Mr Clark on the Pyrophosphate of Soda. 299 



In mixing solutions of nitrate of silver and of phosphate of 

 soda, I was not a little surprised at obtaining a white, instead 

 of a yellow precipitate ; notwithstanding that Doctors Thom- 

 son and Berzelius speak with decision of the precipitate being 

 yellow, and that in my former experience, I had always found 

 it yellow. A question immediately arose, Why the precipitate 

 was white instead of yellow ? 



As I had taken considerable pains, by repeated crystalliza- 

 tions, to render the phosphate of soda pure, it was natural to 

 conjecture, that this salt, in its ordinary state, contains some 

 impurity to occasion the yellowness of the precipitate. To 

 try the accuracy of this conjecture, I dropped a solution of 

 commercial phosphate of soda into one of nitrate of silver. 

 The precipitate was yellow. But when, in order to contrast 

 this effect with that produced by pure phosphate of soda, I 

 dropped a solution of purified crystals into nitrate of silver, 

 I was surprised to get a precipitate, which was likewise yel- 

 low. These effects were not altered by the reverse operation 

 of dropping the nitrate of silver into solutions of the ordinary, 

 and the purified phosphates. 



Such were the experiments, which first pointed out to me 

 the circumstance, upon which depended the production of a 

 white precipitate. In order to avoid any uncertainty in my 

 experiments as to the quantity of dry phosphate of soda, I had 

 used a phosphate dried at a red heat, instead of crystals. It 

 was a solution of this phosphate dried at a red heat, which 

 had produced a white precipitate ; whereas it was a solution 

 of the crystals, which had produced a yellow precipitate. To 

 be quite sure that the alteration was produced by heat, I 

 divided a fine large crystal of purified phosphate of soda into 

 two parts. One of these, I dissolved in water: the other I 

 dried, heated red hot, and then dissolved in water. Dropped 

 into a solution of nicrate of silver, the solution of the undried 

 part of the crystal gave a yellow precipitate; whereas that 

 of the dried part gave a white precipitate. The crystal 

 which I used had been purified, indeed ; but perfect purity 

 is not at all necessary for the success of the experiment. The 

 commercial phosphate of soda answers quite well ; so that this 



