Westward across the valley from St. Albans, the horizon is today fringed by a tract of secluded woodland. A part of this woodland has acquired the name of Prae Wood from a small medieval nunnery, dedicated to St. Mary de Pré, that once stood half-a-mile away beside the derelict Watling Street. The nunnery is now a patch of nettles in the meadows by the Ver, and the Street is a channel from which the metalling has been torn; but both are better known to the historian and the antiquary than are the banks and ditches which lie hidden beneath the summer-jungle of Prae Wood. For all that, these banks and ditches are relics of a city which was at one time no less than the metropolis of a considerable part of Britain. It is nearly nineteen centuries since they passed into oblivion, and there is something of ironical irrelevance in the words of a Georgian milestone which chance has set up in their midst: ‘I mile from Gorhambury, 20 miles from Hyde Park Corner The shade of Sir Thomas Browne might find a happy perch upon that milestone.

About the year 30 B.c., king Tasciovanus, suzerain (it would seem) of a large region in central and eastern Britain, struck coins which bear upon them, in one form or another, the name of Verulamium and are, incidentally, amongst the first inscribed coins minted in these islands. His predecessor, Cassivelaunus, had been the elected leader of the British tribes against Julius Caesar in 54 B.c.; and it is unnecessary here to discuss again in detail the plausible identification of Verulamium with the unnamed stronghold of Cassivelaunus which Caesar's legions stormed in that year.

It is a great help to the understanding of a piece of country, a town or a building to get to some high point in the vicinity and look down upon it, and the more nearly vertical one's point of view the more accurate is one's impression. In the ‘nineties, as a boy, I used to spend hours at the top of the spire of Rouen Cathedral gazing down upon the city below, and it was only a step from this to record such a view photographically; but it was some time before I took my first truly vertical photograph.

About 1904, a friend who was engineer on the construction of the Rothesay Dock near Glasgow asked me to photograph for him the concrete foundations of a part of this dock: I could not get a proper view of these from anywhere on the ground, so I asked if I might utilize a great crane which stood nearby. It was probably against all the rules of the Clyde Trust, but I was thereupon swung high in the air over my subject, and plate 1 shows the result obtained with a no. 2 Bullseye Kodak.

Sir John Rhŷs, whose name is justly honoured by all students of the history of Britain, is nevertheless indirectly responsible for much vague conjecture concerning the Keltic problem by English archaeologists. I write ‘English’ advisedly, for Irish archaeologists—with a profound knowledge of Keltic philology—have definitely refuted the theories propounded by Rhys and maintained in various modified forms by subsequent archaeologists in England. In Wales, the other Keltic country concerned, archaeology unfortunately remained up to recent times the happy hunting-ground of antiquaries whose knowledge of both archaeology and philology was very restricted. Her professional archaeologists, in more recent years, have been hampered by an ignorance of the philological problem and have therefore naturally subscribed to the point of view of the modern English school of thought. On the other hand, some Welsh philologists, often with little or no knowledge of the archaeological evidence, have ridiculed the theories advanced by Rhŷs and the later English writers on the subject. My present purpose is to submit to the readers of ANTIQUITY a case for relinquishing entirely the generally accepted attitude of modern English archaeology towards the Keltic problem.

The first person to introduce these bowls to the archaeological world as a group, and with appropriate comment, was that profound student of Celtic art, the late Mr J. Romilly Allen; subsequently, Mr Reginald Smith set himself the task of making a complete list of all the bowls, or parts of bowls, that were known, and he has since placed us further in his debt by the care he has taken in seeing that every new example, as it came to light, was adequately recorded. Since Mr Allen wrote his paper, however, we have not made very much headway with the main problem of deciding their position in the cultural history of this country; even the recent discovery of the beautiful Winchester bowl, that excited us so much in 1930, was allowed to pass without any revision and revaluation of the really astonishing material that had been collected by Mr Smith, and it seemed that the chief interest aroused by the bowls was for ever to be focussed on the apparently insoluble problem of their use. But last year my friend Mr J. D. Cowen published a paper that constitutes in many respects a very important advance; it was the stimulating kind of article that often provokes other people into rival activities, and I hope it will not be counted against him if I admit that one of his results was an immediate decision on my part to put into print the notes I had collected on this same subject. It is Mr Cowen's distinction that he is the first to isolate and appreciate the ‘heater’ group of bowls, to which I shall refer below, and there are several other points in his very vigorous paper that strike me as being valuable and sagacious observations. But what I have to say here and now is that if Mr Cowen is right throughout, much that I am going to say below is completely wrong.

When the Editor of ANTIQUITY approached the authors and suggested that some of his readers would welcome a visual table showing the occurrence and sequence of the different prehistoric cultures, the matter did not seem to be one of outstanding difficulty. When the time came, however, to produce the work, it was found to be quite otherwise. Perhaps the ideal method would have been to prepare a series of gigantic distribution maps of the area to be covered (Europe, the Mediterranean Basin and adjacent regions) during the different prehistoric periods. But the preparation and publication of such maps on a scale large enough to be at all helpful would have presented insuperable difficulties. The more diagrammatic method which has been adopted divides up the map into several geographical areas which will be found heading the various columns of the table. These areas have been chosen and arranged as conveniently as possible in the light of the general distribution of the early and later prehistoric cultures.

It must always be remembered, however, that the spread of a culture from its area of origin corresponds to the ever-widening rings formed by a stone which is cast into a pond—its influence lessens as we get farther from the centre of diffusion. It is not easy to represent this lessening influence diagrammatically. Of course when the industries of a culture have spread widely over a neighbouring area, its name appears in the appropriate column of the table. Where, however, they are only very rarely found, perhaps as the last, faintest ripple of the circles, or where, as with the Solutrean culture in Spain south of the Pyrenees, they are only present in a small corner of the area adjacent to another thickly sprinkled area (France) whence they have clearly percolated, it would have presented an erroneous picture to have cited the culture-name.