(3) The third element is perhaps somewhat more tangible than the second. If, during the retreat of the ice, the climate of a region which was twice glaciated became as temperate as that of the present day in the same locality, we should be inclined to regard the preceding and succeeding glaciations as distinct ice epochs, especially if the intervening recession were great and its duration long. Unfortunately for simplicity and ease of determination, there are difficulties in determining with precision how far the ice retreated between successive maxima of advance, how long the interval during which it remained in retreat, and the extent to which the climate was ameliorated, as compared with that which went before and that which followed. (4) If changes of any sort which interrupt the continuity of geological processes intervened between successive maxima of advance of the ice, the separation of the later advance from the earlier, as a distinct ice epoch, would be favored. How great the intervening changes should be in order to constitute the re-advance a distinct ice epoch, is a point concerning which there might be difference of opinion. But it is altogether possible that such changes might intervene as alone to give sufficient basis for the separation. Orographic movements, resulting either in continental changes of altitude or attitude are among the events which might come in to separate one ice epoch from another. Changes of this sort have often furnished the basis for the major and minor divisions of time in other parts of geological history, so that there can be no question as to their adequacy, if they were of sufficient magnitude. We hold that the intervention of orographic or other important geologic changes might reduce to a minimum the amount of recession, the duration of the recession, and the warmth of the intervening climate necessary to constitute the separate ice advances separate ice epochs. The absence of great orographic or other changes in glaciated regions between successive advances of the ice would be no proof that such advances should not be regarded as separate epochs. Divisions of equal importance have often been made without evidence of such changes. From the foregoing discussion, brief as it is, it will be seen that within certain narrow limits the definition of a glacial epoch, as distinct from other glacial epochs, must be more or less arbitrary. It is less important that an arbitrary definition should be accepted, than that the same meaning should be attached to technical terms in common use among geologists. In the interest of harmony and of a common understanding, and without the violation of any truth of science, we believe it would be well if the conception of a glacial epoch, as framed by those who are our leaders in position and in fact, were made the basis for our usage of the term. III. =The Criteria of Distinct Glacial Epochs.=