The "Hydrogen Economy" fades ever farther into the future.

Two years ago, Bush launched a five-year, $1.2 billion program to develop a commercially viable hydrogen fuel-cell car.

Congress has been even more generous. The House energy bill authorizes $4 billion over five years for hydrogen research and another $1.3 billion for a new-generation nuclear reactor that would produce hydrogen for cars as well as electricity. The Senate, which is at work on its version of the measure, allocates $3.8 billion to hydrogen.

However, a growing number of scientists and energy experts believe that it will take decades to overcome the technological and infrastructure hurdles facing commercialization of hydrogen cars - if they can be overcome at all.

Not the least of these hurdles is that fact that present fuel cell designs require platinum, which is an extremely rare element. If all the cars on the road were outfitted with fuel cells it would exhaust all known supplies of platinum. And present versions of the fuel cell only last for a couple of years.

Studies last year by the National Academy of Sciences and the American Physical Society concluded that commercially viable hydrogen cars would take considerably longer - about 20 to 30 years - and cost more to develop than had been anticipated.

Similarly, researchers are about 20 years away from producing the kind of high-temperature nuclear reactors envisioned by the Department of Energy for the large-scale production of hydrogen - again assuming technical hurdles can be overcome.

Critics of the spending on fuel cell research say that raising fuel-economy standards for today's cars, increasing incentives for hybrid-gas-electric cars, funding research to allow "plug in" hybrid cars powered primarily by electricity and promoting alternative fuels like ethanol and biodiesel would reduce foreign oil dependence faster.

"We don't object to using R&D to do either basic research or research into something that may be in the distance," said Dan Becker of the Sierra Club. "What we object to is failing to do something now and using the R&D as a shield against doing something responsible today."