The world has been agog these last few years as evidence has mounted to suggest that Mars still has abundant water buried beneath its dusty surface, in the form of permafrost. Early in its history there are many signs of vast floods that carved huge meandering canyons and may have filled basins with moderately long-lived seas. Yet Mars has probably always been pretty cold, as it is now, and the most likely form that surface water would have taken is in glaciers; that is, if there was ever sufficient atmospheric water to precipitate snow. As on Earth, the likeliest places to look are in mountainous regions, and Mars is not lacking in very high places. By far the largest, and indeed they are the highest mountains in the Solar System, are the shield volcanoes of the Tharsis Rise, topping out around 18 km above the Martian version of the geoid. The volcanoes have gnarled surfaces, which until recently have been regarded by most as the result of volcano-related processes. Imaging of the Martian surface has stepped up several notches in resolution in recent years, and details of the small-scale features of the volcanoes are very clear. Above all else, they resemble aspects of the nearest analogue to Martian conditions on Earth – the Dry Valleys of Antarctica. Although the Dry Valleys are now largely free of ice sheets, they show many features of former glaciation, perhaps extending back 30 Ma to the Oligocene. Their frigidity has ensured that any glaciers there were frozen to the surface, rather than having zones of incipient melting at their bases. Such cold-based glaciers move sluggishly, and produce peculiar features. Among these are moraines produced by sublimation rather than melting of the ice – they evidence no reworking by melt water – and rock glaciers that are also products of sublimation and sometimes rest on relics of former glaciers. Probable examples of both occur on the flanks of the Tharsis volcanoes, together with weird track-like assemblies of concentric ridges, that are likely to have formed on the flanks of ablating glaciers as they reached a standstill and then retreated. (Head, J.W. & Marchant, D.R. 2003. Cold-based mountain glaciers on Mars: Western Arsia Mons. Geology, v. 31, p. 641-644). Interestingly, the relationship of the glacial features to impact craters suggests that glaciation took place during the period since about 1.8 billion years ago (the Amazonian phase of Mars’ history) when bombardment had slackened to almost terrestrial rates and liquid water was unable to form on the red planet. Of course, glaciers do not have to be made of water ice, and there is still a possibility that at such immense altitudes any glaciers might have been made of solid carbon dioxide. Head and Marchant speculate that some of the features might still sit upon relics of the glaciers. It could be a bit of a disappointment if future explorers of Mars landed there expecting a water supply.