Value of Black Oxide of Manganese. 415 



Loss of weight 35 grains. Here also the error was in 

 excess, and amounted to 0-755 grains. 



Let us take the mean of these three experiments : 

 Loss of weight by 1st . . . 32-5 grs. 

 2nd . . 34-5 

 3rd . . 35-0 



3)102 



Mean 34 grains. 



Here the error amounts to 0-245 grains, which is consi- 

 derably less than 1 per cent. If, therefore, three trials 

 be made, the error will be under 1 per cent. ; so that the 

 method is quite sufficient to indicate very nearly the quan- 

 tity of binoxide of manganese in any ore. Now, it is the 

 binoxide of manganese alone that is useful to the manufac- 

 turer ; the sesqui-oxide and red oxide availing very little m 

 the preparation of chlorine, for which almost alone the ore 

 is used by manufacturers. 



I tried various other proportions of the ingredients, but 

 found the preceding the best. I tried, also, the effect of 

 rubbing up in a mortar the oxalic acid and black oxide. 

 But the error is least when the oxalic acid is merely poured 

 into the water, and the black oxide added before the acid is 

 dissolved. Unless the sulphuric acid be added last, we 

 cannot be sure of our weights. 



Article III. 



Description and Analysis of Emmonite, a new Species of Car- 

 bonated Strontian from America. By Thomas Thomson, 

 M. D., F.R. S., L.andE. Professor of Chemistry, 

 Glasgow. 



The only specimen of this mineral which I have seen I got 

 from Mr. Sowerby, who had received it from Massachu- 

 setts, under the name of carbonate of strontian, sanctioned 

 by the authority of Professor Emmons, of William's Col- 

 lege. It was obvious, from the external characters of the 

 mineral, that it was specifically different from all the va- 

 rieties of carbonate of strontian that I had seen. I there- 



