442 REPORTS ON THE STATE OF SCIENCE. 



at 637° it consists of a mixture of two-thirds potassium iodide and 

 one-third potassium bromide. In a similar manner the sum of the 

 solubility of potassium iodide and potassium chloride equals that of 

 potassium iodide alone, and at 638° the mixture consists of 80 per 

 cent, potassium iodide and 20 per cent, potassium chloride. Potassium 

 bromide and potassium chloride in presence of each other show a 

 similar behaviour. 



From a study of the solubility of cupric chloride potassium chloride 

 ■,oc)o compounds Meyerhoffer 1;>7 concluded that each individual sub- 

 stance has a definite solubility temperature curve ; the change of 

 solubility at the transformation point is considered to be due to the dis- 

 appearance of that individual substance from solution, of which the 

 curve becomes changed. Roozeboom 159 examined the solubility and 

 transition points of the hydrates of thorium sulphate. 



It has been already mentioned that more recent work * had thrown 

 doubt on the values obtained by Bunsen and his pupils f for the absorp- 

 tion of gases by liquids; this prompted Tintofejew 163 to re-investigate 

 the relation between the solubility coefficients of oxygen, hydrogen, and 

 carbon dioxide, &c, in water, also in alcohol, and the temperature, 

 and, contrary to the observations of Bunsen and Carius, evidence was 

 obtained of a temperature coefficient for these gases. 



The solubility of hydrogen in water at temperatures between D and 

 1K01 ; " j(l ° was determined by WinHler 168 : the values found were in 

 close agreement with those published by Timofejew. It was also 

 observed that the absorption coefficient of hydrogen decreased with rise 

 of temperature. A little later Bohr and Bock 167 published fresh de- 

 terminations of the solubility of oxygen, hydrogen, and nitrogen in 

 "water, together with a further verification of the statement that the 

 solubility coefficient of hydrogen was not independent of the tem- 

 perature. 



This was followed by another paper by Winkler, 16 * in which he gave 

 in tabular form the found solubility values for nitrogen and oxygen in 

 water at temperatures ranging from 0° to 80°, and the calculated values 

 between 80° and 100°, the results here recorded for nitrogen being 

 considerably larger than those given by Bunsen. 



With the object of gaining some knowledge of the state of salts in 

 solution Etard 173 not only followed the change of solubility as the 

 temperature rose, but also the alteration of colour of the variously 

 coloured hydrated compounds of cobalt iodide and chloride; although 

 of considerable interest, no conclusions of a general character were 

 possible. 



™ In a paper, entitled 'Theory of Solubility Curves,' Fan 



Deventer and Van d. Stadt 187 discussed how far it is possible to 



deduce a general law T connecting solubility and temperature. They 



discussed the differential equations given by Le Chatelier and Van't 



* Vide this Section, 1875S0. 



f In a paper on the absorbing power of water for pases of the atmosphere (Pet- 

 tersson and Sonden, Ber. 1889, 22, 1439) is to be found a fairly complete record of 

 the work of Bunsen and his pupils on this subject. 



