The Shrine of St.

Menas in the Maryût

Author(s): J. B. Ward Perkins
Source: Papers of the British School at Rome, Vol. 17 (1949), pp. 26-71
Published by: British School at Rome
Stable URL: http://www.jstor.org/stable/40310474
Accessed: 13/09/2010 21:14

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=bsr.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

British School at Rome is collaborating with JSTOR to digitize, preserve and extend access to Papers of the
British School at Rome.

http://www.jstor.org
 THE SHRINE OF ST. MENAS IN THE MARYUT
PAGE
Bibliography . . . . . . . . . .26
A. Alexandria
Christian and theshrineof St. Menas . . . .27
B. The historical
record . ....... 30
C. The survivingremains. ....... 37
D. Architectural of thesuccessive
features churches. . . • 51
E. Notes on the sculpturalornamentand on the relationof Abu Mina to
and to CopticEgypt .......
Alexandria 60

BIBLIOGRAPHY AND ABBREVIATIONS

The followingabbreviationsare used throughout:

AttiIII Congr.Arch,Crist. Attidel III CongressoInternationale Cristiana,Ravenna1932 (Studi
di Areheologia
di Antichitacristianapubblicatipercuradel PontificioIstitutodi Areheologia
Cristiana,No. VIII, Roma, 1934).
AttiIV Congr.Arch.Crist. Atti del IV Congresso di Areheologia
Internationale Cristiana,Roma 193Ó (same
series,No. XVI, Roma, 1938).
B.S.R.A. Alex. . . Bulletinde la SociétéRoyaled*Archcologie
d*Alexandrie.
Cabrol, Dictionnaire . Cabrol and Leclerq, Dictionnaire chrétienne
d'archéologie et de liturgie.
De Cosson, Mareotis . A. De Cosson, Mareotis;a shortaccountof thehistory and ancientmonuments of the
north-western desert
ofEgyptandofLakeMareotis, London, 1935.
Drescher,Apa Mena . JamesDrescher, Apa Mena(Publications de la Société Copte, textes et docu-
ments,No. 1) Cairo, 1946.
Kaufmann,Menasstadt . K. M. Kaufmann, Die Menasstadtund das nationalheiligtum der altchristlichen
Aegypter in derwestalexandrinischen
Wüste,Vol. I, Leipzig, 19 10. No further
volumes were issued.
Kautzsch . . . R. Kautzsch, Kapitellstudien (Studien zur spátantiken Kunstgeschichtedes
deutschenarchaologischenInstituts,No. 9), Berlin, 1936.
Kitzinger . . . E. Kitzinger,'Notes on earlyCoptic sculpture',Archaeologia, LXXXVII, 1938,
pp. 181-215.
P.O. .... ed. Graffin
Orientalis,
Patrología and Nau, containing ofthePatriafchs
TheHistory
ofAlexandriaof Severusibn el-Muqaffa,ed. Evetts.
White, II and III . . H. G. Evelyn White, TheMonasteriesof theWadCn-Natrun, New York, 1932:
of Nitriaand Scetis.Vol. III. Architecture
Vol. II. TheHistoryoftheMonasteries
andArchaeology.
In addition to the above, the followingworksare of special relevanceto the historyof the shrineof St.
Menas and of the Maryut:
ad Aegyptum,
Breccia,Ev., Alexandrea Bergamo, 1922; in particularpp. 345-9.
Breccia,Ev., Annuairedu Museegréco-romain 19 3 1-2, pp. 23-4, pls. viii-x.
d*Alexandriet
Budge, E. A. Wallis, Textsrelatingto SaintMtna.ofEgyptand thecanonsofNicaea in a Nubiandialect,London,
1909.
De Cosson, A., 'The Desert City of El Muna\ B.S.R.A. Alex., No. 30, pp. 247-53.
De Cosson, A., and Oliver, F. WM 'Note on the Taenia Ridge', B.S.R.A. Alex., No. 33, pp. 162-220.
Dcichmann, F. W., 'Zu den Bauten der Menastadt', Archaologischer An^eiger,1937, 75-86.
 THE SHRINE OF ST. MENAS IN THE MARYÜT 27
Drescher,James,'St. Menas's Camels once more', Bulletinde la SociitiCoptcvii, 194 1, pp. 19-32.
Drescher,James,'More about St. Menas', Aúnalesdu Servicedesantiquités de l'Egyptexii, 1942, 53-70, dis-
cussingprimarilythe life of St. Menas as recordedin the so-called Encomium.
Kaufmann,C. M.t Die Ausgrabung derMenasheiligtümerin derMareotiswüste.Three provisionalreports,Cairo,
1906, 1907,and 1908.
Kaufmann,C. M., La découverte de Menasdans le desertde Maréotis,Alexandria, 1908. French
dessanctuaires
versionof the above.
Kaufmann, C. M., Der Menastempel von Karm Abu Mina in der agyptiscben
und die Heiligtümer Mariütwüste,
Frankfurt,1909. An abridgedguide.
Kaufmann,C. M., Zur Ikonographieder Menasampullen,Cairo, 1910.
Schmitz, A. L., 'Die Formengeschichte der agyptischenMenapolis: ein Beitragzu der Frage Orient oder
Rom', ArchaologiscberAn^eiger,1930, 503-16. Based in the main on Kaufmann's attributions,but
correctlyre-evaluatesan earlyimperialsarcophagusfragment.

%
A. CHRISTIAN ALEXANDRIA AND THE SHRINE OF ST. MENAS
It is one of the saddestlosses of historythatof Antiochand Alexandria,two of the
greatearlycentresof Christianart and learning,hardlya stoneshould now be standing
above ground. Of the two, Alexandriais in the worsecase. Antiochcan at least boast
the magnificentseries of mosaics unearthedbefore the war by the Princeton Ex-
pedition.1 At Alexandriaon the otherhand thereis little chance that excavationcan
everrevealany substantialremainsof the classical city,which was ruthlesslydestroyed
in the last centuryto make way forthe expansionof its modernsuccessor;and whereas
in the case of the minorarts the contributionof Alexandriato the contemporary art of
the Mediterranean,whetherin its relationto the Byzantineworld or to the nascent
Coptic art of Upper Egypt,can at least be usefullydiscussedin the light of surviving
ivories,textiles,manuscriptillustrations,and the like,2in the fieldof architecture and
of architecturalornamentany such enquiryis hamperedby a vacuum at the heartof
the problem. The questionsso well definedby Kitzingerin regardto Coptic sculpture3
are capable of widerapplication. In particular,theextentof Alexandria'scontribution to
the developmentof Christianarchitecture during the fourth,fifth,and sixth centuries
is a matterof thefirstimportanceto studentsof Coptic and of late-classicaland Byzantine
archaeologyalike.
A possible line of enquiry is suggested by the actual sculptural remains from
Alexandriaitself,now collectedin the city's Greco-Romanmuseum. These have never
been seriouslystudied; and thoughperhapstoo scantyto affordin themselvesa basis for
any far-reaching conclusions,theyhave nevertheless a decided and distinctivecharacter,
which warrantsstudy.
A second, and perhapsmore fruitful,source of informationis to be soughtin the
archaeologicalexplorationof theMaryut,the regionwhich lies to the west and south-
west of Alexandria,in the angle betweenthe sea and the westernedge of the Delta.4
1 Doro Levi, AntiochMosaic Pavements,Princeton, see notablyreviewby MissDer Nersessian, ArtBulletin
1947;C. R. Morey,TheMosaicsofAntioch , Londonand XXV, 1943,80-6; andreplyby Morey,op,cit.pp. 160-6.
New York. i<n8. See also sectionE. below,do. 6o-7l.
* A recent • See bibliography, p. 16; and sectionE, below, pp.
studyis thatofC. R. Morey,
andfundamental
EarlyChristian ArtyPrinceton,1942. A majorpointstill 60-71.
4 The history
to whichthecityofAlexandria
indisputeis theextent con- oftheMaryüt
andtopography iswellsum-
as it had beentheformative,
tinuedto be thefertilising, marisedby A. de Cosson,Martods,see bibliography,
p.
centreforthe'Alexandrian',impressioniststyle.On this 26.
28 THE BRITISH SCHOOL AT ROME

This region,which may forconveniencebe definedas stretchingas faras el-Hammam

to the west and southwardsto the belt of sand-dunesto the north of the Wadi'n-
Natrün,is now desert,occupied only by nomadictribesmen.For a periodof antiquity,
however,and as late as Roman times,it carrieda large settledpopulation,the remains
of whose farmsand villagescan be seen in verylarge numbersover the whole area. Its
abandonmentwas not due to any major climaticchange,althoughthereis evidenceof
a gradualand continuousdesiccationsince earlydynastictimes. The recurrentrecords
of flocksand herdscapturedby the earlyPharaohsin areas that are now uninhabitable
desertare too numerousand too consistentto allow of any otherconclusion.5However,
the primarycauses of its eventualdesertionwere other. Firstly,and perhapsdecisively,
therewas the collapse of the strongPtolemaic and Roman government,which alone
could ensurethe securityof these settlementsagainstthe ravagesof deserttribesmen.
Secondly,and largelyas a consequenceof thisdecreaseof security,therewas a slackening
in the methods of water conservation. Wells and cisternswere neglected,and the
kurutn (sing, karm, literally vineyard'), the rectangularbanked enclosuresfor the
'a
capture of the torrentialwinterrains and an essentialfeatureof the local economy,
were allowed to lapse. Thirdly,therewas the disastrousdrop in the level of the fresh-
watertable,which took place with the desiccationof Lake Maryüt.This lake,whichin
classical times stretchedwell beyond the modern Burg el-Arab, and carried large
river-vessels to the quays which can still be seen on the sites of the desertedlake-side
cities of Taposiris andMareotis,relied forits replenishment on the annual floodingof
the Canopic branchof the Nile. Afterthis became silted up, probablyin the fifthor
sixth century,the position was for a time maintained by the Alexandriacanal; but
despite constantattentionthis, too, passed out of use.6 The losses fromevaporation
were no longerreplaced,and the lake rapidlyshrankto the dimensionsof a brackish
marsh. It was the resultantdrop in the fresh-watertable throughoutthe regions
adjacent to the lake which dealt a final blow to the lingeringremains of classical
settlement.7
The farmsand villagesof the Maryütmust have been almost entirelydependenton
Alexandria,to which theysent wine, fruit,and market-produce, and in returnreceived
a part of the merchandisethat flowed throughthe twin harboursof that greatcity.
With the advent of Christianity,although the famousearlymonasticsettlementsof
Nitria and Scetis lay outside the Maryütproper,withinit are to be seen the remains
of a numberof contemporary buildingsand communitieswhich, thoughless austere,
no doubt reflectmore trulythe life and art of the capital itself:at Taposiris(Abu Sir,
near Burg el-Arab), a monasterybuilt within the enclosureof the great Ptolemaic
• De Cosson,op. cit.pp. 19-23. Matruhthe Romanstappedthissupplyby an ingenious
• Ammianus (xxii.15. 10) in thelatefourth saw ofrock-cut channelssitedjustabovesea-level, prov-
century system
theCanopic ('Heracleotic')branch.The Arabwriters, on ing incidentally thathereat any ratetherehas been no
the otherhand,all speak of the Alexandriacanal. The appreciablechangeof sea-levelsinceRomantimes(G. F.
processwas probablygradualand it seemsto havebeen Walpole, An AncientSubterranean Aqueduct¡Vest of
completeby thetwelfth century.PrinceOmarToussoun, Matruh,Survey of Egypt, Paper No. 42, 1932). In
MimoiresurUs anciennes branches du Nil (Mémoirespre- Alexandria detailedrecordof thewater-level in theKom
sentesá la Société Archéologiqued'Alexandrie, vol. I, el-Chogáfacatacomb,whichhassunkslightly andis now
fase. 2, époqueárabe,pp. 195-7); De Cosson, op. cit. partlywater-logged, has revealeda periodicvariationof
chapterXII. levelthatis in directrelationto thelevel oftheNfleflood
7 Throughoutthepermeablesandstonesof thecoastal (C. AudubeauBey, *Note sur l'afTaissement du Nord
regionto thewestofAlexandria, thefresh-watertablerests du Delta égyptiendepuisTempireromain',Bull. Inst.
at sea-leveldirectlyupon a bed of salt water.At Mersa d'Egypte,vol. 1, 1919,pp. 118,134).
 THE SHRINE OF ST. MENAS IN THE MARYÚT 29

temple;8at Karm Abu Girg, south-eastof Amriyah,a chapel which containeda fine
seriesof frescoes,now in the Greco-RomanMuseum at Alexandria;9nearby,at Kururn
et-Tuwal, the remainsof a small church;10south of Amriyah,at Alam Shaltüt, two
distinctbuildingswithChristianfrescoes;11 and farto thesouththelonelychapelof Qasr
el-Qatagi.12 In addition there is hardlyan ancient site throughoutthe Maryütin which
it is not possible to pick up the sherds of late Roman sigillatapottery,which testify
to an intensiveoccupationbetweenthe fourthand seventhcenturies.
In the centreof this region,eightmiles south of the lake and about fortymiles
fromAlexandria,lay the shrineof St. Menas. It is not easy nowadaysto realise the
place which this shrineonce held in the minds of contemporaryChristians,so com-
pletelydid it later vanish fromhuman record. Yet in its heydaynot only was it the
national shrine of Coptic Egypt, but foreignpilgrimscame to it from the farthest
endsof Christendom.The flasks,in whichtheycarriedawaya dropof oil fromthelamp
that burned beforethe saint's tomb, have been found fromProvenceto Asia Minor,
from Hungary to Ethiopia.13When, in the ninth century,it was threatenedwith
destructionon the ordersof the Caliph (see below, p. 35), the patriarchof Alex-
andria begged that it be spared in these words: 'Behold, all the churchesunder my
jurisdictionare beforethee. This churchalone I desire of thee that thou injure not.'
His plea was unavailing. As his enemies remarked,'None of the churchesis like this
one/ It was no pale replicaot the churchesof Alexandria,but itself the standard by
whichtheycould be judged. Justas in Syriathe greatpilgrimagechurchof St. Simeon
Stylitesgivesus a glimpseof the vanishedgloriesof ChristianAntioch,so in thedeserts
of Maryütwe can still see somethingof the Alexandriaof St. Athanasiusand of his
successors.
The site was identifiedand partiallyexcavatedearly in the centurybyMonsignor
Kaufmann,who clearedthe churchand some of the surroundingbuildingsand removed
largequantitiesof sculptureand othermaterialto Frankfurt.In the wordsof Professor
Monneretde Villard it was 'badly excavatedand worsepublished'. Kaufmann issued
severalpreliminaryreports,but of the finalreportonly the firstvolume ever appeared
in print. The plan whichappearstherein14 has, however,passed into the standardworks
of reference;and exceptfora briefcampaignundertakenby Dr. Ev. Brecciain 1926-7
primarily forthepurposeof conservation,15 Kaufmann'sconclusionswereforthirtyyears
accepted without further They firstreceivedcriticalexaminationin 1936
verification.
whenDr. F. W. Deichmann,accompaniedbyDr. A. von Gerkan,visitedthesite,and sub-
sequentlycalled attentionto some of themanifestabsurditiesof Kaufmann'saccount.16
• J. B. Ward Perkins,'The Monasteryof Taposiris 13See Cabrol,Dictionnaire9s.v. Ampouleand Menas;
Maena\B.S.R.A. Alex.*No. 36, 1945,pp. 3-8. andKaufmann, Zur IkonographiederMenasampullen. For
• Ev. BrecciainMunicipalized*
Alexandrie:rapport Hunearv.AttiJIICongr.Arch.Crist,p. 303andfie.12.
surla
marche duservice du musée.1012. 14Kaufmann. Menasstadt* abb. 32.
10R.Eilmann,A. Langsdorff, and H. E. Stier,'Bericht15Ev. Breccia,Municipalitéd*Alexandria le Musée
überdie Voruntersuchungen auf den Kurumel-Tuwál gréco-romain1931-2, pp. 23-4, pls. VIII-X. As will be
bei Amrije', Mitteilungen des deutschen seen fromthe presentarticle,the new featuresthenre-
Instituísfür
Aegyptische Altertumskunde vealedwereelements
in Kairo*Vol. I, 1930, pp. ofa structure farmorecomplexthan
106-20. he imagined.
11A. Adriani, Annuaire du Musée gréco-romain 16F. W. Deichmann,'Zu den BautenderMenasstadt',
d*Alexandrief1935-9,p. 151 ff.The second building, Archdologischer An&iger, 1937, 75-86. Deichmann's
discoveredin 1942 by Mr. C. Musgrave,awaitspubli- observations acute,butinthelightof
arecharacteristically
cation. all theevidence(someof it onlyavailablesince)his con-
11De Cosson,op.cit.pp. 141-4. clusionsarenotacceptable.
3o THE BRITISH SCHOOL AT ROME

In Julyand August 1942 the writerhad an opportunityto re-examineat leisurethe

site in the companyof Prof,(then Brigadier)R. E. M. Wheeler. Recent casual digging
by personsunknownhad revealedseveralnew featuresof interest;and it was at once
apparentthat the structuralhistoryof the tomb of the saint and of the churchimme-
diatelyabove it was farmorecomplexthanhad previouslybeen appreciated. Excavation
was, of course,out of the question. But withthe aid of superficialclearingand cleaning,
it was possibleto make an accurateplan of the existingremains,and fromthis,coupled
with a detailed examinationof the visible signs1
of structuralsequence,to deduce with
some confidencethe historyof the developmentof the shrineand of its buildings.
By a happycoincidencea ¿optic versionof the historyof the shrine,writtenat some
date beforeA.D. 892-3, has recentlybeen the subjectof a studyby Mr. JamesDrescher
of the Fuad I Universityat Cairo, who verykindlyallowed the writeraccess to his then
unpublishedmaterial. A comparisonof the resultsof the two enquiriesis sufficiently
strikingin itselfto warranta freshpublicationof the existingremains. In its initial
preparationthe writeris much indebtedto the help of Mr. Drescher;of Prof. R. E. M.
Wheeler,in conjunctionwith whom the structuralremainswereexaminedand planned;
of Mr. Alan Rowe, Curatorof the Greco-RomanMuseum of Alexandria;of JudgeC.
Brinton,Presidentof the Royal ArchaeologicalSocietyof Alexandria;and of Major W.
Jennings-Bramly of Burg el-Arab. The plan is largelythe workof Mr. D. H. White.
The writer'ssubsequentdebt to ProfessorMonneretde Villard will be evidentin the
pages which follow; and thanksare due to all those,and notablyto Dr. Erik Sjoqvist,
ProfessorG. De AngelisD'Ossat, Dr. R. Krautheimer, and Dr. F. W. Deichmann,who
have since helped the writerwith informationand criticism.

