III Roman London
Ralph Merrifield
The Origin of Londinium were unfamiliar with the ford they were obliged to swim, but some others got over by a bridge a little way upstream.4 It BRIDGE-BCILDI:--:G \vas a particular accomplishment of seems likely that this was a temporary military bridge hastily Roman military engineers, and the main roads that were soon thrown across the river by an army that had come prepared to to be centred on the first London Bridge were almost deal with this obstacle. Ifso, it is likely to have been a floating certainly laid out by the sun'eyors of the Roman army bridge, and it was not necessarily on the same site as the later primarily for a military purpose the subjugation and control permanent bridge, which was presumably a piled structure. of the new province. It was the army therefore that provided Recent work in Southwark suggests that the south bank the conditions from \\'hich London was born. Can \ve go opposite the Roman city was by no means an obvious place to further, and suggest that Londinium itself was a creation of select as a bridgehead. It is true that sandy gravels here rose a the Roman arm\) A.. centre of land communications that fnv feet above the less stable silts on either side, but they were could also be reached b\' large sea-going ships, taking intersected by creeks and river-channels, forming a series of advantage of the tides that probabh' then just reached the islets \vhich themselves had to be linked by small bridges. bridge, would haH been an ob\'ious site for a major supply ~IoreO\'er, no e\'idence of any occupation beside the ap base. \X'e have learnt that mam'-perhaps most-towns of proach roads of the southern bridgehead has been found that Roman Britain originated as military sites, and it would be is likely to be earlier than about AD 50.5 This raises the surprising if the most important strategic centre of all did not question whether the Roman London Bridge and Londinium fall into this category. Some support is given to the hypothe itself originated during the earlier phase of the conquest or sis of a military origin for London by the existence from the se\'eral years later. beginning of a planned street la\'-out, \\'ith an east-west street It is inconcei\'able that a regular crossing-place of the remarkably like the ria priJlcipalis of a fort underlying the Thames \vas not maintained by the Romans from the begin western end of Fenchurch Street, particularly now that we ning; indeed, the continued use of the Richborough base know there were open gra\'elled areas immediately to the implies also a regular use by the military of the land route north of this in the earliest phase of regular occupation.! through Kent, probably mainly for personnel travelling by Nevertheless, no military equipment has been found asso the short sea passage and for the imperial post. It is therefore ciated with the earliest occupation of the central nucleus of reasonable to assume that \\'atling Street south of the river Londinium on Cornhill, and the one undoubted early military was one of the first permanent roads to be constructed. The structure lay a\\'ay to the east on a completely different old argument based on the alignment of \\Tatling Street north alignment. This \\'as a \'-shaped ditch found near Aldgate in and south of the river on a putative crossing at Westminster is 1972. It contained a sword-grip in its till, and appears to ha\'e therefore revi\'ed, and is considerably strengthened by the been part of the northern defences of a short-lived fort,2 discovery in Southwark of a pre-Flavian roadway aligned as if which ma\' ha\'e ceased to exist before the construction of to link the London Bridge crossing with that of Westmin London Bridge ga\'e birth to Londinium. ster. (, The existence of a Roman crossing-place, perhaps a Precisely when this occurred it has so far been impossible ford, between Lambeth and \\'estminster, must therefore to determine. ,\rchaeolog\' cannot help \'en' much; pottery now be considered probable. \\'as this the first regular and coins from early contexts may suggest a Claudian date, crossing-place, which was supplanted by a permanent bridge but give no indication when in the first ten years or so after in Southwark only after se\'eral years had elapsed? Ifso, it can the innsion of AD 43 they are likely to ha\'e been deposited.' hardly have remained unprotected, and a garrisoned staging History is equally inconclusi\'e. Cassius Dio, writing more post at least would be expected in \\'estminster. \'(1e have as than 150 years after the e\'ent, informs us that the initial yet no e\'idence of this, though negative evidence means little crossing of the Thames by the Roman auxiliary canlry in hot in the City of \\'estminster, where archaeology has until pursuit of the enemy took place at a difficult ford, near where recently been sadly neglected. The other question arising the ri\'er empties into the ocean and at flood-tide forms a from this hypothesis is why Westminster did not develop into lake-presumably not far from the site of London. Since they the Roman cit\'. Gi\'en the normal human resistance to
B . .r. Philp, Hri/illi/lla. R ([9'-). -: R. , Cassius Dio, HiS/Or), Loeb edn. (1914), \'ii. 419. H. Chapman and T. Johnson. FL\[ II l4 ([9-,),1 - ~ IOIl/hn-ark EWIIl'a/ions '9.-:2 --I (1978), i. 24-7. Philp (Ioe. cit.) R. \!errificld and H. Sheldon, The London Archae%,gist, 2, :\0. 8 (1974), 187 f.
10 RO'-.fAN LONDON unnecessary change, why was the permanent bridge not built of the clay walls had been decorated and redecorated with in Lambeth, and why did \'Vestminster not become the centre painted wall-plaster no fewer than four times.'! This suggests of the Roman road system? Answers can probably be found a life-span of at least five years, and ten would perhaps be a to this question in the relatiye instabilitv of the Lambeth and more likely estimate. The earlier phase, which might be \X'estminster banks, and in the need to de\-elop a port further militan, is itself likely to haH lasted for some time, so that a downstream at the tidal limit. date not later than about AD ) ° for the origin of London In our present state of knowledge we must therefore seems likely. resenT judgement concerning the exact date of the building The only historical evidence we have of the nature of pre of the first London Bridge and the beginning of London. Boudican London is a brief and ambiguous phrase from Although events at the south end of the bridge may be Tacitus, whose father-in-law, Agricola, was sen'ing in Britain expected to mirror those at the north end, \ve clearly need at the time. He tells us that the governor, Suetonius Paul much more dating and other evidence of the first phases of linus, learning of the revolt while campaigning in l\orth Londinium from the City itself. Moreover, it is necessary to \'V'ales, pressed on with his cayalry to Londinium-'a place be wary of all attempts at precision from purely archaeolo not indeed distinguished by the title of colonia sed copia gical dating; it is always subject to revision. ne/!,otiatorNIJ/ et commeatuum JJuz).:ime celebre. ,1n The last phrase is often translated 'crowded \\ith merchants and filled with merchandise', evoking an image of retail trading that may be Londinium before AD 60 misleading. Tacitus significanth- does not use the commoner \\ord mercator, cut negotiator, \\'hich primarily means a banker Fortunately we have in the City of London one early or wholesale dealer. In London the negotiatoreJ were probably archaeological horizon that does seem to be gi \'en a fixed date mainly concerned with the financing of import and export by history. East of the stream of Walbrook, which bisects the business on a large scale. Certainly goods from O\'erseas, such later Roman city, is a fairly concentrated area south of as -fine pottery from southern Gaul, were already coming into Cornhill in which a laver of burnt daub datable to the mid the port of Londinium, and were reaching inland towns such flrst centurY is found. It also occurs in scattered areas to the as \'erulamium, no doubt by road traffic along \X'atling west of the \X/albrook, especially in the neighbourhood of the Street.!! main Roman roadwa\' to the west. There is little doubt that The need for such a port may well have been the reason for this was produced by the first great fire of London-a the creation of Londinium and the centring of the road deliberate act of arson perpetrated in AD Go by the rebel Iceni system upon it, subsequent to the building of the first and Trinovantes under the leadership of Queen Boudica. It London Bridge. This can now be located with fair confidence has been shown that immediately east of Gracechurch Street at the foot of Fish Street Hill, where a rectangular timber pier and north of Fenchurch Street there were two distinct phases found in 198 I may be part of an early bridge. 