archaeological report - killeisk, co. tipperary (ireland)

Issue 11 [ISSN 2009-2237]

Archaeological Excavation Report

E3587 - Killeisk, Co. Tipperary

Early Bronze Age pits, medieval enclosure and associated field enclosures

Eachtra Journal

December 2011


Co. Tipperary

Laois County Council and National Roads Authority

Simon O'Faolain

Killeisk

EACHTRAArchaeological Projects

E No:

:

Project:

Client:

Date:

E3587

N7 Castletown to Nenagh (Contract 1)

Simon O'FaolainExcavation Director

Written by:

Early Bronze Age pits, medieval enclosure and associated field enclosures

CORKThe Forge, Innishannon, Co. Cork

tel: 021 4701616 | web: www.eachtra.ie | email: [email protected]

GALWAY Unit 10, Kilkerrin Park, Liosbain Industrial Estate, Galway

tel: 091 763673 | web: www.eachtra.ie | email: [email protected]


KilleiskCo. Tipperary

Excavation Director

Simon O'Faolain

Written By

Simon O'Faolain

EACHTRAArchaeological Projects

© Eachtra Archaeological Projects 2011 The Forge, Innishannon, Co Cork

Set in 12pt Garamond

Printed in Ireland

i

Table of ContentsSummary�� 1

Acknowledgements�� 2

1 Scopeoftheproject�� 3

2 Routelocation�� 3

3 Receivingenvironment�� 5

4 Archaeologicalandhistoricalbackground�� 6

5 SiteLocationandTopography�� 9

6 Excavationmethodology�� 9

7 Excavationresults�� 13Description�of�Prehistoric�Pit�Group�1��13

Description�of�Prehistoric�Pit�Group�2��19

Description�of�Enclosure�and�Associated�Field�system��23

Description�of�kiln�and�associated�features��41

8 Discussion�� 52

9 References��56

Appendix1 StratigraphicIndex�� 59

Appendix2 StratigraphicMatrix��60

Appendix3 Groupsandsub-groups�� 61

Appendix4 LithicArtefactReport��83

Appendix5 QuernstoneReport��85

Appendix6 PlantRemainsReport��88

Appendix7 AnimalBoneReport��99

Appendix8 GeophysicalSurvey��105

ii

List of FiguresFigure1: Portion of map of Ireland showing the route of the N7 Castletown to Nenagh

(DerrinsallaghtoBallintotty)RoadScheme(Contract1)�� 4

Figure2: DiscoveryseriesOrdnancesurvetmapshowingtherouteoftheN7CastletowntoNenagh(DerrinsallaghtoBallintotty)RoadScheme(Contract1)andthelocationofallexcavationsites�� 8

Figure3: Portionof the IsteditionOrdnanceSurveyMapOF47showingthe locationofKilleiks��10

Figure4: LocationandextentofKilleiskE3587ontheN7CastletowntoNenagh�� 12

Figure5: Post-excavationplanofKilleiskE3587�� 14

Figure6: Post-excavationplanofprehistoricpitgroup1atKilleisk��16

Figure7: SectionsofpitC245andassociatedstake-holesatKilleisk�� 17

Figure8: Post-excavationplanofprehistoricpitgroup2atKilleisk��20

Figure9: SectionsofpitC464andC466atKilleisk��21

Figure10: Post-excavationplanofenclosureandassociatedfieldsystemsatKilleisk��26

Figure11: SectionsofenclosureditchL1C�4andwellC�135atKilleisk��28

Figure12: SectionsofL9C522,L8C92andL4C11��32

Figure13: SectionsofL5C13andL8C24��34

Figure14: SectionsofL1C4,L5C13andL10C111��35

Figure15: Post-excavationplanofkilnandassociatedfeaturesatKilleisk��37

Figure16: SectionsofkilnC358andpitC364atKilleisk��43

Figure17: SectionsofL7C100��45

Figure18: PlanofgeophysicalshowingextentofenclosureatKilleisk��49

Figure19: PrehistoricsitesonandintheenvironsofN7CastletowntoNenagh��51

Figure20: MedievalsitesonandintheenvironsofN7CastletowntoNenagh��53

iii

List of PlatesPlate1: AerialviewofKilleisk�� 11

Plate2: ViewofpitC245andassociatedstake-holesatKilleisk�� 15

Plate3: ViewofpitC281atKilleisk�� 15

Plate4: ViewofcharcoalenrichedfillsofpitsinBronzeAgePitsGroup2atKilleisk�� 19

Plate5: ViewofpitC466andassociatedstake-holesatKilleisk��22

Plate6: ViewofenclosureL1atKilleiskfromnorth-west��24

Plate7: ViewofmaincutofenclosureditchC4bottomedoutandinsectionrectutC7atKilleisk��24

Plate8: View of stone layer C88 in enclosure ditch C4 at Killeisk� Note quernstoneE3587:88:1abovewhiteboard��27

Plate9: View of main cut of enclosure ditch C4 and the narrower recut C7 inscisedthroughitsbase��27

Plate10: ViewofL8C24,C58andL1C4fromlefttorightatKilleisk��29

Plate11: ViewofL2C39andC90fromNEandLIC4inforegroundatKilleisk��30

Plate12: ViewofL4C11andL5C13fromNWatKilleisk��31

Plate13: QuernstoneE3587:130:1��33

Plate14: QuernstoneE3587:125:1��33

Plate15: ViewofkilnandassociatedfeaturesfromNWatKilleisk��42

Plate16: ViewofkilnC538atKilleisk��42

Plate17: ViewofcharcoalincupolaofkilnC538atKilleisk��44

Plate18: HonestoneE3587:1:1��47

List of TablesTable1 LinearNumbersandcorrespondingcutnumbersusedinthetext��25

Table2 Radiocarbondates��50

KilleisK-e3587

1

http://eachtra.ie/index.php/journal/e3587-killeisk-co-tipperary/

SummaryThe excavation of the site at Killeisk comprised a mixture of Early Bronze Age activity and high and late medieval activity. Two separate groups of pits dated to the Early Bronze Age. A large elliptical enclosure and associated linear features, which formed field enclo-sures and droveways, dated to the high and late medieval period. A kiln and associated pit and enclosure also dated to the late medieval period. The kiln was located to the south-west of the enclosure. Two fragments of decorated rotary quern stones were recovered from the enclosure and associated linear features. A small assemblage of animal bone was recovered from the ditch of the enclosure.

Road project name N7 Castletown to Nenagh Site name KilleiskE no. E3587Site director Simon Ó FaoláinTownland KilleiskParish BallymackeyCounty TipperaryBarony Upper OrmondOS Map Sheet No. TN 21National Grid Reference 194519 / 179498Elevation 101 m O.D.

issUe 11: eachtra JoUrnal - issn 2009-2237 archaeological excavation report

2

AcknowledgementsThe project was commissioned by Laois County Council and was funded by the Na-tional Roads Authority under the National Development Plan (2000-2006). The project archaeologist was Niall Roycroft. Kildare County Council supervised the archaeological contract with RE staff of Pat Dowling and Colum Fagan. Kildare County Council Sen-ior Executive Engineer was Joseph Kelly and Kildare County Council Senior Engineer was John Coppinger. The senior archaeologist was John Tierney and the post-excavation manager was Jacinta Kiely. Illustrations and GIS are by Maurizio Toscano, photographs by John Sunderland and Eagle Photography and aerial photography by StudioLab. Spe-cialist analysis was carried out by Anne Carey, Mary Dillon, Penny Johnston, Margaret McCarthy, Farina Sternke and the 14 Chrono Centre at Queen’s University Belfast.

KilleisK-e3587

3


1 Scope of the project Eachtra Archaeological Projects were commissioned by Laois County Council and the National Roads Authority to undertake archaeological works along 17.1 km (Contact 1) of the 35km N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) national road scheme (EIS approved in November 2005). The scheme runs from the eastern junction of the present N7 Nenagh Bypass, North Tipperary a tie in to the M7/M8 Portlaoise-Castletown scheme to the south of Borris-in-Ossory in County Laois. The scheme is ap-proximately 191 hectares. Contract 1 comprises the western half of the scheme and runs from Clashnevin to Castleroan passing along the Tipperary North and Offaly county border regions. The Ministers Direction Number is A38.

It was funded by the Irish Government under the National Development Plan 2000-2006. The total archaeological cost was administered by the National Roads Authority through Laois County Council as part of the Authority’s commitment to protecting our cultural heritage. The purpose of the archaeological services project was to conduct ar-chaeological site investigations within the lands made available for the scheme and to assess the nature and extent of any new potential archaeological sites uncovered.

Phase 1 of the project (archaeological testing of the route) was carried out in 2007 under licence E3371, E3372 and E3375-8 issued by Department of the Environment Her-itage and Local Government (DoEHLG) in consultation with the National Museum of Ireland. The principal aim of this phase of the project was to test for any previously unknown sites by a programme of centreline and offset testing and to test sites of archaeo-logical potential identified in the EIS.

Phase 2 of the project (resolution) involved the resolution of all archaeological sites identified within the proposed road corridor prior to commencement of the construction of the road. This phase of the project was carried out from June 2007 to February 2008 and excavations were conducted under the management of a Senior Archaeologist. A total of 27 sites were excavated during this phase of works under separate licences issued by DoEHLG.

A post-excavation assessment and strategy document was prepared in Phase 3 of the project to present a management strategy for dealing with post-excavation work aris-ing from archaeological works along the route of the new N7 Castletown to Nenagh. It included a proposal for post-excavation and archiving work and a budget for the works.

2 Route locationThe route of the N7 Castletown to Nenagh road is located in Counties North Tipperary and Offaly (OF) (Figure 1). The project (Contract 1) involves the construction of c. 17.5 km of the N7 from Clashnevin east of Nenagh to Castleroan south-east of Dunkerrin. It passes through the townlands of Clashnevin, Derrybane, Newtown, Lissanisky, Killeisk, Garavally, Derrycarney, Garrynafanna, Gortnadrumman, Kilgorteen, Falleen, Knock-ane, Clash, Park, Rosdremid (OF), Clynoe (OF), Cullenwaine, Moneygall, Greenhills,

De

rg (

Lo

ug

h)

1825

50

1825

50

1989

00

1989

00

2152

50

2152

50

172600

172600

182950

182950

193300

193300

! (N

enag

h

±0

510

Kilo

met

res


4

Figu

re1

:Po

rtio

nof

map

ofI

rela

nds

how

ing

the

rout

eof

the

N7

Cast

leto

wn

toN

enag

h(D

errin

salla

ghto

Bal

linto

tty)

Roa

dSc

hem

e(C

ontr

act1

)�

KilleisK-e3587

5


Drumbaun, Busherstown (OF), Drumroe (OF), Moatquarter, Loughan (OF) and Cas-tleroan (OF). The townlands are located in the parishes of Ballymackey, Cullenwaine, Castletownely, Rathnaveoge, Finglas and Dunkerrin and the baronies of Upper Ormond, Ikerrin and Clonisk,

The route begins at the eastern end of the Nenagh bypass at Clashnevin c. 5 km east of Nenagh and continues eastward on the northern side of the existing N7 in Co. Tip-perary. It crosses a number of third class roads to the north of Toomyvara and 0.7 km east of Clash crossroads crosses the Ollatrim River. It extends into County Offaly directly east of Park. From here it crosses the R490 0.6 km north of Moneygall. It extends back in County Tipperary and through the demesne of Greenhills before crossing the existing N7 at the junction of Greenhills and Drumbaun townlands. It crosses back into County Offaly and climbs east into Busherstown and Drumroe. It crosses the Keeloge Stream into Moatquarter in County Tipperary and extends northeast back into County Offaly through the townlands of Loughan and Castleroan 1.4 km southwest of Dunkerrin.

3 Receiving environmentNorth Tipperary is bounded on the west by the River Shannon and Lough Derg with the Silvermines, to the south, and small hills extending towards Devilsbit and Borrisnoe Mountains to the east. The mountains are composed largely of Silurian strata and Old Red Sandstone. Copper, silver and lead deposits have been mined in the Silvermines. The geology of the lowlands consists of Carboniferous limestone covered by glacial drift in addition to tracts of raised bog.

The western portion of the study area is drained by the Ollatrim River which flows westwards into the River Ballintotty which in turns drains into the River Nenagh. The eastern portion is drained by the Keeloge Stream and other small water sources. These rise in the foothills of the Silvermine Mountains and flow north. The Keeloge drains into the Little Brosna River c. 1 km south of Shinrone, Co Offaly. The Brosna turns north and drains into the Shannon south of Banagher.

The largest population centre in the area is Nenagh. The smaller population centres, are Toomyvara, Moneygall and Dunkerrin.

The soils on the route are characterised by 80% grey brown podzolics, 10% gleys, 5% brown earths and 5% basin peat. They are derived from glacial till of predominantly Car-boniferous limestone composition. These soils occur in Tipperary and Offaly and have a wide use range being suitable for both tillage and pasture (Gardiner and Radford 1980, 97-99). Land use along the route was a mix of grassland devoted to intensive dairying and cattle-rearing and tillage.


6

4 Archaeological and historical backgroundArchaeological sites of numerous periods were discovered along the route of the new road (Figure 2). The periods are referred to as follows: Mesolithic (c. 8000 to 4000 BC), Neo-lithic (c. 4000 to 2000 BC), Bronze Age (c. 2000 to 600 BC), and Iron Age (c. 500 BC to AD 500), early medieval period (c. AD 500 to 1100), medieval period (c. AD 1100 to 1650), post-medieval period (c. AD 1650 to the present).

Mesolithic (c. 8000 to 4000 BC)The earliest known human settlement in Ireland dates from the Mesolithic period (c. 8000 BC - 4000 BC). The majority of the evidence (flint scatters) for Mesolithic occupa-tion has come from the river valleys. No evidence for the Mesolithic was recorded on the route.

Neolithic (c. 4000 to 2000 BC)The Neolithic Period is characterised by the introduction of agriculture and the begin-nings of the clearance of the woodlands. The population increased and became more sedentary in nature. The most important Neolithic site in the vicinity was at Tullahedy recorded on the route of the Nenagh by-pass. It was a specialist chert arrow manufactur-ing site.

No evidence for a Neolithic site was recorded on the route but stone tools dating to the Neolithic were recorded at Busherstown E3661, Clash E3660, Cullenwaine E3741 and Greenhills 2 and 3 E3637 and E3658. Stone tools dating to the late Neolithic/Early Bronze Age were recorded at Busherstown E3661, Castleroan E3909, Cullenwaine E3741, Derrybane 1 E3585, Drumroe E3773, Greenhills 1 E3638 and Moatquarter E3910. Neo-lithic pottery was recorded at Cullenwaine E3741 and Drumbaun E3912.

Bronze Age (c. 2000 to 600BC)The Bronze Age is characterised by the introduction of metallurgy and an increase in settlement and burial sites. Copper ores were mined and copper, bronze and gold items manufactured. The range of burial site types includes cist graves, pit and urn burials, cremation cemeteries, barrows, ring-ditches and wedge tombs. Stone circles and stand-ing stones also date to the Bronze Age. Both enclosed and unenclosed settlement sites are known. The most prolific Bronze Age site type is the fulacht fiadh. These monuments survive as low mounds of charcoal rich black silt, packed with heat-shattered stones, and generally situated close to a water source. Fulachta fiadh are generally classified as ‘cook-ing places’, whereby stones were heated in a hearth and subsequently placed in a trough of water, the water continued to boil with the addition of hot stones and wrapped food was cooked within the hot water. The trough eventually filled with small stones, ash and charcoal that were removed, forming the basis of the familiar mound.

KilleisK-e3587

7


Two new fulachta fiadh / burnt mounds were recorded at Clashnevin 1 E3586, Cullen-waine E3741 and six at three separate locations in Greenhills, E3638, E3637 and E3658. Evidence of nine roundhouses or partial round structures were recorded; two at Cas-tleroan E3909, Derrybane 2 E3591 and Drumbaun 2 E3912 and one at Clash E3660, Drumroe E3773 and Moatquarter E3910.

Iron Age (c. 500 BC to AD 500)Up to recently there was little evidence of a significant Iron Age presence in Munster. Settlement sites are few and far between as well as being difficult to identify (Woodman, 2000) while the material culture of this period is limited. Linear earthworks, believed to have marked tribal boundaries, and hillforts are two of the most visible monuments of the period. Ten percent of sites excavated on NRA road schemes in recent years have produced Iron Age dates. The dates have led to the identification of 30 new Iron Age sites in Munster from road schemes in counties Cork, Limerick and Tipperary (McLaughlin 2008, 51). These include a ditched enclosure in Ballywilliam and a wooden trackway in Annaholty Bog excavated on the route of the N7 Nenagh-Limerick (Taylor 2008, 54).

Three Iron Age dates were returned from pits in Castleroan E3909 and Drumroe E3773 on the route of the N7 Castletown to Nenagh (Contract 1).

Early medieval period (c. AD 400 to 1100)The early medieval period is characterised by the arrival of Christianity to Ireland. The characteristic monument type of the period is the ringfort. Ringforts are the most nu-merous archaeological monument found in Ireland, with estimates of between 30,000 and 50,000 illustrated on the first edition of the Ordnance Survey 6” maps of the 1840’s (Barry 1987). As a result of continued research, the construction of these monuments has a narrow date range during the early medieval period between the 7th and 9th centuries AD. Although there are some very elaborate examples of ringforts, they often take the form of a simple earth or stone enclosure functioning as settlements for all classes of secu-lar society (Stout 1997).

North Tipperary is rich in early ecclesiastical sites and the remains of these religious centres are at the core of some of the towns and villages. Roscrea, for example, was chosen by St Cronan as a location for his monastery in the seventh century as it was located at the crossroads on the Slighe Dála, an important roadway in early medieval times (NIAH 2006, 4-8).

Early medieval activity was recorded at five sites on the route of the N7 Castletown to Nenagh (Contract 1). A series of corn-drying kilns were recorded at Busherstown E3661. A denuded ringfort (OF046-013) was excavated at Clynoe 2 E3774. An area of iron-working and associated pits was recorded at Drumbaun E3912. Iron working activity, corn-drying kilns and settlement activity was recorded at Park 1 E3659. A group of pits and associated ditch were recorded at Drumroe E3773.

Cla

shn

evin

1E

358

6C

lash

nev

in 1

E 3

586

Cla

shn

evin

2E

359

0C

lash

nev

in 2

E 3

590

Der

ryba

ne

1E

358

5D

erry

ban

e 1

E 3

585

Der

ryba

ne

2E

359

1D

erry

ban

e 2

E 3

591

Kil

leis

k 1

E 3

587

Kil

leis

k 1

E 3

587

Gar

rava

llyE

358

9G

arra

vally

E 3

589

Der

ryca

rney

1E

374

0D

erry

carn

ey 1

E 3

740

Kil

gort

een

1E

373

9K

ilgo

rtee

n 1

E 3

739

Cla

sh 1

E 3

660

Cla

sh 1

E 3

660

Par

k 1

E 3

659

Par

k 1

E 3

659

Par

k 2

E 3

772

Par

k 2

E 3

772

Cly

noe

2E

377

4C

lyn

oe 2

E 3

774

Cu

lleen

wai

ne

1E

374

1C

ulle

enw

ain

e 1

E 3

741

Mon

eyga

ll 2

E 3

635

Mon

eyga

ll 2

E 3

635

Gre

enh

ills

1E

363

8G

reen

hill

s 1

E 3

638

Gre

enh

ills

2E

363

7G

reen

hill

s 2

E 3

637

Gre

enh

ills

3E

365

8G

reen

hill

s 3

E 3

658

Dru

mba

un

2E

391

2D

rum

bau

n 2

E 3

912

Bus

her

stow

n 1

E 3

661

Bus

her

stow

n 1

E 3

661

Dru

mro

e 1

E 3

773

Dru

mro

e 1

E 3

773

Moa

tqu

arte

r 1

E 3

910

Moa

tqu

arte

r 1

E 3

910

Lou

ghan

1E

40

00

Lou

ghan

1E

40

00

Cas

tler

oan

1E

390

9C

astl

eroa

n 1

E 3

909

190400

190400

196200

196200

202000

202000

207800

207800

177200

177200

181800

181800

186400

186400

±

Kil

leis

k 1

Kil

leis

k 1

03

6

Kilometres


8

Figu

re2

:D

isco

very

ser

ies

Ord

nanc

esu

rvet

map

sho

win

gth

ero

ute

ofth

eN

7Ca

stle

tow

nto

Nen

agh

(Der

rinsa

llagh

toB

allin

tott

y)R

oad

Sche

me

(Con

trac

t1)a

ndth

elo

catio

nof

al

lexc

avat

ion

site

s�

KilleisK-e3587

9


High and later medieval periods (c. AD 1100 to 1650)This period is characterized by the arrival of the Anglo-Normans and the building of tow-er houses. The Anglo-Normans obtained charters in the thirteenth century for the towns of Nenagh, Roscrea, Thurles and Templemore and established markets. Nenagh grew rapidly in the aftermath of the granting of the lands of Munster to Theobald fitzWalter in 1185 (ibid. 8). Moated sites represent the remains of isolated, semi-defended homesteads in rural areas. They were build mainly in the late thirteenth and early fourteenth cen-turies in counties, such as Wexford, Kilkenny, Tipperary, mid-Cork and Limerick, that were colonised by English settlers (O’Conor 1998, 58). The Archaeological Inventory for North Tipperary lists 39 moated sites (2002, 298).

A medieval enclosure and associated field systems were recorded at Killeisk E3587. A newly recorded moated site was excavated at Busherstown E3661. A series of ditches and settlement activity was recorded at Park 1 E3659.

Post-medieval period (c. 1650 to the present)The post-medieval period is characterised by mills, limekilns, workhouses, country hous-es and associated demesnes, vernacular buildings and field systems (Figure 3). A small demesne associated with a county house was recorded in the townland of Greenhills.

5 Site Location and TopographyThe site at Killeisk was located in level pasturage at between 100 and 110m OD (Plate 1). A small tributary stream of the Ollatrim River (Abhainn an Chalatroma) ran across it from SE to NW, but had been diverted into an underground course in a deep ditch in modern times. It lay in the NW corner of Killeisk townland adjacent to the boundary with Garravally at W and Lissanisky at N.

6 Excavation methodologyThe site was mechanically stripped of topsoil under strict archaeological supervision. Stripping was done with a tracked machine with a flat toothless bucket. Topsoil stripping commenced in the areas of identified archaeology and continued radially outward until the limit of the road take was reached or until the limit of the archaeological remains was fully defined. A grid was set up in the excavation area(s) and all archaeological features were sufficiently cleaned, recorded and excavated so as to enable an accurate and mean-ingful record of the site to be preserved. The excavation, environmental sampling, site photographs, site drawings, find care and retrieval, on-site recording and site archive was as per the Procedures for Archaeological works as attached to the licence method state-ments for excavation licences.

LISSANISKY

KILLEISK

GARRAV

ALLY

BALLINREE

DERRYCARNEY

Ollatrim River Tributary

0150

300

Meters

¥


10

Figu

re3

:Po

rtio

nof

the

Iste

ditio

nO

rdna

nce

Surv

eyM

apO

F47

show

ing

the

loca

tion

ofK

illei

ks�

KilleisK-e3587

11


The site was excavated from 9 June to 25 August 2007. Only areas within the LMA (lands made available) were resolved. The full extent of the area of excavation measured 10,000 m sq (Figure 4).

The full record of excavated contexts is recorded in the context register (Appendix 1) and the stratigraphic matrix (Appendix 2). Detailed stratigraphic descriptions are found in the groups and sub-groups text (Appendix 3). The context register and site photographs maybe viewed in the EAPOD (Eachtra Archaeological Projects office database) in the accompanying CD.

0 30 60Meters ±Killeisk 1 (E3587)Killeisk 1 (E3587)

Plate1: AerialviewofKilleisk�

3400

3300

3200

3100

3000

2900

2800

2700

2600

2500

2400

2300LIS

SANIS

KY

LIS

SANIS

KY

KIL

LEIS

KKIL

LEIS

KGARRAVALLY

GARRAVALLY

Ollatrim River Tributary

194132

194132

194502

194502

194872

194872

179315

179315

179545

179545

179775

179775

±0

100

200Metres

Kil

leis

k 1

(E35

87)

Kil

leis

k 1

(E35

87)


12

Figu

re4

:Lo

catio

nan

dex

tent

ofK

illei

skE

3587

on

the

N7

Cast

leto

wn

toN

enag

h�

KilleisK-e3587

13


7 Excavation resultsArchaeology on the site at Killeisk consisted of four main groups of features and a rela-tively small group of ungrouped features (Figure 5). The four groups, each dealt with in detail below, were:

[1] Pit Group 1: A number of mostly irregular pits some of which were filled with ‘burnt mound’ type charcoal enriched soil with some heat-shattered stone. Dating to the Early Bronze Age, this group formed a fairly diffuse cluster slightly to SW of the centre of the site.

[2] Pit Group 2: A discrete cluster of mostly irregular pits and cuts, some of which were filled with ‘burnt mound’ type fills. These were located near the N end of the site and again dated to the Early Bronze Age.

[3] Enclosure and Associated Field-system: Part of a large elliptical enclosure and as-sociated linear features which themselves form part of a system of enclosures, droveways and fields apparently dating from the late medieval period. This system occupied much of the site except along its NW edge.

[4] Kiln and Associated Features: The sub-surface remains of a limestone-built kiln, also apparently of late medieval date, along with three associated smaller features. This group lay near the SW extremity of the site.

Description of Prehistoric Pit Group 1This group consisted of some 30 cut features, mostly pits of some sort, which lay scattered throughout the W central area of the site (Figure 6). The main concentration of these pits lay in a cluster N of the droveway ditches. No finds were recovered in any of the features forming part of this group. Six of the pits in this group were filled or partially filled with black charcoal-rich ‘burnt mound’ type deposits with fire cracked stone. A radiocarbon date for oak charcoal from this material gave a range of BC 2276–2047.

