Excavations at Fremlin Walk Maidstone

73 Excavations at Fremlin Walk, Maidstone catherine edwards With contributions by Lyn Blackmore, Frances Raymond, Charlotte Thompson, Ian Betts, Melissa Melikian, Jackie Keily, Alan Pipe, Sylvia Warman, Tony Grey and Katherine Roberts. The evaluation and excavation at Fremlin Walk was conducted by AOC Archaeology Group in advance of the redevelopment of the site by Centros Miller Ltd. The site is located on a large piece of roughly rectangular land, approximately 3.35ha, in central Maidstone, TQ 759 559 (Fig. 1). The site was previously occupied by the buildings and yards College of Art Maidstone Art Gallery Museum and Hall Church Almshouses Centre Adult Education Library Mote Park M20 A20 A229 A26 R. Medway Maidstone Site 0 1 km 1:50 000 N Mount Villa Fig 1. Site Location. InFseigr.t 1 s hSiotew losc laoticona.tion within Maidstone CAT cat HE RIN E EDWARDS 74 of Whitbread Fremlin’s brewery depot, as well as a public house, two car parks, derelict buildings and numerous commercial properties. Work began with an initial watching brief (Phase 1), undertaken on the site in 1998 during geotechnical investigations. The investigations revealed extensive truncation to the west of the site by the basement and cellars of the Fremlin Brewery; however, the results for the eastern side of the site revealed a higher potential for preserved remains. T his phase of work was followed by two phases of preliminary eval-uation trenches (designated as Phases 2 and 3). This involved the machine excavation of 27 trenches spread over the eastern section of the site. Due to the archaeological features observed in Trenches 6, 7, 8, 13 and 16, it was suggested by the KCC Archaeological Officer that the area around each trench be expanded which led to the excavation of four open areas (designated as Phases 4, 7, 8 and 9). This was run simultaneously with a watching brief on the remaining ground work carried out on site. A Level 2 Historical Building Recording survey was also conducted on existing properties fronting Earl and Week Streets. The survey revealed that despite the insertion of recent shop fronts, a significant proportion the properties still retained many aspects of their eighteenth-century construction. T he archaeological work was completed in June 2004. All on-site archaeological work was recorded using the site code FWM03 whilst the finds and environmental data have been recorded under Maidstone Museum accession number MNmnEMG 2003.4, where they will be deposited and available for consultation upon request. T he site generally sloped westward down towards the River Medway from a maximum height of 18.54m od on Week Street down to 5.46m od at the western end of Faith Street and 6.40m od at the western end of Earl Street. The underlying deposits are alluvial deposits from the Medway flood plain above Atherfield and Weald Clays. During Phases 2 and 3, where undisputed natural ground was located it comprised of clay and sand with varying concentrations of flint gravel and sparry limestone blocks. T he riverside zone, which occupies the western strip of the site, had been badly truncated by the construction of the nineteenth-century Fremlin’s Brewery buildings but during the watching brief a small amount of peat was exposed below the levels of truncation. A sample taken from the peat contained fragments of charcoal as well as 48 well preserved and identifiable mollusc shells. The molluscan assemblage suggests that at the time of the peat formation, the adjacent body of water featured emergent and marginal vegetation, which may also have been liable to seasonal desiccation. The water would have been permanent with a slow moving current, which probably represents an early course of the Medway EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 75 p rehEHistoric A small amount of residual flint, some of which was probably of Mesolithic date, was retrieved from contexts mainly in the Phase 4 and Phase 9 areas (Fig. 2). Although these are residual finds, they indicate a pre Iron-Age presence in the area. L ittle is known about prehistoric activity in Maidstone, the only evidence being from stray finds. Excavations near the County Council offices in Springfield revealed a small quantity of lithics (SMR Ref. TQ 75 NE 26) and towards the opposite end of town near Lockmeadow Market, two undated flint flakes were recovered (TQ 75 NE 391-KE17883). The former finds are the only other flint assemblages that have been specifically dated to the Mesolithic period in Maidstone. The flint assemblage from the site is therefore locally significant. Its proximity to the Medway would have provided an ideal environment for Mesolithic people, providing water and locally available stones for tools (Clark and Murfin 1995). R elatively little evidence of Iron-Age activity was previously known from the Maidstone area. To the east of the site, in 1889 a bronze spiked roller, part of an Early Iron-Age horse’s bit, was recovered from a garden on Union Street (TQ 75 NE 68-KE1943). Further east, in 1884 a rubbish pit containing pottery and animal bone was dated as Iron-Age and Roman. In 1963, across the Medway to the west, an Iron-Age cremation urn and associated pots were excavated during trial trenching at Haynes Garage on the Ashford Road. The group comprised of a pedestal urn, a fragment of a globular jar with cordon and chevron, containing a cremation and the spring of a fibula. These were dated as first century ad. Although these burials were located some distance away from the Fremlin Walk site, it is clear there were local settlements and cemeteries in the area that were contemporary with the activity at the Fremlin Walk site. Evidence for Iron-Age activity on-site consisted of pits and occasional linear features (Fig. 3). The features and finds associated with this period can be split into two sub categories based on pottery fabrics – Late Iron Age (pre-Roman Conquest) and Late Iron Age/Early Roman (‘Belgic’ Ware). The Late Iron-Age fragments are mainly composed of glauconitic wares, whilst the Belgic wares are mainly grog-tempered. The former can be artificially separated because the fabrics are known to have been present in Britain before the Roman conquest, whereas the latter occurs throughout the transition from Late Iron Age to the early Roman period. Due to the limited assemblage however, the date ranges for both remain lengthy, spanning the first century bc to first century ad for the former and the first century bc to the second century ad for the latter. Late Iron-Age (pre-Roman Conquest) activity consisted of five pits that contained pottery and butchered animal bone. One such feature, pit [568] contained 34 sherds of pottery, derived from at least eight vessels, which CATcatHERIN E EDWARDS 76 Fig. 2 Phased archaeological work. EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 77 Fig. 3 Iron Age features. CATcatHERIN E EDWARDS 78 Fig. 4 Iron Age Pottery. 1. Pit [568], P3. A necked bowl or cup with a horizontal cordon and a burnished exterior made from grog and organic tempered ware, (Fabric Goup 5) equivalent to B25 (CATcat). 2. Pit [568], P1. A bead rim with a notably thickened triangular section made from a coarse flint tempered ware (Fabric Group 1a), equivalent to 1A1 (Lyne 2003). 3. Pit [568], P2. A corrugated biconical jar made in a glauconitic fabric with a burnished exterior (Fabric Group 9: Pollard 1988, Form 23; Thompson 1982, Form B2-2) equivalent to B9.1 (CATcat). were largely glauconitic ware. Featured sherds within the fill included a bead rim with a notably thickened triangular section (Fig. 4, no. 1) made from a coarse flint tempered ware. The only rim in a glauconitic fabric is from a corrugated biconical jar with a burnished exterior (Fig. 4, no. 2) paralleled locally by a vessel from Quarry Wood Camp, Loose and Court Lodge Farm, Teston. Other sherds in similar wares (not illustrated) include examples of ‘combed’ horizontal rustication; deep horizontal lines above an area of fine criss-crossed ‘combed’ rustication; shallow-tooled horizontal lines; deep horizontal corrugations or furrows; and bands of shallow-tooled chevrons below a deep horizontal line. The grog and organic tempered ware, included in the assemblage, was used for a necked bowl or cup with a horizontal cordon and a burnished exterior (Fig. 4, no. 3). Two fragments of sandy sherds are decorated with nine closely spaced rows of point-toothed comb impressions (not illustrated). Comb stabbed motifs are relatively unusual, occurring on a small per-centage of ‘Belgic’ vessels from some of the east Kent sites including Canterbury, Richborough, Wye, Eastry and Highstead (Pollard, 1988). EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 79 The assemblage from feature [568] has been dated to the first century bc to the second century ad. T he ‘Belgic’ period