In Hole 1114A, 67 m of metamorphosed and brecciated igneous basic rocks was encountered below 286 m of Pliocene-Pleistocene claystone, siltstone, and sandstone (see "Lithostratigraphy"). Core recovery below the sediments averaged 13%. Pebbles of igneous material were first encountered in a tectonic breccia in Core 180-1114A-31R. Massive metamorphosed dolerite was recovered from 295.4 mbsf (Core 180-1114A-32R) downward to the base of the hole at 352.8 mbsf. This metadolerite has been subjected to intense brittle deformation and hydrothermal alteration. In general, the extent of both decreases in the dolerite toward the base of the hole. See "Structural Geology" for more details regarding the deformational history of these rocks.
Dolerite clasts were recovered in the tectonic breccia (Section 180-1114A-31R-1) overlying the massive dolerite. The clasts in the upper part of the breccia (e.g., interval 180-1114A-31R-1, 25-74 cm) exhibit quartz and epidote veins crosscutting the rocks and late-stage, massive clay alteration. The matrix itself is free of this generation of veins. A weakly developed layering is defined by the clasts elongated into the matrix. Secondary veins filled with calcite occur at the boundary between the clasts and the matrix. This tectonic breccia evolves downward in intervals 180-1114A-31R-1, 74-130 cm, and 31R-CC, 0-18 cm, into a very fine grained zone in which the structural relationship between the different generations of veins and the nature of the clasts (massive replacement with clay) are difficult to observe.
The rocks recovered at Site 1114 can be classified as one lithologic unit referred to as a brecciated metadolerite. Classification as metadolerite is based upon the patchy presence of fresh to slightly altered dolerite within the same lithologic unit and the presence of relict plagioclase and clinopyroxene in thin section, occurring as typical intergrowths of clinopyroxene and plagioclase, reflecting the original ophitic or granular texture. The intergrowths and texture, in addition to the chemistry (see "Chemistry"), are reminiscent of the fresh dolerite from Site 1109 (Figs. F16A, F16B, F17). These rocks are fine to medium grained. They originally contained 20%-40% plagioclase and 40%-60% clinopyroxene and had an ophitic to granular texture, similar to fresh dolerite from Site 1109 and throughout the metadolerite from the talus at Sites 1108 and 1110-1113. A similar chilled margin to that described in Hole 1109D (Section 180-1109D-51R-4) is observed in Section 180-1114A-36R-1. The actual contact between the glassy margin and the dolerite is brecciated, which indicates that the original contact (veins of glassy material intruding upward into the dolerite) was reactivated during the deformation (Fig. F18). The glass had a variolitic texture but is now completely replaced by very fine grained material, which could not be fully identified optically, but consists largely of an intergrowth of sericite, zoisite, and albite (saussurite) replacing plagioclase, and a minor amount of epidote replacing clinopyroxene. This penetrative hydrothermal alteration is associated with the brecciation of the dolerite. Indeed, epidote, chlorite, and quartz (related to the hydrothermal alteration) fill veins up to 10 mm wide, which crosscut the dolerite, developed contemporaneously to the brecciation because they crosscut the breccia or are fragmented into the breccia (Fig. F19). Subsequently, chlorite, clay, and trace amounts of pyrite, natrolite, and celadonite occurred in veins and in the vesicles. The clinopyroxene remained relatively stable during this stage; we observed this phase to be well preserved in the brecciated dolerite. In the most highly altered examples all minerals in the dolerite are replaced by patches exhibiting a plumose or spherulitic texture, consisting of a very fine intergrowth of fibrous tremolite, chlorite, and clay that radiate from a common center.
The rocks also show a series of subsequent crosscutting relationships between a series of intersecting fractures, which give the rocks an intense cataclastic texture (Fig. F19). Specific crystallization is not observed along these fractures, but the fracturing is intensified, the grain sizes decrease, and a matrix of very fine grained, dark brown clay occurs. In some areas the fractures reactivate previous veins (see "Structural Geology" for more details). In this case, quartz grains that fill the inner part of the veins are ductilely deformed (Fig. F20), which indicates the presence of fluids until the very latest stages of the metamorphic and structural evolution.
Only a minimal amount of analytical work was undertaken on board for Site 1114 because of the highly veined nature of the rocks. We restricted our analyses to two samples, and results are reported in Tables T4 and T5, along with the analytical results of a metadolerite pebble from Site 1108, an epidote-rich schist from Site 1111 (presumed to be an extremely altered dolerite), and the average of results for the dolerite cored at Site 1109 (the less altered material). These analyses have been arranged in descending order of volatile content to allow a clearer view of the effects of alteration. The dolerite from Site 1108 is moderately altered; it contains fresh clinopyroxene but the plagioclase is totally replaced by sericitic material. One of the samples from Site 1114 (Sample 180-1114A-36R-2, 40-44 cm) is massive and relatively unaltered with only restricted plagioclase sericitization visible in thin section. The other (Sample 180-1114A-36R-1, 70-72 cm) is highly sheared with a clear foliation visible in hand specimen and, again, the original feldspar has been entirely replaced.