The central and most obviously archaeological part of the group is the central cluster formed of seven fairly large cuts – three of them intercutting – and seven stakeholes. Cut C245 [l. 2.34m, w. 1.26m, d. 0.18m] was irregular in plan with a flattish base and had three stakeholes cut into its base, [C323, C325 and C327] (Figure 7, Plate 2). The cut and all three stakeholes were filled with the same material, C244, a black charcoal-rich ‘burnt mound’ type fill. Adjacent to the pit and forming an arc on its SE side, were four fur-ther stakeholes [C354, C352, C356 and C350]. The fill in all four cases was a mid greyish brown silty sand with no inclusions. The stakes are probably associated with the activity at pit C245. The stakes cut into the base of the pit and those forming an arc on one side of it would suggest some form of associated superstructure suspended on the stakes.

Cut 247 [l. 2.76m, w. 1.16m, d. 0.35m] was a large sub-rectangular pit. It had a single original fill, mid grey sandy silt C293. A further, quite irregular cut [C268] had been in-cised through the original fill on the N side of C247. There were three fills in this cut, the uppermost of which [C246] was a black stony, charcoal-rich silt of ‘burnt mound’ type

4

11

13

92

9

100

24

19

102

317

80

358

76

111

90217

221 445

39

128

281245

421

305255

364

464

365

54

442

219

491

213

402253

522

474

334

456

452

509

249

381

135

409

173

415

307

390

518

437

484

251

273

397

502

499

261

435

512

259229

401

448

179

344

188

204

412

225

417

270

288

289

183

282

280

313

181

187

466

171

472

194469

194469

194540

194540

1794

58

1794

58

1795

64

1795

64

Enclosure

Droveway

Kiln

Well

±

0 30 m

)O101 m O.D.

L8

L1

L3

L2

L4

L5L6

L7 L9

L10

Sub-enclosure

Bronze Agepits 1

Bronze Agepits 2

L8


14

Figure5: Post-excavationplanofKilleiskE3587�

KilleisK-e3587

15


Plate2: ViewofpitC245andassociatedstake-holesatKilleisk�

Plate3: ViewofpitC281atKilleisk�

11

317

217

221

445

281

421

305

255

365

219

402

253

334

452

249

381

409

390

259

288

±

010

m

245

323

350

327

352

325

354

356

Bron

ze A

gepi

ts 1

L4


16

Figu

re6

:Po

st-e

xcav

atio

npl

ano

fpre

hist

oric

pit

grou

p1

atK

illei

sk�

C.2

44

C.2

45

C.3

25

C.3

24

Kill

eisk

E3

587

NW

faci

ng p

rofil

e of

C.2

45

C.3

52C

.354

C.3

51C

.353

Kill

eisk

1E3

587

NE

faci

ng p

rofil

e of

C.3

52 a

nd C

.354

050

cm

10 c

m

C.2

44

C.2

45

Kill

eisk

E358

7N

E fa

cing

sect

ion

of C

.245

KilleisK-e3587

17


Figu

re7

:Se

ctio

nso

fpit

C245

and

ass

ocia

ted

stak

e-ho

les

atK

illei

sk�


18

which was also the main secondary fill of original cut C247. A further large sub-oval cut, C317 [l. 1.9m, w. 1.32m, d. 0.43m] had been cut adjacent to C247 on its E side. This itself was truncated by a further oval cut, C345 [l. 0.44m, w. 0.39m, d. 0.43m], which had been cut into the bottom of C317. Both fills post-dated the recut.

Some 3m to NE of the four intercutting pits lay C281 [l. 3.2m, w. 2.15m, d. 0.89m], a large, deep cut, sub-circular in plan with steep sides and a flat base (Plate 3). It contained six fills, five of which appeared sterile. The second lowest fill, C278 [d. 0.73m], was the exception, a fine soft dark bluish silt with inclusions of charcoal lumps and which yielded a sample of waterlogged wood. This pit may have been used as a water-hole or well.

Two further pits in the immediate vicinity contained ‘burnt mound’ type fills, ir-regular pit C249 and sub-circular pit C219, oak charcoal from the fill of which gave a radiocarbon date range of cal BC 2276-2047 (UB–15091).

About 10m to W of the above features lay a scattered group of three quite irregular pits, C381, C390 and C402. The fills of these features contained a certain amount of charcoal but other wise their form and content were uninformative.

Interpretation of Pit Group 1.This group is interpreted as a cluster of pits some or all of which relate to some form of pyrotechnical activity.

Of the nine cut features considered to be definitely archaeological, all but one (C219) were found in a tight grouping. Here we had four pits the fill of which was typical of burnt material from prehistoric pit groups, consisting of black charcoal-enriched silt with frequent inclusions of fire-shattered stone. One of the pits had three stakeholes cut into its base, with a further four stakeholes forming an arc immediately to SE. This suggests the likelihood of some form of superstructure having existed in this case. Another pit, C247, had a distinctly rectangular shape in plan. Pits C247, C268 and C317 intercut each other and the latter had a further recut, C345, in its base, all of which suggest some reuse or alteration of this arrangement of pits. None of the pits containing ‘burnt mound’ type material exhibited any signs of in situ burning, and so it is not certain that they were originally intended for use in relation to pyrotechnical activity. Nothing in their fills or shapes indicated what their original purpose might have been, and only the usual sug-gestion of use for storage may be made. A single grain of barley was recovered from the fill of pit C247, but while interesting, a single grain cannot be deemed sufficient to sug-gest anything as regards the function of the pit and could easily have blown in from the surrounding area. Nonetheless, it does at least suggest that agricultural and/or domestic activities were taking place in the immediate vicinity of the pits.

Adjacent to these pits at NE was a large, deep sub-circular cut, C281, the lower fill of which was waterlogged and which produced samples of waterlogged wood. This may have been a well or water-hole. Its proximity to the pits filled with burnt material and stakeholes suggest it might have been a water source used to supply the needs of whatever ‘hot stone’ process was taking place here.

KilleisK-e3587

19


Other than this grouping just discussed, the other features in Pit Group 1 are less certainly linked. Small pit or post-hole C259 had a shape which suggested deliberate for-mation, but was likewise isolated from any other definitely archaeological feature. Other than the ‘burnt mound’ type fills referred to, the other fills of the features in this sub-group, though often containing traces of charcoal, are mostly grey and brown silts of the type which occur all over the site and which seem to result from natural silting.

C219 is another flat-based sub-circular pit filled with ‘burnt mound’ type material, but lies isolated from the others some 9m to N. It is of particular note, however, because it is the only feature in this group to produce a radiocarbon date. This was for oak char-coal from the upper ‘burnt mound’ type fill C218 and gave a dating range of cal BC 2276–2047 (UB–15091). Such a date, in the Early Bronze Age, is fairly typical for such pit groups with their burnt fills.

Description of Prehistoric Pit Group 2This group consists of a fairly discrete cluster of 16 cut features, all but one lying within the same 10m grid [120,120] and the exception lying immediately outside of this (Figure 8). Of these cuts 9 are pits and 7 stakeholes. The overall appearance of the group was very similar to that of Group 1 above, i.e. a number of pits clustered around a central large pit which featured a number of stakeholes within and/or around it. There were likewise a significant number of features which were filled or partially filled with ‘burnt mound’

Plate4: ViewofcharcoalenrichedfillsofpitsinBronzeAgePitsGroup2atKilleisk�


20

type fills [6 of the pits] (Plate 4). Again, no artefacts or bone were recovered from any of the features. A radiocarbon date of cal BC 2286–2140 (UB–15092) for Pomoideae [family including apple, and hawthorn] charcoal, showed that the two pit groups at Killeisk are roughly contemporary.

The central feature was a large flat-based circular pit, C466 [l. 1.92m, w. 1.6m, d. 0.2m], the lower fill of which [C465] was ‘burnt mound’ type material and which had 7 stakeholes cut into its sides and base [C525-531] (Figure 9, Plate 5). The fills of these stakeholes were in all cases free of the burnt material filling the lower part of the pit, a fact suggesting the stakes had been removed or had rotted a long time before the dumping of the burnt fill in the pit.

The W side of C466 was truncated by an elongated irregular cut, C468 which had the same burnt fill as the former. It was itself truncated by a post-medieval field-drain. Three more quite substantial but irregular pits [C472, C474 and C464], two of which contained ‘burnt mound’ type material, lay in the immediate vicinity of the central round cut. A further spread of this material was excavated immediately to W of C468.

About 0.8m to E of C466 was C464 [l. 1.27m, w. 1.01m, d. 0.24m], a pit of irregular plan with gentle to steeply sloping sides and an irregular base. The single fill was C463, a black charcoal-enriched silty sand with fragmented stones. A radiocarbon date range of cal BC 2286–2140 (UB–15092) was returned for charcoal from this fill.

491

474

464

509

472

468

518

484

502

499

512 507

517

±

0 5 m

466

468

528

530

525

526

529

527531

464

Bronze Agepits 2

Figure8: Post-excavationplanofprehistoricpitgroup2atKilleisk�

050

cm

10 c

m

C.4

96

C.4

65

C.4

66

C.4

63

C.4

64

Kill

eisk

E358

7Ea

st fa

cing

sect

ion

of C

.464

Kill

eisk

E358

7SE

faci

ng se

ctio

n of

C.4

66

#

#

##

#

#

##

##

##

##

##

##

##

KilleisK-e3587

21


Figu

re9

:Se

ctio

nso

fpit

C464

and

C46

6at

Kill

eisk

�


22

Four other smaller cuts [C499, C507, C512 and C517] lay scattered between 1 and 5m to N of the central cluster, two of which contained ‘burnt mound’ type material, and which are interpreted as small pits or post-holes.

Interpretation and Discussion of Pit Group 2.This group was similar to Pit Group 1 in that it consisted essentially of a cluster of pits centred on a single large pit with stakeholes incised into its base and sides. In both groups there was ample evidence of the use of ‘hot stone’ technology for some purpose, in the form of fills of black, charcoal-rich silts with quantities of heat-shattered stone. Also, in both cases the densest distribution of cut features occurred around the central pit, thus confirming its central function and importance, with the burnt fills found almost exclu-sively in the central feature and the cut features immediately adjacent to it.

The seven stakeholes in the base of circular cut C466 were the best structural evidence in this group and must represent the remains of some form of superstructure relating to the pit’s use. However, the fills of the stakeholes indicate that they were either rotted or withdrawn and naturally silted up before the pit was used as a location to deposit the burnt stone fill. Alternatively the stakes might have been in situ and still standing when the burnt fill was deposited, but this seems unlikely. In the latter scenario one would expect some of the burnt fill to have entered the upper fill of the stakeholes as the stakes rotted, and this was not found to be the case. These facts, coupled with the absence of any evidence in the sides or bases of any of the pits for in-situ burning (another fact paral-

Plate5: ViewofpitC466andassociatedstake-holesatKilleisk�

KilleisK-e3587

23


leled in Pit Group 1), emphasizes that the deposition of the ‘burnt mound’ type fills was – despite being the most notable feature of the group – peripheral and subsequent to the original usage of the pit group, whatever that may have been. Large regular-planned pits with numbers of stakeholes cut into their base are frequently found in association with Fulachta Fiadh in which situation they are generally interpreted as water troughs.

Other than the stakeholes referred to, there are two other features which could have served a structural purpose; cuts C499 and C512. Both of these were sub-oval with round, flat bases tilted at a slight angle, and of similar dimensions. They are considered possible post-holes and lie 3.6m from each other. The possibility that they were both part of the support for the same structure is noted, although in the absence of further post-holes a suggestion of what this might have been would be highly speculative. The fills of both these cuts are completely free of any charcoal content, which would suggest they were not contemporary with the phase of pyrotechnical activity which is noted in the ‘burnt mound’ type fills of six cuts. The absence of any form of post-pipe suggests that if C499 and C512 held post-holes that they were withdrawn at some stage and allowed to silt up naturally.

A shared angular but irregular morphology is noted in the plans of two adjacent fea-tures, cuts C472 and C474, but no plausible functional explanation can be advanced at present.

Overall, then, it is only possible at present to say that some process involving ‘hot stone’ technology was taking place in the immediate environs of Group 4, but that this was not central to the reason for which these cuts were originally excavated. In terms of dating, a single radiocarbon sample of Pomoidae charcoal from the ‘burnt mound’ type fill of pit C464 gave a dating range of cal BC 2286–2140 (UB–15092). This range is simi-lar to the Early Bronze Age date from Pit Group 1 and suggests the two pit groups may have been contemporary.

Description of Enclosure and Associated Field systemDuring excavation each linear feature on site was allocated a Linear Number for ease of discussion [see Table 1 and Figure 5]. The majority of these [L1, L2, L4, L5, L8, L9 and L10] form part of a field system associated with an enclosure while two [L3 and L6] were modern field boundaries.

The majority of the ditches on the site were excavated as a series of substantial sections rather than in a single-context manner. As a result of this, and adhering to correct archae-ological recording methodology, within each linear feature several widespread contexts received several context numbers which were subsequently conflated under the lowest of the context numbers given; e.g. the main cut in enclosure ditch L1 was given, in different sections, the numbers C4, C5, C32, C38, C43, C59, C97, C106, C122, C143, C153 and C160, but for purposes of post-excavation analysis they are all called C4.


24

Plate6: ViewofenclosureL1atKilleiskfromnorth-west�

Plate7: ViewofmaincutofenclosureditchC4bottomedoutandinsectionrectutC7atKilleisk�

KilleisK-e3587

25


Linear Type Main cut nos.1 Enclosure ditch C42 Sub-enclosure ditch C9, C,39 and C903 Modern field boundary4 N droveway ditch C115 S droveway ditch C136 Modern field boundary 7 Kiln windbreak C1008 Field boundary C24 and C929 Field boundary C102 and C52210 Field boundary C111

Table1LinearNumbersandcorrespondingcutnumbersusedinthetext�

Main enclosure ditch [L1] and internal features This formed an ellipse in plan, aligned with its long axis running NW-SE (Figure 10, Plate 6). Only approximately two fifths of its area was exposed within the road-take, the rest of its extent being traced in the adjacent field to SE. The enclosure was associated with a number of linear cuts which are considered on the basis of horizontal stratigraphy and finds to be coeval and which form part of a system of enclosures, sub-enclosures, drove-ways and field-boundaries. The excavated portion of the main cut of the enclosure ditch, Linear 1 (C4), had an overall circumference of c.120m, with a width varying from 1.4m up to 2.5m and a depth between 0.16m and 0.72m (Plate 7). In general, it was wider and shallower on its SW side and narrower and deeper on its NE side. The base varied from flat to flattish irregular and the sides were mostly slightly concave.

Some 33 different layers/deposits were excavated within the cut. Of these, 11 showed clear signs of human activity, mainly through the presence of animal bone and/or a good degree of charcoal flecking. Analysis of the bone recovered from the various fills of the enclosure ditch has identified those of cattle, horse, sheep/goat, pig, cat, dog, deer, rab-bit and some type of bird. Bone was recovered from the uppermost context [C1] down to some of the lowermost [C104, C81, C36], indicating that whatever was producing the bone waste – probably domestic food preparation – was practised for the duration of the enclosure’s use period. Of particular note in this regard was C6, a widespread and thick fill which was heavily charcoal flecked and contained a significant number of animal bone fragments. This fill is probably a habitation layer washed into the cut. Some iron artefacts were recovered from the enclosure cut, including a horse-shoe and several nails. These were confined to three of the uppermost fills of the enclosure cut [C1, C155 and C149]. Also near the surface, on the S side of the enclosure cut where it borders the sub-enclosure between L1 and L2 [see below], was a deposit of large stones and rocks, C88, which included part of a rotary quern fragment (E3587:88:1) (Plate 8). C3 was a layer of silty clay which occurred along the inner edge of the cut of the ditch throughout the ma-jority of its exposed length and which is thought to be material slumped from an internal bank associated with the ditch (see Figure 14).

4

9

24

19

92

11

80

76

13

90

39

54

442

213

456

135

419

173

415

434

437

251

273

397

261

435

229

401

448

177191

204

412

417

270

289

183

294

282

179

280

187

171

134

185

431

188175

193

196

Enclosure

Well

±

0 10 m

307 225

201

205

230

234

311

210

285

207

240

313

181

232242

296291

263

237

Subenclosure

L1

L8

L8

L2

L4

L5


26

Figure10:Post-excavationplanofenclosureandassociatedfieldsystemsatKilleisk�

KilleisK-e3587

27


Plate8: View of stone layer C88 in enclosure ditch C4 at Killeisk� Note quernstone E3587:88:1 abovewhiteboard�

Plate9: ViewofmaincutofenclosureditchC4andthenarrowerrecutC7inscisedthroughitsbase�

C.1

03

C.1

49

C.1

04

C.1

44

C.1

35

Kill

eisk

E358

7N

W fa

cing

sect

ion

of C

.135

C.1

36

050

cm

10 c

mL1 E

nclo

sure

Ditc

h

Wel

l

C.4


28

Figu

re1

1:S

ectio

nso

fenc

losu

red

itch

L1C

�4a

ndw

ellC

�135

atK

illei

sk�

KilleisK-e3587

29


A cow tibia from C104, a basal fill of the ditch, was dated to AD 1292–1404 (UB–15737).

Two recuts [C123 and C7] at different stratigraphic levels demonstrate that the ditch was cleaned out on more than one occasion, most likely due to natural silting. Both were flat-bottomed, with C123 present in the W and SW sections of the ditch and C7 confined to the W section (Plate 9). The two were separated by the abovementioned thick deposit of habitation surface C6, with C123 cut into this deposit and C7 underlying it, at one point cutting sterile fill C37, at another cutting through the original enclosure cut C4 and into the natural subsoil. At the point along the main enclosure cut C4, at NW where it intersects with field boundary L8, there is a subsidiary cut [C58] parallel with and both adjacent and exterior to main enclosure cut C4, traceable for a distance of 2.4m (Plate 10). The purpose of this extra cut in this area is unknown.

At one point at N the enclosure cut C4 has a deep pit cut into it’s base (Figure 11). This pit had two fills, the uppermost being C136 [d. 0.84m], a soft mid grey silty clay which was not confined to the cut but lay above it in the immediately surrounding area of C4. The lower fill was C144 [d. 0.35m], a soft brownish grey silty clay containing a large piece of animal longbone and pieces of semi-decayed wood [probably tree root]. A number of medium and large stones lay at the base of the pit and may have been delib-erately placed. The cut itself, C135 [diam. 0.54, d. 0.7m] tapered slightly from top to base and had a fairly flat bottom. The purpose of the pit is unclear, but a form of water-hole or sump might be suggested.

Plate10:ViewofL8C24,C58andL1C4fromlefttorightatKilleisk�


30

At no point was any gap in the enclosure which could be considered evidence of a causeway or entrance noted, although the deposit of rocks C95 at SW noted above might possibly be considered a secondary effort to make such a causeway. Nonetheless, it seems certain that an original entrance must be present somewhere along the greater, unexposed part of the enclosure ditch lying outside the road-take to SE.

Of a total of 45 possible cut features excavated within the enclosure defined by ditch L1 only 12 are considered likely to be archaeological. The internal features do not form any recognizable pattern in plan. Almost all of these are of a similar depth (between 0.1m and 0.15m) and are considered to be heavily truncated. The most notable features here were C213, a large but very shallow rectangular pit [l. 1.8m, w. 0.8m, d. 0.12m] with two fills, the upper [C211] consisting of a layer of large angular stones, many of which are long and flat and C80, an oval pit with two fills, the upper of which appeared rich in organic matter and which yielded samples of the tooth and bone of pig and sheep/goat. Ecologi-cal analysis of samples from C83, a fill of pit C80, yielded 26 carbonized cereal grains of which 9 were oat, 4 wheat and 3 barley, while 10 were unspecified. A radiocarbon date for hazel charcoal from C83 gave a date range of AD 1516–1647 (UB–15088).

Curved linear ditch [L2] and internal featuresLinear feature L2 was discontinuous in nature, consisting of three cuts [from N to S they are C39, C90 & C9] arranged end-to-end with narrow gaps between. Its two ends termi-

Plate11:ViewofL2C39andC90fromNEandLIC4inforegroundatKilleisk�

KilleisK-e3587

31


nated in relation to L4 and L1. The S end stopped some 0.3m short of L4, one of a pair of parallel ditches forming a droveway [see below], while its N end intersected with the W side of main enclosure ditch L1 (Plate 11). For most of its length it was fairly straight, running N-S, but turned abruptly to NE near its N end to meet perpendicularly with L1. It therefore appears to have been excavated in deliberate relation to L1 and L4 and is therefore at least partially coeval with them.

The S of the three cuts, C9, had three fills. From top to base these were C18, C27 and C8, which yielded samples of dog bone [C18] and cattle bone [C27 & C8]. No finds were recovered from the ditch. Central cut C90 had a single fill and N cut C39 two fills. At a maximum width of 0.6m, the cuts of L2 were all quite narrow when compared to the other linear cuts in this group, with the exception of L10 [see below].

Although not completely exposed, the area between linear ditches L1, L2 and L4 ap-pears to form a sub-enclosure adjacent to the main elliptical enclosure. The interpretation of this bounded space as a separate sub-enclosure is confirmed by the presence within it of a number of features while the immediate external area, to W of L2, is almost featureless.

The internal features do not form any recognizable pattern in plan. Four small to medium sub-circular or oval pits or post-holes were excavated within the sub-enclosure [cuts C401, C435, C397 and C417]. Of these, the lower fill of pit C397 yielded a sample of cattle tooth and a single grain of wheat, while the only other possible evidence of human activity were flecks of charcoal and burnt bone in C401. As well as the cut features, two discrete areas of stone in a hard gritty matrix were excavated within the sub-enclosure,

Plate12:ViewofL4C11andL5C13fromNWatKilleisk�


32

C384 [l. 3.3m, w. 3m, d. 0.06m] and C386 [l. 2.46m, w. 1.9m, d. 0.05m]. They are inter-preted as possible areas of deliberate metalling.

The droveway [L4 & L5] This sub-group consists of two nearly-parallel linear ditches L4 (C11) and L5 (C13), which formed a droveway aligned running roughly W-E (Plate 12). They are co-terminous at their W end, while both run under the baulk at E. The area between the two ditches nar-rows slightly from E to W. At the E end this is 4.8m and at the W terminal of the ditches, some 38m away, it is 3.4m.

The N of the two ditches, L4 consisted of two separate cuts aligned end-to-end with a short gap [w. 0.4m] between them. While the full extent of C11, the W cut, was exposed [l. 41.5m, w. 1.3m, d. 0.34m], only c.1m of the E cut [C54] lay within the excavation area, it disappearing under the SE baulk some 1.1m away. The shape of the cut in profile varied considerably over its length, with slope of sides varying from gentle to moderate and base from rounded to irregular to flat-bottomed (Figure 12). Four fills were recorded. The uppermost was C10, from which a piece of a cross inscribed quernstone (E3587:130:1)

C.20

C.21

KilleiskE3587SE facing section of C.21

0 50 cm10 cm

C.521

C.520

C.522

Killeisk 1E3587North facing section of C.522

C.151

C.152 Modern field drain

Killeisk E3587NE facing section of C.152

Linear 9

Linear 8 Ditch C.92

Linear 4 Ditch C.11

Figure12:SectionsofL9C522,L8C92andL4C11�

C.20

C.21

KilleiskE3587SE facing section of C.21

0 50 cm10 cm

C.521

C.520

C.522

Killeisk 1E3587North facing section of C.522

C.151

C.152 Modern field drain


Linear 9

Linear 8 Ditch C.92

Linear 4 Ditch C.11

KilleisK-e3587

33


(Plate 13) was recovered as well as a single sample of animal bone. The lower two fills were sterile.

The S ditch of the pair, L5, consisted of a single cut, C13 [w. 1.25m, d. 0.41m] which ran under the SE baulk of the site. Some 35.5m of its length was exposed within the ex-cavation area. Like the cut of L4, its profile varied greatly between gentle and moderately sloping sides and a base which was flat, rounded and irregular in different places along

Plate13:QuernstoneE3587:130:1�

Plate14:QuernstoneE3587:125:1�

C.2

5

C.2

4

Kill

eisk

E3

587

East

faci

ng se

ctio

n of

C.2

5

C.1

2

C.1

3

C.1

17

C.1

18

C.1

19

C.1

27

C.1

28

Ove

rcut

t

Kill

eisk

1E3

587

SE fa

cing

sect

ion

Kill

eisk

1E3

587

SE fa

cing

sect

ion

of C

.13

050

cm

10 c

m

Line

ar 5

Ditc

h C

.13

Line

ar 5

Line

ar 8


34

Figu

re1

3:S

ectio

nso

fL5

C13

and

L8C

24�

C.1

12

C.1

11

Kill

eisk

1E3

587

NE

faci

ng se

ctio

n of

line

ar C

.111

C.1

29

C.1

20

Nat

ural

C.1

21(C

.13)

Kill

eisk

1E3

587

SE fa

cing

sect

ion

of li

near

ditc

h 4

050

cm

10 c

mLine

ar 5

Ditc

h C

.13

Line

ar 1

0

Line

ar 1

Ditc

h C

.04

C.7

3

C.7

7

C.7

8

C.0

5

Kill

esik

1E3

587

NW

faci

ng se

ctio

n of

C.0

5

KilleisK-e3587

35


Figu

re1

4:S

ectio

nso

fL1

C4,L

5C1

3an

dL1

0C1

11�


36

its length (Figure 13 and see Figure 14). Two fills were recorded within the cut with the lower, C12 yielding part of a decorated quern stone (E3587:125:1) (Plate 14) and several samples of animal bone at least one of which was of cattle.

No other features considered to be archaeological were excavated within, adjacent to or between the two droveway ditches.

