Proc Soc Antiq Scot, 124 (1994), 391^32, fiche 3:E1-G6

Excavations at Pluscarden Priory, Moray Finbar McCormick* with contributions by Paul Buckland, Stephen Carter, Ruby Ceron- Carrasco, Richard Fawcett, Barbara Ford, Dom Ambrose Flavell, Sheila Hamilton-Dyer, Daphne Home Lorimer, Coralie Mills, Charles Murray, Thea Gabra-Sanders PenningtoB J , n (wit MiddletonP hJ , C Nicholson & A K G Jones) & P Skidmore

ABSTRACT This report, prefaced briefa by history Pluscardenof descriptiona Priory,by church,the and of describes excavations on the western side of the Priory. Features investigated include the western nave,the end seriesa of of burials, stone-lineda containingpit richa organic deposit, wallsand associated with the medieval monastic complex. The excavation was arranged and funded by Historic Scotland.

INTRODUCTION The site was excavated by what was then AOC, Historic Scotland, in August and September of 199 anticipation 0i e buildinth f residentiana o f go l buildin presumee th n go e wesd th sit f to e claustral range of the monastery. The excavation was organized and financed by Historic Scotland. repore Th t begins with brief description histore foundatio e th architecturae th f th so f yf o o d nan l development of the church.

OUTLIN PRE-REFORMATIOE TH F EO N HISTOR PLUSCARDEF YO N PRIORY, 1230-1560

Ambrosm Do e FlavelB OS l The Valliscaulian Order was a late expression of the eremitical ideal which swept across Europe c 1075-1200. Unlike similar ascetic movements in France at this time, such as the Grandmontines, it never enjoyed great celebrit diffusionr yo . Onl house1 MignardyP 2 A listee J sar y 19the db th , - century historian of the Order, and although Val des Choux retained its independence until 1761 there would appear to have been no new foundations after c 1255 (Folz 1959, 100).

: Department of Archaeology and Palaeoecology, Queen's University, Belfast 392 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

three Th e Scottish houses (Pluscarden; Beauly, Inverness-shire Ardchattand an ; , Argylle ar ) very much outlier fairla f o sy close-knit group centree mother-housth n do f s VaChoueo de l x (Cote d'Or), which is situated about 20 km south-east of Chatillon-sur-Seine. Besides the Burgundian houses, there were two in Normandy, one of which served as a hospice for monks travellin r froo mo gt Scotland. Another, Remonvaux s establishewa , dc 121 Lorrainn 0i e which was then stil Germaa l n duchHole th yf y o Roma n Empire. There is a contemporary description of the early Valliscaulian monks in Chapter 17 of Jacques de Vitry's Historia Occidentalis (Folz 1959, 104). It records a visit made in 1216, when monke th s were still followin austere gth e rule drawViary b ratifie d p dnan u Popy db e InnocenI II t in 1205/6 copA . f thiyo s Bulbees ha l n preserve Chartulare th n di f Morayo connection yi n with Pluscarden (Macphail 1881, 193) but the original rule had been somewhat mitigated before the Scottish foundations, bringing the fledgeling Order more into line with Cistercian practices without altogether abandoning the Carthusian customs and ideals of their founder, Viard, who had receive s monastidhi c brothey trainin e Charterhousla th a t s a rga f Lugnyeo e resultinTh . g legislative text, knowOrdinale,e th s na gives precise detailliturgye th f so , dutie f officialso d an s customs of the Order and allows some degree of reconstruction of monastic life at Pluscarden unde Valliscauliane rth s (1230-1454). Folz (1959, 95; see also Vermeer 1954), analysing the sources of the Ordinale, has shown that the bulk of the text is drawn from the Officia Ecclesiastica (Choisselet & Vernet 1989) of Citeaux (written before 1137), modified in places by statutes taken from the Carthusian Consuetudines Guigonis c 1127( ; Guigues 1984). Ver ydocumene littlth f eo s peculiai t e th o t r Valliscaulians mose Th . t evident departure from Cistercian practice would have bee little nth e cells (probably a subdivided dormitory on the Grandmontine model: Hutchison 1989, 93, 338-9) and private gardens whic brethree hth n were fre cultivato t e e afternoonth n i e s whe t otherwisnno e engaged. This preservation of certain conspicuous elements of the Carthusian ideal, together with the small size of the community (the Bull of 1205/6 specified that no Valliscaulian house was to have more than 20 monks, including lay brothers), would have served to distinguish Pluscarden from her Cistercian neighbour at Kinloss, 10 km away to the north-west. It is surely significant that Walter Bower, Abbot of Inchcolm, should have mistaken the three Valliscaulian houses for Carthusians in his Scotichronicon of 1437 so soon after the foundation of the first, and only, Scottish Charterhous t Pertea 142n hi 9 (Beckett 1988 Bowe; xi , r 1987, vo , 275)8 l . Pluscarde royaa s lnwa foundation, taken unde protectioe th r locae th f l no bisho n 1233pi , Kine Pope th th 123n g1263n i ei d pre-127e 6an mosn th I . f o t 5conventuae th charters n o d an ,l seal house th , s generallei y referre s Vallisa o t d Sancti Andree whic s probablhwa s originayit l name before variants of the place-name Pluscarden became the norm. The dedication of all Valliscaulian monasteries was to Our Lady and St John the Baptist (Birch 1900, 94). King Alexande generousld ha I I r y endowe priore dth y with land holdingd an s s scattered alon coastae gth l strip fro Rivee mth re SpeyNes th lane forestd o th t s: dan f Pluscardeo s d nan Auchtertyre, fishing rights (for salmon) on the Spey and Findhorn; the mills of Pluscarden, Elgin, Dunkinedir, Molen, Forres and Dulpotin with their multures; the church of Dores with its lands, tithes, offering e iro rightsd th n l an stithed extracteal an ,n o s d from minee foresth f n o i ts Pluscarden. The extent of these endowments is important, as the Bull of 1205/6 had specified that the Valliscaulians were to live on their revenues. Unlike the Cistercians they did not have an army of lay brothers to bring wild places into cultivation: their monastic economy depended on rents, tithe paymentd an s kindn i s . Some indicatio relative th f no e prosperit f Pluscardeyo gives ni y nb Bagimond's Tax Roll of 1274: Beauly and Ardchattan are each set down for £200, Pluscarden for £533 (Macphail 1881,80). McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY,3 MORA39 | Y

There are no contemporary annals. The so-called Liber Pluscardensis is a history of Scotland (heavily dependent on Bower, 1987; and Fordun, 1871-2) written at Pluscarden in 1461 under the patronag e Abboth f f Dunfermlineeo o t mona seculaa t t e authono kbu s Th r. cleriwa r c named Maurice Buchanan, and the book contains no internal monastic sources. A history of Pluscarden befor e 15tth e h century, from documents alone, would inevitabl heavile yb y weighte recordy db s of litigation over multures (1300, 1367, 1370, 1390) and similar disputes. One of the few events on record is that the priors of Pluscarden and Beauly attended the Scottish parliament, at Birgham, on 17 March 1289/90. Tradition assign fireo t Pluscardetw sa s e 14tth ho nt century 130n i e 3 on ,whe n Edward I's army marched through Moray, and the other at the hand of Alexander Stewart, Earl of Buchan, the Wolf of Badenoch, in 1390. That there was a major fire at some time about the late 14th or early 15th century is evident from the fabric of the church; the fact that the damage to the transepts was. never fully repaired suggests declining fortunes. Prior Thomas Fullonis resigne n 139di 8 afte year1 2 r officen i s e bisho s letteth Hi .o t rp speaks of ruling through hard times and the same document contains the decree of election of Prior Alexander (1398-c 1417) who is charged 'to repair the ruins of the church and dwelling' (Macphail 1881, 217). Evidently thi t completes no wor s kwa unioe d th for f Pluscardet no a , d nan Urquhart Priory in 1454, the Abbot of Dunfermline's delegate reported that he found Pluscarden in great need of repair; the result of almost 60 years of neglect was that the vaulted roofs of choir and crossing threatened to fall (Webster 1948, 197). e B15tyth h centur generae yth l conditio house th becomd f enha o e desperate. There were frequent disputes over the priorship, revenues were failing, and contact with France had become difficult becaus Hundree th f eo d Year r (1337-1453)Wa s undaten A . d statut exempted eha e dth Scottish houses from attending the annual General Chapter at Val des Choux (Birch 1900, 119), but we do not know what degree of contact the priory managed to maintain with its mother-house. A letter, dated 18 December 1506, from Prior-General James Courtois to the Prior of Beauly, notes that on account of the distance (1300 km - this letter took 40 days to reach Scotland) his predecessors had extended the intervals between the Prior's visits to Val des Choux from four to six years t eve t beebu , nno nthid observedha s . Courtois lameno goet n o s t thae promiseth t d consignment of Scots salmon, which should have been sent to his agent in Bruges for transmission to Dijon, had also failed to appear (Batten 1877, 157). On 12 March 1453/4 Pope Nicholas V issued a Bull uniting the priories of Urquhart and Pluscarden, and, from this date unti Reformatione th l , Pluscarde dependena s nwa t e housth f eo Benedictine abbe t Dunfermlineya , having severe s linkdit s Chouxsde witl hVa . This movd eha been initiate Johy db n Benale, Prio f Urquharto r petitioneo wh , Pope dth e tha bots a t h housed sha suffered material and spiritual reduction (Urquhart had two monks, Pluscarden six) and as Val des Choux was too remote to offer adequate help and guidance, the two houses should be united (Macphail 1881, 223) joine Th . t Pluscarde a communit e b building e o t th s s na ywa s there were considere finee mord b o an drt e extensive wels a , easies a l brino rt g into repair Valliscauliane .Th s were require mako dt e fresh professio Benedictiness na . This union marke improvemenn da priory'e th n i t s fortunes Pluscarded an , n probably benefited materially and spiritually from being placed under the jurisdiction of Abbot Richard de Bothwell (1445-70 tima t a )e when Dunfermlin s flourishinwa e centra s a g f learnin artse o e th Th .d gan second Benedictine Prior of Pluscarden, William de Boyis, had charge of the church fabric at Dunfermline throughout the time of Abbot de Bothwell's restoration and building work at the west end of the abbey nave c 1450 (Webster 1948, 231) and it is tempting to assign the re-ordering of the Pluscarden priory church to within the period of his priorate (Hannah 1936, 190-1). 4 39 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

PLUSCARDEN PRIORY MORAY GRAMPIAN

MORAY FIRTH Branderburg

MONAUGHTY FOREST

Pluscarden Priory

,16

1km

ILLU 1 S Location maps. Based upon Ordnancee th CrownSurvey© p ma copyright McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY,5 MORA39 | Y

Signature e charterth n o s s show thae communitth t y numbered fou n 1345i r t leasa ; t seve 1500n ni ; nin1558n i 1508 n 152n ei 1 thi i o 15481 d T .3 s4an d 1 ; figur an ; muse d w e ad t an unknown number of resident servants. At the time of the Reformation, the Rental of 1561 (Macphail 1881, 254f) takes account of five monks; a chamberlain with two servants; a master-cook; porter; master-baker and gardener. Earlier figures were probably higher rather than lower black-monn i s a , k house servante sth alwayd sha s take brothersy place la nth f eo A . ratio of one layman to one monk was not uncommon even in the llth century, but by the 16th century the proportion of employed servants had risen dramatically (cf Knowles 1971, 260). Certainly the Prior of Pluscarden was in a position to give hospitality to King James IV and his retinue in September 1506 and it is interesting to note that the king's expense sheet on that occasion included 'drink silver to masons working on the priory, 15 shillings' (Anson 1959, 101). The last two priors of Pluscarden, George Learmonth (1509-29) and Alexander Dunbar (1529-60), both owed their position to the patronage of the king. George Learmonth was a cleric of the diocese of St Andrews (Cal Papal Letters, XVIII, no 51), and was later (1529) nominated coadjuto Bishoo t r p Gavin Dunba f Aberdeeno r , with righ f successioo t n (Reg Episc Aberdon,, I havy ma e394 t diebu )d befor s consecrationehi . About this time, certainl 1535y yb , Pluscarden was erected int oregalitya . Learmonth's successor, Prior Alexander Dunbar, died in the year of the Reformation. His long priorate is commemorated by the building of the Dunbar Vestry, and by the extensive alienation of monastic property to members of his own family. To what extent this was a response to the crippling taxation of the Church under James V, and to what extent vested self- interest, is not yet apparent. He was actively involved in affairs of Church and State, attending at least five parliaments between 1540 and 1558, and had been summoned to a Privy Council in St Andrews in 1546. He probably attended three provincial chapters called between 1549 and 1559 which were part of John Hamilton's efforts to reform the Church. Although secular clerics, both these priors took the Benedictine habit on appointment and served as regular superiors (Dilworth 1986, 63). After Dunbar's death in 1560 there followed a series of Commendatory priors who drew the monastic revenues and pensioned the remaining monks. The last definite reference to a monk of communite th y dates from 1587 yeae Pluscardee th , th r n estates were secularize erected dan d into a free barony by Act of Parliament. It is a grant of fishings on the Spey made to James Dunbar of Cumnock in March 1586/7 endorsed with the approval and consent of Thomas Ross, the sole remaining membe e communityth f o r bearind an , e monk'gth s signature beneate h th tha f o t commendatory prior, Alexander Seton (Macphail 1881, 245). e priorTh y fell into ruins som ee 17t timth hn ei centur 168n i s usestonea d s 2wa da yan - quarr provido yt e material e reconstructioth r fo s t GileS f no s Church, Elgin. Paymente th o t s hauliers suggest that demolition of the monastic buildings must have been extensive (Fasti Ecclesiae Scoticanae, II, 286 (Scott 1915-28); Anson 1959, 158). Further decay was arrested by Earh Fiff 4t o l e e th who 1821n i , , converte ease dth t range int oshootina g lodge. John Patrickd 3r , Marques f Buteo s , bough properte th t n 189 yi begad 7an n complete restoratio churce th t f no hbu work was halted on his death, and resumed only when Benedictine monks returned to Pluscarden in 1948. Ther severae ear l account priorf e visitso th o st y 18te ruin19td th han hn si centuriesf o e On . these, writte Catholia y nb c pries 1761n i t particularls i , y valuabl t givei s earliese ea th s t known versio traditioe th f no t completedn no thanav e s th t ewa t readsI . somn i : m e'..a DoubI . t whether or not the Western Part [of the church] has ever been finished, the Ruines of it being now only | SOCIET 6 39 ANTIQUARIEF YO S OF SCOTLAND, 1994

thre r foueo r feet high wherea othee Walle th sth f r so Part almose sar Inhabitantte entirth d ean s there hav evulgaa r tradition that whe l finisheChurce al nth s hwa d excepting that Western Pare th t Builders should have hear Voicda e saying

The Monastery of Pluscartie Begun; but ne'er shall ended be

and upon this they should have desisted' (Humphries 1952, 108).