12 Adjacent to of settled occupation before the destruction. The first in the north side of this, and of later construction, was the cluded the major east-west granlled road underlying Fen western end of a substantial wooden quay, which had been church Street, to which reference has already been made: a built about AD 90 on the site of an earlier quay or landing minor north-south grayelled road: a gravelled area in the stage, which had been erected in the open riYer, well to the west part of the site: traces of two wooden buildings and a south of its original north bank (marked by a double row of partition wall. The second phase involved major replanning, piles) but lOO m. north of the present river-side. Contempor although the east-west road continued as a constant ary stone buildings, belieyed to be warehouses, apparently feature-as it apparently did throughout the Roman period. with an open frontage, stood north of the late fIrst-century New and more massi\'e wooden buildings were constructed quayside. A long stretch of a similar quay of the later first across the line of the former north--south rm.dway, and a new centun', with associated buildings, was found further west, east west roadway was laid 28. j m. (96 pedes, 4/5 actus) to the west of I,-ing William Street, indicating that the port facilities north of the main east-west road and parallel with it. The of Londinium extended on both sides of the bridge. U west part of the site bordering on Gracechurch Street The massive engineering \york recluired for the port and continued to be a gravelled area as before.- The buildings of bridge, together with the planned Ia\'-out of the centre, this phase do not support the idea that the central part of pre indicate that Londinium was the deliberate creation of the Boudican London was still a military base, although their Roman administration, making full use of the skills of the purposeful planning and the use of military techniques in army. If its main purpose was the economic exploitation of their construction suggest that they were built as a result of the new pnwince, this was very much the concern of the official initiative. \'\'ith their veranda or portico fronting on procurator, who was probably based at Londinium from its the main road, they are by no means unlike the contemporary beginning. We know that Decianus Catus, holder of this buildings at Verulamium, which housed workshops and office in AD 60, was not at the capital, Camulodunum, at the perhaps shops.H The existence of semi-basements in the time of Boudica's revolt, since he sent two hundred soldiers London buildings suggests the prm-ision of storage space. there from elsewhere; and was later able to escape by ship. Another interesting feature was a clear indication of a piped His immediate successor, C. Julius Classicianus, died in office water supply beside the main east-west roadway. This phase in Londinium -as we know from the discovery of fragments was brought to an end by the fire of AD 60, but only after one of his tombstone, reused ;;is building material in a bastion of
Philp, op. cit. n. 1 ab()\c, - 12. [,hrlp, ()p. cit. 1 \ . " S. Frere, I "erulamiutJI L.'Jardl!r)f1J i \ 1<)""72), 1..+-19. TlCltLI',-JnnaIJ '1\', ". i""rcrc, '_.\Jal'alioflJ, i. 2IH 2,. - (;. \lrlnc, Britannia,!l (1982), 2-1 6: Flit j)orto//{otJIan London, 19 8 \, 3+-11. Loui,e \1 !ller, London _-Jr(lJafo!(),~iJt, .f 199z), q.,-7. Brief note, on Pudding Lane and I'i,h Street Hill exca\'ations, ibid., 161 1.
I I RALPH MERRIFIELD the city wall near Tower Hill.14 The continued presence of the forum could be dated epigraphically to AD 79. 21 A similar this branch of the provincial administration, probably at a date in the earlier Flavian period might be expected for the somewhat later date, is indicated by a wooden writing-tablet London structure. Marsden has argued for a date about ten from the Walbrook, with a branded stamp showing that it years later, on the grounds that the adjacent north-south was an official issue to the procurator's staff. 1j roadway to the east is likely to be contemporary with the new The early timber-framed and clay-walled buildings on the building, and this overlay a deposit of rubbish which seems to east side of Gracechurch Street did not extend to its \vest side, have accumulated until about AD 85-90.22 Philp, however, where there is clear evidence of a very early building of quite has pointed out that these deposits did not extend to the a different character, with ragstone walls and tile courses. 16 western part of the site below the new building,23 and it is Unfortunately there is no dating evidence for it, but it is more credible that the construction of a minor roadway was stratigraphically earlier than a great ragstone building which delayed than that this whole site in the heart of Londinium succeeds the pre-Boudican buildings east of Gracechurch was abandoned for a quarter of a century, at a time when all Street, and it no longer existed when this was built in the other evidence points to rapid gro\vth. earlier Flavian period. Unless, therefore, it had a very brief There are also problems arising from the identification of life indeed, it may well be pre-BoudicanY The discovery of a this structure as a basilica and forum, for it is tiny if compared massive rags tone base surmounted by a squared ragstone with all others in Britain. The basilica, in particular, suppos block underlying Boudican fire debris east of Gracechurch edly the centre of local government in what was rapidly Street clearly indicates that the Kentish quarries were already becoming the largest and most important city in the province, in operation before AD 60. 1S It might be expected, hown-er, is onl y 44.5 m. (146 ft.) long, and is not ewn extended to that this building material would then have been the near occupy the entire north side of the structure. This may be monopoly of officialdom. The early building of stone and contrasted with Verulamium, about 122 m. (400 ft.): Ciren tiles is therefore likely to have been for some public purpose, cester, 100 m. (p8 ft.): Silchester, 85 m. (280 ft.): Wroxeter, and the hand of the procurator may be suspected. 