With equal cleverness, the Triton refugees now hoped to establish the seeds of a new colony, undetected beneath the clouds of Neptune. They waited on what seemed to be an alpine snowfield that was stained an impossible pink under Neptune’s ghostly blue light.

Some ninety years earlier, just ahead of the first Voyager flyby, Richard Tuna had predicted an entirely different landscape. He reasoned that Triton’s orbit was so far flung from the Sun that liquid nitrogen would not evaporate and blow away on the solar wind, and that lakes, perhaps even oceans, of nitrogen might lie pooled on the surface. As Voyager II approached, he had expected to see a world very similar in appearance to his own Earth. “But nothing will be quite as it appears,” he had told CNN. “The continents will be ice. The icebergs will be methane. And if—incredibly—carbon and hydrogen have become life, the ‘water’ that flows within the tree trunk will be nitrogen.”

This was nonsense, of course, and Tuna would have been the first to admit that eighty percent of his ideas were nonsense. “Eighty percent of everyone’s ideas are nonsense,” he had said. “The trick is in knowing which eighty percent and in being quick to abandon them.”

That, too, was nonsense—but only half nonsense. What made Tuna different was that he had twenty times more ideas than everyone else; and while this produced twenty times more nonsense, he had learned never to cling stubbornly to a wrong idea—even if it had been a pet one.

“There’s another complaint I have about God,” Tuna joked when the first Voyager photos of Triton came through and proved his ocean theory wrong. “If we are to believe God really did have a hand in the creation of the universe, why would he make a mistake like that? All he had to do was place Triton ninety million kilometers closer to the Sun—a matter of just five light-minutes—and we would have seen my beautiful nitrogen seas. So close to a miracle and yet so far.”

Like the case of Venus, it was a perfect example of the right world in the wrong place. Because Triton orbited four light-hours from the Sun, its surface temperature was 230 degrees below the freezing point of water. So not even nitrogen could flow liquid. Indeed, one would have expected nitrogen and water to have formed a solid ball of ice almost from the moment of creation.

As it turned out, Tuna had been too quick to abandon a perfectly good theory on the basis of a single “bad” fact—a mistake that was nonsense in its own right. Had he looked more closely, he would have seen that his idea was wrong only quantitatively, not qualitatively.

In two large, natural basins, nitrogen had indeed flowed like water and waves had indeed lapped at Tritonian shores … more than 600 million years before Tuna’s ancestors crawled out of the sea. To judge from the few craters scattered over the glasssmooth flood plains, Tuna had missed Triton’s nitrogen seas by “only” a billion years—a margin small enough to be considered a very close call by the measure of solar system time. Some powerful and still unidentified tidal force—the same force, perhaps, that had caused the partial melting of Miranda and Ariel, thrown the entire Uranian system on its side, flung Triton into its backward orbit around Neptune, and tom Pluto and Charon away from Neptune—had dumped enormous loads of frictional energy into Triton, sending seas of nitrogen bubbling to the surface. The Tritonian and Mirandan meltings were characteristic of a sudden accumulation of heat inside, and an almost equally rapid release of heat at the planetary surface.