118 Mr. W. B. Hardy. 07h the Conditions which 



The relation really is not as simple as this ; it is complicated by the 

 change which the specific molecular conductivity of a salt undergoes 

 with change in concentration. The theoretical considerations have 

 been dealt with elsewhere.* For convenience of description, how- 

 ever, I will call this relation the relation of the square and cube. 



Action of Acids and Alkalis. — The values for K furnished by these 

 substances show relations to valency even more interesting than that 

 found with salts. As in the case of salts, their action is entirely 

 dependent upon the electric properties of the colloid particles. 



When the colloid particles are electro-negative, alkalis either do 

 not cause precipitation at any concentration, or if they do cause pre- 

 cipitation the value of K does not vary in any simple way with varia- 

 tions in valency. 



When the particles are electro-positive, K increases with valency, 

 but the relation of the square and cube does not hold. Instead, one 

 finds that K varies directly with the chemical activity of the solution. 



Acids have the reverse relations. When the particles are electro- 

 negative, the value of K varies directly with the chemical activity of 

 the solution ; while if these particles are electro-positive, acids either 

 have no precipitating power, or if K has any value, then (in the par- 

 ticular case measured) the value varies with valency according to the 

 square and cube. 



The various measurements are brought together in the following 

 table. The specific conductivities were calculated from the British 

 Association tables.! 



* Hardy and Whetham, 'Journal of Physiology,' vol. 24, 1899, p. 288, and 

 Whetbara, ' Phil. Mag.,' November, 1899. 



t " The Electro-chemiCal Properties of Aqueous Solutions." T. C. Fitzpatrick 

 ' Brit. Assoc. Report,' 1893. 



