80creeplawofolivine(CarterandAveLallemant,L9701.PDFVIP

80 creep law of olivine (Carter and AveLallemant, L9701,Raleigh and Kirby, .].972a. 1970; Weertman, 1970; Green and Radcliffe, b; Kďtlstedt and Goetze, 7974; Buiskool Toxopeus and Boland , 1976). The relation between the differential stress and dislocation density of olivine determined experi- mentally by Kohlstedt and Goetze (L974) has offered a new possibility for estimating the stress acting within the upper manUe. However, in orcier to apply this relation it is necessaryto demonstrate that the dislocation density and structures formed in the upper mantle have not been altered by sub- sequent anneďing. Goetze and Kohlstedt (1973) tried to anneal naturally deformed olivines, but were unable to conclusively show that dislocation annihilation had occurred. The recovery process involves dislocation annihilation, collapse and shrinkage of dislocation loops, and formation of subgrains and cell walls (Cottrell, 1953). These processes involve dislocation climb which requires diffusion of atoms (Cottrell, 1953; Nabarro, L967: Suzuki, 1967). Conse- quently it is necessary to conduct anneďing experiments at high temperature for severď hours in order to study the annihilation rate of dislocations in olivine. These experiments have important implications for deducing the mechanical state of the upper mantle. PRocEDUREs EXPERIMENŤAL Samples were prepared from ďpine type peridotites from Horoman in Japan and Red Hill in New Zealand, and from lherzolite nodules from Eifel in West Germany and from San Carlos in Arizona. A diamond saw was used to cut samples 10 X 5 X 2 mm in size. Thechemical compositions of the olivines fall