Basic Principles of the General Oceanic Circulation 575 



atmospheric conditions it was assumed at an early date that there would be a simple 

 water circulation in a vertical plane between the equatorial zone and the polar oceans. 

 This opinion was first expressed by Humboldt (1814, 1845, p. 322) who also offered 

 a more detailed reason for it. He pointed out that the very low temperatures in the 

 deeper water layers at low latitudes could only be regarded as a consequence for the 

 cold water transport in the deeper layers from the poles towards the equator, which 

 would also imply a surface water transport towards the poles. The entire mass of the 

 oceans between the equator and the poles including the water at very great depths 

 would thus be in constant motion, Humboldt considered the differences in density 

 between equatorial and polar water masses as the cause for this closed circulation 

 system. Since the circulation is in accordance with the given temperature distribution, 

 he concluded that the distribution of salinity was not such as to disturb the thermally 

 produced circulation. Humboldt's ideas were adopted by many investigators and for 

 three-quarters of a century formed the basis of a generally accepted view on ocean 

 circulation. Lenz (1847) found that already for small depths, temperatures in the 

 equatorial regions are much lower than in the subtropics, and he concluded that the 

 almost horizontally flowing deep current coming from higher latitudes must assume 

 an upward directed component near the equator. He deduced from this that, sym- 

 metric to the equator, there must therefore be two major vortices in a vertical plane, 

 one on either side of the equator with the cold deep currents rising and merging in the 

 equatorial region; cold deep water would thus be found nearer the sea surface here 

 than further north or south. He found support for his conclusion in the salinity 

 minimum of the equatorial zone. 



Ferrel (1856) took the Coriolis force into account and proposed a modified form of 

 Lenz's vortices limited not in the polar regions, but only in middle latitudes, but 

 followed by another vortex in the polar regions of each hemisphere with a rising 

 movement near the poles. The analogy between the atmospheric and the oceanic 

 circulation is particularly evident in Ferrel's model; he completely ignores the differ- 

 ence due to heating of the ocean from above, and of the atmosphere from below, and 

 also disregarded the effects of the salinity distribution and winds. 



The wide adoption of the thermal circulation theory is due to the circumstance that 

 it has been included in an important oceanographic work of that period by Maury, 

 The Physical Geography of the Sea (1st edition, New York, 1856). Croll (1870-71, 

 1875) refused it, but took another extreme viewpoint, since he regarded each vertical 

 circulation as produced by the wind. Also Carpenter (1870-77) tried to conclude 

 from the "Challenger" observations that a thermal circulation was present. Both agree 

 on the existence of a major vertical circulation and differ only on its cause. 



Detailed analysis by Buchanan (1885) and Buchan (1895) of data from the 

 "Challenger" expedition showed that the actual spatial distribution of temperature 

 and salinity is incompatible with a vertical circulation of the type suggested by Lenz. 

 In all oceans there are alternating layers of different temperature and salinity under- 

 neath a relatively shallow top layer. This excludes the possibility of a single closed 

 circulation system with two vortices syrmnetrically placed on either side of the 

 equator. According to Sandstrom's proposition (p. 491) a thermo-haline circulation 

 is substantially promoted and intensified if the heat source is at a lower level than the 

 cold source, particularly when the effects of heat conductivity and turbulence are of 



