almost contemporary Cæsar were in vogue among the British Celts of the time. In view of this substantial coincidence of the strata of legend and historic fact in Ireland (save that what should be the earliest stratum, the mythologic, was "salted" with biblical and classic gems that mislead as to its true character), let us turn back to Mr. Borlase's theory. He postulates a tremendous disturbing influence during a period roughly corresponding to the second stage of what the Germans call the Völkerwanderung, the Barbarian Invasions. Assume for one moment that he is right, that this mass of alien story was poured into Ireland some time between 300 and 600 a.d. What wonderfully clever fellows the bards of the sixth and seventh centuries must have been to refit the whole to Irish circumstances, to Hibernicise the topography, and to marshal the sequence of the sagas so as to correspond with historic facts. For, be it noted, Mr. Borlase draws his conclusions from legends which the traditional Irish chronology places long before the events glorified in the Ulster cycle, and also from legends assigned to a period two, three, and even four hundred years later. But, ex hypothesi, these legends must have come into the hands of the sixth-seventh century bards from the same source and at much the same date. What reason induced them to throw some back into the remotest past, whilst others were dated only a few centuries before their own time, and associated with Irish chiefs of whose historical existence there is practically no doubt? In especial, how comes it that this artificial saga-chronology corresponds substantially to what the natural saga-chronology would have been, that the traditions assigned to the earliest period wear that vague, impersonal, unindividualised aspect which everywhere distinguishes mythical from heroic legend, that the earliest heroic legends are in circumstances of place, custom, setting, &c., exactly what they should be, whilst the later legends fall precisely into place in a chronological sequence which we know from unimpeachable historic testimony (that of Roman and Romano-British records) to be in the main accurate?
It may be urged that this is a priori reasoning, and cannot prevail against Mr. Borlase's facts. Against facts undoubtedly not; but, as we shall see presently, what the author relies upon are inferences from facts, and from facts of an extremely uncertain and disputed character. Here, again, I note that the evil communications of the physical anthropologist and the monumental
