260 ANNUAL KEPOET SMITHSONIAN INSTITUTION, 19 3 2 



lies deeper on the right side (in the northern hemisphere). In the 

 case of the Gulf Stream, this means that the warm water must lie 

 deeper on the southeastern side. In other words, half of the current 

 consists of relatively cold water and half of relatively warm water, 

 because the isotherms slope sharply across its path. In the case of 

 the 10° isotherm, for example, this means a slope of from 200 meters 

 on the western side of the stream to 800 meters on the offshore edge, 

 the current being about 50 miles wide off New York. The Gulf 

 Stream then is not a river of warm water flowing through colder 

 seas, but the boundary between a body of relatively cold water and 

 the mass of warm, central, Atlantic water. It follows then that the 

 Gulf Stream, by which we mean the whole of the convectional cur- 

 rent along the eastern North American seaboard and not just the 

 surface layer, does not transport the warmest water northeastward. 

 Even at the surface, the warmest water is not over the stream unless 

 displaced by easterly winds. From a technical point of view, it is 

 wrong to think of the Gulf Stream as being a river of warm water, 

 although it is the existence of warm water in the southern North 

 Atlantic which is its cause. The current, because of the earth's 

 rotation and the stable arrangement of the water layers, only ac- 

 complishes what it sets out to do in a most inefficient manner. 



But besides this horizontal circulation, at the same time there exists 

 much slower vertical movements. Some of the cold, relatively heavy 

 northern water sinks and gradually moves southward along the 

 bottom, only to rise again near the equator. That these two sys- 

 tems exist can be clearly and easily demonstrated from the observa- 

 tions already recorded, and some progress has been made by study- 

 ing the pressure field of the ocean in much the same way as is done 

 in meteorology. The region south and east of Bermuda, known as 

 the Sargasso Sea, corresponds to a permanent high-pressure area on 

 a weather map. Such southward-moving tongues of cold water as 

 the Labrador current can be compared to the polar fronts of the 

 meteorologists. But sea water, in contrast to air, is nearly incom- 

 pressible and is usually found to be well stratified and therefore 

 (luite stable. Moreover, it is out of the question for oceanographers 

 to secure any such instantaneous picture of the pressure field as in 

 a weather map and besides it is not known to what extent the ocean 

 circulation is wind driven. 



Thus it is an unfortunate fact that as we look more closely into 

 the ocean currents and attempt to find reasons for their paths and 

 characteristics, we must admit that this phase of oceanography 

 is at a most awkward stage. The existing observations are too few 

 to explain more than the broadest picture, and as each expedition 

 brings back modern data, that is, sections made up of closely spaced 



