192 PROCEEDINGS OF THE AMERICAN ACADEMY 



Therefore, the solution saturated at 15° contains the following per- 

 centages of the anhydrous salt. 



I. n. 



0.50 0.49 



The calcium salt when heated with an excess of lime gives off a 

 combustible vapor, probably difluorbeuzol, as only 4 per cent of fluorine 

 was removed from the acid in this way, but we do not intend to try to 

 isolate this product for fear of interfering with Paterno. This obser- 

 vation was made in an attempt to determine the fluorine in the acid 

 by Piria's method,* which showed that a much larger layer of alkali 

 than could be obtained in this way was needed to remove the fluorine 

 from the organic compound. We propose in our subsequent work to 

 make another effort to solve this difficult problem in organic analysis. 



Jt is to be observed that the large percentages of hydrogen ob- 

 tained in the combustion of the acid correspond to the formula 

 CgH^F^COoH better than to C^HgF^COJI, and that our other analyt- 

 ical work is not capable of deciding between these formulas. The 

 formula CgH.^FoCO^H would be that of an oxybenzoic acid containing 



two atoms of fluorine instead of one of oxygen ( '^ ^p u j^ and 



although the difficulty of removing the fluorine from the acid rendered 

 this formula highly improbable, we have thought it necessary to prove 

 that it was incorrect. For this purpose we treated the acid with a 

 large excess of potassic hydrate in alcoholic solution, and analyzed the 

 potassium salt, which can be precipitated by ether in broad silvery 

 scales, when we obtained an amount of potassium corresponding to 

 the formula CeH^F.COsK. 



* H. Schiff, Ann. Chem., cxcv. 293. 