is represented on-site by a linear ditch with associated postholes and several pits. The ‘Belgic’ assemblage on-site was composed of 107 sherds. As previously mentioned, due to the small assemblage and the lack of more diagnostic sherds, the dating for this period spans over three centuries, from the first century bc to the second century ad. The ditch and the postholes may have formed some sort of fence line or linear structure. The backfill of the ditch contained nine sherds of Belgic pottery dated as first century bc to second century ad, and an assemblage of animal bones all of which were identified as livestock species. No visible signs of butchery were observed on the bones, however it remains likely that these animals would have been used for the both meat consumption as well as farming bi-products such as milk and wool. T he majority of the remaining pottery assemblage of this date was recovered from features within the Phase 9 excavation and is likely to be residual. However one feature, grave [990] contained a single fragment of ‘Belgic’ pottery, the date of which may coincide with the date given to the burial practice associated with the inhumation, in this case, the deposition of hobnailed boots, which dates the burial at the earliest to the second century. T he evidence for activity of this date is limited but perhaps points to land and stock management on a modest scale, rather than settlement. The timber structure recorded in the Phase 9 excavation, mentioned below, although likely to be Roman, could possibly date to the Late Iron Age period; however, the lack of good dating evidence or associated features means an accurate date cannot be established. roman T he Fremlin Walk site is located to the west of the Roman road that would have linked Maidstone to Lympne and the Greensand hills. The road would have been used for trade and transportation links to the south of Kent, especially to the iron producing areas of the Weald. Maidstone is thought to have been a dispersed and wealthy agricultural settlement throughout the Roman period (Clark and Murfin 1995; Pollard 1988). This is echoed by the number of villas in the area, such as Mount Villa (Houliston 1999) (TQ 757 562), the recently discovered villa on Bower Lane (TQ 751 552), and the large villa at Teston (TQ 690 530). Other substantial Romano-British buildings have been identified at Barming (TQ 72 53), East Barming (TQ 72 54), Boughton Monchelsea (TQ 77 51) and Chart Sutton (TQ 80 49) (Pollard 1988). The Roman archaeological features on-site dated to between the first and the third/fourth century ad. The dating of the features was based on CATcatHERIN E EDWARDS 80 pottery and other typologically dateable finds. There is little stratigraphic evidence on site, and in areas where only cut features survived, it remains difficult to determine the depositional sequence. The Roman features were mostly located within the western part of the site in Phases 2, 4 and 9 areas and consisted of mainly ditches and postholes forming possible boundary lines or enclosure plots, as well as a portion of a Roman cemetery and the remains of large timber building (Fig. 5, and see below). This evidence suggests that the local area was probably rural during the Roman period and the activity that took place may have been related to either Mount Villa to the north or to the Roman road from Rochester to Hastings that followed the line of the present Week Street to the west. Early Roman (1st - 2nd century) O nly three early Roman features were recorded on site, these were recorded as a solitary oval pit and two linear ditches. The features contained pottery dating to this period as well as cow and goat/sheep Fig. 5 Plan of the Roman cemetery within the Roman landscape. EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 81 bone. The linear features may have been used as boundary ditches, or for the drainage of cultivated plots. Later Roman (2nd - 3rd century) Several features were dated specifically to this period including two linear features, one recorded in the eastern limits of Phase 4, and the other recorded in the eastern part of the Phase 9 area. The former ditch contained fragments of cow bone and 17 fragments of pottery, dating the feature to 120-300, whilst the latter contained 34 sherds of Thames Kent Ware pottery, all dating to 180-300. It is probable that this ditch formed a boundary to either the timber building to the west, as it ran on a similar alignment or as a boundary between the three burials that were also located in the Phase 9 excavation area and the five burials in the Phase 4 excavation. Two other pits in the Phase 4 excavation also contained small amounts of pottery, dating to 120-250, and mid first to early third century respectively. Roman (not closely dated) T he majority of the Roman features on the site could not be closely dated mainly due to the lack of diagnostic pottery sherds and other dateable finds. A possible ditch or linear pit [342] ran ne-sw towards the eastern limits of the Phase 4 excavation area. The ditch contained ceramic building material, animal bone and a pottery assemblage composed of 5 sherds of Roman pottery dated to mid first to early third century, along with 12 sherds of Iron Age and Belgic pottery. Included in this assemblage was a rare sherd of samian rouletted variant Dragendorff form 30 bowl, which was dated to ad 70-150; and a small piece of a Roman figurine, unfortunately too degraded to ascertain any form or symbolic relevance. The range of pottery from this single fill might be explained in a number of ways: the ditch was open and in use during both the Iron Age and the earlier Roman period; or the ditch may have originated in the Iron Age and was backfilled during the Roman period; or the Iron-Age pottery is purely residual, having been deposited during the backfilling in the Roman period. T wo further ditches were recorded on site; one was located within the Phase 4 excavation area, whilst the other was recorded during the watching brief towards the north of the site. Both ditches contained butchered animal bones as well as assemblages of pottery that could only be loosely dated as Roman. Included in the finds assemblage from both ditches were residual sherds of Iron Age pottery, including examples of flint tempered ware bead rim vessels. The inclusion of so many sherds of Iron-Age and ‘Belgic’ pottery within this fill is interesting as it questions whether the pottery was CATcatHERIN E EDWARDS 82 merely residual or whether some of this pottery was in continual use during the early Roman period on site. The ditch itself was probably part of a boundary or enclosure, and its disuse indicates a change in land use and division or the changing of settlement boundaries. T he remaining features dating to this period were pits, of varying sizes. Seven small pits, all approximately the same size, were recorded either side of the linear feature in Phase 4. Only three pits contained pottery which dated them generally to the Roman period. Two large pits within the Phase 4 excavation and three within the Phase 9 excavation were dated to this period. Again there was a mixture of Late Iron-Age and Belgic pottery amongst the Roman assemblages, although these are likely to be residual. Also recovered from one of the features in Phase 4 was a copper alloy brooch pin <6>, which has been identified as characteristically Roman. The function of these pits remains obscure; however the large pits within the Phase 9 excavation may relate to the timber building located approximately 1m to the southeast (see below). Environmental samples taken from the Roman period features revealed slightly more fruit remains than the preceding period, including strawberry/cinquefoil, figs and elder, although the last two were only present in small quantities. Only occasional charred plant remains were present in samples from this period. Roman pottery T he Roman pottery assemblage consists of 285 sherds, dating between the mid first and the late third or mid fourth century ad. As with the earlier pottery this includes a series of small single period assemblages, which may be in situ but are just as likely to be residual. The assemblage is dominated by coarse wares, 30 per cent of which are grog tempered including a significant proportion of ‘Patch Grove’ sherds with an origin in the mid first century ad. The few featured fragments are likely to be from large storage jars, a type with a longer history of production than other forms continuing into the early third century (Pollard 1988). L ocally produced reduced sandy fabrics and black-burnished wares with a second-century origin are also prominent within the assemblage. These include BB2 of Hadrianic date (125 to 145) and Thames-side Kent ware produced between 180 and 300. These fabrics were also in the majority amongst the coarse