’ 406 ve CORALS AND CORAL ISLANDS. 
ocean, just as if the subsidence increased in that direction. 
Finally, the Atlantic beyond is water only, as if it had been 
made a blank by the sinking of its lands. 
Thus the size of the islands, as well as the existence of 
coral banks, and also the blankness of the ocean’s surface, all 
appear to bear evidence to a great subsidence. 
The peninsula of Florida, Cuba, and the Bahamas look, as 
they lie together, as if all were once part of a greater Florida 
or southeastern prolongation of the continent. The northwest- 
ern and southwestern trends, characterizing the great features 
of the American continent, run through the whole like a warp 
and woof structure, binding them together in one system; the 
former trend, the northwest, existing in Florida and the Ba- 
hamas, and the main line of Cuba; and the latter course, the 
west-southwest, in cross lines of islands in the Bahamas (one 
at the north extremity, another in the line of Nassau, and 
others to the southeast), in the high lands of northwestern 
and southeastern Cuba, and in the Florida line of reefs, and 
even further, in asubmerged ridge between Florida and Cuba. 
This combination of the two continental trends shows 
that the lands are one in system, if they were never one in 
continuous dry land. 
We cannot here infer that there was a regular increase of 
subsidence from Florida eastward, or that Florida and Cuba 
participated in it equally with the intermediate or ad- 
joining seas; for the facts in the Pacific have shown that 
the subsiding oceanic area, had its nearly parallel bands of 
greater and less subsidence, that areas of greatest sinking al- 
ternated with others of less, as explained on page 326; and that 
the groups of high islands are along the bands of least sinking. 
So in the Atlantic, the subsidence was probably much greater 
between Florida and Cuba than in the peninsula of Florida it 
