Limyra

INSTITUT FRANÇAIS D’ÉTUDES ANATOLIENNES GEORGES-DUMÉZIL-ISTANBUL XXXI 20 23 A N A TO L I A ANT I Q U A E S K İ A N A DOL U ANATOLIA ANTIQUA ESKİ ANADOLU XXXI OFFPRINT ANATOLIA ANTIQUA ESKİ ANADOLU XXXI Recueil de travaux publiés par l’Institut Français d’Études Anatoliennes Georges-Dumézil Istanbul édité par Deniz GENCEOLU INSTITUT FRANÇAIS D’ÉTUDES ANATOLIENNES GEORGES-DUMÉZIL-ISTANBUL CNRS USR 3131 ZEROBOOKS Édition - Diffusion ISTANBUL 2023 Ce volume a été composé par les soins de l’Institut Français d’Études Anatoliennes-Georges Dumézil et Zero Prod. Ltd. © 2023, Institut Français d’Études Anatoliennes Georges-Dumézil – Istanbul ISBN 978-2-36245-088-4 ISSN 1018-1946 e-ISSN 2823-1090 Éditions Ege Yayınları - Numéro de certificat : 47806 Illustration de couverture : Photo zénithale du secteur froid des bains. La saison 2022 à Labraunda. Secrétariat de rédaction IFEA - Palais de France Nur-i Ziya Sok. No. 10, 34433 Beyoğlu / Istanbul - Turquie Tél +(90) 212 244 1717 E-mail : publications@ifea-istanbul.net ; www.ifea-istanbul.net Production et diffusion Zero Prodüksiyon Kitap-Yayın-Dağıtım San. Ltd. Şti. Abdullah Sokak, No. 17, Taksim 34433 Beyoğlu / Istanbul - Turquie Tél. : +90 (212) 244 7521 Fax. : +90 (212) 244 3209 E-mail : info@zerobooksonline.com www.zerobooksonline.com ; www.egeyayinlari.com.tr Imprimé par Fotokitap Fotoğraf Ürünleri Paz. ve Tic. Ltd. Şti. Oruçreis Mah. Tekstilkent B-5 Blok No: 10-AH111 Esenler / Istanbul - Turquie Tél. : +90 (212) 629 0385 - +90 (541) 231 1887 Numéro de certificat : 47448 La loi du 11 mars 1957 n’autorisant, aux termes des alinéas 2 et 3 de l’article 41, d’une part, que les “copies ou reproductions strictement réservées à l’usage privé du copiste et non destinées à une utilisation collective” et, d’autre part, que les analyses et les courtes citations dans un but d’exemple et d’illustration, “toute” représentation ou reproduction intégrale, ou partielle, faite sans le consentement de l’auteur ou de ses ayants droit ou ayants cause, est illicite (alinéa 1er de l’article 40). Cette représentation ou reproduction, par quelque procédé que ce soit, constituerait donc une contrefaçon sanctionnée par les articles 425 et suivants du Code Pénal. Direction Directeur de la publication : Philippe BOURMAUD, Directeur de l’Institut Français d’Études Anatoliennes Directrice du comité de lecture : Anaïs LAMESA, Responsable du pôle archéologie Responsable d’édition : Deniz GENCEOLU Comité de rédaction Laurence ASTRUC, CNRS - ArScan (VEPMO) Olivier HENRY, Université Lumière Lyon 2 - HiSoMA (UMR5189) Éric Luc JEAN, Université Hitit (Çorum) Anaïs LAMESA, IFEA B. Tolga UYAR, Université Nevşehir Hacı Bektaş Veli Comité de lecture Catherine ABADIE-REYNAL, Université Lumière-Lyon 2 Anna ANGUISSOLA, Université de Pise Dominique BEYER, Université de Strasbourg Wolfgang BLÜMEL, Université de Cologne Isabella CANEVA, Université de Salento (Lecce) Olivier CASABONNE, Societas Anatolica (Paris, Istanbul, Çorum et Louvain-la-Neuve) Jacques des COURTILS, Université Michel de Montaigne Bordeaux 3 Roberta FABIANI, Université Degli Studi Roma Tre | UNIROMA3 Véronique FRANÇOIS, CNRS, LA3M (UMR 7298) Marcella FRANGIPANE, Université de Rome, “La Sapienza” Marie-Henriette GATES, Université Bilkent İlgi GERÇEK, Université Bilkent Martin GODON, Attaché de coopération scientifique et universitaire/ Directeur de Campus France-Turquie/SCAC Ambassade de France Winfried HELD, Université de Marburg Olivier HENRY, Université Lumière Lyon 2 - HiSoMA (UMR5189) Éric Luc JEAN, Université Hitit (Çorum) Francis JOANNÈS, Université Paris 8 Koray KONUK, CNRS, Ausonius (UMR5607) Kemalettin KÖROĞLU, Université Marmara Catherine KUZUCUOĞLU, CNRS, LGP (UMR8591) Caroline LAFOREST, Institut Royal des Sciences Naturelles de Belgique Vasilica LUNGU, Institute of South-Eastern European Studies, Académie Roumaine Catherine MARRO, CNRS, Archéorient (UMR5133) Philipp NIEWÖHNER, Université Georg-August Alastair NORTHEDGE, Université Paris IV-Sorbonne Mihriban ÖZBAŞARAN, Université d’Istanbul Christine ÖZGAN, Université Mimar Sinan des Beaux-Arts Raffaella PIEROBON-BENOIT, Université de Naples Francis PROST, École Normale Supérieure Cemal PULAK, Université du Texas A&M Haluk SAĞLAMTİMUR, Université de l’Égée Recai TEKOĞLU, Université du Dokuz Eylül Régis VALLET, CNRS, IFPO Amélie VIALET, Muséum national d’Histoire naturelle, UMR 7194 - UPVD, CERP de Tautavel Sommaire Olympia BOBOU, John HEALEY et Rubina RAJA Revisiting Edessa’s Funerary Portrait Habit ...................................................................................................................................................... 1 Olivier CASABONNE (Avec les amicales collaboration et complicité de Didier LAROCHE) Caryatides/Karyatides en Asie Mineure et Grèce Femmes porteuses et plantes bulbeuses : L’imaginaire religieux et la réversibilité de la vie et de la mort .................................................................................................................................................................................................................................. 53 Mehmet Ali KOCABAŞ, Stephen MITCHELL et Philipp NIEWÖHNER Filling in a Gap: Roman Votives and Gravestones as well as Early Byzantine Churches at Tavşanlı in Phrygia ........................................................................................................... 63 Yusuf SEZGİN et Gözde ŞAKAR A Plastic Vase in the Form of an Actor from the New Bouleuterion of Aigai ................................................................ 87 CHRONIQUES DES TRAVAUX ARCHÉOLOGIQUES EN TURQUIE 2022 Claire BARAT, Lorenzo d’ALFONSO, Emine KÖKER GÖKÇE, Jean-François PICHONNEAU, Alvise MATESSI, Alessio MANTOVAN, Braden CORDIVARI, Chamsia SADOZAÏ et Louise BRUNEAU Porsuk - Zeyve Höyük : Rapport préliminaire de la campagne 2022 .................................................................................. 111 Olivier HENRY, İpek DAĞLI et Görkem ÇİMEN, Anaïs LAMESA, Lucas SABATIER, Isabelle HASSELIN ROUS, Ragnar HEDLUND, Jesper BLID, Jeanne CAPELLE, Anna SITZ, Raphaëlle CHEVALLIER, Clémence PAGNOUX, Baptiste VERGNAUD, Naomi CARLESS UNWIN, Damla ÜSTÜNEL, Christophe BOST, Fabrice CHARLIER, Nicolas LAMARE, Frédérique MARCHAND-BEAULIEU, Mohamed SASSI, Alina STREINU, Alexandra SPÜHLER Labraunda 2022 – Rapport préliminaire ....................................................................................................................................................... 145 Çiğdem MANER, A. Emre KURUÇAYIRLI, Rula SHAFIQ, Hasan PEKER et Gökçe ÖZTÜRK A New Start at Kayalıpınar: 2022 Field Season ....................................................................................................................................... 179 Martin SEYER, Alexandra DOLEA, Philip M. BES, Dávid Zs. SCHWARCZ, Gerhard FORSTENPOINTNER, Danai KAFETZAKI, Nikolaus SCHINDEL, Ceyda ÖZTOSUN, Hakan ÖNIZ, Mercan HELVACIKARA, Zeynep KUBAN, Bilge AR, Umut ALMAÇ et Görkem GÜNAY The Excavations at Limyra (Lycia) 2022: Preliminary Report ................................................................................................... 205 Berkay DİNÇER, Serkan ŞAHİN et Göknur KARAHAN West Afyonkarahisar (Turkey) Palaeolithic Survey 2022 Field Methods and First Results ......................... 243 Abdulkadir ÖZDEMİR Elazığ Prehistoric Archaeological Survey (EPAS) 2022 Harput Plateau and Elazığ Plain: A Preliminary Report ........................................................................................................................................................................................................ 261 Asil YAMAN, Koray KONUK, Anna SITZ, Taylan DOĞAN, Merve YEŞİL, İlayda ALİKAYA, Dilara TAKI, Münire Rumeysa ÇAKAN, İbrahim ULU, Koray GÜNYAŞAR, Mutlu KARADAĞ, Alper Serhan DALTEKİN et Mehmet Serhat AYDEMİR A Preliminary Report on the 2022 Fieldwork at Phoenix ............................................................................................................. 277 Anatolia Antiqua XXXI (2023) 205–241 The Excavations at Limyra (Lycia) 2022: Preliminary Report Martin SEYER, Alexandra DOLEA, Philip M. BES, Dávid Zs. SCHWARCZ, Gerhard FORSTENPOINTNER, Danai KAFETZAKI, Nikolaus SCHINDEL, Ceyda ÖZTOSUN, Hakan ÖNIZ, Mercan HELVACIKARA, Zeynep KUBAN, Bilge AR, Umut ALMAÇ, Görkem GÜNAY* The excavation season at Limyra lasted from 1 August until 30 September 2022, with the permission granted by the Ministry of Culture and Tourism. We would like to express our gratitude to the state representative Yunus SUSAM from the Antalya Museum for his support and effort. Introductory Remarks (M. Seyer) In the 2022 campaign, the program for the study of the urbanistic development of Limyra (for a plan of the ancient city: Seyer et al., 2020, p. 220, Fig. 1) was continued. After an enforced interruption of two years, it did not seem appropriate to immediately realise new research projects but to initially focus the work on the archaeological research of the Western City, as in the campaigns before (Fig. 1). In this area, the excavations of recent years had already yielded valuable findings on changes in the urban structure from the Hellenistic to the Byzantine period (Seyer, Dolea, Kugler, Brückner & Stock, 2017; Seyer et al., 2019; Seyer et al., 2020)1. According to the findings to date, the Western City underwent a change in significance from a representative city district in the Hellenistic and Roman periods to a crafts quarter in the Byzantine era (Dolea, 2020, p. 228). In order to get a deeper understanding of these changes, the excavation of the past years was continued. Results to date have been primarily for the Early (mid-4th to early/mid-7th century) and Middle (early/mid-7th to 11th century) Byzantine periods, but 2022 brought unexpected new findings for the Roman Imperial period as well. The work in this campaign concentrated on a number of prominent areas within the existing excavation area that promised to yield significant insights into both the urban development of the site and its chronological sequence. These included, first and foremost, the central area with the foundation walls of a massive square building measuring approximately 10 x 10 m, as well as a water installation and a stairway immediately to the east of it. The second focus was on the eastern part of the site with an already partially excavated square and a building that might have been a tavern or a public eating house in Early Byzantine times. * M. Seyer, Ph. M. Bes, D. Zs. Schwarcz & N. Schindel, Austrian Archaeological Institute – Austrian Academy of Sciences, Vienna; A. Dolea, independent researcher, Vienna; G. Forstenpointner, University of Veterinary Medicine, Vienna; D. Kafetzaki, Universities of Leuven and Hasselt; C. Öztosun, Polish Academy of Sciences Archaeology Ethnology Institute, Warsaw; H. Öniz, Akdeniz University Underwater Research and Application Centre, Antalya; M. Helvacıkara, Middle East Technical University, Ankara; Z. Kuban, B. Ar, U. Almaç & G. Günay, Istanbul Technical University, Faculty of Architecture. 1 All dates are AD and circa unless indicated otherwise. 206 Martin Seyer et al. A special focus was on the documentation and scientific processing of the archaeological find material since it was an important goal to obtain new information on the relative and absolute chronology of the urban structure through the stratigraphic evaluation of the find material. In addition to the processing of the pottery, metal finds, and animal bones, the 2022 campaign also allowed for the first comprehensive study of the numismatic material from the 2016-2022 excavations, which so far includes approximately 1,000 coins. Limyra’s most important building from the Roman Imperial period is the Cenotaph of Gaius Caesar, the grandson, adopted son and presumptive successor of the Roman Emperor Augustus. Gaius Caesar died in Limyra on his way back to Rome from a diplomatic mission in eastern Asia Minor on the 21st of February, 4. The few preserved remains of this structure were scientifically researched in earlier phases of the excavation and published in an exemplary manner. Due to the difficult hydrological situation in Limyra, as a result of which large parts of the city are flooded (Öner, 2013, pp. 367-378; Rantitsch, Prochaska, Seyer, Lotz & Kurtze, 2016; Wiegand, 1973), its foundations lie underwater today. This is the reason why they could be measured and drawn in the course of the exploration, but not adequately documented photographically. This scientific gap could be closed in the 2022 campaign by a team of underwater archaeologists using modern equipment. The two circuits of the Late Antique/Early Byzantine fortification walls certainly are a landmark in Limyra, not least due to the impressive remains, particularly those encircling the Western City. In spite of that, they have not been investigated intensively to date. A project undertaken in the scope of a Ph.D. dissertation at Middle East Technical University (METU/ODTÜ) that started in 2022, began in its first phase, the documentation, analysis, and description of this huge structure. This work will form the basis of an extended study on how city walls were built in Lycia and Anatolia during Late Antiquity and the Byzantine period, which should be considered a source of information for conservation approaches. Apart from the scientific studies on the city’s urbanistic development, the Limyra archaeological project also carried out events promoting the education of junior researchers. In continuation of the activities of 2019 a summer school was carried out as a cooperation project between the Istanbul Technical University (ITU/İTÜ) and the Austrian Archaeological Institute. During this one-week training program students of architecture from ITU/İTÜ were given an insight into the necessary knowledge and the skills which are expected from an architect at archaeological excavations. Excavations in the Western City of Limyra (A. Dolea)2 During the 2022 campaign, the excavations took place between 2 August and 26 September. The aim was to continue the work within the large area opened in 2019 in the Western City (Dolea, 2020, pp. 222-228), and to obtain more information regarding the development within this part of the city from the Hellenistic to the Byzantine periods (Fig. 1). A total of 96 layers and remains of 18 structures were excavated and documented, which are summarised below3. 