Enclosure ditch or field boundary [L9 and L10] This consists of two linear ditches considered likely to have originally formed part of a single ditch but which have both been truncated by a modern field boundary [L6].

L9 (C.388 and C.522) is discontinuous, consisting of two cuts with a narrow gap be-tween them and aligned NW-SE. Only 1.7m of the W tip of SW cut, C522 [w. 1.03m, d. 0.26m], is exposed, it running under the baulk at E (see Figure 12). The NW cut, C102 [w. 1.4m, d. 0.35m], was separated from C522 by a gap 0.9m wide and ran for 10.3m before being truncated by L6. No finds or animal bone were recovered from either cut of L9.

L10 consisted of a single cut, C111 [w. 0.64m, d. 0.07m], running SW-NE which was truncated at SW by modern feature L6 and which terminated at NE some 0.75m short of L5, the S ditch of the droveway [see above]. A sample of animal bone was recovered from C112, the single fill of C111 (Figure 14).

L9 and L10 may originally have formed a single curved linear which, along with L5 at N, would have formed three sides of an enclosed area or field (the fourth side would have been out in the unexcavated area to SE).

Field boundary [L8] This linear ditch enters from under the NE baulk of the excavated area and runs in a SW direction for approximately 45m until it intersects with main enclosure ditch L1, where it terminates. L8 (C92 and C24) is composed of two identically aligned cuts with a gap some 3.9m wide between them. The NE part of L8 consisted of a single flat-bottomed cut C92 [l. 27m to point where it exits under baulk, w. 0.88m, d. 0.11m] and sterile fill C93 (see Figure 12). The SW part of L8 consisted of a single cut, C24 [l. 16m, w. 0.94m, d. 0.18m], again flat-bottomed, with three fills, all sterile (see Figure 13). A single stakehole was cut into the base of C24 near where it intersected with enclosure ditch L1 at SW.

Interpretation of Enclosure and Associated Field systemThis is interpreted as a high medieval domestic enclosure with sub-enclosure and a sur-rounding field system. On the basis of the horizontal stratigraphy alone the direct rela-tionship between L1 and the other linear ditches seems clear. The ditches either respect each other or join, with one failing to emerge on the far side of the other. They are also linked by similar shape and dimensions [except for L2 and L10], by the nature of their fills and by the recovery of animal bone from each [with the exception of L9 and L8]. L1, L4 and L5 are also linked by finds of rotary quernstone fragments in each. The existence of gaps or causeways in the ditches is a further common feature.

100

358

364

344

Kiln

±

0 5 m

L7

KilleisK-e3587

37


Figure15:Post-excavationplanofkilnandassociatedfeaturesatKilleisk�


38

It seems quite clear that an internal bank originally accompanied L1, the main en-closure ditch and was most likely composed of the upcast from the ditch. This is dem-onstrated by fill C3, which occurred throughout most of the excavated circumference of the cut, visible in section as a deposit slumping down into the ditch from its internal edge and generally composed of sterile clay similar to the natural subsoil. Nothing indicating the existence of a palisade atop the bank was noted during excavation, although it is quite possible that such evidence would not survive given the apparent level of truncation on site.

Evidence of periodic recutting and/or cleaning out of the ditch following natural silt-ing was identified in the form of at least two recuts [C7 and C123] which stratigraphy demonstrated were clearly not coeval. The presence of these recuts suggests that either

[1] continuous habitation was of sufficient duration to necessitate periodic clearing of natural silt and domestic waste from the ditch, or

[2] that habitation was sporadic, with corresponding periods of abandonment during which the ditch silted up.

Given that a high medieval date is suggested for this site and what we know of domes-tic habits at that time, the former suggestion would seem is more likely.

Despite the ditch and internal bank, it seems unlikely that these bounding elements were defensive in any military sense of the word, but were more likely of similar function and scale to those of the average ringfort, serving to bound the space, to contain domestic stock and to exclude wild animals. The depth of the ditch on its NW and W sides is quite shallow, as little as 0.16m where it joins field boundary L8, although truncation may be partially responsible for this as the latter linear ditch is also far shallower than the other examples excavated.

There are elements of the ditch which are not well understood. The deep pit C135 on its NE side may be a well or water-hole. Although it seems likely that runoff from the ditch into this would not be a clean source of potable water, it may have been a drinking hole for livestock or, more likely given its depth and position in the base of the ditch, a source from which water could be removed to a suitable receptacle for livestock. The divi-sion of the ditch into two adjacent parallel cuts for a distance of 2.4m in the area where it joins field boundary L8 and is at its shallowest, is a feature for which a convincing expla-nation has yet to be formulated.

Of the 12 archaeological features excavated within the main enclosure most seem to be pits, probably serving a variety of domestic uses such as storage and/or refuse disposal. Almost all of these are of a similar depth (between 0.1m and 0.15m) and are considered to be heavily truncated. The fills in most of the 12 cuts contain some level of charcoal content, although mostly low enough to be explicable simply through having been open in the general area where fire was being used. However, a number of fills – in cuts C251, C261 and C229 – had significant concentrations of charcoal and burnt clay/ash which may have resulted from the deliberate dumping of residues from fire-related domestic processes.

KilleisK-e3587

39


Pit C213 was considered initially on shape and size as a possible grave but nothing to confirm this was recovered, although the high level of truncation and the presence of a probable stone lining suggest this possibility should not be entirely ruled out. Nonethe-less, the generally domestic character of the find and sample evidence from the Group 1 features suggest it is unlikely that a formal burial would occur here. The fill of several pits was noted to be particularly rich in organic content, such as C80, which might result from rotted down domestic refuse. C80 is the only cut in the interior to yield animal bone/tooth which links it to the finds of such material in most of the linear ditches exca-vated. Although a late medieval date was obtained from hazel charcoal from the pit C80 and a high medieval date from animal bone from the ditch.

One or two of the cuts – C229 and C188 – might be interpreted as post-holes, par-ticularly the former which appeared to have separate post-pipe and packing. However, no alignment or other indication of a structural pattern could be discerned.

Thus the evidence for human activity from within the main enclosure was modest with little to indicate the purpose of the pits here or even to clearly demonstrate that hu-man habitation took place. Truncation is clearly partially responsible for this, but more significant is the fact that less than half the enclosure was excavated. This suggests that much of the evidence lies in the unexcavated greater part of the site to SE [c.1,150 sq.m. of a total of c.2950 sq.m. was excavated, thus giving a ratio of excavated to unexcavated area of 23:59].

Finds from the enclosure ditch included two undecorated fragments of quernstone [one of which was ex situ, as recovered during initial monitoring] and an iron horse-shoe and spike-nail. Many bone samples, though mostly of small quantity, were recovered, with the following species identified: cattle, horse, sheep or goat, pig, dog, rabbit and bird [unspecified].

Between the main enclosure and linear ditch L2 was a sub-enclosure which was prob-ably enclosed also at S by a continuation of N droveway ditch L4. L2 was relatively nar-row in relation to the majority of the other ditches [L10 being the exception], as well as having three fairly closely spaced gaps along its length, making it more discontinuous than any of the other ditches. Nonetheless, the existence of a number of archaeological features in the area it encloses and the absence of any in the area immediately outside, indicate that it formed an effective boundary, at least in relation to human activity. An existence of an accompanying palisade might be suggested here although there is an ab-sence of evidence for such in the form of stakeholes or such like. The features within the enclosed space consisted of several pits, post-holes and discrete areas of metalled surface. Like the features within L1, there was little to indicate a definite purpose for those within L2, and significant structural evidence was likewise absent. A sizeable deposit of large rocks [including a quern fragment] was excavated within ditch L1 where it runs between the main enclosure and the sub-enclosure. This deposit [C88] is interpreted as a deliber-ate deposition intended to form a causeway between the two enclosures. It also indicates that a causeway between the two enclosures was not planned for during initial excavation of the L1 ditch [which appears to be earlier on the basis of horizontal stratigraphy]. This


40

might mean one of two things, either that L2 and thus the sub-enclosure are significantly later additions, or that the two enclosures were originally planned together, but that the function of the sub-enclosure changed in some way at a later stage which made direct ac-cess between the two enclosures desirable.

It is also worth noting that C88, the stone deposit in question, is quite high up in the stratigraphic sequence, with significant silting and slippage from the internal bank – as well as evidence of two separate recuts – below it, implying that it was laid down at an advanced stage in the habitation of the site.

No linear features were recorded to either W or N of the W end of the droveway [L4 and L5] and it would seem that this was either unenclosed or part of a very large enclosure, the bounding elements of which lie outside the excavated area. The small gap between the W and E cut of the droveway’s N linear, L4, lies immediately E of linear ditch L2 and may be interpreted as an entryway into the sub-enclosure formed between linears L2 and L1. The size of the entrance, 0.4m, would seem sufficient only for human entry, not for that of stock.

The gaps and gateways in the ditches vary greatly in width. Some, like that in L8 which is 3.9m wide, were clearly suitable for driving cattle. However the majority, such as that between the two cuts of L9 [w. 0.9m] or the two parts of L4 [w. 0.4m] were clearly only suitable for allowing humans to pass through. Although no post-holes or other such suggestive structural evidence were excavated next to any of the gaps it is nonetheless envisaged that some form of gate existed in each case.

The existence of an infield/outfield system of land management is a possibility. Three enclosed areas are noted; [a] within L1, [b] between L1, L2 and probably L4 [c] between L9, L10 and L5. In contrast, the area to NW of all these enclosures appears to be free of coeval boundaries and could be seen as an outfield or unenclosed pasturage. The area to SE of L8 may either be another unenclosed pasture/field are part of a large enclosed area bounded on two sides by L1 and L8 and, presumably, on the others by further ditches not lying within the excavated area. Ditches L1, L4, L5, L9 and possibly L2 all continue into the unexcavated area to SE and it is likely that other significant elements of the field system lie here also.

In looking at the dating evidence for the enclosure and its related field system a number of obvious facts presented themselves. The fact that one of the decorated ro-tary quern fragments features part of a carved cross demonstrated the Christianity of the inhabitants, while the fact that a rotary disc quern was in use obviously ruled out a prehistoric date. No pottery whatsoever was recovered from the site. Two radiocarbon dates were obtained from the enclosure, a high medieval date from animal bone from the ditch and a late medieval date from charcoal from a pit in the interior. The date towards the very end of the medieval period is surprising, given the absence of pottery on the site but it is interesting that a similar date range was obtained from a sample from the kiln, albeit slightly earlier. This raises the likelihood that the later occupation of the enclosure is contemporary with the kiln.

KilleisK-e3587

41


As regards the economy of the site, the evidence makes a connection with agriculture clear, both in terms stock rearing and cereal cultivation. The system of field boundaries and droveway, with accompanying gateways and gaps strongly suggests a stock-control system the widespread recovery of livestock bone – of cattle, sheep or goat and pig, as well as rabbit – confirm the availability of cattle meat to the inhabitants. Dog bones were also recovered from the main enclosure ditch and from L2, suggesting that they may have been used for stock control, as well as other purposes such as hunting, security and com-panionship. Cat bones, also from the main ditch, show that this species was present and although the possibility that these bones are those of wildcat cannot be ruled out it seems more likely that they were domestic and an aid to control of rodents and as pets. The recovery of a horseshoe and several pieces of horse bone from from the main enclosure ditch indicate that this species was in use here, probably for traction in agricultural work and transport also. Numerically, there were more sheep/goat bones identified than any other species [76], with cattle slightly less [72], followed by horse [30], with small numbers of pig [5], cat [3] and dog [2] and single fragments of deer, bird [unspecified] and rabbit also recovered. The latter three bones may be evidence of hunting, but the numbers are so small that chance occurrence cannot be ruled out.

The four pieces of rotary quern recovered from L1, L4 and L5 are clear evidence of cereal use. None of the four fragments formed part of the same disc, thus constituting the remains of between two and four querns. That the nearby cereal-drying kiln is coeval seems likely given the similarity of the radiocarbon dates from the kiln and the pit in the interior of the enclosure, and so it seems likely that the inhabitants of the enclosure [if indeed they lived there at all] were the same as those utilizing the kiln. Ecological analysis of soil samples from the enclosure and associated ditches identified the seeds of a number of plant types. Cultivated cereals were most common, with the majority coming from a single large pit, cut C80, which yielded 9 oat grains, 4 wheat grains, 3 barley grains and 10 unspeciated cereal grains. From the various fills of the main enclosure ditch were re-covered a further 2 barley grains, 1 grain each of wheat and oats and 10 unspeciated cereal grains. Further cereal grains were found in the fill of droveway ditch L5 [2 unspeciated grains] and post-hole or pit cut C397 in the sub-enclosure [1 wheat grain]. As well as the obvious practice for crop cultivation which these grains imply, a number of weed seeds such as plantains, legumes and various grasses were recovered. These are types of weed which are commonly associated with arable fields and waste ground and they were prob-ably brought to site along with the cereal grains.

Description of kiln and associated featuresThe kiln and three cut features associated with it lay close to each other in the SW ex-tremity of the site in an area where no other archaeological evidence was present. The group consisted of the kiln, a pit, a stakehole and a linear ditch which ran around the other three features from SW through W to NW (Figure 15, Plate 15).


42

Plate15:ViewofkilnandassociatedfeaturesfromNWatKilleisk�

Plate16:ViewofkilnC538atKilleisk�

KilleisK-e3587

43


The kiln was visible prior to the commencement of excavation as an elongated, rough-ly rectangular area with its long axis oriented N – S. On the surface alone at least seven different fills could be seen. The limestone structure of the kiln was discernible on both of its longer sides, although it did not survive above ground level. The suspicion of a roughly circular cupola in the centre was confirmed as contexts were removed and the shape be-came clearer. This was flanked to N and S by open-ended cavities which gave access to the cupola. These cavities flared outwards from the cupola to the ends of the furnace, so that the entire structure had a shape in plan somewhat resembling a bow-tie (Plate 16).

The fills within the structure were numerous, 27 in all, and were disposed in an in-tricate series of overlaps (Figure 16). Unsurprisingly, almost all of the fills displayed signs of burning, oxidatization or exposure to extreme heat in the form of charcoal, burnt or baked clay and fire-reddening. Many of the fills were confined to either the N or the S cavity, having obviously been raked out on that side from the cupola. 9 deposits were confined to the N funnel, 11 to the S funnel and 6 to the interior of the cupola. One of

C.370

C.371

C.372


Killeisk 1E3587SW facing section of C.372

C.371

C.370

C.372

C.347

C.346

C.348

Killeisk E3587SE facing section of C.348

0 50 cm10 cm

Linear 7

Linear 7

Linear 7

Figure16:SectionsofkilnC358andpitC364atKilleisk�

C.370

C.371

C.372


Killeisk 1E3587SW facing section of C.372

C.371

C.370

C.372

C.347

C.346

C.348

Killeisk E3587SE facing section of C.348

0 50 cm10 cm

Linear 7

Linear 7

Linear 7


44

the uppermost fills visible before excavation began, C373, during environmental analysis yielded 29 cereal grains, 15 of which were of wheat and the other 14 unspecified. The context immediately below this, C3374 yielded a further 10 grains, 3 of wheat and un-specified. C398 yielded a single grain of wheat. The above would suggest the kiln served to dry cereal.

Five of the fills [C385, C408, C423, C424 & C515] consisted mostly or substantially of crushed limestone grit. The lowest of these, C515, lay directly upon subsoil and was the only deposit present in all three parts of the kiln and formed a very hard, very thin layer of light yellowish grey crushed limestone with occasional inclusions of charcoal and burnt clay. It is likely that this results from dissolved limestone carbonate percolating through the porous deposits and accumulating on the impermeable natural subsoil. However, the other deposits of limestone cannot be explained as the result of natural processes in this way and the possibility that limestone was being deliberately burnt in this kiln must be considered despite its small size and the fact that it seems to have been used for drying cereal [see above].

No artefactual material was recovered from the kiln but, surprisingly, pieces of cat-tle tooth were recovered from two contexts [C357 & C424]. One deposit in the cupola, C453, was of particular interest in that it constituted a number of pieces of carbonized roundwood and thus appeared to represent the remains of an in-situ firing of the kiln

Plate17:ViewofcharcoalincupolaofkilnC538atKilleisk

C.3

63

C.3

76

C.3

64

Kill

eisk

1E3

587

East

faci

ng se

ctio

n of

C.3

64 C.4

26C

.425

C.3

73C

.424 C

.453

C.3

74C

.375 C.3

85C

.423

C.3

82 C.4

08C

.387

C.4

27C

.357

C.3

98C

.383

C.5

13C

.515

C.4

62C

.514

C.4

50C

.453

C.4

61C

.358

C.3

58

Kill

eisk

E3

587

Wes

t fac

ing

profi

le o

f C.3

58

050

cm

10 c

m

KilleisK-e3587

45


Figu

re1

7:S

ectio

nso

fL7

C100

�


46

(Plate 17). The roundwood was identified as Birch [Betula sp.] and a radiocarbon sample of this gave a dating range of AD 1420–1617 (UB–15090).

To W of the kiln with its alignment turning to enclose it on this side, was linear ditch cut C100 [cumulative l. 25.5m, w. 1.2m, d. 0.35m]. Its S terminal lay 7.5m due S of the kiln, from which point it ran 7.2m in a WNW direction before turning and running N for a further 16m, at which point it turned and ran a further 1.5m to ENE, where it terminated. The shape of this cut varied a fair deal along its length (Figure 17). The ditch contained three fills, all of them apparently sterile.

A metre to the W of the furnace and 3m E of the ditch lay a large pit, cut C364 [l. 1.83m, w. 1.2m, d. 0.3m]. This was sub-rectangular in plan with vertical sides and a sub-rectangular base which was concave in plan (see Figure 16). There were two fills [C363 and C376], both very rich in charcoal. The lower of these, C376, contained several frag-ments of animal bone and tooth which were identified as those of sheep or goat. Hazel charcoal from this fill gave a radiocarbon dating range of AD 1161–1262 (UB–15089), which would seem to suggest that it is not coeval with the kiln.

Some 0.4m to N of the kiln lay a single small cut, C344 [diam. 0.35m, d. 0.13m], the fill of which contained some charcoal and which could be a post-hole or small pit.

Interpretation of the kiln and associated features.The function of the kiln is unclear. During initial cleaning of the feature before excava-tion a number of carbonized cereal grains were recovered, which would suggest use as a corn-drying kiln. Environmental samples from contexts within the kiln also yielded further cereal grains, of which those identifiable were of wheat. However, the presence within the kiln of a number of deposits apparently made up almost entirely of crushed limestone (C424, C385, C423) might suggest use as a lime-burning kiln. That the kiln served both uses, perhaps at different stages in its use-life, is a possibility.

The numerous and diverse range of function related deposits within the kiln suggest a sustained period of use, even without invoking the fact that it was probably periodi-cally cleared out and the waste disposed of elsewhere. It is certain that the kiln originally extended some height above ground level and was roofed in some fashion. The fact that ash and charcoal did not occur at all in cut C100, the nearby ditch, suggest that this was not open in the period of kiln use. It may be that the ditch had been excavated, a palisade or some such windbreak erected within it and then backfilled before the kiln was put into use. The absence of stake or post-holes within the ditch creates something of a problem in terms of suggesting how such a windbreak might have been erected.

The overall provisional interpretation of the kiln and associated features was as a work-area relating to the use of the centrally located kiln for cereal drying and/or lime burning, with ditch C100 forming part of a windbreak shielding the kiln from the pre-vailing W and SW winds and with large pit C364 and small pit/post-hole C344 likewise being involved in some capacity. However, although the kiln and ditch are still considered coeval and the small pit may be so, the large pit now seems unlikely to be contemporary with the other features. This is on account of the disparity between the dating range for

KilleisK-e3587

47


the radiocarbon sample from within the pit [hazel charcoal AD 1161–1262 (UB–15089)] and that from the kiln [birch charcoal AD1420–1617 (UB–15090)]. As neither hazel or birch are particularly long living trees it seems unlikely that this disparity results from the ‘old wood effect’ and it therefore seems likely that the pit represents an earlier high medieval phase of activity.

Only further dating of samples from these features would clarify this matter, which is not made any clearer by comparison of the cereal grains recovered from within them. In the kiln, of the 39 grains recovered 18 were identified as wheat and the other 21 un-specified. Only two grains were recovered from pit C364, and these were both identi-fied as oat. The different species emphasis might be taken to support the suggestion of a separate phase. However, the strong possibility remains that some of the 21 unspeciated cereal grains from the kiln may be oat, in which the comparison might suggest the two features are contemporary. This spatial arrangement is paralleled at nearby Derrybane 1 E3585, also excavated as part of this project. At this site a dumb-bell-shaped kiln dating to AD 1181–1269 (UB–15040) also had a large oval pit directly adjacent to it. Here, as at Killeisk, oat and wheat were recovered from the kiln and pit, but unlike Killeisk oat predominated. The date from the Derrybane kiln is coeval with the date from the pit beside the kiln at Killeisk. But it is earlier than the high medieval date from the ditch of the enclosure.

Lithic findsThe lithic artefacts were examined by Farina Sternke (Appendix 4). A single hone stone (E3587:1:1) was recovered from the surface of the site (Plate 18).

Plate18:HonestoneE3587:1:1�


48

Rotary quernstones The quern stones were examined by Anne Carey (Appendix 5). Three main types of ro-tary querns, beehive, disc and pot querns, have been identified by Seamus Caulfield. The rotary querns from Killeisk 1 are all Disc C querns and are fragmentary. The disc quern was first introduced to Ireland in the first or second century A.D. and continued in use until modern times. It has a widespread distribution. This type of rotary quern consists of two flat, thin circular discs of large diameter. The two decorated examples from Kil-leisk are interesting. The spiral design on quern E3587:125:1 is interesting and unusual (see Plate 14). The pattern on quern E3687:130:1 is likely to form part of a cross shape, with one of the arms surrounding the handle hole (see Plate 13). Decorated stones are known throughout the country with dates ranging from the thirteenth to the seventeenth centuries.

Plant remainsThe plant remains were identified by Penny Johnston (Appendix 6). The charred plant remains included the remains of both cultivated and wild plants. Cereals were the most common cultivated plant. At least one half of the total grains were not identifiable to type. Of the identifiable remains, wheat was the most common cereal type recovered, followed by oat and then a small quantity of barley. The plant remains from the kiln were almost exclusively wheat. The majority of the oats were recovered from the pit C80 located within the enclosure. Comparative assemblages from ringfort sites are generally unlike those obtained from Killeisk. The wheat-rich assemblage is generally more com-mon in deposits from the later medieval period.

Animal bone reportThe animal bone was identified by Margaret McCarthy (Appendix 7). In all 1076 bones were examined, over 90 % were recovered from the ditch fills. The sample is dominated by the remains of cattle and sheep/goat and to a lesser extent by horse. The cattle and sheep/goat account for two thirds of the identified assemblage. Single fragments of pig, dog, rabbit, deer and bird bone were also identified from the ditch fills. The animal bone as-semblage suggests that the bulk of the waste generated by the occupants of the enclosure was discarded into the enclosure ditch.

Geophysical surveyA geophysical survey was undertaken by Earthsound Archaeological Geophysics (Ap-pendix 8). A continuation of the ditch of the oval-shaped enclosure was identified by geo-physical survey at Killeisk (Figure 18). The mapped components, excavated and surveyed, of the ditched enclosure measured 95 m NW/SE by 41 m NE/SW and enclosed an area of 0.335 hectares. A continuation of the part of the drove-way was also identified along with a number of other possible ditches and a relic field boundary.

1944

54

1944

54

1946

30

1946

30

179445

179445

179550

179550

Encl

osur

e

Dro

vew

ay

Kiln

±

010

0m

Gra

diom

eter

inte

rpre

tatio

nD

itch

Hea

rth o

r bur

ning

Indu

stry

Pits

Plou

gh fu

rrow

s

Sub-

encl

osur

e

Bron

ze A

gepi

ts 2

Bron

ze A

gepi

ts 1

KilleisK-e3587

49


Figu

re1

8:P

lan

ofg

eoph

ysic

als

how

ing

exte

nto

fenc

losu

rea

tKill

eisk

�


50

CharcoalThe charcoal was identified for radiocarbon dating by Mary Dillon.

Radiocarbon datesRadiocarbon analysis was carried out by the 14 Chrono Centre in Queen’s University Belfast. Dates were calibrated using Calib Rev5.0.2 (©1986-2005 M.Stuiver & P.J. Re-imer) and in conjunction with Stuiver & Reimer 1993 and Reimer et al. 2004.

Lab code

Context Material Un-cali-brated date

δ 13 C

1 sigma calibration 2 sigma calibration

UB 15088

83 Hazel charcoal from pit C.80

304 +/- 18 -28.3 AD 1524-1558 1564-1570 1631-1643

AD 1516-1596 1618-1647

UB 15089

376 Hazel charcoal from pit C.364

832 +/- 29 -29.5 AD 1184-1229 1231-1252

AD 1161-1262

UB 15090

453 Birch charcoal from kiln C.358

428 +/- 33 -28.7 AD 1434-1474 AD 1420-1515 1600-1617

UB 15091

218 Oak charcoal pit C.219

3754 +/- 21 -29.8 BC 2200-2157 2155-2140

BC 2276-2254 2227-2224 2209-2128 2087-2047

UB 15092

463 Pomoideae charcoal from pit C.464

3782 +/- 22 -24.1 BC 2277-2252 2228-2222 2210-2195 2173-2145

BC 2286-2246 2224-2140

UB 15737

104 Animal bone from enclosure ditch C.4

614+/-35 -25.6 AD 1288-1328 1341-1369 1380-1395

AD 1292–1404

Table2Radiocarbondates

1912

32

1912

32

2082

32

2082

32

178059

178059

184059

184059¢

02.

