Earthmovers dumped the last load of dirt on Oroville Dam in 1967, completing the backbone of the State Water Project.

That same year, the fish hatchery about five miles downstream welcomed its first bewildered run of salmon.

Like their ancestors, the fish sought the upper reaches of the Feather River watershed. But instead of thrashing upstream into ever smaller creeks and rivulets to seek a mate, members of the spawning class of 1967 found themselves leaping up an aquatic staircase to an indoor holding tank. There, after being stupefied with carbon dioxide, they were killed. Hatchery workers stripped the eggs from females and squeezed milt from the males, mixed it all in a bucket and transferred the fertilized eggs to plastic trays.

Forty years later, this process continues at Oroville, as it does at hundreds of other sites throughout the West. Artificial reproduction may lack the dignity, mystery and drama of its natural analog, but it is efficient: Hatcheries, constructed mainly to offset the fish-killing propensity of modern dams, release a collective 5 billion juvenile Pacific salmon and steelhead annually. The one at Oroville, about 60 miles north of Sacramento, produces millions of young salmon and steelhead each year from as few as 11,000 adults.

The hatchery system is based on the utilitarian premise that a fish is a fish: As long as there are enough to be caught for fun or profit, the process and the habitat that produced them are irrelevant.

New research, however, confirms what critics of the hatchery system have long argued: The process matters. Hatchery fish are not the same as wild fish, and the difference is not solely spiritual or aesthetic. Breeding fish in captivity, besides eliminating their awe-inspiring reproductive journey, reduces their fitness to survive in the wild.

The limitations of fish hatcheries, particularly those constructed as compensation for giant dams built in the middle decades of the 20th century, have been apparent for many years. Despite construction of a vast network of these fish factories — there are about 100 hatcheries in the Columbia River basin alone — every major western watershed has seen its salmonid populations dwindle, some to the point of extinction.

Those fighting to restore imperiled salmon and steelhead populations have long argued that hatchery fish are more prone to disease and less fit to survive than wild fish. In their view, hatcheries make things worse, not better, diluting the gene pool and rendering the species increasingly dependent on artificial breeding to maintain the fiction of a viable population.

Those who benefit from the water, energy and transportation services provided by dams argue that hatchery fish are indistinguishable from wild fish and should be counted when regulators try to determine whether a species is imperiled.

Federal courts have ruled both ways, one judge declaring that hatchery fish should be counted, another one saying they should not.

Last month, however, the journal Science published a study by researchers at Oregon State University that comes down squarely on the side of those who believe captive-reared fish are fundamentally different from their wild cousins.

The research, involving steelhead on Oregon's Hood River, found that fish reared in a hatchery are nearly 40 percent less likely to produce offspring than wild fish. Hatchery-born fish whose parents also were reared in a hatchery had only half the reproductive success of wild fish. The study results were published Oct. 5.

The conclusions should not surprise anyone familiar with the concept of natural selection. The spawning journey of wild salmon and steelhead — an arduous, dangerous trip past predators, waterfalls and other obstacles — is not merely poetic in its rigor. It is a ruthlessly effective test of ecological fitness. Those that survive it produce offspring that must endure their own peril-filled journey back downstream. Only the strongest and wariest survive both ends of this migratory journey to pass along their DNA.

Hatchery fish, on the other hand, pay no price for being weak and slow-witted; food pellets fall from the sky on the robust and the wimpy alike. If held in captivity and allowed to breed, they are all equally likely to produce offspring, regardless of how well adapted they might be to life in the outside world.

Dumping the offspring of such fish into the hostile natural environment is like tossing handfuls of sardines into the seal tank at the zoo. It is no wonder that of the billions of hatchery fish released into western rivers each year, only a relative handful ever return and attempt to spawn.

— John Krist is a senior reporter and Opinion page columnist for The Star. Visit www.johnkrist.com. His e-mail address is jkrist@VenturaCountyStar.com.