182 Dr. Wilson on the Decomposition of Water by Platinum 



hot globules of oxide of iron, was hydrogen mingled with a 

 small quantity of air, previously no doubt in solution in water. 



As only the hydrogen, then, of the water decomposed was 

 obtained, it became necessary to account for the absence of 

 the oxygen. I was tempted for a moment to think it pos- 

 sible that the black oxide of iron might have changed into the 

 red oxide of the same metal, by combining with the oxygen 

 not obtained in the elastic form : ex. gr. thus 2 Fcg 04 + = 

 3Fe2 O3. 



But the proto-peroxide of iron is known to be a very stable 

 compound, little if at all prone to become the peroxide ; and 

 it seemed more likely that unoxidized iron might be present 

 in the fused globules, which occasioned the evolution of hy- 

 drogen when it came in contact with water. To ascertain this 

 point, portions of the globules were dissolved in dilute muriatic 

 and sulphuric acids, and were found in most cases to evolve hy- 

 drogen. Some specimens of the globules gave off not a trace 

 of gas when they dissolved, and must have consisted of the 

 definite oxide ; a point of interest in connection with the 

 fact already mentioned, that globules were frequently ob- 

 served to drop into water without any bubbles of gas rising 

 from them. 



The volume of hydrogen however given off in some of the 

 trials, when the product of combustion was placed in acid, 

 was very considerable. A graduated gas jar was filled with 

 dilute sulphuric acid, and inverted over a small capsule con- 

 taining 100 grains of the crushed globules, which was placed 

 in a basin also containing dilute acid. By this arrangement 

 the gas was collected and measured at the same time, Mithout 

 risk of mixing with air, or necessity for watching the process, 

 which is a slow one. 100 grains treated in this way gave off 

 16 cubic inches of hydrogen, corresponding to 9 grains of 

 iron. The experiment was accidentally stopped at this point 

 whilst the gas Mas still rising in undiminished quantity. 



Metallic iron, then, was certainly present in many of the 

 globules, and of this I had direct ocular demonstration. On 

 crushing some of them in a mortar, they were found to sepa- 

 rate into a shell of pulverizable oxide, and a core of iron 

 which formed a nearly spherical pellet. In one case 50 grains 

 of the globules were crushed, the pellets separated, and the 

 residue placed in diluted sulphuric acid. It did not evolve a 

 trace of hydrogen in the course of twenty-four hours. The 

 pellets were then added to the same acid, and gave off 12 

 cubic inches of gas = 13*6 per cent, of iron in the globules*. 

 The shell of oxide is frequently imperfect or perforated, so 



• In none of the experiments was the thermometer or barometer spe- 

 cially observed, as minute accuracy was not aimed at. 



