IV DEPTHS AND DEPOSITS OF THE OCEAN 191 



of water surcharged with the substance, whence deposits have 

 taken place on whatever nuclei offered, forming a hard radial 

 aggregation, which would continue to grow until either the 

 solution was exhausted or the supersaturation was relieved by 

 external causes. The shape of the concretion must depend on 

 the shape and number of its nuclei and the evenness of concen- 

 tration in the surrounding solution ; in the ideal case of a small 

 single nucleus and a uniform supply of substance from all sides, 

 the concretion becomes an almost perfect sphere, like the 

 manganese nodules met with in certain localities. 



Iron and manganese depend for the formation of super- Concretions 

 saturated solutions in bottom-waters on the change of valency n\°nganesc. 

 of which these elements are capable. Iron is brought into 

 solution as ferrous bicarbonate by the decomposition of minerals; 

 or again a solution of the bicarbonate may be produced locally 

 by the action of decaying organic matter on ferric compounds. 

 Now ferrous oxide is a base of strength comparable to, but 

 rather less than, that of calcium oxide, and is subject to 

 analogous conditions of solubility as bicarbonate. If oxygen 

 were absent, and it the solubility were diminished, e.g. by with- 

 drawal of carbonic acid, we should expect a deposition of ferrous 

 monocarbonate (such as has often taken place on a large scale 

 on land). As it is, the ferrous solution, diffusing out of the 

 mud, meets with dissolved oxygen, and the change of valency 

 to ferric iron rapidly supervenes. Ferric oxide, however, is a 

 much weaker base, and the hydrolytic dissociation of its salts 

 with a weak acid like carbonic is so complete as to render a 

 ferric carbonate practically incapable of existence in presence of 

 water. That is, the substance now in solution is ferric hydroxide. 

 But this is a vastly less soluble body than ferrous bicarbonate ; 

 therefore the iron in solution is now supersaturated. 



Non - manganiferous ferric concretions are comparatively 

 rare, and have been reported only from the North Atlantic and 

 the polar seas,^ where the terrigenous bottoms are poor in 

 manganese. They attain no great size or hardness, contain 

 much silica, and are rather balls of clay cemented with hydrated 

 ferric oxide. 



As for manganese, the manner in which supersaturated 

 solutions come into being is the same, in2itatis imitaiidis, as in 

 the case of iron. The deposited peroxide has approximately 

 the composition MnO., in deep-sea nodules, but shows notable 



1 Schmelck, Norwegian North Atlantic Expedition, No. IX. p. 52, 1882 ; Eoggild, 

 Norwegian North Polar Expedition, Scientific Results, vol. v. No. XIV. p. T)%, 1906. 



