the Thermo- Electric Properties of Liquids. 



537 



of it, and are not reversed in direction during the cooling, it is evident 

 that they cannot be due to the act of diminution of cohesion or adhe- 

 sion produced by the heat. As, further, all the substances employed 

 are in precisely the same physical and chemical states after an experi- 

 ment as they were before it, and as the portion of heat expended in 

 raising the temperature, expanding the substances, and diminishing 

 their cohesion and adhesion, must have been given out again on the 

 cooling and contracting of those bodies, that portion cannot have been 

 expended in producing the currents, From these considerations I con- 

 clude that neither the force of cohesion, nor that of adhesion, acts as 

 an intermediate cause of the currents, and that the latter are, in an 

 ordinary sense of the word, direct effects of the heat, (It may be 

 further remarked that if a change of the force of cohesion, or of that 

 of adhesion, gave rise to the currents, then such a change when pro- 

 duced by other means than heat (such as enormous pressure) must be 

 capable of causing similar effects.) 



As heat was the true, immediate, and only cause of the currents, the 

 action may, in accordance with usage in similar cases, be properly 

 termed thermo-electric ; and as heat was the only force expended, we 

 may also, in accordance with the great law of conservation of energy, 

 safely conclude that heat disappeared in the process. And as it is the 

 nature of the liquid which most determines both the existence of the 

 current and its direction, it is true thermo-electric action of liquids, 

 and liquids possess true thermo-electric properties. 



The heat must, however, be subject in its operation to some guiding 

 condition which determines the kind of effect to be produced, and both 

 the cause, heat, and the effect, electricity, must be essentially related 

 to that condition. Such a condition appears to be the acid or alkaline 

 nature of the liquid ; as, however, chemical action is not necessary, and 

 different liquids of acid reaction (or of alkaline reaction) give a reverse 

 direction of current, the chemical conditions of acidity or alkalinity 

 are probably only concomitant circumstances, and do not really, but 

 only apparently, determine the kind of effect ; and as those conditions 

 are usually of an opposite kind in liquids which yield an opposite 

 direction of current, and are in most cases stronger as the currents are 

 more powerful, they are probably themselves dependent upon some 

 more general condition which itself determines the direction of the 

 currents. 



A single clear exception invalidates a general inference, and indi- 

 cates the existence probably of the action of a second law, or of a more 

 general cause or condition. That the thermo-electric-negative pro- 

 perty of a liquid is separable from and may exist without the alka* 

 line property, is clearly proved by the exceptional direction of the 

 current in solutions of selenious acid (No. 157), sulphate of nickel 

 (No. 155), nitrate of silver (No. 147), chrome alum (No. 156), sul- 



VOL. XXVII. 2 K 



