SCIENTIFIC RESULTS

We drilled two holes at Site 1185 (Table 1), the first of which was a pilot hole to determine the depth to basement and the length of casing necessary to attach to a reentry cone at the second hole. In Hole 1185A, we started coring sediments at 250.6 mbsf, contacted basaltic basement at 308.5 mbsf, and cored basement rocks to 328.7 mbsf. In Hole 1185B (20 m west of Hole 1185A), we started coring at 308.0 mbsf, contacted basaltic basement at 309.5 mbsf, and cored basement to 526.1 mbsf. We recovered 14.1 m of the 57.9-m sedimentary section cored in Hole 1185A (Fig. 25). The dominant lithology is middle to upper Eocene radiolarian nannofossil chalk, which gradually darkens downward from white to light gray. Maximum bulk density is 1.6 g/cm³. The most distinctive features of the chalk are its abundant siliceous microfossils and a highly variable abundance (in places, a virtual absence) of planktonic foraminifers, suggesting that deposition was often below the foraminifer lysocline. Foraminifers preserved in the chalk indicate a middle to upper Eocene unconformity that may be associated with a rise in the CCD. Eocene siliceous chalk similar to that at Site 1185 has been found at other DSDP and ODP sites on the Ontong Java Plateau.

The basement sequence consists of pillow basalt and massive basalt flows. We did not recover the sediment-basalt contact in either Hole 1185A or 1185B, but rare intercalations of limestone are present between lava flows and in fissures within the basalt. The limestone is composed of micritic calcite with very rare and poorly preserved nannofossils, foraminifers, and radiolarians that provide only rough age control but reveal that limestones of two ages are present. Extremely rare nannofossils in limestone within the upper 15 m of basement indicate a latest Cenomanian to Albian (possibly late Albian) age, whereas recrystallized planktonic foraminifers in thermally metamorphosed limestone 126 m below the top of basement suggest a late Aptian age. This difference in age between the upper and lower parts of the basement section corresponds to differences in basalt petrography, composition, and alteration. The entire basement sequence exhibits normal magnetic polarity, compatible with this range of ages, and paleolatitudes derived from paleoinclination data agree well with those for Site 1183. The ~50-m.y. hiatus between the lower sediments and basement suggests that this site may have been below the CCD much of the time from the Late Cretaceous to middle Eocene.

In Hole 1185A, we cored 20.2 m of pillow basalt (308.5 to 328.7 mbsf; 55% average recovery). The pillows have glassy rims, spherulitic chilled margins, and fine-grained interiors. We divided the basalt into five units (Fig. 26) on the basis of apparent limestone interbeds, some of which may be only interpillow fill. The 216.6 m of basaltic basement penetrated in Hole 1185B (309.5 to 526.1 mbsf; 42% average recovery) exceed the previous maximum on the plateau of 149 m of lava flows penetrated at Leg 130 Site 807 (Kroenke, Berger, Janecek, et al., 1991). We divided the 216.6-m basement section of Hole 1185B into 12 units ranging in thickness from 1 to 65 m. Units 1, 3, 4 and 6-9 were identified as pillow basalt on the basis of glassy rims and grain size variations, and Units 2 and 5 are more massive lava flows with pillowed tops and bases. Units 1-9 are separated by thick (as much as 70 cm) intervals of hyaloclastite breccia composed of pillow-rim fragments cemented by carbonate and clay. Units 10-12 are massive flows. The flow tops of Units 10 and 12 are marked by carbonate- and clay-cemented breccia; the top of Unit 11 was not recovered but was inferred from the presence of vesicles, a pronounced change in alteration, and a marked increase in drilling rate over an interval of about 3 m.

The basalt in Hole 1185A and in Units 1-9 of Hole 1185B is sparsely to moderately olivine phyric and generally highly veined. Olivine, the only common phenocryst phase, varies from fresh, in the glassy and aphanitic rims of pillows, to completely replaced by smectite, Fe oxyhydroxide, or calcite. Tiny octahedral crystals of chrome spinel are present, often as inclusions in the olivine phenocrysts (Fig. 27). Aphanitic pillow margins display a prominent spherulitic texture (Fig. 28) that grades into variolitic texture in fine-grained pillow interiors. The massive units also have variolitic texture and are less heavily veined. Units 10-12 in Hole 1185B contain small, sparse phenocrysts of plagioclase and clinopyroxene in addition to olivine. These rocks are similar in appearance to the basalt flows at Sites 1183 and 1186 and, like them, contain plagioclase-rich xenoliths. Shipboard ICP-AES analyses show that Units 10-12 are also very similar in composition to basalt at Sites 1183 and 1186 (Fig. 10, Fig. 11, Fig. 12, Fig. 13). For example, all have TiO₂ 1.1 wt%, Cr200 ppm, and Zr 60 ppm and appear to belong to the widespread Kwaimbaita magma type. In contrast, the overlying basalt flows (those in Hole 1185A, and Units 1-9 in Hole 1185B) have the lowest concentrations of incompatible elements (TiO₂ ; 0.7 wt%; Zr ; 38 ppm; Fig. 12) yet recorded in basalt from the Ontong Java Plateau. Furthermore, Units 1-8 are composed of the most primitive, magnesium-rich basalt (MgO 8-10 wt%; Cr 460 ppm) yet found on the plateau (Fig. 11, Fig. 13). This combination of elemental characteristics appears to indicate that the magmas formed by even higher total fractions of partial melting than other Ontong Java Plateau basalt (see "Background" section above).