pottery from the Mount Roman villa (Kelly 1992; and Savage 1999) and are a notable component within the assemblage from Churchfields, Snodland (Seager-Smith 1995). It is very likely that the north Kent ceramics were reaching Maidstone up the Medway, explaining their prominence within this area (Pollard 1992). T he earliest of the Roman imports, dated to c.ad 50-80, was a rim fragment from a Central Gaulish white ware butt beaker. A rim from a EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 83 relatively rare colour-coated roughcast folded beaker, similar to examples from Sinzig (Fig. 6) was also recovered from the site. The central Gaulish samian includes a fairly rare sherd from a rouletted variant of a Dragen-dorff Form 30 bowl, dating between ad 70-150 and a fragment from a Les Martres-de-Veyre dish. Fig. 6 Imported Roman pottery. Pit [112] P8. Rim of rare colour coated roughcast folded beaker (Symonds 1990, 8, No. 40) equivalent to R80 (CATcat). Cemetery Eight burials were excavated on the site, five in the Phase 4 excavation and three in the Phase 9 excavation (Fig. 5). The graves demonstrated several different orientations varying between north-south and east-west. Three of the eight burials were provided with grave goods, which illustrate clearly the distinction between items worn on the body at burial and those placed with it as equipment for the afterlife. In grave [948] the skeleton was adorned with two pieces of jewellery, a shale armlet <14> (Fig. 7) and a copper alloy ring <13> (not illustrated). Both objects were placed on the right hand (middle finger) and wrist of the body. The position of the body itself was also very interesting as the arms from both sides were raised up and placed behind the head. 05cm1:21:410cm0Fig 7. Shale Armlet1231. Grave 948 <14> [947] Shale armlet (Skeleton 947) Complete, in two fragments. Diam (ext) 83 mm; (int)59 mm; H 21 mm; W 13mm. Plain ring with D-shaped section. All surfaces, including the interior, are well finished and smooth. Fig 6. Imported Roman potteryPit [112] P8. Rim of rare colour coated roughcast folded beaker, (Symonds 1990, 8, No. 40) equivalent to R80 CAT. 05cm1:21:410cm0Fig 7. Shale Armlet1231. Grave 948 <14> [947] Shale armlet (Skeleton 947) Complete, in two fragments. Diam (ext) 83 mm; (int)59 mm; H 21 mm; W 13mm. Fig 6. Imported Roman potteryPit [112] P8. Rim of rare colour coated roughcast folded beaker, (Symonds 1990, 8, No. 40) equivalent to R80 CAT. Fig. 7 Shale armlet. Grave 948 <14> [947]. Shale armlet (Skeleton 947). Complete, in two fragments. Diam (ext) 83mm; (int) 59mm; H 21mm; W 13mm. Plain ring with D-shaped section. All surfaces, including the interior are well finished and smooth. CATcatHERIN E EDWARDS 84 Grave [244] only contained the remains of a right arm along with twelve fragments of copper alloy which may have formed an armlet <18> (not illustrated). Whether this is the entire burial is unclear. It is thought that the arm may belong to a truncated burial or a burial that has been previously excavated. There have been occurrences, dating to the Iron Age, of burials that have contained disarticulated human bones (Drewett et al, 1988). I tems of jewellery such as the copper alloy ring are among the grave goods most commonly recovered and when worn, it is suggested that they are objects which belonged to the deceased and were part of their dress, implying also that these bodies were fully clothed for burial and not simply wearing shrouds. One of the reasons for the burial of jewellery, whether or not it was worn, was likely to be apotropaic. Personal possessions were thought to be imbued with the owner’s spirit which could become a malign influence after death, and were best placed safely underground. The choice of a shale armlet for the burial of the mature female in Grave [948] may be significant. In the later Roman period jet was both fashionable and of significance in a burial context for its apotropaic and mystical qualities, noted for example by Pliny (NH 36). Shale armlets are found over a longer period but when oiled and polished may have appeared little different from jet. Although both shale and jet were popular in the third and fourth centuries, the size of this armlet might suggest an earlier date. Grave [990] was very badly horizontally truncated, probably during the original construction of the road. Within the burial were the remains of a pair of hobnailed boots <25-28> (not illustrated) laid directly on top of the lower legs of the burial. Only the nails remain but the outline of the nails suggest that the boots were placed side-by-side but facing opposite directions. Whether this deliberate act has a symbolic relevance is unclear. The placing of shoes in graves is a well-attested custom in Roman Britain, although it can be debated whether they were provided for the journey to the underworld or for use in the afterlife. In this case, where the shoes were not worn as part of the burial costume, but placed side by side in the area of the (disturbed) lower left leg, their deposition is clearly a ritual act. Dating evidence within the grave fills was varied, ranging from residual Iron Age through to ‘Belgic’ and Roman. The range of pottery dates from the cemetery indicates that it may have been in use in the late Iron Age/early Roman period and became disused around the third century. The burial rite of inhumation and the provision of personal ornament and grave goods, falls within the Roman-British tradition and makes it probable that these burials date, at the earliest, to the second century. This date overlaps slightly with the dates assigned to the pottery assemblages from the majority of the graves. The identification of a copper-alloy EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 85 object <18> in one grave might date this burial to a slightly later date, in the third or fourth century. This would also tie in with pottery recovered from two graves which contained pottery dated as mid first century ad+ and 250 ad+. Timber structure A structure composed of ten stone-packed postholes was located within the Phase 9 excavation area (Fig. 8). The postholes formed a right angle corner of a building (measuring at least c.15 x 5.5m) with two posts on a second, possibly internal line. The cuts were sub-circular, with steeply sloping, almost vertical sides. Seven had a flat base whilst three examples contained the remains of a post pipe in the centre of the cut. Fig. 8 Roman timber structure. CATcatHERIN E EDWARDS 86 S ix of the ten postholes contained pottery. Of particular interest was a large handmade Patch Grove thick walled tub. Attached to the inside of the largest fragments were iron nails. The nails were large, approximately 50-60mm in length and radiography shows that they were all of the same form. It has been suggested that this type of nail was used for building construction. This object may have been deposited during works on the timber structure, such as its demolition or repairs; or the nail-filled pot may have been damaged or broken during transportation and then thrown away as rubbish or damaged goods. D ating this structure is complicated by the wide date range observed in the pottery recovered from the six postholes. Several fragments date from the first to second century whereas other fragments have been dated from ad 50-400. Although the building appears to date to the Roman period, no specific date [closer dating?] within that period can be established. Other finds recorded from these features included iron fragments, mainly nails; fragments of cow, pig, sheep and goat bones and a few fragments of daub, some showing signs of a wattle impression. medieval Evidence for the medieval period in Maidstone is limited in both its volume and location. The majority of excavated sites dating to this period are limited to the southern area of Maidstone, in particular the areas surrounding the Archbishops Palace and All Saints Church. The Palace was constructed between 1348-1366, whilst the church is dated to 1395-1398. The investigations that have taken place on both structures and the local area, have uncovered previously demolished sections of the palace and various outbuildings. A small assemblage of pottery was recovered the palace during investigations by the Maidstone Area Arch-aeological Group; as yet these investigations remain unpublished. Little archaeological evidence has been uncovered relating to the settlement of the population or evidence of their everyday activities. T he northern area of Maidstone which has undergone more development in recent years when compared to the south has yielded further evidence relating to the general population of medieval Maidstone. The first discovery, in 1889, comprised a collection of thirteenth-century jugs, cooking pots and bowls recovered during construction works for the Bentlif wing