510 Proceedings of the Royal Society 
the deposits of other parts of the ocean. Another circumstance 
influences the bathymetrical distribution of these surface shells. 
When there is a complete and free oceanic circulation from the top 
to the bottom, these dead shells are found at greater depths in the 
deposits than where the circulation is cut off by submarine harriers. 
The agent by which these shells are removed is, as Sir Wyville 
Thomson suggested, carbonic acid. Analysis shows that carbonic 
acid is most abundant in sea water, and especially so in deep 
water. Pteropod and Heteropod shells are very much larger than 
the Foraminifera, yet are very much thinner ; and hence, for the 
quantity of lime contained in them, they present a much greater 
surface to the action of the sea water. This seenjs to he the 
reason why all large and thin shells are first removed from the de- 
posits with increasing depth, and not the fact that some shells are 
composed of arragonite and some of calcite, as has been suggested. 
There is a continual struggle in the ocean with respect to the 
carbonate of lime. Life is continually secreting it and moulding it 
into many varied and beautiful forms. The carbonic acid of ocean 
waters attacks these when life has lost its hold, reduces the lime to 
the form of a bicarbonate, and carries it away in solution. In all 
the greater depths of the ocean these surface shells are reduced to 
a bicarbonate either during their fall through the water or shortly 
after reaching the bottom. 
In the shallower depths — on the tops of submarine elevations or 
volcanoes — the accumulation of the dead silicious and calcareous 
shells is too rapid for the action of the sea water to have much 
effect. Long before such a deposit reaches sufficiently near the 
surface to serve as a foundation for reef-forming corals, it is a 
bank on which flourish numerous species of Foraminifera, Sponges, 
Hydroids, deep-sea Corals, Annelids, Alcyonarians, Molluscs, Polyzoa, 
Echinoderms, &c. All these tend to fix and consolidate such a 
bank, and add their shells, spicules, and skeletons to the relatively 
rapid accumulating deposits. Eventually coral-forming species attach 
themselves to such banks, and then commences the formation of 
Coral Atolls — Mr Darwin has pointed out that “ reefs not to be 
distinguished from an atoll might be formed ” * on submerged 
banks such as those here described. However, the improbability of 
* Coral Beefs, p. 118. 
