consider what is perhaps the most remarkable feature of the whole structure. Only a space of 6 feet of clear waterway was left for purposes of navigation. It has been suggested that in the spaces marked on Plate IV., A and A1, there might have been also passages. But it appears that the same old piles were found in those spaces as in those marked B and B1. If then the 6-ft. passage were the only free waterway, what explanation can be given of such a method of construction? There could have been no difficulty in making the opening wider. It certainly looks as though the constructors of this plankway either cared nothing for the use of the river, or else deliberately preferred to obstruct the passage for all but small boats. A structure characterised by indifference to the use of the waterway might well have been the work of the Romans in the height of their power if this river-passage fell in the line of one of their great "streets." This may have been so. I have elsewhere1 drawn attention to a cobbled way seen by the side of Wensum Street and supposed to be sloping towards a ford, as possibly a branch of a Roman road passing over the site of Norwich and recognised in the later names of the roads called "Berstrete" and "Holmstrete." If this plankway formed part of a Roman road it must manifestly have been an important road; and we might expect to find a

1 Norfolk Archæology, vol. xii., p. 31. It will be observed that a fragment of a similar cobbled way is shewn on the north side of the river in Plate II. But I am informed that the stones were hardly sufficient to justify any conclusion, and as they appear to have lain on the top of the plank-way, they must belong to some later road construction. The position of those formerly found on the south side is given on the authority of a neighbouring resident. They were found more than forty years ago under a house by the side of the road, and, as I was credibly informed, at a depth of 12 ft. below the level of the road. In any case their importance seems much diminished by the present discovery.

"strete" on its line. It is to be noticed that Kirkpatrick (Streets and Lanes of Norwich, p. 84) says of the street on the north of the river:-"This street was called in the time of John of Oxford, Bishop of Norwich, 'Fibrige Strete' and 'Fibrigge Strete."" John of Oxford was Bishop between 1175 and 1200. Kirkpatrick gives no reference for this statement, and Dr. Bensly, who has kindly searched in the Monastic Registers, has at present not been able to verify it. In the existing City documents the name of the street is always "Fibriggate." Among the enrolled deeds, however, in the City Record Room is one of 16th Edward I., in which Muspol Street (which runs north from Colegate Street, shewn in Plate I.) is described as "via regia vocata Muspolstrete," and in another of 23rd Edward I., two streets in that immediate neighbourhood (one of them probably Colegate Street) are described as "regia strata." I can only say that in these City deeds, which number 1000 of the thirteenth century, these instances quoted are, I believe, almost unique in the use of these terms "strete," instead of the Danish "gate," and "strata," instead of "via" or "vicus"; except only in the words Berstrete and Holmstrete, where the use is constant.1

It seems to me, however, that the theory that the builders of this timber causeway had a deliberate intention to block the river for all but small boats is the best explanation of the narrow waterway. Two epochs might be suggested when such a line of action might have been taken. One when the Romans towards the close of their occupation had hard work to ward off the attacks of the Saxon pirates; the other when the Angles, having outlived their seafaring skill, were equally anxious to block the way of the fierce Northmen. On the evidence of local history the former

1 On this subject, see the publications already referred to.

epoch seems the most likely. We know extremely little about Norwich in Angle times, before the coming of the Danes. But we can hardly suppose that the inhabitants of what must then have been comparatively small settlements on the north and south of the river could have accomplished so great a work, or have found it worth while to preserve a roadway, at the cost of so great labour, for the mere purpose of communication between each other. Nor is there any apparent reason why such a work should be attributed to some chieftain who may have occupied the "burh." Moreover, if the blocking of the river was intended for defence, it would have been useless in the times of the Angles. Their chief settlement was in the district of Conesford, where invaders could easily land without ascending the river so high as the site of this bridge.