PRIORE TH Y CHURCH

Richard Fawcett Ther s virtuallei documentatioo yn whicy nb constructioe hth churce th f f nPluscardeo h o e b n nca dated and reliance, therefore, must be placed almost entirely on the evidence of the architecture. This evidence indicates that, althoug unduln ha t buildinno y s protractegwa d operationa s wa t i , complex process involving several stages, with subsequent damag rebuildind ean g addine th o gt difficultie f understandino s precise gth e sequence e followinTh . g paragraph thereforn ca s e b e offere moro n s eda tha verna y provisional interpretation. Presumabl e plas determinedyth nwa footinge th d basd an an ,s e course laid outt lonno , g afte foundatioe th r priore th f aboun nyo i t 1230 (Cowa Easson& n 1976, 84-5), e althougb y ma t hi assumed that a temporary oratory would be provided for worship pending the completion of the church. The plan chosen, with an aisle-less rectangular presbytery, and a two-bay chapel aisle on the east sid eacf eo h transept typa f eo moss i , t commonly associated wit Cisterciane hth s (illu. s2) Scotlandn I , however, this plan typalss oewa adopte severay db l other Orders, presumably because it represented a relatively economical arrangement, and variants of it were used for Valliscaulian Ardchattan, Augustinian Cambuskenneth and the Tironensian houses of Kilwinning and Lindores. The nave was set out with a single aisle on the side towards the cloister, although it was more common for single aisles to be on the side away from the cloister, in which position they often represent later augmentation. The architectural evidence points to the transepts having been completed in advance of the presbytery, presumabl alloo yt e gatherinwth f adequatgo eeaster n funda r fo sn lim f greatebo r pretensions than was at first possible. The internal eastern elevations of the two transepts present differing solution e probleth o t s f devisinmo g two-storeyed designs which could accommodate vaulted aisles together with a mural passage at clearstorey level. What is seen in the two transepts was clearly the result of several campaigns (illus 3). The pier and responds of the south transept arcade are of stepped-and-shafted profile, as is als towere oth south-wese pieth t a r t cornecrossinge th f northero e r Th . n respon norte th f hdo transept arcade, whic courses hi witn di outee hth relatera wallf o s di , type l thiAl .s suggests that the south transept arcade was built along with the outer walls of the transept. By contrast, since the piesoutherd an r n respon northere th f do n transept arcade, together north-ease wit piee th t hth a r t cornecrossinge th f clusteref o o r e ar , d shaft typ typa e( e alssouth-wese pieoe th t th usea rr dfo t corner of the crossing), work on the main body of the north transept seems likely to have been carried out later than on its southern counterpart. In the south transept, the chapels and the eastern clearstorey each have windows in the form otriaa f f lancetdo s embrace singla y db e hood mouldin storeyo eacn tw gi e f thahso th baytd an , transept thus seem r aparunlikelfa daten i e t b .o yt Internally e clearstoreth , y windowe ar s McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, 7 MORA39 I Y

• standing wall .- burial HZ surveyed wall t Opi 222 excavated wall _._ limit of excavation — land drain _o pipe connected to downpipe O soakaway

10m

ILLU S2 Site plan before restoration (after Mackintosh 1893) 398 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

East facf eo East face of north transept south transept

20m

ILLUS 3 North/south section through transepts, showing east elevation before restoration (after Mackintosh 1893)

contained withi arcn n a f similaho r two-centred segmenta lexternae forth s ma l hood mould, with flanking arches opening onto the clearstorey passage. The floor of this passage, of course, runs at a lower level tha windoe nth w t abov sillsse pitchee e eb th , whico t dd roohha f ove aisle rth e vaults, anflankine dth g arches consequentl muce yar h lower tha centrae nth l arch. Althoug ease hth t conventual range, which adjoin extendd an s s southwards fro e soutmth h transept, falls outsid e remieth f thio t s paper s perhapi t i , s worth mentioning thae originath t l sacristy, immediatel soute soute th th f hho o y t transept detailins ha , g very lik transepte thath n i t , and was evidently built as part of the same campaign. It could be that the need to provide the all- important accommodation in the east range at an early stage of the operations was a factor in encouragin completioe gth adjoinine th f no g south transept befor northe th tha f o .t Internallya , fine featur soute th f heo transep nighe th s ti first-floote staith o rt r dormitor ease th t n yrangei e th ; vaulting which carries the upper part of this stair across the sacristy appears to be a later intrusion presens init t form. norte Th h transept clearstorey, whiclase probabls th t transept e f hpartwa o th e f e so b y on o st completed simplea s ha , r design tha southe n th tha n o ,t witcuspeo htw d lancet window eaco t s h bay, framed by single arches on the inner side of the clearstorey passage. Unfortunately, the McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY I 399

20m

ILLUS 4 East/west section through transept, showing north gable before restoration (after Mackintosh 1893)

relationship between the two levels of the north transept has become less clear as a result of the later modification of the eastern windows of the chapel aisle. The chapel window reveals themselves appear to be original, however, and the wide arches could have been designed for tracery fro starte mth . Externall e mosyth t strikin gable th g s e i featur crose elevatioth m f ar e norts o eth f hno transep ts superimpose(illuIt . 4) s d rank f lanceto s s could indicate some transepe debtth o t s t gables designe Elgir dfo n Cathedral c 1220s, mose th pars a ,f t o tprestigiou s campaign underway in the north-east in that period. Such an approach to the design of terminal walls had a long history, however, and Pluscarden shows its independence of Elgin in many respects. The north doorwa notee particularlb a y s da yma y refined example openinn a f o g with carefully detailed arch mouldings emerging directly from broadly chamfered jambs. A related doorway was provided as the entrance of the warming house in the east range (now used as the refectory). The place of the nave in this complex chronological sequence is now difficult to determine, since so little of it remains, but, as with the transepts, its surviving details suggest it was completed befor presbyterye th e s alreadha t I . y been said thaclustered-shafe th t t south-west crossing pier belongs to the same intermediate phase of works as the north transept arcade, and this pier also incorporate thud san easterdictated- e th for- e nf dth mo respon soute th hf do nav e arcade. Perhap mose th s t interesting aspec f thio t s respon s thadi t apparentli t y reste solia n do screen wall intende down ru t leasna o dt nave t th par ef o tarcade . This suggests thamonastie th t c choir | SOCIET 0 40 ANTIQUARIEF YO SCOTLANDF SO , 1994

Section through choir towed an r looking south

Section through choir towed an r looking north

20m

ILLU S5 East/west sections through presbytery/choi crossind an r g before restoration, showing (above) south wall, (below) north wall (after Mackintosh 1893) McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 401

originally extended down into the nave, with this wall being provided as a backing to the stalls. Suc arrangementn ha , which continue particularle b o dt y commo Cistercian ni n churches whicn i , h divides westere th wa m dn ar betwee brethrenchoiry e monke nla th th d f so stiln s an seee ca ,b l n ni the late medieval rebuilding at Melrose. Supporting evidenc earlien a r nave fo r th dat er thae fo presbytere nth provides yi e th y db design of the handsome round-arched processional doorway, which opened from the east bay of the south aisle into the cloister (illus 5). This entrance had five alternately projecting and recessed free-standing shafts to the jambs. The great west doorway of the nave must have been even more imposing on the fragmentary evidence of its surviving bases, which indicate jambs with at least four orders, each apparently composed of a triplet of shafts separated by individual shafts. When the presbytery was eventually started, it would seem to have had nothing of that tentative quality which resulted from the various changes of design in the transepts; it was, indeed, structura highese th f eo t calibre s completioIt . s nunlikeli havo yt e been befor e 1260th e r so 1270s, since its side windows - originally of great scale - have stumps of bar tracery related to that first used at Glasgow Cathedral and Elgin Cathedral and at Sweetheart Abbey, in those decades (Fawcett 1984, 152, 162-4). Along the flanks the surviving springings of the tracery show that the windows wer fouf eo r lights groupe pairso tw n ,d i wit h circlets withi betweed nan sub-archese nth ; tha , essentiallis t e samyth ee wes forth ts m a windo f Glasgowo w Cathedral before modern restoration e desige originath Th . f no l east windo s unknownwi e higth ht qualitbu e , th f o y detailin tracere presbytere th th f go n yi whola stily s ya appreciatee b lema d fro delightfue mth l 'spherical triangles' abov aisle eth e roofs, which contain triplet foilef so d circlets. Althoug e transepte aisleth hth f s o navd san e were stone vaulteds commowa e s th a , n ni greater Scottish churche periode th f so , wit exceptioe hth crossine th f no g area highee th , r main vessels of those parts were not. By contrast, the presbytery was given particular prominence by being vaulted - a distinction shared with few other Scottish buildings under construction in the 13th century. (On es temptei suspeco dt t that havy thima se bee nresula f rivalro t y with Elgin Cathedral, wher t first eveea aislee no t nth s were vaulted. rooe Th )f space abovt li vaule s eth wa t by a large vesica and a trifoliated triangle, which pierced the east gable one above the other (illus 6). Barely rising above the crossing of the four arms of the church was a squat tower, each face of which was decorated by bifurcated-cusped trefoils in the triangles to each side of the four roofs. In a church which owes something to Cistercian ideas it may be that the tower was never intended to rise more than minimally above the roofs. In its final form, the tower was capped by a characteristically Scottish Late Gothic saddle-back roof, although a low spire possibly was intended originally. Auncertain a t n staghistorys it extensivele b n ei churce o t th , d hha y remodelled. Mos f thio t s reconstruction could post-dat change eth allegiancf eo housee th f eo , fro Valliscauliae mth e th o nt Benedictine Order 145n i , 4 (Cowa Easson& n 1976 , 85),61 . Certainly 1457n ,i buildinge ,th s were statd ba ea (Cal.n i e b saiPapal o dt particularls Letters,wa t , 330)i XI d an ,y stated tha t thaa t t tim choivaulte e crossined th th f ran so g were under threa f collapseo t . However, although thers ei definito n e record that Pluscarde s affecte depredatione nwa th y b d Wole th f Badenoc o f f so n hi 1390 s believei t i , d traditionally tha ravagee h t d Pluscarde s welna s Elgina l . This theory would possibly accord better with the evidence. In support of this the resignation of Prior Thomas, and the election in his stead of Brother Alexander in 1398, seems to have been on the basis that a firm hanneedes dwa brino dt buildinge gth s back into repair (Macphail 1881, 83-4) additionalls i t I . y significant that it was specifically stated in the 1450s that it was nearly 60 years of inaction which thee th no t existin d le g stat pendinf eo g collapse.

McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 403

What is certain is that at some stage the high vault of the presbytery was removed, and the windows were greatly reduce sizn di orde n ei strengtheo rt damagee nth d walls. Single lancets- Y , tracer r intersectinyo g tracery rather dispiritingly replaced what must finese havth f et o bee e non ensemble f first-generatioso r tracernba Scotlandn yi three-lighw . Onlne e yth t upper east window f moro s e wa tha e mosnth t minimal pretensions, with dagger r trefoilo s s withi intersectine nth g matrix (illus 6). It may be noted, incidentally, that the way in which the eastern windows fit within the earlier window t dissimila arcno s hi r fro mannee mth r that windows were inserted int wese oth t window arc t Sweethearha t Abbey, presumably afte fire th er ther f 1397eo . Possibly as part of the same operation of stabilization and repair, the monastic choir must have been moved eastwards so that it was fully contained within the eastern limb, since the eastern arch of the tower was partly blocked by an inserted wall. This wall also served as a pulpitum, pierced onlrelativela y yb y small central doorway flanked an stai,a y drb leadinlofta o t . Sincp gu e the western tower arch was also built up it is likely that these measures were an attempt to increase the stability of the tower. The implications of these operations for the nave are difficult to assess, again because of the paucity of the remains. However, the mouldings of the single window reveal at the eastern end of nave th e north wall suggest this windo rebuildina s we 14twa th hr f 15tgo o h centuries, indicating that significant structural works also had to be carried out here in the later Middle Ages. Nevertheless, there t leasmusa e suspicioa tb t n tha nav e neves tth ewa r fully rebuilt. lase tTh significant identifiable additio sacrist e churce th th s treasurd o nt hy wa an y blockn ,o norte th hpresbyterye sidth f eo , which took ove functione rth earlie e th f o sr sacristy adjacene th o t t south transept. This square structure, wit polygonaha l stair s nortturreit n ho t face s databli , o et between 1533 and 1560 on the basis of arms on its vault which are assumed to refer to Prior Alexander Dunbar tracere eass Th .it f t yo window f uncuspeo , d loop form characteristis i , e th f co first half of the 16th century (Fawcett 1984, 182). One other late medieval insertion was the sacrament norte housth presbyteryhe n ei th wal f o l , althoug suggestioe hth bees nha n made- perhap littla s e over-ingeniousl y- tha t composita thi s i s e piece whic presenhs it coul e dow t appearance to post-Reformation reconstruction (McRoberts 1965, 52-3).

THE EXCAVATION

Finbar McCormick

NAVE (ARE , AILLU6 ) S2 reture Benedictine Prioth th f o nt ro e communit Pluscardeo yt 1948n i , nothin nave s th f ewa g o visible except for small segments of walls protruding from the crossing. This led to speculation thanave neved th t eha r been t builtuncommono a , n featur f Scottiso e h medieval churchest S t a , Monan Fifn Setod i s an e Easn ni t Lothia r instancenfo . However e clearancth , e areath f ,eo undertaken by the current resident Benedictines, uncovered the first course of dressed stone of the nave and the buttresses along its northern wall, as well as the loose base stones of the western doorway (referred to in R Fawcett, above). It was therefore clear that the building of the nave had at least been started, although the possibility that the walls had not extended higher than their basal course could not be discarded. Excavation was limited to the south-west corner of the nave, and nothing of this survived abov foundationse levee th eth f o l excavatione Th . s reveale foundatione dth f abouo s e t th hal f o f nave's west wall, a short segment of the foundations of the southern aisle, and the foundations of 404 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

two buttresses. The northern, and smaller, of these two buttresses presumably flanked the southern end of the west doorway, while the foundations of a larger buttress were present at the south- western cornechurche th foundationf e o r Th . s were 0.5-0. deep6m . They were mad digginy eb ga pit with rounded edges into which concret poureds ewa . This concrete consiste mixtura f do f eo morta smalld ran medium-sizo -t e stones. The upper surface of the foundations was sometimes faced with flat stones and these generally were shattered acutely, indicating that the foundations originally supported a great weight. This further implies that high walls had formerly existed on the foundations and, on balance, that the nave of the church had been completed.

