If the great salmon runs of the Pacific Northwest become another dodo bird, you can't blame the Columbia River Indian tribes or their allies in the scientific world. But the portents of the summer of 2005 are ominous. The snowpack in the northern Cascade Range is near a record low, and the National Weather Service predicts the third-lowest runoff in a century. Fisheries biologists warn that endangered salmon stocks swimming home from their five-year migration to Japan face a tragic die-off.

Determined to avoid another disaster made possible by federal fisheries policy, three Columbia River tribes last week joined state and federal scientists and conservationists in calling for a legal showdown with the Bush administration over the fate of the salmon stocks. They asked federal courts to force the government to provide enough water this summer for the fish to migrate to their upstream spawning beds. The administration is equally determined to use the water to generate hydroelectric power, a move that would, in effect, dismantle provisions of the Endangered Species Act.

The contest is anything but a routine battle between old adversaries. Beyond the dams, the fish counts and the blizzard of court documents, this is an Orwellian struggle between two worldviews: one that would sacrifice economic advantage for coexistence with all species and one that would claim dominion over all species and earthly resources.

For three decades, from Northern California to British Columbia, tribes and fisheries biologists have battled to protect the salmonid's riverine habitat and high-country spawning grounds. Politicians have fought back to safeguard dam-based economic interests and their constituents. The impasse has jeopardized the Endangered Species Act and kept the salmon extinction clock wound tight and on schedule — 2017.

The clock began ticking when the Army Corps of Engineers closed the flood gates on four Snake River dams in the early 1970s. Fish counts soon plummeted. As salmon-recovery costs exploded in the 1980s, the anadromous icon of the Pacific Northwest morphed into a political football with fins.

Phase 1 of the battle came to a symbolic fork in the river in 1991. That autumn, a lone sockeye reentered the Columbia River near Astoria, Ore., and made the 700-mile swim upstream to its place of birth at Redfish Lake, in the Sawtooth Wilderness of central Idaho. Just before reaching its destination, the fish flicked its tail and swam into a net. Biologists scooped it up, nicknamed the fish "Lonesome Larry," then whacked it on the head and collected its sperm. The fish's varnished carcass was mounted on a pine board and hung in the governor's office in Boise, Idaho.

The parable of Lonesome Larry eventually proved that half-measures can fail at an astonishing cost. During Phase 2, approximately 50 of Larry's descendants, reared and released under the auspices of a federally funded program, returned to Redfish Lake as adults — at an estimated cost of $500,000 per fish. The price tag for restoring one spawning lake was so high that federal biologists, in the mid-1990s, started to talk openly about removing the Snake River's dams. The only surefire way to restore vitality to the fisheries, they contended, was to let the rivers flow like rivers. But if the "best available science" was pushing for dam removal, the political gloves were coming off.

Once upon a time, both sides in the salmon wars agreed that "the best available" science would play the lead role in designing recovery plans. This compact assured all disputants that, regardless of how heated the argument became, any solution would be anchored to biological realities in the rivers and lakes.

Then one day in spring 2004, officials of the Bonneville Power Administration, a semi-autonomous agency responsible for distributing power generated by dams on the Columbia and Snake rivers, asked for a meeting with the Columbia River Inter-Tribal Fish Commission.

This commission, the official voice on issues pertaining to salmon, holds treaties that guarantee tribal fishing rights in perpetuity. The Bonneville Power Administration's dams constitute the single largest threat to those treaty rights. At the meeting, the agency informed the tribes that the annual "summer spill," an event timed to help juvenile salmon reach the sea from mountain lakes in Idaho, was not going to happen. The water would instead be used to generate electricity.

Phase 3 of the battle had begun.

To the tribes, this phase displayed all the earmarks of an endgame. Although time was running out for the fish, the power people had little to lose. If the gambit worked, President Bush could take credit for lower utility rates in his reelection campaign. If it failed, he could accuse judges of caring more about fish than people.

In early July, the commission asked a federal court to reject the power agency's "emergency request." Later that month, U.S. District Judge James A. Redden ordered the Bonneville Power Administration to carry out the planned spill. The ruling, upheld on appeal, was hailed by scientists and tribes as a landmark victory for salmon and treaty rights.

The consensus among the region's aquatic biologists and natural-resource economists is that money alone will not restore salmon runs. Their reports and recommendations to the National Fisheries Service have underscored the results of a yearlong investigation of the science and economics of salmon recovery conducted by the Idaho Statesman in 1997. The newspaper concluded that the billions of dollars spent to restore salmon runs would be money down a rat hole until the Snake River dams were removed. Also, the region's economy would derive greater benefits from a healthy fishery than from hydropower and irrigation receipts.

In November 2004, the Bush administration unveiled its long-awaited salmon recovery plan. Earlier plans recognized that native fishing was an integral feature of the "environmental baseline" of the last 10,000 years. Dams were considered "impacts" to that baseline.

The Bush approach turned these biological models upside down: Dams pose "no jeopardy" to salmon because they are natural features of the landscape, and native fishing should be regarded as an "impact."

"This thing doesn't even meet the straight-face test," said Rob Masonis, regional director for American Rivers, a conservation organization. "What will they say next? That the future of the passenger pigeon looks bright?"

Administration officials contend that new "fish-friendly weirs" will indefinitely forestall the salmon's extinction. Fisheries biologists counter that these weirs, which cost $10 million each, have increased fish counts by less than 1%. Even under optimal conditions, the biologists say, this cannot begin to sustain the fish population at its current levels.

If politics trumps science, the Indians will not go away empty-handed. A 1995 treaty between the Pacific Northwest tribes and the Canadian and U.S. governments guarantees tribes a lump-sum payment of $10 billion for the destruction of their livelihoods. Biologists, with one eye on the salmon extinction clock, have urged that Snake River fish counts be posted every morning in the Oval Office and on the floor of Congress. Not a bad idea.

The courts will probably throw out Bush's salmon recovery plan and order fisheries scientists to come up with another rescue scheme. The fix will cost millions of dollars and years of time. And while the extinction clock ticks toward 2017, a catastrophic die-off in the summer of 2005 could push wild salmon stocks beyond the point of no return.

While we await the outcome, we can humor ourselves with Mark Twain's verdict on politicians: If they started out dead, we'd all be better off because they could set off on the road to honesty that much sooner.