Thermo-electric Action of Metals in Electrolytes. 261 



thermo-electric sign also of single metals occurred, and in all these, 

 viz., 64, with some intermediate strength of liquid, and the two 

 pieces of metal at 60 and 160 F. respectively, no thermo-electric 

 current would be produced. Magnesium suffered a very great depres- 

 sion of electro-motive force from near the top to quite the bottom of 

 the series in solution of potassic fluoride by increased strength of the 

 liquid ; and many other great changes of position of metals in other 

 solutions were similarly produced. 



The depression of magnesium to about the bottom of the thermo- 

 electric series in nine different liquids in Table I, and about the same 

 number in Table IV, may have been due to the already -mentioned 

 (p. 255) circumstance that in all the solutions, except those of acids 

 and salts of acid reaction, magnesium becomes coated with an 

 adherent film of white basic salt. In not one of the solutions of acids 

 or acid salts was magnesium thus coated, or its position depressed. 

 On the other hand, the coating of basic salt had no such effect in the 

 solution of weak potassic fluoride ; if also the coating caused the 

 magnesium to be thermo-electro-negative in so many liquids, the hot 

 magnesium would probably have been positive at the first instant of 

 immersion, before the coating had time to sufficiently form, but it was 

 not. (See Table I, Notes.) 



Both in weak and strong solutions, silver, copper, palladium, and 

 cadmium usually gave thermo-currents of the least electro-motive 

 force, and aluminium and nickel were conspicuously thermo-positive 

 and platinum and magnesium frequently the most thermo-negative. 



Many of these results agree with the circumstance that with un- 

 equally heated metals in liquids, great changes of strength, and even 

 reversals of current, frequently occur without any apparent cause, but 

 probably in consequence of alteration of chemical composition and 



Table V. 



strength of the liquid by irregularities of diffusion at the surfaces of 

 the metals ; this is largely proved by the great effect which is usually 

 produced by stirring the liquid. Bouty refers such variations partly 

 to the presence of dissolved air and partly to movements of flow of 