55 Km

Cre

mat

ion

(2)

Pit g

roup

(3)

Burn

t spr

ead

(2)

Link

ards

tow

n bu

rial (

2)M

ound

(6)

Settl

emen

t site

(9)

Barro

w (1

1)C

airn

(1)

Stan

ding

sto

ne (9

)Fu

lach

t Fia

dh (1

5)M

egal

ithic

tom

b (3

)

KilleisK-e3587

51


Figu

re1

9:P

rehi

stor

ics

ites

ona

ndin

the

envi

rons

ofN

7Ca

stle

tow

nto

Nen

agh�


52

8 DiscussionTo begin with the earliest activity on site, both Pit Groups 1 and 2, on the basis of the radiocarbon determinations (2276-2047 BC and 2286-2140 BC respectively) clearly date to the same period, the Chalcolithic (Early Bronze Age). The prehistoric dates for these two clusters of cut features were unsurprising given that, despite the absence of finds from them, they conform to a well-established and recognizable pattern often encountered during excavations. Unfortunately, to say that they are easily recognizable and frequently encountered is by no means to say that they are well understood.

Both groups share a similar arrangement in that the majority of the cut features clus-ter about a large central pit which has possible superstructural evidence in the form of a number of stakeholes cut into the sides and base of the pit and, in one example, around its edge. Another commonly occurring feature and one that is shared by both Killeisk groups, is that the fills of many of the cuts in both cases are composed of a mixture of charcoal-enriched soil and fragmented stone. The similarity of this material with that forming the mounds of fulachta fiadh is often noted, and the general form and features of the central pits mentioned above are such that they would, if encountered below or adja-cent to a fulacht fiadh mound, be interpreted as troughs used to hold water and involved in whatever processes utilizing heated stones and water were taking place at the site. On that basis it is necessary to expand the definition of fulachta fiadh to exclude requirements of ‘classic horse-shoe-shaped’ mounds and – in the absence of good indicative evidence – to keep an entirely open mind on the processes pursued on these sites. As is often the case, the cuts of both pit groups at Killeisk in no case displayed any evidence for in-situ burning and raise the possibility that their original purpose was not directly related either to the burnt fill or to the central pit with it’s stakeholes.

The possibility of early medieval activity on or in the environs of the site at Killeisk was raised by a hone stone recovered loose on the surface by the director in the NW part of the site. Specialist analysis considers this to be diagnostically early medieval in date (Appendix 4), although otherwise there is nothing to indicate a date earlier than high medieval for any of the archaeological features of historical date on the site.

To turn to the medieval enclosure and associated field system, the absence of compa-rable high and late medieval sites in the published literature makes meaningful compari-son difficult. None of the standard texts on medieval Ireland contain any example of a site of similar morphology to that at Killeisk. It must also be acknowledged that the very partial nature of excavation at Killeisk may militate against a proper discussion. As out-lined in the interpretation above, less than one third of the area of the main enclosure fell within the road-take and we can only guess at what may lie within the unexcavated area. In the case of the field system we cannot know what proportion was excavated, except to say that a significant part also lies outside the excavated area. Nonetheless, there are some observations and suggestions which can be made.

Firstly, the shape of the enclosure makes it clear that it is too far beyond any approxi-mation of circularity to be considered a ringfort of any type. This is not surprising given

1920

00

1920

00

2080

00

2080

00

177100

177100

184800

184800¢

02.

55 Km

CAS

TLE

CH

UR

CH

ECC

LES

IAS

TIC

AL

ENC

LOS

UR

E

FOR

TIFI

ED

HO

USE

KILN

- C

OR

N-D

RYI

NG

ME

TALW

OR

KIN

G S

ITE

MO

ATE

D S

ITE

REL

IGIO

US

HO

US

E

RIN

GFO

RT

RIT

UA

L SI

TE

SETT

LEM

EN

T D

ESER

TED

SOU

TER

RA

IN

KilleisK-e3587

53


Figu

re2

0:M

edie

vals

ites

ona

ndin

the

envi

rons

ofN

7Ca

stle

tow

nto

Nen

agh�


54

the late dates for the site, but prior to the receipt of the dates, given the aceramic nature of the site and the fact that North Munster is a zone of high ringfort density (Stout 1997, 82), it had been necessary to consider this possibility. Using Stout’s method of calcula-tion (ibid. 14) the Killeisk enclosure has a circularity index of 0.45 and cannot really be considered even oval, not to speak of round.

In rejecting the possibility of the enclosure being a ringfort or related site type, we stray into the area of the postulated ethnicity of the inhabitants. Killeisk lies in an area which in the late Middle Ages was the subject of conflict as the Gaelic Resurgence took place. As early as the middle of the 13th century the Anglo-Norman Butlers had lost much of north Tipperary to the Irish (Loeber 2001, 279). Presumably this was to the Uí Chin-néide (O’Kennedys), the main Irish sept in the area. The absence of pottery on the site at Killeisk and the fact that the site does not conform to any of the known settlement types of the Anglo-Normans (whose archaeology is far better understood than that of the na-tive Irish) may suggest this was a Gaelic site. Given how commonly pottery occurs on, for instance, medieval moated sites – an Anglo-Norman type occurring commonly in North Tipperary – the complete absence of such at Killeisk seems meaningful. A further element relating to this matter is the cereal assemblage from the enclosure and kiln.

It seems on the basis of the current dating evidence that the later phase of occupation of the enclosure and kiln are contemporary at Killeisk. In their paper on drying kilns Monk and Kelleher (2005, 88) note that wheat is generally only significant where it is the dominant cereal type. Although only 17 of 39 grains from the Killeisk kiln were identi-fied to species, all of these were wheat. As Johnston notes in the specialist report on plant remains from Killeisk (Appendix 6), McClatchie (2003, 398) suggests that cereal prefer-ence may have been used as a badge of ethnicity in Ireland the medieval period, with wheat the choice of the English settlers. However, this assertion seems difficult to accept as fact in the absence of a large body of statistical evidence or contemporary textual mate-rial to back it up. In any case, the Killeisk evidence is not so clear as the remains from the kiln alone might suggest, with oats and barley recovered from within the enclosure and from a pit directly adjacent to the kiln, albeit again in small amounts.

An intriguing aspect of the Killeisk kiln is that it does not fit into any of the plan forms listed and described by Monk and Kelleher; figure of eight, dumb-bell, keyhole and angled L or comma-shaped. It is clearly, however, closer to the keyhole type than the others, perhaps being best described as double-keyhole in shape. It has the same regularly flared flue and circular chamber as the keyhole type but has two opposed flues, one on either side of the chamber. None of the kilns of any type listed by Monk and Kelleher fea-ture the provision of two flues. The similarity between the Killeisk kiln and the keyhole type in general is supported by similar dating. Monk and Kelleher note that excavated examples of the keyhole type tend to date to the late medieval and early post-medieval period, a range into which the dates of AD 1420-1617 for the Killeisk kiln fit comfortably. The latter authors also note the “relatively significant presence of wheat” (ibid. 105) in sev-eral keyhole-type kilns in comparison with those of the early medieval period, a pattern also shared by the Killeisk example.

KilleisK-e3587

55


The size and extent of the probable windbreak feature running around the kiln at Kil-leisk from S through W to NW is also more considerable than anything noted by Monk and Kelleher, although despite the ditch cut being substantial, the absence of any identifi-able post or stakeholes within it remains somewhat puzzling.

A further unparalleled feature of the Killeisk kiln is that here the central chamber, which in drying kilns is generally accepted as the drying chamber and is not used for fire-setting, had been used to house a fire on at least one occasion, as the discovery of in-situ carbonized round-wood pieces here demonstrated. One method of explaining this is to suggest that the above-surface portion of the kiln, which did not survive, provided for a drying chamber of a design hitherto unrecorded. Perhaps more likely, and certainly sim-pler, is the suggestion that this recorded instance of the central chamber for fire-setting is aberrant and contrary to the original design of the kiln. That this atypical use was uncon-nected with the drying of cereal seems likely. It may, in fact, fit in with the possibility that the kiln was also used at some stage – probably secondarily – to burn lime, a possibility noted in the interpretation and for which there is some contextual evidence.

It is interesting also, although it’s significance remains unclear, that carbonized grain was only recovered from two of the topmost contexts of the kiln at Killiesk and that none was found in any of the other two dozen or so deposits in a stratigraphically inferior position to those two contexts, this despite many or most of those deposits being clearly related to pyrotechnical activity.

An interesting possibility is that the kiln is significantly older than the date range of AD 1420-1617 (obtained from the carbonized round-wood discussed above) would sug-gest. If, as suggested, the wood for which this date was obtained resulted from an aberrant use of the kiln then that in turn might suggest a late use or re-use of the kiln in a period when it was no longer being utilized for its original purpose of cereal drying. Such a hy-pothesis would certainly help explain a date of AD 1161-1262 obtained for charcoal from a large pit directly adjacent to the kiln and which also yielded two carbonized oat grains. This spatial arrangement is paralleled at nearby Derrybane 1 E3585, also excavated as part of this project. At this site a dumb-bell-shaped kiln dating to AD 1181–1269 (UB–15040) also had a large oval pit directly adjacent to it. Here, as at Killeisk, oat and wheat were recovered from the kiln and pit, but unlike Killeisk oat predominated. The date from the Derrybane kiln is very similar to that from the pit beside the kiln at Killeisk and this might support the suggested explanation of the later date from the Killeisk kiln as result-ing from a late re-use episode.


56

9 ReferencesBrindley, A.L. and Lanting, J.N. (1990) ‘The dating of fulachta fiadh’, in Buckley, V.

(ed.) Burnt Offerings. International contributions to burnt mound archaeology, 55-56. Dublin, Wordwell.

Farrelly, J., and O’Brien, C. (2002) Archaeological Inventory of County Tipperary Vol. 1 - North Tipperary, The Stationery Office Dublin.

Frazer, W. (2009) Archaeological Assessment Report Nenagh NRA Service Area Park townland, North Co. Tipperary and Roshedrid and Clynoe townlands, Co. Offaly 09E122. Margaret Gowan & Co. Ltd. Unpublished report.

Gardiner, M.J. and Radford,T. (1980) Soil Associations of Ireland and Their Land Use Potential. Dublin, An Foras Talúntais.

Loeber, R. (2001) ‘An Architectural History of Gaelic Castles and Settlement’, in Duffy, D., Edwards, D. and Fitzpatrick, E. eds. Gaelic Ireland c.1250-c.1650: Land, Lordship & Settlement Four Courts Press, Dublin.

McClatchie, M. (2003) ‘Section 12- The plant remains’, pp. 391-413 in Cleary, R.M. and Hurley, M.F. (eds.) Cork City Excavations 1984-2000 Cork: Cork City Council.

McLaughlin, M. and Conran, S. (2008) ‘The emerging Iron Age of South Munster’ in Seanda, Issue 3, 51-53. Dublin.

Monk, M. and Kelleher, E. (2005) An assessment of the archaeological evidence for Irish corn-drying kilns in the light of the results of archaeological experiments and archaeobotanical studies, Journal of Irish Archaeology Vol. XIV, 77-114.

McQueen, A. (2008) Architectural Survey N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme Contract 1. Unpublished Eachtra Archaeological Projects Report.

National Inventory of Architectural Heritage (2006) An Introduction to the Architectural Heritage of North Tipperary. Government of Ireland.

O’Brien, C. (1997) Archaeological Inventory of County Offaly, The Stationery Office, Dublin.

KilleisK-e3587

57


O’Conor, K.D. (1998) The Archaeology of Medieval Rural Settlement in Ireland, Discovery Programme Monographs No 3, Discovery Programme/Royal Irish Academy Dublin.

O’Flannagan, Rev.M (Typescript) (1930). Ordnance Survey Name Books for Kings County. (Aghanoon to Durrow). Bray.

O’Kelly, M.J. (1954) Excavations and experiments in Irish cooking places. Journal of the Royal Society of Antiquaries of Ireland, Vol 84.

Ó Néill, J. (2003/2004) Lapidibus in igne calefactis coquebatur: The historical burnt mound “tradition”, Journal of Irish Archaeology Vol. XII & XIII.

Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C., Blackwell, P.G., Buck, C.E., Burr, G., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hughen, K.A., Kromer, B., McCormac, F.G., Manning, S., Bronk Ramsey, C., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J. and Weyhenmeyer, C.E. (2004) ‘IntCal04 Terrestrial Radiocarbon Age Calibration, 0–26 Cal Kyr BP’, Radiocarbon 46, 1029-1058.

Roycroft, N. (2006) A theory on Boiled Bull and Burnt Mounds, Seanda Issue 1, 38-39, National Road Authority, Dublin.

Roycroft, N. (2008) Before, during and after the Kingdom of Ely, Seanda, Issue 3. 34-35, National Road Authority, Dublin.

Sternke, F. (2009) More than meets the eye; an appraisal of the lithic assemblages from the route of the N7 Castletown to Nenagh (Contract 1). Seanda. Issue 4, 30-31,National Road Authority, Dublin.

Stout, M. 1997 The Irish Ringfort Four Courts Press, Dublin.

Stuiver, M., and Reimer, P.J. (1993) ‘Extended (super 14) C data base and revised CALIB 3.0 (super 14) C age calibration program’, Radiocarbon 35, 215-230.

Stout, M. (1997) The Irish Ringfort. Dublin, Four Courts Press.

Taylor, K. (2008) ‘At home and on the road: two Iron Age sites in County Tipperary’ in Seanda, Issue 3, 54-55. Dublin.

Tourunen, A. (2008) Fauna and fulachta fiadh: animal bones from burnt mounds on the N9/N10 Carlow Bypass. In J. O’Sullivan and M. Stanley (eds.), Roads,


58

Rediscovery and Research. Archaeology and the National Roads Authority Monograh Series No. 5. Wordwell.

Valuation Office. (1848) Valuation Office House Books, vol. 1592, Parish of Cullenivane. Public Records Office.

Woodman, P.C. (2000) ‘Hammers and Shoeboxes: New Agendas for Prehistory’., pp. 1 -10 in Desmond, A., Johnson, G., McCarthy, M., Sheehan, J. and Shee Twohig, E. New Agendas in Irish Prehistory. Papers in commemoration of Liz Anderson. Bray, Wordwell.

KilleisK-e3587

59


Appendix 1 Stratigraphic Index

Please see attached CD.


60

App

endi

x 2

Stra

tigra

phic

Mat

rix

KilleisK-e3587

61


Appendix 3 Groups and sub-groupsThe archaeological features excavated at Killeisk can be mostly accommodated in four principal groups. These groups are assembled primarily on the basis of a known or likely functional and chronological link existing between the contexts within each group and, more closely, within each sub-group:

• Group 1, consisting of the main elliptical enclosing ditch L1 (of which only the N two fifths were exposed and excavated within the CPO area), along with a number of associated linear field boundaries (L2, L4, L5, L8, L9 & L10) and those non-linear features – cuts and deposits – judged to be associated with the aforementioned en-closure and field boundaries.

• Group 2, consisting of the kiln excavated near the SW extremity of the site along with associated linear ditch (possibly a windbreak) L7 and several peripheral associ-ated features.

• Group 3, consisting of a cluster of cut features centring on grid point [70,80] and extending in a radius of approximately 12m around than point in each direction. These cuts frequently contained material (‘burnt mound’ or fulacht type burnt stone fragments in a matrix of charcoal-rich black/grey silt) which indicated pyrotechnical activity in the vicinity.

• Group 4, consisting of a cluster of cut features and spreads mostly, though not com-pletely, contained with the limits of 10m grid square [120,120]. As with Group 3, many of these features contained material resulting from pyrotechnical activity.

• The following abbreviations are used below in describing the contextual evidence: • d. = depth, diam. = diameter, l. = length, w. = width.

Linear Group Type Cut nos.1 1 Enclosure ditch C4, C7, C24, C39, C58, C123, C1352 1 Sub-enclosure ditch C9, C39, C903 Modern field boundary4 1 North Droveway ditch C11, C545 1 South Droveway ditch C136 Modern field boundary7 2 Kiln windbreaker C1008 1 Field boundary C24, C929 1 ?Field boundary C102, C52210 1 ?Field boundary C112

Table1Linearsandassociatedcutnumbers

Group 1Group 1 consisted of elliptical enclosing ditch L1, associated linear field boundaries (L2, L4, L5, L8, L9 & L10) and associated non-linear features (for cut and deposit numbers corresponding to each linear see table 1). These are considered to form a valid grouping on the basis of stratigraphic relationships (almost exclusively of the horizontal type) and on


62

the basis of the similarity of the type of finds and samples recovered. All of these suggest a significant level of contemporaeniety and related function. Group 1 is further divided into the following seven Sub-groups.

Group 1, Sub-group 1: The main enclosure ditch L1, its fills and cuts and any other features within those cuts. Consists of 34 contexts: C1, C3, C4, C6, C7, C35-7, C41, C47, C51, C58, C67, C69, C77, C81, C85-8, C95, C98, C104-5, C110, C123, C135-6, C144, C149, C155, C162-3 & C166.

Group 1, Sub-group 2: Interrupted ditch L2 which runs roughly NNW-SSE from the W side of the main enclosure to L4, the more northerly of two parallel linear ditches. In so doing, L2 creates a sub-enclosure between itself and L1. Consists of 8 contexts: C8-9, C18, C27, C39, C46, C90 & C91.

Group 1, Sub-group 3: Parallel and co-terminous ditches L4 and L5, forming a drove-way aligned running roughly W-E. Consists of 12 contexts: C10-3, C54-6, C60-2, C68 & C117.

Group 1, Sub-group 4: Two partially truncated linear features, L9 and L10, which were cut through by a post-medieval field boundary [L6] but which were originally prob-ably connected. Lies immediately S of G1, SG3. Consists of 9 contexts: C101-2, C107-8, C111-2 and C520-2.

Group 1, Sub-group 5: Interrupted linear ditch L8, one end of which terminates at, and is connected to, the NW part of enclosure L1. It runs NNE from that point to the edge of excavation. Consists of 5 contexts: C24-5, C92-3 & C158.

Group 1, Sub-group 6: The various features excavated within the main enclosure ditch L1 and deemed to be related to it. Many of these were deemed on consideration to be non-archaeological. Those contexts deemed probably archaeological were 24 in number: C72, C76, C79, C80, C83, C172-3, C184-5, C188-90, C209-13, C227-9, C251-2 and C260-1.

Group 1, Sub-group 7: The various features excavated within the sub-enclosure lying between interrupted ditch L2 and main enclosure ditch L1. Of the 9 cut features exca-vated in this area 4 were considered archaeological, consisting of the following contexts: C394-7, C400-1, C435-6 and C416-7. Also within this area were two discrete areas of possible metalled surface, C384 and C386.

Description of Group 1, Sub-group 1

The uppermost fill over the majority of the excavated area of enclosure ditch L1 was C1, a widespread layer of firm brown silt which yielded a good amount of animal bone, as well as an iron nail [Find no. 2]. Charcoal occurred in C1, but only in certain parts of its ex-tent. The depth of this deposit within the ditch varied considerably from 0.15m to 0.50m.

The only other fill occurring in the uppermost stratigraphic position along with C1 was C163 [d. 0.14m], an apparently sterile deposit of soft yellowish grey silty clay restrict-ed to the NW part of the ditch. This lay over C162, a bluish grey silt which in turn lay above C36 [d. 0.15m], a brownish grey clay containing a number of pieces of animal bone. C36 lay on the bottom of the main ditch cut, C4, for a description of which see below.

KilleisK-e3587

63


The main uppermost fill, C1, lay directly above six other fills – soft dark grey clayey silt C155 [d. 0.2m], which contained animal bone, an iron horseshoe [Find no. 4] and iron ?nails [Find no. 5], soft brownish grey clayey silt C44 [d. 0.11m] with a few animal bone fragments, soft yellowish grey sandy clay C95 [d. 0.32m], a deposit of large stone fragments C88, which yielded a piece of a rotary quern [Find no. 3], soft yellowish brown clayey silt C110 [d. 0.12m] and firm pebbly silt C67 [d. 0.4m]. The first of these six, C155, lay upon C149 [d. 28m], a soft light grey clayey silt which contained a number of animal bone fragments and an unidentified iron fragment [Find no. 7]. Under this was C77 [d. 0.25m], a soft dark grey silty clay containing a single bone fragment. This latter lay over two other fills, C3, which is described below, and firm yellowish orange silty clay C85 [d. 0.3m]. Below C85 was C86 [d. 0.15m], a soft light grey silty clay which may be overcut natural subsoil or have lain directly upon main ditch cut C4 (see below). Like C77, C44 also lay upon C3, a layer of silty clay which occurred along the inner edge of the cut of the ditch throughout the majority of its exposed length and which is thought to be material slumped from an internal bank associated with the ditch. The colour of this deposit varied considerably but tended to be grey, and varied in depth from 0.07m to 0.55m in depth. Charcoal flecking occurred locally in one or two places within C3 and a single animal bone fragment was recovered from it.

Contexts C95 and C88 lay directly upon C6 [d. 0.2 to 0.5m], a widespread layer of firm mid to dark grey silty/sandy clay which contained many animal bone fragments and frequent charcoal flecking and which is thus considered likely to be related to human activity associated with the enclosure. C110 also lay upon C6, but was separated from it stratigraphically by cut C123 [w. 0.74m, d. 0.12, but probably truncated], which occurred along the W and SW sides of the enclosure cut. This flat-bottomed cut is thought to be a recut resulting from cleaning out of the ditch following natural silting. C67, the last of the 6 contexts lying directly under C1, lay in turn above C87 [d. 0.18m], a firm brown silt which was itself over stiff brownish grey silt C105 [d. 0.21m]. Below this was firm mid grey clayey silt C69 [d. 0.23m], which was also overlapped by C47, the uppermost fill of the SW cut of linear ditch L8 (for which see Group 1, Sub-group 5). Like C95, C88 and C123, context C69 lay atop bone and charcoal-rich layer C6.

C6 in turn itself lay over three contexts – fills C98 and C51 and cut C7. C98 [d. 0.08m] was a yellowish grey sandy clay with some charcoal flecking, which lay directly upon the main ditch cut C4 (see below). Cut C7 [w. 0.84m, d. 0.18m] was found in section at two points along the W side of the enclosure ditch. At one point it was cut into C37 and at another all the way down through the original cut C4 and into the natural subsoil. Like C123, it was flat-bottomed and considered to be a recut resulting from cleaning or re-digging the ditch due to natural silting, and although both C123 and C7 occurred on the W side of the enclosure, they were not the same cut, being separated stratigraphically by widespread fill C6, which fills C7 but into which C123 is cut. C37 [d. 0.5m], into which C7 was cut, was a soft orange clayey silt. It lay upon internal bank slippage C3 (see above for description).


64

The third context underlying C6 was C51 [d. 0.24m], a firm dark grey silty clay which lay within cut C58 [w. 0.62m, d. 0.26m] at the NW extent of the enclosure ditch. This cut only existed in this area and was a subsidiary cut running parallel with and both adjacent and exterior to main enclosure cut C4, traceable for a distance of 2.4m in that part of the enclosure ditch where it intersects with linear field-boundary L8. The purpose of this extra cut in this area is unknown. It is likely to be coeval with or later than main cut C4, but this is not proven stratigraphically, with both cut directly into natural subsoil and the disturbed nature of the fills in section at this point.

Directly below C3 were four contexts – C81 [d. 0.14m], a soft light brown silty clay containing charcoal flecking and a single piece of animal bone, C166 [d. 0.18m], a stiff mid grey clay, C35 [d. 0.55m], a firm light brown silty sand, and C104 [d. 0.39m], a soft dark grey clayey silt containing numerous fragments of animal bone occurring in the N part of the enclosure ditch. All four of the preceding contexts lay directly upon the main enclosure cut C4, with C104 also overlying C136 (see below).

The excavated portion of the main cut of the enclosure ditch, C4, had an overall length of c.120m, with a width which varied from 1.4m up to 2.5m and a depth between 0.16m and 0.72m. In general, it was wider and shallower on its SW side and narrower and deeper on its NE side. On the wider, shallower SW side the slope of the sides was cor-respondingly gentler and the base flatter than on the NE side.

The degree of truncation is difficult to assess, but appears to be quite considerable where very shallow at the northernmost part of the cut, adjacent to where it intersects with the cut of field boundary L8. The paucity of features excavated on the interior of the enclosure might also suggest a high level of truncation (see Group 1, Subgroup 6).

On the N part of the enclosure a deep pit had been excavated into the base of C4. This pit had two fills, the uppermost being C136 [d. 0.84m], a soft mid grey silty clay which was not confined to the cut but lay above it in the immediately surrounding area of C4. The lower fill was C144 [d. 0.35m], a soft brownish grey silty clay containing a large piece of animal longbone and pieces of semi-decayed wood [probably tree root]. A number of medium and large stones lay at the base of the pit and may have been deliberately placed. The cut itself, C135 [diam. 0.54, d. 0.7m] tapered slightly from top to base and had a fairly flat bottom. The excavation of this feature was problematic due to water-logging and col-lapsing sides, and recording was impeded by these factors also. The purpose of the pit is unclear, but a form of well or sump might be suggested.

Analysis of the animal bone recovered from the various fills of the enclosure ditch has identified those of cattle, horse, sheep/goat, pig, dog, rabbit and some type of bird.

Provisional interpretation of Group 1, Sub-group1

This sub-group is interpreted as an enclosure ditch which originally had an internal bank and which was coeval with habitation and agricultural activity either within the enclosure or elsewhere in the immediate vicinity, as illustrated by several archaeologically rich fills within the ditch, particularly C6 which contained frequent pieces of animal bone and strong charcoal flecking throughout.

