332 Oil a new Variety of Wolfram. [May, 



■26"945, the tungstic acid ought to make up 100, that is to say, 

 73*886 ; this comes very near the result obtained by BerzeUus, 

 who estimates the proportion of tungstic acid in woH'ram at 

 74-666. Tiiese 73-885 would contain 34-96 of oxygen, 100 of 

 this acid consequently contain 47-20. 



In making this calculation, I suppose that 100 of protoxide 

 of iron contain 2-2-57 of oxygen, and that it is in the state of 

 protoxide that it exists in the tungstate of iron ; I also allow 

 that 100 of protoxide of manganese contain 28* 107 of oxygen, 

 according to Arfwedson ; these results do not, however, at all 

 agree with the analyses of MM. Delhuyar, Bucholz, and Berze- 

 lius, all of whom allow only a fifth of oxygen in the tungstic acid. 

 I nevertheless beheve that I did not lose any thing in my expe- 

 riments, the metallic mass obtained being very even, and 

 perfectly entire. I shall, moreover, mention how I operated : 

 the wolfram was put into a small charcoal crucible, placed in 

 another of charcoal, and with a stopper of the same material, 

 and these were introduced into a Hessian crucible, and heated 

 in a forge for an hour. 



Not knowing to what to attribute this difference, I imagined 

 that part of the tungstic acid had been volatiUzed. To satisfy 

 myself respecting it, 1 heated some tungstic acid in a moderate 

 fire, but sufficient to reduce it, and I had actually a loss of only 

 20 per cent. ; but having another time submitted it to a very 

 strong and long continued heat, it lost 30 per cent. A portion 

 of the metals was, therefore, volatilized. 



Although I have supposed in my calculation that the iron in 

 the wolfram is in the state of protoxide, yet some appearances 

 which this metal exhibited in the course ot my analyses left some 

 doubts on this subject; I thought it proper, therefore, to subject 

 them to the test of experiment, and the results obtained were as 

 follows ; 



Into a solution of one gi-anune of crystallized sulphate of iron, 

 I poured a solution of gold, as neutral as possible ; there were 

 obtained 22 centigrammes of gold ; I also added solution of gold 

 in excess to the solution of iron obtained by muriatic acid from 

 five grammes of wolfram. The quantity of metallic gold was 

 37 cent. ; but in a gramme of crystallized sulphate of iron, there 

 is about 26 cent, of protoxide, and in five grammes of wolfram 

 there is at least 72 cent, of the same oxide ; thisquantity of iron, 

 if it were in the state of protoxide, according to the above pro- 

 portions, ought to precipitate 76-154 grammes of metallic gold, 

 and it gave only 37 : there is, therefore, only about half the iron 

 contained in the wolfram which is in the state of protoxide. It 

 will be, perhaps, objected, that during the decomposition of the 

 wolfram, which can be effected only with the assistance of ebul- 

 lition, a part of the iron is peroxidized ; but this operation 

 beiug effected by muriatic acid, ^Yhich is concentrated, and in 



