During the last two decades, there have been important Lower Palaeolithic discoveries in western Asia in regions such as the Levant, the Caucasus, and Pakistan pushing the evidence for the earliest hominid occupation in the region back to about 1 to 2 million years ago (Reference Bar-yosefBar-Yosef 1998 and references therein).

In terms of Lower Palaeolithic occupation, Iran is one of the least-known regions in western Asia with only a handful of evidence including some core-chopper assemblages from gravel deposits along Ladiz, Mashkid, and Kashafrud rivers in eastern Iran (Ariai & Thibault 1975; Hume 1976) and some surface occurrences and isolated finds of both core-chopper and Acheulian industries from the west and north-western parts of the Zagros region (Reference BraidwoodBraidwood 1960; Reference Sadek-koorosSadek-Kooros 1976; Singer & Wymer 1978; Reference MortensenMortensen 1993; Biglari et al. 2000).

Figure 1.

Rostamabad plain and the location of Ganj Par, looking north.

Figure 2.

A core-chopper from Ganj Par.

The evidence for the Acheulian industry in Iran consists of a few examples of bifaces as isolated finds or in association with chopping tool industries from the Zagros region. Two of these bifaces are single finds from Quri Goll north-east of the Lake Urmia, and Gakia in the Kermanshah region. Two other are surface assemblages from Amar Merdeg in the Mehran Plain, and Pal Barik at Holailan. Of these, Holailan bifaces are in doubt and seem to be Levallois cores (Jacques Jaubert 2003 pers. comm.).

Against such limited background, a recent discovery of an Acheulian occurrence with a rich lithic assemblage in northern Iran provides us with an opportunity to examine the strongest evidence for Lower Palaeolithic occupation so far discovered in Iran.

Figure 3.

Biface and proto biface from Ganj Par.

The evidence for the Acheulian industry in Iran consists of a few examples of bifaces as isolated finds or in association with chopping tool industries from the Zagros region. Two of these bifaces are single finds from Quri Goll north-east of the Lake Urmia, and Gakia in the Kermanshah region. Two other are surface assemblages from Amar Merdeg in the Mehran Plain, and Pal Barik at Holailan. Of these, Holailan bifaces are in doubt and seem to be Levallois cores (Jacques Jaubert 2003 pers. comm.).

Against such limited background, a recent discovery of an Acheulian occurrence with a rich lithic assemblage in northern Iran provides us with an opportunity to examine the strongest evidence for Lower Palaeolithic occupation so far discovered in Iran.

During the second season of joint Iranian-Japanese archaeological excavation in September 2002 at the historic site of Jalaliyeh in the western edge of Rostamabad plain, two of the authors (Biglari & Heydari) had the opportunity to carry out a brief survey of the surrounding area. While examining the surface of the Jalaliyeh mound, Biglari found a chert flake, prompting him to have a closer examination of the area. In an almost flat open area located just to the south of the mound a lithic scatter was found which yielded a lower Palaeolithic assemblage characterised by the presence of large cutting tools in association with cores and other debitage.

Figure 4.

Location of Ganj Par and some key Lower Palaeolithic sites in the Near East.

The site, called Ganj Par, is located on the western edge of the fluvial plain of Rostamabad at an elevation of about 235m asl and at about 36° 53' 38" N, 49° 28' 57" E , commanding the deep dissected valley of Kaluraz tributary river, where it opens to the Sefidrud valley. The Sefidrud River, which in an almost south-north direction flows into the Caspian Sea, is about 2km to the east of the site.

The geomorphology of the Rostamabad region has been briefly described by Maeomoku (2003). He distinguished five terraces at the western part of the plain. These terraces are between 150-800m asl. Ganj Par lithic scatter is located at altitudes between 225-235m asl, higher than terrace IV, and may have been reworked from deposits of older terraces.

During three visits to the site in 2002-2003 more than 100 artefacts were collected in an area of about 0.5 hectare. Almost half of the assemblage is made from limestone. Of other pieces, large proportions are made from silicified sandstone and volcanic rocks such as tuff, andesite and basalt. There are also some cherts and other rock types.

The assemblage is composed of high frequencies of core-choppers, along with heavy-duty scrapers, bifaces (handaxes, cleavers, and a trihedral pick), cores, and flakes. Bifaces are generally thick and crudely made and some of them have a form intermediate between chopper and biface. The collection lacks spheroid and polyhedron.

Lithic assemblages similar to Ganj Par are unknown from Lower Palaeolithic sites of Iran and its neighbouring regions. But there are some similarities between Ganj Par and those known from the early Acheulian assemblages from Ubeidiya in the Levant (Bar Yosef and Goren-Inbar 1993) and East African sites such as Olduvai Gorge, upper bed II (Reference LeakeyLeakey 1971) and Konso-Gardula (Asfaw et al. 1992).

The Ganj Par evidence suggests that migrating Acheulian groups probably made their way into the northern Alborz foothills from the north-west along the ancient Caspian shores. However, the possibility of a southern arrival from the Zagros range and the Central Plateau of Iran can not be excluded as the narrow Sefid Rud valley provides a major route from the north-western part of the Iranian plateau to the western Caspian lowland.