Mr. E. J. Mills on a Defect in the Theory of Saturation. 365 



C 2 , C 2 H, C 2 H 2 , C 2 H 3 , C 2 H 4 , C 2 H 5 , 



having the standard weights 



24, 25, 26, 27, 28, 29, 



are saturated respectively by 



6, 5, 4, 3, 2, 1 



unit weights of hydrogen. Complete saturation, with respect to 

 either, is represented in the formula 



C 2 H 6 . 



Owing, however, to the difficulty of always obtaining hydrogen- 

 compounds, or to the absence of them, the following point has 

 been allowed in practice — that a constant weight of any other 

 element, equivalent to a unit weight of hydrogen, shall be ac- 

 cepted in the place of the latter, and be considered, equally with 

 it, a measure of the saturability of the given substance. This 

 concession has been most frequently made in the case of chlo- 

 rine-, bromine-, and iodine-compounds — an equivalent of either 

 of the elements mentioned being supposed to function, with 

 respect to saturation, in precisely the same manner as one part 

 by weight of hydrogen. It is to this point that I wish briefly 

 to direct attention. 



The question as to interchangeability of saturating function 

 between any elements must depend not only on their being 

 capable of transposition in terms of equivalent value, but also 

 on their affinity for the substance to be saturated. For it would 

 be impossible to attribute to the vicarious element (as, for ex- 

 ample, chlorine used in the place of hydrogen) the power of satura- 

 tion at all, unless it had an affinity for the substance employed ; 

 nor could it conveniently be taken, if, as is sometimes the case, 

 the affinity were variable in its nature. Furthermore this ex- 

 change of function cannot be considered an equal one unless 

 the two elements are precisely alike in their affinity for the third 

 body. 



Let us suppose, for illustration's sake, a radical X' combined 

 with chlorine, bromine, and iodine, the last two being successively 

 weaker compounds. The measure of the full saturability of X' 

 will be the largest possible quantity (say, of one of these three 

 elements) which has the greatest affinity for it. Hence it is 

 obvious that X'Cl will only be fully saturated ; X'Br and X'l 

 will be deficient by some portion of a saturability, a positive 

 number, which may be termed x in the former, and {x+y) in 

 the latter case. Each of these quantities might of course ap- 

 proach or exceed unity if X were poly-equivalentic ; and it is 

 hardly necessary to say that these remarks apply to any radicals 



