to his senior with thankful deference. And this it is that makes the distinction between inference and evidence so important. To mistake the former for the latter is to overvalue antiquity and exclude ourselves from a legitimate and fertile field of research. To confound the latter with the former, is to raise ourselves into criticism when our business is simply to interpret. Proceeding to details, we find that the _Historia Gildæ_ and the _Epistola Gildæ_ are the two earliest works upon Anglo-Saxon Britain. For reasons which will soon appear, these works are referred to A.D. 550. The class of facts for which the evidence of a writer of this date is wanted, is that which contains the particulars of the history of Britain during the last days of the Roman, and the beginning of the Anglo-Saxon domination. Amongst these, the more important would be the rebellion of Maximus, the Pict and Scot inroads, the earliest Germanic invasions, and the subordination of the Romans to the Saxons. But all these are deeds of devastation, and, as such, unfavourable to even the existence of the scanty literature necessary to record them. Again, there were two other changes, equally unfavourable to the preservation of records, going on. Pagan or Classical literature was becoming Christian or Medieval, whilst the Latin or Roman style was passing into Byzantine and Greek. Ammianus Marcellinus, the last of the Latin Pagan historians, was cotemporary with the events at the beginning of the period in question. Procopius, one of the last Pagan writers of Byzantium, died about the same time as Gildas.