A 
934 The American Naturalist. [October, 
the results of those voyages, the most important (excepting, perhaps, 
the influence of the experiences of travel upon the growing minds of 
the great naturalists themselves) was the theory of the origin of barrier 
reefs and atolls, independently developed by Darwin in the study of 
the coral formations of the Indian Ocean, and by Dana in the study 
of those of the Pacific. 
The problem of barrier reefs and atolls may be briefly stated. The 
reef-forming corals grow only in shallow water, seldom, if ever, reach- 
ing a depth of much more than one hundred feet. Their skeletons are 
broken by the waves, and their comminuted fragments consolidated 
into the reef rock. A belt of reef is thus naturally formed immediately 
adjoining the shore of a continent or island, or separated from the 
shore by a shallow channel, having a width of a mile, more or less. 
Such a channel is very apt to exist, since the water immediately in con- 
tact with the shore is apt to be too impure for the luxuriant growth of 
corals, while the conditions are most favorable for such growth on the 
outer margin of the reef. Such a reef is called a fringing reef, and its 
formation presents no perplexing problems. But the case is very dif- 
ferent with the barrier reefs, which are separated from the shore of the 
continents which they border, or of the islands which they surround, by 
a deep channel, ten, twenty, one hundred, or more miles in width; 
and with the atolls, which are more or less irregular rings of reef rock, 
entire or broken by channels, standing out in mid-ocean away from 
any land, and enclosing lagoons, which are sometimes small and shal- 
low, but which sometimes have a diameter of scores of miles anda 
depth much exceeding the limit of coral growth. ' On the outside of 
both barrier reefs and atolls the bottom generally slopes off pretty 
rapidly into truly oceanic depths. 
The solution of the problem was given by Darwin and Dana ina 
theory which may be expressed in one word,—subsidence. Darwin had 
the priority in the formulation and publication of his views, and the 
ory is most commonly called by his name; but Dana’s work was 
equally independent, and he was able to illustrate the theory with a 
much more extensive series of observations than Darwin had the oppor- 
tunity to make. It seems, therefore, most just to link together the two 
illustrious names, and to call the theory the Darwin-Dana theory. Let 
the earth’s crust in a region of coral reefs undergo a subsidence not 
more rapid than the rate of coral growth, and fringing reefs will ob- 
viously be converted into barrier reefs. Since the corals always grow 
most rapidly at the outer margin of the reef grounds, the inevitable 
effect of subsidence will be the widening and deepening of the channel 
