456 REPORTS ON THE STATE OF SCIENCE. 



solubility of potassium hydrogen tartrate was published by Noyes and 

 Clement . 207 Their results may be briefly summarised as follows : 

 ' (i) The three potassium halogen salts decrease the solubility 

 of the tartrate to about the same extent; (ii) potassium nitrate in con- 

 centrated solutions causes a smaller decrease in the solubility of the 

 tartrate than is occasioned by the potassium halides : potassium chlorate 

 to even a smaller extent than the nitrate ; (iii) potassium acetate causes 

 a decided increase ; and (iv) potassium sulphate least effect of all other 

 salts ; while (v) hydrogen chloride brings about an increase in the 

 solubility of the KHC 4 H 4 ? . 



With the object of gaining information as to the behaviour of 

 isomorphous salt pairs, Muthmann and Kuntze 217 studied the solubility 

 of monopotassium phosphate and arsenate; potassium perchlorate and 

 permanganate ; and potassium and rubidium permanganates. Although 

 difficult to interpret, the results seem to be in accord with the con- 

 clusions of Roozeboom (vide this Section, 1891). 



Following a general review 214 and amplification of his work, 

 Etard 21S recorded an attempt made to measure the solubility of the 

 potassium halide triple mixture, KCl + KBr + Kl in water. It was 

 found, however, that these salts were not capable of existing simul- 

 taneously in solution ; the potassium chloride remaining undissolved in 

 the presence of the other two salts. 



A further contribution to the study of mixed hydrated crystals was 



Q _ published by Stortenbeker. 23 * The molecular percentage of the 



' two constituent salts are made to serve as ordinates and 



abscissae, and the solubility curves drawn, when the following three 



classes of curves are illustrated: — 



(1) The solubility isothermals do not cut — an example being the 

 case of (MgS0 4 ,7H 2 0; ZnS0 4 ,7H 2 0). 



(2) The isothermals cut, but have no breaks — an example being 

 (CuS0 4 ,5H 2 6; FeS0 4 ,5H 2 0). 



(3) The isothermals cut, but one of them has a break — an example 

 being CuS0 4 ,5H 2 0; Mn(S0 4 ),5H 2 0. 



Van Laar 2i2 investigated thermodynamically the formulae express- 

 ing changes of solubility, and obtained theoretically the following 

 results : When there is one ' ion ' common to both substances the 

 solubility becomes lowered, the least soluble undergoing the greatest 

 relative change. Non-electrolytes were found to have no effect on the 

 solubility, neither are they themselves affected in this respect. When 

 there is no ' ion ' in common, the effects produced are more complex 

 and frequently indeterminable. 



VI. Mutual Solubility and Distribution Coefficient a. 



The first to examine the distribution of a substance between two 

 1872 noil -"iiscible solvents were Berthelot and Jungfleisch.' 19 Ex- 

 perimenting with iodine and bromine in mixtures of water and 

 carbon bisulphide, also various organic and inorganic acids, and 

 ammonia in mixtures of water and ether, they tried to find some 

 relationship between the partition of the substance and its solubility in 



