The high sea-level on the Pacific side is caused by the prevailing north-west winds which blow down the North American coast past California as far south as Central America; while, on the other hand, the south-east trade winds impel the surface waters of the South Pacific along the coast of Peru down to the equator, and so onward 5° to 8° north latitude. Thus the space between the ending of the two ocean winds obtains a high sea-level, corresponding to the high level of the Caribbean Sea. This has been proved from levellings for the Nicaragua ship canal.
Consequently, the Atlantic waters would not run into the Pacific Ocean, even if a channel opened through Central America.
Therefore, the Gulf Stream has never been turned away from the North Atlantic.
Writers, while seeking a cause for the mild climate of ages preceding the glacial epochs, have thought that during such times channels opening through Asia from the Indian Ocean by the way of the Persian Gulf into the arctic seas would be the means of furnishing the Arctic Ocean with warm water. But it is evident that such a movement of water could not be brought about, because the winds would not be favorable for it. For, when we reflect that the prevailing winds would blow in the same direction as now, and that the seas of Eastern Europe and Western Asia were enlarged during the warm epochs, it seems that they would obtain high levels superior to the high level seas of the Indian Ocean.
Besides, we should consider that there is a continuous range of high land separating the Persian Gulf from the northern seas, which probably existed anterior to the ice age. Still, during later periods, while the ice-sheets were being melted from the northern hemisphere and also on the ending of the last ice age, the Isthmus of Suez was submerged, as were all other low lands in that latitude; but it is probable that the waters of the high sea-level of the Indian Ocean abreast tropical Africa did not flow largely into the Mediterranean Sea for the reason that the enlarged European seas, being within the westerly wind-belt, maintained a high sea-level, while at the same time the high level tropical Indian Ocean waters were strongly attracted into the southern oceans through the Mozambique and Agulhas currents in the manner I have previously explained. Yet the waters of the high sea-level of the southern European seas must have been strongly attracted to the low sea-level abreast the Canary Islands.
While considering the causes which brought about the glacial periods, it is well to reflect that the natural mode of action which could have produced a frigid age was as extensive as the surface of the globe; and, therefore, any geographical change that would affect only a comparatively small portion of the earth cannot serve to account for ages of warmth which extended over the globe, or for glacial epochs which were separated by warm periods of time, which seem to have affected all lands and seas.
And it appears from the geographical explanations given in preceding pages of the general movements of the winds and currents of the sea how impossible it is for heat to be conveyed to the antarctic latitudes sufficient to prevent the growth of glaciers on their lands while the Cape Horn channel is in possession of its present capacity.
For, as has been shown, this channel furnishes opportunity for the westerly winds to impel the surface waters of the great southern ocean constantly around the globe, and so largely turns away the tropical currents from the high southern latitudes.
Consequently, there seems to be no method yet devised through nature’s mode of action that can carry sufficient heat into the antarctic latitudes to melt the ice-sheets from the southern continent, or even arrest their growth, while the Cape Horn channel maintains its present width and depth.
Therefore, the increase of glaciers and icebergs will slowly continue until a glacial epoch is perfected.