The Roman Road at Springhead Nurseries

INTRODUCTION THE ROMAN ROAD AT SPRINGHEAD NURSERIES VICTOR T.C. SMITH, B.A., F.S.A. Following the publication in this journal of two short interim notes, 1 this report further comments upon and illustrates the results of the searches undertaken in 1972 and in 1991-92 for the route of the Roman road (R2) through Springhead Nurseries. It also notes some archaeological features observed during 1992 in a cable trench cut by the South Eastern Electricity Board along the southern perimeter of the Nurseries. Current schemes for improvements to the A2 trunk route and a proposed new road infrastructure for the Ebbsfleet Valley Development are likely to encroach upon the areas which were investigated. It is, therefore, an opportune point at which to highlight the results of the investigations and to discuss the archaeological potential of the Springhead Nurseries in general. Moreover, the Nurseries have also been identified as an area for possible further commercial development in the longer term. The route of the Channel Tunnel Rail Link through ground to the east of the Nurseries will also affect areas of archaeological interest. During May and June of 1997, a team from Wessex Archaeology investigated that area and fields to the west of the Nurseries, with an extensive pattern of trenches. There was also limited trenching within the Nurseries themselves. The results of these investigations are unknown to the writer but are probably to be reported. 1 V.T.C. Smith, The Roman Road(R2) at Springhead',Arch. Cant., cix (1991 ), 332-3 and 'Sprlnghead: A Second Interim Note', Arch. Cant., ex (I 992), 383. 51 VICTOR T.C. SMITH BACKGROUND TO THE INVESTIGATIONS It has never been established how the main Roman road or Watling Street between London, Rochester and Richborough and Dover passed through the Roman settlement at Springhead. Both from the London and Rochester sides the road probably made its near approaches on a line similar to that of the modem A2, judging from the evidence of perpetuation through trackways and parish boundaries noted by I.D. Margary.2 It was also observed by Margary that the Roman road to the south of Northfleet was mentioned as having formed a property boundary in a Saxon land grant document of A.O. 963.3 The route subsequently followed by the A2 is identified as the former Watling Street in the Ordnance Survey Map of 1863. In addition, according to local folk memory and the labelling of some old maps, the former track, now a slip-road, which defines the southern perimeter of Springhead Nurseries, is on part of the route of the Watling Street.4 In 1965, the then Director of Excavations at Springhead, William S. Penn, suggested that the Watling Street, which he designated R l, entered from the east on a line approximately under what is now the central reservation of the A2 dual carriageway and then re-aligned south-west across One Tree Field which contained the temple area. His map portrayed this re-alignment in a definite line for a distance of 125 m., with the implication that it continued in that direction for a further (unspecified) distance.5 He also showed a 'branch' road (R2) passing north-west from R l under the A2 to about the entrance of the Springhead Nurseries. R l was first discovered as a cropmark in One Tree Field, initially noted at ground level. Its first 50 m. were confirmed by archaeological excavation. It is not clear whether its presumed line under the central reservation of the A2 had also been proved by excavation. Within One Tree Field, Rl was of considerable thickness and this led Penn to conclude that the ground on which it had been built was marshy during the Roman period.6 R2 was revealed as a cropmark in an aerial photograph. Its existence was confirmed by the finding of its metalling on a prolongation of its line during the building of the southern carriageway of the A2 in 1963/4.7 A later 2 l.D. Margary, Roman Roads in Britain, London ( 1973), 51-2. 3 Ibid. 4 W .S. Penn, 'The Springhead Pleasure Gardens and Plantation, I 805- I 936', Arch. Cant. lxxxi ( 1966), 63. 5 W.S. Penn, 'Springhead- Map ofDiscoveries', Arch. Cant., lxxx (1965), facing I 08. 6 W.S. Penn, 'The Romano-British Settlement at Springhead. Excavation of the Watling Street, Shop and Pedestal, Site B', Arch. Cant. lxxil (I 958), 77-9. 1 Ibid., note 5. 