No doubt the enormous rock-art group of Twyfelfontein forms the most important concentration of engravings of Namibia. It is part of a long N-S running belt of rock-art (Figure 1). This band stretches from the border of Angola to the Orange River in the south, where it merges with the rock-art regions of the Republic of South Africa (RSA). Strikingly, this rock-art belt, running parallel to the coast, clearly separates the coastal Namib Desert, from the upland desert-steppe of the Kalahari; two areas almost completely bereft of rock-art, mostly because these areas lack suitable rock surfaces. The largest part of the rock-art belt is moreover characterised by having an interesting combination of engravings and paintings. In other parts of southern Africa these different techniques often seem to occur separated, although there are other important rock-art regions, for instance northern South Africa and southern Zimbabwe, where the distribution of paintings and engravings clearly overlap. Twyfelfontein (Site 1 in Figure 1 and Figure 3) is situated in the north of the mixed zone of Namibia and is also an outstanding example of a site where both engravings and paintings are found together. Interestingly, cupules also occur at Twyfelfontein, associated with engravings, not with paintings.
In view of the enormous array of rock-art sites in Namibia, cupules and cupule sites seem to be relatively scarce in this vast country (824.290 km2). However, it must be emphasised here that cupules are easily overlooked, also because people often think they are natural features (Bednarik 1993: 138) or grinding hollows. Therefore these small hemispherical depressions usually (and regrettably) get little attention. And although Scherz reported about 60 sites/panels with cupules in Namibia (Twyfelfontein counting as one site with 37 cupule panels), he did not state the exact number of cupules at each site. At Twyfelfontein about 1092 cupules (including all doubtful examples - see Appendix 1, Table 1) were known to exist by 2002, yielding an average between 30 to 40 cupules per panel. If we accept 40 cupules as an average for each known Namibian rock panel bearing cupules, the Namibian total may still be estimated to be less than 3000 cupules. If one compares this with the more than 50.000 cupules (Van Hoek 2001) in the much smaller area of the British Isles (316.000 km2), or the 20.000 cupules on minute Hawai'i Big Island (Lee & Stasack 1999: 186), one can state that cupules are indeed relatively scarce in Namibia. The statistical comparison of these three rock-art regions does not imply however that those cupule traditions also are culturally and chronologically similar. The majority of the cupules of the British Isles roughly date between 6000 BC to 1500 BC, those at Hawai'i date from AD 1200 to AD 1600, whereas some Namibian cupules may even date from 12000 BC to AD 1000.
The number of cupules at individual Namibian rock-art panels may vary considerably, also at Twyfelfontein, where boulder 4.3 has only one cupule (and no engravings or paintings), whereas boulder 4.7 (Figure 21) at Twyfelfontein has more than 200 cupules (and a large number of animal engravings).
Keeping in mind that Namibian rock-art has not yet been completely surveyed and that many researchers do not look specifically for cupules and that many cupule sites may therefore have been overlooked, some remarks about the distribution of Namibian cupules can still be made.
Firstly, there are, up to now, no cupule-sites reported in Namibia within the areas with only paintings (although some sites with only paintings, e.g. the Tsisab Gorge in the Brandberg complex, feature rows of painted dots that could perhaps functionally be compared with some of the rows of cupules at Twyfelfontein). The combination of cupules and paintings on the same panel is very rare in southern Africa (and indeed elsewhere in the world). Only at Site 6 in Figure 1, there is a combination of cupules and paintings on the same vertical panel, although Scherz does not clarify whether the two traditions are also spatially separated (Scherz 1972: 260; 1986: 221). At an outlying site (14 in Figure 1), at least 104 cupules of varying sizes have been recorded on a vertical panel together with remains of black paint, while nearby non-iconic engravings and a painting of a rhinoceros are found (Scherz 1986: 369).
