189o.] Recent Literature. 935 
between the reef and the shore ; and this is exactly what is required to 
transform a fringing into a barrier reef. If the reef was formed 
around an island, continued subsidence will suffice to convert the bar- 
rier reef into an atoll, the original island disappearing beneath the 
waters of the lagoon. Since the outer wall of the reef, though some- 
what steep, is never precipitous, subsidence still continued after the 
formation of an atoll must diminish the size of the atoll and tend to 
obliterate the lagoon. The last stage of a coral island disappearing by 
continued subsidence is therefore a mere dot of coral roc 
The theory, in its charming simplicity, reminds one oF Columbus's 
egg. It accounts for the facts with a marvellous perfection. And the 
correlation which it traces between the formation of those 
“ Summer isles of Eden lying in dark purple spheres of sea," 
and the vast crustal movements involved in continental changes of 
level in late Tertiary and Quaternary times, opens one of those glimp- 
ses of the unity of nature which give to scientific speculation a poetic 
sublimity. The Darwin-Dana theory was immediately and well-nigh 
universally accepted. Probably the first thought of almost every 
geologist, when the theory was announced, was, ** Why did I not think 
of that myself ? ”’ 
The subsidence theory, however, has had a rather curious history. 
After an undisputed reign of a third of a century, its title to the throne 
has been recently questioned, and, in the judgment of some of the 
ablest geologists, its days are numbered. The Duke of Argyll, who often 
(as King James said of Lord Bacon) ‘‘ writes of learning like a Lord 
Chancellor," has made the assumption of the certain falsity of the sub- 
sidence theory the basis of a wholesale charge against the morality of 
scientific men, alleging that the majority of geologists have formed a 
** conspiracy of silence’’ to suppress the new views, in order to main- 
tain before the public the infallibility of the idolized Darwin. It is, 
then, with a special interest that we turn, in this new edition of Dana's 
classical work, to the pages in which he deals with the recent discussion 
of the subject, and, after the mature deliberation of a half-century, 
defends the theory whose discovery was one of the earliest of his great 
achievements 
It is tmdeibtediy true that both barrier reefs and atolls may be 
formed without subsidence. If the water off the shore of a continent 
or island deepens very gradually, the water near the shore line may be 
too impure for coral growth, and the favorable conditions for ch 
growth may be only attained at a distance of several miles d. 
$ 
