1923] 



CAMP — CITRIC ACID AS A SOURCE OF CARBON 215 



out. Handbooks ordinarily give the sodium salt as crystallizing 

 with 11 molecules of water but when it is crystallized out from 

 alcohol as above described the percentage of water seems to be 

 much lower. None of the salts of this acid with the heavy metals 

 are quantitatively insoluble in water but most of them are less 

 soluble in alcohol. The salts of Ca, Pb, and Ba are commonly 

 used in analysis and will be discussed under the head of quantita- 

 tive methods. 



Citric acid is quite readily oxidized, its salts and the acid itself 

 being oxidized in air at less than 200° C. The acid is decomposed 

 by concentrated sulphuric acid, sulphuric and chromic acid mix- 

 tures, by KMn0 4 in acid solution, and by K 2 Cr 2 07 under the 

 same conditions as for KMn0 4 , but more slowly. The general 

 products from such oxidations are C0 2 , CO, acetone, acetaldehyde, 

 acetic acid, and formic acid, depending on the oxidizing agent 

 and the conditions of the reaction. On account of its high carbon 

 content and the ease with which it is oxidized citric acid should 

 be a fairly good source of carbon for those organisms capable of 

 utilizing it. 



QUALITATIVE DETECTION 



The common methods of detection are based largely upon the 

 fact that certain salts are less soluble in hot water than in cold, 

 i. e., precipitates are formed on heating the solution and these 

 disappear when the solution is cooled. Calcium citrate is com- 

 monly used in this test and the acid lead salt has been recom- 

 mended by the Association of Official Agricultural Chemists ('07). 

 This sort of method is likely to be misleading, however, in the 

 presence of other salts or in the case of too great or too small a 

 concentration of citric acid. Stahre's pentabromacetone method, 

 depending upon the formation of a complicated compound, 

 pentabromacetone, when citric acid is oxidized to acetonedicar- 

 bonic acid by KMn0 4 in the presence of Br, has been used exten- 

 sively. This test is probably less sensitive, and much less satis- 

 factory than that of Deniges ('98), which is based upon the 

 formation of a complicated mercury compound with acetone. 

 As this method was used in this work it will be given in detail 

 and follows closely the instructions by Yoder ('11). DenigeV 

 solution is prepared by dissolving 5 gr. of mercuric oxide in 20 



