did. The details of the upper gateway show a re-use of late Norman work.

There has been difference of opinion as to the date of the earliest earthworks at Castle Acre. Mr. Harrod, of whose labours here, and anywhere else in Norfolk, I should wish to speak with the greatest respect, was of opinion that the circular and horse-shoe works were pre-Roman. Many were carried away with this idea who have since abandoned it, and the change is creditable-and I suppose inevitable for archæology of this kind has made great strides in the last thirty years. The story at Mileham, a few miles off, is just the same; there we have the Roman, the Saxon, and the Norman works quite as distinct as here, and each perhaps individually coeval with that at Castle Acre. Many other precisely similar instances could be adduced.

The written history of the castle is very slight-we know, indeed, the descent of the lordship-but we fortunately still have in mound and masonry these great witnesses of a long life, not silent, but more eloquent than the written record. But slight as the written history is, it is something to know that the great Edward was more than once at Castle Acre, and I am willing to believe that he lodged here, and not at the Priory, in February 1297. At any rate he would have visited the castle at that time in its prime, and with its Norman defences just then getting a little obsolete; and, no doubt, he came under the gateway that has fallen, and mounted the now vanished steps into the keep which has nearly perished. And, perchance, it was on this very spot, where we are now standing, that he made answer to the deputation from the clergy in the parliament at Bury, who had refused a subsidy to the king:--" From the moment that you cease to bind yourselves by the homage and on the pledge to me for your baronies, I hold myself to be bound in no respect to you.' This was bold speech, but I think the king had to give way. Fifty years later the castle was in ruins.

ANTONINE'S ITINERARY.

ROUTE IX., BRITAIN.1

By the Rev. Canon RAVEN, D.D.

At the Colchester Meeting of the Institute in 1876 I had the honour of reading a paper on Roman roads in the east of England, in which something was said on the subject treated of on this occasion. The views then enunciated have been modified on some points and confirmed on others by further information and examination. It is not without hesitation that the present remarks are made. They will be found mainly directed towards the first stage on the route, to which I have been able to give some personal attention.

A few words may be said on the document with which we have to deal.

The detail of the work which has come down to us by the name of Antonine's Itinerary of necessity ranges over a great extent of time. From the record of the Appian Way to the mention of Diocletianopolis the mind has to traverse some six centuries, and the mileage of the former as well as of other early roads is probably earlier than their titles, for the words of Livy about the Appian Way "viam munivit." Liv. ix, 29,) leave it quite open to conjecture whether Appius Claudius Cæcus laid out the great road which goes by his name, or merely threw up an agger on that which had long existed as a level road. A compilation embedding in itself the result of earlier work it must of course be; and the question is to whom it owes its name. There are three Emperors who bore the name of Antoninus, to any one of whom the publication of the Itinerary may be ascribed. The claims of

1 Read in the Antiquarian Section at the Annual Meeting of the Institute, at Norwich, August 7th, 1889.

others of the name are but slender. These three are Titus Antoninus Pius, Marcus Aurelius Antoninus, and the elder son of Severus, commonly called Caracalla, but never known formally by that name, his designation on coins being also Marcus Aurelius Antoninus. The name of the first is little connected with road-making, but Julius Capitolinus in his life of the great philosophic Emperor records his care for the roads of the Itinera ("Vias etiam urbis atque itinerum diligentissime curavit, cxi.),' and this is very strong evidence, if the text be trustworthy. To support Caracalla's claim, there is an inscription recorded by Maffei, to the effect that Severus and Caracalla ordered the restoration of "milliaria vetustate conlabsa." The honours thus divide themselves between the philosopher and the fratricide. The Itinerary further contains traces of the days of Diocletian, but nothing later than those of Constantine the Great.

An element of uncertainty is thus introduced by the very title of the book, which does not leave us when we come to the examination of the special route, which is our subject this evening.

66

The text, after a most exhaustive recension of MSS., by the latest editors stands thus :-" Item a Venta Icinorum Londinio.

