Epsom & Ewell History

& Archaeology Society

St Mary's No.5 Churchyard, Ewell


Chapter 3



3. The finds


All finds were recorded by context, and for extensive contexts, by metre square. This will enable the fall-off rate of activity from Stane Street to be studied (see chapter 4).


3.1 pottery


In the pottery catalogue all pottery is classified by fabric and, where possible, by form. For comparison with earlier work in Ewell (e.g. Orton 1997a) the same fabric codes are used. They are based on the Museum of London Department of Urban Archaeology (DUA) fabric codes (Davies et al 1994, 233), with some simplification where difficulties were experienced in distinguishing between types. Some codes were slightly changed and a few new ones were created. To facilitate broader comparisons, codes from the National Roman Fabric Reference Collection (NRFRC) (Tomber and Dore 1998) are also given; further references can also be found there.


For forms, by contrast, an independent system was set up (0rton 1997a, 104), using broad form types for as much of the pottery as possible.


The amount of pottery found, divided by fabric and context, is shown in Tables 2 (eves) and 3 (sherds). The amounts and proportions by broad periods are given below:


period                quantity (eves)      percent      quantity (sherds)      percent     sherds/eve

prehistoric                   0.04                 0.2                  12                          0.5               300

Roman                      20.54               95.6              2484                        94.0               120

Saxon                          0                      0                      3                           0.1                na

medieval                      0.11                 0.5                 40                          1.5               380

post-medieval              0.78                 3.6               103                          3.9               130

total                           21.47               100                2642                       100                120


Thus, about 95% of the pottery is of the Roman period, and the Roman pottery is considerably less broken than the prehistoric, Saxon or medieval material, and slightly less broken than the post-medieval material. The fabric type descriptions (below) show the proportions of the Roman types as percentages of all Roman fabrics; this analysis cannot be done for other periods.


3.1.1 Fabric types


The names, codes and definitions of the types used are given below in alphabetical order. Where the code differs from that in use at the DUA, the latter is given in [  ]; where there is no corresponding DUA code a [*] is shown. Where appropriate, NRFRC codes are in { }. The descriptions given here are abbreviated, and should be seen as references to standard descriptions rather than in their own right.





BA = Bronze Age: thick but small abraded sherds with flint tempering are thought to be of Bronze Age date.


FLIN = flint-tempered ware: a very coarse fabric characterised by its abundant flint inclusions.


GROG = mainly a Roman fabric (q.v.), some may be prehistoric.


Roman period (i) fine wares


BLEZ = black Lezoux ware [CGBL]: a black colour-coated ware from Central Gaul. The only form present is BEA. {CNG BS} Less than 0.1% of all Roman pottery.


COLC = Colchester colour-coated ware: a fine colour-coated ware with a red fabric and brown surfaces. Produced throughout the Roman period, but most common in the 2nd century (Tyers 1996, 167). The only form present is BEA. {COL CC II} 0.3% of all Roman pottery.


FG = 'fine grey' ware [*]: a distinctive smooth light grey ware with no visible inclusions. Almost all examples seem to be closed forms (BEA or jars); there is one possible bowl. 1.4% of all Roman pottery.


FINE = 'other' fine wares: a rag-bag category of all fine wares that were not classified to other fabrics. They appear to be of closed forms, but in the absence of rims it is difficult to be specific. 0.5% of all Roman pottery.


FOX = 'fine oxidised' [*]: similar to FG but oxidised; possible just a variant of it. No forms can be identified. 0.1% of all Roman pottery.


GCC = general (i.e. unidentified) colour-coated ware [CC]: all colour-coated wares not otherwise identified. Almost all examples appear to be BEA; there is one possible bowl. 1.5% of all Roman pottery.


MICA = mica-dusted wares [includes LOMI = London mica-dusted ware]. Fine grey fabric with yellow-brown margins and surfaces (see Green 1980, 69). All examples appear to be from closed forms. 0.1% of all Roman pottery.


MOSL = moselkeramik: a fine black colour-coated ware from the Trier area. The fabric is hard and shows dark grey and red layers; it is dated to AD 150-275. The only example is of unidentified form. {MOS BS} Less than 0.1% of all Roman pottery.


NVCC = Nene Valley colour-coated ware: similar to Cologne colour-coated ware, with which it can easily be confused. All examples appear to be BEA. {LNV CC} 1.0% of all Roman pottery.


