There is no railway to Camapuan. You can only get there by car or bus, driving out on the great straight road that runs north from Campo Grande to Cuiabá, in the cattle heartland of Mato Grosso do Sul. This is open countryside nowadays, stretching into the far distance like some great prairie, known here as cerrado. Only a few small clumps of woodland remain. This is well-fenced cattle land for the most part, thousands of white Cebu dotting the landscape.

Grigori Langsdorff

The road to Cuiabá takes a colossal pounding. An immense and endless stream of 14-wheeler lorries hurtle across the plateau like the waggons of an express train. The narrow tarmac track goes to the horizon straight as an Indian's arrow, and on the crest of the long undulations you can see for miles ahead. Stopping briefly, I counted 18 lorries ahead, and 10 coming the other way - and this was early on a Sunday morning. Driving along in a small car, dwarfed by these monstrous machines, is an unnerving experience, and it was a relief to turn off onto a lesser road to Camapuan. The huge volume of commercial traffic says something about the dynamism of Brazil, and particularly of its backlands. The area feels as though driven by some unseen engine, endlessly beating away, gathering strength over the past 250 years since gold was discovered near Cuiabá in the 1720s.

The heavy lorries that carry today's trade between Campo Grande and Cuiabá travel on the main road that lies to the west of Camapuan. In the gold rush of the 18th century the traffic came to the town by canoe. And there they stopped. Camapuan was the portage point between two great rivers - the watershed between the Paraná and the Paraguay - and a crucial stage on the long journey from the Atlantic to the heart of Brazil. At Camapuan everything was unloaded from the canoes, carried on the backs of black slaves from one river to the other, and reloaded again on the river far below. The work was done by the slaves, but it also gave employment to a small permanent staff of overseers that formed the nucleus of the small town that exists today.

Camapuan is a simple place, the kind of agreeable retreat where people might come for a quiet weekend. I took lodgings at the Grande Hotel, though it's not particularly "grand", just a few rooms round a large courtyard - and electric fans to keep the heat at bay. A cheerful, friendly town, tucked in out of the wind below the plateau, Camapuan has a population of about five thousand. The tree-lined main street has bumps to slow the traffic's speed, and runs down towards a bridge over the Camapuan river, an easily fordable stream. A few bars with billiard tables, a handful of bungalows in the outskirts with discs to receive satellite television, and that's about it. Drifting gently through the streets in the hot sun, I found a small restaurant in a side turning, its clientele spilling out onto the pavement. A party of twelve were celebrating a birthday.

The atmosphere is festive. Everyone has a gargantuan appetite. Nowhere in the world do people eat so much as in the backlands of Brazil. These are dishes to feed an army, and they all arrive at once. A plate of white rice; an earthenware pot of beans; great yellow chunks of yucca; a bowl of chicken, with whole sections freshly crisped; another bowl with bits of another bird bathed in chilli sauce; a dish of salad. With all this, and a bottle of Brahma Chopp, the ubiquitous Brazilian lager, I feel well-disposed to Camapuan by early afternoon.

In past centuries, Camapuan was more significant than it is today. An immense flotilla of canoes arrived here in October 1826, flying the imperial Russian flag. It was one of the first great European scientific expeditions in the 19th century to penetrate into the heartland of Latin America, this one funded by the Russian Imperial Court at St Petersburg. It stayed at Camapuan for several weeks.

The European fashion for exploring South America had been sparked off by the writings of Baron Alexander von Humboldt, who travelled through the continent between 1799 and 1804. The Russian Tsar Alexander I was persuaded to finance what he thought of as a comparable scientific expedition. He had a man on the spot to take on the task. Heinrich Langsdorff, born in 1774, was a Prussian aristocrat long in the service of the Russian imperial court. A naturalist through inclination and interest (he had located 1600 different kinds of butterfly), he was also an experienced traveller. As a member of an earlier Russian expedition in 1803-06, he had circumnavigated the world. He had worked as a botanist at the Academy of Sciences in St Petersburg, and since 1813 was Russia's consul at Rio de Janeiro. His fazenda at Mandioca, outside the city, was the natural port of call for the European scientists and travellers who flocked over to Brazil after 1814 (Auguste de St Hilaire, Johann Spix and Carl Friedrich Martius among them). In 1826, having persuaded the Tsar to finance his expedition, it was the turn of Langsdorff, at the age of 52, to travel into the heart of the continent.