2 I express my gratitude to the following colleagues: B. Orakçılar, D. Üstünel, T. A. Özyazıcı, G. Çimen. 3 The documentation system for recording layers, contexts, constructions and finds are defined by unique individual acronyms and numbers, based on the area of the excavation, the year and the layer number. For example: LiPW followed by 16, 18, 19 or 22 translates to Limyra Polis West, namely excavations carried out in the Western The Excavations at Limyra (Lycia) 2022: Preliminary Report 207 Fig. 1 The excavation sector in the Western City of Limyra at the end of the 2022 campaign (A. Dolea, N. Gail, J. Kreuzer, B. Orakçılar © ÖAW-ÖAI). Water Channel This channel (Fig. 2, centre) was already partly revealed during the 2018 and 2019 campaigns (Dolea, 2019, pp. 233-234; Dolea, 2020, pp. 226227), but only this year we were able to document its fillings and construction. The channel has a north-east to south-west trajectory and continues beyond the southern margin of the trench. Here, the channel narrows considerably and, as far as we were able to observe, was constructed of horizontally placed blocks upon which two rows of vertical blocks rest, which were in turn covered by large horizontal blocks, thus creating a tunnel. North of the channel, several massive stone blocks have been documented and interpreted as being a bridge that (partly) covered this area in order to form a closed passageway. The preliminary data Fig. 2 Detail (drone) photo of the suggests that it was in use during the Roman era water channel, the square construction with and went out of function sometime between the the staircase and the stone pavement at the end of the 2022 excavation campaign end of the 3rd and the beginning or first half of the (A. Dolea, N. Gail, J. Kreuzer, B. Orakçılar 4th century (see the pottery analysis by Ph. Bes be© ÖAW-ÖAI). low). The discovery of this water channel further highlights water management issues that must have been a reality in ancient and Byzantine Limyra (Rantitsch, Prochaska, Seyer, Lotz & Kurtze, 2016, pp. 203-210; Dolea, 2020, p. 224; Dolea, in press). City of Limyra in 2016, 2018, 2019 or 2022. Variations to LiPW occur as follows: LiWT18 (Limyra Westtor – West Gate – 2018) or LiPO19 (Limyra Polis Ost – Eastern City – 2019). The layer follows the acronym and the year, for example: LiPW19-1054, which translates to Limyra Polis West (=Western City) 2019, layer 1054. 208 Martin Seyer et al. Square Construction Immediately east of the water channel, a square structure (ca. 5.5 x 4 m) was revealed already during the previous campaigns (Fig. 2, bottom right) (Dolea, 2019, pp. 233-234; Dolea, 2020, p. 227), but only this year we were able to finalise the work inside it. The western and southern walls were built most probably in the Hellenistic period, while the eastern and the northern walls were added in the Roman era. Inside, a huge fill was excavated in 2018 and 2022, which was deposited in the second or, at the latest, the third quarter of the 3rd century (see Ph. Bes’ contribution). The function of this structure is still unclear, but we have some understanding of how the structure was transformed during the Roman era. The east wall incorporates parts of an older inscription (spolia) dating back to the 2nd century, which mentions Affarūs honouring her late father, Ornimythos IV (Quatember, Leung & Wörrle, 2020, pp. 321-326). Based on the excavated materials inside the square structure, the pottery provides a terminus ante quem for the construction of the eastern and northern walls in the second, the latest the third quarter of the 3rd century, which thus suggests that parts of the monumental inscription were spoliated not later than the third quarter of the 3rd century. Staircase and Pavement The staircase was already partly documented during the 2019 campaign (Dolea, 2020, p. 227). This year we were able to reveal the whole structure and its connection to a paved open space, most probably a public square (Fig. 3). The staircase has four steps (max. dim. 4.95 x 2.44 x 0.175 m) and is accessible from the east, north, and west, and connected a large square building – most probably a propylon (excavated during the 2019 campaign: Dolea, 2020, p. 225, fig. 5; note the massive foundation immediately west of the water channel) with the paved open space (public square) that continues to the east until the Byzantine lime kiln. The stairs and the public square were presumably built in Roman Imperial times. Fig. 3 General view of the water channel, the square construction with the staircase and the stone pavement that was revealed also further to the east, where the Byzantine lime kiln and the public eating space were uncovered (A. Dolea, N. Gail, J. Kreuzer, B. Orakçılar © ÖAW-ÖAI). The Excavations at Limyra (Lycia) 2022: Preliminary Report 209 Public Eating House This building, which is situated to the east of the square, was partly excavated during the 2016 (Dolea et al., 2017, pp. 56-59; Dolea, 2017, pp. 144-152) and 2018 (Dolea, 2019, pp. 292-233) campaigns (Fig. 3, right). The uncovered assemblages of arte- and ecofacts indicate that the building functioned as a public eating house (Forstenpointner, 2020, p. 241 and the contribution of Forstenpointner below). Therefore, we decided to further investigate the area for a better overview concerning chronology and functionality. The processing of the discovered material correlated to the stratigraphy and construction phases, will clarify for how long the public eating space was in use. The preliminary data suggest that the place functioned during the late 5th until the end of the 6th-beginning of the 7th century, when the area was transformed into a proto-industrial district (Dolea, 2020, pp. 227-228). Conclusions In the 7th century, this area underwent a dramatic change from urbanistic and functional points of view. A massive lime kiln (ca. 7 m ext. diam.) was installed on the public square and most of the surrounding buildings went out of function or were transformed (Dolea, 2019, p. 233), presumably in relation to the installation of the lime kiln. Furthermore, new data points to the existence of primary iron production in the vicinity of the lime kiln (Schwarcz, 2020, p. 236; see Schwarcz’s contribution below). The excavation campaign of 2022 brought to light many relevant structures and layers from the Hellenistic until the Byzantine era, with a focus on the Roman Imperial to Early Byzantine periods. The functional variation within the excavation zone is also highly significant for the city life of Limyra and its inhabitants throughout its history (water management, public spaces, proto-industrial activities, etc.). Furthermore, this year’s campaign brought us more information regarding the diachronic use, reuse, and repurposing of structures and features revealing how the urban fabric changed throughout antiquity and the Byzantine era. Pottery Study in Limyra in 2022: An Overview (Ph. M. Bes) In 2022, the study of the pottery from the 2016 and subsequent excavation seasons, including those carried out this year, was resumed – matters such as methodology and consulted literature are summarised in the report of 2019 (Bes, 2020a). The framework4 in which this took place aims to (1) contribute chronological and other information to the understanding of the architectural remains and stratigraphy; and (2) build a ceramic sequence that spans the 3rd to the 7th/8th centuries, key aspects therein being change and continuity regarding typology and provenance. The pottery study5, first of all, concentrated on finishing two large layers – their excavation was completed this year – which were already preliminarily explored in 2018 and 2019 (Bes, 2020a, 4 This research was funded in whole by the Austrian Science Fund (FWF), Grant Number Lise Meitner M 3170-G. Dating of pottery relies on fabric and typological classification, the pottery’s taphonomy and fragmentation as well as a layer’s stratigraphic position. Other sources of chronological information, such as the rich numismatic evidence that has been found, are under study and will be welcomed to complement the stratigraphical and chronological analysis. 5 Image documentation comprised profile drawings and artefact photography. Unless otherwise indicated, Ioana Potra (Cluj, Romania) made all profile drawings – all drawings were digitised by the author. Judith Kreuzer (ÖAWÖAI) photographed selected artefacts. Both are warmly thanked for their efforts and commitment. 210 Martin Seyer et al. pp. 232-233); this pottery is significant as it broadens the diachronic picture in considerable detail. Second, the pottery from all layers that were excavated in 2019 was fully studied; the majority of these layers, however, contained mixed to highly mixed pottery that usually ranged from Classical to Ottoman times. Third, several key layers excavated during the 2022 excavation campaign were studied, which will continue in 2023. Then, research to improve the typological and chronological knowledge of the regional cooking and kitchen wares, the “lyciennes kaolinitiques” (LycKaol henceforth); this will also be discussed in detail immediately following this pottery report. Furthermore, another regional fabric (or fabric group), Fabric 2, was also documented in greater, in particular typological detail, emphasising the centuries-long importance also of this second category for the inhabitants of Limyra. Last, samples representing a broad variety of both regional and imported functional ceramic categories were selected for laboratory analysis6. Broadening Horizons In the stratigraphic record of the excavations that have been carried out since 2011, layers datable to Middle/Late Hellenistic to Middle Roman times (200 BC-300/400) remained elusive. Yet, architectural and epigraphic evidence, as well as coins and pottery fragments that turn up as residual items in younger layers, testify to the existence of an urban settlement. This major chronological gap of close to six centuries can now be partly filled in: 1. A small layer (LiPW19-10547) contained a mixture of mostly Middle to Late Hellenistic pottery – all well beyond secondary deposition – that was deposited sometime during 130/125100/90 BC (for comparanda from Ephesos: Gassner, 1997, pp. 69-71, 112-113; Ladstätter, 2003, pp. 26-41)8. Key are (a) two non-joining fragments of a Hayes Form 22A in Eastern Sigillata A with profiled base and interior rouletting (Slane, 1997, pp. 309-310, pl. 17, no. FW 179) – the second, larger fragment (Fig. 4a) was found in layer LiPW22-1075 which was located beneath LiPW19-1054, both layers presumably representing a single deposition; (b) fragments of an Ionian grey-ware mouldmade bowl with the common motif of meanders and rectangles containing an X (e.g. Gassner, 1997, pp. 77-78, Tafeln 15-17, 84-85, nos. 218, 223-224, 226-227); (c) fragments of a presumably Ephesian white-grounded lagynos; and (d) fragments, including the nozzle, of an Ephesian grey-ware lamp. 2. The preserved, lower courses of a square structure (see the section Square Construction above) held a particularly large layer (LiWT18-1024): it contains 12,030 pottery fragments that in total weigh just over 207 kilograms. The pottery was fragmented and, as quantities of Classical and Hellenistic pottery indicated, strongly mixed. The presence as well as absence of certain categories indicate that this material was deposited sometime in the second or, at the latest, the third quarter of the 3rd century – there are strong similarities with an important foundation fill (Lemaître, 2007, pp. 203-223) that was dated to the late 5th-early 6th century, which will be further explored. Only a single, small rim fragment of a Hayes Form 50A in African Red Slip Ware was attested, otherwise common are Eastern 6 At the time of writing, thin-sections as well as the results of Neutron Activation Analysis (NAA) of previously exported samples were available but were not yet being studied – their analyses and interpretation also take place within the context of the project mentioned in footnote 4. 7 See footnote 3 for details and meaning of the recording system. 8 Similarly dated contexts: respectively, a large context dated to the last third of the 1st century BC and, similarly, a well fill that is dated to 100 BC (110-90 BC?). The Excavations at Limyra (Lycia) 2022: Preliminary Report 211 Fig. 4 Selected fragments: (a) Eastern Sigillata A, Hayes Form 22A, from layer LiPW22-1075; (b) Zeest 72, from layer LiWT18-1024; (c) African Red Slip Ware, Hayes Form 50 (transitional A-B), from layer LiWT18-1025; (d) and (e) Sagalassos Red Slip Ware, Form 1C200 and Form 1B191 respectively, both from layer LiWT18-1024 (Ph. Bes © ÖAW-ÖAI). Sigillata A and D and Sagalassos Red Slip Ware (Fig. 4d-e), while Eastern Sigillata B and C (the latter was presumably manufactured at Çandarlı/Pitane) are represented by small quantities. A wide range of imported amphorae is attested that includes Zeest 72 (Fig. 4b), Agora G 199 (in multiple fabrics), Kapitän II and Pompeii V, among others, and whilst Late Roman Amphora 1 (LRA 1) is common in Limyra from the 4th century onwards (Bes, 2020b, pp. 386, 403, Table 1), neither its predecessors nor prototypes were clearly observed9. Oil lamps are almost exclusively represented by moulded specimens with decorated discs in light-coloured fabrics; presumably present among these are one or a few specimens that were manufactured at Corinth and Athens (Fig. 5) – some base fragments, in fact, preserve parts of names10 – as well as at Tlos (Uygun, Özdemir & Korkut, 2023). Whilst several types Fig. 5 Selection of oil lamp fragments, from layer LiPW22-1017 (J. Kreuzer © ÖAW-ÖAI). 