F14). Following the completion of a comprehensive wireline logging program, Hole 1256D was successfully exited and left clear of equipment with only minor unconsolidated fill at the bottom of the hole. Hole 1256D is in excellent condition and ready for deepening.

Expedition 309 (July–August 2005) will be followed closely by Expedition 312 (November–December 2005), which will continue to deepen Hole 1256D. Despite our grueling pace of advance (~15 m/day), progress with deepening Hole 1256D was steady (Fig. F14). Optimistically anticipating the same benign drilling conditions and good fortune, assured of highly astute rig floor operations, Expedition 312, with 37 days of drilling operations, is set to deepen Hole 1256D by a further 500 m. This is well beyond the depths where geophysical interpretations predict gabbros to occur (~1275–1525 mbsf; Fig. F14; see "Predictions of Depth to Gabbros" and "Background and Objectives").

The first scientific operation of Expedition 309 was to deploy the wireline WSTP to collect a sample of the fluid at the bottom of Hole 1256D and measure the ambient thermal conditions in the hole using the Advanced Piston Corer Temperature (APCT) and Temperature/Acceleration/Pressure (TAP) tools. The WSTP was followed by wireline logging runs using the triple combo and FMS-sonic tool strings to assess the condition and caliper of the hole and check for borehole wall breakouts and unstable regions. There was ~27 m of loose fill in the bottom of Hole 1256D that precluded running the wireline tools to the full depth of the hole (752 mbsf).

The temperature profile made with the TAP tool on the triple combo shows a steady increase in temperature from the base of the casing (269 mbsf) to a maximum temperature of 67.5°C recorded at the deepest logging depth (~725 mbsf). Heat flow measured in the sedimentary section of Hole 1256B to a depth of 158 mbsf was 113 mW/m² (Wilson, Teagle, Acton, et al., 2003). In the basement section in Hole 1256D, TC averaged 2.0 W/(m·K) in the ponded lava flow above 350 mbsf and 1.7 W/(m·K) below 350 mbsf. For uniform heat flow of 113 mW/m² downhole, these conductivities predict a thermal gradient of 0.056 K/m above 350 mbsf and 0.067 K/m below 350 mbsf. These predictions are reasonably close to the observed gradients of 0.067 and 0.071 K/m, probably within uncertainties of the TC measurements and the inevitable slight disturbance of the ambient hole temperature due to the passage of the drill string and wireline tools. This suggests that there is little advection of heat by fluid in the Site 1256 basement or major vertical fluid movements in Hole 1256D.

The 67.5°C deep borehole fluid collected from the bottom of Hole 1256D (~725 mbsf) is chemically very distinct from seawater. Relative to Site 1256 bottom seawater, the borehole fluid is hotter, slightly more neutral (pH = ~7.4), and has significantly lower alkalinity (0.85 mM). Salinity is unchanged (35‰). The largest changes are in the concentrations of dissolved ions with major reductions in the concentrations of boron (–18%), sulfate (–19%), potassium (–41%), lithium (–47%), and magnesium (–55%). In contrast, the strontium concentration is slightly increased (18%), and the calcium content is very strongly elevated (415%). The deep borehole fluid is also significantly different from the composition of the ~35°C uppermost (~250 mbsf) basement fluid estimated from pore water chemical gradients measured at Site 1256 during Leg 206 (Wilson, Teagle, Acton, et al., 2003). The deep borehole fluid has lower Li (–15%), Mg (–25%), and K (–30%) concentrations but higher dissolved silica (14%), sulfate (84%), and calcium (122%) concentrations relative to the uppermost basement fluid.

When the deep borehole fluid is compared to the well-characterized basement fluids from the eastern flank of the Juan de Fuca Ridge (Elderfield et al., 1999), most ions are present in concentrations similar to those predicted for a 67.5°C fluid. The exception is the Mg concentration of the deep borehole fluid, which at ~24 mM is higher than would be expected for a fluid reacted with basement at >60°C. The disagreement between the concentration of Mg in hot deep borehole fluids and that predicted from uppermost basement fluids or laboratory experiments has previously been noted for wireline fluid samples recovered from Hole 504B (e.g., Becker, Foss, et al., 1992). However, the chemistry of the deep borehole fluid from Hole 1256D is closer to equilibrium than the most pristine borehole fluids, which were taken from the deepest available points in Hole 504B during the different drilling expeditions there. Those fluids have Mg concentrations similar to those of the uppermost basement fluid at Site 504. In contrast, the fluid from the bottom of Hole 1256D has Ca (and Li and K) concentrations close to those expected for a fluid in chemical equilibrium with 67.5°C basement and, unlike Hole 504B, the Mg concentration is at least partially decreased toward the predicted composition.