(image source)

The river route to Camapuan had been taken by many before him. The first generation of Spanish conquistadores, exploring the waters of the upper Paraguay river in the 1540s, had hoped for silver, but what they actually found was imported from Peru. Two centuries later, in the 1720s, alluvial gold was discovered on the rivers Cuiabá and Coxim, tributaries of the Paraguay river north-west of Camapuan, and the Portuguese, pressing ever westward, were the beneficiaries of this new discovery. The Spaniards had missed their opportunity. At that time, this vast extent of territory was virtually unknown to Europeans. A few missionaries and slave hunters, and a handful of escaped black slaves, were the only outsiders to have moved into these Indian lands. Now a gold rush of classic dimensions began. The route to the gold fields through Camapuan began to hum. By 1726 more than seven thousand people were living in Cuiabá; a third of them black slaves.

Thousands made the journey. They flocked in from Rio, from São Paulo, and from Minas Gerais. Prospective miners and traders came one way, the gold made the return trip. It didn't always get there. When the heavy chests of gold arrived in Lisbon from Cuiabá in 1728, with the locks intact and the official seals unbroken, they were found to contain lead. In 1730, the Payagua Indians captured nearly 900 kilos - which found its way down to the Spanish city of Asunción. The long journey was a tiring and dangerous affair, but the profits were enticing. For miners, it was a gamble. For traders, bringing the commodities necessary for the expanding population, the rewards were huge. (In the 1860s, a horse worth 10 dollars in São Paulo would sell for 60 dollars in Cuiabá.)

To reach the land of the gold strikes from the Atlantic coast involved months on the rivers. Travellers would embark on canoe flotillas on the river Tiete, a tributary of the Paraná, at Vila do Porto Feliz, some 80 miles west of São Paulo. The flotillas would make the journey every year. Once on the Paraná, the flotilla would travel up the river Pardo, a tributary further north. Negotiating innumerable falls and rapids, intrepid traders would eventually arrive at Lake Sanguesuga, a marshy expanse of brackish water that lies on the plateau above Camapuan.

This is the watershed between the river systems of the Paraná and the Paraguay. The headwaters of two of their tributaries, the Pardo and the Camapuan, meet at the town, divided only by an extraordinary geological formation. The Pardo springs from the pond on top of the plateau, the Sanguesuga lake, while the Camapuan begins in a valley far below. (Eventually these two rivers flow into the Paraná and the Paraguay, which, initially moving far apart, come round in a great bight a few hundred miles to the south, to join forces at the Argentine town of Corrientes, and run on together into the River Plate.)

After my agreeable lunch in a Camapuan side street, I went in search of the headwaters of the river Pardo. Lake Sanguesuga was marked on my map, and I walked up the green sloping hillside above the town to look for it. There, on the edge of the plateau by the side of the road, lay a patch of green meadow with reeds around the edge. It looked more like a marsh than a lake. Past some large agro-industrial sheds, I turned off onto a track to the left, marked with signboards pointing a way to a string of fazendas, huge farms that lie along the banks of the Pardo river. Some way along, the path passes the southern end of the lake.

A small board by the roadside gives notice that the land here belongs to the Fazenda de San Antônio del Desembarque - "the unloading point" - the only sign that I could find with any reference to the historical significance of this obscure spot. Here the canoes that had paddled up the Tiete river were brought ashore, to be placed on wheels and drawn down the hill to the Camapuan by several yoke of oxen. Their cargoes was taken down in separate carts or carried on the backs of black slaves. An armed escort kept a look-out for Indians.

