A PRIME MOTIVE for a recent visit of mine to London was more research into exploration and mapping of the Amazon. The Map Room and Archives of the Royal Geographical Society are easier to access in person than from Canada's west coast - notwithstanding the best of co-operation from the Society's staff.

In particular, I wanted to inspect some maps produced for RGS official periodical publications, such as Journal of the Royal Geographical Society and The Geographical Journal, during the heydays of Victorian expansion and global exploration. Some of these important signposts on the route to geographical knowledge of remote parts of the world are still available for sale. But as yet they do not seem to have attracted the attention from the map collecting public that they merit.

During 1993-94 David McNeill at the RGS took on the task of sorting out seventy-eight boxes of a treasure trove of maps that had accumulated over the last century and a half. He recorded on the Map Room's computer examples of more than 1400 different maps, illustrations and diagrams, that had been stored unsatisfactorily and sometimes mislaid. The surviving items dated from an 1832 chart of Paumbaum in the Gulf of Manar, India, to a map of the Musandam Peninsula in Arabia made by the RGS expedition in 1973.

Most of the maps had been produced to accompany the reports of explorers and expeditions published in the RGS publications and had been printed in quantities deliberately in excess of those needed for folding and stitching into the journal. In this way a stock of "flat maps" was made available for interested persons to purchase. Somehow both knowledge of their existence and availability drifted into obscurity apart from the occasional sale to a new member of the Society.

They are now recorded on a computer database in the Map Room, sorted by region, location, date, author, and printing details. Stocks of some of them are at or below that considered necessary for archive needs, but the surplus has been made available for sale priced between £1 and £60. Francis Herbert, Curator of Maps (who initiated this project and who prices all of them), has identified some uncaptioned items and reports that sales have already contributed almost £25,000 to the Society's coffers.

One of the maps I sought recorded a landmark expedition as far as I was concerned. While it did make a valuable contribution to the map of the Amazon basin, it was of even greater importance to the overall story, due to the significance of the participants.

That landmark journey was the Roosevelt-Rondon Scientific Expedition of 1914. The map that interested me accompanied Colonel Theodore Roosevelt's account to the RGS, presented in person a few months after its completion and published in the February issue of the journal. I knew the story of the expedition; how Roosevelt, accompanied by his twenty-five-year-old son Kermit and a group of scientists from the American Museum of Natural History, had been taken under the wing of a renowned Brazilian explorer and early champion of native peoples, Colonel Candido Rondon of the Engineer Corps in the Brazilian army. A survey team, led by Rondon, had earlier discovered the headwaters of a potentially important water course. Since neither they nor anyone else knew where it emerged, he called it Rio da Duvida – the River of Doubt.

In a fine example of "it seemed a good idea, at the time" the Brazilian authorities proposed that Rondon escort the Americans down that mystery river. Roosevelt, who had recently completed his second term of office as President of the United States, accepted. With his "speak softly, but carry a big stick" bravura, Colonel Roosevelt cut an heroic, tough-guy figure throughout Latin America – where machismo in a leader was much admired. However, the image cloaked a man then in his mid-fifties, thick in the waist, out of condition, and with a heart problem, his eyesight weakened by years of paper work, desk-bound in the Oval Office.

The index map on the printed version of the "Map of the Rio Roosevelt from surveys by the Roosevelt-Rondon Expedition 27th Feb-30th April" and a section. It was a copy of this map which the author bought from the RGS. (By courtesy of the Royal Geographical Society)

The early stages of Roosevelt's expedition were carried out in well-equipped comfort. The Americans were conveyed to their rendezvous with Rondon aboard the president of Paraguay's personal gunboat. At Tapiapoan, Rondon's headquarters, they assembled a lavish safari to traverse the lightly wooded, upland savannahs of the Plan Alto and the Matto Grosso. A train of mules and over seventy pack-oxen were needed to carry their impedimenta.

During their approach march to the river's head, all went well. At night the scientists, well-travelled to a man, shared stories of adventure around comradely camp fires. Rondon recounted tales of piranhas, man-eating catfish, and other dangers of the region. Then they reached the River of Doubt.

As had many a previous expedition into the Amazon region, they found conditions became suddenly perilous. Progress on the river was painfully slow. In the first month they managed barely 65 miles [104 kms]. There was no knowing how long the remainder might take or how much food they would need. The flotilla required to transport the massive array of camping gear of the visiting scientists proved a liability as well. The seven long and heavy dugout canoes needed, proved awkward to manoeuvre down the many waterfalls, rapids, and extensive stretches of white water that they encountered early on.

The journey did not just involve paddling and scrambling down the river; the team carried out survey measurements as it went. Rondon's lieutenant, Joao Lyra, assisted by Kermit Roosevelt, carefully mapped the main stream and the position of significant features such as waterfalls, major rapids, and the streams entering from either side. Ornithologist George Cherrie collected many specimens of birds, including several species new to science.

The Duvida turned out to be truly a major river; almost 1,000 miles [1,609 kms] long, the largest affluent of the Madeira, itself the largest tributary of the world's largest river. It did not appear on the best maps then available. The rigours of a real survey expedition down such an unknown waterway proved nearly fatal for the titular head of the expedition. He suffered from malaria, dysentery, infected insect bites and scratches from thorns and was lame from a septic leg wound threatening to turn gangrenous.

Two of the Brazilians did perish during the trip and a third was lost in the forest, presumed dead. According to Cherrie's autobiography Dark Trails, Roosevelt became acutely aware that his physical condition so impeded progress that it jeopardized the survival of the whole expedition. His depression became so severe that at one point he requested Rondon to leave him and go on. One can imagine the international furore that would have erupted had Rondon emerged having abandoned an ex-President of the USA and recipient of the Nobel Peace Prize to a gruesome, lingering death in the jungle! Fortunately, Rondon and his lieutenant retained their composure, managing to nurse their guests to the lower reaches and safety.

At one point Rondon called a formal meeting, announcing that he had been duly authorised to rename the river in honour of their distinguished co-expeditionary. There exists some confusion over the name that was bestowed, early accounts describe the new name as Rio Teodoro, but the records became adjusted later to read Rio Roosevelt. Side streams were also named after Kermit and Cherrie and for the expedition members who had lost their lives during the journey.

Roosevelt published a book about his adventure Through the Brazilian Wilderness that would become a best seller. It was already in print by the time his report was published in the RGS Journal. The book included two maps illustrating the text, but they were preliminary and at small scale. For the RGS article, the cartographic department of the Society produced a far more detailed map from Lyra's manuscript provided by Roosevelt. It had been adjusted to fit the most reliable astronomically - determined coordinates available. It was this map, "Rio Roosevelt" that I had sought for my research and collection.

The map is in strip form – somewhat similar in format to John Ogilby's well known road maps - in six sections. The tributaries named for Kermit and Cherry (sic) are shown, as well as the features honouring the unfortunate team members. The map is printed in two colours; blue for the water and black for the toponymy and text.