Congo

In 1872, Stanley passed near the Virunga region but did not enter it; in 1899, a German expedition went in, losing more than half its party. A privately financed Italian expedition disappeared entirely in 1911. There had been no more recent searches for the Lost City of Zinj.

"So no one has ever found it," Elliot said.

Ross shook her head. "I think several expeditions found the city," she said. "But nobody ever got back out again."

Such an outcome was not necessarily mysterious. The early days of African exploration were incredibly hazardous. Even carefully managed expeditions lost half of their party or more. Those who did not succumb to malaria, sleeping sickness, and blackwater fever faced rivers teeming with crocodiles and hippos, jungles with leopards and suspicious, cannibalistic natives. And, for all its luxuriant growth, the rain forest provided little edible food; a number of expeditions had starved to death.

"I began," Ross said to Elliot, "with the idea that the city existed, after all. Assuming it existed, where would I find it?"

The Lost City of Zinj was associated with diamond mines, and diamonds were found with volcanoes. This led Ross to look along the Great Rift Valley – an enormous geological fault thirty miles wide, which sliced vertically up the eastern third of the continent for a distance of fifteen hundred miles. The Rift Valley was so huge that its existence was not recognized until the 1890s, when a geologist named Gregory noticed that the cliff walls thirty miles apart were composed of the same rocks. In modern terms the Great Rift was actually an abortive attempt to form an ocean, for the eastern third of the continent had begun splitting off from the rest of the African land mass two hundred million years ago; for some reason, it had stopped before the break was complete.

On a map the Great Rift depression was marked by two features: a series of thin vertical lakes – Malawi, Tanganyika, Kivu, Mobutu – and a series of volcanoes, including the only active volcanoes in Africa at Virunga. Three volcanoes in the Virunga chain were active: Mukenko, Mubuti, and Kanagarawi. They rose 11,000 –  15,000 feet above the Rift Valley to the east, and the Congo Basin to the west. Thus Virunga seemed a good place to look for diamonds. Her next step was to investigate the ground truth.

"What’s ground truth?" Peter asked.

"At ERTS, we deal mostly in remote sensing," she explained. "Satellite photographs, aerial run-bys, radar side scans. We carry millions of remote images, but there’s no substitute for ground truth, the experience of a team actually on the site, finding out what’s there. I started with the preliminary expedition we sent in looking for gold. They found diamonds as well." She punched buttons on the console, and the screen images changed, glowing with dozens of flashing pinpoints of light.

"This shows the placer deposit locations in streambeds near Virunga. You see the deposits form concentric semicircles leading back to the volcanoes. The obvious conclusion is that diamonds were eroded from the slopes of the Virunga volcanoes, and washed down the streams to their present locations."

"So you sent in a party to look for the source?"

"Yes." She pointed to the screen. "But don’t be deceived by what you see here. This satellite image covers fifty thousand square kilometers of jungle. Most of it has never been seen by white men. It’s hard terrain, with visibility limited to a few meters in any direction. An expedition could search that area for years, passing within two hundred meters of the city and failing to see it. So I needed to narrow the search sector. I decided to see if I could find the city."

"Find the city? From satellite pictures?"

"Yes," she said. "And I found it."

The rain forests of the world had traditionally frustrated remote-sensing technology. The great jungle trees spread an impenetrable canopy of vegetation, concealing whatever lay beneath. In aerial or satellite pictures, the Congo rain forest appeared as a vast, undulating carpet of featureless and monotonous green. Even large features, rivers fifty or a hundred feet wide, were hidden beneath this leafy canopy, invisible from the air.

So it seemed unlikely she would find any evidence for a lost city in aerial photographs. But Ross had a different idea: she would utilize the very vegetation that obscured her vision of the ground. –

The study of vegetation was common in temperate regions, where the foliage underwent seasonal changes. But the equatorial rain forest was unchanging: winter or summer, the foliage remained the same. Ross turned her attention to another aspect, the differences in vegetation albedo.

Albedo was technically defined as the ratio of electromagnetic energy reflected by a surface to the amount of energy incident upon it. In terms of the visible spectrum, it was a measure of how "shiny" a surface was. A river had a high albedo, since water reflected most of the sunlight striking it. Vegetation absorbed light, and therefore had a low albedo. Starting in 1977, ERTS developed computer programs which measured albedo precisely, making very fine distinctions.

Ross asked herself the question: If there was a lost city, what signature might appear in the vegetation? There was an obvious answer: late secondary jungle.

The untouched or virgin rain forest was called primary jungle. Primary jungle was what most people thought of when they thought of rain forests: huge hardwood trees, mahogany and teak and ebony, and underneath a lower layer of ferns and palms, clinging to the ground. Primary jungle was dark and forbidding, but actually easy to move through. However, if the primary jungle was cleared by man and later abandoned, an entirely different secondary growth took over. The dominant plants were softwoods and fast-growing trees, bamboo and thorny tearing vines, which formed a dense and impenetrable barrier.