New York Agricultural Experiment Station. 323 



lizations did not alter the composition. Heavy metals other than 

 barium were absent and we could not detect any weighable quantity 

 of alkalies in 0.5 gram of the salt. It was completely free from 

 inorganic phosphate and it was free from nitrogen and sulphur. 

 We believe, therefore, that it represents a pure chemical compound. 



The composition as previously reported 6 agrees very closely 

 with that required by inosite hexaphosphate, CeHisC^Pe- The 

 free acid was prepared and analyzed, which also agreed with the above 

 formula. Silver salts were prepared from the above isolated acid 

 but it would seem that silver salts are not very suitable for the 

 purpose of identifying an acid of the above nature. They are 

 obtained as amorphous precipitates which do not represent homo- 

 geneous salts. They are evidently mixtures of more or less acid 

 silver salts. 



The silver precipitates which we obtained did not agree in com- 

 position with the compounds analyzed by Rather nor did they agree 

 with any definite silver salts of inosite hexaphosphate. As sug- 

 gested above, they are evidently mixtures of more or less acid silver 

 salts of inosite hexaphosphate — for after deducting the amount 

 of silver found, allowing for a corresponding amount of hydrogen 

 and water and calculating to the free acid, the results agree very 

 closely with the percentage composition calculated for inosite 

 hexaphosphate. 



In the isolation and purification of the barium salt we made use 

 of our former method in preference to that proposed by Rather for 

 the reason that we consider our method more simple and convenient. 

 The essential difference in these methods of isolation is that we use 

 barium hydroxide throughout, precipitating the substance with 

 this reagent from dilute hydrochloric acid solutions. Rather 

 used a modification of the method of Patten & Hart 7 — substituting 

 the use of sodium hydroxide with ammonium hydroxide. The use 

 of either sodium or ammonium hydroxide which must be eliminated 

 again is not necessary, for barium hydroxide is equally efficient and 

 by its use the introduction of other basic ions is avoided. 



Since the present work substantiates our earlier results and since 

 all the analytical data agrees with inosite hexaphosphate, C 6 Hi 8 024P6, 

 or with salts of this acid, we believe that the organic phosphoric 

 acid in cottonseed meal must be represented by the formula 

 either of inosite hexaphosphate, C 6 Hi 8 02iP6, or else some- formula 

 isomeric with this. 



It may be noted that the percentage of phosphorus found on 

 analyzing the free acid is somewhat low. In the analyses of the 

 acids previously reported 8 the phosphorus was found to be from 



8 LOC. tit. 



7 Am. Chem., Journ. 31:566, 1904. 



8 LOC. tit. 



