J. H. Long — Polarization of Tartrate Solutions. 279 



rotation of Rochelle salt in presence of excess of sodium car- 

 bonate. 



But the calculated rotations approach those of actual obser- 

 vation only in the cases of the solutions with 0*1 grm. of 

 carbonate, and in those with very large amounts of the latter, 

 observed after long standing or boiling. No simple relations 

 can therefore be traced in this way, and it is evident that a 

 part of the antimony must be held in solution in some other 

 way than as tartrate. 



In mixing the solutious at a low temperature no precipitate 

 was formed and no C0 2 given off. It is well known that the 

 addition of acids to tartrate solutions decreases their rotation, 

 mainly perhaps, from formation of bitartrate, and the idea 

 suggested itself that this may be the action of the carbonic acid 

 in the present mixture. An experiment tried with Rochelle 

 salt after saturation with the gas failed to show, however, a 

 lowered rotation. Slightly lower results were found by mixing 

 the Rochelle salt with sodium bicarbonate. 



That an important reaction takes place in solution can, how- 

 ever, be shown in another way. I prepared several solutions 

 of the potassium antimony tartrate containing exactly 5 grm. 

 in something less than 100 c.c. of cold water. To one of them 

 a little litmus was added, after which a solution of sodium 

 carbonate of exactly 2 per cent strength was run in from a 

 burette. An alkaline reaction did not appear immediately, but 

 the final change of color was not sharp enough for an accurate 

 determination. Another solution was tried with phenol- 

 phthalein, and here good results were found. In several trials 

 between 74 and 78 c.c. of the carbonate solution were required 

 for coloration. If the assumed equation of decomposition is 

 correct, for 5 grms. of the tartrate we need practically 0*8 gram 

 of the carbonate. This amount is contained in 40 c. c. cf the 

 solution used and it required nearly double that volume for 

 coloration, which reaction suggests the formation of a bicar- 

 bonate if we bear in mind that such salts are neutral to 

 phenol-phthalein. If it is also taken into consideration that in 

 our titration a little gas is readily lost by agitation, it will be 

 seen that practically the proper amount of alkali is added to 

 complete the reaction above. From these tests it is apparent 

 that a reaction between the tartrate and carbonate begins 

 immediately on the addition of the latter, and also that the 

 liberated C0 2 is held as bicarbonate or in other form with 

 metal and neutral to the indicator. The following equation 

 may express what takes place : 



2KSbOC 4 H 4 6 + 2Na 2 C0 3 + 2H 2 = 2KNaC 4 H 4 6 + Sb 2 3 . H„0 + 

 2HNaC0 3 ; or = 2l£NaC 4 H 4 6 + 2SbONaCO s + 2H 2 0. " 



Am. Jour. Sci. — Third Series, Vol. XL, No. 238.— Oct., 1890. 

 18 



