Thermo-electric Action of Me.tals in Electrolytes. 271 



compared the relative positions in Table X of the metals which have 

 crossed each other by simultaneous heating to 160 F. with those of 

 the same metals in Table I, as determined by heating them separately 

 to that temperature in the same liquids. I thus found that out of the 

 total 83 instances of crossing 72 are explicable in this manner, and 

 the remaining 9 are but feeble exceptions, both as regards their 

 relative positions in the chemico- and in the thermo-electric series, 

 and also as regards their relative degrees of electromotive force, 

 7 out of the 9 couples being composed of noble metals, which give 

 relatively feeble currents. No instance occurred in which the re- 

 versal was produced by alteration of potential of one metal only; 

 32 happened in which it was due to one metal increasing and the 

 other decreasing in that power; 37 in which it was due to both 

 metals increasing simultaneously, but at different rates ; and 5 in 

 which it was a result of simultaneous but unequal diminution of 

 potential in the two metals* The differences of order of the chemico- 

 electric series at 60 and 160 F. in Table X, and the coincident 

 reversals, are therefore results of the combined effect of heat upon 

 the two individual metals, as shown in Table I. In the cases of re- 

 versal, the number of instances in which a metal becomes positive by 

 rise of temperature was more than twice as great as those in which it 

 became negative ; this agrees with the proportion of thermo-positive 

 and negative members in Table I (see pp. 253 and 254), and with the 

 statement that heat usually increases electro-positive activity. 



It is evident that as in certain cases a given metal is chemico- 

 electro-positive at 60 to another metal, whilst at 160 F. it is negative, 

 it must at some intermediate temperature be neutral to it and pro- 

 duce no current ; the electric potentials of the two metals must at 

 that point be equal and opposite. A number of experiments were 

 therefore made in order to ascertain the temperatures of these neutral 

 points at which reversal took place. The solutions employed were all 

 of them of the same composition as those used in determining the 

 chemico-electric series (p. 269). The following are the results : In 

 Potassic cyanide: Mg and Al, 80 F. ; Mg and Zn, 146; Pb and Pd, 

 84. Potassic fluoride : Sn and Cd, 146. Potassic nitrate : Pb and 

 Fe, 65; Al and Sn, 90; Ag and Pd, 182. Potassic carbonate: 

 Mg and Al, 106. Sodic diphosphate : Al and Cd, 150 ; Al and Zn, 

 124 ; Al and Pb, 100 ; Al and Sn, 120. Potassic iodide : Fe and 

 Pb, 120 ; Fe and Al, 160. Sodic chloride : Fe and Pb, 60 ; Cu and 

 Ni, 72. Potassic chloride : Au and Pd, 65. Potassic sulphate : Pb 

 and Al, 170 ; Ni and Cu, 64 ; Ag and Pd, 100. Oxalic acid: Al 

 and Cd, 152 ; Al and Sn, 122 ; Au and Pd, 62. Formic acid : Al 

 and Sn, 170; Al and Pb, 185; Pd and Pt, 110. Dextro-tartaric 

 acid: Al and Fe, 177 ; Al and Sn, 182; Al and Pb, 154. Potassic 

 hydrate : Cd and Pb, 138 ; Cd and Fe, 68. Sulphuric acid : Al and 



