Wood identification of wooden marine piles from the ancient Byzantine port of Eleutherius/ Theodosius

Abstract

The purpose of this study was to identify the wood species of the marine and filling piles obtained from the ancient Byzantine port of Eleutherius/ Theodosius, Istanbul, Turkey. Anatomical descriptions and identifications of 12 marine and 4 filling piles were performed by microscopic evaluations. In the study, Castanea sativa Mill., Quercus ithaburensis Decne., Quercus pontica C. Koch., and Cupressus sempervirens L. species were identified. No precise identifications were completed for only six samples at the species level; however, those samples showed significant similarity to Quercus spp. and Fagus spp. It was concluded that the economically viable supply of wood was more appropriate than obtaining it from nearby regions. The people living in ancient times had solid knowledge and experience on the utilization of wood species.

Full Article

WOOD IDENTIFICATION OF WOODEN MARINE PILES FROM THE ANCIENT BYZANTINE PORT OF ELEUTHERIUS/THEODOSIUS

Dilek Dogu,* Coskun Kose, S. Nami Kartal, and Nurgun Erdin

The purpose of this study was to identify the wood species of the marine and filling piles obtained from the ancient Byzantine port of Eleutherius/ Theodosius, Istanbul, Turkey. Anatomical descriptions and identifications of 12 marine and 4 filling piles were performed by microscopic evaluations. In the study, Castanea sativa Mill., Quercus ithaburensis Decne., Quercus pontica C. Koch., and Cupressus sempervirens L. species were identified. No precise identifications were completed for only six samples at the species level; however, those samples showed significant similarity to Quercus spp. and Fagus spp. It was concluded that the economically viable supply of wood was more appropriate than obtaining it from nearby regions. The people living in ancient times had solid knowledge and experience on the utilization of wood species.

Keywords: Marmaray Project; Archaeological wood, Wood identification; Port of Eleutherius/Theodosius

Contact information: Department of Forest Biology and Wood Protection Technology, Forestry Faculty, Istanbul University, Bahcekoy, Sariyer, 34473, Istanbul, Turkey; *Corresponding author: addogu@istanbul.edu.tr

INTRODUCTION

Throughout its long history Istanbul has served as the capital city of four great empires, namely the Roman Empire, the Byzantine Empire, the Latin Empire, and the Ottoman Empire, for more than 1,600 years. Istanbul has been a crucial trade center for various goods as well as a ‘metropolitan’ city for more than 2,800 years since the city was not only an administrative, but also a religious center (Keskin and Diren 1992).

In 2004, an enormous project called Marmaray was given a start to construct an underwater tunnel between the Asian and European sides of Istanbul, and thus to ease the city’s traffic problem. The tunnel under the Bosphorus will be the deepest built ever with its deepest point being about 58 m under the water surface. The deep stations and tunnels are being constructed in the area where civilization can be traced more than 7,000 years back in time (Lykke and Belkaya 2005). The ancient Byzantine port of the fourth-century has been recently uncovered under the slums of Yenikapi as the focal point of $4 billion tunnel project, in the European side of Istanbul, (Fig. 1) (Marine Cultural and Historic Newsletter 2006).

The port is a trove of relics dating back as far as the time of Constantine the Great. The Roman emperor Constantine moved his capital from Rome to Byzantium in 330 AD and renamed the city Constantinople, which became Istanbul later (Journal of Indian Ocean Archeology 2006). It finally grew into the busiest trading center in the

Fig. 1. The Marmaray Project along the coast line of the Marmara Sea and Yenikapi excavation area

eastern Mediterranean. The ships from here carried the wine in jars and amphorae from the Sea of Marmara and the cargoes of grain came in from Alexandria. The ancient Byzantine port called Port of Eleutherius / Theodosius, which was built by The Roman Emperor Theodosius I (A.C. 379–395), was an important one until the 7th century; however, it was abandoned since then because grain trade came to end. Afterwards, this ancient port area became filled by alluvium of the River Bayrampasha, was merged with the mainland during the first years of the Ottoman Empire, and was used for vegetable farming at a site called Vlanga (Langa) Bostani (ARIT Newsletter 2006, 2007).

This paper aims to identify the wood species of wooden marine and filling piles from the biggest port of the Byzantine Era (Günsenin 2007). For this purpose, wooden objects excavated from the port were subjected to anatomical examination by using light microscopy analysis.

