1890. | NEW YORK ACADEMY OF SCIENCES. 13 
of earth in the bottom, in which the natives plant their bananas, 
and hence the name. 
In shape and dimensions these holes vary greatly. Some are 
cylindrical, about four feet in diameter, but at the same time 
twenty feet or more in depth. Others are from ten to twenty- 
five feet across, and some even larger, and often of an irregular 
shape and much longer than wide. 
The walls are often excavated below, so that the side becomes 
an overhanging ledge and forms a small cave. In some a cave 
begins at the side of the hole and runs backward. It is hence 
hard to draw the line between these holes and the caves. There 
are also holes that are not called banana-holes, but which may 
be here described, as they differ only in shape. In some the 
opening is barely large enough to allow aman to pass. One 
such I descended, and found that below it was over five feet in 
diameter and cylindrical. The top had been excavated so as to 
form a domed roof. 
Other holes were connected by a horizontal passage through 
which I could craw] from one to the other. One of these I saw 
near Conch Sound, where the passage ran from the bottom of 
one hole to the side of the other, which was much deeper. 
Near by I saw two shallow holes that were connected bya 
horizontal passage, so that they resembled a large tube bent up 
at each end. It is not unusual to find openings in the ground, 
barely large enough to admit an ordinary pail, and sometimes 
much smaller. These are simply openings in the roof of a cave 
or hole of unknown dimensions, and frequently in the bottom 
is a quantity of fresh water that is used by the people. 
The subject of banana-holes has been briefly discussed by Dr. 
C. 8. Dolley,* who accounts for their formation by ‘ the action of 
decaying vegetable matter, that undergoes fermentative changes 
by the products of which the soft, calcareous rock is dissolved 
and leaches away.” There is no doubt that the rock is in many 
places eroded in this manner, as the small, saucer-shaped depres- 
sions so common on the surface, and each often containing 
leavesand water, plainly testify. But I doubt if this agent alone 
would cause the deep vertical cylindrical holes, or those in 
which the sides recede into caves or the horizontal passages. 
And if the holes were formed in the manner described by Dr. 
Dolley, should we not find them in the low level land as well as 
on the ridges? But, as stated before, the holes are found in far 
greater number on the ridges, and in places where the surface is 
such as to indicate that formerly the erosion from the waves was 
1Proc. Acad. Nat. Sci., Phila., 1889, p. 182. 
