The abundances of siderophile elements in the Earth’s silicate mantle are too high for the mantle to have been in equilibrium with iron in the core if equilibrium occurred at low pressures and temperatures. It has been proposed that this problem may be solved if equilibrium occurred at high pressures and temperatures. Experimental determination of the distribution of siderophile elements between liquid metal and liquid silicate at 100 kilobar and 2000 degrees C demonstrates that it is unlikely that siderophile element abundances were established by simple metal-silicate equilibrium, which indicates that the segregation of the core from the mantle was a complex process.