Wattle hurdles and wooden structures during the Iron Age: positives from clay imprints recovered at Castro de S. Vicente da Chã María Martín-Seijo 1, , João Fonte 2 and Inés L. López-Dóriga 3 1 Grupo de Estudos para a Prehistoria do NO Ibérico (GI-1534), Departamento de Historia I, Universidade de Santiago de Compostela, Praza da Universidade s/n, 15782 Santiago de Compostela, Spain 2 Incipit-Instituto de Ciencias do Patrimonio, Avenida de Vigo s/n, 15705 Santiago de Compostela. 3 Wessex Archaeology E-mail address: maria.martin.seijo@gmail.com (María Martín-Seijo) INTRODUCTION Wooden architecture was usual during Iron Age. Most of the structural elements of buildings such as posts and poles were made of timber. Besides them wattle hurdling played an important role during this period. The perishable nature of wood conditioned the preservation of this organic material in the archaeological record. Despite this inconvenience building with wattle and daub techniques favoured the preservation of clay fragments that preserved indirect evidences of the presence of trunks and branches at Iron Age sites of Northwest Iberia (Martín-Seijo and Carballo 2010, Rey et al. 2011 among others). The aim of this research was to study the clay imprints recovered at Castro de São Vicente da Chã (Chã, Montalegre, Northeast Portugal) (Fig. 1). Imprints have been measured, registered and positive moulds have been obtained using silicone. Secondly, these would be the starting point of a collection of positive moulds obtained from clay imprints. This compilation would facilitate the study and comparison of this kind of indirect evidences. Finally our objective was to enlarge our knowledge about wooden architectures during the Iron Age. ARCHAEOLOGICAL BACKGROUND The Castro de S. Vicente da Chã was object of rescue excavations in the 1960s of the 20th century by Joaquim Rodrigues dos Santos Júnior, Agostinho Isidoro and Osvaldo Freire due to the construction of the Alto Rabagão dam (Júnior and Isidoro 1963; Júnior and Freire 1964; Freire 1968), financed by the former Empresa Hidro Eléctrica do Cávado. Several structures were exhumed, namely some quadrangular and circular buildings (Fig. 2. 1) and a monumental stone wall (Fig. 2. 2) (© Centro de Memória de Torre de Moncorvo). Nowadays the site is partially covered by the Alto Rabagão dam. The main results of this archaeological works were published in the journal Trabalhos de Antropologia e Etnologia and the archaeological materials were deposited at the Museu de História Natural in Porto. These materials allow us a generic dating for the site between the 2nd century BC and the 1st century AD. There were also obtained positive moulds of silicone (Shera®Twist 60) from clay imprints, using high-resolution silicone in areas where negatives of the original surface were preserved (Sherasil-Extra Light). The moulding process followed the steps previously described by Camarós et al. 2016 and López-Dóriga et al. 2015 (Fig. 4). This technique is a non-destructive and conservation friendly method for the study of archaeological remains The surface of the fragments was cleaned using smooth brushes in order to remove soil and roots remains. After we started with the moulding process. Three areas for obtaining moulds were selected. Two of them with the objective of obtaining positives of the branches and their disposition in the hurdle structure. The last one was made with the aim of obtaining a high resolution mould of the surface of the imprint. The silicone (Shera®Twist 60) was directly applied to the clay surface. In the case of the high resolution mould, it was first applied a layer of high resolution silicone (Sherasil-Extra Light) followed by a new layer of silicone putty (Shera®Twist 60) to cover it for providing support to the mould. While the silicone was still soft, we wrote the reference of the sample. When the silicone was totally set, it was carefully removed. MATERIAL AND METHODS Several clay fragments with branch and trunk imprints were recovered during the excavation works. There have been selected for this study only two fragments (UP-MHNFCP-154922 and UP- MHNFCP-154923), those clay remains with best preserved imprints (Fig. 3). The rest of fragments with evidences of erosion were discarded because they have lost their original morphology. Each clay fragment was analysed individually. We have counted the number of imprints preserved and measured the maximum diameter of the branch imprints using a vernier caliper (Martín- Seijo and Carballo 2010). There