Human translocation of rodents has been evidenced in the Near Oceania as far as the Late Pleistocene, demonstrating an early role of human agency over the distribution of mammalian species. The most ubiquitous of the commensal rodents is the house mouse, and the widely accepted cause of its worldwide distribution is its stowaway transport within infested grain and foodstuffs cargo, although the data to support this assertion is sparse. The historical process of its biological invasion in the Mediterranean is now well documented thanks to the diachronic mapping of zooarchaeological occurrences showing an invasive mechanism intimately linked with the evolution of maritime trade and exchange. Its first human translocation happened as a by-product of the Neolithic colonization of Cyprus, evidencing indirectly a deep-sea Neolithic navigation covering 10,000 years. Bronze Age maritime trade linking eastern Mediterranean cities and empires was then the main vector of its spread in the eastern Mediterranean. Finally, its adaptation to the western Mediterranean commensal niche had been successful only once the migrant flow of mice by passive transport was impulsed by the commercial and demographic maritime expansions of the last millennium B.C.
Although this historical process relies on the assumption that stowaway transport was the dispersal vector in the Mediterranean, no direct archaeological evidence was available to support it. The only direct archaeological evidence of stowaway transport is the skull of a brown rat (Rattus norvegicus) found in the shipwreck of an XVIIIth century French galleon that sank off the Corsican shores. This discovery could directly highlight the association between the brown rat invasion and the history of modern navigation.
Systematic sieving of sediments from the Late Bronze Age Uluburun shipwreck produced a single small murine (mice) mandible of a little more than one centimeter. This seemingly incidental find among the vast royal cargo of this impressive vessel provides us with a unique opportunity to examine the role of stowaway transports in the biological invasion of commensal rodents in light of Mediterranean Bronze Age trading networks. Two-dimensional geometric morphometric analysis of the in situ first lower molar from the mandible is utilized to identify the species involved in stowaway transport and localise a potential geographical source relying on intraspecific geographical variation of molar phenotype.
If this mandible can be connected to the vessel’s final journey, this anecdotal find will provide direct evidence for the role of the Late Bronze Age maritime trade as a significant factor in the history of the dispersal of the house mouse in Europe, contributing to the global consensus for human activity as a vector for the dispersal of commensal murines, which are among the most threatening invasive species for biological diversity and human health. In addition, if the geographical print on the molar can be inferred from its phenotype, the results could provide a significant contribution to the questions regarding the route followed by the ship before its sinking.
The Uluburun shipwreck is a Late Bronze Age vessel discovered in the Mediterranean Sea lying on a rocky side between 44 and 52 meters in depth, just off the south coast of Turkey near the city of Kas in the province of Antalya. The shipwreck was excavated during eleven campaigns by the Institute of Nautical Archaeology (INA; Bodrum, Turkey) between 1984 and 1994. Other Bronze Age shipwrecks such as the Dokos wreck, third millennium B.C. in the Saronic gulf, and the Cape Gelidonya wreck, 1200 B.C. off the Turkish coast, have also been excavated, but the Uluburun wreck is significant in that it is the oldest one to be excavated using modern underwater archaeological techniques. Dendrochronology has dated the stored firewood to a tree felled between 1316–1305 B.C. which is contemporary with finds from the wreck such as the Mycenaean wares (LHIIIA: 2 type); however, these dates are now considered to be insecure due to the poor preservation of the two samples used.
The vessel itself was 15 meters in length and provided a carrying capacity of 20 tonnes. Its vast cargo has yielded a unique picture of the richness and diversity of eastern Mediterranean maritime trade at the end of the Bronze Age, with items sourced from at least nine different cultures: Canaanite, Mycenean, Egyptian, Cypriot, Nubian, Baltic, Northern Balkan, Babylonian, Assyrian, and Kassite. The main cargo appears to have consisted of metal ingots comprising 10 tonnes of copper ingots in typical oxhide shape and 40 tin ingots. The ship also carried some exotic raw materials such as ebony, amber, ostrich eggs used as containers, elephant ivory, and hippopotamus teeth. Manufactured goods included Mycenaean and Cypriot pottery stored in Canaanite storage jars, tools, fishing equipment, weapons, stone artifacts, gold and silver jewellery, and a unique gold scarab bearing the name of the Egyptian queen Nefertiti. The wealth of the cargo strongly indicates the use of the ship as a Royal vessel rather than that of a private merchant, though still seemingly within the structure of Late Bronze Age international exchange. Evidence from the Amarna letter and the iconography of the Kenamun tombs (Thebes) describes Syrian crews discharging their cargo, and, indeed, the Uluburun cargo probably represents a royal shipment like those exchanged between the Syro-Palestinian coast, Egypt, and Alashia (Cyprus) with destinations further into the Aegean also suggested.