SITE OF WEST RANGE (AREAS 1, 2 & 3; ILLUS 2 & 7) The wall bordering the west range (illus 2) is of unknown date and forms the western wall of a modern dormitory built since the return of the Benedictines in 1948. The wall clearly pre-dates presens i thisd Abbee planan n , o th t f so y mad19te th hn ei century . Ther evidenceo n s ei , however, to indicate that it is of medieval date or formed part of the original monastic west range. e areTh a immediatel e wes th f thi o to yt s wall containe standino dn beed g ha wallna d an s pathway until converted into a garden in the recent past. To the west of the garden was a rockery that abutte fieldda surface th , f whiceo s 1-1. hwa highe 3m r tha gardenne th tha subsoif e o t Th . l was a glacial gravel terrace. The south of the garden coincided with the end of the terrace which fell away steepl levea o yt l river plain smalA . lattee l excavates areth f wa rao produced dan e dth foundations of a medieval building (see below). The domestic range on the southern side of the cloister, and the south-east part of the eastern range, both extended beyond the edge of the terrace on to the river plain, and this difference in height allowed these buildings to have a basement. The glacial till was found to lie about 0.2 m below most of the garden area and, despite extensive trenching, no evidence was found for any substantial buildings in the areas excavated to the west of the cloister. One trench (Area 5) was extended westward into the adjoining field to explore the ground that would have contained the outer wall of a west range, if this had been of similar width to the eastern and southern ranges. No evidence for any buildings was found. It can therefor assumee b e d tha westero n t n range, contemporar same th en ye o scal witth d s ean a h souther easterd nan n ranges, existe sitee th .n do The foundations of an insubstantial wall (F4 on illus 2) were present, however, and extended westwards for a short distance. The wall, which is about 0.3 m wide, has a maximum height of about 0.2 m. It contains some re-used dressed stone and the eastern part consists of a single sill stone (F6), abou longm 1 t , whic thia bouns hi easters y nit b quarrallet d a nen d angle roll jamb stone (F7; illus 7), a type which Richard Fawcett (pers comm) states was most common during the 16th century. This jamb clearly pre-date wale f th sunknow o l n date that extends north/south, definin wese gth presene t sidth f eo t west range. This wall also doin n sealsi d preserveso gs an , a , lin f smaleo l flat paving slabs (F5), 0.2-0. square3m , which exten norte soutd th h o an dt f wal ho l likels i (illut 4 I F y. stha7) t thes remnantee slabpaveth o e tw s ar df so area s partitione wale th ly db (F4), with access betweeareaso tw e beinnth g provide dooe th ry db represente d e silan th l y db jamb. This wall, the door and the paving probably represent the remains of some late medieval building erected against a north/south wall, now superseded, that originally demarcated the western side of the domestic range. In a monastic house of this kind, the west range, when present, was usuall latese yth domestie t parth t constructef o t no s c rangbuiltwa e b d o det an , until pressure createe presencth y db f largo e e number f monko s laymed an s n mad t necessaryei e th s A . Valliscaulian generallt no d sdi y hav brethrey ela n (Flavell, above), havy thisema meant tha wesa t t McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, 5 MORA40 Y

jamb (F7) modern wall modern wall

paving slabs (F5) -sill (F6)

-wall (F4) 2m

ILLUS 7 Area 3 plan

t needeno rang s dewa durin primare gth y perio f occupatiodo t Pluscardenna wesA . t rangs ei similarly absent from Dryburgh Abbey. Ther survivine eth g remain more sar e extensive than those at Pluscarden, and the cloister is bounded on the west by a featureless wall distinguished only by a lavator smald yan l south-westers dooit t a r (Richardsod nen Tabrahan& m 1983 . Som4) , e small buildings were erected against the western face of the Dryburgh Abbey wall in late medieval times thesd an , e provid gooea d paralle structuree th o t l s note Pluscardent da .

ARE) A9 (ILLU4 & 8 , S2 The southern part of the monastic complex was built on the river plain, at a lower level than the remainder of the buildings. As stated above, this included the fragmentary southern domestic rangeeastere th southere s f welth a o ,n s a d lclaustra nen l range. Extending eastwards froe mth latter are a series of ruinous, stone-vaulted buildings which were not included in the incomplete, and slightly inaccurate, 19th-century plan on which illus 2 is based. Prior to excavation, part of the foundation buildina f o s g orientated east/west coul e south-wes e seeth db o e claustrant th f o t l range. Excavatio eastere thif o th sf d nbuildino n en g uncovere dseriea f wall so steps d san . Two periods of construction were represented and can be summarized as follows:

Phase 1: This consisted of the building of wall F24 with a door sill (F25) and a flight of steps (F26) outsid doore eastere th Th . n step e edgflankes th f swa e o wal similaf y whic4 o db s lF3 hwa r width to wall F24.

s widenePhase e additiowa : th Wal 4 2 coursa y db F2 f l no f stoneo e s F28s .wa Wal4 F3 l demolishe stepe replacetopd e th leve e widee f th san s th o f th o lo dy t rd b wall F27, which covered part of the eastern edge of the steps.

The broad chamfered sill (illus 9) can be dated, on stylistic grounds, to the 13th or early 14th century, but in the context of Pluscarden, a middle to late 13th-century date is more likely 406 j SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

jledge of wall (F34)

line of earlier wall paving stone ILLUS 8 Area 4 plan

(R Fawcett, pers comm) t displayI . s practicall weao yn r which suggests that eithe buildine rth s gwa never finished, or that it fell into disuse in a very short period of time. The steps begin immediately outside the door and lead to a level area. It seems likely, however, that a second flight led from this level area up onto the gravel terrace. This hypothesis is supported by the fact that the last 1.5 m of F24 has a rough outer face that originally would have been masked by a second series of steps. A pil f rubbleo e (F30) als againsy ola t thi walse parth lf o t(an d underneat wale hth l thickening F28) and probably represents the foundations of such a flight of steps. The only artefacts from the rubble were several sherds from the same pot (Find 19), which Murray (below) suggests- is of 13th/14th-century date piecee Th . s were found togethe survived an r largs da e fragments compared with other pottery from the excavation. This implies that the pottery was not redeposited, and is probably contemporary wit Phase hth buildinge parth e1 f to . e wal 7 sealee basth F2 lf Th eo grouda stepse th f animaf po .o p to l bonee th n so thay la t These consis f severao t l bones eac f cattleho , sheep/goa pigd an t, along with three oyster shells (see fiche repor r completfo t e list). Withi foun s fabrie nwa th f fragmenwalda 7 o c F2 l f glazeo t d strap handle (Pottery ) whicCat2 o .hN Murray (infra) state s probabli s a 14th/15th-centur f yo y date. e necessite reasoth Th r nfo f widenin yo wallso tw e , gth wit e accompanyinhth g partial destruction of earlier features, is unknown. It is possible that the east/west building was never completed, a hypothesis supported by the absence of wear on the sill.

BURIALE TH S (ARE ; ILLUA5 S 10) A grou f threpo e extended inhumation founds swa , close together, south-wes westere f th o f d o t nen the nave. Further excavation to the south, north, east and west of the inhumations failed to produce McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, 7 MORA40 I Y

' 'Ik*

ILLU S9 Are afro4 m sout beed hha n afte 8 removeF2 r d

any more burials buriaa t s founbu ,wa l d extending intbaule oth k immediately north-wese th f o t group, this burial having been truncated by the stone-lined pit F22. buriale Th s wer vern ei y shallow graves immediately unde gardee th r n soil, with grave fills of very similar soil type e graveTh . s wer l aligneeal d NE/SW with their lattee headth t ra s end. This orientatio s similani abbee th o yt r church sampl(illuA . 2) sf bon eo e froms wa buria 1 F2 l submitte r radiocarbodfo n dating. This provide datda f 490±5eo (GU-2983p 0b ) which calibrates 68.9a o t % probability rang f Aeo D 1405-146 95.9a r 0o % probability rang f Aeo D 1300-1475 (Pearsonera/ 1986). The burials are an isolated group, and do not constitute a cemetery. They are of identical orientation and do not cut each other. In addition, there was no indication of any surface markers site th e n (althougo h consequent disturbance could have removed these) t seemI . s likely, therefore, 8 40 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

\ McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY. MORAY 409

thaskeletone th t s thif represeno se buriaarea us s e aa tth l groun limitea r dfo d period. Indeeds i t i , temptin suggeso gt t thaburiale th t s occurre same th t eda time. Possible suppor thir provides fo tsi d in that two pieces of pottery - one found amongst the bones in Skeleton F16 and one among the bone Skeleton si - wer 1 enF2 adjoining sherd f 13t so r 14th-centur ho y date (Potter& 5 y s CatNo . implicatioe Th . 6) s thani shalloe th t w grave same th st e a wer time g thad spoie edu an th t, l from the graves contained some earlier pottery, whic thes hnwa broke backfilled nan d intdiffereno otw t graves. The radiocarbon date indicates that the burials are almost certainly of pre-Reformation date and probably date to the earlier part of the 15th century. skeletone Th s were reinterre modere th n di n graveyar t Pluscardeda Augus3 n no t 1991.

THE STONE-LINED PIT (F22) (AREA 7; ILLUS 11) Lying immediately to the north of the burials was a partially stone-lined pit. It post-dated the burials as it cut an unexcavated skeleton at the junction of the femur and tibia. The pit survived to a maximum depth of about 1.45 m, and was roughly 2 m long with a maximum width of 1 m. The northern and southern sides of the pit were lined with stone walls to their base. The western edge unlines wa consisted dan straigha f d o vertica d an t l face througt cu , h concreted glacial gravelse Th . easter t edgenpi , again unlined probablroundes s wa wa , d originae dan yth l t (seentrancpi ee th o et below). Most of the layers in the pit were sterile but layer F39 was a striking exception. It contained animal bones, glass and, most unusually for free-draining soils, a small sample of organic matter (F23) uppee Th . r levellayee th f rs o containe articulatee dth d partial skeleto middle doga f Th no . e part contained a fist-sized deposit of organic material, comprised mostly of wool, which did not

N t NW

Pit (F22)

F41 F35 Very dark brown (10YR 2/2) sandy loam F42 Limit of F36 Very dark grey/brown (10YR 3/2) stony sand excavation F37 Dark yellow/brown (10YR 4/4) stony sandy loam F38 Very dark grey/brown (10YR 3/2) stony loamy sand Truncated F39 Dark brown (10YR 3/3) stony loamy sand human skeleton F40 Very dark grey/brown (10YR 3/2) stony silty sand F41 Yellow/red (SYR mixeh 4/6as ) d with sand 0______1m F42 Dark brown (10YR 3/3) sand F43 Dark yellow/brown (10YR 4/4) stony loamy sand

ILLUS 11 Plan and section of Pit F22 410 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

extend either to the sides of the pit or the drawn section. The organic material also contained pollen, parasite ova, beetle and fly remains and fragments of feathers. Within the organic context, surroundine th an n di g sterile soil, were fragment f windoo s w glass .belo m Aboum e wth 0 5 t organic deposit was a partly articulated cat skeleton. Other mammal, fish and bird bones were also presen layere artefactuae th n Th i .t environmentad an l l finds from this layer (F39 discussee )ar d below boneg Do . s from this layer produce uncalibraten da d radiocarbon dat f 425±4eo (AAP 5B - 8792) which calibrate 68.9a o st % probability rang Ao et D 1430-6 95.9%a r 9o probability rango et AD 1410-1614 (Pearson et al 1986). Most of layer F39, however, was sterile with all the finds being in the immediate vicinity of roundee th pite dth .easterf Thio d sn en suggest probabls wa t spi thaye roofetimth e t th et da tha t this materia depositeds wa lentranc e th thad roundes e onld an th ,t wa e an y d meanden f accesso o st pite th , thus accountin vere th yr localizegfo d distributio findse th f n.o

SERVICE TRENCH OUTSIDE NORTH TRANSEPT

Dom Ambrose Flavell OSB In December 198m5 0. wid 7a e service trenc rainwate w carro ht ne ya frog r du mdraina s wa , point just west of the north-west corner of the north transept to a previously constructed soak- north-ease away th eas m f o 6 t, t cornechoie paralleth n norte f ra ro rth t (illuI ho . t l transep2) s t wall at a distance of 0.5 m from the outer face of the buttresses, turning south-east when directly opposite the north-east corner of the Dunbar Vestry. The depth of the trench, which took advantage of the downward gradient of the ground surface, was a fairly constant 0.6-0.8 m. Of the three buttresses along the north wall of the transept, only the one in line with the transept aisle arcade projected into the trench. This was a solid construction of large unworked stone block rubbl d mortarn i san t e se natura t e resteI . th n do l orange-brown sandy-clay subsoil, aboum 1 t below present ground level subsoie Th . l shelves more steeply tha groune nth d surface, increasing from a depth of 0.25 m at the western end of the trench to 0.6 m opposite the Dunbar Vestry. stratigraphe Th y outsid norte eth htransepe followsdoos th a f s ro wa t : 0.1-0.1 f topsoio 5m l overlying loose angular stone rubbl dara n eki brown soil matrix which included modern bottle glass depta t f 0.2-0.2A .h o belo5m surface w th wha s ewa t appeare sealea e b do dt layer , 20-100 mm thick, made up of very compact crushed sandstone chips mixed with earth. Within this layer, on line with the middle lancet of the north transept window, were found window glass and leading, two sherd f green-glazeso d pottery, fragment f mouldeo s d ston fleckd ean f charcoalo s . This layer was almost immediately abov naturae eth l subsoil. Shallower trenches were dug to connect the gutter downpipes to the main drain. One of these revealed thawese th t Dunbate walth f o l r Vestr restins yi g directl foundatione th n yo earliern a f o s , and larger, building. No other structural remains or graves were uncovered.

WATCHING BRIEF REPORT OF WORK UNDERTAKEN IN THE SOUTH-EAST VAULTS

Dom Ambrose Flavell OSB The priory was built on the edge of an old river terrace and consequently the south range and its extension running eastward had to be built on a series of vaults to bring the floor to the level of the cloister foue Th .r collapsed l thavaultal te surviv sar range th f eeo juttin t fro south-ease gou m th t McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 411

corneease th tf rangero second-westernmose Th . thesf o tbees eha n give temporarna ys i roof d an , used as a potato store by the present community. A new floor was laid in 1989 which required the vault to be dug out. The deposit in the northern half of the vault was evidently modern (ash, coal and coal briquettes), and rested directly e naturaoth n l subsoil. There wer indicationo n e f pitso s , post-hole r similao s r features. Archaeological layers surviv southere th n ei vaulne werd halth t removedf an t o feno . A strip 1.5m wide, 4 m from the north wall and parallel to it, was cleaned down to natural. aren a f burn founds ao n i wa t A t se m ,dau fla 4 tb 0. ston (possiblx m e slab 5 n base a 0. y th f c ,eo open fire )thick m aboum .0 2 Arount stone dth e slab, above this surface browa s wa n, silty clay with some animal bone irod snan fragments, also some white-glazed pottery. This vaul s directli t y opposit Prior'e eth s Lodging, which remaine occupation di n untie th l 18th century (Macphail 1881, 177).

