From the motion of the stars, the time between each frame was determined to he about 200 years; the software displayed 50 frames, 10,000 years, per tau. The whole view was heavily stylized, and the image was binary: not even a gray scale, just black and white. But the software had concluded that the vertical lines attached to each star were a kind of luminosity scale, giving the distance at which the energy density of the star's radiation fell to 61 femtojoules per cubic meter, coincidentally or not, the same as the cosmic microwave background. For Voltaire, this distance was about one eighteenth of a light year; for the sun, about one seventh. The orthogonal projection enabled the "luminosity lines" for a few hundred stars to be visible simultaneously, all at the same scale; a realistic perspective from anywhere in the galaxy would have shown all but a few diminished by distance to the point of invisibility, making the intended meaning much more obscure. As the view continued to expand, though, all the stars' lines were soon reduced to identical, single-pixel specks anyway. Yatima was puzzled, but reserved judgment.

When the whole Milky Way was visible, not quite edge-on, the zoom-out stopped. Then a short vertical line appeared suddenly: twelve hundred light years long, pointing up from the plane of the galactic disk, vanishing after just one frame. Yatima had been wondering how the map would portray sources of radiation that shone for less than 200 years; the simplest method would be to match their total energy to an ordinary star's output over two centuries. On that basis, a twelve-hundred-lightyear luminosity line corresponded to a burst of radiation comparable to the output of the sun over fourteen billion years. The kind of burst produced by two colliding neutron stars.

Neutrons to warn of neutron stars? Was that another level of the isotopes' multilayered meaning?

Every two or three hundred thousand years, another burst appeared somewhere in the galaxy. Smaller lines flashed up more frequently, most of them probably supernovae; a few corresponded to known remnants. Orlando asked soberly, "So is this history, or prediction?"

"Well, from the pattern of heavy isotopes in the crust, it looks like the Transmuters processed the atmosphere at least a billion years ago." So if their predictions of these events in their far future were accurate, it would prove that they'd understood the dynamics of neutron star binaries far better than C-Z or gleisner astronomers. It was impossible to judge their record on these ancient bursts, predating even flesher gamma-ray astronomy, but if it turned out that they'd correctly anticipated the time of Lac G-1's collision, they'd have shown themselves to be extraordinarily trustworthy forecasters.

Yatima glanced at Orlando, his eyes locked on the screen. The Transmuters could promise him a flesher's eternity without another Lacerta. They could guarantee a safe return to Earth, and everything he'd once valued.

Around 100,000 years before the present, the scale began to change again. Yatima watched uneasily as the Andromeda galaxy, the whole Local Group, and then ever more distant galactic clusters came into view. Then at 26,000 BP a line appeared, almost two billion light years long, skewering the tiny Milky Way.

The image zoomed back in rapidly, just in time to show a gamma-ray burst at 2000 BP: Lac G-1. The Transmuters had correctly predicted the time of the burst to the nearest 200-year frame, and its position and energy to the nearest pixel.

Orlando remained silent as the map ran on for another twenty million years. In all that time, it showed no more gamma-ray bursts near enough to Earth to harm the biosphere.

But if the map's predictions were all equally reliable, then 26,000 years ago there'd been an event in the galactic core that rendered every ordinary burst irrelevant. In a thousand more years, the consequences would finally sweep through the region—and even if the Diaspora, the gleisners, and the Earth-based polises began to flee at once, when the pulse of radiation finally washed over them it would be thirty million times more intense than Lacerta.

Paolo said firmly, "It's not possible. You'd need six or seven billion solar masses undergoing gravitational collapse to release that much energy."

Yatima had asked to meet him to talk about Orlando, not to debate the meaning of the neutron data for the thousandth time. But Paolo seemed determined to dispose of the core burst itself before he'd listen to a word on any other subject, and maybe that was fair enough. Belief or disbelief in the event formed the ground beneath everything else, now.

"The galactic core contains more than enough mass, depending on where you draw the boundary."

"Yes, but those stars are all in orbit. They're not about to fall together into a giant black hole."

Yatima laughed humorlessly. "Lac G-1's neutron stars were in orbit, too. They weren't supposed to fall together for another seven million years. So I wouldn't stake my life on conservation of angular momentum until I found out where it all went with Lacerta."

Paolo shrugged dismissively. The burden of proof wasn't his. Even if it was being read correctly, the Transmuters' message wasn't necessarily honest; even if it was honestly intended, that didn't mean it was infallibly true. And the failure to explain Lacerta hardly meant that conservation laws could be discarded at will. If it had been a purely theoretical argument, Yatima would have happily conceded every point.

Ve glanced around the Heart, trying to gauge the mood. People were talking quietly in small groups, edgy and subdued, but far from despairing. Since the neutron data had been released, Yatima had seen as wide a spectrum of responses in Voltaire C-Z as ve'd witnessed among the fleshers when they'd heard about Lacerta. Many citizens had simply refused to accept that the core burst was a real possibility—and a few had succumbed to paranoid fantasies to rival any flesher's, declaring that the Transmuters' message had been planted in order to induce a state of panic and decay among "rival" civilizations. Others were searching for ways to survive the event. Arranging to be in the shadow of a planet could shield the polises from gamma rays, but the neutrino flux would be unavoidable, and intense enough to damage even the most robust molecular structures. The most plausible scheme Yatima had heard so far involved encoding every polis's data as a pattern of deep trenches on a planetary surface, and then building a vast army of non-sentient robots on a variety of scales, from nanoware up, so numerous that there was a chance that the relatively few survivors would he capable of reconstructing the polis.

"Suppose this burst really is on its way, and the message is a warning." Paolo settled back in his chair, and regarded Yatima amiably. "Then having gone to the trouble of creating a whole planet's worth of coded neutrons out of the goodness of their hearts, why didn't the Transmuters leave us something more than the unpalatable facts? A few survival tips might have come in handy."

"Don't give up on the rest of the data yet; it might contain all kinds of things. Preferably instructions for shortening traversable wormholes. Failing that, a reliable technique for sealing and reopening their mouths; then we could hide inside one as a stream of nanomachines until the burst is over."

Contemplating that scenario gave Yatima severe claustrophobia, but Gabriel had gone even further and suggested that the undeciphered bulk of the neutron data might be the Transmuters themselves: digital snapshots entombed in the particles in the hope that post core-burst life, once such a thing evolved, would stumble upon them and obligingly restore them to active existence. If that was the case, they'd left no obvious clues for anyone aspiring to join them in their sanctuary—and if they'd known about the burst a billion years ago, it seemed far more likely that they'd set off for another galaxy, whether by wormhole or by more conventional means.