622 



SCIENCE 



[N. S. Vol. XLVI. No. 1199 



then introduced into the solution by breaking the 

 capsule in two with the electro-magnetic hammer. 

 As soon as the reaction was well started, the 

 bomb was sealed ofE and allowed to react at room 

 temperature. The bomb consisted of two com- 

 partments. AVhen the reaction was completed, 

 the solution was poured off from the excess anti- 

 mony, and the antimony washed by distilling the 

 solvent over from the solution. The analysis was 

 completed by distilling off the solvent into 

 weighed water bottles and weighing the anti- 

 mony left behind. Electrolyses of these solutions 

 were carried out at the temperature of boiling 

 ammonia. The electrolyses showed that the com- 

 pounds in solution are electrolytic in nature and 

 that more than one atom of antimony is associ- 

 ated with each negative carrier. Both the analyses 

 and electrolyses showed that there are at least 

 two compounds involved in the final equilibria, 

 one having more than two atoms and one having 

 less than two atoms of antimony for each negative 

 charge. These compounds are in some ways simi- 

 lar to the polyiodides. A detailed exposition of 

 this investigation will be offered for publication 

 to the Journal of the American Chemical Society. 



The effect of acid concentration on the photo- 

 chemical oxidation of quinine hy chromic acid: G. 

 S. Forbes and E. S. Dean. In a previous investi- 

 gation of this reaction by Luther and Forbes, the 

 acid concentration had been constant. In the 

 dark, with concentrations of CrOj and purified 

 quinine constant, the velocity varies as the square 

 of the acid concentration. A shallow cylindrical 

 dish was bisected by a glass partition, and re- 

 volved under a quartz mercury lamp. Provisions 

 were made for stirring, cooling and temperature 

 measurement. Solutions as described above were 

 compared in pairs. After correction had been 

 made for the dark reaction, the velocity of the 

 photochemical reaction was found independent of 

 acid concentration. It was also proved that 

 quinine solution exposed to light does not retain 

 its activation for long in the dark. 



The temperature coefficient of the distribution 

 ratio: G. S. Forbes and A. S. Coolidge. Solubili- 

 ties in two and three component systems involving 

 water, ether and succinic acid were determined or 

 redetermined at 15°, 20° and 25°, also the distri- 

 bution ratio of the succinic acid between two 

 ether-water phases. An equation was derived 

 and verified showing the temperature coefficient of 

 the distribution ratio, with excess of the acid, as a 

 function of the temperature and mutual solubil- 

 ity coefficients of each substance in each layer. 



The distribution ratio, when calculated on the 

 basis of ether-water phases in which the ratioB 

 ether to water are constant, is by no means inde- 

 pendent of the concentration of succinic acid. 

 Evidence was secured that the average degree of 

 association of water dissolved in ether at these 

 temperatures is somewhat less than two. 



The application of palladium as an indicator for 

 silver titrations: L. Schneider. A very dilute so- 

 lution of palladous nitrate, dissolved in an excess 

 of nitric acid, is added to the silver nitrate solu- 

 tion which is then titrated with potassium iodide. 

 The silver nitrate is precipitated by the potassium 

 iodide and the least excess of potassium iodide is 

 converted by the palladous nitrate to paUadous 

 iodide which is visible to the extent of one part in 

 a million. For very dilute solutions, this method 

 gives better precision than the Volhard method. 

 The size of the plus and minus errors have been 

 determined. The constant plus error in concen- 

 trated solutions is due to the palladous iodide be- 

 ing carried down by the silver iodide at the end- 

 point, whereas the negative error is caused by the 

 absorption of silver nitrate by silver iodide. The 

 standard method for overcoming these errors has 

 been applied with such effect that not only good 

 precision but satisfactory accuracy has been ob- 

 tained. The ease and rapidity with which the 

 standard solution and the indicator can be pre- 

 pared recommend this new method. Also the pal- 

 ladous nitrate method can be used to better ad- 

 vantage than Volhard 's in cases where the silver 

 nitrate solutions are colored pink or yellow. Ni- 

 trous acid interferes and must be boiled off before 

 titrating. 



The application of the thermodynamic methods 

 of Giihs to equilibria in the ternary system 

 B.O-K,SiOs-SiO,: George W. Morey and Erskinb 

 D. Williamson. A discussion of Gibbs's deriva- 

 tion of the phase rule and the application of 

 Gibbs's thermodynamic methods to various types 

 of heterogeneous equilibria occurring in the ter- 

 nary system H.O-KoSiOs-SiO.. The slopes of the 

 various P-T curves which proceed from a quin- 

 tuple point are discussed, with special reference to 

 the dependence of the slope of a given curve on 

 the composition of the phases which coexist along 

 it. The change in slope with change in composi- 

 tion of phases of variable composition is dis- 

 cussed in detail. Conclusions reached in the above 

 discussions are applied to typical cases in the ter- 

 nary system H.O-K^SiOs-SiO:. 



{To be continued) 



