A common assumption in modeling dynamic processes in the earth's lower mantle is that thermal conductivity is sensibly constant throughout the lower mantle and equal to approximately 4 W m-1 K-1. This value is the result of extrapolation using a linearized theory with estimated parameters for lattice contributions to thermal conductivity. Furthermore, this value is highly dependent on the assumed geotherm. Excluding theoretical uncertainties in the extrapolation, recent experimental data for the pressure dependence of conductivity coupled with a revised geotherm lead to substantially higher estimated values for the lower mantle. The hypothesis that the seismically resolved anomalous zone at the base of the mantle (D') is a thermal boundary layer can be questioned in this case of a higher estimation for thermal conductivity.