They're cool, macho and powerful, and even cartoon mom Marge Simpson fell under the empowering spell of a sport/utility vehicle (SUV), her beloved fictitious "Canyonero." But the latest round of compatibility crash tests between cars and SUVs is no laughing matter.

The National Highway Traffic Safety Admin. reports that the probability of fatality is two to four times greater when a car is struck by a larger SUV or other light truck than by another car. The agency discloses its findings at the Society of Automotive Engineers International Congress and Exposition in Detroit in early March.

NHTSA offset-crashed a 1998 Chevrolet S-10 pickup, 1997 Dodge Caravan, 1997 Ford Explorer SUV and 1997 Chevrolet Lumina into a 1997 Honda Accord with both vehicles moving at 35 mph (56 km/h).

The Explorer, the heaviest (4,553 lbs. [2,066 kg]) and highest riding of the vehicles tested, imposed the greatest risk of injury to the Accord driver. Among striking vehicles, the Lumina posed the greatest injury risk to its own driver.

The findings bolster earlier research by the University of Michigan Transportation Research Institute, which found that 2,000 fatalities would have been prevented in 1996 if SUVs and other lights trucks and vans were replaced by same-weight passenger cars.

"The Michigan study is testimony to the fact that many lives can be saved if we can use the tools we are creating to build less-aggressive vehicles," NHTSA Administrator Dr. Ricardo Martinez says.

The results demonstrate that geometry (or ride height) and stiffness, in addition to the vehicle's weight, contribute to the "aggressivity" of light trucks. Automakers failing to recognize this and integrate it into future vehicle development programs could face government regulation in the future, says William Hollowell, chief of NHTSA's Crashworthiness Research Division.

Industry analysts argue that highway deaths and fatalities from auto crashes have dropped from 47,878 in 1977 to 41,967 in 1997, even as the number of light trucks on the road has increased.

However, once staunch opponents of NHTSA's past findings, automakers seem to be turning the corner.

"This testing once again proves that mass matters," says Terry Rhadigan, manager of safety communication for GM. Automakers such as Ford have already taken steps to address safety concerns.

The new Excursion SUV (see sidebar, p.39) includes a "BlockerBeam" installed below the front bumper to keep cars from sliding underneath in a collision. "I think, personally, that's the right direction," Mr. Hollowell says.

Keeping the heat on automakers, the agency is planning to add another test in the coming months to examine the crash compatibility of a large pickup with the Honda Accord.

One surprising result came in the Caravan-Accord test. The passenger front air bag in the Caravan deployed with such force - more than three times the acceptable federal standard - that it indicated a high probability of serious injury, perhaps a broken neck, or death for a small female.

NHTSA researchers are now studying the Chrysler minivan air bag system and the crash data and will report their findings, likely by this summer.

A DaimlerChrysler Corp. spokeswoman says the company is outraged that NHTSA waited eight months to share the information about the minivan result. "We are taking NHTSA's concerns seriously and will work with them as we review the data."

Meanwhile, NHTSA is requiring manufacturers to provide more visible warning labels in SUVs about the risk of rollover. SUVs' height, along with other factors, contributes to a rollover rate of 98 fatalities per million registered vehicles compared to only 44 fatalities per million for all other light vehicle types, NHTSA says.

More than 60% of the SUV occupants killed in 1997 died in crashes when the vehicle rolled over.