374 
THE VOYAGE OF H.M.S. CHALLENGER. 
first interpretation, and we shall endeavour to show that this %dew accords best with all 
that is known with respect to the distribution and conditions under which these ferro- 
mang-anic concretions are found at the bottom of the ocean. 
It may be accepted as a fact that all the hydrated oxides of manganese found at the 
surface of the globe as coatings, concretions, or depositions, as well as the greater part of 
the hydmted oxide of iron, have their original source in the decomposition of the 
crystalline rocks containing these two metals in the form of silicates or anhydrous 
oxides. This decomposition takes place generally under the action of water, almost 
always containing carbonic acid. It is not necessary to insist upon the wide distribution 
of iron in volcanic rocks ; manganese is rarer in these rocks, but is found as a constituent 
of pyroxenic, amphibolic, and peridotic minerals, and also of some varieties of magnetic 
and titanic iron, which are all so abundantly distributed on the bottom of the sea. 
It is well known that, under the influence of water charged with carbonic acid, such 
silicates are decomposed at ordinary temperatures, and that the contained protoxides of 
iron and manganese are transformed into soluble bicarbonates, leaving a residue in which 
the hydrated silicate of alumina plays a large part. As soon as the loosely-combined 
carbonic acid is given oflf from the bicarbonates, the carbonates are precipitated and 
rapidly absorb oxygen, becoming hydrates corresponding to higher oxides of the metals, 
the remaining carbonic acid being of course set free. The metals are now in the form of 
compounds which are insoluble in water. Such is the order in which the formation of 
manganiferous and ferruginous minerals takes place on land surfaces, forming coatings, 
concretions, and dendrites on ancient and recent and other eruptive rocks and minerals. 
In the case of the manganese deposits of the deep sea, there are many indications 
that an identical interpretation of the phenomena may be given. It may be granted 
tliat the manganese in solution in the superficial layers and in the whole mass of the ocean 
might contribute a small part to the deposits of great depths, still the great bulk of the 
oxides of manganese in these deposits has evidently had another origin. The greater 
}»art of the manganese and iron of these deposits almost certainly comes directly, along 
with clay, from the alteration of the manganese and iron-bearing silicates scattered 
over the bed of the ocean, and especially from those of volcanic origin. In describing 
the rocks and mineral particles, and in speaking of the formation of clay, we have 
sliown that the alteration which these substances undergo in sea-water may be compared 
in all respects with the general processes just indicated. We have laid special stress on 
the occurrence in the deposits of volcanic materials, generally in a vitreous condition and 
of a spongy nature, or in microscopic fragments. On the other hand, we have shown 
tliat nearly all these rocks and minerals are decomposed or in process of alteration, and 
that they nearly all contain protoxides of iron and manganese combined with silicic acid, 
a.s for instance in augites, hornblendes, magnetite, and the easily alterable basic volcanic 
glasses. It is known besides that sea- water contains free or loosely-combined carbonic 