52 .... .... ' INVESTIGATIONS AT SPRINGHEAD NURSERIES GENERAL MAP ' .... ' .... ' ' .... 50M 􀀄 \ ' .... Fig. 1. General map of the areas investigated, based on the Ordnance Survey VICTOR T.C. SMITH undated map (but earlier than 1979) by S.R. Harker, who succeeded Penn as Director of Excavations, showed Rt 's re-alignment more cautiously: he marked it in definite line for a lesser distance of 50 m. which he had tested by excavation, to the point at which Peon's 'branch' road diverged north-west. He also showed a tentative continuation ofRl to the building Bl, identified by Penn as a bakery.8 Another map by Harker dated 1980 omitted the continuation of the road to the 'bakery' and, following the results of his excavations, confidently showed Rt and R2 to be the same road, formed into a dog-leg which aerial photographs had suggested.9 The investigations by the writer, assisted by members of the Excavations Committee of the Gravesend Historical Society and others, were undertaken to explore the extent to which R2 continued on its alignment through Springhead Nurseries and to look for any evidence of a re-alignment or branching of this road west, towards London. FIELDWORK IN SPRINGHEAD NURSERIES The investigations were a reconnaissance rather than a research excavation. They were conducted by means of resistivity survey, augering and by limited confirmatory trenching along a number of traverses. Most of the investigations in 1991-92 were within a greenhouse shown in Figs. 1 and 2. The commercial operations of the Nurseries and the need for their management to continue routes of access to various parts of the site placed constraints upon the areas that could be selected for search and the way in which they could be investigated. The mainly non-invasive nature of the investigations also limited the extent to which strata encountered could be examined and interpreted. Strata were penetrated in only one place (Traverse 19). Investigation of the ground by a number of traverses shown in Fig. 2 suggested that R2 continued for about 80 m. into the grounds of the Nurseries. Its metalling was found to be about 35-55 cm. below the then existing ground surface. The usable carriageway varied in width 8 S.R. Harker, Vagniacae - The Roman Town at Springhead, Gravesend (undated), facing 5. 9 S.R. Harker, 'Springhead - a brief re-appraisal' in W. Rodwell, Temples, churches and religion, British Archaeological Reports, British Series, 77, 285-8. See also Barry C. Burnham and John Wacher, The 'Small Towns' of Roman Britain, Lendon (1990), Pl. 60. 54 u, u, ' ... LOCATION OF TRAVERSES IN GREENHOUSE ., , I T􀀎r·._ •+·---. __1 T6 T7 l·-.. TS • •• 20M .... '·, T16 Fig. 2. Location of traverses in greenhouse, with inferred line of carriageway in broken lines Traverse 3 = i ====--=-=--=􀀓=:=''-=--=􀀔=-:::::::==--= ===--=====-􀀕-. r--, i--t::-s-:J􀀞---􀀟-􀀠- • . "􀀡- .... ..:::. d ::::=-,;j ·• ···········•Q Traverse 11 Trave= 13 6 PLANS AND PROFILES OF TRAVERSES 3,11AND13 Key ... flint u pebbles G chalk - burnt layer 1 topsoil 2 light brown soil 3 dark brown soil 4 light yellow clay 5 mixed soils 6 Brick•earth SM Fig. 3. Plans and profiles of traverses 3, 11 and 13. Broken lines show metalling confirmed by auguring. THE ROMAN ROAD AT SPRINGHEAD NURSERIES PLATE I Metalling in traverse 19, looking south (I 992) between 5.50 and 6.50 m. and, in places, appeared defined by flints let flush into the metalling. The carriageway rested upon an agger about 12 m. wide. The surface of the metalling consisted of 4-5 cm. pebbles, with traces of gravel in the interstices. In places the surface and edges of the road had been disturbed and were fragmentary. The road had been in a better state of preservation in 1972 when it lay under orchard land. The use of bulldozers and mechanical excavators since that date to modify the ground for the erection of greenhouses appeared to have produced some damage, but the limited archaeological trenching undertaken during the investigations suggested that some interventions to the edges of the road occurred during the Roman period. On a projected continuation of the road line outside the greenhouse to the north-west, augering and trenching at I 00 m. (Traverse 3) revealed traces of a similar but wider metalled surface. A downward slope, perhaps a camber, was found on its northern edge but its southern edge had been cut away by a pit (Fig. 3). Augering in a greenhouse immediately to