Site 1153 basalts were recovered from ~28-Ma seafloor formed within Zone A of the SEIR. One whole-rock powder was analyzed by both X-ray fluorescence (XRF) and inductively coupled plasma-atomic emission spectrometry (ICP-AES); two samples of fresh basalt glass chips were analyzed by ICP-AES only. The results are shown in Table T3. The whole-rock ICP-AES analyses are of poor quality because of dissolution problems and inexperience with the JY2000 instrument early in Leg 187. We reanalyzed the glass samples later in the leg, and both results are presented in Table T3. For the figures, only the averages of the rerun results are used. Note that, in the original analyses, SiO2 content was estimated by difference from 100%. Otherwise, the original and rerun glass analyses agree very well for most elements and there were no systematic biases. Despite the analytical difficulties, XRF and ICP-AES analyses agree well, with the exception of Ni and Cr, which are both lower in the ICP-AES analyses, and Ba, which is below the detection limit of the XRF.
Samples from Hole 1153A are assigned to a single aphyric basalt unit, based on macroscopic and microscopic examination (see "Igneous Petrology"). The single whole-rock sample analyzed contains ~6.5 wt% MgO (Fig. F9), ~0.2 wt% K2O, and ~11 ppm Ba. As noted for Site 1152, the glass samples from Hole 1153A have higher MgO contents (8-8.4 wt%), lower K2O contents (0.08-0.13 wt%), and lower Ba contents (5.2-8.5 ppm) than the associated whole rock (Table T3). All other elements are similar in both whole-rock and glass samples. The compositional differences between the whole-rock and the glass samples cannot result from simple low-pressure crystal fractionation, as the whole-rock glass tie lines are oblique to the fractionation trends for 0- to 7-Ma Zone A lavas (Figs. F9, F10). These observations are similar to those from Site 1152, again suggesting that alteration or nonequilibrium magmatic processes have affected the whole-rock compositions.
Site 1153 glasses have higher Ti, Fe, Zr, Y, and Ba and lower CaO and Cr, compared with axial Zone A lavas of the same MgO content that are not associated with this propagating rift environment. The Site 1153 glasses are displaced from Zone A low-pressure crystal fractionation trends, away from the more evolved compositions typical of Zone A toward the distinct, more primitive, Zone A propagating rift tip (PRT) glass compositions (Figs. F9, F10). Higher Ti, Zr, Y, and Ba in the Site 1153 glasses and Zone A PRT lavas could result from either a more fertile mantle source or a lower degree of melting of the same mantle source. The low Na contents of Site 1153 glasses relative to 0- to 7-Ma Zone A axial glasses suggest a higher extent of melting, more than typical for PRT lavas from Zone A. High Fe2O3 in the Site 1153 glasses suggests a greater mean depth of melting. These compositional variations suggest that Site 1153 and PRT lavas have formed by slightly different degrees of melting, from a deeper, more fertile mantle source than most Zone A lavas. The similarity of Site 1153 and PRT lavas is not surprising, as Site 1153 is in one of a series of en echelon grabens that are likely associated with rift propagation.
The Ba/Zr systematics of Site 1153 basalts suggest that Pacific-type mantle was present beneath Zone A at ~28 Ma (Fig. F11). The Na2O/TiO2 vs. MgO diagram is also consistent with a Pacific-type mantle. Because Site 1153 glasses have higher TiO2 at a given MgO content, they plot at significantly lower Na2O/TiO2 than the 0- to 7-Ma Zone A glasses (Fig. F11B).