FIGURE CAPTIONS

Figure 2.Satellite-derived free-air gravity field of the Ontong Java Plateau region (after Sandwell and Smith, 1997). Solid stars = sites penetrating lava sections. Open star = Site 1184, where a volcaniclastic sequence was recovered. Solid circles = previous ODP and DSDP drill sites that reached basement. Open circles = Site 288, which did not reach basement but bottomed in Aptian limestone, and Site OJ-7, which was proposed for Leg 192 but not drilled (see text). Black lines indicate multichannel seismic surveys on the plateau: Hakuho Maru KH98-1 Leg 2 (1998) and Maurice Ewing EW95-11 (1995). White lines indicate single-channel seismic surveys: Glomar Challenger GC07 (1969), GC30 (1973), and GC89 (1983); Thomas Washington TW88-11 (1988); and JOIDES Resolution JR130 (1990). The bathymetric contour interval is 1000 m.

Figure 3. Summary of pre-Leg 192 age data for Ontong Java Plateau basement basalt, later lava flows and intrusions on Malaita and San Cristobal, and ash or glass-shard-rich layers within the sedimentary section at Sites 288 and 289 (Tejada et al., 2000). Ages derived from biostratigraphy are shown with error bars; the others are ⁴⁰Ar-³⁹Ar plateau or, for the Malaitan alnöites, U-Pb zircon ages.

Figure 4. Total alkalis vs. silica diagram (after Le Bas, 1986) for basement rocks recovered at DSDP Site 289 and ODP Sites 803 and 807 on the Ontong Java Plateau. Data are from Stoeser (1975), Mahoney et al. (1993), and Tejada et al. (1996, 2000). The broken line separates Hawaiian alkalic and tholeiitic basalts (Macdonald and Katsura, 1964).

Figure 5. Primitive mantle-normalized incompatible-element averages for basalts of Malaita, Santa Isabel (plus Ramos), and ODP Sites 803 and 807. Shaded fields indicate range of values for central Malaita. From Tejada et al. (2000).

Figure 6. Initial_Nd(t) vs. (²⁰⁶Pb/^204Pb)t data for basement lava flows of Santa Isabel and Malaita. Heavily outlined fields encompass data for ODP Site 807 basement Unit A and Sites 803 and 807, Units C-G. Data for the Site 289 basalt lie in the latter field. Data fields for the Manihiki Plateau (for dredged and DSDP Site 317 lava flows), Nauru Basin, Pacific midocean-ridge basalts (MORB), the Koolau and Kilauea volcanoes of Hawaii, and the Mangaia Group islands of the South Pacific are shown by light outlines (from Tejada et al., 2000).

Figure 7. Hakuho Maru KH98-1 Leg 2, line 404, multichannel seismic reflection profile across Site 1183 (see Fig. 2, for location). Vertical exaggeration = ~4.2 at seafloor. CDP = common depth point, UTC = Universal Time Coordinated.

Figure 8. Site 1183 log showing core recovery, lithostratigraphic divisions, schematic lithology, color reflectance, magnetic susceptibility, and carbonate content.

Figure 9. Lithologic log of the basement units at Site 1183 showing the distribution of plagioclase rich xenoliths, glassy pillow margins, and phenocrysts. Basalt units with the suffix B were delineated by the presence of thin sedimentary interbeds, which are designated as Subunit A (not shown).

Figure 10. Total alkalis vs. silica diagram (after Le Bas, 1986) for basement rocks recovered at all Leg 192 sites. The broken line separates Hawaiian alkalic and tholeiitic basalts (Macdonald and Katsura, 1964).

Figure 11. TiO₂ vs. Mg# for basement rocks recovered at all Leg 192 sites and ODP Sites 803 and 807. The fields for basaltic lava flows of the >2.7-km-thick Kwaimbaita Formation and the overlying ~750-m-thick Singgalo Formation on Malaita (Tejada et al., 1996, 2000) are shown for comparison. Data for Sites 803 and 807 are from Mahoney et al. (1993).

Figure 12. Zr vs. TiO₂ for basement rocks recovered at all Leg 192 sites, DSDP Site 289, and ODP Sites 803 and 807. Data for Sites 803 and 807 are from Mahoney et al. (1993). The fields for basaltic lava flows of the >2.7-km-thick Kwaimbaita Formation and the overlying ~750-m-thick Singgalo Formation on Malaita (Tejada et al., 1996, 2000) are shown for comparison.