The primary purpose of the initial logging operations was to check Hole 1256D for borehole wall breakouts and variations in hole diameter though comparison with measurements made at the end of Leg 206. The initial phase of wireline logging indicated that, before the commencement of drilling, the borehole conditions in Hole 1256D are excellent, and no ledges or obstructions were encountered. Caliper readings from both the triple combo and FMS-sonic tool strings show good borehole conditions, with a diameter typically between 10 and 12 inches. The FMS-sonic tool string followed a different pathway during the Expedition 309 pass compared with the Leg 206 passes, and consequently, in many intervals the FMS image coverage of the borehole wall has increased. Sonic velocities measured by the Dipole Sonic Imager appear to be of high quality. Several narrow zones (at 517, 597, 602, and 685 mbsf) with strong differences in the orthogonally arranged calipers on the FMS-sonic tool were identified during both Leg 206 and Expedition 309. These intervals may correspond to borehole breakouts, and the north–south orientation of the borehole enlargements suggests a west–east maximum stress direction. A tight spot recognized on Leg 206 at 486 mbsf was also recorded during the Expedition 309 predrilling passes, and a new zone was identified at 472 mbsf (9.3 inches).

Hole 1256D was reentered with a rotary core barrel (RCB) assembly and CC-9 coring bit at 2325 h on 18 July 2005, and ~27 m of loose fill was cleared from the bottom of the hole so that it was open to the full depth achieved during Leg 206 (752 mbsf; 502 msb). Rotary coring of the basement continued until 20 August (~33 days), when the hole was conditioned for wireline logging operations. A total of nine CC-9 RCB hard-formation coring bits were used, and Hole 1256D was deepened by ~503 m to 1255.1 mbsf (~1005.1 msb; Cores 309-1256D-75R through 170R). Drilling proceeded without core liners from Core 309-1256D-91R to stop the jamming of fractured basalt in the core liner sleeve and core liner, which sometimes prevented the capture of additional core. Because of the slow penetration rates, often less than 1 m/h, we recovered the core barrel after every 4–5 m advance from Core 309-1256D-103R. Including wireline time, reaming, hole-cleaning, and other drilling activities, a rough approximation of daily progress at ~15 m/day may be useful for planning future deep drilling of the oceanic basement and a target for progress during Expedition 312 (Fig. F14).

At 1005 msb, Hole 1256D is the fourth deepest hole drilled into oceanic basement since the launch of scientific ocean drilling in 1968 and the second deepest penetration into in situ ocean crust (Fig. F15). Hole 504B, deepened during seven DSDP and ODP legs into 6.9 Ma crust on the southern flank of the intermediate-spreading Costa Rica Rift, remains the deepest penetration of in situ ocean crust. Prior to Expedition 309, Hole 504B was the only hole to sample a complete sequence of extrusive rocks as well as the transition from extrusive rocks to sheeted dikes (Alt, Kinoshita, Stokking, et al., 1993).

To facilitate description and discussion of the crustal stratigraphy at Site 1256 and assist in the interpretation of cores recovered during Expedition 309, we present a preliminary subdivision of the upper crust sampled so far in Hole 1256D (Table T4; Fig. F16). Detailed descriptions are presented in the following sections. These subdivisions were suggested principally from igneous stratigraphy and were made before the wireline logging was undertaken. As such, they should be considered with the proviso that the boundaries suggested are preliminary and that they may change as further information becomes available.

The upper crust at Site 1256 can be portioned into five basement subdivisions, which, in descending order down the hole, are the lava pond, inflated flows, sheet and massive flows, transition zone, and sheeted intrusives (Table T4). The uppermost two zones were cored during Leg 206, as was the upper ~220 m of the sheet and massive flows.

The lava pond caps the uppermost crust at Site 1256. This domain includes Units 1256C-1 through 18 and 1256D-1 (~250–350.3 mbsf). The uppermost lavas were not recovered in Hole 1256D because 16 inch casing was set 19.5 m into basement and the interval was not cored. In Hole 1256C, the rocks immediately below the sediments comprise thin basaltic sheet flows a few tens of centimeters to ~3 m thick separated by chilled margins and rare intervals of recrystallized sediment (Units 1256C-1 through 1256C-17). The massive ponded flow, sensu stricto (Units 1256C-18 and 1256D-1), is defined at its top by a ~75 cm rind of glassy to cryptocrystalline aphyric basalt that overlies ~30 to ~74 m of fine-grained basalt in Holes 1256C and 1256D, respectively. The massive ponded flow becomes abruptly cryptocrystalline ~1.5 m from the base of the flow. Although the massive flow is much thicker in Hole 1256D than in 1256C, it is interpreted as a single lava body whose interior was liquid at the same time in both locations. The dramatic increase in thickness over 30 m of lateral distance and a total thickness in excess of 74 m indicates that there was at least this much paleotopography in order to pool the lava. On fast-spreading ridges, such topography does not normally develop until ~5–10 km from the axis (e.g., Macdonald et al., 1989), and we suspect that these lavas flowed a significant distance off axis before ponding in a faulted depression.