the west (Traverses l and 2) encountered no resistance beyond 110 m. to suggest any solid layers which might have been metalling. However, in 1993 at 150 m. in the field to the west of the Nurseries faint cropmark traces were spotted from the air 57 V. 00 Traverse 16 I Traverse 19 PLANS AND PROFILES OF TRAVERSES 16, 17 AND 19 Key .. Jlint e pebbles .. chalk l topsoil 2 light brown soil 3 yellow cl&y 4 yc11ow silt and loam S Brickeanb !OM Fig. 4. Plans and profiles of traverses 16, 17 and 19. Broken lines show metalling confirmed by auguring. THE ROMAN ROAD AT SPRINGHEAD NURSERIES of a possible continuation of the road. 10 A traverse was cut in this area by Wessex Archaeology in May, 1997 but any results from this are not known to the writer. An illustration of Traverse 11 in Fig. 3 shows a profile of the carriageway of R2 as revealed partly by augering and partly by trial trenching. Trenching in 1972 near the eastern edge of the road (Traverse 13) had revealed the flint comer of a structure of indeterminate size (Fig. 3). This was not further investigated. The profile of the road metalling found in this traverse is also shown in Fig. 3. Traverse 16 revealed a more rounded profile to the carriageway. There was a scatter of chalk blocks on the northern side of the road in this traverse. On the southern side were traces of chalk block and flint building debris and a build up of deposited yellow silt and loam which suggested that an occupation zone had encroached upon the agger of the road itself (Fig. 4). In the topsoil immediately above the metalling were found a bronze Quinarius of Allectus (A.D. 293-96) and a small bronze coin of Constantine I (A.D. 307-37). Within Traverses 17 and 18 was found what appeared to be another pebbled camber (Fig. 4) evidently on an east-west alignment. If it was a road, then it appeared to have had a carriageway about 3.8 m. wide. Cultivation within the greenhouse prevented further investigation. Augering of adjacent areas suggested the presence of flint and/or chalk paving. A small area of metalling outside the eastern end of the greenhouse was excavated in Traverse 19 (Fig. 4 and Plate I). At the northern end of the traverse the slope of the agger was encountered, but at the southern end there appeared to have been an intervention in the form of a cutting away of the metalling which may have occurred during the Roman period. Cultivation and the need for the management of the Nurseries to retain areas for the storage and movement of their products prevented further lateral exploration but the camber was sectioned. The natural brickearth had been worked to help form the base of the agger and had been capped with pebbles and sand, with an admixture of clay to form a hard layer. Another distinct layer of compacted pebbles, with patches of rammed and puddled chalk was added above this. The carriageway of the top layer of metalling was defined by larger flints let into the smaller pebbles. Although the surface had been damaged by the use of a bulldozer at some point since 1972, several repairs and re-laminations appeared visible. 16 Observation by the writer. 59 °' 0 3 1----- 5----;,7: 􀀋􀀌,....caooca,c»o,c-o-a,.,«.CDoo 0o, & aooo􀀈o Chalk block layer 9 10 11 _ ,-------- ___ ,...,. 􀀂 --- R9 SECTIONS OF FEATIJRES DISCOVERED IN CABLE TRENCH SM Key m. pebbles ® chalk - modern tarmac l modern concrete road foundation 2 orange/brown clay 3 compacted pebbles and soil 4 compacted pebbles burnt layer light brown clay compacted pebbles and soil -·-----...... 8 brick earth 9 topsoil 10 deposited clay ( c. 20) 11 dark soil Fig. 5. Sections of features discovered in the cable trench < n 6 ::0 :-l 0 (I) 􀀁 THE ROMAN ROAD AT SPRINGHEAD NURSERlES THE CABLE TRENCH OF 1992 The existence of a trench cut by the South Eastern Electricity Board along the outside of the southern perimeter of Springhead Nurseries was notified to the writer too late for it to be fully examined but some features were observed in the line of the trench east of the entrance to the Nurseries. A useful, though partial section through the metalling of R2 was revealed immediately outside the entrance to the Nurseries. This indicated that at this point the base of the road was chalk blocks laid upon the natural brickearth (Fig. 5). To the west of this was a horizontal surface of chalk blocks laid on brickearth, surmounted by a layer of compacted pebbles and