9 One or at most a few fragments may be classified as late Pompeii V or proto-LRA 1. Also, see the channel fill discussed in this contribution. 10 Joseph Rife (Vanderbilt University, Nashville, United States) generously shared his knowledge on these oil lamps. 212 Martin Seyer et al. of pans manufactured in Phokaia were noted – some, however, do not obviously resemble the Phocaean fabric and may originate from one or more other sources – southeast Lycian LycKaol make up the bulk of the cooking and kitchen vessels, which indicates that by this time the workshops that manufactured these vessels were well-established. In spite of its rather poor preservation, the pottery from this layer provides us with a rich and important signal regarding the categories that were in circulation and in use in Limyra particularly during the 1st/2nd to mid-3rd centuries. 3. Part of a channel (see the section Water Channel above) contained a large fill (mostly layers LiWT18-1021, LiWT18-1025, LiPW22-1008 and LiPW22-1017; various smaller layers from the 2022 excavations await study). This channel likely was part of ancient Limyra’s water infrastructure; water presented the city with considerable challenges (Rantitsch, Prochaska, Seyer, Lotz & Kurtze, 2016). This fill was rich in rather well-preserved pottery and hence suitable for detailed documentation and analysis by means of Minimum Number of Vessels11. Matching fragments from thirteen vessels were found in multiple layers throughout this fill – a number that is expected to increase with further study – which combined with the general preservation indicates that much of this pottery is in secondary/tertiary deposition: no complete and/or fully restorable vessels have yet been observed. A number of vessels was damaged in a rather severe fire, indicated by (1) blackening on the surfaces and often also the breaks, which suggests the heat/fire continued after these vessels cracked; and (2) small sets of joining fragments belonging to closed vessels that have unusual, wavelike breaks (Fig. 6). Even if no traces of such a fire have been identified in the current excavation zone, the preservation and matching fragments strongly suggest that this fire took place in rather close proximity to the channel. Early versions if not in fact so-called protoversions of LRA 1 (Marchand, Reynolds & Le Bomin, 2022), dozens of fragments belonging to multiple vessels in African Red Slip Ware, all of Hayes Form 50 except one (Fig. 4c), fragments of a 3rd-century Almagro 51C (Fig. 7)12, a largely restorable Gauloise 4-style amphora Fig. 6 Fragments of various closed vessels, from layer LiWT18-1025 – note the unusual, wavy breaks (Ph. Bes © ÖAW-ÖAI). Fig. 7 Base and wall fragments of an Almagro 51C, from layer LiPW22-1017 (J. Kreuzer © ÖAW-ÖAI). 11 Some younger, mostly single fragments – e.g. an African Red Slip Ware base, presumably Hayes Form 99 – in the upper part of this large fill are regarded as modest contamination that occurred in Late Antiquity or later. No intrusive fragments were observed in the main body of the fill. 12 I express my gratitude to Inês Vaz Pinto (CEAACP-Universidade de Coimbra, Portugal), Horacio González Cesteros (Universidad Complutense Madrid, Spain) and Paul Reynolds (Universitat de Barcelona, Spain) for confirming this identification (from a photograph). The Excavations at Limyra (Lycia) 2022: Preliminary Report 213 [not in (any of) the typical fabric(s)], but also the complete absence of Late Roman C and D for example, suggests that this channel was filled in sometime during the late 3rd if not the first quarter/half of the 4th century. Typology and Chronology of Southeast Lycian Cooking and Kitchen Ware: the “lyciennes kaolinitiques” For their cooking and kitchen requirements, throughout the 3rd to 7th century, the inhabitants of Limyra almost exclusively made use of regionally manufactured vessels in so-called LycKaol (Lemaître, 2007, p. 214; Lemaître et al., 2013; Bes, 2020b, pp. 382-384, 399-401, figs 6-10). Not a single scrap of primary archaeological evidence of its manufacture has been found, yet, archaeometric clues as well as its abundance in any 3rd- to 7th-century layer indicate that the origin of LycKaol almost certainly needs to be sought in southeast Lycia, possibly in the environs of Limyra proper. Vessels in LycKaol were widely distributed particularly during Late Antiquity – though it appears in modest quantities – and are found for instance at Sagalassos (Bes accepted), Paphos (Hayes, 2003, pp. 511-512, fig. 35.376) and as far away as Vigo on Spain’s Atlantic coast (Fernández, 2014, pp. 363-365, fig. 188). This underlines that this aspect of the southeast Lycian economy had its (small) share in regional and long-distance exchange. While the typological and chronological parameters of some of the main types have been outlined (Yener-Marksteiner, 2019), the morphological in casu typological repertoire is very extensive, itself a further indication that LycKaol was presumably manufactured in rather close proximity to Limyra. Besides some rare type-variants, 54 distinct types could already be defined in 2022, with more to follow. These represent a variety of shapes and functions, ranging from chytra-style vessels to lids (which are notably rare) and lopades, as well as cooking bowls with horizontal handles that plausibly were influenced by southern Levantine morphological traditions. Chronologically, already in the 3rd century vessels in LycKaol make up the bulk of the cooking and kitchen vessel repertoire (Fig. 8) (cf. supra), which shows that the manufacturing and distributing frameworks were fully functioning. Interestingly, the Late Hellenistic layers (LiPW19-1054 and LiPW22-1075) highlighted above contained a handful of fragments that macroscopically show strong similarities with the LycKaol, albeit having an admixture of larger inclusions and thus appears coarser. Whilst this certainly requires further research, in particular laboratory analyses, this could suggest that the LycKaol appeared around that time. Indeed, a Late Hellenistic appearance has been suggested elsewhere (Arqué, Lemaître & Waksman, 2012, p. 143)13. Fig. 8 Selection of cooking vessels in “LycKaol”, all from layer LiWT18-1024 (Ph. Bes © ÖAW-ÖAI). 13 Fragments have been observed at Xanthos, dated to the “fin de l’époque hellénistique”. 214 Martin Seyer et al. In order to improve typological and chronological knowledge of the LycKaol, and eventually to be able to use it as a dating element in more and better detail, a seriation program was initiated. A first set of raw data was obtained by documenting 521 individual fragments. These data are being explored preliminarily (see 4 below) which nevertheless shows the potential of this line of research. Data collection will therefore continue during the next fieldwork campaign. A Second Regional Fabric: Fabric 2 A second fabric (Fabric 2 in the Limyra fabric classification) of quantitative importance that is regarded as a fabric of regional, southeast Lycian origin is discussed elsewhere in detail (Bes, 2020b, pp. 384-385, 401, figs. 11-12). It is recognised in a broad array of functional shapes – lekanai, pithoi and pithoid vessels, one-handled jugs, thick-walled bowls (mortaria?), imbrices, waterpipes (Fig. 9b), etc. – whose common presence illustrates these were widely used by the people in Limyra (Fig. 9). One specific shape was recently discussed in detail based on evidence from Arykanda, a “table amphora” labelled type AF1 (Yaman, 2022; also Lemaître, 2007, p. 215, fig. 7.1-5). The Limyra and Arykanda specimens appear to share the same fabric – i.e., Fabric 2 in the Limyra classification. Combined with the common occurrence of this shape at both sites, and the fabric in Limyra, this fabric (group?) can quite confidently be regarded as a regional, southeast Lycian fabric, as Yaman also suggests for his type AF1 (Yaman, 2022, pp. 325, 327, figs 4-5). Rim diameters of such ‘table amphorae’ from Limyra that were thus far measured range from 7 to 14.9 cm, a much broader range than that observed for the specimens from Arykanda (7-10 cm; Yaman, 2022, p. 325). More measurements are required to test the current idea that sizegroups existed – or at least that diameters concentrate around certain values; different sizes could have fulfilled different functions and/or held different contents. Furthermore, the two specimens illustrated (Fig. 9a, c) make clear that there are typological differences – these are, at present, considered as type-variants rather than as distinctly different types. Also, the handle profile of the smaller specimen (Fig. 9c) bears some resemblance to Cretan as well as Roman Imperial Knidian (Pompeii 38) amphorae, whereas the rim appears more ‘regional’, southeast Lycian, in conception14. Whilst no complete vessel profile has been restored so far, these vessels were almost certainly equipped with rather high ring bases (Fig. 9d), which also occur in a wide range of sizes and diameters. Their use as table amphorae – taken literally – might be considered for the smallest specimens (note the smaller range in diameters observed for the Arykanda specimens). In my opinion, however, they are more appropriately regarded as transport vessels for the intraregional distribution and storage of agricultural produce – the occasional find beyond southeast Lycia notwithstanding (Hayes, 2003, pp. 477-478, fig. 17.169). While ceramic lids that could be associated with these vessels were not identified, such lids may in fact not have been necessary: while the mouth could have been sealed with a lid or rather stopper made of an organic material, the omnipresent flange on the exterior rim created a large groove below, which would (theoretically) have allowed to cover the mouth with a ‘lid’ made for instance of textile or leather, wrap a piece of rope or string around it, fix this within this groove, and tie it. 14 Paul Reynolds observed (pers. comm., 16 February 2023) that the larger specimen (Fig. 9a) is not unlike late variants of amphora type Pompeii V – I thank him for this relevant observation. The Excavations at Limyra (Lycia) 2022: Preliminary Report 215 Fig. 9 Examples of vessels in Fabric 2 showing some of its functional variety: (a) and (c) amphorae, both from layer LiWT18-1025; (b) waterpipe, from layer LiPW22-1017; (d) ring base of a closed vessel, from layer LiPW22-1017 (I. Potra, Ph. Bes © ÖAW-ÖAI). Yaman dates this vessel type to 375-450 based on contextual evidence and considers that their manufacture halted soon after. Yet, rims of this vessel type are common in layer LiWT18-1024 and throughout the channel fill (cf. supra), which indicates that they were already manufactured and played a significant role during the 3rd and 4th centuries. If future evidence corroborates a chronological scenario in which this type was manufactured/in use during the period 200-450, then it looks as if for the remainder of Late Antiquity southeast Lycia ceased to have a tradition of amphora manufacture for regional use. Indeed, all amphorae in Late Antique layers from Limyra appear to have been imported from beyond the region (Bes, 2020b, pp. 385-386). While this seems significant in itself, it also raises questions regarding the organisation and (re)orientation of the regional agricultural economy – an absence of regionally-manufactured amphorae can hardly have signified an altogether absence of the production and circulation of agricultural produce – particularly changes that southeast Lycia may have gone through during the 4th and 5th centuries. If future research finds support for these various observations, this amphora provides an interesting parallel with as well as contrast to Sagalassos, where the manufacture of local amphorae only began sometime in the mid- to late 4th century. The initiation of the Sagalassos Amphora is considered an indirect effect following the founding of Constantinople and the ensuing gravitational shifts in political, military, economic and other matters, and the effects on the local/regional level (Poblome, Corremans, Bes, Romanus & Degryse, 2008; Corremans, Poblome, Bes & Waelkens, 2010; Bes & Vanhecke 2015; University of Southampton, Sagalassos). This may suggest that southeast Lycia and the environs of Sagalassos followed different trajectories, which in turn may have tied in with regional differences in agricultural production and organisation against a broader regional-historical background. 