The period of two thousand years or so which we set out to cover here— roughly 1500 bc to ad 500—begins at a time when the evidence available to us is purely archaeological, untainted by the vagaries of history, and ends when the gleanings from archaeology have to be reconciled with a rich historical tradition and the varied interpretations of linguists. Thus, in spanning the millennia, we bridge the disciplines. The first historian to consider the tribes of the British Isles from a truly informed position was Tacitus. Writing towards the end of the first century ad he had access not only to the vague and anecdotal writings of the Posidonian tradition and the observations of Julius Caesar on the tribal situation in the south-east, but he was also able to draw upon the reminiscences of his father-in-law, Julius Agricola, who had spent many seasons campaigning in Britain first as a legionary commander and later as governor of the province. Agricola travelled from one end of the island to the other and, incidentally, was probably responsible for killing more Britons than any other Roman. Assessing the varied array of evidence available to him in an attempt to characterize the British population, Tacitus showed the commendable restraint of an historian in his famous summation . . .who the first inhabitants of Britain were, whether native or immigrant remains obscure: one must remember, we are dealing with barbarians. (Agricola, 11). . . After several centuries of hard archaeological endeavour the situation has changed little. Forty years ago, in considering the formation of the British people, we would have been much more confident. We would have talked of a series of ‘invasions’ bringing in successive waves of new people from the Continent— Deverel-Rimbury folk in the Late Bronze Age about 1000 bc, Hallstatt overlords resplendent on their horses and wielding long slashing swords in the seventh century, invaders from the Marne region around 400 bc and Belgae first raiding and then settling in the south-east in the first century bc (Hawkes and Dunning 1931).