174 



substance. For example, a curdy white precipitate of silver 

 chloride is obtained not only by adding to a solution of silver 

 nitrate a solution of hydrochloric acid (equation 1) or sodium 

 chloride (equation 4), but by addition of any solution containing 

 the radical Ag to any solution containing the radical Cl. If we 

 suspect that a solution contains Ag, we add, therefore, a solution 

 of any chloride. If we suspect that a solution contains Cl, we 

 add a solution of any soluble silver salt. In either case a negative 

 result is conclusive evidence of the absence of the radical tested for. 

 If a positive result is obtained, the precipitate must be examined 

 further, to prove whether or not it is silver chloride. 



Even when no precipitate appears, however, some interaction 

 takes place. Thus a solution obtained by mixing sodium nitrate 

 and potassium chloride solutions is identical in all its properties 

 with a solution obtained by mixing sodium chloride and potassium 

 nitrate solutions. The appearance of a precipitate is not, there- 

 fore, an essential feature of a double decomposition. The most 

 important point for us to notice is that, in all such reactions, each 

 substance present behaves exactly as if it consisted of two dis- 

 tinct radicals. We may therefore regard double decompositions 

 as a simple result of the liberty of radicals to interchange partners. 

 Normally, this exchange will not be complete. Thus in the 

 instance cited immediately above we have the reversible reaction: 



(Na)(N0 3 ) + (K)(C1) ^ (Na)(Cl) + (K)(NO 3 ). 



Whichever pair of salts we start with, we reach the same result 

 on mixing their solutions. Every double decomposition is sim- 

 ilarly reversible in theory, and gives an equilibrium mixture. But 

 because, in many mixtures, one of the four possible compounds 

 withdraws from the exchange of partners by disappearing from the 

 solution either as a gas (p. 126) or as a precipitate (p. 127), the 

 reaction in such cases becomes practically complete in one direction. 

 Another significant point may now be noted. Not only does 

 every acid, base and salt in solution behave, in double decom- 





