of elective Attraction's. 57 



corrected order of the simple elective attractions, I have also 

 inclosed it in a parenthesis. Such an apparent inconsistency 

 may perhaps in some cases be unavoidable, as it is possible 

 that the different proportions of the masses concerned, in 

 the operations of simple and compound decomposition, may 

 sometimes cause a real difference in the comparative magni- 

 tude of the attractive forces. Those nuuibers, to which no 

 asterisc is affixed, are merely inserted by interpolation, and 

 they can only be so far employed for determining the mutual 

 actions of the salts to which they belong, as the results 

 which they indicate would follow -from the comparison of 

 any other numbers, intermediate to the nearest of those, 

 which are more correctly determined. I have not been able 

 to obtain a sufficient number of facts relating to the me- 

 tallic salts, to enable me to comprehend many of them in 

 the tables. 



It has been usual to distinguish the attractions, which pro- 

 duce the double decompositions of salts, into necessary and 

 superfluous attractions J but the distinction is neither very 

 accurate, nor very important : they might be still further 

 divided, accordingly as two, three, or the whole of the four 

 ingredients concerned are capable of simply decomposing 

 the salt in which they are not contained ; and if two, ac- 

 cordingly as they are previously united or separate; such 

 divisions would however merely tend to divert the attention 

 from the natural operation of the joint forces concerned. 



It appears to be not improbable, that the attractive force 

 of any two substances might, in many cases, be expressed 

 by the quotient of two numbers ap[)ropriate to the substances, 

 or rather by the excess of that quotient above unity ; thus 

 the attractive force of many of the acids for the three prin- 

 cipal alkalies might probably be correctly represented in this 

 manner ; and where the order of attractions is different, 

 perhaps the addition of a second, or of a second and third 

 quotient, derived from a different series of numbers, would 

 afford an accurate determination of the relative force of at- 

 tractioi!, which would always be the weaker, as the two 

 substances concerned stood nearer to each oiiicr in these 

 • orders 