Immediately underlying the lava pond is a sequence of massive flows, pillow lavas, and sheet flows (Units 1256D-2 through 15; 350.3–533.9 mbsf) grouped together as the inflated flows. Although rocks exhibiting a number of eruptive styles are included here, the critical criterion for subdivision is the occurrence of subvertical elongate fractures filled with quenched glass and hyaloclastite (e.g., Sections 206-1256D-21R-1 and 40R-1) at the top of the lava flows. These features indicate flow-lobe inflation that requires eruption onto a subhorizontal surface at less than a few degrees (Umino et al., 2000, 2002).

The bulk of the extrusive lavas at Site 1256 are included in the sheet and massive flows (Units 1256D-16 through 39b; 533.9–1004.2 mbsf). A total of 218 m of this subdivision was drilled during Leg 206, with a further 252 m of penetration during Expedition 309. This sequence consists of sheet flows tens of centimeters to ~3 m thick with subordinate massive flows >3 to 16 m thick and uncommon breccias. The flows are aphyric to sparsely phyric, cryptocrystalline to microcrystalline basalts. Units are distinguished by the presence of chilled margins or by grain-size variations. Throughout this interval, glassy chilled margins are common.

It is the very essence of a transitional sequence that boundaries are loosely defined and subjective. In Hole 1256D, the transition zone from Units 1256D-40 through 44a (1004.1–1060.9 mbsf) is identified by the occurrence of a number of criteria and different rock types as opposed to the appearance of one specific feature. Shore-based analysis of wireline logs and further petrographic and geochemical investigations will help refine the boundaries of this zone. Most of the rocks within the transition zone are aphyric, cryptocrystalline sheet flows. The top of the transition zone is marked by a cataclastic massive unit (Section 309-1256D-117R-1, 85 cm). This comprises subvertically oriented cryptocrystalline basalt clasts hosted within a very highly altered fine-grained basalt that has been incipiently brecciated and deformed along numerous fine veins and cataclastic stringers. Core 309-1256D-120R (~1018 mbsf) includes the first sign of a subvertical intrusive contact other than the single occurrence further upcore in Section 206-1256D-32R-2 at ~475 mbsf. Dike chilled margins become more common downhole, although extrusive textures and vesicles are still encountered. It should be noted that subvertical fracture sets possibly indicative of diking into the host rocks near Hole 1256D are common from ~900 mbsf. Breccias of various styles are common in the transition zone, including a spectacular mineralized volcanic breccia that comprises Unit 1256D-42a (interval 309-1256D-122R-1, 20 cm, to 122R-2, 30 cm; ~1028 mbsf). In the transition zone, secondary mineral assemblages (chlorite-smectite, albite, chlorite, actinolite, anhydrite monite) indicative of hydrothermal alteration at subgreenschist to greenschist facies temperatures start to become more common.

The upper boundary to the sheeted intrusives is defined by a change from sheet flows to massive basalts at 1060.9 mbsf (Unit 1256D-44a; Core 309-1256D-129R). From that level, subvertical intrusive contacts are common, and these can be sharp or irregular and lobate, the latter style indicating the intrusion of magma into hypersolidus rocks (e.g., interval 309-1256D-149R-1, 30–97 cm; 1156 mbsf). Extrusive rocks could be present in this domain, although there are no unambiguous indicators of eruption. Groundmass grain sizes vary from glassy to microcrystalline, although a few samples deeper in this subdivision are fine grained with holocrystalline doleritic textures. No fresh glass was found in the sheeted intrusives, but altered glass is present along some dike chilled margins and associated breccias. There is a step change in physical properties with significant increases in average TC (from 1.8 and seismic velocity (5.4 salts decreases from 4% the massive basalts that are the dominant host rock of the sheeted intrusives are dikes or sills remains uncertain. Unfortunately, an unambiguous subvertical contact that grades continuously from a glassy chilled margin to a microcrystalline to fine-grained massive basalt has so far not been recovered.

Basement rocks recovered during Expedition 309 in Hole 1256D from 752 to 1255 mbsf were divided into 39 igneous units (Units 1256D-27 through 65), labeled continuously from the last rocks recovered during Leg 206 (Table T5; Fig. F17) (Wilson, Teagle, Acton, et al., 2003). The basement cored during Expedition 309 has been divided into three crustal sections: the sheet and massive flows, which continue from Leg 206, a lithologic transition zone, and the sheeted intrusives (Fig. F16).

