260 Messrs Burton and Phillips, 



Susceptibility of Iron in Colloidal Solution. By E. F. Burton, 

 Emmanuel College, and P. Phillips, Emmanuel College; 1851 

 Exhibition Scholars. (Communicated by Professor J. J. Thomson.) 



[Read 29 January 1906.] 



Introduction. 



Some years ago Townsend* made an exhaustive series of 

 experiments on the susceptibility of aqueous solutions of various 

 iron salts of both the ferric and ferrous type and was led to 

 formulate two empirical formulae connecting the susceptibility 

 of any such solution with the quantity of iron per cubic centi- 

 metre in the solution. If k is the susceptibility of a solution, w 

 the weight of iron per c.c, then, taking the susceptibility of pure 

 water equal to — 7*7 x 10 -7 , we have for ferric salts 



10 7 k = 2660 w- 7-7, 

 and for ferrous salts 



10 7 k = 2060 w- 7-7. 



Further Townsend found that, roughly, the magnetic properties 

 of the various salts in the dry state are the same as when they 

 are in solution. 



These results show clearly that the magnetization is entirely 

 due to the iron, and is accurately the same whatever the acid 

 radical connected with it so long as the iron remains either in the 

 ferrous or ferric state. Assuming a similar formula to hold for 

 pure iron in a solution, a simple calculation shows that the 

 constant coefficient for ' w' would be much larger than that in 

 the above equations. Again, Townsend found that when iron 

 appears in the acid radical, e.g. in ferrocyanides, its permeability 

 is less than -^ of the value when the iron plays the part of a 

 metal in a salt. From these results we may conclude that, so far 

 as its magnetic properties are concerned, iron exists in four 

 different states : (1) as pure iron, (2) as ferric iron in salts, (3) as 

 ferrous iron in salts, and (4) in the acid radical in salts. 



Iron exists in still another form — in colloidal solution — for 

 which determinations of magnetic properties have been wanting. 

 The purpose of the experiments recorded in the present paper 

 was to find the susceptibility of a colloidal solution of iron in 

 methyl alcohol and to deduce the formula analogous to Townsend's 

 for iron in this state. The results show that the iron in the 

 colloidal solution is over thirteen times as magnetic as that in 

 ferric (or ferrous) salts — a conclusion which has an important 

 bearing on the theory of the constitution of colloidal solutions. 



* "Magnetization of Liquids," Phil. Trans. 187, A, 1896, p. 533. 



