Plug-in hybrids, which use electricity from the grid to replace gasoline for daily driving, would cut gas consumption and save commuters from high fuel prices. But some experts have been concerned that switching from gas to electricity, much of which is generated from fossil fuels, would actually significantly increase pollution in some parts of the country, as opposed to decreasing it.

A study released last week by the environmental group National Resources Defense Council (NRDC) and the largely utility-funded Electric Power Research Institute shows that plug-ins, once they're on the market, will significantly cut greenhouse gases. Across the country, the vehicles will on average also decrease other pollutants, but the impact in local areas will depend on the source of electricity.

In plug-in hybrids, a large battery pack that is recharged by plugging it in stores enough energy to power a car entirely, or almost entirely, with electricity for the first 40 miles or so of driving. For longer trips, the car reverts to conventional hybrid operation, relying largely on gasoline for power but improving efficiency: by storing energy from braking in the battery and using it for acceleration, for example.

The study shows that if plug-in hybrids are adopted widely in the United States, and if measures are taken to clean up power plants, by 2050, plug-in hybrids could reduce carbon-dioxide emissions by 612 million metric tons, or roughly 5 percent of the total U.S. emissions expected in that time frame, according to Marcus Sarofim, a researcher at MIT's Joint Program for the Science and Policy of Global Change. That's a significant amount, he says, considering that transportation accounts for only about a third of the total greenhouse-gas emissions.

But if plug-in hybrids account for only a small part of the total vehicle sales in 2050 (about 20 percent, compared with 80 percent in the first scenario), and if little is done to improve pollution from power plants, the vehicles will still reduce greenhouse emissions by about 163 metric tons, according to the study.

The picture for other pollutants is more complicated. The study found that on average, air would be cleaner. But in some areas of the country, such as those downwind from coal plants and having certain atmospheric characteristics, there could be a slight increase in levels of mercury, ozone, or particulate matter in the air by 2030, due to increased demand on power plants from plug-in hybrids. The best results would be in areas that use the cleanest electricity, such as in the Northwest, which relies heavily on hydroelectric power.

One implication of the study, says Luke Tonachel, an NRDC analyst, could be that federal incentives for encouraging the sale of plug-in hybrids should be adjusted by region. In some areas, the study shows, conventional hybrids may do almost as well as plug-ins at reducing greenhouse-gas emissions, while resulting in overall cleaner air. It would make sense to promote plug-ins more in other regions, where electricity sources are cleaner.

The report makes clear that turning to plug-in hybrids will mean changing the way the government thinks about vehicles. The standard miles-per-gallon EPA ratings will only give a part of the picture. What's needed, Tonachel says, is a new standard that conveys the impact of both burning liquid fuels and using electricity, such as a measurement of the overall amount of carbon that using a car will release.

(Offshore wind in the U.S.: We're getting close)