CHEMISTRY. 243 



grees of ionization in the same concentrations and react with different 

 velocities with methyl iodide, and yet these three salts, in concentrations 

 varying from N/i to N/32, give the same value for K^, namely, 0.0282, 

 0.0283, and 0.0284, respectively. The values of K^ are 0.00474, o-oo37. 

 and 0.0040, respectively, this wide variation harmonizing very nicely with 

 our theory because of the differences in the nonionized sodium, potassium, 

 and lithium phenolates. 



Dr. J. Chandler proved, as did also Dr. Sidney Nirdlinger and Dr. F. M. 

 Rogers working together, that ethyl iodide reacts with both the anion and 

 the nonionized portion of sodium phenyl thiourazole in ethyl alcohol at 25°. 

 The solutions were varied in concentration from N/3 to N/64. Dr. Chand- 

 ler found the values Kj^ 0.465 and K^=o.i6. Dr. Nirdlinger and Dr. 

 Rogers found the values Kj=o.43 ^^d K^^o.17 in their work; the solu- 

 tions contained small quantities of water, which lowers the reaction velocity. 



The work of Stieglitz on the catalysis of imido esters has been interpreted 

 unfortunately by him and gives excellent evidence for our theory. We find 

 that both the cations and the nonionized salts are decomposed by the 

 molecular water, Stieglitz believing that the cations alone are decomposed 

 and that the water reacts through its ions alone. 



The work of Tubandt on the inversion of menthone by sodium ethylate 

 has been recalculated by us and found to give fine evidence that the ethylate 

 anion and the nonionized sodium ethylate are about equally active in this 

 transformation. 



The recent work of H. Goldschmidt on the esterification of formic acid, 

 acetic acid, phenylacetic acid, and butyric acid by the catalyzer trichlorbu- 

 tyric acid, especially in the presence of the aniline salts, shows very clearly 

 that there are esterifications of cations as well as of nonionized complexes 

 taking place. 



We are extending the study of these problems to very dilute solutions and 

 hope to measure some of the reactions accurately in N/iooo concentration. 

 In such a case it will be interesting to see whether the above equations hold 

 accurately in all concentrations beyond say N/50, and especially to learn 

 whether 10 or 20 molecules of an added salt change the reaction velocity of 

 a N/500 or N/iooo solution of the reacting salt exactly as is demanded by 

 the above equations. Our evidence seems to show that in such dilute solu- 

 tions these equations should hold accurately, and that there should be no 

 appreciable "abnormal effect" produced by the added salts. This matter 

 will be decided definitely after further careful studies. 



Baxter, Gregory P., Harvard University, Cambridge, Massachusetts. Grant 

 No. 763, allotted December 15, 191 1. Determination of atomic zveights. 

 (For previous reports see Year Books Nos. 3-10.) $1,000 



With the assistance of the Institution the following researches were car- 

 ried on under Professor Baxter's direction : 



