Discovery and Rivalry on the World's Greatest River
On the evening of February 18, 2012, while preparing for a six-month kayaking expedition in the Peruvian Andes, James "Rocky" Contos, Ph.D., discovered the true source of the Amazon River.

Jeff Moag


Erich Schlegel
Rocky Contos

Contos was at home in San Diego, marking every kilometer, known rapid and point of interest on the four upper Amazon tributaries he intended to run from top to bottom. Among these were the Apurimac, long celebrated as the most-distant source of the Amazon, and the Mantaro, until now distinguished only as the most-polluted river in Peru. Applying vector graphics to scanned topographic maps, Contos measured the Apurimac and the Mantaro from their most-distant sources to their confluence. The Mantaro was 50 miles longer.

Contos, a slender 41-year-old, is telling me this at a Peruvian restaurant in a San Diego strip mall. He's handsome in a way that doesn't attract notice, with a voice

that is high-pitched and slightly nasal. He speaks in neutral tones, as if he's observing our conversation rather than a part of it. When I ask him to describe the moment of his discovery, a touch of animation creeps into his voice.

"I tell my wife, 'Barb, I think I found something important here. The source of the Amazon is in a completely different place than people have thought for the past half-century.' And she says, 'I don't believe it. How could something so important be overlooked by so many people?'"

Contos is not a geographer (his doctorate is in neuroscience), but he was keenly aware that, if his calculations were correct, the experts had misplaced the source of the world's largest river. He confirmed the measurements using Google Earth, and began drafting a scientific paper titled Correct Placement of the Amazon Source in the Rio Mantaro Basin.

Substantiating his claim would require fieldwork to pinpoint the new source, and Contos wanted to make complete descents of the Mantaro and Apurimac in order to obtain GPS tracks of both rivers. It was a dangerous undertaking, comprising nearly 1,000 miles of paddling on two of the steepest and most remote big rivers in the Andes. Nearly half of that distance would be first descent in a region known for its sievey, dead-end rapids.

Contos had been preparing for just such a challenge since 1991, when he hitchhiked to Sacramento, bought a Perception Mirage for $300 and paddled 120 miles home. He's been exploring rivers ever since, most notably in Mexico, where he has completed more than 80 first descents. He published the definitive whitewater guide to Mexico and has become an underground legend for putting up big trips on minimal budgets.

Peru would be no different. Contos recruited paddlers for a cost-sharing descent of the Marañón, and emailed a select list of acquaintances seeking partners for the more challenging Apurimac and Mantaro. On March 23, barely a month before he was to leave for Peru, Contos came across the website of West Hansen, a 50-year-old marathon canoeist from Texas who was planning a speed descent of the Amazon from source to sea. Contos emailed him, and over the next few hours the men exchanged a series of messages under the heading "Apurimac Collaboration." Both raised the question of a potential rivalry and dismissed it, but scheduling remained a problem. Finally, Hansen wrote three lines that Contos could not ignore:


It's too bad you couldn't hold off your expedition until August. I could probably fund most of your trip in return for your company and gear collaboration. Just a thought. -w

Contos now had a dilemma. Was it fair to join Hansen's expedition and accept his help, knowing that he planned to start from the wrong source? Contos brooded on the question for more than a week. Then, in a Skype conversation on April 1, 2012, after securing Hansen's agreement not to tell anyone or to start his expedition from the new source in the Mantaro basin without his consent, Contos revealed his discovery to Hansen.
ccording to the most recent published study, the precise source of the Amazon is a spring at an elevation of 16,962 in the Apurimac

drainage of southern Peru. The place is called Quebrada Apacheta, and is marked with a plaque mounted on a cross of steel pipe. Some distance below this point, it is possible to float a kayak in a tiny rivulet that eventually becomes the Apurimac, and then the Ene, Tambo, Ucayali and, ultimately, the Amazon River.

The first person to paddle that route to the sea is a Polish émigré named Piotr Chmielinski, who began in August 1985 as part of a 14-person expedition. He finished seven months later with only the journalist Joe Kane, whose 1989 account of the journey became a bestseller. Running the Amazon describes the unraveling of an expedition in the face of difficult whitewater, hostile guerilla fighters, financial duress, and, most critically, the insecurities of its leader, Francois Odendaal. Reading between the lines in Kane's book, it's clear that Chmielinski's triumph owes as much to his masterful diplomacy as to his gritty toughness and smoldering ambition.

West Hansen first encountered the book in 2008, at the home of his friend, David Kelly. They were en route to Peru to compete in the Great River Amazon Raft Race, in which teams build balsa rafts and race them 87 miles through the jungle. Hansen finished the race, devoured the book, and in a matter of weeks decided to paddle the Amazon from source to sea. He began contacting others who had completed the route, and was delighted when Chmielinski suggested they speak on Skype.

"He had his bottle of wine on his end, and I had my bottle of wine on my end, and we talked for four hours. I was nobody," Hansen says. "But he took me as serious as the next explorer." Chmielinski provided Hansen the benefit of his advice and extensive connections, as he has for other adventurers, cultivating a network that seems to give him a sixth sense about the state of Amazon exploration.

Less than three weeks after Contos shared his discovery with Hansen, the Texan received a text from Chmielinski, asking to speak immediately about an important matter. Hansen called Contos. "I said, 'Rocky, if Piotr asks me point-blank about the source, what do you want me to say?'"

Contos decided to contact Chmielinski immediately. He knew that he was sitting on an extraordinary discovery, one that potentially put the first full descent of the Amazon back on the table. He also was acutely aware that Chmielinski had claimed that prize 27 years earlier, after unseating Odendaal as expedition leader. Contos was wary of the Pole, and in a tense telephone conversation, he told him so.

