of the Oxides of Manganese. 99 



From the result of both analyses it is apparent that man- 

 ganite, in relation to manganese and oxygen, is a deutoxide. 



Also as 89*90 : 10-10 :: 40 : 4-494.. 

 The fourth number is so near 4*5, half an equivalent of water, 

 that we may safely regard manganite as a compound of 80 

 parts or two equivalents of the deutoxide of manganese, and 

 9 parts or one equivalent of water. 



The material for the preceding analysis was taken from a 

 very fine crystallized specimen from Ihlefeld. The result of 

 Gmelin's analysis of the same variety is as follows : — Red oxide 

 87*1, oxygen 3*4, water 9*5. The water is here certainly un- 

 derrated. 



The grey oxide from Undenaes in West Gothland, analysed 

 by Arfwedson, is a similar compound. 



Analysis of the Brachytypous Manganese-ore or Braunite, — 

 The colour of this ore, both in mass and in powder, is nearly 

 black. With sulphuric acid it yields no distinct odour of 

 chlorine. It dissolves in muriatic acid, leaving a trace of si- 

 liceous matter. The solution gives a precipitate of sulphate 

 of baryta with sulphuric acid, but does not contain any other 

 impurity. Of all the native oxides this is the most easily re- 

 duced to the state of protoxide by the action of hydrogen gas. 

 The material for analysis formed part of a specimen from 

 Elgersburg. 



As a mean of two closely corresponding experiments, this 

 oxide contains 0*949 per cent of water. 



To ascertain the quantity of oxygen, 16*634 grains were 

 exposed for half an hour to the action of hydrogen gas at a 

 red heat. The residue weighed 14-837 grains, and had the 

 light green tint of the protoxide. The total loss was 1*797 

 grains, or 10-80 percent; and subtracting 0*949 for water, 

 there remains 9*851 per cent as the loss in oxygen. 



The baryta was precipitated by sulphuric acid from a solu- 

 tion in muriatic acid of 42*09 grains of the mineral. The pre- 

 cipitate after being heated to redness amounted to 1*44 grains, 

 equivalent to 0*951 of a grain or 2*26 per cent of pure baryta. 

 According to this analysis, 100 parts of the ore contain 



Protoxide 86*94 



Oxygen 9*851 



Water 0*949 



Baryta 2*260 



Silica a trace. 



100*000 

 Now 86*94 : 9*851 : : 36 : 4*079 ; and as the presence of wa- 

 ter and baryta, from the small quantity of these substances, 



O 2 must 