B. THE HISTORICAL RECORD

The Menas literaturehas recentlybeen the subject of a valuable study by James
Drescher,in the introductionto his edition of a Coptic manuscriptfoundat Hamouli
in the Faiyumand now in the PierpontMorgan Library(Cod. M. 590).17 This manu-
script,which is dated 893, containsthreecompositions,the Martyrdomof St. Menas,
the Miracles of St. Menas, and an Encomium on St. Menas compiled afterthe Arab
Invasion of 640 and attributedto John,Archbishopof Alexandria.18It owes its im-
portanceto its manifestly greaterreliabilitythananyof theversionspreviouslyavailable,
whetherCoptic, Greek, Arabic, or Ethiopic. The Martyrdomand Miracles follow
familiarlines, and are of greaterinterestto the hagiographer
than to the historian.The
Encomiumon the otherhand, afteran accountof the life and martyrdom of the saint,
relateshow his remainswere broughtfor burial to the Maryüt,how theywere later
discoveredand a shrineerected,and how the shrineprosperedand developed under
various emperorstill the Arab Conquest. It is a document which requires inter-
pretationin the light of the clearlyexpressedprejudicesof its author; but it remains
17James
Drescher, Apa Mena (Publications de laSodété andthewriter
takingresearches; is further forthe
indebted
Copte: Texteset Documents, No. i), Cairo,1946. Earlier loanofportionsofMr.Drescher'smanuscript beforepubli-
anides by thesameauthorare: 'St. Menas'sCamelsonce cationandformuchheloruladvice.
more',Bulletinde la Sociiti CopteVII, 1941,19-32; and 18JohnIII, 677-86,or
perhapsmoreprobably JohnIV,
'MoreaboutSt. Menas',Annalesdu ServicedesAntiquités 775-89,who had been oeconomus of the Maryütshrine.
d*EgypteXII, 1042,53-70. Itwillbeobviousinthefollow- />.O.X,p. 381; XI, p. 605.
ingparagraphs nowmuchis owedto Dr. Mrescher's pains-
 THE SHRINE OF ST. MENAS IN THE MARYÚT 31

unquestionablythe most importantsurvivingliteraryevidencefor the foundationand

earlyhistoryof the shrine.
The originsof the cult of St. Menas are shroudedin legend,and fromthe tangleof
improbabilities,contradictions,special pleading, and downrightfalsehood, which
constitutethe Menas literature,it is often hard to extractany coherentthread of
historicalfact. It seems probable that the issue was at the outset confused by the
existenceof a Phrygiansaint of the same name, who was martyredat Cotyaeum,
and that the two legends early became assimilated. Certain it is that such of the
ancient versionsas tryto accountfor the presence,in a remote desert village of the
Maryüt,of theremainsof a saintmartyred in Phrygia,fail evenby the modeststandards
of credibilityrequiredof such literature.Accordingto one accountcurrentin Egyptthe
saint was a simple camel-herd,who was born,lived, and died in the Maryüt. A second
versionclaimed him as an Egyptianof noble birth,whose parentshad migratedto
Phrygia,where aftera period of militaryservicehe sufferedmartyrdom.His body,
accordingto this account,was broughtback to Egypt,and aftera seriesof miraculous
happeningswas buried in the desert. The authorof the Encomium,who writeswith
a strongaristocraticbias, berates 'foolish men seemingto do him honour but rather
utteringthingsunfitabout him, some that he belongedto Nepaeiat (usedin Coptic, for
theGreekAi(3úti),othersthat he came fromMareotes,and othersthat he was a camel-
herd'.19It is characteristic of the earlyCoptic Church,a bodyremarkableratherforits
zeal in the faith than for any pedestrianstandardsof veracity,that it should have
preferred insteadthe strangetale of camels and sea-serpents, and of miraculousjourneys
by sea and land, which constitutes the first half of the Encomium.
On the whole the versionthat the Encomium is at embarrassedpains to refute
commandsrespectby its verysimplicity.St. Menas of the Maryüt may well indeed
have been a local saint,who lived and died in modestobscurity;and only later,as the
healingfameof his remainscame, by some chance,to spreadout of all proportionto his
fame duringlife,did he absorb the characterand title of an, until then,better-known
St. Menas of Cotyaeum.
Attemptsby moderncommentators to show thatthe saintwas a fictionalbeing,who
replaced a local pagan cult, carry no conviction.20 Extensiveexcavationhas revealedno
traceof an earliershrine,let alone of a shrinesufficiently importantto accountforthe
later popularityof the saint;21and in general terms,as Drescher remarks,such an
hypothesis,howeversuperficially attractive,singularlyfailsto explainthe practicalsteps
whereby in the space of a few yearsa pagan deity could be convincinglytransformed
into an allegedlyhistoricalmartyr.It would have been a verydifferent matterfromthe
often-attested occupation of a pagan shrine a
by being such as St. Michael, who was not
required to have earthly connexions, or from the absorption of established pagan
attributesand festivalsby an equallyestablishedlocal saint. This hypothesisraisesmore
problemsthan it solves.
From the extensiveliteraturewhich purportsto describethe life, martyrdom, and
burialof the saint, all thatwe can reallysaywith certaintyis that,at some date before
the middle of the fourthcentury,the body of a certainMenas was foundnear Este, an
19Drescher,Apa Mena,p. 120. 1om, chapterV.
t0NotablyR. Miedema,De HeiligeMenas,Rotterdam "See below,p. 38.
32 THE BRITISH SCHOOL AT ROME

outlyingvillage of the Maryüt;and that it came to acquire a reputationformiraculous

powers of healing,which soon spread beyondthe bordersof Egypt,and withina few
yearswas bringingpilgrimsfromthe farthestends of the Christianworld. The identity
of this Menas was earlythe subject of conflictingaccounts,and about the only point
upon which all of these agree is that he was in some way connectedwith the camels
with which he was thereafter invariablydepicted.
With the discovery,however,of the body, and the spreadingfameof its miraculous
healingpowers,we pass fromlegendinto history.Apartfromminorcontradictions and
obscurities,the accountof the growthof the shrineand of its buildingscontainedin
the Encomium is clear and credible. It distinguishesa successionof churcheson the
site of the grave. The firstof thesewas small and simple:

'And so theybuilt over the gravea small oratorylike a tetrapylon.They hung

a lamp in its midst like the firstone. The lamp remainedburning,withoutever
goingout, day and night. All who took awayof the oil of the lamp to distantlands
received healing; so that a great concoursegatheredthere as well as countless
multitudescoming thitherat all times unceasingly.'22

for the
It was not long, however,before this little oratorybecame insufficient
growingnumber of pilgrims:
'And theysuffereddistressbecause the place was desert,and theylacked water
and the benefitof the Holy Mysteries.Accordingly, the chiefcitizensof Alexandria
and those of Mariotes and all the archonsof Egyptbesoughtthe holy Athanasius,
the archbishop,to build a wondrousmemorialchurchto the gloryof God and the
holy Apa Mena. • . . God raised up the just and pious king,Jovian.Then the
holy Athanasiusundertookthe carrying-out of the people's requestto the gloryof
God and His blessed martyr.And when the God-lovingking, Jovian,heard, he
wroteto the stratelatesof Alexandriathat he should help him with moneyforthe
building of the church. ... He broughtit to completionin all beauty,adorning
it with preciousmarblesglisteninglike gold.
'In the days of the just kings,Valens and Valentinus,his brother,the sons of
the kingJovianof happymemory,theywroteto the august al of Alexandria,Tatian.23
He proclaimedto all the bishops of Egypt the combat( ?) of the holy Apa Mena.
And so the bishopsmet togetherand depositedthe remainsof the holy Apa Mena
in the cryptwhich had been made forthem. And theycelebratedthe feastof the
consecrationof the churchon the firstof Epép/24

This second building,the memorialchurchof Athanasius(Archbishopof Alexandria,

326-73), was, it seems,begun under Jovian(363-4). It was substantiallycompleted
beforethe death of ValentinianI in 375, and was consecratedshortlyafter. It included,
it will be noted,a cryptforthe body of the saint.
So rapid, however,was the growthof the pilgrimagethat soon this building,too,
11Drescher.Apa Mena*o. 144. IV. A 2. 2461-7.
188-02. Paullv-Wissowju
MFlavius Éutolmius *4Drescher,
Tatianus,comessacrarum¿argi- Apa Menaypp. 144-5.
tionumin Egypt,c. 375-8, and laterPraetorianPrefect,
 THE SHRINE OF ST. MENAS IN THE MARYÜT 33
needed enlargement.The third church was the work of Theophilus (Archbishop of
Alexandria, 385-412):
'When some time had passed until the days of Theodosius the great, with
Arcadiusand Honorius, his sons, in the days of the ArchbishopTheophilus, there
beinggreatpeace and prosperityin theirreign,(it befellthat) when the feastof the
blessed martyrcame round, on the fifteenthof Hathor, many great multitudes
assembled. And they suffereddistress because the church could not hold the
multitudesbut theywerestandingoutsidein thedesert. And theblessedarchbishop,
Apa Theophilus,was there. At thesightof thepeoples' distresshe wroteto Arcadius,
the king. And the king orderedthe buildingof a spacious memorialchurch. And
theylabouredwith royalpowerand might,with decorativeskill like the Temple of
Solomon. And he made it one withthe memorialchurchwhichthe holyAthanasius
had alreadybuilt. And when he had broughtit to completionin all beauty, he
conveneda synodof bishops and all the archonsof Egypt. They consecratedit in
gloryand honour.'26
AlthoughTheodosius is mentioned,the text indicatesclearlythat the new churchwas
builtunderArcadius,39 5-408. It enlarged,but apparentlydid not whollyreplace,that
of Athanasius.No mentionis made of thecrypt,althoughit is clearfromthe surviving
remainsthat this, too, was enlargedand radicallyremodelled.
Beforethedeathof Theophilusin 412 the structureof the greatchurchof St. Menas
stoodsufficientlycompletefortheceremonyof consecration.The workof ornamentwith
mosaics and marble panelling no doubt continued for many years afterwards.Two
centurieslater,in the time of ArchbishopMichael I (744-68), a dispute aroseover the
possessionof the shrinebetween the Coptic Church and the orthodoxChristiansof
Alexandria,a dispute whichwas finallydecided in favourof the Copts afterthe Arch-
bishop had given the followingevidence:
'Then the Patriarchansweredand said, "Verily my fatherTheophilus and his
successorTimothywere the buildersof this church,and it is theyalso who testify
to me thatTheophilusfoundedit and set up its rowsof pillars,forthereis his name
writtenupon them,and when he died Timothyerectedthe remainder,since there
is his name inscribed.These are my witnesseswho testifyto thisday"/28

Timothywas Archbishopfrom458 to 480, and his inscriptionpresumablyrecordedthe

finalcompletionof the church.
Timothy's contemporary, the EmperorZeno, 474-91, was, for political reasons,a
patron of the Coptic Church. He made benefactionsto the monasteriesof Scetis, in
the Wadi'n-Natrün,and at the shrineof St. Menas:
'Then the Archbishop,Timotheus,told the king, Zeno, about the barbarians
who came over Mariotes, afflictingthe shrineand all the churchesin Mariotes.
Then the kingorderedall thoseof senatorialrank in the kingdomto build each of
thema palace there. He also wroteto thearchonsof Alexandriaand thoseof Egypt,
that each of them throughoutthe land should build himselfa house there until
•• Drescher,Apa Mena,pp. 145-6. *P.O.X,p. 132.
D
34 THE BRITISH SCHOOL AT ROME

they made it a city. And so it was built and given the name Martyroupolis.
Multitudesgatheredto it fromeveryland and residedin it. And the king,Zeno,
establishedalso a garrisonof 1200 soldiersto guardthatplace againstthe inroadsof
the barbarianhorde. And the God-lovingking did this as an aid to the whole of
Mariotes and the shrine likewise. And he provided their maintenancefrom the
revenuesof Mariotes. And he also gave to the shrinesome eparchiesof Egypt,
remittingtheirtaxes that the moneymightbe used forthe expensesof the church
and the hospiceswhich he built at it.'27
The dangersfrommaraudingdeserttribesmenand theestablishment of a garrisonwere
ominousportents of the time when central authoritywould no longerbe able to afford
As
such protection. early as the middle of the fifthcenturythe monasteriesof Scetis,
which were more isolated and exposed,had alreadythricebeen devastated.28For the
present,however,the shrine of St. Menas was very prosperous,and around it was
growinga large civil settlement,which included quartersfor the accommodationof
visitingpilgrims,similar to those, for example, which can still be seen at Kala'át
Sim'án, the greatshrineof St. Simeon Stylitesin Syria. Under Anastasius,491-518,
provisionwas made also for the well-being of pilgrims on the road. A Praetorian
Prefect,Philoxenus,whose name is not elsewhererecorded,
'. ♦ . built hospices by the lake and rest-housesfor the multitude to stay at.
And he had the market-placeestablishedamong themin orderthat the multitudes
mightfindand buy all theirneeds. He had spaciousdepositoriesconstructedwhere
the multitudescould leave their clothes and baggage and everythingwhich they
broughtto the shrine. When he had completedeverything, he called it Philoxenite
afterhimself. He also set up porticoesat different
places wherethe people might
rest. And he establishedwatering-placesalong the roads/29
With the close of the fifthcenturythe period of growthand developmentcame to
an end. For the next two centurieswe have verylittle direct informationabout the
historyof the shrine. The times were troubled. The increasingweaknessof Byzantine
rule, culminatingin the Arab conquest of Egypt in 640, is clearlyreflectedin the
recurrentrecordof bedouinforays.About 570 the monasteriesof Scetis weresacked for
the fourthtime and remaineddesolate for nearlyfortyyears.30The shrineitself of
St. Menas escapedviolence,but the generalstateof insecuritymust have had an adverse
effecton thevolumeof thepilgrimage.On theotherhand it would be easyto exaggerate
this aspect. As late as the beginningof the ninthcentury,the revenuesderivedfrom
the pilgrimswere still a considerablefactorin the financesof the patriarch:
'But famineand plunderbegan to reappearin Alexandria;and the patriarch(James,
819-30) could not find that which he was wont to give to the churches,for
nothingwas leftto him. And the visitsof the faithfulfromall partsto the martyr
St. Mennas at Maryüt were interrupted;and with them the patriarchused to
trade.'31
17Drescher, Apa Mena,pp. 146-7andnotesadloc. The " In 407,434,and444. WhiteII, pp. 154-5.
two Ethiopicversionsgive thefiguresof thegarrisonas •• Drescher.Aoa Mena. d. i ¿8.
12,000andas 123,000,figureswhichfairly
mirrortherela- 80WhiteII, p. m5.
oftheseversions.
tivereliability » P.O. X, 451.
p.
 THE SHRINE OF ST. MENAS IN THE MARYÜT 35
A second passage speaks of

'• ♦ . the povertyof the churcharisingfromthe interruptionof the pilgrimages

to the churchof St. Mennas, the Martyr,in consequenceof the continualwars/32

It was these revenuesno doubt which lay behind the dispute for the shrinealready
quoted, in the timeof the ArchbishopMichael I, 744-68.
The attitudeof the Arab conquerorswas at firstconciliatory.The monksof Scetis
receiveda very favourablecharterfrom 'Amr ibn-el'Asi, respectingtheir rightsand
confirming theirrevenues.33It was not until the beginningof the eighthcenturythat
the requirements of State revenuebroughtabout a changeof policy. It maywell be that
on this occasion the shrineof St. Menas lost some of its regularincome. But it was
anotherhundredyearsbeforeit suffered majordisaster. In 8 33 the Caliph, al-Mutasim,
decided to removehis capital to Samarra,and to build the new city worthilyhe sent
agents far and wide to collect marble columns and otherprecious building materials.
At Alexandriathe task was entrustedto a certainLazarus (describedwith characteristic
gusto by Severus ibn al-Muqaffa as 'a malignanthereticof the Nestoriansect'), and
this agentappearsto have done his workwith a will:
'The People of his foul sect ♦ . . led him out to the churchof the martyr,St.
Mennas, at Maryüt,in theirgreatjealousy againstit, and then theysaid to him:
"None of the churchesis like this one, forall thatthou hast come to seek thou wilt
findherein". So that Nestorianhastilyarose, by the advice of the informers, and
entered the church of the martyr,St. Mennas. And when he looked at the building
and its ornaments,and saw the beautyof the columnsand colouredmarbleswhich
it contained,he marvelledand was amazed and said: "This is what theprinceneeds.
'
This is here and I knew nothingof it!"

The protestsof the patriarch,Joseph(830-49), were of no avail:

'Yet the hereticwould not listen to the patriarch'swords nor to his request and
robbed the churchof its colouredmarblesand of its unequalled pavement,which
was composedof all coloursand had no equal, norwas its value known/

Whereupon,we are told, the patriarch

\ . . did not cease to mournnightand day forthe calamitythat had befallenthis
For he sent for surface
holy church. And he took care to restoreit quickly.
decorationfrom Misr and Alexandria, and began to repair with all beautiful
ornamentthe places from which the pavements has been stripped,until no-one
who looked at themcould perceive that any thingwas gone fromthem/34

Bravewords! Josephdid indeed succeed in rebuildinga churchwithinthe shell of

theold churchof Athanasius;and thecryptbeneathit had perhapsbeen leftundisturbed.
But the greatchurchof Theophilus was abandoned to ruin, and some of its materials
used in the new building. The gloriesof the shrinewere gone.
3*P.O. X, pp. 467-8. "White II, p. 268. »*P.O. X, p. $12ff.
36 THE BRITISH SCHOOL AT ROME

It was soon to sufferfreshmisfortune.In 852, as a resultof disordersat the annual

festivalof the martyr,the governorof Alexandriaimprisonedthe patriarchand seized
the entirerevenuesof thefeast.35A fewyearslater,duringthepatriarchate of Shenouda,
859-80, it was to receivewhat may well have proveda mortal blow at the handsof the
bedouin,who fora timegainedcompletecontrolof the Maryut,levyingtaxesand seizing
the lands and propertyof the churchof St. Menas, and besiegingin theirconventsthe
monks of the Wadi'n-Natrün. 'The churchof the martyrSt. Menas at Maryut', we
learn,'which was the delightof all the Orthodoxpeople of Egyptbecame a desert. No
one was able to reachit in thosedays in whichtherewas the siege aroundAlexandria'.36
It does not appear that the shrineitselfwas destroyed;but it may well be doubtedif it
everrecoveredanythingof its formerimportance.
A small communitylingeredon; and by a twistof fortuneAbu Mina, once itselfthe
goal of one great pilgrimage,now became a halting-placeon another.Through it
passed the Darb el-Haj, the caravan-roadthat broughtMuslim pilgrimsfromTripoli
and the West on their long journeyto Mecca. One of these pilgrims,el-Bekri,an
Arab from Cordova,writingtowards the end of the eleventhcentury,has left us a
descriptionof his journey. The second stage westwardfrom Cairo broughthim to
El-Muna (or El-Mena), a ruinedcityon the westernfringeof the Delta, and the stage
beforeAbu Mina itself.37

'The next stage afterTerremontis El-Mena, a districtwhich includes three

desertedtownswith theirbuildingsstill standing. Several finecastles can be seen
out in the sandydesertwherethe caravansrun the risk of attackby nomad arabs.
These castles, which are verystronglybuilt and have remarkablyhigh walls, are
almostall raisedon brickvaults. El-Mena has severalwells of sweetwater,but the
supply is limited.
The next stage after El-Mena is Abu Mina, a great church containingvery
strangeimages and statues. Lamps burn thereday and nightand are neverextin-
guished. Below the buildingcan be seen a greatdome and withinit the image of a
man standingupright,eitherfootupon a camel, with one hand open and the other
closed. The group is carvedin marble,and is said to representSt. Menas. In the
same churchcan be seen also figuresof all the prophets(God's blessingbe with
them). That of Zacharias is beside that of John. The figureof Jesusstands on a
greatmarblecolumn,on the rightas you enter,guardedby a locked door. A double
curtainhangsin the frontof the figureof Mary.
Outside the church are images of all kinds of beasts and of craftsmen.The
figureof a slave-merchantstandssurroundedby the objects of his trade,and in his
hand a purse with a hole in the bottom. This symbolisesthe impossibilityof a
slave-merchantmaking a fortune.In the middle of the churchthereis a domed
tabernaclecontainingimageswhichpurportto be angels. In one partof the church
can be seena mosque,perfectlyoriented,wherethemuslimspray. The surroundings
>5HistoryofthePatriarchs oftheEgyptianChurch,ed. althoughthesimilarity
are clearlydistinguished, ofnames
Aba al-MaslhandO. H. E. Burmester,
YassA* vol.II, partI, has at timesled to someconfusion.For theidentification
Cairo 1943,p. 3. of El-Muna,see A. de Cosson,B.S.R.A. Alex.,No. 30,
**
Op. cit.p. 56 ft.andp. 59 ff.
17The 1936,pp. 247-53.
twoplaces,El-Muna(or El-Mena)andAbuMina,
 THE SHRINE OF ST. MENAS IN THE MARYÜT 37
of the churchare full of fruit-trees,especiallysmooth-barkedalmond-treesand
carob-trees,the fruitof which while still green tastes like honeyand is used for
makingsyrups.There are also manyvines which providegrapesand wine to send
to Cairo/38

Abu Salih, writingrathermore than a centurylater, recordsthat the shrinestill

existedand thatit still containedthe bodyof the saint.39It was probablysoon afterthat
it was finallyabandonedand the body broughtto Cairo, whereit seems to have been
installedby the.middle of the fourteenth century.40None of the laterArab historians
mentionthe shrinein the desert. It passed into oblivion, leaving only heaps of sand-
coveredruinsand a name, Boumnah,bywhichit was knownto the bedouin,to markthe
place of whathad been one of the greatsanctuariesof ancient Christendom.

C. THE SURVIVING REMAINS

The cityof St. Menas consistsnowadaysof a tangleof sand-coveredmounds,rising
in places thirtyfeetand coveringan area of manyacres. Kaufmann'sexcavationscom-
prised not only the churchitselfbut the whole quarteradjacent to the north;a large
outlyingchurchin the middle of a cemetery;a second,smallerchapel; a bath-building
with a basilical hall; wells and cisterns;wine-vats;potteriesfor the manufactureof
figurines,ampullaeand othervessels;and a numberof dwelling-houses.Stone fromthe
lake-sidequarries and, in places, brick were used for public buildings; while private
houses were of plasteredmud-brickwith stone door-jambs.
Withinthechurchitself,fivemainstructural periodscanbe recognised.The first
consists
simply of an earth-cut catacomb. If therewas at first
any corresponding structure above
ground, no traceof it nowremains. Later there was built overthecatacomb a smallchapel
of mud-brick.Then came a stonechurch,and at the same timea part of the catacomb
was replacedby a stone-builtcrypt. The fourthphase saw the greatestexpansionof the
shrine. Churchand cryptwererebuilt,with the additionof a baptistryto the west,and
of a new, and larger,churchto the east. Finally the new churchwas abandonedand a
smaller,simplerbuilding erectedon the site of the earlierchurch. It can hardlybe
doubtedthatthe grandiosereconstruction of the fourthphase representsthe workof the
ArchbishopTheophilus and of his successor Timothy,underwhom we know that the
shrineattainedits maximumdevelopment.With this fixedpoint,we get the following
identifications:
I. The originalburial-place. An earth-cutcatacomb, pagan in origin.
II. 'An oratorylike a tetrapylon',built over the grave. A mud-brickchapel built over the catacomb.
Mid-fourthcentury.
III. The memorial churchand crypt,built by Athan- The firststone churchand crypt.
asius, and completedsoon after 375.
•* AftertheFrenchversionof G. de Slane,Description What was lackingwas security.
conditionspermitted.
de VAfrique par El Bekri(second edition,
septentrionale Kaufmann'sexcavationsrevealedthe remainsof wine-
Paris,1913). FromAbuMinathenextday'sjourneyledto presses.
8fTheChurches
el-Hammam.The accountof conditions in theMaryutis andMonasteriesofEgypt. . . attributed
veryvaluable.Settledlifehad ceased,althoughtherewas toAbüSáüh,theArmenian,ed. EvettsandButler,p. 103.
whereother *BDrescher,Apa Mena, p. xxxi,n. 2.
evidentlystillwaterenoughforagriculture
38 THE BRITISH SCHOOL AT ROME
IV. The great new church,'made one with*its prede- The firststone churchand cryptremodelled. A
cessor by Theophilus, 385-412. Finished by new, and larger,churchadded to the east; also a
Timothy,458-80. baptistryat the west end.
V. The pillage of the churchby Lazarus and its re- The churchof Theophilus abandoned;and a small
building by Joseph,8 33-49. churchrebuilton the site of that of Athanasius.

