EXPERIMENT STATION BULLETINS. 607 



I. THE PREPAKATIOX OF AMMONIUM CITRATE SOLUTIONS 

 HAVING A DEFINITE REACTION OR COMPOSITION. 



a. Truly neutral solutions; 



b. Solutions of triammouium citrate; 



c. Solutions of empirical composition. 



II. THE RELATION BETWEEN COMPOSITION AND REACTION. 



III. THE RELATION BETWEEN THE REACTION OF THE 

 REAGENT AND ITS SOLVENT ACTION ON CALCIUM PHOS- 

 PHATES. 



a. In commercial fertilizers; 



b. In pure calcium phosphates. 



I. THE PREPARATION OF AMMONIUM CITRATE SOLUTIONS 

 HAVING A DEFINITE REACTION OR COMPOSITION. 



The original description called for a neutral solution and for this 

 reason as well as the fact that in most subsequent variations of this 

 original procedure a neutral solution is first obtained and then made 

 alkaline by the addition of a definite quantity of ammonia, the neutral 

 solution has come to assume a position of paramount interest and, in 

 fact, the means of preparing it have obscured to a large extent, the 

 more fundamental consideration of the actual value of the method itself. 



Fresenius and his colleagues make no suggestion in their original 

 article as to their method of obtaining neutrality in their reagent. From 

 Luck's (36) later communication it appears that such a solution was 

 never actually obtained except perhaps by chance or but for a short 

 time as it was being evaporated on the water bath to a density of 1.09 

 after being saturated with ammonia. In view of these facts as well 

 as of the state of development of the more accurate methods of physical 

 chemistrj', it seems safe to assume that the neutrality of the solution, 

 if it was determined at all, was tested by some indicator, probably 

 litmus, used in the ordinary way. In the light of subsequent experience 

 it may be concluded with equal certainty that the "vollkommen neutralen 

 Losung" of Luck was only approximate at best. 



Its preparation promptly gave trouble, judging from statements in the 

 literature. Herzfeld and Feuerlein (2G) claimed that' it was impossible 

 to get a neutral solution with litmus. Rosolic acid gave better results 

 but they prepared their reagent by adding to an aqueous solution of 

 citric acid an excess of ammonium carbonate, and allowing it to stand 

 for an hour with frequent stirring after which it was diluted to a 



