at the Contact of Mutually Reacting Liquids. 537 



electrolytic or voltaic effect of any action between the liquid 

 and this immersed pole has been ignored, and an error of 

 unknown amount has thus been introduced. There are, how- 

 ever, two series of experiments besides those of Exner and 

 Tuma mentioned above in which this error has been avoided. 

 In 1880 Ayrton and Perry* described an apparatus for the 

 investigation of the potential of liquids in contact, by their 

 inductive action on plates of gilt brass held close above their 

 surfaces. In 1883 Bichat and Blondlott published an 

 ingenious method involving the dropping of one of the liquids 

 from a funnel whose jet was within a cylinder kept moistened 

 by the other. 



Partly from consideration of the large and striking character 

 of the results obtained by these authors with strong acids, and 

 partly in the search of a possible simplification of the problem, 

 I have tested a series of pairs in which acids and strong- 

 aqueous solutions of various kinds were tried against water ; 

 the object being to ascertain whether the observed effects have 

 any connexion with the heats of solution or dilution of these 

 substances. The results are given in Table III., in which the 

 first column contains the formulae of the substances tested, 

 with the water of solution, if any ; the second column con- 

 tains the results of the experiments, stated separately ; and 

 the third the heats of solution or dilution for the number 

 of H 2 molecules stated, as given by Thomsen. The heats of 

 dilution for potassium nitrate and chlorate are not available 

 but are probably negative since the heats of solution of the 

 anhydrous salts are negative. 



A comparison between the numbers in the third and fourth 

 columns shows that in the case of positive heats of solution 

 there is a rough agreement between their amounts and those 

 of the observed electromotive forces, while with negative 

 heats the results appear variable and indefinite. 



7. Table III. gives also a test of an hypothesis current 

 among German physicists, which, premising that the ions 

 are oppositely electrified, states that " water must show against 

 every electrolytic solution the potential of the faster ion." 



I have placed after each radical a number proportional to 

 this " ionic velocity," according to F. Kohlrausch, as quoted 

 in the Eeport of the British Association Electrolysis Com- 

 mittee for 1887 J. It appears that out of the fourteen pairs 

 tried, in the four cases numbered 26, 27, 36, 37 water takes 

 what would be assumed according to the above hypotheses as 



* Phil. Trans. 1880, p. 15. 



t Comptes Rendus, xcvii. pp. 1202, l'Jdo, 



X B. A. Report (1887), pp. 838, 354. 