OXRC = Oxfordshire (red) colour-coated ware: a fine reddish-brown fabric, often with a grey core, and red or brownish coat. Dated late 3rd or 4th century. The only identifiable forms are MORT and bowl, but it is possible that some of the unidentified sherds are from closed forms. {OXF RS} 1.2% of all Roman pottery.


RBGL = Roman glazed ware [SEGL]: smooth fabric with applied white slip decoration and green/orange glaze. Late 1st or early 2nd century (Tyers 1996, 178). The only form is a hemispherical bowl (fig 10). {SOB GL} 0.3% of all Roman pottery.


SDG: smooth dark grey ware [*]: a very smooth dark grey to black fabric with a distinctive conchoidal fracture. Probably the fine component of the north Kent ware, sometimes known as Upchurch ware (Monaghan 1987, 173). The only example is of unidentified form. {UPC FR ?} Less than 0.1% of all Roman pottery.


SWCG = Central Gaulish samian ware: samian from the Central Gaulish kilns, probably of 2nd century date.


SWEG = East Gaulish samian ware: samian from the East Gaulish kilns, probably of mid 2nd to mid 3rd century date.


SWSG = South Gaulish samian ware: samian from the South Gaulish kilns, probably 1st century in date.


The only samian forms identified were single examples of Drag. 33 and Drag. 37. There were also body sherds of MORT, and sherds of bowl, cup and dish forms. Total of all samian ware is about 2.2% of all Roman pottery; the total of fine wares is about 8.2%.


Roman period (ii) amphorae


DR20 = Dressel 20 amphorae: globular amphorae from southern Spain, thought to have originally contained olive oil. They are very common and the form is well known. {BAT AM 1} 0.6% of all Roman pottery.


AMPH = all other amphora fabrics. 0.2% of all Roman pottery.


Roman period (iii) mortaria


OXWW = Oxfordshire white ware [OXMO]: an off-white sandy fabric with distinctive trituration grits of red, grey and clear rounded quartzite. {OXF WH} 0.5% of all Roman pottery.

Other mortaria are catalogued under their respective fabrics, e.g. VRW, SWEG.


Roman period (iv) coarse wares (reduced)

AH = Alice Holt ware [AHSU + AHFA]: thoroughly described by Lyne and Jefferies (1979). The dating was revised by Millett (1979) on the basis of seriation. Produced nearly 50 km west-south-west of Ewell from the 1st to 4th centuries. {ALH RE} 48.2% of all Roman pottery.

The main forms present are:

(a) dishes and bowls: FB (16%); PB/PD (7%), RRB/RRD (4%); REB (3%); ATB (1%); other (3%).

(b) jars: ERJ (17%); BBJ (10%); FRJ (7%); BRJ (6%); LSJ (6%); HRJ (4%); other (9%).

(c) other: BEA (3%); FL (2%); LID (1%); MUG (1%).


BB2 = black burnished ware type 2: defined by Farrar (1973) and further described by Green (1980). According to Williams (1977) it was produced at sites in Kent, as well as at Colchester. The ware is generally dated to AD 120 – early 3rd century. The only identified form is the RRB. 1.3% of all Roman pottery.


GROG = grog-tempered wares: these are generally early in the Roman period (late 1st century, e.g. HWB (Brown and Sheldon 1974, 224)) or late (4th century). Large closed forms predominate; the only identifiable rims are ERJ. There is also one PD rim. 1.3% of all Roman pottery.


HIGH = Highgate ware type C [HWC]: the standard sandy Highgate ware, it seems to be at the limit of its distribution here. The only form present appears to be BEA, some with white slip and/or barbotine decoration. {HGW RE C} 0.4% of all Roman pottery.


SAND = 'other' grey sandy wares: this is a rag-bag heading not easily categorised. They date from all parts of the Roman period, but this reflects a number of probably short-lived sources rather than a single continuous tradition. 22.2% of all Roman pottery.

The main forms present are:

(a) dishes and bowls: PB/PD (7%), FB (6%); RRB (5%); other (5%).

(b) jars: ECJ/ERJ (25%); BBJ (6%); BRJ (3%); LSJ (1%); FRJ (1%); other (26%).

(c) other: BEA (7%); FL (5%); LID (1%); unidentified (4%).


