PROGRESS IN PHYSICAL CHEMISTRY. 299 



Ponsot (110) describes another method, in which he 

 adds pounded ice to the solution, finds the temperature 

 when equilibrium is attained, pours off the solution and 

 determines its concentration by analysis. 



Two new solvents have been added to the list. Paterno 

 and Montemartini ( r 1 1 ) show that paraxylene has a 

 molecular depression of 43. The majority of substances 

 when dissolved in it give normal molecular weights, hydroxy 

 compounds, just as in the case of benzene, being exceptions. 

 According to Lespieau (112) the hydrate H 2 S0 4 H 2 melts 

 at 8*53° and behaves as a solvent with a definite molecular 

 depression of 48. 



Noyes and Whitney ( 1 r 3) find that in aqueous solution 

 alkaline aluminates have the formula M Al 2 . Borates 

 have a similar formula unless excess of boric acid be present 

 when acid salts are formed: 



The behaviour of solutions in benzene of some 80 

 oxy-compounds has been studied by Auwers (114). If the 

 molecular weight in solutions of medium concentration is 

 abnormally large, he concludes that the substance is a 

 hydroxy compound. Certain derivatives of acetoacetic 

 acid, diacetylacetone, and the oxy-methylene ketones ap- 

 peared to be ketonic ; the acid derivatives of the primary 

 organic bases gave abnormal molecular weights, but the 

 secondary and tertiary compounds gave normal values. 

 Mercaptans differ from the alcohols in showing' no aggrega- 

 tion. (Compare pp. 227, 228 and 234.) From the de- 

 pression of the freezing-point of acetic acid by water in 

 presence of sulphuric acid/ H. C. Jones (115) obtained 

 evidence of the existence under the experimental conditions 

 of the hydrates H 2 S0 4 , H 2 and H 2 S0 4 , 2H,0. Similar 

 observations rave no indication that water and alcohol 

 combine, but action of some kind seemed to be exerted by 

 water on sodium acetate. 



Conductivity. — In a lengthy paper, which is practically 



a review of present knowledge of ion migration, Bredig 



(116) calculates from conductivity observations the velocities 



of some 300 different ions. He finds that the velocity of 



elementary ions is a periodic function of the atomic weight. 



21 



