170 CARNEGIE INSTITUTION OF WASHINGTON. 



that such a f ringing-reef must necessarily change into a barrier reef through 

 waste and solution of limestone from its floor, for as the reef-flat deepens the 

 currents must lessen, and a uniformly sandy or muddy bottom is not so favor- 

 able for holothurians as are the small patches of limestone sand largely sur- 

 rounded by dead and broken coral, such as characterize the Aua reef-flat 

 to-day. Thus as the reef-flat deepens the factors which destroy it probably 

 diminish and a balance may be attained between the accession of limestone 

 due to growth of corals and other organisms and its loss due to mechanical 

 and organic causes. 



It is interesting to see that although Pontes heads are more numerous than 

 Acropora, yet Acropora (due to its remarkable growth-rate) is the most impor- 

 tant coral genus in building up the Pacific reefs. 



We find that the Samoan corals grow fully twice as fast as do those of 

 Florida and the Bahamas according to Vaughan. As I have stated elsewhere 

 in this report, this is probably due to the better food-supply in the Pacific. 



Our results explain the disappearance of the Uthothamnion ridge over the 

 shoreward parts of the reef-flat. 



Alkalinity of Surface Waters of the Atlantic and Pacific Oceans, hy 



Alfred G. Mayor. 



The alkalinity of the surface-water of the Atlantic was tested each day at 

 noon upon voyages from New York to Key West and returning in February 

 1918, and also in going from New York to Trinidad, British West Indies, in 

 March and returning in May 1918. The results show that the cold, dull- 

 colored water which drifts down the Atlantic coast of America from the Gulf 

 of St. Lawrence to Cape Canaveral, Florida, is less alkaline than the warm 

 deep-blue water of the so-called Gulf Stream. Thus the alkalinity of the Gulf 

 Stream, tropical Atlantic, and Caribbean Sea ranges from 8.2 to 8.25 PH, 

 being usually about 8.22, while that of the shore current along the coast from 

 New York to middle Florida is about 8.1 PH, ranging from 8 to 8.17. 



Moreover, the shore current is relatively dilute, its salinity ranging from 

 30.77 to 35.71 grams per 1,000 grams of water, being more brackish close to 

 shore than it is along its outer edge near the Gulf Stream. In the Gulf Stream 

 and tropical Atlantic north of the Caribbean Sea the salinity is well above 

 36 grams of salts per 1,000 grams of water, while in the Caribbean it is slightly 

 lower, ranging in our tests from 35.28 north of Grenada to 36.08 near Saba. 

 This lowered salinity may be due to the effect of the great rivers of South 

 America, for the muddy surface current at Mayaro Bay, off the east coast of 

 Trinidad, British West Indies, had a PH of 8.17 at 26.1° C, and its salinity 

 was only 31.85 on March 20, 1918. 



Upon leaving the Gulf Stream and entering the shore current off the Atlantic 

 coast of the United States, the alkalinity of the water suddenly changes from 

 about 8.2 to 8.1 PH or lower, and this easily effected test might serve to inform 

 ship captains of the fact that they had entered the region of the coastal drift. 



Studies in the Pacific were made upon a voyage from San Francisco, Cali- 

 fornia, to Honolulu, and thence to Samoa, and returning in 1917; and also in 

 June 1918, when we steamed from San Francisco to Honolulu and thence to 

 Pago Pago, Samoa. We also went from Samoa to Fiji in August, and from 

 the Fiji Islands to Vancouver, British Columbia, in September 1918. It 

 appears that the cold current which sets southward along the Pacific coast 

 of the United States is relatively acid and of 8.05 PH about 100 miles from 

 shore and 8.17 PH about 1,000 miles out between the American coast and the 

 Hawaiian Islands. Beyond this, in the tropical regions of the mid-Pacific, 

 the water of the prevailing westerly drift is more alkaline, ranging from 8.2 

 to 8.25 PH. 