SHEL = coarse shell-tempered wares: these appear to be mainly early Roman in date, and may include North Kent Shelly Ware [NKSH]. They are tempered with sand as well as coarse shell. The only identifiable forms are BRJ and PB. 0.5% of all Roman pottery.


VCS = very coarse-sandy ware [*]: miscellaneous sherds of large vessels, characterised only by the coarse nature of their inclusions. Large closed forms predominate; the only identifiable rim is BRJ. 0.6% of all Roman pottery.


Roman period (v) coarse wares (oxidised)


HOO = Hoo-type ware: a red ware usually with a white slip, though to be produced in north Kent in the 1st and 2nd centuries (Monaghan 1987, 173). All examples appear to be from FL. 0.2% of all Roman pottery.


OXID = 'other' oxidised wares: the oxidised counterpart to the SAND category, and a similar rag-bag. 11.1% of all Roman pottery.

(a) dishes and bowls: FB (2%); RRB (2%); other (7%).

(b) jars: ERJ (19%); HRJ (11%); BBJ (7%); FRJ (5%); BRJ (3%); LSJ (1%); other (9%).

(c) other: BEA (15%); MORT (1%); FL (1%); LID (1%); unidentified (15%).


TILF = Tilford ware, aka Portchester 'D' ware [PORD]: a red or yellowish sandy fabric made in the same general area as AH. It was first identified at Portchester (Fulford 1975), but has since been found elsewhere, e.g. London. It is diagnostically late, e.g. AD 350+ in London. All are from closed forms; the only identifiable rims are ERJ and HRJ. {OVW WH} 1.8% of all Roman pottery.


VRR = Verulamium region red ware [VRR + VCWS]: a red/grey sandy fabric with white slip, usually dated to the 1st and 2nd centuries by analogy to VRW, and once known as Staines- London ware (Green 1980, 60) (SLWS). All examples appear to be from FL. 0.1% of all Roman pottery.


VRW = Verulamium region white ware: an off-white or buff sandy fabric with a distinctive fracture, dated to the 1st or 2nd century. Most examples appear to be from FL; there is also one ERJ and one MORT. {VER WH} 1.8% of all Roman pottery.


(vi) Saxon fabrics


CHAF = Saxon chaff-tempered ware: a dark hand-made fabric, probably middle Saxon in date.


There is also one sherd of SAND which is of Saxon date and is stamped (fig. 11), and one similar but unstamped sherd.


(vii) medieval fabrics


CHEA = Cheam white ware: forms cannot be identified.


EARL = Earlswood ware: only small body sherds of decorated jugs are present.


KING = Kingston ware: a late 13th–14th century 'white' ware (Pearce and Vince 1988). Sherds from jugs, bowls and a possible cooking-pot are present.


LOND = London ware: a wheel-thrown sandy ware, mainly of the 13th century but also produced earlier and later (Pearce et al 1985). Sherds are all from jugs.


MCS = medieval coarse sandy ware: one sherd from a cooking-pot.


(viii) post-medieval fabrics


BORD = border ware (Pearce 1992): divided into BORDG (green-glazed) and BORDY (yellow-glazed). Forms present are bowls and cooking-pots.


CHPO = Chinese porcelain: imported, mainly in the 18th century. Forms cannot be identified.


FREC = Frechen stoneware bottles: imported from Germany in the late 16th or 17th century.


GUYS = Guys ware: red or brown earthenwares with extensive zones of white slip, made in the London area in the late 15th and 16th centuries (Orton 1988, 297). Only bowl sherds are present.


LONS = London stoneware: one sherd from a mug.


NOTS = Nottingham stoneware: one unidentified base sherd.


PMR = post-medieval red ware [PMCR + PMFR + RBOR]: glazed redwares of various sources. LSJ appear to predominate; there area also bowls and a mug.


RAER = Raeren stoneware: one sherd, probably from a mug.


STBU = Staffordshire butter-pot ware: self-explanatory, late 17th or 18th century.


STON = unidentified stoneware.


TGW = tin-glazed ware: probably produced in London, 17th to mid-18th century. The only identified sherds are from wall tiles.


TUDB = Tudor brown ware: one rim sherd, probably from a cooking-pot.


TUDG = Tudor green ware: fine white ware with glossy green glaze, 15th to 16th century. One unidentified body sherd.


