1891.] ^OJ [Brown. 



Besides these two experiments, pyrlte was heated for one hour at a red 

 heat in a stream of the gas. A copious sublimate of ferrous chloride was 

 found in the tube, with a trace of ferric chloride and sulphur. This lime 

 the loss was 46.47 per cent, of sulphur. It seems evident from these 

 experiments that, as above staled, the iron in pyrite is in a condition that is 

 more readily acted on by hydrochloric acid than is the iron in marcasite. 

 It will be pnwed that the iron in raarcasite is all ferrous, while part of that 

 in pyrite is ferric, and this is probably the explanation of the above phe- 

 nomenon. All of the iron in each case described above would form fer- 

 rous chloride (FeCl^) on account of the reducing action of the hydrogen 

 sulpliide formed. Under the conditions of the above experiments, the 

 critical point developed in the oxidation of pyrite was not reached, but it 

 is not likely tliat it exists with this reagent, or if tiiere be a critical point 

 it is not 21 percent. Tiie thought suggested itself to me that perhaps 

 some sulphur would be lost in pyrite if it were heated to d2o^ in a neutral 

 atmosphere, and that this might account for the ditiereuce shown in tlie 

 loss of sulphur in the two minerals. This proved not to be the case. 

 Pyrite healed in this way in an atmosphere of nitrogen gave no appre- 

 ciable loss after one hour at a temperature of 335°. 



Instead of hydrochloric acid gas, the action was tried of ammonium 

 chloride at temperatures up to 835^ and in an atmosphere of nitrogen. 

 Under these conditions the sulphur was combined as ammonium sulpliide 

 probably and did not exert such a reducing action on the iron. These 

 experiments were conducted as follows : 0.3 gram of the finely pulverized 

 mineral was mixed with 0.5 gram dry ammonium chloride and introduced 

 (in a porcelain boat) into a glass tube. Test samples of NaHSO^- HgO 

 and KCIO3 in sealed tubes were used to regulate the temperature. All 

 air was displaced in the tube by nitrogen and a slow current of nitrogen 

 passed tlirough the tube before heating. Under these conditions with 

 niarcasile, sulphur and ammonium sulphide were tound sublimed in the 

 tube along with ammonium chloride, and in the boat there was found 

 much ferrous chloride without any ferric chloride, but in the case of the 

 pyrite there was formed a large proportion of ferric chloride, which sub- 

 limed on the tube towards the end of the operation. The healing was 

 conducted slowly in each case and continued until all ammonium chloride 

 was sublimed from the boat, the temperature of 335° not being exceeded 

 during this time. The entire operation lasted about twenty-five minutes 

 in each case. Three trials of each mineral were made and with the same 

 result in each case ; with marcasite only ferrous chloride was found in the 

 boat and no iron in the tube, pyrite always gave much ferric chloride and 

 little ferrous. The amounts of sulphur removed are probably not very 

 significant ; they showed the following results : 



Pyrite heated with NH^Cl lost sulphur (a) 7.03 



(6) 7.10 



Marcasite " " " («) 9.50 



PROC. AMER. PHILOS. SOC. XXXIII. 145. 2 E. PRINTED JUNE 19, 1894. 