The sheet and massive flows (Units 1256D-16 through 39; 533.9–1004.1 mbsf) are mainly composed of sheet flows and massive flows. Sheet flows (with individual cooling units ranging from tens of centimeters to <3 m thick) make up 80% of the total sheet and massive flows cored during Leg 206 and Expedition 309. However, in the portion of this subdivision drilled during Expedition 309, sheet flows account for only 65% of the rock recovered, indicating a greater prevalence of massive flows deeper in the section. Individual flows are commonly separated by chilled margins containing altered or fresh glass. Where contacts were not recovered, individual flows could be distinguished by systematic changes in grain size. Using these criteria, minimum thicknesses of individual flows or cooling units range between 0.11 and 1.68 m with an average thickness of 0.55 m (<1% phenocrysts), and grain size ranges from glassy at the chilled margins to cryptocrystalline or microcrystalline (Fig. F18). Rare sheet flow interiors are fine grained. The groundmass of sheet flows generally consists of plagioclase and clinopyroxene microlites, with interstitial titanomagnetite and altered glass, similar to those described during Leg 206. Where phenocrysts occur (for example in Units 1256D-28, 35b, and 37), these rocks have plagioclase, clinopyroxene, and olivine phenocrysts, in order of decreasing abundance, commonly clustered in a glomeroporphyritic texture (Fig. F19). Unit 1256D-35c contains three small (0.5–2.2 cm) holocrystalline gabbroic xenoliths (intervals 309-1256D-107R-1, 44–52 cm, 108R-1, 20–36 cm, and 108R-1, 132–138 cm) consisting of fine-grained olivine, plagioclase, and clinopyroxene (Fig. F20).

In contrast to the thinner sheet flows, minimum thicknesses of the massive flow units vary from 3.2 to 11.3 m with an average of 6.3 m (cumulative thickness calculated using only the pieces recovered). The thickest, Unit 1256D-31, consists of a single cooling unit of fine-grained basalt below a 12 cm cryptocrystalline to microcrystalline upper contact (Fig. F21). A total of 26 m of this unit was cored, of which 11.3 m was recovered. In contrast to the sheet flows, fine-grained rocks are more common in the massive lavas (Fig. F17). The massive flows are aphyric and nonvesicular, with the exception of Subunit 1256D-39a. This basalt is sparsely clinopyroxene-olivine-plagioclase-phyric and is moderately vesicular (8%) (Fig. F22). Thin section observations show that the most finely grained rocks collected from the massive flows have intergranular to intersertal groundmass textures (Fig. F23).

The transition zone (Units 1256D-40 through 43; 1004.1–1060.9 mbsf) is characterized by increasing abundance of volcanic breccias interbedded within sheet flows. The top of this zone is defined by the cataclastic massive unit (Unit 1256D-40; interval 309-1256D-117R-1, 85 cm, through 118R-1, 66 cm). The upper part of this unit (interval 309-1256D-117R-1 [Pieces 9–14, 97–142 cm]) has a complex structure with fine- to medium-grained basalt in contact with brecciated clasts of cryptocrystalline basalt (Fig. F24). The fine- to medium-grained basalt contains highly altered glass clasts and is disrupted by an intensive network of thin chlorite-smectite veins imparting an incipient cataclastic texture. Thin section examination of these rocks (Sample 309-1256D-117R-1, 122–125 cm) shows that fractured crystals have a seriate texture, deformed and cemented by a banded matrix that shows flow structures. With increasing distance below the top of the unit, the igneous texture is better preserved and more homogeneous, mesostasis is less abundant, and crystals are less fractured (intervals 309-1256D-117R-2, 9–72 cm, through 118R-1, 0–66 cm). In this lower part of Unit 1256D-40, the cataclastic massive unit consists of fine-grained dolerite with a partially developed subophitic texture (interval 309-1256D-117R-2, 23–26 cm). A few pieces, similar to the disrupted rocks of the upper part of this cataclastic unit, occur in intensively veined sheet flows in Units 1256D-37 and 41. This may support the interpretation that these disrupted rocks indicate the nearby presence of a dike or fault zone. Interval 309-1256D-120R-1, 9–26 cm, of the transition zone, however, captures the first unambiguous subvertical intrusive contact.

A second type of breccia is present in interval 309-1256D-122R-1, 25 cm, through 123R-1, 109 cm, where 2.8 m of mineralized volcanic breccia and breccia intercalated with basalt was recovered and defines Unit 1256D-42 (Fig. F25). This unit can be further subdivided based on abundance of basaltic rocks. The upper part, Subunit 1256D-42a, consists solely of volcanic breccia (interval 309-1256D-122R-1, 25–149 cm, through 122R-2, 0–30 cm), but in Subunit 1256D-42b, the breccias are intercalated with aphyric, cryptocrystalline to microcrystalline basaltic sheet flows. These breccias comprise angular to subangular aphyric cryptocrystalline basaltic clasts (0.5–4.5 cm) and subangular to elongate clasts of altered glass with rare flame-shape clasts (0.1–1.5 cm), cemented by chalcedony, saponite, calcium carbonate, albite, anhydrite, and sulfides (Fig. F25).