3.1.2 Forms


A simple division into broad groups of forms, based on the ratio of height to diameter (complete profiles) and the apparent open or closed nature of the pottery, was adopted as follows:

B = bowl; BEA = beaker; D = dish; J = jar; LID = lid.

In addition, specifically functional forms were recognised: AMPH = amphora; CUP = cup; FL = flagon; MORT = mortarium.

It was found that B, D and J had to be qualified by extra codes, viz.:

FD = flanged dish,                                      BBJ = jars in the 'BB1' style,

PD = plain-rimmed dish,                             BRJ = bead-rimmed jars,

RRD = rolled rim dish,                                ECJ = everted and cordoned jars,

ATB = 'atrebatic' bowl,                                ERJ = everted-rimmed jars,

DB = deep bowl,                                         FRJ = flat-rimmed jars,

PB = plain-rimmed bowl,                             HRJ = hook-rimmed jars,

REB = reeded-rim bowl                               LSJ = large storage jars,

UCB = unclassified (i.e. 'other') bowl.         UCJ = unclassified (i.e. 'other') jar.


Well-known form typologies have also been used where appropriate (e.g. Dragendorff, Ludowici, Curle, Déchelette, Walters).


3.1.3 Decoration


Codes for decoration are based on a combination of

(i) method of application (A = applied; B = burnished; C = combed; G = grooved) and

(ii) pattern (A = arcs or arcading; B = band; D = diagonal line(s); H = herringbone; L = lattice; P = 'other' pattern; S = 'scribble'; SL = straight line(s); V = vertical line(s); W = wavy line(s); Z = zone).

Examples are: BB = burnished band; CL = combed lattice; GSL = grooved straight line(s).

Not every possible combination actually occurred.

Other codes, which do not fit this system, are: COR = cordon; MCR = multiple cordons; ROU = rouletting; RUS = rustication, SL = slashed; STC = stabbed with comb; WS = white slip.

Codes can be combined with a /, e.g. BZ/ROU = burnished zone and rouletting. When all else failed a * was entered and the decoration was described in free text under 'comments'.


3.2 Other ceramic artefacts


Excluding building materials (see 3.7), there were very few other ceramic artefacts. They comprise:


Spindle whorls

Fragment (20%) of disc in grey sandy fabric, probably AH. One face flat, other slightly domed. Central hole made before firing. Diameter 40 mm. Max thickness 7mm. Sf. no. 41. From <103>, 217/100.


Fragment (40%) of cut-down base of jar or beaker in grey sandy fabric, probably AH. Central hole drilled. Diameter 40–45 mm. Max thickness 8 mm. Sf. no. 51. From <103>, 213/100.



Irregular fragment of badly wedged fired clay, with impression of stick c. 15 mm in diameter. No other finished surfaces. Max dimension 45mm. Similar to fragments from the King William IV excavation (Orton 1997, 104). Sf. no. 75. From <103>, 211/100.


Fragment of badly wedged fired clay with apparently two finished surfaces at about 60to each other; one appears to have been laid on a flat surface while the other may be hand-finished. Max dimension c. 60 mm.


Clay tobacco pipe

Stem fragments of clay tobacco pipes were found in <101>, <102> and <103>. No bowls were found, but an 18th-century foot marked IG was found in <101>, 222/100.


3.3 Glass


There were sixteen fragments of glass, from <101> (7), <102> (4), <103> (4) and <104> (1). Five are from wine bottles of 18th century or later date, and four are bluish fragments, probably of window glass. There is a small piece of blue cullet from <104> (sf. no. 28). The rest are fragments of vessel glass of unknown date (including sf. nos 5, 6, 64).


3.4 Metal


3.4.1 Coins

Kris Lockyear and Adrian Popescu


Twenty-eight objects where submitted for examination. 26 were Roman coins; one was a coin of Charles I and the remaining object a lead weight. Of the 26 Roman coins, 25 were identifiable to the level of Richard Reece’s 21 periods (see, e.g., Reece 1987). The catalogue is presented below in date order.