74 m. (245ft.): and even Caerwent, 56 m. (180 ft.). If the London building is in fact a basilica and forum, it should therefore be of an earlier date, before Verulamium set the Post-Boudican London proper standard in A.D 79. It seems unlikely that a governor of the 80S, such as Agricola, who was endeavouring to persuade There is no doubt about the public character of the great local authorities to invest in adequate fora as a measure of building that succeeded both the burnt timber buildings east Romanization, could have permitted such a bad example to of Gracechurch Street and the stone building to the west of it. be set bv London. This consisted of four ranges of rooms about a central courtyard, oblong in shape, with ragstone walls on a mor tared flint foundation. The external north, east, and west Londinium as the Capital walls were buttressed at the base as if to support applied columns. A recently revised plan gives a north-south length In the Flavian period, Londinium was not on1\- rapidly of 104.5 m. (343 ft.) and an east-west width of 52.7 m. expanding but probably approaching its zenith of prosperity. (172 ft.). The northern range is wider than the others, but is It was also, it seems, being transformed into the capital of the shorter than the overall width. A small contemporary build province. In the last decade or so of the first century a great ing to the west of it, with walls constructed of roof-tiles, is complex of public buildings was constructed on terraces of almost certainly a temple. The identification of the large the steep slope rising from the Thames just east of the mouth oblong building, which straddles the line of modern Grace of the \'V'albrook, a site more than I 12m. (no ft.) wide, of church Street, is less certain. Marsden, who has produced a which the western portion is now occupied by Cannon Street com-incing reconstruction of its plan, has argued strongly Station. It was curiouslv irregular in shape, probably because that it is the first forum of London, and that the north range is of the difficulty of the terrain, but had architectural preten the basilica. 19 This implies local government and a defined sions, with an impressive ornamental pool in a central garden. status, which London did not apparently possess in AD 60. It The rooms included great reception halls and suites of smaller is clear that some years elapsed between the Boudican rooms that may have been offices or staff quarters.24 It has destruction and the building of this structure, which on been plausibh' interpreted as the gm'ernor's palace, although evidence obtained by F. Cottrill in 1934 cannot be earlier than we ha\'e as yet no epigraphic evidence of the presence of a the Fla\-ian period.2o As at Verulamium and Camulodunum, governor until a much later date. 25 There is, however, an there seems to have been a considerable delay before rebuild indication of the probable transference of the headquarters of ing after the Boudican destruction. ;\t Verulamium, there the Imperial Cult to London from Camulodunum before was considerable coin and potterv evidence to suggest a lapse about the end of the first century, and this would suggest a of about fifteen years before rebuilding in Insula XIV, and formal recognition of London as the provincial capital by that
14 Roman IflScriptioflS oj Britain (RIB) )-6, Il. .'1 Frere, Excal'atiofIJ, 40. I; British ~[uscum, GlIide 10 the ,'lntlquities 0/ Roman Britain (19) I), 48-9, fig. 22, 5. " ~[arsden, Co!!ectanea I JmdiniefIJia, 100-2. 1<. R, ~[errifield, Roman Cit)' oj J,ondon (19Gi), IJ8, fig. 24, Z, ZI, fig. 2,C. 2; Philp, op. cit. 19.
1- Through a misunderstanding it has been suggested that this building was " P. \1arsden, TLjL-LI' 26 (19'5), I 102. part of the later forum structure (P. ~[arsden, Collectanea J,ondiniensia, L~!AS C' ""n inscription reused as building material in the late Roman ri\'erside wall Special Paper :\0. 2 (1978\ 92). This is stratigraphicalh impossible. relates tu the rebuilding of a temple b\ a ,rd-cent. g()\'ernor of Britannia Superior I> Phllp, op. CIt. I j, 1G, jig. 8. (after the dIvision ()f Britain into two provinces). R. P. \\'right, ~!. \\'. c:. Hassall, 19 ;\Iarsden, Collec!anea Londinimsia, 9(,- 8: The Roman Forum site in London, 1987, and R. S. O. Tomlin, Britannia, , (1976), 3'8 -9. 22 ,8. \lcrrifield, op. cit. 2)" (229)'
12 RO\[A:--: LO"JDO"J
date. 2() The governor \vould certainh' ha\'e had a headquarters walls, cm'ering an area of nearly twelve acres in a hitherto in the capital, and the Cannon Street building is as yet the undeveloped district to the north-west of the city, on the west only possibility. side of the \X!albrook. It was to have a lasting effect upon the Other maior public buildings were being constructed shape of London, and its Ilia praetoria and Ilia principalis \vere about this time; a public bath-house on the slope above to sUr\'ive in the medieval street-plan; while its north gateway Cpper Thames Street and, at last, a great basilica that was survived as Cripplegate-the only early medieval city gate worthy of Londinium. This was an exceptionally large aisled that did not lead to a maior road, since it owed its origin not building of ragstone and brick, with an eastern apse which to the need for access to a Roman road, but merely to the marks the position of the tribune. On the northern side was a con\ention that a Roman fort had a gate in each of its four double row of square rooms, presumably offices. It lay to the sides. Part of its west gate has been preserved as excavated north of the supposed earlier basilica, and in marked contrast beneath the modern roadway of London \'Vall, and its south with it was more than ]) 0 m. () 00 ft.) long, extending west corner with corner turret can be seen west of Noble between \'{lhittington :\venue in the east and St i\Iichael's Street, where it was exca\'ated beneath the shallow basements Alley in the west. It formed the north range of a square forum of buildings destnwed by bombing. 12 occupying a new lnJula about ]70 m. () 60 ft.) square, \vhich There is little doubt that, in spite of its com'entional \\'as laid out around the earlier structure, the main cast-west defensi\'e adjuncts of ditch and turrets, the prime purpose of road remaining its southern limit as before. The earlier the fort was as a barracks rather than a stronghold for the structure probably continued in usc while the new building defence of London. The pnwision of this accommodation for \vas laid out around it, but was e\'entuallv demolished, and a troops seems to mark the achievement of status as the le\'elling operation took place m-cr the whole site by dump prm'incial capital almost as certainly as the construction of ing great quantities of brick-earth and other material over the the great basilica marks sclf-gcJ\'erning status as a maior demolished walls. There can be no doubt that the status of city although both were doubtless achieved before these Londinium as a city had now been defined, and that it had were actually built. A,uxiliary troops were required in the received its charter, almost certainly as a colonia. The date at capital for guard, escort, and ceremonial duties, and for work which this occurred is best indicated by the date of the as grooms and orderlies in the governor's household.33 There basilica, the seat of local government, for which good were also legionaries who were seconded for special duties to evidence was found by G. C. Dunning beneath the western the headquarters' staff, though these are perhaps more likely part of the building in 1929. Its floor overlay brick-earth from to have been accommodated in the riverside palace. One of the foundation trenches of the walls and rubbish including these, of a later date, was a legionary of the Second Legion pottery, all of the Flavian period.27 This admittedly only gives Augusta, named Celsus, who was seconded as a speculator, a us a terminus post quem, as is usual with archaeological staff officer concerned with the execution of justice. He died evidence, but the cumulative evidence of the absence of in London, where his tombstone was set up by three of his second-century material antedating this and other parts of the comrades, who were also speculatores. 