The transition zone also hosts the last occurrence of a glassy margin not associated with either a dike contact or clastic brecciation. This margin was recovered in the lower half of sheet flow Unit 1256D-43 (~1060 mbsf).

The upper boundary to the sheeted intrusives (Units 1256D-44 through 65; 1060.9–1255.1 mbsf) is defined by a distinct change from sheet flows to massive basalts at 1060.9 mbsf. Extrusive rocks could be present in this section, but evidence for eruption remains ambiguous. The massive basalts are most commonly aphyric and nonvesicular. Most rocks are microcrystalline (Units 1256D-44 through 46, 49 through 50, 52A, 54 through 55, and 59 through 65) and fine-grained (Units 1256D-47, 48, 51, and 53a) basalts, but rare units are cryptocrystalline to microcrystalline (Units 1256D through 56a, 57a, and 58) basalt. Thin sections of the massive basalts show holocrystalline and commonly doleritic groundmass textures (Fig. F26).

In contrast to shallower in Hole 1256D, subvertical intrusive contacts, thought to be dikes, are common, suggesting that the massive basalts of the sheeted intrusives represent the beginning of the sheeted dike complex. In general, two types of contacts can be distinguished: sharp or irregular direct contacts and brecciated contacts. Most contacts belong to the latter category, with brecciated zones one to several centimeters wide along the contact. All contacts have developed chilled margins. The chilled margins of the dikes are composed of glassy to cryptocrystalline aphyric basalts that have quenched against cryptocrystalline to fine-grained, massive basaltic hosts. Breccias at the contacts comprise fragments of altered glass initially quenched at the chilled margin with subordinate, angular to subangular clasts of the host rock and cemented by anhydrite, chlorite, and sulfide (Fig. F27). One spectacular example is a >50 cm long vertical contact in interval 309-1256D-140R-1, 26–80 cm (Unit 1256D-47), with a sulfide-impregnated dike margin breccia with complex intrusive relationships and intricate multiple margin-parallel sulfide veins crosscut by anhydrite veins.

In interval 309-1256D-155R-1 (Piece 20, 84–90 cm), the chilled margin forms a convex lobe, indicating that the host rock was not rigid during the intrusion (Fig. F28A). Further evidence for multiple intrusions are seen in Sections 309-1256D-161R-1 to 161R-2 (Unit 1256D-56b), where at least two intrusions are present. An inner sparsely clinopyroxene-olivine-plagioclase phyric dike has intruded a sparsely clinopyroxene-olivine-plagioclase-phyric spherulitic cryptocrystalline rock, itself chilled against an aphyric microcrystalline host rock (Fig. F28B). Another example of dike intrusion into a ductile host rock is observed in interval 309-1256D-163R-1, 113–122 cm. This contact is lobate and highly complex with fractured pieces of the chilled margin dispersed in the host rock (Fig. F28C).

The overall mineralogical characteristics of basement drilled during Expedition 309 are similar to Leg 206 basalts, although there are some important differences. More than 60% of the basalts drilled during Leg 206 are sparsely phyric with olivine, plagioclase, and clinopyroxene phenocrysts (Wilson, Teagle, Acton, et al., 2003), and the proportion of aphyric basalts is <40% for Leg 206. In contrast, the vast majority of basalts recovered during Expedition 309 are aphyric (>80%) (Fig. F29). This difference is demonstrated by the downhole variation of phenocryst contents and by a decrease with depth of the total phenocryst content (Fig. F17). More than half of Leg 206 basalts have three major phenocryst phases (clinopyroxene, olivine, and plagioclase), whereas only rare samples (eight) of the Expedition 309 basalts have more than two phenocryst phases (Fig. F29). Phenocryst-bearing basalts collected during Leg 206 are dominantly olivine-phyric (>80%), whereas in Expedition 309 basalts, plagioclase is the most common phenocryst phase and olivine is the least abundant among the three phenocryst phases (Fig. F29). The different observation teams during Leg 206 and Expedition 309 may account for some of this difference, but such bias should be relatively minor. The general change from sparsely phyric to aphyric and from dominantly olivine to plagioclase phenocrysts appears to be a real downhole trend.