I. TheEarlyCatacomb.
Period
Of the catacomb in which the saint's body was originallylaid, less than half now
remains. The rest has been destroyedby later building. The soil here is a compact,
sandy gravel,and the catacomb consistedof a connectedseries of earth-cutchambers
with radiatingloculi, of a typealreadyfamiliar,in varyingdegreesof elaboration,from
the Christiancatacombs at Alexandria,41and derived ultimatelyfromthe Ptolemaic
and Roman tombs of the same region. It would have been enteredeither by a shaft
or by steps, and theremay well have been also some formof antechamberin the part
now destroyed.It was lined throughoutwith white gypsum-plaster, and when first
excavatedthis bore tracesof simple ornamentand of graffiti.Fromthe morenortherly
of the two chamberswhich lie to the east of the PeriodIV crypt(Pi. VI, 2), Kaufmann
mentions'the Constantinian monogram'(not morepreciselyspecified)and a graffito which
recordsa certainFatherJoseph.The second chambercontainedthe body of a Father
Paul.42 In the westernrangeKaufmannobservedfurtherexamplesof the Constantinian
monogram,and a crosswith the lettersIC XC above the horizontalstroke. The latter
mayalmostcertainlybe completedas IC XC NIKA, inscribedwithinthe fourquadrants
of the cross,a formulawhich is recordednot only from the Karmoüz catacomb at
Alexandria,43but also on an inscribed boss now in the Greco-RomanMuseum at
Alexandria,fromBahig, a village in Maryütdue northof Abu Mina.
Kaufmannstates furtherthat severalof the graves,thoughrifled,showed signs of
re-use,and thatsome had originallyservedforpaganburials. It is evident,however,from
the productsof the potters'workshopsof the fifthand sixthcenturiesat Abu Mina that
the cult of St. Menas long retaineda decidedlypagan flavour.At Ehnasiya,too, Petrie
noted the survivalof large numbersof pagan terra-cottafigurines well on into thefifth
century.It would be veryrashtherefore on the availableevidenceto argue,as morethan
one writerhas done,44thatthe cult of St. Menas grewout of thatof a pre-existing pagan
shrine. All we can say forcertainis thathere,as elsewhere,older beliefsdied hard and
were only graduallyabsorbed; and that here at the end of the thirdcenturywas the
cemeteryof the little village of Este. It was in such a humble gravethat the body of
St. Menas was firstlaid.

PeriodIL ThePrimitive Oratory.

Whetherany structureabove groundmarkedthe primitivecatacombwe cannotnow
say. If it did, all tracehas since been sweptaway; and the accountin the Encomiumis
41See Cabrol,Dictionnaire.
s.v. Alexandria 43Cabrol,Dictionnaire,
fig.284.
41Kaufmann,Menasstad 44R. Miedema,De HetltgeMenas, Kotterdam1913,
t, p. 66. The inscriptionof
FatherPaul containstheformulaXMF,forwhichsee G. chapterV; MetaE. Williams'WhencecameSt. George?',
Lefebvre, Recueil des inscriptions grecques-chrétiennes B.S.R.A. Alex., No. 30, 1936, pp. 79-109.
The only
d'Egypte,Cairo,1907,p. xxxiiand No. 32. It is common specifically paganobjectsquotedbyKaufmann area terra-
on papyriand ostrakaof thefifthand sixthcenturies.At cottafigurine ofBesfromdie catacomb, anda smallsteleof
Alexandriait occurson datedinscriptions
ofa.d. 537and Horus-Harpocrates (Menasstadt,fig.24).
570.
 THE SHRINE OF ST. MENAS IN THE MARYÜT 39
heretootingedwithlatercolourto carrymuchweight.The first buildingof whichwe
haveanycertainremainsis the 'oratory whichwas erectedafterthe
like a tetrapylon',
ofthebodyandofitsmiraculous
rediscovery healingpowers.Littleofthishassurvived;

Fig. 1.- Successive Tomb-Structures Beneath the Church

butfortunatelythatlittleis bothdistinctive and clearlydatedin relationto

in character
the buildingswhichsucceededit.
.. Immediately to theeastofthefooting,whichin PeriodIV carriedthewestern arcade
of thenarthexof theArcadianbasilica,are two smallapsidalplinths. These are con-
structedof mud-brick, roughly strengthenedwithlimestone chippings.The upperand
4o THE BRITISH SCHOOL AT ROME
outerfacesarefinished in smoothgypsum-plaster, witha bevelat thejunction, whilethe
innerfaceis roughand can neverhavestoodfree.At eitherendtheyareinterrupted by
wallsof PeriodsIII and IV, whilethetrampleof builders'rubbishassociatedwiththe
construction ofthesouthern apseis cutbythefoundation-trench ofoneofthetransverse
wallsof thePeriodIII sanctuary (Fig. 2).
Of thestructure whichstoodupontheseplinthsnothing remains;noris anyother
part of this building now visible, although it is verypossiblethatsomeportionof it
may survive beneath the later floor,to the west of the semi-circular confessio(ifsuchit be)
whichmarksthepositionof thetomb-chamber at theeastend of thePeriodIII nave.
It is a reasonable presumption thatthetwoapsesflanked a third,centralapse,and this
wouldnaturally a of
suggest chapel simple basilical form. Unfortunately the nearest
local analogy,the mud-brick chapel at Kurum et-tuwál45 is incomplete. has three
It
aisles, but the east end is missing. It is, moreover, undated.
A possiblealternative is suggested by the description in the Encomium,'a small
oratory, like a tetrapylon'. This phrase would well describe the typeof four-square
domedtomb,sometimes open in the form of a four-way arch, sometimes enclosinga
chamber, which was early current in Syria46 and in Egypt and was to become sucha
commonfeature in theIslamicworld.In theremarkable of
cemetery El-BagawSt in the
Kharga oasis the commonest form of well-to-do tomb chapel is a square structure of
mud-brick withfourshallow,archedrecesseswithin,and a well-developed spherical
vaultwithpendentives reaching downintothespandrels of thearchesbeneath.47 There
is onlyone door,but falsearcadingin stuccois usedexternally to givean architectural
effect.Severalof the buildingshavein additiona singleprojecting apse on the side
the
facing entrance, although none has more than one. The cemetery was excavated
manyyearsago by the Metropolitan Museum of New York. No finalreportwas ever
published, but in a brief interim report the excavators assigned it in generalterms
to the fifthcentury,48 although it no doubt embraced in all a considerably longer
period.
The literary evidenceand generalprobability alikeinclineto suggestthatit was in
facta buildingof thissortwhichfirstmarkedthesiteof thesaint'sgraveand housed
the lamp thatburnednightand dayin his honour.The actualremainswouldmore
naturally be takento belongto a smallbasilicalstructure. Therecan be no certainty
withoutthe excavationof the undisturbed area immediately to the west of the
confessio^9
45R. Eilmann,A. Langsdorf, and K. E. Stier,'Bericht deuxpilastres dontun esttoujoursplusgrandque l'autre,
iiberdie Voruntersuchungen aufden Kurum-el-tuwál bei se détachent á angledroitdu murantérieur.Les ^oupoles
Amrije*,Mitteilungen desdeutschen furAegyptische reposentsurdes aresde souténement
Instituís adossésaux quatre
Altertumskunde in Kairo,Vol. I, 1930.pp. 106-29. murs,et leurhauteurdependde Jahauteurde ees aresqui
46E.g. thewell-known tombat Brad,Northern Syria, sont tantótsoubaissés,tantótplein-cintre, ou mémeen
H. C. Butler,Ancient ArchitectureinSyria,1920,pp. 299- ogive. Les coinsentreles arcs sontremplispar les pen-
300, ill. 329. See also S. Guyer*ZweispatamikeGrab- dentifs des coupoles'.
monumente und deráltesteMartyr- 48Bulletinof theMetropolitan Museumof Arts,New
Nordmesopotamiens
grabtypusder ChristlichenKunst', Fünf Jahrtausend York, III, 1908,pp. 203-8.
4i It has notseemednecessary
Morgenldndischer Kultun Festschrift Max Freiherrn von to rebutin detailKauf-
Oppenheim, Berlin,1933,pp. 148-56. mann'smanystrange theories.Itis perhapswell,however,
47W. de Bock,Matirtauxpourservird I archéologte de to pointout thatthe allegedexistenceof an octagonal
VEgyptechrétieriy St. Petersburg, 1901. The cemetery lies Constantinian buildingis basedupon nothingmorethan
5 km.northofel-Khargeh.De Bock'sdescription isworth twoofthecolumn-bases ofthePeriodIV columnar exedra
quoting: 'La plupart destombeauxsontcarrésá uneporte, above theconfessio (see pp. 48-9).
etontunecoupole.Des deuxcótésdelaporte,á l'intérieur,
 THE SHRINE OF ST. MENAS IN THE MARYÚT 41
Period ofSaintAthanasius
III. TheChurch soonafter
(completed J7j).
(a) TheTomb.At the same time as Athanasiusbuilt a 'spacious memorialchurch1
in honourof St. Menas, he also converteda part of the earlycatacomb into a crypt,
withinwhich,in the presenceof the assembledbishops,the saint's body was solemnly
laid. This cryptwas largelydestroyedsome thirtyyearslaterwhen,in PeriodIV, church
and tomb alike were rebuilt. The east wall, however,of the later tomb-chamber
incorporates a fragmentof earliermasonry,includingthe greaterpart of an arch,which
has been blocked by the brick vaulting of the later corridor(see Pis. V and VI, 1).
This arch opened on to a barrel-vaultedpassage, lined with plaster, which sloped
upwardstowardsthe east; and fromits centralposition on the axis of the overlying
church,it was probablythe entranceto the actual tomb-chamber.The floorof the
tomb was then some threeor fourfeetabove that of Period IV, and approximatelyon
a level with that of the earliercatacomb.
The rest of the Period III tomb-chamberhas vanished. To this period, however,
probablybelongs also the low-pitchedfoundationarch,50which marks the western
boundaryof the presentsemi-circularconfessio. It is clearlydesignedto give stabilityto
some heavystructureat thepointwhereit crossesthe underlying catacomb; but it seems
an unnecessarily elaboratedevice if it werein factcontemporary with the PeriodIV walls
which lie directlybeneathit. These would appear ratherto be a later insertion;and it
was by theirconstructionthat the primitivecatacombwas finallysealed. In Period III
on theotherhand thecatacombwas apparentlystillaccessible,to judge fromthe number
of graffiti which Kaufmannfound,some of which name the writeras 'Abba' and must
belong to a time when therewas an organisedcommunityat hand. One of theserefers
probably the EmperorsValens and Valentinianand can be dated thereforebetween
to
the years 364 and 375.61
One furtherfeaturecan be assigned to this period. Within the extremenorth-
easterncornerof the ninth-century churchlies partof a shaft,a metresquare and nearly
two metresin depth. The northernface of this shaftconsistsof one of the transverse
foundationsof the sanctuaryof the Period III church,which it can thereforehardly
antedate. On the otherhand to the east it has been cut shortby a foundationof Period
IV. Except on this latereasternface it is lined with gypsumplaster,and to the west it
opens,down a low step,on to a similarlysurfaced,barrel-vaulted corridor.The corridor
is cut shortand blocked by the masonryof the easterncorridorof the Period IV crypt.
Togethershaftand corridorconstitutea familiartype of tomb-entrance, of which at
least one example has alreadybeen recordedfromthe Maryüt,at Kurüm et-tuwál,to
the south of Amriyah.52But neitherin characternor in its relationto the churchdoes
it suggestthe formalentranceto the tomb of the saint, which probablylay on the
axis of the church. It was perhapsthe entranceto what remained of the earlycata-
comb, which,as we have seen, was only finallysealed by the architectsof the follow-
ing period.
(b) The Church.Over the. crypt Athanasius built a 'spacious memorial church'.
Less than half a centurylater his successorTheophilus rebuiltit almost in its entirety
50Visiblein Plate VIII, 4, extreme
left.The original 61Kaufmann,
Mtnasstadt,p. 65.
is nowmuchobscuredbytherestoration, under- MR. Eilmann,A. Langsdorf,and K. E. Stier,op. at.
masonry
takenby Brecciain 1926-7. pp. 106-29.
42 THE BRITISH SCHOOL AT ROME
and to-daythe foundationsonlyof the earlierbuildingremain.While therecan be
littledoubtthatexcavation
wouldrecovertheentireplan,at presentmuchof it must
remainconjeeturaL
necessarily

Fig. 2.- The Church of St. Athanasius, and Remains of Earlier Structures

Like its successor,

it seemsto have been built on a T-shapedplan, thoughon a
smallerscaleand withlessmarkedly projecting
transepts (Fig»2). To theeastextended
semi-circular apses,probablythreein number.These weredestroyed, foundations
and
all, forthe masonry theycontainedby the buildersof PeriodIV, who leftonlythe
stumpsof thecentralapse to carrythebroaderfootings of thewesternentranceto the
 THE SHRINE OF ST. MENAS IN THE MARYUT 43
newchurch,and thatpartof the northern apse whichcouldbe incorporatedintothe
continuousfootingof the northernarcade. Of a southernapse no traceis visible,
althoughthescarof its northern
abutment on themaineasternfoundationmighthave
beenexpectedin theareaexposed.It is mostunlikely, thatthechurchwould
however,
in facthavebeenasymmetricalin thisimportantrespect.

Fig. 3.- Structures in the Central Area Above the Tomb (Shown in Dotted Outline).
Ninth-Century Additions Are Omitted

The lateraland westwardextentof the transepts cannotbe determined beyond

questionwithoutexcavation;but theycan be deducedwithsomedegreeof probability
of theworkof thePeriodIV builders.It has alreadybeennoted
froma consideration
thatthe shallow,archedfoundation, whichbounds to the west the presentsemi-
circular
shaftbeneaththeconfessio, appearsto be of an earlierdate thanthePeriodIV
wallswhichunderlieit (seep. 41). This foundation can be seento extendsouthwards,
44 THE BRITISH SCHOOL AT ROME

thoughit hasbeenmuchinterrupted bylaterwork.Northwards it awaitsexamination.

In bothdirections it is alignedon theprojecting shoulders whichat thispointinterrupt
theouterwallsofthePeriodIV building.Theseshoulders servenoveryclearpurposein
theArcadianstructure, whichelsewhere followsa remarkably coherent and consistent
plan, and it is reasonable to suppose thatthey are, in fact, the western anglesof the
Athanasian transept, which have been incorporated into the laterdesign.
A singularfeatureof theplan of the easternpartof thischurchis thenumberof
foundations, fourin all, whichdivideit internally fromeastto west. Whiletheouter
pairprobably delimit side-chapels, the inner pair are perhapsto be associatedrather
withthecryptand its entrance.Unfortunately bothare maskedat theirwesternend
by the overlying masonry of two succeeding periods.The depthof all thefoundations
in the easternhalfof the church(in places2 metres)is sufficiently explainedby the
existence of thecatacombbeneath;and thesameconsideration mustbe heldresponsible
fortheconsiderable irregularity of theplan in its detailedlay-out.
Of thenaveofAthanasius' churchitis hardtospeakwithanycertainty. The problem
can be simplystated.On the site of the navethereexistthreepairsof foundations
running fromeastto west.53Of thesetheoutermost pairbelong,at leastfromground-
levelupwards,to PeriodIV. In PeriodV theintermediate paircarriedtheouterwalls
ofthenave,whiletheinnerpairformed thecontinuous footing ofthenave-arcade. The
intermediate at
pair any rate belongoriginally to Period III; foron thesouth side they
can be shownto be contemporary withthepresumed westwall of thetransept.What
is not certainis whether at thistimethis intermediate pair carriedthe outerwalls
of thenaveorthecolumnsofthenave-arcade. The plan(Fig. 2) illustrates conjecturally
thefornier alternative, theproportions ofwhichare,on thewhole,themoreconvincing.
Excavation wouldno doubtdecidethematter.
A further possiblelineofenquiryis suggested bytheexistence towards thewestend
ofthechurchofa cistern, in theformofa passage80 metres long,which crossestheline
of the navefromnorthto southand to whicha well-shaft gave accessimmediately
withinthe centralwesterndoor. At three unspecified points withinthis cistern
Kaufmann notedtheexistence of foundations, designedto supportthefootings of the
churchabove. Two ofthesehe statesthathe partially destroyed; but there can be little
doubtthata re-examination of thiscistern(whichwas not practicable to the writer)
wouldyieldevidenceas to therelativedatesof theoverlying structures.

PeriodIV. ThechurchofTheophilus, c. 400-10.

consecrated
(a) The Tomb.The Encomiumspeaksonly of the 'wondrousmemorialchurch'
whichthePatriarch Theophilusaddedto Athanasius' church.It is clear,however, from
a studyof theactualbuildingsthatat thesametimehe rebuiltand enlarged thecrypt
to meettheneedsofthepilgrimage, andthatin theirpresentformthesubstructures are
almostentirelyhiswork.Theyconsistofa lofty, barrel-vaultedstaircase (Pi. V) leading
downfromthenorth church
sideoftheprevious tomb-chamber
toa central on thesiteof
theearliertomb;to the westof thetomb-chamberis a small,
square, domed chapel,with
MIn additionthereare possibletracesof a rangeof comment, fig.22). On
a secondon his plan(Menasstadt,
column-basesimmediately withinthe line of the inter- thesouthsidecanbe seenthematrix fromwhicha similar
mediatepairof footings.One of theseappearsto be still base mayhavebeenremoved.
insituon thenorthside,and Kaufmann without
indicates,
 THE SHRINE OF ST. MENAS IN THE MARYÜT 45
an antechamber at its northernend, while fromthe oppositeside of the tomb-chamber a
barrel-vaulted corridorleads to theeast,thenturnsnorthparallelwiththe entrance-stair-
case, and finallyeast again,up a second flightof stairs,into the north-west cornerof the
newchurch.The walls throughout areofgood ashlarmasonrycoveredwitha thickcoating
of coarsestucco,which servedas a basis fora revetmentof marbleslabs. The vaulting
is carriedout in kiln-bakedbrick,widelyjointedin a hardmortarand coveredoriginally
with stucco or mosaic ornament.54Kaufmannrecordsa stucco patternof rectangular
cofferswith centralrosetteson the barrel-vaultof the staircase;while a fragmentof the
figuredgold mosaic fromthe domed chapel is preservedin the Greco-RomanMuseum
at Alexandria.56No doubt the tomb-chamber itselfcarrieda similarmosaicdecoration.
The building techniquethroughoutis uniform,and the whole cryptwas evidently
planned as a unit. Its relationto the buildingsabove groundis, moreover,certain;for
incorporatedinto the barrel-vaulting at two points, and integralto it, are two stone
relieving-arches. These carried the footingsof the centraland northernsemicircular
exedrae,respectively, of the narthex(see pp. 48-9), both of which certainlybelonged
to the Period IV structure.
v
When the shrinewas finallydesertedthe cryptwas ransacked;and subsequently,as
the churchabove fell into ruin, quantitiesof masonryand rubble collapsed into the
vaults beneath. It is, however,still possible to obtain a fairlycompletepictureof the
tomb as it was in the fifthcentury.The chapel to the west of the tomb-chamberis
structurally all but intact,althoughthe centreof the dome had fallenand has since been
restored.The dome, whichis of brick,is supportedon sphericaltriangularpendentives,
based on four irregularbut roughlyrectangularcorner-pilasters, which in turn carry
the archesof fourshallow recesses(Pi. VI, 3). The curvatureof the dome is distinct
fromthat of the pendentives.56The southernmost of the recessesopens on to a small
annexe with a brick half-domecarried on squinches(Pi. VI, 4).67 This chapel is in
manyrespectsthe counterpart,below ground,of the square tomb-chapelsof the El-
a hole
Bagawátcemetery(see p. 40). A deep shaftsunk in the floorof the annexeand
leading into the early catacomb are the work of later robbers. The irregularity the
of
masonry was concealed by figured mosaics on a gold ground, which probablycovered
the whole surfaceof the vault and possiblyof the walls also.
The tomb-chamberitself has fared worse, and the survivingremains have been
furtherobscuredby the necessaryworkof restoration, undertakensome fifteen yearsago.
The whole of the northernhalf of the west wall and much of the southernpart of the
east wall are of modernmasonry.It was originallydivided into two unequal portions,of
whichthe southern,and larger,part constitutedthe actual tomb-chamber, while at the
northernend was a corridorby whichthe pilgrimwas led fromthe footof the staircase
54The of thevaultis clearlyvisible ofbrickwork.
lifeofa localtradition
peculiarstructure long
**Kaufmann,
in Plate V. The lowercoursesare laid flatand corbelled Menasstadt,fig.16. Fromthecharacter
outwards, whilethecrownof thevaultconsistsof bricks of the masonryit is certainthatthismosaicsurfacewas
laidon edge. Thismethodavoidstheuseoftimber center- envisaged at themoment ofconstruction.
Monneret de Villardhaspointedoutto the 66See K. A. Creswell,Early MuslimArchitecture, 1,
ing. Professor
thewriterthatthetechniqueis exactlyparalleledin the thevaultafterrestoration.
p. 319 and fig.394,illustrating
aqueductofBasatinat Cairo,builtby Ibn Tulün; K. A. The lattershould be comparedwith PI. VI, 3, after
Creswell,Early MuslimArchitecture, voh II, Oxford, Kaufmann, Menasstadt,pl. 8. Creswell(p. 319) is mis-
1940,p. 330,pl. 95b. But thedate of the vaultsof the takenin sayingthatdomeand pendentives areofthesame
Period IV cryptcan hardlybe questioned,and the curvature.
oftechnique mustbe takenas an instanceofthe 67Kaufmann, Menasstadt,pl. 9. 6 and 8.
similarity
46 THE BRITISH SCHOOL AT ROME

