THE PLEISTOCENE OR GLACIAL PERIOD. 441 



But as soon as the great depths were filled, and these super-car- 

 bonated waters themselves rose to the surface in the warm zones, they 

 must have given forth not only the super-charge of carbon dioxide 

 they then retained, but, because the coefficient of absorption was 

 lowered by the rise of temperature, they must have given forth a por- 

 tion of what was their normal content in the cold zone. It is obvious, 

 therefore, that as soon as the new circulation was well established, 

 its output of gas in the lower latitudes must have equaled or sur- 

 passed its intake in the higher, incidental qualifications aside. The 

 circulation was then no longer a source of atmospheric depletion. The 

 whole ocean body had been raised to the higher state of carbonation 

 required by the lower temperature. Not only this, but the process 

 was reversed; for the intake in the high latitudes had been decreasing 

 since the carbon dioxide of the atmosphere had been declining as the 

 result of the very process of loading up the ocean, and the surface- 

 waters that entered the freezing zone were lower in carbon dioxide 

 than they had been at the start, and hence the concentration by freezing 

 was less effective. This was not true of the salts, so far as this process 

 was concerned, and hence the circulation was not effected by the 

 reduced carbonation. At this stage, therefore, the atmosphere began 

 to be enriched in carbon dioxide, and the reverse swing of the oscil- 

 lation was inaugurated. 



If this reasoning be valid, the length of the previous stage of higher 

 carbonation of the ocean becomes a matter of concern. It is prob- 

 able that the deep-sea circulation is affected by other factors than 

 those of low temperatures and increased salinity in the polar regions. 

 It has been thought that the winds of the North Atlantic tended to 

 heap up the waters in the Arctic Ocean, and thus to induce a return 

 current below, in addition to the recognized Labrador current at the 

 surface. While this may be true in this instance, because of the con- 

 figuration of the North Atlantic, it is not obvious that, for the whole 

 world, the pole-ward winds would be more effective on the ocean sur- 

 face than the. opposite winds. Rather might one suppose that the 

 colder air moving equator- ward would, on the whole, flow more largely 

 at the bottom of the atmosphere, and be the more influential on the 

 currents of the ocean. If the winds, on the whole, promote deep-seated 

 circulation from high to low latitudes, they would shorten the periods of 

 carbonation and decarbonation; if the opposite, they would lengthen them. 