The freshest rocks from each igneous unit were selected for elemental analysis by shipboard inductively coupled plasma–atomic emission spectroscopy (ICP-AES) to obtain a downhole record of primary magmatic compositions (Fig. F30). All samples have chemistries within the range of MORB with SiO₂ = 48–55 wt%, FeO = 9.4–14.0 wt%, MgO = 6.2–8.9 wt%, CaO = 7.1–12.8 wt%, Na₂O = 1.8–5.0 wt%, Cr = 21–367 ppm, Sr = 71–129 ppm, Zr = 56–133 ppm, and Ba = 1–37 ppm. The range of the Mg# is 45–62, with an average value of 53. These values broadly overlap the results from Leg 206 (Fig. F31) and correspond to typical values for MORB (Su and Langmuir, 2003). Although there is considerable scatter in the data, linear magmatic fractionation trends are present for TiO₂, FeO, CaO, Na₂O, and Zr versus MgO (Fig. F31).

There are subtle variations in the basalt chemistry downhole, with a number of step changes or reversals of fractionation trends possibly indicating cycles of fractionation, replenishment, and, perhaps, assimilation (e.g., at ~600, 750, 908, and 1125 mbsf). Analyses of cryptocrystalline basalts that are unambiguously dikes are chemically indistinguishable from massive basalts into which these dikes were intruded. There do not appear to be any systematic geochemical differences between sheet flows, massive flows, and dikes (Fig. F30).

Similar to analyses of basalts from Leg 206, TiO₂ and Y show a good positive linear correlations with Zr, due to their similar geochemical behavior (Fig. F32). Leg 206 basalts above 750 mbsf were classified into three groups based on a distinct gap in the TiO₂-Zr concentration: high Zr-TiO₂, low Zr-TiO₂, and high Zr groups. With additional data analyzed during Expedition 309, this data gap disappears and the three-fold subdivision is probably not valid. Rare samples from the inflated flows in Hole 1256D fall off the Y versus Zr and TiO₂ versus Zr trends, suggesting some minor variation in source composition.

Basalts from different igneous subdivisions in Holes 1256D and 1256C all have MgO in the range 6–9 wt%, and when trace element compositions of Site 1256 basalts are compared to compilations of EPR MORB, they are within one standard deviation of the average, albeit on the relatively trace element–depleted side of average EPR MORB (Fig. F33). Note that Site 1256 basalts have higher Zr/Y and Zr/TiO₂ ratios than the strongly trace element–depleted MORB from Hole 504B (Fig. F34). From all the basalts from Site 1256, the lava pond is the only unit that is relatively enriched in V and depleted in Cr compared to the EPR MORB. The lava pond includes the rocks with the highest incompatible element (Zr, TiO₂, Y, and V) concentrations and the most depleted in compatible elements (Cr and Ni), suggesting that it is more evolved than the rest of the basalts from Hole 1256D (Fig. F33).

Compared with other first-order mid-ocean-ridge segments along the EPR, the basalts from Site 1256 have very low Zr/TiO₂ and Zr/Y ratios (Fig. F35). Although there is overlap among segments and large scatter in the data for each segment, Zr/TiO₂ and Zr/Y ratios appear to decrease with increasing spreading rate. The origin of this relationship is unclear, but spreading rate may affect magma fractionation or partial melting of the mantle, or else reflect regional-scale mantle heterogeneity.

One of the principal objectives of the combined missions (Leg 206 and Expeditions 309 and 312) is to investigate the alteration processes that occur in a section of upper crust that formed at superfast spreading rates (200–220 mm/y) to test whether these differ from those documented in crust formed at slow and intermediate spreading rates. Of particular interest during Expedition 309 is the opportunity to observe the transition between low-temperature alteration and high-temperature hydrothermal alteration in a continuous section of oceanic crust. To date, this transition has only been described in Hole 504B. All rocks in the ocean crust intersected by Hole 1256D are partially altered to secondary minerals, and the products of fluid-rock exchange are manifest by background alteration, alteration halos related to veins, isolated alteration patches, and veins and breccias (Figs. F36, F37, F38).

