COMPOSITION OF OCEANIC AND LITTOEAL MANGANESE NODULES. 477 



Also, subtracting 0*61 from 4*06 (the average of the iron and alumina estimations), 

 we find 3*45 per cent, of Fe 2 3 4- A1 2 3 . 



In 50 c.c. of the above-mentioned 250 c.c. solution 0*00676 grammes of iron were 

 found by titration with stannous chloride. This is equivalent to 2*31 per cent, of Fe 2 3 

 in nodules. Subtracting from 3*45, there remains 1*14 per cent. A1 2 3 . 



In the filtrate from the barium carbonate precipitate the manganese was not estimated, 

 but the cobalt was. It was separated by means of sulphide of ammonium and acetic acid, 

 as usual, then precipitated as Fischer's salt with nitrite of potash, and this precipitate 

 dissolved in HC1, and the cobalt reprecipitated with a few drops of carbonate of soda 

 solution. It should be mentioned that, before treatment with sulphide of ammonium and 

 acetic acid, sulphuric acid was added to precipitate the excess of barium ; to the filtrate 

 from BaS0 4 sodium carbonate was added to precipitate both manganese and cobalt, and 

 this precipitate dissolved up and the cobalt separated. "00049 grammes of Co 3 4 were 

 found, representing 0*025 per cent, of Co 2 3 in the nodules. 



E. Estimation of Carbonic Acid. — 0'1792 grammes of nodules were treated with 

 solution of phosphoric acid, the carbonic acid gas liberated passed into baryta water of 

 known strength, and the excess of baryta determined by titration with hydrochloric acid. 

 '01 5905 grammes of Co 2 were found, representing 8*87 per cent, of carbonic acid in the 

 nodules. 



Another estimation was made; 0*1704 grammes of nodules were found to contain 

 0*015285 grammes of carbonic acid gas, that is, 8*91 per cent. 



F. Estimation of Phosphoric Acid. — 0'7162 grammes of nodules were converted into 

 chlorides by treatment with HC1 (the insoluble residue being removed in the usual way), 

 and the solution made up to 250 c.c. 



125 c.c. of the solution was treated with acetate of soda, and the phosphoric acid 

 obtained as a precipitate of ferric phosphate. This precipitate w T as fused with bisulphate 

 of potash and the fused mass dissolved in water. There was just a mere trace of insoluble 

 matter. To this acid solution a very slight excess of pure sodium carbona/te was added 

 to neutralise any free sulphuric acid present, and then excess of nitric acid. The solution 

 was then evaporated down, large excess of ammonium molybdate added, and the solution 

 allowed to stand in a warm place till all the phosphoric acid was precipitated. The 

 precipitate was filtered off and dissolved in ammonia, and the phosphoric acid precipi- 

 tated with magnesia mixture as usual. The precipitate was ignited and weighed. It 

 weighed 0*00319 grammes (Mg 2 P 2 7 ), representing 0*57 per cent. P 2 5 . 



Estimation of Lime and Magnesia. — 120 c.c. of the above-mentioned 250 c.c. solution 

 (equivalent to 0*3438 grammes nodules) were treated with sulphide of ammonium, 

 filtered, the ammonium sulphide in the filtrate destroyed by means of hydrochloric acid, 

 evaporated and filtered from sulphur, and the lime precipitated twice with ammonium 

 oxalate as usual. After the second precipitation the calcium oxalate (which seemed to 

 contain alumina) was redissolved, and the lime precipitated a third time with ammonium 

 oxalate. The precipitate was filtered, ignited, and dissolved in hydrochloric acid. It 