The ship’s point of departure is still debated as having a Cypriot or Syro-Palestinian coastal origin or alternatively coming from both these areas; however, current opinion based on Lead isotope characteristics suggests that the vessel was outbound from Cyprus with copper ingots quarried from its own mines. Its final destination, according to the vast wealth of the cargo which may have included offerings to Egyptian Pharaohs, may have been the Nile River, as it was one of the main trading centres of the time. The nationality of the ship itself is, nevertheless, still questionable.
The right mandible (L 1⁄4 1.4 cm, l 1⁄4 0.5 cm) was recovered by wet sieving loose sand and sediment and fragments of encrustation originating from the vicinity of the fourth row of copper ingots; it is well-preserved partly due to corrosion from the copper ingots evidenced by the greenish encrustation on the mandible. The third molar (m3) was missing and the coronoid process was broken. According to molar tooth wear, the mandible comes from a young individual (circa 3–6 months old). Preliminary macroscopic observation of the first lower molar (m1) allowed its positive taxonomic identification as Mus sp. (mice). However, morphoscopic examination alone could not provide its determination to species level due to the genus Mus comprising several sibling species.49 Therefore, the m1 morphology of the Uluburun mouse has been compared with genotyped specimens using molar shape analysis. This technique has already demonstrated its suitability for interspecific and intraspecific distinction within the Mus genus.50 The genotyped reference set includes the three Mus species of the eastern Mediterranean considered as prime candidates for taxonomic identification of the Uluburun mouse. The data set includes the Macedonian mouse (Mus macedonicus), which is represented by two sub-species with parapatric geographic range; the Cyprus mouse (Mus cypriacus), a recently discovered endemic species; and the house mouse (Mus musculus domesticus), the only species which has developed a commensal relationship with humans. Mus macedonicus macedonicus occurs in Syria, Turkey, and Greece, whereas Mus macedonicus spretoides has been discovered only in Israel. Feral house mice do live in sympatry (i.e. share the same habitat) with other mice, such as M. macedonicus in the eastern Mediterranean or the endemic M. cypriacus in Cyprus, but remain the only Mus species in Crete and Egypt.
It should be noted that the modern data set of house mouse populations does not include samples from coastal Anatolia or Egypt, two potential sources of stowaway transport; however, historical sources allow these locations to be ruled out as useful for this study. Starting with the Anatolian coast, there are no historical or archaeological sources for a port or harbour on the maritime route of the Bronze Age Mediterranean trade. Antiphellus, the oldest Anatolian harbour situated under the modern city of Kas, was only founded during the VIth century B.C., becoming an important trading port during the IVth century B.C., exporting timbers from Lycia. Egypt, as previously stated, is more likely to have been the final destination of the vessel rather than a step on its route.
The most efficient way to describe the morphological variation of murine molars is to quantify the two-dimensional projection of its occlusal view, providing information regarding the relative position, and the swelling of the cusps, which are an important factor on the genetic identity of the taxon. The external outline of the murine molar does not vary with the wear on the attrition surface, which prevents morphological parameters being defined depending on the age of the specimen.
The external outline of each m1 is acquired using 64 equally spaced points semi-automatically sampled using an optical image analyser (Optimas 6.2). To obtain the shape descriptors of the external outline of the first lower molar (m1) occlusal surface, an Elliptic Fourier Transform (EFT) was performed. This method is based on a separate Fourier decomposition of the incremental changes of the x- and y- coordinates as a function of the cumulative length along the outline provided for both the periodic functions, an infinite sum of trigonometric functions with decreasing-length waves called harmonics (Ha). Each Ha corresponds to four coefficients: An and Bn for x and Cn and Dn for y, defining an ellipse in the x-y plane. The first harmonic, describing the best-fitting ellipse of the original outline is used to perform the size, orientation, and starting point standardizations necessary to allow the shape comparisons between all outlines. The coefficients of the first harmonic corresponding to residuals after standardization have not been included in subsequent statistical analyses. Shape changes are visually depicted by reconstructed outlines. An outline can be reconstructed from any set of Fourier coefficients following the inverse Fourier method. ‘‘Mean outlines,’’ reconstructed on the averaged Fourier coefficients per taxon, were used to visualise shape differences between the groups.
Digitizing and acquisition of the outline coordinates have been performed using an image analyser (Optimas v.6.2 by Imasys).
The size of the m1 was estimated using the area of the ellipse of the first harmonic, and the difference between the Uluburun mouse and Eastern Mediterranean species was tested using analysis of variance (ANOVA) with pairwise mean comparisons assessed with Tukey post hoc test.