Two main alteration types were encountered in the section of Hole 1256D drilled during Expedition 309. From 752 to 965 mbsf, rocks that had reacted with seawater at low temperatures, similar to the range of conditions encountered in Leg 206 cores, are present. The background alteration is uniform, from 85% to 100% dark gray, because of the presence of saponite filling vesicles and replacing olivine and plagioclase; clinopyroxene phenocrysts and chalcedony and calcium carbonate filling vesicles; and miarolitic voids (Fig. F36). The predominant vein mineral phases related to low-temperature alteration in Hole 1256D include saponite, celadonite, iron oxyhydroxides, chalcedony, and minor pyrite (Figs. F37, F38). Celadonite is commonly intergrown with iron oxyhydroxides with later overgrowths of saponite. Specific vein-related alteration types identified in Hole 1256D include black halos, brown halos, mixed halos, simple light green, light gray, dark green, and light gray halos, and discontinuous pyrite halos (Fig. F36). Black, brown, and mixed halos and dark patches are common throughout the rocks from 752 to 918 mbsf and are related to veins filled by saponite, celadonite, and iron oxyhydroxides. These halos result from the more pervasive replacement of the host rock groundmass, as well as olivine and plagioclase phenocrysts. The formation of black halos derives from an early low-temperature seawater–basalt interaction under anoxic conditions, which initiated during cooling of the lava within 1–2 m.y. of basalt emplacement (Böhlke et al., 1980; Honnorez, 1981; Laverne, 1993; summary in Alt, 2004). Subsequent interaction of the basalts with cold oxidizing seawater produces brown halos characterized by replacement of primary phases by saponite and iron oxyhydroxides. From 918 to 964 mbsf, black, brown, and mixed halos are absent (Fig. F36) and dark gray background alteration with abundant saponite and pyrite is ubiquitous. These rocks, as well as saponite- and pyrite-bearing intervals cored during Leg 206 (e.g., 554–562 mbsf), result from the interaction of basalt with low-temperature basement fluids that have chemically evolved from seawater through water–rock reactions.

The interval from 964 to 1028 mbsf is an alteration transition zone and is characterized by the presence of pyrite-rich alteration halos and mixed-layer chlorite/smectite instead of pure saponite (Fig. F39). We also observe an increase in the occurrence of anhydrite in this zone. This alteration mineral assemblage suggests slightly more elevated temperatures (100°–200°C) than are found higher in the crust. The alteration transition zone ends at ~1028 mbsf with the occurrence of the mineralized volcanic breccia. From ~1028 mbsf, green and dark green background alteration, particularly in the coarser grained rocks, occurs as a consequence of the moderate to complete replacement (up to 100%) of basaltic clasts and glass in the mineralized volcanic breccia to saponite/chlorite and minor talc. The first occurrences of actinolite, prehnite, titanite, and epidote are recorded at 1027, 1032, 1051, and 1095 mbsf, respectively (Fig. F39). These minerals are indicative of hydrothermal alteration under subgreenschist to greenschist facies conditions. In this part of the crust, alteration halos occur both as simple dark gray, dark green, light gray, and light green halos and composite halos in which every combination of these colors is possible. These halos comprise 10%–100% secondary minerals with chlorite, actinolite, titanite, albite, and pyrite as the main secondary minerals replacing plagioclase and clinopyroxene and filling interstitial spaces along with minor quartz, chalcopyrite, epidote, and prehnite.

Although there are a few spectacular examples of highly altered and partially mineralized rocks (e.g., the mineralized volcanic breccia at ~1029 mbsf), the rocks from Hole 1256D are less altered compared to most other basement sites (e.g., Sites 417 and 418, Holes 504B and 896A). Within the extrusive lavas, Hole 1256D contains a much smaller amount of black, brown, and mixed alteration halos compared to Holes 504B and 896A, and this alteration style is not systematically present within the uppermost region of the crust. Instead, these alterations by relatively oxidizing fluids occur irregularly with depth and are most commonly associated with well-developed steeply dipping vein networks. As observed during Leg 206, the amount of calcite within Hole 1256D is very low compared to other basement penetrations.

Although pyrite is abundant in the Expedition 309 cores, intense quartz-epidote-Fe, Cu, Zn, Pb sulfide stockworklike mineralization as was intersected in Hole 504B was not intersected in the alteration transition zone. However, anhydrite, which is sparse in Hole 504B (Teagle et al., 1998), is abundant at Site 1256.

The basalts recovered during Expedition 309 exhibit brittle structures and minor brittle-ductile structures. The main structural features are represented by veins, vein networks, cataclastic zones, shear veins, microfaults, and breccia (Fig. F40). Primary igneous structures include syn- to late-magmatic structures, partially linked to flow and solidification of lava. Three main types of breccia were recovered during Expedition 309: incipient breccia, hyaloclastite, and hydrothermal breccia.

In the sheet and massive flows (752–1004.2 mbsf), structures and fracturing are heterogeneously partitioned and are most intensely developed at the top of the massive flows. Vertical sets of veins, cataclastic zones, and shear veins are present in massive units, whereas breccias (incipient breccia) are more common in sheet flows. On the whole, the vertical vein sets become more common from ~900 mbsf. Most structures are related to the cooling of lava and are represented by curved, radial, Y-shaped, and irregular veins filled with secondary minerals.