Context      Easting    SF no.      description                                                                                     date

101             263.3        26             dupondius or as of Vespasian or Titus.                                      AD 69–81

                                                                                      otherwise illegible

 99             200           23             as, RIC 2, Hadrian 617, minted in Rome                                    AD 121–122

103             218.8          8             radiate, Claudius II Gothicus. Cunetio 2019,                              AD 268–

                                                    Issue 1

103             217.8        10             radiate, Victorinus or Tetricus I, rev. Cunetio Fides I                 AD 269–271/4

 99             210           14             radiate copy, rev. Cunetio Victoria 3A                                prob AD 271/4+

103             226.8        50             radiate copy of Tetricus I, rev. Cunetio Pax 1b                  prob AD 271/4+

 99             208           34             radiate copy of deified Claudius II Gothicus                      prob AD 271/4+

                                                    rev. Cunetio Consecratio Eagle 2

 99             244           36             AE radiate copy, otherwise illegible                                   prob AD 271/4+

 99             210           15             radiate copy, otherwise illegible                                         prob AD 271/4+

 99             225           30             AE RIC 6 London 264, Constantine I                                         AD 312–313

103             223.5        24             AE Constantinopolis, copy of RIC 7 Lyons 241                         AD 300–331+

 99             218           20             AE Constantinopolis, copy of RIC 7 Lyons 241                         AD 300–331+

 99             226           49             AE Constantinopolis, copy of RIC 7 Lyons 256                         AD 332+

 99             242           44             Gloria Exercitus two standards, RIC7 Arles 365                        AD 335–341

103             220.8          4             copy of RIC 8 Arles 182, Decentius; mint mark                         AD 351+


 99             208           13             AE4 copy of Fel Temp Reparatio Falling Horseman 3,             AD 353/4+

                                                     Constantius II

 99             218           18             AE4 copy of Fel Temp Reparatio Falling Horseman 3              AD 353/4+

                                                     or 4,otherwise illegible

 99             216          19              AE4 copy of Fel Temp Reparatio Falling Horseman 3,             AD 353/4+

                                                    Constantius II. Half coin, perhaps broken

 99             209           33              AE3 copy of Fel Temp Reparatio Falling Horseman 3,            AD 353/4+

                                                    Constantius II with pearl diadem

 99             243           35              AE4 copy of Fel Temp Reparatio Falling Horseman 3,            AD 353/4+

                                                    Constantius II

 99             244           37             AE4 copy of Fel Temp Reparatio Falling Horseman 3,             AD 353/4+

                                                    Constantius II. NB cast copy as shown by surviving flange

 99             206          43              AE4 copy of Fel Temp Reparatio Falling Horseman 3,             AD 353/4+

                                                    Constantius II

103             209.7       11              Gloria Romanorum, as RIC 9 (mint illegible), Valens                AD 364–378

 99             213          17              Gloria Romanorum, RIC 9 Lyons 20a, Valentinian I                  AD 367–375

 99             203          31              Gloria Novi Saecvli, RIC 9 Arles 15, Gratian                             AD 367–375

 99             203          47              Third or fourth century, accretion prevents identification

101            262           27              Charles I Rose farthing                                                              AD 1636–1644



RIC = Roman Imperial Coinage volumes 1–10, London 1923 to 1994.

Cunetio =Besly and Bland 1983.



There is no obvious correlation between date and easting (i.e. distance from Stane Street), though it is curious to note that the earliest coin <26> is furthest east, and the latest, <11>, <17> and <31> are in the western end of the trench, closest to Stane Street. In view of the small numbers of coins, little weight should be placed on these observations.


This assemblage can be compared to the broader assemblage from Ewell as a whole published by Abdy and Bierton (1997, 140). The following table compares the distributions of the two assemblages across Reece’s ‘issue periods’ (Reece 1987).


Reece period          date                AB count            AB %           ECY count         ECY %

 1                         to AD 41                5                      3.4                    -                       -

 2                             41–54                2                      1.4                    -                       -

 3                             54–68                1                      0.7                    -                       -

 4                             69–96              17                    11.7                    1                     4.0

 5                             96–117              5                      3.4                     -                       -

 6                           117–138              5                      3.4                    1                     4.0

 7                           138–161              4                      2.8                     -                       -

 8                           161–180              2                      1.4                     -                       -

 9                           180–192               -                         -                     -                        -

10                           192–222              3                      2.1                    -                        -

11                           222–238              1                      0.7                    -                        -

12                           238–259              1                      0.7                    -                        -

13                           259–275            32                    22.1                    2                     8.0

14                           275–294            10                      6.9                    5                   20.0

15                           294–317              4                      2.8                    1                     4.0

