276 J. II. Long — Polarization of Tartrate Solutions. 



the antimony. Other salts produced a marked decrease, but 

 less than that by the carbonate. 



As a problem of chemical dynamics this phenomenon pos- 

 sesses considerable interest as bearing on the general question 

 of the nature of solutions, and I concluded to make it the 

 special subject of an investigation, the chief points of which 

 are given below. 



Experiments with sodium carbonate. 



For each test five grms. of the tartrate were dissolved' in about 

 70 c.c. of water by aid of heat and then cooled to 20° C. or 

 below. Definite amounts of the carbonate, pure and dry, were 

 dissolved in water and added to the tartrate, the mixture being 

 then diluted to exactly 100 c.c. at 20° C. By careful work it 

 is possible to prepare clear solutions containing, with the five 

 grms. of the tartrate, a gram, or more, of the anhydrous car- 

 bonate. With much greater amounts precipitation usually 

 takes place before the polarization can be observed. 



In the table below the solutions numbered 1 to 5 inclusive 

 remained perfectly clear during several hours and were polar- 

 ized in two to four hours after preparation. Numbers 6, 7 and 

 8 became slightly turbid and finally deposited a precipitate. 

 The rotations were observed after twenty-four and forty hours. 

 In nearly all cases some differences were observed in the tests 

 made twenty-four hours apart. Such differences became very 

 plain when the first test was made as soon as possible after the 

 preparation of the solution. In order to show this I made 

 cold solutions of the tartrate and carbonate, brought them to 

 20°, mixed and diluted quickly with the small amount of water 

 necessary to make 100 c.c. After gentle shaking, the mixture 

 was poured into a 100 mm tube, kept at 20° by water, and polar- 

 ized within five minutes. The results found were very singu- 

 lar and interesting. They are given under numbers 9, 10 and 

 11. The three solutions remained perfectly clear during 

 several hours. At the end of twelve hours number 9 had 

 deposited nothing, but after twenty-four hours an unweigh- 

 able trace separated out. Number 10 deposited nothing until 

 after twenty-four hours. Number 11 was clear during two 

 hours but in twelve hours a precipitate settled out, leaving the 

 supernatant liquid perfectly clear. In all the above cases the 

 precipitate appeared very much less than called for by the 

 equation : 



2KSbOC 4 H 4 6 . H 2 + Xa 2 C0 3 = 2KNaC 4 H 4 6 + Sb 2 3 . H 2 + C0 2 



Analysis of the precipitate showed it to have the composi- 

 tion given when dried at 100°. For the five grms. of the 