soil (Fig. 5). The pebbled layer appeared to have been a crude metalling forming a surface next to, and perhaps connecting with, the road itself. A chalk block feature orientated north-east/south-west (possibly part of a wall) was found 20 m. to the east of the entrance to the nurseries (location given in Fig. 2). There was associated evidence of occupation debris (building materials and pottery) in the section but this was left in situ. Further along, about 140 m. west of the former railway bridge was found what appeared to be a linear pebbled surface aligned north-west/south-east (Figs. 2 and 5). This was approximately on the line of the road designated R9 by Penn and which he showed heading in the direction of the pebbled surround at the source of the Ebbsfleet, which he described, perhaps grandiosely, as a 'harbour' .11 OTHER AND EARLIER DISCOVERIES WITHIN SPRINGHEAD NURSERIES There is a history of the discovery of structures and artefacts within the area of the Nurseries originating from soon after the cultivation of the area of the Ebbsfleet stream bed for the growing of watercress in 1805. In 1807, coins, brooches and pottery were found and, separately in the same year, 'horse shoes and clinkers' .12 Around 1814, excavations for a hut (possibly a few metres north of the stream) revealed brickwork and masonry, claimed at the time to be the remains of a 11 Ibid., note 5. 12 A.J. Dunkin, Memoranda ofSpringhead, 1848, and unpublished notes of Dunkin referred to by Penn in note 4, 65-6. 61 VICTOR T.C. SMITH Roman bath. 13 The cultivation of watercress evolved into the creation of a garden with various attractions for visitors. When Mr Silvester took over the gardens in 1834 he was 'surprised at the quantity of coins. etc .• his labourers were continually picking up .. .' By around 1837, he had been persuaded to establish a museum for his finds on the site. When the antiquary Dunkin visited in 1844 he was shown a cabinet of antiquities, including over 400 coins. It may be that not all of these items had been found within the gardens and some might have come from nearby fields. During a visit of 1849 excursionists were shown: '150 coins from Vespasian to Honorius A handmill of granite Two urns taken from nearby A Roman building just laid open One beautiful figured specimen of Upchurch Ware Personal ornaments' 14 In 1844 'some workmen. whilst grubbing up the shrubs in the bank immediately above the now [temporarily] dried up springs ... struck upon a foundation ... picked up a beautifully executed piece of Samian ware ... seized for [his] share of the spoil two silver coins.'15 During the summer of that same year 'whilst engaged in recovering further portions of the bog [workmen] came upon heaps of fictile ware ... bottoms of urns. amphorae. Samian ware .. .' 16 It was concluded at the time that there had been a pottery kiln nearby. Penn noted that a nearby field was called tiler's field. 17 'A perfect half of a quern was dug up on 16th May, 1845'18 About the same year a British coin was found. An inhumation and a cremation burial were found this same year on the north side of the coach road which ran past the southern boundary of the gardens and which forms a slip road to the A2 today. 19 These were described as having been found in a bank above the road and this points to the rising ground to the east of the pebbled head of the stream as the find spot, and where other burials have been discovered. Finds of various kinds appear to have been made at intervals down 13 Ibid. 14 Ibid., note 4, 68. :! Ibid., 70. Ibid. 17 Ibid. 18 Ibid. 19 Ibid. 62 THE ROMAN ROAD AT SPR.INGHEAD NURSERIES PLATE II Column section and bases at south corner of the car park at Springhead Nurseries to the present day. The gradual replacement of orchard land with greenhouses on the south slope down to the course of the stream since the l 960s, resulted in a spate of new discoveries, through the process of digging foundations, terracing ground to make flat surfaces and the actual cultivation of the ground. These were not recorded at the time but are known anecdotally to have included traces of structures, surfaces, pottery and coins. 20 Possible Roman flint terracing lower down the slope to the stream was noted by the writer in 1970. Excavations by the Gravesend Historical Society in 1964 at the head of the stream revealed the already noted pebbled surround. The most generous description to be assigned to this is that it might have formed a small dock. Behind this were traces of a building (B l 7). 