of Maidstone Museum. Little is known about what type of contexts these finds were collected from though they are probably from pits (Spillett et al. 1942; Maidstone Museum acc 150.1989/11-12/13). Similar finds were recovered from pipe- laying trenches excavated in 1959 under the Old Reference Library. Again the contextual information was not recorded (Maidstone Museum acc 103.1959). Nearby in Week Street, a medieval pottery kiln, pottery and kiln furniture were discovered EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 87 in 1921 (Grove 1967; Maidstone Museum acc no.4.1921). No other features are known to have been excavated or recorded. In 2003, a watching brief conducted by AOCaoc Archaeology on service trenching in Brenchley Gardens adjacent to the Museum and Art Gallery (site code BGM03) revealed 73 burials dating from the medieval to post-medieval periods. These remains were probably associated with St Faiths Church which dates to the thirteenth century. A large quantity of unexpected medieval archaeological features was uncovered at Fremlin Walk, as well as the largest single assemblage of medieval pottery in Maidstone and the first to be studied in detail. The majority of features from this period were identified as rubbish and cess pits, located in the north and eastern areas of the site (Fig. 9). This suggests that rather than the site being used for occupation and settlement, it is likely that the areas were utilised by the local community for their waste disposal. The sequence of features in the Phase 8 excavation are likely to be related to buildings that would have fronted Week Street and are likely to have been used over a considerable time for the dumping of domestic waste. Medieval (late 11th to early 13th century) S ix features dated to the earlier medieval period. Two large pits excavated in the Phase 4 excavation contained pottery dating from the eleventh to the thirteenth century and represent the first phase of cess/rubbish pits on site. Towards the eastern limit of the site (Phase 8), was a layer which might be the remains of an occupational deposit or remnant of garden soil. Five sherds of pottery were recovered from the layer dating to 1100-1250. Above this deposit was a rectangular feature filled with un-mortared limestone cobbles which may have provided a foundation or base for a post or structure. Unfortunately there were no obviously related features associated with this possible structural remnant. The pottery recovered from this feature could only be dated approximately to 1175-1350, due to the lack of diagnostic sherds. These features represent the development of Week Street during this period, with the construction of road side structures, occupancy and the remnants of daily life. Medieval (13th - 14th century) T he majority of medieval features on-site were dated by pottery to 1200-1350/1400. Some of the features have a closer date range of 1200-1250. Features of this period were spread across the entire site in the form of linear features and pits. Several of the linear features appear to have been small drainage gullies, such as the two narrow linear features excavated in the Phase 4 area. Both features contained animal bones, peg tile and pottery dating to this period. Other larger linear features excavated during CATcatHERIN E EDWARDS 88 Fig. 9 Medieval features. EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 89 the watching brief phase, formed an interrupted right angle, which are the result of removed or robbed-out wall lines of a possible structure. An excavated slot through these linear features contained a fragment of a jar handle that was decorated with a series of oblique slash marks down one side (Fig. 10, no. 1). Chemical analysis performed on this sherd and ones of a similar fabric type and composition from the site suggest that they are closely linked to the samples that have been taken previously from the medieval pottery kiln on Week Street (SMRsmr TQ 75 NE136 – KE1992) and the other samples from Maidstone and Rochester. This does not suggest that they were produced at this kiln, as thinned sections of the pottery shows that they are fabricaly different, but it does suggest a similarity between the clays that have been used to make the pots, suggesting a local source and an as yet undiscovered kiln. T his period represents an increase of rubbish and cess pitting on-site as well as the size of the pits. These pits were recorded across the site in Phases 4, 7, 6 and 8 and the general site watching brief. The majority of features contained examples of butchered animal bone, glass, tile and varied amounts and styles of pottery. Of particular note was a large and deep pit, excavated during the watching brief phase to the north of Phase 4. The pit contained an intact, whole cooking pot (Fig. 10, no. 2), and over 20 sherds of another, both dated to 1200-1250. Inside the whole pot were cereal grains and grass seeds. I n both Phase 7 and Phase 8 and the adjoining watching briefs, a number of rubbish and cess pits were excavated dating to this period. The circular Fig. 10 Medieval pottery. 1. Linear [541], M11, decorated jug handle with a series of long, oblique slash marks down each side (Kent fabric code M38A2). 2. Pit [526], M1, whole shell-tempered jar (Kent fabric code EM36). 21Fig 10. Medieval potteryFig 11. Worked horse bone. The metatarsal has a smooth and shiny shaft with a deep groove running diagonally across the centre of the bone 255mm 415cm01:415cm01.Linear[541],M11,Decoratedjughandlewithaseriesoflong,KentFabricCodeM38A2)2.Pit[526],M1,Wholeshell-temperedjar(KentFabricCodeEM36). CATcatHERIN E EDWARDS 90 pits in the Phase 7 area contained peg tile, chicken bone and pottery as well as a large quantity of seeds, grains and pulses, which suggests the pit was used to dump domestic waste such as cess. The cluster of activity towards the northern limits of Phase 8 represented domestic dumping, possibly thrown away from the back of a property on Week Street. The five pits recorded in this area contained occasional peg tile, cat remains, butchered animal bones and pottery sherds including an almost complete early medieval jar and 14 sherds of another, both dated to 1100-1250. The features recorded in these phases, represents domestic activity on what is now St Faiths Street and Week Street, such as back yard or waste ground domestic dumping of both rubbish and cess waste. Medieval (15th to early 16th century) T he site continued to be used for domestic waste dumping throughout the late medieval period. Three circular rubbish pits within the Phase 4 excavation area contained domestic waste. Similar features were recorded in Trench 16 in the Phase 2 evaluation and the Phase 7 excavation just to the north, where pit [403] contained fragments of an imported Langerwehe drinking vessel. This pit also contained a horse metatarsal with a smooth and shiny shaft suggesting extensive handling. The plantar surface of the bone has a deep groove running across it diagonally slightly downwards from the medial to lateral sides, (Fig. 11). Metapodials (cannon bones) from cows and horses are very strong dense elements and thus are often used in bone working. No other example of this has been found in any standard works. The most reasonable suggestion thus far is that the bone may have been used as a tool in leather working to stretch straps. This find may be the first indication of industry on site. Although the bone is an isolated find, it may represent the development of Week Street and the beginning of small, possibly domestic, industry. Fig. 11 Worked horse bone. T he metatarsal has a smooth and shiny shaft with a deep groove running diagonally across the centre of the bone (255mm long x 44mm wide). EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 91 p ost-medieval Post medieval activity was recorded in almost all parts of the site and this period is the best-represented on-site (Fig. 12). This reflects the changes within Maidstone town itself during the late medieval and throughout the post-medieval period. The archaeological remains dating to this period were mainly ditches, rubbish and cess pits, backfilled wall footing and basements, soakaways and wells. This evidence shows large-scale use of the site by the inhabitants for both habitation and for the deposition of waste. The majority of features contained a number of finds including pottery, glass, clay pipe, brick and tile. The archaeological and cartographic evidence shows that the site has been subject to many phases of construction and demolition. The earliest map of 1650 shows that open areas mainly occupied the site; one such plot was possibly an orchard owned by Thos Fletcher. To the east of Havock Lane, which was already established at this point, the land was split between at least 16 different owners, each with a plot and a property; and