COMPOSITION OF OCEANIC AND LITTORAL MANGANESE NODULES. 481 



A. Estimation of Moisture driven off at 180°. — 0*8078 grammes of nodules (unex- 

 tracted) were heated in an air-bath to 180°. The heating was continued for about three 

 hours, two hours being between 160° and 180°. The loss of weight was found to be 

 0*0363 grammes (4*50 per cent.). 



B. Estimation of Matter soluble in Water. — 8*8078 grammes were (after the moisture 

 had been driven off) extracted with hot water. The solution was evaporated to dryness, 

 ignited, and weighed. It weighed 0*0169 grammes, that is, 2 "09 per cent, of the 

 nodules. 



The dry residue was redissolved and tested for potash, and found to contain a little. 



C. Estimation of Insoluble Residue, Manganese, Iron, and Alumina. — Treated what 

 was insoluble in water of the above 0'8 07 8 grammes with hydrochloric acid, and separated 

 the silica and silicates as usual. The insoluble residue weighed 0*26106 grammes, equal 

 to 32*32 per cent. 



The filtrate from the insoluble residue was made up to 100 c.c. 50 c.c. of the 

 solution (equal to 0*4039 grammes of nodules) were used for manganese estimation, and 

 0*1349 grammes of Mn 3 4 found. This gives 3 I'll per cent, of MnO. 



The iron and alumina were precipitated by barium carbonate in 100° c.c. of the 

 solution. (It was the filtrate from this precipitate that was used for the manganese 

 estimation.) The precipitate was redissolved in hydrochloric acid, the baryta present 

 precipitated with sulphuric acid, and iron and alumina precipitated from the filtrate from 

 the barium sulphate. 0*0146 grammes of Fe 2 3 + A1 4 3 were found, representing 3*60 per 

 cent, in the nodules. This was fused with bisulphate of potash, and found free from 

 silica. 



To make the figures comparable with those of X, we have 2*09 soluble matter and 

 4*05 moisture, together 6*14 ; from which we deduct 1*47 moisture in X, leaving 4*67 ; 

 whence 95 33 of Z are equivalent to 100 of X. We must therefore multiply the figures 

 obtained in Z by 1*049. We have then MnO = 32*62 per cent., insoluble residue = 33*90 

 per cent., and Fe 2 3 + Al 2 3 = 3*77 per cent. After making all reductions we have — 



Insoluble Residue. — In X 34*05 per cent., and in Z 33*90 per cent. As the soluble 

 silica was determined in X, and the value of the total residue in Z agrees closely, we 

 take the values for X, namely, soluble Si0 2 6*32 per cent., and remainder of residue 

 27*73 ; together, 34*05 per cent. 



Fe 2 3 and A1 2 3 . — As the figure found in Z for the sum of these constituents agrees 

 sufficiently with that found in X, we take the value for the individual constituents found 

 in it, namely, Fe 2 3 , 2*31 per cent., and A1 2 3 , 1*14 per cent. 



Manganese. — The results obtained in Y and Z agree well, and we take the mean, 

 3.2'43 per cent, of MnO. 



Loss on Ignition. — The amount of substance taken for this determination was 

 0*1263 grammes. Eecent experiments show that it would lose about three-sevenths of 

 its available oxygen on ignition, and it may therefore be safely concluded that all the 



