A gasoline truck, a potential inferno in its belly, was amazingly intact, resting beneath a section of the Nimitz Freeway's upper level that miraculously stood while other parts of the roadway fell.
Nearby, a BMW sat unscathed, as shiny as a showroom special.
These were the exceptions, the remnants of normality amid an almost unimaginably grotesque scene of crushed bodies, mangled steel and crumbled concrete.
As many as 250 vehicles may have been trapped under the rubble. Bodies are only beginning to be extricated; it could be several days before the death toll is known.
There was no clear pattern to the Nimitz's collapse. Some columns along the one-mile stretch crumbled where they supported the freeway's upper level, about 60 feet above Cypress Street; others broke off closer to the first level or to the ground.
Dick Cademartori, a businessman who was at work next to the freeway, said the earthquake that "shook you like God was coming" leveled the Nimitz "like a sand castle."
He wasn't just being descriptive. The Nimitz was built atop mud and water-saturated soil that can shake like Jell-O in an earthquake, intensifying the temblor's wave--a key factor in its collapse. Its columns were constructed of "nonductile" concrete, a pre-1971 technology that leaves pillars too brittle to withstand much lateral sway.
"It's a well-known type of hazardous construction," said Fred Turner, a staff structural engineer for the California Seismic Safety Commission.
While the Nimitz's double-deck height probably contributed to its instability, other double-decked freeways in the Bay Area, such as the Embarcadero in San Francisco, withstood the earthquake because they were newer and built on more solid ground, engineers said.
Still, while the dangers of the Nimitz were known, engineers were stunned by the magnitude of the structural failure.
"We thought it might be damaged, but we did not think it would collapse," said William E. Schaefer, chief engineer for Caltrans.
When it was completed in 1957, the double-decked Nimitz Freeway was the pride of Oakland--a forerunner of other freeways that would ring and traverse the city. It was a major artery to the San Francisco-Oakland Bay Bridge, and carried 195,000 vehicles a day.
But as it was reduced to a graveyard, the Nimitz accented its outdated technology, and focused attention on an aging road system that is simply too expensive to fix all at once.
"Where do you start?" wrote Ian G. Buckle, deputy director of the National Center for Earthquake Engineering Research at the State University of New York at Buffalo, in a newspaper editorial. "There are more than 13,000 bridges in the (California) state system. It is simply not possible to upgrade all of them simultaneously."
Could it have been fixed? Some scientists say yes, if there had been a bigger push to correct problems that everyone knew existed. Gov. George Deukmejian, calling for a full inquiry into the freeway collapse, said he was "displeased and disappointed" to discover that not all of California's freeways are seismically safe--and wondered aloud if the Nimitz's construction had been substandard.
"You can tell that obviously something went wrong," Deukmejian said.
But other engineers, from within the state and around the country, said California's highway engineering is already the most advanced in the nation, and a solution to the problems posed by aging, massive structures such as the Nimitz is simply unavailable.
The Nimitz was shored up with steel cables in 1977 as part of a Caltrans effort to reinforce more than 1,200 bridges, overpasses and elevated road structures. The $58-million effort, spurred by the 1971 Sylmar earthquake, was completed two months ago.
But the steel cables only help keep spans of freeway decks from slipping off their supports. They do not strengthen the brittle concrete columns, spaced about every 75 feet, which must support 500-ton sections of freeway.
In the second phase, a $65-million project to improve structures, Caltrans plans to place steel wrapping around the bases of elevated freeways that are supported by single columns. That project, expected to run through 1992, would not have included the Nimitz.
The third phase would bolster multicolumn structures such as the Nimitz and Bay Bridge--but engineers at UC San Diego, where tests are being conducted, are still not sure what the best method of doing that would be.
Caltrans engineer Schaefer said: "We really don't have the expertise to know what to do."