In common practice thermopowers of metals or conductors are determined from thermo e.m.f. of the thermocouples. In the conventional method, reported in the literature [I.C.T.⁹, Barnard¹], thermopower of the thermocouple is obtained from the temperature derivative of thermo e.m.f. Thermopowers of individual metals are obtained by using thermopowers of the reference metals³. Previous method as well as thermopowers of reference metals have the vital role⁸ in the determination of thermopowers of the test metals. Some points in favour of the need of test of these factors are essential. Moveover, there is a poor agreement between theory and experiment of thermopowers of metals. Either values of thermopowers of a metal reported by two authors or valves obtained with respect to two reference metals or valves of thermopowers of two sample of a metal, hardly agree with each other [I.C.T.⁹, Seitz¹³, Wilson¹⁵, Barnard¹, MacDonal¹¹].

In the present communication we are going to discuss the complicated anomalies, we also test the conventional method for determination of thermo-powers of metals form thermo e.m.f. and accuracy of thermopowers of some reference metals reported in the literature³ and determine the thermopowers of lead. Thermopower of the most popular reference metal⁷ has also been suggested.