Fortunately, too, we are not left in this matter to mere conjecture or analogy: thanks to the researches of Heer and others, we have positive geological facts to guide us which show conclusively that up to the Miocene period Europe was covered by forests of large-leaved evergreen trees, of what we should now consider distinctively tropical types. Ever since the Miocene, and on to the culminating point of the great Ice Age, the European climate has been growing steadily colder, and the European flora has been at the same time steadily adapting itself to the new conditions, and to assuming what we now consider a typically northern aspect. During all that time, the large-leaved evergreens gave way before the deciduous trees and the chillier conifers, beginning at the north pole and spreading gradually southward, as the cold deepened and widened its range. Since the end of the great Ice Age, and the subsequent slight amelioration of the climate in Northern Europe, a reverse process has begun to set in; the Arctic types have begun to recede slightly once more, and the comparatively southern or temperate types have pushed their way northward to occupy the place from which they were previously dispossessed by the newly evolved kinds. It is not necessary for us to inquire here into the causes of this great cycle; the facts are there, and for our present purpose they are quite sufficient. They show conclusively, when one follows them out in detail, that the evolution of deciduous trees was concomitant with the growth of cold conditions around the two poles; and that such trees now exist only where winter, for part of the year, renders the evergreen condition an undesirable one. Even in the tropics, indeed, we find on high mountains a belt of deciduous forest, stretching above the belt of large-leaved evergreens, which itself succeeds to the lowland palms and tree-ferns of the thorough-going equatorial plains.