SPECIALIST AND ARTEFACT REPORTS

HUMAN BONES

(A full report is in microfiche)

Daphne Home Lorimer The three burials all appear to be females, aged about 17-18 years (F21), 25-35 (F16) years, and 35-45 years (F20 t possibl t timno a ) s f deathewer6 o calculato wa et F1 t eI . d bot staturan e heth 0 abouf F21F2 eo t t,bu 1.6 (5f2m t 4in. heightn )i ageine Th .sexind skeletone gth an f wore go s base th f Baswa sko n do s (1987); Gray (1977 uses generar )wa dfo l anatomical reference. The bones were very fragmented and abraded. Few non-metrical variations could be recorded in the skull (see fiche), and almost none in the post-cranial skeleton. All three skeletons, however, exhibited platymeri r anterio-posterioo a r flattenin femure th Poirier'd f o an g, se e neck faceth s th see f n wa o stno F16d an . femor0 ThesF2 f eo a were relatively common feature formen i s r population consideree ar d san d indicative of activity over rough or hilly ground (Angel 1971, 85); squatting facets on the right tibia of F20 suggeste activite dth y might have involved acute dorsi-flexio fooe th t f (Chestermano n 1983; Bacon 1990). The abraded condition of the bones precluded observation of many superficial bone changes, but there was possible reactive bone tissue on the right superior articular facet of the second cervical and possible hyperostosis on the spine of the fifth lumbar vertebrae of F21 (Ortner & Putschar 1985). Owing to the fragmented nature of the facial bones, many teeth were loose, but caries was only found teet3 8 e h th iretrieved nf o eigh t teeto beeou d t Tw hha . n lost ante-morte alveolae th d man r margin remodelled (F16). There was no evidence of calculus or periodontal disease. In F21, however, hypoplasia was seen on the upper incisors and all canines, suggesting a systemic disturbance at about three or four years of age.

MAMMAL BONES

(A list of the bones recovered is in microfiche)

Finbar McCormick All the bones in the stone-lined pit (F22) were found in layer F39. The bones for the most part consisted of a small grou f cattlepo , thos f sheep/goaeo occasionan a d an t boneg pi l . These were generally chopped into small fragments indicating that they represent discarde boneg ddo foosd presentan d t refuseca e , howeverTh . , 412 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

ILLUS 12 skeletog Partdo f so 2 nF2 fount pi n di

unusuan a e ar l group. There wer remaine e th articulate n a f so d skeleto doga f no , comprisin heade gth , vertebral column, sacrum, ribs, both sides of the pelvis and one scapula. The four legs and the tail are missing (illus 12). beed ha n g killedo beine y db Th g struck severel foreheae th n yo d (illus ISA). Knife marks alonge nexth d , an ,to t pelvie th lip f so acetabula indicate thaanimae lege th th ts f werof l t (illuecu woulg s do 13s) e do t hav Th . d eha skinnee b orden di facilitato t r e cutting t seemi t bu ,s t likeleviscerateno s y thawa t i td before depositioe th s na penis bon stils situ.n ei lremovewa s Evewa t nca d e morpriodepositios th it f eo ro t pit e th . n i The removal of the limbs of the cat and dog cannot be accounted for simply by the suggestion that they had been incomplete skinnedth y b r o , entireles excavatiowa t ypi featuree e emptiedth th f no s a , . Skinninn gca sometimes account for knife marks on the pelvis. Such marks were present, for instance, on a dog pelvis from medieval levels associated wit Benedictine hth e monaster t lonya a (McCormick, forthcoming) than I . t instance, they wer iliume shafe th founrepresend th f o an ,tn do skinner'e th t s cuttin pele th t f aroungo sphinctee dth o rt avoid contact with excrement. Knife marks have also been noted on the greater trochanter of a cat femur from Rattray Castle in Aberdeenshire (Murray & Murray 1993, 204). The cutting at Pluscarden, however, represents the deliberate separation of the legs from the torso. It is difficult to avoid the conclusion that the limbs of the cat werang ddo e being remove culinarr dfo y purposes othey limbe an possible th r f purposeTh .sfo o e eus , sucs ha the utilization of the bones as a raw material, is so remote that it can be discounted. Both the front and hind limbs would have contained a relatively substantial quantity of meat compared with other parts of the carcass. muse On t assume that this practice reflect perioa s extremf do e food shortage.

| SOCIET 4 41 Y OF ANTIQUARIES OF SCOTLAND, 1994

BIRFISE D TH HDAN BONES

Sheila Hamilton-Dyer Eigh bir4 1 t d fisd bonehan s were recovered from layer F39 fise .hTh t werleasa f eto three speciesd Co : (Gadus morhua), Hake (Merluccius merluccius) and Gurnard, probably Grey (Eutrigla gurnardus). The latter was represented by a single dorsal spine matching comparative material for a fish of about 300 mm total length. The Hake remains consisted of two, probably adjoining, caudal vertebrae. One of the Hake vertebr verticalls awa t mid-centrumycu remaind Co e sTh . wer caudao e alstw f oo l vertebra. There were four other fragments, including three righe boneth f t o spectora l girdl gadoia f o e d fish, probably also Cod. bire dTh bones consiste f fivdo e domestic fowl bones goose on , e bon eighd ean t passerine bones. e fowTh l bones were fro t leasma individualso tw t e probablon , ymalea , wit tarsometatarsua h s total length of 77.5 mm. Another tarsometatarsus was from a much younger bird. There were also an ulna, femu sternumd an r femue sternud Th . an r m both showed knif t markse cu goos e Th . e humerus could either have been fro mdomestia c Greylaa bir r do g (Anser anser).severad ha t I l small cuts roun e distadth l articulation passerine th f o e e On femubonea . Thrush-sizea s f o wa rs d bird othere th , l beinal s g corvid. These were probably from one individual, and compared favourably with Jackdaw (Corvus monedula). Bot hwerlefe g wingle th t ed representedsan t therbu , e wer footo en , hea r trundo k bonese vien th I . f wo butchered dog and cat bones, the possibility that the bird was eaten must be considered. Only the breast of corvid consideres si d otheediblee th l r al part, s being discarded before cooking. However heae d th dan s a , feet were missing, and there were no cut marks present, it seems more likely that the bones do not represent food waste.

FEATHERS FROM DEPOSIT F23 IN THE STONE-LINED PIT (F22)

Ruby Ceron-Carrasco organie Th c deposit fro stone-linee mth t containedpi d large number f fragmentso f possiblo s y chopped feather, perhaps originally the contents of a cushion or some padding in clothing. Initial comparison with reference material donated by the Natural History Division of the Royal Museum of Scotland suggested that the feathers were probably those of a Pheasant. The results were then sent to Dr Karin Perremans of the Laboratory of Ecology & Aquaculture in the Zoological Institute in the Katholieke Universiteit, Leuven for her comments. Initial study indicates that the feathers displayed downy barbules possessing ringlike annuli whic r PerremanD h s stated '...coul e encounteredb e th n i d Tinamidae (Tinamiformes) e Phasianidaeth , Tetraonidaee th , e Numididaeth , e Meleagridaeth , e th , Cracidae (Galliformes Musophagidae'e th d an ) vers i t yI . unlikely, however, that mos f theso t e would have been encountered in a Scottish context. The remains were then studied by means of a scanning electron microscop r PerremanD resulte y b eth d ssan corresponded completely wit Pheasanhe thath f o t t (Phasianus colchicus). The presence of pheasant remains in a 15th-century Scottish context is extremely interesting. As e writeth fas s awarei a rr , their bones have never been foun a Scottis n i d h medieval e contexth d an t earliest reference to their presence is in 1578 when Bishop Lesley stated that pheasants were present, but scarce, in Scotland (Lever 1977, 337-8). A Scottish Act of Parliament passed in 1594 stated that no one should kill 'Phesanis, foullis, partrikis' in or near the King's hunting parks (ibid). Pheasants were essentially park faunahuntina d an , g forest existe t Pluscardeda n fro e mid-13tmth h century (Gilbert 1979, 339). Lever (1977, 335-6) noted thapheasane th t introduces wa t d into England befor Normae eth n conquest and seems to have become quite common by the 12th or 13th century. In Ireland, the Anglo- Norman introduced ha s e pheasan dth e 13t th hy b centurt y (McCormick 1991, 49)s quiti t I .e likely, therefore, that examples of the pheasant, earlier than that noted at Pluscarden, will be discovered in Scotland. McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, MORAY 415

MICROFOSSIL ANALYSIS OF THE ORGANIC DEPOSIT F23 IN THE STONE-LINED PIT (F22)

Corali MilleM s A complete report on the material, which includes a description of the methodology, is presented in microfiche and the results of the analysis are given in Table 1. Preservation was generally very good and the concentratio f polleno spored sample nan th abous n si pollee ewa Th t . 67,00n ml spectru r 0pe mdominates i d by cereal-type pollen, mostlHordeume th f yo grou t witpbu h some attributabl Avena-Triticume th o et group (groups as defined by Andersen 1979). The size of some damaged Gramineae grains indicated that they were of cereal-type, but these could not be attributed to a particular 'Andersen' group. In total, cereal-type pollen represented 71.0% of the Total Land Pollen (TLP), and wild grass pollen represented a further 13.2% TLP. About half of the pollen attributed to the wild grass group was at the upper extreme of the size range for this group (Andersen 1979), and it is possible that some pollen attributed to the wild grass group may be of the Hordeum group. A rang f taxhabitaed o aan t type representee sar remaindee th n spectrumde i th f o r , with trees, shrubs, heathland plants, weeds, ferns aquaticd an , l presensal smaln i t l quantities.

Discussion The organic deposit investigated here was found in close association with fragments of cloth, and close to a deposit of animal bones. The analysis of parasite ova (Pennington et al, infra) implies that human faecal matte importann a s ri t componen organie th f o t c deposit. Cereal-type pollen is the predominant type in the sample, representing 71% TLP. Most of this was attribute Hordeume th o t d group (Andersen 1979) which comprise cultivateo tw s d species, Hordeum vulgare (barley) and Triticum monococcum (eincorn), and a number of wild grasses. The Avena-Triticum group (Andersen 1979) includes only one wild grass (Avena fatua), a weed of arable fields, and several cultivated specie f wheao s t (Triticum. t (Avena sppoa d )an spp) e presenc .Th Avena-Triticum e th f eo type pollen in the Pluscarden deposit therefore indicates that at least some of the cereal-type pollen derives from arable activities. Given the presence of parasite ova indicative of human faeces (Pennington et al, infra), mose th t likely origicereal-type th f no e polle s froni m human food suc broths ha . Other pollen typey sma derive fro humae mth n faeces; aquatic havy sma e entered through drinking water othed an , r havtaxy ama e been 'contaminants' of food and drink. Microscopic charcoal particles may also have been ingested with food. The absence of macroplant remains (Buckland, pers comm) means it is not possible to be certain which, if any, cereal plants are represented by the pollen. Given the large number of cereal-type pollen present, it is surprising that no bran fragments were present. High percentages of cereal-type pollen, in the rang f 34-50eo % TLP, were foun peata n di y latrine deposi t Worcestera t . Bran, however alss owa , presenn i t the Worcester example (Greig 1981, 267). The Pluscarden deposit contained a much higher proportion of cereal-type pollen tha humae nth n coprolites from Warebeth Broch, Orkney, where Hordeum group pollen represented 1.1-5.6% total pollen and spores (Dickson 1989, 117, and Table 1). These coprolites contained animal hairs and bone fragments, and were interpreted as representing a largely meat-based diet (Dickson 1989). It is tempting to interpret the Pluscarden deposit as representing a restricted vegetarian diet at the Priory; however, the absence of the meatier parts from the dog and cat carcasses, also found in the pit, may indicate otherwise (McCormick, infra). s possibli t I e that some pollee Pluscardeth n ni n deposit derives from sources other thae th n human faeces. These other sources could include water-seepage throug air-falpite d hth an , f polleo l n fro surroundine mth g vegetation e rang Th .f taxeo a indicate some heathland, woodland, grassland dan vicinitye t areath we n e possibiliti s Th . s considereywa d thadecayine th t g e pitrefusth , n includini e g the dog and cat corpses, may also have contributed some pollen. It is known that in medieval times, hounds were usuallbrean o r sometimed do fe y brotn o s h (Cummins possibls 1988i t i ,o s 26-7e d an ) that cereal-type pollen could derive from the dog's digestive tract. Such a find is not unprecedented; 416 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

TABLE 1 Results of microfossil analysis of the organic deposit (F23) from the stone-line pit at Pluscarden. Percentages of the Total Land Polle Totae th nf l o (TLP Polled Spored an ) nan s (TPS givene ar ) :

COUNT PERCENTAGE TREES AND SHRUBS TLP TPS A/nus 4 1.0 Betula 1 0.2 cf. Castanea 3 0.7 Corylus/Myrica 13 3.2 Pinus 1 0.2

HEATHS Ericaceae undiff. 17 4.2 Calluna vulgaris 6 1.5 cf. Erica cinerea 1 0.2 cf. Erica tetralix 10 2.5

GRASSE SEDGES& S Gramineae 53 13.2 Cereal-type, Hordeum group 239 59.3 Cereal-type, Avena-Triticum group 37 9.2 Cereal-sized, crumpled & undiff. 10 2.5 Cyperaceae 1 0.2

HERBS Caryophyllaceae 2 0.5 cf. Chrysosplenium 2 0.5 cf. Crassulaceae, Umbilicus type 1 0.2 Lotus type 1 0.2 Plantago lanceolata 1 0.2

TOTAL LAND POLLEN (TLP) 403

AQUATIC Potomogeton, section Eupotomogeton 0.5

SPORES Filicales undiff. 0.5

UNIDENTIFIED POLLEN Broken 11 2.5 Crumpled 13 3.0 Degraded 4 0.9

TOTAL POLLEN & SPORES (TPS) 435

CHARCOAL PARTICLES 10-2n 5ur 163 25-50 |lm 92 50-7n 5ur 80 >75 (in 58

Key: undif undifferentiate= f d McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 417

Hordeum-lype polle bees nha n founaniman a n i d l coprolite from Warebeth Broch (Dickson 1989, 118). The evidence from the Diptera (Skidmore, infra), however, suggests that the organic deposit was not in close contact with r formeo , , rottindby g carcasse s unlikeli t i o s y thapollee th t s deriveni d from such a source. s alsi t oI possible tha t Pluscardee somth f eo n pollen derives from refuse whic s sinchha e decayed completely; the former presence of crop-processing waste, for example, could lead to an enrichment of cereal-typ ee introductio polleth d nan f otheno r pollen taxa (Robinso Hubbarn& d 1977) e absencTh . f eo macroplant remains at Pluscarden means this cannot be proved or disproved.

