And found that the lock had been cycled since I left.

McAndrew was sitting in the pilot’s chair, staring at the controls. I came quietly up behind him, patted him on the shoulder, and slipped into the copilot’s seat. He turned toward me, straggly eyebrows raised.

“It’s now or never,” he said at last. “But what about Tallboy? What will he do to the Institute?”

I shrugged. “Nothing. Not if we make it clear that it’s our fault.”

I reached out and called for a destination reading. When I left, the coordinates had all been set to zero. Now they carried precise values.

“Do you think that anyone else suspects?” I said. “I checked the experimental logs in your lab today, and they were all current up to this afternoon — and you’re always months behind. If I noticed that, maybe one of the others will.”

He looked surprised. “Why should they? We’ve been careful not to talk about this when anyone else could hear.”

There was no point in telling Mac that he was probably the world’s worst person you’d want to keep a secret. I tapped him on the shoulder. “No point in worrying about it once we’re on our way. Come on, Mac, move over — you’re sitting in my chair. And think positively. We’ll have a nice, long trip, just the two of us.”

He stood up, rubbing at the back of his head the way he always did when he was embarrassed. “Och, Jeanie,” he said. But he was smiling to himself as we changed seats.

The calculations were elementary, and I could do them as well as he could. The Merganser would reach the rogue planet in about sixty days of shipboard time if they kept close to maximum acceleration all the way. We could be there in thirty-five days of shipboard time, but that would pick up only ten days of inertial time. We would reach Vandell a couple of days after them. For me, that was two days too late.

Our drive wake left an ionization track across the whole width of the Solar System. Mac checked that there were no ships directly behind for us to burn a hole through, and while he was doing it I had a new idea and sent a message back to External Affairs. I said that we were about to perform a brief high-gee test of the Hoatzin’s drive before we took her in and decommissioned her. With luck, Tallboy’s group would assume we had been the unhappy victims of a nasty accident, shooting out of the Solar System on a one-way journey when some control element of the drive unit had failed. Limperis and friends at the Institute wouldn’t believe that, not as soon as they checked our destination coordinates — but they would never tell their suspicions to Tallboy. Maybe they could even get some mileage from our disappearance, pointing out the need for more funds for reliability and system maintenance. Limperis could play that game with his eyes closed.

Perhaps everything would work out fine — until McAndrew and I came back. Then the truth would come out, and we’d be roasted for sure.

Neither of us could get too worried about that possibility. We had other things on our minds. As we raced out along the invisible scintillation of the Merganser’s drive, Mac dumped the data bank for information about Vandell’s rogueworld. He didn’t get much. We had coordinates relative to the Sun, and velocity components, but all they did was make sure we could find our way to the planet. Wicklund had been able to put an upper limit on its diameter using long base line interferometry, and estimated that we were dealing with a body no bigger than Earth. But we were missing the physical variables — no mass, internal structure, temperature, magnetic field, or physical composition, not even an estimate of rotation rate. Mac fumed, but at least I’d have a lot more information for him as soon as we got close. In the week before we left the Institute, I had put on board the Hoatzin every instrument that wasn’t nailed down, anything that might tell us something useful about Vandell without having to go down there and set foot on its surface.

At a hundred gees acceleration you head out of the Solar System on a trajectory that’s very close to a straight line. The gravitational accelerations produced by the Sun and planets are negligible by comparison, even in the Inner System. We were bee-lining for a point in the constellation Lupus, the Wolf, where Vandell lay close in apparent position to an ancient supernova fragment. That explosion had lit up the skies of Earth more than a millennium ago; an interesting object, but we wouldn’t be going even a thousandth of the way out to it. Wicklund was right; Vandell’s rogueworld sat in Sol’s backyard.

Without a complicated trajectory to worry about, I went round and round with a different problem. When the drives were on, both the Merganser and the Hoatzin were blind to incoming messages, and drowned out any of their own transmissions. Thus we had a chance to get a message to Sven Wicklund and Jan only when their drive was turned off, while they were coasting free to rubberneck or study the starscape scenery from a slightly different point of view. Even though they might not be listening for an incoming signal when the drive was off, their computer would, and should notify them of anything important.

But now see my problem: to send a message, we had to switch our drive off, and that would delay our arrival a little bit every time we did it. Our signal would then take days or weeks to reach the Merganser — and to receive it, their ship had to have its drive off at just the right time. DON’T LAND was all I wanted to say. But how would I know when to switch off our drive and send an urgent message, so it would get to them just when their drive was not operating?

I wrestled with that until my brains began to boil, then handed it over to McAndrew. He pointed out that we had knowledge of the occasions when their drive had been switched off, from the gaps in their drive wake. So making a best prediction was a straightforward problem in stochastic optimization. He solved it, too, before we had been on our way for a week. But the solution predicted such a low probability of successful contact that I didn’t even try it — better to leave our drive on full blast, and try to make up some of their lead.

With the shields on to protect us from the sleet of particles and hard radiation induced by our light-chasing velocity, we had no sense of motion at all. But we were really moving. At turnover point we were within one part in ten thousand of light speed.

If I haven’t said it already, I’ll say it now: the 100-gee balanced drive is nice to have, but it’s a son of a bitch — you travel a light-year in just over a month of shipboard time. Two months, and you’ve gone fifty light-years. Four shipboard months, and you’re outside the Galaxy and well on your way to Andromeda.

I calculated that two hundred days would put you at the edge of the Universe, 18 billion light-years out. Of course, by the time you got there, the Universe would have had 18 billion more years to expand, so you wouldn’t be at the new edge. In fact, since the “edge” is defined as the place where the velocity of recession of the galaxies is light-speed, you’d still be 18 billion light-years away from it — and that would remain true, no matter how long you journeyed. Worse still, if you arranged a trajectory that brought you to rest relative to the Earth, when you switched off the drive the Galaxies near you would be rushing away almost at light speed…

An hour or two of those thoughts, and I felt a new sympathy for Achilles in Zeno’s old paradox, trying to catch the tortoise and never quite getting there.

Travel for a year, according to McAndrew, and you’d begin to have effects on the large-scale structure of space-time. The vacuum zero-point energy tapped by the drive isn’t inexhaustible; but as to what would happen if you kept on going…