58 A Retrospect of Oceanography 



the carbonic acid. But this was not what was wanted. It 

 subsequently turned out that by the addition of chloride of 

 barium, I had taken exactly the best means of avoiding the 

 action of the chloride of magnesium on the neutral carbonates 

 which vitiated Jacobsen's results; for it had been observed 

 by H. Rose, though at the time I was ignorant of it, that the 

 precipitate formed by chloride of barium in a mixed solution 

 of carbonates and sulphates is a double salt of the carbonate 

 and sulphate of baryta, which is extremely stable even in 

 presence of strong acids, to which it yields its carbonic acid 

 only with great difficulty. It was, therefore, practically 

 quite proof against the action of a weak acid like chloride 

 of magnesium. My results, therefore, gave quite accurately 

 what was wanted, namely, all the half-bound and free carbonic 

 acid, and none of that engaged as neutral carbonate. 



Part of my regular work was to examine the samples of 

 bottom which came up in the sounding tube or the dredge. 

 Amongst them the most interesting were the ochreous deposits, 

 of which the best example is the manganese nodule which was 

 found to be widely distributed over the bottom of the ocean. 



The experience of the "Challenger" was that manganese 

 nodules are found all over the ocean, but principally in the great 

 depths where calcareous deposits are rare or absent. On Sep- 

 tember 23, 1878, while prosecuting oceanographical researches in 

 the Firth of Clyde 1 on the steam yacht "Mallard," which I had 

 built in that year expressly for such work, I found in Loch Fyne 

 in water of about 100 fathoms a rich deposit of mud which con- 

 tained over 20 per cent, of its bulk of manganese nodules, which,, 

 in outward appearance and characteristics as well as in chemical 

 composition, were not to be distinguished from oceanic nodules. 

 This was a very important discovery. Some years later these 

 nodules were found in other parts of the Firth of Clyde. The 

 submarine manganese nodules are a distinct geological forma- 

 tion. Their essential constituent is an ochre, that is, a higher 

 hydrated oxide of one or more of the metals of the iron group. 

 The hydrates of the peroxides of manganese and of iron are 

 present in preponderating quantity, and they are always 

 1 See Paper No. 8, p. 160. 