216 Martin Seyer et al. Similarity Matrices of Lyciennes Kaolinitiques from Limyra as a Means of Seriation and Chronological Modelling (D. Kafetzaki, Ph. M. Bes) Introduction As mentioned in the pottery report (cf. supra), during the 2022 fieldwork campaign considerable attention was paid to southeast Lycian cooking and kitchen vessels, the LycKaol15. In Limyra, these dominate in layers dated from the 3rd to the 7th century, while it almost certainly appeared (much) earlier, possibly as early as the Late Hellenistic period (cf. supra). Its archaeometrical characterisation and aspects of its regional significance have been previously explored, yet in spite of this, a detailed typological and chronological study is lacking. This is regrettable, as both its dominance as well as observed diachronic changes in its typological repertoire provide elements for constructing a framework, allowing the LycKaol to be used as an important tool in the chronological and functional analysis of stratigraphic layers from Limyra and other sites in southeast Lycia, more than is now feasible. To help address this matter, in 2022 a start was made with data collection for use in seriation analysis (Table 1). The aims of this data are (1) to perform seriation analysis on the collected, layer-based data for creating a relative chronology of types and assemblages; (2) to transform this into an absolute chronology using stratigraphic and artefactual information; and (3) to document and present the typological and chronological repertoire of LycKaol from Limyra. In terms of methodology, each sherd is numbered individually (being a unique identifier) and the following parameters are recorded: layer; type-identification; exterior and interior colour (descriptive, i.e., not using a coding system such as Munsell); rim diameter (in mm and % preserved); weight (in grams); handle section (descriptive); handle width (in mm) and number of ridges on the outer handle. For constraints of space, this preliminary report explores only typefrequency. Data collection and recording as well as the topic’s basic parameters are considered to meet the conditions and assumptions that lie at the basis of seriation (Dunnell, 1970, pp. 307-311). Table 1 Quantities of recorded fragments in LycKaol per layer (Ph. Bes, D. Kafetzaki © ÖAW-ÖAI). Layer Number of Recorded Fragments Types 1 to 54 Variations & Uncertainties OT12 1039 38 2 PW19 1056 10 5 WT18 1021 53 1 WT18 1024 269 31 WT18 1025 105 7 Total 475 46 The data that was collected in 2022 comes from five layers (two are considered to belong to one stratigraphic unit, the ‘channel’ fill (cf. supra) which the analyses below allow to test), and while not all LycKaol fragments in each of these five layers have yet been recorded, the current sample 15 This collaboration is part of a two-year project, Pottery and Transformation in Roman and Byzantine Limyra, at the Austrian Archaeological Institute in Vienna; see footnote 4 above. The Excavations at Limyra (Lycia) 2022: Preliminary Report 217 nevertheless illustrates this study’s aims and potentials. These five layers come from four different findspots: the ‘tower’ and ‘channel’ fills (cf. supra); a fairly small, slightly mixed layer (LiPW19-1056) with pottery datable to 450-500/525; and part of a waste dump (LiOT12-1039) dated to 575-650 (Bes, 2020b, pp. 380-381). A shape is assigned a type-number when it is considered to differ morphologically from what is known so far – in other words, it is ‘new’. As such, the typology as it currently exists is preliminary and requires further study, organisation, and functional grouping. The reason for this approach is rather simple and practical: the variety of types is wide; in 2022 alone, 54 types (from the five layers mentioned) were documented. One hindrance is that fully restorable profiles from the excavations in question are rare, which prevents a fuller understanding of a number of shapes regarding their morphological grouping and function (for (more) complete profiles, see Lemaître, 2007; Yener-Marksteiner, 2007; Yener-Marksteiner, 2009; Yener-Marksteiner, 2019). True, one can only be certain that all typological variety is covered once all excavated pottery has been studied. This cannot be the aim, however, and a classification will be devised that can more easily accommodate new types. Quantitative Analysis Approach Data and Processing. Following the research questions and recording activities as outlined above, the collected data resulted in an abundance data matrix with n layers in rows, m types in columns and cell values equal to the sherd frequency of each type found in each layer. The quantities of the recorded profiles are summarised in Table 1. Types 1-54 were included in the analysis while variations of type 1 (1 var. a, 1 var. a/10, 1 var. a/5, 1 var. b) and uncertain labels were excluded16. Sherds of type 1 variations were found majorly in LiWT18-1024 (16 sherds) while one sherd was found in LiWT18-1025. Following the original data format, the cell values are transformed to relative abundances of types within each layer, representing the percentage frequency of each type, so that each row sums to 100. Similarity Computation. The dataset of relative abundances, henceforth ‘normalised dataset’, is used to compute both the layers’ similarity matrix and the types’ similarity matrix. Such similarity matrices indicate numerically how strongly each pair of layers or each pair of types is related, and are also useful for seriation applications, since (a) a strong similarity in a dataset intended for seriation implies chronological proximity; and (b) in some cases serve as input to seriation algorithms. Typically, similarity is quantified using either a similarity-based metric or a distance-based metric, though distance-based metrics can be numerically reversed such that they reflect similarity. Here, Spearman’s ρ is used to compute the association of types, while Robinson’s IA is used to compute the association of layers. Similarity-based metrics include, but are not limited to, correlation metrics. Such is the Pearson’s correlation coefficient which measures the strength of a linear relationship between two types. Another is Spearman’s rank correlation coefficient (Spearman’s ρ), which assesses the monotonic relationship between two numeric vectors, this relationship being either linear or non-linear. The metric is computed for the sample ranks17 of the values in the normalised 16 These are presently excluded mostly because of the small size of the dataset, even if such ‘variants’ potentially illustrate that more was happening typologically. Also, because of the preliminary character of the typology, fragments that have now been classified as variants might turn out to be ‘real’ types once I more contexts and data have been processed. 17 The average method is used to compute the sample ranks of the normalised data values for any type A. 218 Martin Seyer et al. data for any two types A and B in the n layers. Mathematically, it is given by the covariance of the rank variables over the product of the standard deviations of the rank variables, Spearman’s ρ = . A Spearman’s ρ value for a pair of types is equal to 1 when the relative abundances of the two types are maximally related monotonically, in other words, each relative abundance that is larger in a layer compared to the rest of the layers in one type, will also be larger in the other type. A Spearman’s ρ value equal to –1 indicates the opposite, namely, each relative abundance that is larger in a layer compared to the rest of the layers for one type, will be smaller in the other type. A Spearman’s ρ value that equals 0 signifies there is no relationship between the relative abundance for a pair of types. Archaeological applications of distance-based metrics traditionally include Robinson’s Index of Agreement (Robinson’s IA) (Robinson, 1951) Robinson’s AIXY = 200 - ∑|Xi – Yi|, with i ∈ {1, 2, …, m}, for two layers X and Y, based on all m types. The second part of the equation measures absolute differences in the layers collection and the first part of the equation subtracts the differences from 200, so as to represent similarity. For Robinson’s IA, a value of 200 for a pair of layers indicates that the distribution of types in these two layers is identical, while a value of 0 indicates that the distribution of types is completely unrelated. Several distance metrics exist and could be applicable in this setting according to the data as well as any specifics of the research question. A traditional distance metric is the Euclidean distance, which likewise Robinson’s IA, could be reversed to represent similarity, Euclidean similarity = , where d(X,Y) = √∑(Xi – Yi )2, with i ∈ {1, 2, …, m}, is the Euclidean distance between layers X and Y, based for all m types. Euclidean similarity will take the value of 1 for identical layers and values smaller than 1 for layers with less similarity. Seriation Methods. In frequency seriation, using the Kendall model (Kendall, 1971a; Kendall, 1971b), the layers are expected to be arranged such that the relative abundances of types change over time following a pattern of growth-peak-decline (see Lund, 2018 for an archaeological case study) and competing with other types in a so-called battleship curve. The fact that the Kendall model does not permit fluctuations in relative abundances along with other limitations is discussed elsewhere (Banning, 2020). Using similarity matrices as input converts the seriation problem into a matrix permutation problem, however, this also does not guarantee a globally optimal solution (e.g., Brusco, Köhn & Stahl, 2008). Currently, several algorithms that deal with the normalised dataset or the similarity matrix as input are openly available, for instance in the R package ‘seriation’ (Hahsler, Hornik & Buchta, 2008; Hahsler, Buchta, & Hornik 2022). Although in principle every algorithm can provide a sequence for types or layers, to our knowledge, there is no systematic way in archaeological seriation to select the best applicable algorithm for the data under study. Such a research endeavour is part of our future research plans. Indicatively, Principle Component Analysis (PCA) (Pearson, 1901) is one of the algorithms to perform seriation using the normalised dataset as input, while a hierarchical clustering algorithm (GW) as developed by Gruvaeus and Wainer (1972) is proposed when using the similarity matrix. Both algorithms are used in our case to provide results. In order to visualise in two dimensions, the multivariate relationships in the collection of layers, PCA is performed using the function ‘prcomp’ from R package ‘stats’ (R Core Team, 2022), while visualisation of the PCA results uses the function ‘fviz_pca’ from the R package ‘factoextra’ (Kassambara & Mundt, 2020). PCA is not used for sequencing the layers but for mapping The Excavations at Limyra (Lycia) 2022: Preliminary Report 219 and interpreting the relationships between layers. Seriation is performed using the R package ‘seriation’ and specifically, the GW approach18 with Robinson’s IA to compute (dis)similarity between layers. The sequence is visualised and evaluated from algorithmic and archaeological perspective in the rearranged Robinson’s IA matrix. Similarity matrices are visualised using the R package ‘corrplot’ (Friendly, 2002). Typically, a larger number of layers would be needed to provide a solid seriation result. Nevertheless, archaeological reality often involves drawing preliminary inferences with limited sample sizes, at the same time helping to set out and/or adjust future data collection/research plans. Preliminary Results and Discussion Relationships between Types. Having obtained the normalised data, Spearman’s ρ matrix is computed. There are five layers for evaluating the monotonic relationships between types, therefore, a relationship that is stronger than 0.9 in magnitude can be considered different from zero in a 10% significance level. As visualised (Fig. 10), 12.65% (181/1431) of the pairwise relationships are found to differ from zero. All relationships between types are positive, except for one negative relationship which occurs between types 3 and 32. Types that exhibit a positive relationship are expected to be found in the same layers and to occur in layers of similar chronology, whereas types with a negative relationship are expected neither to be found in the same layers nor in layers of different chronology. Regarding positively related types – i.e., contemporaneity is supposed – five groups are observed that consist of types that occur in a single layer. Specifically, the first group includes types 29, 30, 33, 35, 36, 37, 38, 39, 40, 41, 42, 43 and 44 which are found in layer LiOT12-1039. Types 2, 4, 5, 13, 14, 15, 16, 17, 18, 19, 21 and 22 are also positively related and found only in layer WT18 1024. Similarly, for types 27, 45, 46, 47, 48 and 49 that are found in LiWT18-1025. Another group includes types 11, 23, 24, 25 and 26 where one profile of each type is found in LiWT18-1021. Finally, types 50, 51, 52 and 53 are positively related and one profile of each is found in LiPW19-1056. Next, positive relationships are observed between types that are found in two or more layers. Type 8, which is found mostly in LiWT18-1025 but less so in LiWT18-1021, is positively related to three other types, namely 6, 7 and 10. Type 32, which is found mostly in LiOT12-1039 and less in LiPW19-1056 and LiWT18-1025, is positively correlated with type 31, found mostly in layer LiOT12-1039 and less in LiPW19-1056, while it is negatively correlated with type 3, which is mostly found in LiWT18-1024 (12 rims) and less in LiWT18-1021 (2 rims) and LiWT18-1025 (1 rim). Finally, types 9 and 12 are positively related and are found in LiWT18-1021, LiWT181024 and LiWT18-1025 with rim counts of 13, 4, 22 and 4, 3, 5 respectively. There are also five types for which no relationship with other types could be established, namely, types 1, 20, 28, 34 and 54. Type 1 is the most abundant, with 116 rims found in LiWT18-1024 and 2 rims found in LiWT18-1025. Type 20 is also found in the aforementioned layers with rims counts being 2 and 1 respectively. Type 28 is found mostly in LiWT18-1021 (9 rims) while 1 rim is found in each of the layers LiPW19-1056 and LiWT18-1025. One example of type 34 is found in LiWT18-1025 and two examples in layer LiOT12-1039. Type 54 is mostly found in LiWT181024 (29 rims) while 1 rim is found in LiPW19-1056. 