21 In the 1960s during excavations by the South Eastern Gas Board, just to the south of the entrance to the Nurseries, were found the remains of a small unidentified building (B14), a well (F2) and an area of flint and chalk paving (F l ). 22 20 Personal comments by various members of staff of Springhead Nurseries. 21 Ibid. note 5, 113. 22 Ibid., 112-3. 63 VICTOR T.C. SMITH Finally, should be mentioned the stone column base, column section and rectangular base deposited near the south corner of the car park of the Nurseries (Plate 11).23 The column section was, with another, apparently first noted by the Gravesend Historical Society in 1964 embedded on either side of the stream, having been adapted to form the supports of an extemporised timber footway or bridge. If these items are of Roman origin, it is at least possible, or even probable, that they originated from a placement within the area occupied by the Nurseries. The implications of this need to be considered. DISCUSSION AND THE ARCHAEOLOGICAL POTENTIAL OF SPRINGHEAD NURSERIES The investigations supplied evidence for the continuation of R2 into Springhead Nurseries. Where sections were exposed, the metalling of the road appeared to be only 35-40 cm. thick. However, the areas investigated provided no clear evidence of how Watling Street passed out of the settlement towards London. This might be looked for as a turning west from R2, but a possible second camber found in the greenhouse needs further investigation to confirm that it was a road. This might represent a candidate. Failing that, there is a range of other possibilities for a turning in uninvestigated areas within Springhead Nurseries and, indeed, under the A2 carriageways. The evidence of interventions along the southern edge of R2 strongly suggests later invasive occupational activity. It has been suggested elsewhere that the dog-legging of R l/2 might have been to avoid marshy ground around the head of the strearn.24 If, indeed, the Roman settlement at Springhead may be identified as 'Vagniacae', one interpretation of that name is that it referred to a 'marshy place' or a 'marshy place estate' .25 That description might plausibly fit a low-lying water-logged situation near to springs. If that is how the Romans found it when they built Watling Street (perhaps within several years of their invasion in A.D. 43), it may be that the stream became managed, with some attempt to control flooding or ground saturation to allow development nearby. The evident revetting of the stream26 might have been part of such a regime of management. 23 Observed by the writer and Mr Charles Oliver. 24 Barry C. Burnham and John Wacher, op. cit. in note 9,343. 25 A.L.F. Rivet and Colin Smith, The Place Names of Roman Britain (London}, 1981, 485. 26 Ibid., note 6, 113. 64 THE ROMAN ROAD AT SPRINGHEAD NURSERIES In any event, the road R9 was aligned to connect the head of the stream with the area around the temples. While key-hole glimpses of the kind reported here have added useful evidence, larger areas need to be examined to gain clearer information about the pattern of development and to make more informed judgements. The long-running excavations mounted at Springhead from the early 1950s to the early 80s centred on the temple area in One Tree Field. It is these which have been the focus of most academic attention but other possible areas of development such as Springhead Nurseries need also to be addressed by archaeological enquiry. Investigations by the Kent Archaeological Rescue Unit at intervals during 1991-94 in the 16 Acre Field, to the south of the A2 and to the east of the former Gravesend-Longfield railway line, have been helpful in broadening our knowledge of the site by revealing a previously unknown and unsuspected metalled road. This was apparently linked to Rl, and was aligned south-east towards the partly excavated site of a building within an enclosure which has been referred to as an extra-mural temple. Traces of some masonry buildings and of less substantial structures, possible post-holes of fences defining property boundaries, and some burials were found on either side of this road. Other structural evidence has probably been removed by ploughing.27 Rl/R2 was clearly a spine for