538 



Mr. G. Gore on 



phate of manganese (No. 148), and probably also snlpliate of iron 

 (No. 159) ; and in a less conspicuous degree in solutions of sulphate of 

 copper (Nos. 15 and 113), sulphate of didymium (No. 154), and other 

 substances. 



These anomalous results indicate the existence of a deeper seated 

 condition of the currents than that of the acid or alkaline nature of 

 the liquid, because the currents in those instances are not merely 

 altered in quantity but actually reversed. And this deeper condition 

 must also affect the quantity of the current in different cases, because 

 a diminution of quantity is a step towards a reversal of direction, and 

 an actual reversal is probably only an extreme degree of the same 

 change. 



This more abstruse condition consists probably of the structural 

 arrangement of the molecules of the liquid ; for we know as a matter 

 of course that tridimensional geometrical, as well as mechanical condi- 

 tions of the molecules must be fundamentally involved in the action. 

 It is also probable that as the chemical and electrical phenomena in 

 these experiments are usually coincident in their changes, this hidden 

 condition is of a more general kind than either, and determines not 

 only the acid or alkaline reaction of the solution, but also the produc- 

 tion and direction of the currents (i.e., it determines whether the heat 

 shall produce a current at all, and also the direction of the current pro- 

 duced) ; and that these two effects, the chemical and electrical, are but 

 concomitant ones. It is, however, evident that heat acting upon a 

 saline conducting liquid in the manner described is not alone sufficient 

 to cause a current, but that another condition is necessary, because 

 with some good conducting solutions, such as were used in Nos. 4<7, 

 49, 51, 57, 67, 69, 75, 90, 122, and especially No. 146, little or no cur- 

 rent was produced. 



The heat then acts as the real cause of the currents, and a par- 

 ticular molecular state of the liquid probably determines the kind and 

 direction of effect the heat shall produce, and as we also know that the 

 molecular structure of all bodies is profoundly altered by change of 

 temperature, we may reasonably conclude that whilst a portion of the 

 heat is absorbed in temporarily changing the molecular structure, and 

 is again set free when the structure returns to its original state during 

 the process of cooling, another portion is expended and lost in pro- 

 ducing the current. 



According to this interpretation the phenomenon of the current is a 

 product of three circumstances, viz., 1st, a particular molecular struc- 

 ture of the liquid ; 2nd, a change of the latter by alteration of tempe- 

 rature ; and 3rd, a direct conversion of heat into dynamic electricity; 

 such conversion being dependent upon the two prior conditions. The 

 second circumstance also agrees with the fact that in certain liquids, 

 for instance, Nos. 130, 139, 155, 156, 157, and 162, the current does 