As the primary objective of this study was to identify the specific status of the Uluburun’s mouse, a multiple group form of discriminant analysis of the shape variables called Canonical variate analysis (CVA) was utilized. EFT computed a very large amount of shape variables (256 coefficients in our case, 64 harmonics × 4 coefficients). CVA requires more specimens than the sum of the number of groups and measurements per specimen. Therefore, a dimensionality reduction based on Principal Component Analysis (PCA)78 was performed. The scores of the Principal Components that optimize the rate of correct assignments obtained through cross-validation (Jackknife) were used rather than the coefficients set.
Differences between mice groups (species, geographic localities) were assessed using Multivariate Analysis of Variance (MANOVA; test considered: Wilks’ Lambda). Mahalanobis distances (D2) were computed inside the canonical molar shape space for the group assignment (species, geographic locality) of the Uluburun mouse. The posterior probabilities for group membership associated to these distances will allow us to decide which specific status and geographical area can be allocated to the Uluburun mouse. Allocations with posterior probabilities below 0.75 will be rejected.
Statistical analyses were performed using Systat v.11 Systat Software Inc. (SSI) and NTsys-PC v.2.1Ó 2002 by Applied Biostatistics, Inc.
Molar shape differences are highly significant between the three extant mice species (Wilks’ Lambda 1⁄4 0.1108, ddl1 1⁄4 2, ddl2 1⁄4 143, P < 0.0001). The classification matrices indicate that 97% of the specimens are correctly classified; using the cross-validation (Jackknife), the assignment drops to 89%, confirming the robustness of the comparative model. The canonical scores plot clearly shows that the Uluburun mouse falls within the M. musculus domesticus molar shape variation range. This graphical classification is confirmed by the D2 with a 1.0 probability associated with the assignment to the M. musculus domesticus group. Its molar size falls within the variation range of house mouse populations from both mainland and island regions, with no significant differences. Therefore, the mandible found in the Uluburun shipwreck can be unambiguously identified as a house mouse (M. musculus domesticus).
Extant house mouse populations from the eastern Mediterranean display highly significant molar shape differences (Wilks’ Lambda1⁄40.1481; ddl11⁄44; ddl21⁄4114; P<0.0001). These populations represent five potential geographical sources for the Uluburun house mouse origin. Nevertheless, the classification matrices indicate that only 65% of the specimens are correctly classified (using the Jackknife procedure, this percentage drops to 45%). The highest probability of group membership associated to the D2 is the sample from Syria (P 1⁄4 0.89). This means that among the phenotypic diversity of house mouse populations in the eastern Mediterranean basin, Uluburun’s house mouse is most similar to the modern Syrian morphotype. This suggests that this particular house mouse boarded the ship with cargo from a harbour situated somewhere along the Northern Levantine Coast. However, this specific geographic attribution is somewhat tentative due to the percentage of correctly assigned modern specimens among their geographical group and must be further discussed in light of other archaeological and historical evidence.
Because shipwrecks are not sealed deposits, they can suffer from later contamination.88 The greenish encrustation on the mandible, demonstrating the long contact between the mandible and the copper ingots among which it had been found, provides clear evidence that the mandible does not represent an intrusive element and can then be considered as contemporary with the shipwreck. The mandible is, therefore, the remnant of a house mouse living on board when the ship sank.
The molar of this mandible does not present any peculiar morphological characteristic compared to extant house mice, whereas allometric increase of the tooth size (macrodontie) and morphological drift are common features for murines (rat and mice) populations adapted to isolated environments like islands95 and, potentially, boats. However, there is currently no known biometric study of house mouse populations living on boats that could confirm that this type of isolation will lead to the same morphological adjustment as the ones found on island populations. Nevertheless, we can hypothesize that the high level of commensalism on a boat leads to increased population density, which is in favour of body-size increase and a concomitant macrodontie. The biometric analysis of the brown rat (R. norvegicus) skull discovered in the ‘‘Ça ira’’ shipwreck supports this hypothesis. Indeed, this skull presents allometric proportions (a large cranium and jugal teeth raw with a reduced rostrum length) interpreted as the morphological adaptation to a life spent in the galleon. Therefore, both the age at death (juvenile) and the lack of morphological drift of the Uluburun house mouse lends strongly in favour of this particular mouse being loaded just prior to the vessel’s final journey. The house mouse in the Uluburun shipwreck is, therefore, the oldest direct evidence so far for stowaway transport of a commensal species and clearly illustrates how Bronze Age maritime traffic was a vector of house mouse dispersal.