18 The seriated sequence is retrieved using the function ‘seriate’ specifying the method as ‘GW’. 220 Martin Seyer et al. Fig. 10 Spearman’s ρ matrix for recorded types. In the upper triangle, all values are present while in the lower triangle only the values higher than 0.9 are present (D. Kafetzaki © ÖAW-ÖAI). Revisiting this analysis after collecting more data will be useful if not in fact necessary for refining current as well as revealing additional relationships between pairs of types. For instance, a negative relationship between type 3 and type 31 could be tested in the future. Negative relationships could potentially also arise between type 31 and types 6, 9, and 12, or between type 7 and type 54. Relationships between Layers. To analyse the complete collection of types within the five layers, first, Robinson’s IA is computed and visualised (Fig. 11). Large values reflect similarity which signifies proximity in time, while lower values signify a contrast in the type collections attributed to layers and therefore their proximity in time. Layers LiWT18-1025 and LiWT18-1021 exhibit the largest similarity in their typological composition, and are therefore expected to be close in time with respect to the other layers. This can be corroborated by both the stratigraphic relation as well as the overall spectrum and preservation of ceramic finds in LiWT18-1021 and LiWT18-1025. The second largest similarity occurs between layers LiPW19-1056 and LiOT121039 which nevertheless have a considerably lower similarity index than the first pair of layers. As a result, they are expected to be close in time but their proximity will be smaller compared to the first pair. The typological composition recorded for layer LiWT18-1024 seems to be mostly unrelated to all other layers, having the ‘largest’ similarity index of 20/200 with LiPW19-1056, followed by 16/200 with LiWT18-1025 and LiWT18-1021. Whereas the small similarity index of LiWT18-1024 compared with the other four layers deserves further study, it nonetheless finds The Excavations at Limyra (Lycia) 2022: Preliminary Report Fig. 11 Robinson’s IA matrix for layers, rearranged according to the inferred chronological sequence (D. Kafetzaki © ÖAW-ÖAI). 221 Fig. 12 Graph of individuals in the first two components of PCA (D. Kafetzaki © ÖAW-ÖAI). support in LiWT18-1024’s overall ceramic composition, which more generally allows to see it apart from the other four layers (cf. supra). The observations resulting from Robinson’s IA matrix are in line with the PCA results. Plotting layers in the first two principal components together explains 85.28% of the total variance observed in the dataset (Fig. 12). Layers LiWT18-1025 and LiWT18-1021 are closer to each other than layers LiPW19-1056 and LiOT12-1039, while layer LiWT18-1024 is located in the middle and seems mostly unrelated to the other four layers. In the second principle dimension, LiWT18-1024 is closer to LiWT18-1021, whereas in two dimensions, LiWT18-1024 is closer to LiPW19-1056. This latter point, as mentioned, requires further study. Following, the GW seriation results are obtained based on Robinson’s IA matrix. The resulting reconstruction of the relative chronological sequence of the five layers is LiWT18-1024 – LiWT18-1021 – LiWT18-1025 – LiPW19-1056 – LiOT12-1039. Although different algorithms would produce different seriation results, GW places the layers in a sequence that is in line with the available ceramic information, with LiWT18-1024 being the oldest and LiOT12-1039 the youngest assemblage. A richer and more varied dataset as well as other configurations of the data, understanding of applied components that affect this interdisciplinary application, customisation of algorithms and parametrisations in line with archaeological theory, are needed to build on the understanding of the current results and to further explore this application. Referring back to Robinson’s IA (Fig. 11), when the matrix is rearranged according to the chronological sequence, higher cell values are expected to cluster close(r) to the diagonal while cell values further away from the diagonal should be lower/decrease. Although the proposed sequence is not optimal, it can also be observed that no other proposal would satisfy this premise. Continuing with the data collection during fieldwork on the one hand, and enriching the quantitative analysis approach on the other, are both essential for further researching the typological and chronological repertoire of LycKaol. 222 Martin Seyer et al. Metal Finds in Limyra (D. Zs. Schwarcz) In 2022, after two years interruption, the documentation and research on the metal finds from previous as well as ongoing excavations in Limyra continued. The database was updated with 309 new inventory entries (over 1900 objects in various states of preservation) from the excavation conducted in the Western City in 2022. Besides the usual domestic items (i.e., nails, knives, spindle hooks, styli, sewing needles, etc.), more than 20 kg of metalworking remains deriving from iron smelting – including slags and blooms – were found also this year (for the material evidence on primary ironworking, see Schwarcz, 2020a, pp. 219-263). Due to the intensive restoration campaign in this season, numerous fragments of bronze statues could be identified in the archaeological material of both the previous and the recent excavations. A few pieces can be determined as fragments of drapery; however, the rest is too small (approximately 5-10 cm in length) for an exact classification. Nonetheless, most of these smaller fragments bear traces of cutting as they were probably prepared for recycling. One of the primary research subjects in 2022 concerned the use of horse, donkey and mule shoes (types A, B and D: Gaitzsch, 2005, pp. 125-129) in Limyra in the post-Antique periods, based on the results of the excavations between 2016 and 2022. 35 such objects were selected for a detailed study19, which aims to further refine form-typological details in accordance with functional and chronological indications. Another important task was the natural scientific analysis of 40 objects by an Olympus InnovX Delta Premium 6000 handheld XRF spectrometre. The results of the analysed 25 slags, blooms and furnace fragments, all from various archaeological contexts of the excavation in 2022, support the previous macroscopical studies concerning primary ironworking in Limyra. Besides an Early Byzantine copper-alloy coin weight and two loop chains (so-called “Schlaufenkette”), the remaining 15 objects belong to the find group of jewellery and dress accessories. Two Early Byzantine belt buckles are worth mentioning: a buckle (type D4: Schulze-Dörrlamm, 2009, pp. 156-158; cf. a recent summary on this type of buckles proposing a refined typo-chronology as Class D4/1 or Class Olba, see Bollók & Tepper, 2021, pp. 455-461), already published in a previous report (Schwarcz, 2019, pp. 240-241) and a rectangular buckle (type A14: Schulze-Dörrlamm, 2009, pp. 2930; for another such buckle from Ephesos: Pülz, 2020, cat. no. T 9). For further analogies and information, see Pülz, 2020, p. 29), both made of leaded bronze (major element: Cu, minor elements: Pb, Sn; Fig. 13). Although only two buckles were studied, the analysis results provide additional information concerning tendencies in the use of different copper alloys during the Early Byzantine period (Schwarcz, 2020b, pp. 293-294)20. The complete Fig. 13 Limyra Western City (2022), front and back side of an Early evaluation of the natural scientific data is planned for Byzantine leaded bronze rectangular the final publication of the metal finds from the recent buckle analysed by portable XRF excavations in Limyra. (P: point of measurement; D. Schwarcz © ÖAW-ÖAI). 19 The research topic is studied in collaboration with G. Forstenpointner, A. Dolea and Ph. Bes. 20 Concerning a possible change in the use of different copper-alloys (i.e., brass and bronze) in the Eastern Mediterranean between the Early and Middle Byzantine periods, based on the XRF-analysis of metal objects from Ephesos. 223 The Excavations at Limyra (Lycia) 2022: Preliminary Report Faunal Remains from Areas Polis West (PW) and West Gate (WT) – Preliminary Results (G. Forstenpointner) Introduction While archaeozoological analysis in the frame of the project “Urbanistic Development of Limyra in the Hellenistic Period” initially had focused on faunal remains from the building near the lime-kiln (Forstenpointner, 2020; see also for a short overview on available reference data), a two-week research campaign in August 2022 yielded new evidence on faunal exploitation in the settlement of Limyra from Late Classic until Early Byzantine times. Several samples are too small to allow neither statistical testing nor reasonable interpretation, but a larger assemblage from the sondage at the West Gate, dating to the 3rd-4th centuries (Table 3) and a smaller sample from Early Byzantine layers from the sondage Polis West (6th-7th centuries; Table 4) deserve closer examination. Early to Middle Hellenistic layers from the West Gate (Table 2) yielded only a small amount of faunal remains, barely interpretable, yet to be presented anyway. In the course of this preliminary report, rough descriptions of the mentioned samples will be given in chronological order, highlighting particular finds. Synoptic considerations will be the task of a comprehensive study on all significant archaeozoological findings of the project. Table 2 Faunal remains from LiWT18-1019. Abbreviations: B – cattle; O-C – sheep/goat; O – sheep; C – goat; S – domestic pig; Ur - brown bear; m – medium-sized mammal; l – large sized mammal; NISP – Number of Identified SPecimens; NSP – Number of SPecimens (G. Forstenpointner © ÖAW-ÖAI). WT Early-Middle Hellenistic B O-C O S Ur m l Total Head cranial fragment 1 Mandibula 4 Dens sup. 3 Dens inf. 4 Total 12 0 1 1 0 27.3 0 100 16.7 0 NISP species % 1 1 5 3 1 5 Trunc Vert. cerv. 1 Vert. thor. 3 Vert. lumb. 1 3 1 1 Costa 8 Costa cartil. 2 Coxa 1 Total 15 1 0 1 0 34.1 12.5 0 16.7 0 NISP species % 1 9 2 1 Legs Scapula 1 Humerus 3 2 Radius 1 2 3 Femur 4 2 6 Tibia 5 Total 14 6 0 2 0 31.8 75 0 33.3 0 NISP species % 1 2 1 6 5 224 Martin Seyer et al. WT Early-Middle Hellenistic B O-C O S Ur m l Total Auto-/Metapodials Metacarpus 1 1 2 Metatarsus 3 Total 3 1 0 2 0 6.8 12.5 0 33.3 0 NISP species % 1 4 Toes Phalanx prox. ant. dig. I 1 Total 0 0 0 0 1 NISP species % 0 0 0 0 100 1 Unidentified long bone 5 compacta Total NSP % Total NISP % (NISP=60) 44 8 1 6 1 73.3 13.3 1.7 10 1.7 8 13 4 4 5 12 17 6.5 15.6 22.1 5 12 77 Table 3 Faunal remains from LiWT18-1021, 1023, 1024, 1025. Abbreviations: B – cattle; O-C – sheep/goat; O – sheep; C – goat; S – domestic pig; Ea – donkey; Ee - horse; Ss – wild boar; Ce – red deer; Ur – brown bear; Lu – Eurasian otter; Le – hare; Er – hedgehog; m – medium-sized mammal; l – large sized mammal; NISP – Number of Identified SPecimens; NSP – Number of SPecimens (G. Forstenpointner © ÖAW-ÖAI). WT 3rd-4th centuries B O-C O C S Ea Ee Ss Ce Ur Lu Le Er m l Total Head antler 2 cranial fragment 20 4 Dens sup. 8 6 Dens inf. 1 6 1 5 6 5 12 Dens M ind. 16 Stylohyoid 1 NISP species % 1 68 26 3 3 1 1 26 1 Mandibula Total 37 2 16 2 8 1 29 1 72 1 46 26 3 24 85 16.7 15.1 15.8 34.8 54.8 3 4 100 100 2 2 1 0 0 0 40 50 100 0 0 0 Trunc Atlas 1 Axis 1 2 2 2 Vert. cerv. 8 1 1 10 Vert. thor. 10 2 1 13 Vert. lumb. 7 5 3 15 Costa 41 25 21 87 Coxa 15 3 3 25 Os sacrum 1 4 1 225 The Excavations at Limyra (Lycia) 2022: Preliminary Report WT 3rd-4th centuries B O-C O C S Ea Ee Ss Ce Ur Lu Le Er m l Total Trunc Total 83 38 0 4 30 0 0 0 0 0 0 0 0 30.2 22.1 0 5.8 19.4 0 0 0 0 0 0 0 0 Scapula 35 12 1 3 2 Humerus 20 14 1 6 11 Radius 22 19 2 3 2 Ulna 5 6 Femur 15 16 Tibia 10 25 2 5 6 Total 107 92 6 19 32 0 0 1 1 0 1 0 1 NISP species % 38.9 53.5 31.6 27.5 20.6 0 0 20 25 0 100 0 100 NISP species % Legs 53 1 53 1 49 5 2 16 6 1 1 41 48 Auto-/Metapodials Talus 4 1 5 2 Calcaneus 3 2 1 1 Tarsus 1 Metacarpus 5 5 4 7 1 Metatarsus 13 11 2 8 1 13 1 8 1 22 1 35 Metapodium 3 Total 29 16 9 21 5 0 0 1 1 0 0 1 0 10.5 9.3 47.4 30.4 3.2 0 0 20 25 0 0 100 0 1 2 NISP species % 1 4 Toes Phalanx prox. 4 Phalanx med. 2 Phalanx dist. 4 1 7 1 1 5 4 Total 10 0 1 1 3 0 0 1 0 0 0 0 0 NISP species % 3.6 0 5.3 1.4 1.9 0 0 20 0 0 0 0 0 Unidentified cranial fragment 53 9 62 Vertebra 2 3 5 3 3 Coxa long bone 67 66 133 compacta 64 123 187 spongiosa 11 3 14 Total 197 207 404 NSP % 17.7 18.6 36.3 197 207 1114 Total 275 172 19 69 155 3 4 5 4 1 1 1 1 NISP % (NISP=710) 38.7 24.2 2.7 9.7 21.8 0.4 0.6 0.7 0.6 0.1 0.1 0.1 0.1 Materials and Methods Faunal remains had been sampled consistently by means of manual picking as well as dry-sieving (mesh size 2.5 mm). While only a few specimens showed signs of corrosion or sinter, the state of preservation was primarily good. Determination was carried out on-site, quantification based mainly on counting the Number of Identified Specimens (NISP). Calculation of MNI (Minimum 226 Martin Seyer et al. Number of Individuals) tends to level the proportions of the taxonomic sample composition by exaggerating the frequency of actually infrequent species (Lyman, 2008), however, MNI provides a valuable tool in order to highlight selective patterns of the assemblage that might enable the recognition of specific functions. All analytical steps obeyed the methodical rules of archaeozoological research. Estimation of culling ages was based on rating the wear stages of teeth (Payne, 1987, modified by Hongo, 1997) as well as the state of epiphyseal closure (Noddle, 1974; Habermehl, 1975). Osteometry was performed according to the standards of Von den Driesch (1976). Data recording, including significant bone modifications like corrosion, butchering, burning or gnawing marks was accomplished by means of an appropriately