Lundstrom This project is an exploratory investigation of the effect of an external melt gradient on the behavior of peridotite melting. Peridotite melting experiments will be undertaken in a piston cylinder apparatus at 1 GPa with the peridotite in diffusive contact with a silica undersaturated melt (in a sense, simulating a melt conduit-peridotite interface). The motivation for these experiments stems from two fundamental discrepancies between natural peridotite samples and experimental results. First, abyssal peridotites, inferred to be residues of MORB generation, record that orthopyroxene contributes more to melting than clinopyroxene, an observation inconsistent with experimental results that predict clinopyroxene to dominate the melting mode. Second, melt compositions observed within peridotite xenoliths are dominated by silica rich (55-65%) melts, not basaltic melts as would be predicted by peridotite melting experiments. If an external melt gradient is able to influence how peridotite melts and thus explains these observations; future directions in experimental melt determinations and numerical modeling of mantle melting would be profoundly affected. The potential influence of external melt gradients also will be evaluated through further analyses of peridotite xenolith-melt relations including examining in-situ partitioning of trace elements between melt and clinopyroxene by ion probe analysis. This exploratory study is particularly timely given the rapidly changing state of ideas regarding the genesis of igneous rocks based on both experiments and observations.