The biometric data have yielded evidence of a stowaway house mouse loaded on the vessel prior to the ship sinking. Its tentative Syrian origin can contribute to the debate regarding the route of the vessel before its fateful day. The question concerning the point of departure or the origin of the vessel is impossible to answer using items from the cargo alone. As in the case of the Uluburun vessel, they could have been loaded from Cypriot, Syro-Palestinian, or Aegean ports.
In general, the best evidence for the ship’s origin comes from the stone anchors stored in the bow of the shipwreck. However, until petrography results refine the geographical source of their limestone, the ones of the Uluburun vessel could potentially come from Cyprus, the coasts of Israel, or Ugarit. The ship galley wares represent the best potential indicators of the nationality of the boat. Among the artefacts discovered were saucer-shaped ceramic oil lamps presumably of Syro-Palestinian origin that displayed charring along their nozzles indicating their use during the journey, and, therefore, leaning in favour of the argument for the Syro-Palestinian ethnicity of the ship. Unpublished petrography results of the ship galley wares yield a geographic source around Haifa in Israel for these ceramics, which would refine the ethnicity of the crew as Canaanite. It is well known that Canaanite merchants who settled along the Syro-Palestinian coast played a major role in eastern Mediterranean maritime trade during the Late Bronze Age.
However, the most dramatic direct evidence of a port along the route of the vessel is undoubtedly the geographical source of unwelcome guests like commensal rodents, especially if the specimens are young and newly loaded, like the house mouse of Uluburun shipwreck. In contrast to non-cargo pottery or personal items, the geographical origin of a stowaway commensal cannot yield the ethnicity of a ship, but it can inform us about the route of the vessel and ports visited.
This can be inferred when combining behavioural, archaeological, and historical information.
We know from behavioural ecology that, although omnivorous, the house mouse is mostly fond of cereals and that the transport of infested grain and hay are the most likely vector for their passive transport. The rate of stowaway transports of one mouse per 7 tonnes of grain/hay suggests that huge amounts of grain/hay have to be shipped to allow a successful loading of commensal mice. The presence of a house mouse in the shipwreck would then plead in favour of a grain shipment for trading purposes. Direct evidence of cereal crops among the plant remains only rely on a few fragments of charred wheat and barley (rachis and chaffs). Furthermore, the possibility of a grain shipment for trading purposes has been considered unlikely since no space was available on the ship for traded cereal crops cargo. However, the tenuousness of cereal crops evidence should not rule out the hypothesis of a grain shipment for trading purposes if one considers that their storage could have relied on basketry or matting, whose fragments have been recovered, rather than on jars. Such containers would not have been able to prevent the cereals from dispersal both during and after sinking. In addition, it cannot be denied that such long-distance trade would have necessitated extensive food provisions for the voyage. Hence a large amount of grain is highly probable, even if not for trading purposes, to ensure that the needs of both crew and passengers were met.
The city of Ugarit (Ras-Shamra) seems to be the most likely Canaanite port that allies with the Late Bronze Age east–west copper and tin maritime trade route where large amount of grains could have been loaded. Situated on the Syrian coast, this city was the largest international emporium between the XIVth and the XIIIth centuries B.C., benefiting from its fertile hinterland which produced an abundance of wheat, barley, wine, and olive oil. It owed its prosperity to its strategic position, its well sheltered port of Minet el Beida, and to a valley situated north–east of the city. All these factors converged to ensure control of the trade routes which connected Egypt, Mesopotamia, the Hittite Empire, Cyprus, and Crete. Of relevance to the understanding of the origins of the Uluburun cargo, its strategic position also made Ugarit the main tin port of the Levant. According to Carol Bell, the Uluburun ship most likely took on its tin cargo at Ugarit because it was situated at the end of the shortest route between the tin mines of Central Asia (Iran) and the Mediterranean coasts, along the Euphrates River and through the Syrian Desert.
The port of Minet el Beida as the potential home port of the Uluburun ship was already mentioned by Pulak, after the discovery of 80 canaanites amphoras similar to those from Uluburun as well as 1000 pieces of mostly Cypriot vessels possibly awaiting export by sea. The biometric identification of a stowaway house mouse with a Syrian phenotype provides further support for the hypothesis considering Minet el Beida as a most likely port on the trading route of the Uluburun vessel.
The tiny, single mandible of a house mouse discovered among the prestigious and vast Uluburun shipwreck cargo represents the earliest direct evidence for the stowaway transport of commensal rodents. This anecdotal but important discovery has highlighted the role of Bronze Age shipping in the invasive process of house mouse dispersal. By combining biometrical data with ecological, historical, and archaeological sources, this study has also been able to pinpoint that a grain shipment from a storehouse infested with mice most probably originated in the servicing port of Ugarit. A seemingly insignificant discovery within a vessel crammed full of exquisite and exotic goods provides a useful piece of evidence for the ship’s route 3500 years after its final and fateful journey.