adapted input mask that is based on the software package PASW Statistics which allowed the direct application of all relevant statistical assay methods. Results WT – Early to Middle Hellenistic Finds. LiWT18-101921 Doubtlessly, the composition of the very small Hellenistic sample is highly subject to taphonomic coincidences (Table 2). On the other hand, the striking predominance of cattle bones that cover almost three quarters of the assemblage, representing all body parts of several individuals, appears more than an elusive trend. Two of these finds show particular butchery marks (Fig. 14), characterised by shearing off planar splinters of the bone surface that is associated with the removal of preserved meat, which became firmly attached to the bone in the course of smoking or salting (Lauwerier, 1988). The occurrence of “chipping” in a Hellenistic context appears surprising as this bone modification is widely assigned to Roman meat processing. However, together with the extraordinarily high density of cattle bones, this feature might indicate the functional assignment of the sample as butchery waste, apart from domestic garbage. A singular, nevertheless very particular find is a proximal forepaw phalange of a small brown bear that most probably represents a bear’s skin. WT – Roman Finds (3rd-4th centuries). LiWT18-1021, 1023, 1024, 1025 Four stratigraphic units from the area West Gate yielded a considerably large sample that provides valuable insights into the consumption scape of Roman Limyra (Table 3). In comparison to available contemporaneous reference data (Galik, Forstenpointner & Weissengruber, 2012, table 1) a ratio of almost 40% cattle bones appears surprisingly high. However, again a frequent occurrence of “chipped” bones is discernible (Fig. 14) which proves true most notably in LiWT18-1024 with more than 10% corresponding cattle bones. Supposedly, the area of the West Gate housed a kind of abattoir for cattle butchering and processing since Hellenistic times, producing correspondent butchery waste. An isolated bovine horn core from LiWT18-1024 (Fig. 15) provides an impressive hint at Roman traditions of harnessing draught animals. A distinct impression on the front face of the core indicates the jacking point of the wooden yoke, doubtlessly causing consistent pain when dragging a heavy burden. The ratios of other economically relevant domesticates cover expectable ranges, with an elevated frequency of pig bones (21%) which is a typical feature of Romanised 21 See footnote 3 for details and meaning of the recording system. The Excavations at Limyra (Lycia) 2022: Preliminary Report Fig. 14 Cattle bones with “chipping” marks, indicated by arrows. Upper row, left and middle: femur (left) and humerus (middle) from Hellenistic layers; upper row, right: Costa from Roman layers; below: Coxa (Ilium) from Byzantine layers (A. Dolea, G. Forstenpointner © ÖAW-ÖAI). 227 Fig. 15 Left bovine horn core, dorsal aspect. Impression caused by yoke indicated by arrows (A. Dolea, G. Forstenpointner © ÖAW-ÖAI). settlements of the Eastern Mediterranean and the well-known Lycian ratio in ovicaprines of 8 goats to 2 sheep. LiWT18-1024 again produced a remain of brown bear (Fig. 16). A left lower fourth premolar tooth of a small individual appears worn and polished, probably this object served as a talisman, carried along in a pocket or wallet. Fig. 16 Brown bear, left mandibular 4th premolar tooth, buccal aspect (A. Dolea, G. Forstenpointner © ÖAW-ÖAI). PW – Early Byzantine Finds (550-650). LiPW18-1052, 1074 The small, yet acceptably significant Byzantine samples from the area Polis West quite closely resemble finds from contemporaneous strata in the thermal complex near the theatre of Limyra (Schuh, 2012a; Schuh, 2012b) that had been addressed as “aberrant layers”, due to very low percentages of pig bones and a high frequency of carnivore gnawing marks (Forstenpointner & Gaggl, 1997) (Table 4). Both parameters contrast highly with the respective findings of other over- or underlying Byzantine layers and have been associated with alimentary taboos and more careless garbage management of an early group of Islamic invaders. The sample from Polis West comprises 6% of pig bones, and 5% of all remains show gnawing marks. In comparison, the Roman sample from the West Gate yielded 21% pig remains, while only 2.5% of all specimens appear to have been gnawed on. The effective percentage of domestic pigs might be even lower, as four teeth from LiPW18-1052 quite likely represent one individual. Therefore, a hypothetically similar generation of waste in the Byzantine layers of Polis West appears likely, however, this hypothesis requires thorough further studies. 228 Martin Seyer et al. A very small sample from Early Byzantine layers in the area West Gate (LiWT18-1018) yielded 25 identifiable specimens, too small to allow statistical analysis. Nevertheless, two observations deserve being mentioned at least. On the one hand, the sample comprised only one pig bone, which might be very easily caused by taphonomic chance. On the other, two cattle bones showed “chipping” marks (Fig. 14), therefore probably indicating continuous activities of butchery and meat processing near the West Gate from Hellenistic up to Byzantine times. Table 4 Faunal remains from LiPW18-1052, 1074. Abbreviations: B – cattle, O-C – sheep/ goat; O – sheep; C – goat; S – domestic pig; Cn – domestic dog; Ss – wild boar; Cd – fallow deer; m – medium-sized mammal; l – large sized mammal; NISP – Number of Identified SPecimens; NSP – Number of SPecimens (G. Forstenpointner © ÖAW-ÖAI). B O-C 1 2 O PW Early Byzantine C Cn S 1 2 Ss Cd m l Total Head cranial fragment Mandibula 6 Dens sup. 6 1 4 1 6 1 10 1 7 Dens inf. 1 Dens M ind. 1 Total 3 22 1 7 0 5 2 1 9.7 28.9 50 53.8 0 62.5 50 100 NISP species % 8 1 4 17 1 Trunc Atlas 1 Axis 1 1 5 6 Vert. cerv. 1 Vert. thor. 1 Vert. lumb. 3 Vert. caud. 2 Costa 3 Coxa 1 Os sacrum 1 1 2 1 6 1 14 1 10 3 1 2 1 Total 1 11 21 0 0 1 2 0 0 35.5 27.6 0 0 100 25 0 0 Scapula 5 6 Humerus 1 8 NISP species % Legs Radius 11 1 10 3 Ulna 3 1 Femur 3 1 7 1 11 Tibia 3 4 1 1 Total 12 29 1 3 0 0 0 0 38.7 38.2 50 23.1 0 0 0 0 NISP species % 9 Auto-/Metapodials Talus 2 Calcaneus 2 2 1 Metatarsus 2 4 5 2 Metacarpus Total 1 3 1 1 3 2 0 3 0 1 1 0 229 The Excavations at Limyra (Lycia) 2022: Preliminary Report NISP species % B O-C O 12.9 3.9 0 1 1 PW Early Byzantine C Cn S 23.1 0 12.5 Ss Cd 25 0 m l Total Toes Phalanx prox. Phalanx dist. Total NISP species % 1 1 2 1 1 0 0 0 0 1 0 3.2 1.3 0 0 0 0 25 0 Unidentified cranial fragment 3 Scapula 3 13 13 long bone 38 11 49 compacta 28 31 59 spongiosa 1 3 4 Total 70 58 128 26.5 22 48.5 70 58 264 NSP % Total NISP % (NISP=136) 31 76 2 13 1 8 4 1 22.8 55.9 1.5 9.6 0.7 5.9 2.9 0.7 Additional Finds. Apart from mammal bones which formed the vast majority in all studied samples, also some avian remains and mollusc shells have been identified. The Roman sample yielded a total of 20 avian remains, comprising 14 chicken bones, 3 remains of the Chukar partridge and 1 specimen each of domestic goose, mallard and dove. The Byzantine layers produced 3 chicken bones and, surprisingly, a wing bone of a pelican. Marine resources played a very low or even no part in the alimentary supply of the settlement. Apart of 4 not identifiable fish bones the three mentioned assemblages only produced 11 marine mollusc shells, comprising Bolinus brandaris, Cerastoderma glaucum and Donax trunculus. As all shells are complete, these finds rather testify to a walk along the beach than to the remains of a sea-food lunch. Coin Finds from Limyra, 2016-2022 (N. Schindel) The excavations in the Western City of Limyra conducted between 2016 and 2022 have yielded almost 1,000 coins so far. An overview is given in Table 5, the distribution by periods from which the coins originate in Table 6 and Fig. 17. This overview contains all kinds of coins. After a preliminary attribution, about 40% of the total could be given to one of the historical periods listed below. Table 5 Number of coin finds per campaign year (N. Schindel © ÖAW-ÖAI). Campaign Year Number of Coins found 2016 255 2018 202 2019 273 2022 262 Total 992 230 Martin Seyer et al. Table 6 Number of coin finds per period (N. Schindel © ÖAW-ÖAI). 2016 2018 2019 2022 Total 5 4 1 6 16 6 4 2 12 Dominate 52 44 36 78 210 Byzantine 20 10 8 24 62 Islamic 9 Classical Hellenistic Principate 24 89 6 128 Medieval 1 1 0 2 Modern 1 0 1 116 431 Total 86 90 139 One has to emphasise that these results are provisional; the final count might differ a little bit, but the overall picture will not change much. Almost 50% of the coin finds from Limyra date to the Dominate, here defined as the numismatic period from the reign of Emperor Diocletian (285-305) until the demise of Emperor Zeno (474491). Also, Islamic coins were found in large quantities, especially during the 2019 campaign. A detailed analysis by rulers and mints will offer new insights into the economic history of the city. For the ancient periods, comparisons Fig. 17 Pie chart showing the number of coin finds per period (N. Schindel © ÖAW-ÖAI). with other excavations in Western Asia Minor are easily possible. Things are slightly different with the Islamic finds for which much fewer comparative data is available. While the main interest of coin finds lays in their value for statistical analysis (as indicated above), also individual specimens can be of interest. One example is a silver denarius struck by Emperor Domitian (81-96) (Fig. 18). Imperial precious metal coins of the Principate are not commonly found in Asia Minor. Another remarkable coin is a small copper nummus struck by Fig. 18 Denarius of Emperor Domitian (N. Gail © ÖAW-ÖAI; enlarged). Fig. 19 Nummus of King Athalaric (N. Gail © ÖAW-ÖAI). Fig. 20 Mangır of Beyazid I (N. Gail © ÖAW-ÖAI). The Excavations at Limyra (Lycia) 2022: Preliminary Report 231 the Ostrogothic King Athalaric (526-534) in the mint of Rome (Fig. 19). While Vandal coins from Northern Africa occur rather frequently in the Eastern Mediterranean region, Ostrogothic coins from Italy are much rarer. Finally, an example of an Ottoman copper coin of Bayezid I (13891402) is shown (Fig. 20). The spread of Ottoman base metal coins during the 14th century will be one of the many interesting topics of a detailed study of the Islamic coin finds from Limyra. Underwater Archaeological Studies in Limyra 2022 (C. Öztosun, H. Öniz, M. Seyer) Underwater archaeology studies have been carried out by the authors and their team in 2019 and 2022 under the directorship of Martin Seyer and Hakan Öniz. In 2019 the Roman colonnaded street south of the Ptolemaion was investigated, one of the nowadays submerged areas of the ancient city (Fig. 21). The second part of the work, which had to be postponed for two years, was started in 2022. During these studies, documentation of the submerged colonnaded street continued, though the major effort was the cleaning and documentation work at the south, west and east sides of the foundation of the Cenotaph of Gaius Caesar. It is the aim of this study, in the framework of the ancient city plan of Limyra, to better understand the roads and structures that are below the modern water level (Öztosun, Seyer & Öniz, 2020). Fig. 21 Colonnaded Street, looking northwest (H. Öniz © ÖAW-ÖAI). After having cleaned the vegetation and alluvial deposits at the colonnaded street important results were gained concerning its adjacent architecture with a roofed colonnade and several buildings behind it, most probably of mercantile character. Anyway, the main part of the work in 2022 was carried out around the Cenotaph of Gaius Caesar, who died in Limyra on Febuary 21st, 4. As many structures in Limyra this building is surrounded by groundwater nowadays. Extensive work at the Cenotaph had already been carried out in 1971-1974 and 1980-1981, including excavations, a detailed architectural survey as well as the graphic documentation of 232 Martin Seyer et al. the few surviving fragments of the building’s sculptural decoration, which were also published in an exemplary way (Ganzert, 1984; Borchhardt, 1993, pp. 85-97; Borchhardt, 2002). During these studies, the depositional layer was excavated with a special technique using a water pump, so that a depth of more than 3.80 m could be reached. The socle of the building consists of eight rows of stone blocks on the south side, which protrude the upper part of the building at 0.2 m. Therefore, there are two possibilities for the position of the Cenotaph’s foundation: according to the first one, the building was built on the debris cone and the foundation was put directly on the bedrock, which would mean that all the eight rows of blocks have to be regarded as the foundations of the building. The second option is that there was no debris cone at all and that the Cenotaph was built directly above the ground with one layer of blocks as the foundations and the remaining seven as part of the podium (Borchhardt, 1993, p. 92). Anyway, due to this great depth, it was not possible to focus carefully on the results and archaeological details. After an interruption of forty-one years underwater archaeology in this part of the ancient city was resumed in 2022. This new study is a resumption of, in a sense, a detailed understanding of the underwater studies carried out in the 1970s with the use of modern technologies. However, the first impression obtained from the area is that the water around the Cenotaph is covered with more filling material and that the water level has increased, as the comparison of old and new photographs suggests. This shows that a major part of the area studied in the 1970s and early 1980s was covered again by fillings, and despite regular clearing, numerous plants grow in this area again (Fig. 22). Fig. 22 Submerged area of the Cenotaph of Gaius Caesar (H. Öniz © ÖAW-ÖAI). The first work done during the underwater archaeology excavation in 2022 was the documentation of the current situation around the Cenotaph by the team members, from land and from below the water level. In addition, a photo-scanning technique was used. Due to the fact that the water around the Cenotaph is covered with plants and filled with debris and rubbish, this documentation was an important component for the underwater survey and underwater photography. For this reason, in order to facilitate underwater vision and to enable the imaging of the wider area around the building, cleaning of the underwater vegetation and the area around the Cenotaph was the first task undertaken. During the underwater work, a horseshoe and brick fragments from an obviously late deposit layer could be documented. After having completed this, photo scans were carried out below water level at a depth between 0.1 and 2.4 m (Figs 23-24). In addition, aerial photographs of the sector were made with the use of a drone. The Excavations at Limyra (Lycia) 2022: Preliminary Report Fig. 23 Photoscan of the submerged southwest corner of the foundation/podium of the Cenotaph, before cleaning (H. Öniz © ÖAW-ÖAI). 