I §73-] Coral Reefs and the Glacial Period. 173 



long ere the beginning of the glacial period in the northern 

 hemisphere, and continued uninterruptedly all through that 

 period, the atolls being slowly built up throughout the 

 whole of that time, and that, as the northern climate 

 became finally milder, they were gradually extended further 

 north. Dana estimates the subsidence in the Pacific area 

 as not less than 6000 feet, and taking the rate of subsidence 

 and the upward growth of a reef as 1 ft. per century, this 

 would give a period of 600,000 years for the formation of 

 the Pacific atolls, without allowing for any time of inter- 

 mittent upward movement, or times during which there 

 might be little or no movement in either direction. The 

 instance of the Florida reefs was also brought forward ; 

 here some 10 reefs, one within the other, have been formed, 

 probably since the ice-sheet disappeared, only 9° farther 

 north, and each reef being taken at 70 ft. in thickness, and 

 the rate of growth as above, a period of 70,000 years is 

 given for their formation. Hence, on the whole, the present 

 distribution of coral reefs, especially of atolls — for the most 

 part south of the equator — would seem to favour the idea of 

 a glacial climate having prevailed in the northern hemisphere 

 at a much more recent period than in the southern. But, 

 on the other hand, it may be argued, that atolls, perhaps, 

 do not occur north of the equator in any abundance, because 

 the requisite sinking land was not present, and this argu- 

 ment may hold good to a certain extent, especially as it is 

 the very existence of atoll reefs that marks in great measure 

 the broad land of subsidence in the Pacific. 



2. Let us now take the case of an extreme glacial climate 

 in the southern hemisphere, and a proportional increase of 

 heat in the northern. 



Agassiz, in the results of his South American Expedition, 

 has just shown that an ice -sheet probably enveloped the 

 southern part of South America, down to the latitude of at 

 least 37 , and even supposing the sheet not to have extended 

 so far as that in the north did, on account of the less amount 

 of continental land round the southern pole, are we not 

 justified in concluding that the southern isotherm of 68° 

 and the equator of heat itself would be shifted considerably 

 north, and the growth of coral reefs rendered as generally 

 impossible south of the equator as they probably were 

 north of it during the undoubted glacial period? Supposing 

 no glacial period to have visited the northern hemisphere 

 from the time of extreme southern glaciation until now, we 

 should have expected to find north of the equator coral reefs 

 of great thickness, and atolls in great abundance, provided 