EXPERIMENTAL

Materials

In 2007, 16 wood samples (12 marine and 4 filling piles) in varying sizes and characteristics were obtained from the Yenikapı Marmaray site (Table 1), where excavations continue, led by the Istanbul Archaeological Museum (Fig. 2). The site showed barren sandy sediment characteristics; however, no soil analyses were done to determine site characteristics. Since wood may shrink, fragment, and collapse into small pieces, all samples were stored in water at 4ºC, and the water was renewed at two- or three-week intervals (Blanchette 2000).

Fig. 2. General view of the excavation site and wooden marines (marine and filling piles)

Methods

Anatomical descriptions and identifications of each sample were performed based on the microscopic studies of cross (CS), radial (RS), and tangential sections (TS). For these purposes, approximately 10 by 10 by 20 mm blocks were cut from the samples. Well-preserved samples were cut to thin sections (about 20 to 30 µm) using a Reichert sliding microtome; however, heavily decomposed samples were not suitable for cutting with a microtome. Such samples were hand-cut with a razor blade.

Table 1. Information on Wood Samples Obtained from the Ancient Port of Eleutherius / Theodosius

Sections from the blocks were stained with safranine and safranine–picro–aniline blue and were then observed by means of an Olympus BX51 Light Microscope. Images were taken by using analySIS FIVE software and a DP71 Digital Camera installed and adapted on the microscope. The terminology for the wood descriptions generally conforms to the format by the International Association of Wood Anatomist (IAWA Committee 1989, 2004).

Wood identification was done considering not only quantitative features but also qualitative anatomical characteristics of the samples observed under the microscope. Tangential vessel diameters, vessel frequencies (only in diffuse porous wood), ray heights/widths, and intervessel pits size were measured for hardwood identification. Tangential tracheid diameters, ray heights, and cross-field pits number per field were determined for identification of softwood samples. On the other hand, the presence of growth ring boundaries, porosity, vessel arrangements, shape of the solitary vessel outline, type of axial parenchyma patterns, cellular composition of rays, intervessel pits arrangement, and types of perforation plates were used as qualitative anatomical characteristics for hardwood identification. Presence of growth ring boundaries, transition from earlywood to latewood, presence of axial parenchyma, arrangement of earlywood tracheid pitting in radial walls, ray composition, horizontal and end walls of ray parenchyma cells, cross-field pitting, and end walls of axial parenchyma cells were also used for softwood identification. Quantitative and qualitative features of the wood samples were then compared with microscopic slides in the Xylarium of Forestry Faculty, Istanbul University, Turkey. Atlases, websites and publications were also used as references for comparisons.

RESULTS AND DISCUSSION

Restrictions in Examinations

There were some restrictive factors in the study. One was the selection of the areas to measure anatomical characteristics of wood samples and number of measurements due to distortions and decompositions of wood during burial. Despite the negative structural features of the wood samples, measurements were carried out on the cells without decomposition. Another restrictive factor was unknown tree age and growing conditions of the wood samples used in the construction of the port. Since changes in wood structure depend on age, growing conditions, and location height that the specimen is taken from the tree, even the same wood species can show more or less variations in anatomical properties such as growth ring width, cell size, and cell wall thickness (Bozkurt and Erdin 2000).

Uncertainties in anatomical features of the samples for wood identification depending on the restrictive factors mentioned above can be explained as follows.

There was obvious distortion in the shape of the some vessels because the wood was compressed during burial in samples of MRY’07 J 142, MRY’07 L 140, and YKM’07 2Ea2.

Since there was obvious distortion in sample of MRY’07 J 144, the number of earlywood vessels in radial multiples; tangential diameter of vessels was not determined. The shapes of latewood vessels and scanty paratracheal parenchyma were indis-tinguishable.

Since there was obvious distortion in the structures of the growth ring in the sample of YKM’07 2Ec4 the shape of the vessels, types of the axial parenchyma, appearance of flames in latewood were indistinguishable, and the diameters of the vessels were not measured. Only apotracheal diffuse type axial parenchyma was observed in this sample.

Sample MRY’07 I 10 showed distinct distortion and decomposition in the structure of the growth rings. Therefore, types of the axial parenchyma were not distinguished and only apotracheal diffuse type axial parenchyma was observed.

Anatomical Descriptions and Identification of Wood Samples

The qualitative results of the samples and comparisons to literature are summarized below, and the quantitative results are given in Tables 2 and 3, respectively.

