Contrast this persistence of the old type in its old home, and in certain lands which it had conquered, with its utter disappearance in certain other lands where it was intrusive, but where it at one time seemed as firmly established as in Italy—­certainly as in Spain or Gaul.  No more curious example of the growth and disappearance of a national type can be found than in the case of the Graeco-Roman dominion in Western Asia and North Africa.  All told it extended over nearly a thousand years, from the days of Alexander till after the time of Heraclius.  Throughout these lands there yet remain the ruins of innumerable cities which tell how firmly rooted that dominion must once have been.  The over-shadowing and far-reaching importance of what occurred is sufficiently shown by the familiar fact that the New Testament was written in Greek; while to the early Christians, North Africa seemed as much a Latin land as Sicily or the Valley of the Po.  The intrusive peoples and their culture flourished in the lands for a period twice as long as that which has elapsed since, with the voyage of Columbus, modern history may fairly be said to have begun; and then they withered like dry grass before the flame of the Arab invasion, and their place knew them no more.  They overshadowed the ground; they vanished; and the old types reappeared in their old homes, with beside them a new type, the Arab.

Now, as to all these changes we can at least be sure of the main facts.  We know that the Hollander remains in Holland, though the greatness of Holland has passed; we know that the Latin blood remains in Italy, whether to a greater or less extent; and that the Latin culture has died out in the African realm it once won, while it has lasted in Spain and France, and thence has extended itself to continents beyond the ocean.  We may not know the causes of the facts, save partially; but the facts themselves we do know.  But there are other cases in which we are at present ignorant even of the facts; we do not know what the changes really were, still less the hidden causes and meaning of these changes.  Much remains to be found out before we can speak with any certainty as to whether some changes mean the actual dying out or the mere transformation of types.  It is, for instance, astonishing how little permanent change in the physical make-up of the people seems to have been worked in Europe by the migrations of the races in historic times.  A tall, fair-haired, long-skulled race penetrates to some southern country and establishes a commonwealth.  The generations pass.  There is no violent revolution, no break in continuity of history, nothing in the written records to indicate an epoch-making change at any given moment; and yet after a time we find that the old type has reappeared and that the people of the locality do not substantially differ in physical form from the people of other localities that did not suffer such an invasion.  Does this mean that gradually the children of the invaders have