of Edinburgh, Session 1879 - 80 . 
511 
so many submerged banks existing in the open ocean caused him to 
reject this mode of formation for atolls. As here stated, recent 
deep-sea investigations have shown that submerged banks are con- 
tinually in process of formation in the tropical regions of the 
ocean, and it is in a high degree probable that the majority of 
atolls are seated on banks formed in this manner. 
Mr Darwin has also pointed out that the corals on the outer 
margin of a submerged bank would grow vigorously, whilst the 
growth of those on the central expanse would be checked by the 
sediment formed there, and by the small amount of food brought to 
them.* Very early in the history of such an atoll, and while yet 
several fathoms submerged, the corals situated on the central parts 
would be placed at a disadvantage, and this would become greater 
and greater as the coral plantations approached the surface. When, 
the coral plantation was small there was a relatively large periphery 
for the supply of food to the inner parts, and also for the supply of 
sediment ; and hence, in small atolls the lagoon was very shallow, 
and was soon filled up. For the same reasons coral islands situated 
on long and narrow banks have no lagoons. An atoll one mile 
square has a periphery of four miles. In an atoll four miles square 
— the periphery increasing in arithmetical progression and the area 
as the square — we have for each square mile only a periphery of one 
mile over which food may pass to the interior, and from which 
sediment is supplied for filling up the lagoon. 
With increasing size, then, the conditions become more and more 
favourable to the formation of lagoons, and as a consequence we 
have no large or moderate sized coral islands without lagoons. Tow- 
net experiments always showed very much less Pelagic life (food) in 
the lagoon waters than on the outer edge of the reef. The lagoon 
becomes less favourable for the growth of all the more massive kinds 
of coral as the outer edge of the reef reaches the surface, and cuts off 
the free supply of ocean waters. Many species of corals die. f Much 
dead coral, coral rock, and sediment is exposed to the solvent 
action of the sea water. Larger quantities of lime are carried 
away in solution as a bicarbonate from the lagoon than are 
* Coral Reefs, p. 134. 
t There are no living corals or shells in some small lagoons, the waters of 
which become highly heated, and in some cases extremely saline. 