Faience beads found at Uluburun vary widely in form and comprise eight distinct categories. While the surface glaze remains in rare patches only, most faience beads exhibit a blue undertone. Other colors, while less common, include red, yellow, white, and turquoise. The glass beads found at Uluburun may be loosely grouped into two categories, small and large. Many of the large glass beads exhibit yellow and white spot or crumb decoration. There is a possibility that all the large glass beads were decorated in this way, but surface deterioration masks the decoration.
From the 16th to 18th centuries, European merchants brought millions of Venetian glass beads to Africa, where the beads were traded for gold and slaves, consequently becoming a vital part of African economy and society.
The metal ingots from the wreck comprise the “single largest assemblage of Bronze Age copper and tin ingots.” Items found include gold jewelry, elephant and hippopotamus ivory, ostrich eggshells, and African blackwood, a wood that the ancient Egyptians called hbny.
Found in the wreckage were beadmaking materials, including glass, faience, Baltic amber, quartz, ostrich eggshell, bone, agate, and carnelian.
The Syro-Palestinian origin for several shipboard items, including lamps, balance weights, stone anchors, and the ship’s gold-foil clad, bronze deity figurine, confirms a west Asian port of origin, possibly even Ugarit, a city frequently mentioned in the Amarna letters.
Faience, a siliceous ceramic, consists primarily of silica with small amounts of soda and lime; its use in beadmaking dates back to the fifth millennium B.C. Faience beads found at Uluburun vary in shape and represent both cargo and personal items. Glass is a soda-lime-silica compound that, unlike faience, is fully melted to a liquid state and sets upon cooling. Two broad categories of glass beads were recovered at Uluburun: wire-wound glass beads and glass relief-beads. The relief-beads will be excluded from this study.
Beads possessed multiple values, among them amuletic value. Another use was personal adornment, and “...even the poorest inhumations do not lack a string or two of beads around the neck or arm of the deceased.” An easily-recognized form of Egyptian beading is the usekh, or multi-strand, beaded broad collar. The usekh design, developed during the Old Kingdom, remained widely worn by both men and women throughout the New Kingdom. Ceremonial beaded girdles with animals’ tails were widely worn by high-ranking men in the Old Kingdom; during the Middle Kingdom, beaded girdles, often incorporating cowrie-shell or acacia-seed beads, were worn by women, including servants. Tiny beads were also sewn onto cloth at Ur.
Beads are important during this late Bronze age and in other times; although originally associated with priestesses of the goddess Hathor, the menyet collar eventually evolved into a symbol of high rank for a woman wearing the counterpoise. A specific type of royal reward during the 18th-22nd Dynasties was the shebyu collar, given to both men and women for military or civil service. Strands of beads were also in dowry lists of Mitanni princesses. Moreover, Middle-Assyrian graves of high-status individuals include numerous beads of semi-precious stones as well as glass and faience. Another example is strands of glass and faience beads comprising part of the grave goods in Pit-Cave 66 at Knossos. Thousands of clay molds were used in the production of faience amulets at Amarna. Many varieties of eye beads and pendants were also found in Bronze Age Egypt, which perhaps protected the wearer against the evil eye, also known as “fascination.” Faience symbolises rebirth and its association with the god Osiris rather than its inherent value. A range of glass and faience eye beads of first millennium B.C. were found at Alishar Hüyük in Anatolia, again serving as amulets against fascination.
Numerous beads, possibly part of a votive offering or the equipment of a priest, were also found at the Late Bronze Age temple at Lachish near the Syro-Palestinian coast, which were set into bricks, sewn onto decorative textiles, and strung along the walls, supported by glazed wall-nails.
After the flood in The Epic of Gilgamesh, “Then, at last, Ishtar also came, she lifted her necklace with the jewels of heaven that once Anu had made to please her. ‘O you gods here present, by the lapis lazuli round my neck I shall remember these days as I remember the jewels of my throat; these last days I shall not forget.’”
A set of metric dial calipers provided bead dimensions accurate to one-tenth of a millimeter. I will need to mention that some categories include one or more subcategories representing beads exhibiting slight differences in decoration, proportion, or technique of manufacture.
Roughly 72,000 tiny faience beads were recovered from the Uluburun shipwreck. These beads range in diameter from two to five millimeters; a sample of 100 tiny faience beads yielded the following dimensions:
Average diameter: 0.27 cm;
Average length: 0.12 cm;
Average perforation diameter: 0.12 cm.
Tiny beads possess flat ends and a profile that is either straight or slightly convex, and their length-to-diameter ratio approaches that of disk beads.
