540 



Mr. W. W. J. Nicol on the 



Sodium Nitrate. 



(1) M.V., NaN0 3 solid 



(2) M. V., NaN0 3 in saturated solution 



(3) M. V., NaN0 3 in 1 molecule solution 

 „ (1-2) 3-4 „ ri 

 Here (T^3) = M = " 36L 



Potassium Nitrate. 



(1) M. V., KN0 3 solid 



(2) M. V., KNO3 in saturated solution 



(3) M. V., KNO3 in -5 molecule solution 



= 37-6. 

 = 34-2. 



= 28-2. 



=48-5. 

 =41-4. 

 = 38-1. 



Here 



(l-2)_ 7-1 



= •682. 



(1-3) 10-4" 



Now, to compare these results with the solubility, it must 



(1 — 2) 

 be kept in mind that the more nearly the fraction ^- — ^ 



approaches to a whole number the greater the attraction of 

 the salt-molecules for one another, and conversely. 



Table III. 



Salt. 



(1-2) 

 (1-3) 



Solubility. 



Q ^XsoLxMol.wt. 

 (1— 6) 



KNO3 



KCl 



NaCl 



NaN0 3 ... 



•682 

 •631 

 •590 

 •361 



5-63 



8-38 



10-99 



18-45 



387 

 394 

 38-0 

 66-6 



We find that the less the solubility the larger the fraction, 

 holds good in all four cases; and I have added to the table, 

 without intending to lay much stress on the result, the product 

 of the fraction by the solubility by the molecular weight. In 

 the three first cases the result is nearly a constant number ; 

 the mean difference is 2 per cent, only, which is little when 

 the number of determinations involved is considered ; but 

 NaN0 3 is entirely exceptional. I have failed to find any 

 error anywhere, and must leave it at present unexplained. It 

 will be found that this is not the only instance in which NaN0 3 

 behaves differently from ihe majority of salts. My reason for 

 introducing the molecular weight as a factor in the above cal- 

 culation is that, other things being equal, the cohesion of the 

 salt is evidently a function of its molecular weight. 