throughthe antechamberand into the domed chapel, beforepassing into the actual
tomb. The stump of the narrowpartition-wallwhich separatedthe corridorfromthe
tomb-chambercan be seen high up in the easternwall (PL VI, i); also the springof
the vault overthe tomb-chamber, while the northernfaceshows tracesof a barrel-vault,
similarto that of the staircase,but at rightangles to it and at a lower level (visible in
PL VIII, 4). The semicircularshaftdown to the tomb-chamber, so marked a feature
of the survivingremains,had no visible counterpartin antiquity. It was a concealed
structuralfeature,designedto ensure the stabilityof the exedra and of the confessio
above, and has only since been revealedby the collapse of the underlyingvaults.
The tomb itselfwas originallyvaulted,possiblywith a brickdome similarto thatof
the adjacent chapel, but moreprobably,in view of the limitedheightavailable,with a
sphericalvault. This would have been the 'greatdome below the building', which El
Bekri saw (p. 36). A fragmentonly remains, in the south-westerncorner;but the
characterof this fragmentis distinctive,for it includes portions of the archivolts
spanningtwo of the recessesand of the pendentivebetweenthem.58It is possible that
in thiscase the brickcorner-pilasters,of whichthereis no trace,werereplacedby marble
colonettes.In the centreof the southernwall of the tomb-chamberis the matrixof a
large slab, i-8o metresin breadthand recessed 2 centimetresmore deeply than the
surrounding marblerevetment.59Here withoutdoubt was thegreatimageof St. Menas,
standing with hands upraisedin prayerbetweentwo camels. Thus he is depictedon the
small-marblerelief,verypossiblya copyof thatin the parentshrine,fromthe associated
monasteryof St. Thekla at Ennaton,near Alexandria(PL VIII, i).60 Thus, too, on an
earlyfrescofromthe chapel of Karm Abu Girg in the EasternMaryüt(PL VIII, 2\61
on the ivoriesof Alexandria,62 and universally on theampullae. It is true thatEl Belcri,
a most intelligentobserver,describedit as the statueof a man standingupon, and not
between,two camels; but in the face of the consensusof ancientrepresentations, it is
hardlylikely that the archetype should have so differedfrom all its copies.
It is probablyno accidentthat the staircaseleadingdown to the tomb is so aligned,
askew to the axes both of the tomb and of the church,as to have given the descending
pilgrimfreevisionof the site of the imagefromthe momentthathe reachedthe head of
the stairs. Kaufmannrecordsthe discoveryof quantitiesof charredwood in the tomb-
chamber,and the partition-wallbetweenthe tomb-chamberand the entrance-corridor
to the northmay well have consistedin part of an open-workwooden screen.
Of theotherinternalfittings of the tombit is impossibleto speakwithanycertainty.
The only light,otherthan that fromthe entrancestaircaseand fromthe sacredlamps
to which the textsrefer,came froma small shaftleading down fromthe back of the
exedra above to the centreof the east wall of the tomb-chamber.The tomb itselfhas
vanishedwith the image,but fromthe presenceof the exedraabove, it is reasonableto
conjecturethat it stood freenear the screenwall, oppositethe image and axiallyto the
48For the
surviving angleof thevaultsee Kaufmann, Museumat Alexandria, decorateda mud-brick building,
Menasstadt,pl. 9* i. whichwaslaterrebuiltinstone.Theyrevealcertain analo-
*• Kaufmann. MenasstadudI. k. with the at
frescoes Bawit.
60Now in the museumat Alexandria.See B.S.R.A. gies
62An
ivorypyxisin the British
Museum,B.M. Cata-
Alex., No. o, 1907,pp. 3-12. logueoftheIvoryCarvings oftheChristian Era, p. n, pl.
61Ev. Brecciain Muncipalitéd*Alexandriei sur
rapport vii; a panelfromthe Bishop'sthroneat Grado,now in
la marchedu servicedu musée,1912; see also De Cosson, MilanMuseum,Cabrol,Dictionnaire, vol. 1,1114,fig.270.
Mareotispp. 147-9. The frescoes, whichare now in the
 THE SHRJNE OF ST. MENAS IN THE MARYÜT 47
churchabove. Add the marblefloorand the walls and the gilded mosaicsof the vault,
and it is not difficultto imaginethe chamberwhich,fornearlynine centuriesuntil the
finaldestructionof theshrine,held therelicsof St. Menas.
(b) The Church.The church itself was completely excavated and described by
Kaufmann,and the plan whichhe publishedhas since passed into the standardworksof
reference.Though devisedon opulentlinesit presentsfewcomplexities,and Kaufmann's
account is here least unsatisfactory. The ground-planis that of a T, with a single,
projectingapse. The nave is of normal three-aisledbasilical form. The transepts,also
of three-aisledbasilical form,have in additiona rudimentary terminalaisle, whichgives
the effectof an ambulatory.The terminalaisle is a commonfeatureof Coptic churches
elsewhere,63 as also is the use of decorativeapsidal recesses(at Abu Mina characteristicof
PeriodIV), one ofwhichcan be seenat theend of each transept.Nave and transeptsalike
were composedof a double order,with matronea over the lateralaisles.64 On the north
side of the nave a door leads into the adjacent domesticquarters,and on the south a
tripledoorway,flankedby columnson octagonalbases(Pi. VII, 2), constitutesthe main
entrancefromthe southerncourtyardand fromthe streetbeyond.
The masonrythroughout is somewhatroughlycoursed,and is less finethanthatof the
Athanasianchurch. Characteristicalso of all the Period IV work is the use of a brick
bonding-course,made up of kiln-bakedbricksmeasuring25 by 12 by 6 centimetres.
The rathersummarycharacterof the masonryis readily explained by the elaborate
decorativeveneerwith whichit was covered. As in the crypt,the walls werefacedwith
panelsof marbleset in a thickbed of coarsestucco,oftenmixed with potsherds,and the
floorwas paved in part at least with largesquare slabs of the same material. Both were
removedin the ninthcentury,and onlya fewbrokenfragments now remain. These are
all of a fine-grained island marble,white banded with irregulargreystreaks;but the
Encomiumspeaksalso of colouredmarblesand of the 'unequalled pavement,whichwas
composedof all colours'; and elsewheretwo of the decorativeniches of Period IV still
retaintracesof polychromemarblepanelling.65The columnsof the two ordersof the
naveand transeptswereremovedat the same timeas the pavement,but it would appear
thatthesewere throughoutof the same white,island marble.
The roofwas undoubtedlyof timber. The span of the nave (14*8 metres)was con-
siderable. The roofingof the junction of nave and transeptmust have presenteda
considerableproblem,forthe corner-columns are uniformwiththerestof thearcadeand
no provisionis made to carryany extraweight. Kaufmannrecordsthatat this point he
foundconsiderabletracesof fallengold mosaic. The positionof the high altar, in the
centreof thecrossingand at the focalpointof thewhole architectural plan, is markedby
the fourbases of the canopythat shelteredit. Around it, occupyingalmost the whole
area within the four angle-columnsof the crossingand extendinga short distance
towardsthe apse, ran a rectangularchancelscreen. At the back of this chancel,with its
back to the apse and facingthe altar,stood a raised dais, gentlycurvedand steppedon
w See below,sectionD, p. 55. ofwhichthereis nowno trace.Butperhapstheiniquitous
"The elementsof the double order are summarily Lazarus did his work too well. For the column-bases,
by Kaufmann
illustrated (Menasstadt>p. 89,fig.39), who see PL VII, i, and p. 63.
removedthe bestof thesurviving architecturaldetailto 68In thenarthex, bythePeriodV builders(PL
retained
Frankfurt. ProfessorMonneretde Villard,who has in- III, 1, and Fig. 4). Tracesof colouredmarblerevetment
spectedthismaterial,questionstheuseofa flatarchitrave, werealso foundin thebaptistry.
48 THE BRITISH SCHOOL AT ROME

the westernface. It was presumablya bench forthe officiating clergy,and may in very
general terms be compared with the free-standinginternalapse of the Pre-Eufrasian
basilica at Parenzo;66but no close parallelis forthcoming.Behindthe dais, and outside
the chancel,the octagonalbases of an open, colonnadedscreenfollow the same gentle
curve to demarcatethe main apse. Two tombs, the remains of which it overrides,
belong to an earlier phase of the site's history.67From the centre of the western
side of the chancel a narrowpassage extended down the axis of the nave. It led
undoubtedlyto the ambon, the seatingsof which would perhapsbe revealedby careful
excavation.
The narthexoccupied the site of the east end of the Athanasianchurchand was
itself supersededin the ninth centuryby the sanctuaryof the Period V church. It
consisted,in effect,of a porticolinkingthe old and the new buildings. The westernside
was colonnadedand opened directlyon to the nave of the Athanasianchurch,while on
the east a greatcentraldoor, flankedby ornamentalrecesses,formedthe main west door
of the Theophilan church,with two smaller doors opening into the side-aisles. The
northernand southernends of the rectangularspace so enclosedwere shut offby two
apsidal,arcadedscreens;and a thirdscreen,of the same formbut slightlylarger,stood in
frontof the westernarcade immediatelyover the tomb-chamberof the saint. The
characterof these screens is hardlyin doubt, for at Deyr el-Abiad, Sohag, a similar
screenblocksone end of the najjhex. The latterexamplecarriesa brickhalf-dome,and
may be consideredas an elaborationof the shell-headed,apsidal recessesof the narthex
at Dendera.68 There can be little doubt that at Abu Mina, too, these screenswere
half-domedin brickover a marblecornice,in conformity with the survivingPeriodIV
brickworkin the cryptand in the baptistry;and thatabove thema facingof brickwas
carriedup to roof-height.It has alreadybeen noted that at the two points where the
foundationsof the narthexrun across the lines of the underlyingcrypt,the brick
vaultingof the latteris strengthened by the additionof stone relieving-arches,
proving
beyond doubt that the two structuresabove and below groundare contemporary.
To the west of the narthexlay the nave of the Athanasianchurch. Without excava-
tion, and withouta furtherknowledgeof the plan of the earlierbuilding, it is not
possible to say how far the work of Theophilus conformedwith, and incorporated,
existingstructures, and how far it involvedthe completerebuildingof the Athanasian
nave. The textof the Encomiumcertainlysuggeststhata substantialpartof the earlier
churchremained'to be made one with*the new building. On the otherhand it is quite
clear that in theirpresentformthe existingouter walls of the nave are of Period IV
workmanship,with a brick bonding-courseand ornamental recesses (Kaufmann,
Menasstadt, pl. 10). This would seem to implya moreradicalreconstruction, and theplan
is certainlysuggestiveof a conventionalatrium-narthexlay-out. Where so much is
uncertain,however,analogy is a dangerousargument;and pending excavationit is
perhapswiserto leave the matteropen.
To this period certainlybelongsthe semicircularscreenwhich marksthe site of the
•• B. Molajoli,La BasilicaEufrasiana
diPárenlo,Padova, nothingto do withthePeriodIV structure, as Kaufmann
I943> PP« I7~28; P- Verzone, Varchiutturareligiosa suggests,butprecedeit.
" U. Monneret
dellalto medioevo Italia settentrionale,
nett* Milan,1942,p. de Villard,Les Couvents
présde Sohag,
49. Cf.also S. Croce,Ravenna,Verzone,op.cit.pp. 9-10. Milan,1925,vol.I, fig.52(plan). AmlVCongr.Arch,Crist.
•'Kaufmann,Menasstadt,pl. 59. These tombshave p. 303,fig.9.
 THE SHRINE OF ST. MENAS IN THE MARYUT 49
69
confessioand of the tombbeneath;perhapsalso the remainsof a squarescreenof
marblepostsand slabs,doubtlesstheenclosure foran altar,in thecentreof thenave
immediately to thewest of the semicircular screen» it couldbelongequally
Superficially
well to PeriodV, but bothin character and in positionit belongsfarmorenaturally
to thefifth century thanto theninth.It mayalternatively be a relicoftheAthanasian
church.Here againexcavation woulddoubtlessdecide.
At the extremewest end of the whole complexstood the baptistry. The brick
bonding-courses, the use of ornamental recesses,and thetechnique of thebrick vaulting,
all betraytheworkofthePeriodIV architects; whileon thenorthsideit can be shown
to be ofonebuildwiththePeriodIV outerwallofwhathadbeentheAthanasian nave.
It was considerably modified at a laterdate,but the laterworkis structurally clearly
distinguishable, and nowhereare the mainlinesof the earlierbuildingin doubt.It
consisted ofa centralbaptistry chamber, flanked on thesouthbyan antechamber witha
single row of columns running down thecentre,70 on thewest by passage, on the
a and
northbya chamber of uncertain character, sincemuchmodified.The baptistry itselfis
octagonalinternally and was covered by a brick dome, of which substantial partswere
foundby the excavators, fallenintact.71 The crownof the dome was markedby a
crossworkedintothepatternof thebrickwork, whichcanneverhavebeen
a jeudfesprit
visiblein the completedwork.The alternate facesof theinternal octagoncontained
recessedniches,and thereweresimilarbut smallerexternalnichesfacingon to the
passageon thewestside.72In thecentre, beneatha canopy,wasa sunkencircular basin,
enteredfromtheeastandfromthewestbyflights ofsteps73andidenticalin designwith
thebasinin thebaptistry of the cemetery-church.74 Internally, the baptistry-chamber
waselaborately ornamented with gold mosaic on the dome and with whiteandcoloured
marble on the wallsand floor. Here, too, were found portions of threesmallcapitals,
witheaglesdisplayed at each corner flanking a chalice; and of three low-reliefpilaster-
capitalswithsimplefoliateornament.75
PeriodV. Patriarch ChurchA.D. <3j3-49'
Joseph's
AfterLazarus' workmenhad done theirworkand strippedthe sanctuary of its
and the
columns,its pavements, marble from its walls,thePatriarch Joseph faced
was
so completethat
witha destruction there could be no of
question rebuilding thegreat
PeriodIV churchofTheophilus.True, the Encomium speaksas if thiswere done;but
thesurvivingremainsshowthatin factit was abandoned, and that he had to be content
witha farmorehumblestructure,on thesiteof,andfollowing thesamegenerallinesas,
but slightly
thePeriodIII churchof Athanasius, reduced in scale. This is in themain
the buildingwhich Kaufmannidentified and published as the 'Grave-Church'
of Athanasius.The most superficial
(Gruftkirche) examination proves,however,that
whilethis'Grave-Church' some
incorporates earlier structures,
particularlyofPeriodIV,
69If indeedit be a confessio 71tfiV.pl.38.
proper.The presentsemi- " ibid,pi 16.
circularshaftis a structuralfeatureonly, and was in-
belowwasvaulted.There 78ibid.pl. 37.
visiblewhenthetomb-chamber 74ibid.fig.46.
are,however, tracesof a shaft
small down
leading intothe
fromthe base of the columnarexedra 76Capitals,see Kaufmann, Menasstadt% pl.66>nos. 2, 4,
tomb-chamber
above. For the significance screen, and 6; thebestpreserved
of the semi-circular of theseis in theGreco-Roman
see p. 59- Museumat Alexandria (Inv. No. i744*)« Pilaster-capitals,
70Kaufmann, Menasstadt,pl. 40. see p. 64, pl. ix. 4.
E
5o THE BRITISH SCHOOL AT ROME

Fig. 4.- The Ninth-Century Church

 THE SHRINE OF ST. MENAS IN THE MARYÜT 51
it is in factthe latestbuildingon the site and isftherefore, beyondanyreasonabledoubt,
the workof the PatriarchJosephbetween833 and 849. Both the plan and the character
of the buildingare in accordancewith this identification.
This latestchurchis built of verymixed,and mainlybad, masonry.The northand
south walls of the Sanctuaryare partitionsor rubble. The survivingstretchesof the
nave-wallsare also of rubbleconstruction, but morerobustand withan unevenplastered
surface.They incorporate severalmarblebases takenfromthePeriodIV church(PL VII,
3). These bases were alreadyin damagedconditionwhen used by the PeriodV builders,
a circumstancewhich no doubt explains why they were not taken by the rapacious
Lazarus. In curious contrastto the generallypoor quality of the masonryis the pro-
jectingeasternapse, a stout ashlar structure,betterthan anythingbuilt in Period IV
and equalled onlyby theremainsof the PeriodIII church. The relativedate of thisapse
is, however,beyondquestion. Not only does it block the greatwest door leading from
the narthexinto the centralnave of the PeriodIV church,but its footingscan be shown
to be trenchedinto the make-up of the Period IV floor,which in turn overlies the
survivingfoundationsof the centralPeriod III apse.
The plan of this latestchurchis extremelysimple. The Sanctuaryconsistsin effect
of the narthexof the Period IV church,slightlynarrowedby the insertionof two
two lateral exedrae. To
partitionwalls, which cut rightacrossthe foundationsof the
the west the intercolumnations of the narthex-arcade were partlyblocked to cut offthe
fillsthe
Sanctuaryfromthe bodyof the church;while to the east a singleprojectingapse
where was formerly the main west door of the earlierchurch. By a coincidence, of
gap
which the builders were certainlyunaware, the apse lies almost directlyover the
foundationsof the centralapse of the Period III Athanasianchurch.
A curiousfeatureof this laterapse is that it does not fillthe entiregap* On either
side, a narrowpassageleads into the bodyof the;old churchbeyond,and the wear of the
thresholdshows that these passages were much used. Presumablythey led to the
monasticquarters,whichstill lay to the northof the site of the PeriodIV church. The
two decorativeniches,whichhad flankedthe main doorwayin PeriodIV, now in effect
became apsidal recesseson eitherside of the main apse. By everyanalogythe Sanctuary
would have been divided longitudinallyinto threeby screens,probablyof wood, but of
these no trace now remains.
Of the body of the churchvery little can be said. Time and the excavatorhave
with a door shows that the
strippedit bare. On the southernside a stretchof walling
outer walls lay within the span of the outer Period IV walls and were built upon the
of
intermediatepair longitudinal foundations described above (p. 44). That the inner
carriedan arcade in Period V cannot be regardedas proved,but by analogy with
pair
contemporary Coptic Architecture it is highlyprobable.

