It's cold. Here on the North Dakota-Montana border they're predicting rain, then freezing rain, then snow, up to 15 inches. Not ideal fishing weather. Still, I have traveled to the confluence of the Missouri and the Yellowstone rivers with 15 wildlife conservationists and government scientists to search for a species perilously close to extinction. We're fishing for Scaphirhynchus albus, the pallid sturgeon.

They call it the "Dinosaur of the Missouri," although whether for its age or its appearance is a tossup. The pallid looks more at home in a natural history museum than on the end of a fisherman's line. It has a flat, upturned shovel of a nose; long, fleshy whiskers called barbels; a knobby back; and bony protrusions called scutes, rather than scales, lining the gray skin of its body. The fish, which can weigh 80 pounds, can live 60 years or longer. For about 70 million years, since the height of the dinosaur age, the pallid sturgeon and its ancestors reigned as a top carnivore in the vast river system that drains the middle portion of the North American continent. Pallids were harvested for their flesh and their caviar, like their Russian cousins, until they were listed as endangered in 1990.

The fish is perfectly adapted for cruising the nearly opaque waters. Its eyes have shrunk to the size of beans; instead of using vision, sensors in its head detect the electric fields of prey burrowed in shallow muck. Its scoop of a snout stirs up mud in search of small fish and aquatic insects. Tucked into the fish's flat white underside is a toothless mouth that folds in like a telescope, ready to shoot outward in a flash to suck up prey.

The trouble is, the river in which pallids flourished no longer exists. In Mark Twain's day, "the turbulent, bank-caving Missouri," as he called it, changed its path capriciously, especially during spring floods, which could be devastating. The river tore up trees and hid them below the surface, where they snagged passing boats. It was dangerous for people to live near, and sometimes fatal to navigate.

Twain wouldn't recognize today's Missouri, tamed as it has been by the U.S. Army Corps of Engineers over the past century. It's been deepened and straightened—which has shortened it by hundreds of miles. Its banks have been lined with dikes and rocky barriers called revetments. Six major dams punctuate its path. Flooding is far less common. The seasonal high and low waters are gone, as is much of its turbidity: the Big Muddy has cleared considerably.

The high waters of spring signal the pallid to spawn, but in these meticulously managed waters, this very endangered species has not reproduced successfully for years. There are fewer underwater tree snags where they can find food and fewer shallows where their larvae are safe from predators. It's estimated that fewer than 10,000 live in the Yellowstone, Missouri and Mississippi rivers from Montana to the Gulf of Mexico; those that survive are oldsters, and they are not being replaced by new generations. The population here, in the South Dakota and Montana river system, is down to about 200 aging adults.

"Plus or minus 70," says George Jordan.

Jordan, of the U.S. Fish and Wildlife Service, is the leader of the Pallid Sturgeon Recovery Team, the crew braving the weather today. The team's goal is to catch up to 38 pallids—ideally at least a dozen females—and transport them to hatcheries where hormone treatments will induce them to breed.

Five shiny aluminum motorboats are launched. Jordan teams up with Fish and Wildlife Service colleagues Steve Krentz and Bill Bicknell, and I hop aboard with them. Krentz pilots the boat, and Jordan and Bicknell will fish. All wear waterproof neoprene hip waders. In addition, Jordan sports nifty neoprene gloves with no fingertips. He's prepared to haul nets, draw blood delicately with a syringe or jump in the river up to his chest to rock a grounded boat free.

Yesterday, a boat with a radio telemetry antenna detected signals up the Yellowstone River from pallids that had been previously tagged with transmitters. Krentz steers the boat in that direction.

"Let's catch a fish," Bicknell says. To one end of a long net he ties a basketball-size orange buoy, then swings it in a circle and lets it fly. He and Jordan begin paying out net. Krentz cuts the motor. Ten minutes is as long as a pallid can be left in a net before it gets stressed. On Krentz's signal, Bicknell and Jordan haul in the net. Empty.

Drift after drift is cast, bringing up nothing but debris. A golden eagle flies overhead. Pelicans watch the boat from sandbars. Onshore, a single oil rig pumps crude. The roar of another boat at full throttle cuts through the quiet. As it speeds past us, naturalist Mark Nelson from Montana Fish, Wildlife & Parks smiles and holds up an index and middle finger. They've caught two pallids.