233 Fig. 24 Photoscan of the submerged southern wall of the foundation/podium of the Cenotaph, after cleaning (H. Öniz © ÖAW-ÖAI). Fig. 25 Drone photo of the Cenotaph of Gaius Caesar, looking northeast (H. Öniz © ÖAW-ÖAI). As a result of the 2022 study on the Cenotaph of Gaius Caesar, it has become clear that the depth of the survey in the 1970s could not be reached due to the fact that all cleaning works done using SCUBA equipment has to be done at clear visibility. Therefore, the careful cleaning work reached a depth of 2.4 m only, which corresponds with the fifth row of the blocks. A continuation of the underwater excavations around the Cenotaph of Gaius Caesar (Fig. 25) is planned for the next campaign, with the aim to reach the lowest row of the podium and the foundations using various archaeological methods. The Fortifications of Limyra (M. Helvacıkara) The fortifications of Limyra have become one of the topics of research in the 2022 excavation season at Limyra. In the scope of a Ph.D. dissertation at Middle East Technical University, Architecture Department’s Graduate Program in Conservation of Cultural Heritage, Limyra’s Byzantine-period fortifications are the main research topic (Fig. 26). 234 Martin Seyer et al. Fig. 26 Western City fortifications in 2021-2022 (M. Helvacıkara © ÖAW-ÖAI). The Byzantine fortifications, with their complex structure and position, layout, logistic concept, and construction technique and materials, mirror not only the history and topographical transformation of the Byzantine Middle Ages but also regarding the different phases of spolia incorporated in its walls. They are important information sources that should be systematically studied, recorded, and documented, and information related to fortifications should be carefully managed as a basis of any conservation project. This theoretical framework will include understanding, documentation, recording, and information management of fortifications as heritage sites based on international policy documents of heritage organisations. The first phase includes the documentation, analysis, and description process as a generative act of the fortifications of Limyra. In the second stage, it will be studied how the city walls built in Lycia and Anatolia during Late Antiquity and the Byzantine period, should be considered a source of information for conservation approaches in the light of the information obtained from Limyra. Therefore, this thesis proposes to create an approach via Limyra fortifications for perceiving cultural heritages as information sources via getting, understanding, storing, sharing, transferring and using the information as a crucial part of the conservation process. The documentation work of the city walls was completed in September 202222. In the studies, measurements were made to include plan/section/facade diagrams with the precision that can be expressed in 1/50 scale, which will provide the expression of the building in the architectural (vectorial) language to prepare the current situation drawings. In the measurement studies, devices were installed in sixty places, and a three-dimensional laser scanning system (3DLS), a digital-based documentation technique, was used (Fig. 27). The ‘FARO Laser Scanner Focus 3D’ device was used in the measurements. The current state of the walls was documented in three dimensions. Thanks to the system that creates a three-dimensional digital image in photographic reality, a three-dimensional model of the building, called ‘point cloud’, was created, and all plan and facade diagrams, sections and land elevations were determined by this model. The visual accessibility of construction techniques, keeping original materials and construction characteristics are the main criteria for selecting the Late Antique and Byzantine fortifications 22 The author wants to express her sincere thanks to K. Gülcen (Ankara) for his valuable help with the photogrammetric documentation of the walls. The Excavations at Limyra (Lycia) 2022: Preliminary Report 235 Fig. 27 Three-dimensional scan of the Western City fortifications, the interior east façade, created via a three-dimensional laser scanning system (3DLS) carried out in 2022 (M. Helvacıkara © ÖAW-ÖAI). within this study. Therefore, some examples were out of the scope because they are just ruins, and also, some examples could not be taken as a case due to the fact that they had been exposed to intense conservation implementations for many years. Limyra is the most proper one from the Lycia region for this study. A Summer School at Limyra 2022: Architects on Archaeological Sites (Z. Kuban, B. Ar, U. Almaç, G. Günay) A summer school, “Arkeolojik Alanda Mimar/Architects on Archaeological Sites,” organised by the authors took place at Limyra between 3-11 September 2022. The objective of the summer school is the qualification of architecture students for archaeological field studies. The summer school is free of charge for students and their expenses are covered by external sponsorships. This year Limyra Excavation hosted the summer school providing free lodging and food. In addition to the facilities provided by the official excavation budget, travel expenses of the students and organisers were financed by a project grant of AKMED (Suna & Inan Kıraç Research Center for Mediterranean Civilizations). A one-day excursion to other nearby excavation sites and examples of vernacular architecture was made possible with the help of the municipality of Finike, which provided a bus. There is a great need for architects specialised in archaeological field studies for the numerous excavations and surveys throughout Turkey. None of the national architectural schools give any specified education that teaches their students what an architect is supposed to do and know on an archaeological site. This summer school tries to fill the gap with a specified education. The summer school series, annually recurring since 2018, was organised for a third time this year after a two-year break because of the pandemic, and it took place for a second time in Limyra this year. The students are exclusively İTÜ undergraduates. The announcement of the summer school is made during the semester. The students apply with a motivation letter. Their interest and 236 Martin Seyer et al. performance in the courses related to History of Architecture are also taken into consideration during the selection of successful applicants. Among the participants this year, nine out of 16 students were in their first year of architecture education, and seven were in their second and third years. The mixture of younger and older students is very fruitful as the older students share their knowledge with the younger ones, especially on digital representation techniques, computer programs for drawing, etc. The 2022 summer school was carried out as a one-week program in September. The practical field works were executed during the day, while in the evenings the knowledge was deepened and intensified with theoretical lectures. The members of the Limyra archaeological team contributed with presentations on the objects and topics they have been working on, such as the introductory talks of the specialists on ceramics and metals as well as the archaeologist responsible for the fieldwork and the restoration team. The director of the excavation gave an overview of archaeological studies on Limyra as well. The first practical task that the participants encountered was the architectural documentation of a masonry structure, partially preserved with its walls, located at the eastern part of the ancient settlement. Before this task, the students experienced traditional survey techniques and instruments through practice on site, since a learning-by-doing approach is essential to achieve the goals of the summer school. Through this practice, the students exercised plan-elevation and section drawing of a structure on a scale of 1:50 and came into close contact with historic structures. They practiced sketching, measuring, recording, and drawing in a CAD environment. During the survey process in the field, traditional building techniques and materials have also been introduced benefiting from the close-up view to our case study. The second task was about the documentation of small-scale architectural elements such as column capitals, architrave blocks, details of reliefs, etc. on a scale of 1:5. This practice paved the way to think and discuss construction techniques, architectural styles, and reconstitution of structures in archaeological sites. For the third task the students were organised in groups to experience working together. In Necropolis V of Limyra each group chose a rock-cut tomb, which provided them with a good example of a tiny building with a complete room with a ceiling, door and facade. In the fourth task, a stretch of the Byzantine city wall was documented (Fig. 28), which moved the practical study to a larger object, where the use of total station and computer-aided design software were introduced in this context. In this part, the students were divided into different groups, and each group was responsible for one segment of the wall. A new experience at the summer school this year was a one-day extra brainstorming for the use of design practices in archaeological sites. As in many other archaeological excavations in Turkey, the facilities that the excavation team employs for accommodation and scientific work in Limyra are located in a limited section within the archaeological site. Of course, this situation offers many advantages, but it also brings some limitations. The design practice of the summer school in which the older students participated, aimed to meet some of the daily basic needs of the excavation team such as the improvement of the wet areas containing sanitary fixtures and appliances (Fig. 29). Of course, this is just another one of the architect’s responsibilities from the wide variety of tasks that fall within the discipline of design in archaeological sites. The enthusiasm of the students towards this practice encouraged us to initiate a new module of the summer school that will be devoted to design issues in archaeological sites in the upcoming versions with contributions from colleagues working in the field of architectural design. The Excavations at Limyra (Lycia) 2022: Preliminary Report 237 Fig. 28 Fourth Task-Measuring the Byzantine Wall (AAMR-Archive) (© ÖAW-ÖAI). Fig. 29 Sketch for extended washing possibilities proposal (Photo and proposal: N. Kıyak) (© ÖAW-ÖAI). 238 Martin Seyer et al. Bibliography Arqué, M.-C., Lemaître, S., & Waksman, S. Y. (2012). Les céramiques de cuisson en Lycie romaine: essai de définition des faciès morphologiques et d’approvisionnement en Méditerranée orientale. In C. B. Vallet (Ed.), Les céramiques communes dans leur contexte régional: faciès de consommation et mode d’approvisionnement. Actes de la table ronde organisée à Lyon les 2 et 3 février 2009 à la Maison de l’Orient et de la Méditerranée. Travaux de la Maison de l’Orient et de la Méditerranée 60 (pp. 141-153). Lyon: Maison de l’Orient et de la Méditerranée Jean Pouilloux. Banning, E. B. (2020). The Archaeologist’s Laboratory: the Analysis of Archaeological Evidence. Cham: Springer. Bes, Ph. M. (2020a). Fourth- to eighth-century pottery at Limyra: three short case studies. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, G. Forstenpointner, K. Löcker, R. Totschnig, I. Uytterhoeven, E. Cayre, H. Öniz, C. Öztosun, & Z. Kuban, The excavation at Limyra/Lycia 2019: preliminary report. Anatolia Antiqua, XXVIII, 228-235. Bes, Ph. M. (2020b). Early Byzantine pottery from Limyra’s West and East Gate excavations. Adalya, 23, 377407. Bes, Ph. M. (accepted). Overland connections. Roman-period ceramic exchange between Limyra (Lycia) and Sagalassos (Pisidia), southwest Asia Minor. HEROM. Journal on Hellenistic and Roman Material Culture, accepted. Bes, Ph. M., & Vanhecke, L. (2015). Turning over a new leaf. Leaf impressions of Styrax officinalis L. and Vitis vinifera L. on Late Roman Sagalassos amphorae. HEROM, 4(1), 107-166. Bollók, Á., & Tepper, Y. (2021). The changing fashion of belt accessories in the southern Levant: sixth- to seventh-century belt accessories from Shivta. Liber Annuus, 71, 441-492. Borchhardt, J. (1993). Die Steine von Zẽmuri. Archäologische Forschungen an den verborgenen Wassern von Limyra. Vienna: Phoibos Verlag. Borchhardt, J. (2002). Der Fries vom Kenotaph für Gaius Caesar in Limyra. Forschungen in Limyra 2. Vienna: Phoibos Verlag. Brusco, M. J., Köhn, H.