THE CORAL-REEF PROBLEM. 71 



one form or another come to, or beyond, the surface, and real- 

 ize how few of them are in a condition of activity, it is diffi- 

 cult to believe that many of these peaks are to-day in a course 

 of volcanic construction, or that other submarine peaks, scat- 

 tered between these, are undergoing a similar process of for- 

 mation. It seems far more natural to assume that these peaks 

 or islands have been for a long time fully formed, and that 

 they were formed at a time when their relations to the sur- 

 rounding sea were more nearly those which govern the posi- 

 tions of by far the greater number of the active volcanoes of 

 to-day. In other words, they were probably continental or 

 sub-continental, and their present positions are the indices of 

 continental subsidence; the vast mass of overflowing water 

 may have extinguished the fires that at one time supplied the 

 material for eruption. 



The recent discovery of a large number of submarine peaks, 

 whose existence had not previously been even surmised, rising 

 to within a comparatively short distance of the surface, seems 

 to support the general conclusion of subsidence. The sound- 

 ings of telegraph ships indicate that between the latitude of 

 Lisbon and the island of Teneriffe there are not less than seven 

 peaks over which the depth of water varies from not more than 

 12 to 500 fathoms. From the entire oceanic basin it is claimed 

 that there are already known about 300 such "submarine 

 cones, rising from great depths up to within depths of from 500 

 to 10 fathoms from the surface "*. 



Probably the greatest difficulty that lies in the way of the 

 acceptance of the subsidence theory of coral structures is the 

 fact that there are not more islands which are in a condition 

 of semi-formation i. e., peaks, partially submerged and sur- 

 rounded by an encircling barrier reef. This is not an in- 

 superable objection, and might be treated by some geologists 

 in the nature of negative evidence. But the fact is of signifi- 

 cance, and must be taken into account for whatever it may be 



*Murray : Nature, Feb. 28, 1889, p. 425, 