occupation and development as portrayed in successive maps of discoveries at Springhead, and as supplemented by the features found within Springhead Nurseries in 1972 and 1991-2. In addition, pottery and other evidence of occupation were observed by the writer in trenching within the Nurseries by Wessex Archaeology in 1997. Added to this, the evidence deriving from nineteenth-century reports of discoveries in Springhead Nurseries strongly point to this general area having a special archaeological importance. Most of the find-spots in the nineteenth century seem to have been close to the then still-running stream. The excavation of chalk pits nearby from around 1900 gradually lowered the water-table and the stream dried up by 1936. The evidence, both actual and implied, points to development on either side ofR2 at the top of the gentle slope down to the stream, to other structures possibly terraced in that slope and to occupational activity close to the stream itself. Potentially there is evidence of a whole aspect of the settlement at Springhead to be found here. 27 Brian Philp and Maurice Chenery, A Roman Site al Springhead (Vagniacae) near Gravesend (Canterbury), 1996. 65 VICTOR T.C. SMITH Indeed, recovery of archaeological infonnation from the area of Springhead Nurseries may throw new light upon, and give a more balanced view than hitherto, of the nature of the settlement at Springhead. An examination of the area at the head of the stream, might also provide some evidence for the role of the Ebbsfleet in the life of the settlement. It is hard to avoid the conclusion that the placement of the apparently early (perhaps Claudian) ditch ofV-shaped military profile nearby (seen by the writer during its excavation) was to secure, if temporarily, this waterway.28 Further investigation of the depth and width of the stream during the Roman period is still needed but, as has been suggested elsewhere, it is not improbable that it was used as a supply route for the invading Roman army.29 It must later have had an economic role in the movement of goods and perhaps agricultural products to and from the settlement. Any significance of the springs there and the watercourse for the apparent religious attraction and focus of the site should also be considered for investigation. The remains of related structures as well as those concerned with the secular use of the waterway should also be sought. The discovery of evidence of pre-Roman Iron Age activity under Romano-British occupation in One Tree Field might also be paralleled by similar traces closer to or around the head of the stream. The possible role of the springs and the stream in encouraging the development of cult-of-source worship in the pre-Roman period should also be considered. Given that the road R l/2 was a spine for development, evidence of buildings next to, or set back, from the road should be looked for. The planned improvements for the A2 will drive through the area where key evidence for this might be found. The writer hopes that beyond contributing a modest addition to knowledge of the road infrastructure at Springhead, one of the results of this report will be to focus attention on the wider archaeological possibilities for Roman and pre-Roman occupation of the area presently occupied by Springhead Nurseries. In addition, any remaining evidence for the cultivation of watercress or of other modifications of the area in the early nineteenth century for use as a pleasure garden should not be overlooked for recording purposes. Sufficient and comprehensive investigations must precede any development of the site. 28 This is also suggested as a possibility by Dr Alec Detsicas in The Cantiaci (London), 1983, 61-2. 29 Ibid., 62. 66 THE ROMAN ROAD AT SPRINGHEAD NURSERIES ACKNOWLEDGEMENTS The writer would like to thank Mr and Mrs D. Elliot for their permission to carry out archaeological investigations at Springhead Nurseries. For assistance with the investigations in 1972, thanks are due to Mr F.K. Holder, Mr R.M. Lyne and Mrs L. Smith. In 1991-92, help was given by Messrs N. Baxter, P. Bayldon, Mrs D. Coker, Messrs. L. Coker, S. Coker, J. Elford, A. Giles, D.G. Miall, C. W. Oliver, Mrs. D. Reeve, Messrs J. Reeve, M. Reeve, A. Ridgers, Mrs M. Ridgers, Messrs F. Suckling and J. Waugh. He also thanks Ms L. Dyson, Mr A. Ridgers and Mrs S. Soder for reading a draft of this report. 67