The portage in the middle of the 18th century was organised by Andreas Alvarez, a Portuguese landowner, who supplied the carts and provided the travellers with provisions from his own land: maize and beans, bacon and dried meat. The portage provided a moment of welcome rest, for apart from the danger from Indians and from rapids on the rivers, disease and famine was always a threat. The fazenda of Camapuan, wrote Spix and Martius, two Bavarians who travelled through Brazil shortly before Baron Langsdorff, "is often a place of refuge for the crew, who are frequently attacked with malignant fevers.” These, they noted, were caused “by incessant hardships and the damp foggy climate they have travelled through." A century earlier, in 1720, the entire flotilla had perished. Those who came after, wrote Robert Southey in his History of Brazil, found the canoes "with the goods spoiled inside them, dead bodies lying along the river banks, and hammocks slung with their owners lying dead inside them."

Langsdorff’s expedition, with forty people on seven canoes, had left the Tiete river in June 1826 and arrived at Camapuan four months later. Hercule Florence, an artist from Nice who recorded the expedition’s findings, described its “decaying state". The village had barely 300 people; a hundred were black slaves. Yet the portage system was still working: "the canoes are carried across by dry land across the hills for two and a half leagues, borne on enormous two-wheeled carts drawn by seven pairs of yoked oxen."

Camapuan was still a nice little earner, although by now the fazenda owner was an absentee landlord, a trading company with headquarters in São Paulo. Langsdorff sent a rueful report to the Academy of Sciences in St Petersburg, detailing the extent to which he had been exploited: "travellers barter at monstrous prices for salt, iron, gunpowder, shot, victuals."

Langsdorff's expedition remained in Camapuan for six weeks, waiting for the rains to fill the river. The canoes when reloaded were only able to take half the cargo they had carried up the Pardo. In November they moved on to the river Paraguay and on to Cuiabá. Like so many expeditions before and since, it ended in disaster. Everyone quarrelled, one of the three artists drowned, it split into two, and the younger scientists accused Langsdorff of being unduly preoccupied with a blonde German girl he had brought with him. Suffering from malaria, Langsdorff had a nervous breakdown. The expedition did eventually reach the Amazon, and paddled down to the mouth at Belem, from where they took ship to Rio. Langsdorff returned to Europe and lived on in Germany for another twenty years, and he died in Freiburg in 1852. Yet he never recovered sufficiently to publish an account of his travels, and since the other members of the expedition had agreed not to publish anything before he did, nothing ever appeared.

His scientific specimens were shipped off for eventual exhibition in St Petersburg, but the papers and the drawings were never published. Nearly a century later, Henry Manizer, a Russian ethnographer who came with his students to the river Paraguay in 1914, described how he had been fired with enthusiasm by the impressive collection of South American material he had found in the Academy of Sciences. But no one in St Petersburg had any idea how it had got there. No one had heard of Langsdorff.

I walked back down the hill from Lake Sanguesuga, following the route taken by the canoes and the slaves. The journey seemed less than the two leagues mentioned by Florence or the ten miles suggested by Southey. The road goes steeply downhill, the lake and the river being on two completely different levels. Fine for the portage down; not so good for the journey back. The river Camapuan at the bottom of the village street is a miserable rivulet today, just as shallow as it was two hundred years ago.

A party was held in the town square that evening, with staid young couples listening to a folk-rock band and occasionally venturing onto the dance floor. An old man sat outside a bar playing the guitar along the street, singing a Brazilian version of country music and joined by a crowd of enthusiastic admirers. Camapuan today is a lively and friendly place, but a day and a night is enough in these remote backwaters. Langsdorff and his expedition were here for six long weeks. Early the next morning, I took the busy road back to Campo Grande.

Richard Gott's book, Land Without Evil: Utopian Journeys Across the South American Watershed, is published by Verso at £18.95.