most of which front onto Week Street. Some of these properties were stores, small industries and commercial properties. Unfortunately the cartographic evidence between 1650 and 1821 is not clear enough to identify the site and/or any changes within it. By 1821 the site had developed into a densely packed area of buildings, with the beginnings of what was to become Market Street, visible as an unnamed lane. Brewery buildings occupied the western part of the site whereas more buildings with associated backyards and gardens occupied the eastern half of the site. The development of Museum Street and the widening of Havock Lane by 1936 promoted the demolition of a number of properties both commercial and domestic. It is these phases of development and demolition that are reflected in the archaeological record of the site. Late 15th - 17th century In the Phase 4 excavation area, five large, four small pits and three postholes were recorded. The pits contained domestic and commercial debris including a copper alloy thimble (not illustrated), which would have originated from the properties that previously occupied the site. The postholes may have been used as a structural frame for a building or as a fence or division line. No further evidence for a structure of this date was recorded on site M ore interesting, however, are a series of pits located towards the northern centre of the site (Phase 7) and a series in the north-eastern section of the site (Phase 8). These two clusters of pits represent continual dumping of cess and domestic waste within a small area. The series of nine pits recorded in the Phase 8 excavation contained various finds CATcatHERIN E EDWARDS 92 Fig. 12 Post-medieval features. EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 93 associated with domestic waste such as various forms of pottery including a rim sherd of a pipkin in Wealden type redware, a knife blade <33> (not illustrated), butchered animal bones including fragments of sheep, cow and goat and fragments of copper alloy, one of which may have been part of a vessel <21> (not illustrated). It is likely that these pits would have been backyards or gardens of the properties fronting Week Street. The lack of industrial waste within these features suggests that these properties are more likely to be residential than commercial or industrial properties. O ne of these pits mentioned above was cut by four blocks of green sandstone which formed a small uneven structure. The blocks were badly eroded and several blocks were apparently missing. The blocks measured 0.40m long and 0.10m wide. It is likely that this small structure is related to deposits of burnt material below them, suggesting that they were part of an outside hearth or oven. It is possible that this could have been related to some small-scale industrial process, although no industrial residues were recovered. T he pits recorded in the Phase 7 excavation included four large intercutting pits that not only contained cultural finds but also provided interesting environmental data such as seeds, pulses and grains in the waste, either from the cess being dumped or from food waste. Finds recovered from the pit fills included pottery, butchered animal bone and fragments of a small container lid made from copper alloy, <5> (not illustrated). The pottery assemblage from pit [427] was dated to 1575-1650 and it contained the best example of redware from the site, a virtually complete black glazed conical tankard (Fig. 13, no. 1). It measures 249-250mm in height and would have had a 2 pint capacity. The conical shape is probably derived from the more elaborately decorated sixteenth/seventeenth-century German stoneware Schnelle forms made at Siegburg, Raeren, Cologne, Frechen and elsewhere. This in turn would have been influenced by leather forms. The origin of this piece remains unclear but it may be a product of Staffordshire, Harlow or Wrotham. Six other smaller pits were excavated in this area, each containing fragments of butchered animal bone, tile and pottery including a tripod pipkin dated to 1550-1700. The organic fills in these pits suggest that they were being used as cess pits and possibly places to dump domestic waste. As these pits cut earlier medieval pits of similar character, it is possible that this area was continually used as a location of cess pits from the medieval and into the early post-medieval periods. Fourteen features dating to this period were excavated during the watching brief phases on site. Seven of these features were linear whilst the remainder were circular or rectangular pits. The majority of the linear features were wall foundation trenches or related to drainage. Several of the pits were the remains of backfilled soakaways; however, the CATcatHERIN E EDWARDS 94 majority of the pits were the truncated remains of rubbish pits. All of the pits contained animal bones, tile, rubble, fragments of copper alloy and various fragments of pottery generally dating to 1550-1725. Interesting examples of pottery included a fragment of a slip coated jug with five stab marks, three closely spaced, inside the attachment of the oval handle dated to 1450-1550 and an almost complete red earthenware type 3 Fig. 13 Post-medieval pottery. 1. Pit [427] PM29. Complete black glazed redware tankard, 249-250 H. There are three raised cordons just above the base, and four below the plain rim, from which the strap handle springs (Kent Fabric Code PM1). 2. Pit [143] PM24. Complete Spanish olive jar of Goggin type C, c.337mm L. Dated to c. 1575-1625 (Kent Fabric Code PM22). 3. Pit [193] PM10. London stoneware tankard with excise mark of William IIIiii datable to 1700-1702 (Kent Fabric Code PM25). 2131:420cm01.Pit[427]PM29.Completeblackglazedredwaretankard,249-250mmH.andfourbelowtheplainrim,KentFabricCodePM1).2.Pit[143]PM24.CompleteSpanisholivejarofGoggintypeC,c.337mmL.1625(KentfabricCodePM22).3.Pit[193]PM10.LondonstonewaretankardwithexcisemarkofWilliamIIIdatableto1700-1702(KentFabricCodePM25).Fig 13. Post-medieval pottery EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 95 Martincamp flask which was unfortunately discovered broken in situ. Also recovered from one pit were fragments of leather, which included part of a stacked heel from a shoe. The thickness is made up of a number of layers of leather fastened together with both wooden and metal tacks. Pottery recovered from the feature was dated to 1550-1700. Human Remains A post-medieval burial was excavated in Trench 16 during the Phase 3 evaluation. The individual was buried in a well-preserved shouldered and tapered wooden coffin. This measured 1.69m long, 0.29m wide (shoulders) and 0.25m deep and was orientated west to east. Dating evidence, indicated by two small sherds of pottery, suggests a date of 1200-1725. The structure of the coffin itself suggests a date of post-1660. Located on the coffin was a wooden ovoid vessel measuring 0.39m long, 0.24m wide and 0.16m deep. The vessel was constructed of wooden vertical staves with one piece of wood for the base. It is believed that the vessel would have contained salt. This practice is well-documented; the salt was believed to take any sins away (pers comm. Julian Litten). The individual was buried with a bone hair-pin placed under the head (probably within the hair) but unfortunately this disintegrated on excavation. The individual was buried supine in the coffin with the arms and legs extended. T he skeletal remains were of a juvenile aged between 12 and 15 years at death. The remains displayed a good level of surface preservation and was represented by 90 per cent of the skeleton. The left ulna was truncated during the mechanical excavation of the trench. Schmorl’s nodes were observed in T6-10. No other vertebral pathology, joint disease, enthesopathies or gross pathology was observed in the skeleton. It is unclear why this burial has been located away from the known cemeteries in Maidstone. It is possible that superstition surrounding an illness, forced the juvenile to be buried individually. 