An estimated 68,000 tiny faience beads were found as a concreted mass.
They are 0.21-0.31 centimeters in diameter (average 0.27 cm) and 0.10-0.17 centimeters in length (average 0.13 cm). Similar beads from other sites have been labeled as disk, flat-ring, or annular beads. They have been found at Bronze Age sites throughout the Levant in faience as well as gold, ostrich eggshell, and various stones. Jewelry recovered from the grave of the 12th-Dynasty princess Sit Hat-Hor Yunet at Lahun includes bracelets composed of hundreds of tiny beads of gold, turquoise, and carnelian. An 18th-century B.C. jar containing the tools and materials of a goldsmith also held nearly 3,000 tiny faience beads, and another 4,500 were recovered at a contemporary site nearby. Tiny beads were still in use in the region during the 14th century B.C. at Tell Zubeidi. Tiny faience beads of Late Bronze Age date were found at Cypriot sites including Hala Sultan Tekke16 and Toumba tou Skourou and at Kissonerga dating to the 3rd millennium B.C. Tiny beads of ostrich eggshell and black and green faience have also been found at Tarxien on Malta.
The segmented beads at Uluburun have a diameter of around 0.2 centimeters and possess two, three, four, or, in one instance, five segments. Some segmented beads exhibit a taper at one end at which the perforation or bead profile is usually distorted. The dimensions and color of the segmented and regular tiny beads are so similar as to suggest manufacture in close association.
Different forms, however, are contained within the spheroid group, such as the I.B.1.b beads with slightly flattened ends. Globular beads frequently exhibit staining, causing beads to appear in a range of colors including shades of gray, brown, and blue. Magnification reveals the actual color of each globular bead to be blue, varying only in value and chroma, certainly resulting from copper contamination:
Average diameter: 0.54 cm;
Average length: 0.42 cm;
Average perforation diameter: 0.18 cm.
Beads nearly identical to the Uluburun globular beads in both form and dimension occur at Aidonia, Dendra, Prosymna, and Pylos. They are common at Mycenae and correspond to Xenaki-Sakellariou’s type 1 (spherical) or type 2 (rounded). Similar beads dating to LC III are found on Cyprus. Globular beads are also common along the Syro-Palestinian coast, with close parallels to the Uluburun beads found at Tell Abu Hawām66 and Megiddo. Such beads are also present in 12th-century B.C. graves at Tell Zubeidi in Mesopotamia.
The number of oblate flutes present varies independently of the bead’s diameter or profile. Erosion along bead perimeter primarily affects ribs:
Average diameter: 0.68cm;
Average length: 0.52 cm;
Average perforation diameter: 0.21 cm;
Average number of flutes: 5-6.
The cogwheel group contains a special sub-group known as collared-cogwheel beads, so-called due to the presence of a collar or distinct convex band around the perforation at each end. These collared cogwheel beads occur infrequently at Uluburun—only 17 have been identified—and are sometimes difficult to distinguish from plain cogwheel beads.
Irregularity in rib size and spacing as seen at Uluburun occurs on plain cogwheel beads from Megiddo on the Syro-Palestinian coast and Toumba tou Skourou on Cyprus. Collared-cogwheel beads are also present at Megiddo but are roughly made in contrast to those found at Uluburun. Collared-cogwheel beads occur frequently in gold in Mesopotamia and constitute much of the well-known Dilbat necklace. Closer collared parallels exist at Beth Pelet, Amman, and Lachish.
Diam. 0.72 cm;
min. diam. 0.4 cm;
length 1.51 cm;
diam. of perforation 0.13, 0.15 cm;
diam. of end 0.23, 0.27 cm;
Length-to-diameter ratio: 2.1 (long).
Flattened amygdaloid form with five grooves forming five ribs or gadroons on one side, and four grooves on the opposite side. Central groove extends to bead end on only one face.
Diam. 0.62 cm;
min. diam. 0.43 cm;
length 1.39 cm;
diam. of perforation 0.18, 0.17 cm;
diam. of end 0.3, 0.25 cm;
Length-to-diameter ratio: 2.24 (long).
Amygdaloid form, slightly flattened, with two grooves on each side forming three large, rounded gadroons. The gadroons closest to the bead edge curve around to join the next gadroon on the opposite face, so that four lobes are present in transverse section.
Diam. 0.53 cm;
length 1.72 cm;
diam. of perforation 0.19 cm;
diam. of end 0.37, 0.34 cm;
Length-to-diameter ratio: 3.25 (long).
Fusiform bead with five longitudinal flutes that do not extend to bead ends.