Jordan and Bicknell pull in a fish. It's not a pallid, but a smallmouth buffalo, perhaps two feet long. As Jordan works to free it, the fish gives a mighty flop, driving itself farther into the net.

"We're the government," Jordan tells it. "We're here to help."

It will be the only fish he'll catch today.

Later, onshore, two pallids caught by another team are brought in, and the scene takes on the atmosphere of an emergency room at the arrival of a heart attack patient. To minimize stress, the fish must be quickly transferred from tubs of water into a large transport truck. But first, a biologist waves an electronic scanner over one of the fish, and the scanner lights up with a ten-digit code. Rob Holm, manager of the federal fish hatchery in Garrison, North Dakota, flips through a thick notebook.

"We know this fish," Holm says. It's a male that was captured and bred a year or two ago, with offspring from the breeding program already in the river. To cut down on inbreeding, the team has to throw him back.

The other fish has no tag. He's going to become a daddy this year. Tape measures flash. Numbers are called out and recorded. Snout to tail, 143 centimeters (a little under five feet). A tag about the size of a carpet tack is injected into the base of the dorsal fin. Now it's pallid number 465043745B.

Reproductive physiologist Molly Webb makes an incision with a scalpel, then inserts an otoscope to find out whether the fish is male or female. "I see nice, white testes," Webb says before suturing the fish closed.

Meanwhile, somebody dips a bucket into the tub and pours the water back in, over and over, oxygenating it. Somebody else injects the pallid with an antibiotic. A third snips two tiny wedges of webbing from its dorsal fin with scissors—a tissue sample for genetic testing.

The scutes are hard and bony, and when I rub them the wrong way, they snag my fingers. Its white belly feels like a smooth, tough sheet of wet rubber. Its gills flare bright red, a sign of stress. A clear plastic stretcher is slipped under him, and he's lifted into a white fiberglass tank on the back of a truck. To make him feel at home, the tank has been filled with water pumped from the river ten feet away. The sturgeon dives into the brown water and vanishes.

Darkness falls. The day's catch: four pallid sturgeons, two of which had already been bred in previous years and had to be thrown back. Now the other two will travel for two hours through the night, in a driving rain, to the state hatchery in Miles City, Montana, where for the next three months they'll live in a 24- by 10-foot fiberglass tank. After two weeks, the recovery team will have collected 4 females and 11 males, far short of its goal.

From the outside, the hatchery looks like an airplane hangar, a vast structure with bay doors big enough to accommodate a sizable truck. Inside, a warehouse-like space is lined with a tangle of pipes and scattered with tanks ranging in size from large barrels to small swimming pools. It's clean but smells powerfully of fish. Our two pallids are now in a fiberglass tank filled with crystal clear water, probably the cleanest they have ever been in.

The team has been capturing and breeding pallids, producing as many as 100,000 young per year, since 1997. Most of them don't survive the first few days. Those that do are released, sometimes a few weeks after hatching, sometimes when they're older, six to nine inches, and less vulnerable. Over the years, the biologists have also been collecting data to determine the best age, time and places to release. As of now, there are more questions than answers.

Ultimately, the success of the recovery effort hinges on a bigger question: Will the Missouri River ever again be a viable place for pallids to reproduce on their own? Since the middle of the 20th century, the Army Corps of Engineers, which can turn the river on and off like a faucet, has managed it largely to make barge traffic reliable and safe. Over the years, though, most freight has switched from barges to trucks. Many believe the river can soon be restored to a more natural state.

Last year, under pressure from the Fish and Wildlife Service, conservation groups and the courts, the corps agreed to open the faucet a little. This past May, a small, controlled spring rise was released, in imitation of the river's historic cycle, in hopes that the mock flood would spur the pallid sturgeon to spawn. Conservationists are anxiously waiting to see if it worked.

For now, though, the Dinosaur of the Missouri is largely relegated to reproducing indoors. The next generation of pallids will grow up not in mud, but in fiberglass. Whether there will be other generations is uncertain. The pallid's future, like the waters it dominated for millennia, is opaque.

Sam Hooper Samuels is a freelance writer and a fundraiser for Smith College. He lives in Brattleboro, Vermont.