386 Mr. H. C. Jones on the Cryoscopie Relations 



to the most dilute solution. Had the series been extended to 

 any very considerable dilution (say 005 n), with the same 

 rate of increase molecular lowerings would have been found 

 differing unmistakably from the calculated constant. This 

 can be seen most accurately by plotting the results in a curve. 



While the correction-term for ethyl alcohol was small 

 enough to be neglected, with cane-sugar it assumed consider- 

 able dimensions, amounting to o, 006, where the quantity 

 measured was only o, 0277, being thus somewhat more than 

 one fifth of the whole. Unless a correction-term enjoys 

 mathematical exactness, its introduction, when amounting to 

 more than 20 per cent., would not tend to increase our con- 

 fidence in the final result. The results throughout with 

 cane-sugar are not very convincing, being only very rough 

 approximations to the theoretical value 1'87. It further 

 seems desirable to know how Nernst and Abegg assured 

 themselves of the purity of the compounds which they 

 employed, since one searches their paper in vain for evidence 

 that this, the chemical side of the problem, has received any 

 attention. 



The experimental work of Nernst and Abegg as recorded 

 in the paper in question *, as it appears to me, may be fairly 

 summarized thus. When measurements were made of the 

 lowering of the freezing-point of water by dilute solutions of 

 cane-sugar and ethyl alcohol, and when the former were cor- 

 rected by a term amounting to as much as 22 per cent, of 

 the quantity measured, and the latter not corrected at all, 

 values were obtained for molecular lowerings which were not 

 more than 6 per cent, greater (1'97 for ethyl alcohol) than 

 the theoretical, 1*87; nor more than 7*5 per cent, less (1'73 

 for cane-sugar). From their results the authors concluded 

 that non -electrolytes give theoretical results. 



A valuable suggestion, however, which can be tested expe- 

 rimentally, has been made in this paper by Nernst and Abegg; 

 and the object of the work herein described is to study the 

 effect of the temperature of the freezing-mixture on the 

 results. The authors thought that the difference between the 

 results found by the different experimenters, working with 

 cane-sugar for example, was caused by the use of freezing- 

 mixtures of different temperatures. How far this was of 

 influence will appear from the following work. 



In this investigation cane-sugar solutions were used for 



the most part, since with this compound so many have worked 



and have obtained such widely different results; and, further, 



it was with this substance that the above-mentioned correction 



* Zcit.f. phys. Chcm. xv. p. 680. 