D. ARCHITECTURAL FEATURES OF THE SUCCESSIVE CHURCHES

The purpose of the presentarticle is ratherthe presentationof certainfreshfacts

the detailed discussion of the problems of
regardingan importantmonumentthan
comparativearchitecturalhistoryto which it gives rise. Nevertheless,it may prove
52 THE BRITISH SCHOOL AT ROME
usefulin thefollowing sectionto call attention briefly to certaindistinctive elements,
ofwhichaccountmustbe takenin anymoredetailedstudy.
Transepts. The Athanasian churchand its successorare alike distinguished by the
additionofa transept totheordinary basilicalplan. There,however, thesimilaritybegins
andends. The transept oftheAthanasian churchis subdivided intoa seriesofcompart-
mentsbywallsorscreens running parallelto themajoraxisofthechurch.The roof-plan
musthave suggestedan organicunity,focusedat the crossing, whichis not in fact
reflected in the ground-plan. The Theophilanchurchis by comparison a thoroughly
and
homogeneous purposeful building, in which every accent converges upona single
the
point, high altar in the centre of the crossing of nave and The
transepts. distinction,
whichis oneofconsiderable for
importance comparative studies, is withoutdoubtbased
a
upon divergence of function.
The transept,whethercontainedwithinthe rectangular perimeter of the main
or
building projecting to form a T, but in either case distinguished by a roofsetat right
angles to the main axis of the building, had a wide, though scattered, currency in the
earlychurch. Its origin has been the subject of a considerable literature, in which the
issuehasoftenbeensadlyconfused by the to
attempt squeeze all the surviving examples
withina singlemould.76The mostrecentdiscussion ofa partofthefieldis a character-
isticallylucid articleby Krautheimer, based upon his researches in the churchof S.
Pietroin Vincoliin Rome.77In thishe showsthattheoriginalfabricof thepresent
churchof S. Pietro,built in the earlyfifthcentury, had a well-marked projecting
transept, which was clearly divided internally into a central bay and wings.In the
originalschemethis divisionwas to have been achievedby screen-walls withthree
openings in theirupperparts and some form of arcade below. This scheme was in fact
modifiedduringconstruction, but the distinctionbetweencentralbay and wings
remained.This 'tripartite transept'is, as Krautheimer shows,certainly derivedfrom
Greece,whereit is one of the commontypesof earlychurch.78 In some cases the
subdivisions aremarkedbyarcades,in othersbyprojecting pilasters, whichmayor may
nothavecarriedan archsimilarto thatbetweenthecentralcompartment ofthetransept
and the nave. In bothtypesalike the tripartite divisionis further emphasised by the
detailedarrangement of the internalfittings. Chancel-screens flankthe centralcom-
partment to forma kindof forechoir or continuation of the apse towardsthe nave,
whereasthewingsarein effect separate rooms, set apart alikefromthecentralbayand
fromthenave-aisles, withwhichtheycommunicate onlythrough well-marked doors,
oftenwithsecondary screens.Therecan be no doubtthatthisdistinctive formwas
evolvedin response to therequirements ofa developing churchritual;orthatin function
it broadlycorresponds to thepastophories whichflankthechancelin Syrianand North
Africanearlychurcharchitecture.
76This literatureis well
surveyed by J. P. Kirsch, *I1 near Athens. The churchat Dodona also may probablybe
transettonella basilica cristiana antica', Scrittiin onoredi considered as a later (sixth-century)derivative. The evi-
BartolommeoNogara* Rome, 1937, p. 205 ff. dence forthesebuildingsis summarisedby G. A. Soteriou,
77 R. Krautheimer,'S. Pietro in Vincoli and thetripartite **Al TraXonoxptonotviKal ftaaiXiKalTfft'EAAABos",Arch,
transeptin the Early ChristianBasilica', Proceedingsof the Eph.> 1929, pp. 161-248; see also 4Die altchristlichen
AmericanPhilosophicalSociety LXXXIV, 1941, 353-429. Basiliken Griechenlands*,Atti IV Congr. Arch. Crist.,
78 The churches in
question are churches A and B at pp. 355-80; and Krautheimer,op. cit. pp. 417-23. For
Nikopolis; and the churches at Epidauros; at Arkitsa- Corinth, see Hesperia^ xii, 1943, pp. 166-189.
Daphnousia, near Lokris; at Corinth; and by the Ilissos,
 THE SHRINE OF ST. MENAS IN THE MARYÜT 53
OutsideGreecea solitary instanceof thetripartite transept canbe seenin ChurchI
at Bosra,in Syria.79Otherwise S. Pietroin Vincolistandsalone,and is rightly regarded
as an intruder into the Romanseries,introduced fromGreece.None of the Greek
churches ofthisformis precisely dated;butseveralarecertainly ofthefifth century, and
thereseemsno reasonto doubttheconventional attribution of thewholegroupto the
samecentury, or in somecasespossiblyto theclosingyearsof thefourth.80 It may,or
may not,prove to be significant that the church at the
Epidauros, only one in whichthe
transept is further subdivided, as in the Athanasian church at Abu Mina, into five
compartments, is considered to be an early member of the series.
With the greatPeriodIV churchwe are on verydifferent ground.It has nothing
whateverto do eitherwith the Greektripartite transept with the tall, narrow,
or
continuoustranseptinterposedbetweenthe nave and the apse of the two great
Constantinian basilicasin Rome,St. Peter'sand St. JohnLateran.Whateverbe the
purpose of the latter,it is quiteclearthat,outsideRomein thefourth century, it was
singularly without influence on contemporary church architecture, and only came into '
itsownfourcenturies laterwhenit was adoptedas partof theself-conscious revivalof
classicaltypescharacteristic of the Carolingian renaissance.81
In markedcontrast to eitherof theabove-mentioned typesthePeriodIV churchis
distinguished by its architectural unity. Nave and transepts arenotjuxtaposed elements
butpartofan integrated wholeradiating froma commoncentre.This unityofplanis a
characteristic sharedwiththecruciform memoria, a well-attested earlytype,whichseemsto
havebeenassociated particularly with the martyr-cult.82 Guyermaybe rightinattribut-
ing to thecruciform memoria 2í
decisive initialinfluence in thedevelopment ofthetypeof
transept under discussion. In the present state of knowledge it is, however, prudentto
recallchatthereis stilla linkmissing in theevidence, foras yetnosymmetrically planned,
cruciform, basilicalchurchhas beenidentified fromthefourth or fifthcenturies.All-
Salona,thepre-Justinian churchof St. Johnat Ephesus,thechurchof the Martyrs and
at Gerasa, and H. Demetrios at Salónica ^ - to a or lesserextent
Apostles perhaps greater
emphasise themajoraxis,andto thisextentdeviatefromthepurecruciform plan. The
earliestdatedexample,theTheophilanchurchof St. Menas,lackstheeasternarmalto-
gether.It is opentoargument thattheinfluence ofthecruciform memoria wasrather that
of thelaterassimilation of an established type than the provision of theinitial impulse.
79H. C. Butler, inSyria,Princeton, 83Earlycruciform basilicalchurches:
EarlyChurches 1929,
p. n8f. Salona (490-140): R. Egger and W. Gerber, For-
80Soteriou,op. cit. passim; Krautheimer, op. cit. pp. schungen in Salona,I, Vienna,1917,p. 23 ff.;Dyggve,
AttiIV Congr.Arch.Crist.,pp. 391-414.
81R. Krautheimer, The CarohngianRevivalof Early Ephesus(notprecisely : J. Keil,
dated,butpre-Justinian)
ChristianArchitecture',ArtBulletin,XXIV, 1942,pp.1-38. 'Vorlaufiger Benefitiiberdie Ausgrabung in Ephesos',
82Most recentlyand fullystudied by A. Grabar, Jahreshefte d. Oesterreichischen Arch.Inst., Beiblatt
Martyrium, Paris, 1946 (available to the writeronly XXVII, 1932; Guyer,AttiIII Congr.Arch.Crist.,p.
when this articlewas in proof). See also S. Guyer, 438,fig.5.
'BeitragezurFragenachdemUrsprung des Kreuzformig- Gerasa, ch. of the Martyrsand Apostles(464-46$):
basilikalenKirchenbausdes Abendlandes',Zeitschrift J. W. Crowfoot,Gerasa,New Haven 1938,p. 256 ff.;
für Schwei^erisches Archáologie und Kunstgeschichtey andEarlyChurches inPalestine,London 1941,pp.85-8,
VII, 1945,pp.73-104.Thisworkis a preliminary studyfor fig.8.
a longerworkwhichthisauthorhasinpreparation. Guyer Salónica, H. DemetriosQ412-413): I follow Krau-
is followedby Krautheimer (*S. Pietroin Vincolietc.', theimer, *S. PietroinVincoli'etc.,pp.415-16andfoot-
pp.413-17),whostrangely quotestheTheophilanchurch note216.
of St. Menas as the best exampleof the basilicawith The groupis discussedby Guyer,AttiIII Congr.Arch.
centraltomband radiating transepts. Crist.,pp. 433 ff.;andby Crowfoot,Gerasa,pp. 190ff.
54 THE BRITISH SCHOOL AT ROME
It is well also to be cautiousin assuming,formallyor by implication,a uniformity of
liturgicalusage and thought in an age which was still very much feeling its way.
Of thetwoknownfourth-century examplesof thecruciform memoria,withfourradiating
arms of equal length,the one at Antioch-Kaoussié,dated 379-80, commemoratesa
local martyr,St. Babylas;84the other housed Jacob's well at Sichem.85 It should be
noted that whereas in the latter it was the object of venerationwhich occupied the
centreof the building,at Antiochthis position was reserved,not for the actual tomb
of the martyr,but forthe altarof the church.
We findthe same diversityin the big earlytranseptalbasilicas. In the pre-Justinian
churchof St. Johnat Ephesus the tomb of the saintstood withina centraltomb-chapel,
which opened to the north,south, and west upon threeradiatingbasilical naves; the
fourthside was closed by an apse, excludingthe wide five-aislednave which projected
eastwardsto completethe cruciform plan. In thegreatTheophilanchurchof St. Menas,
on theptherhand, it was the chanceland the highaltarwhichoccupiedthe centreof the
radiatingplan; the body of the Saint lay rightoutsidethe church,in a cryptto the west
of the main door, a strangepositionindeed if employmentof the cruciformor of the
transeptalplan be held to depend upon an architectural tradition,or liturgicalusage, in
which the object of venerationcoincided with the point of maximum architectural
emphasis.
At Abu Mina the purposeis surelyclear. This was a greatchurch,built,we are told,
because its predecessor'could not hold the multitudes,but theywere standingoutside
in the desert'(p. 33). The lay-outwas a simple and effectiveway of ensuring,forthe
maximumnumberof pilgrims,a close and uninterrupted participationin the ritesand
ceremonies of the central sanctuary.Precisely the same considerationsfaced the
architectof anychurchthatwas the object of a popularpilgrimagewitha specificmartyr-
cult. It is hardlysurprisingthatthe same solutionshould have been adopted elsewhere,
and forthe same reason.The churchat Ephesus,whichin plan mostnearlyapproximates
to theTheophilanchurchof St. Menas, was itselfthegoal of manypilgrims. Fiftyyears
lateran identicalexpedientwas to be adopted in anothergreatcentreof pilgrimage,the
churchof Kala'át Sim'án near Antioch,with its four basilical arms radiatingfroma
centraloctagon, in which stood the object of veneration,the column of St. Simeon
Stylites.88Here, as at Ephesus,the easternmostarm was separatedfromthe restof the
building,forthe use of the officiating clergy.
Locally in Egypt the transeptalplan had little or no influenceoutside the narrow
coastal belt of which the sanctuarywas the naturalcentre. At Abu Mina itselfit was
followedin miniatureby the buildersof the small detachedchapel to the east of the
shrine;87and its influencecan certainlybe detectedat Abu Sir (Taposiris), ten miles
away on the coast near Burg el-Arab, in the plan of the firstchurchto be built (c.
390-400 ?) on the site of the formertempleof Osiris.88 To it can probablybe traced
84J.Lassusand G. des
Downeyin Anúoch-on-the-Orontes, Krenckerand R. Naumann,Die Wallfahrtskirche
vol. II, Princeton,1938,pp. 5-48; AttiIV Congr.Arch, SimeonStylites inKala'át Sim'an (Abh.d. preuss.Akad.d.
Crist.,p. 340,fig.4.
88Knownfroma Wissenschaften,1938,PhiL-Hist.Klasse,41) Berlin,1939.
seventh-century drawingby Arculf, The centraloctagonalmostcertainly carrieda wooden
reproducedin Migne,Patrologíalatina,LXXXVIII, col. dome.
802; andAtti111Congr. Arch.Crist.,p. 439,fig.6. 87Kaufmann, Menasstadt.fig.46-7.
88Most and convincingly 88J. B. Ward Perkins,'The Monasteryof Taposiris
recently studied,on thebasis
of «excavationin the east churchand octagon,by O. Magna',B.S.R.A. Alex.,No. 36, 1945,pp. 3-8.
 THE SHRINE OF ST. MENAS IN THE MARYÜT 55
also the use of a transeptin the earlyfifth-century churchat Apolloniain the Cyrenaican
Pentapolis, a districtwhich was at this time stronglyunderthe influenceof Alexandrian
In
Christianity.89 upper Egypt, on the other hand, the transeptis neverfound.
Nave Ambulatories. In contrastto the incorporationof a transept,a clear link with
later Coptic architectureis affordedby the provision,albeit apparentlyin a rather
specialised form,of a thirdaisle at the end of each transeptin addition to the two
lateral aisles of the normal basilical plan. This is a typical featureof later Coptic
church-planning, found already in the mid-fifth-century monasteriesat Sohag and
frequently adopted later. At Abu Mina, too, it was used also in the cemeterychurch.90
The plan of theterminalaisle givenin PL XI is based on thatpublishedby Kaufmann
and on theremainssuperficially visible. It raisescertainstructural difficulties,
particularly
in relationto the upper galleries of the transept.These would naturallyhave been
expectedto conformto the ambulatoryplan of the groundfloor.Instead, if the plan
as given is correct,the centralsection of both ordersmust have been occupied by a
wide and loftyarch, reachingnearlyto roof-heightand giving in a more open form
somethingof the impressionof the 'triumphalarch' customaryat the junctionof nave
and transepts.Confirmationof this feature,as indeed of several other details of the
Theophilan church,would require carefulre-excavation.In the meantimea note of
caution is desirable.
The ambulatoryis not common in earlyChristianbasilical planning. Monneretde
Villard quotes severalSyrianexamples. It was, however,a formthoroughlyestablished
in civil architectureand so well adapted to the special needs of a pilgrimagechurch,
that it is hardlynecessaryto look fartherafieldin searchof parallels. It remainsin that
case problematicalwhetherits subsequentemploymentin Coptic architectureis due to
directborrowingfromÁbu Mina or whetherboth alike derive froma common local
traditionof civil architecture.
ThreeProjecting Apses.Projectingfromthe east wall of the fourth-century Athanasian
churchwere two, almost certainlythree,apses. The plan is markedlyindividual. But
evenin a morestereotyped formtheuse of threeprojectingapses is rareand sporadic,and
nowheredemonstrably early. Examples known to the writerof earlybasilical churches
with this featureare:
Greece
Considered
Hypate:G. A. Soteriou,Arch,Eph.t 1929, 186-7, and fig. 19. Known fromsketch-planonly.
by Soteriouto be post-Justinian.
has two external
Nikopolis:G. A. Soteriou, Atti IV Congr.Arch.Crist., p. 361, fig. 6. Central apse
buttresses.Consideredby Soteriouto be post-Justinian.

Roumania
Ibida: D. Ciurea, AttiIV Congr.Arch.Crist., p. 388 and pl. 11. Not dated.

Syria
KaVát Sim'an, the east church,c. 460-90. See p. 54, n. 86.
89P. Romanelli, Atti IV Congr. Arch. Crist., 1938, p. p. 315, lists all the Egyptian examples of this form and
274, fig. 23; J. B. Ward Perkins, Bulletin de la Société illustratesseveral, including the cemeterychurch at Abu
d%ArchéologieCopteIX, 1943, pp. 126-31. Mina (afterKaufmann, Menasstadt, fig.46).
90U. Monneretde Villard, Am IV Congr.Arch. Crist.,
56 THE BRITISH SCHOOL AT ROME
Palestine
Yarún(Galilee): C. R. Conder and H. H. Kitchener,TheSurveyof Western Palestine,1, London, 188 1,
p. 258. Broughtto my notice by ProfessorMonneret de Villard. See, however,S. W. Crowfoot,AttiIV
Congr.Arch.Crist.,p. 324, and Early Churchesin Palestine,London, 194 1, who states that the projecting
triapsidalplan does not occur in Palestine.
Transjordania
Pella (KhirbetFahil), two churches,both with threeprojectingapses. J. Richmond, Pal. Explor.Fund
QuarterlyStatement 1934, pp. 26-9 and pl. V; also A. M. Schneider,'Sudjudaische Kirchen', Zeitschr.L
Deutsch.Pal. VereinsLXI, 1938, 106, fig.20. Broughtto my notice by ProfessorMonneretde Villard.

