Bearing in mind that the most serious conservation threat to rock art is the one posed by humans, it is interesting that our most sustained conservation efforts seem to be directed at the eradication of lichens. The principal reason given is that they interfere with our aesthetic appreciation of rock art. It is also claimed frequently that they might affect the conservation of rock art, although the evidence presented for this seems to be largely anecdotal. The chemical effect of lichen on most rock substrates is probably negligible, and their principal contribution to the deterioration of rock is probably their longer retention of moisture relative to lichen-free rock. While the degradation lichens are claimed to cause remains largely unproven (Cochran and Berner 1993; Drever 1994), archaeologists and site managers around the world have waged chemical warfare on lichens, or scrubbed them away wherever they encountered them. Although their stated main motivation concerns ‘aesthetics’ (and one might reasonably argue that lichens also have aesthetic qualities), the subconscious reasons are perhaps more complex. In my view they are related to the perception of rock art as art, which must be kept free of ‘blemishes’, and to the phenomenon identified by Freeman (1994). Freeman found interesting connections between some aspects of rock art studies and religious practices concerning shrines and ‘sanctuaries’. Psychologically, rock art sites need to be kept neat and free of the threatening anarchy of Nature, particularly in countries whose societies have been conditioned by religious values such as those of puritanism. However, lichens can assist in the scientific study of the rock art (see Lichenometry) and rock art researchers have begun to speak out against the vandalistic practices of lichen removal (Bahn et al. 1995; Tratebas and Chapman 1996; Walderhaug and Walderhaug 1998).The methods used in the eradication of these ‘pests’ are particularly questionable. After all, the easiest and most obvious way of killing lichens is by depriving them of light. They can be covered by black plastic sheeting or, as has been done on Scandinavian pavements, by covering them with a thin layer of sand. Yet most publications on rock art conservation recommend instead a variety of highly toxic fungicides, requiring the wearing of protective clothing and masks (e.g. orthophenyl phenol in dehydrated ethanol), ammonium hydroxide (which destroys most forms of life), and fluorosilicates of zinc and magnesium. Some archaeologists have used bleach, compressed water jets and vigorous manual scrubbing ‘with wooden tools’ to remove even very light lichen growths on hundreds of quite fragile schist panels. Obviously this would have eliminated not only the lichens, but everything else on the rock surface, including paint traces, patination, mineral accretions, and even poorly anchored rock fragments. In addition, it compromised the research potential of the rock art so much that meaningful analytical work may no longer be possible at these sites.

Unless there is unambiguous evidence that the lichen concerned is threatening the survival of the rock art, specifically rock paintings, it is preferable to ignore lichens, or to remove only loose and exfoliating thalli and discourage growth indirectly. Dead lichens can be removed without vigorous mechanical action, and contamination with chemicals should always be avoided. Most importantly, all lichen thalli must be fully recorded before they are removed, and the species must be identified. Tratebas and Chapman (1996) have provided detailed guidelines for the correct procedure in lichen treatment. Mosses, however, are much more effective in retaining moisture than lichens, but they are also easier to remove without causing chemical or mechanical damage to the rock surface. At some sites in the Sydney sandstone region, organic material (presumably lichens, fungi and algae) has been mechanically removed from the grooves of petroglyphs to highlight them for the benefit of site visitors. This practice was, however, questioned and discontinued. Young and Wainwright (1995) have shown that algal growth can depend significantly on tree shade, and experimented with seven algicides on several species they identified on the marble pavement of the Peterborough site in Canada.

Plant roots can have corrosive effects on rock, especially through the organic carbon dioxide exhaled by the micro-organisms living on the root surfaces. In some cases tree roots have caused extensive damage to rock art panels, particularly in limestone caves. A panel of finger markings in Prung-kart Cave, South Australia, was invaded by the root system of a pine tree (an introduced species in Australia) which followed the interface between the primary rock and a precipitated travertine layer bearing the rock art. This resulted in the exfoliation of much of the travertine lamina (Bednarik 1998). Other trees above the cave were then felled as a preventative measure. The activities of trees growing above limestone caves need to be monitored, even where the roof thickness is 20 or 30 m, and this applies particularly where a new species has been introduced in an areaorder Flomax.

Similarly, all other conservation threats posed by plants are relatively easy to deal with. Plants whose parts rub against decorated rocks can be removed, and it is in any case preferable to remove most plant growth within a few metres of rock art. By far the greatest danger plants pose to rock art is through the fire threat they pose, particularly in regions where plant matter tends to be high in resins and oils.

There are two strategies to deal with termite infestation of rock art sites: to either destroy the central nest, which eliminates the integrative mechanisms on which the colony depends for its survival; or to place a chemical barrier at the base of the panel (a strip of soil is soaked with an insecticidal emulsion). The former method is clearly preferable, however. It is advisable to destroy all nests within 50 m of the art, and to return after a year to check whether any structures have been rebuilt (Watson and Flood 1987). order BentylThe galleries traversing the art panel should not be removed. Nests of mud-daubing insects, however, should be removed (after first wetting them, or during the wet season in the tropics; but see below), because it has been found that existing nests attract new ones (Naumann and Watson 1987).

Some birds, too, construct mud nests at rock art sites in many parts of the world. Reports of damage to rock art are rare, however. In contrast, damage occasioned by large mammals is very common throughout the world. By far the most frequent culprits are domestic and feral grazing species. They favour rockshelters because of the shade and shelter they offer, they rub their bodies against the walls, depositing organic and sedimentary materials, they churn up archaeological deposits and walk on petroglyph pavements. In Australia, the significance of this threat is well appreciated and stock fences have been erected at numerous sites, particularly in the north of the continent.

REFERENCES Bibliography of Rock Art Conservation