Igneous rocks were initially represented by a conglomerate containing generally highly rounded clasts of glassy basalt and dolerite in a clayey silt matrix with volcaniclastic material or sparry calcite (Fig. F37). This conglomerate of lithostratigraphic Unit X, which was recognized in Sections 180-1109D-42R-CC through 45R-3 (see "Lithostratigraphic Unit X") and is thought to extend from 737.2 to 772.9 mbsf, is underlain by the dolerite (lithostratigraphic Unit XI), which was first encountered in Section 45R-3 and continued downward to the base of Hole 1109D at ~802 mbsf. This is a compound intrusion, but otherwise extremely homogeneous.
Rounded dolerite clasts from lithostratigraphic Unit X have 1- to 2-mm concentric weathering rings that are interpreted as Fe oxidation penetrating into the clast. This weathering suggests transport and deposition of the clasts into an oxidizing environment. Some clasts near the bottom of the conglomerate unit exhibit red alteration products rimmed by green alteration products (Fig. F38). This may represent a change from an oxidizing to a reducing environment. The basaltic clasts are variolitic, with quench crystals of plagioclase and microphenocrysts of olivine (Fig. F39), and were probably extruded under water. Alternatively, these glassy rocks may be from the upper contact of the intrusive body although, in this case, we would expect the glass to be less extensive, giving way to microcrystalline material after a centimeter or so.
The topmost samples of dolerite (lithostratigraphic Unit X) have a fine-grained granular texture (Fig. F40), suggesting that they come from close to a chilled margin, but this passes rapidly downward into the typical ophitic texture almost ubiquitous in such rocks. Poikilitic plates of augite, ~1 mm in size, enclose numerous plagioclase laths with a length of about half this size (Fig. F41). A few percent of opaque iron-titanium oxides are present along with about 5% alteration products that appear to have replaced interstitial glass, although some may pseudomorph original olivine crystals (Fig. F42). No entirely unequivocal olivine pseudomorphs were identified.
The massive dolerite (Unit XI), of which 34.93 m was recovered, is extremely homogeneous in terms of mineralogy and texture and, apart from fractures and veins (see "Structural Geology"), contains no features of note, except toward the bottom of the hole. In Section 180-1109D-51R-3 at about 100 cm (799 mbsf), the grain size begins to decrease, suggesting the approach of a lower contact. In Section 180-1109D-51R-4 (at about 20 cm) there is a chilled margin with thin veins of glassy material intruding upward into the fine-grained base of the upper component. This is clear evidence for the intrusion of a second component of the sill, which, in turn, becomes coarser downward to the bottom of the hole. However, material from the lowermost recovery does not attain the ophitic texture seen in the most slowly cooled parts of the dolerite higher up.
Along the contact between the glassy chilled material and the crystalline dolerite, small "pillows" of the glassy material are enveloped in the dolerite, suggesting that the latter was still plastic at the time of the new intrusion (Fig. F43).
Fractures include both joints and faults (Figs. F44, F45). Pyrite, greigite (XRD identification), tetronatrolite (XRD), cristobalite (XRD), and calcite (identified in thin section) fill these fractures, which are described more comprehensively in "Structural Geology."
The X-ray fluorescence (XRF) analyses (Tables T4, T5) of the dolerite confirm its general homogeneity, but show that there is a small systematic variation with depth. The data is plotted against depth in Figure F46A and F46B. In general, with decreasing depth (i.e., height in the body) MgO falls while Fe2O3 rises (seen most clearly in the Fe2O3/MgO ratio). This is the pattern to be expected from fractionation of the original magma after emplacement and is seen in countless examples of sill-like bodies worldwide. The upward-increasing pattern is also shown by other incompatible elements, such as Na2O (marginally), TiO2, Zr, and Ce (although Zr and Ce both show anomalously low values around the middle of the profile, perhaps not significant because they do not coincide). Ni, like Co, also shows a decrease with height, typical of compatible elements. The more equivocal is Cr, with high values in the middle of the section. At the uppermost end of the range for basaltic rocks, SiO2 also shows low values around the middle of the profile, reinforcing the indication that there is indeed a difference here.