Hansen found Contos's tone-deaf approach to his mentor extremely disrespectful. The exchange seemed to punctuate reservations that Hansen already had developed about teaming with Contos. For his part, Chmielinski could not ignore the impact of Contos's discovery. When Contos told him he'd applied for a grant from the National Geographic Expeditions Council and had not received so much as a reply, Chmielinski was incredulous. He began to work his contacts at Geographic, and pledged that if the grant did not come through, he would personally contribute $3,000 toward Contos's field research. Six days after Chmielinski told editors of the potential new source, the Expeditions Council routed $7,500 through his nonprofit, Canoandes, to support Contos's source research. The organization also soon funded Hansen's Amazon Express expedition for $35,000. Members of both teams were required to give National Geographic magazine exclusive first rights to their stories. (Citing this agreement, Chmielinski declined to comment for this story.) The stipulation had an unintended side effect: It gave Contos the impression that the magazine would publish his account of the discovery.

Chmielinski's involvement had been like gasoline on a fire, injecting money and enthusiasm into both expeditions. His diplomacy also seemed to seal an alliance of sorts between the two men. On April 21, just two days after sharing his find with Chmielinski, Contos gave his blessing for Hansen to run the Amazon starting from the new source. Contos would depart immediately to pinpoint the source and complete the first descent of the Mantaro, and then join Hansen's whitewater team in August.

Maps published with Contos' paper in the journal Area depict the Mantaro and Apurimac rivers. According to Contos' scholarship, the most-distant source of the Amazon is Cordillera Rumi Cruz in the Mantaro headwaters, some 50 river miles farther from the Atlantic than the previously recognized source at Nevado Mismi in the Apurimac basin. Maps courtesy Rocky Contos
The explorer at the source.

Contos knew that he was sitting on an extraordinary discovery, one that potentially put the first full descent of the Amazon back on the table.

The Mantaro is the longer river, but that doesn't settle the question. Geography can be as much a political subject as a scientific one.

t the beginning of May, less than two weeks after their first fraught conversation, Chmielinski and Contos traveled together to the headwaters

of the Mantaro, where Contos ascended a horseshoe-shaped ridge above Lago Acucocha and erected a cairn of stones, called an apacheta. It marks the most distant point in the Mantaro basin and thus, according to Contos's scholarship, of the Amazon itself.

The next day, Contos began a 29-day, alpine-style first descent of the Rio Mantaro with James Duesenberry, a 27-year-old carpenter from New Hampshire who was one of two paddlers to act on Contos's invitation. The other, Boris Trgovcich, joined them on the easier section in an inflatable kayak. They boated conservatively, stopping frequently to scout, sneaking rapids they could easily have run through the center, and portaging often. They lunched on canned tuna and dry crackers and resupplied in villages every few days.

Ten days below Lago Acucocha Contos and Duesenberry entered the sheer-walled Cañon de Izcuchaca, and the river began to bare its teeth. There, as the kayakers scanned for a suitable campsite late in the afternoon, the river swept them around a bend and into a Class V maelstrom they would later call The Blur. Both men reached the bottom shaken, but unscathed. The next bend revealed an unbroken horizon line. "The entire river was just dropping down into this giant hydraulic," says Contos. "All I could do was line up to charge it."

Duesenberry was following a few boat lengths behind. "I saw Rocky drop in and get flipped and just get eaten by the hole," he says. "I hit it farther left and got cartwheeled once or twice and somehow was lucky enough to come out."

He estimates the hydraulic held Contos for no fewer than 60 seconds, and possibly twice that long. Duesenberry had time to catch an eddy, get out of his boat and, finding no vantage from which to get a rope to Contos, climb back into his kayak and snap the skirt. Contos was still in the hole, still in his boat. More seconds passed. Finally the boat flushed.

It was dusk in the bottom of a canyon; Duesenberry couldn't see much, certainly not Contos. He began to chase after the boat, which was barreling through a train of exploding waves toward God knows what. Contos flushed after it, his body starved of oxygen, his worn-out drysuit now full of water and dragging him down. "I thought this could be it," he recalls. "I didn't know what was downstream, and if I'm going to drop into another hole, I could black out."

Duesenberry glimpsed Contos's helmet from the crest of a wave and made a snap decision to let the boat go. The two had a breathless consultation in the eddy, Duesenberry asking, "What should I do? Do you want me to go after the boat?"

Contos, still stunned, said, "I don't know."

Duesenberry stayed. The Pyrahna M3 and its cargo—Contos's camping gear and food, his GPS receiver, MacBook Air, a video camera and copious river notes—drifted downstream through the night, as Duesenberry and Contos huddled among the rocks on river-left and talked about the risks they take to see such places.

The road was about 1,000 feet above them, on river-right. At dawn, Contos began scrambling along the left bank to the remains of a cable bridge. He crossed it hand-over-hand, climbed out of the canyon, and hitched a ride to the town of Izcuchaca, 22 river miles below the drop he had decided to name Contos's Hole.

Duesenberry continued alone, through miles of Class III and IV rapids. The next day, he found Contos's boat pinned on a rock and freed it with a Z-drag. Contos hiked in, recovered the boat, and continued downriver using Trgovcich's spare paddle, a 220-centimeter plastic model marketed to beginning sea kayakers. The most challenging whitewater was still to come.
Continue to Part 2

Contos hiked in, recovered the boat and continued downriver using a plastic paddle marketed to beginning sea kayakers. The most challenging whitewater was still to come.
Hole-riding in the Abyss.
Box-canyon roulette
Part 2