Moreover, according to the statements of General Cowell, the present period of mildness should be on the increase, and obtain perfection in the year 2,295, or about 400 years hence; while, on the contrary, according to the explanations we have given in the preceding pages, there is much to show that an ice age is advancing, and has made considerable progress in the high latitudes of both hemispheres. Furthermore, if the second rotation, as claimed by General Cowell, is able to perfect a glacial period at regular intervals of 31,600 years, it seems that traces of frigid epochs should not be confined to late geological records, as there appear to be little or no traces of glacial work prior to the Quaternary or Post-tertiary periods. It appears that explanations so far given, which depend on the astronomical theory to account for the ice age, are not in harmony with well-known geographical facts. The explainers neglect the attention due to the great prevailing winds which since the earlier geological ages have, in connection with continents, moved the surface waters of the ocean from torrid latitudes to colder zones, and from the colder zones to the warmer latitudes. This exchange of ocean waters between the zones is as old as the continents which shape their courses. The important change wrought in the ocean currents sufficient to have caused the glacial age which ended the early warm epochs was brought about through the action of the prevailing winds, which, in connection with the form of continents, became able to move the ocean waters from the northern hemisphere into the southern sufficient to submerge the low lands of the southern hemisphere, causing a great diversion of the tropical currents from the high southern latitudes, such as I have pointed out in preceding chapters. Those writers who believe that ocean currents have been the cause of great climatic changes have suggested that the existence of an ancient channel through the isthmus of Panama would have caused a frigid period on lands bordering on the northern shores of the Atlantic by turning the head-waters of the Gulf Stream into the Pacific Ocean. Professor Agassiz thinks that such a channel existed during some remote geological age, judging from the semblance of the fauna pertaining to the Caribbean Sea and Pacific Ocean. Yet it may be said that an open channel through Central America would have connected two high sea-levels. For this reason there would be little or no exchange of water between the Caribbean Sea and the Pacific Ocean.