were identified 3 imprints at UP-MHNFCP-154922 and 4 at UP-MHNFCP-154923. At the first clay fragment the maximum diameter of the branches ranges from 2.1 to 2.8 cm. At the second one only two measures could be done, one of 2.3cm and the other one that preserved the complete diameter measured 1.4cm. These measurements are coincident with those known for other Iron Age sites, although it is significant the presence of braches of more than 2cm in comparison with the other sites (Fig. 5). The moulds obtained allowed us to document the disposition and size of the branches (Fig. 6.5), and to register the detail of the surface of one of the vertical elements of this wattle hurdles (Fig. 6.3). Regarding wattle and daub construction, coarse grained clay was used to recover the wattle hurdles, covering the complete diameter of the branches (Fig. 6.5). The clay was mixed with small stones and small fragments of slag clearly visible to the naked eye in both fragments studied (Fig. 6.1 and 6.2). The external surface of the wattle and daub was smoothed and probably harden with fire (Fig. 6.6). CONCLUSION The fragments of clay with imprints of perishable materials offers a high potential of using inorganic material remains to study wooden architecture. Obtaining moulds is crucial to register and compare this kind of perishable structures. Coarse grained clay silicone putty offered good results and we recommend only to apply high resolution silicone when details of the original surface of the branches or the timber has been preserved. REFERENCES Camarós, E., Sánchez-Hernández, C., Rivals, F. (2016). Make it clear: molds, transparent casts and lightning techniques for stereomicroscopic analysis of taphonomic modifications on bone surfaces. Journal of Anthropological Sciences, 94: 1-8. Freire, O. (1968). O castro de S. Vicente da Chã (Montalegre). Campanhas de escavações de 1965 e 1966. Trabalhos de Antropologia E Etnologia, 20(3-4): 368-375. Júnior, J. R. dos S. & Isidoro, A. (1963). Escavações no Castro de S. Vicente da Chã (Barroso). Trabalhos de Antropologia E Etnologia, 19(2): 178-186. Júnior, J. R. dos S. & Freire, O. (1964). O Castro de S. Vicente da Chã (Barroso). Campanha de escavações de 1964. Trabalhos de Antropologia E Etnologia, 19(3-4): 336-371. López-Dóriga, I., Camarós, E., Cueto, M., Teira, L., Sureda, P. (2015). Aproximación a la explotación de los recursos vegetales en Cap de Barbaria Ii: resultados preliminares. In: VI Jornades d'Arqueologia de les Illes Balears. Formentera, 26-29 September 2014 , pp. 143- 150. Martín-Seijo, M., Carballo Arceo, L. X. (2010). Le travail du bois et les pratiques d’élagage à l’Âge du fer : le site de Castrovite (Galice, Espagne). In: C. Delhon, I. Théry-Parisot, and S Thiébault, Des hommes et des plantes. Exploitaton du Milieu et Géstion des Resources Végétales de la Préhistoire à nous jours. XXXe Rencontres Internationales d’Archéologie et d’Histoire d’Antibes, Ed. APDCA. P. 343-353. Fig. 1. 1) Location of Castro de S. Vicente da Chã in Northwest Iberia. 2) The site is now partially covered by the Alto Rabagão dam. 3) Two views of S. Vicente da Chã before (SPLAL 1949) and after the construction of the dam (ORTOS_DGRF_2004_06). 1 3 2 Fig. 2. 1) Quadrangular and circular buildings with stone walls. 2) Stone wall that surrounded the site during the development of the archaeological excavation. 1 2 Fig. 3. 1 and 2) Clay with negative imprints of branches with the code UP-MHNFCP- 154922. 3 and 4) Clay with negative imprints of branches with the code UP-MHNFCP- 154923. 1 2 3 4 1 2 3 4 5 6 Fig. 4. 1) Surface cleaning. 2) Application of the first layer of high resolution silicone. 3) Application of the second layer of silicone putty. 4) Drying process. 5) Removing the moulds. 6) Two moulds obtained from UP-MHNFCP-154922. The orange is the high resolution mould of the surface of the imprint. 1 2 3 4 5 6 Fig. 6. 1) Coarse grained clay. 2) Slag framgents mixed with the clay. 3) Imprint of a vertical wood piece. 4) Branches imprints. 5) Complemente diameter of a branch completely recovered with clay. 6) Smooth finish of the external surface. RESULTS AND DISCUSSION There were identified two kind of imprints: 1) longitudinal and parallel with oval or semicircular section; 2) transversal to the previous imprints with quadrangular section. The first type correspond with the weaved branches and the second one with the vertical elements of this kind of structures. Fig. 5. Maximum diameter (in cm) measured in clay imprints from Iron Age sites of Northwest Iberia.