16                           317–330            18                    12.4                    -                       -

17                           330–348            10                      6.9                    4                   16.0

18                           348–364              5                      3.4                    8                   32.0

19                           364–378            10                      6.9                    3                   12.0

20                           378–388              3                      2.1                    -                       -

21                           388–402              7                      4.8                    -                       -

Total                                               145                                             25


There is a clear mismatch between the two assemblages, with only 16% of the ECY assemblage but over 50% of the Ewell assemblage dating to before AD 275. However, it would be unwise to interpret these figures uncritically, because of the different circumstances of retrieval. Most of the ECY assemblage was found by metal-detecting (which appears not to have been used in previous work) and comprises small nummi coins which could easily have been missed in earlier work.


Although with only 25 coins it would obviously be foolhardy to develop complex interpretations of this assemblage, it is none-the-less necessary to place this group into a wider setting. A wide variety of methods have been developed over the last 40 years by Casey, Reece, myself and others, all of which take as a starting point the conversion of the full list to a series of ‘issue periods’ (Reece 1987). In this case the periods used by Reece have been adopted due to the availability of comparative data, especially from southern England.


Fig. 12 is a simple bar chart of this data converted into permilles (‰). The graph clearly shows a typical Romano-British loss pattern with low levels of coin loss up to period 12, high loss of low value radiate coinage in periods 13 –14, lower loss in periods 15–16, and high loss in the mid-4th century, particularly in period 18, the Fel temp reparatio coinage and its copies.


Reece (1974, 1993) further grouped his periods into four phases: A (1–12), B (13-14), C (15–16) and D (17–21). He found that by plotting phases B against D patterning by site type could be discerned. For example, towns tended to have equal amounts of B to D, whereas rural sites and especially temples tended to have far higher quantities of phase D coinage. Fig. 13 plots the B:D figures for the 140 sites published by Reece with the addition of Ewell. Although there is no clear division between urban and rural sites on this graph, there is a trend for civitas capitals to be to the lower right hand side of the graph and rural sites to be to the upper left. The data for Ewell is in a relatively central part of the graph along with many civitas capitals. The site list appears, therefore, to fit nicely into the main run of Romano-British coin lists and is particularly similar to urban sites.


Recently, more detailed work on site lists, particularly from the Portable Antiquity Scheme data and a larger selection excavated assemblages by Walton (2011) has shown that more detailed regional and period patterns exist but as yet there is insufficient properly published data from London to allow this type of detailed analysis.


3.5 Bone artefacts


There were no bone artefacts as such, but a few fragments suggestive of bone-working:

small fragment, broken at both ends, trapezoidal cross-section, polished surfaces. Length 15 mm, width 5 mm, possibly tooth from a comb. Sf. no. 3, from <101>, 209/100.

Fragment, broken at both ends, hexagonal cross-section, polished surfaces. Length 46 mm, max diameter 4 mm, possibly rough-out for a pin. Sf. no. 39a, from <103>, 212/100.

Fragment, broken at both ends, triangular cross-section, notched. Length 40 mm, max. width 5mm. Sf. no. 39b, from <103>, 212/100.

Long bone, sawn at both ends and slightly flattened on two faces. Length 83 mm, diameter of shaft 10 mm. Sf. no. 67, from <103>, 209/100.

Flat fragment with one square end and marks of sawing (?) on one face. 12 mm square. Sf. no. 68, from <101>, 220/100.


3.6 Stone


3.6.1 Building stone


The only potential building stone that was at all common was chalk. Thirty fragments were retained as showing evidence of working, although as chalk is not part of the natural geology here, all examples, whether worked or not, must have been brought onto the site. The majority (20 fragments) was found in the topsoil layers <103> and <104>. Two were found in <105>, five in <107> and one in <108>. The surface <117> was composed of squared chalk blocks; a sample of two was retained.


One shaped block of unidentified sandstone was found in <129>.


3.6.2 Abrasive stone


The most notable example was a large fragment of a circular quern stone from <126>, where it appeared to serve a secondary function as post-packing (see fig. 4). It has an estimated diameter of 400 mm; its thickness varied from 40 mm at the edge to 25 mm near the centre. Sf. no. 46.


Forty-two very small fragments of Mayen lava quernstone were found in the topsoil <103> at 225/.