This form does not appear to be found in the westernMediterranean,In southern

Palestineand Sinai the formwith threeinscribedapses is certainlyboth relativelyearly
and common; but only exceptionallyis the line of the outerwall broken,and then only
by the partial projectionof the centralapse.91
The sourcefromwhich the architectof St. Athanasius'churchderivedthis unusual
featureis happily beyond dispute. The one certainfact about the mud-brickchapel,
which earliermarkedthe graveof St. Menas, is that it possessedtwo projectingapses
flankinga destroyedcentralfeature.It is not unduly fancifulto conjecturethat the
remoterorigins,too, are to be soughtlocally.
It was not, however,until a verymuch later dater that the triapsidalbasilical plan
was establishedin Coptic Egypt,and theninvariablyinscribedwithina plain rectangular
exterior.The Athanasianchurchstood barelythirtyyearsbeforeit was supersededby
the Theophilan buildingwith its singleprojectingapse, and it appears to have left no
markon contemporary or later Egyptianpractice.
ButtressedApse. The singleprojectingapse of the Period IV church,while unique in
the earlyChristianarchitecture of Egypt,is in itselfa featurefamiliarenoughelsewhere
to call for little comment. It is, however,in this case furtherdistinguishedby the
addition of radiating external buttresses.These buttressesare characteristicof the
fifth-century architectureof the EasternAdriaticcoast.92 Elsewheretheyare unusual at
this early date. An example in Asia Minor is the church of St. Thekla, r. 470, at
Meriamlik,Cilicia.93
SphericalTriangularPendentives. The originof the pendentive,as a means of adapting
a dome to a square space, is a subject on which scholarshave lavishedmuch attention,
but haveoftencontrivedto generatemoreheatthanlight. This resultmustbe attributed
in no small degreeto the misleadingantithesisinherentin the question 'Rome or the
Orient?\ whichhas dominatedso much enquiryin this fieldin the last fewdecades,and
has givenrise to a greatdeal of incautiousgeneralisation.Much of thismighthave been
avoided by a morescrupulousregardforthearchitectural grammarof the subject,and by
91Sincethisarticlewas inS alona,Vol. Ill, Vienna,1939;Atti
preparedthewriterlearnsthat Egger,Forschungen
recentworkin S. Pietroin Vincoli,Rome,hasshownthat IV Congr.Arch.Crist.,p. 402, fig.15. The apse of the
the two smallflanking apses are contemporary withthe is wrongly
Basilicaat Sabrathain Tripolitania shownas of
transeptand are the work of SixtusIII (432-40). Riv. thisformby Bartoccini, Guidadi Sabratha,1927,repro-
Arch.Crist.XXI, 1944-5,P« 324. ducedby Romanelli,AttiIV Congr.Arch.Crist.,p. 247,
•a
Nikopolis,inEpirus,BasilicasA (fifthcentury) andB fig.1. . . ... ....
(secondhalfoffifth Soteriou,Arch.Eph., 1929, 98E. Herzfeldand S. Guyer,Mertamhk undKorykos
century),
figs.33and37; Paramythia, inEpirus,ibid.fig.36; Philippi, (Monumenta AsiaeMinorisAntiqua,Vol. II, Manchester,
secondBasilica,Bull. Corr.Hell. LXI, 1937,463 ff.;Atti 1930),p. 14, figs.8 and 14. The authorsquote a second
IV Congr.Arch.Crist. , p. 369, fig.15; Marusinac,main exampleat Miletus.
church(mid first-half fifth century),E. Dyggve and R.
 THE SHRINE OF ST. MENAS IN THE MARYÚT 57
a recognitionof the factthat,save in a fewthoroughly exploredareasof the field(such as
Italy), the available materialis not yet to
sufficient warrant a comprehensive judgement.
The immediateneed is fora largerbody of agreedand carefullystudiedevidence.94
At Abu Mina, the sub-structures of the PeriodIV church,c. 400, containone certain
instance of a square chamber with a brick dome springingfrom fully developed,
spherical-triangular pendentives (PL VI, 3). This somewhat cumbrous phrase is
necessary if we are to distinguishbetweenthose more elementarypendentives,which
follow the curveof the simple sphericalor domical vault withoutinterruption and so
constitutea uniformdownwardprolongationof the curvedsurfaceinto the anglesof the
squarespace beneath,and thosewhich,while themselvessegmentsof a sphericalsurface,
followa curveat a sharpangle to thatpf the dome above, and thus retaintheiridentity
as separatetriangular elementsdistinctfromtheactualdome.95The simplerformbelongs
to the continuousvault(yoltaa vela,Hdngekuppel), which ist relativeto the truedome, an
elementary form of construction that had a wide currencyin the classical Roman world
as a means of vaultingsquare rooms of modest dimensions.The spherical-triangular
pendentive,on the other,hand, belongs to a group of devices of a more ambitious
character,wherebya true dome, of greaterheightand span than the sphericalvault,
could be accommodatedto a squareplan. Failure to observethe distinctionbetweenthe
two formsoftenresultsin obscuringthe elementary fact thatthe formerwas in general
use throughoutthe Mediterraneanworld long beforethe earliestauthenticatedappear-
ance of the developedspherical-triangularpendentive;and a good deal of the evidence
oftenadduced forthe originsof the latteris in consequenceof verylittledirectrelevance
to the questionat issue.
It may well be that the vault over the actual tomb-chamberwas of the simpler,
continuoussphericalform,whichwould have been betteradapted to coverthe relatively
largerareawithoutprojectingthroughthe pavementabove. At Karánis,in the Faiyum,
vaultsof this formwere foundin the second of threedistinctbuildinglevels,of which
the third,and latest, had been finallyabandoned about 450.96 The shallow pitch of
some of these revealsa level of technicalcompetencethat is the more notable in that
Karániswas a modestprovincialtownwith fewarchitectural pretensions.The vaults in
the cemeteryof el-Bagawát(see above, p. 40) are of the same general construction,
thoughrathermorelofty.There can, in fact,be no doubt thatthe technicalbackground
of the brickvaultingof the PeriodIV cryptand baptistryat Abu Mina is to be soughtin
the indigenouscraftsmanship of Roman Egypt. What is new to this tradition,at least
in so faras its remainshave so farbeen recorded,is the appearance,in the small chapel
of the crypt,of two elements,the developed spherical-triangular pendentiveand the
squinch (see followingsection), both of which belong properly a more ambitious
to
architecture.In the contexttheycan most naturallybe explained as elementsalready
currentin the architectureof Alexandria;and their use in a building that can be so
securelydated to the openingyearsof the fifthcenturyis a factof no littlesignificance.
94See J. B. Ward Perkins,TheItalianElementin Late »6A. E. R. Boak and E. E. Peterson,Karánis1924-28
RomanandEarlyMedievalArchitecture (BritishAcademy (Universityof MichiganStudies,Humanistic Series,No.
AnnualItalianLecture,1947),London,1949,pp. 17-8. XXV, 1929),pl.XIV, fig.28; cf.pp.23-4,pls.XIII, XVIII.
95In rejectingthisusefuland generallyaccepteddis- For thechronology see also D. B. Harden,Roman Glass
tinction,Creswell{Early MuslimArchitecture^ II, 1940, fromKaránis(Same series,No. XLI, 1936), pp. 24-34.
p. 322) adds a quiteunnecessaryelementofconfusionto The terminaldateis notindispute.
the controversy.
58 THE BRITISH SCHOOL AT ROME
The squínch(Pennacchio
Squinches. or a mezgo
a tromba padiglione, can takea
Ecktrotnpe)
number of forms,but all consist in essence of some form of small structuralarch
springingfromthe upper part of two adjacent walls and spanningthe angle between
them,to bridgethe transitionfroma square or polygonalspace to a dome. The angle
within this small corner-archis often itself half-domed.In its simplest form,it is
structurally less satisfactory than the sphericalpendentive,forit softens,b.utdoes not
wholly resolve, the awkward junctionbetweencircleand square or polygon. But it was
capable of considerable and ingenious development;and it had a long and varied
currency in the later architecture of both East and West.
If the Palace of Firüzabád were rightlydated to the thirdcentury,the case for a
Sassanian originwould be strong.97 What is in any case beyonddispute is that it was
already in use, both in Italy and in the EasternMediterranean,. in the fifthcentury.The
fourwell-developedsquinchesin thebaptistry of S. Giovannia Fonteat Naples, thework
of Bishop Soter(465-86), have hithertoenjoyedpride of place as the earliestcertainly
attestedexamples in the Roman world.98Those which supportthe half-domein the
chapelof thecryptat Abu Mina are halfa centuryearlier. In themselvestheyare modest,
but theyareenoughto showthat,like thependentive,the squinchwas alreadyfamiliarin
Alexandriaat the beginningof the fifthcentury.
NarthexwithApsidalExedrae.A distinctivefeatureof the Period IV complexis the
narthexwithitslateralpair of columnarexedrae. At firstsighttheseexedraecall to mind
the columnarexedraewhich were to play so importanta part in the developed sixth-
centuryarchitecture of NorthItaly and of Constantinople.Their employment, however,
is entirelydifferent. Whereas in the serieswhich leads up to Haghia Sophia and to San
Vítale, the exedrae performan essential architecturalfunction,here they have no <
structuralsignificance whatever.It is temptingto regardthem as a derivationfroman
establishedfunctionaltype;but it mustbe admittedthatin the disputedquestionof the
originsof the lattertheirevidentialvalue is slight.
In moregeneraltermsthe narthex,withits two opposedapsidal exedrae,should per-
haps be relatedto a groupof Italian monuments,at the head of which in the Christian
seriesstandstheMausoleum of Santa Costanza at Rome.99 Othersurvivingexamplesare
the baptistrybuilt by SixtusIII, 423-40, in replacementof the Constantinianbaptistry
ofSt. JohnLateran;100 S. Ippolito, Milan, thirdquarterof the fifthcentury;101San Vitale
97The squinchesof Firüzabádare well alia
developedand 1 piazza
del Collegio Romano',Acta Instituti Romani
betraya competent tradition
(accessiblyillustrated
byG. T. RegniSueciae,XII, 1946,138). Earlierthanthisbaptistry
Rivoira,Le OriginidellaArchitettura Lombarda,Rome, or the chapel at Abu Mina is the small Caldarium,
1901, fig. 124, and Lombardic Architecture) 2nd edition, addedc. a.d. 300 to theHuntingBathsat LepcisMagna
Oxford,1933,vol. I, fig.290). Theirdate,however,is far (Archaeologia XCII, 1949); and to the fourthcentury
fromcertain.The palace is oftenassignedto the third belong the well-developedsquinchesin the towersof
century a.d. (e.g. by O. M. Dalton, East Christian Art, the AurelianWalls,notedby Dr. L. Cozza (Lectureat
Oxford,1925,p. 78, n. 1, and p. 82, n. 1; and by K. A. theMuseo di Roma,8 April,1948).
Creswell,Early Muslim Architecture,
II, 1940,pp. 101-5); •• wellsurveyed Roemische Mitteil-
Recently byStettler,
but the evidenceis farfromconclusive.It maywell be ungenLVIII, 1943,pp. 76-86; forthemostrecentbiblio-
two centuries later.The palace of Sarvistanis certainly graphy see cit. n.
notearlierthanthe fifthcentury.For bibliographyon 100O. B.Sjoqvist, op. p. 144, 3.
Giovenale,//BatusteroLateranense (Studidi
thesquinch,see Creswell,op. cit.tp. 101,n. 1. Antichitápubblicatiper cura del PontificioIstitutodi
•• E. Bertaux, L *artdansVItalieméridionale, Archeologia No. L io2o\
Cristiana,
Paris,1901,
p. 40; G. T. Rivoira,RomanArchitecture, 101p^
Oxford,1925, Verzone,IS architetturareligiosadeWaltomedio-
fig.311. It maynotbe withoutsignificance in thiscontext evo neW Italia settentrionale,
Milan,1942,pp. 79-99. GS
thattherectangular formof thisbaptistry, familiarin the Chierid,in PalladloII, 1938,3,illustratesa similarfeature
EasternMediterranean and NorthAfrica, is uniquein the inthenarthex ofS. LorenzoatMilan,butthiswouldappear
EarlyChristian seriesofItaly(E. Sjoqvist,*Studiintorno to be a paper-restorationunwarranted bytheknownfacts.
 THE SHRINE OF ST. MENAS IN THE MARYÚT 59-
at Ravenna,begun between 520 and 530 and consecratedin 547; and the churchof S,
GiovanniBattistaat Canosa.102A porchwithtwoopposedexedraewas added in the fifth
centuryto theso-calledTemple of MinervaMedica in Rome;103and thesame featurecan
be seen in the plans, made by an anonymousFlorentineartistbefore1 514, of the two
imperialmausolea, later the churchesof S. Andrea and of S. Petronilla,which were
destroyedalongsideold St. Peter's in the earlysixteenthcentury.104
Outside Italy the onlyexamplesknownto thewriterareto be foundin Asia Minor105
and in the churchesof Coptic Egypt,whichpresumably,in thisrespectat anyrate,stem
directlyfromAbu Mina. The derivationis particularlyclear in Deyr el-Abiad, Sohag,
c. 440,106which repeatstheotherwiseunique, columnarformof the Abu Mina exedrae.
At Dendera107theexedraeare, moreconventionally, apsidal recesses,so inscribedwithin
the thicknessof the wall as not to break the characteristically plain, rectangularouter
At
perimeter. Deyr Abu Hennis and at Medamüd thereis one such exedraonly;and the
churchof Abu Sargehin Old Cairo mayrepresent a tardysurvivalof thesame tradition*108
The semi-circularscreenoverthe tomb.Whetheror not the nave of the Athanasian
church was an open courtyardin Period IV, the semi-circularscreen over the tomb
seems to have been a free-standingfeature;and there can be little doubt that, like
the exhedraeof the narthex,it carrieda light half-domeof brick. The significanceof
the free-standing apse as an earlyformof martyrionhas recentlybeen fullydiscussed
by A. Grabar (Martyrium, Paris, 1946).
Octagonal So
Baptistry. manyconflicting claims have been made about the originsof
the architecturalformrepresentedby the Period IV baptistryat Abu Mina, that it is
usefulto call attentionto the factthat,alreadyby the close of the fourthcentury,it was
firmly establishedin thecivilarchitecture bothof theeasternand of thewesternMediter-
ranean. In Italy we findit as earlyas the close of the firstcenturyin the Flavian Domus
Augustanaon the Palatine;109and withveryslightmodification it persisted,notablyas an
elementof bath architecture, until it was adopted by the ChristianChurch of North
Italyas a standardbaptistry plan.no In the Near East we findit alreadyestablishedin the
101R. Bartoccini,Felix Ravenna,n.s. Ill, 1934, 158, 108U. Monneret de Villard,Les Couventsprés de Sohag,
fig.1. Milan,1925-6,vol. I, figs.3 and 18; and AttiIV Congr.
103G. Caraffa,
La cupoladellasala decágonadegliHorn Arch.Crist.,p. 300,fig.7.
Uciniani: restauri1942, Rome, 1944. 'Die (as yet un- 107U. Monneret de Villard,op.cit.pp.47-8,fig.52; and
published)survey,undertaken shortlybeforethewarby Atti IV Congr.Arch.Crist. , p. 303,fig.9.
Dr. F. W. Deichmann,has shownthattheoriginalcon- 108 Deyr Abu Hennis: Somers Clarke,Christian An-
structionofthisbuildingdatesfromtheseconddecadeof tiquities intheNile Valley, Oxford,1912, pp. 181-7,fig.4.
thefourth century;and thattheporchbelongsto a third AttiIV Congr. Arch.Crist.,p. 297,fig.4. Medamüd F. : M.
structuralperiod. Bissonde la Roque, Rapportsurlesfouilles de Médamoud
104H. Koethe, RoemtscheMttteilungen XL VI, 1931, (1931-1932),Cairo,1933,pp. 13-20. AttiIV Congr. Arch.
9-26, and plan,abb. 2; reproduced also by C. Cecchélli, Crist.,p. 312,fig.19. Abu Sargeh:AttiIV Congr.Arch.
Atti III Convegno na\ionaledi Storia dell* Architettura,Crist.,p.313,fig-"•
109G. Lugli,Attidel III Convegno Naponaledi Stona
Rome, 1940,p. 149.
105Gulbagdsche, nearSmyrna:Karl Michel,'Die alt- deWarchitettura, Rome, 1940,p. 97; sketch-plan in G. T.
christliche
Kirchenanlage vonGülboghdsche', Forschungen Rivoira, Romana,1921,fig.120.
Architettura
Kunst{Festschrift 110The Italiandevelopment
und^iirchristlichen
iurKirchengeschichte hasbeenwellstudiedby G.
Ficker),Leipzig,1931,pp. 180-200,probably secondhalf De Angelisd'Ossat,whosework,withearlier bibliography,
*
of the fifthcentury;see also G. Weber, Basilikaund is summarised in two articles:*Sugliedificiottagonalia
in
Baptisterium Gülboghdsche', By^antinische Zeitschrift, cupola nell'antichitá
e nel medio evo', AttidelI Convegno
1901,pp. 568-73;andJ.Strzygowski, Kleinasien einNeu- Naiionale di storia dell* architettura, Rome, 1938, pp.
landderKunstgeschichte,Leipzig,1903,p. 49, fig.35 (the 14-24; and Romanitádelle CupolePaleocristiane, Rome
planwrongly a transept).Aphrodisias,
indicates churchin (Istitutodi StudiRomani),1946,p. 13,fig.3. See also P.
thetempleof Aphrodite:L. Crema,Monumenti Antichi Verzone,L*Architettura religiosadelValtomedioevo nell*-
XXXVIII, 1939,179,% 34. Italia Settentrionale,
Milan,1942.
6o THE BRITISH SCHOOL AT ROME
secondcentury, in a bath-building(BathC) at Antioch;111 and it had no inconsiderable
influence on theNearEasternChristian architectureofthefourth andlatercenturies.112
The baptistry at Kal'át Sim'án,113 c. 460-90, with its flanking hallsand entrance
corridor,affords an unusually closeand interestingparallel to thatof Abu Mina.
TheNinth-century Church.The churchbuiltby thePatriarch Josephdiffers markedly
fromthe roughlycontemporary architecture of theWadi'n-Natrün.The sack of the
lattersite,c. 817, hadsweptawayalmostentirely theearlierbuildings, andthesurviving
remainsof theninth-century reconstruction reveala standardisation ofpractice approxi-
matingcloselyto laterCopticusage.114 Indeed,theonlyimportant difference is thatthe
naveis notyetvaulted,butretainsthetraditional timberroof.Thereis alreadya rigid
divisionbetweennaveand chanceland,withinthechancel,betweenthethreehaikals or
chapels. In the ninth centurythe central was
haikal normally square and covered with a
domeseton sleeperbeamslaid acrosstheangles;andit wasflanked by two tall,narrow,
barrel-vaulted lateralhaikals»
Suchan arrangement presupposesinternal wallssufficiently solidto carrythedome
overthecentralhaikal, andofthesethereis no traceat Abu Mina. The retention ofthe
twoPeriodIV nichesflanking thecentral apse indicates that the chancel was probably in
some way subdivided;but the divisioncan hardlyhave been moresubstantial than
woodenpartitions, and thewholechancelwas probably roofedas a unit. The apsesare
another unusualfeature.The ordinary Coptic haikalis square-ended. In otherwords,the
ninth-century church at Abu Mina reveals less
considerably progress thanits contem-
poraryneighbours in the Wadi'n-Natrün towards the idiosyncrasies of laterCoptic
church architecture; andtherecanbe littledoubtthat,in this,it reflects itsclosercontact
withAlexandria.

E. NOTES ON THE SCULPTURAL ORNAMENT AND ON THE

RELATION OF ABU MÍNA TO ALEXANDRIA AND TO COPTIC
EGYPT
The foregoing notescoversomeonlyofthemorestriking ofthearchitecturalfeatures
revealedat Abu Mina. Theyarein no senseexhaustive; andlittleorno mention is made
ofotherfeatures, whichareequallyimportant fortheappraisaloftheshrine'ssignificance
in relationto contemporary architectural on which,however,any judgment
currents,
would at presentbe premature. The developedplan of the PeriodIV Theophilan
for
complex, example, forall that(likethecathedralcomplexat Gerasa115)itowessome-
thingin detailto thearchitectural of
accidents theshrine's
growth,remains an important
document forthestudyofthedevelopment oftheprinciples bythe
oflay-outestablished
111Antioch-on-the-Orontes,
I, excavations
0/1932, Leipzig,1903,pp. 74-90; forlaterbibli-
Prince- Kunstgeschichte,
ton, 1934: C. S. Fisher,'Bath C, pp. 19-31,planpl. V. ographysee S. Guyer,Byiantinische XXXIII,
Zeitschrift
C. R. Morey,The MosaicsofAntioch, London and New 89-90). Survivingexamplesfromtheearlysixthcentury
York,1938,planp. 12. The bath was built the
after earth- are the well-known churches at Esra and Bosra,and the
quakeof 115,and rebuilt, on thesamelines,in thefourth churchofSt. TohntheBaDtistat Gerasa.
113De
century. Vogüé, SyrieCéntrale, pl. 149; reproducedby
118The churchwhich
GregoryofNyssa,c. 335-95,built Cabrol, Dictionnaire*II, fig.1161.
u* White,III, p. 15 ff.
at Nyssa,was certainly of thisgeneralform,somewhat
elaborated(Gregoryof Nyssa, Letterto Amphilochius 116Gerasa, Cityof theDecapolis,ed. Kraeling,New
(Migne,PatrologíaGraeca,XL VI, 1093-1100); see recon- Haven,1938.
structionin J. Strzygowski, Kleinasienein Neulanddes
 THE SHRINE OF ST. MENAS IN THE MARYÜT 61

greatcommemorative foundations of Constantine, such as the churchof the Holy

in
Sepulchre Jerusalem; and it may be relevant also to thehistory ofthedevelopment of
thecanonicalatrium-narthex-basilica complex. But detailed discussion of either point
canbe oflittlevalueuntilit is clearlyestablished byexcavation whether in PeriodIV the
naveof theAthanasian churchwas,in fact,replacedby an atrium,or whether instead
somepartoftheearlierstructure wasretained in use. Excavation is equallyrequiredfor
a fullerknowledge oftheprimitive shrineandoftheAthanasian church, andto establish
in detailthe character of the internalfittings not onlyof the latterbut also of its
Theophilan successor.
Thereare,moreover, severalaspectsofthesite,whichareopento immediate enquiry
withoutfurther excavation, on which,however, thewriter had neither thetimenorthe
facilitiesadequately to supplement theexiguousand misleading accountspublishedby
Kaufmann.No discussionof the architecture of the Sanctuary can, forexample,be
complete without a comparative examination of the other important excavated buildings,
of the church 116and of the with its basilical
notably cemetery bath-building two-apsed
hall.*17
Another aspectof theremains, whichmustawaitmorefavourable conditions fora
detailedstudy,is thearchitectural ornament. The bestofthiswasremoved by Kaufmann
to Frankfurt-am-Main, wherecircumstances havenotpermitted its inspection; norwas
thewriter ina positiontosecurethenecessary documentation of thefragments surviving
on thesiteitself.It may,nevertheless, be usefulto pointoutin thisconcluding section
certain general features,which have an important bearing on the relation between theart
of Christian Alexandria andthatof CopticEgypt.
A masterly statement, notonlyofthewiderproblems presented byCopticsculpture
butalso of the chronology andcontent ofthesculpture itself,is to be foundin an article
tenyearsago byErnstKitzinger.118 He startsfroma re-examination of the
published
sculpture from Annas, the classical
Heracleopolis.119 This forms a continuous
important
series,ranging fromtheearlier,'soft'group,in whichtheclassicalelementis stillrela-
strong, tothemorerigidstylisation ofthelater'hard'group.On theevidence ofthe
tively
associated capitals,the central dates forthese two groups can be shown to be r. 400 and
themiddle of thefifthcentury, respectively. This Annas is in
sculpture only partCoptic,
in thestrictsenseof theterm,formuchof it,probably themajority, is paganwork.120
But it is the immediate and
iconographic stylistic forerunner of the developedCoptic
sculpture of Bawitm and of Saqqara,122 whichcan be securelydated,againon the
evidence oftheassociated capitals,to themid-sixth century.
The source, from which muchoftheAnnassculpture was itselfdeveloped, hasbeen
the excavation of a of
group comparable, but earlier, sculpture at
putbeyondquestionby
116Kaufmann, 120For thelatesurvivalof paganismin Egyptsee Mon-
Menasstadt,planfig.46 (reproducedby
s.v. Menas,fig.7964).
Cabrol,Dictionnaire, neretde Villard,op,ciup. 52,n. 1,citingtheearlierbiblio-
117Kaufmann, Menasstadt,planfig.49 (reproduced by graphy.See also Cambridge MedievalHistory,vol. I,
s.v. Menas,fig.7955)- D- Krencker 1911, P- 112 if.
Cabrol,Dictionnaire,
and E. Kriiger,Die TriererKaiserthermen, 121E. Chassinat,Fouillesd Baouit(Mémoires deTlnstitut
Augsburg,
1929, p. 235. franc.aisd'Archéologieoriéntaledu Caire, vol. 13,
118É. Kitzinger,'Notes on Early Coptic Sculpture , 1911).
122J.E. Quibell,Excavations at Saqqara,vol. Ill, 1909,
Archaeologia LXXXVII, 1938,pp. 181-215.
119Previouslystudiedby U. Monneret de Villard,La andvol. IV, 1912.
SculturaadAhnds,MUan,1923.
6z THE BRITISH SCHOOL AT ROME