Still less can we think that the Anglo-Danish or Anglo-Norman inhabitants at a later time, if they wished to substitute a bridge for a ford, would have had any object in leaving so narrow a waterway. With the weakened Romans the case was different. The preservation of their roadways was everything to them. A free waterway up a river was at best a secondary object. When it meant a weak point in their defence against sea pirates it was a manifest disadvantage. In every way this kind of passage over the river, instead of a ford inaccessible except at low tide and at all times exposed to attacks from seaward, is just a work which the Romans might naturally have thought advisable and even necessary to preserve their inland communications when their hold upon the country was becoming more difficult to maintain from year to year. The absence of distinctive Roman finds may be explained by the consideration that there was probably no Roman station nearer than Caistor, and that the Angle settlements, which afterwards became the

city of Norwich, were not commenced till perhaps 200 years after the Romans left Britain.

In reference to the suggestion that these piles supported a Roman road, it may be observed that the whole subject of Roman roads in the north of Norfolk is wrapped in obscurity. A line leading out of Norwich to the north would come upon a place called "Stratuna" in Domesday Book, now Stratton Strawless, a name which indicates a Roman Street; and would approach Brampton, a place reputed to be a Roman settlement (Blomefield, History of Norfolk, 8vo. edn., vol. vi., p. 430), though this is doubted by Mr. Fox in a Paper on "Roman Norfolk" in the Archæological Journal, vol. xlvi., p. 349.

I am not able to confirm this view by any similar instance of Roman work in England. A somewhat parallel case may, perhaps, be found in the curious set of piles called "Cowey Stakes," in the Thames near Walton Bridge. These piles have frequently been found in the bed of the river, and are described as blackened with long exposure like those in Norwich. They were, strangely enough, identified by Camden with those described by Cæsar as having been hurriedly set by the Britons to bar his passage over the Thames. Other authorities, as Wright (The Celt, the Roman, and the Saxon) incline to attribute them to the Romans either as a fishing weir or even as connected with a Roman road. An interesting account of these stakes may be found in Brayley's History of Surrey, vol. ii., p. 342. Their antiquity is undoubted, for they are mentioned by Bede as being as thick as a man's thigh, and immovably fixed in the bottom of the river. They differed from those in Norwich in being shod, or even covered, with lead.

After, however, the foregoing observations were already in proof I was informed that a description of the recent discovery of a similar series of piles in Holland had been

received by the Society of Antiquaries. Unfortunately, their copy was not at the time accessible. But, on my application, the Author courteously sent me one, and the perusal of it seems to add confirmation to the theory of the Roman origin of the Norwich pile bridge. The Paper,1 by Dr. W. Pleyte of Leyden, Director of the Royal Netherlands Museum of Antiquities, relates to a discovery of ancient piles at Zuilichem on the river Waal.

In digging clay for brickmaking in a field on the south bank of the river was discovered a double row of piles at a depth of 450 metres below the surface. They extended for a length of 180 metres, and were in groups separated from each other by an interval of 3:30 metres. They had evidently formed the supports of a bridge, the centre of which had rested on an island still existing. Like the Norwich piles, they all had their tops decayed or broken off. The flooring of the bridge which they supported had entirely disappeared. Although similar in character to the Norwich piles, their arrangement was different. Instead of being irregularly scattered all over the ground, they were systematically arranged in two distinct rows. consisted of groups of four supporting piles, set in pairs close together, each group being ponding group of the other row. not sunk perpendicularly, but sloping towards each other. Dr. Pleyte points out how this construction differs from that of the medieval Continental bridges built after the model of known Roman bridges (such as one at Coblenz) which were supported by stone piers resting on piles. But he also shews how entirely it agrees with the construction of a military bridge described by Cæsar (Gallic War, book iv., ch. 17). That bridge, which was

Each row

opposite to the corresThe rows of piles were

1 "Jets ower de oude Brug te Zuilichem, door W. Pleyte." Overgedrukt uit de Verslagen en Mededeelingen der Koninklijke Akademie van Wetenschappen, Afdeeling Letterkunde, 3de Reeks, Deel xii. 290.