One hundred ninety-eight faience grain-of-wheat beads were found on the shipwreck: 153 beads of variant A, 43 beads of variant B, and two beads of variant C. Although variant C is both less common and easily identified, variants A and B can, at times, be indistinguishable, confirming Effinger’s statement, “Varianten dieses Typus lassen sich aufgrund der nur geringfügigen Unterschiede nicht klar abgrenzen.” This being the case, the aforementioned variants should be used as a general guide to understanding the forms and frequency thereof.
Nearly 50 beads of variants A and B from Panaztepe in western Anatolia are on exhibit at the Archaeological Museum in Izmir. Grain-of-wheat beads were also found at the Dictaean Cave on Crete, Pylona on Rhodes, and Hala Sultan Tekke on Cyprus. A solitary grain-of-wheat bead was found at Tell Abu Hawām. The grain-of-wheat type corresponds with Petrie’s type 122 at Gaza and Riis’ type 18 at Hama.
Diam. 1.83 cm;
length 0.96 cm;
diam. of perforation 0.31;
Length-to-diameter ratio: 0.52 (short).
Wear is heavier around perforation on gadrooned side.
Diam. 1.83 cm;
length 0.96 cm;
diam. of perforation 0.31;
Length-to-diameter ratio: 0.52 (short).
Fifty-four biconical faience beads were found on the Uluburun shipwreck: 47 of variant A and only two of variant B; five beads are poorly preserved and of indeterminable variant. Both variants are short bicones alike in dimension and would appear similar when strung; they are, however, distinguishable by several characteristics. The variant A beads possess 17 gadroons of varying width on one face only, hereafter referred to as the molded face due to their likely means of manufacture. The reverse, hereafter referred to as the incised face, is shorter and is decorated with 17 radially incised lines forming segments similar to petals. Although, the number of gadroons is indeterminable on roughly one-third of the variant A bicones due to attrition, the remaining beads possess 17 gadroons without exception. The level of preservation of the molded face varies from good to very poor. The incised face is, almost without exception, better preserved, in terms of both color and attrition. None of the variant A beads retain their original surface glaze; beads appear to be various shades of gray or blue, although all 52 beads appear blue when viewed under low magnification. The two variant B bicones, unlike variant A, are radially grooved on either face.
Diam. 0.64 cm;
length 1.17 cm;
diam. of perf. 0.18, 0.20 cm;
diam. of end 0.37, 0.38 cm;
Length-to-diameter ratio: 1.83 (long).
Grooved-barrel beads exhibit a long, convex profile terminating in flat, circular ends. Over a dozen grooved-barrel beads similar to those at Uluburun were found at grave VIII at Prophitis Elias. Ten similar beads at Elateia-Alonaki are somewhat longer and fusiform rather than barrel-shaped; however, Nightingale notes that each possesses between 11 and 13 incised lines.
Diam. 1.06 cm;
length 0.25 cm;
diam. of depression around perforation 0.6 cm;
diam. of perforation 0.22 cm;
Length-to-diameter ratio: 0.24 (disk).
The perforation is set in a central depression with a buff colored body. There are five beads of this type, only three of which are sufficiently preserved to allow detailed observation; the remaining beads deteriorated significantly after excavation. They have been labeled as button beads due to their similarity to the base of typical Mycenaean shanked, campaniform buttons. Examples of these were found in a tomb at Beth Pelet that were dated to the 19th Dynasty. Yellow, gray, and dark red faience beads of this type were found with other types, including tiny and globular beads, near the neck of a body at a burial at Gurob in Egypt. Vertically pierced rosette disks have been found in limited numbers at Mycenae, Troy, and on Crete.
Diam. 1.10 cm;
length 0.69 cm;
diam. of perforation 0.41 cm;
Length-to-diameter ratio: 0.63 (short).
Melon beads have oblate, faience spheroids with 13 straight, longitudinal gadroons extending nearly to the bead end. This is the only bead of its kind found at Uluburun. It is called a melon bead due to this type’s similarity to some fruits of the Cucurbitaceae (squash) family. A finely carved melon bead of rock crystal was found on the Uluburun shipwreck and possesses fifteen gadroons.
The type is found frequently in faience at sites along the Syro-Palestinian coast, from Alalakh to Gaza and numerous points between.
The excavation yielded a large number of wound glass beads of varying size. Of these, many were in such an advanced state of deterioration that they disintegrated in the process of excavation, often crumbling before reaching the surface. Still others had disintegrated completely, leaving only their impressions in encrustations. Due to their extended period underwater and the leaching out of various elements by the seawater, the present color of these beads no longer accurately reflects their original color.
Diam. 0.78 cm;
length 0.67 cm;
diam. of perforation 0.26 cm;
Length to diameter ratio: 0.86 (short).