In the Tsodilo hills of Botswana cupules co-occur in the same area featuring numerous paintings (Campbell, Denbow & Wilmsen 1994: 153) but those cupules have not been recorded in direct association with the paintings. In the rest of southern Africa (South Africa and Zimbabwe) there are only about eight sites where paintings of animals are associated with cupules. In this respect, mention must also be made of a newly discovered site at Tshikongomoti in the north of South Africa, where cupules are found on the same panel as paintings and engravings (mostly in the form of animal spoor - at least seven different species). Most of these cupules however, occur high up (up to 3.8 m above surface level), and there are no cupule-animal superpositionings. Three of the cupules are, however, found underneath thin, lenticular humanly abraded grooves (Sven Ouzman 2001: pers. comm.).
Secondly, it is remarkable that cupules seem to be almost absent in the south of Namibia, whereas the art in that area, roughly south of the line A-B in Figure 1, is dominated by non-iconic engravings. The only possible exceptions reported by Scherz so far seem to be the Sites 9 and 14 (see Figure 1). Site 9, however, features only doubtful cupules amongst more than 160 non-iconic designs. This general absence of cupules in southern Namibia seems to be an anomaly, as at Twyfelfontein and other sites in the north of Namibia cupules are an important element of some non-iconic designs. This may point to different authorship of this kind of geometric rock-art of (southern) Namibia and southern Africa as a whole.
Thirdly, the cupule sites seem to concentrate in two lesser bands that roughly run W-E. One band is formed by several sites near the capital of Windhoek (7 in Figure 1) and a group further east (8 in Figure 1). However, the cultural depressions at these sites often involve grinding hollows and depressions used as game boards.
The other band runs from Site 18 (see Figures 1 and Figure 3) via Twyfelfontein and another cluster of cupule sites (Sites 3, 4 and 6 in Figure 1) to a group near the town of Grootfontein (Site 5 in Figure 1).
The sites with cupules at Piet Alberts Kopjes near Kamanjab (10 in Figure 1 and Figure 3) and Site 11 (see Figure 1) may be part of a rock-art band that runs towards the north, where, near the Epopa falls (Site 12 in Figure 1), Scherz reported some cup-and-ring features that could also comprise true cupules. Much of the enormous area between Kamanjab and Angola, called Kaokoland, is unexplored and may yield many more rock-art sites, and some may possibly feature cupules as well.
It is not known whether these W-E bands indicate some sort of route from the coast to the interior, or whether this distribution pattern is related to the geomorphology of the area or is just accidental. Although this huge country has not yet been completely surveyed, it looks as if cupules are unevenly distributed within the Namibian rock-art zones.
This also seems to be the case for the distribution of cupules in Damaraland (Figure 3). In this vast area, Scherz recorded only few sites with cupules. I already mentioned Site 6 (see Figure 1), which has cupules together with paintings on a vertical face in a cave, but the relation between the paintings and the cupules is uncertain.
One cupule site, which overlooks the Ugab river valley (Site 3 in Figure 1), is exceptional for Namibia for not being associated with other forms of rock-art (although other rock-art sites occur in the vicinity). Site 18 (see Figure 1 and Figure 3) has nine animal engravings, but also features cupules on a nearby boulder. Also at Piet Albert Kopjes or Kamanjab (Site 10 in Figure 1 andFigure 3there is a spatial distinction (though not as strict as at Site 18) between the major cupule outcrop and the rocks with iconic art.
In one of the many rock-art gorges of the Brandberg (2 in Figure 1 and Figure 3), Scherz reported also a number of "Punktkreise" (cup-and-ring designs), that, like at Site 12, could comprise true cupules, but this is uncertain, as Scherz does not describe the nature of the "Punkte" (cupules or dots).
In the Erongo Mountain area (13 in Figure 1 ) there are numerous sites with rock paintings. At one site, Scherz also mentioned many small cupules on an isolated outcrop just north of a concentration of painted sites (his site 131N), possible cupules at another site (his site 147C), and many "ground depressions" at his site 152G. The few crude depressions at Site 17, SE of Twyfelfontein, are doubtful as cupules and only associated with non-iconic designs.