A BEETLE FAUNA FROM THE ORGANIC DEPOSIT F23 IN THE STONE-LINED PIT (F22)

P C Buckland The sample was decidedly rich in insect remains, with large numbers of fly puparia (Skidmore, below), and many well-preserved beetle remains e largon ; e spider, probabl a yTeganaria sp. s als,wa o notede Th . Coleopterous fauna (Tabl dominates i ) e2 threy db e species: Tipnus unicolor, Rhiwphagus parallelocollisd an Quedius mesomelinus. The presence of larval remains and ill-formed adults further suggests that this assemblage forms the core of the indigenous fauna. The flightless spider beetle, Tipnus unicolor, is represented by a minimum of 126 imagines. The species is largely synanthropic, ie lives in association with e northermanth n i , nranges it par f o t , although Crowson (1962 )leak recordoa f n littei t woodlani sn i r n di south-west Scotland. Zacher (1933) suggested tha . unicolorT t feeds largely upo e faecenth f rodento s d an s small birds, but Woodroffe (1953) failed to find it in his extensive study of the insect fauna of birds' nests. Its frequent occurrence in barns, in straw and hay waste (Koch 1970), as well as in cereal debris (Howe 1955), may reflect its association with small mammals, but its fossil record, in moderate numbers in archaeological contexts, where ther synanthropiy fewe an ar e f ,i c rodent r examplfo - s southern ei n Iceland (Perr t atye 1985 - suggest) more b y es toleranthama dietart s i tit n i t y requirements mann I . y archaeological deposits, particularly from the less liquid domestic ones, T. unicolor is usually the most frequent Ptinid, and its presenc t Pluscardeea s likelyni t leasa , t initially resulcasuas a it e f b o t o lt , impor many b t . Osborne (1983, ibid)s suggesteha d that archaeologican i , l contexts mors i t i , e frequently foun featuren di s interprete cess da s pits. Similar problems atten e habitath d t data availabl e nexth tr mosfo e t abundant species, Rhizophagus parallelocollis, with 56 individuals. Its vernacular name, the coffin beetle, associates it clearly wit s mosit h t frequent occurrence ,e actuaeveth f i nl food source remains ope r discussiofo n n (Peacock 1977). The most widely quoted primary source, that of Megnin (1894), notes that the beetle appear n largi s e number n exhumeo s yearo d tw corpse so t old p u s , wher t probabli e y preys upoe th n larvae of Phorid flies (Johnson 1963). It is also recorded from beneath bark and in fungi (Peacock 1977). beetle'e Th s apparent association with decaying bodie alss si o eviden archaeologicae th n i t l occurrences, wher t appearei largn si e number ranga n si f medieva eo l burials (Buckland 1979; Girling 1981; Stafford s presenc1971)it e t Yorth bu , kn i e Roman sewer (Buckland 1976) would suppor s predatioit t n upon other larval Diptera, in this case probably the sewer flies, Psychodidae. The beetle, in fact, is not infrequen urban i t n archaeological contexts where particularly foul conditions migh e assumeb t d from other lines of evidence, for example in Anglo-Scandinavian York (Hall et al 1983), and medieval Winchester (Grove, pers comm) e abovTh . e medieval l burialwithial e nar s standing buildings and, despit e foueth l conditions whic a rottinh g corpse might offer e indicatiof relativelo th , e on y s i nydr decay. R. parallelocollis is associated with the rove beetle Quedius mesomelinus in the lead coffin of Archbishop Greenfield (obit 1316 Yorn i ) k (Buckland unpublished) thid an s, fairly eurytopic predatos i r also frequen Pluscardee th n i t n material. Species of Aleochara predatore ar larvan yo l Diptera (Good& Giller 1991) and the mould-feeding Lathridiids and Cryptophagids are widespread in a range of mouldy plant debris rather than human waste. Whilst the comparative data for the Pluscarden fauna largely concerns human burials, the insect fauna doe t discriminatno s e othed betweean rn animalma n s burie n similadi r conditions. Taken wite th h 418 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

TABLE 2 Beetle fauna from F23 in the stone-line pit

COLEOPTERA Carabidae Trichocellus sp. (burnt) 1 Leiodidae 2 Catops . sp Ptiliidae 1 ?Ptenidium . sp Staphylinidae ?Acrolocha sulcula (Steph.) 1 Xylodromus concinnus (Marsh.) 1 Philonthus sp. 1 Quedius mesomelinus (Marsh.) 16 Aleochara spp. 6 Aleocharinae gen. indet. 30 Rhizophagidae Rhizophagus parallelocollis Gyll. 56 Cryptophagidae Cryptophagus scutellatus Newm. 3 Cryptophagus spp. 7 Atomaria spp. 6 ?Ootypus globosus (Waltl.) 1 Lathridiidae Lathridius minutus (L.) (group) 4 Corticaria . sp 13 Ptinidae Tipnus unicolor (Pill. & Mitt.) 126 Scarabaeidae Aphodius rufipes (L.) 1

archaeological evidence greatee th , coleopteroue r parth f o t s faunal assemblage coul related e partlde b th o yt butchered dog and cat found in the pit, but Skidmore's evidence from the Diptera is as much diagnostic by its absence as its presence. Any significant accumulation of carrion, even where buried, attracts a characteristic assemblage of flies (cf Smith 1986). Both relatively dry and wet stages of carrion decay (Diptera, Calliphorids, Phorids and Piophilids) are absent from the pit, suggesting that the butchered animal bones were relatively clea disposaln no insece Th . t fauna appear supporo st interpretatioe th t n thastone-linee th t d feature was used as a cess pit. Amongs remaindee th t faunae th f o r, onl burne yth t elytron (wing case f Trichocelluso ) y ma . sp provid littlea e additional archaeological informatio thatwetlann a i n s a , r peatlando d species likels i t i , o yt have been introduced in peat or turves for burning.

ANALYSIS OF FLY REMAINS FROM F23 IN THE STONE-LINED PIT (F22)

P Skidmore An interestin highld gan y specialized assemblage comprisin identifiabl9 g43 e dipterous specimens wa s present in the organic deposit (Table 3). Virtually all the material consisted of puparia and there were few identifiable fragment f adultsso . By far the most abundant dipteron present was the troglophilous leptocerine sphaerocerid, Terrilimosina racovitzai (Bezzi). The sphaerocerid puparia present all appeared to be identical, but it is possible that other species were present sinc puparie eth f moso a t species remain undescribed eved an ,n several which have been figured were clearly based upon misidentifications puparie Th . f manao y species, in any case, are probably not distinguishable by external characteristics. The identification of the McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, MORAY 419

Pluscarden material rests primarily upon onlunemergeo ytw d males amongst five unhatched examples from a total of 403 puparia. The spermathecae of the three females agreed with T. racovitzai, but these are of less diagnostic significance since several species possess very similar structures. Three sphaerocerid wings, and 18 adult head capsules, also agree well with description and figures for T. racovitzai in Rohacek (1983). The wing size and venation, notably the slight prolongation of the costal vein, agree very closely with Rohacek's figure. T. racovitzai a specialize s i d species havin reducee gth d eyes typica mano s f o yl troglophilous acalypterates. The male is unique amongst described European leptocerine sphaerocerids in the remarkable bifurcate median process and lateral curved teeth of sternite 6, features clearly visible on the male puparia from Pluscarden. Whilst originally described from caves, T. racovitzai has also been recorded from mole burrows, but accordin Rohaceo gt k (1983 largese th ) t populations occu cellarn i r s containing large amount f decayino s g vegetable material, althoug alss i t hoi recorded from carrion. Heleomyzids were presen modesn i t t numbers (19 empty puparia t thei)bu r generic identity coult dno be ascertained positively. Two species appeared to be present, both very close to Heleomyza serrata, but four wer densela ed largeha yd spiculatran e 'anal segment', somewha Scathophagan i s a t Hylemyar o (but lacking the perispiracular papillae). Several heleomyine f subterraneao e ar s n habi Scoliocentra,g (e t Eccoptomera, etc)eved more an ,n th e eurytopic Heleomyza serrata (L.) often abound largee caven i sth r ro s burrowf o s mammals. Lik . racovitzai,eT heleomyzine t predominantlno e sar y associated with carrio t exceptionnbu e sar know Neoleriae n(i possibld an y Scoliocentra species).

TABLE 3 Dipterous remain stone-linee th sn i fro3 t mF2 d pi Specimens present

1 Trichocera larval head capsule 6 Psychodid pupal fragments 403 Leptocerine Sphaerocerid puparia (5 unhatched) Terrilimosina racovitzai (Bezzi) 3 Leptocerine wings, identica abovo t l figures e(a Rohacey db k 1983) 18 Leptocerine adult head capsules identical to above 19 Leptocerine thoraces, probabl abovs ya e 1 Phorid puparium (empty) 14 Heleomyzine puparia (empty) smaller with non-spiculate 'anal segment' 5 Heleomyzine puparia (tempty) larger with spiculate 'anal segment' 5 Fannia puparia (empty) (probabl Scalaris. yF (F.) 3 Muscid puparia, probably a Helina sp. close to H. reversio (Harris)

Fragments of eight Fannia puparia all appeared to belong to the same species, probably F. scalaris (F.). This species is a polysaprophage, which is very doubtfully recorded from carrion, but which probably prefers dune th f omnivoresgo populae th s A . r nam f 'latrineo e fly' suggests, thi commonls si y recorded from human faeces, and formerly abounded in privies, before the advent of the flush toilet. Whilst the adults of T. racovitzai, and possibly the heleomyzines, could survive and breed in total darkness, Fannia knowns i s speciesa r ,fa requiro s , e acces sunlighto st malee th , s havin highlga y developed nuptial swarming habit. psychodix si e Th d pupae infe t leasa r partialla t t organiywe c biotope, whils three th t e puparif ao predatory muscids, tentatively referre genue th o dst Helina, bree ricn di h humus soil. It is noteworthy that, apart from the Fannia, there is a total absence of any heliophilous polysaprophagous elements in the fauna such as, for instance, Piophila, Madiza, Drosophila, Calliphora, Sarcophaga, Muscina, Hydrotaea, Musca r Stomoxys.o Indeed habitae th , f taccumulate o inferre e on s i d d faecal or decaying plant material lying in permanent darkness. None of the taxa present unequivocally points e presencth o t s possiblf i carrioneo t i d ean , thawhicy havy an t hma e been presen havy ma te been uncontaminated by flies when it was cast into the pit. 420 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

PARASIT FROA ORGANIE EOV M TH CSTONE-LINEE DEPOSITH N I 3 TF2 (F22T DPI ) (A complete report, including the methodology, is presented in microfiche) J B Pennington, J P Middleton, C Nicholson & A K G Jones A sampl takes ewa n from organic material associated with some woolle teso e t n th t2 clot F2 ht founpi a n di hypothesis that the cloth was used as an anal wipe before being discarded. The sample contained abundant Trichuris ova. The counts averaged 20 ova per sample. The ova were not very well preserved - none had polar plugs. The mean size of the ova was 50.6 microns long (length minus polar plugs 27.y b ) 5 microns wide (sample size 10). Egg f thio s s sizproducee ar ehumae th y db n whipworm, T. trichiura. The whipworm of rats and mice, T. muris, and that of pigs, T. suis, also produce eggs in this range. Other trichurid parasites found in the British Isles (71 vulpes, dogs, T. ovis and T. globulosa, sheep, goat cattled san ) produce much larger eggs Ascariso N . wera ov e identified. numbee observea Th ov f ro d indicates thasample th t e contains faecal material Trichurise size th f eTh .o a ov indicate that faece moss shumawa f t o likel e b nyo t origin , although other specie havy sema been responsible. t seemI s likely thaclote th t h associated wit sample hth uses r sanitarewa dfo y purposes. Howevers i t i , possible tha e clothavy th t ma he been introduce e featurth o dt e independentl e organith f yo c matted an r subsequently became incorporated wit. hit

THE SOILS FROM PIT F22 (The analytical method tabla d f resulteo san s appea microfichen i r ) Stephen Carter Of the nine samples analysed, eight are fills of the stone-walled pit, and one (F43) is from the backing to the wall face , los .ignition ResultpH so e phosphatd th nf an so e test similae alkalins i l samplessar al H p n ri e e Th . (7.7-8.4), los ignition o s n (and therefore organic matter content (2.4-6.w lo s i ) 4 easild %an ) y available phosphat uniformls ei y high CaCOe Th . 3 conten variables i t highese th ; t fillpi t s (F3 F365& calcareoue )ar s and very calcareous respectively F3d slightls 8an ,i y calcareous. Because these are samples of pit fills, it is not surprising that their chemical characteristics do not match surroundine thosth f eo g natural soid an l parent material priore terracbuils a Th . n ywa o t f fluvio-glaciaeo l gravel which supports freely drainin gCorbe soilth f yso Association (Soil Surve f Scotlandyo , 1:63360 Soil map, 95)d sheetan . 5 Thes8 s e soil e typicallar s y acid thereford an , e non-calcareous, witavailablw hlo e phosphate (Glentworth & Muir 1963, 276). High, easily available phosphate ratings in the pit will be at least partly caused by the alkaline pH which promotes the creation of soluble forms of phosphate. Inputs of phosphate-rich organic materials are also present in F39. High CaCO3 levels in certain layers are probably caused by the presence of mortar, whic notes hwa d durin excavatioe gth F36n ni . Thi turn si t fill npi .explaine th f o hige sth H hp The chemical similarity of the wall backing (F43) to the pit fills reflects the mobility of soluble soil components; in this case out of the pit fill and into the pit walls. This mobility has probably removed chemical differences between the pit layers that existed when they were deposited; it is therefore difficult to determine the original composition of the pit fills. The fact that only the upper fills are calcareous suggests that building debris (mortar) was dumped only late in the filling of the pit. High pH conditions throughout the pit mean that bone should have been preserved in all contexts; thereforfouns d ian wa t d9 onlF3 e onln yi y deposite that da tinfilling e stagth f cleaes o i t I .r tha apparene tth t chemical uniformit sedimente th f yo s doe reflect s no changin e th t g naturmateriae th f eo l being dumped int pite oth .