Hardwoods

Family: Fagaceae

Castanea sativa Mill. (Sample code: 2Hb3, 2Hd1, MRY’07 J 144, MRY’07 J 145)

CS―Growth ring boundaries distinct. Wood ring-porous. Earlywood vessels mostly solitary and in radial multiples of 2–5 (in 2Hb3) and 2–6 (in 2Hd1 and MRY’07 J 145). Tyloses common. Latewood vessels in radial and / or diagonal pattern (in MRY’07 J 144, MRY’07 J 145) and almost in dendritic pattern (in 2Hb3, 2Hd1), angular in outline. Axial parenchyma scanty paratracheal and apotracheal diffuse, diffuse-in-aggregates (Figs. 3a―d).

Fig. 3. Castanea sativa samples obtained from the ancient Port of Eleutherius/Theodosius

a) Cross section of MRY’ 07 J 145, b) Cross section of MRY’ 07 J 144, c) Cross section of 2Hd1,

d) Cross section of 2Hb3, e) Radial section (2Hd1)

Fig. 3 (continued). f) Vessel-ray pits (2Hb3), g) Scalariform perforation plate in latewood (MRY’ 07 J 145), h) Tangential section (2Hb3)

RS―Perforation plates simple, rarely scalariform in latewood. Rays composed predominantly of procumbent cells, sometimes with one row of upright and / or square marginal cells. Vessel-ray pits with much reduced borders to apparently simple: (i) pits rounded or angular, (ii) pits horizontal (scalariform) (Figs. 3e―g).

TS―Rays exclusively uniseriate. Intervessel pits alternate (Fig. 3h).

Comparison to literature―Earlywood vessels mostly solitary (Schoch et al. 2004) and in radial multiples of 2–6 (Merev 1998a; Bozkurt and Erdin 2000). Latewood vessels in radial and / or diagonal and sometimes in dendritic pattern (Merev 1998a; Bozkurt and Erdin 2000). Tangential diameter of earlywood vessels 240–400 µm (Bozkurt 1967), 120–394 µm (Merev 1998a), ≤ 300 µm (Bozkurt and Erdin 2000), 100-200 µm (The Xylem Data Base). Tangential diameter of latewood vessels 19–99 µm (Merev 1998a), 30–40 µm (Bozkurt and Erdin 2000). Axial parenchyma paratracheal, apotracheal diffuse, diffuse-in-aggregates (Merev 1998a; Bozkurt and Erdin 2000; Schoch et al. 2004; The Xylem Data Base) and in marginal (The Xylem Data Base). Perforation plates simple (Bozkurt and Erdin 2000; Akkemik et al. 2004; Schoch et al. 2004; The Xylem Data Base) and scalariform in latewood (Merev 1998a). Rays composed exclusively of procumbent cells (Merev 1998a; The Xylem Data Base), uniseriate (Merev 1998a; Bozkurt and Erdin 2000; The Xylem Data Base) and rarely biseriate (Bozkurt 1967; Akkemik et al. 2004; Schoch et al. 2004), 1-20 (9) cells high (Merev 1998a), ≤ 30 cells high (Akkemik et al. 2004) and 10–30 cells high (Schoch et al. 2004). Intervessel pits alternate (Merev 1998a; The Xylem Data Base).

Fagus spp. (Sample code: 1Bd3)

CS―Growth ring boundaries distinct. Wood diffuse–porous. Although there is a gradual change to narrower vessels in the latewood, vessel diameter is uniform throughout most of the growth ring. Vessels mostly solitary, angular in outline, 108–145 per square mm. Larger rays distended at the growth ring boundary. Axial parenchyma apotracheal diffuse, diffuse-in-aggregates (Fig. 4a).

RS―Perforation plates simple in large vessels and scalariform in narrow vessels. Rays composed predominantly of procumbent cells, sometimes with one row of upright and / or square marginal cells. Vessel-ray pits with much reduced borders to apparently simple: (i) pits rounded, (ii) pits horizontal (scalariform) (Fig. 4b and Fig. 4c).

TS―Rays of two distinct sizes, smaller rays 1–6 seriate, larger rays > 10-seriate. Intervessel pits opposite and scalariform (Fig. 4d).

Fig. 4. Fagus spp. sample obtained from the ancient Port of Eleutherius/Theodosius a) Cross section

Table 2. Some Quantitative Anatomical Features of Hardwood Samples

^a Mean values are given in the first parenthesis in italics

^b Min. and max. values are given in the second parenthesis

^c Values outside parenthesis indicate the numbers of observation

^d Medium

^e Small

^* Could not be measured