Abstract
It has been suggested in the past that the thermoelectric power of expanded fluid mercury near the critical density rapidly decreases to zero. A new experiment has been performed to check whether this is really so or whether the observations can be attributed to the effect of high temperature on the other materials in the measuring system. It is concluded that the thermoelectric power of Hg really does cease to vary monotonically with temperature near the critical point and that it changes sign twice. This behaviour is tentatively connected with the critical point itself and the situation is discussed. The critical point is found to be at pc=0.1675+or-0.0025 GPa and Tc=1495+or-8 degrees C.