A fractured dolerite unit (Unit VIII) was first encountered within Section 180-1118A-70R-3, at 869 mbsf, below a packstone/grainstone unit with scattered rounded basaltic clasts and fragments from a lagoonal environment and a paraconglomerate containing basalt and dolerite clasts set in a calcareous matrix (Units VI and VII; refer to "Lithostratigraphic Unit VI" and "Lithostratigraphic Unit VII"). The FMS images show that at least the upper part of Unit VIII is also a conglomerate (see "Log Unit L9"), although it was originally interpreted as a fractured dolerite with cracks infilled with sedimentary material. Clasts in the conglomerate reach a large size; several unbroken pieces of dolerite were recovered that were tens of centimeters in length. Fifty-eight meters of this conglomerate were cored from 869 to 920 mbsf of which 13.6 m (23.4%) were recovered (Fig. F34). The rocks range from slightly altered, unbrecciated dolerite clasts (Fig. F35) to pervasively altered breccia rocks. The recovered breccia rocks contain angular dolerite clasts within a dark brown to black, clay-sized, iron oxide-rich matrix thought to be derived from extensive weathering of dolerite (Fig. F36). Among the clasts recovered, brecciation occurs in zones (see "FMS Images" for spatial attitudes and trends of brecciation).
The clasts within the conglomerate have varying degrees of subaerial iron oxide weathering and brecciation that create a dark brown, dolerite- derived matrix. Angular clasts of dolerite were recovered, some within this brown matrix, others without matrix (see "Domain III: Dolerite"). Many of the angular clasts within the matrix material have onion-skin iron oxide weathering, which indicates weathering of the clasts before incorporation into the conglomerate. One particular example shows a clast with a network of iron oxide weathering separating angular zones within a dolerite (see "Structural Geology"). This most likely represents a partially brecciated and weathered dolerite clast, intermediate between the two types observed. The close association of iron oxide weathering to fractures suggests that fluid flowing within these veins is a likely agent of weathering. Thin-section analysis shows veins filled with calcite and zeolite with iron oxide weathering on the inside, suggesting that weathering may be associated with the fluids responsible for calcite precipitation (Fig. F37). Many slickensided surfaces can be seen on chips of this clay-sized, brown matrix material, which suggests that the unit has undergone some degree of faulting (refer to "Structural Geology").
Apart from the degree of weathering and brecciation, there is no mineralogical difference between fractured and unfractured dolerite rocks within this unit. The rocks are generally slightly to moderately altered and fine- to medium-grained with a granular to ophitic texture. Mineralogy includes largely fresh clinopyroxene and partially sausseritized plagioclase, often intergrown along with dendritic magnetite, iron oxide, and less commonly, pyrite. Chlorite occasionally replaces clinopyroxene (Fig. F38) but also occurs interstitially, possibly replacing interstitial glass. Clusters of greenish layer silicates ("iddingsite") represent pseudomorphed olivine at least in some cases, but this never makes up more than ~5% of the rock. Grain size varies moderately throughout zones within cores, but no systematic change can be noted with increasing depth. Mineralogy of the dolerite at this site is similar to that at Sites 1109, 1114, and 1117, although there are textural and grain-size variations, and the small amount of gabbro recovered at Site 1117 has appreciable amounts of quartz, which was not seen except in occasional small amounts in the gabbro rocks from the other sites. Two pegmatitic zones occur at intervals 180-1118A-74R-2 (Piece 5, 120-149 cm; 900.24-900.53 mbsf) and 76R-2 (Pieces 11 and 12, 77-89 cm; 917.71-917.83 mbsf), which have dendritic magnetite crystals even better developed than those at Site 1117 (Fig. F39). Very fine grained (basaltic) clasts, similar to the chilled margin described in the Site 1109 dolerite are also present in interval 180-1118A-74R-1 (Piece 1, 124-126 cm; Fig. F40). There is no significant difference in primary mineralogy between the pegmatite and the dolerite, and no sharp contact can be observed between the rock types.
X-ray fluorescence (XRF) analyses are reported in Tables T4 and T5. One sample is from a basaltic pebble in the overlying Unit VII. Compositions generally resemble those of other dolerites and gabbros from Leg 180, but there is significantly more variation than reported at Site 1109, although this is largely due to the presence of a pegmatitic patch (analysis 5 in Table T4). Variation is shown in Figure F41 where some elements are plotted against the iron/magnesium ratio, showing a general pattern to be expected from igneous fractionation: alkalis and Ce rise, MgO, Ni, and Cr fall, although Ba and SiO2 are equivocal. The latter, in particular, generally show a slight rise with increasing differentiation, although the pegmatite is anomalous with high Fe*/MgO and a relatively low silica content. Possibly, processes other than crystal fractionation that also involve volatiles can account for this anomaly.
The precise nature of the dolerite body is not entirely agreed upon (see, e.g., "Domain III: Dolerite") and the chemistry may shed some light on the problem. It is clear from the data in Tables T4 and T5 that dolerite compositions are not randomly distributed. The most primitive compositions are at the top and bottom, the more evolved compositions toward the middle. This is consistent with the samples coming from a single dolerite sheet rather than clasts in a conglomerate, but analyses would need to be more numerous to give a definitive answer.
Our preliminary interpretation is that this is a conglomerate body consisting largely of dolerite and subordinate basaltic clasts in a matrix composed largely of weathered dolerite-derived material, most of which was probably not recovered (see "Downhole Measurements"). The latter type of clast may be samples of the dolerite chilled margin, although the relatively large amount of fine-grained basalt in these overlying units suggest that there were lava flows in the hinterland as well as dolerite.
The dolerite clasts have been fractured and subjected to subaerial alteration, most likely aided by fluid circulation within fractures and prior to incorporation in the conglomerate, although this may have taken place along cracks and fissures in a dolerite body. The interpretation of the overall nature of the unit relies heavily on FMS logging data, which extends down to Core 180-1118A-72R and suggests that the dolerite clasts recovered from Cores 70R to 72R are part of a large conglomerate unit from which only minimal matrix material has been recovered. However, no FMS data is available below Core 72R, and the dolerite recovered below may either be part of the same conglomerate or could be in situ brecciated dolerite. The alternative interpretation (i.e., it is a fractured dolerite body) cannot be ruled out and is, in fact, supported to some extent by the chemical data.
The mineralogy and chemistry suggest that dolerite clasts within this breccia are similar to those of the dolerite unit encountered at Site 1109. Further comparison of petrological and chemical evidence, such as mineral chemistry, trace element and isotope geochemistry, and radiometric dating between Sites 1109 and 1118 from the hanging wall and Sites 1114 and 1117 from the footwall of this fault system, will be required to clarify the relationship between rocks from each site and to determine the tectonic settings in which these rocks were formed.