Site 1158 basalts were recovered from three holes (1158A, 1158B, and 1158C) ~78 km south of Site 1157. The site is located on ~20-Ma crust within Zone A at the eastern margin of the depth anomaly. One whole rock from each hole was analyzed for major and trace elements by X-ray fluorescence (XRF) and ICP-AES (inductively coupled plasma-atomic emission spectrometry). Basalt glasses from Holes 1158A and 1158B were analyzed by ICP-AES only (Table T3). Whole-rock Ni and Cr contents determined by ICP-AES are consistently low, and MgO, Fe2O3, and Al2O3 are consistently high when compared with XRF data.
Samples from Hole 1158A are assigned to one lithologic unit, an aphyric to sparsely olivine-plagioclase phyric pillow basalt (see "Igneous Petrology"). The Hole 1158A glass is primitive (i.e., ~9.1 wt% MgO) with relatively high Na2O, Fe2O3, Zr, and Y contents and low CaO and CaO/Al2O3 values (Figs. F10, F11). The corresponding whole rock has ~0.5 wt% less MgO than the glass, as observed throughout Leg 187 probably as a consequence of alteration (see "Geochemistry"). However, based on similarities in most major elements, Hole 1158A whole rock and glass are probably genetically related. The whole rock is lower in Na2O, slightly lower in Y, and slightly higher in Cr than the associated glass. Alteration of olivine and the consequent loss of Mg from the whole-rock sample could explain the constancy of Ni with decreasing MgO. Accumulation of plagioclase or olivine phenocrysts cannot explain the relatively low Na2O, Y, Zr and high Cr contents. These variations could indicate retention of clinopyroxene and/or spinel in groundmass while mesostasis ± olivine is lost to alteration. Note that Sr and Ba are unchanged.
Hole 1158B samples are assigned to a single unit of aphyric to sparsely olivine-plagioclase phyric basalt similar to that of Hole 1158A. Unlike Hole 1158A, the 1158B whole rock is much lower in MgO (~6.7 wt%) than the associated glass (~8.2 wt%). Major and trace element variations are inconsistent with the whole rock being derived from the glass by simple low-pressure crystal fractionation. CaO, Al2O3, TiO2, Zr, and Y are relatively similar in the glass and whole rock, even with a 1.5 wt% decrease in MgO. Furthermore, Na2O and Fe2O3 decrease rather than increase, as would be expected with simple crystal fractionation. Loss on ignition for whole-rock Sample 187-1158B-2R-1, 55-58 cm, is slightly high (1.24 wt%), consistent with MgO loss during alteration.
Two lithologic units are recognized in Hole 1158C. Unit 1 is similar to those described in Holes 1158A and 1158B, and Unit 2 is a diabase. Only Unit 2 was analyzed aboard ship. Although lower in MgO, the diabase is similar to Hole 1158B whole-rock Sample 187-1158B-2R-1, 55-58 cm, in many respects. Slightly higher CaO and Al2O3 contents could result from plagioclase accumulation, and lower MgO suggests that the diabase has evolved magmatically relative to the Hole 1158B whole rock. However, the scale of variations in MgO introduced by alteration makes petrogenetic interpretations based on whole-rock data highly speculative. The compositional pattern described above (Holes 1158A and 1158B) appears to apply equally to the diabase, suggesting that the unchanging CaO, Al2O3, TiO2, Zr, and Y contents with decreasing MgO may be a complex interplay between in situ crystal growth and alteration effects. Neither of the Site 1158 glass compositions appears to be a simple parent to either the whole rock or the diabase.
Site 1158 basalts are compared to two Zone A propagating rift segments (A2 and A3) in Figures F10 and F11. The Site 1158 glasses exhibit some of the major and trace element characteristics of lavas dredged from propagating rift tips (PRTs) of Segments A2 (127.5°E) and A3 (131.0°E) (Pyle, 1994). In particular, the primitive Hole 1158A glass has low CaO/Al2O3 contents, high Na2O, TiO2, Zr, and Y, and slightly elevated Sr and Ba, indicating low-degree partial melts. It also has high Fe2O3, which suggests a high mean melting pressure (Klein and Langmuir, 1987). These features are less pronounced in the Hole 1158B glass, but its composition overlaps trends defined by PRT glasses from Segments A1 and A2; note the increasing CaO and CaO/Al2O3 and decreasing Zr, Y, and TiO2 with decreasing MgO of Segment A1 and A2 glasses and the general similarity of Site 1158 glasses. The unevolved nature of the Site 1158 glasses (>8.0 wt% MgO) and the variability in major and trace elements indicate that these compositional variations are caused by partial melting with little influence from low-pressure crystal fractionation in subaxial magma systems. This similarity of Site 1158 lavas to lavas recovered in the PRTs of Zone A suggest proximity to a PRT environment for Site 1158 basalts.
The Zr/Ba systematics of Site 1158 (Fig. F12A) suggest a Pacific-type mantle. The Hole 1158A glass, Hole 1158B glass, and Hole 1158C diabase have high Zr/Ba and vary within the compositional range of Pacific-type mid-ocean-ridge basalt (MORB) from Southeast Indian Ridge (SEIR) Segments A2 and A3 (Fig. F12B). The Site 1158A whole rock lies below the main Pacific-type MORB field, toward the PRT field of Segment A3, consistent with low-degree melting (i.e., higher Ba) and with the inference that these lavas are associated with the rift-tip environment. Site 1158 lavas also have Na2O/TiO2 systematics that suggest a Pacific-type source. Comparing the Site 1158 results with those from Sites 1153 and 1157 (see "Geochemistry" in the "Site 1153" chapter and "Geochemistry" in the "Site 1157" chapter), we identify Pacific-type mantle beneath western Zone A at 28 Ma (Site 1153) and 20 Ma (Site 1158), with intervening Indian-type mantle at 22 Ma (Site 1157). These fluctuations in mantle source are on a time and space scale similar to that of the recent Pacific-type source migration across Segment B5 (Pyle et al., 1992).