Tables T4 and T5 do show some variability between the analyses, although not spectacular, and how much to ascribe to hydrothermal alteration and how much to ascribe to normal variations in the dolerite composition is not immediately apparent. There is, however, no consistent trend, and we conclude that chemical alteration is minimal.
The data in Tables T4 and T5 show the following consistent changes (i.e., the altered rocks are all changed in the same sense relative to the unaltered dolerite): Al2O3 and CaO loss; Fe2O3, MgO, Na2O, K2O, LOI, Ba, and Rb gain. However, we need to be critical of these results because they do not always correspond to previous results (e.g., those of Humphris et al., 1998), and they do not often follow the LOI values, which are thought to be the best index of alteration. Of the added components, only Fe2O3 and MgO correspond at all or closely with this criterion, and none of the lost components does. We also note that Humphris et al. (1998) did not find significant variations in Al2O3, alkali metals, or Ba in their samples. Ni appears to show a significant increase with alteration, as found by Humphris et al. (1998), but Sr increases, the reverse of findings by Humphris et al. (1998). We conclude from this discussion that the rocks may have gained some components and lost others through alteration, but the evidence is highly equivocal and, in general, the chemical changes have not been overwhelming in spite of the considerable alteration of mineral assemblages and mechanical deformation.
In conclusion, we emphasize that this study has not included the most altered rocks, which we have termed epidosite and is well represented in the talus at Sites 1110-1113 (see "Igneous and Metamorphic Petrology" in the "Sites 1110-1113" chapter) and only occasionally at Site 1114.
The differentiation state of these rocks is very similar (i.e., the FeO*/MgO ratio, where FeO* = total iron as FeO; see Table T4), making the above comparisons relatively simple. In fact, only one rock is significantly different from the others (i.e., Sample 180-1111A-16R-CC, 8-15 cm; Table T4). This rock is more evolved and also has a lower silica content, which suggests that it may follow the Skaergaard type differentiation of falling silica with increasing Fe/Mg ratio (Wager and Brown, 1968). Unfortunately, this is the most altered sample, and the lower silica content may be because of the alteration, as found by previous workers, but the data are insufficient to allow a distinction between these alternatives.
If an account is taken of the possible hydrothermal addition of K, it is likely that the Site 1114 rocks, just as at Site 1109, fall in the field of mid-ocean ridge basalts, although this overlaps with neighboring fields: island arc tholeiites and other low-K tholeiites; therefore, we can only conclude that they belong to the low-K tholeiites and may well be the same as other rocks of this type in the area, which have previously been interpreted as primitive island arc crust (Rogerson et al., 1993).
This unit was formerly dolerite, which was metamorphosed under low-grade greenschist facies conditions before the unroofing of the Moresby Seamount and was subjected to considerable deformation, fracturing, and permeation by fluids. It is likely that this material is the source of the dolerite recovered in the talus. Fluid circulation is inferred from the numerous veins filled with quartz, epidote, and pyrite. This fluid is responsible for the brecciation (fluid-assisted fracturing) of the rocks, although there is no evidence from the existing analyses for pervasive chemical alteration. However, such alteration might be found in the most altered rocks, which were not sampled because of the intensity of brecciation and veining.
The decreasing alteration with depth in the dolerite suggests that it is caused by the proximity of the upper levels to a fault zone along which fluids were channeled. Although it seems likely that these rocks were originally dolerites similar to those at Site 1109, it is not easy to visualize the precise geological nature of the terrain we have sampled. Dolerites generally occur in isolated bodies as sills or dikes, although they can also be found in sheeted dike complexes such as those of ophiolites like Troodos, Cyprus (Robertson and Xenophontos, 1993). Massive sill complexes are also known, although the grain size of these dolerites does not suggest very large intrusions because sills (and dikes) more than a few tens of meters thick tend to develop a gabbroic texture. These are, therefore, minor intrusions that collectively seem to make up a substantial areal extent. It is possible that at Site 1114 we have encountered a sheeted dike complex, perhaps related to the Papuan ophiolite (e.g., Davies and Jaques, 1984). In other areas the ophiolite is represented by the Lokanu Volcanics and equivalents (such as the Loluai Volcanics of Woodlark Island, referred to in the "Site 1109" chapter) that were encountered at the bottom of the Nubiam well in the Cape Vogel Basin at about 2300 mbsf. These rocks generally have the character of oceanic basalts and related rocks. It is not impossible that these are the rocks at the top of Moresby Seamount, which have been uplifted and unroofed by movement along the detachment fault to the north of Site 1114 from a depth where greenschist facies conditions prevailed.