356 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. [June 9.
site hemisphere after half that interval, or 12,500 years. Now no
geological evidence could detect an interval of 12,500 years between
the dates of glacial scratches or of beds of drift in widely separated
regions, such as Scotland and Patagonia. But according to M. Martins
there is geological evidence of at least two glacial periods, with
a geologically short interval between them; and such recurrent
glacial periods ought to be expected on either Mr. Croll's theory or
mine.
It is to be observed, in conclusion, that Mr. Croll's reasoning and mine leave untouched the arguments by which Sir Charles Lyell has endeavoured to show how all changes of climate may be referred to geographical causes, especially to changes in the distribution of land and water, and changes in the direction of the ocean-currents produced by these. The climate of any region at any period is due to a complication of causes, some of which are geographical, and some astronomical. Sir Charles Lyell has dealt with the geographical ones, Mr. Croll and I with the astronomical ones. There is no doubt that the geographical causes of changes of climate are veroe causoe; but it is not so certain that they are fully adequate to account for the facts. It is equally certain that the astronomical causes also are veroe causoe; and I have endeavoured to show what their mode of operation must be.
Discussion.
Prof. Ramsay remarked that Prof. Dana and himself had both referred the origin of many fjords to the same cause as the author.