An estimated 9,000 small glass beads were found on the Uluburun shipwreck. Beads of this category are small, evenly formed spheroids, either spherical or just slightly oblate. Striations perpendicular to the bead axis suggest manufacture through winding and thus a Karklins classification of WIb. Intact beads in this category generally exhibit a well-preserved surface with occasional surface spall. Devitrification, however, is predominant, and the translucent blue patches on the beads of KW 1550 are a rare exception. As a result of devitrification, the small glass beads are rendered opaque and exhibit a range of muted colors, including shades of yellow, brown, and light green. Bead form is generally well-preserved in this category due to the relatively infrequent occurrence of spallation.
While most of the small glass beads possess a plain, or approximately straight, perforation, several beads possess a conical (Type III) perforation that is exceptionally small on one end or fails to penetrate the bead surface altogether.
There is a possibility that a conical or incomplete perforation was intentional if these objects are pinheads rather than beads. However, it is far more likely that this feature is a manufacturing flaw resulting from winding on a tapered mandrel.
Beads nearly identical to those found at Uluburun in both form and dimension were found in tomb XIX at Prosymna (97 beads) and chamber tomb 524 at Mycenae (over 110 beads). Close parallels were also found along the Syro-Palestinian coast at Lachish116 and Tell Abu Hawām. Factory waste at Amarna provides evidence of glass bead manufacture in Egypt,118 and small, blue glass beads were found at both Gurob and Lahun.
Several of the small glass beads found on the Uluburun shipwreck exhibit a striped surface in which bands of differing colors, usually light brown and cream, encircle the bead axis. The bands are very narrow and their boundaries often ill-defined.
Glass beads decorated with spiral stripes have been found at several Late Bronze Age sites. However, this decoration is usually a thick, white spiral contrasting a black or dark bead body, often a barrel rather than a spheroid; beads found on the Late Bronze Age shipwreck near Cape Gelidonya were decorated in this style.
The small bead size and narrow stripes detract from the theory that the Uluburun beads were intentionally decorated with stripes and instead supports the idea that the striations represent characteristic deterioration of a wound glass bead. The segmented, glass beads always possess two segments joined end-to-end and sharing a single axis. In some, the two beads are nearly separated, exhibiting a deep score that does not meet but produces a Z-shaped line around the middle of the bead. A similar segmented glass bead has been published.
Average diameter: 1.25 cm;
Average length: 0.83 cm;
Average perforation diameter: 0.40 cm.
The beads in this category are both significantly larger and less uniform than the small glass beads. Approximately 500 large glass beads were found on the Uluburun shipwreck, although many more were likely present on the ship. They are classified as I.B.1.a or I.C.1.a and are generally spheroid, although some are too squat to be classified as such. These beads are often lopsided and frequently exhibit a small peak at one end, suggestive of manufacture through winding and thus of Karklins classification WIb. Many suffer from extensive surface spallation, effectively destroying the bead surface and precluding accurate measurements.
Some large glass beads from Tiryns are lopsided, again suggesting manufacture through winding. Glass beads found at Kāmid el-Lōz are very similar to those found at Uluburun, exhibiting both lateral cracks and a lopsided profile. A greenish-blue glass bead was found at Stratum IV at the Anatolian site of Alishar Hüyük (1400-1200 B.C.) and exhibits a white coating inside its perforation. Oblate spheroids in green and blue glass are also noted at Gurob.
Well-preserved, Uluburun, large glass beads are recognizable as eye beads, which are beads possessing spots or “eyes” of a different color of glass applied to the original glass matrix.
Simple spot glass eye beads utilize two colors of glass: that of the matrix or bead body and that of the spots, which are drops of glass applied then marvered or heated to lie flush with the bead surface.
Eye-ring beads had dropped out of the bead, leaving a hollow imprint where they once were.134 The same phenomenon may be observed in some of the eye beads found at Uluburun. Some beads do not retain eyes but possess pits approximating the size and shape of eye spots.
Other beads possess both eyes and pits, the pits occurring in approximately the same size and location as the eyes on complete beads. It should be attributed to differing coefficients of expansion of the colored glasses used. Flush-spot eye beads, then, may have been found in larger numbers than publications suggest.
Instead of rounded eyes, some of the large glass beads from the Uluburun shipwreck possess flush, yellow, and white spots that are both smaller and more angular than those of the traditional spot eye beads.
A glass crumb bead is also considered a form of simple glass eye bead. The spots on a glass crumb bead are characterized by random placement and often overlap. Many of the large glass eye beads found on the Uluburun shipwreck appear to possess both rounded spots and smaller, irregular crumbs.
Another, albeit abnormally large, crumb bead was noted at Elateia-Alonaki.