34 new basilica/forum complex, except perhaps in the south There is a strong indication that the provincial government western area, remains convincing. Similar evidence in fact was closely involved with the impressive developments that comes from beneath both of the new north-south flanking took place in Londinium between about AD 90 and 130. On roads east and west of the forum,28 and from the make-up the sites of the great basilica, the riverside palace, the public over the demolished earlier building in the south-eastern part baths in Upper Thames Street, and the Cripplegate fort, tiles of the forum. 29 Only from one site, in the southern range of stamped' P P BR LON', or some variant of this, have been forum buildings west of Gracechurch Street, is there evidence found. Apart from a few other sporadic finds, and a small of second-century construction that may be Hadrianic.30 concentration north of St Paul's, where some may have been Marsden has closely re-examined the entire dating evidence made, they are in fact limited to these areas, with the addition and concludes that the great second forum and basilica of another significant concentration west of the baths in the complex as a whole was built 'within a few years either side of south-western part of the city, where, as will be seen, there AD 100,.31 There seems little doubt that the new insula was laid are indications of massive public works at a somewhat later out, the basilica begun and most of the site levelled for the date.'s A variant 'P PR BR' strongly suggests that the second new forum by about AD 100. Moreover, the small first forum P stands for PROVINCIA, rather than a plural form with the is unlikely to have been demolished until sufficient of the east first P such as Procuratores or Portitores (port officials) as has and west ranges of the second forum had been completed to sometimes been proposed. The first P might well in fact stand take its place. The south range, however, may have been for Procurator, since a state brick-works in London is likely to occupied by an open colonnade or arcade resting on piers have been under the control of the procurator's civilian until a more enclosed structure was built much later in the branch of the administration. \X'hat is quite clear is that it was second century. the provincial government, and not any local authority, that To the early second century rather than the end of the first took steps to provide tiles and no doubt other necessary can be attributed the construction of a fort \\'ith ragstone materials for this great programme of public building, which
,\ funlTan n1
", G. C Dunning, "ll1llq. r. 2\ (19+\), \2 ('c. Sec also \lcrntield, 91, 1m , R. \lcrriticld, .-ll1llq. J. +2 (19112), ,8p. additional SItcs. " II. Shc]d))n, LOlld",! lr i, I:l. Hobin' and J. Schoheld, "'ll1llq . .J. 17 (ICp7), q- 8. S. Roskams, LOlldon .~lr(/li1rologj.r/, \ (19 7 () Bo), 199- 20+,403-7, RO:-'fAN LONDON Instead, two much smaller and slighter Roman structures thickness. In all, at least 86,000 tonnes of ragstone had to be were built there at a subsequent but unknown date. 42 quarried and brought to Londinium. The immensity of the Nevertheless, the public baths were not left derelict but were task cannot be iudged in terms of transport and unskilled deliberately demolished as if in readiness for a major redevel labour alone. The wall was constructed b,' skilled builders opment, that either did not take place or mainly affected a \\ ho laid regular courses of SCI uared ragstone blocks to form much larger area, probably to the we5t. both faces, before filling the interior between them with It would be premature, while the evidence is still so scanty, rubble concrete, consisting of laYers of unworked lumps of to speculate on the reasons for an apparent set-back in the ragstone set in lime mortar. This work was carried out in later second century. Historical sources gi\e no clear indica horizontal stages, separated \w double or treble levelling tion whether this might have been due to political troubles, courses of flat tiles, each providing a level platform on which economic recession, or a \'isitation of plague, and it may be a the stage above could be securely based. Skilled work was long time before archaeological research can produce an also required at the quarries, where more than a million explanation. ,Morem'er, we have no confirmatory evidence as squared facing blocks had to be shaped. Finally the battle yet from the more important eastern half of the city, which ments at the top were cO\'ered with coping-stones, carefully ma\, present quite a different picture. It was, after all, onh' a shaped with convex upper surfaces and at least 3,500 of little later that the port of Londinium was pnwided \\'ith a the5e had to be produced.45 Other essential building included new and massive quayside. the construction of four (possibh flye) gate-houses where existing roads crossed the line of the wall, and an unknown number of internal turrets c()ntaining wooden stairways to The City Wall the ramparts, There was a \'-shaped external ditch 35m. wide, and 1.35-2 m. deep, 3-4.5 m. from the wall; and earth, It is quite clear, also, that the Roman authorities still con presumably from the ditch and the foundation trench, was sidered Londinium to be of the greatest importance, since finally piled as a bank against the inner face of the wall. they took impressive measures to safeguard it. For it was Onh' digging, earth-mm"ing, and general muscular assist about this period, or very little later,4) that the city was ance can have been entrusted to forced labour, and much of provided with its great wall-a fortification that was to the work demanded specialized skills that were probably determine the shape of London for more than a thousand mainh' to be found in the arm\'. This was clearly a task that . . ' years. Moreover, there is no suggestion that it was a defence drew on much more than local resources, and it can only have for a shrunken city, like the late Roman city walls of Gaul. been undertaken when these were not req ui red elsewhere. Instead, it enclosed the whole area of Londinium at its ::\Ioreover, the uniformin of the \\'hole of the landward wall greatest expansion, extending from the Tower of London in suggests that it was constructed as a single continuous the east to Blackfriars in the west, and incorporating the operation, though one that can hardly ha\'e been completed earlier Cripplegate fort in the north-west by meeting its in a fe\\ months. \\'ithin the peri()d between about AD 190 and north-eastern and south-\\'estern corners. The north and west 225, indicated for the building b\ archaeological evidence sides of the fort now became part of the city wall, thereby (see n. 4')), the most likely time seems to be between AD 194 saving a great deal of stone, which had to be brought from and 196, when Clodius Albinus, gm"ernor of Britain and the ragstone quarries of Kent, the nearest source of good usurping emperor, was preparing to make his final (and building-stone. unsuccessful) bid for power against his rivals. For this, he was Even so, an immense quantity of ragstone had to be compelled to strip Britain of its military strength in order to transported by barge down the ::\Iedway and up the Thames. fight on the Continent, thereby putting the prm'ince at risk The wreck of a large barge that, from its date, may well have from barbarian attack. The towns \\"ere the bastions of been taking part in this operation, was found at Blackfriars in Roman ci"ilization, from which la\\' and order could be 1962.44 On an estimated capacity of 2() m.', it can be calcu restored, and there is evidence that a number of them were lated that at least 1300 similar barge-loach \\'ould have been fortitlecl with earth ramparts about this time-work that required to build this landward \\all of more than 3 km. It could be carried out b,' local forced labour with a minimum \\'as 2.7 m. thick at ground level, where it was faced externally of supenision.46 Londinium recei\ed special treatment, pre with a plinth of sandstone (from the same part of Kent), and sumably because it might be required as the principal base for just over 2-4 m. thick above this. Parts of it haH sun'ived in the recovery of the prcwince. The northern defences of recent times to a height of 4-4 m., and it must have been at Britain were in fact m'errun and partly destroyed, so that, for least 6 m. high, with a narrower parapet wall above that. the next fifteen years after the defeat of ,-\lbinus, the principal Even in the fort area more material had to be added, since the concern of Severus and his sons \vas the re-establishment of original fort walls were only half the thickness of the new the northern frontier, and it is unlikeh that they would haye wall, and were therefore reinforced on the north and west been able to deYC)te building resources to the prcn'ision of a sides b,' an internal thickening to bring them to the standard wall for Londinium. 