This is a collaborative project between James Natland of Scripps Institution, and Fred Frey of MIT. It is an integrated textural, mineralogical, and geochemical study of ultramafic xenoliths and their host lavas from the Samoan, Society, and Marquesas Islands in the southwest Pacific. The xenoliths are in existing collections, and preliminary mineralogical and isotopic work demonstrates that they are related to processed of partial melting and magmatic differentiation beneath these volcanoes. The island chains contain some of the most enriched basaltic lavas in the ocean basins, which are hypothesized to be related to a regionally widespread, perhaps hemispheric, geochemical anomaly in the mantle. Xenoliths should provide complementary information on the compositions of mantle sources, the action of metasomatic processes in influencing or modifying source compositions, and the processes of melt extraction and magmatic differentiation in and near the source regions of ocean-island basalts. Studies so far indicate that both residual (Type 1) and "cumulate" (Type 2) xenoliths are present in all three island chains, and that substantial degrees of partial melting produced possible tholeiitic magmas which crystallized olivine-spinel- pyroxene assemblages en route from the mantle. Such lavas may comprise much of the voluminous, submerged, but still unsampled portions of these volcanoes. Principal techniques to be applied to the xenoliths will be 1) petrography; 2) electron-probe microanalysis of mineral phases; 3) ion-probe analysis of trace-element abundances in minerals, and 4) a combination of mass spectrometry (conventional solid- source and ICP-MS), XRF, and instrumental neutron-activation analysis for whole-rocks and mineral separates. Steps 1 and 2 will be used to select the appropriate range of samples for detailed ion-probe, isotope, and trace-element studies. Bulk chemical, trace-element, and isotopic measurements will be made on host lavas and these related to temporal successions on the islands, which are already known. Comparisons will also be made to published information on continental and Hawaiian xenolith suites.
