440 GEOLOGY. 



possible may be inferred from the fact that the range of density-effects 

 for a range of temperature of 30° C, is about 0.004, while the range 

 due to salinity may be 0.028 or more. The probable ranges were, 

 however, much less wide apart, and this circulation is not a deduction 

 wholly beyond question. 



The water thus thought to be carried down and poleward from 

 the equatorial regions was carbonated under the conditions of equi- 

 librium then prevalent in the low latitudes. Because of the high 

 temperature there, the carbonation of this poleward flowing water 

 was relatively low, and the main body of the ocean would be sub- 

 carbonated, i.e., carbonated below an ideal equilibrium for the aver- 

 age temperature, for the average content of carbon dioxide in the 

 air, and for the average carbonates in the sea. In the glacial period, 

 when freezing in high latitudes was brought on by the general lower- 

 ing of temperatures, the salts and gases of the sea- water must have 

 been largely forced out of the ice, and passed into the layer of water 

 next below, which thus became super-charged with salts and carbon 

 dioxide. 1 



In being cooled before freezing, the sea-water, under normal con- 

 ditions, absorbed carbon dioxide, because the coefficient of absorption 

 for carbon dioxide was raised by the cooling. The sea-water should, 

 therefore, have been more highly charged with this gas than the aver- 

 age ocean even before the freezing took place, and hence was specially 

 super-carbonated. 



The layer of water below the sea-ice, thus super-carbonated and 

 rendered heavy by super-salinity, tended to descend and flow toward 

 the equator. Thus the depths of the ocean were slowly filled with 

 cold, super-carbonated water, displacing the previous warm, sub- 

 carbonated water. 



1 A portion of the carbon dioxide thus concentrated probably escaped into the 

 air when opportunity was afforded by seams and lanes in the ice, but the greater 

 part doubtless followed the course of the dense water in which it was dissolved. An 

 illustration of the incidental effects of this process is probably given in the exception- 

 ally high content of carbon dioxide found in the air at certain times in Grinnell Land 

 and Greenland. (Moss, Notes on Arctic Air, Proc. Roy. Dublin Soc, Vol. II, 1880, 

 and more fully, Krogh, Abnormal C0 2 percentage in the air of Greenland, etc., Med- 

 delelser om Gronland, Vol. XXVI, 1904, pp. 409-411.) At present the Arctic 

 ice drift is concentrated toward Greenland and the islands* west of it, and the waters 

 below are doubtless more or less carried with the ice and discharge some of their 

 super-charge of carbon dioxide into the air. 