The sum of these distances is a mile short of the total. It will be unnecessary for me, in the presence of so many members of the Norfolk and Norwich Archæological Society, to reiterate the arguments by which the identity of Norwich with Venta Icenorum has been established. I have been for years a convert to them. The extremities being thus fixed, let us first take the half-way house, Camolodunum, better known under the form Camulodu

1

1 Parthey and Pinder question the correctness of the text.

num, the Kapovλódovvov of Dion Cassius. Now we have as a guide to this part of our road the fifth route in the Itinerary; the obscurities in which are comparatively trivial till we get past Essex. In that route from London to Luguvalium ad vallum, on the Roman wall near Carlisle, via Colchester and Lincoln, we have the following mileage :

A Londinio Cæsaromago mpm. XXVIII.
Colonia

mpm. XXIV.

I quite assent to the identification of Cæsaromagus with Billericay, but would draw attention to the difference of mileage in the two routes, twenty-eight miles in the fifth route to thirty-one in the ninth. Now, in the latter there is an intermediate station, Durolitum, which according to Reynolds and Mannert is Romford. I follow them from etymology and mileage; and suggest that a shorter road was made by cutting through the scrub and forest further east. If this be the true solution of the discrepancy, then the ninth route is older than the fifth. This, I think, is confirmed by Camolodunum appearing in the ninth route, but Colonia in the fifth. That the two places are not identical is shown by the fact that the ninth route gives twenty-four miles between Cæsaromagus and Colonia, whereas the fifth gives twenty-one between Cæsaromagus and Camulodunum; and the fifth, which as yet we have found the longer road, has an intermediate station, Canonium, which would be unlikely to shorten the distance. Thus we are taken by the mileage to Prebendary Scarth's conclusion that Lexden, not Colchester, is the Camulodunum of Tacitus, Dion Cassius, and the ninth route in Antonine's Itinerary, the city of Cunobellinus and Boadicea.

Standing now at Lexden we have to deal with the distance from that place to Norwich, seventy-five Roman miles in the Itinerary, but under fifty as the crow flies. How is this to be accounted for? It seems certain that there must have been a great deflection either to the east or to the west. First of all the Stour had to be crossed, no doubt near the station ad Ansam, presumably the lowest point where the river would be usually fordable, still 1 ὁ Πλαύτιος—τὸ Καμουλόδουνον τὸ τοῦ Κυνοβελλίνου βασίλειον εἷλεν. Dion Cassius, LX. 21.

known by the suggestive name of Stratford S. Mary. The name seems to have arisen from the gathering up as in a clamp the various tracks which came together at the ford. Once across the river we must use our judgment for the east or for the west. The next station is Combretonium. It we adopt the eastward course this will be Burgh near Woodbridge, if the westward it will be Brettenham, about half-way between Lavenham and Stowmarket. I regret that I have been unable to visit either place, but I am informed that both possess earthworks. The syllable Bret has, of course, proved attractive, but it ought not to weigh against the entire absence of roads of any note radiating from Brettenham in any direction. In the Peutinger Tabula Convetni which no doubt represents Combretonium is close to the coast. Written against it is xv. the Antonine mileage between it and ad Ansam.

Suckling's remark that the adoption of the eastward course would charge the Romans with having left the heart of the county of Suffolk unprotected may be disposed of by the fact that the fifth route traversed that very district. Camden's preference for the westward course has no other basis but the supposed identity of "Sit" in Sitomagus, with "Thet" in Thetford. He speaks of the river Sit or Thet, but there is no proof of other existence of the first name.

The balance of evidence seems to me to incline eastward, and such remarks as I have to make from local knowledge are based on that theory. Assuming this, let us look to the first stage. And first of all its length (32 miles) is remarkable, being rarely surpassed. We have 35 mile stages twice in the very obscure Iter v., and one 36 mile stage on Iter xv. between Durnonovaria and Muridunum, on the road from Silchester to Exeter, and these are the only British instances in excess of the stage between Venta Icinorum and Sitomagus. And as it was undoubtedly long, so it was presumably difficult. Three rivers, the Tase, the Waveney, and the Blyth, had to be forded. On the inland side lay, for the greater part of the way, an ancient and deep forest, which also extended occasionally beyond the road on the sea side. The lighter lands on the sea side were covered with thickets and scrub, and