192 On the Composition of Oceanic 



alumina. In the filtrate precipitated the manganese, as usual, 

 as MnO 2 by means of bromine, filtered off the MnO 2 , dissolved 

 it in hydrochloric acid, removed any baryta as BaSO 4 , evapor- 

 ated solution to dryness, took up with water, and precipitated 

 the manganese with sodium carbonate as MnCO 3 . Filtered, 

 ignited to Mn 3 O 4 , and weighed. The Mn 3 O 4 weighed 0-19646 

 gramme (or 33-29 per cent.). This is equivalent to 30-96 per 

 cent, of MnO (including cobalt) in the nodules. 



BaCO 3 Precipitate. Dried this precipitate, redissolved it in 

 hydrochloric acid, and precipitated the baryta with sulphuric 

 acid. Evaporated the solution over water-bath to get rid of 

 excess of acid. Took up with water, filtered from traces of 

 insoluble matter, diluted, and precipitated with a slight excess 

 of ammonia, boiling till all excess was expelled. The precipitate 

 of Fe 2 O 3 + A1 2 O 3 when ignited weighed 0-0266 gramme. 

 Fused it with bisulphate of potash in order to see if silica was 

 present. 0-00194 gramme of the fused mass was insoluble, 

 leaving 0-0246 gramme of Fe 2 O 3 + A1 2 O 3 , i.e., 4-17 per cent. 



To make the figures obtained for the alkalies and manganese 

 comparable with those obtained in X, we have in both cases 

 used substances extracted with water, and therefore free from 

 soluble matter ; but in X the substance was used dried at 100 C., 

 while in Y it was exposed to the air after extraction, and was 

 therefore in an air-dry condition. X contained 1-47 per cent, 

 moisture, which was driven off at 162 C. ; Y contained 5-53 

 per cent., the difference being 4-06. Therefore 95-94 parts of 

 Y are equivalent to 100 parts of X, and we must multiply the 

 results in Y by 1-042, and we have MnO = 32-23 per cent., 

 K 2 O = 0-47 per cent., and Na 2 O = 0-43 per cent. 



Some of the same air-dried samples of nodules were examined 

 without being extracted with boiling water, or rather the 

 extraction formed part of the analysis. 



A. Estimation of Moisture driven off at 180 C. 0-8078 

 gramme of nodules (unextracted) were heated in an air-bath 

 to 180 C. 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 gramme (4-50 per cent.). 



B. Estimation of Matter soluble in Water. 8-8078 grammes 



