COMPOSITION OF OCEANIC AND LITTOEAL MANGANESE NODULES. 467 



A. The nodules were reduced to an impalpable powder in an agate mortar, and were 

 not dried previous to analysis. 



B. In two separate portions (of about 2 decigrams each) the available oxygen was 

 determined by Bunsen's iodometric method. 



C. Two portions, of about 0*5 grams each, were dissolved in about 5 c.c. of pure 

 strong hydrochloric acid, in a small covered beaker, heat being applied cautiously at first 

 to prevent loss by effervescence of chlorine. The mixture was then evaporated to 

 dryness over a water-bath, covered with a glass shade ; the residue, when free from all 

 excess of acid, moistened with a few drops of strong hydrochloric acid, and the insoluble 

 residue filtered off, ignited, and weighed. 



(a) Insoluble Residue. — This insoluble residue was kept in a corked tube, and after- 



wards fused with (NaK)C0 3 (in the proportion of about four parts of carbonate 

 to one of residue) for the estimation of silica. The filtrate from the silica was 

 examined qualitatively for the bases present. 



(b) The filtrate from the insoluble residue was treated with a moderate excess of 



"pure carbonate of baryta. In practice about 2 grams were required. This 

 was added in a state of cream to the liquid in a small flask. The flask was 

 corked and allowed to stand some hours, being shaken occasionally. The 

 liquid was then filtered off, and the precipitate, after thorough washing, tested 

 for manganese with carbonate of soda on platinum foil. 



D. To the filtrate from the carbonate of baryta precipitate excess of sodium acetate 

 was added, so that there should be not less than 3C 2 H 4 2 to HC1 in the solution (in 

 practice 15 to 20 c.c. of a 1 in 10 solution of NaA were added). Before adding the 

 acetate it is well to drop in a little acetic acid, otherwise an immediate precipitation of 

 barium salt occurs. Bromine in excess was then added to the sufficiently dilute fluid, 

 and the flask corked and placed in a moderately warm place for some time. Gentle 

 warming at first greatly aids the precipitation of the Mn0 2 . When the manganese had 

 begun to deposit, and the liquid in the flask was at a temperature of about 50°, it was 

 removed to a colder place (at the ordinary temperature of the room), and allowed to stand 

 over-night. If necessary, a little more bromine was first added ; this is regulated by 

 practice. Next day the flasks were heated in order to drive off all superfluous bromine, 

 a few drops of alcohol being added towards the end to get rid of the last traces. When 

 the solution became perfectly colourless it was filtered, all the Mn0 2 that came easily out 

 of the flask being thrown on the filter and washed with boiling water. The filtrate was 

 tested first for more Mn0 2 with NaA and bromine, and then for nickel and copper (vide 

 below). 



(a) The filter, on which was the bulk of the Mn0 2 precipitate, was dried. When 

 dry, the Mn0 2 was returned to the original flask (on the sides of which some 

 Mn0 2 remained sticking), the filter was ignited, and its ash also dropped into 



VOL. XXXVI. PART II. (NO. 17). 4 A 