The lowermost analysis appears to be from a separate magma injection, and the uppermost part of the dolerite is lacking. The former may explain the different position of the lowermost sample in some plots (e.g., Fe2O3, Na2O, and TiO2) and the fact that these variations are not in concert (Fe2O3 is high, whereas Na2O and TiO2 are low, in spite of the fact that they are all to some degree incompatible) suggests a slightly different magma batch
Ideally, it should be possible to see from the chemical compositions the tectonic affinities of these rocks (Wilson, 1989), because volcanics from collisional environments tend to be enriched in the large ion lithophile elements (K, Rb, Ba, and Sr) and depleted in high field strength elements (Ti, Zr, etc.). Figures F47, F48, and F49 show the position of this dolerite in the discriminatory diagrams of Pearce et al. (1977), Pearce and Cann (1973), and Pearce (1980), which have been widely used for this type of discrimination. In the TiO2-K2O-P2O5 diagram, the dolerite plots clearly in the oceanic basalt field, whereas in the two remaining diagrams it is also in the field of oceanic basalts, although there is substantial overlap with the field of calc-alkali or island-arc tholeiite; therefore, no unequivocal distinction can be drawn. These diagrams do, however, clearly rule out any within-plate basalt affinities for this dolerite. A similar conclusion can be drawn from the Ti-Zr-Sr diagram of Pearce and Cann (not shown here), where the dolerite also plots in the ocean-floor basalt field. In comparison, data from nearby Woodlark Island (Ashley and Flood, 1981) is shown. Both the Hole 1109D dolerite and the data from the Woodlark low-K tholeiites plot in the fields of the low-K tholeiites and are quite distinct from the later high-K, or shoshonitic, rocks from this island. The latter are also included in Figures F46 and F47 to show the dramatic compositional difference.
Although the interpretation of this igneous body remains equivocal, our present view is that it is a sill. Arguments against a thick, ponded flow are mainly the lack of any vesicles (which are almost ubiquitous in flows) and also the lack of any flow-top breccia (which could, however, have been eroded away). We cannot, of course, rule out the possibility that we encountered a dike and have been following it down, as dikes are frequently tens of meters thick. The magnetic properties of the dolerite offers some help with this problem.
Regarding the time between the intrusion of the two dolerite components, the evidence outlined above suggests that the first component was still plastic at the time of the new intrusion and, hence, there was no significant time gap between them.
We picture an outcropping sill from which the uppermost part had already been removed. This subsided and became overlain by a deposit of rounded pebbles and boulders derived from the sill with some basalts and sandstones perhaps washed in from elsewhere (see "Lithostratigraphy"). These are the sediments of lithostratigraphic Unit X, probably derived from a hinterland that had previously been uplifted, as witnessed by the presence of submarine basalts among the clasts. Initial lithologic interpretation of cores just above the conglomerate unit suggests an anoxic, reducing environment. The interpreted transition in weathering from oxidation to reduction in dolerite clasts fits well with subsidence from a subaerial to a lagoonal setting. This scenario suggests that there may be a considerable time difference between the intrusion of the sill and the deposition of its overlying sediments, although this cannot be quantified at the present time.
Low-K basalts, resembling both mid-ocean ridge and arc tholeiite types, are widespread in the Papuan Peninsula-Milne Bay area (e.g., Smith, 1982). These are generally of Cretaceous to Paleocene age. Additional Paleocene volcanics include the Cape Vogel boninites, which are chemically quite distinct from the dolerite discussed here. High-K rocks also occur, although these are appreciably younger, extending up to subrecent. In the Dobu Island area there is a bimodal suite, including transitional basalts and comendites clearly related to the recent extensional tectonics in the Woodlark Basin. At the moment, we can only say that the Hole 1109 dolerite is related either to the low-K tholeiites or to the low-K calc-alkaline basalts. It has no affinities to the high-K rocks or to the basalts and rhyolites (comendites) of the Dobu Island area.
Ashley and Flood (1981) considered that the low-K tholeiites on Woodlark Island were of ocean-ridge affinity and were part of the Papuan ophiolite, which is of Paleocene age. Although it does not seem likely that the Hole 1109D dolerite is part of an ophiolite, radiometric dating will hopefully throw considerable light on this problem.