Chemical Observations upon Spathic Iron. 521 



In order to ascertain if it really was owing to this earth 

 that the refractory properties of spathic ores are owing, I 

 made the following experiment: — I took equal quantities 

 of the ore of V r aunaveys and that of Allevard which I had 

 analysed, and after having pulverized them, and added 

 enough of oil to make a paste, I introduced them into a 

 lined crucible, and I heated them like a specimen of iron. 

 The specimen of Vaunaveys had been a little more heated 

 than that of Allevard, which, as will be recollected, had 

 yielded a very notable quantity of manganese. The latter 

 was perfectly melted,, and covered with a green scoria, while 

 the experiment with the ore of Vaunaveys only presented a 

 mass of little coherence, through which was diffused a mul- 

 titude of small globules of cast iron. 



This difference could not arise from the state of oxidation 

 of the iron, since in the specimen of Vaunaveys this metal 

 was reduced into globules; but as it might be supposed that 

 the manganese contained in the specimen of Allevard had 

 some influence on its complete fusion, and that it would not 

 have taken place if it had been exempt from it, I took a new 

 quantity of the ore of Vaunaveys, and, after having taken 

 the magnesia from it by means of the nitric acid, I treated 

 the oxide of iron which remained as I had treated the ore, 

 without continuing the fire so long. I obtained from this 

 trial a button covered with a little brownish scoria ; and this 

 button was not only well melted but even a little ductile, 

 and presented in its fracture a gray textare composed of 

 spherical cells like native iron, which, as far as we know, 

 is a first degree of refining in the furnace. 



This experiment left me no more doubt upon the cause of 

 the little fusibility of certain spathic ores. In fact, it is 

 evident that the reduction of the oxide of iron is effected 

 with facility. But the metallic globules cannot unite, owing 

 to the earthv particles not becoming sufficiently fluid, and 

 which therefore choke the furnaces. One may still conclude 

 that the ores are more or less fusible according as they con- 

 tain more or less magnesia; and those which do not contain 

 it at all, and which, on the contrary, contain manganese, 

 pught to be verv fusible 



Vi^. 25. No. too. Sept. 1806. Y But 