Bulk densities are higher (>2.4 g/cm³) in Units 2, 5, 10, and 11 than in Units 3 and 6-9 (<2.3 g/cm³). Both grain and bulk density decrease downhole in Units 4-9, corresponding to a change from massive to highly veined pillow basalt, and bulk density increases with the change from veined to massive basalt from Unit 9 to Units 10-12. P-wave velocities are generally >5000 m/s in the dense basalts of Units 2 and 10-12 and generally <5000 m/s in the veined basalts of Units 3 and 6-9.

Seawater-derived fluids have interacted pervasively at low temperatures with the basaltic basement, and we can divide the basement section into two groups of flows with different alteration characteristics. One group consists of all the basement units of Hole 1185A and Units 1-9 of Hole 1185B. Alteration in these units occurred under highly oxidizing conditions and with high water:rock ratios, as indicated by light and dark yellow-brown colors near glassy and aphanitic pillow margins; these colors fade to gray-brown and dark gray in coarser-grained pillow interiors. The yellow-brown colors are a result of the complete replacement of olivine and pervasive alteration of groundmass by smectite (saponite and nontronite) and Fe oxyhydroxide. Although olivine phenocrysts are commonly completely replaced, rare unaltered olivine is present in aphanitic, dark gray to black areas interpreted as pillow rims. These characteristics are significantly different from the style of alteration in basalt at Site 1183.

The other alteration type comprises Units 10-12 at Hole 1185B. The top of Unit 10 marks a dramatic change in alteration character, and the brecciated top of this unit consists of angular basalt fragments that are the most pervasively altered yet observed on Leg 192. Such severe alteration is likely to be the result of exposure of very permeable basaltic seafloor to bottom seawater for an extended period of time (several million years). Groundmass alteration in Units 10-12 is characterized by pervasive celadonite. Unaltered glass is much rarer in these units than in Units 1-9, but this may be a consequence of the greater flow thickness rather than the level of glass alteration. Broad (centimeter scale) green-gray halos developed near veins; wider reduction fronts consisting of scattered pyrite grains in the groundmass extend a few millimeters to a few centimeters beyond these halos. Smaller (millimeter scale) olive halos similar to those at Site 1183 also are developed near veins, both within green-gray halos and where green-gray halos are absent.

Veins throughout the basement sequence predominantly contain calcite, zeolites (phillipsite), smectite, Fe oxyhydroxide, and rare celadonite and pyrite. Some veins were probably produced by sediment filling open fractures in the basalt; these veins are a few centimeters to <5 mm wide, are filled with pink carbonate and Fe oxyhydroxide, and contain recrystallized foraminifers. They are present both near the boundaries and within the interiors of basaltic units.

The major results of drilling at Site 1185 are summarized as follows:

1. The 60 m of white to light-gray radiolarian nannofossil chalk immediately above basement are of middle to upper Eocene age. The most distinctive feature is the abundance of siliceous microfossils and, in places, a nearly complete absence of planktonic foraminifers, suggesting that many of the sediments were deposited below the foraminifer lysocline. Recovered foraminifers indicate a middle-to-upper Eocene unconformity that may be associated with a rise in the CCD.

2. A hiatus on the order of 50 m.y. separates the sedimentary succession from the basaltic basement.

3. Extremely rare calcareous nannofossils recovered from limestone between basalt flows in the upper 15 m of basement indicate an Albian to latest Cenomanian age. Recrystallized planktonic foraminifers found within thermally altered limestone 126 m below the sediment-basement contact in Hole 1185B tentatively suggest a late Aptian age.

4. The basaltic lava flows in Hole 1185B can be divided into two major groups. The upper group appears to have been erupted sometime between the latest Cenomanian and Albian; the lower group is Aptian in age.

5. The two basalt groups have distinct chemical compositions. The older group is very similar to the basalt at Sites 1183 and 1186 and appears to represent the abundant Kwaimbaita magma type. The younger group has much lower concentrations of incompatible elements and includes the most primitive (magnesian) basalt yet found on the plateau.

6. Different styles of alteration in the two groups of basement lava flows suggest that the older group was in contact with seawater for a long period before the eruption of the younger group. This observation is consistent with the paleontological evidence that suggests a >15-m.y. hiatus between the two groups.