4' P. ~farsden, TL,\1.,lS Z7 (I97(,),I9-23.:'.n ancient stone building known as " p, .\larsdcn, ,11!11/, oj !lit ROll/an Periodfrom Black/i'ian, Gui[dhall \!uscum Hn'CFftmmdes stan and described as a courtyard seems to ha\'e sun'iyed on this site (19(n). until the 9th cellt., howe\'er, when It 1:; mentioned in a c1uner ('I'. Dyson, ibId., " The coping,stones seem to ha \'e .\\craged ah()ut 1 m. in length for rhe Z(,), It seems IIkeh' that part at least of this was the .~re'lt carly Roman terrace wall cn1hLhurc<." and 0.67 m. for the meri(J!1 .... (or \-icc-\-cr~-J', The e~tin1ated total is for north of the baths which can be seen abo\c gwund on the west side of Huggin the wall ('"eluding the two tort walls, \\'htch \\ou[d presumably alread\' ha\'e been IIrl!. Inttlcmented. [t is almost certainh' an undcrL'stimate, smee it does not allow for " There is abundant e\'idence, from deposits ante(Lltln~ the wall and contem the additional coping-stones required tor gate-h(Juses and internal turrets, and the poran wtrh it, that the wall was not burlt earlier than about the last decade of the e"tstence ot coping,stones with !ateLl[ proJecttons suggests that the latter at lea't 2nd cent., and debris deposited b\' a coin foq:cr In a turret ()f the wall shows that it \\ ere pr()\'I(kd with them. had been in existence for some time when Ill' was workl!1g apparent!\' about ::;, frere, Britannia. a Historr oi RMlllti Bn/(Illi ',1(;/)-), Z50 1. AD 220-5 (R. :\lerrilield, Roman London (1969), 118-19), RALPH "fERRIFIELD Outside the Walled City been found in these areas, they are more likely to have been elaborate tombs or memorials than dwelling-places. One The only true suburb of Londinium was Southwark, where underlies St Bride's Church, and another was in the neigh some occupation is likely to have begun at the southern bourhood of St Andrew's Holborn. 5li It is likely that, in the bridgehead as early as in the City of London itself. As we case of St Bride's at least, the Roman structure determined the have seen, archaeologists working there have been unable to position of the Saxon church, though this does not neces find evidence of occupation earlier than about AD 50.4~ There sarily imply continuity with any late Roman Christian cult of are indications of bronze and iron~working in Flavian times the blessed dead, The ,\ldgate road is exceptional in that the in the neighbourhood of Toppings \~/harf, and these are main cemetery area lies well to the south of it, in the associated with somewhat flimsy houses of timber and neighbourhood of Alie Street. There is reason to believe, unbaked clay, which have been found bordering the two however, that the original road to Camulodunum lay further approach roads to the bridge. Such buildings occupied the south, and that it was moved to Aldgate during the late first sites in at least two phases until about the middle of the century replanning of Londinium, when the cemetery was second century, when they seem to have been abandoned. It presumably already established.sl was in fact in Southwark that this phase of apparent local There is little doubt that the adjacent countryside was desertion, suggesting a more general decline, was first clearh exploited for food production, but it was probably farmed by demonstrated in the London area (see n.40). ()nlv on one residents of Londinium, who returned to the city when the site, in St Thomas Street, is there evidence of continuous day's work was done, One must go considerably further occupation to the end of the second centun'. ;\bout the afield into the outer boroughs of Greater London to find middle of the third century or a little later, recovery seems to definite traces of probable farm-houses. 52 l\t \X'estminster, have commenced, and substantial stone-built structures however, there are indications of Roman buildings that were began to appear on both sides of Borough High Street, the clearly residential, although we do not know enough about northern stretch of which overlies a Roman road. This passes them to identify their purpose.53 to the west of Borough High Street just south of Union Street, where it changes direction to the south, again crossing the modern street. Another Roman road has been found Londinium in the Third and Fourth Centuries north of Southwark Cathedral, heading in a south-\vesterly direction, presumably to the \X'estminster ford. The align Ironically, the completion of the city wall, Londinium's ments of both roads converge on the site of Old London greatest surviving monument, was soon followed by some Bridge, on or near which the Roman bridge must ha\'e been diminution of the city's pre-eminence, The political reorgan located. 4H ization under Sen.:rus divided Britain into two provinces, North of the river major Roman roads left Londinium b,' with York (F:!JNra{um) the capital of Britannia In/frior while four of its gates. The main road to Colchester passed through London became the capital only of Britannia Supn·ior. To Aldgate, continuing along the line of Aldgate High Street some extent the seat of power also moved to York as the and presumabh \Yhitechapel High Street. Ermine Street, the centre of militan' acti"in', and an imperial palace was built main road to the north, passed through Bishopsgate and is there. ;\ century later, under Diocletian, a further division perpetuated by the modern street of Bishopsgate. The Sil into four small provinces took place, but Londinium, in chester road, the main road to the \\'est, passed through addition to being the capital of one of these, was probably Newgate, and, after crossing the Fleet, presumably by a also the seat of the lJicarius, who was the link between the four bridge, continued on or near the line of Holborn and Oxford prm'incial governors and higher authority. Londinium also Street, with a fork leading to \'erulamium along Edgware housed the British mint until its abolition in )26, and until the Road. An alternati\e route to the west seems to have passed very end seems to ha \'e been regarded as the financial centre through Ludgate and along Fleet Street and the Strand to of Britain.s1 Between ~26 and 383, when the London mint meet the road from the \'('estminster crossing, probably was briefly re-established by the usurper ~lagnus ""laximus, joining the main Silchester road in Chiswick f Iigh Road, It Londinium had been officially renamed Au,guJta, a mark of seems likely that there \\as a minor gate in the neighbour imperial esteem that was perhaps earned by the city's new hood of the later Tower Postern, giving access to a road function as a strategic base for the restoration of law and skirting the river along the line of The Highway, Stepney, order after barbarian incursions. It gave good service in this beside which a late Roman military site, probably a signal role on at least two occasions-in 360, when Lupicinus was tower, was found in 1973-4.'1() In general, no extra-mural sent b,' Julian, and in the gra\'er crisis of 367-8, when Count dC\'elopment followed the roads out of Londinium; instead, Theodosius was sent by Valentini an, and subsequently res as was the Roman custom, the roadside beyond the cin' limits tored the defences of Britain. was usually de\'oted to the dead. There were burials in the This work ma\ well have included the strengthening neighbourhood of f Iolborn Viaduct and Holhorn, Fleet of London itself by the addition of bastions to the city Street and Bishupsgate, and although Roman buildings have wall on its more yulnerable east side and by the rebuilding, ,- \1. Hammerson has pointed out that thc pattern of COln finds in Southwark Crimes, 18, 3; RCH,\[, iii, 147, also ~uggests an ()rigin f(n the '-'ctrlcn1cnt of about AD ~o. (So/t/!I}J'ilrk J.'