14 DR, GLADSTONE ON THE REFRACTION-EQUIVALENTS OF THE ELEMENTS. 
Composition of solution. 
Befraction- 
equiyalent 
of Na Cl. 
Variation 
from mean. 
1 atom Chloride of Sodium 4- 10-74 atoms Water 
15-33 
-007 
+ 12 „ 
15-45 
+005 
+ 12 „ 
15 51 
+0-11 
+ 12 „ 
15 51 
4-011 
+ 14 „ 
15-26 
-014 
+ it; „ 
15-32 
-008 
+ is , 
15-47 
+0-07 
+ 20 
15-43 
+0-03 
+ 22 
15-55 
+ 015 
+ 24 „ 
15-51 
+ 0 11 
„ + 26 „ 
15-37 
-0-03 
„ + 34 „ 
15-30 
-010 
This shows also that under favourable circumstances a refraction-equivalent may he 
depended on to the first place of decimals, but not to the second. 
4th. The calculated refraction-equivalent is the same whether water or alcohol be the 
solvent employed. This was tested in the following cases, the actual observations for 
which are given in Appendix III. 
Table IV. 
Substance. 
Aqueous solution. 
Alcoholic solution. 
Cobalt Chloride 
3202 
3238 
Copper Chloride (line D) 
3400 
34-72 
Mercuric Chloride 
41-23 
35-59 
Potassium Iodide 
35-72 
351 
Potassium Sulphocyanide 
33-47 
33 -7 
Ammonia 
9-49 
8-97 
The mercuric salt appears to be exceptional, but this metal will be seen later on to be 
anomalous. 
But whatever may be the worth of these considerations, an examination of some cor- 
responding series of salts in solution, viz. the chlorides, bromides, and iodides, convinced 
me at once that we thus obtain numbers made up of two component parts, the one due 
to the base, the other to the radical with which it is combined ; and the multiplication 
of these experiments on a large variety of salts has only served to deepen this conviction. 
The actual observations will be found in the Appendix, but the refraction-equivalents 
thus arrived at are given in the following Table. 
As the determination of the refraction- equivalent of a salt in solution depends on the 
difference between it and the refraction-equivalent of water, it is evident that experi- 
mental errors will be multiplied undesirably if the water be large in quantity as com- 
pared with the salt. Hence the most soluble salts give the most trustworthy results. 
Iu some instances the solubility of the salts depended on the addition of some other salt 
or acid to the solution ; in such cases the refraction due to the salt or acid, as well as 
that due to the water, has been deducted, and in the following Tables the number so 
arrived at has been marked with an asterisk (*). 
