Internet Archaeology 9. Smith et al.

5.0 Discussion

5.1 Provenance

In Group 1 the size and roundness of the quartz; accessory minerals plagioclase, zircon and pyroxene; limestone and microfossils, including the coraline algae Amphiroa (Sivan 1996, 76), indicate a coastal provenance. These constituents are all present in the 'Coastal Sand', a relatively high-quartz sand extending from coastal Sinai to the Israel/Lebanon border (Horowitz 1979, 25).

In these fabrics, the Coastal Sand is combined with a small proportion of chert, chalk and basalt, indicating sources further inland. The most probable area is the seaward portion of the Jezreel Valley to the north-east and south-east of the Carmel Ridge, where the Kishon river, draining from an area of basalts inland near Affula, from Eocene formations to the north and south, and from the Carmel Ridge Cenomanian and Senonian formations, passes into the region of Coastal Sand (Map 2). This provenance may be related to a Bronze Age port, Tell Abu Hawam, north-east of Haifa. We also found good matches with samples from nearby Tell Nami (samples provided by Michal Artzy).

Group 2 is less homogeneous than Group 1. Some examples overlap with Group 1, by containing Coastal Sand with calcareous inclusions and a small amount of volcanic rock fragments, including basalt. The degree of similarity is insufficient to restrict the source of Group 2 to the same area as the former, having a lower percentage of basaltic fragments and sometimes a highly silty and ferruginous matrix perhaps from a terra rossa soil. This occurs with Cenomanian and Eocene formations, from which the calcareous components, including chalk and nari, and chert could be derived, in the Shephela region of central Israel. A similar combination also occurs further north into Lebanon so either source is possible (Map 3).

In Group 3 the basalt, limestone, chalk and chert, plus the absence of Coastal Sand, indicate an inland, not a coastal, origin. The limestone and chalk inclusions are well-rounded, so appear to have been transported, but their size shows they are not far from their source. This appears to indicate incorporation of a wadi sand and basalt as temper. A likely origin is the central Jezreel Valley in to eastern Galilee (Map 2). However, a similar combination of constituents exists elsewhere, around the north Lebanon coast from which no comparative material was available at the time of writing. Further work is being undertaken to determine which area is the most probable.

For Group 4, the igneous rock fragments, including those showing alteration to iddingsite or serpentinite, together with quartzite, schist and both types of chert, indicate a source where components of an ophiolite complex exist. The ophiolite-related constituents occur with pelagic and shallow water carbonates, including planktonic and benthic Foraminifera. This combination is consistent with a possible origin in the Antalya or Diyarbakir areas of Turkey, the Baër-Bassit area of north-west Syria or the area of the Mamonia and the Troodos complexes of west and south-west Cyprus (A. Robertson pers. comm. 1999); more work is required to choose solely on geological grounds between these areas, particularly Baër-Bassit and Cyprus, see Map 3. However, local pottery from the coastal site of Ras Shamra (Ugarit) immediately south-west of the Baër-Bassit contains the same ophiolite-related suite of minerals and lithic fragments with components indicating a coastal provenance. Further support comes from a similar set of constituents in thin sections of two Amarna Tablets, known to have been sent from Ugarit (Y. Goren pers. comm. 1999). It is intended to examine further one current subgroup (4.2.2) to determine whether this may represent a Cypriote provenance.

No closely-circumscribed source area has yet been identified for Group 5. The chert includes opaques, which often occur as a replacement after dolomite, indicating a source in a Cenomanian chert region. Preliminary examination of the constituents, by N. Porat and L. Grossowicz, showed that the microfossils include shallow-water algae and planktonic Foraminifera, indicating that they are derived from near the coast. The high content of carbonates rather than quartz is characteristic of the region north of Israel. Thus the source may lie in coastal Lebanon or Syria (Map 3). Yuval Goren (pers.comm.) has recently identified similar fossils in thin sections of Amarna Tablets sent from Amurru, suggesting a provenance within Lebanon, north of Byblos.

5.2 Contents

This has been approached by studying inscribed jars (Serpico 1996) and by residue analysis (Serpico and White 2000; Stern et al. 2000). It is clear that Groups 1 and 2 are linked to the trade in pistacia spp. resin, while 4 and 5 are associated with oil. Excavations at Ras Shamra have provided evidence of olive oil production in commercial quantities in the Late Bronze Age.