18th - 19th century T he wealth generated from the industrialisation of Maidstone and the growth of the town as a residential as well as a commercial centre is reflected in the large quantity of archaeological features of this period. In the Phase 4 excavation, nine large sub-square pits were recorded which probably represent demolished and backfilled cellars or basements. These are likely to relate to properties fronting Earl Street. The backfill of these features contained fragments of tile, brick, animal bone and pottery. Of particular note, was an unusual and complete carrot-shaped Spanish olive jar of Goggin type C, dated to 1575-1625 (Fig. 13, no. 2). Also of interest CATcatHERIN E EDWARDS 96 was a fragment of porcelain with overglaze floral design painted in red and green (not illustrated). The style of the pottery is very English, though the porcelain is almost of Chinese quality, which may suggest a European piece or a Chinese piece created specifically for the European market. I n Trench 16 a circular pit contained several horse long bones. It is estimated that the horse would have stood at approximately 16.2 hands tall. The bones showed signs of well-developed muscles and this combined with the estimated size suggests that the bones may have been from a draft horse, possibly a shire horse that was commonly used by breweries. Other horse bones recovered from another pit within Trench 16 have been estimated as pony size. These features were located close to the Fremlin’s Brewery but were also close to a veterinary surgery and several stables that are indicated on the 1912 Goad Fire Insurance plan. T he remaining features dating to this period represent the dumping of both domestic and commercial waste, in the form of large and small pits; the remains of buildings previously occupying the site; over 39 postholes, soakaways and wells. The postholes had no obvious structural pattern but probably represent the truncated remnants of property plot layouts and fence lines from tenements that occupied the site during the post-medieval period, either for habitation or for small-scale industry. There were also several wood-lined ditches filled with dark brown or grey blue silty clay. These deposits had a very strong odour and contained leather and animal bone suggesting they may have been associated with an industrial process or possibly stabling. T he pitting observed in Phase 8 in the previous period continued into the eighteenth and nineteenth centuries. The same backyard domestic deposits were recorded and one feature contained over 47 sherds of pottery, including a large handled bowl, a large dish and a flowerpot. Pit [193] Pit [193] in the Phase 4 excavation, contained an extensive assemblage of household debris including 15 items of glass dated 1725-1780. These included pharmaceutical phials (Fig. 14, nos 1 and 2); wine glasses (Fig. 14, no. 4); wine bottles, including a complete example of unusual type with a flattened octagonal section – and the body and the base from two other multi-sided bottles (Fig. 14, nos 5 and 6); decorated glass bases, (Fig. 14, no. 3) and other household vessels. As well the large array of glassware, over 57 sherds of pottery from an estimated 21 different vessels where recovered from this feature. An example of London stoneware had a stamped excise mark to the left of the handle (Fig. 13, no. 3). The stamp is very faint but is shows the letters WR under a crown. It dates to between 1700, when the Act for ascertaining the measures of ale and beer was introduced, and 1702, the year in which William IIIiii died. Not only EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 97 Fig. 14 Eighteenth-century glass from pit [193]. 1. Pit [193] G10-15. Complete; cylindrical; Ht 78mm; rim Diam 24mm, base Diam 29mm; pointed basal kick. 2. Pit [193] G10-15. Complete; cylindrical; Ht 64mm; rim Diam 21-22mm, base Diam 25mm; pointed basal kick. 3. Pit [193] G2. Remains of vessel body and solid glass base with foot. Decorated with optic-blown moulded decoration; possibly the base of a jug, jar or bowl. 4. Pit [193] G1 lower half of a trumpet bowl. Solid stem with central teared inverted baluster knop. 5. Pit [193] G3. Complete wine bottle with moulded eight-sided body, long neck, high string rim and almost flat oval eight-sided base. 6. Pit [193] G4. Rectangular eight-sided base from an eight-sided wine bottle. CATcatHERIN E EDWARDS 98 does this fragment connect Maidstone to London but it is also evidence of the trade of ale and beer between the two. Clay pipe waster pit [501] T o the north and east of the Phase 4 excavation area, one rectangular pit [501] of particular note was excavated to the south of Market Street. It was filled with a large assemblage of clay pipes, both whole and broken and fragments of a smithing hearth bottom, slag and hammerscale spheres. Several of the fragments were marked with a makers stamp. Four pipes were marked with CB (Fig. 15, nos 2 and 4), which is as yet unidentifiable whilst 64 fragments were stamped with HH (H. Hunt), a maker in Maidstone 1845-7 (Fig. 15, nos 5 and 7). All the decorated examples of clay pipes found on-site were located in this one context. The decoration varies from foliage, to leaf steams (Fig. 15, no. 8) to vertical ribbing. There is little evidence that any of the pipes had been smoked and with the inclusion of wasters and overfired examples and kiln debris it is likely that the manufacture took place close by and that the production waste was dumped on site. Pottery recovered from the feature [501] was dated to 1820-1900 and the entire clay pipe assemblage has been dated to 1820-1880. Other examples of clay pipes with the makers mark were recovered from Trenches 6 and 16. One example from Trench 6, was marked with RB (Fig. 15, no. 1), which is thought to signify the pipe maker Robert Hornsby, recorded in Canterbury in 1715 (ibid., 75). Pit feature 16/004 in Trench 16 contained a fragment of clay pipe marked with the initials EM moulded in relief on the sides of the heel (Fig. 15, no. 3). These initials probably represent Elizabeth Middleton, recorded as a pipe maker in Maidstone in 1724-32 (Oswald 1975, 175).This collection is both locally important as well as of a wider significance as Maidstone was a known centre of pipe manufacture. conclusion T his excavation, the largest to date in Maidstone, has expanded our knowledge of the town and its development. The results have provided more detailed archaeological evidence for the early origins of human activity in the area; expanded our knowledge about the nature of the Roman occupation of the town and demonstrated the expansion of settlement and commerce throughout the medieval and post-medieval periods in line with historical and cartographic sources. T he presence of Iron-Age activity on-site was unexpected. Iron-Age and Belgic sites are well known throughout Kent, with large pottery assemblages recorded at sites locally such as Mount Villa and also sites at Aylesford and Loose. The occurrence of Iron-Age and ‘Belgic’ pottery EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 99 Fig. 15 Decorated clay pipe bowls. 1. Pit 6/039. C2 Type AO 25. Maker’s mark Robert Hornsey. 2. Pit 501. C4 Type AO 28. mark B (unknown) 3. Pit 16/004. C1. Type OS 12. Maker’s mark Elizabeth Middleton. 4. Pit 501. C3 Type AO 28. Maker’s mark B (unknown). 5. Pit 501. AO 28. Maker’s mark H Hunt. Pipe decorated. 6. Pit 501. C6 Type AO 28. Maker’s mark H Hunt. Pipe decorated. 7. Pit 501. Type AO 29. Maker’s mark H Hunt. Pipe decorated. 8. Pit 501. C8 Type AO 29. Pipe decorated with leaves and foliage on bowl. 1234567802.5cm1:1Fig 16. Decorated clay pipe bowls1.Pit6/039.C2TypeAO25,maker’smarkRobertHornsey2. CATcatHERIN E EDWARDS 100 in a number of the Roman burials and in the postholes of the timber structure is curious. It is possible that the cemetery had its origins in the Late Iron Age and that it was still being used during the early Roman period. The mixed dating evidence from the timber structure can be interpreted two ways. The structure is probably a Roman building with some earlier residual pottery incorporated into the posthole fills however it is also possible that the structure was originally an Iron-Age building, which was demolished and removed during the Roman period to allow for the expansion of the cemetery. The phenomena of Iron-Age rectangular buildings has been documented: ‘there is some evidence that rectangular buildings, perhaps became more common by the end of the Iron Age, and in the south-east there is a correlation between the demise of circular huts and emergence of the Belgic dynasties’ (Drewett et al. 1988). However, an Iron-Age date for this structure is speculative. D ata retrieved from the excavation indicates that it is likely that the area would have been open land remaining unutilised or used only for animal grazing. The latter seems most likely due to the animal bone evidence recovered from the Iron-Age features. The assemblage for this period comprised of 43 specimens identified as horse, cattle, sheep/goat sheep, pig, dog, cat and chicken. Cattle are represented by a wide range of elements, and sheep/goat by head, forelimb and hind limb parts. All of which, could have been utilised for food consumption. For the other taxa a more restricted range of body parts are present, cat and chicken being represented by single bones, suggesting their presence was not through consumption but via disposal. The majority of the species indicated above were of adult age at their death. There are very few juvenile or sub-adult bones, which suggest that the animals were kept to produce bi-products such as milk, wool and eggs, and were killed