Site 1164 basalts were recovered from two holes that sampled 18- to 19-Ma-old crust from Segment B5 of the AAD. Two whole-rock samples were analyzed for major and trace elements by XRF only, and two samples of fresh glass chips were analyzed for major and trace elements by ICP-AES only. The results are shown in Table T4.
The samples from Hole 1164A are from a single unit of aphyric pillow basalt (see "Igneous Petrology"). Both whole rock and glass are relatively evolved, consistent with the presence of euhedral clinopyroxene phenocrysts (see "Igneous Petrology"). However, the glass has ~1.75 wt% more Fe2O3, ~1 wt% more MgO, ~0.1 wt% less K2O, ~25 ppm less Sr, ~100 ppm less Ni, and ~100 ppm less Cr than the whole rock. Lower Ni and Cr contents in the glass could reflect analytical bias between the XRF and ICP-AES analyses, although ~100-ppm differences have not been observed in other whole-rock XRF-ICP sample pairs. There are smaller differences in other elements between the glass and the whole rock that appear to be mostly caused by lower MgO content in the whole rock. Nonetheless, it is difficult to attribute the differences in whole-rock and glass compositions to removal of Mg by alteration when loss-on-ignition values are <1%. Taken at face value, the glass and whole rock are not genetically related by simple low-pressure crystal fractionation.
Samples from Hole 1164B are from a single unit of pillow basalt rubble (see "Igneous Petrology"). Although aphyric basalt predominates in this hole, moderately plagioclase-olivine phyric basalt occurs at the top of the hole. Our samples are from this part of the sequence. The glass has ~0.30 wt% more TiO2, ~1.5 wt% more Fe2O3, ~1 wt% more MgO, ~10 ppm more Zr, and ~0.10 wt% less K2O than the whole rock. Smaller differences in composition between the glass and whole rock also exist for other elements. Again, the glass and whole rock compositional trends are not indicative of simple low-pressure crystal fractionation. Either Mg loss is responsible for this lack of a genetic relationship or the whole rocks are derived from parental melts that differ from the associated glass.
Site 1164 samples are compared to 0- to 7-Ma lavas from Segment B5 of the AAD in Figures F30 and F31. The glass samples plot within the range of younger Segment B5 basalts. The whole-rock samples are displaced from the glass trends in the TiO2, Na2O, Ni, and Cr vs. MgO plots. The glasses from Holes 1164A and 1164B are identical to younger Segment B5 glass; therefore, no significant difference in melting or mantle composition is inferred.
On the Zr/Ba vs. Ba diagram, Hole 1164A glass lies within the Pacific-type field and Hole 1164B glass lies within the Indian-type field. Note that both glass samples plot on the low-Ba side of the gray tie line connecting Segment B5 axial glasses that define the transition between Indian and Pacific types along the present SEIR (Fig. F32A). On the Na2O/TiO2 vs. MgO diagram, Site 1164 glasses also parallel samples from the B5 spreading axis but are displaced toward the Pacific-type region on the diagram (Fig. F32B). These observations suggest that Site 1164 basalts are derived from a transitional mantle, similar to that present beneath the Segment B5 axis today.