.\"t"ill";llIOIJ.l lI, Chapman and T. lohns()n, TLIJAS 24 (197,), I, 'r2 J (19 7R ) ii, \ R~ If) lI, Sheldon ,1nd I .. :-;ch,uf, ( oilertanfa Londinimsia, L\I.\S Special Paper :\0,1 ., R, \!crrifield and II. :-;heldon, London .'lrd"lelJ/o,~i.r/, z (19~2 (,), Ig~ f, The (197H),74 8, precise point on which the ro'Hls eOn\'nge is near the western side ()f the medicI'al " I"pocamt tiles arc said to ha\'e been found under the nalT (It the ,'\bbey bridge, but it is llkell that the prinCipal road continued on Its knll\\n alIgnment Church (Ren\{, Ill. q' bCl'ond the juncti()n and thcn changed alignment to that ()f the hndge beforc' '" The Solil/a f)(~l!i!iilll!IJ, ,1 list of o/tices, compiled ah()ut -\LJ ;()\ and subse reaching it. This II ()ule! hJI"L' tJken place on the considerable p,ll"! ()f the ",uth cluenth' amended. Illciudes,ln ollicer in charge of the Treasun' at I.ondon, bank that has since becn eroded hI the ri\"Cr, ", T. ,i()hn,,)n. London ,·lrcoae%,?isl, 2 (1972-(,). d'l, RO~IAN LONDON completion, or repair of the riverside wall. It has been shown Destruction of Third-Century Magnificence that at least one, and probably all, of the hollow bastions in the neighbourhood of Cripplegate are of medieval origin.55 A tantalizing glimpse of the south-western part of the city in There is, however, a strong probability that the solid bastions later Roman times is afforded bv the massive blocks of on the eastern side of the cin' are late Roman. They contain masonry reused in the riverside wall at Blackfriars. These much reused Roman material, mostly derived from funeran have been shown by T. Blagg to come from a monumental monuments, but also including coping-stones from the arch and a screening wall with an ornamental end. 58 Both Roman wall itself, packed to form solid platforms on which were richly decorated with figures of deities, and strongly artillery (ballistae or catapults) could be mounted. They are a suggest the presence of a religious precinct. Since the archi characteristic late Roman fortification intended to cO\'er the tectural fragments are massive, they are likely to have been approaches to the \vall with a field of projectiles, and their reused in a stretch of riverside wall not far from their source. absence from the defences of Londinium would be surpris Excavation at St Peter's Hill in 198 I revealed massive ing. Moreover, it is clear that these bastions were constructed foundations of oak piles, rammed chalk, and stone blocks, for at a time when the Roman monuments were still visible in the the west and south walls of a large building set on the lower cemeteries and the original battlements of the Roman wall of two terraces. The retaining wall of the upper terrace served were intact, until they were superseded in places by the new also as its north foundation.") Here as elsewhere in this area, bastions themselves. there was a lack of domestic debris, as might be expected in a The defensive riverside wall can also be attributed to a late precinct of public buildings. It is Ii keh' that the arch and phase of the history of Londinium. ~o trace of the distinctive screen also belonged to this. There is little doubt that both are City wall of the late second or early third century has been of the third century, as are a unique relief of four Mother found on the waterfront, and it seems likely that at this time goddesses and two inscribed altars commemorating the the river itself was considered sufficient protection, as it was rebuilding of temples (one of Isis), found in the same stretch throughout the Middle Ages. A continuous defensive ri\er of wall.60 The rebuilding of one temple by a governor or side wall was in fact probably incompatible with the use of acting governor and the participation of an imperial freedman the riverside for the wharfs of an acti\'e port. Traces of in the rebuilding of the other are clear indications that, massive stone walls have, however, from time to time been whatever setbacks Londinium had suffered in the third observed under Lpper and Lower Thames Streets, and century, it was still the home of officialdom. These inscrip recently investigations at both western and eastern ends of tions relate to rebuilding, however, probably in the second the Roman city have shown that there was a late Roman half of the centurv,(l! whereas the great monumental arch, at ri\'erside wall that stood well above the contemporary river least, is more likely to have been constructed when the !e\'el and was therefore undoubtedly defensive, and not precinct was laid out. A major public work seems to be merely an embankment. Where this was built on gravel rather indicated, and one, moreO\'er, apparently devoted mainly to than clay, a foundation consisting of a chalk platform laid on religion (perhaps accompanied by entertainment, as was usual wooden piles was constructed. Preliminary results from a in the Roman world). In a city that appears to have been in a (( )mbination of Carbon 14 dating and dendrochronology, state of some decline, there must surely be a political carried out at the l'niversity of Sheffield by Ms Ruth Morgan explanation for this rather surprising development. It is ,1l1d Ms Jennifer Hillam on piles from Blackfriars, New hesh tempting to attribute it to the presence of the dOJ1JUS dit'ina \X'harf, and the Tower of London, suggested a date not itself in Britain from 208 to 2 I I, and especially to the earlier than the third quarter of the fourth century, but influence of Julia Domna, a strong-minded lady \vith a keen "ubsequent work by dendrochronologists, based on the dated interest in religion. \X'hat seems to be certain is that these German sequence, indicates a date a century earlier. The third-century pagan buildings were still available as a quarry h[,ron' of the riverside wall is e',idently complex, for another for building-stone, not necessarily in a ruinous condition, in -;ection near Blackfriars contained as building material altars the Christian Londinium of the later fourth century. .lnd sculptures from pagan temples, including one that had Our sparse archaeological evidence for late Roman Lon heen rebuilt in the mid-third century;')() manifestly this section don comes almost entirely from the neighbourhood of the Clnnot be earlier than the Christian fourth century. At the river, where buildings continued to be occupied between St Tower, moreover, a later defensive wall was found to the Peter's Hill in the west and the Tower of London in the cast, ;wrth of the first riverside wall, and a deposit of earth laid and were even rebuilt, as at Pudding Lane after 37 j. .lgainst its inner face, apparently immediately after construc ti()n, contained late fourth century coins, including one of \[) 388-92.57 It would seem, therefore, that building and The End of Roman London rebuilding the riverside wall extended (wer a long period, the :,ltest refortification belonging to the final phase of the The condition of London after the abandonment of Roman Roman occupation, and possibly the work of Stilicho's responsibility for Britain in AD 4 IO has been much debated. mIlitary expedition of 396-8. Did it continue to be a stronghold of sub-Roman civilization Grimes, 71-7. ;, Hill, .