KilleisK-e3587

65



Linear feature L2 is discontinuous, consisting of three cuts arranged end-to-end with narrow gaps between. Its two ends terminated in relation to two other linear features in Group 1; The S end stopped some 0.3m short of L4, one of a pair of parallel ditches forming a droveway, while its N end intersected with the W side of main enclosure ditch L1 and does not continue any further. From its southern end L2 ran NNE for a distance of 19m before turning ENE and continuing a further 3.5m to intersect with L1. At both ends the alignment of L2 was perpendicular to the linear features it respected (in the case of L4) or intersected (in the case of L1). L2 formed a sub-enclosure of the area between itself and the main enclosure L1.

The S cut of L2 [C9] had three fills, the uppermost of which was C18 [d. 0.10m], a soft dark grey sandy clay with some charcoal flecking from which a piece of dog bone was recovered. Below this was C27 [d. 0.2m], a stiff light grey sandy clay which yielded a single sample of cattle bone. The bottom fill of the S cut was a firm yellowish grey silty clay, C8 [d. 0.19m], which also yielded a piece of cattle bone. The cut, C9 [l. 6.65m but truncated at N by post-medieval field drain L3, [w. 0.57m, d. 0.26m], varied from round to flat-based in different places and had moderately-sloped concave/flat sides.

The central cut of L2 had only a single cut and fill. Fill C91 [d. 0.27m] was a firm mid grey clayey silt, while cut C90 [l. 10.8m, w. 0.42m, d. 0.27m] had moderate to steep sides and a rounded base.

The N cut of L8 had two fills. The uppermost of these was C1 (see Subgroup 1 above), the uppermost main fill of enclosure ditch L1, which overlapped L8 at the point where the two ditches intersected. Below this was C46 [d. 0.08m], a firm yellowish grey sandy clay which filled sharp-sided, flattish-based cut C39 [l. 2.4m ,width 0.6m, d. 0.2m].

Provisional interpretation of Group 1, Sub-group 2

The interpretation of L2 is complicated by the fact that it is of intermittent, discontinuous form, consisting of three cuts separated by narrow causeways no more than 0.3m across. It does seem to form an effective boundary of some form in that smaller cut features and some areas of possible metalled surface in the immediate vicinity were confined to the interior of the sub-enclosure formed by L2 with L1 (i.e. to SE of L2). On this basis L2 is considered to be a bounding element for human activity, probably associated with that within L1. Although the stratigraphy gives no support to the postulated existence of a bank, hedge or palisade along either edge of L2 it is thought likely that such must have existed in order to form a significant impediment, the ditch itself being quite narrow. The reason for the discontinuous nature of the cut remains unexplained for the present.


This Sub-group consists of two nearly-parallel and co-terminous linear ditches L4 and L5, which formed a droveway aligned running roughly W-E. The area between the two ditches narrows slightly from E to W. At the E end this is 4.8m and at the W terminal of the ditches, some 38m away, it is 3.4m.


66

The N ditch, L4, consisted of two separate cuts aligned end-to-end with a short gap [w. 0.4m] between them. Only the very W tip of the E of these two cuts lay within the excavation area, it disappearing under the SE baulk some 1.1m away. This cut had three fills, two of which were contained within the third. C55 [d. 0.23m] which was the enclos-ing fill, a firm light grey silty clay which contained a single piece of animal bone and some charcoal flecking. It enclosed C61 [d. 0.17m] and C62 [d. 0.12m], with firm orangish yel-low C61 lying over compact dark orange clayey sand C62. What little was visible of cut C54 had gentle to moderately sloping sides and a flat base.

The W cut of L4, C11, was visible in its entirety, with a length of 41.5m, a width of 1.3m and a depth of 0.34m. The shape of the cut in profile varied considerably over its length, with slope of sides varying from gentle to moderate and base from rounded to irregular to flat-bottomed. Four fills were recorded. The uppermost was C10 [d. 0.25m], a soft mid grey sandy silt from which a piece of a decorated quernstone was recovered, as well as a single sample of animal bone. Below this were two contexts, firm yellowish orange sandy clay C56 [d. 0.07m] and soft dark orange clayey sand C60 [d. 0.15m]. The former lay over C68 [d. 0.05m], a friable light grey clayey silt, while the latter lay directly upon cut C11.

The S ditch of the pair, L5, consisted of a single cut, C13 [w. 1.25m, d. 0.41m] which ran under the SE baulk of the site. Some 35.5m of its length was exposed within the ex-cavation area. Like the cut of L4, its profile varied greatly between gentle and moderately sloping sides and a base which was flat, rounded and irregular in different places along its length. Two fills were recorded within the cut. These were upper fill C117 [d. 0.16m], a soft brownish grey clayey silt, and lower fill C12 [d. 0.28m], a compact mid grey sandy clay with occasional charcoal flecks and which yielded part of a decorated quern stone (not from the same disc as that in L4, C10) and several samples of animal bone, some of which were identified as those of cattle.


The near-parallel and coterminous nature of the two linear ditches in this sub-group, as well as their similar fills and the presence in both of decorated quern fragments and ani-mal bone demonstrates their contemporary use and shared function. They are interpreted as a droveway, probably for moving livestock between enclosed and unenclosed areas.


This Sub-group consists of two linear features, L9 and L10, which may once have formed a single linear ditch, but are now truncated by large post-medieval field boundary L6.

L9 consisted of two linear ditches arranged end-to-end with a short gap, 0.9m wide, between them. The visible portion of the N of the two cuts was aligned running NW-SE, while not enough of the S cut was visible to allow assessment of its alignment.

From its NW terminal, only 1.7m of the S cut of L9 was exposed before it ran under the baulk at SE. This cut, C522 [w. 1.03m, d. 0.26m] had gentle to moderately sloping sides and an irregular base and contained two fills. The upper fill, C520 [d. 0.22m], was

KilleisK-e3587

67


a soft brownish grey clayey silt, while the lower fill, C521 [d. 0.21m], was a soft yellowish grey sandy silt.

From its SE terminal, the N cut of L9 ran for a distance of 10.3m before being truncat-ed by post-medieval field boundary L6. There were three fills in this cut, the uppermost of which was firm dark grey silty clay C107 [d. 0.35m]. Below this was C101 [d. 0.32m] a mid to light grey silty clay, and below this again the bottom fill, C108 [d. 0.35m], a mixture of C101 and redeposited natural subsoil. The cut, C102 [w. 1.4m, d. 0.35m], and moder-ate to steeply sloping sides and a base which was quite irregular due mainly to decaying limestone and water action.

L10 consisted of a single truncated cut aligned running SSW from its terminal for a distance of 5.7m before being truncated by post-medieval field boundary L6. The cut, C111 [w. 0.64m, d. 0.07m], had gently sloping sides and a flat base. Its marked shallow-ness in comparison to other linears is not likely to result from truncation because it lies so close to other, deeper linear ditches that such truncation would have to have been unfea-sibly concentrated. The single fill of L10 was C112 [d. 0.07m], a soft mid grey clayey silt from which a sample of animal bone was recovered.


It is considered probable that L9 and L10 may originally have formed a single curved lin-ear which, along with L5 at N, would have formed three sides of an enclosed area or field (the fourth side would have been out in the unexcavated area to SE). However, if L9 had continued straight on until aligned with the current L10 and made a sharp turn onto that alignment it should have emerged on the far side of the truncating field boundary. As this is not the case, it seems likely that L9 curved gradually round to the alignment of L10.

Description of Group 1, Sub-group

This Sub-group consists of a single field boundary, L8, which enters from under the NE baulk of the excavated area in 10m grid [160,120] and runs in a SW direction for ap-proximately 45m until it intersects with main enclosure ditch L1, where it terminates. L8 is composed of two identically aligned cuts with a gap some 3.9m wide between them. Both cuts are very shallow, but this is particularly the case at the NE end of the feature.

The NE part of L8 consisted of a single cut and fill. The fill was C93 [d. 0.11m], a soft brownish grey clayey silt. Nothing of a possibly archaeological nature was recovered from this during excavation. The cut was C92 [l. 27m to point where it exits under baulk, w. 0.88m, d. 0.11m], a flat bottomed cut with moderately sloping sides and a well defined break of slope at base. This cut got progressively shallower as one moved E, with a depth of only 0.07m at the point where it met the E baulk.

The SW part of L8 consisted of a single cut with three fills. The uppermost of these fills was C47 [d. 0.13m], a firm dark greyish brown clayey silt which occurred at the W extremity of L8 at the point where it intersected with enclosure ditch L1, so that as well as overlying soft orange silty sand C158 [l. 3m, w. 0.55m, d. 0.08m] in L8, it also overlapped C69 in L1 (for which context see Group 1, Subgroup 1 above). Below C158 was C25 [d.


68

0.18m] a firm grey clayey silt which was the bottom fill of the cut. The cut itself, C24 [l. 16m, w. 0.94m, d. 0.18m], was similar in profile to the E cut of L8, being flat bottomed with moderately sloping sides and a well defined break of slope at base. The shape of the cut became somewhat irregular where L8 joined with L1 at west, C24 merging with C58, the subsidiary cut of L1 which occurs only in this area of the enclosure ditch. Cut into the base of C24 near the point of connection with L1 was a single small stakehole, C134 [diam. 0.1m, d. 0.1m] the fill of which contained a moderate amount of charcoal and which may have been burnt in situ.

Provisional interpretation of Group 1, Sub-group 5L8 is interpreted as a field bound-ary. No archaeological material was recovered from within this feature during excavation, but a number of factors, mainly in the nature of comparison with other linear features on the site, suggest this interpretation is correct. The dimensions and shape of the profile of L8, as well as the nature of the main fills, are similar to other linear features on site, although L8 seems to have experienced greater truncation than these latter, resulting in a significantly lesser surviving depth. The presence of a deliberate gateway or gap between the two cuts of L8 is also suggestive of a field boundary and is paralleled on a number of the other linear features on site.


The features included in this Sub-group are considered together primarily due to their shared position within the enclosure defined by L1 and not by virtue of perceived shared function or date, although morphological similarity might be taken in some cases to imply this. These features are a mixture of discrete cuts varying in size from small stake-hole to substantial pit. They are considered below in four spatially and/or morphologi-cally defined groupings. Due to the perceived central importance of understanding the enclosure it was considered imperative to excavate all features even vaguely likely to be archaeological. Many, or even most, of the possible features excavated in the interior of L1 were subsequently considered to be of no archaeological significance, as outlined below.

[a] Substantial pits. Seven cut features are included here [see matrix]. They are described beginning with the southernmost and ending with the northernmost.

Cut C213 [l. 1.8m, w. 0.8m, d. 0.12m] was roughly rectangular in plan with gentle to moderately sloping sides and a sub-circular base. Initially considered to be a possible grave cut, excavation produced nothing to confirm this theory and the cut is obviously truncated. There were two fills, the upper being C211 [d. 0.14m], a layer of medium to large stones, mostly flat and elongated, which may have been a stone lining of the cut, with further fills above this prior to truncation. C211 lay over C212 [d. 0.12m], a mid brown silty clay. No finds or samples were recovered from the fills.

Cut C251 [l. 68m, w. 0.5m, d. 0.14m] was oval in plan with moderate to steeply slop-ing sides and an irregular base. Its single fill was C250 [d. 0.14m], a dark brown silty clay with occasional inclusions of charcoal and burnt clay. Although yielding no finds or vis-

KilleisK-e3587

69


ible samples, this fill appeared to have a high organic content and the feature may have been a storage or refuse disposal pit.

Cut C261 [l. 0.54m, w. 0.44m, d. 0.11m] was oval in plan with steeply sloping sides and a sub-circular base. It held two fills, the upper of which was C260 [d. 0.07m], a dark brownish grey sandy clay with frequent inclusions of charcoal and ash. Under this was C262 [d. 0.06m], a light yellowish grey silty clay with moderate inclusions of charcoal.

C76 [l. 2.84m, w. 2.53m, d. 0.15m] was a large pit cut with an irregular shape in plan. It had gentle to moderately sloping sides and a flattish base. The single fill of this was C72 [d. 0.15m], a mid yellowish grey clayey sand with occasional inclusions of decayed stone and charcoal flecks.

C273 [l. 0.75m, w. 0.35m, d. 0.12m] was a cut which is irregular in plan with moderate to steeply sloping sides and an irregular base. It contained two fills, the upper of which was C271 [d. 0.09m], a mid grey sandy clay, and the lower of which was C272 [d. 0.06m], a dark brown clayey silt.

Cut C173 [l. 0.1.5m, w. 0.8m, d. 0.1m] was sub-rectangular in plan with gently sloping sides and a sub-oval base. Its single fill was C172 [d. 0.1m], a mid greyish brown clayey sand with occasional flecking of charcoal and burnt clay.

C80 [l. 3.1m, w. 2.37m, d. 0.19m] was a large, oval-planned pit with gently sloping sides and flattish base. It contained two fills, the upper of which was C79 [d. 0.09m], a mid brownish grey sandy silt relatively rich in organic plant matter and which yielded samples of animal bone and tooth. Below this was C83 [d. 0.1m], a mid greyish black silty sand which yielded a single sample of animal bone. Like C79, this contained a lot of organic material.

(b) Cluster of possible small cut features in grid [140,65].A total of 12 possible cuts were excavated here [see matrix], but consideration of the results has led to the conclusion that almost all of these are non-archaeological and result from the excavation of colour-differentiated patches of soft, decayed limestone. The shapes of the profiles of these features, as well as their sterile fills, mostly agree with a non-archae-ological origin.

The only feature deemed definitely archaeological here was C210 [l. 0.27m, w. 0.19m, d. 0.32m], a cut of oval plan with steeply sloping sides and an oval base. The single fill of this cut was C209 [d. 0.32m], a mid brownish grey sandy clay with occasional inclusions of charcoal. Probably a pit, use(s) unknown.

(c) Cluster of possible small cut features in grid [145,60].A total of 7 possible cuts were excavated here[see matrix], but, as with the previous group-ing, overall consideration of the results by the director has led to the conclusion that all of these are non-archaeological and result from the excavation of colour-differentiated patches of soft, decayed limestone.


70

(d) Diffuse grouping of small cut features in and next to grids [145,70] and [150,70].A total of 12 possible cuts were excavated here [see matrix], but consideration of the results has led to the conclusion that most of these are non-archaeological on similar grounds to the previous two groupings. Two cut features were considered possibly archaeological.

The first of these was C188 [l. 0.29m, w. 0.27m, d. 0.1m], a circular-planned pit or post-hole with steep to vertical sides and a round base. It held two fills, the upper of which was C189 [d. 0.1m], a dark brownish grey clayey silt which lay over C190 [d. 0.1m], a mid brownish orange clayey silt.

The other was C185 [l. 0.23m, w. 0.21m, d. 0.12m], a cut of sub-oval plan with steeply sloping sides and an oval base. Its single fill was C184 [d. 0.12m], a mid greyish brown silty clay. Probably a pit, use(s) unknown.

(e) Apart from the above groupings some seven other isolated cut features were excavated in the area enclosed by L1 [see matrix], of which only one is currently considered likely to be archaeological.

This was C229 [diam. 0.4m, d. 0.1m], a cut of near circular plan with gentle to mod-erately sloping sides and a sub-circular base. It contained two fills, of which the upper was C227 [d. 0.1m], a dark brown silty clay with occasional charcoal flecks, considered to be a possible post-pipe. The lower fill was C228 [d. 0.05m], a mixed light/mid grey clayey/pebbly sand with moderate charcoal lumps and flecks. This may be post-hole packing, in which case the stratigraphic order of the two fills would be reversed.


A total of 45 possible cut features were excavated within the enclosure defined by ditch L1. However, on serious consideration only 12 of these were considered likely to be archaeological.

Of these 12, most seem to be pits, probably serving a variety of domestic uses such as storage and/or refuse disposal. Almost all of these are of a similar depth (between 0.1m and 0.15m) and are considered to be heavily truncated. The fills in most of the 12 cuts contain some level of charcoal content, although mostly low enough to be explicable sim-ply through having been open in the general area where fire was being used. However, a number of fills – in cuts C251, C261 and C229 – had significant concentrations of char-coal and burnt clay/ash which may have resulted from the deliberate dumping of residues from fire-related domestic processes.

One or two of the cuts – C229 and C188 – might be interpreted as post-holes, par-ticularly the former which appeared to have separate post-pipe and packing. However, no alignment or other indication of a structural pattern could be discerned.


As noted above, 9 cut features were excavated in this area, the sub-enclosure formed by L1 at NE, L2 at W and SW and probably L4 at S. Of these 9 cuts, 4 were considered likely to be archaeological in origin. These are the following;

KilleisK-e3587

71


Cut 401 [l. 0.42m, w. 0.3m, d. 0.38m], which was sub-circular in plan with steeply sloping sides and a bluntly pointed base. Its single fill was C400 [d. 0.38m], a mid brown silt with occasional inclusions of charcoal and burnt bone flecks. This is interpreted as a post-hole.

Cut C435 [l. 0.6m, w. 0.4m, d. 0.2m], which was oval in plan with moderately slop-ing sides and an oval base, filled with C436 [d. 0.2m], a dark brown/black silty clay with occasional charcoal flecks. Probably a small pit.

Cut C397 [l. 0.72m, w. 0.6m, d. 0.45m] was sub-circular in plan with steeply sloping sides and sub-circular base. Within it were three fill, the uppermost of which was mid brownish grey silty clay C394 [d. 0.13m] which lay over C395 [d. 0.27m], a mid bluish grey silty sand, under which was the bottom fill, C396 [d. 0.45m], a dark bluish grey silty clay from which a single cattle tooth was recovered. Small pit or post-hole.

Cut C417 [l. 0.4m, w. 0.3m, d. 0.15m] was oval in plan with steeply sloping sides and a flat oval base. Its single fill was C416 [d. 0.15m], a mid brownish grey silt. Possibly a pit, use(s) unknown.

As well as the above cut features, two discrete areas of stone in a hard gritty matrix were excavated within the sub-enclosure, C384 [l. 3.3m, w. 3m, d. 0.06m] and C386 [l. 2.46m, w. 1.9m, d. 0.05m]. They are interpreted as possible deliberate metalling which lie 5m apart and which may originally have been separate areas of metalled surface or part of a wider layer, the rest of which is now destroyed.


The archaeological evidence for this area consists of two small pits, two possible post-holes and two areas of stone metalling. While this is a quite limited range and number of fea-tures, it is sufficient to differentiate this sub-enclosure from the area immediately exterior to it at W and NW, where there is little or no archaeological evidence, and thus suggest-ing that this area was enclosed for a specific use probably relating to domestic activity. Nonetheless, the scattered and partial nature of the features, which reveal no pattern, are at present insufficient to tell us much about the nature of that activity.

Provisional overall interpretation of Group 1

Group 1 is provisionally interpreted as an Early Medieval domestic enclosure with sub-enclosure and a surrounding field system. The direct relationship between L1 and the other linear ditches is clear from a look at the horizontal stratigraphy where they either respect each other or join, with one failing to emerge on the far side of the other. They are also linked by similar shape and dimensions [except for L2 and L10], by the nature of their fills and by the recovery of animal bone from each [with the exception of L9 and L8]. L1, L4 and L5 are also linked by finds of rotary quernstone fragments in each. The existence of gaps or causeways in the ditches is also a common feature.


72

Group 2Group 2 consists of an elongated kiln with its in-situ function-related deposits [Sub-Group 1] and several associated cut features in the immediate area of the kiln [Sub-Group 2].

Group 2, Sub-Group 1: The kiln [cut and limestone superstructure] and its fills, the nature many of which demonstrates some involvement with pyrotechnical activity. There are 29 contexts in this Sub-Group: C357-8, C373-5, C382-3, C385, C387, C398, C408, C420, C423-7, C447, C450, C453, C461-2, C482-3, C513-5, C532 and C538.

Group 1, Sub-Group 2: Three cut features situated in the area immediately surround-ing the kiln, consisting of the following 7 contexts: C99-100, C343-4, C346, C363-4 and C376.


Prior to excavation the kiln was visible as an elongated, roughly rectangular area contain-ing, on the surface alone, at least seven distinguishable fills. It was defined along both its long sides by a limestone structure, although the latter did not survive above ground level. It was aligned with its long axis oriented N – S. The suspicion of a roughly circu-lar cupola in the centre was confirmed as contexts were removed and the shape became clearer. This was flanked to N and S by open-ended cavities which gave access to the cupola. These cavities flared outwards from the cupola to the ends of the furnace, so that the entire structure had a shape in plan somewhat resembling a bow-tie. The fills within the structure were numerous and disposed in an intricate series of overlaps. Many fills were confined to either the N or the S cavity, having obviously been raked out on that side from the cupola.

The uppermost surviving layer consisted of three contexts which did not overlap and thus could not be differentiated from each other in stratigraphic terms. These were C373, C375 and C382. C373 [l. 0.56m, w. 0.5m, d. 0.06m] lay near the N end of the kiln and was a discrete deposit of loose mid greyish brown sandy silt with occasional inclusions of charcoal and burnt clay. C375 [l. 0.8m, w. 0.44m, d. 0.12m] lay a short distance to S and was a compact mid yellowish red sandy clay. Nearer the S end of the structure, C382 [l. 0.92m, w. 0.48m, d. 0.1m] was a loose mid yellowish red sandy clay. The similarity be-tween the latter two deposits of oxidised clay is noted and it seems likely they result from the same deposit originally within the central cupola, C375 being raked out into the N cavity and C382 into the S cavity. Lying beneath and overlapped by both C373 and C375 in the N cavity was a deposit of loose dark grey sandy silt with occasional inclusions of charcoal flecks and small lumps, C374 [l. 1.1m, w. 0.64m, d. 0.11m]. Beneath C382 in the S cavity was C387 [l. 0.48m, w. 0.32m, d. 0.08m], a loose light yellowish brown sandy silt with moderate inclusions of burnt clay lumps.

Essentially confined to the cupola, but slightly overlapped at N by C375 and S by C382, was C383 [l. 0.98m, w. 0.68m, d. 0.13m], a deposit of loose mid reddish grey sandy clay with occasional inclusions of charcoal flecks and burnt clay. Beneath C374 in the N

KilleisK-e3587

73


cavity was C424 [l. 0.8m, w. 0.45m, d. 0.08m], a hard layer of mid yellowish grey stony sand essentially consisting of heat-fused limestone grit and pieces. A single piece of cattle tooth was recovered from this context. Within the cupola on its N side, overlapped by C374 and C382, was C385 [l. 0.36m, w. 0.26m, d. 0.06m], a band of loose light reddish grey sandy clay containing frequent lumps of limestone. Its composition seemed much the same as C424 (i.e. crushed limestone grit), but was very soft in comparison to the latter cemented layer. In the S cavity C383 and C387 both lay over C408 [l. 0.77m, w. 0.36m, d. 0.05m], a spread of mid reddish grey sandy clay with inclusions of charcoal and crushed limestone. This in turn was over C398 [l. 0.88m, w. 0.45m, d. 0.38m], a thick deposit of loose mid yellowish brown clayey sand with moderate inclusions of burnt clay and fire-reddened stones.

Under hard limestone-rich C424 in the N cavity were two deposits. The first of these was C447 [l. 0.4m, w. 0.36m, d. 0.04m], a very soft dark grey sandy clay with frequent inclusions of charcoal. The other was C425 [l. 0.85m, w. 0.44m, d. 0.08m], a loose mid brownish grey clayey silt with occasional inclusions of charcoal and burnt clay. Under C425 was C426 [l. 1.4m, w. 0.8m, d. 0.1m], a layer of soft mid brownish grey clayey silt containing occasional inclusions of burnt clay and charcoal. This layer covers the entire area of the N cavity and seems to be the same as C427 (see below) in the S cavity. C426 lay over C483 [l. 1.12m, w. 0.85m, d. 0.11m], a hard layer of light greyish brown sandy silt with occasional inclusions of organic material which may have been decayed twigs or small bone fragments. This was likewise confined to the N cavity of the kiln.

In the S cavity C398 lay over two contexts, C420 and C423. C423 [l. 2.9m, w. 0.76m, d. 0.14m] was a relatively widespread layer of loose mid yellowish grey sandy silt with moderate charcoal inclusions. It was essentially a loose fill of comminuted limestone which covered most of the S cavity and the central cupola and as well as C398 it was also under C447 and C385 (see above). Three thin layers existed within the thickness of C423, with the latter lying both above and below. These were C450, C453 and C461, all three being confined to the area of the central cupola. C450 [l. 0.67m, w. 0.6m, d. 0.02m] con-sisted of a very soft black silt with moderate charcoal inclusions, while C453 [l. 0.67m, w. 0.6m, d. 0.04m] was a deposit consisting solely of carbonized wood. It was the remains of a layer of round-wood branches, all laid parallel with each other and aligned N – S, prob-ably a layer of fuel used in one of the firings of the kiln. Below this was C461 [l. 0.8m, w. 0.54m, d. 0.02m], a loose dark brownish grey sandy silt with occasional charcoal.

The other context below C398, C420 [l. 0.5m, w. 0.47m] was a discrete layer of large stones near the mouth of the S cavity. Its origins are unclear but it may be a collapsed part of the superstructure. C420 lay over C357 [l. 0.98m, w. 0.74m, d. 0.11m], a deposit of friable black silty clay with occasional inclusions of charcoal and burnt clay lying in the S cavity. A piece of cattle tooth was recovered within this. Under C357 was C482 [l. 0.42m, w. 0.4m, d. 0.02m], a discrete deposit of mid greyish black sandy clay with frequent inclu-sions of charcoal which lay near the S end of the S cavity.

Directly below widespread layer C423 were three deposits; C462, C514 and C427. Loose mid greyish brown gravelly sand C462 [l. 0.54m, w. 0.4m, d. 0.03m] lay in the in-


74

terior part of the N cavity over C483 (see above), while loose light yellowish brown sandy silt with frequent burnt clay fragments and charcoal, C514 [l. 1.1m, w. 0.64m, d. 0.03m] was confined to the cupola. This latter is very similar to C423, only differing in terms of burnt inclusions. C427, the last of the three contexts beneath C423, was also below C483 (see above). C427 [l. 1.92m, w. 1.36m, d. 0.2m] was a relatively widespread layer covering most of the S part of the kiln. It was a soft mid brownish grey clayey silt with occasional inclusions of charcoal and burnt limestone grit.