Oxyrhyncus.128 An importantexceptionmustbe made of the figuredmythologicalcom-

positions which are the most strikingand individualfeaturesof the Ahnás sculpture.
Whetheror not Kitzingeris rightin regardingtheseas a local Egyptiandevelopment,124
it is clear that theyhad only a briefcurrency.They are foundin rudimentary format
Oxyrhyncus; theyand were dropped from the sixth-century Coptic repertory. The other
characteristic features,however,of earlyCoptic sculpture,capitals,niches,and gables,
almost without exceptiondisplay a continuousand coherentdevelopmentfrom the
sculptureof Oxyrhyncus,throughthat of Ahnás, to that of Saqqara and of Bawit. The
Oxyrhyncussculptureis not itselfdated; but, as a group and allowing for individual
exceptions,it can hardlybe earlierthan the beginningof the fourthcentury;and the
Ahnásgroupprovidesa terminus antequernat the end of thesame century.125
While the termCoptic can be properlyapplied onlyto the laterChristianmembersof
theseries,thecontinuity withintheseriesis suchthatit is to thewhollyorpredominantly
pagan sculpture Oxyrhyncusand of Ahnás thatwe are bound to look forcomparison
of
with the survivingfourth-and fifth-century sculptureof Abu Mina. The overwhelming
conclusionfromsuch a comparisonis thatneitherin stylenorin contentis theretheleast
similarity.The two seriesbelongto different culturaland stylisticworlds.
The largestsurvivingbody of comparativematerialis affordedby the capitals. A
numberof theseare illustratedby Kaufmann;126 and Kautzsch,in his valuableanalysisof
the late-classicalcapital,describesand in partillustratessome of the Corinthiancapitals
fromAbu Mina in the collectionat Frankfurt.127 to
Of the latterit is here sufficient
remark that they belong without exception to the generaltraditionof late-Roman
craftsmanship in the easternMediterraneanand are ascribedby Kautzsch to the fourth
and fifthcenturies.Of the characteristically Egyptian late-Roman and early Coptic
Corinthiancapital thereis no trace. With its upper row of acanthusleaves so widely
spaced as to exposethecaulicoliand sheath-leaves, the junctionsof whichare furtherem-
phasised by conspicuousround knobs, the latter is a strikingand easily recognisable
form:128 it isfmoreover,peculiarto Egypt,so thatits absenceis all themorenoteworthy.
A furtherpoint of markeddivergencelies in the treatmentof the detail. At Abu Mina
thereis no traceof thatsimplification and of thathardeningintoa sharpblack-and-white
pattern,whichcan be feltas earlyas the fourthcenturyat Oxyrhyncus and was increas-
ingly to become the salient characteristicof the succeeding Coptic sculpture.The
relaxationof classical formsis here a process of dissolution,unaccompaniedby any
positivestylisticdevelopment.
The pointsof contactbetweenthetwoserieslie, significantly enough,in thosecapitals
which are demonstrablyintrusiveto the local Egyptianseries. Such, for example,is a
fineCorinthiancapitalof Proconnesusmarble,now in the Coptic Museum at Cairo,with
mEv. Breccia,Municipalité<TAlexandria,le Musée 82,83,100,108,109,112,122,128,129,133,138,152,153;
gréco-romain t$25-3t,p. 60 ff;and 1931-2,p. 36 ff.For pls. 6-9. These examplesembracehis types2, 3, 5-7,
scatteredsculpturefromOxyrhyncus see Kitzinger,op, 9-13, 16.
cit.d. 200.n. 1. 128Kitzinger, and
pp. 187-8,pl. lxviii,1, 2, 4; recently
x*4For a see morefullyillustrated
contrary opinion Monneret de Villard,op, by £. Drioton,Les sculptures copies
cit,passim,who stresses theparallelswiththeartofSyria, du Nilomitrede Rodah, Cairo, 1942. Drioton accepts
Mesopotamia, andIndia. Kitzinger'smainconclusionand chronology.See also J.
186Kitzinger, p. 200. Kunst{Cataloguegeneraldesanti-
li§ Menasstadt. Strzygowski,Koptische
pls. 66-71. quitéségyptiennesdu Musée du Caire,vol. XII), Vienna,
187Kautzsch, (Studienzür spátantiken
Kapitellstudien 1904,no. 7348,fig.101.
Kunstgeschichte, No. 9), Berlin,1936,nos. 50,60, 73,79,
 THE SHRINE OF ST. MENAS IN THE MARYÚT 63
an eagle in the centreof the abacus.129It is one of six formerly at Ahnás,and it is un-
a
questionably stranger to the local series. At Abu Mina, on the otherhand, similar
capitals fall naturally into place within the main seriesof capitals, manyof which are
themselvesof island marble,while all reflectthe same generalstylisticbackground.130
The onlyotherformof capital which KautzschrecordsfromAbu Mina is thathere
illustratedin PL IX. 3,fromthesiteof theCoptic Patriarchate in Alexandria,131 a familiar
easternMediterraneantypewith a pedigreethatgoes back to theTower of the Winds at
Athens,the Traianeumat Pergamon,and the Severan Forum at Lepcis Magna, to cite
threefamiliarexamplesonly. Of thislate flutedformthereare a numberof examplesin
Cyrenaicaand in the museumat Alexandria.It is not a commonCoptic type,althougha
paintedversionis represented at Saqqara.132
Two othergroupsofcapitals,foundat Abu Mina in considerablenumbersbut omitted
by Kautzsch,deservebriefmention. The firstof thesecomprises,in varyingdegreesof
debasement,Ionic and compositeIonic capitals, in the latterof which the Corinthian
elementis alwaysexiguousand maybe reducedto fourleafletsat the anglesbeneaththe
volutes.133 Most are of modestdimensions.There is no example of the related Ionic
impost-capital,a Greek type of which two examples,fromAlexandriaitself,are pre-
servedin the Alexandriamuseum.134The secondgroupcomprisesa wide varietyof small
fromsmall but fully
capitals fromscreens,windows,and ornamentalniches,ranging
articulatedCorinthianto diminutivedrumswithfourshallowleavesscoredat the angles.
A numberof theseincorporatethe wreathed-cross motif(see below pp. 66-8), and in at
least one instancethishas displacedall otherornament.135
With the exceptionof thecapitals,whichare at least sufficiently numerousto be con-
sideredas representative, all too little of the architecturalornamentof Abu Mina has
survived.The fragmentsof gold mosaic associated with the Period IV buildings of
has gone. The surviving
ArchbishopTheophilus are a tantalisingreminderof what
column-bases, some of them securely dated to Period IV, are mainlyof the formillus-
trated in PL VII, 1 with the Attic base and the moulded rectangular pedestalcarvedin a
block. This form originated in the eastern Mediterranean, where it is common
single
from the second century onwards. It was later only,and then sporadically, it foundits
that
into the west.136 At Lepcis Magna it is used, with base and plinthcarvedseparately,
way
in thegreatSeveranForum;and,carvedin a singleblock,in a fourth-century porticobeside
the Theatre. In Rome it is hardly found before the fourth century. In the museumat
Alexandria there are a number of scattered examples; and in Coptic Egypt is not un-
it
common,thoughusually in a somewhat evolved form.137 Pedestals of the same general
m Strzygowski, op.cit.no. 7350,fig.102; cf.no. 7351, tratesone in theAlexandria Museum,no. 737,pl. 44. See
fromAschmunein. also Kaufmann, pl. 69. 6.
180Kaufmann, 184Kautzsch,op.at. no. 536,pl. 33.
Menasstadt, pl.66.1and68. 8; Kautzsch, 186Kaufmann,
no. 79; cf.Kaufmann, Menasstadty pl. 68. 5, and 70. 1 Menasstadtypl. 69. 2; cf.pl. 69. 1 and 5;
and s- 136See J. B. Ward Ferkins,JJCS. AAAV111,1948,
131Kautzsch, no.753,pl.44; seealso Kautmann, Menas-
stadt,pl. 69. 6. D. 7O.
187AtBawit,forexample,thecolumn-bases areraisedas
182J. E. Quibell,Excavationsat Saqqara 111,1909,pl.
xxii.4-6; IV, 1912,pl. xl, 1-2; pl. xliii.3; Duthuit,Van a
a regularpractice.Quibell,op. cit.pl. XVIII, illustrates
witha pedestalreminiscent oftheAbu Minaform.
copteypl. xliv.b; cf.pl. xlv.a, fromCairo. The specimen pilaster
ofplainrect-
illustratedin pl. ix. 3, fromthesiteof theCopticpatriar- The majority oftheunder-bases are,however,
chateinAlexandria, is nowin theGreco-Roman Museum, angularshape(e.g. pl. LXVI), in one caseelaboratedinto
inv.no. n«8i. a crouchinganimal(pl. XXIV).
188Kaufmann, Menasstadty pl. 68. 2. Kautzsch illus-
64 THE BRITISH SCHOOL AT ROME

form, butoctagonal insection(PL VII, 2) arelessgenerally commonthantherectangular

type. An example, similar to those in the Period IV church at Abu Mina, is recorded
fromSaqqara.138
Thereremainfivefragments onlywhichbetray a certain relationship to contemporary
Coptic work. Three of these belong to a set of pilaster-capitals found in thebaptistry
(PL IX, 4).1^9 All three echo Corinthian forms in which, as at Ahnásand Saqqara,the
voluteshavebeenabsorbedandunitedwiththesheath-leaves. Moreimportant perhaps,
theyillustrate in a loosebutunmistakable formtheWedelranken, oracanthus shootswith
leavesspringing frombothsidesof thestem,whichplayso largea partin earlyCoptic
ornament.Kitzingerhas convincingly demonstrated thatthisis a local Egyptian^con-
vention, theearlystagesofwhichcanbestudiedat Oxyrhyncus.140 The substance ofthese
capitalsis, then, common to the Coptic tradition; and a further element common to the
localEgyptian School,thoughitwasrejected the
by sculptors of Annas, is the longnaked
ofthefourth
caulicoli also a
fragment, pilaster-capital, which was found in the buildings
lyingto thenorthoftheTheophilanchurchandidentified byKaufmann as themonastic
quarters (PL X, 1). The samenakedcaulicoli canbe seenona fourth-century capitalfrom
Oxyrhyncus.141 Even at this earlystage, however, the comparison reveals the essential
difference betweenthesoftfluidity oftreatment at Abu Minaandthehardangularity of
theCopticandpre-Coptic series.Thesefourpilaster-capitals at Abu Minareveala cer-
taincommunity oftradition, butnoneoftreatment. Atmosttheyindicate a reciprocal in-
fluence fromcontemporary sculpture. The fifth fragment (PL IX, 2)142is all that remains
of a chancel-slab. The littledetachedsprayof threeleavesin thesurviving triangular
compartment wouldnotbe outofplacein theCopticseries.
Fromthisbriefsurveyit is possibleto drawtwo conclusions at any rateof some
generalsignificance. The firstof these is the essentially conservative character ofall the
surviving fourth- and fifth-century material at Abu Mina. A more detailed studyreveals
localcharacteristics anddetailsoftreatment todistinguish itfromthecontemporary work
ofothereastern Mediterranean centres.Butbycontrast withthemannerisms ofthecon-
temporary Egyptian pre-Coptic andCopticstylerepresented at Oxyrhyncus andat Annas,
itrepresents a decided,ifdecadent, classicism.The secondpointofinterest is that,apart
fromrepairsandreplacements, thereis nothing herethatis obviously laterthanthefifth
century.In particular, thereis no tracehere,and verylittlein Alexandria itself,of the
Justinian imperial art, which is so striking a feature in the reconquered Vandal provinces
of NorthAfricaand in Italy. This absenceis a revealing commentary on thescopeand
purposesof the policyof deliberate culturalre-absorption, whichwe see reflected not
only in the of
pages Procopius, but in the remains of so manysixth-century buildings,
greatandsmall,in thecentralandwestern Mediterranean.
We can supplement thispicturebya glanceat theChristian remainsof Alexandria
188Quibell, cit.vol. Ill, V. For itsuse elsewhere Hersha,Svria,third-centurv templein antis).
op. pl.
see L. Crema,'I monumenti architettoniciafrodisiensi', 189AfterKaufmann, Menasstadt, pl. 65. 8. The other
Monumenti Antichi,XXXVIII, 1939,col. 223,figs.57,58, twoareillustrated op.cit.pls.65. 9 and 68. 7. Cf.notably,
and 60 (Aphrodisias,in Caria,a latermodification to the Duthuit,L*artcopte,pl.lx.b, panelfromDeir el-Ganadla,
Tiberianporticoof the agora); A. von Gerkan,Das nearAboutig;H. PeirceandR. Tylor,Van byiantin, vol.
Stadwn(Milet,ed. Th. Wiegand,vol. II. 1), pl. Ill and V I, Paris,1932,no. 139.
140Kitzinger,
pp. ios-6; pls.lxxi-lxxii.
(Miletus,the latethird-centuryeast entranceto the sta-
141Kitzinger,
dium); D. Krenckerand W. Zschietzmann, Romische p. 188,pl.lxviii.1. .
142 Menasstadt, pl. 68. 6.
Tempelin Syrien,Berlin-Leipzig, 1938, p. 245 ff.(Ain AfterKaufmann,
 THE SHRINE OF ST. MENAS IN THE MARYÜT 65

itself,as represented by the collectionsof the city'sGreco-RomanMuseum. These not-

ably confirm the impressionof a somewhatconservativeclassicism. The majorityof the
capitals are of Corinthianor compositetypes,idéntica/in form,material,and craftsman-
ship with those of Abu Mina. There are, however,in additiona certainnumberof more
exoticforms,whichindicatewiderhorizons. There are, forexample,two Ionic impost-
capitals fromrue Sidi Abu Dardá.143They are carvedin local stone, and the wreathed
crosson the impoststampsthemas of local workmanship (see p. 67). But whiletheform
fallsreadilyintoplace withina long seriesrepresented in the Greekmainland,it is found
nowhereelse in the Egyptianseries. These two capitals are unquestionablyintruders.
The same is trueof a remarkablefifth-century Corinthiancapital with foliatedmasksat
theanglesin place of thevolutes.144It belongsto a widelydistributedand variedgroup
of capitals with a long classical ancestry,of which other examplesare recordedfrom
Egypt. In this instance,however,the hard, drilled treatmentof the acanthus leaves
and of the foliageof the hair and beard leaves no doubt as to its direct metropolitan
affinities.
To the same generalorbitas the capitalsand othersculpturalremainsof Abu Mina
belonga groupof antiquitiesfoundduringquarryingoperationsto thewestof Alexandria
nearthe villageof el-Dekhela,and now in the Greco-RomanMuseum. This is believed
to be the site of one of thegroupof monasteriesof Ennaton,a communitywhichtook its
namefromits proximityto the ninthmilestoneout of the city. The remainsconsistof a
homogeneousgroupof fourteenepitaphsfroma monasticcemetery,twelveof themdated
between524 and 590;145and a groupof sculpturewhichpresumablyrepresents thescanty
remainsof the fittingsof the conventchurch. The latter comprisesthe column of a
canopyor of a chancel-screen, flutedspirallyabove and verticallybelow; the lowerpartof
a similarcolumn,slottedto takea transenna;sevenCorinthianor flutedcapitalsof various
sizes; an open-worktransenna(Pi. IX, 1 and 5); two smallchancel-posts (Pi. VIII, 3); and
a bas-reliefof St. Menas betweentwokneelingcamels(Pi. VIII, i).146 The last-namedis
almostcertainlya copyof the cult-figure at Abu Mina (see p. 46); and the capitalsare in
thesame simpletraditionas themajorityof thesmallercapitalsat Abu Mina. The trans-
enna, and in particularthe foliateornamentin the anglesof the centralpanel, offeran
interestingparallel to the pilaster-capitalfrom the monastic quarters illustratedin
Pi. X, I. The evidenceof the inscriptions,which may well be a chance groupfroma
largerseries,tells us no more than that the communitywas alreadyestablishedin 524»
The sculpturemaywell be earlier;and whateverits date, thetraditionis clearlythatillus-
tratedin the earlyyearsof the fifthcenturyat Abu Mina.
The sixthcenturyis represented by a singleimportedgroupof threeimpost-capitals of
varying sizes in the Greco-Roman Museum; a also
fourth, from Alexandria, in the Cairo
Museum; and a fifthin the en-Nasirmosque at Cairo.147All areof thetype,bestknown
of
perhapsfromits use in San Vítale at Ravenna,with an overallpattern basket-work
143Kautzsch,no. 536,pl. 33. 147Alexandria, Greco-RomanMuseum,inv.nos. 3, 5,
144Kautzsch, ad
no. 762,pl.45; Ev. Breccia,Alexandrea and17013; Kautzsch, no.632,pl.38; E. Breccia,Alexandrea
Aegyptum, Bergamo,1922,p. 289,fig.200 (above). ad Aegyptum, Bergamo,1922,p. 289, fig,200 (below).
146G. Lefebvre,'Epitaphesde moinesalexandnns, B.S.R.A.Alex.y No. 9, 1907,pp. 108-9;No. io, pp. 232-3,
B.S.R.A. Alex.No. 8, 1905,pp. 11-19. fig.34. Cairo,fromAlexandria,J.Strzygowsky, Kopúsche
148Ev. Breccia,"Dun édihced époaue chretienne a el Kunst.y no.7352,fig.105; Duthuit,L*artcopti,pi xlvii.c.
Dekhélaet l'emplacement du Enaton',B.S.R.A. Alex.No. Cairo, ^n-Nasir Mosque; Kautzsch, no. 630,pl. 38.
9, I9O7,PP-3-".
F
66 THE BRITISH SCHOOL AT ROME
abouta central rectangular panelofstylised lotus-ornament.148 froma num-
It is familiar
berofwidelyscattered sites. KautzschlistsRavenna,Parenzo,Constantinople, Antigoni,
Brussa,Nicaea,Pergamon, andTrebizond.To thislistmaybe addedat leastthreeother
NorthAfrican sites:Kairouan,in themosqueof Sidi Okba, fromCarthage; Tripoli,a
damagedspecimen now in the Castle grounds; and Barce in Cyrenaica. Not onlyis this
formofcapitalclearlydatedto thetimeofJustinian; butit is quiteevidentthatin most
casesat anyrateitwaseitherexported ready-made froma common whichisusually
centre,
assumedtohavebeentheimperial quarriesof Proconnesus, or it was completed locallyby
workmen trainedin a commoncentraltradition.The factthatin Egypttheseimported
exampleshave been foundonlyin Alexandria acquiresan addedsignificance whenwe
recallthatit was fromthis,and fromrelatedmetropolitan Justinian formsof capital,
thatseveralofthemostdistinctive members ofthelaterCopticrepertory werederived.149
It looksverymuchas if,despitea divergence ofstyleandofdetailedtreatment thatwas
sufficiently marked to maintain its individuality unimpaired, it was nevertheless to
Alexandria thatCopticEgyptlookedforcontactwiththeouterworld.
The samerelationship emerges fromthestudyofwhatwasoneofthecommoner, and
certainly one of the more distinctive, of thesculptural motifsemployed at Abu Mina and
in theregionof Alexandria.This is theequal-armed crosswithina wreathboundwith
shortcurlyribbons(see Pis. IX and X). Whetheror not it is foundelsewhere in pre-
ciselythis form, it is hard to its on
say. Certainly appearance capitals that are normally
regarded as importsfromabroad(e.g. thewell-known capitalfromAnnasin theCairo
Museum,listedbelow)suggests thatit oughtto be foundelsewhere; indeed,ifnot,it will
be necessary to reconsider theacceptedviewsas to thesourceof thesecapitals,and to
examine ingreater detailthepossibility thattheyweremade,orcompleted, locally.In any
case,it is certainly foundin theareaof Alexandria in suchlargenumbers relativeto the
amountof surviving material, thatitmayrightly be regarded as a peculiarlyAlexandrian
motif.The following listis doubtlessincomplete, butit givesat anyratea fairviewof
therelativeincidence.

Abu Mina: presentlocation1 Marble capital, akin to thatfromAnnas(see below), with a wreathedcrosson
the abacus. Kaufmann,Menasstadt, pl. 68. 5; cf. (but without the wreathedcross), pl. 68. 7, now
in Frankfurt(presumably= Kautzsch, no. 79, pl. 6). Kaufmann,pl. 72. 3, illustrateseitheranother
similar capital or anotherface of the same capital.
: presentlocation? Marble capital, similar to the above but less deeply cut; wreathed cross on
abacus. Kaufmann,Menasstadt, pl. 70. 2.
: in the Greco-Roman Museum, Alexandria,¿nv. no. 17008. IncompleteCorinthiancapital with
wreathedcrosson the abacus (PL X, 4).
: present location? Colonnette-capitalwith acanthus leaves at angles and wreathedcross in the
centreof one face. Kaufmann,Menasstadt, pl. 69. I.
: presentlocation? Colonnette-capital,plain save for a wreathedcross with up-swept ribbons in
the centreof each face. Kaufmann,Menasstadt, pl. 69. 2.
: presentlocation? Colonnette-capital,withrudimentary acanthus-leavesat the anglesand wreathed
cross in the centreof one face. The cross has a hooked P terminal to the upper arm. Kaufmann,
Menasstadtypl. 69. 3.

