If the first batch of battles occurred in the north, it is fair to assume that they were fought against the Angles, and this practically fixes them in the north-east. There is good reason for believing that the Angles had very early established themselves in this quarter. Now, it is worthy of note that there is to-day a streamlet called the Dunglass at the entrance of Cockburnspath, the pass through the eastern corner of the Lammermuirs, south-east of Dunbar, which played an important part in the Great Civil War over eleven centuries later. Cockburnspath is just the place at which one might expect battles between rival forces moving from north to south and vice versa. The hills around are still covered with the remains of fortifications, many apparently of great antiquity. It may be that the course of the war in this neighbourhood was as follows: The English moving south from their settlements on the Firth of Forth were met and defeated by Arthur, perhaps on the Glen, and retreated to Cockburnspath. After a succession of engagements, they were driven thence, again defeated on the Bassas (locality unknown), and finally pursued into the Caledonian Forest.

The seat of war then appears to shift southward. Gwinnion may have been fought to drive off some of the invaders who had got into the rear of the Britons while the latter were fighting at the Forth. Urbs Legionis is probably in this instance York. The Ribruit and Agned cannot be identified; Mons Badonis or Badonicus is supposed to have been Bath, but there is no reason for this identification; it probably arose out of the similarity between Badon and the English appellation of Aquae Sulis. Wadon Hill, near Avebury, and Badbury (? Baden-burh) Rings in Dorset have both been identified with this mysterious site. It seems from Gildas that it was a fortress. Probably it was besieged by the invaders, and relieved by Arthur in a battle which stayed the English advance for many years.

BRITAIN FROM ABOUT 500-570.

Showing the probable effects of the Romano-British rally under Ambrosius Aurelianus and his successors. No towns are indicated within the English areas, as it is probable that all had been deserted or destroyed.

The date of these events is doubtful. Gildas, in a sentence which is the despair of all Latinists, seems to say that the battle of Mons Badonicus was fought forty-four years and one month before he wrote his book. We know that he wrote some years before the death of King Mælgwn of Gwynedd (A.D. 547). This fixes the date of the battle at about 500. The ‘Annales Cambriæ’ put it in 516, but the ‘Annales’ are as late as the ‘Anglo-Saxon Chronicle.’ Mr. E. B. Nicholson has suggested that the forty-four years are to be reckoned from the appearance of Ambrosius to 516. It is quite possible that the great leader did commence his campaigns about 472, but Gildas hardly gives us the impression of a man capable of very accurate chronology. But if he were born on the day of the battle, as he says, it would be a simple matter. On the whole, we may reasonably say that the battle was fought about the year 500, and its effect was to stay the advance of the English in the south for at least forty-four years.

One would willingly hear more of the men whose efforts for a while stayed the flowing English tide; but practically all that is known is here set down. The ‘Annales Cambriæ’ state that twenty-one years after Mons Badonicus Arthur and Medrant fell in a battle at Camlan. Arthur is repeatedly mentioned in the bardic poems of Wales, but we cannot tell to what extent these works may have been altered in later ages.

Whatever the fate of Arthur, and whatever the extent and nature of his influence in Britain, his victories staved off ruin for half a century, but no longer. The curse of Britain was that there were too many dynasts for ever warring among themselves. The picture that Gildas has painted may be highly coloured, but there is no reason to doubt that when the danger appeared to be over, the old tribal quarrels began again. Strathclyde and Reged were divided among the different branches of the Houses of Coroticus and Coel. About 540 the energetic but unscrupulous and dissolute Mælgwn, King of Gwynedd, succeeded in getting rid of his related competitors. He may also have asserted a primacy in the south and west, for Gildas calls him ‘Insularis Draco’ (Island Dragon), though it is true that this may simply refer to the fact that the seat of the Cunedda dynasty was at Aberffraw in Mona. Gildas overwhelms him with turgid invective and garbled quotations from Scripture—in fact, his so-called history is, on the face of it, nothing but a sermon directed at Mælgwn. He also denounces two other Welsh dynasts—Vortipore of the Demetæ, and Cuneglass of Powys(?), and two princes with Latin names—Constantinus of Damnonia and Aurelius Caninus. The first name of the latter suggests that he may have been one of those degenerate grandsons of Ambrosius Aurelianus of whom Gildas speaks.

It is in the interval between Mons Badonicus and 577 that the Saxon Chronicle places the conquest of Wessex. To discuss this matter in detail would be to occupy far more space than is here available, but it may be said, in short, that there is every reason to believe that the Chronicle’s chronology is wrong; that it is highly improbable that the West Saxons ever came up Southampton Water, the shores of which were occupied by Jutes at an early date; finally, that Cerdic is a very doubtful figure. The West Saxons were also, and commonly, called Gewissæ, ‘the Confederates.’ Cerdic is a Celtic name; it is, indeed, the same as Coroticus, or Caradoc, and this fact, coupled with the curious name of the kingdom, suggests that Cerdic may have been a Celtic prince, who founded a kingdom with the aid of English mercenaries or allies, and became so identified with them that his origin was forgotten. The latest opinion is that Wessex did not start from Hampshire at all. Some of the battles mentioned—i.e., that with Natan-leod—may be authentic, but they were the work of Jutes. Any delimitation of the territory occupied by the invaders at the time must necessarily be a very vague one; but it is probable that when Gildas wrote they fell into three sections. South of the Thames—largely south of the Weald—a long English-Jutish strip of territory stretched from Southampton Water to Thanet, never reaching far inland except in Kent, which was solidly occupied, and, perhaps, always the wealthiest of the new States.

North of the Thames indications are even vaguer. Professor Oman is of opinion that all the Romano-British cities in Eastern Britain perished very early, mainly on the authority of Gildas, who appears to give ‘a picture of a Celtic Britain which does not extend anywhere towards the east coast.’