Under C427 were two deposits. The first of these was C523, a scattered roughly semi-circular arc of medium to large limestone rocks immediately outside the S opening of the kiln. These lay directly upon the cut of the kiln, C358 (see below) and did not appear to be structural. The other deposit below C427 was C513 [l. 0.88m, w. 0.82m, d. 0.04m], a spread of loose mid brownish grey sandy silt lying near the S opening of the S cavity.

C515 [l. 3.3m, w. 0.72m, d. 0.03m] was a very hard, very thin layer of light yellowish grey crushed limestone with occasional inclusions of charcoal and burnt clay which was found throughout both the N and S cavities, as well as the central cupola. It lay under C483, C514 and C513 (see above for all three) and was interpreted as a deliberately laid floor level or work surface in the kiln interior.

Below C515 was the structure of the kiln, C538 [l. 3.61m, w. at N end 14.m, w. at S end 1.5m H. 0.29m] consisting of two sections – E and W – with no structural stone at the open N and S ends. These sections consisted of one to two courses of roughly squared and faced limestone blocks which appeared to have been placed directly into the cut and not bonded with any mortar or soil but wedged in places with small stones and spalls. Although crushed lime did occur in places, it seemed to originate from one of the crushed lime layers within the kiln rather than an original deliberate wall-bonding. The structure must certainly originally have projected above ground level, but to what height can now only be guessed at.

C538 and C523 lay in the original cut made in the subsoil to house the kiln. This cut, C358 [l. 5.58m, w. 1.98m, d. 0.35m], had a steeply sloping cut on all sides except at N, where it was gradual. The base was flat with a sharp break of slope at E and W, gradual at S and imperceptible at N. The overall ‘bow-tie’ shape of the kiln structure in plan was repeated in the shape of the cut which housed it. Provisional interpretation of Group 2, Sub-group 1

This structure is interpreted as a kiln. Its overall structure and the fire-related nature of the majority of the fills make this clear. The absence of any naturally accumulated deposits amongst the upper contexts excavated is noteworthy and might suggest that the structure had been heavily truncated, possibly during pre-development stripping which immediately preceded the excavation.


This Sub-Group consists of the three cut features which lie in close proximity to the kiln (Group 1, Sub-Group 1) and which are deemed likely to have had a functional relation-ship with it.

KilleisK-e3587

75


The first of these is linear ditch cut C99 [l. as the crow flies 21.3m, cumulative l. 25.5m, w. 1.2m, d. 0.35m]. Its S terminal lay 7.5m due S of the kiln, from which point it ran 7.2m in a WNW direction before turning and running N for a further 16m, at which point it turned and ran a further 1.5m to ENE, where it terminated. The shape of this cut varied a fair deal along its length. The sides were moderate to steeply sloping with a concave tendency and a sharp break at top. The base varied from flat to rounded to somewhat irregular, with break of slope at base often imperceptible. Three fills were identified. The two upper fills, which both lay independently over the lower fill, were soft mid brown sandy silt C346 [d. 0.13m], found along the central part of the ditches length, and C370 [d. 0.12m], a soft dark brown sandy silt found at the N terminal of the ditch. The lower fill, which was the main fill throughout the ditch, was C99 [d. 0.28m], a soft brownish grey sandy silt. None of the fills had any form of possibly anthropogenic material noted within them.

A metre to the W of The furnace and 3m E of the ditch lay a large pit, cut C364 [l. 1.83m, w. 1.2m, d. 0.3m]. This was sub-rectangular in plan with vertical sides and a sub-rectangular base which was concave in plan. There were two fills, both very rich in charcoal. The upper fill was C363 [d. 0.21m], a soft dark grey sandy silt, and the lower was C376 [d. 0.12m], a soft black silty sand which contained several fragments of animal bone and tooth identifed as those of sheep or goat.

Some 0.4m to N of the kiln lay a single small cut, C344 [diam. 0.35m, d. 0.13m], which was circular in plan with moderately sloping sides and a sub-circular base. Its single fill was C343 [d. 0.13m], a very soft mid greyish brown silty clay with occasional inclu-sions of charcoal. This could be a post-hole or small pit.


Of these three features – a ditch, a large pit and a small pit/post-hole – it cannot be sug-gested at present that they share a related function or date except on the basis of their proximity to each other and isolation from any other features [other than the kiln]. Also, of these three features, only the ditch [cut C100] can be considered at present to be defi-nitely associated with the central feature in this group, the kiln. This connection is made on the basis of horizontal stratigraphy alone. As the ditch turns throughout its length to enclose and/or shelter the S, SW and W sides of the kiln, the spatial arrangement of the two features strongly suggest that the ditch was excavated in relation to the latter. It seems most likely that the ditch was part of a windbreak intended to shield the kiln from the prevailing winds from W and SW.

The purpose of the pit [cut C364] is unclear. The burnt nature of its fills initially sug-gested a functional relationship with the kiln, as its position in proximity to the latter obviously does. However, the disparity between the radiocarbon dates for the kiln [AD 1420 – 1617] and the pit [AD 1161 – 1262] cast doubt on this. As birch and hazel respec-tively – two species not particularly long-living – were the sources of the charcoal used to obtain the two dates, an old wood effect seems unlikely here.


76

Possible post-hole or pit [cut C344] is likely to relate to activity at the kiln, but no obvious function presents itself on the basis of the current evidence. The fairly charcoal free nature of its fill, despite being within half a metre of the kiln, strongly suggests it was not open during the use period of the latter.


The overall interpretation of Group 2 is as a work area relating to the use of the central kiln for cereal drying and/or lime burning, with ditch C100 forming part of a windbreak shielding the kiln from the prevailing W and SW winds and with large pit C364 and small pit/post-hole C344 likewise being involved in some capacity not currently understood.

The numerous and diverse range of function related deposits within the kiln suggest a sustained period of use, even without invoking the fact that it was probably periodically cleared out and the waste disposed of elsewhere. The fact that ash and charcoal did not occur at all in ditch L7 suggest that this was not open in the period of kiln use, and it may be that the ditch had been excavated, a palisade or some such windbreak erected within it and then backfilled before the kiln was put into use. The absence of stake or post-holes within the ditch creates something of a problem in terms of suggesting how such a wind-break might have been erected.

Group 3This group consisted of some 30 cut features, mostly pits of some sort, which lay scattered throughout the W central area of the site, between 55 and 100 on the X axis [easting] and 55 and 90 on the Y axis [northing]. They lie in an area which was at time of excavation defined by post-medieval field boundaries L3 at E and L6 at W, and to N of early medi-eval droveway ditch L4. However, this is not to imply that they were originally bounded by these features, all of which they probably predate. The main concentration of these pits lie in a cluster centred around grid point [85,65], these constituting Group 3, Subgroup 1. No finds were recovered in any of the features forming part of Group 3.


This cluster consisted at time of excavation of 23 cuts and their associated fills. Of these, five are considered non-archaeological, comprising 16 contexts in all [C216-7, C222, C252-8, C287-8, C303-6], and these are not discussed here. The other 18 cuts and their fills are thought likely to be archaeological, comprising 45 contexts in all [C218-21, C239, C244-5, C246-9, C259-62, C265, C274-9, C281, C315-7, C323, C325, C327, C334-6, C345, C349-56], and these are discussed hereunder.

Of the 18 possibly archaeological cuts in this sub-group, some 9 can be considered to have good evidence of human use [C219, C245, C247, C249, C268, C281, C323, C325, C327], mainly though not exclusively on the basis of the fills they contain. Morphology and samples of organic material recovered play a part in several cases. The evidence of human use in the case of the other 9 cuts is weaker [C221, C259, C317, C334, C345, C350,

KilleisK-e3587

77


C352, C354, C356] with only slight indications of use present. In almost all cases the pri-mary indicator of human use is in some form of burnt material.

The features are described starting with the northernmost and moving southwards. Cut C259 [l. 0.57m, w. 0.5m, d. 0.3m] was of near circular plan with steep sides and

a flat base. It contained three fills, the uppermost of which was C360 [d. 0.08m], a mid brown sandy silt with occasional charcoal flecks. Below this was C361 [d. 0.12m], a mid grey silty clay which lay over dark grey silty sand with occasional charcoal flecks C362 [d. 0.16m]. On the basis of shape it is possibly a post-hole.

Cut C219 [l. 1.62m, w. 1.4m, d. 0.24m] was sub-circular in plan with steep sides and a flattish base. It contained two fills. Black sandy clay C218 [d. 0.18m] with moderate charcoal inclusions was a typical ‘burnt mound’ type fill, mixed with fragments of heat cracked stone. This lay over C239 [d. 0.1m], a light grey silty sand with occasional char-coal flecks. Pit.

Cut C334 [l. 1.39m, w. 0.76m, d. 0.26m] was irregular in plan with moderate to steep sides and an irregular base. It contained two fills, the uppermost being C335 [d. 0.1m], a light grey silty sand with occasional inclusions of charcoal flecks. The lower fill was C336 [d. 0.17m], a greyish yellow sandy clay with occasional inclusions of charcoal flecks. Pos-sible pit.

Cut C221 [l. 2.65m, w. 1.23m, d. 0.27m] was very irregular in plan with moderately sloping sides and an irregular base. It contained two fills; lower C220 [d. 0.27m], a mid grey sandy silt, and upper C265 [d. 0.18m], a dark grey sandy silt with inclusions of char-coal flecks. Possible pit, but shape very irregular.

C281 [l. 3.2m, w. 2.15m, d. 0.89m] was a large, deep cut, sub-circular in plan with steep sides and a flat base. It contained six fills which are described from the top down-wards. C274 [d. 0.24m] was a light grey sandy clay. C275 [d. 0.29m] was a light yellowish grey sandy silt. C276 [d. 0.39m] was a mid bluish grey sandy silt. C277 [d. 0.51m] was a mid brownish grey sandy silt. C278 [d. 0.73m] was a dark bluish grey sandy silt with mod-erate inclusions of charcoal flecks and lumps and which yielded a sample of waterlogged wood. The bottom fill was C279 [d. 0.89m], a mid orangish grey sandy silt. Possibly a water-hole or well. Otherwise a pit.

Cut C245 [l. 2.34m, w. 1.26m, d. 0.18m] was irregular in plan with moderate to verti-cal sides and a flattish base. It had three stakeholes cut into its base, C323 [Diam. 0.2m, d. 0.17m], C325 [diam. 0.1m, d. 0.12m] and C327 [diam. 0.16m, d. 0.13m]. The cut and all three stakeholes were filled with the same material, C244, a black charcoal-rich ‘burnt mound’ type deposit with fire cracked stone. A pit, possibly with some form of associated superstructure suspended on the stakes.

Adjacent to Cut C245, forming an arc on its SE side, were four further stakeholes. They were, from W to E; cut C354 [diam. 0.07m, d. 0.1m], with fill C353, cut C352 [diam. 0.06m, d. 0.08m], with fill 351, cut C356 [diam. 0.07m, d. 0.1m] with fill C355 and cut C350 [diam. 0.09m, d. 0.13m] with fill C349. The fill in all four cases was a mid greyish brown silty sand with no inclusions. The stakes are probably associated with the activity at pit C245.


78

Cut C249 [l. 1.67m, w. 1.08m, d. 0.14m] was irregular in plan with gently sloping sides and an irregular base. It contained a single fill, ‘burnt mound’ type C248, a black sandy clay with charcoal and some angular fire cracked stone. Pit.

Cut 247 [l. 2.76m, w. 1.16m, d. 0.35m] was a large sub-rectangular pit with moder-ate to vertical sides and an irregular base. It had a single original fill, mid grey sandy silt C293. A further cut had been incised through the latter original fill on the N side of C247. This cut was C268 [l. 1.23m, w. 0.6m, d. 0.56m], of irregular shape in plan and with steep to vertical sides coming down to an irregular base. There were three fills in this cut. The uppermost was C246 [d. 0.35m], a black stony, charcoal-rich silt of ‘burnt mound’ type which was also the main fill of original cut C247. The middle and lower fills of C268 were not found in the original cut. These were greyish yellow sand with occasional charcoal flecks C266 [d. 0.1m], and C267 [d. 0.13m], a yellowish grey silty sand. Intercutting pits.

Cut C317 [l. 1.9m, w. 1.32m, d. 0.43m] was sub-oval in plan with moderately sloping sides and a base which had been truncated. This truncation was the result of a further cut, C345 [l. 0.44m, w. 0.39m, d. 0.43m], which had been cut into the bottom of C317. This later cut was oval in plan with vertical sides and an oval base. Both fills post-dated the recut. The upper fill was C315 [d. 0.2m], a light grey sand, and the lower C316 [d. 0.23m], a mottled grey/brown sand with occasional inclusions of charcoal flecking. Recut pit.


This group is interpreted as a cluster of pits some or all of which relate to some form of pyrotechnical activity. Four of the pits contained a fill typical of burnt material from pre-historic pit groups, consisting of black charcoal-enriched silt with frequent inclusions of fire-shattered stone. One of the pits, cut C245, had three stakeholes cut into its base, with a further four stakeholes forming an arc immediately to SE, suggests the likelihood of some form of superstructure having existed in this case. There was no evidence of in-situ burning in any of the cuts and nothing in the fills or shapes of the pits indicate what their original purpose might have been. Waterlogged pit C281 may have been a water-hole or well.


Group 3, Subgroup 2: This is a scattered grouping of four cut features to W of and sepa-rated from the rest of the subgroup. It comprises 17 contexts in all [C377-81, C390-3, C399, C402-6, C409-10]. One of these cuts [C409, fill C410] is considered non-archae-ological. The other three are described hereunder, from northernmost to southernmost.

Cut C402 [l. 1.4m, w. 0.94m, d. 0.54m] was irregular in plan with moderately slop-ing sides and an irregular base. It had four fills, the uppermost of which was C403 [d. 0.12m], a mid brown silty clay with occasional flecks of charcoal. Below this was C404 [d. 0.17m], a light greyish yellow sandy clay which lay over mid grey silty clay with occasional charcoal flecks C406 [d. 0.14m]. This turn lay over bottom fill C405 [d. 0.1m], a mid grey sandy clay with occasional inclusions of charcoal.

KilleisK-e3587

79


Cut C381 [l. 1.22m, w. 0.61m, d. 0.37m] was pointed oval in plan with steep sides and an oval base. It contained four fills. The uppermost of these was C377 [d. 0.17m] a grey sandy silt with occasional flecks of charcoal. Below this was C378 [d. 0.12m], a dark grey sandy silt with occasional charcoal lumps. This was over C380 [d. 0.12m], a light grey sandy silt which in turn lay upon the bottom fill, mid brown sandy silt C379 [d. 0.17m].

C390 [l. 0.9m, w. 0.6m, d. 0.15m] was oval in plan with gently sloping sides and an oval base. It contained four fills. From top to bottom these were mid grey silt with occa-sional charcoal lumps C393 [d. 0.04], mid greyish brown silty sand C392 [d. 0.08m], black charcoal-enriched sandy silt C391 [d. 0.09m] and mid grey silt with moderate charcoal flecks and lumps C399 [d. 0.03m].


The grouping of these three features is essentially on the basis of a shared location out on the W side of the excavation area some 10 to 15m from the majority of the Group 3 features. Neither morphology nor the nature of the fills suggests a clear function or use-history.

Two of the cuts are roughly oval in plan, the other irregular, but none is regular enough to of itself indicate they were deliberately dug by humans.

Cut C390 has a single substantial burnt fill [C391], other than which charcoal content in the Subgroup 2 material is essentially restricted to occasional lumps and flecks in water deposited silts. C391 could result from deliberate dumping of burnt material, conceivably from the pyrotechnical activity going on in Group 3, Sub-group 1. The other fills indicate only having been open in the general vicinity of fire, whether natural or anthropogenic.

Ungrouped features

Three further ungrouped cut features are included in Group 3.[1] Cut 365 [l. 1.82m, w. 1.13m, d. 0.3m] lay c.7m to NW of Sub-group 1. It was ir-

regular in plan with moderately sloping sides and an irregular base. It contained four fills. From top to bottom these were; C366 [d. 0.15m], a light grey silty sand; C367 [d. 0.18m], a dark grey silty sand; C368 [d. 0.15m], a very dark grey sandy silt, and C369 [d. 0.2m], a yellow sandy silt. All four fills contained occasional charcoal flecks. Possible pit.

[2] Cut C421 [l. 2.2m, w. 0.7m, d. 0.25m] lay c.10m to N of the rest of Group 3. It was irregular and elongated in plan with gently to steeply sloping sides and an irregular base. Its single fill was C422, a mid grey silt. Unlikely to be archaeological.

[3] Cut C452 [l. 1.2m, w. 1.2m, d. 0.13m] lay c.10m to E of the rest of Group 3. It was irregular in plan with gentle to moderately sloping sides and an irregular base. Its single fill was C451, a mid grey silty sand. Possible pit.


As can be seen, consideration of Group 3, Sub-group 2 and the few ungrouped features in Group 3 adds little or no information to that dealt with in the discussion of Group 3, Sub-group 1, for which see above. A Final Neolothic/Early Bronze Age date of BC 2276


80

– 2047 for charcoal from pit C219 confirmed initial impressions that these features were prehistoric.

Group 4This group consisted of some 21 cut features, mostly pits of some sort, 5 of which are considered to be non-archaeological [cuts C491, C502, C509, C518, C484] and are not discussed.

The other 16 cuts with their fills constitute 37 contexts in all [C463-8, C471-4, C487, C496-7, C498-500, C504-7, C511-2, C516-7, C525-37] which were considered archaeo-logical or possibly archaeological. All of these bar one [C499] were located within 10m grid [120,120], and the former exception lay immediately outside this square. No finds or samples other than charcoal were recovered from any of the contexts in Group 4.

As these contexts form a single cluster and, insofar as the limited evidence goes, ap-pear to serve a related date and function, they are considered as a single grouping without dividing Group 4 into sub-groups.

Description of Group 4.

The 16 cuts are described starting with the northernmost and working southwards. Cut C499 [l. 0.6m, w. 0.6m, d. 0.3m] was circular in plan with steep to vertical sides

and a base which was flat but slanted somewhat to one side. It contained three fills, the uppermost being C498 [d. 0.11m], a greyish brown silty clay which lay over middle fill C500 [d. 0.12m], a brownish grey clay under which was bottom fill C504 [d. 0.11m], a yellowish grey sand. Probable pit or posthole.

Two metres to S lay C517, a shallow cut [l. 0.4m, w. 0.27m, d. 0.08m] of irregular plan with gentle to moderately sloping sides and an irregular base. Its single fill was black charcoal-enriched silty sand with fragmented stones C516, one of a number of fills in this group of typical ‘burnt mound’ type. Possible pit.

Cut C512 [l. 0.62m, w. 0.45m, d. 0.28m] was located c.1m SE of C517 and was an oval planned cut with moderate to vertically sloping sides and a round flattish base. It was filled with C511, a stony brown silty sand. Possible pit or posthole.

Cut C507 [l. 0.54m, w. 0.4m, d. 0.14m] lay slightly to E of C512 and was sub-rectan-gular in plan with gentle to steeply sloping sides and an oval base. Its upper fill was black charcoal-enriched silty sand with fragmented stones C505 [d. 0.02m] and its lower fill stony yellowish grey sand C506. Possible pit.

About 2m to E of C507 was C464 [l. 1.27m, w. 1.01m, d. 0.24m], a cut of irregular plan with gentle to steeply sloping sides and an irregular base. The single fill was C463, a black charcoal-enriched silty sand with fragmented stones. Pit cut.

Located about half a metre to SW of the latter feature was C466 [l. 1.92m, w. 1.6m, d. 0.2m], a large pit-cut of round plan with moderate to vertical sides and a flat base. It held two fills, upper dark greyish brown clayey silt C496 [d. 0.13m] and lower and main fill, black charcoal-enriched silty sand with fragmented stones C465 [d. 0.2m]. Cut into the

KilleisK-e3587

81


base of C466 were seven stakeholes, three of which [cuts C529-31, fills C535-7] had been inserted into the sides of the cut, and the other four of which [cuts C525-8, fills C532-4] were in the base. The cuts of the stakeholes varied from 0.05m to 0.15m in diameter and from 0.15m to 0.2m in depth. All the fills were greyish brown clayey silts and contained none of the burnt fill of the main pit cut C466, perhaps implying that the stakes rotted in situ and filled in before the pit was used as a location for dumping the ‘burnt mound’ type material.

Immediately adjacent to C466 at SW was cut C468 [l. 2.4m, w. 0.5m, d. 0.1m], ir-regular and elongated in plan with gently sloping sides and an irregular base. The single fill was C467, a black charcoal-enriched silty sand with fragmented stones. This feature had been truncated by a NW – SE running post-medieval field-drain.

About 1m to W of C468 lay a discrete spread of black charcoal-enriched silty sand with fragmented stones, C469 [l. 1.59m, d. 0.07m] which lay directly upon the natural.

About a metre S of C468 lay cut C472 [l. 0.82m, w. 0.6m, d. 0.15m] which was irregu-lar in plan with gentle to steep sides and an irregular base. It contained two fills; upper grey clayey silt with occasional charcoal flecks C471 [d. 0.05m] and lower fill C487 [d. 0.08m], a brownish grey clayey silt.

Half a metre to SE of the latter feature was similar cut C474 [l. 0.89m, w. 0.73m, d. 0.18m], which had an irregular shape in plan, gentle to steep sides and an irregular base. Its two fills were upper ‘burnt mound’ type charcoal rich sandy silt with stones C473 [d. 0.08m] and lower C497 [d. 0.1m], a mid grey clayey silt.


This group consisted essentially of a cluster of pits centred around a single large pit with stakeholes incised into its base and sides. There was ample evidence of the use of ‘hot stone’ technology for some purpose, in the form of fills of black, charcoal-rich silts with quantities of heat-shattered stone.

The seven stakeholes in the base of circular cut C466 were the best structural evidence in Group 4 and must represent the remains of some form of superstructure relating to the pit’s use. The absence of any evidence in the sides or bases of any of the pits for in-situ burning emphasizes that the deposition of the ‘burnt mound’ type fills was – despite being the most notable feature of the group – peripheral and subsequent to the original usage of the pit group, whatever that may have been.

Other than the stakeholes referred to, there are two other features which could have served a structural purpose; cuts C499 and C512. Both of these were sub-oval with round, flat bases tilted at a slight angle, and of similar dimensions. They are considered possible post-holes and lie 3.6m from each other. The possibility that they were both part of the support for the same structure is noted, although in the absence of further post-holes a suggestion of what this might have been would be highly speculative.

Overall, then, it is only possible at present to say that, as in the case of Group 3, some process involving ‘hot stone’ technology was taking place in the immediate environs of Group 4, but that this was not necessarily central to the reason for which these cuts were


82

originally excavated. Such round or rectilinear pits with stakeholes in their bases are fre-quently found on excavations and generally turn out to be prehistoric, a fact confirmed by the Final Neolithic/Early Bronze Age date [BC 2286 – 2140] given for charcoal from cut C464.

Ungrouped Features.A number of possible features excavated do not fit into any grouping.

The following five features – four cuts and a spread – might have been considered a sub-group on the basis of their all lying immediately exterior to main enclosure ditch L1. However, they all lie a considerable distance from each other [<10m], and have no shared characteristics other than a doubt as to whether they are archaeological at all. They are from N to S:

C448 [l. 0.3m, w. 0.3m, d. 0.08m] , a shallow cut of roughly circular plan with gently sloping sides and a round base. Its single fill was dark brown clayey silt C449. In grid square [140,85]. Possibly a small pit or the base of a post-hole.

Cut C442 [l. 1.7m, w. 1.3m, d. 0.58m] was sub-rectangular in plan with steep sides and a sub-rectangular base. It contained five fills which were, from top to bottom; grey silt with occasional charcoal flecks C459 [d. 0.25m], brownish grey clayey silt with oc-casional charcoal flecks and small lumps C460 [d. 0.1m], yellowish brown silt C439 [d. 0.09m], dark grey silty clay with moderate to frequent charcoal lumps and occasional flecks of burnt clay C441 [d. 0.23m] and light grey sandy silt C440 [d. 0.17m]. In grid square [120,80].

The sub-rectangular cut shape in plan along with charcoal-rich fill C441 and possibly it’s proximity to enclosure ditch L1 all suggest this pit may have been related to domestic activity within the enclosure.

Spread C444 [l. 1.8m, w. 0.6m, d. 0.04m] consisted of an area of the natural in grid square [120,75] where stones appeared to have been densely set into it to form a metalled surface, although it might also be natural in origin.

Cut C456 [l. 1.3m, w. 0.92m, d. 0.2m] was irregular in plan with gentle to steeply sloping sides and an irregular base. It contained two fills; upper mid grey silt C457 [d. 0.1m] and lower yellowish grey sand C458 [d. 0.1m]. In grid square [165,60]. Probably not archaeological.

Cut C437 [l. 0.94m, w. 0.5m, d. 0.16m] was oval in plan with gentle to steeply slop-ing sides and an irregular base. Its single fill was light grey sandy silt C438. Probably not archaeological.

KilleisK-e3587

83


Appendix 4 Lithic Artefact Report

Farina Sternke

IntroductionOne lithic find from the archaeological investigations of a possible prehistoric site at Kil-leisk 1, Co. Tipperary was presented for analysis (Table 1). The find is associated with a number of cut features and their fills, e.g. interrelated linear and curvilinear ditches, pits, post-holes and a kiln. A ringfort is situated nearby.

