Unless, of course, Lisa thought, she had her remaining stock taken out while she was in her twenties and stored in liquid nitrogen.

“What I tried to do,” Morgan went on, “was to introduce retroviruses into pregnant mice, aiming them specifically at the eggs within the fetal ovaries. The idea was to secure a vast collection of ready-transformed pre-oocytes, which could then be extracted from the aborted fetus. It would have been authentic mass production, on a time scale measurable in days rather than weeks, let alone the years it takes to bring transformed sheep and cows to adulthood. You can see what a boon a system like that would have been to my search for the ideal addressable vector.

“Unfortunately, it wasn’t as easy as it sounded. Nature’s genetic engineers are unreliable slaves—they have their own agendas, and a lot of those agendas are what the man in the street calls diseases: colds, colics, and cancers. The womb has it own agenda too. It has a system programmed into it, and when you have wombs within wombs, things can get very complicated. I couldn’t get effective transmission across the placenta. I had to switch my attention to newboms, although it seemed like a terrible waste. So many eggs have already gone by the time a mouse is born, and the rest are dying in droves day by day. I thought it might at least be possible to do something about the latter problem, so I modified my retroviruses yet again, incorporating a control gene that was supposed to stop the oocytes from committing suicide.

“That one worked. In fact, it worked far better than I’d hoped. In coupling it with the rest of the package, I’d somehow contrived to produce a synergistic effect—one of those million-to-one shots of which I’d always been so flagrantly contemptuous. When you have a hundred thousand genetic engineers trying out hundreds of novel gene combinations every year, though, the laws of probability will give you a million-to-one shot every month. Mine was the only one I ever got in forty years of trying, but it was a big one.

“In those days, we were only beginning to get used to the first principle of genetic engineering—you can never do just onething—so I hadn’t figured multiplicity of effect into my plans, let alone synergy, but they sure as hell came out in my results. Do you ever come across genetic mosaics in your police work?”

“Occasionally,” Lisa confirmed. Mosaics had first attracted attention when biologists contrived to fuse the embryos of two different species. The first sheep/goat hybrids had been produced in the 1990s, and the revelation had prompted people to wonder how often the same thing happened in nature. Whenever a single fertilized egg divided into two to produce identical twins, the result was obvious, but when two fertilized eggs fused to produce a single individual, there was no easy way of telling that the resultant individual was a mosaic. Until DNA analysis came along, there was no way of knowing how many cows in the bam or people walking the streets were actually patchworks of two distinct but closely related genomes. Human mosaics were even rarer than pairs of identical twins, but a world of nine billion people had to contain millions. Lisa had run across half a dozen human mosaics while conducting DNA analyses in the police lab.

“In that case, you probably know that animal mosaics were often created mechanically back in the 1990s. It was an early alternative to cloning that lost fashionability when nuclear-transfer techniques improved. The mosaics I created with the aid of my trusty retroviruses were a kind that nature had never contrived, though. My retroviruses produced a strain of mice whose egg-filled ovaries became benign cancers—not merely benign in the accepted sense that the cancers were harmless, but in a much stronger sense. The transformed eggs became capable of fusing with one another to produce zygote-like bodies that then began to grow, but not like fetuses, and not like commonplace tumors. What they did was to emit a slow but steady stream of new stem cells that could be—and were—distributed throughout the body and gradually integrated into the organs of the mothers. The mothers became, in consequence, a complex mosaic. Their complexity didn’t show up readily in the kinds of DNA analysis that Ed and I taught you to do, because the sum total of all the pesudezygote types was delimited by the original female genotype. I didn’t figure out exactly what was happening for quite a while, and I might have missed it altogether if I hadn’t started working with newborns, but that made it obvious enough that something very weird was happening.

“The long and the short of it is that the process of mosaic reconstruction stopped the aging process in its tracks. The transgenic mice were rejuvenating themselves. Initially, of course, that did my specimens more harm than good because the newborns, which remained newborns by virtue of their new power of self-renewal, couldn’t survive the interruption of their developmental processes. They died of superabundant youth. Once I’d figured out what was going on, though, I soon found out that the retrovirus could also be used to infect adults. Although the effects were variable, some of the inoculated adults were stabilized by the transformation. Their life spans were dramatically extended—and I’m not talking thirty or forty percent. In time, I found that a substantial minority were living ten or twenty times as long as their parents. A few lived a hundred times as long—and the current record holders were still extending the multiplier two days ago. Were the angels of wrath telling the truth when they said they’d torched Mouseworld?”

“Yes, they were,” Lisa confirmed.

Morgan Miller sighed again, but this time there was an element of theatre in the sigh. “It was a long time, of course, before I was convinced that even a few of the mice were authentically emortal, but the cream of the crop has stayed stable, fit, and healthy for forty years. A few were sterile, but not all. The real champions didn’t cannibalize all the fused oocytes; every now and again they gave birth to litters of daughters. Most of the offspring failed to develop, like the newborns I’d transformed myself, but a few grew to maturity before stabilizing. The selective regime progressed by degrees to the inevitable terminus: a population of emortal female mice whose daughters were likewise emortal. It took time, but when the potential’s there and the regime is stern, natural selection is no slouch.

“Long before I was convinced they were authentically emortal, I’d begun introducing the mice to the cities, for exactly the same reason that Chan wanted to introduce his augmented specimens: to see how they’d fare in a stressful and competitive situation. Mine did a little better than his—obviously, or Stella would never have found the transformed mice—but not thatmuch better, and not for a long time. When you came along in 2002,1 only had half a dozen potentially emortal mice, and nineteen of the twenty offspring they had so far produced had died paradoxical deaths of superabundant youth. By the time I moved on to experiment with other species in ’09 or thereabouts, I had a hundred adult mice and the survival rate among the new litters was up to one in three. Even then, you see, I couldn’t be surethey’d live significantly longer than normal. If I had been, I might have told you … maybe.

“It was all so gradual, so uncertain, so surprising. You should be able to imagine how tentative my conclusions were when I first knew you, how much more needed to be done before I could be confident. Stella came in on the hind end of things, when everything was set and fixed, and she never tried to imagine how it must have been in the long and confusing beginning. All she saw, when she tumbled to what was going on in London and Rome, was a secret that I had kept for forty years. And all she cared about was the obvious—she and her friends didn’t pause long enough to wonder whether there was more.”