The realisation that some of the bones from the Negev siteswere of horses (Grigson, 1991 and 1993) came as a completesurprise, no one would have expected them at such an earlydate in this area. It has even been suggested that they must beintrusive. However the presence of horse bones in all the siteswith large sample size—Gilat, Shiqmim and Grar, plus thosenewly recognized from the 1993 excavations at Shiqmim andin the smaller assemblage of Abu Matar, is incontrovertible.As we shall see horses of similar size are also present in someearly Bronze sites in the northern Negev.
Figure 3 compares the sizes of the equid bones from theNegev sites and Ghassul with those of the domestic donkeysfrom Maadi (Boessneck, von den Driesch and Ziegler,1989), the wild asses from Kom Ombo in Egypt (Churcher,1972) and the other sites noted above, and with the equidsfrom Shams-ed Dinn, almost all of which Uerpmann (1982and 1986) considered to be onagers. The graph shows a verywide range of variation for the sizes of bones from the Negevsites, and that many of the bones are much larger that thoseof donkeys or onagers; the only explanation is that they are ofhorses, or mules.As stated above the assumption that these Negev horseswere domestic, has been questioned by Wapnish (1997), Levine(1999) and Olsen (2006).
One of the equid teeth at Gilat, Fig. 3, tentatively identified as horse because of its great length, wasa lower second premolar. The crown was worn down past theinternal enamel patterns, the mesial face was so worn that itsloped distally and was polished all over. It is possible that thisis not merely the result of old age, but also of bit-wear, or ratherwear against a rope, signifying that the animal was under closehuman control (Grigson, 2006: 286, pl. 6.1b). The bevel on themesial face is much like that described by Anthony and Brown(1991; 2000), as indicating bit-wear. However Olsen (2006)has observed similar wear on a few of the second premolars ofhorses from the Pleistocene in North America.