Find Num

ber

Context

Material

Type

Cortex

Condition

Length (mm

)

Width (m

m)

Thickn. (m

m)

Com

plete

Retouch

E3587:1:1 1 Quartzitic Sandstone Hone Stone n/a Heavily Weathered 133 38 22 No NoTable 1 Composition of the Lithic Assemblage from Killeisk 1 (E3587)

MethodologyAll lithic artefacts are examined visually and catalogued using Microsoft Excel. The fol-lowing details are recorded for each artefact which measures at least 2 cm in length or width: context information, raw material type, artefact type, the presence of cortex, arte-fact condition, length, with and thickness measurements, fragmentation and the type of retouch (where applicable). The technological criteria recorded are based on the terminol-ogy and technology presented in Inizan et al. 1999. The general typological and morpho-logical classifications are based on Woodman et al. 2006. Struck lithics smaller than 2 cm are classed as debitage and not analysed further, unless they are retouched or of specific significance, e.g. cores etc. The same is done with natural chunks.

QuantificationThe lithic (E3587: 1:1) is a worked piece of quartzitic sandstone.

ProvenanceThe find was recovered from the topsoil.

Condition:The lithic survives in heavily weathered, incomplete condition.


84

Technology/Morphology: The artefact is a hone stone. It is of an elongated, rectangular type and is heavily polished and worn on two opposed edges. The hone stone measures 133 mm in length, 38 mm in width and 22 mm in thickness.

Dating:The artefact is typologically diagnostic and dates to the Early Medieval period.

ConservationLithics do not require specific conservation, but should be stored in a dry, stable environ-ment. Preferably, each lithic should be bagged separately and contact with other lithics should be avoided, so as to prevent damage and breakage, in particular edge damage which could later be misinterpreted as retouch. Larger and heavier items are best kept in individual boxes to avoid crushing of smaller assemblage pieces.

ConclusionThe lithic find from the archaeological excavation at Killeisk 1, Co. Tipperary is a rectan-gular, elongated Early Medieval hone stone made of quartzitic sandstone.

BibliographyInizan, M.-L., M. Reduron-Ballinger, H. Roche and J. Tixier, 1999. Technology and

Terminology of Knapped Stone 5. CREP, Nanterre.

Woodman, P. C., Finlay, N. and E. Anderson, 2006. The Archaeology of a Collection: The Keiller-Knowles Collection of the National Museum of Ireland. National Museum of Ireland Monograph Series 2. Wordwell, Bray.

KilleisK-e3587

85


Appendix 5 Quernstone Report

by Anne Carey

Three stones were examined from Killeisk, as part of the stone tool specialist analysis. They comprised three rotary querns. The rotary querns were represented in largely frag-mented form, to the extent that the diameters of each could not be estimated. Two of the stones have decoration and central perforations survive, in fragmentary form, on all three examples.

Rotary querns are generally represented in the find inventories of medieval sites in Ire-land and throughout Europe. In terms of their occurrence and use in Ireland, the rotary quern was introduced from the first or second century A.D. (Caulfield 1969, 61).

Some work that has been carried out on rotary quernstones in Ireland (Bennett and Elton 1898, Curwen 1937, Caulfield 1966 and 1969) and the results are valuable espe-cially in their dating and typological discussions. In terms of technological innovation, the rotary quern represents a significant departure from earlier grain processing methods. Previously a to- and -fro grinding motion was employed, where the grain was placed on a broad lower stone and crushed, and ground to flour by a smaller hand held rubbing stone. The most common implement employed for carrying out this process was the saddle quern, known from agricultural communities as early as 7000 B.C. in the Near East. The introduction in the Iron Age of the rotary quern, with its revolving upper stone, was to have far reaching implications for the processing of grain, although the adoption of the new technology did not see the immediate dispatch of the less advanced quern. They were to work simultaneously, sometimes on the same sites, as at Cahercommaun and Ballinderry 2 for a short period (Connolly 1994, 32).

The rotary principle, employed in both the rotary quern and the mill, involves the crushing of grain between two circular stones, with the upper one revolving upon the lower. The upper stone was perforated centrally and it was through a spindle (originally wooden but later with metal components), set in the lower stone and fitted into the cen-tral perforation of the upper stone, that the mechanism was securely linked. Much work has been done on the classification of rotary querns. Three main types of rotary quern have been identified (Caulfield 1966). These are beehive, disc and pot querns.

The standards of dressing on quernstones varies considerably. Some are exquisitely finished with added decoration of either secular or religious significance, while others are neatly finished but quite plain. The preparation and maintenance of the grinding surfaces was of greatest priority. The grinding surfaces had to be rough in order to grind effectively and this was done by pecking the surfaces giving random pockmarks. Con-tinual use wore down the working surfaces, giving a smooth finish, which necessitated a re-pecking, though many stones still retain a smooth/worn strip around the outer edge of the grinding surface.

Decorated stones are known throughout the country and are preserved in many arte-fact collections. They are also known from excavation contexts with dates ranging from


86

the thirteenth to the seventeenth centuries. The reasons why quernstones were inscribed with cross shapes appears not to be merely decorative but to be related to restrictions on milling among the tenants of medieval monasteries or manors. A tax was exacted from the tenants, who were obliged to bring their grain to be ground at the local watermill and raids to discover unauthorized private querns were carried out. Any querns discovered would be either broken or brought to the monastery. A cross inscribed quern indicated the quern was in the ownership of the monastery.

The rotary querns from Killeisk 1 are all Disc C querns and are fragmentary. The two decorated examples give an incomplete sample of very interesting workmanship. The spiral design on Killeisk 1 E3587:125:1 is especially interesting and unusual. The pattern on Killeisk 1 E3687:130:1 is likely to form part of a cross shape in the ‘3’, ‘6’, ‘9’, ‘12’ posi-tion, with one of the arms surrounding the handle hole.

Disc QuernsThe disc quern was first introduced to Ireland in the first or second century A.D. and continued in use until modern times. It has a widespread distribution. This type of ro-tary quern consists of two flat, thin circular discs of large diameter. The upper and lower stones are further subdivided on two bases. The first relates to the spindle setting in the lower stone. The spindle setting can be a complete perforation, where the lower stone is bored through completely, or it can be a socket, that holds the spindle on which the upper stone revolves. The second subdivision concerns the handle-hole in the upper stone, of which three types have been identified. The first, called Disc A, has an angled perfora-tion from the top of the stone to the side, which would possibly have been strung through with rope. The second type is called Disc B, with a socket bored partly into the upper surface into which a handle would have loosely fitted. The third type is Disc C, where the handle-hole is a complete perforation into which a handle would have been fixed.

Killeisk 1 E3587, C 125, Find 1

Rotary quern fragement. Roughly rectangular shaped fragment of the upper stone of a rotary quern. The working surface is worn, with extensive wear sheen and the sides and top are dressed and even. There is an incised line running along the top of the stone at the outer edge and a portion of the central perforation survives, defined by a raised rim. Adjacent to the raised rim is an incised spiral decoration. Dimension: Diameter: Inesti-matable. L 230mm, Wth 115mm, Th 40mm.


Rotary quern fragment. Roughly triangular shaped fragment of the upper stone of a rotary quern. The working surface is flat and well dressed with neat pockmarks and the top and sides are dressed. A portion of the central perforation survives as a well dressed, broad arc, defined by an incised rim. The incomplete perforated handle hole (Diam. 20Mm, Depth 25mm) forms the centre of a two concentric circles, incised and curving to

KilleisK-e3587

87


link with the incised rim, forming the arm of a cross. Dimension: Diameter: Inestimat-able. L 180mm, Wth 290mm, Th 600mm.


Rotary quern fragment. Roughly d-shaped fragment of the lower stone of a rotary quern. The working surface is flat and well dressed with neat pockmarks, rising to a slight rim around the narrow central perforation, which is hour glass shaped. The sides and base of the stone are roughly dressed. Dimension: Diameter: Inestimatable. L 180mm, Wth 370mm, Th 100mm.

BibliographyBennett, R. and Elton, J. (1898) The History of Corn Milling, Vols. I and II, London.

Caulfield, S. (1966) The Rotary Quern in Ireland. Unpublished PhD. Thesis, N.U.I., Dublin.

Caulfield, S. (1969) ‘Some Quernstones in Private Possession in Co. Kerry’, JKAHS. Vol. 2, 59-73.

Connolly, A. (1994) ‘Saddle Querns in Ireland’, Ulster Journal of Archaeology, Vol. 57, 26-36.

Curwen, E.C. (1937) ‘Querns’, Antiquity Vol. 11, 129-51.

O’kelly, M. (1951) ‘An Early Bronze Age ringfort at Carrigillihy, Co. Cork’, JCHAS 56, 69-86.


88

Appendix 6 Plant Remains Report

By Penny Johnston

IntroductionThis report details the results of plant remains analysis from Killeisk, Co. Tipperary (E3587). The site comprised a portion of a ringfort or a medieval enclosure, pits, a small number of post-holes and a kiln. Radiocarbon dates indicate both Bronze Age and late medieval phases of occupation at the site, with most of the plant remains probably derived from the later phases of occupation.

MethodologyThe samples were collected on site as bulk soil and were processed using machine-as-sisted floatation (following guidelines in Pearsall 2000). The floating material (or ‘flot’) from each sample was collected in a stack of geological sieves (the smallest mesh size was 250mm). When all the carbonised material was collected the flot was then air-dried in paper-lined drying trays prior to storage in airtight plastic bags. There was also some waterlogged material from this site. These were saturated in water and passed through a graded series of sieves (small mesh size 250mm) and then stored in water (to maintain preservation conditions) and stored in air-tight plastic bags. The processed samples were scanned under low-powered magnification (x 10 to x 40) using a binocular microscope. Nomenclature and taxonomic order follows Stace (1997).

ResultsThe results of preliminary scanning are presented in Table 1 at the end of this report. A total of 128 samples from Killeisk were processed, of these 101 samples were scanned (27 samples did not produce flots). Charred plant remains were present in 20 of the samples and waterlogged seeds were noted in two samples. The identifications are presented in Tables 2 and 3.

There are some anomalies in the results presented in this report and in those found in the assessment report on the environmental material from this site (see Johnston 2009). Some samples were incorrectly listed as having seeds present when this was not the case, a mistake that appears to have been down to the inexperience of the person scanning the samples; modern, un-charred seed and molluscan remains were frequently mistaken for seeds. The correct scanning results are presented in Table 1 below.

Charred plant remainsThe charred plant remains included the remains of both cultivated and wild plants. Of the cultivated plants, cereals were the most common. A large portion of these were not

KilleisK-e3587

89


identifiable to type (Ceralia), making up almost one half of the total grains from recov-ered from the site (see proportions in Figure 1).

0

5

10

15

20

25

30

35

33 31 36 30 44 36 63 51 83 74 166 108 246 373 374 363 398

Contexts

Oat Barley Wheat Ceralia

Figure1:CerealcountsinsamplesfromKilleisk(E3587)

(E3587)

N = 44

Oat 30%

Barley 14%

Wheat56%

Figure2:CompositionoftheidentifiablecerealassemblagefromKilleisk


90

Of the identifiable remains, wheat was the most common cereal type recovered, fol-lowed by oat and then a small quantity of barley (Figure 2). Although preservation was not very good, it was possible to ascertain that 3 of the 6 grains present were hulled barley.

In general, the plant remains from the pits and the ditches produced similar propor-tional diagrams, and contained many different types of plant remains. In contrast, the plant remains from the kiln were almost exclusively wheat (see Figure 3) and it appears that the predominance of wheat was biased by the recovery of wheat from the kiln C.358 (from the fill C.373), as demonstrated in Figure 1. It also appears that most of the oat from the entire assemblage was taken from the pit C.80 (from the fill C.83), and this accounts for most of the oat grains in the assemblage.

0%

20%

40%

60%

80%

100%

Ditches Pits Kiln

Weeds Oat Barley Wheat Ceralia Hazelnut shell frags

Figure3:Proportionsofcharredseedsfrom

Weeds were present in Killeisk in small quantities, and these included knotgrass, wild radish, legumes, plantains, cleavers, daisy family and grass family seeds. These are weed seeds that are commonly associated with arable fields and waste ground and they were probably brought to site along with the cereal grains.

The remains of other plants were found only in small amounts, including hazelnut shell fragments and a tuber. These were recovered only in the smallest quantity and it is unlikely that they are significant features of the plant economy at this site. Hazelnut shell fragments are extremely common in deposits from Irish archaeological sites, from all pe-riods from the Mesolithic to the post-medieval period and they are widely preserved (as outlined in Monk 2000, 74 – 75). The charred tuber, only a single example was recovered from the site, may have been charred accidentally.

KilleisK-e3587

91


The plant remains from Killeisk were taken from pit fills, kiln fills and most samples with seeds were obtained from ditch fills. The ditch fills were taken from the ditches C.30 (C36 S19), C.32 (C.31, S.16), C.38 (C36 S.19, C.36 S.27 and C.44 S26), C.48 (C.33 S.15), C.58 (C.51 S45), C.59 (C.74 S.48), C.64 (C.63 S.34), C.80 (C 83 S.47), C.102 (C.108 S.112)C.143 (C.149 S.79), C.160 (C.166 S.92) and C.372 (C.371 S.201). As the plant remains were widely dispersed and sparse throughout the ditch fills it is likely that these remains reflect the scattered remains of accidentally charred seeds that were used during occupa-tion of the site at Killeisk.

The pits C.80 (C.83 S.47), C.247 (C.246 S.157), C.364 (C.363 S.203) and C390 (C.391 S.219) also contained plant remains. These were richest in the fills of C.80, but in gener-ally the quantities of seeds found in these deposits were quite low.

The kiln fills were all taken from the kiln C.358: from C.373 (S.198), C.374 (S.199) and C.398 (S.213). Although seeds are often widespread in kiln deposits, at this kiln seeds were only present in relatively small quantities in these samples, apart from perhaps the sample C.373 (S.198).

Only two samples were notably rich in cereal grains (C.83, fill of pit C.80 and C.373, fill of kiln C.358), with almost half of the assemblages from each of these samples being indeterminate. The pit contained a collection of cereals, predominantly oat grains, but including small quantities of barley and wheat. The identifiable cereals from the kiln were solely classified as wheat, indicating that this was the only crop that was dried in the kiln, at least in the final phase of its use.

Comparative assemblages from potential ringforts are generally unlike those obtained from Killeisk. For example, six ringforts in south-west Ireland contained assemblages where wheat was rare and oat and barley were the most common cereal types found (Monk et al. 1998, 69-70). Oat was predominant in the samples from two ringforts in Galway, with some barley, wheat and rye also recovered (Dillon et al. 2007). These results are quite unlike the results obtained from Killeisk and it is possible that these represent later plant remains, as the dates from this site indicate that much of the occupation was later than the early medieval period. The wheat-rich assemblage is generally more com-mon in deposits from the later medieval period. For example, wheat was the standard food crop in the area of the Pale in the middle ages but the native Irish tended to eat oats (Nicholls, 2003; 133) and there are suggestions that choice of food (oat or wheat) was used as a badge of tribal or cultural identity (e.g. McClatchie, 2003; 398.), with wheat interpreted as the English choice.

Waterlogged plant remains were recovered from two contexts at Killeisk (see Table 3). These were taken from a ditch fill (ditch C.38, fill C.36) and a pit fill (pit C.135, fill C.136). These were possibly deep features that cut the water table, or were subject to long periods of being waterlogged. As a result some sees from plants growing in the surrounding areas survived. The remains indicate buttercup, bramble and common, wind blown seeds such as those from the daisy family. These are common plants that are often associated with the margins of human settlement.


92

ReferencesDillon, M., Johnston, P. and Tierney, M. (2007) ‘Reading the ashes: charred plant

material from two ringforts in County Galway’, Seanda 2, 27 – 29.

Johnston, P. (2009) ‘Assessment of environmental remains from the N7 Castletown to Nenagh road scheme’. Unpublished technical report for Eachtra Archaeological Projects.

McClatchie, M. (2003) ‘Section 12- The plant remains’, pp. 391-413 in Cleary, R.M. and Hurley, M.F. (eds.) Cork City Excavations 1984-2000 Cork: Cork City Council.

Monk, M. (2000) ‘Seeds and soils of discontent,’ pp. 67-87 in A. Desmond, G. Johnson, M. McCarthy, J. Sheehan and E. Shee Twohig (eds.) New Agendas in Irish prehistory. Bray, Wordwell.

Monk, M., Tierney, J. and Hannon, M. (1998) ‘Archaeobotanical studies in early medieval Munster’, pp. 63-75 in Monk, M. and Sheehan, J. (eds) Early Medieval Munster. Archaeology, History and Society. Cork, Cork University Press.

Nicholls, K. (2003) Gaelic and Gaelicized Ireland in the Middle Ages. (2nd edition) Dublin: Lilliput Press.

Pearsall, D. (2000) Paleoethnobotany: a Handbook of Procedures. New York, Academic Press.

Stace, C. A. (1997) New Flora of the British Isles. (2nd edition) Cambridge, Cambridge University Press.

KilleisK-e3587

93


Table1:ScannedsamplesfromKilleisk1,Co�Tipperary(E3587

Table 1Sample Context Charcoal Seeds % Scanned2 1 L Absent 1003 3 L Absent 1004 6 L Absent 1006 10 L Absent 1007 12 L Absent 1008 20 L Absent 1009 22 L Absent 10011 25 L Absent 10012 26 L Absent 100

13 28 L Absent 10015 33 L L 10016 31 L L 10017 34 L Absent 10018 35 L Absent 3319 36 H L* 10022 30 M L 10023 47 L Absent 10024 45 L Absent 10025 50 L Absent 10026 44 L L 10027 36 L L 10028 40 L Absent 10032 56 L Absent 10034 63 M L 10038 73 L Absent 10045 51 M L 10046 79 L Absent 10047 83 H L 6048 74 L L 10049 93 M Absent 10051 101 L Absent 10052 103 L Absent 10053 104 M Absent 8054 112 L Absent 10060 120 M Absent 10061 125 L Absent 10062 94 L Absent 10063 95 L Absent 10064 96 L Absent 10070 136 L L* 10071 137 L Absent 10072 140 L Absent 10075 110 L Absent 10076 144 L Absent 3377 145 L Absent 10079 148 L L 10080 151 M Absent 10083 154 M Absent 100


94

Table 1Sample Context Charcoal Seeds % Scanned84 155 M Absent 10085 159 L Absent 10091 161 L Absent 3392 166 L L 10094 170 L Absent 100103 194 L Absent 100112 108 L L 100113 209 L Absent 100119 218 H Absent 100128 241 L Absent 100132 258 L Absent 100146 248 M Absent 75157 246 H L 100159 267 M Absent 100165 277 M Absent 100166 278 L Absent 20172 144 H Absent 100179 316 L Absent 10191 347 L Absent 100198 373 M M 100199 374 H L 100201 371 M L 100202 375 M Absent 100203 363 M L 100204 376 H Absent 100205 377 H Absent 100209 382 L Absent 100210 383 L Absent 100212 387 L Absent 100213 398 L L 20219 391 L L 65222 403 L Absent 100223 404 L Absent 100224 405 L Absent 100227 389 L Absent 100228 408 H Absent 100235 357 H Absent 100236 398 H Absent 100246 439 M Absent 100254 447 H Absent 100263 453 H Absent 70265 482 H Absent 100266 461 H Absent 50278 465 H Absent 100282 496 L Absent 100289 463 H Absent 100290b 473 H Absent 100291 497 L Absent 50292 498 L Absent 100298 483 L Absent 100

KilleisK-e3587

95


Table 1Sample Context Charcoal Seeds % Scanned301 513 L Absent 100303 514 M Absent 100309 521 L Absent 100

*indicatespresenceofsomewaterloggedseeds(allotherseedsarecharred)

Key:H=Highfrequency,M=Mediumfrequency,L=Lowfrequency


96

Table2:IdentifiedcharredplantremainsfromKilleisk1,Co�Tipperary(E3587)

Context 33 31 36 30 44 36 63 51 83 74Sample 15 16 19 20 26 27 34 45 47 48Hazelnut shell fragments (Corylus avellana L.) 2 Indeterminate seeds from the Knotgrass family (Polygonaceae)

2

Wild radish (Raphanus raphanistrum L.) capsule

1

Indeterminate seeds from the Legume family (Fabaceae)

1 1 3

Plantain (Plantago L. species) 1 1 Cleavers (Galium aparine L.) 1 Indeterminate seeds from the daisy family (Asteraceae)

1

Oat grains (Avena L. species) 1 1 7 Possible oat grains (cf Avena species) 2 Barley grains (Hordeum vulgare L.) 1* 2* Possible barley grains (cf Hordeum vulgare) 1 Wheat grains (Triticum L. species) 1 4 Indeterminate cereal grains (Ceralia) 1 2 1 1 2 10 3Indeterminate grass seeds (Poaceae) Indeterminate tuber

*indicateshulledbarleygrains

KilleisK-e3587

97


Table3:IdentifiedcharredplantremainsfromKilleisk1,Co�Tipperary(E3587)continued

Context 149 166 108 246 373 374 371 363 398 391Sample 79 92 112 157 198 199 201 203 213 219Hazelnut shell fragments (Corylus avellana L.)

3

Indeterminate seeds from the Knotgrass family (Polygonaceae)

1

Wild radish (Raphanus raphanis-trum L.) capsule

Indeterminate seeds from the Legume family (Fabaceae)

1 1

Plantain (Plantago L. species) Cleavers (Galium aparine L.) Indeterminate seeds from the daisy family (Asteraceae)

Oat grains (Avena L. species) 2 Possible oat grains (cf Avena species)

Barley grains (Hordeum vulgare L.)

1 1

Possible barley grains (cf Hor-deum vulgare)

Wheat grains (Triticum L. species)

1 15 3 1

Indeterminate cereal grains (Ceralia)

14 7

Indeterminate grass seeds (Poaceae)

1

Indeterminate tuber 1


98

Table4:IdentifiedwaterloggedplantremainsfromKilleisk1,Co�Tipperary(E3587)

Context 36 136Sample 27 70Meadow/Creeping buttercup (Ranunculus acris L./repens L.) 1Bramble: blackberry drubes (Rubus fructicosus L.) 25 6Indeterminate daisy family seeds (Asteraceae) 1

KilleisK-e3587

99


Appendix 7 Animal Bone Report

Margaret McCarthy

IntroductionThe excavations produced a relatively small collection of 1,064 animal bones the bulk of which were recovered from the enclosure ditch and represent the butchery refuse and table waste of the medieval occupants of the site. Other smaller samples of bone with less secure provenance but which are probably later in date were also examined. These include bones from a kiln and a pit located outside the perimeter of the enclosure, some inde-terminate fragments from a field system and animal bones from the fills of two parallel ditches that formed a droveway.

Condition of the materialBone survival has been poor and fragmentation rates throughout all the excavated fea-tures are high resulting in large numbers of fragments that can only be classified into large and medium mammal size categories. Most bones appear to have been subjected to a certain degree of erosion and weathering due presumably to being left scattered on the living surface for sometime before permanent disposal into the enclosing ditch. A number of specimens are burnt to a chalky white texture suggesting that food waste was occasionally cast into the fire and remained there for a sufficient period of time to take on the white cracked appearance of heat-shattered bone. Most of the bones from C74, one of the fills of the enclosing ditch, were calcined and this deposit may represent rakings from a fireplace. Traces of blackening on a few cattle limb bones are presumed to relate to cooking meat over a spit and gnawing marks by a dog are visible on four specimens.

AnalysisThe distribution of identified species is shown in Table 1 and the predominance of the main livestock animals suggests various episodes of food waste disposal primarily associ-ated with the medieval occupation of the enclosure. In all, 1,064 bones were presented for examination with 87% of these alone coming from various fills of the enclosure ditch. The results of the analysis are described below according to the different context groupings recognised by the excavators.

Enclosure DitchThe density of bones recovered from the ditch fills was high representing over 87% of the entire assemblage. A total of 934 bones were examined and these were spread relatively evenly across the various fills with the largest individual sample (141 fragments) coming from C77. Among the mammalian remains, most of which come from domesticated


100

species, those of sheep/goat are most abundant, accounting for 37% of the identified as-semblage and contributing 70 bones out of an identifiable sample of 187 fragments. On the basis of distinguishing features all of the sheep/goat bones are identified as sheep with no evidence for goat. The sample comprises butchered fragments of vertebrae, scapula, humerus, metapodia, pelvis, femur and tibia. Most upper limb bones show some traces of crude butchery and almost half of the fragments are chopped through with no complete bones surviving intact. Ageing data is very limited but epiphyseal fusion data indicate that most of the sheep kept ranged in age from between two and four years with no evi-dence for young lambs or individuals kept to an advanced age. In the absence of skulls and horn cores little can be said about the appearance or sex of the sheep although it is likely that both males and females were kept.

The bones of cattle account for 36% of the total number of faunal remains identified and these represent a selection of meat-bearing and peripheral elements from two adult individuals. A wide range of elements were recovered including teeth, vertebrae, pelvis, scapula, metapodia, sacrum, humerus, radius and ulna indicating the slaughter and con-sumption of animals within the enclosure. Limited ageing evidence indicates that cattle were selected for slaughter at their prime stage for eating with only one late fusing element having all epiphyses fused; most have at least one end still growing, a few have both ends still growing and on this basis the cattle at Killeisk probably ranged in age from two to no more than four and a half years of age. Estimating age by mandibular wear was not possible due to the absence of mandibles. Evidence of butchery consists mainly of rough chop marks associated with disarticulation and jointing and as no complete cattle bones were found, no assessment can be made concerning the size and sex of the animals.

The proportion of horse bones at the site is high with a total of 30 elements being recovered from six fills of the medieval enclosure ditch. The anatomical elements present include radius, femur, tibia, metapodia, astragalus, splint bone, mandible and loose teeth. The recovered elements represent two adult individuals with fully fused distal femur and distal radius being identified. The teeth were mostly found as single elements in the ditch fills and all belong to adult horses around eight years of age. None of the horse bones are complete; some display fresh breaks caused during the excavation of the ditch deposits, others have evidence of ancient horizontal chopping through the midshaft of the bone associated with dismemberment and perhaps the extraction of marrow.