REPOR TEXTILEN TO S Thea Gabra-Sanders The organic material in layer F39 yielded 24 textile fragments and two plucks of raw wool. A catalogue of the material, along with a more detailed report, is contained in microfiche but the main attributes of the fragments are summarized in Table 4. The numbers of the fragments in the text refer to the catalogue and the summary in this table. McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY I 421

Description Woven cloth of wool - plain weave Ten of the fragments woven from wool are in the simplest weave, tabby, also known as plain weave (Table 4). These consist of either S-spun single yarn, of which there are four examples combiner o , S-spud an - ndZ single yarn f whico , h ther sixe samplee ear .Th s woven from combined Z/S-spun single yarn hav loweea r thread-count. The medium/coarse yar e woollens f everydao , y quality, with a low thread-count, but of suitable weight for clothing or blankets. Similar textiles are known from Fast Castle (Ryder & Gabra-Sanders 1992, 6) where fabrics made of entirely S-spun single yarn predominate same Th . e featur bees eha n note textilen di s from Newcastle upon Tyne (Walton 1981, 193) and Perth (Bennett 1987, 159) while the material from Lubeck (Tidow 1982, 272) contained vera y large amoun f Z/o t S spun spunS tabbie t relativel S/ spuZ bu d , Z/ n san w tabbiesyfe . Likewise, material from Haarlem, in the Netherlands (Vons-Comis 1980, 66; 1982) and from London (Pritchard 1982, 199, 200, 205), contain both Z/S, Z/Z and S/S spun tabbies with Z/S fabrics the most common variety. Feltin s note gwa n severa e textildo th f o el fragments. Feltine primarb n gca r secondar yo s i t i t ybu often difficult to differentiate between the two. Primary felting can arise by fulling, ie during pounding while we orden i t shrinko t r hencd an , thickeo et condensd nan fabrice eth . Frequent washing result secondarn i s y felting. Table 4 provides details of the finishing of the individual fragments.

2/1 Twills Nine wool textile twil1 2/ ln si (Tabl , alse4) o know three-shafs na same th t f etwillo qualit e ar , s ya the plain weave. Four (5, 8, 16 & 21) have Z-spun single yarn in system 1 and S-spun single in the other. In case on e (21 Z-spue )th n system form warpe th s , identifie reinforcea n di d selvedge (illus 14s). Another four workee ar sample) d22 fro& sm0 2 (11 , Z-spu,14 n singl ef which o yarn e on ,, textile reinforce(14)a s ha , d selvedge. There is only one (13) textile woven from S-spun yarn. Details of finishing are shown in Table 4. Similar 2/1 twills worked from Z/S, and of mediocre quality, are familiar finds on Scottish medieval sites mighd an ,describee b t s typicalda t leas a ,r easter fo t n Scotland r exampleFo . , comparable fabrice ar s known from Fast Castle, from 13th 14th-centurd -an y level Aberdeen i s n (Bennett 1982, 197-8)n i d an , considerable quantity from Perth (Bennett 1987). Twenty fragments hav t edgeecu s (Tabl numbea , e4) f whico r beed hha n oversew preveno nt t fraying. Eleven have sign f sewinso g althoug stitchine th l hdisintegrateds al gha . This almost certainly indicates that flax thread was used. The stitch-holes and thread impressions indicate even stitching ranging from three to five stitche r centimetrepe s f interesO . vera s i yt worn fragmen ) whic (4 s patchet hwa d wit a hsecon d fragment (5) using a hemming stitch (illus HA). Measures have been taken to strengthen the selvedge with extra paired warp threads. In sample 14 the outermost six warp threads are paired; two pairs are in a dark colour and one pair in a lighter colour. Sample alss i o1 2 reinforced, with three paired war psame threadsth l e al colourt the e bu ,y ar . These strong selvedges would have been necessary for cloths which were to be heavily fulled and stretched on cloth tenters. ) displa 25 o sample & Tw y6 weavin(1 s g faults. Sampl probabl6 1 e wefa s t yha fault causey db incorrect treadle sequence sampln I . eithe5 e2 wefa rbees ha tn accidentally passed twice throug hsheda r o , the warps were incorrectly tied. Of the two faults the former is the more common. Sample 23 consists of two plucks of unspun wool. No staple formation (locks) is apparent, but both have formed a layer of loose felt about 1-2 mm thick, which appears to be neither manufactured felt nor natural cotting of a fleece. A possible explanation is the 'pilling' of clothing.

Conclusion textileAle th l s from Pluscarde woollene nar , occurrin almosn gi t equal number f plaiso n twill 1 weav2/ d . ean The similae clote yar th ho t r types foun Scotlandn di , London, German Netherlandse th d yan 15th-centurA . y dat possibls ei l piecesal r efo . samplee Mosth f fragmenteo e t sar wornd majorite dan Th . y show clear sign f havinso g been cut, sewn, and sometime sf the o re-used mw fe als A .o sho edgw wra e a remain overcastr o seama f m e o s he ,Th . threadcoun averags i t r everydaefo y clothing presence th d an ,f weavin eo g faults suggests thafabrie s th t cwa woven at home and, in this case, possibly by the religious community. 422 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

20mm

fragmento Tw ILLU A 14 Sf brow o s n woollen worn fabric e smalleth : r fragmen) (4 t s sewwa n ont e largeoth r fragmen ) (se(5 t e fiche repor r fulfo tl description)

ILLUS 14s Fragment of brown woollen fabric with reinforced selvedge. It is felted on both sides (see fiche report for full description). The scale is the same as illus 14A

generae Th l standar fragmente th f do moderates i s t woulI . d appear from contemporary reports (Innes 1867, 193) that in the 15th century bolts of cloth were being sent from east Scotland to the Low Countries to dyede b presumabld an , y finished. Thi e 15tth s h f suggest o centur d en ye s th thatextil t a t e aree skillth an i s were poorly developed e textileTh . bese ar s t interprete s discardeda d remnant f worn-ouo s t clothind gan tailors' waste. Since faecal materia bees ha ln pite founth , containinn di g wha almoss wa t t certainly parasites of human origin t seemi , s likely that these textiles were use r sanitardfo y purposes. McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 423

TABLE 4 Catalogu f wooeo l cloth fragments Threa value d warr otheon coune numbee : efo th weftr th px rfo s i t mm thread f .r0 o 1 r spe When known, the warp count is given first and the same applies to the direction of the spin. amoune Th f finishino t indicates gi followss da :

x slightl yside felteon e n do xx slightly felte botn do h sides xxx felted on one side xxxx felte botn do h sides

± approximately

Sample Dimensions Thread Spin Weave Finishing Colour Cut Sewn Other details m m n i o N count

1 45x43 9x9 SxS plain xxx It brn x x Seam stained 2 5 ±53 0x 6x6 ZxS plain brown 3 65x40 7 Zx? 7 xxxx brown x 4 80x33 7 Zx? 7 xxxx brown x x was sewn into No 5 with a hemming stitch; stitch holes present 5 110x59 9x10 ZxS 2/1 xxxx brown x x raw edge overcast-stitch twill holes present 6 50x40 8x8 SxS plain xxx It brn x x raw edge overcast stained 7 ±70 7 0x 5/6x5 ZxS plain It brn ?x 8 7 ±32 5x 6x6 ZxS 2/1 It brn x raw edge overcast twill 9 75x10 6x6 ZxS plain xxxx brown x 10 102xlO->4 7x7 ZxS plain xxx brown x x stitch holes presentw ra , edge overcast 11 ±70 x 25 10x10 ZxZ 2/1 x It brown x hole present twill 12 6 9->7x 3 7 Zx? 7 xxxx brown x 13 0 101 8x 11 xlO SxS 2/1 XX brown x x 2 small hems-raw edge twill overcast stitch holes present 14 120 x 19->5 10x8 ZxZ 2/1 xxxx brown x high twist spin-reinforced twill selvedge 15 140x40->20 9x9 SxS plain xxx It brown x x raw edge overcast stained 16 0 14 9 0x 8x 12 ZxS 2/1 X It browx n high twist spin-weaving twill stained fault 17 140 x 110 7x7 ZxS plain x browx n raw edge overcast, 2 holes stained present 18 185x85->7 6x6 ZxS plain brown x yarns unevenly spun 19 380 x 26->19 7 ZxZ 7 xxxx reddish x torn brown 20 370x50->10 7x7 ZxZ 2/1 xxxx reddish x x seam-stitch holesw ra , twill brown edge overcast 21 185x30->5 8x6 ZxS 2/1 xxxx brown x reinforced selvedge twill 22 150x25->10 10x8 ZxZ 2/1 xxxx reddish x high twist spin twill brown 23 65 Z It brown length of single yarn 24 70x5 7 ZxZ 7 xxxx reddish brown 25 85x25 9/10x7 SxS plain grey/ x weaving fault brown 4 42 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

WINDOW GLASS

(A complete repor catalogud an t glase th f se o fragment presentes i s microfichen di )

Barbara Ford Thirty-five fragment f windoo s w glass were recovered f whico 3 3 ,h e camstone-lineth n ei fro9 mF3 d pit. One fragment (35) was from topsoil, and a piece of painted glass (34) is a stray find, found in the vicinity of the steps at the south-west end of the claustral range, several years ago, by one of the monks of the community. weatherede ar fragmente t mose th pi th l e r tAl th fo par, n i s t manifeste fore th f varyinm o n di g degrees of brown staining. This is enamel-type weathering which often appears in patches, and in many cases is probably a result of contamination by iron or manganese compounds (Newton & Davidson 1989). The weathering gradually penetrate t goei glase ) th scasee s (4 s righth d untilf (1.n so i an )t s a throug, coree th o ,ht often renderin glase gth s opaque. Many of the fragments of glass from the pit show weathering on the broken edges, suggesting that some of the pieces were broken before burial. Further breakages have taken place after burial. Where weathering allows identification fragmente th e seef l excepb o al havo ,t n e ca seon t been originally plain clear glass wit greeha r palno e green tinte greeTh . n tin s causei t iroy db n impuritie glasse th e n Th i s. remaining fragment (5) is a pale blue glass. It has been 'pot-coloured', a metallic oxide having been added to moltee th n glass, producin uniforga m colour throughou glasse th t . Combination f copperso , iron cobald an , t were used to produce different shades of blue. The dullness of the blue suggests a 14th-century date. Muted tone f bluso e were popula 14te th hn i rcentury , contrasting wit rice hth h previoue toneth f so s period (Graves 1985). Example bluf so e 'pot-metalled' glass have been recovered fro mnumbea Scottisf o r h sites. Fragments from Lindores have been dated to the late 13th/14th century, while several fragments from Dunfermline have been assigned to the 14th century (Graves 1985). There are two methods of manufacturing window glass: the crown process, and the cylinder or muff process . fragmente Nonth identifiee f b eo n ca s beins da g manufactured usin crowe gth n method (see Frank 1982) t severae fragmentbu ,th f o l s were made usin cylindee gth r method. This process involves blowing molten glass int obubbla e whic s lengthenehi rolline y db th swingingd d gan an f ende of thee t Th .sar ncu cylinder split longitudinally by a hot iron, and reheated. The glass is then flattened with iron tongs and a 'flattener'. Cylinder glass can be recognized by the shape of the air bubbles, which are elongated. However, r bubble duai Pluscarde e weatherino ey t th t possiblan n si no e s se i n t o g ei t glass lone Th .g e edgeth f o s cylinder were rounded in the flame, and fragments with fire-rounded edges, such as 1, 14, 15, and 35, are indicative of cylinder glass (Harden 1961). There are no complete panes, but 3 is part of a rectangular pane. Rectangular panes were often used along the borders of windows. Many of the fragments have grazed edges formed by trimming the glass with a 'grazing' iron. Several fragments have darker staining along the edges (9, 12, 14, 17 & 25). This is possibly as a result of a reaction with traces of lead staining left from lead window cames. Only one fragment was decorated (34) and, as stated above, this was a stray find from the site (illus 15). It has brown enamel painted decoration, a technique used since the llth century (Newton & Davidson 1989). The decoration comprises part of a stylistic trefoil and curved stems, a style commonly used in grisaille typa , f decoratioo e n forme paintiny b d g natural foliate motifs suc s leaveha s onto plain glass. Grisaille originated in the 12th century and became popular in the 13th and 14th centuries. The style of decoratio s typicai 4 3 f tha o ln no tlate useth e n d13ti e see b hfragmentn no n centur ca d f latyan o s e 13th- century glass from Coldingha Elgid man n (Graves 1985). Windows decorated with grisaille ofted ha n medallion f colouso r suc13th-centurd blues hmi a red d e seee san th s b than n i ca ty 'Five Sisters' windot wa York largese th , t surviving exampl grisaillf eo e (Archer 1985; Gibson 1979). glase Th s fro stone-linee mth t cannodpi datee b t d very accurately lace decoratiof kTh .o n makes dating difficult 14th-centurA . y dat bees eha n suggeste fragmene th r d fo f blu o t eremainine glasTh . s5 g glass from t variee thickepi Th e thicknes n . f i sthesth o r mm e 3 fragments so t fro 5 m1. , tha s thosi t n i e m ovem 2 r McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORAY 425

36

30mm

ILLU 5 Glas1 S metald an s: fragmen 34 . f corrodeo t d window glass with milky brown surface painted an s d decoration representing part of a stalk and trefoil 36: fragments of beaker; the rim sherd has grooved decoration and the body sher smals dha l applied purnt coppe: s2 r alloy strip

thickness, have been more affected by weathering (1-4). These fragments could also date to the 14th century. e remaininTh g fragment vera e yar shomogenou s group, varyin thicknes n , gi alf ver o l mm y 2 so t fro 5 m1. similar appearance and weathering patterns. The quality of the glass suggests a later date, possibly in the 15th r 16to h centuries e weatherinTh . g e glascrustth sn o sindicat e that e glas mucs broketh f wa sho n before deposition. This suggests that they were broken fragment f disuseo s d glass, from window f variouso s dates, 15te whicth r hbee16tn d o i ht ha hn pi century disposee th n i f .do

Vessel glass Two very small fragments of a vessel with a straw-coloured tint were recovered (36) from unknown contexts b membeya presene th f ro t Benedictine community excavation e priorth o t glase Th .vers si y finheavild ean y bubbled and the fragments appear to be parts of a beaker (illus 15). The yellowish tinge is due to impurities in the glass. The body of the vessel is decorated with small applied prunts. The technique of dropping molten blobs of glass onto the body of a vessel was brought to Europe in the 14th century (Vose 1975). It is thought that these blobs may have had some religious significance, being put on glass to ward off evil spirits (Vose 1980). Apart from being decorative they also improved the drinker's grip on the vessel. The beaker was probably imported into Scotlan Europeae th df o fro e nmon glass houses.

