of. 



Dilute Solutions of Cane-Sugar and Ethyl Alcohol. 39 L 



the same effect on the lowering produced bj ethyl-alcohol 

 solutions as on that of cane-sugar. Here also the molecular 

 lower ings found are considerably higher than the theoretical, 

 1-87. 



A comparison of the results which I have found under 

 these conditions for cane-sugar with those obtained by 

 Raoult * for this compound, will show that the two series 

 agree well for the more dilute solutions, while my results are 

 somewhat lower for the more concentrated. Since this work 

 was undertaken, Raoult has kindly informed me that he has 

 repeated his work with cane-sugar, using an improved cryo- 

 scope f, constructed so as to prevent direct radiation from 

 the freezinor-mixture to the bulb of the thermometer. This 

 was accomplished by interposing a sheet of platinum, which 

 served at the same time as a stirrer. Results were obtained 

 which agree well with those found earlier by himf, strengthen- 

 ing him in the belief that the molecular lowerinos of cane- 

 sugar in dilute solution are greater than the calculated, and 

 that this increases with the dilution. 1 take the liberty to 

 quote a line from the letter of this distinguished authority, on 

 problems of this character : — "Pour conclure, je reste con- 

 vaincu que les abaissements moleculaires de l'alcool en dissolu- 

 tion aqueuse, comme ceux du sucre, deviennent de plus en plus 

 superieurs a la valeur normale a mesure que les liqueurs 

 deviennent plus etendues.'" And who deserves to be heard 

 in reference to the freezing-point method if it is not Raoult ? 

 since it was he who showed, by his extended investigations §, 

 the significance of " molecular lowerings," and gave us in 

 this method the means of determining the molecular weights 

 of substances in solution jj. What has been done subsequently 

 in this line either bears an evident theoretical relation to the 

 work of Raoult, or consists in devising better apparatus and 

 more accurate methods for carrying out the measurements. 



To summarize : — In the above tables are recorded the results 

 of measurements of the freezing-point lowerings of water by 

 solutions of two non-electrolytes. The work was not extended 

 to dilutions greater than O'Ol normal, since for greater dilution 

 the experimental errors necessarily increase. But for the 

 dilutions used molecular lowerings were found, which rise 

 very appreciably above the theoretical value. Methods of 

 work were employed which seem to eliminate experimental 

 errors arising from too cold a bath, too vigorous stirring, or a 

 failure to establish equilibrium in the solution, and still the 



* Zeit.f.jjhys. Chem. ix. p. 343. 



t Revue Scientifique, Sept. loth, 1895. 



\ Zeit.f. phys. Chem. ix. p. 343. 



§ Compt. Rend. xciv. p. 1517; xcv, pp. 188, 1030. 



|| Co)npt. Rend. ci. p. 1056. 