All other species are only nominally represented. Pig is attested from four metapodial bones present in one of the lower fills (C104) of the ditch. These represent a single adult individual and are assumed to be from a domestic pig. A fractured dog tibia was recov-ered from fill (C44) representing an adult medium-sized individual, similar to a modern sheepdog. The partial remains of an adult cat were found in C104 and a radius and ulna of another adult cat were found in fill (C94). Also recorded from the enclosure ditch is a red deer (Cervus elaphus) humerus indicating that venison was consumed at the site from time to time. A domestic fowl coracoid was found in fill (C74) and one of the upper fills (C31) contained the leg bone of an adult rabbit. The domestic fowl bone is totally cal-cined from being in contact with intense heat for a prolonged period of time.

KilleisK-e3587

101


PitsAnimal bones were recovered in small amounts from two of the excavated pits. One of these (C80) was located within the enclosure and it contained a piece of a pig mandible and two sheep bones identified as a mandible and a pelvis. It was not possible to estimate the age of the individuals present as the teeth did not survive in the mandibles. The only other bone-producing pit (C364) was located outside the ringfort close to the kiln and it contained four adult pig teeth, five fragments of bone from a medium-sized mammal and seven indeterminate fragments.

Posthole- C397An adult cow tooth was recovered from a posthole (C397) within the interior of the ringfort.

Kiln – C359Two fills (C357, C424) of the kiln located outside the ringfort produced negligible amounts of animal bone. A sheep tooth was found in fill (C357) and the other fill con-tained a portion of a sheep scapula.

Linear feature 2A total of 24 animal bones were recovered from this linear feature thought to be con-temporary with the ringfort. The bones are extremely fragmented and just three positive identifications can be made. Dog is attested from the recovery of a single vertebra from C18 and there are two cattle bones; the midshaft section of a horncore from C140 and a tooth from C145. The remaining 21 bones are not determinate to species.

Droveway ditchesThe droveway consisted of a double-ditched linear feature measuring c.40m in length. Animal bones were recovered in small amounts from the fills of both the north and the south ditches. The only identified species are two fragmented cattle teeth from a fill (C117) of the southern ditch. The remainder of the sample is not determinate to species but 21 fragments can be placed into a medium-mammal sized category.

Field boundary – C112Twenty three small fragments of bone were recovered from this field system none of which are diagnostic to species. Seven pieces of long bone are sufficiently large to indicate that they originated from sheep-sized mammals and 16 fragments are indeterminate.

ConclusionsThe collection of animal bones from Killeisk 1 represents the dumping of domestic

refuse primarily by the medieval occupants of the ringfort. Bones of a variety of species were found in the enclosing ditch including cattle, sheep, pig, horse, dog, cat, red deer,


102

rabbit and domestic fowl. Species ratios based on both minimum numbers of individuals and fragment counts indicate that sheep are slightly the most frequent species represented although cattle had an almost equal place and they would certainly have been the most important meat source by virtue of their larger size. Pigs do not seem to have played an important role in the livestock economy and as these animals are normally associated with woodland usage their rarity at Killeisk may indicate the limited occurrence of wood-land at the time the site was occupied. The figures for horse are quite interesting in view of the small sample size and the data suggest that these animals may have held an important position at the site. Analysis of the bones indicates that the two individuals represented were killed for their meat once they had outlived their usefulness as beasts of burden. The only red deer bone from the site came from the post-cranial skeleton and presumably represents an individual that was hunted for its meat. In common with other medieval sites domestic fowl also makes an appearance albeit as a single find and the rabbit bone found in the ditch is probably not intrusive as this species was introduced into Ireland in the medieval period. While the identified sample is too small for valid comparisons to be made with other contemporary sites, the faunal material from Killeisk 1 seems to con-form to the type of bone debris one would expect in an medieval secular settlement with a preponderance of the main livestock animals and little evidence for the exploitation of the surrounding natural resources to augment the meat supply.


103

Con

text

No.

Hor

seC

attle

S/G

Pig

Dog

Cat

Rab

bit

Dee

rB

ird

LMM

MIN

DET

IND

ETfil

l of e

nclo

sure

ditc

h1

412

3551

fill o

f enc

losu

re d

itch

22

1314

29fil

l of e

nclo

sure

ditc

h3

52

1410

31fil

l of e

nclo

sure

ditc

h6

77

from

line

ar 2

C19

181

23

fill o

f ditc

h C

1926

77

fill o

f enc

losu

re d

itch

3013

3245

fill o

f enc

losu

re d

itch

311

11

1125

1049

fill o

f enc

losu

re d

itch

331

23

fill o

f enc

losu

re d

itch

3410

6272

fill o

f enc

losu

re d

itch

361

1fil

l of e

nclo

sure

ditc

h37

12

47

fill o

f enc

losu

re d

itch

421

23

fill o

f enc

losu

re d

itch

441

1fil

l of e

nclo

sure

ditc

h51

52

7fil

l of d

rove

way

ditc

h 53

1410

24fil

l of d

rove

way

ditc

h 55

716

23fil

l of e

nclo

sure

ditc

h73

11

135

2141

fill o

f enc

losu

re d

itch

745

61

2425

1778

fill o

f enc

losu

re d

itch

772

330

6046

141

fill o

f enc

losu

re d

itch

781

1pi

t 80

in in

terio

r of e

nclo

sure

791

1fil

l of e

nclo

sure

ditc

h81

18

211

liste

d as

voi

d co

ntex

t82

24

6pi

t 80

in in

terio

r of e

nclo

sure

832

2fil

l of e

nclo

sure

ditc

h94

42

62

14fil

l of e

nclo

sure

ditc

h96

715

22fil

l of e

nclo

sure

ditc

h10

41

118

41

208

1467

fill o

f ditc

h C

111

S of

dr

ovew

ay

112

716

23


104

Con

text

No.

Hor

seC

attle

S/G

Pig

Dog

Cat

Rab

bit

Dee

rB

ird

LMM

MIN

DET

IND

ETfil

l of d

rove

way

ditc

h C

119

117

22

fill o

f dro

vew

ay d

itch

125

1212

from

line

ar 2

C14

614

01

1fro

m li

near

2 C

146

141

55

fill o

f enc

losu

re d

itch

144

11

from

line

ar 2

C14

614

51

25

8fil

l of e

nclo

sure

ditc

h14

95

321

1746

fill o

f enc

losu

re d

itch

154

11

2fil

l of e

nclo

sure

ditc

h15

55

13

121

2517

73fil

l of e

nclo

sure

ditc

h15

91

423

823

1271

fill o

f enc

losu

re d

itch

161

13

197

1949

fill o

f enc

losu

re d

itch

165

21

35

11lis

ted

as v

oid

cont

ext

168

24

6fil

l of k

iln 3

5935

71

1fil

l of p

it 36

4 ne

ar k

iln37

64

57

16fil

l of p

osth

ole 3

9739

61

1fil

l of k

iln 3

5942

41

1

TO

TAL

3072

765

23

11

121

329

138

110

76

KilleisK-e3587

105


Appendix 8 Geophysical Survey

By Earthsound Archaeological Geophysics

Introduction to the Sites Earthsound Archaeological Geophysics were commissioned by Eachtra Archaeological Projects to execute a series of geophysical surveys over pre-selected sites located outside the C.P.O. boundary of the new N7 Road Scheme. The survey areas were located adjacent to sites containing archaeological remains that were excavated within the C.P.O. boundary and extend beyond it.

The new N7 road development extends from the eastern edge of the present N7 Nenagh Bypass and ties into the M7 / M8 Portlaoise-Castletown Scheme. The new de-velopment terminates south of Borris-in-Ossory. The scheme in total covers a length of 35 km and the work undertaken for this report comprised of land adjacent to Contract 1 (Clashnevin to Castleroan), a 17.1 km section of the road located on the western half of the overall development.

The bedrock geology along the scheme consists of a mixture of greywacke, siltstone, sandstone and mudstone all of which are suitable for magnetic susceptibility and mag-netic gradiometer geophysical surveys. However, the majority of archaeological features were magnetically very weak, which reflects a poor contrast between the fill of cut mate-rials and the parent geology. Magnetic susceptibility data were also generally very weak.

Permissions to undertake the geophysical surveys were obtained from the Department of the Environment, Heritage and Local Government (License R179). The geophysical survey was requested to assess the extent of the archaeological remains

located outside the C.P.O. boundary. All sites were assessed using magnetic gradiom-eter and magnetic susceptibility surveys with the exception of Drumbaun 2 where high vegetation prevented the use of the magnetic susceptibility meter.

Killeisk 1Located within the townland of Killeisk, the northwest corner of the site (Figure 1) lies at Ordnance Survey of Ireland Irish National Grid (ING) Reference E194488 N179462, adjacent to Chainage 2800-2900.

Fieldwork was conducted on the 19th

November 2008. The survey area encompassed flat land, which is relatively low lying and was under stubble at the time of survey. In the week preceding the geophysical survey, the climatic conditions were cold and wet and this continued during the fieldwork.

The excavated archaeology at Killeisk consisted primarily of a large number of cut features and their fills, including a number of interrelated linear and curvilinear ditches, pits of various shapes and sizes, a small number of post-holes and a kiln, the basal courses of which had been inserted into a specifically shaped cut. Truncation was evidenced


106

throughout the site and was locally severe, with the result that few or none of the cut features excavated could be considered whole or even substantially intact.

A portion of a ringfort or medieval enclosure was located within the Lands Made Available (LMA) at Killeisk. It measured 35 m north-south by 33 m within the area of the LMA. The area of the entrance was not located within the LMA. The ditch was 1.6 m to 2.2 m wide by 0.4 m to 0.5 m in depth. A well was cut into the base of the ditch. There were a number of fills in the ditch, some of which contained animal bone. Two quern stones were recovered from the ditch. The ditch was re-cut in places and there was evidence of an internal bank. There were a small number of pits, post-holes and stake-holes in the interior of the site but there is no evidence of a structure. There were three sets of features outside the ringfort, a field system and droveway, a group of pits and a kiln (Eachtra 2008d).

Aims & Objectives The geophysical survey was requested to assess the extent of the archaeological remains located outside the C.P.O. boundary. Specific objectives were to:

• Assess the extent and location of the archaeological remains associated with those uncovered in the excavation

• Determine the presence or absence other associated archaeological features A meth-odology was developed to allow multiple techniques to systematically investigate the site. Detailed magnetic gradiometer and magnetic susceptibility surveys were carried out within the survey area. These techniques have been used in commercial and research archaeological projects for many years and are considered the most appropriate techniques for a detailed investigation of the underlying archaeology (Clarke 1996, Scollar et al. 1990). Where possible, the use of multiple geophysical techniques allows a greater confidence to be placed in the interpretation of detected anomalies, which is especially useful on small sites such as this. Their combined application can be used to determine the geometry, compositional material and the extent of an archaeological target.

Methodology The fieldwork was carried out by J. Bonsall, D. Jones and H. Gimson of Earthsound Ar-chaeological Geophysics. A magnetic gradiometer survey was carried out using two Geoscan Research FM256 fluxgate gradiometers. Two pairs of sensors were mounted on a CF6 carry frame.

A magnetic susceptibility survey was carried out using a Bartington MS2 Magnetic Susceptibility meter and MS2D search loop interfaced with a Trimble Pro-XRS Differ-ential GPS.

A rectangular grid was laid out using a Trimble Pro-XRS Differential Global Position-ing System (see Technical Appendix 2), and divided in to 40 × 40 m sub-grids for the gradiometer survey.

KilleisK-e3587

107


Magnetic Gradiometer Survey The survey was undertaken along lines parallel to the sub-grid edges, walking approxi-mately south to north, starting in the southwest corner of each grid. Subsequent lines were surveyed in alternate directions (‘zigzag’).

Data were recorded using an FM256 at a spatial resolution of 1 m intervals between traverses and 0.25 m intervals along those lines. The instrument was positioned facing north, parallel to the Earth’s magnetic field, to allow increased geo-magnetic resolution.

The instrument was set to a recording sensitivity of 0.1 nT. Prior to the beginning of the survey and after the completion of every two sub-grids, the electronic and mechanical setup of the instrument were examined and calibrated as necessary over a common refer-ence point. The magnetic drift from zero was not logged.

Data were collected automatically using an internal sample trigger while the opera-tor walked at a constant pace along the traverse. The data were stored in an internal data logger and downloaded to a field computer using Geoscan Research Geoplot v.3.00a software.

Data Processing

Preliminary Data Treatment The data were pre-processed in Geoplot 3.00. Spurious high intensity anomalies, com-monly statistical outliers, are referred to as geophysical ‘spikes’. In magnetic data, an ‘iron spike’ is a response to a buried ferrous

object, often in the topsoil. Iron spikes are generally not removed in geophysical data; although often modern in origin, they can be indicative of archaeological material. The raw data contained some poorly matched sub-grids, caused by the internal drift of the

fluxgate gradiometer and the gradual misalignment of the fluxgate sensors between calibration episodes. To compensate for this, a zero mean traverse (ZMT) function was employed. The use of ZMT alters data to adjust the mean of each traverse to zero by in-creasing or decreasing data as necessary. This alters the statistical properties of the data to give a uniformly bipolar background, centred around zero. Post-ZMT plots were com-pared with raw data to analyse the potential removal of geophysical anomalies along the line of a traverse. Further Processing No further processing functions were applied due to the high quality of the data collec-tion. A low pass Gaussian filter was applied, reducing the variability of the data whilst

improving the visibility of weak archaeological features. This also had a smoothing effect on the data. A sine wave interpolation function was applied to provide a smooth, aesthetically pleasing

image for presentation. For a given point x, the contribution of adjacent readings to the interpolated point is given by the function sinc (x) = sin πx/πx (Scollar 1990). This function is used as a sliding window along each transect, resulting in an interpolated im-age, expanding the resolution of the data from 1 m x 0.25 m to 0.5 m x 0.125 m. This func-


108

tion was chosen as giving a clearer interpolated image than linear interpolation (which assumes a direct linear change between each point) or bicubic interpolation (taking the surrounding sixteen values into account).

Graphical Display

Pre-processed data are displayed in XY traceplot format in Figures 2, 7, 10, 15, and 20. An XY traceplot presents the data logged on each traverse as a single line with each succes-sive traverse incremented on the Y-axis to produce a stacked plot. The data have not been clipped at –3 and +3 nT. The main advantage of this display option is that the full range of data can be viewed, dependent on the clip, so that the ‘shape’ of individual anomalies can be discerned and potential archaeological anomalies differentiated from iron ‘spikes’.

Processed data are shown in Greyscale format in Figures 3, 8, 11, 16 and 21. The grey-scale plot presents data as pixels on a linear grey shaded scale, increasing or decreasing dependent on the values of the maximum and minimum clip. The magnetic gradiometer data have been clipped at –2 (white) and +2 nT (black). Data values beyond the clip limits are shown as ‘pure’ black or white. The main advantage of this display option is that the data can be viewed as a base map.

An interpretation plot is presented in Figures 4, 9, 12, 17 and 22.

Magnetic Susceptibility Survey A geophysical grid baseline was not established for the Magnetic Susceptibility survey; a Differential Global Positioning System (DGPS) data logger, a Trimble TSC1, displayed a graphical representation of the survey area, with traverses spaced every 5 m. The graphi-cal representation was utilised by the geophysicists to navigate and collect data points at predetermined intervals.

The topsoil volume magnetic susceptibility survey was undertaken along lines parallel to the survey grids, walking approximately south to north. Subsequent lines were sur-veyed in alternate directions (‘zigzag’). Data were recorded at a spatial resolution of 5m intervals between traverses and 5m intervals along those lines.

The MS2 was set to a recording sensitivity of 1 SI unit to obtain (infinite) volume specific magnetic susceptibility (Volume MS or κ). When measured in SI units, the data is expressed as 1 x 10-5 κ. The MS-DGPS recorded northing and easting within the Irish National Grid to a minimum accuracy of ±1 m, and altitude to an accuracy of ±2 m.

Prior to surveying each survey station, the MS2 was calibrated according to the man-ufacturers guidelines, by ‘zeroing’ whilst holding the sensor approximately 3 m in the air. The positive and negative data presented in this report are the κ value of the survey area compared to the κ value of the air, being, theoretically, zero.

Data were collected and stored automatically in the TSC1 data logger by using a push button trigger on the MS2. The geophysicists walked at a constant pace along each traverse, pausing only briefly at each survey station to obtain a measurement of magnetic susceptibility. The data were downloaded to a field computer using Trimble Pathfinder Office 2.9 software.

KilleisK-e3587

109


Data Processing

Preliminary Data Treatment The data were exported from Pathfinder Office 2.9 to Microsoft Excel. The Excel data were gridded in xyz format as northing, easting and κ, using Golden Software Surfer 8.00. Further Processing A natural neighbour interpolation function was applied to the data to provide a smooth, aesthetically pleasing image for presentation. No further processing functions were ap-plied due to the high quality of the data collection.

Graphical Display

Contour plots can be shaded to emphasise particular regions between lines. Processed data are shown in interpolated colourscale contour plot format in Figures 5, 13, 18 and 23. The colourscale plot presents data as pixels on a linear colour shaded scale, increasing or decreasing dependent on the values of the maximum and minimum clip. The geophysical data have not been clipped. The main advantage of this display option is that the data can be viewed as a base map. A disadvantage is that the interpolation process can exaggerate isolated high or low data (this is noticeable over areas where no data has been collected,

e.g. in the space occupied by a field boundary); to compensate for this, each survey station has been marked by a small black dot, creating a point cloud, to an accuracy of ±1 m, so that exaggeration between points can be visualised. An interpretation plot of the magnetic susceptibility data is presented in Figures 6, 14, 19 and 24.

Reporting, Mapping & Archiving The geophysical survey and report follow the recommendations outlined in the English Heritage Guidelines (David et al. 2008) and IFA Paper No. 6 (Gaffney et al. 2002) as a minimum standard. The conditions of the Detection Licence issued by the Licensing Section of the Department of the Environment, Heritage and Local Government require a copy of this report.

Geophysical data, figures and text are archived following the recommendations of the Archaeology Data Service (Schmidt 2001).

Field boundaries were mapped and drawn based upon data gathered by the DGPS. Technical information on the equipment used, data processing and methodology are given in Appendix 1. Appendix 2 details the survey geo-referencing information and Ap-pendix 3 describes the composition and location of the archive.


110

Results & Discussion The interpretation figures should not be looked at in isolation but in conjunction with the relevant discussion section and with the information contained in the Appendices. Features are numbered in the relevant figures listed below [G1=gradiometer anomalies, M1=magnetic susceptibility anomalies] and are described and interpreted within the text.

In magnetic gradiometer data, a dipolar anomaly or ‘iron spike’ is a response to buried ferrous objects, often in the topsoil. Iron spikes generally are not removed in geophysical data, although often modern in origin, they can be indicative of archaeological material.

Magnetic Gradiometer Survey Figure 15 – Pre-processed Magnetic Gradiometer Data Figure 16 – Magnetic Gradiometer Data Figure 17 – Magnetic Gradiometer Interpreta-tion The magnetic gradiometer data is very weakly enhanced at this site, making the ringfort or

medieval enclosure ditch features difficult to visualise. Nonetheless, dipolar anomalies and trends of magnetic noise mark out the extent of a number of ditches and enclosing elements. There are also a number of possible pits and possible isolated areas of burning, which could suggest the presence of hearths and a possible kiln, both within and beyond the enclosure.

Anomaly [G24] is an arcing ditch which represents the southern extent of the ringfort or medieval enclosure ditch revealed in the excavation area. The anomaly is comprised of weakly magnetic ditch fill and a number of possible pits, or isolated magnetic components (including some ferrous responses) within the ditch fill. The anomaly measures 123 m in length. Taking the ditched enclosure as a whole, the feature measures 95 m NW-SE and 41 m NE-SW, enclosing a total area of 0.335 hectares.

Anomaly [G25] comprises two parallel ditch features. The eastern example matches a linear boundary ditch revealed in the excavation. The western example matches a small annex enclosure ditch, revealed in the excavation.

Anomaly [G26] is a linear trend which traverses the survey area. Measuring 83 m in length this anomaly crosses anomaly G24 and is a probable continuation of the drove-ways’s southern ditch, as revealed in the excavation. The remains of the northern ditch of the droveway could not be seen in the magnetic gradiometer data.

Anomaly [G27] is a linear trend which runs in a northerly direction across the survey area. Measuring 100 m in length this anomaly extends beyond the survey area to the south and its orientation suggests that it continues into the excavation area although no feature was detected during the excavation. This anomaly runs parallel to the existing field boundaries and is likely to represent a former field division, drainage ditch or mag-netically strong plough furrow.

Anomaly [G28] comprises of two trends which are located on the southern edge of the survey area. The trends measure 20 m and 47 m respectively. The larger of the two may represent a relict field boundary visible on the 1st

Edition OS map of the area.

KilleisK-e3587

111


Anomaly [G29] is a small cluster of possible pits which were detected between G27 and G28. A possible area of industrial deposits or modern dumping can be seen in the southern corner of the survey area, which include possible pits and ferrous responses.

Anomaly [G30] represents two possible pit clusters located in the centre of the survey area. One cluster of three possible pits is located within the enclosure. The other compris-es of two possible pits and is located 20 m to the east, on the outer edge of the enclosure ditches.

Anomaly [G31] is an area of plough furrows which were detected across the eastern edge of the survey area.

Magnetic Susceptibility Survey Figure 18 – Magnetic Susceptibility Data Figure 19 – Magnetic Susceptibility Interpretation The magnetic gradiometer anomaly G24, which represents the southern extent of the

enclosure ditch, has been superimposed on to Figure 18. The magnetic susceptibil-ity data are reasonably quiet which was also apparent in the weakly enhanced magnetic gradiometer data. The data generally falls between 1 and 4 SI units. The mean of the data (3 SI units) reflects the slightly stronger data on the western side of the site, no doubt in-fluenced by the stronger magnetic susceptibility of the enclosure and the droveway.

Anomaly [M7] marks the location of a boundary between low magnetic susceptibility values to the southwest and slightly higher values to the north. This anomaly probably marks the boundary of the archaeological activity zone.

Anomaly [M8] is located in the northwest corner of the survey area and represents a coherent area of slightly raised magnetic susceptibility values. Located close to the new M7 C.P.O boundary this anomaly could be associated with the droveway and kiln activ-ity revealed during the excavation.


112

Conclusion

Achievement of Objectives The geophysical surveys have assessed land adjacent to archaeologically significant sites which were identified within the C.P.O boundary of the new road. The magnetic gradi-ometer surveys have identified and mapped possible archaeological remains within these areas which will help to enhance the interpretation of the sites’ extent and composition. The magnetic susceptibility survey across the sites was useful in that it reinforced certain trends identified in the magnetic gradiometer data.

Summary of Results A continuation to the medieval oval-shaped enclosure ditch was detected in the geophysi-cal results. The northern half of the enclosure ditch and a south-western possible annex were revealed during the excavation -the geophysical results have mapped the southern extent of the enclosure. The mapped components of the ditched enclosure, from both ex-cavation and geophysics, measures 95 m NW-SE and 41 m NE-SW, enclosing a total area of 0.335 hectares. 37% of the enclosure lies within the CPO of the N7. A continuation of parts of the droveway were also identified along with a number of other possible ditches and a relict field boundary.

KilleisK-e3587

113


AcknowledgementsFieldwork: James Bonsall BA (Hons) MSc PIFA Daniel Jones MA Heather Gimson BA (Hons) MSc MIAI Report: Heather Gimson James Bonsall Graphics: Heather Gimson


114

Bibliography CLARK, A.J. 1996 Seeing Beneath the Soil, London, Batsford

DAVID, A. LINFORD, N. & LINFORD, P. 2008 Geophysical Survey in Archaeological Field Evaluation, Second Edition, English Heritage

EACHTRA 2008a Busherstown Preliminary Archaeological Report, N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme, Unpublished Archaeological Report by Eachtra Archaeological Projects

EACHTRA 2008b Drumbaun 2 Preliminary Archaeological Report, N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme, Unpublished Archaeological Report by Eachtra Archaeological Projects

EACHTRA 2008c Drumroe 1 Preliminary Archaeological Report, N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme, Unpublished Archaeological Report by Eachtra Archaeological Projects

EACHTRA 2008d Killeisk 1 Preliminary Archaeological Report, N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme, Unpublished Archaeological Report by Eachtra Archaeological Projects

EACHTRA 2008e Park 1 Preliminary Archaeological Report, N7 Castletown to Nenagh (Derrinsallagh to Ballintotty) Road Scheme, Unpublished Archaeological Report by Eachtra Archaeological Projects

GAFFNEY, C., GATER, J. & OVENDEN, S. 2002 The use of Geophysical Techniques in Archaeological Evaluations, IFA Paper No. 6, Institute of Field Archaeologists

SCHMIDT, A. 2001 Geophysical Data in Archaeology: A Guide to Good Practice, Archaeology Data Service, Oxford, Oxbow

SCOLLAR, I., TABBAGH, A., HESSE, A. AND HERZOG, I. 1990 Archaeological Prospecting and Remote Sensing, Cambridge, Cambridge University Press. Topics in Remote Sensing Vol. 2

The following texts are referenced in the Technical Appendix:

WALKER, R. 2000 Geoplot Version 3.00 for Windows, Instruction Manual, Version 1.2, Clayton, West Yorkshire

archaeological report - killeisk, co. tipperary (ireland)

Education

associated pitand enclosure

large elliptical enclosure

late medieval period

n7 castletown

simon ofaolainwritten

late medieval activity

portion of map

tipperaryearly bronze