68 G. J. BURROWS AND C. E. FAWSITT. 



It does not appear likely that the oxygen and water 

 together act simultaneously on the iron. It is more prob- 

 able that the iron first dissolves to a very limited extent 

 in the water. It is known to be capable of dissolving to a 

 small extent in water as ferrous ion, displacing at the 

 same time an equivalent quantity of hydrogen. The 

 hydrogen in the absence of free oxygen polarises the iron 

 surface and the function of the oxygen is to remove the 

 layer of polarising hydrogen. According to this view of 

 the rusting process, iron should dissolve faster if it be in 

 contact with a more electro-negative metal like platinum, 

 for then the hydrogen would tend to collect more on the 

 platinum plate than on the iron one. The acceleration of 

 corrosion caused by contact with platinum is shown by 

 the result of the experiment given in Table I. 



Uniform steel strips weighing about 64 grams were cut 

 from steel sheet of the composition : — Oarbon 0*35, man- 

 ganese 0*61, phosphorus 0*06, silicon 0*01, sulphur # 01°/°. 

 They were fully immersed in unstirred distilled water. 

 Two of the strips were connected to strips of platinum. 

 The surfaces of all the strips of steel were approximately 

 the same. Also the temperature of experiment may be 

 regarded as being the same for these comparative experi- 

 ments. (Room temperature.) 



Note as to the Method of Experiment. — In all the 

 experiments given here, those in any series were placed 

 together in one part of a room. The temperature of experi- 

 ments was the temperature of the room and therefore 

 was not constant, but it varied in the same manner for 

 each individual test in any series. The pieces of steel in 

 comparative experiments were always placed wholly under- 

 neath the water and at the same distance from the surface. 

 The corrosion was determined by removing the steel after 

 an appropriate time, scrubbing the surface well with a 



