284 CARNEGIE INSTITUTION OF WASHINGTON. 



of water," would be found less destructive to the membranes than 

 the potassium cation. The surmise proved to be correct, and we 

 were able to measure satisfactorily the osmotic pressure of a large 

 series of solutions of lithium chloride. In one ''endurance test" 

 under a constant osmotic pressure of nearly 19 atmospheres, the 

 membrane gave no evidence of deterioration until after 112 days. 



In regard to the magnitude of the osmotic pressure of solutions of 

 lithium chloride, it is to be stated that it was found to be somewhat 

 larger than would have been anticipated from the dissociation of the 

 salt as determined by the conductivity method. 



Dr. Frazer and Dr. Holland, both of whom may be said to have 

 grown up with, and into, this investigation, have given their whole 

 time to the problems mentioned above; the former to mannite and 

 to the electrolytes, and the latter to glucose and to the preparation 

 of new cells for the resumption of the high-temperature work on 

 cane-sugar. In the capacity of volunteer. Dr. J. B. Zinn has con- 

 tributed materially to the progress made during the past year. 



Noyes, Arthur A., Massachusetts Institute of Technology, Boston, Massa- 

 chusetts. Grant No. 848, allotted December 13, 1912. Researches 

 upon the properties of solutions in relation to the ionic theory. (For 

 previous reports see Year Books Nos. 2-11). $3,000 



During the past year an extended article on" The Relation between 

 the Concentration and Conductance of Electrolytic Solutions" has 

 been published by Dr. C. A. Kraus and Dr. W. C. Bray (Jour. Amer. 

 Chem. Soc, 3, v. 35, pp. 1315-1434, 1913.) This article consists 

 in a critical study of existing conductance data in a great variety of 

 solvents. This study, which has established the important principle 

 that, in all the solvents where the data enable any conclusion to 

 be reached, the ionization of uniunivalent salts changes with the 

 concentration in accordance with the mass-action law, provided the 

 ion-concentration be smaller than about 0.005 normal. The inves- 

 tigation has, moreover, shown that the deviations which in the case 

 of ionized substances begin to be appreciable at only slightly higher 

 concentrations than this can be represented in all cases by an expres- 

 sion of the same general form, namely, {cyy/c{l-y)=K+D{cy)'", 

 in which c is the concentration, y is the ionization, K is the mass- 

 action constant, and D and m are other empirical constants which 

 determine the magnitude of the deviations from the mass-action law. 

 The work has also led to many other, more specific, conclusions. 



In further confirmation of the conclusions reached in the article 

 just referred to, new experimental work has been done by Mr. W. J. 

 Winninghoff on the conductance of solutions of iodides in amyl 

 alcohol and propyl alcohol. 



The measurements on the vapor-pressure, specific volume, and 

 compressibility of water at temperatures from 200° to the critical 