much later in their life. There is no evidence for industrial preparation of meat on-site which suggest that the assemblage is the result of small scale domestic husbandry or that the meat was brought in from elsewhere. T his landscape continued into the Roman period as the majority of features recorded on-site dating to this period were ditches, pits and postholes. There was little building debris on-site such as roof tile, floor material or building stone, which suggests a paucity of structures on site. The Fremlin Walk site borders a Roman road, Week Street, at the eastern limit of excavation. Evidence of Roman occupation in the immediate area has been limited to a single Roman structure, excavated at the corner of Week Street and the High Street and to the south east of the site (TQ 75 NE 139-KE1995) (Kent County Council SMRsmr). The presence of a timber structure on-site is interesting but there was very little evidence for the purpose or use of the structure, mainly due to extensive horizontal truncation that would have removed all traces of floor surfaces. A number of timber structures were excavated on the Mount Villa site, which was EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 101 located 0.5km to the north of the Fremlin Walk site (Houliston, 1999), some of which were believed to be animal pens or wheat stores, whereas another larger timber-built structure is believed to have been an early build of the villa, a ‘proto’ villa. The Mount Villa timber structure has been dated to the third phase of activity on that site, ad 175-225. The structure was formed with postholes approximately 0.3-0.4m in diameter. They were set in large pits filled with compacted clay and ragstone fragments (Houliston, 1999), which is a similar construction technique to the Fremlin Walk timber structure. The lack of associated evidence means that it is difficult to draw a firm conclusion regarding the nature, date and use of the structure. The Mount Villa had a ‘proto villa’ on-site and it seems unlikely that one would be built this distance away from the final site, unless the Fremlin Walk site was the original construction site that was abandoned. If the timber structure is a ‘proto villa’ not associated with Mount Villa then it may be associated with a fully-developed villa, which is as yet undiscovered. T he burials from the Fremlin Walk site are undoubtedly associated with several cremations that have been discovered in Maidstone. In 1850 a Roman cremation was excavated in St Faiths Street. This find follows a similar discovery, 20 years before, of a number of small urns during the excavation of foundations further east along the street. Unfortunately no record of these finds can be found and it seems no further analysis of the urns and their contents took place. A similar assemblage of cremation urns were recorded in 1715 to the south of the site in Earl Street (then Bullock Lane) and Pudding Lane. The urns are said to have contained ashes with human remains scattered nearby. In 1932 a Romano-British burial group was uncovered during service trenching along the southern end of Havock Lane. The group consisted of a large cinerary urn containing the ashes of a child, a jug, a Drag 32 plate and a small globular glass pot. These were dated to the mid second century ad. In the same year a second group of possible burials was excavated during construction work on Museum Street. This group consisted of a large jar, a flagon, a dish, and a samian cup stamped Macerati. This assemblage was thought to date to the early second century. Finally a Romano-British cemetery was excavated in 1959-60 at Westborough, located towards Maidstone West Station. The cemetery contained about 25-30 skeletons and some 150 cremation urns. No evidence of further work has been carried out on these finds and no date other than Roman has been assigned (Kent County Council SMRsmr Records). O f particular interest with regards to the burials recovered from the Fremlin Walk site, is their proximity to the Mount Villa. Elsewhere in the Maidstone area several of the Roman villas have been associated with walled cemeteries, such as East Barming, Boughton Monchelsea and Westborough. Houliston (1999), suggests that ‘although romanized CATcatHERIN E EDWARDS 102 the cemeteries may have had their origins in the burial practices of the late pre-Roman Iron Age. As each of them was associated with a specific villa site it is possible that the Mount Villa also had its own walled burial ground’. Although there is no evidence of a wall surrounding the burials or the cremations, these are the closest burials to the villa yet discovered. Without firmer dating evidence it remains unclear whether the timber structure pre- or post-dates the cemetery and whether it was constructed, used or demolished during the lifetime of the Mount Villa. The stratigraphical relationship between the cemetery and the timber structure, which overlap each other, means that they were not in use at the same time. During the Roman period it was custom to establish cemeteries away from the homes of the local population. This suggests that at the time that the cemetery on-site was in use, the local population and their dwellings would not have been situated within the boundaries of the site, suggesting that the timber structure either pre- or post-dates the cemetery. T he animal bone assemblage changed little from the Iron-Age period suggesting continuity in the types of animal exploitation across the two periods. Environmental samples taken from the Roman period features revealed slightly more fruit remains than the preceding period, including strawberry/cinquefoil, figs and elder. A t the end of the Roman period it appears that the site was abandoned for some time. This may have coincided with a shift of focus towards the south of the current town assembling around land that All Saints Church would later be built upon. N o further activity was recorded on the site until the medieval period. The various medieval features produced a range of different post-dates, but when these are considered across the site it would seem that most have a central date of c.1200-1250. There is very little early medieval pottery; only ten contexts could date to before 1100, and it is much more likely that they are of twelfth-century date. The main period of activity seems to have been in the twelfth to thirteenth centuries (52 contexts), with only ten later medieval features (dated to between c.1300 and 1450) recorded on site. The majority of the pottery assemblage comprises of shell-tempered cooking pots and jars that appear to be from a single source. In form these are typical of the twelfth to earlier thirteenth centuries across the county, but the petrology and chemical composition point to a source in north or west Kent, probably on the northern side of the Medway valley. Production of sandy wares in Maidstone is indicated by a kiln and wasters found nearby in Week Street (no shell-tempered wasters were found). There is, however, no evidence for pottery production from the Fremlin Walk site. T he site at this time was being used sparingly by the local community for their waste disposal. No definite medieval structures were excavated on site, although a series of features in the Phase 4 excavation area may EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 103 have been the remnants of a trench-built structure that had subsequently been demolished and removed. The sequence of pits across the site especially those in the Phase 8 excavation are likely to be related to building that would have fronted Week Street. There is evidence for on-site agricultural processing in the form of small scale cereal threshing which would have produced wheat grains for use in either in the production of bread, animal feed or brewing. The majority of the pits show evidence for the disposal of human waste and food remains, suggested by the large quantities of small seeds that would have been ingested when the fruit was eaten, together with the presence of mineralised and waterlogged fly puparia in some of them. The fact that the seeds of stinging nettle and elder and other nitrogen loving plants were relatively common in most of the waterlogged assemblages in this area at this date, suggests that nitrogen levels were high in and around the features, probably from accumulated organic rubbish. The presence of waterlogged sedge seeds and other wetland plant seeds may be explained by their use as a flooring or roofing material which, when soiled, was later disposed of in rubbish and cesspits. The finds assemblage reflects the fact that by the thirteenth century, Maidstone was a developed medieval market town. By this time the town had begun to expand from its origins surrounding the Archbishops Palace and other ecclesiastical building to the south of the town and