148W. F. Volbach,'Intornoai capitelli

a fogliadi lotoin
sky'stacitassumptionthatin Egypttheseformscould
S. Vítalea Ravenna',Felix Ravenna,n.s. fase.II, 1934,
precedetheByzantine prototypeswillnotbearexamina-
pp. 124-9. tion; see O. Wulff,Byiantinische
Zeitschrift
XIII, 1904,
149Kautzsch,pp. 232-4; Kitzinger,
p. 190.Strzygow- p. 564 ff.
 THE SHRINE OF ST. MENAS IN THE MARYÚT 67
: in the Greco-Roman Museum, Alexandria,¿nv.no. 17444. Marble pilastercapital with central
wreathedcross(Pi. X, 1). Kaufmann,Mtnasstadt, fig. 36.
: in situ. Impost-blockof limestone,plain save for wreathedcross on each of the four faces (PL
X, 3).Kaufmann,Menasstadt, fig. 31.
: in situ,now in crypt. Fragmentsof a heavy marble impost-block( ?) with wreathedcross on one
face.
Annas. In the Cairo Museum, ¿nv.no. 7350. ImportedCorinthiancapital of Greek marble,one of a set of
at least six, fiveof which were still to be seen in the churchat Annas in 1902. On the foursides
of the abacus are an eagle, a wreathedcross, a formalspray,and a twined boss. J. Strzygowski,
Koptische Kunst(Catal. generaldesantiquités igypticnnes du Musicdu Caire, vol. xii)t Vienna, 1904, pp.
75-6, fig.102-3. Kautzsch, no. 81, pl. 6.
Alexandria:rue Sidi Abu Darda. In the Greco-Roman Museum, Alexandria,inv. no. 11901. A white
marble lintel or ornamentalcornice-mouldingwith a wreathed cross againsta shallow projecting
bracket(PL X, 5).
: rue Sidi Abu Darda. In the Greco-RomanMuseum, Alexandria,inv. no. 1 1791» A pair of Ionic
impost-capitalsin local stone,witha wreathedcrosson the impost-block. Kautzsch, no. 536, pl. 33.
Seep. 65.
: provenanceuncertain. In the Greco-Roman Museum, Alexandria. Bracketof whitelocal marble,
the projectingarchitrave-element of an engaged order,with a wreathedcrosson one face.
: Agnewcatacomb. Incised slab withinscriptionof the fifthor sixthcentury. H. C. Agnew,Archae-
ologia,XXVIII, 1840, p. 152 ff.,pl. XIV. Cabrol, Dictionnairc, s.v. Alexandriefig.292.
: Hagar-Nawatieh,nearthe canal at Ramleh. Greco-Roman Museum, Alexandria,¿nv.no. 14 158.
Marble stele(m. v6^ X 0-70 X 0*67) with metricalinscriptionrecordingthe clearanceof the canal
by Alexander,probablythe praefectus alis of 467. Above the inscriptionis a wreathed cross
august
(PL X, 2). G. Lefebvre,Recueildes inscriptions d*Egypte,
grecques-chritiennes Cairo, 1907.
Aschmunein. In the Cairo Museum, inv. no. 7351- As no. 7350 fromAhnas. Strzygowski,op. cit.
Cairo: Mosque el Mu'aijad. Capital similar to that fromAnnas (see above). Kautzsch, no. 43, not illus-
trated,but describedpp. 25-6.
: Mosque Salih Tala'i. Capital similarto that fromAhnás(see above). Kautzsch, no. 96, pl. 6.
: provenanceuncertain. In the Coptic Museum. Engaged half-columndivided into two zones:
below, overallpatternof lozenges and rosettes; above, spiral flutingwith superimposedwreathed
cross.
: provenanceuncertain. In the Coptic Museum. Small marble Corinthian capital with wreathed
cross.
: fromFostat. In the Coptic Museum, ¿nv. no. 4 181. Small altar-shapedbase with a wreathed
crosson one face. The ribbons are unusuallylong and terminatein ivy-leaves. Marcus H. Simaika
Pacha, GuideSommaire du MusicCoptc,Cairo, 1937, pl. xlii.
: found re-usedin the sub-structures of the Nilometer at.Rodah, and believed to come originally
fromHeliopolis. A broad, shallow pilaster-capitalwith a wreathedcross in the centreof the face.
E. Drioton, Lessculptures coptesdu Nilometre deRodah,Cairo, 1942, pp. 88-91, no. 19.
: in the churchof el-Moallaqa, Old Cairo. Marble ambon-panel, re-used in the later medieval
ambon, with a wreathedcrossbeneatha shell-headedcanopy.
El-Dekhéla: fromthepresumedsiteof themonasteryat Ennaton. In theGreco-RomanMuseum,Alexandria.
Small pilaster-capitalwith wreathedcross on one face. The ribbons in this case are omitted; but
in all otherrespectscf. Saqqara, below. B.S.R.A.Alex., No. 9, 1907, pp. 3-12, fig.4.
InscribedGreekstele with wreathedcross. W. E. Crum, Coptic
Faiyum. In the Cairo Museum, no. 8506.
Monuments (Catal giniraldes antiquitésigyptiennesdu Musicde Caire, vol. lV)t Cairo, 1902, pl. XIX.
Rosetta. In the Greco-Roman Museum, Alexandria, inv. no. 22234. Re-used Ptolemaic lintel of black
marble,re-carvedwith threewreathedcrosses(PL IX, 6).
almost identicalwith that fromel-Dekhela.
Saqqara: fromthe monasteryof Apa Jeremias.Pilaster-capital,
J. E. Quibell, Excavations at SaqqaraIV, 19 12, pl. XXXV. I.

It will be seen thatoutsidethe immediateneighbourhoodof Alexandriathe motifis

found, in the firstplace, on a group of importedcapitals (Ahnas, Aschmunein,the
local pre-Copticand Coptic series. But it is
mosquesof Cairo) whichare intrudersto the
68 THE BRITISH SCHOOL AT ROME

foundalso sporadicallyon objects of local workmanship,sometimesrelativelyunaltered

(half-columnand capital in the Coptic Museum, ambon-panelin the church of el-
Moalláqa, colonnette-capitalat Saqqara), more often modifiedto suit local taste. An
interesting exampleof suchmodification is a funeralstelein the Coptic Museum, inwhich
the equal-armedcrossis replacedby theankh-cross.160More oftenthemotifis simplified
by the omissionof the ribbons,as in the large seriesof Middle and Upper Egyptian
funeralstelae with plain wreathedcrosses. A derivationfromthe Alexandrianribboned
formseemsclearfromthe existenceof such examplesas no. 844 3 (Greek stelafromthe
Faiyüm,listed above). The ribbonsfirstdegenerateintolooped dots in the lowerangles,
and thesedots are then treatedas purelydecorativedevices,applicable to all fourangles
or even to the two upper angles only. In other words,the studyof this motifshows
that Coptic sculptorswerefamiliarwith,and drewupon, the resourcesof contemporary
Alexandria;but that,while they sometimesreproducedtheir models unaltered,more
often these underwenta more or less radical transformation to accord with Coptic
capabilities and standardsof taste.
In the finalsectionof his articleKitzingerdiscussesthe questionwhetherthe sculp-
turalevolutionrepresentedat Oxyrhyncus,at Annas,and at Bawit and Saqqara, can be
regardedas typicalof theentiredevelopmentof Egyptianartduringthefourth,fifth, and
sixthcenturies,or whetherit appliesexclusivelyto themorepopularspheresof thehinter-
land as distinctfromAlexandriaitselfwhich, accordingto one theory,remainedthe
centreof a livingand cultivatedHellenistictraditionthroughoutthe Christianperiodin
Egypt.151In a reasonedrejectionof the latterview, Kitzingerremarksthat purelygeo-
graphicalargumentsare notconclusive;forwhile it is truethatall theprincipalgroupsof
Coptic sculpturediscussedlie in the Nile Valley southof Cairo, thereis no comparable
bodyof materialfromthe soil of Alexandriaitselfwithwhichthesemaybe comparedor
contrasted.The largestsurvivingbody of Christiansculptureavailableforcomparisonis
the seriesof capitalsfromAbu Mina and fromthe mosquesof Cairo; and these,he main-
tains,are internationalin character,and have no moreconnexionwith Alexandriathan
they have, for example, wrthSalona, Greece, or Ephesus. In other words, the dis-
tinctionthatcan be drawnis valid onlyin termsof a local and of an officialart,notof the
artsof two geographicalentities,the hinterlandand Alexandria;and of the two it is the
sculptureof the Coptic Churchesthatis the moredeeplyrootedin the Hellenisticartof
Egypt.
It is, however,on the evidenceof Early Christianivories,ratherthan on sculpture,
that the theoryof a survivalof Hellenism in Alexandriais based.152There are many
studentswho maintainthata greatdeal of the finestsurvivingEast Roman ivory-carving
of the fifthand sixthcenturies,outstandingamongthembeingtheThroneof Maximian
at Ravenna,are the workof craftsmenworkingin Alexandria,or at anyrateworkingin
the closest contactwith Alexandria. Kitzingerdoes not discuss any of the arguments,
primarilyiconographic,which are advanced in supportof this view. In discussingthe
characterof the sculpture,the Egyptianoriginof these ivoriesmust,he maintains,be
regardedratheras the object of the whole enquirythanas an argumentin favourof one
160MarcusH. SimaikaPacha,GuideSommaire duMusée 1MFor a
concise,up-to-dateexpositionof thisview,
Copte,Cairo,1937,p. 16,no. 32, pl. xliii. withearlierbibliography,
see C. R. Morey,EarlyChristian
181Kitzinger.pp. 210-15. ArtyPrinceton.1942,pp. 89-97.
 THE SHRINE OF ST. MENAS IN THE MARYÜT 69
or the othertheory.We have, instead,to look forindependentevidenceof the survival
of a classical traditionin Egypt's fifth-and sixth-century sculpture;and this we do,
in fact,findin a few isolated but importantworks which afforddefiniteevidenceof
its persistence.
One suchworkis thecarvedwoodendoorfromthechurchof St. Barbarain Old Cairo.
It has been publishedas a fourth-century work;153but a comparisonwiththeivorycarv-
ings, to which it is closelyrelated, shows veryclearlythatit can hardlybe earlierthanthe
close of the fifthcentury.The vine-scrolls,which have nothingwhateverto do with
Coptic sculpturaltradition,recall thoseof the Throne of Maximian; while the pairs of
flyingangels,upholdinga bust withina ribbonedwreath,bear a strikingresemblanceto
thoseon the celebratedBarberiniDiptychin the Louvre. The same relationshipemerges
fromthe studyof a sscond work,a limestonepilasterfromBawit,now in the Louvre.154
It is carvedon one facewitha geometricornamentofhexagons,on a secondwitha delicate
medallionvine-scrollwith birdsand a chalice,whichcloselyresemblesthe scrollson the
Throneof Maximian. To clinchthe relationship,at the top of eitherfaceof the pilaster
therestandsa figureof an angel, which unmistakablyechoes the archangelof the cele-
bratedArchangelDiptych in the BritishMuseum. In otherwords,therewere working
in Egyptat the turnof thefifthand sixthcenturies,theapproximatedate of the ivoriesin
ratherthanthe
question,sculptorswho followedthe classicalmannerof the ivory-carvers
Coptic style of Annas.
Are we entitledto assume that the work of these sculptorsrepresentsa surviving
Hellenistictraditionestablishedin Alexandria? Kitzingermaintainsthatwe are not. In
the sculptureof Oxyrhyncushe recognisesa typicallylate-antiquestyle,basicallyinter-
national but with certainwell-markedlocal peculiarities.Members of the same late-
Africa. The artof
antiquefamilycan be seen in Syria,Asia Minor,Dalmatia, and North
is
Oxyrhyncus provincialindeed; but it tendencies
reflects widespread the late-antique
in
Mediterraneanworld;and the sourcefromwhichit was immediatelyderivedcan hardly
be otherthan contemporary Alexandria. By contrast,it cannot be shown that the St.
Barbaradoorsor the Bawit pilasterhave any forerunners in Egypt; it is theytherefore,
and not the Coptic sculptures,which representan element foreignto the Egyptian
tradition.This elementis not an Alexandrianstyleat all, but a courtstyle,an essential
characteristic of whichis thatit transcendedprovincialboundaries. 'The basketcapitals
foundat Bawit and Saqqara are an obvious example of Byzantinecourtart influencing
and of the Paris(i.e. Bawit)
Egyptin thesixthcentury.The styleof the St. Barbaradoor
must be explained in a similarway. We cannotcall it an Alexandrianstyle. The
pilaster
artists who did such work in Egypt in the sixth century must have been trainedin the
of
courtatmosphere Byzantium. If we want to see what had by that time happened to
tradition of Hellenistic art, we must not look at these few and isolated
Egypt's own
monuments, but at the genuinelyCoptic sculptures in the monasteriesof the Nile
Valley.155
It is not withinthe scope of the presentarticleto examinein any greaterdetail than
does Kitzinger,the argumentswhich have been advanced for an attributionto Alex-
1MA. Patricoloand U. Monneret de Villard,La chiesa Early MuslimArchitecture, II, 1940,fig.483-
164Kitzinger,
pl. lxxvii,1.
di Santa Barbaraal VecchioCairo,Firenze,1922.Illus- 156Kitzinger,
tratedby Kitzinger,pl. lxxvi,1; and by K. A. Creswell, p. 215.
 3 THE BRITISH SCHOOL AT ROME
andriancraftsmen of suchivoriesas theThroneof Maximianor the BritishMuseum
Archangel Diptych.In generalterms,it maybe remarked thattheiconographic method
undoubtedly has its pitfalls, and the validity of itsconclusions must restultimately upon
a hardcoreofindependently established fact:butthat,so faras thepresent writerisaware,
nocritichasattempted todisposeindetailoftheelaborate structure oficonographic argu-
mentuponwhichthisthesisis based;noris sufficient regard alwayspaidto theverycon-
siderablebodyof independent evidencebywhichit can be shownto be linkedto such
knownEgyptianantiquitiesas the frescoes of the Alexandrian catacombsand of the
Coptic monasteries or to the peculiarly Egyptianliterary versions of thefamiliar bible
stories.156 It maywell be thatnotall theivoriesclaimedas Alexandrian weremadein
Alexandria itself;it is arguable, forexample,thattheThroneof Maximianwasmadein
Constantinople by Alexandrian craftsmen working underimperialpatronage.But even
if thiswereso, it wouldstillseemto be necessary to postulatea continuous traditionin
Alexandria itselfto accountforthecontinued and consistent of
employment seemingly
Egyptian elements in thishypothetical, eclectic,courtschool.
It is,however, withtheevidence afforded bysculpture in woodandstonethatwe are
directly concerned; and here there are two important questionsof interpretation and of
terminology on which it be that
may suggested Kitzinger's conclusions requiremodifi-
cation.In thefirst placewe haveseenthat,whileadmitting theexistence in Egyptofa
considerable of
body Early Christian sculpture, mainly in the form of capitals,whichis
quite distinct from the Coptic work of the fifth and sixth centuries, he maintains that
thissculpture hasno specialconnexion withAlexandria orwithanyEgyptian Hellenistic
tradition, butbelongstoan international, court-inspired art,theproducts ofwhichfound
theirwayindiscriminately all overtheEasternMediterranean world.This viewsurely
misinterprets both the rangeand the character of this non-CopticEgyptianEarly
Christian sculpture.Antiquity the
regarded sanctuary of St. Menasas an Alexandrian
shrine;andwe haveseenthattheremains fromthesoilofAlexandria itselfandfromthe
Maryüt confirm this evaluation. The body of material is not large; it is consistent,
but
andquitelargeenoughto showus something ofthecharacter offifth- andsixth-century
Alexandrian sculpture and at the same time to point the absence of anytraceof con-
temporary Coptic work. There are a few fragments from Abu Mina itself, whichindicate
contactwithCopticideas. But theselack precisely thosemannerisms whichare most
typicalofactualCopticworkandso serveonlyto underline thefundamental divergence
of thetwostyles(see p. 64).
To describethesculpture of Oxyrhyncus, of Annas,and of Bawitand Saqqaraas in
anysenserepresentative of thewholeof Egyptis, then,to inviteconfusion. That its
inspiration came in no small degree from Alexandria seems to be demonstrated. Butit
does not necessarily followthatit was a mereprovincial dialect,or thatin it we can
identify, in rudebutrecognisable form,theartofAlexandria itself.The capacity ofthe
Copticartistto selectandto transmute borrowed forms in termsofhisownvigorous and
strongly individual tastes and conventions is well demonstrated in the sixthcentury in
thecaseof severalfamiliar Justinian forms of capital. This capacity was not new. We
see it, forexample,in wall-paintings of theearlierEmpirefromtheFaiyümand else-
where.It wasthecenturies-old reaction ofthenativeEgyptian to thealiencultureofthe
156Of
particularinterestin the presentcontext are the representationsof the Menas cycle cited on p. 46.
 THE SHRINE OF ST. MENAS IN THE MARYÚT 71
Delta;andwhileit maybe trueto describe thesculpture ofOxyrhyncus andofAhnásas
a memberof a largeand widelyscattered late-antique family, the element whichit
possesses in common with other late-antique sculptural styles can be amplyexplainedin
termsoflikereaction underlikeconditions to a commonsourceofinspiration: it is not
to
necessary postulateany closer or more immediate connexion.
The architectural remains tellthesamestory.Atleastonefeature represented in later
Coptic architecture, the use of columnar exedrae in the narthex, can be shown to be
derived fromAbuMina;andit maybe thatothers, suchas thetriconchos chancel,whichit
sharedwithEarlyChristian architecture outsideEgypt,157 foundtheirwayintoCoptic
architecture byway of the vanished churches of Alexandria. ButifAbu Mina be rightly
as a
regarded representative Alexandrian church, it is abundantly clearfromtheevidence
discussedabovein sectionD (pp. 51-60) thatthe architecture of Alexandria and of
CopticEgypt followed sharply divergent paths, and that, as in the case of sculpture, there
werein EgypttworelatedbutdistinctschoolsofEarlyChristian architecture.
This bringsus to oursecondquestionof interpretation andof terminology. Can the
Alexandrian artofthefifth andsixthcenturies properly be called international ? Bycom-
parison with that of CopticEgypt it was undoubtedly cosmopolitan in its connexions;
but it is an over-simplification of a complexpictureto describeit as 'international',
' andqualified.It is truethatindi-
unlesstheuseofthattermbe veryclearlyunderstood
viduallocalschools,andinthepresent instance theconservative HellenismofAlexandria,
haveat timesbeenallowedto bulklargerin thehistory of.EarlyChristian artthanthe
knownfactswarrant.The extreme reactionagainstthisexcessivelocalisation is a bald
andrather characterless internationalism, in which localtraditions are recognised as little
morethanunimportant variantsofthecentralmelody.It is surelytruerto regardthem
as the themesof a greatcontrapuntal fugue.Each was colouredand modifiedby its
relation to theothers:buteachin turncontributed itsuniqueandindividual quotato the
harmonic whole.The capitalsof Abu Mina and of the mosquesof Cairo belongto
familiartypes,representatives of whichare scatteredthroughout the contemporary
Christian world.Manyof them,however, werenotonlymadein Alexandria butreveal
characteristics whichappeartobe peculiarto Egypt.The individual architectural features
ofAbu Mina,too,canwithoutexception be paralleledoutsideEgypt.Buttheresultant
wholeis uniqueand makesit veryclearthatthearchitecture ofAlexandria wasfarfrom
being a colourless copy of models developed elsewhere. The earlyappearanceof such
features as the spherical-triangular pendentive and the squinch thislightacquiresadded
in
significance. Indeed the importance of Abu Mina lies verylargely in theevidencewhich
itaffords of the continued individuality of Alexandria as a centre of thearts.It doesnot
tellus as much as we should liketo know: stillless does it afford anydirectevidence as to
of
thesource thedisputed ivories. It does,however, help to clear the ground, byputting
the sculptureof Coptic Egyptinto its true perspective; and it does suggestthat
further research at Abu Mina itselfand in themanyChristian sitesof theMaryütmay
yetthrowfreshlight on one of the fundamental outstanding problems oflateantiquere-
search, thecontribution ofAlexandria to the Koine of Early Christian art andarchaeology.
J. B. Ward Perkins
167U. Monneret de Villard, Les couventsprés de Sohag, Milan, 1925, chapter III.
 PLATE III

(/. B. WardPerkins)
i. Abu Mina : The Chancel of the Ninth-century Church.

(J. B. War. Ains)

2. AbuMina : General View, Looking East. In the Foreground, the Baptistry,and beyond
it {surrounded the Shaft over the Tomb. The White Mounds in the Distance
byrailings)
Cover the Ruins of the City.
 PLATE IV

{AfterKaufmann)
i. Abu Mina : The Baptistry during Excavation (p. 49)

(/. B. WardPerkins)
2. Abu Mina : Remains of Earlier Apse beneath the Footings of the North Nave
Arcade of the Period IV Church (p. 43)
 PLATE V

\sljter JXaujmann)
Abu Mina : Staircase Down to the Crypt. The Stone Arch, incorporated obliquely
into the Brick Vault, carries the Footings of the Semi-Circular Exedra over the
Tomb (p. 44)
 PLATE VI

i. Remainsof Earlier Masonry incorporated 2. East Corridor of the Crypt, showing

in the Period IV Crypt (p. 41) loculi of the primitivecatacomb (p. 38)

3. Crypt: Dome on Triangular Pendentives, 4. Crypt: Half Dome on Squinches (p. 45)
before Restoration (pp. 45, 57^
{AfterKaufmann)
 PLATE VII

(/. B. WardPerkins) (/. B. WardPerkins)

i-2. Abu Mina : Marble Bases in the Period IV Church (p. 47)

{J. £S. Ward JTerKins)

3. Abu Mina : Burnt Marble Base of Period IV, Re-used in the Ninth-century Church
(P- 50
 PLATE VIII

i. Dekhela : Bas-relief, figuring St. Menas

(p. 46)
2. Karm Abu Girg : Wall-painting,
figuring St. Menas (p. 46)

3. Dekhela :
Chancel- 4. Abu Mina : Semi-Circular Shaft over the Tomb, during Excavation. Left {above
post photographer's Spring of Barrel-Vault (p. 46)
shadow)
(p. 65) (1-3, bycourtesy Museum,Alexandria
oftheGreco-Roman ; 4, afterKaufmann)
 PLATE IX

i. Dekhela : Screen-panel (p. 65)

2. Abu Mina : Screen

Fragment (p. 64)

5. Dekhela : Detail of Fig. i.

3. Alexandria: Capital (p. 63)

4. ABU JV1INA
: riLASTER-CAPITAL (p. 64) 6. Rosetta : Detail of Lintel (p. 67)
(1,3 and 5, bycourtesy Museum,Alexandria
oftheGreco-Roman ; 6, /. B. WardPerkins)
; 2 and 4, afterKaufmann
 PLATE X

3- Abu Mina : Impost Block (p. 67)

1. Abu Mina : Pilaster (pp. 64, 67)

4. Abu Mina : UnfinishedCapital (p. 66)

2. Hagar Nawatieh : Stele (p. 67) 5. Alexandria : Detail of Lintel (p. 67J
oftheGreco-Roman
(1, 2, 4 and 5, bycourtesy ; 3, /. B. WardPerkins)
Museum,Alexandria
ABU /AINA PERIOD IV, EARLY FIFTH CENTl
FIFTH CENTURY
PLATE XI