One fragment of a laminated sandstone slab that had been used as a sharpening stone (sf no. 54) was found in the topsoil <103> at 227/.


3.6.3 Flintwork


Seventy-one pieces of flintwork were recorded. Six appeared to be debris from the preparation of flints for use as a building material (from <101> (3), <103> (2) and <112> (1)), and two may be gun-flints (from <101> and <103>).


Leaf-shaped arrowhead from <112>, sf no. 55.


Three possible cores from <103>, 212/100 (2) and 218/100 (1).


Of the remaining 59 flakes and blades, sixteen showed signs of working: six were notched and the rest had one or both edges blunted. A high proportion of both worked and unworked examples had been snapped, perhaps indicating post-depositional damage.


3.7 Other building materials


3.7.1 Roman brick and tile


A total of 180 fragments of brick or tile were identified as being probably of Roman date. The majority (136) was from the topsoil and associated spreads (<100> to <104>). The rest were from phased contexts: one from <117> (phase 1); one from <113>, four from <110> and four from <130> (phase 2); five from<118>, one from <125>, five from <106>, two from <107>, five from <108> and four from <127> (phase 3).


Of these, 73 were identified as fragments of tegula and a further seven as imbrex (roof tile). The majority (88) were too thick to be from tegulae, and were simply identified as brick/tile. There were seven possible tesserae (all from <101> to <103>) and five fragments of box tile, of which Ernest Black writes:

“Sf no. 62 <127>: part of unkeyed side (T16) and one face (T16) of box tile. The face carries substantial traces of mortar and is keyed with die 4.

Sf no. 66 <103>, 220/100: two joining fragments (Tc. 13) from the face of a box tile keyed with die 4.

Sf no. 71 <103>, 223/100: part of face of box tile (Tc. 15), apparently burnt, with three intersecting impressions of straight combing (i.e. not relief-patterned).

Sf no. 72 <101>, 242/100: fragment from the face of a box tile (Tc. 15) keyed with die 1.”


All these fragments are small (none has a measurement exceeding 100 mm). The two dies represented here belong to a series dating to the middle decades of the second century (see Betts et al. 1997). Die 4 is known from the mansio baths at Chelmsford in Essex and from postulated baths at the roadside settlement at Alfoldean in Sussex which, like Ewell, lies on Stane Street. Die 1 was previously found by Lowther at Purberry Shot, Ewell while a virtually complete tile keyed with die 4 and other fragments were reported from Ewell Churchyard Extension in 1960. Dies 5, 14 and 66 are also known from Purberry Shot and die 66 from Tayles Hill, Ewell. All the dies represented at Ewell were found at the villa on Ashtead Common, just over 5 km to the south-west of Ewell.”


3.7.2 Daub and mortar


A total of 34 fragments of daub were found in a relatively small area between 219/ and 226/. Nine of them were from the topsoil <103> and rest from the surface <105> (six fragments) and the dumped deposits <108> (eleven fragments), <127> (four fragments) and <129> (two fragments). In addition, two were found in the cut feature <130>. Traces of wattle impressions were found on some examples. Six fragments of mortar were found: five in <103> and one in <113>. The identifications of one sample of daub and one of mortar were confirmed by soil micromorphology (McPhail 2001).


3.7.3 Plaster


Three fragments of Roman wall plaster were found, all in the topsoil <103>. One (sf no. 40, found at 212/) showed evidence for having been re-coated; the others (sf no. 32, from 212/ and sf no. 60, from 208/) were plain.


3.7.4 Opus signinum


One fragment of opus signinum was found, in <101>.


3.7.5 Medieval and post-medieval roof tile


Some 100 fragments of medieval or later roof tile were found, all from the topsoil (<101> to <103> and <111>). All were small and none had more than one corner present.


3.8 Animal bone


Some 412 fragments of animal bone, weighing just over 7 kg, were found. The majority (361 fragments = 88%; 5.4 kg = 77%) is from the topsoil layers <101> to <104>. Most of them are very small (average weight 15 g) and many are worn. Because of this, and because of the difficulty of assigning them to a definite period, they were not catalogued by species or body part. By contrast, a group of only six bones (including one skull) from the spread <129> weighed over 1 kg (15% of the total weight), and a group of five bones from the cut feature <130> weighed 180 g (2.5% of the total). There were also small groups from <110>, <112>, <113>, <118>, <125> and <126>.