-F., & Stahl, S. (2008). Heuristic implementation of dynamic programming for matrix permutation problems in combinatorial data analysis. Psychometrika, 73(3), 503. DOI: org/10.1007/s11336007-9049-5 Corremans, M., Poblome, J., Bes, Ph. M., & Waelkens, M. (2010). The quantification of amphorae from Roman Sagalassos, southwest Turkey. In B. Horejs, P. Pavúk, & R. Jung (Eds.), Analysing Pottery. Processing – Classification – Publication (pp. 285-303). Bratislava: Comenius University. Dolea, A. (2017). The preliminary results of 2016 Limyra excavation campaign. In M. Seyer, A. Dolea, K. Kugler, H. Brückner, & F. Stock, The excavation at Limyra/Lycia 2016: preliminary report. Anatolia Antiqua, XXV, 144152. Dolea, A. (2019). Excavations in the West City. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, S. Baybo, A. K. L. Leung, U. Quatember, M. Wörrle, H. Brückner, F. Stock, A. Symanczyk, G. Stanzl, K. Kugler, & B. YenerMarksteiner, The excavation at Limyra/Lycia 2018: preliminary report. Anatolia Antiqua, XXVII, 232-234. Dolea, A. (2020). Excavations in the lower city of Limyra. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, G. Forstenpointner, K. Löcker, R. Totschnig, I. Uytterhoeven, E. Cayre, H. Öniz, C. Öztosun, & Z. Kuban, The excavation at Limyra/Lycia 2019: preliminary report. Anatolia Antiqua, XXVIII, 222-228. Dolea, A. (in press). Urban pragmatism and adaptation in Byzantine Limyra. In Proceedings of Contested Heritage: Adaptation, Restoration and Innovation in the Late Antique and Byzantine World 21st International Graduate Conference, Oxford University Byzantine Society, Leiden. Dolea, A., Anton, S., Hangartner, J., Kainz, K., Lotz, H., & Orakçılar, B. (2017). Excavation in the West City of Limyra – Limyra Polis West. In M. Seyer, Limyra 2016. ANMED. News Bulletin on Archaeology from Mediterranean Anatolia, 15, 56-59. Dunnell, R. C. (1970). Seriation method and its evaluation. American Antiquity, 35(3), 305-319. Fernández, A. (2014). El comercio tardoantiguo (ss. IV-VII) en el noroeste peninsular a través del registro cerámico de la ría de Vigo. Roman and Late Antique Mediterranean Pottery 5. Oxford: Archaeopress. The Excavations at Limyra (Lycia) 2022: Preliminary Report 239 Forstenpointner, G. (2020). Faunal remains from the 2016 and 2018 West City excavations – Preliminary results. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, G. Forstenpointner, K. Löcker, R. Totschnig, I. Uytterhoeven, E. Cayre, H. Öniz, C. Öztosun, & Z. Kuban, The excavation at Limyra/Lycia 2019: preliminary report. Anatolia Antiqua, XXVIII, 238-241. Forstenpointner, G., & Gaggl, G. (1997). Archaozoologische Untersuchungen an Tierresten aus Limyra. Jahreshefte des Österreichischen Archäologischen Instituts Beiblatt, 66, 419-426. Friendly, M. (2002). Corrgrams: Exploratory displays for correlation matrices. The American Statistician, 56, 316-324. Gaitzsch, W. (2005). Eisenfunde aus Pergamon. Geräte, Werkzeuge und Waffen. Pergamenische Forschungen 14. Berlin: De Gruyter. Galik, A., Forstenpointner, G., & Weissengruber, G. (2012). Archäozoologische Befunde zur Jagd und Viehwirtschaft in Limyra. In M. Seyer (Ed.), 40 Jahre Grabung Limyra. Akten des internationalen Symposions, Wien, 3.–5. Dezember 2009 (pp. 163-168). Vienna: Phoibos Verlag. Ganzert, J. (1984). Das Kenotaph für Gaius Caesar in Limyra. Architektur und Bauornamentik. Istanbuler Forschungen 35. Tübingen: De Gruyter. Gassner, V. (1997). Das Südtor der Tetragonos-Agora. Keramik und Kleinfunde. Forschungen in Ephesos XIII(1)(1). Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Gruvaeus, G., & Wainer, H. (1972). Two additions to hierarchical cluster analysis. British Journal of Mathematical and Statistical Psychology, 25(2), 200-206. Habermehl, K.-H. (1975). Die Altersbestimmung bei Haus- und Labortieren, 2. Auflage. Berlin: Verlag Paul Parey. Hahsler, M., Hornik, K., & Buchta, C. (2008). Getting things in order: an introduction to the R package seriation. Journal of Statistical Software, 25, 1-34. Hahsler, M., Buchta, C., & Hornik, K. (2022). _seriation: Infrastructure for Ordering Objects Using Seriation_. R package version 1.3.5. Retrieved from https://CRAN.R-project.org/package=seriation. Accessed 10 March 2023. Hayes, J. W. (2003). Hellenistic and Roman pottery deposits from the ‘Saranda Kolones’ castle site at Paphos. The Annual of the British School at Athens, 98, 447-516. Hongo, H. (1997). Patterns of animal husbandry, environment and ethnicity in central Anatolia in the Ottoman Empire period: faunal remains from Islamic layers at Kaman-Kalehoyuk. Japan Review, 8, 275-307. Kassambara, A., & Mundt, F. (2020). factoextra: extract and visualize the results of multivariate data analyses. R package version 1.0.7. Retrieved from https://CRAN.R-project.org/package=factoextra. Accessed 10 March 2023. Kendall, D. G. (1971a). Abundance matrices and seriation in archaeology. Zeitschrift für Wahrscheinlichkeitstheorie und verwandte Gebiete, 17(2), 104-112. Kendall, D. G. (1971b). Seriation from abundance matrices. In F. R. Hodson, D. Kendall, & P. Tauta (Eds.), Mathematics in the Archaeological and Historical Sciences (pp. 215-252). Edinburgh: Edinburgh University Press. Ladstätter, S. (in collaboration with Bezecky, T., Czurda-Ruth, B., Giuliani, A., & Rogl, C.) (2003). Keramik. In C. Lang-Auinger (Ed.), Hanghaus 1 in Ephesos. Funde und Ausstattung. Forschungen in Ephesos VIII(4) (pp. 2285). Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Lauwerier, R. C. G. M. (1988). Animals in Roman Times in the Dutch Eastern River Area. Nederlandse Oudheden 12. Project Oostelijk Rivierengebied I . A m e r s f o o r t : R i j k s d i e n s t v o o r h e t O u d h e i d k u n d i g Bodemonderzoek. Lemaître, S. (2007). Vaisselle antique à Limyra en Lycie orientale (SO 3A et 3B). In T. Marksteiner, S. Lemaître, & B. Yener-Marksteiner, Die Grabungen der Jahre 1982-1986 am Südtor von Limyra. Jahreshefte des Österreichische Archäologischen Instituts, 76, 197-235. Lemaître, S., Waksman, S. Y., Arqué, M.-C., Pellegrino, E., Rocheron, C., & Yener-Marksteiner, B. (2013). Identité régionale et spécificités locales en Lycie antique: l’apport des céramiques culinaires. In P. Brun, L. Cavalier, K. Konuk, & F. Prost (Eds.), Euploia. La Lycie et la Carie antiques. Dynamiques des territoires, échanges et 240 Martin Seyer et al. identités. Actes du colloque de Bordeaux, 5, 6 et 7 novembre 2009. Ausonius Éditions Mémoires 34 (pp. 189-212). Bordeaux: Ausonius. Lund, J. (2018). Towards an explanation of the dearth of ceramic fine wares in Cyprus and parts of the Levant between AD 150 and 300. Acta Archaeologica, 89, 159-172. Lyman, R. L. (2008). Quantitative Paleozoology. Cambridge: Cambridge University Press. Marchand, J., Reynolds, P., & le Bomin, J. (2022). Proto-LRA 1 amphorae from Taposiris Magna: a typological and petrological study. Bulletin de Liaison de la Céramique Égyptienne, 31, 87-120. Noddle, B. (1974). Ages of epiphyseal closure in feral and domestic goats and ages of dental eruption. Journal of Archaeologic Science, 1, 195-204. Öner, E. (2013). Likya’da Paleocoğrafya ve Jeoarkeoloji Araştırmaları. Edebiyat Fakültesi Yayın 182. Izmir: Ege Üniversitesi Yayınları. Öztosun, C., Seyer, M., & Öniz, H. (2020). Underwater archaeology at Limyra. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, G. Forstenpointner, K. Löcker, R. Totschnig, I. Uytterhoeven, E. Cayre, H. Öniz, C. Öztosun, & Z. Kuban, The excavation at Limyra/Lycia 2019: preliminary report. Anatolia Antiqua, XXVIII, 254257. Payne, S. (1987). Reference codes for wear states in the mandibular cheek teeth of sheep and goats. Journal of Archaeological Science, 14, 609-614. Pearson, K. (1901). LIII. On lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 2(11), 559-572. Poblome, J., Corremans, M., Bes, Ph. M., Romanus, K., & Degryse, P. (2008). It is never too late… The Late Roman initiation of amphora production in the territory of Sagalassos. In I. Delemen, S. Çokay-Kepçe, A. Özdizbay, & Ö. Turak (Eds.), Euergetes. Festschrift für Prof. Dr. Haluk Abbasoğlu zum 65. Geburtstag Volume 2. Suna & Inan Kiraç Research Institute on Mediterranean Civilizations 1 (pp. 1001-1012). Istanbul: Suna & Inan Kiraç Research Institute on Mediterranean Civilizations. Pülz, A. M. (2020). Byzantinische Kleinfunde aus Ephesos. Ausgewählte Artefakte aus Metall, Bein und Glas. Forschungen in Ephesos XVIII(1). Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Quatember, U., Leung, A. K. Y., & Wörrle, M. (2020). Ornimythos von Limyra und sein frühkaiserzeitlicher Ehrenbogen. Jahreshefte des Österreichischen Archäologischen Institutes in Wien, 89, 219-370. R Core Team (2022). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing. Retrieved from https://www.R-project.org/. Accessed 10 March 2023. Rantitsch, G., Prochaska, W., Seyer, M., Lotz, H., & Kurtze, C. (2016). The drowning of ancient Limyra (southwestern Turkey) by rising groundwater during Late Antiquity to Byzantine times. Austrian Journal of Earth Sciences, 109(2), 203-210. DOI: 10.17738/ajes.2016.0015 Robinson, W. S. (1951). A method for chronologically ordering archaeological deposits. American Antiquity, 16, 293-301. Schuh, U. (2012a). Die Ausgrabungen in den Theaterthermen von Limyra, Vorläufige Ergebnisse 2007-2009. In C. Reinholdt, & W. Wohlmayr (Eds.), Akten des 13. Österreichischen Archäologentages, Salzburg, 25. – 27. Februar 2010 (pp. 161-167). Vienna: Phoibos Verlag. Schuh, U. (2012b). Die Theaterthermen von Limyra: Ergebnisse der Ausgrabungen 2007-2009. In M. Seyer (Ed.), 40 Jahre Grabung Limyra. Akten des internationalen Symposions, Wien, 3.–5. Dezember 2009 (pp. 287-300). Vienna: Phoibos Verlag. Schulze-Dörrlamm, M. (2009). Byzantinische Gürtelschnallen und Gürtelbeschläge im Römisch-Germanischen Zentralmuseum, Teil I. Die Schnallen ohne Beschläg, mit Laschenbeschläg und mit festem Beschläg des 5. bis 7. Jahrhunderts. Mainz: Verlag des Römisch-Germanischen Zentralmuseums. Schwarcz, D. Zs. (2019). Preliminary results of the study on the metal finds in the West City of Limyra. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, S. Baybo, A. K. L. Leung, U. Quatember, M. Wörrle, H. Brückner, F. Stock, A. Symanczyk, G. Stanzl, K. Kugler, & B. Yener-Marksteiner, The excavation at Limyra/Lycia 2018: preliminary report. Anatolia Antiqua, XXVII, 239-241. The Excavations at Limyra (Lycia) 2022: Preliminary Report 241 Schwarcz, D. Zs. (2020a). Preliminary results of the study on the metal finds. In M. Seyer, A. Dolea, Ph. M. Bes, D. Zs. Schwarcz, G. Forstenpointner, K. Löcker, R. Totschnig, I. Uytterhoeven, E. Cayre, H. Öniz, C. Öztosun, & Z. Kuban, The excavation at Limyra/Lycia 2019: preliminary report. Anatolia Antiqua, XXVIII, 235-237. Schwarcz, D. Zs. (2020b). Erste Auswertung der Materialanalysen byzantinischer Kleinfunde in Ephesos mit Bezug auf Archäologie, Technologiegeschichte und Herstellungstechniken. In A. M. Pülz, Byzantinische Kleinfunde aus Ephesos. Ausgewählte Artefakte aus Metall, Bein und Glas. Forschungen in Ephesos XVIII(1) (pp. 243-295) Vienna: Verlag der Österreichischen Akademie der Wissenschaften. Seyer, M., Dolea, A., Kugler, K., Brückner, H., & Stock, F. (2017). The excavation at Limyra/Lycia 2016: Preliminary report. Anatolia Antiqua, XXV, 143-160. Seyer, M., Dolea, A., Bes, Ph. M., Schwarcz, D. Zs., Baybo, S., Leung, A. K. Y., Quatember, U., Wörrle, M., Brückner, H., Stock, F., Symanczyk, A., Kugler, K., & Yener-Marksteiner, B. (2019). The excavation at Limyra/Lycia 2018: Preliminary report. Anatolia Antiqua, XXVII, 231-252. Seyer, M., Dolea, A., Bes, Ph. M., Schwarcz, D. Zs., Forstenpointner, G., Löcker, K., Totschnig, R., Uytterhoeven, I., Cayre, E., Öniz, H., Öztosun, C., & Kuban, Z. (2020). The excavation at Limyra (Lycia) 2019: Preliminary report. Anatolia Antiqua, XXVIII, 219-263. Slane, K. W. (1997). The Fine wares. In S. C. Herbert (Ed.), Tel Anafa II,i: the Hellenistic and Roman Pottery. Journal of Roman Archaeology Supplementary Series 10 (pp. 247-406). Ann Arbor: Journal of Roman Archaeology. University of Southampton. (2014). Roman Amphorae: a Digital Resource [Data-Set], York. Retrieved from https://archaeologydataservice.ac.uk/archives/view/amphora_ahrb_2005/. Accessed 31 January 2023. Uygun, Ç., Özdemir, B., & Korkut, T. (2023). The lamp molds and lamp production of Tlos in the Roman period. Olba, XXXI, 199-232. Von den Driesch, A. (1976). Das Vermessen von Tierknochen aus vor- und frühgeschichtlichen Siedlungen. Munich: Ludwig-Maximilian-Universität. Wiegand, G. (1973). Physisch-geographische Veränderungen im küstennahen südwestkleinasiatischen Raum und deren Bedeutungen für die Entfaltung und für den Niedergang der antiken Siedlungen zwischen Limyra, Arneai, Phellos und Antiphellos. In J. Borchhardt et al., Limyra: Bericht der III. Grabungskampagne 1971. Türk Arkeologi Dergisi, 20(1), 40-43. Yaman, A. (2022). Some thoughts on a table amphora type (AF1) in Arycanda. Cedrus, X, 323-331. DOI: 10.13113/cedrus.202217 Yener-Marksteiner, B. (2007). Keramik aus der Sondage 5 in der Weststadt von Limyra. In T. Marksteiner, S. Lemaître, & B. Yener-Marksteiner, Die Grabungen der Jahre 1982-1986 am Südtor von Limyra. Jahreshefte des Österreichische Archäologischen Instituts, 76, 236-276. Yener-Marksteiner, B. (2009). Ausgewählte spätrömisch-frühbyzantinischen Keramik aus einem Gebäudekomplex in der Weststadt Limyras. Jahreshefte des Österreichische Archäologischen Instituts, 78, 227252. Yener-Marksteiner, B. (2019). Arkeolojik kontekstler ışığında Limyra yöresel-yerel mutfak seramiğinin ana formlarına diakronik bir bakış: tencere ve güveçler (ms I–VII. yüzyıl). Cedrus, VII, 345-364. DOI: 10.13113/ CEDRUS/201914 zerobooksonline.com