The EDGES of the Mantle

The core-mantle boundary region is often considered to be the source of narrow upwellings which drive or influence plate motions and continental breakup, fuel large igneous provinces and generate volcanic chains. The plume hypothesis has influenced most fields of geochemistry, petrology, geodynamics and mantle evolution. The key axioms underlying the plume paradigm are identified: Axioms are self-evident truths and are seldom stated explicitly. When they are, they are sometimes not so self-evident. The role of the surface boundary layer is discussed in connection with large igneous provinces and volcanic chains. Partial melting is the expected natural state of the upper mantle and only abnormally high seismic velocities imply absence of melting (slabs, cratons). Plume theoreticians have underestimated the average temperature of the mantle and have overestimated melting temperatures. Extensive melting in the upper mantle does not require abnormal temperatures or plumes. The dynamics and chemistry of midplate volcanism are explainable by near-surface processes. "Midplate" volcanism starts at plate boundaries or discontinuities, at lithospheric "edges", in regions of extension generated by plate processes. The chemistry of "hotspot" basalts implies contamination by processes and materials that occur near the surface of the Earth. A buoyant metasomatised layer at the top of the mantle, near the melting point, removes the need to import heat and chemical inhomogeneity from D" to explain midplate volcanism. D" is an interesting and important region, but any connection to surface processes or chemistry is speculative.