The transition zone (1004.2–1060.9 mbsf) is characterized by the appearance of steeply dipping chilled dike margins and the concurrent presence of cataclastic zones, breccias (mostly hyaloclastite), and vertical veins. The cataclastic massive unit in Section 309-1256D-117R-1 consists of rounded to angular clasts of dolerite and glassy spherulitic to variolitic basalt. Three to four centimeters of cataclasite separates doleritic basalt fragments from chilled fragments (Fig. F41). The damaged zone is characterized by a complex network of tiny veins, mostly dark green, dark brown, and light green, on the cut surface of the core. The light green veins have an aphanitic vitreous luster and disturb and cut across dark brown cataclastic saponite-bearing bands. These bands are composed of cataclasite and protocataclasite cut by veins of ultracataclasite and gouge. The crosscutting relationships between the different type of "fault-rocks" are visible in thin section (Fig. F41). Vein networks and cataclastic banding have caused incipient brecciation of the host rock, and larger fragments show only minor relative rotation. Flow-related microstructures and laminations are observed in very narrow (0.2–0.5 mm wide) veins. In thin section, fragments of plagioclase show intergranular and intragranular deformation. Clasts are surrounded by a banded matrix that displays flow textures and is made up of subangular and rounded fragments of minerals and altered glass of variable grain size.

The intensity of fracturing downhole is mostly slight with the exception of the mineralized volcanic breccia (Unit 1256D-42; Sections 309-1256D-122R-1 and 122R-2), a hyaloclastite with abundant sulfide minerals. This volcanic breccia consists of aphyric basalt clasts with subangular to subrounded shapes (ranging in size from 2 mm to 7 cm), volcanic glass clasts, glassy shards, and subrounded to rounded altered glassy shards (Fig. F42). Basalt clasts exhibit the textural features of sheet flows, such as spherulitic to variolitic textures (see Fig. F42) and lava flow-related folding. Clasts are embedded in a scarce fine-grained clay matrix cemented by sulfides, carbonate, and silica. In interval 309-1256D-122R-1, 52–125 cm, the mineralized volcanic breccia grades from an almost pure hyaloclastite with rare sulfides to mineralized hyaloclastite. There is a concomitant increase in basalt clasts and matrix volume with respect to glassy clasts (Fig. F43).

Numerous chilled margins were recovered in cores from the transition zone and the sheeted intrusives, and these contacts are increasingly common with depth. From ~1004 mbsf, where such contacts are subvertical, they are interpreted as dike contacts. Chilled margins range from lobate and interfingered to sharp. In the sheeted intrusives, the occurrence of dike chilled margins becomes very common. Many of these dike chilled margins are associated with, or highly disrupted by, diffuse veining and brecciation (Fig. F44). Multiple dikes and banded dikes also occur. The true dip of the chilled margins ranges from 50° to 90° with a mode at ~70°–75° (Fig. F45). Preliminary interpretation of FMS and UBI images indicates that these features dip steeply to the northeast. The sheeted intrusives are also characterized by the first notable occurrence of systematic conjugate veins. From 1090.7 mbsf (Unit 1256D-45) downhole, all the structural features, except shear veins, are common and more abundant. Shear veins are present only in the uppermost portion of the massive basalt (Unit 1256D-44).

The primary goal of paleomagnetic studies is to assess the roles of different rock types that make up the upper oceanic crust in generating marine magnetic anomalies. Magnetic remanence data were collected before and after progressive alternating-field or thermal demagnetization.

Because of the strong drilling overprint and uncertainty about how completely the overprint has been removed by demagnetization, we cannot yet make strong statements based on the paleomagnetic results from Expedition 309. Rocks from the lower parts of Expedition 309 recovery have higher coercivities and a pronounced increase in the apparent quality of data that occurs over the interval 970–1030 mbsf. Because of the equatorial paleolatitude of the site, polarity remains ambiguous until absolute declinations can be obtained based on orienting pieces relative to the downhole logging images of the borehole wall. The component of the drilling overprint that may remain would affect inclination more than declination, so for samples for which data analysis suggests that much of the drilling overprint has been removed (e.g., Fig. F46), generally from >1000 mbsf, declination values will be reliable enough to determine polarity in oriented pieces. If the number of oriented pieces is small but the polarity pattern is clear from those pieces or from measurements of the downhole magnetic field, the declinations from the more stable unoriented samples should be adequate for orienting pieces for structural purposes once the polarity has been determined separately.

The generally positive inclinations are not what is expected for the low paleolatitude. The most obvious possibility is that a significant portion of the drilling overprint remains on nearly all of the samples. A potential alternative is that there is a pervasive present-field overprint. Another alternative, tectonic tilting, cannot be entirely discounted. However, any tilting must predate deposition of the ponded lava flows at the top of the section, and the nearly north–south original strike of the ridge axis does not provide a favorable orientation for changing inclination as a response to slip on ridge-parallel faults.