Dissociation of Compounds in Water. 485 



his method of curve-drawing to my results from sodium 

 chloride, which differ from his to the extent of more than 

 50 per cent., and with it claims to have found the same 

 " breaks " as in his own results. He has thus shown the true 

 value of his method, which seems to be largely independent 

 of the experimental data. 



I have since * shown that his curve contains the following 

 fundamental errors : — 



I. At least some of the points in his curve are wrong. 



II. The choice of points through which he has drawn the 

 curve is purely arbitrary. 



Organic Compounds. 



Many of the organic compounds are known to undergo 

 little or no electrolytic dissociation. These would be expected 

 to give molecular lowerings of the freezing-point not widely 

 different from the constant for the solvent, at least in fairly 

 dilute solutions. The chief object in mind at first in studying 

 some of the organic substances was to determine experimen- 

 tally and with some care the constant for water. The work 

 was begun with cane-sugar, and it was found that the most 

 dilute solutions gave a molecular lowering much greater than 

 the constant calculated for water, and that this high molecular 

 lowering decreased fairly regularly with increase in concen- 

 tration. It was thought at first that this was probably due 

 to some inorganic impurity in the sugar, but that this was not 

 the cause was shown when the same results were obtained 

 from sugar as pure as it could be prepared. It was then 

 thought that such results might be peculiar to sugar and bodies 

 of similar chemical constitution. It was possible that the 

 cane-sugar molecules conducted themselves in the very dilute 

 solutions as if inverted into two molecules. Dextrose was 

 then investigated, the molecules of which could not apparently 

 conduct themselves in this way. When results of the same 

 general character were obtained from the dilute solutions of 

 dextrose, it was decided to study a number of organic com- 

 pounds whose chemical composition and constitution were as 

 widely different from each other as possible. One could then 

 see whether this property were more or less general among 

 the organic compounds, or limited to certain classes of them. 

 In addition to cane-sugar and dextrose, urea, phenol, ethy] 

 and propyl alcohols have been studied in reference to this 

 point. Acetic and succinic acids have also been investigated, 

 but the organic acids are known to undergo electrolytic dis- 



ss 

 sociation. 



Jones, Ber. d. deutsch. Chem. Gesell. xxvi. p. 1685. 



