100 The American Geologist. Fib. isoo. 
of the coast lines of the Bermudas have a general application 
to a classification of ring-shaped coral islands. 
While it is customary to call all ring-shaped coral islands 
atolls, it seems to the author that the practice has led to a 
misconception in assigning the same causes for their similar 
shapes. Ring-shaped coral islands may be true atolls, as 
those of the Pacific, or ring-shaped like the Marquesas group 
of the Florida Keys. Still others may be typified in ring- 
shaped islands derived from previously existing coral islands 
by erosion. Of these Bermuda is a type and an excellent 
illustration. What its original form was when it emerged from 
the sea we do not know. Speculation and comparison with 
other coral islands would lead us to regard it as circular. But 
since the appearance above the surface of the sea it has been 
profoundly affected by other forces which have greatly mod- 
ified it and changed its contours. The island has passed 
through a geographic development or a life-history in which 
the work of erosion fills much of the last chapter, for coral 
islands have an initial form characteristic of growth and a 
final form resulting from erosion. The former are coral 
islands in the process of active development ; the latter of 
denudation or decay. 
NOTE ON SOME OF THE CAUSES OF EXTINCTION OF 
SPECIES. 
J. M. McCreery, Akron, 0.* 
Facts concerning the life history of the various forms of life 
whose fossil remains are found in every stratum of the more 
than 5,000 feet of limestones, shales and sandstones that make 
up the geological column as exposed in Ohio, are always of 
interest to students of Ohio geology. 
Some species made their appearance and continued to exist 
while hundreds of feet of sediment were deposited. Other 
species seem to have come abruptly upon the scene and after a 
brief existence — geologically speaking — but which represents 
vast cycles of time as men reckon years — disappeared, and were 
replaced by later forms. The list of American palaeozoic 
fossils now numbers not less than 10,000 species and is being 
*ThiR paper is extracted from an address delivered before the Akron 
Scientific Chib on Jan. 8, 1890, on "Losers in Life's Race." 
