444 ftfcPOltTS ON THE STATE OE SCIENCE. 



above 0° to be a linear function of the temperature ; below this it was not 

 so: on the other hand, the solubility of ether in aqueous hydrogen 

 chloride is approximately inversely proportional to the temperature. 



Arctowslci i09 determined the solubility of mercuric halides in carbon 

 bisulphide over the range of temperature from — 10° to + 30°. The 

 solubility temperature curve in each case consisted of two straight lines, 

 the points of change of direction in these graphs lie on a straight line. 



The solubility of the iodide was found to be greater than that of the 

 bromide, and increases at a greater rate as the temperature is raised : 

 the same is true of the bromide as compared with the chloride. 



A further communication 210 contained values for the solubility of 

 iodine in carbon bisulphide between — 94° and +42° C. ; the graph in 

 this case being composed of no fewer than six straight lines connected 

 by short curves. 



Although nothing very definite can be gleaned from Roelofsen's 213 

 measurements of the solubility of potassium hydrogen tartrate in various 

 strengths of aqueous alcohol, it is of interest to find that as the con- 

 centration of the alcohol becomes greater the solubility of the tartrate is 

 less increased by rise of temperature : in 90 per cent, aqueous alcohol 

 a rise of temperature causes a decrease in the solubility of the tartrate. 



Etard 214 published a complete summary of his more recent work on 

 solubility, and included the results of determinations of the solubility of 

 sulphur in carbon bisulphide, benzene, ethylene dibromide, and in 

 hexane. Except in the last-mentioned solvent, the range of temperature 

 was from the freezing-point of the solvent to the melting-point of the 

 solute. 



Later on he recorded Z1S the limit of solubility temperature for various 

 salt mixtures: that found for AgN0 3 + KN0 3 = 198° (the melting-point 

 of AgN0 3 ) ; for Ba(NO,) 2 + BaCl 2 =474° (the melting-point of Ba(N0 3 ) 2 ; 

 and several others. The composition of the salt mixtures at these tem- 

 peratures was also given. 



Board's previous work on solubility of substances in carbon bisul- 

 phide at low temperatures ivas called into question by 

 Arctowski. 2[ii The solubility was not found to become zero at 

 the freezing-point of the solvent; neither were the solubility tempera- 

 ture graphs found to tend to cut the temperature axis, but were 

 asymptotic with it. 



It had been independently shown by Schroder 202 and by Le 

 Chatelier 221 that the solubility of a substance could be calculated from 

 certain mathematical expressions. 



In the light of this relationship Linebarger 237 discussed data for the 

 solubility of inorganic salts in normal organic liquids, but could not 

 recognise the applicability of the expression above referred to. The 

 substances he used were mercuric chloride and cupric chloride dissolved 

 in ethereal salts, mercuric halides in carbon bisulphide, and mercuric 

 chloride, cadmium iodide, and silver nitrate in benzene. 



