The test which Zeuss used for establishing the age of these documents is a scientific test, the test of orthography and of declensional and syntactical forms. These matters are far out of my province, but what is clear, sound, and simple, has a natural attraction for us all, and one feels a pleasure in repeating it. It is the grand sign of age, Zeuss says, in Welsh and Irish words, when what the grammarians call the ‘_destitutio tenuium_’ has not yet taken place; when the sharp consonants have not yet been changed into flat, _p_ or t into _b_ or _d_; when, for instance, _map_, a son, has not yet become _mab_; _coet_ a wood, _coed_; _ocet_, a harrow, _oged_. This is a clear, scientific test to apply, and a test of which the accuracy can be verified; I do not say that Zeuss was the first person who knew this test or applied it, but I say that he is the first person who in dealing with Celtic matters has invariably proceeded by means of this and similar scientific tests; the first person, therefore, the body of whose work has a scientific, stable character; and so he stands as a model to all Celtic inquirers. His influence has already been most happy; and as I have enlarged on a certain failure in criticism of Eugene O’Curry’s,—whose business, after all, was the description and classification of materials rather than criticism,—let me show, by another example from Eugene O’Curry, this good influence of Zeuss upon Celtic studies. Eugene O’Curry wants to establish that compositions of an older date than the twelfth century existed in Ireland in the twelfth century, and thus he proceeds. He takes one of the great extant Irish manuscripts, the _Leabhar na h’Uidhre_; or, _Book of the Dun Cow_. The compiler of this book was, he says, a certain Maelmuiri, a member of the religious house of Cluainmacnois. This he establishes from a passage in the manuscript itself: ‘This is a trial of his pen here, by Maelmuiri, son of the son of Conn na m’Bocht.’ The date of Maelmuiri he establishes from a passage in the _Annals of the Four Masters_, under the year 1106: ‘Maelmuiri, son of the son of Conn na m’Bocht, was killed in the middle of the great stone church of Cluainmacnois, by a party of robbers.’ Thus he gets the date of the _Book of the Dun Cow_. This book contains an elegy on the death of St. Columb. Now, even before 1106, the language of this elegy was so old as to require a gloss to make it intelligible, for it is accompanied by a gloss written between the lines. This gloss quotes, for the explanation of obsolete words, a number of more ancient compositions; and these compositions, therefore, must, at the beginning of the twelfth century, have been still in existence. Nothing can be sounder; every step is proved, and fairly proved, as one goes along. O’Curry thus affords a good specimen of the sane mode of proceeding so much wanted in Celtic researches, and so little practised by Edward Davies and his