“…Eruption of some picritic and ultrabasic liq uids has been reported [Clarke, 1970;Viljoen and Viljoen, 1969], but there has been a real ization that ultrabasic and picritic magmas are generally too dense to erupt readily, either through the crust or through an existing magma chamber [Stolper and Walker, 1980;Sparks et al, 1980]. The mixing process whereby picritic mag ma enters a basaltic magma chamber involves the equilibrium, not fractional, crystallization of most of the excess olivine [Huppert and Sparks, 1980], thus greatly reducing the pre dicted nickel depletion, which has in any case ceased to be a major issue because the origi nal calculations were based on incorrect as sumptions [Cawthorn and Mclver, 1977;Clarke and O'Hara, 1979;Elthon and Ridley, 1979].…”