Figure 8.3 Solar orientation of Middle and Upper Palaeolithic cave and rock-shelter sites in the northern Perigord region, as documented by Duchadeau-Kervazo (1982). In both cases the sites show a strong tendency to be oriented towards the south -presumably to obtain the maximum solar radiation.
south of the Dronne (Duchadeau-Kervazo 1982, 1986; Geneste 1985) (Figs 8.1, 8.5). It would seem that in many contexts these smaller tributary valleys offered the most attractive habitats for Middle Palaeolithic groups - perhaps because of the better degree of climatic protection provided by these sheltered narrow valleys, or perhaps because of local vegetational or other micro-habitats which would have attracted concentrations of animals or other economic resources to these locations (see Chapter 2 and White 1985). It may be significant, however, that most of these sites are usually located only a few kilometres from the confluence with larger valleys and would therefore have allowed easy access to the economic resources available within the wider and more exposed floodplains of the major rivers.
A more universal pattern is the tendency for sites to occur predominantly along the south or south-east facing flanks of valleys, in preference to the north or north-west-facing slopes (Fig. 8.3) (Buchadeau-Kervazo 1982, 1986; White 1985). The explanation in this case is almost certainly related to simple climatic factors. South-facing locations inevitably benefit from the maximum exposure to sunlight, and enjoy correspondingly higher temperatures in all seasons of the year. Protection from winds was no doubt an equally important factor, in an area where the prevailing wind direction is mainly from the west, and where the coldest and harshest winds come mainly from the north (White 1985). This would no doubt have been an especially crucial factor during the winter months, when local temperatures during the colder periods of the Upper Pleistocene could well have fallen below — 20°C (see Chapter 2). Even today local temperature differences of up to 25°C can occur between the north and south facing slopes of certain river valleys in the Perigord (Duchadeau-Kervazo 1986: 57). Thus it is not surprising that almost all the known cave and rock-shelter sites, from both the Middle and Upper Palaeolithic periods, are located specifically on northern flanks of principal river valleys (see Fig. 8.3; Ducha-deau-Kervazo 1986; White 1985). The only clear shift in this settlement pattern occurs during the final stages of the Upper Palaeolithic sequence (late Magdalenian and Azil-ian), when year-round temperatures improved rapidly during the closing stages of the last glaciation (Duchadeau-Kervazo 1986:57).
A third, more variable pattern can be seen in the altitudes at which cave and rock-shelter sites occur above the floor of adjacent valleys. Some sites were located very close to the floodplain of valleys - in some cases in locations which were evidently subject to flooding during periods of high river level (for example in the lower shelter at Le Moust-ier and at Trou de la Chevre, Sandougne, Les Ourteix and La Quina). More frequently, however, the sites were located well above the river floodplains, beyond the reach of seasonal flooding. Several factors could have been involved in the choice of these higher locations. As Duchadeau-Kervazo pointed out (1982, 1986; see also White 1985) most of these sites would have benefited from longer exposure to direct sunlight than those in the valley-bottom locations (an especially significant factor during the late autumn and winter months) and would have commanded more extensive views over local valley habitats. As vantage points for observing the distribution and movement of game or the location of other economic resources, these high-level locations would have offered obvious advantages (Fig. 8.4).
There are sporadic sites at even higher elevations, in positions which would have required quite a strenuous climb from the adjacent valley floors. Some of the best documented examples have been recorded in the Céou valley, immediately to the south of the Dordogne - most notably the sites of Grotte Vaufrey and Grotte XVI, which stand at heights of around 100 metres above the adjacent valley floor (Rigaud 1982, 1988). These and similar sites (such as the Roc de Marsal and La Ferrassie, close to the Vézére valley) would probably have been more easily accessible from the neighbouring plateaux than from the valley habitats immediately below (Turq 1988b; Delporte 1984). As Rigaud (1988) pointed out, some of these (most notably the Grotte Vaufrey and Grotte XVI) seem to indicate much more sporadic occupation than sites in the more accessible valley locations.
4. Finally, there can be little doubt that the immediate availability of good-quality flint supplies was a major factor in the location of many if not the majority of documented cave and rock-shelter sites in the Perigord area. As noted earlier, both Turq (1988a, 1989a) and Duchadeau-Kervazo (1982, 1984, 1986) have pointed out that the distribution of Middle Palaeolithic sites seems to mirror in many ways the distribution of good quality flint supplies, while Geneste (1985, 1989a; Gen-este & Rigaud 1989) has maintained that the presence of raw material supplies within a distance of at most 2-4 km is an almost invariable feature of the sites studied by him in the central and northern Perigord. In several cases there are examples of cave or rock-shelter sites which are located effectively on the source of high quality flint outcrops, for
example at Combe Capelle in the Couze valley (Peyrony 1943), and at Moulin du Milieu in the Lot (Turq 1989a: 196). Perhaps more significantly, studies by Duchadeau-Kervazo (1982, 1986) and Turq (1988a, 1989a) have suggested that in areas where flint supplies are either scarce or of poor flaking quality, Mousterian sites tend to be more sparsely distributed, for example on the outcrops of Jurassic limestone in the northern Dordogne and on similar outcrops between the valleys of the Dordogne and the Lot (Figs 8.8-8.10). The clarity of these patterns leaves little doubt that the immediate accessibility of suitable raw material was a critical factor influencing the location of the majority of cave and rock-shelter locations in the southwestern French region.
Was this article helpful?