Then, in order to show the mode in which the general oceanic circulation takes place, he supposes two cylindrical vessels, W and C, of equal size, to be filled with sea-water. Cylinder W represents the equatorial column, and the water contained in it has its temperature maintained at 60°; whilst the water in the other cylinder C, representing the polar column, has its temperature maintained at 30° by means of the constant application of cold at the top. Free communication is maintained between the two cylinders at top and bottom; and the water in the cold cylinder being, in virtue of its low temperature, denser than the water in the warm cylinder, the two columns are therefore not in static equilibrium. The cold, and hence heavier column tends to produce an outflow of water from its bottom to the bottom of the warm column, which outflow is replaced by an inflow from the top of the warm column to the top of the cold column. In fact, we have just a simple repetition of what he has given over and over again in his various memoirs on the subject. But why so repeatedly enter into the _modus operandi_ of the matter? Who feels any difficulty in understanding how the circulation is produced? _Polar Cold considered by Dr. Carpenter the Primum Mobile._—It is evident that Dr. Carpenter believes that he has found in polar _cold_ an agency the potency of which, in producing a general oceanic circulation, has been overlooked by physicists; and it is with the view of developing his ideas on this subject that he has entered so fully and so frequently into the exposition of his theory. “If I have myself done anything,” he says, “to strengthen the doctrine, it has been by showing that polar cold, rather than equatorial heat, is the _primum mobile_ of this circulation.”[77] The influence of the sun in heating the waters of the inter-tropical seas is, in Dr. Carpenter’s manner of viewing the problem, of no great importance. The efficient cause of motion he considers resides in _cold_ rather than in _heat_. In fact, he even goes the length of maintaining that, as a power in the production of the general interchange of equatorial and polar water, the effect of polar cold is so much superior to that of inter-tropical heat, that the influence of the latter may be _practically disregarded_. “Suppose two basins of ocean-water,” he says, “connected by a strait to be placed under such different climatic conditions that the surface of one is exposed to the heating influence of tropical sunshine, whilst the surface of the other is subjected to the extreme cold of the sunless polar winter. The effect of the surface-heat upon the water of the tropical basin will be for the most part limited (as I shall presently show) to its uppermost stratum, and may here be _practically disregarded_.”[78]