610 STATE BOARD OF AGRICULTURE. 



tions which may be either truly neutral or solutions of the "neutral" 

 salt. The use of the latter was recommended. 



McCandless (38) shortly afterwards republished his method for the 

 preparation of a solution by analysis. He also reported the preparation 

 of the solid salt, triammonium citrate, which when dissolved in water 

 gave an alkaline solution gradually turning acid with loss of ammonia. 



R. G. Hall (24) subsequently prepared this salt and proposed its use 

 for the preparation of the reagent under consideration. 



In his report to the A. O. A. C. in 1914, Walker (50), the associate 

 referee on phosphoric acid, reported results comparing the titration 

 method and the litmus method of Hand (25). He also published the 

 results of the determination of insoluble phosphoric acid using four 

 solutions of citrate, prepared in as many different laboratories, together 

 with the composition of these solutions. These latter results illustrated 

 the wide discrepancies which can and undoubtedly do arise from the 

 use of reagents of varying composition prepared in different laboratories 

 and demonstrated the need of a more rigid and accurate method. 



Shorey (48) has lately published an account of a method of prepara- 

 tion of this reagent by calculating the theoretical amount of ammonia 

 necessary to neutralize the desired quantity of citric acid. 



Washburn (52) varies the usual routine of autoneutralization by 

 drawing a current of air through his solution in order to remove the 

 excess ammonia. 



Such are the most important methods which have been proposed for 

 gaining the desired end. It will be seen that from the very first pro- 

 posal of the use of this reagent there has existed a confusion of the term 

 neutral, as applied to the solution itself, and to the salt. This was due 

 of course to a failure to recognize and emphasize the fact that, owing 

 to dissociation and hydrolysis, a solution of the "neutral" or better, 

 normal salt of ammonia and citric acid is not neutral at all but alka- 

 line, i. e., the concentration of the hydroxyl ions exceeds that of the 

 hydrogen ions. As it developed later the originators of the method 

 while intending to use a solution of the normal salt actually specified 

 a "neutral" solution. Methods subsequently offered for the preparation 

 of the reagent reflect this confusion and naturally fall into three classes, 

 (a) those whose sponsors attempt to prepare a truly neutral solution, 

 i. e., one in which the concentrations of hydrogen and hydroxyl ions are 

 equal, (b) those which give a solution of the triaraonnium citrate, and 

 (c) those which yield solutions of a strictly empirical character, viz. 

 are merely "acid" or "alkaline." 



(a) Methods for the preparation of truly neutral solutions: I\v this 

 class are all those which are adjusted by means of indicators except in 

 the case of that of Eastman and Hildebrand (15) as hereinafter dis- 

 cussed. There is the serious objection to this procedure that there is 

 no indicator which gives a sharp end-i)oint witli the acid and alkali 

 used. Of those indicators which have been tried, rosolic acid, corraline 

 and litmus apparently give the best results in about the order named. 

 Yet, througliout the literature, one continually finds statements to the 

 effect lliat the best is not satisfactory. From the comments of otliers as 

 well as from his own experience, the writer is forced to conclude that 

 only with the procedure devised by Lord (35) and modified by Hand 

 (25) can anything like accurate results be obtained. This procedure 