THE MEDIEVAL POTTERY

(A catalogu stratifiee th f eo d materia presentes i l microfichen di )

Charles Murray

Pottery from excavationthe quantite Th f potteryo y fro excavatioe mth t Pluscardena exceedingls ni y small, amountin onlo g t sherds 3 y2 , 10 of which make up most of the base and some of the lower portion of a single jug (Cat No 1). This was foun rubbln di e (F30) that probably originally supporte flighda f step o remaininte Aren si Th . a6 g sherde sar undiagnostic, being mainly body sherds lowee , th althougs r straparwa a f e to p hon handl e. (Ca2) o tN With the exception of one sherd which is possibly East Anglian in origin (Cat No 4), the remaining pottery would appear to be of local manufacture. However, the absence of kiln evidence and the lack of | SOCIET 6 42 Y OF ANTIQUARIE SCOTLANDF SO , 1994

scientific analysis makes this conclusion speculative fragmene On . s similai ) tfabria 2 (Ca d o o t r an cN t vessel type noted from excavation Elginn si . The dating of the pottery would appear to range between the 13th and 15th centuries, although it must be borne in mind that this dating is based on the fabric, glaze and appearance of the sherds, rather than on the more diagnostic portions of pots such as rims, handles and bases. Similarly, it is difficult to be definite about vessee th l types most bu , t appea havo rt e been jugs. In onlinstance yon sherdd edi s from different site features ,F21d burialan , 6 conjoiF1 s , n 6) (Ca o N t demonstratin contemporaneite gth thesf yo e features.

Unstratified pottery largA e assemblag f unstratifieo e d potter s gathere e currenywa th y db t Benedictine community at Pluscarden between 1948 and 1990. Although this cannot be related to any particular area of the Abbey complex, it sets the pottery from the excavation in a wider ceramic context and demonstrates that the quantity, range and quality of pottery was greater than the excavated material would appear to signify. In all, the assemblage comprised some 650 sherds. The majority of these were very fragmentary, amorphous body sherds, probably local in origin. About 60 sherds, however, merited closer examination.

Local products The main vessel type represented is the jug, but a range of other forms was found. Portion f possiblso mans ya thres ya e urinal evidente sar . Ther s alsei ofragmena drippina f o t g panA . very fine rim sherd appears to derive from a beaker-type vessel also recognized in Aberdeen (Murray 1982, 146, Illus 81, Nos 276-281). Highly decorative forms are represented by parts of two very distinctive anthropomorphic face-mask jugs stylistin o ; c ground f theso e e son woul d appeae dato th t r o et 13th century, while the other could be as late as the 15th/16th century, although both are of a similar fabri glazd can e type. Lik excavatee eth d pottery, this unstratified local material appear rango st datn ei e betwee e 13tnth h and the 15th/16th centuries; but in this instance the dating can be based on the rim and handle forms present.

Imported pottery smalA l amoun assemblagee th f o t , som sherds5 e1 , derive from imported vessels l beinal ; g from Scarborough Ware products. It is hardly surprising that Scarborough Ware should be present in this assemblag ubiquitous i t i s ea almosn so l excavateal t d site f medievaso l dat north-easn ei t Scotland. Several Scarborough Ware types are evident, ranging from the distinctive thick-grooved rod handle type associated with face-mask jugs, to a shield piece from a Knight jug. The sherds include both the fabrics identified by Farmer (1979) and date to the 13th century.

METAL AND STONE OBJECTS 1 A roughly shaped shale disc, with a depression in the centre of one side, probably the beginning of a perforation. Diam 48 mm; Perf 7 mm; th 6 mm From layer F39 from the stone-lined pit.

(joining o 2 Tw (illu ) 15 )s fragment f thio s n copper alloy sheet, possibly strappara smallee f o te Th . th f ro two pieces is curved, and the larger has a perforation made by piercing with a pointed object from the largee th frontf r o t eac fragmentd A . hen aren a f solder ao s i s , wit hcoppea r alloy patch affixe oneo t d , resule attempn a th f o t meno t t dbreaa o strengthe t r ko aren a n a whic s crackinhwa g (Amanda Clydesdale, pers comm). Larger fragment: 36 mm x 11 mm. Smaller fragment: 13mm x 10 mm. From stone-linen i laye 7 rF3 d pit. McCORMICK: EXCAVATION PLUSCARDET SA N PRIORY, MORAY 427

DISCUSSIO CONCLUSIONN& S

Finbar McCormick The principal objectives of the excavation at Pluscarden were to ascertain if the nave of the church anwestere dth n cloistee rangth f eved eo ha rinvestigat ro t bee , so n f i built d e an suc, h remains, prior to the construction of a modern building in the area. Substantial foundations, in places shattere weigh e walle th th y f dsb o t they once supported, indicated thachurce th t h nav indeed eha d been constructed, thus ending speculatio contrare th o nt y whice begur bacd th fa hha s ka s n a middle of the 18th century (Flavell, above). Extensive excavation, however, demonstrated that there was no evidence for the construction of a western claustral range in the monastic complex. This is unsurprising given the rule by which the Valliscaulians lived. This Order founded Pluscarden which remained in its possession until its adoptio Benedictine th f no e idea 1453/4n i l . Flavell (above shows )ha n tha Valliscauliane th t s were not allowe keeo dt p large number brothery la f thasd o 1205/6n san i t r instancefo , house o n ,th f eo Order was allowed to have more than 20 monks, including lay brothers. Western domestic ranges generalle ar y though havo t e been buil accommodato t large eth e number brothery la f so s which were a feature of other Orders. When one considers that the present community of 25 Benedictine monks are accommodated in the eastern wing of the medieval claustral range at Pluscarden, it can e appreciateb d that ther s littlewa e necessit buildine th r westera yfo f go n medievae rangth n ei l period. e excavationTh , however, produced evidenc fairla r efo y insubstantial buildin e areth an gi usually occupied by a west range. All that remained of the building was the basal course of a wall (F4) which separate paveo dtw d areas wale th sil lA . n i indicatel d direct access between both areas. A jamb stone suggest sa 16th-centur structure y th dat r efo e (see above) absence Th . f eitheeo r foundation r foundatioo s n trenche e remaindeth r buildine fo sth f o r g confirm e unsubstantiath s l nature of this edifice. Its function is unknown, but it could be similar to the small, late medieval building s builthawa t t agains e outsidth t wala f eo l which demarcate e westerdth ne th sid f eo cloister at Dryburgh Abbey. Given its probable 16th-century date, it is likely to be part of the building campaign undertaken by Alexander Dunbar, prior at Pluscarden in the years immediately precedin Reformatione gth responsibls wa buildino e nortth e vestr e wh ,r th th e fo f hf gyo of sid f eo the choir (Flavell & Fawcett, above). A further building comple investigates xwa south-eastere th t da n easter e edgth f eo n ranget I . phaseo buils tw wa n i st wit hshera f 14th/15th-centurdo y pottery being found withi fabrie nth f o c the secondary wall. A group of four burials, three of which were excavated, lying to the west of the remains of the 16th-century building, is problematical. The burial areas of monasteries were usually located to the east and south-east of the choir, and expansion of the graveyard would be expected in the area churche e nortth th e servicf ho o t Th . e trench excavate Doy db m Ambrose Flavell (above) produced no evidence for burials in the latter area. Radiocarbon dating suggests a 15th-century burialse datth r efo . Of the four skeletons noted three were excavated. All were women, ranging in age between year5 4 d sabou an (se 7 1 et Lorimer, above) bonee Th . s wer relativeln ei y poor conditiod di d nan t providno evidency ecause an th f deathr eo e fo t seemI . s likely, however, that they we'r l burieeal d at the same time and this suggests that death was sudden, due, perhaps, to a contagious and fatal disease, or to some natural disaster that left no traces on the skeletons. A 15th- or 16th-century stone-lined pit post-dated the burials. It was probably roofed, with 428 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1994

an opening at its rounded eastern end. The absence of stones within the pit fill suggests that this f woodo alss roos i t oI .wa f likely tha openine th t originalls gwa y covered with somr o e d forli f mo trap-door. Thi s implie i sface th t y thadb greae th t materiaty fl majorit e th lf o ypresen e th n i t organic from the entrance area consisted of Terrilimosina racovitzai (Bezzi) which generally inhabits dark environs suc s caveh a mol d an s e holes whicd an , especialls hi y abundan cellarn i t s containing large amounts of decaying vegetable matter (Skidmore, above). Once the pit became partially silted, but still retained its roof and probable trap-door, it was used as a refuse dump and cess pit. Its original use, however, is unknown. It can be assumed that the monastic complex at Pluscarden, with its adjacent river, would have had a reredorter system in its original design and it is extremely unlikely that the pit was constructed as a cess pit. Even if the reredorter system became blocked at a later stage in the Priory's history, it is unlikely that the pit would have been constructed for this purpose as it is too far from the domestic buildings. It is more likely that the pit was constructed originally for storage. The use of a hole for such a purpose suggests that coolness, rather than drynesse primarth s wa y, attribute being soughs it y b t construction case storage th eth thif s I . sdairf wa eo y produce, suc cheess ha butterd havey an ma e, bee primars nit y function. foue useTh s lt whe defecationr naturinteriopi e wa dfo e t th ndumpini e th f eth o f ro d an ,f go animal carcasses, is clearly reflected in the beetle and fly assemblage present (Buckland, above & Skidmore, above). The presence of Trichocellus, which is a wetland or peatland species, suggests that peat burningr fo , alsy oma , hav epite beeth . n i e dismemberedTh sometimed an , s butchered musg do td represenan , t remainca a a tf o s period of acute hunger at the site. It should be added, however, that a small quantity of bones of usuae th l meat-providing domesticated animals wels a , thoss a l f birfishd eo dan , were also present layere inth . An organic deposit of human faecal material contained a range of environmental material including pollen. This indicate presence dth f cerealeo , most likely representing typ e brothth et bu , of cereal could not be identified (Mills, above). The faecal material included 24 scraps of woollen material, presumably used as anal wipes (Gabra-Sanders, above). Most of the fragments were well worn, and clearly represented old clothing or blankets, cut up and used for this toiletry purpose. In, and around, the faecal material was a scatter of broken window glass fragments, originally clear glass with a green tint (Ford, above). The only exception to this was a fragment of pale blue glass, whic probabls hwa y als f 14th-centuroo t alsy pi date oe containeTh . fragmentw fe da f 14tho s r -o 15th-century pottery. Stone-lined pite alsar so known from other Scottish medieval sites t ThreavA . e Castle, Galloway, a stone-lined pit was found immediately outside the wall of a house within the outer castle enclosure (Good & Tabraham 1981, 102). Its function was unknown but it had been used as rubbisa t afte hpi s origina it r l functio f usefalled o t Auchindount ha n.A nou Morayn i , stonea , - discoveres linewa t dpi d underneat cellae hth r flootowea f o r r house excavatoe Th . r suggested that it was too small to be a pit-prison and that one possible use was that 'it may have served as a food store, where perishable food stuffs were kept' (Wordsworth 1990, 171). Like the Pluscarden pit, it seemed ultimatel havo yt e been useduma s da p and, coincidentally, containe almose dth t complete skeleton of a cat and the remains of two dogs as well as some bones of the usual domesticates, ie sheep and cows. e excavatioTh n produced very little stratified pottery largeA . r assemblag f potteryo e , however, has been collected by the present Benedictine community since 1948. The great majority of the assemblage was locally made, and dates from the 13th to the 15th centuries (Murray, above). e potterTh y consisted mostl f jugsyo , although beakers, urinals drippina d an wern , gpa e also McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY. MORAY 429

present. Nearlimportee th l yal d pottery consiste f Scarborougdo h Ware typa , e foun almosn di l al t medieval site north-easn si t Scotland. The excavation provides little information concerning the diet and livestock economy of the inhabitants of Pluscarden. The bones from the stone-lined pit, as well as providing evidence for severe food shortage, indicate tha s inhabitantit t s rearedaccesd ha , cattler sto o , , sheep/goatd an , pig. A pig mandible from a very young individual was, at most, a few weeks old at time of death; this implies that pigs were being reared on the site. Chickens were also kept; the goose present in the pit may have been either Greylag or domesticated stock. A small bird, possibly a Jackdaw, was alst certaiono presens i thif ni t i st representebu t d food refuse. As Pluscarde onls ni y t surprisin aboufro no m coase k s m i th 9 t t i t g that marine produce were included in the diet. The stone-lined pit produced bones of Cod, Hake and Gurnard, while an oyster shell was present under the base of wall F27. Cod seem to have been commonly traded to inland rural site medievan i s l s alsScotlanwa o t foundi s da , along with Haddock t Smailhola , m Tower, Roxburghshire a (Barnetso se ,fro m some k m th 0 e4 n 1988, Fich G.I)e3 . Shellfish were also traded with inland rural sites and oyster has been noted at Coull Castle, Aberdeenshire, which a (Simpso se fro m e k m th 5 i4 sn 1924, 92) e faece.Th s presen e stone-lineth n i t t providedpi d evidenc consumptioe th r efo cerealf no sitee th .t sa unlikels i t I y thaValliscauliae th t n monk t Pluscardesa n ever engage farminn di largs a n ego a shee e scalth s pea rearingwoollee th r fo ,n trade Bordee , th carrie y b r n monasteriesdo . Flavell (above s noteha ) d t hav tha Ordee large no th etth d e di r brethrey armla f yo n neede r sucdfo h activit thad monastery'e yan th t s income depende rentsn do , tithe paymentd san kindn si . Amongst these were mining and fishing rights as well as the ownership of several mills.

ACKNOWLEDGEMENTS I would lik thano et Benedictine kth e communit t Pluscardeya r theinfo r kindnes held san p during the excavation. I would also like to thank the excavation team: Martin Crowther, Chloe Dondos, Dom Ambrose Flavell, Jamie Hamilton Hendersonn Jo , Ibbetsoy Ka , Cairind nan e Sutherlande Th . illustrations are by Christina Unwin and Sylvia Stevenson. Ruby Ceron-Carrasco would like to than r KarikD n Perreman providinr sfo g identificatio featherse th r nfo . Richard Fawcett adds preparinn :I finae gth l versio architecturae th f no l repormuce ow o ht I t the kindnesAmbrosm Do f o s es tireles Flavellwa o showinn i s wh , e aroungm e abbedth d yan discussing its architecture - although the views expressed here are entirely my own. The report was edited and submitted by AOC (Scotland) Ltd. Historic Scotland arranged and provide dexcavatione fundth r post-excavatioe fo s th r fo , publicatione n th analyse r fo d san .