Figure 13. Cr vs. TiO₂ for basement rocks recovered at all Leg 192 sites, DSDP Site 289, and ODP Sites 803 and 807. Data for Sites 803 and 807 are from Mahoney et al. (1993). The fields for basaltic lava flows of the >2.7-km-thick Kwaimbaita Formation and the overlying ~750-m-thick Singgalo Formation on Malaita (Tejada et al., 1996, 2000) are shown for comparison.

Figure 14. Close-up photograph showing a 1.5 cm x 3 cm plagioclase-rich xenolith on the outer surface of the core (interval 192-1183A-59R-2 [Piece 10A, 112-119 cm]) .

Figure 15. Paleolatitude, calculated from paleomagnetic inclination, for cores from Hole 1183A. Data are plotted against biostratigraphic age.

Figure 16. Seismic reflection record for Site 1184. Hakuho Maru KH98-1 Leg 2, line 101, multichannel seismic reflection profile across Site 1184 (see Fig. 2, for location). Vertical exaggeration = ~4.2 at seafloor. CDP = common depth point, UTC = Universal Time Coordinated.

Figure 17. Summary of lithologic characteristics in the subunits of Unit II at Site 1184, based on the facies defined in Table 2. "Wood" indicates approximate intervals in which pieces of wood were found. The uppermost interval is represented by two pieces, 1-2 cm long and <2 mm thick. At the other three wood-bearing levels, pieces of wood are as long as 6 cm. In the chromaticity plot, a = the variation between green (negative) and red (positive), and b = the variation between blue (negative) and yellow (positive) (Blum, 1997).

Figure 18. Close-up photograph of wood fragments in lithic vitric tuff near the base of Subunit IIE (interval 192-1184A-45R-7, 48-65 cm).

Figure 19. Photomicrograph showing a wide range of lithic and vitric clast types present in a lithic vitric tuff from Subunit IID (Sample 192-1184A-31R-7, 40-43 cm). The clasts include glass shards (light brown), basalt, and tachylite (black). The glass is completely altered to smectite, and the matrix is composed mainly of zeolite (field of view = 5.5 mm wide; plane-polarized light).

Figure 20. Close-up photograph of accretionary lapilli, both whole (round) and broken fragments, in gray lithic vitric tuff, Subunit IIA (interval 192-1184A-14R-1, 58-66 cm).

Figure 21. Close-up photograph of coarse (¾2 cm) angular clasts in lapilli tuff, Subunit IIB (interval 192-1184A-21R-6, 60-80 cm). The large pale clast is diabase.

Figure 22. Schematic interpretation of the eruptive setting of the volcaniclastic rocks at Site 1184. These reworked pyroclastic deposits probably formed on top of a large seamount as it grew to within ~200 m of the sea surface. The presence of basalt at depth is conjectural. A = the presence of blocky, nonvesicular glass shards, suggests fragmentation of rapidly quenched magma in hydromagmatic eruptions under shallow water; abundant tachylite clasts in parts of the succession suggest that this process occurred in an environment that was at times subaerial. B = accretionary and armored lapilli form in the atmosphere, in steam-rich columns of volcanic ash. C = the absence of blocks, bombs, or lapilli >20 mm suggests that the primary pyroclastic deposits did not form close to the eruption center(s). D = wood fragments found at the bases of four of the five subunits indicate proximity to land. E = deposition or redeposition of the volcaniclastic material in a marine setting is indicated by the presence of nannofossils throughout the unit. F = redeposition by turbidity currents is suggested by the presence of rip-up clasts and broken accretionary lapilli and by the absence of the layering that would be expected from material settling through water.

Figure 23. Photomicrograph of lithic vitric tuff from Subunit IIE (Sample 192-1184A-42R-1, 147-150 cm). Rims of glass shards are altered to smectite (field of view = 2.8 mm wide; plane polarized light).

Figure 24. Hakuho Maru KH98-1 Leg 2, line 401, multichannel seismic reflection profile across Site 1185 (see Fig. 2, for location). Vertical exaggeration = ~4.2 at seafloor. CDP = common depth point, UTC = Universal Time Coordinated.

Figure 25. Site 1185 log showing core recovery, lithostratigraphic divisions, schematic lithology, color reflectance, carbonate content, and components of sediments. In the color column, L = the total reflectance and indicates lighter shades to the right; a and b quantify the hue as chromaticity, where a = variation between green (negative) and red (positive) and b = variation between blue (negative) and yellow (positive) (Blum, 1997). Component percentages are from observation of smear slides and thin sections.

