“No,” Pitt replied. “ The Inuit account, though vague, seemed to indicate that the Erebus was farther south. Before I left Washington, I had some folks in the climatology department do some modeling for me. They attempted to re-create the weather conditions in April 1848, when the ships were abandoned, and predict the potential behavior of the sea ice.”

“So the ice didn’t just melt and the ships dropped to the bottom where X marks the spot?”

“It’s possible but not likely.” Pitt pointed to a large body of water north of King William called Larsen Strait.

“The winter freeze propels the pack ice in a moving train from the northeast, down Larsen Strait. If the sea ice off King William didn’t melt in the summer of 1848, which the climatologists suggest, then the ships would have been pushed south during the winter freeze of 1849. They might have been reboarded by a small party of survivors, we just don’t know. But it is consistent with the Inuit record.”

“Swell, a moving target,” Giordino said. “Doesn’t make for a compact search zone.”

Pitt drew his finger down the western shore of King William Island, stopping at a conglomeration of islands located twenty miles off the southwest coast.

“My theory is that these islands here, the Royal Geographical Society Islands, acted as a rampart against the southerly moving ice pack. That rock pile probably diverted some of the ice floe, while breaking up a good deal more piling up on its northern shore.”

“That is a pretty direct path from your X,” Giordino noted.

“ That’s the presumption. No telling how far the ships actually moved before falling through the ice. But I’d like to start with a ten-mile grid just above these islands and then move north if we come up empty.”

“Sounds like a good bet,” Giordino agreed. “Let’s just hope they dropped to the bottom in one piece so they’ll give us a nice, clean sonar image.” He looked at his watch. “I better rouse Jack and get the AUV prepped before we get on-site. We’ve got two aboard, so we can lay out two separate grids and search them simultaneously.”

While Pitt laid out the coordinates for a pair of adjoining search grids, Giordino and Dahlgren prepared the AUVs for launching. The acronym stood for autonomous underwater vehicles. Self-propelled devices that were shaped like torpedoes, the AUVs contained sonar and other sensing devices that allowed them to electronically map the seafloor. Preprogrammed to systematically scan a designated search grid, they would cruise a few meters above the seabed at nearly ten knots, adjusting to the changing contours as they ran.