REFERENCES Andersen, S T 1979 'Identification of wild grass and cereal pollen', Danmarks Geologiske Unders0gelse Arbog (1978), 69-92. Angel, J L 1971 The People ofLema. Washington. Anson, P F 1959 A Monastery in Moray. London. Archer, M 1985 An Introduction to English Stained Glass. London. Bacon M 199 W , 0 'The dependenc f humaeo n bone structur life-style'n o e Land,c ,0 So Proc24 y Ro (1990), 363-70. Barnetson, L 1988 'Animal remains', in Good, G L & Tabraham, C J, 'Excavations at Smailholm Tower, Roxburghshire', Proc Antiqc So Scot, (1988)8 11 , 258-6 fiche0& : F8-G2. Bass, W M 1987 Human Osteology: a laboratory and field manual, 3rd edn. Missouri Archaeological Society of Columbia. 430 SOCIET ANTIQUARIEF YO SCOTLANDF SO , 1994

Batten, E C H 1877 The Charters of the Priory of Beauty. London. BeckettM 198 N 8W , 'The Perth Charterhouse before 1500', Analecta Cartusiana, 128, Salzburg. Bennett, H 1982 'Textiles', in Murray, C J (ed), Excavations in the medieval Burgh of Aberdeen 1973-1981, Edinburgh, 197-20 Antiqc So = 0 ( Scot Monogr. 2) r Se Bennett, H 1987 'Textiles', in Holdworth, P (ed), Excavation in the medieval burgh of Perth 1971-1981, Edinburgh, 159-72 (= Soc Antiq Scot Monogr Ser 5). Berry, A C & Berry R J 1967 'Epipenetic variation in the human cranium', J Anat, 191 (1967), 361-78. Birch, W De Gray 1900 Ordinale Conventus Vallis Caulium. London. Boucher J 195B , 5 'Sex difference foetae th n i ls sciatic notch' Forensic,J Medicine, (1955)2 , 51-4. Bower, W 1987 Scotichronicon, Watt, D E R (ed). Aberdeen. Boyd B ,198 8 'Cereal Scottisn si h Antiquity', Circaea, (1988)5 , 101-10. Brothwell D ,198 1 Digging Up Bones. Oxford. Buckland, P C 1976 The Environmental Evidence from the Church Street Roman Sewer System (= Archaeology of York 14/1). Buckland C 197P , 9 Thorne Moors: palaeoecologicala study oBronzea f sitee contributiona ( Ag e th o t historyBritishe th f o insect fauna). Birmingha Departmentm= ( f Geographyo Occasional Publication 8). Cal. Papal Letters = Calendar of entries in the Papal registers relating to Great Britain and Ireland, Bliss, alt e (eds)H W . 1893. London. Chesterman T 198J , 3 'The human skeletal remains' Hedgesn i , , Isbister:W J , chambereda tombn i Orkney, Oxford, 73-12 Brit115)r R Se BA . 3= ( Choisselet, D & Vernet, P (eds) 1989 Les Ecclesiastica Officia Cisterciens du Xlleme siecle, Abbaye d'Oelenberg, F-68950 Reiningue. Consuetudines Guigonis = see Guigues, 1984. Cowan, I B & Easson, D E 1976 Medieval Religious Houses: Scotland. London. Crowson A 196R , 2 'Observation Coleoptern so Scottisn ai h Oakwoods', Glasgow Naturalist, (1962)8 1 , 177-95. Cummins J , 198 Hawk.e Hounde th 8Th d an London . Dickson, C 1989 'Human coprolites', in Bell, B & Dickson, C, 'Excavations at Warebeth (Stromness Cemetery) Broch, Orkney', Proc Antiqc So Scot, (1989)9 11 , 101-31. Dilworth, M 1986 'The Commendator System in Scotland', Innes Rev, 37 (1986), 51-72. Farmer, P G 1979 An introduction to Scarborough ware and a re-assessment of Knight jugs. Hove. Fasti Ecclesiae Scoticanae = see Scott, 1915-28. Fawcett, R 1984 'Scottish medieval window tracery', in Breeze, D J (ed), Studies in Scottish Antiquity presented to Stewart Cruden, Edinburgh, 148-86. Finnegan, M 1978 'Non-metrical variations of the infra-cranial skeleton', J Anat, 125 (1978), 23-37. Folz 195R , e monaster9'L s ChouVau ed de l premieu xa r histoire'n sieclso e ed , Bulletin Philologiquet e Historique du Comite des Travaux Historiques et Scientifiques, 1959, 91-115. Fordun J , 1871-2 Historia Gentis Scotorurn, Skened e (HistorianF W , f Scotlano s . d4) vol& 1 s Edinburgh. Frank, S 1982 Glass and Archaeology. London. Gibson, P 1979 The Stained and Painted Glass of York Minster. Norwich. Gilbert 197M J , 9 Hunting Huntingd an Reserves Medievaln i Scotland. Edinburgh. Girling, MA 1981 'Beetle Remains', in Thorn, J C, 'The Burial of John Dygon, Abbot of St Augustine's', in Detsicas (ed)A , , Collectanea Historica: essays memoryn i of Stuart Rigold, Maidstone, 82-4. Glentworth, R & Muir, J W 1963 The soils of the country round Aberdeen, Inverurie and Fraserburgh. Edinburgh. Godwin H ,197 5 History British e oth f Flora. Cambridge. Good, G L & Tabraham, C J 1981 'Excavations at Threave Castle, Galloway, 1974-78', Med Archaeol, 25 (1981),90-140. Good, J A & Giller, PS 1991 'The diet of predatory Staphylinid beetles - a review of records', Entomologist's monthly magazine, 127 (1991), 77-89. McCORMICK: EXCAVATIONS AT PLUSCARDEN PRIORY, MORA | Y43 1

Graves P , 1985 Scottish Medieval Stained Paintedd an Glass, thesisA M , Universit Glasgowf yo . Gray H , 1977 Anatomy descriptive applied,d an Pickering-Pick Howden& , T , (eds)R , , 15th ednw Ne . York. Greig, J 1981 'The investigation of a medieval barrel-latrine from Worcester', J Archaeol Sci, 8 (1981), 265-82. Guigues ler 1984 Coutumes Chartreusee d Consuetudine= [ s Guigonis] LaportM m eDo . (ed), Sources Chretiennes 313, Paris. , HallKenwardR A ,, Williams K H , Greig& A 198 P ,R J , 3 Environmental Livingd an Conditionso tw t a Anglo-Scandinavian Sites Archaeology= ( f Yorko 14/4). Hamond, F W 1983 'Phosphate analysis of archaeological sediments', in Reeves Smythe, T & Hamond, F W (eds), Landscape Archaeology in Ireland, Oxford 47-80 (= BAR Inter Ser 116). Hannah, I C 1936 'Screens and lofts in Scottish churches', Proc Soc Antiq Scot, 70 (1936), 181-201. Harden B 196D , 1 'Domestic window glass, Roman, Saxo Medieval'd nan Jopen (ed)i , M E , Studiesn i Building History: essays recognitionn i worke o th O'Neill,fJ t S ofB H London, 39-63. Hillson S , 1986 Teeth. Cambridge. Historia Occidentalis = Vitry's Historia Occidentalis, quoted in Folz, 1959. Hodgson M 197 J , 6 Soil survey field handbook. Harpenden (= Soil Survey Technical Monograph. 5) o N Howe W 195 R , 5 'Studie Beetlen Familse o th f so y PtinidaBiologe Th : f Tipnuse12 yo unicolor (Pill& . Mitt.)', Entomologist's monthly magazine, 91 (1955), 253-9. Humphries R 195 W , 2 'A Visi Pluscardeo t 1761'n ni , Innes Rev, (1952)3 , 107-9. Hutchison, C E 1989 The hermit monks ofGrandmont. Kalamazoo MI (= Cistercian Studies Series 118). Innes C (ed) 1867 The Ledger of Andrew Halyburton 1492-1603. Edinburgh. Jacque Vitre sd y 1597 Historia Orientalis t Occidentalis,e Moschus (ed)F , . Douai. Johnson C , 1963 'The British Specie Genue th f o s Rhizophagus Hbst. (Coleopatera, Rhizophagidae)', Proc & Trans Manchester Entomological Soc, (1961-3), 3-9. Knowles D ,197 Religiouse 1Th Orders England,n i vol 3. Cambridge. Koch, K 1970 'Subfossile Kaferreste aus romerzeitlichen und mittelalterlichen Ausgrabungen im Rheinland', Entomologische Blatterfilr Biologie und Systematik der Kafer, 66 (1970), 41-56. Lever, C 1977 The Naturalised Animals of the British Isles, 1979 edn. London. McCormick F , 1991 'The Anglo-Normae effecth f o t n settlemen Ireland'n o t s wildomesticated dan d fauna" Crabtreen i , Ryan& J (eds)K P , ,Cultural d Animalan e Us Change. Philadelphia. McCormick, F forthcoming 'Excavations at lona: 1988'. Mackintosh A ,189 3 'Measured drawing Pluscardyf so n Priory, Elginshir plates)'6 e( , Trans Aberdeenshire Ecclesiological Soc, (1890-32 , parts 5-8), PI 1-5. Macphail, S R 1881 History of the Religious House of Pluscardyn. Edinburgh. McRoberts D ,196 5 'Scottish sacrament houses', Trans Scott Ecclesiological Soc, (1965)5 1 , 33-56. Megnin, P 1894 La Faune des Cadavres. Paris. Mignard, JAP 1864 'Historic des principales fondations religieuses du Bailliage de la Montagne en Bourgogne, 'Memoires de la Commission des Antiquites du Departement de la Cote-d'Or, 6 (1861-4), 412-76. Moore, P D & Webb, J A 1978 An Illustrated Guide to Pollen Analysis. London. Murray Murray& K C 199 H , J , 3 'Excavation t Rattraysa , Aberdeenshire Scottis:A h deserted burgh'd Me , Archaeol, 37 (1993), 109-218. Murray, J C (ed) 1982 Excavations at the medieval burgh of Aberdeen 1973-81. Edinburgh (= Soc Antiq Scot Monogr Ser 2). Newton, R & Davidson, S 1989 Conservation of Glass. London. Officia Ecclesiastica = Les Ecclesiastica Officia Cisterciens du Xlleme siecle, see Choisselet & Vernet (eds), 1989. Ortner Putchar & J 198D G , W , 5 Identification of pathological conditions humann i skeletal remains. Washington. Osborne J 198P , 3 'An insect fauna fro mmodera n s comparisocesspiit d an t n with probable cesspit assemblages from archaeological sites', J Archaeol Sci, 10 (1983), 453-63. | SOCIET 2 43 ANTIQUARIEF YO SCOTLANDF SO , 1994

Peacock R 197 E , 7 Rhizophagidae: handbooks identificatione th r fo of British Insects, 5(5a). London. Pearson, G W, Pilcher, J R, Baillie, M G L, Corbett, D M & Qua, F 1986 'High Precision 14C measurement of Irish Oaks to show the natural 14C variations from AD 1840 to 5210 BC', Radiocarbon, 28(1986), 911-34. , PerryBucklandW Snaesdottir& D , C P , M ,198 5 'The Applicatio f Numericano l Technique Inseco t s t Assemblages from the Site of Storaborg, Iceland', J Archaeol Sci, 12 (1985), 335-45. Pritchard F , 1982 'Textiles from Recent Excavation Cite f London'th yo n si J0rgensenn i , Tidow& B L ,K , . (eds), Textilsymposium Neumiinster, 193-208. Episcg Re Aberdon Registrum= Episcopatus Aberdonensis, Spalding Club. 1845. Edinburgh. Richardson Tabraham& S J , J 198C , 3 Dryburgh Abbey. Edinburgh. Robinson, M & Hubbard, R N L B 1977 'The transport of pollen in the bracts of hulled cereals', J Archaeol Sci, 4 (1979), 197-9. Rohacek, J 1983 A monograph and reclassification of the previous genus Limosina Macquart (Diptera, Sphaeroceridae) of Europe, Beitrag Entomologier ezu (19832 t P , ) 33(1). Berlin. Ryder, M L & Gabra-Sanders, T 1992 'Textiles from Fast Castle, Berwickshire, Scotland', Textile History, 23(1) (1992), 5-22. Scotichronicon Bowere se = , 1987. Scott H ,1915-2 8 Fasti Ecclesiae Scoticanae (new edn vols,8 ) Edinburgh. Simpson, W D 1924 The excavation of Coull Castle, Aberdeenshire', Proc Soc Antiq Scot, 58 (1923-4), 45-99. Skene (edH J )F , 1877-80 Liber Pluscardensis (Historian f Scotlandso , vol10)& s7 . Edinburgh. Smith, K G V 1986 A Manual of Forensic Entomology. London. Stafford F ,197 1 'Insect medievaa f so l burial' Archaeol,& i Sc , (1971)1 , 6-10. Stewart D 197T , 9 Essentials of Forensic Anthropology. Springfield, 111. Tidow K , 1982 'Untersuchunge Wollgeweben na s einenau n BrunneSchrangem de f au n Liibeck'n ni , Lubecker Schrifien Archaeologier zu und Kulturgeschichte (LSAK)) (1982)6 d B , , 251-84. Rudolf Habelt Verlag GMBH - Bonn. Vermeer, P 1954 'Citeaux - Val des Choux', Collectanea Ord Cist Ref, 15 (1954), 35-44. Vons-Comis, S Y 1980 'Laat Middeleeuwse textielvondsten in het Brinkmann-Complex te Haarlem', Haarlems Bodemonderzoek, 12 (1980), 59-66. Vons-Comis, S Y 1982 'Medieval textile finds from the Netherlands', in J0rgensen, L B & Tidow, K (eds), Textilsymposium Neumiinster, 151-61. Vose, R H 1975 Glass. London. Vose H 198 R , 0 Glass. London. Walton, P 1981 The textiles', in Harbottle, B & Ellison, M (eds), 'An excavation in the Castle Ditch, Newcastle upon Tyne 1974-1976', Archaeologia Aeliana, 5 ser 9 (1981), 190-228. Webster J , 1948 Dunfermline Abbey. Dunfermline. Woodroffe, G E 1953 'An ecological study of the insects and mites in the nests of certain birds in Britain', Bull Entomological Research, (1953)4 4 , 739-72. Wordsworth J , 1990 'Excavation t Auchindousa n Castle, Moray District 1984'n i , , Proc Antiqc So Scot, 120(1990), 169-71. Zacher, F 1993 'Haltung und Zuchtung von Vorratsschadlingen', Abderhaldens Handb biol Arbeitsmeth, 9 (1933), 389-592.

This paper is published with the aid of a grant from Historic Scotland