\Iillett, and Blagg, or. cit. 12 j -69. C. Hill, ;\1. Millett, and T. Blagg (T. Dyson, ed.), The Roman RII'mlde II"all ;<; Britannirl 13 (1982), 3'4; London Archrleologist, 4 (Spring 1982), 162 . ... \['i!lllrJII'ntrl! Arch in London, L.\[AS Special Paper "0. 3 (1980), 198. fjrl/IJh lrrbaeologlcal Report, 4' (ii) (19'''), ;R;-(); Britannia 7 (1976), pH 9; (L Parncll, London AuhaeologiJt, i (I 'riJ KOI, ')" H. !fdl, \!ilktt, and Blagg, op. cit. 169 '1,19\ R. The tcmple of Isis was rebuilt In (he reign of (wo .·11(~IIStl, probably \'aicri,ln ;lI1el (~allicnus (.\D 2\3 Go), though the hrIef ]"int reign of Caracalla and Geta ',.\Il II I I is perhaps a possibilin, particuLlrl\' in \'iew of the partiality of the Sc\'CLln ,lInast\· to the Eg\'ptian cult. Since the prccinct itself was probabh then ne\\, lwwe\'cr, the rebudding would ha\'c been on ,l nn\' site, like the new Iscum burlt tor Cuacalla on the Quirinal; j( would therefore prcsumabh' ha\'c been regarded as a nc\\' temple and gi\'en a dedicaton' inscriptIon in a different form. RALPH :\IER RI FIELD until the end of the century at least, as might be suggested by brooch in the debris of the bath-house roof. The date of this the curious absence of late fifth-centur\' Saxon material from a applied disc-brooch is clearly crucial. It appears to be an considerable territory to the north of Londinium?62 Or was a English derivation of a Continental type of the early fifth city that had been essentially a trading and administrative century, and is likely to have been made, pOSSibly in Surrey, centre unable to maintain its population when trade was in the first half of the fifth century. An identical specimen, disrupted and g<)Yernment no longer existed? however, comes from a grave at :Mitcham that has been Archaeology has not Yet answered these questions, but it attributed to the second half of the fifth centun'.('- The has giyen us a glimpse of one small corner of Londinium in brooch from [.ower Thames Street is therefore likeh' to have the relevant period, and from this certain tentative conclu been lost bet\\een AD 410 and )oo-in all probability before sions can perhaps be drawn. A large house, with a bath-house 475 and the building had then been derelict for some time. in a central courtyard, stood in late Roman times on a terrace It would be rash to generalize from the fate of one in the hill-slope just above the riHr-front, about 210 m. to the building; nevertheless it stood on a central site near the river, east of the bridge. 63 Part of the bath-house has long been where occupation would be as likely to continue as anywhere familiar as the hypocaust preserved beneath the late Coal in the cit\'. Our one piece of e\'idence does therefore favour Exchange in I,ower Thames Street. 64 The house was occu the hypothesis of a period of abandonment, perhaps only pied until about AD 400 or later, on the e\'idence of a coin partial, not so long after the traditional date (AD -!-17) of the hoard of AD 3 S S 402 found scattered on the floor of a furnace Battle of Crecganford when, according to the Anglo-Saxon room, from which the east wing \\'as heated by a s\stem of Chronicle, the Britons of J.-:.ent fled for safety to London. underfloor heating. The ash of the final firing of the furnace \,\'h\' should am'one ha\e wished to lea\'e the cin' with its , , itself contained a fragment of an amphora of a distinctive powerful defences in those unsettled times? Economic pres eastern :\fediterranean type, probably from Gaza, of the sure may ha\'e been a strong influence, since even refugees fourth or fifth century. It was presumably a container for a must C\'entually seek a living, and a London without trade or standard :\fediterranean product such as wine or olive oil, bureaucracy could have supported only a tiny population commonly imported into Britain in earlier times from the under a subsistence economy-a population too small to man western :\lediterranean..\n import of this kind from the east the extensi\e defences etfectiveh'. would be more likely after the disruption of the nearer It has been suggested that a major cause of the abandon sources of supply by the barbarian incursions into Spain in ment of Romano-British towns may have been plague,68 409, and similar imports in the west of Britain have been possibly the epidemic of about AD 443, described by Hyda attributed to the second half of the fifth centurv.. 6S The onlv. tius, or one described by E\'agrius that spread through the other stratified sherds of amphorae of this type from London, Roman world about AD 4) ). The scanty evidence from however, came from a deposit dumped on the Cannon Street London does not conflict with this hypothesis, which might palace after its demolition, and the pottery associated with also account for the apparent avoidance of the cin' and its these was of the fourth century.66 Kevertheless, in the tinal territory, together with \'erulamium and Camulodunum, by phase of the house in JJ)\\'er Thames Street, which on coin the Anglo-Saxon settlers of the later fifth centun'. It ma\' be e\'idence cannot be earlier than about the end of the fourth that the\' had a greater danger to fear than a sub-Roman army, century, luxury imports were apparently a\'ailable, and there for which we ha\'e as \'et no firm evidence in this area.6'J must ha\'e been exports to pay for them. Since in these Sir Mortimer Wheeler once argued that the Middlesex disturbed times most other commodities \\"(JUld ha\'e been in Grim's D\'ke and the Faestendic in west J.-:.ent were boundar short supply, it is possible that the eastern traders came to ies containing the territor\' of fifth-sixth century London, purchase slaves. Londinium, accessible from the Continent and marking it off from the lands of the new settlers.70 An but temporarily safe from its troubles, may possibh' have exca\'ation in 1973 of what appears to be an eastern extension served as entrepclt for the traffic in captives. of Grim's Dyke at Brockley Hill showed that, contran' to It is unlikeh', however, that much of the cit\"s prosperity more recent \'iews, it could not be earlier than the fourth remained. The Lower Thames Street house shows evidence century.~l \,\'heeler was therefore almost certainh right in his of significant social change in its final occupanC\. ;\lterations assessment of date and probably right in his interpretation of to the heating system suggest multiple tenancy, with separate purpose. It is significant that these earthworks are dC\'ised for furnaces heating small suites of rooms, while the bath-house containment of London's territory, not its defence. Neverthe was no longer used for its original purpose. less it seems unlikeh' that such a massi\'e work was under The ash containing the amphora sherd was never cleared taken merely to restrict the wanderings of plague-stricken awa\', and at this point the building was abandoned. The refugees. \\'indows were broken, giving access to wind and rain, and in \X'hate\'er the reason for the abandonment of the house in due course the roofs collapsed O\'er both house and bath Lower Thames Street, it is unlikely to have been an isolated house. \(Thile the building stood as a roofless shell, it recei\'ed phenomenon, and the roofless ruin by the Thames is a fitting a visit from a group of Anglo-Saxons, probabh' on a symbol of the end of the distinctive civilization of Roman scavenging expedition, and one of their women lost her London. Ko attempt seems to have been made to repair and ,,2 J.:-';. I.. \1\res, Anglo-Sa,,:on potten and tlH .