Figure 26. Lithologic log of basement units at Site 1185. The upper and lower groups of units in Hole 1185B are distinct in their petrography, composition, state of alteration, and biostratigraphic age (see text for details).

Figure 27. Photomicrograph of olivine phenocrysts and tiny octahedral crystals of chrome spinel in a quenched pillow margin (Sample 192-1185A-8R-1 [Piece 2, 15-18 cm]). The groundmass consists of glass with dendritic microlites (field of view = 0.28 mm wide; plane-polarized light).

Figure 28. Close-up photograph showing spherulitic texture in an aphanitic pillow margin, basement Unit 1 (interval 192-1185B-3R-1, 54-63 cm). The spherulites are highlighted by alteration.

Figure 29. Hakuho Maru KH98-1 Leg 2, line 403, multichannel seismic reflection profile across Site 1186 (see Fig. 2). Vertical exaggeration = ~4.2 at seafloor. CDP = common depth point, UTC = Universal Time Coordinated.

Figure 30. Lithologic log and selected properties of sediments at Site 1186. There was no coring from 0 to 697.4 mbsf. In the color column, L = the total reflectance and indicates lighter shades to the right; a and b quantify the hue as chromaticity, where a = the variation between green (negative) and red (positive) and b = the variation between blue (negative) and yellow (positive) (Blum, 1997). Component percentages are from observation of smear slides and thin sections. Higher values of magnetic susceptibility generally correlate with bands rich in volcanic ash (e.g., Core 192-1186A-13R) or with claystone (e.g., Core 192-1186A-26R and base of sedimentary sequence in Core 192-1186A-30R). Natural gamma-radiation intensity is from the Formation MicroScanner tool run; medium-penetration resistivity and caliper of borehole diameter are from the geophysical tool run.

Figure 31. Close-up photograph of limestone conglomerate between lava flows in Hole 1186A (interval 192-1186A-32R-3, 65-93 cm).

Figure 32. Lithologic log of basement units at Site 1186 showing the distribution of plagioclase rich xenoliths, glassy pillow rims, and vesicles. Dashed line = unit boundary. The location of the Unit 3/Unit 4 boundary is uncertain because of poor recovery. However, formation microscanner logging tool images suggest that the top of Unit 4 consists of ~9 m of pillow basalt.

Figure 33. Photomicrograph of plagioclase laths and sprays of clinopyroxene crystals in basalt from the interior of massive flow Unit 2 (see Fig. 32) (Sample 192-1186A-34R-2, 143-146 cm; field of view = 2.8 mm wide; crossed polars).

Figure 34. Thomas Washington 88-11 single-channel seismic-reflection profile across Site 1187
(see Fig. 2, for location). Vertical exaggeration = ~6.67 at seafloor. CDP = common depth point, UTC = Universal Time Coordinated.

Figure 35. Lithologic log of basement units at Site 1187 showing the thickness of individual cooling units defined by the presence of glassy or aphanitic margins. Cooling units <1.5 m thick are probably pillows, but the four units that are 2-3 m thick could be thin massive flows.

Figure 36. Stratigraphic sections drilled during Leg 192 and at the three previous DSDP and ODP Ontong Java Plateau basement sites. Seven sites are arranged on a transect from the crest of the main plateau (Site 1183) eastward to the plateau rim (Site 1185) and then north and northwestward to Site 807 on the northern flank. Site 1184 lies off the transect, 590 km to the southeast of Site 1185. Basement ages for the previously drilled sites are from ⁴⁰Ar-³⁹Ar dating of basalt (Mahoney et al., 1993). For the Leg 192 sites, basement ages are estimated from biostratigraphic evidence.

Figure 37. Basement sections drilled during Leg 192 and at the three previous DSDP and ODP Ontong Java Plateau basement sites (as in Fig. 36), with water depths corrected for sediment loading. The corrected basement depth (D_c) is obtained from the equation of Crough (1983): D_c = d_w + t_s (_s - _m)/(_w - _m), in which d_w is water depth in meters, t_s is sediment thickness in meters, _s is average sediment density (1.90 g/cm³), _m is upper mantle density (3.22 g/cm³), and _w is seawater density (1.03 g/cm³).