were beginning to expand north up Week Street, which would culminate in the establishment of construction of St Faiths Chapel, for the convenience of the new inhabitants (Russell, 1978). The earliest record of the church is in 1268 although it is thought to have been founded before 1200 (Clark and Murfin 1995). T he Fremlin Walk site continued to develop from the medieval period through to the post-medieval period. From the earliest cartographic evidence (1650) it is clear that development in Maidstone was progressing and that the site which was once empty of properties was now starting to become encroached by properties. The area later occupied by the Fremlins Brewery is mainly occupied by open plot areas, whilst to the east of Havock Lane, the land was split between at least sixteen different owners, each with a plot and a property. By this point Week Street was lined with properties, such as seven Inns and although the map does not define the other properties it is likely that some of these properties were also stores, small industrial and commercial properties. Evidence on-site dating to this period varies between pits, both large and small, and linear features. The pits are likely to represent domestic dumping associated with these properties and the surrounding area, whilst the linear features may relate to small scale agriculture or horticultural management. T he occurrence of a post medieval burial on-site was not anticipated. There are no records of there being a post-medieval cemetery on site, so it CATcatHERIN E EDWARDS 104 is likely that this burial, although interred formally in a coffin, is an isolated one. The deposition of a wooden bowl which might have contained salt, also suggests that this burial was isolated and segregated due to an illness or taboo behaviour. Pottery recovered from the grave fill was dated to 1200-1725 whilst the coffins construction and design has been dated as post- 1660. Unfortunately due to the lack of more diagnostic dating evidence it is not possible to date the burial more specifically. In 1650, the area where the burial was located was owned either by Caleb Bankes or Robert Salmon. There is no further cartographic evidence after this date until 1779 which is too general to show the area of the site. The only church located within close proximity to the burial is the Unitarian Chapel in Market Buildings, approximately 120m to the south which was established in 1736 (Humphrey, 1991). It is unlikely that this burial relates to this church as it is not known to have had a graveyard. It is more likely that the burial was an isolated event, not on consecrated ground, an idea perhaps supported by the superstitious nature of the rite surrounding the burial itself. By the eighteenth and nineteenth century the site had developed further with yet more properties and the establishment of alleys and lanes. Brewing had been long established in the area with the Fremlin’s Brewery established on-site by 1790 (Lawson Price 1998). The large pits located in the Phases 9 and 4 areas possibly suggest a much larger scale of domestic dumping rather than the previous individual dumping. The content of the features also reflects the changing nature of society with an increase in available material goods such as pottery fragments and glass; as well as dietary changes such as the growth in the consumption of fruits, herbs and spices. Evidence gathered from several features also implies the existence of specific industries such as a local clay pipe industry and a pharmaceutical dispensary. There is also a noticeable increase in the wealth of the finds recovered from site, such as wine glasses, unusual and elaborate bottles and imitation Chinese porcelain. The pits that contained domestic waste and rubbish likely belonged to the properties fronting St Faiths Street, Market Street, Week Street and Earl Street. I t is not until the 1912 Goad Fire Insurance map, that we are able to see the exact nature of the properties that occupied the site. The western limits were occupied by various buildings of Fremlin’s Brewery, such as cold stores for hops, malt floors and bottle washing rooms. Also noted on this map are the concrete arched cellars under the main brewery complex which were observed during the watching brief on the western section of the site. By now the eastern section of the site was occupied by numerous small properties both commercial and domestic. Some industries that can be identified on the map might be linked to the archaeology observed on site. Several of the linear features observed in the central area watching brief were recorded as wood-lined ditches or trenches which contained EXCAVATIONSxcavations at FREMLIN WALK , maidstonE 105 high levels of organic waste, with inclusions of leather and animal bone, which suggests a small scale industry involved in either tanning or leather work or an industry associated animal husbandry. T he archaeological sequence on this site appears to be a microcosm of the development of Maidstone, featuring localised activity in the Iron Age, Roman development and occupation, the establishment of the medieval market, the expansion of the town, the establishment of the brewery trade and the development of the town as a residential and commercial centre in Kent. acknowledgements T he author and AOCaoc Archaeology Group would like to thank Centros Miller for commissioning and generously funding the evaluation, excavation analysis and publication. Special thanks are due to Wendy Rogers and Simon Mason of Kent County Councils Heritage Conservation unit for their advice and support during the field work and the post excavation analysis. Thanks are also due to Hugh Crawford and the team at Miller Construction for their help and understanding during the fieldwork and to Careys Construction for their on-site assistance. T he author would like to thank the excavation team (Phases 4-9): Magnus Alexander, Sophia Adams, Syann Brooks, Chris Clarke, Raksha Dave, Andy Leonard, Dan Eddisford, Jenny Giddins, Shane Maher, Denise Mulligan, Neil Hawkins, Caz Madden, Melissa Melikian, Gary Evans, Ron Humphrey; the archaeologists from the previous Phases of work 2-3, Catherine Cavannagh, Paul Fitz, Les Capon. Thanks to the specialists for their contributions, Lyn Blackmore, Alan Pipe, Charlotte Thompson, Melissa Melikian Frances Raymond, Jackie Keily, Sylvia Warman, Tony Grey, Liz Barham, Katherine Roberts, and Ian Betts; with special thanks to Fiona Seeley and Penny MacConnoran, for the project management of the specialists. The site was surveyed by Alasdair Harper and Jonathan Moller of AOCaoc Archaeology, and Duncan Lees from Molas. Finds illustration was by Les Capon and the site drawings were completed by Chris Adams, Alasdair Harper and Jonathan Moller of AOCaoc Archaeology. T he author would like to thanks the Centre for Kentish Studies for their help and advice. The site work was managed by Darryl Palmer and the post excavation was managed by Ron Humphrey. bibliography Clark, P. and Murfin, L., 1995, The History of Maidstone. The making of a modern county town, Alan Sutton, Stroud. D rewett, P., Riding, D. and Gardiner M., 1988, The South East to AD 1000, London. Grove, L.R.A., 1967, ‘Researches and discoveries in Kent: Maidstone’, Archaeologia Cantiana, lxxxii, 294-6. CATcatHERIN E EDWARDS 106 Houliston, M., 1999, ‘Excavations at the Mount Roman Villa, Maidstone’, Archaeologia Cantiana, cxix, 71-172. Humphrey, S.C., 1991, Churches and Chapels of Southern England. Jessup, F., 1974, A History of Kent. K elly, D.B., 1992, ‘The Mount Roman Villa, Maidstone’, Archaeologia Cantiana, cx, 177-235. L awson-Price (co.), 1998, ‘Archaeological Desk Based Assessment of Land Bounded by Earl Street, Week Street, St. Faith’s Street and New Spine Road, Maidstone, Kent’. Lyne, M., 2003, ‘Pottery’, in K. Parfitt, ‘A Belgic-early Roman site at Great Mongeham, near Deal’, Archaeologia Cantiana, cxiii, 138-148. O swald, A., 1975, Clay pipes for the Archaeologist, Oxford, BARar 14. Pollard, R.J., 1988, The Roman Pottery of Kent, Monograph Series of the Kent Archaeological Society, 5. Pollard, R.J., 1992, ‘The Mount Villa, Maidstone: the evidence of the Roman pottery for trade and economy’, in D.B. Kelly, 232-234. R ussell, J.M., 1978, The History of Maidstone. S avage, A., 1999, ‘Ceramic’, in M. Houliston, 113-129. S eager-Smith, R., 1995, ‘Pottery’, in V. Birbeck, ‘Excavations on a Romano-British Villa at Churchfields, Snodland, 1992-94’, Archaeologia Cantiana, cxv, 97-109. S pillett, P.J., Stebbing, W.P.D. and Dunning, G.C., 1942, ‘A pottery kiln site at Tyler Hill, near Canterbury’, Archaeologia Cantiana, lv, 57-64. S ymonds, R. P., 1990, ‘The problems of roughcast beakers and related colour-coated wares’, JRPS, 3, 1-17. T hompson, I., 1982, Grog-Tempered ‘Belgic’ Pottery of South-Eastern England, BARar, British Series, 108.