For students of the Big Rollover - the fast-approaching moment when demand for oil outstrips supply, with consequences politicians don't dare think about, far less mention - the recent upheavals at Shell were a taster of things to come.

Angry shareholders called for the chairman's head when he announced that 20 per cent of reserves Shell had classified as readily extractable were not so definitely there. Before the revision, Shell's figure for replacing the oil it was extracting was 105 per cent - slightly in surplus. Following Sir Philip Watts's announcement, this figure fell to 57 per cent - Shell's assets are being used up faster than they are being replaced. Watts and head of exploration Walter van de Vijver were forced to resign over the row.

For the moment Exxon, with a replacement figure of 116 per cent, and BP, with 152 per cent, are more comfortable. But this cannot last long.

The oil production curve follows the discovery curve, with about 40 years' delay; and the discovery curve peaked around 40 years ago. However many chairmen resign, nothing will alter this stark geological fact.

New oilfields are being found, but we need to find reserves equivalent to those in the recently discovered Angola field every 15 months, just to stand still. And we are not standing still - far from it. China, the world's most populous nation, is only just starting to motorise. So what is to be done?

The motor industry has been anticipating the situation. In February, James Rosenstein, a Toyota vice-president, told the House of Commons transport committee that fuel-cell vehicles running on hydrogen would be the mainstream cars of the future. The technology is already there: and "the finite nature of oil would spur on awareness of the need to develop this".

Car manufacturers have no problem in admitting the approaching end of oil: it merely represents a challenge. As scarcity kicks in, fuel economy will become a vital selling-point. Toyota's Prius, with its electric/internal combustion engine, gives 58.5 mpg. And hydrogen waits in the wings.

There is, however, a problem with hydrogen. Unlike oil, which gushes from the ground, hydrogen must be manufactured. This is currently achieved by mixing fossil fuels such as gas with steam to strip the hydrogen out: a wasteful use of resources, and hardly conserving hydrocarbons.

In the 1970s, oil companies became energy companies, concerned with all sources. Then North Sea gas came on-stream, the OPEC hold was broken, and the 1980s and 1990s roared away. Now, once again, an energy gap looms.

People are working to close it. Engineers at the University of Minnesota have discovered a reaction that turns ethanol into hydrogen, the beginning of what will become important research.

But how will it be distributed? The oil distribution network of pipelines, depots and filling-stations will need to be replaced: you can't pipe hydrogen long distances.

While oil remains cheap, there is no sense of urgency about all this. But when the Rollover comes, it will get very expensive, very fast. A lead-time of 25 years is generally taken to be realistic. If that is the case, then planning should already be well under way. As, of course, it may be: but if so, they -- whoever "they" may be -- are being uncharacteristically discreet about it.

Ruth Brandon is the author of 'Automobile: How The Car Changed The World', published by Macmillan

Search for used cars