\'ettlnJll'nt 0/ England (1969), map 8 \1. G. Welch, Antiq. j. 55 (197\),91-2. facing p. IIi. ,,' J. \X'acher, FoulI' o( RI)!!hW iJritain (1975), 414-18. 6) Excanted b\' P. R. \, \lar,den in 1969-70 and 1976. ,," It has been suggested that "jm/Jros- place-names, a numher of which occur (,4 The whole hath-house ,md part of the east wing of the house ha\T heen north-east of London. 111;\\' he dcrl\'cd from Ambrosius, the succe,,[ul sub-Roman preserved under the new hudding that has been crected on the site. leader of the pcriod. and c()uld be sites oi his garrisons (,1. \lorns. 1/" /-(~f of (,; L. Alcock, ",lrtIJllr·.,. Bri/alll (19~1), 208 H. Artbllr (197,), ICS f.) ,,', P. R, \'. ~1arsdcn, n ..\1.-1\ 2() (1971), tig.40, :-';OS.24; 4. l.'nstratiticd -" R. E. ~L \\hceicr, Londoll dlld tbe Sa,Yo/ls (19'1), (,2 ~".j, sherds of Class B amphoLlc ha\'e ,liso been found on the sites of 5t Dlonls S, A, Castle, iLl!.1\' 2() (19~5), 274 7. Backchurch and St hathcnnc Coleman Street. :\lthough all are irum the Eastern Empire the\' arc of var\'ing wares and irom \'arious sources, 18 = ROMAN LONDO~ reoccupy the derelict house, and the first indication of Fortunately, although only a very limited portion of the purposeful activity on the site was the deliberate demolition total plan has so far been recovered, it is sufficiently dia and removal of the walls-perhaps many years later. Else gnostic for a positive identification to be made. There is a where in the city occupation may perhaps have continued, or close parallel in the amphitheatre at Cirencester, where the may have been renewed after a brief period of time; but it is north-east entrance to the central arena is similarly flanked by likely to have been of a non-urban character, with a subsis two chambers. There is also evidence for t,vo chambers tence economy in which the specialized skills of Roman flanking the east entrance to the amphitheatre at Dorchester material culture could not survive. In a period of anarchy and (Maumbury Rings). It is suggested that these were used as extreme fragmentation London was merely a fortress unable cells for prisoners or wild animals, and possibly also as ante to support a garrison. It could only resume its proper rooms for gladiators and other performers. At least one of function as a centre for collection and redistribution after the each pair probably also contained an altar to Nemesis and development of new factors tending towards reunification served likewise as a shrine. the dominance of powerful kingdoms, the return of the The curved retaining walls of the raised seating would Christian Church, and the revival of overseas trade. have continued round the oval arena, with the church of St Lawrence Jewry overlying the south-western quadrant, Postscript on Roman London, March 1988 and the Guildhall itself built over the northern half. The Publication of the London Atlas has coincided with a great proximity of the south-east corner of the contemporary fort expansion of archaeological investigation in the City of suggests that the amphitheatre was used for military purposes London, reflecting the acceleration of development due to the as a ludus or training-ground as well as for entertainment. The revolution in financial marketing nicknamed 'Big Bang'. New legionary ludus at Chester is similarly placed outside the technological demands ha\re made existing accommodation south-east angle of the fortress. In the case of London, the out of date, and rebuilding proceeds apace. Fortunately a military function would probably have been training the more enlightened attitude by public authorities and devel governor's guards in ceremonial drill as much as in the use of opers alike has made it possible for archaeologists to meet this weapons. new challenge on a scale that would have been unimaginable The discovery of the amphitheatre is of great topograph fifteen years ago. There are at the time of writing more than ical interest, for it not only influenced the curves of Basing IOO full-time archaeologists employed within the historic hall Street and Aldermanbury, but may well have determined 'Square Mile' alone, largely at the expense of the developers the site of the Guildhall itself. Guildhall Yard occupies the themselves, who can usually be persuaded to allocate a small central part of the arena and has evidently been an open space percentage of their development costs to the recovery and and natural place of assembly throughout London's history. publication of the archaeological evidence that would other If early civic leaders took the central place of honour in the wise be destroyed without record. raised seating at the northern end of the short axis of the New information on Roman London is therefore still amphitheatre, they would have occupied the very place ,vhere being won, and will, it is hoped, continue to be for perhaps the thirteenth-century Guildhall would eventually be built. another ten years. This has made it necessary to revise the Another important recent discovery, somewhat over map and text to the latest possible moment, and I am indebted shadowed by that of the amphitheatre, was an octagonal to the Department of erban Archaeology of the Museum of Romano-Celtic temple outside the city wall, south of the road London, and in particular to John Maloney and David to the west from Newgate. It was built fairly early in the Bentley for their willing co-operation in this task. Roman period on a site previously occupied by pottery kilns, It is symptomatic of the dramatic advance in our know and was subsequently replaced by a large masonry building ledge that the amphitheatre of Londinium should have been containing at least nine rooms. identified only within the last few weeks, on a site where there The finding of the amphitheatre was hailed by some had previously been no suspicion of its existence. Excavation newspapers as the discovery of the last important building of on the east side of Guildhall Yard, formerly the site of the Roman London, but this is almost certainly untrue. It would medieval Guildhall Chapel and subsequently of the Art not be surprising if Londinium also had at least one theatre, Gallery, has revealed an unbroken 8 m. stretch of curved and quite possibly a circus for chariot-racing. Major temples wall, which leads to an imposing buttressed entrance-way, and government buildings may also remain to be discovered, 10 m. wide with timber thresholds still surviving. The north to say nothing of market halls and horrea. We know a fair end forms the western wall of a small chamber adjacent to the amount about one large bathing-complex, the public baths on entrance, to the north of which a more fragmentary wall Huggin Hill, but scattered early finds indicate that there were continues the curve, with equally fragmentary traces of a others, of which we know virtually nothing. Important work similar small chamber on the north side of the entrance. The has already been done on the water front, but we need more curved wall is the internal wall of the amphitheatre enclosing information over a wider area, and we remain in ignorance of the central arena, and other fragmentary walls previously pre-Boudican riverside facilities, as well as of much else about recorded on another site to the south may mark the position the earliest Londinium. It will be surprising if the intensive of the external curved wall that enclosed the tiered rows of work now being undertaken does not fill many of these gaps • seats. The two chambers flanking the eastern entrance to the within the next ten years or so, and some future finds may be arena each had doot\vays opening on to the entrance-way and quite as dramatic as anything in the past. through the curved wall into the arena.